Report No. 39021-TZ Tanzania Sustaining and Sharing Economic Growth Country Economic Memorandum and Poverty Assessment (In Two Volumes) Volume I: Main Report March 1, 2007 The World Bank The United Republic of Tanzania Poverty Reduction and Economic Management Unit 2 Ministry of Planning, Economy, Africa Region and Empowerment Document of the World Bank Sustainingand SharingGrowthinTanzania i Table of Contents ACKNOWLEDGEMENTS ............................................................................................................. iv SUMMARY ....................................................................................................................................... 1. POVERTYREDUCTIONAND GROWTH. RECENTPERFORMANCEAND A Decadeof Structural Reforms, Macro-economicStabilityandEconomicGrowth.................1 PROSPECTS .............................................................................................................................. 1 Economic Growth, Poverty, andInequality .............................................................................. 10 Outlook on GrowthandPovertyReduction .............................................................................. 14 ProgressTowards the MillenniumDevelopment Goals ............................................................ 16 2. ENHANCINGINTERNATIONALCOMPETITIVENESSAND ACCELERATING DIVERSIFICATION .................................................................................................................. 19 FosteringGrowth, ExportCompetitiveness, andEmploymentinTanzania's Manufacturing Sector......................................................................................................................................... 19 Improvingthe Business Environment ....................................................................................... 20 FosteringInnovation,Productivity, andTechnologicalChange ............................................... 28 3. MAKINGGROWTHPRO-POOR ............................................................................................ 38 SupportingAgriculture Sector Development ............................................................................ 38 HarnessingNatural Resourcesfor SustainableGrowth SupportingThe Informal Sector andMicro, Small, andMediumSize Enterprises..................47 ............................................................ Strengtheningthe Capacity of the Poor to ParticipateandContributeto Economic Growth...59 53 4. MANAGINGPOLICIESAND RESOURCESFOR SUSTAINED SHAREDGROWTH ......65 StrengtheningInstitutionsto Steer the Development andImplementationofTanzania's Strategy for GrowthandPovertyReduction ............................................................................. 65 Scaling up Public Spendingfor the Implementationofthe NSGRP ......................................... 67 REFERENCES ................................................................................................................................ 73 ANNEX 1: LIST OF BACKGROUND STUDIES ......................................................................... 78 Sustainingand Sharing GrowthinTanzania 11 .. LIST OFBOXES Box 1. Overview o f Structural Reforms inTanzania ..................................................................... .................................................................... 2 5 Box 4. Review o f Experience with Export Processing Zones (EPZs).......................................... Box 3. BenchmarkingTanzania inthe Global Context .................................................................. Box 2. Government Spending and Economic Growth 8 Box 5. Underlying principles for the design o f an agricultural technology system...................... 10 Box 6. Examples o f voluntary formalization................................................................................ 41 57 LIST OFFIGURES Figure 1 Growthrates ofGDP incl and excl government spending. 1990 .2005 Figure 2 Contribution o fpublic and private expenditure to economic growth. 1990 .2005 ........45 . . . . ....................... Figure 3. Decompositiono f economic growth per worker into contribution o f human and physical capital accumulation and total factor productivity (TFP). 1985.2005 .............................. 6 ................................................. Figure 6. Cost o f inefficienciesinbusiness environment as percent o f sales, various countries..21 Figure 5. Projected changes inincome poverty and malnutrition Figure 4. Simulated changes inpoverty. 1992-2002 .................................................................... 12 17 Figure 8. Domestic credit and interest rates, 1990-2005 Figure 7. L o w levels o f infrastructure development constrain economic growth......................... 22 . .............................................................. 24 Figure 10. Requests for bribes are especially common duringtax inspections ............................ Figure 9 Savings and Investment (% o fGDP), 1990-2005 .......................................................... 24 Figure 11. Adult Literacy Rates, 1960-2000 ................................................................................ 28 Figure 12. Average Years o f Schooling, 1970-2002 .................................................................... 29 Figure 13. Marginal social returns per year o f education based on integrated labor force survey31 30 Figure 14. Predicted earnings inmanufacturing sector based on manufacturing firm surveys ...- 31 Figure 15. Labor productivity levels inTanzania and comparators ............................................. 39 Figure 17. Size o f the informal economy for selected countries (as % o f GNI)........................... Figure 16. Cost components o f marketingmargins vary significantly with crop and location....44 Figure 18. Median value addedper worker .................................................................................. 53 Figure 19. Changes inthe distribution o f accessto education, rural Kilimanjaro ........................ 57 61 Figure20 Multilateral credit disbursements andDebtSustainability: Net Present Value o fDebt- . to-Export Ratio, 2006-26 .............................................................................................................. 70 LISTOFTABLES Table 1. Elements o f a growth strategy ..................................................................................... xvlii ... Table 2. Key Economic Indicators. 1995-2005 .............................................................................. Table 3. Sources o f Growth (production), 1990 - 2005 .................................................................. 1 Table 4. Sources o f Growth (expenditure), 1990 - 2005 ................................................................ 3 Table 5 Zonal Poverty Status....................................................................................................... . 4 .............................................................................. 11 Table 7. Policy basedgrowth projection....................................................................................... Table 6. Decomposition o f change inpoverty 13 Table 8. Scenarios for economic growth and structural transformation ....................................... 14 15 Table 9 MDGbaseline, most recent estimate and target . ............................................................. Table 11. Role o f the Public Sector inFostering Innovation........................................................ Table 10.Estimated size o f the informal economy (%-of GDP). various countries and regions...26 16 Table 12. I C T Indicators (per 1.000 people). 2002....................................................................... 33 36 Table 14. Institutional Framework for Sustainable Development o f Smallholder Irrigation Table 13. Land use and potential for land expansion (mid 1990s) ............................................... 39 Systems .......................................................................................................................................... 43 Sustaining and Sharing GrowthinTanzania ... 111 ............................................................... no-education .................................................................................................................................. Table 16. Increase inper capita consumption relative to households headedby individuals with60 Table 15. Typology o f forms of enterprise inTanzania 54 62 Table 18. Differences inhealth outcomes by quintile .................................................................. Table 17. Number of days missed due to illness, by quintile ....................................................... 62 Sustaining and SharingGrowthinTanzania iv Acknowledgements This report was prepared in collaboration between the Tanzanian Government, Ministry o f Planning, Economy, and Empowerment (MPEE) and the World Bank with support from the African DevelopmentBank and various local and international researchers. It was prepared under the overall supervision o f Dr. Enos Bukuh (Permanent Secretary, Ministry o f Infrastructure Development, previously Permanent Secretary, MPEE), Charles Mutalemwa (Permanent Secretary, MPEE), Judy O'Connor (Country Director for Uganda and Tanzania), and Kathie Krumm(Sector Manager, Poverty Reduction and Economic Management Unit for East Africa), who provided substantive inputs,comments, and support at all stages o f the preparation process. The task manager and principal author o f the report i s Robert Utz. The Government team was initially led by Arthur Mwakapugi (Permanent Secretar, Ministry of Energy and Mining, previously Director for Macroeconomics, MPEE), followed by Laston Msongole (Director o f Macroeconomics, MPEE). Substantive inputsand background papers were preparedby Johannes Hoogeveen, (poverty, nutrition, and strengthening the capacity o f the poor to contribute to growth), Louise Fox, and Marianne Simonsen (poverty assessment), Meera Shekar (nutrition), Henry Gordon (agriculture), Anuja Utz and Jean-Eric Aubert (education, innovation, and ICT), Michael Wong (business environment), Ravi Ruparel (financial sector development), Vandana Chandra, Pooja Kacker, Ying Li (manufacturing sector analysis), Philip Mpango (spatial dimensions o f growth), Allister Moon (institutional reform), Jack Ruitenbeek, Cynthia Carter (natural resource management) and Emmanuel Mungunasi (data and statistical annexes). Peter Mwanakatwe (African DevelopmentBank) contributed the analysis o f infrastructure. Mary-Anne Mwakangale and Arlette Sourou provided dedicated logistical support. Arlette Sourou was also responsible for word processing and physical production o f the report. background papers by Annabella Skof (consultant - tourism), Kirsten Pflieger (COW1 consultants The Governments o f Austria, Denmark, Netherlands and Sweden financed the preparation o f -naturalresourcebasedgrowth), MeinePietervanDijk(UNESCOInstituteforWaterEducation -Analysis oftherole ofurbanareas), andJorgenLevin(consultant - CGE analysis). InfoDev financed a study on growth, competitiveness, and ICT carried out by OTF consultants. Annex 1 presents the list o f background studies prepared for this report by local and international researchers and by World Bank staff. Insightfuland challenging comments were provided at various stages o f the preparation process by peer reviewers BennoNdulu(World Bank), Josephat Kweka (ESRF), ErikThorbecke, Steven Younger, and David Sahn (Cornel1 University), and Dani Rodrik (Harvard University). During the consultations held in Dar es Salaam, Haidari Amani (ESRF), Brian van Arkadie (ESRF), Amon Mbelle (ERB), Peter Noni (BOT),Adolf Mkenda (University o f Dar es Salaam), Robert Mbelle (University o f Dar es Salaam) served as discussants o f the draft background papers. Detailed comments were also provided by members o f the World Bank Country Team, including Keith Hinchliffe (education), Dieter Schelling (transport), Karen Rasmussen, Arun Sanghvi, and Duncan Reynold (energy), Mavis Ampah (ICT), Mathew Glasser (decentralization and local government), and Denyse Morin (institutional reforms). Sustaining and Sharing GrowthinTanzania V The C E M benefited from the participation in the "growth path" project led by Roberto Zagha (World Bank) in cooperation with Harvard University that helped to sharpen the growth diagnostic. The preparation o f the report included several rounds o f consultations in Tanzania organized by the Ministry o f Planning, Economy, and Empowerment. Initial consultations in September 2003 and July 2004 served to define and agree on the scope and focus o f the study as well as on collaborative arrangements. The main mission took place in November 2004 and included field visits to Kigoma, Lindi, and Mtwara. In March 2005, a series of workshops in Dodoma, Morogoro, Moshi, and Dar es Salaam were organized to obtain feedback and inputs on the draft background papers prior to the drafting o f the main report. At that stage, the Ministry of Planning, Economy, and Empowerment organized a review meeting with Permanent Secretaries and senior officials from a large number o f ministries for a briefing on the consultations and the discussion o f emerging main messages and recommendations. The team would like to express its sincere gratitude to all who have provided valuable comments and inputs during these consultations. The report i s presented in two volumes. Volume Isummarizes the main findings and recommendations. Volume I1contains the main report. Note: Government i s currently revising its National Accounts data, including changing the base year for constant price data series from 1992 to 2001. The revised National Accounts data are expected to be released in2007. Sustaining and SharingGrowthinTanzania vi Summary Tanzania's National Strategy for Growth and Reduction o f Poverty (NSGRP) sets an ambitious target o f 6 to 8 percent annual economic growth to achieve rapid reduction inpoverty. This report focuses on three issues that are central to the success o f Tanzania's poverty reduction efforts: 0 What factors explain Tanzania's recent acceleration ineconomic growth? 0 Has the accelerated economic growth translated into reducedpoverty? 0 What mustbe done to sustain economic growth that i s pro-poor? The report presents evidence from the macroeconomic, sectoral, and firm and household levels that shed light on these questions. This summary provides an overview o f the main findings and recommendations. What Factors Explain Tanzania's Recent Acceleration inEconomic Growth? The average annual growth o f Tanzania's gross domestic product (GDP) o f 6.0 percent during 2000 to 2005 has been high, not only compared with its own historical growth performance but also compared with international growth rates. Growth rates increased across all sectors, with industrygrowingby 8.7 percent, services by 5.9 percent, andagriculture by4.8 percent duringthe same period. Mining(growth rate o f 15.2 percent), construction (10 percent), manufacturing (7.0 percent), and trade hotels and restaurants (6.9 percent) were the fastest-growing sub-sectors. The contribution o f the various sectors to growth, which depends on both the growth rate o f the sector and its share in the economy, shows that agriculture contributed 2.3 percentage points, services 2.1 percentage points, and industry 1.6 percentage points o f the average annual growth o f 6.0 percent during 2000 to 2005. The analysis o f the sectoral contributions to the increase in the average GDP growth rate from 2.5 percent during 1990 to 1994 to 6.0 percent during 2000 to 2005 confirms that growth accelerated in all sectors. Growth in the service sector contributed 1.4 percentage points to the increase, industry 1.3 percentage points, and agriculture 0.8 percentage points. The implementation o f a comprehensive set o f macroeconomic and structural reforms laid the foundation for the recent growth acceleration. These reforms enhanced the incentives for private sector activities and led to improved efficiency o f resource allocation and use in the economy. The domestic and foreign private sectors as well as Tanzania's development partners reacted to the improvements in the economic and incentive regime in a variety o f ways that explain the increase in economic growth. A central element o f Tanzania's recent growth performance i s large inflows o f private and public capital that were triggered by the reforms undertaken by government. The transition o f Tanzania to a market economy began inthe mid-1980s with an initial focus on the liberalization o f the economy through the removal o f constraints on private sector activities and the abolition o f controls on prices and exchange and interest rates. The reforms also included Sustaining and SharingGrowth inTanzania vii a restructuring o f the public sector and an ambitious privatization program. Inthe mid-l990s, the reform agenda was augmented by a strong focus on macroeconomic stability and the quality o f public financial management. Initially, this effort involved sharp cuts in government expenditures to minimize the government's domestic and nonconcessional borrowing. These cuts served as the basis for a prudent monetary policy that reduced the rate o f inflation to well below 10 percent. Subsequently, reform efforts focused on improving Tanzania's tax system and public financial management to improve allocative and operational efficiency o f public expenditures and to minimize resource leakages. An important result o f prudent monetary and fiscal policy, combined with financial sector reforms, is the recovery o f credit to the private sector which grew by more than 30 percent annually in recent years. The environment for economic growth i s thus vastly improved, and current government efforts are targeting higher levels o f investment in human capital and physical infrastructure, improvements in the business environment, and strengthening o f government capacity. The intensification o freforms since 1995 and improvements inthe business environment, as well as sector-specific reforms-especially in the miningsector-have triggered an increase in foreign direct investment (FDI) and aid inflows. FDI has increased rapidly since the mid-1990s and reached about US$542 million or 5 percent o f GDP by 1999, partly driven by large investments inminingand privatization-related investments. Following the completion o f major investments inthe miningsector and the major privatizations, FDIdeclined to US$473 million or 3.9 percent o f GDP by 2005, a level that i s still highin comparison with that o f most other African countries. The sectors that received the bulk o f the FDIshowed the highest growth rates, including mining, manufacturing, and trade and tourism, which together attracted about 75 percent o f FDI during 1999 to 2001. The reforms implemented by the government also triggered a continuous increase in aid inflows that, together with improved domestic revenue collection, supported the increase in government spending from 16 percent o f GDP in 1999/2000 to 26 percent in 2005/06. National accounting statistics suggest that this increase in government spending contributed significantly to the acceleration in economic growth. Inthe short term, the increased demand for goods and services by the government led to increaseduse o f available capacity. For example, the rehabilitation and expansion o f administrative, economic, and social infrastructure are reflectedinthe fast growth o f the construction sector by about 10 percent annually during 2000 to 2005. Fast growth in the service sector i s also partly related to increased government expenditures. In addition to these direct effects o f increased government spending, traditional multiplier effects translate increases in government spending into increased demand for goods and services in all sectors. In the medium to long term, if government spending contributes effectively to the building o f human capital and the expansion o f economic infrastructure, then sustained levels o f increased government spending have the potential to expandthe productive capacity o f the economy. A noteworthy development is the rapid growth o f the informal sector-particularly in Dar es Salaam-as the result o f various factors. These include the liberalization o f the economy, the tolerance o f many informal sector activities that were previously illegal, the need for laid-off government workers and migrants to generate new income-earning opportunities, and the increased demand for informal sector products and services as a trickle-down effect from growth inthe formal economy. Another significant economic development during the past decade has been the rapid expansion o f miningand gold exports, whose share in total exports increased from 4 percent in 1998 to 56 percent in 2005. However, the contribution o f miningto overall growth was only 0.4 percentage points, reflecting the relatively small size o f the mining sector, as well as the high import Sustaining and Sharing GrowthinTanzania ... VI11 dependence o f the sector for machinery and its very limited domestic backward and forward links.Aside from gold, fish, and tourism, the value of exports remains low and volatile. Between 1995 and 2001, the real effective exchange rate appreciated by almost 50 percent and then returned to its 1995 level. The real appreciation reduced the competitiveness o f Tanzania's tradables sector and merchandise exports declined between 1995 and 2001. Despite a subsequent recovery in parallel to the recent real depreciation, to date, exports other than gold and fish have played a relatively small role as a dynamic source o f growth and have seen little diversification. Thus, a key challenge for the Tanzanian economy is to strengthen its export competitiveness. Doing so would ensure that, aside from the dynamic growth effects o f a strong export sector, exports will provide an important demand stimulus for the economy, especially because the scope for continued increases o f government spending as the primary demand stimulus i s clearly limited. The analysis o f factor inputssuggests that the acceleration ineconomic growth is primarily due to an increase in cultivated land in the agriculture sector and increased factor productivity for the other sectors. The increase in total factor productivity reflects both increased capacity use in response to increased aggregate demand and economic efficiency gains in the wake o f the removal o f economic distortions. Innovation and technological change have so far played only small roles in improving Tanzania's total factor productivity, mainly in the form o f FDIbut also as some encouraging innovations emerging from the agricultural research system. At the firm level, there i s some evidence that the structural reforms have resulted in a more dynamic and competitive private sector. Increased competition in the private sector i s evidenced by an increasing number o f firms exiting and entering the market. The fact that firms entering the market are typically more competitive than those that exit i s an important driver o f the increase in total factor productivity registered at the aggregate level. Public investment has recovered from an average o f about 3 percent o f GDP duringthe late 1990s to about 8 percent of GDP in recent years. The analysis o f public investment suggests, however, that only about one-third o f it was used on public infrastructure such as roads or electricity, while the remainder was devoted to the rehabilitation and expansion o f administrative and social infrastructure. Private sector investment had been stagnant at about 11 percent until 2002 but increasedto 14percent by 2005, reflecting increased investor confidence inresponse to sustained implementation o f investor friendly reforms and increased demand. Although the contribution of human capital accumulation to economic growth has been relatively small, the recent increases in school enrollment can be expected to be reflected inhigher economic growth inthe future. Drawing on the review o f Tanzania's recent growth performance, the report assesses the prospects for sustained high growth and the key challenges that need to be addressed. Policy- based growth projections suggest that growth o f 6 to 8 percent per year i s feasible. However, some o f the factors behind the recent growth acceleration are unlikely to be sustainable in the medium to long term. The demand-side impulses o f foreign aid and government spending depend on ever-increasing amounts o f aid and government spending. There i s also a clear limit to the extent that agricultural production can be increased solely by increasing the land under cultivation. Signs o f environmental and social stress (especially between pastoralists and agriculturalists) o f increased land use already exist in some areas o f Tanzania. Similarly, the effect o f reform-induced efficiency gains on economic growth will diminish when the higher level o f efficiency has beenreached. Thus, for Tanzania to achieve sustained high growth, increases in government spending and expansion o f land under cultivation need to be gradually replaced by increased productivity, saving, and investment by the private sector as primary drivers o f growth. Sustained economic SustainingandSharingGrowthinTanzania ix growth will depend on the ability o f the economy to diversify and to increase its international competitiveness. Diversification requires efforts both to enhance the capacity to innovate and to find new areas o f economic activity where Tanzanian enterprises can successfully compete. Enhancing international competitiveness requires measures that enhance productivity and reduce the cost o f doing business at the microeconomic level and macroeconomic policies that ensure a competitive exchange rate as well as interest rates and access to capital that are not distorted by highpublic demand for funds. Has the Accelerated Economic Growth Translated into Reduced Poverty? Sustained economic growth i s critical to achieving progress in poverty reduction. The mechanisms through which the poor contribute to and participate in economic growth include the following: 0 Increased incomes from the main sources o f livelihood o fthe poor; 0 New income-generating opportunities for the poor; 0 Reduced vulnerability to shocks that affect the incomes o f the poor; 0 Increased government revenue for pro-poor expenditures; and 0 Increasedprivate transfers and strengthened social safety nets. Inaddition, the report examines the effectiveness ofmeasures that support the poor inefforts to accumulate human and physical capital, which would enhance their prospects o f contributing to economic growth. Modest per capita GDP growth rates during the early and mid-1990s resulted in equally modest poverty reduction. In 2001, government estimates show 35 percent o f the population living in poverty. The potential effect o f the recent GDP growth acceleration has not yet been captured in available poverty data. Ownership o f assets such as improved housing, radios, and bicycles by the poor has also increased. The expansion o f access to free primary education has also clearly benefited the poor. The analysis o f growth incidence suggests that expenditures o f all income groups grew at about the same pace, probably because growth in agriculture, which i s the source o f income for most o f the poor, was similar to growth in other sectors during 1991 to 2000. Since 2000, growth in the industry and service sectors has been higher than in the agriculture sector, which may have caused an increase ininequality. The Household Budget Survey data show large regional differences in poverty reduction. Although poverty dropped from 28.1 percent to 17.6 percent in Dar es Salaam, in other urban areas poverty declined only from 28.7 percent to 26 percent and in rural areas from 40.8 percent to 38.7 percent. The faster pace o f poverty reduction in Dar es Salaam reflects Tanzania's pattern o f growth. Inparticular, Dar es Salaam accounts for about 50 percent o f the FDIstock and flows, and as the seat of central government and most donor agencies, it also benefits disproportionately from the increase in aid inflows. Although central government expenditures increased from 18 percent to 25.6 percent o f GDP between 2000 and 2005, transfers to local governments increased only from 2.9 percent to 3.3 percent o f GDP during that period. Growth in the formal sector in Dar es Salaam also supported an increase in the size and incomes in the informal sector, which contributedsignificantly to poverty reduction duringthe period from 1991/92 to 2000/0 1. More than 80 percent o f Tanzania's poor derive their livelihoods from agriculture. Between 1991 and 2000, the agriculture sector grew by an average o f 3.5 percent, which suggests per capita Sustainingand SharingGrowthinTanzania X growth o f less than 1 percent. The increase in per capita expenditure by farm households i s equally modest at 7.3 percent during the period from 1991/92 to 2000/01. Nonetheless, because most o f the poor derive their livelihood from agriculture, this modest increase explains more than half o f the total decline in poverty observed during that period. Between 2000 and 2005, growth in the agriculture sector accelerated to an average o f 4.8 percent annually, which according to poverty simulations i s likely to have generated a further drop in rural poverty. The study argues that given Tanzania's agricultural potential, there i s significant scope for reducing poverty by measures that would foster growth inagriculture and thus the incomes o f farmers. Another path out o f poverty i s the movement from agriculture to other sources o f income, possibly combined with migration from rural to urban areas. Data suggest that the shift from agriculture to nonagriculturalactivities inrural areas has been an important contributor to poverty reduction. Informal sector activities have been an important entry point for the poor to engage in nonagricultural activities. Rural-urban migration has also contributed to poverty reduction. However its quantitative significance was less than that o f the other channels, probably because most o f the migrants are from households above the poverty line. However, migration i s only one path in which fast urban growth can benefit the poor in rural areas. Indirect channels include higher demand for rural products, wage effects, and transfers. However, the fact that rural growth and poverty reduction lag significantly behind urban growth and poverty reduction suggests that these links are still weak. However, urban-rural links can also result in a deepening o f urban-rural differences. There i s evidence that rural migrants are typically better educated than the average rural population, leading to a widening o f the education gap as these migrants move from rural to urban areas. A large share o f financial savings collected by banks in rural areas flows toward Dar es Salaam, funding credit to the private sector and government in Dar es Salaam, as well as overseas investments by the banking sector. Although the mobility o f human and financial resources toward opportunities where the returns are highest i s supportive o f high economic growth in Tanzania, measures that counteract an increasing marginalization o f the rural poor in Tanzania's growth process are needed. Such measures would include enhanced rural access to quality education and policies that support agriculture. The report highlights instances in which the lack o f integration o f rural areas in the economy significantly reduced rural growth. Key among these instances i s the access o f rural areas to markets as well as to agricultural inputs.For example, surveys carried out in the Kilimanjaro and Ruvuma regions suggest that lack o f access to agricultural inputs results in low agricultural productivity and, consequently, limited progress inrural poverty reduction. This limited access to agricultural inputs is the result o f two equally important problems: (a) limited access to input credit; and (b) lack o f a rural input supply infrastructure that would allow farmers to purchase these inputs. These results suggest that rural development and informal sector activities are the primary direct drivers o f poverty reduction in Tanzania, where the informal sector has been an important transmission mechanism that allowed the poor to participate in economic growth opportunities originating in the formal and pubic sectors. This interpretation i s reinforced by the fact that although economic growth was significantly higher inurban areas than inrural areas inthe period from 1990/01 to 2000/01, modest rural growth has clearly dominated the faster urban growth with respect to its effect on poverty reduction. Furthermore, even in an environment o f relatively high growth differences between rural and urban areas, the contribution o f migration and other urban- rural linksto poverty reductionhas been relatively modest. Sustaining and Sharing GrowthinTanzania xi Appropriate tax and public expenditure policiesplay an important role in fostering shared growth. As the report highlights, enhancement of the domestic revenue base through sustained economic growth i s central to the sustainable financing o f public expenditures and reduction o f aid dependence in the medium to long term. Inturn, tax policies have a direct influence on the level o f investment and economic activities. Similarly, public expenditures play an important role not only in improving the environment for economic growth, but also in enhancing the access and quality o f public services for the poor. Overall, the Tanzanian tax policy i s assessed as being sound and not inimical to growth. However, several measures could enhance the contribution o f the tax system to fostering shared growth. First, there remains an urban bias in the tax system: effective tax rates are higher for farmers than for businesses, which are mostly urban. Inparticular, the crop cess collected by local authorities imposes a relatively heavy tax burden on agriculture. Second, the presumptive tax regime for small businesses i s one o f the most sophisticated in the region. Nonetheless, it i s regressive for small businesses that do not keep records. Third, there are significant weaknesses inthe taxation ofnatural resources, which result inboth distortions to the sustainable exploitation o f natural resources and suboptimal collection o frevenue. Social sector expenditures have seen significant increases in recent years. However, incidence analysis suggests that only in the education sector have public expenditures been pro-poor. In other sectors, such as water, primarily nonpoor households benefited from improved quality and access to services. The focus on social expenditures also limited the availability o f funds for growth-enhancing expenditures. What MustBeDoneto SustainEconomicGrowthThat I s Pro-Poor? The review o f Tanzania's recent growth performance suggests that enhancing the pace of structural change and diversification and increasing the international competitiveness o f the economy remain the key challenges for sustaining growth. The poverty analysis highlights the importance o f a productive agriculture sector and o f a conducive environment for the activities o f small and medium enterprises (SMEs) as key elements o f a shared-growth strategy. It also emphasises that participation o f the poor in the growth process requires that policies support the accumulation by the poor o f primarily human capital and physical and financial capital. Finally, the report underscores the importance o f appropriate policies and institutions to manage the design and implementation o f a shared-growth strategy, as well as resources for their implementation. Here the report discusses not only the management o f economic policies and public finances, but also the equally important management o f natural resources. Improved governance o f Tanzania's natural resources, strengthened capacity to ensure that tax and expenditure policies are supportive o f a shared-growth agenda and better institutional coordination framework for development and implementation o f a growth strategy require attention inTanzania's quest for sustainable shared growth. To sustain and share economic growth across all income groups o f society, Tanzania will need to preserve achievements, consolidate ongoing reforms, and strengthen institutional capacity for both policy advice and program implementation. It will also be important to guard against backsliding in the face o f pressures from vested interests or impatience with the pace o f poverty reduction. The sectoral distribution o f growth has a significant effect on the pace o f poverty reduction and inequality. Tanzania's comparative advantage in agricultural production and its large potential to enhance agricultural productivity provide a good basis for a focus on agriculture and agriculture- Sustaining and SharingGrowthinTanzania xii related activities as the central element o f its efforts to reduce poverty. Agricultural activities are the source o f livelihood for 75 percent ofthe population, more than 40 percent ofwhom are poor. Efforts to reduce poverty must focus on measures that will help the poor to (a) generate more income from their current agricultural products; (b) shift their production to more profitable agricultural products; and (c) shift to income-generating opportunities outside o f agriculture in both rural and urban areas. In addition, a decline in prices attributable to productivity increases benefits poor people who are net buyers o f agriculturalproducts. Increasing agricultural incomes requires policies that target both improvements in market access and increases inagricultural productivity. Enhancing rural infrastructure remains critical to ensure market access for farmers. It i s equally important to ensure that institutional arrangements are in place that link farmers to domestic and international markets. An example o f reforms in this area would be to ensure that the crop boards function efficiently, with a clear separation o f public and private functions, and that they are accountable to farmers. Regulations such as mandatory auctions and single license rules for coffee potentially harm the efficiency o f markets and reduce farm incomes. Scaled-up investment in agricultural research and a reform o f Tanzania's extension service have important roles to play in supporting farmers in the move to raising crops that yield higher returns. The study highlights the large productivity losses attributable to human diseases that make health intervention an important element o f efforts to increase agricultural productivity. In addition to improving farm-level productivity, the focus needs to be on promoting upstream activities such as agroprocessing and enhancing links to domestic and foreign markets. Efforts toward raising agricultural productivity must also take into account the general question o f the impact that risks have on agricultural activity in general and the more specific consideration o f the capacity o f the poor to grow out o fpoverty. Increased income-generating opportunities in nonagricultural activities, especially in rural areas, are also important for poverty reduction and for the medium- to long-term structural transformation o f the economy. Providing opportunities for Tanzanians to move out o f the agriculture sector can be expected to improve the labor productivity in the sector and to provide higher incomes for those moving out o f the sector. MSMEs, often in the informal economy, provide an important entry point for the poor to engage in industrial and service sector activities. Measures that support MSMEs are thus important, including and public recognition, tolerance, and support for informal sector activities. Formalization should be primarily incentive based in the case o f microenterprises. The report highlightsthe importance o f the manufacturing sector as a potential dynamic driver o f diversification and growth. Analysis o f firm-level data from the enterprise survey suggests that, in order o fpriority, the leading five factors that affect firm growth and that deserve special attention by policy makers are (a) access to and cost of financial capital; (b) access to technology to improve productivity; (c) infrastructure, especially energy; (d) skilled labor; and (e) the regulatory environment for business activities. Because firm growth i s intricately tied to growth in exports, an aggressive and proactive policy stance promoting manufactured exports i s likely to have the greatest effect on manufacturing growth in Tanzania and i s recommended. The rationale for this selective approach i s motivated by today's global reality: if a firm cannot compete in the global market (that is, if it cannot export), it i s unlikely to survive too long in Tanzania's domestic or Africa's regional markets, which are flooded with cheaper imports from low-cost and high-skills producers such as those from East and South Asia. The policy implications o f an export-oriented stance have several Sustaining and Sharing GrowthinTanzania xiii overlaps with factors that promote growth in non-exporting firms, but an aggressive focus on incentives that facilitate the expansion o f existing firms and promote new entrants in the export sector i s likely to yield the most benefits. Priority areas that require improved policies or scaling up o f expenditures include investing in infrastructure; enhancing access to finance, notably for the rural and SME sectors; and continued efforts to streamline regulation, reduce bureaucracy, and fight corruption. Focus on these issues promises increased economic activities in areas that the private sector can readily support. This advice may seem tantamount to recommending everything-that is, redressing all barriers to production presently facing all manufacturing firms in Tanzania. But it i s not. To circumvent the high financial and time costs and the government's limited implementation capacity, the report recommends focus and pragmatism in catering to existing and potential exporters. A sound strategy for delivering physical inputs (such as infrastructure) and financial inputs (which will make bank finance more accessible) includes the identification o f spatial locations where export activity i s most prevalent and where exporters most likely to locate. The government i s pursuing this strategy through export processing zones and special economic zones. However, in rolling out this strategy, it will be important to sequence these activities by initially addressing problems o f existing zones before new ones are created. A review o f Tanzania's export processing zones highlightsinfrastructure weaknesses, especially reliable access to electricity and water, as a main constraint for firms located inthose zones. The targeted improvement o f infrastructure services to the manufacturing sector thus needs to be a priority in Tanzania's efforts to spur growth and structural transformation. Spatial targeting helps intargeting exports. This approach would render public support in a financially feasible and timely manner for fast-growing exporters and potential new entrants into the export business. Sustained economic growth will increasingly depend on the capacity o f economic actors to innovate, to produce an increased array o f goods and services, and to accelerate the pace o f technological change. It will require foremost a greater focus on investment in human resource development (and vast improvement in secondary, technical, and tertiary education), as well as strengthening o f the innovation environment and Tanzania's fledgling information and communicationtechnology (ICT) infrastructure. Tanzania's natural resource endowment could be an important source o f growth and poverty reduction. Strengthening the governance o f natural resource use and the backward and forward links to other activities is critical to ensure that Tanzania benefits from the exploitation of its natural resources. In addition, improved governance arrangements are important for the sustainable exploitation o f renewable resources such as fish or forests and are necessary to minimize the impact o f negative externalities such as that o f commercial fishing and mining on their artisanal counterparts. A shared-growth strategy requires also a focus on the capacity o f the poor to contribute to-and participate in-economic growth. This strategy includes opportunities to build human capital through measures that center on equitable access to, and improved quality of, education (primary, secondary, and technical), nutrition, and health services; to reduce the burden o f communicable diseases; to improve child nutritional status; and to reduce maternal mortality by helping women achieve their desired family size. Other measures include supporting household savings and investment through development o f appropriate finance and ancillary institutions, especially in rural areas. Limited social protection measures can also play a role in mitigating the impact o f large shocks that may create poverty traps for the poor. SustainingandSharingGrowthinTanzania xiv Regional differences in economic performance reflect not only inherent differences in economic potential, but also factors such as past investments in infrastructure and human resources, local governance, and connectivity. Data suggest some degree o f convergence in per capita incomes across regions. However, in most regions, the growth performance remains intimately linked to the fortunes o f individual crops. Several policy lessons emerge from the analysis o f subnational growth patterns. First i s the importance o f capturing local knowledge to fully exploit growth opportunities. This lesson suggests that decentralization not only i s a means for improved service delivery, but also has an important role to play in the implementation o f Tanzania's growth strategy. Local knowledge about growth opportunities i s critical for a series o f public interventions to foster growth, ranging from infrastructure investments to targeted, crop-specific interventions inthe agriculture sector. Second, local governments have a significant effect on the business environment, ranging from their attitude toward the informal sector to local tax policy and administration, licensing, land management, and so forth. Finally, attention to subnational growth i s important to identify successful strategies that can be scaled up and replicated inother parts o f the country. The study argues that the regional distribution o f public investment should be determined by the growth opportunities, whereas distributional objectives should be primarily pursued through targeting o f access to education. As the report highlights, Tanzania has been successful inestablishing a sound basis for sustained and shared economic growth through the implementation o f a broad reform agenda. Tanzania's strategic frameworks for growth and poverty reduction, including the National Development Vision 2025, the Medium-Term Plan for Growth and Poverty Reduction, and the NSGRP, adequately identify core interventions that are needed to sustain economic growth. These interventions include the strengthening o f economic infrastructure, scaling up o f human resource development from an initial focus on primary education to secondary and higher education, and the implementation o f reforms to strengthen the business environment. In many key areas, specific reform programs are in place, including the Business Environment Strengthening Program in Tanzania, the Agricultural Sector Development Program, the Primary and Secondary EducationDevelopment Program, and the Second-Generation Financial Sector Reform Program. The report broadly endorses this reform program. Table 1 summarizes key issues o f the growth agenda, many o f which are already being addressed by government. The report highlightsthree elements o f the reform agenda that deserve increased attention: 0 Enhancing international competitiveness and accelerating diversification; 0 Making growth pro-poor; and 0 Managing policies and resources for shared growth. Enhancing International Competitivenessand AcceleratingDiversijkation Enhancing international competitiveness and accelerating diversification with a focus on macroeconomic management, infrastructure, access and cost o f credit, the regulatory environment for private sector activities, human resource development, the innovation environment, and the rollout and use o f ICT involves the following: 0 Resolve infrastructure bottlenecks. Recurrent energy shortages are the most visible constraint to economic growth. However, general underinvestment in the development and maintenance o f transport infrastructure, especially in the rail sector and for rural roads, also holds back growth, particularly in rural areas. In addition to scaled-up investment, there i s an urgent need to get appropriate policy and regulatory frameworks in place, which would support a greater participation by the private sector in selected areas o f infrastructure development, operation, and maintenance. Sustaining and Sharing GrowthinTanzania xv e Devote greater attention tofostering structural transformation. To sustain its economic growth, Tanzania will increasingly need to broaden the range o f goods and services it produces. The role o f government in the process o f structural transformation i s not to identify new growth opportunities, but rather to support the identification and exploitation o f opportunities by the private sector. This effort will require scaling up of investment in higher education, with the availability o f skilled labor already being a constraint to economic growth. FDIand the import o f technology are the primary sources o f new technology for Tanzania. However, a strengthening o f Tanzania's research and development systems, especially in agriculture, also has an important role to play in the adaptation and dissemination o f new technologies. Finally, greater access to ICT i s an important tool to accelerate the acquisition o f technology and knowledge. In addition, microeconomic evidence supports a direct link between greater use o f ICT, especially cell phones, and the productivity o f farms and rural enterprises. In addition to economywide support, there i s also some scope for more direct support at the sector and firm levels. An example of such support would be the establishment o f a matching grant scheme for the introduction o f business activities that are new to Tanzania. e Develop an urban strategy. Urban areas play an important role in the economic growth process. Thus, an important challenge i s not only to sustain the good performance o f Dar es Salaam, but also to enhance the role o f regional urban centers as hubs o f economic activity. Such an urban strategy would define the role o f urban areas in Tanzania's economy, as well as the necessary investments and an appropriate institutional and fiscal framework that would allow the implementationo f such an urban strategy. Making Growth Pro-Poor Making growth pro-poor highlights reforms that would accelerate growth o f the agriculture sector; expand income-generating opportunities in micro-, small-, and medium-size enterprises; and strengthen the capacity o f the poor to participate and contribute to economic growth: Ensure that reforms benefit rural areas. To date, the effect o f reforms undertaken has been visible primarily in urban areas, especially Dar es Salaam, while rural areas have seen much less systemic improvements in growth performance and poverty reduction. The report highlights five areas in which reform programs should scale up the focus on rural areas. First, improved access to markets i s central to increasing real incomes o f the rural population. Such access requires investments in road infrastructure and in storage and market facilitation (standards and business climate).The second area i s financial sector reform. Despite the overall progress made, rural areas remain largely cut o f f from access to financial services, which, in turn, have been identified as a key constraint to both farm and off-farm activities in rural areas. The third area i s public expenditure reform. Inthis area, the increases inthe overall resource envelope have accrued primarily to central government agencies and to a much lesser extent to local governments. Progress in the development o f a sound intergovernmental fiscal framework i s a necessary precondition to strengthening the development, operation, and maintenance o f local infrastructure and delivery o f local services, which, in turn, directly affect rural growth and poverty reduction. Fourth are government policies and expenditures to support agriculture. Here the report suggests that public expenditures on research and extension or irrigation can play an important role in supporting the sector. However, scaling up the public support for agriculture needs to be grounded ina careful analysis o f the effectiveness o f expenditure programs. Finally, a better integration o f rural areas in the growth process will require an effort to draw effectively on local knowledge in the Sustaining and Sharing GrowthinTanzania xvi development and implementation o f regional and district growth strategies. This effort will require an increased focus o f Tanzania's decentralization program on economic growth, inaddition to the provision o f social services. e Encourage micro-, small-, and medium-size enterprise activities. Microenterprises, mostly in the informal economy, not only are an important source o f income for many Tanzanians, but also are an important entry point for Tanzanians into entrepreneurial activities. The primary focus should be on facilitating and supporting such activities. Facilitating the transition to the formal sector i s important, but the move should be voluntary unless the primary motive for being in the informal sector i s clearly tax evasion. Devote greater attention to the effect of social expenditures. The report highlightsthat the substantial increase in public expenditures since the late 1990s has directly benefited the poor to only a limited extent. This situation calls for a revision o f spending priorities and targeting approaches to ensure that public expenditures do benefit the poor. Managing Policies and Resourcesfor Shared Growth Managing policies and resources for shared growth means strengthening institutions to develop and implement a growth strategy and to harness both public finances and natural resources toward the objective o f sustained shared growth: Strengthen the capacity for the implementation of Tanzania's reform agenda. As highlightedthroughout the report, Tanzania has made great strides inestablishing a vastly improved economic and incentive regime for economic activities. However, in many areas where appropriate policies and regulations are in place, their effectiveness i s hindered by limited implementation capacity. Enhancing implementation capacity will rely on the continued implementation o f public sector reforms, but the increased use o f the private sector, when appropriate, i s an important way to realize reform objectives. Examples range from an increased role o f the private sector in the provision o f agricultural support services to the contracting ino f capacity at the National Audit Office. Strengthen the management of the growth process. The report highlightsthe importance o f developing and strengthening institutions that are able to coordinate the formulation and implementation o f a shared-growth (pro-poor) strategy in Tanzania. This effort includes the ongoing strengthening o f the budget process to examine and prioritize investments ininfrastructure. Given the large regional differences inproductive potential, strengthening institutions that are able to respond to these differences will be critical. Strengthen analytic underpinnings andparticipation in the design of growth policies. The NSGRP suggests a very welcome greater focus on economic growth, supported by a scaling up o f expenditures that would support accelerated growth, ranging from increased infrastructure investments to subsidies for agricultural inputs or credits to targeted segments o f the economy. In addition, the government i s implementing a range o f + specific programs, such as Tanzania's Mini Tiger Plan 2020, the Property and Business Formalization Programme, and the National Economic Empowerment Policy. For these measures to be effective, they must be grounded ina solid analytic basis that supports the choice and prioritization o f specific strategies and expenditure policies. Strong processes for stakeholder inputs, adequate governance arrangements for expenditure programs at the sectoral level, and a strong monitoring and evaluation system are o f central Sustaining and Sharing GrowthinTanzania xvii importance. Although these issues have received significant attention in the social sectors, they are yet to be fully developed for growth-enhancing expenditure programs. 0 Ensure the effectiveness of government interventions through appropriate governance arrangements. The report highlights a range o f government interventions and supports the government's proactive efforts that complement the focus on creating an enabling environment for private sector activities. Recent government interventions-such as targeted credit guarantee schemes, export processing zones, and targeted agricultural subsidies-are innovative efforts with the potential to foster growth, but they also carry significant risks. These risks include the potential for governance problems that could make such interventions ineffective or even counterproductive. The report suggests a range o f measures that would improve the likelihood o f success o f government interventions. 0 Pay greater attention to the management of natural resources. Tanzania's natural resource endowment could be an important source o f growth and poverty reduction. Governance o f natural resource use, as well as the backward and forward links to other activities, must be strengthened if Tanzania i s to benefit from the exploitation o f its natural resources. Moreover, governance arrangements must be improved for the sustainable exploitation o f renewable resources such as fish or forests. Strengthening governance can minimize the impact o fnegative externalities such as commercial fishing and mining on their artisanal counterparts. Finally, improved governance should aim at an equitable sharing o f natural resourcerents. Table 1provides an overview o f key issues for the implementationo f a shared growth strategy: c B Y 3 3 3 3 3 !3 E!3 3 Y E!3 Y 8!3 Y 3 3 8!3 e, Y 3 !3 2.aE E E E E E E E E ----- --- E E x E E 5: E E E E E x x E E E E E - c .r P a 8 .38 .3 Y Y e, 8E 4 .3 & c1 e, .r Y EE. cd b c- E sE! .r E f$ .e 8 Y 3E cd a 'E eE 8 s v: .-M 0 T .3 3 ke 0 e E .r E 8 E e, B Ir 3 +c1 !u 5z 0 3 8 % 8 8 U 8 2- 55 E11 1 x Eh 4E e x x ...cc 1 3 c1 .r E- Et.E 7 i I $ e .r r t 1I $7 ! i 3 I E i 'ErV I TE i i ...F 3 I $ .r +c1 ?e 5I! I i i I ? l e?e 5 Y Eg E - E E E E .r .C 44 0 .r EF E F .r E I f cE t '5I E W TE tS .r S .r + 0 5: Sustaining andSharing GrowthinTanzania 1 1. POVERTYREDUCTIONAND GROWTH-RECENT PERFORMANCEAND PROSPECTS A DECADE STRUCTURAL REFORMS, MACRO-ECONOMIC AND ECONOMIC OF STABILITY GROWTH Much progress has been made in implementing structural reforms in Tanzania (Box 1). Since the mid 199Os, the country has successfully pursued macro-economic stability as its main policy objective. Monetary and fiscal policy have been designed and implemented with the objective o f price stability. The privatization process i s well advanced. Government has withdrawn from most commercial activities and has encouragedprivate sector participationinthe utility sector. Trade policy has been reformed in the context o f the East African Community, where a customs union became effective in 2005. Although some improvements have already taken place, bureaucracy and corruption remain a source o f concern to current and potential investors. Table 2. Key Economic Indicators, 1995-2005 Indicator 1995 2000 2004 2005 Actual Actual Actual Prel. Est. Annual inflation(percent,period average) 27.4 6.0 4.1 4.4 National accounts, percent growth rates Agriculture 5.7 3.4 5.8 5.2 Industry -2.0 6.7 10.0 10.6 Service 2.9 6.4 6.3 6.9 GDP at marketprice 3.6 5.1 6.7 6.8 National accounts(percent of GDP at market price) Gross domestic investment 19.8 17.4 20.9 21.9 Public investment 3.4 6.0 7.7 7.5 Privateinvestment 16.2 11.4 13.2 14.4 Gross domestic saving 0.8 9.2 11.4 8.4 Balanceof payments(percentof GDP at market price) Currentaccount balance -9.9 -5.3 -3.9 -7.9 Current account balance(excluding current official transfers) -19.9 -10.0 -9.0 -11.8 ExternalIndicators Debt ServiceIExportingof Goods and Services (After rescheduling 21.3 13.2 1.8 2.3 Reserves inmonths of imports 1.5 5.2 5.9 4.4 GovernmentFinance(percentof GDP at market price)* DomesticRevenue 12.5 11.3 12.7 13.3 Public Expenditure 18.3 18.6 22.0 24.4 Overall Deficit (excluding grants) -5.9 -7.8 -9.3 -11.1 Overall Deficit (including grants) -4.2 -3.3 -3.2 -3.4 Externalborrowing (net) 0.9 1.6 3.8 3.0 Domesticborrowing (net) 3.3 0.1 -0.4 1.1 data are for fiscal years, e.g., 2006- 2005106 Source:URT. Economic Survey.Various years. Sustaining and Sharing GrowthinTanzania 2 Box 1. Overview of Structural Reforms in Tanzania Financial Sector. During the past decade, the financial sector has seen significant change. Frombeing the sole preserve o f state-owned financial institutions, it has gone through a process o f privatization and been opened up to new entrants. The successful privatization o f the two largest state-owned banks (sale o f a 70 percent stake o f National Bank o f Commerce and a 49 percent stake o f National Microfinance Bank (NMB)) has contributed to increased efficiency and competition inthe banking sector as well as narrowed interest spreads and a fairly rapid increase in credit to the private sector. Financial sector reforms also included the strengthening o f the legal and regulatory framework, including for micro-finance, as well as a strengthening o f supervision by the Bank o f Tanzania. The recent establishment o f a credit rating agency is a further step inenhancing the efficiency o f financial intermediation inTanzania. The main challenges for the sector include firther reduction in interest spreads and enhanced access to credit by the private sector, especially in rural areas. Government has prepared a comprehensive Action Plan for Second Generation Financial Sector Reforms (SGFSR) which is now under implementation. Parastatal Sector. Tanzania has been aggressively implementing its privatization agenda. The privatization o f manufacturing and commercial parastatals was virtually complete by 2000 and a solid success. The reform and development o f public private partnerships in the infrastructure sector proved much more difficult. At the end o f the 199Os, Tanzania launched the privatization o f its infrastructure enterprises. By 2003, five key infrastructure enterprises (TANESCO, power; DAWASA, water; TTCL, telecom; TICTS, the container terminal; and Air Tanzania) had some form o f private participation. However, by early 2007, only the private sector participation in TICTS i s still in place and considered as being successful, while private sector participation in DAWASA, and Air Tanzania have been dissolved and that o f TANESCO was not renewed as a result o f performance. For the two railroads, TRC i s under negotiations while TAZARA has not yet started and so i s the harbor authority (TPA). Trade Policies and Institutions. Reforms o f trade policies have taken mainly place in the context o f regional agreements, including SADC and EAC. Tanzania adopted the Common External Tariff (CET) o f the East African Community (EAC) inJanuary 2005, lowering its average tariff from 13.8 to 12.3 percent, but further raised the dispersion o f protection. The lowering o f the maximum tariff o f the CET from the current 25 percent to 20 percent, as expected to happen in 5 years in accordance with the Customs Union Protocol, should help correct some o f the dispersion o f protection. On the export side, the main issue pertains to export taxes. International experience has shown that export taxes and bans have generally failed to achieve industrial development objectives, led to informal trade, and frequently hurt small-holders who receive lower prices as a result. Factor Markets (Labor and Land). The revision o f land and labor legislation has been completed, with most o f the emphasis on the reform o f institutions. The main challenge i s now to implement this new legislation, which inthe case o f land reforms is expected to be a lengthy and costly process. Infrastructure (Power Sector and Transport). The establishment o f the executive agency TANROADS with responsibility for the trunk road network has been a major step forward for the transport sector. However, a clear separation o f responsibility between the Ministry o f Works, TANROADS, and districts has not yet been implemented and this hampers effective road maintenance and development activities. The current formulation o f a new Road Act provides a real opportunity to establish a more appropriate policy and institutional framework andprovide the basis for accelerated infrastructure development. Detailed work on the restructuring o f the power sector has been carried out but the implementation o f the restructuring has been delayed. The performance o f the parastatal power utility TANESCO had improved under a performance based management contract, but remains highly vulnerable as the power crisis during 2005/06 has demonstrated. Reforms o f the policy and institutional framework for the power sector are essential to ensure the effectiveness o f future investments inthe sector. Public Institutions Interfacing with the Private Sector. Red tape, corruption, and overly burdensome regulatory and licensing requirements are among the main constraints to private sector development in Tanzania. The government has started reviewing regulations, focusing on removing obstacles and re- organizing the most important tasks o f government. In practical terms this means (1) harmonization o f local government taxation to remove excessive tax burden on private enterprise; (2) streamlining o f work permit procedures; (3) review and amendment o f licensing legislation to reduce the cost o f business Sustaining and SharingGrowthinTanzania 3 establishment and continuation; (4) review and revision o f export-import procedures to reduce time costs and corruption-related costs; and (5) design and implementation o f a program for enhancing access to commercial courts by SMEs. Tanzania has reformed the legal framework for regulatory institutions. The effectiveness of these newly established regulatory institutions, especially given the current oversight arrangements, needs to be closely monitored. Reformshaveresultedinsignificant improvementsinthe economicfundamentals (Table 2). Net domestic borrowing by the central government has been tightly controlled through a cash budget system since the late 1990s. This has allowed the pursuit o f a monetary policy which contained inflation at about five percent during the past five years. This has also created space for a rapid expansion o f credit to the private sector. Credit to the public sector declined from five percent o f GDP in 1998 to 1.9 percent by the end o f 2005, while credit to the private sector increased from four percent to 10.4 percent. Domestic saving increased from 1.5 percent of GDP in 1998 to 9.7 percent in 2005, and the real interest rate on savings became positive. The real effective exchange rate has depreciated by about 30 percent since 2001, which eliminated the overvaluation o f the exchange rate. The current account deficit (after grants) has narrowed from around ten percent o f GDP in 1998 to less than five percent in2004 but increased subsequently to 9.3 percent o f GDP in2005/06, reflecting an increase in aid inflows and debt relief. International reserves stood at about 5 months o f import coverage by the end o f 2006. Table 3. Sources of Growth (production), 1990 2005 - Avg. Ann. GrowthRate Avg. Contr. To Growth ECONOMIC ACTIVITY 1990-94 1995-99 2000-05 1990-94 1995-99 2000-05 Agriculture 3.1% 3.6% 4.8% 1.5% 1.8% 2.3% Industry 2.0% 5.4% 9.0% 0.3% 0.9% 1.6% Mining and Quarrying 11.8% 14.8% 15.2% 0.1% 0.2% 0.4% Manufacturing 0.4% 4.6% 7.3% 0.0% 0.4% 0.6% Construction 2.2% 3.5% 10.3% 0.1% 0.2% 0.5% Services 1.9% 3.8% 6.1% 0.7% 1.3% 2.1% Trade, Hotels andRestaurants 2.0% 4.5% 7.1% 0.3% 0.7% 1.2% Total GDP (factor cost 2.5% 4.0% 6.0% 2.5% 4.0% 6.0% Source: Author's calculahons based on data from URT. Economic Survey. Various Years. Macro-economicstability and structural reforms have created a favorable environment for higher economic growth, which was 6.0 percent during the period 2000 -2005 comparedto 4.0 percentand 2.5 percentinthe precedingtwo five-year periods. Table 3 shows the growth rates o f the main sectors o f the economy as well as the contribution o f these sectors to overall growth. The increase in the growth rates occurred across all sectors, with industrygrowing by 9 percent, services by 6.1 percent, and agriculture by 4.8 percent duringthe period 2000-2005, with mining,manufacturing, and construction being the fastest growing sub-sectors. The contribution of the various sectors to growth, which depends bothon the growth rate o f the sector and its share in the economy, shows that agriculture contributed 2.3 percentage points, services 2.1 and industry 1.6 percentage points of the average annual growth of 6.0 percent during the period 2000-2005. 2.5 3.8 7.0 2. I 6.0 5.4 2. I 6.2 5.6 2.6 5.7 4.0 2.1 5.0 3.4 -0.3 7.3 10.9 0.0 1.2 2.2 5.8 --1.3 15.3 1.2 -0.3 3.5 - 1.4 2.i 9,I 0.4 0.4 -1.7 12.1 5.4 19.3 1.6 1.0 5.0 3.2 1.4 15.0 f .3 0.4 6.7 -37.4 -285.0 4.3 " 1.2 2,6 -0.4 2.1 "4.3 18.9 0.2 -0.4 f .8 9.9 1,o 12.5 1.0 0.1 f .6 20.5 0.3 11.6 2.3 0.0 1.9 Figure I,Growthrates of GDP incl. and excl, ~ o ~ spending,~1940n 2005 ~ - ~ ~ e ~ ~ Increased g ~ ~ ~ e ~ ~pendi~ghas been an i n r n ~ ~ t ~ p engine ~ ~ ~ c ~o growth, ~ ~ of t~ o ~ ~ c o ~ t3.8~percentage~points to the overall growth of 6.8 percent of GBP at market ~ ~ t ~ ~ ~ prices during the period 2 ~ ~ ~ - 2 ~Figurc 0 5 . 1 shows ~ r a m a ~ ~~ ic ~ ~~trajectcmesnfor the a ~e~ r ~ ~ growth rates of C3DP at market prtccs including and ~ ~~ ~~ ~ l ~~ Overall~~ u ~ ~c ~ ~~ GDP at niarkct prices shows the ~an~1l~arc e l e r ~ ~of~eo ~~ o n growth~in~Tarmania over the a c o ~ past fiftccn years. I:Iowever, GUP net of public sector e ~ ~ ~ ~(r.e,, private ~~ or e ~ uand ~ ~ ~ i o ~ ~ d r ~ ~ n ~ ~ sho%sr.sa n ~ a r kdifferent growthpath.~There, ~o~~~~duringthe period 1995-2005 ~ e ~ ~ ~ ~ c ~ ~ n ~ ~ Sustaining and SharingGrowthinTanzania 5 was 4.5 percent and thus actually lower than growth during the first half o f the 199Os, when it was 4.8 percent. This suggests that demand side effects o f variations in government spending contributed significantly to the recent growth acceleration. Most o f the increase in government spending i s financed from increased aid inflows to Tanzania. Inthe National Accounts data, this i s reflected ina wideningo f net-exports duringthe period 2000-2005. Box 2. Government SpendingandEconomicGrowth Government spending affects economic activity both though its demand side and supply side effects. The magnitude and time pattern o f demand side effects depends on whether there are unused capacities in the economy and whether changes in government spending are temporary or permanent. Inthe case o f excess capacity, increases ingovernment expenditure add to economic activity directly through the added demand inthe form o f government purchases as well as through multiplier effects onprivate sector consumption. Empirical estimates o f this demand side effect for developing countries (Hemming et. al. 2002, Schclarek 2005) suggest that these demand effects are typically larger indeveloping than in developed countries. If the increase in government expenditure exceeds unused capacity, it is likely to exert inflationary pressures inthe short runand - if increases in government expenditure are consideredto be permanent - induce an expansion inthe productive capacity o f the economy. Public expenditures also affect aggregate supply. Government spending, especially government investment in economic infrastructure, human capital, or research and development, can increase the productive capacity o f the economy. While the demand side effects typically occur in the short term, supply side effects have a longer gestation period. In addition, while demand side effects only affect the level o f economic activity, supply side effects have the potential to leadto a sustained increase inthe growth rate. Separate from the impact o f government spending o n economic activity is the impact o f the way government spending i s financed. In general, higher taxation and domestic borrowing have a negative impact o n economic activity. The impact o f donor financing-separate from the impact o f the associated increase in government spending-is primarily through its impact on the exchange rate, if the associated foreign exchange inflows are not sterilized. ~~ ~ Figure 2. Contribution of public andprivateexpenditure to economic growth, 1990 2005 - 6.0% 1 5 0% 1 4.0% 3.0% 2 0% 1.O% 0.0% -1 0% -2 0% -3 0% 1990.94 1995-99 20[-05 0 PrivateConsuption and Investment 0 Public Consuption and Investment Net exports and statisticaldiscrepancy Source: Author's calculation'sdbased on data from URT. Economic Survey. Various Years. Sustainingand SharingGrowthinTanzania 6 Figure 2 presents the contribution o f private and public expenditure to economic growth and shows a similar picture as Figure 1. The contribution of private expenditure to growth increased only from 4.5 percentage points during the period 1990-94 to 5.3 percentage points during the period 2000-05. On the other hand, the direct contribution o f government spending to economic growth increased dramatically during that period, Public sector reforms during the first half o f the 1990s involved a significant reduction in the size o f government and thus slowed down growth of GDP. The period 1995-1999 saw as slight recovery o f government spending and the contribution to GDP growth was one percentage point. The period 2000-2005 saw a sharp increase in government spending and its direct contribution to overall GDP growth increased to 3.8 percent.' The contribution o f the sum o f net exports and unrecorded trade and the statistical discrepancy has been negative and increasing during the three periods. This suggests that the demand side impulses emanating from increasedgovernment spending have only partly translated into higher domestic production, but have also contributed to increased imports and a widening current account deficit. A rapid expansion of exports of gold and fish, as well as a recovery of exports of manufactures has led to rapid export growth, while agricultural exports have been in decline since the mid 1990s and have only started to recover very recently. Even though exports o f gold rose from virtually nothing to about five percent o f GDP, their contribution to economic growth has been only around 0.4 percentage points. There i s concern that both the gold and fishing industries are reaching the limits of expansion o f natural resource extraction, with only limited prospects o f future growth, coupled with concern about their environmental impact. Exports o f manufactured goods have recovered in recent years, from about US$30 million in 1999 to US$156 million in 2005, which i s only 23 percent higher than the value o f export o f manufactured goods that had already been achieved in 1996. A key challenge for the Tanzanian economy i s thus to strengthen and diversify its export base. Figure3. Decompositionof economic growthper worker into contributionof human and physical capitalaccumulationand total factor productivity(TFP), 1985-2005. 2.0% 1.5% 1.O% 0.5% capital Mucation 0.0% -0.5% -1.0% -1.5% -2.0%' I 1985-89 1990-94 1995-99 2000-05 Source: Author's calculations based on data from Bosworth and Collins (2003) and URT.Economic Survey.Various Years. Sustaining and Sharing GrowthinTanzania 7 Growth accounting calculations suggest that the recent acceleration o f economic growth i s primarily due to more rapid accumulation o f physical capital and increased factor productivity, as shown in Figure 3. Duringthe period 1995-1999, the ratio o f investment to GDP declined from 19.9 to 15.5 percent and the contribution o f capital accumulation was negative. After 1999, investment increased significantly, reaching 22.2 percent o f GDP by 2005. During that period, public and private sector investment increased by 2.1 and 4.4 percentage points, respectively. However, national accounts data may not fully reflect the significant increase in foreign direct investment (FDI) that Tanzania has witnessed since the mid 1990s. While there was virtually no FDIuntil the early 199Os, by 2000 FDIwas more than US$ 500 million or about 5 percent o f GDP. An Investment Reportpreparedby the Banko fTanzania (2004) shows that most o f the FDI flows go into mining (30 percent), manufacturing (31 percent), wholesale and retail trade, including tourism (14 percent), while agriculture (7 percent) receivedonly a relatively small share o f total FDI.Z The sectors that where the main beneficiaries o f FDI flows showed the highest growth rates in the past five years. The FDI survey also shows the increasing importance o f regional FDI flows from SADC and EAC countries (primarily South Afhca, but also Kenya) accounting for 42.2 percent of total FDI flows in 2001 compared to 52.2 percent from OECD countries. In 2005/06, FDI had declined slightly to about 3.8 percent o f GDP, as the major investments inthe miningsector hadbeen completed and privatization related FDIdeclined. Recent investments in education have a relatively long gestation period until they lead to an effective increase in the human capital o f Tanzania and the contribution o f human capital to growth i s still small. However ifrecent trends inexpanding primary and secondary education are sustained, the contribution o f human capital to growth i s likely to increase. The share o f growth that cannot be explained by human and physical capital accumulation i s labeled "total factor productivity," but contains a variety o f different factors that contributed to growth. Firstly, it captures the increase in land under cultivation in response to improved incentives for agricultural production, which has been a major contributor to growth in the agriculture sector. It also includes increases in capacity utilization as the result o f increases in aggregate demand. Enhanced efficiency o f resource allocation and utilization i s another important component o f the observed increase infactor productivity inTanzania. Analysis o f the manufacturing sector suggests that increased productivity in the sector i s the result o f accelerated exit and entry o f firms inthe sector (Harding et. al. 2002). As older, inefficient firms leave, new, more productive firms enter the sector. All these factors contribute towards bringingTanzania's economy closer to its production frontier. Productivity increases that lead to an expansion o f the production frontier, such as the adoption o f improved technologies have been o f lesser importance, but a number o f cases were new technologies have been successfully introduced demonstrate the scope for future growth impulses. This includes innovations introduced through FDI,but also the results o fTanzania's agricultural researchsystem. Various international benchmarking exercises highlight weaknesses in Tanzania's competitiveness, both in comparison to other African countries and even more pronounced, in comparison to some o f Tanzania's potential key competitors inAsia (Box 3). Sustaining and Sharing GrowthinTanzania 8 Box3. BenchmarkingTanzaniain the GlobalContext World Economic Forum's Global Competitiveness Report 2006-2007 highlights its new Global Competitiveness Index (GCI) that provides an overview o f factors critical for driving productivity and competitiveness, grouped in9 pillars: institutions, infrastructure, macro-economy, health andprimary education, higher education and training, market efficiency, technological readiness, business sophistication, and innovation. Tanzania i s ranked 104" (of 125 countries) o n this new 2006 GCI, behind Kenya (94), but ahead o f Uganda (113). WEF's Business Competitiveness Index (BCI) is another index that focuses on microeconomic factors that determine economies' current productivity and competitiveness. Tanzania i s ranked 731d(of 121 countries) on the BCI in2006, ahead o f Uganda (88) but again behind Kenya (68). World Economic Forum's Africa CompetitivenessReport 2004 highlights the prospects for obstacles to improving competitiveness in 25 African economies. In this report, Tanzania ranks 9 out o f 25Powth and countries on the overall GCI index, surpassing Uganda, which is ranked 14" and Kenya, which is placed 15*. World Bank Institute's Knowledge Assessment Methodology (KAMp: Figure A compares Tanzania's performance with the African region, with neighbors, as well as with Botswana and South Africa, as well as with well performing East Asian economies, such as Malaysia and Thailand on the knowledge economy index (KEI) It shows that between 1995 and the most current period (2004-2005), Tanzania has made a substantial improvement in its overall knowledge readiness, as evidenced by positive changes in the KEI, particularly for the economic and incentive regime and the innovation pillars, as well as some improvement in ICTs,. Uganda has also made strides in improving its economic incentive regime and the ICT pillars, and Kenya has also strengthened its information infrastructure over the past decade or so. Kenya's performance also surpasses that o f the African region, while Tanzania and Uganda do not. Botswana and Malaysia have slightly improved their recent performance over that in 1995, while South Africa and Thailand have not. Thus, this relative comparison shows that even though a country can make progress, it can still fall relatively behind because the world as a whole may have made a much more significant improvement in the variables that are used to track knowledge and innovationrelated performance. FigureA: Tanzania and Comparators,1995 andMostRecent Period MaIaysia 11995 South Africa 1995 Thailand 1995 Botswana 1995 Kenya 1995 Africa 1995 Uganda 1995 Tanzania 1995 I I I I 0:o 2.0 4.0 6.0 1-.J 10EconomicRegime 0 Innwation 0 Education 0 ICT Note; In Figure A, the two bars representthe aggregate KEI score for a selectedcountry for most recent year and for 1995, split into four pillars. Eachcolorbandrepresents the contributionof aparticularpillar to a country's overallknowledgereadiness. Sources: World EconomicForum's GlobalCompetitivenessReport 2006-2007; World Economic Forum's Africa Competitiveness Report2004, andthe World Bank's KnowledgeAssessment Methodology(KAM): httD://www.worldbank.or&am. Firm level data suggest access to financial capital, improvedtechnology, infrastructure,and labor skills, and the investment climate as the main constraints to higher investment and firm growth. The analysis also shows that firms which export typically grow faster than firms which do not. The strong linkbetween the proportionof sales exported and growth indicates that Sustaining and SharingGrowthinTanzania 9 exports are a critical source o f growth for globally competitive firms as they relax the demand constraint that otherwise limits the size o f the market for nonexporters. Younger firms tend to grow faster than older firms, which highlightsthe importance o f facilitating market entry, but also exit, as an important element o f a strategy to raise productivity and growth. The analysis of subnational patterns of income and growth reveals large differences across regions. About 52 percent o f the annual national GDP (1992-2003) was produced in only six regions, namely Dar es Salaam, Mwanza, Shinyanga, Arusha, Mbeya and Iringa out o f the 20 regions. The Dar es Salaam region alone, which i s home to less than eight percent o f Tanzania's population, contributes about 18 percent o f Tanzania's GDP, equal to the contribution to national GDP o fthe bottom six regions (Coast, Lindi,Kigoma, Mtwara, Mara and Dodoma) ~ombined.~ The past decade has seen significant differences in the economic fortunes of individual regions. Interms o f GDP per capita, the top-five regions for the period 1992-99 were Dar es Salaam, h s h a , Rukwa, Ruvuma and Iringa. Those with lowest GDP per capita included Kigoma, Dodoma, Tanga, Kilimanjaro and Mara regions. However, the ranking changed during 2000-03. Mtwara and Mwanza regions replaced Ruvuma and Rukwa among the top five regions in terms o f per capita GDP. Similarly, Kagera, Tabora and Coast regionsjoined the ranks of regions with the lowest GDP per capita while Mara, Kigoma and Tanga regions made marginal gains. The reversals in the regional ranking in terms o f average GDP per capita partly reflects new investment in mining, fishing and related services around Lake Victoria, pick-up o f the cashew nut industry in Mtwara, and collapse o f the coffee industry (Kagera, Kilimanjaro and Ruvuma). Various investment climate assessments (e.g., World Bank 2004b, World Bank 2006b) carried out for Tanzania show significant regionaldifferencesin the business environment, which are relatedto the economic performanceof enterprises inthese regions. This suggests that differences in regional growth rates are partly due to factors under the control o f local authorities and not only due to external factors such as price movements for region specific products or natural resources. Policies o f local authorities that affect growth include local taxation, the quality and implementation o f regulatory and administrative procedure, or the quality o f public expenditure decisions o f local authorities for rural roads, agriculture support, and economic infrastructure such as markets. Tanzania's "Mini-Tiger Plan" (URT, 2004) presentsan important effort to create regional growthpolesthat would capitalize on regionspecific economicpotentials. Many elements of the Mini Tiger Plan, such as the targeting o f infrastructure services based on specific economic potentials would clearly serve to enhance regional economic growth. However, some proposals, such as the roll out o f Special Economic Zones with expanded tax incentives need to be considered carefully in light o f the experience to date with Export Processing Zones in Tanzania and internationally (Box 4). Sustaining and Sharing Growth inTanzania 10 Box 4. Review of Experience with Export ProcessingZones (EPZs). Tanzania Since the establishment o f the Mainland Tanzania EPZ Program in 2002, 7 developers and 9 operators have been granted EPZ licenses. EPZs currently provide employment for about 1,600 people, compared with 2,500 in 2004. The "Tanzania - Diagnostic Trade Integration Study" (World Bank 2005a) suggests that Tanzania's legal and regulatory regime i s broadly adequate. The time and cost involved in obtaining an EPZ license are competitive with neighboring countries such as Kenya and Zambia. However, Tanzanian EPZs have higher leasing cost per square foot than Kenya, while at the same time the services provided are more limited. EPZ firms continue to face infrastructure constraints, especially limited access to power, transport and water, EPZ firms also face institutional constraints, including lack o f an `on-site' customs office in the largest EPZ (Millennium Park), and cumbersome reporting requirements by the Tanzania Revenue Authority and National Development Corporation. These factors have contributed to low occupancy and limitedpositive impact o f EPZs to date. International Experience A recent World Bank report "Free Zones: Performance, Lessons Learned and Implications for Zone Development" (FIAS, 2004) documents that Free Zones have been an important tool to create jobs, generate exports and attract foreign direct investment in many countries, especially Asia. Successful zones are characterized by some or all o f the following features: Streamlined regulatory framework 0 Public-private partnership approaches for zone development Largely private sector-led; leadrole for one developer Clear zone designation and development criteria Top level, integrated support o f government e.g., Jordan, UAE Competition on the basis o f facilitation and services rather than incentives Zone authority i s autonomous, flexible, and focused on regulation, 0 Regulatory authority capabilities are built up Minimization o fpublic expenditure by locating zones carefullyhsing existing facilities The performance o f Free Zones in Africa to date i s rather disappointing, with only a few well known successes such as Mauritius and Madagascar, but a large number o f schemes that are performing well below expectations, or have failed completely. This suggests that Special Economic Zones (SEZs) and EPZs are not a panacea to overcome weaknesses in the business environment that prevail in many African countries. Economic Zones cannot be a viewed as a substitute for a country's larger trade and investment reformefforts. They are one tool ina portfolio o f mechanism commonly employed to create jobs, generate exports and attract foreign direct investment-through the provision o f incentives, streamlined procedures and custom built infrastructure. However, maximizing the benefits o f zones depends on the extent to which they are integrated with their host economies. The static and economic impacts o f zone development are suppressed when zones are operated as enclaves. They are multiplied when accompanied by economic policy and structural reforms that enhance the competitiveness o f domestic enterprises and facilitate the development o f backward and forward linkages. ECONOMICGROWTH, POVERTY, AND INEQUALITY With an estimated per capita incomeof US$340(Atlas method)in 2005, Tanzania is among the world's poorest countries. A household budget survey carried out in2001 shows that 35.3 percent of the population lives below the national poverty line. However, the national poverty line (at US$0.32 per day) is low both in terms of regional and international comparisons. Using the international dollar a day poverty line increases the share of the population living in poverty dramatically to 57.5 percent. Povertylevels are highestamonghouseholdsthat dependon agriculturefor their livelihood. About 40 percent of such households are below the poverty line and they make up about 81 Sustaining and SharingGrowth inTanzania 11 percent o f the poor in Tanzania. Poverty levels are lowest among households whose head i s a paid private or public sector employee, followed by households whose head i s self-employed in a non-agricultural occupation (the urban informal sector). Consumption per capita in non- agricultural households i s about 1.5-1.8 times higher than in agricultural households, and these households have experienced a faster growth o f income over the last decade, especially in Dar es Salaam. The household budget survey revealed fairly large regional differences in poverty levels in Tanzania (Table 5). Dar es Salaam experienced a major reduction in poverty, triggered by growth o f employment and incomes in the formal private sector, which spilled over to the informal sector, raising incomes there as well. Meanwhile, poverty worsened in the Northern Highlands area, especially in the rural areas and i s lowest in the Southern Highlands. Thus, Singida has the highest incidence o f poverty (55 percent), while it stands at only 21 percent for Mbeya. Despite these regional variations, the level o f inequality in Tanzania-between regions, between rural and urban households, and for the whole mainland-is still relatively low compared to other African countries. Table 5. Zonal Poverty Status Population share Poverty headcount 1991/1992 2000/2001 1991/1992 2000/2001 Tanzania 100.0 100.0 38.6 35.3 Rural 82.1 79.0 40.8 38.6 Other urban 12.6 13.6 28.7 25.9 Dar es Salaam 5.3 7.4 28.1 17.6 Coastal 13.7 14.6 40.0 34.7 N"Highlands 10.3 9.9 20.2 36.1 Lake 26.2 30.2 37.0 39.0 Central 7.3 10.3 48.8 42.4 S" Highlands 10.9 14.7 46.6 25.8 South 8.3 14.8 43.9 43.2 Note: Boldindicates that the differencebetweenthe two surveys i s significant at the 5% level. Source:Author's calculations using data from HBS 1991/92and2000/01. Poverty simulations suggest that the link between poverty reduction and economic growth i s strong in Tanzania. The decline in poverty between 1992 and 2001 was limited because o f modest economic growth and not because o f a missing link between growth and poverty. Real GDP grew at an average by 3.6 percent, which combined with a high annual rate o f population increase o f 2.9 percent resulted in annual per capita growth o f only 0.7 percent. The HBS data provides only two snapshots in time and fails to represent the full evolution o f poverty between 1992and 2001. By applying macroeconomic growth data to the micro-level HBS data, changes in consumption are simulated year-by-year using unit-record survey data, under varying assumptions for growth rates and inequality changes. The simulations suggest that poverty rose between 1992 and 1994 and then fell in the final years o f the decade. The fall in poverty was faster inDar than in other parts o f the country. Sustaining and Sharing GrowthinTanzania 12 Figure 4. Simulated changesin poverty, 1992-2002 Simulated changes inpoverty .45 2 survey approach. using within stratum distribution Fractionbelow Dovertv line , , I , I I I 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year I-......-.--- MainlandTanzania -----Dares Salaam Other urbanareas -. -. - Rural areas Source: Author's calculations based onNational Bureau of Statistics. 2002. and URT.Various Years. Growth has clearly reduced poverty but at the same time, higher inequality hasworked in the opposite direction. Outside o f Dar es Salaam, consumption growth was small and the increase in inequality almost negated it. InDar es Salaam, inequality increased, but household consumption growth occurred as well, and the poor benefited substantially from this growth. Thus, despite the muchlarger increase ininequality inDar, poor households gained more than in other areas where the increase in inequality was more modest. In other words, while the rising inequality in Tanzania i s somewhat o f an issue, increasing overall income growth across the country inthe poor households i s an even greater issue, and income inequality does not seem to have been a drag on growth inDar es Salaam. Most of the poverty reduction during the 1990s has happenedwithin areas. Table 6 shows the decomposition o f the change in poverty into regional composition effects. There are inter- regional effects, population shift effects, and interaction effect^.^ Duringthe 1992-2001 period, poverty declined substantially inDar es Salaam and other urban areas. This attracted people from rural areas, resulting in a 4.6 percent population growth rate in urban areas, compared to 2.5 percent in rural areas. Despite these population inflows and the substantial urban poverty reduction, only 0.4 percentage points o f the total drop in national poverty o f three percentage points can be attributed to a shift in the population from poorer rural to wealthier urban sectors. While rural-urban migration may indirectly promote poverty reduction, e.g. by facilitating stronger rural-urban linkages, migration itself i s not a major contributing factor. More generally, the spill-overs from urbangrowth to ruralpoverty reductionappear to have beenmodest. Sustaining and SharingGrowthinTanzania 13 Table 6. Decomposition of change inpoverty Tanzania Dar es Salaam Otherurban Rural Poverty 1991 38.6% 28.1% 28.7% 40.8% Poverty 2001 35.4% 17.6% 26.0% 38.7% Total change inpoverty 2001-1991 -3.2% -10.5% -2.7% -2.1% Growth impact -8.4% -18.4% -6.6% -5.3% Inequality impact 5.5% 12.4% 4.0% 2.7% Residual -0.2% -4.5% -0.2% 0.6% Urban- rural decomposition Change inpoverty inDar es Salaam -0.6% Change inpoverty inother urban -0.3% Change inpoverty inrural -1.7% Total intra-sectoral effect -2.6% Population-shift effect -0.4% Interaction effect -0.2% Types o f employment Change inpoverty fanninglfishing -2.0% 1.1% -0.2% -2.5% Change inpoverty paid employment -0.4% -7.9% -0.9% 0.2% Change inpoverty self-employment -0.6% -5.8% -0.9% -0.2% Change inpoverty family employment 0.1% -0.2% 0.2% 0.0% Change inpoverty non economic activity 0.0% 0.1% -0.1% 0.1% Total intra-sectoral effect -2.9% -12.6% -1.8% -2.5% Population-shift effect -0.6% 0.6% -0.8% -0.7% Interaction effect 0.4% 1.5% -0.2% 1.2% Source: Author's calculations using data from HBS 1991/92 and 2000l01. Table 6 shows the power of a rural poverty reduction strategy for Tanzania. Even slight income growth for the rural poor has a large impact on overall poverty in Tanzania. The same holds true for households engaged in farming and fishing, as they are a large share o f the population and their income i s low. However, the last section o f shows that the movement o f households out o f agriculture has also played a strong role in poverty reduction. In particular, moving from agriculture to other occupations in rural areas has contributed significantly to poverty reduction. In fact these structural changes in rural employment had a more significant impact on poverty reduction than rural urban migration. Considering that urban-rural growth and income differentials were large during the period under consideration, it i s unlikely that the poverty impact o f sustained high urban growth will significantly increase in the future. On the other hand, the impact o f higher rural growth, both in agriculture as well as other income generating activities, i s likely to be significant. This analysis underlines the importance o f increasing rural incomes as the primary instrument for successful poverty reduction in Tanzania. Urban growth remains important for Tanzania, especially with respect to its role as an engine of structural transformation o f the economy. However, the impact o f urban growth on poverty reduction i s clearly limited. Measures that support mobility, primarily mobility from agriculture to other sectors, but also geographic mobility are nonetheless important elements of the structural transformation process. Sustaining and SharingGrowth inTanzania 14 OUTLOOKONGROWTH AND POVERTYREDUCTION Tanzania's National Strategy for Growth and Poverty Reduction (NSGRP) has set ambitious goals for poverty reduction, which even exceed the MDG of halving poverty by 2015. An average growth rate o f at least 6.1 and 5.0 percent would be necessary to halve poverty by 2010 and 2015, respectively. The pattern o f growth and inequality i s also important for Tanzania's prospects o f halving poverty over the next five to ten years. If growth i s biased towards urban areas, higher average growth rates will be necessary to reach the overall poverty reduction targets. Achieving such poverty reduction targets will also require policies and investments that ensure that economic growth creates income earning opportunities for the poor and enables them to take advantage o fthese. Within the framework of current policies and institutions, sustained per-capita GDP growth rates of at least four percent per annum seem to be achievable, which would translate into overall GDP growth rates between 6-7 percent. This assessment is based on policy based growth projections, where improvements in the institutional and policy framework translate into an improved outlook for economic growth. Measures o f the quality o f policies and institutions are provided by four independent sources: the World Bank's Country Policy and Institutional Assessment, the Institutional Investor assessment, the ICRGassessment and the Euro money assessment, which all suggest significant improvements between 1999 and 2006 (Table 7). Based on the country ratings for 2006, the projections o f per-capita growth range between 3.8 and 5.0 percent, with an average o f all four projections o f 4.4 percent. If there are further improvements in institutions and policies, the projections suggest that even per capita growth rates o f over 5 percent per annum are feasible. Table 7. Policy based growth projection Per capitaGrowthProjections Rating 1999 Rating2006 Constant Improve+O.S CPIA 3.5 3.9 5.0 6.1 EUROMONEY 2.2 2.8 4.1 4.8 ICRG 3.8 4.1 4.6 5.1 INSTITUTIONAL 2.0 2.3 3.8 5.0 AVERAGE 2.9 3.3 4.4 5.3 Note: Ratingsare ona scale of 1- 6 Source: Author's calculations based on data from Euromoney. Various Years. Institutional Investor. Various Years. ICRG.Various Years, andWorldBank. Various Years. The input requirements for such growth rates in terms of investment,increasesin the level of education in the work force, and sustained productivity increasesare large, but seem to be achievable. For example, sustaining a growth rate of output per worker o f four percent could be achieved with an investment to GDP ratio o f around 18 percent, an increase in the average years o f schooling by one year by 2011, and sustained increases in factor productivity. The expansion inthe average years o f schooling per worker would require sustaining further progress inprimary school enrolments and especially completion rates, as well as a significant expansion insecondary school enrolments. Development and economic growth is characterized by structural transformation of the economy. The share of agriculture typically declines, while that o f industry and services increases, and higher economic growth i s typically associated with faster structural transformation. Table 8 illustrates the likely development o f the sectoral composition o f GDP Sustaining and SharingGrowthinTanzania 15 under various growth assumptions, reflecting cross country experience on the association o f economic growth and structural change. The baseline scenario projects current sectoral growth rates into the future. This scenario i s compared to one with higher aggregate growth and faster structural transformation and one with lower growth and limited structural transformation. Table 8. Scenariosfor economic growthand structuraltransformation Slow Growth MediumGrowth Fast Growth Share Avg. real Share in Avg. real Share in Avg. real Share in inGDP growthrate GDP growthrate GDP growthrate GDP 2005 2006-2025 2025 2006-2025 2025 2006-2025 2025 Agriculture 45.6% 3.3% 40% 4.8% 34% 5.0% 26% Industry 19.7% 4.8% 23% 9.0% 32% 10.7% 32% Services 34.8% 4.4% 37% 6.1% 33% 9.0% 42% Total 100.0% 4.0% 100% 6.3% 100% 8.0% 100% Source: Author's calculations. Under the medium growth scenario, the share of agriculture in GDP dropsto 34 percentby 2025 while the shares of industry and services increaseto 32 and 33 percent respectively. Comparing these projections with international experience suggests that these sectoral projections are consistent with the pattern o f structural transformation observed in other economies. In the high growth sceniario, annual average growth of 8 percent is the result o f accelerated growth in the industry and services sector. Under this scenario, agriculture would only account for 26 percent o f GDP in2025, while the share o f the industryand services sector would have increased to 32 percent and 42 percent respectively. In order for the economy to achieve the NSGRP growth targets, growth will have to accelerate in all sectors. In the medium term, the agriculture sector (including agro-industry) will remain central to Tanzania's economy and growth performance. However, as discussed below, growth in the agriculture sector should be primarily driven by productivity increases and labor saving technologies, which would increase per capita incomes in the agriculture sector. At the same time, an accelerated growth and poverty reduction scenario will require a gradual movement o f labor from the agriculture sector to the more remunerative secondary and tertiary sectors. The combination o f these factors suggests that growth inagriculture i s likely to be lower than that o f other sectors. Nonetheless, sustaining average agriculture sector growth o f about 5 percent i s both necessary and achievable over the medium term and will constitute the foundation o f overall accelerated growth o f the Tanzanian economy. A continuation of the strong growth performance of the manufacturing sector is critical. Growth in the manufacturing sector i s important to realize Tanzania's poverty reduction objectives, as it provides access to higher paying jobs and strengthens the demand for primary products. The contribution o f the manufacturing sector to GDP has been stagnant at around eight percent since independence. But, more recently, with an average growth rate o f nine percent duringthe past five years, it has outperformed the rest o fthe economy. The growth potential of the manufacturing sector derives primarily from linkages to Tanzania's agriculture sector as well as the country's natural resource base, especially forestry, minerals and fish. Strengthening the backward linkages o f the manufacturing sector with its resource base as well as the forward linkages, especially to export markets, is therefore essential for the development o f the manufacturing sector. Sustaining and Sharing GrowthinTanzania 16 Analysis of enterprise survey data covering growth, investment, exports, and employment identifies the key constraints to more rapidgrowth. In2002, inorder o f importance the four leading constraints were (i)access and cost o f finance; (ii) access to superior technology; (iii) infrastructure; and (iv) supply o f skilled labor and its productivity. Other constraints such as demand and the business environment were significantly less important for firm growth. The service sector, especially tourism, also has the potential to sustain high growth rates in the medium term. Official statistics suggests that Tanzania's strategy o f marketing itself as an upmarket destination has been relatively successful. Tanzania's vast natural resources provide ample potential for further expansion, such as through the development o f the Southern Circuit which includes the Selous Reserve. However, in order to sustain Tanzania's up-market image, investments insupportive infrastructure and humanresources will be critical. PROGRESSTOWARDSTHE MILLENNIUMDEVELOPMENTGOALS It will be challengingfor Tanzania to meet all MillenniumDevelopmentGoals (MDGs),yet there is room for cautious optimism in some areas. Data from the 2004 Demographic and Health Survey suggest that considerable progress was made in the reduction o f malnutrition and child mortality. Policy simulations show that the income poverty and hunger MDGs may be attainable if Tanzania i s able to sustain o f medium highgrowth and with its improvements inthe social sectors. Table 9. MDGbaseline,most recent estimate and target. Year Most Year of most Baseline recent Target baseline recent Goal 1: eradicateextremepoverty and hunger Reduce extremepovertyby half-nationalpovertyline 38.6% 35.6% 19.3% 1991 2000 -dollaradaypovertyline 61.1% 57.5% 30.6% 1991 2000 Reducehungerby half 29% 22% 14.5% 1991 2004 Goal2: Achieveuniversalprimary education Net enrollmentinprimary school 51% 91% 100% 1990 2004 Goal3: Promotegender equality and empower women Equalgirls'enrollmentinprimary school 1.01 0.99 1 1990 2004 Equalgirls'enrollmentinsecondary school 0.70 0.81 1 1990 2000 Goal4: Reducechild mortality Reduce child mortality of under fivesby two thirds 141 112 47 1991 2004 Goal5: Improve maternalhealth Reducematernalmortality by three quarters* 529 578 132 1996 2004 Goal6: Combat HIV/AIDS, malaria and other diseases Halt andreversespread of AIDS n.a. 7.0% 2003 Halt andreversespread ofmalaria 21% 36% 1999 2004 Goal7: Ensureenvironmentalsustainability Halveproportionwithout improveddrinkingwater inurban areas 13% 10% 7% 1991 2000 Halveproportionwithout improveddrinkingwater inrural areas 65% 54% 33% 1991 2000 Halveproportionwithout sanitation inurbanareas 2% 4% 1% 1991 2000 *Halveproportionwithouta sanitation inrural areas 9Yo 8% 5% 1991 2000 Maternal mortality i s so called low frequency event. Low frequency events are difficult to measure accurately in surveys like the DHS which serve as source for these estimates, as few cases of maternal mortality are Sustainingand SharingGrowthinTanzania 17 registered. Consequently, maternal mortality rates are associated with large confidence intervals and the observed maternal mortality rates o f 529 in 1996 and 578 in2004 are statistically not different. With respect to the eradication of extreme poverty and hunger, the latest householdsurvey carried out in 2000/01shows that incomepoverty has declinedfrom 38.6 percent in 1991to 35.6 percent in 2000/01, using the national poverty line. If the international dollar a day poverty line i s used, income poverty declined from 61.1 percent to 57.5 percent. Halvingpoverty by 2015 would thus require poverty to decline to 19.3 and 30.6 percent respectively. With the limited progress made during the 1990s (Table 9) this appears unattainable, but the data obscure that the early 1990s were characterized by low growth. The period o f positive growth only materialized from 1995 onward so that by the end o f the 1990s most o f the damage from the earlier period was undone. The period of highgrowth since 2000/01 i s not captured in the survey data. Yet poverty simulations (Figure 5), suggest that the income poverty MDG i s attainable. Muchwill depend on the pace o f rural income growth as most poor are found inrural areas. Figure5. Projectedchangesinincome poverty and malnutrition Prevalenceof income poverty($ day poverty line) Prevalenceof underweightchildren under 5 65.0 30.0 29.0 55.0 ' 25.0 45.0 h h 5 ~x *. 45.734.8 20.0 ' 35.0 ,E, 30.6 4M .C 25.0 +Slowgrowth .*...'.. 15.0 14.5 15.0 * 19.0 -0 e, --31tSlowgrowth -Medium growth 5 10.0 -Medium growth 5.0 , .......Fastgrowth - ...... Fast growth , , , , , 5.0 -5.0 -3 m \ I -c 0 - 0o 0m l0 . m- l 3 0 0 0 0 0 3 - 0 N W N N N N N N Note: Growth scenarios allow for different rural-urban growth rates. Inthe slow growth scenario GDP growth is 4.( percent per annum (3.8 percent in rural areas, 4.6 percent in urban ones). In themedium and fast growth scenarios this i s 5.4 percent and 8.0 percent respectively with rural growth rates o f 5.0 percent and 6.5 percent and urban growth rates o f 7.4 percent and 9.7 percent. Population growth i s assumed to be 2.9 percent per annum. The income poverty projections are determined by applying GDP growth rates to unit record consumption information in the HBS.The nutritionprojections are basedonan income-nutrition elasticity o f0.51.The horizontal lines inbothgraphs present the baseline and the MDGobjective, respectively. After a decade of stagnancy in which the share of underweight children remained unchangedat 29 percent, it declinedto 22 percent between 1999and 2004, so that Tanzania i s on track with respect to attainingthe hunger MDG. Income growth can explain at most a third of the decline. Malnutrition appears more responsive to changes innon-income factors such as a more effective management o f malaria and vitamin A supplementation. Policy simulations (Figure 5) suggest that if income growth and especially improvements in nutrition related interventions can be sustained, the hunger MDGmay be attained by 2015. Progress towards attaining the MDGon universal primary education has beenimpressive. As soon as universal primary education was introduced in2001, net enrollment rates jumped and are now close (91 percent) to the target o f 100percent enrollment. In primary education gender inequalities are small (Table 9), but in secondary education gender inequalities persist. However, the recent increase in secondary education was much more gender balanced than previous enrolment patterns: in 2004 the girl-boy ratio in Form Iwas Sustaining and SharingGrowthinTanzania 18 0.98 while it was 0.55 inFormIV. Other non-MDG indicators such as the ratio o f female to male prevalence o f HIV/AIDS(of 1.22) show that gender inequality persists. Another encouragingfact is the observedreductionin child mortality from 141per 1000in 1991to 112in2004. This drop occurs after stagnancy inchildmortality rates duringthe 1990s (it was 137 in 1996 and 147 in 1999) and appears to be a reflection o f improved malaria management (improved treatment as well as increased use o f bed nets) and stronger health systems in general. The MDG target o f 47 deaths per 1000 i s ambitious however, and it will remain a challenge to meet this objective. With respectto the reductionof maternalmortalityno progresshasbeen made. The data in Table 9 even suggest an increase inmaternal mortality rates but the difference i s not statistically significant. The lack o f progress i s reason for concern as the MDG i s ambitious (a 75 percent reduction) and maternal mortality rates high(578 per 100,000 births). A recent blood sample survey suggests that the prevalence rate of HIV/AIDS is 7 percent which is less than was expected based on tests performed on blood donors. There is much variation in infection rates by age group, gender and location. Women (7.7 percent) are more affected than men (6.3 percent). Prevalence i s 5.2 percent in the 20-24 age cohort, but 10.9 percent in the 30-44 age cohort. Prevalence rates vary from a low o f 2.0 percent inManyara and Kigoma to a higho f 13.5 percent inMbeya and Iringa. Finally some limitedprogresswas made in improving access to drinkingwater mainly due to increasedaccess to protectedwells and springs, Still some 54 percent o f all rural households do not have access to improved drinking water and much needs to be done to attain the MDG objective o f 33 percent. As far as sanitation i s concerned almost all urban (96 percent) and rural (92 percent) households have access to a toilet. In urban areas, one observes an increase in the share o f households withwt a toilet from 2 percent on 1991 to 4 percent in 2000, possibly as a result from rapid urbanization. Sustainingand SharingGrowthinTanzania 19 2. ENHANCINGINTERNATIONAL COMPETITIVENESS AND ACCELERATINGDIVERSIFICATION The analysis of macro-economic developments clearly indicated that sustaining high economic growth will require a strengthening of Tanzania's international competitiveness and accelerated diversification of the range of products and services produced by the Tanzanian economy. This part o f the report first presents an analysis o f the factors that determine growth in Tanzania's manufacturing sector. Expansion o f this sector i s considered a central driver o f sustained growth and structural diversification o f the Tanzanian economy. This discussion o f the manufacturing sector provides the basis for an in depth discussion o f policy priorities for enhancing international competitiveness and for accelerating diversification. To achieve increased international competitiveness, improvements in infrastructure, access to finance, and the business environment are central. Accelerated diversification will require a strengthening o f post-primary education, improvements in Tanzania's innovation environment, and increased access and use o f ICT. FOSTERING GROWTH, EXPORT COMPETITIVENESS, AND EMPLOYMENTTANZANIA'S IN MANUFACTURING SECTOR A growth strategy basedon supporting the manufacturing sector, and within it larger firms, especially exporters, has non-trivial poverty implications for the large proportion of unemployed and relatively less skilled Tanzanians. Faster rates o f growth in the sector contributed to higher growth in the overall economy and the largest proportion o f better paying jobs relative to employment in the agricultural sector where incomes typically fall below the poverty line. For example, in 2001, manufacturing generated about one-third o f nonagricultural private employment. H o w can policy makers maximize the sector's growth potential? Empirically, larger firms are the drivers o f growth and employment within the manufacturing sector. In absolute terms, in a sample o f 264 firms, larger firms with more than 30 employees generated 29,000 jobs relative to only 1700 in smaller firms. Employment in larger firms grew at 8.5 percent per year relative to only 3.0 percent in smaller firms. Unsurprisingly, larger firms systematically paid significantly higher wages. The median wage for professionals in larger firms was Tsh. 240,000 relative to Tsh. 100,000 in smaller firms; for skilled workers, it was Tsh. 90,000 inlarger firms compared to Tsh. 60,000 insmaller firms; and for unskilledworkers, it was Tsh. 60,000 inlarger firms relative to Tsh. 50,000 insmaller firms. Empirical analysis of the determinants of manufacturing firm growth shows that the two leading contributors to firm growth are a technically skilled labor force and growth of large firms who also represent the majority of the exporters. A 1percent increase inthe technical skills o f the workforce increases firm growth by 0.70 percent and a 5 percent increase in the output exported delivers almost a 1 percent increase in firm growth, which, in turn, raises employment growth. The third important determinant o f growth in manufacturing firms i s investment growth (growth elasticity i s 0.08). As the jump in firm growth i s intricately tied to growth inexports because o f limited purchasing power inthe domestic market, an aggressive and proactive policy stance for promoting manufactured exports i s likely to have the biggest effect on manufacturing growth in Tanzania. The rationale for that selective approach i s motivated by SustainingandSharingGrowthinTanzania 20 today's global reality: if a firm cannot compete inthe global market (that is, export), it i s unlikely to survive for too long in Tanzania's domestic or Africa's regional markets, which are flooded with cheaper imports from low-cost, high-skills producers such as those from East and South Asia. Firmsize is a criticaldeterminant of firm growthinTanzania. As many larger firms are also exporters, and most exporters are larger firms, policies targeting larger firms should have large payoffs in helping to expand existing firms and promote new entrants into the export sector. Although policies that favor large firms also favor exporters, export promotion strategies are important in their own right, given the limited purchasing power in the Tanzanian market. This recommendation requires a strategic, two-pronged approach: one that targets large firms, and another that targets existing and potential exporters. Policies need to address domestic supply constraints associated with disproportionately higher transaction costs of investment faced by large firms and exporters, and reduce obstacles associated with finance, infrastructure,technology, and skills, which emerged as the key constraints to growth from the analysis of the investment climate assessment.All of that may seem tantamount to recommending everything-that is, redressing all barriers to production currently facing all manufacturingfirms inTanzania. It i s not. To circumvent the high financial and time costs and the government's limited implementation capacity requires focus and pragmatism in catering to larger firms. The targeted delivery o f physical inputs such as infrastructure would be facilitated by the identification o f spatial locations where manufacturing and export activity i s most prevalent. Spatial identification also helps in targeting exports. Similarly, to attract potential exporters, the government could identify special areas, such as industrial districts and export processing zones, where larger firms interested in exporting would be able to benefit from improved infrastructure and service delivery. That approach would render public support in a financially feasible and timely manner for fast-growing exporters and potential new entrants into the export business. The manufacturing sector cannot afford to wait untilthe constraintsto investment are resolvedeconomy-wide. Althoughthe entry of firms intothe export sector canbefacilitated,muchmoreis neededto sustain them in the business. Creating and nurturing firms' ability to export implies grappling with the challenge o f improving productivity and spurring competitiveness to export to non- African markets, where competitiveness i s best tested. That challenge requires a set o f externally oriented policies that exploit Tanzania's latecomer advantage to leapfrog into the global marketplace. The starting point could be direct public support to facilitate the delivery o f the two key public goods: (a) superior technologies o fproduction through adaptation and (b) development o f technical, tertiary, and managerial skills needed to apply them. Public-private partnerships have served as the best mechanism to deliver those two public goods. The ability o f Fundaci6n Chile to promote technology transfer and adaptation and that o f the Indian Institutes o f Technology and Management to deliver critical skills offer useful lessons. Additionally, financial incentives are needed to reduce the high fixed cost o f entry into export markets and to attract firms and sustain them in the export business. Special incentives to promote exports out o f Africa are likely to have the highest payoff and sustainability. Lessons from East and South Asia are good starting points, especially inthe area o f agroprocessing and light manufacturing. IMPROVINGTHE BUSINESS ENVIRONMENT The quality of the business environment affects the cost of doing business, and is an important determinant of a country's attractiveness for investors and its international competitiveness. For Tanzania, the cost o f an inefficient business environment are estimated to Sustainingand SharingGrowthinTanzania 21 be very high, amounting to 25 percent o f sales, including the cost o f contract enforcement difficulties, regulation, bribes, crime, and unreliable infrastructure (Figure 6). Figure 6. Cost of inefficienciesinbusinessenvironmentas percentof sales, various countries Source: World Bank, 2004a. The report focuses on three aspects o f the business environment in Tanzania that promise the highest gains in terms o f accelerated economic growth if appropriate actions are taken. These include the provision o f complementary infrastructure, the cost and access to finance, and the cost arising from bureaucracy and corruption inthe interaction o f the public with the private sector. Infrastructure Tanzania's weak infrastructure, especially in transport and the power sector, impedesthe competitiveness of the private sector and limits economic expansion. The levels of infrastructure provision in Tanzania are among the lowest in Africa (Figure 7) and returns to infrastructure investment are generally high, although they vary considerably by region, type o f investment, and time period o f the investment. The highest rates o freturn on selected projects are 104 percent in the road sector, 23 percent in the power sector and 50 percent in telecommunication. Sustaining and Sharing Growth inTanzania 22 Figure7. Low levels of infrastructuredevelopment constraineconomic growth Roads, paved (YOof total roads) Electric power consumption (kwh per capita) 100 90 80 70 60 50 40 30 20 10 0 Source: World Bank.DevelopmentDataPlatform. A reliable power supply is of particular importance for economic growth in industry and the service sector and priority should be given to enhancing power supply in urban areas. Businesses inTanzania suffer highcosts from an unreliable power supply due to the need to have their own back-up facilities and interruptions o f production due to power outages. Due to the changed situation in international energy markets, the current energy sector policy may require revision to be more focused on a long term vision for the growth o f the sector, including rural electrification and connection to regional grids. This vision should be underpinned by a realistic investment plan based on the financing expected to be available from various sources (Government, development partners and the private sector). It would be reasonable for the new policy to include a much less complex plan for the restructuring and possible divestiture o f TANESCO. An improved road network is critical for enhancing agricultural incomes by increasing access to localmarketsand reducingcosts. Inaddition, any efforts to promote ago-processing will require enhanced linkages between rural areas where agricultural production and processing takes place, and national and international markets. The maintenance and rehabilitation o f existing roads based on economic criteria needs to be a priority. The passing o f an effective Roads Act, defining the various responsibilities in the sector and the financing mechanisms, and the transformation o f TANROADS into a Road Authority, are critical to establish an appropriate framework for increased expenditures in the road sector and ensuring its effectiveness. While financing for the road sector will primarily be in the form o f public and donor financing, rehabilitation and expansion o f the rail and port infrastructure offers greater potential for private sector involvement. Recommendations. In order to enhance infrastructure services in Tanzania to meet the long- term growth targets, increased spending for both investment and maintenance o f infrastructure i s required. The following principles are important to ensure that increased infrastructure expenditure yields the needed impact on economic growth: 0 Establishmentof an appropriatepolicy and institutionalframework is critical to ensure the efficacy of infrastructure expenditure. In particular, the enactment o f appropriate Sustaining and Sharing Growth inTanzania 23 electricity and road acts are critical for the development o f infrastructure in Tanzania and improving efficiency. Tanzania's infrastructure requirements are large and must be properly prioritized. This prioritization should be primarily based on the expected rates o f return to infrastructure and poverty impacts. Infrastructureneedsand constraintsinruralareas vary, reflectingthe diversity intypes of economic activities, population densities, existing infrastructure endowments and remoteness: The policy implication o f this diversity is the need for regionally targeted infrastructure investments that respond to local priorities. Part o f the strategy could be to cluster infrastructure investment in specific areas which would accelerate growth by enabling producers to exploit economies o f scale. Appropriate balance needs to be maintained between new investment, rehabilitation, and recurrent maintenance. Currently, road sector spendingis still insufficient to maintain the existing road network. Building new roads is expensive and, thus, appropriate maintenance o f existing roads i s necessary to ensure cost effectiveness and sustainability o f the growth andpoverty impacts. Strengthening regulatory institutions and ensuring their independence is critical to attract private sector participation inthe infrastructure sectors. Scaling up of investment in infrastructure may have adverse short-term macro- economicconsequences:The highdomestic content o flarge scale infrastructure construction such as roads, may affect relative prices between tradables and non-tradables. Inthe longer run, however, the impact is likely to be more than compensated by the economy wide productivity gains from the investment. Nonetheless, this i s a potential down side risk that would needto be managed carefully. To avoid and mitigate Dutch disease effects inthe short term, attention needs to be paid to the import content o f infrastructure expenditure and its gestation period. A higher import content reduces the likelihood and magnitude o f Dutch disease effects while expenditures with a shorter gestation period would accelerate the compensating productivity gains expected from infrastructure investment. Expandingaccess and reducingthe cost of credit The current depth and efficiency of Tanzania's financial system falls well short of what is needed to help support economic growth and ensure the efficient allocation of resources. The econometric analysis o f the enterprise survey data suggests that access to finance i s the main constraint for growth and investment o f firms. Analysis o f the agriculture sector similarly suggests that access to credit i s an important constraint for agricultural production. Access to credit hasseen dramatic changes duringthe past decade. Credit as a share o f GDP has declined dramatically from about 35 percent to GDP in 1993 to only 7 percent o f GDP in 2002, has, however, recovered subsequently and reached 12 percent in 2005 (Figure 8). Most o f this decline was due to the fiscal consolidation, which resulted in a reduction in credit to the public sector from 23 percent o f GDP in 1993 to 0.2 percent in2004. Credit to the private sector contracted from about 15 percent o f GDP to only three percent in 1996. However, since then it has steadily recovered and stands now at ten percent o f GDP. Sustaining and Sharing GrowthinTanzania 24 Figure 8. Domestic credit and interest rates, 1990-2005 40 1 1 20 35 n g l5 J 10 e i 5 I E o 1 -5 1990 1992 1994 1996 1998 2000 2002 2004 ECentralgovernmnt E Rwatesector 0 m e r ----tT-bilk --c-Lending ---*---Saving Source:Bank of Tanzania. Quarterly Economic Review. Various Issues. Figure 9. Savings and Investment (YOof GDP), 1990-2005 25- 20. B -. 15- c1 10- F- ' # 5 - m. ,- -. 1 `\r 0 - *', :`.- I ,,` I , , , , , , , , , , I 5 - -10- Formal credit in rural areas is very limited. For commercial banks operating inrural areas, most o f the collected savings are transferred to Dar es Salaam to be invested inassets that have a more attractive return risk profile than rural investment opportunities. Credit by traders and marketing organization plays an important role in agriculture, with interest rates significantly higher than commercial lending rates. Domestic saving has shown a dramatic increasefrom minus five percent of GDP in 1993 to about elevenpercent in 2005 (Figure 9). The increase insaving is almost entirely the result o f increased public sector saving. Between 1990 and 2005, public sector consumption declined from 18 percent o f GDP to seven percent. Private sector consumption increased initially from 83 percent of GDP in 1990 to 88 percent o f GDP by 1999but has declined subsequently to about 80 percent in2005. The difference between saving and investment, which corresponds to the current account deficit, declined dramatically since the early 1990s. Sustaining and SharingGrowth inTanzania 25 A key issue is the distribution of liquidity. Although aggregate indicators suggest excess liquidity, this is not the case for all the banks. Ninety percent o f the excess liquidity i s concentrated in three banks, and a large part o f this liquidity comes from government deposits. Moreover, a large portion o f this liquidity i s invested ingovernment securities. Financial markets are affected by the sterilization of aid inflows by issuing T-bills. The interest rate appears to be quite sensitive to changes inthe supply o f T-bills. At the same time, a fairly highrate o f monetary expansion has been consistent with declining rate o f inflation. This suggests that the economy may be able to absorb aid inflows without the need to fully sterilize them. Recommendations. Enhancing the access to finance requires (a) policies that foster saving; (b) monetary and fiscal policies that support growth incredit to the private sector; and (c) policies for the development o f the financial sector. With regard to the latter, government-in collaboration with the Bank and Fund-has prepared a Second Generation Financial Sector Reform Program. Implementation o f the reform program involves the following elements: Strengthening the lending environment and financial infrastructure. This includes completing the task o f divestiture o f state-controlled entities in banking and insurance; strengthening the legal and judicial framework supporting lending; clarifying and deepening the regulatory, information and technology infrastructure for households and micro enterprises, and encouraging long-term pension and insurance funds to finance longer-term private investments. Facilitating the increase in SME and Long Term Lending. Government is taking initiatives in five areas: (a) establish a SME Credit Guarantee Scheme to encourage commercial bank lending for SMEs; (b) launch a Development Finance Guarantee Facility (DFGF), to be managed initially by the BOT, which will provide partial government guarantees to commercial banks for their loans to development- and export-oriented projects; (c) facilitate the creation o f a privately owned and managed long-term financing facility (LTFF) that would channel funds from non-banks or banks and potentially development partners (without government guarantee) to be on-lent to commercial banks on a long-term basis; (d) introduce a Development Finance Institution (DFI), most likely incorporating the Tanzania Investment Bank. The DFIwould channel multilateral and bilateral donor funds and perhaps utilize government seed money from the budget, but not take any new deposits from the public; and (e) advance reforms inthe pension fund sector inorder to have a unified legal and regulatory framework for all pension funds, along with investment guidelines. It i s expected that this effort, particularly the development o f investment guidelines, will facilitate the channeling o f pension funds' resources into longer term lending through commercial banks. Direct support to providers of financial services for micro and small enterprises. The Second Generation Financial Sector Reform Plan includes initiatives related to micro and rural finance. These initiatives are designed to respond to the government's vision for the development o f pro-poor finance in Tanzania, as articulated in the National Microfinance Policy. Some o f these initiatives, such as the strengthening o f the regulatory framework will be addressed as part o f the strengthening the lendingenvironment and financial infrastructure. However others will require direct support to the providers o f financial services. It i s expected that a large part o f this support will be provided by the Financial Sector Deepening Trust (FSDT) funded by four bilateral development partners.6 The FSDT will provide assistance for the transformation o f microfinance NGOs, the strengthening o f networks o f Sustaining and Sharing Growth inTanzania 26 Savings and Credit Cooperatives (SACCOs), and the development o f links between banks and microfinance institutions. Achieving the objectives of the Second Generation Reform Program requires the timely implementation of the action plan. Considerable efforts have gone into developing the strategies and buildingconsensus. The challenge is now to begin implementing the plan. Enhancingthe Public-PrivateSector Interface Regulatory agencies, tax revenue authorities including customs, business and land registries, and the judicial system all form part of the public interface with the private sector that have an important bearing on the cost, risks and barriers to business in Tanzania. These costs will influence the range o f opportunities that are profitable, because investments are forward looking, risks and uncertainty will determine the types and nature o f investments, and entry restrictions will limit innovation and the efficient provision o f goods and services. Aside from the impact on the cost o f doing business, the quality o f the public private interface i s critical. This relates to the flow o f information between the private and the public sector, which allows government to play a supportive role to the private sector by removing obstacles, collaborating in the identification o f growth opportunities, and ensuring that the provision o f public goods and services (especially infrastructure) i s well aligned with private sector needs. An efficient public private interface i s part o f the second generation reforms that will determine the private sector's response to opportunities that are productivity based. The size of the informal economy is estimated to be large in Tanzania, suggesting high barriers to entry (Table 10). However, it also suggests the existence o f entrepreneurial potential which could become an important driver o f growth, if an environment for growth and formalization o f informal activities i s established. Among the main causes for informality are relatively high cost o f registration and licensing as well as tax obligations implied by formal registration. Recent reforms o f business licensing which abolish fees for small businesses are an important step to reduce barriers to entry. Table 10. Estimated size of the informal economy (% of GDP), various countries and regions ICountry ISize o f Shadow Economy IRegiodCountry I Size o f Shadow Economy 1 ("A of off. GDP) Group (% of off. GDP) Tanzania 60% Africa 42% Kenya 36% Asia 31% Uganda 45% South America 43% Mozambique 42% Central Europe and 38% SouthAfrica 30% IOECD 16% Poor customs administration and overly restrictive trade and customs regulations discourage enterprises from exporting. Trade and customs regulation partially explain why enterprises in Tanzania export less than similar enterprises in Kenya. Since exporting has been linked to improvedproductivity, these delays and restrictions can have a real impact on enterprise performance. The 2003 enterprise survey revealed that among enterprises that engaged in foreign trade, the median firm reported that it took 14 days on average for imports and 7 days for exports to clear customs, once they had reached the point o f entry or exit.' Managers also reported that clearance times were unpredictable, forcing them to hold additional inventory in anticipation o f worst-case scenarios. Port and customs delays are considerably longer inTanzania than inany o f Sustaining and SharingGrowthinTanzania 27 the comparator countries. The median delays for imports and exports in China were five and three days respectively-less than half the delays facing firms in Tanzania. Similarly, reported delays for imports and exports were seven and four days respectively in Kenya and seven and three days respectively inIndia. Reform of the country's legal sector to support and enhance private sector led economic growth and efficiency has not taken place at the same pace as economic reform. Problems affecting the legal sector identified by the Legal Sector Task Force in 1996 include inordinate delays in resolving disputes and dispensing justice in the justice system, inaccessibility o f the justice system for the majority o fpoor and disadvantagedTanzanians, low levels o fpublic trust in the justice system and excessive prevalence o f unethical behaviour.' The key challenges derive from the rapid social, political, economic and technological changes taking place in the country and internationally, and the need to for the legal sector to catch up with these changes to facilitate the efficient development o f a private-sector and market-led economy, and for due protectiono f consumer rights. Reforms o f the legal system to date have been slow and piecemeal. The establishment o f commercial courts in 2000 initially resulted in faster dispute resolution for businesses. However, even the commercial courts became quickly overburdened with large caseloads, which suggest the need for speeding up overall reforms and strengthening o f the judicial system. Grand and petty corruption have a negative impact on competition and impose high cost on businesses. There has been some progress inreducing corruption since 1996, when a presidential commission ledby Judge Warioba produced the Warioba Report. Government i s implementing a National Anti-Corruption Strategy and has strengthened the institutional framework-notably through the Finance Act o f 2001 and the Public Procurement Act o f 2004-and adopted a clear zero-tolerance position on corruption. Nonetheless, corruption continues to have an important impact on businesses. Enterprise managers who participated in a recent enterprise survey (World Bank, 2004b) see grand corruption-payments made to policymakers or senior bureaucrats in order to win government contracts and influence lawmaking-as a serious problem. Enterprises are also affected bypetty corruption-payments made to lower-level government officials to "get things done" in connection with customs, taxes, licenses, and other services. About 35 percent o f enterprise managers said that informal payments were typically needed for firms like theirs. Sustaining and Sharing GrowthinTanzania 28 Figure 10. Requests for bribesare especiallycommonduring tax inspections 30% 375 400 AIn v) .-% E 300 20% $a :10% c1 a C 200 CI E0 100 6 I- 0% 0 W Note: "Percentage reporting bribe requests" only includes enterprises that were inspected; "median account" only includes enterprises that reportedpositive amounts Source: WorldBank 2004b. Recommendations. Establishing a business environment which encourages the entry o f businesses into the formal sector and which supports competitiveness by minimizing transaction cost i s critical for Tanzania's economic growth. Priorities o freform include: 0 Revision of licensing and registration legislation to consolidate recent reforms of business licensing; 0 Review o f tax regime for micro and small enterprises; 0 Strengthening o f customs and trade regulations and administration; 0 Accelerating legal sector reforms and strengthen capacity inthe sector; and 0 Revision o f anti-cormption legislation to provide more effective instruments for the control o f grand and petty corruption. FOSTERING INNOVATION, PRODUCTIVITY, AND TECHNOLOGICAL CHANGE Tanzania's prospects for sustained economic growth depend significantly on the capacity of the private sector to innovate and to translate innovation into new business activities. Innovation i s the key driver o f economic growth rather than a by product o f the growth process. Inthe Tanzanian context, innovation covers the expansion of the range and quality of products and services and the adoption o f technologies that enhance the competitiveness o f the Tanzanian economy, at the firm, industry, and national level. At the micro level, there is the adoption and adaptation o f available technologies for use by firms, individuals or households which helps to improve their productivity, welfare, living conditions, and so on; at the sectoral level, there i s the development o f new industries, generally based on foreign technologies that can be a source o f Sustaining and Sharing GrowthinTanzania 29 newjobs, income and exports. There are four pillars that are generally considered to be important for countries to support the process o f structural change: An economic and institutional regime that provides incentives for the efficient use o f existing knowledge, creation o f new knowledge, and the flourishing o f entrepreneurship. An educatedand skilledpopulation that can create, share, anduse knowledge well. An eficient innovation system of firms, sciencehesearch centers, universities, think tanks, consultants and other organizations that can tap into the growing stock o f global knowledge, assimilate and adapt it to local needs, and create new knowledge. A dynamic information infrastructure that can facilitate the effective communication, dissemination, and processingo f information. The following sections discuss these key drivers o f innovationand productivity and provide recommendations aimed at enhancing the pace o f innovation and productivity inTanzania. Investing in education Tanzania's economy does not have a sound base of adequately qualified and trained work force that is essential for rapid economic growth and effective diversification of its production and export bases. But Tanzania has made some progress in education. Figure 11 shows that in 2001, Tanzania's adult literacy rate at 77 percent was higher than that o f Uganda (69 percent), but lower than that o f Kenya (84 percent), Botswana (79 percent), or South Africa (86 percent). In addition, according to Cohen and Soto, in 2000 (Figure 12), Tanzania's average years o f schooling (3.4) was higher than that of Uganda (3.22), but lower than in Kenya (5.08), and far from that inSouth Africa (7.22). Figure 11. Adult Literacy Rates, 1960-2000 Adult Literacy Rates I I Tanzania I 2 6 7 , , , , , , , , , , 1970 i972 1914 is76 le78 tM0 1MZ i 9 M iM6 1911 1990 1992 1994 i996 1998 2004 2002 Source: World Bank.DevelopmentDataPlatform. Sustaining and Sharing Growth inTanzania 30 Figure12. AverageYears of Schooling, 1970-2002 Average Years of Schooling (Cohen-Soto) 1960 1970 Source: Cohen and Soto, 2001. Tanzania's gross enrollment ratio (GER) for primary education increasedfrom 78 percent in 2000 to 113 percent in 2006, following the abolition of primary school fees and the roll out of the primary education development program. A recent assessment (SACMEQ 11) suggests that the quality o f primary education i s not worse than in other African countries. A key challenge i s to maintain and enhance the quality o f primary education while the sector expands. Tanzania has still one of the lowest secondary net enrollment ratiosin Sub-SaharanAfrica, despite rapid progress in recent years. Only about 13 percent o f the relevant age group attended secondary education in 2006, compared with an average o f 27 percent for Sub-Saharan Africa. To improve access and quality o f secondary education, the government has launched the Secondary EducationDevelopmentProgram(SEDP) in2004. Levels of tertiary education are also very low in Tanzania. The tertiary GER stood at 1.2 percent in 2006, as compared to 3.24 percent for Uganda and 3.52 percent for Kenya in 2001, while Botswana and South Africa had tertiary GERs o f 4.69 and 15.05 percent, respectively in 2002. The government has been the major financier o f technical/vocational education and training (WET), but the system in Tanzania faces several problems in terms o f inefficient resource utilization, inequitable distribution o f educational opportunities, poor labor market linkages, and a lack o f coordination between donors and the government. Analysis of the Integrated Labor Force Survey data suggest that marginal returns are around eight percent for every year of primary education, but increase sharply for additional years of lower secondary education. This data provides strong support for government's investment in expanding primary education and the current focus on secondary education (Figure 13). Sustaining and Sharing GrowthinTanzania 31 Figure 13. Marginal social returnsper year of educationbasedon integratedlabor force survey 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% Primary Low Secondary Up Secondary Univ Source: Author's calculation based on ILFS data Analysis of rates of returnto education in the manufacturingsector suggests that marginal returns to education increase significantly for higher education compared to primary and secondary education (Figure 14). The difference in the profile of marginal social returns to education for workers in the manufacturing sector and the overall laborforce may suggest that at the tertiary level there are certain degree programs that are well rewarded by the manufacturing sector, but that many other degree programs result in lower paying jobs. The implication ffom this would be to shiftthe supply o fhigher educationto those programs that seemto be indemand by the manufacturing sector. The difference in the profiles o f social returns to education also suggests that a limited supply o f workers with relevant tertiary education i s a constraint for manufacturing, while for other sectors the limited supply o f Tanzanian's with secondary slulls may be more o f a constraint. Figure 14. Predictedearnings in manufacturingsector basedon manufacturing firm surveys '1 Source: Soderbom et al. (2004) Sustaining and SharingGrowthinTanzania 32 Tanzania is no stranger to the brain drain phenomenon.The most vulnerable occupations at the national level include medicine, accountancy, law, engineering, and science based occupations. L o w salaries for doctors are the principle reason driving the brain drain-and those that remain seek higher wages in private hospitals in large urban centers, leading to a lack o f doctors in some o f the country's district hospitals, Inaddition, HIV/AIDSposes significant risks for humanresource development inTanzania. Recommendations, Key challenges facing Tanzania in education and human resource development include: Sustaining and improving the quality o f education as enrollments increase by recruiting teachers, constructing classrooms, increasing preservice teacher training, and providing subsidies for purchasing teaching and learningmaterials. Ramping up secondary education, including improving its quality and relevance to the needs o f the economy. Inhigher education, strengthening the governance and administration of the country's three public universities in terms o f financial sustainability, up-to-date content, and teacher training. Usingthe potential o f distance education to expand access to education services while improving equity. The Open University o f Tanzania offers degree programs by correspondence and also in regional centers. The costs are low, because the state covers tuition. But enrollments are low, partlybecauseo f lack o f content andpartlybecause o f a dearth o fpartnershipswith international academic institutions that could provide degree programs online. Combining distanceeducation modalities with extended face-to-face interactions with Tanzania's other public universities may be one way to boost enrollments andincreaseaccessto higher education. Reforming teaching methods and the curriculum at all levels to include skills and competencies (communication skills, problem-solving skills, creativity, and teamwork) to meet the new needs o f the economy. Increasing the interface between industry and education and offering differentiated curricula that better meet the new skill demands o f industry, generated by changing markets and technologies. Harmonizingthe technical education offered in secondary schools with that offered intechnical colleges and then linking these schools with the proposed zonal and regional institutes and colleges. These institutes and colleges should offer differentiated products to meet the differing needs o f industries, such as mining, fisheries, major cash and food crops, external trade, and metal. Devising strategies to proactively deal with problems o f skills lost throughbrain drain. Strengtheningthe innovationenvironment A recent study by the World Bank (Chandra 2006) identifies a host of government actions that are essential to the development of competitive industries (Figure 15). In order of importance, firstly, these include actions that make available appropriate skills and provide Sustaining and Sharing GrowthinTanzania 33 support for technology acquisition and development. This i s followed by actions o f a regulatory nature, such as relating to standards and quality control. Then come various types o f support measures that can be provided to enterprises and industry organizations for export promotion, investment, and so on. Table 11. Role of the Public Sector in Fostering Innovation development Yource: Based on information inChandra (2006). In Tanzania, government institutions in support of innovation lack adequate resources, infrastructure, equipment and trained personnel to respond to the increased needs of the local entrepreneurial society. The low level o f R&D as percentage o f GDP-only 0.2 percent (comparable to other African countries), reflects the modest nature o f Tanzania's research and innovative effort. There are currently about 62 R&D institutions in Tanzania covering agriculture, including livestock and forestry (28), industry (1l), medical (1l), and wildlife fisheries (4) and Universities and other higher learning institutions (9). Inaddition to the lack of resources for R&D institutes, mechanisms for technology diffusion are modestly developed, with a quasi absence o f decentralized structures. FDI is an important source of technological upgrading in Tanzania. FDI-which i s relatively high in Tanzania compared to other African countries-has played a key role in the modernization o f important sectors o f the economy such as trade (retail), banking, tourism and the telecom networks. Ithas also been crucial inthe take-off and growth o fnew industries such as the fishing and the gold mining industries. However, international experience suggests that it takes time to build an indigenous innovative capability through foreign investment and that this requires explicit mechanisms such the employment o f large contingents o f local cadres -at managerial positions as well as programs to closely link local suppliers of components and materials to upgrade their equipment and the quality o f their products. Mechanisms such as these do not exist in Tanzania, and consequently, the transfer o f knowledge and technology from foreign sources remains modest. Recommendations. In addition to improving the overall business environment and the upgrading the education system, specific actions for the promotion o f innovation and technology diffusion are needed in order to put Tanzania's R&D infrastructure in the service o f the country's development. There i s a need for a systemic approach that provides complementary support on three basic aspects: financial, technical and regulatory. Sustaining and Sharing GrowthinTanzania 34 Financial support: There is a crucial lack o f resources on the demand side for accelerating the design, testing, use and dissemination o f technologies in the Tanzanian context. It i s suggested that Tanzania establish two complementary schemes, based on simple matching fund principles, that provide-in grant form-50 percent o f the funding required for the development (R&D phase) o f small and medium sized projects (for example, up to US$20,000): 0 User scheme: This scheme would allow particular groups and communities to buy needed technologies, and provide, if appropriate, complementary in-kind funding (labor for community purpose). 0 Developer scheme: This scheme would fund 50 percent o f technical services or R&D projects undertaken by SMEs with R&D institutes (public, academic, etc), thereby helping indirectly, but more effectively, the R&D institutes use their competences to serve communities (note that this type o f scheme i s in place in a number o f advanced countries). Technical support: On the basis o f experience accumulated in industrial countries, it i s suggested that a network o f locally based and owned structures be established to serve the needs o f rural and urban communities for technical advice, information, and assistance (in design, marketing, and so forth). These structures should be adapted to different sectors (for example, extension services for agriculture and design and manufacturing workshops for industry).They should also be conceived and operated as antennas o f central bodies to which they would be strongly connected through information technology (IT), databases, and the like. They should be established on a clearly expressed demand from local communities and funded on a 50/50 cost-sharing basis, with local organizations (municipalities, business or farmer associations, and so forth) matchingthe resources put inbythe central government. Regulatory support: Regulatory-relatedactions should address several issues: Proposed actions that aim to stimulate service-based contracts and formalize new links between the business sector and the R&D infrastructure require the establishment o f clear legal and administrative procedures. We therefore recommend reviewing, adjusting, and standardizing appropriate models for such relationships. Many issues may be involved, including the use o f public or university laboratory equipment and personnel by firms, the temporary employment o f university researchers by business enterprises, and intellectual property rights. Inmany sectors, there is a need to develop quality awareness and quality control, as well as related accreditation and certificationprocedures. A program should be implementedto raise awareness o f these issues, because doing so could yield important results ina short time span. Firmsor individuals that are first producers face major financial problems and frequently do not get access to credit from the banking system. The government's recently established credit guarantee mechanism, which i s supposed to mobilize the banking and financial sector, does not seem to be working well. An audit needs to be undertaken to examine in detail the mechanisms that can be put in place to complement this incentive, such as microcredit schemes; equity investment procedures (such as the Dutch Program for Cooperation with Emerging Markets, which supports 30 percent o f the investment o f individual firms in the flower industry); and the like. Sustaining and Sharing GrowthinTanzania 35 These reforms could be supported through a multipurpose innovation facility that would target specific priority industries, including the tourism sector and the agro-food industry. The first i s already an important source o f income for Tanzania; the second would take advantage o f the large agriculture base o f the country. Systemic, well focused action would be needed in both cases. The involvement of foreign enterprises is crucial. These enterprises provide access to foreign markets and management competencies, and introduce up to date technologies. It i s important to establish inliaisonwith Tanzanian Enterprise Associations, efficient organizations for negotiating and partnering with foreign businesses. Clear contracts regarding technology licensing, personnel training, and access to export markets should be developed. On the whole, a two pronged action strategy combining both the upgrading and developing domestic capabilities and involving foreign actors would be essential for a successkl innovation intervention to help improve Tanzania's growth prospects. Enhancingthe roll-out and use of ICTs Rapid advances inICTs are dramatically affectingeconomic and social activities, as well as the acquisition, creation, dissemination, and use of knowledge. These advances are affecting the way in which manufacturers, service providers, and governments are organized and how they perform their functions. As knowledge and innovation become increasingly important elements o f competitiveness, the use o f ICTs i s reducing transaction costs, and time and space barriers, thereby allowing the mass production o f customized goods and services and substituting for limited factors o f production. Compelling evidence suggests that strengthening telecommunications infrastructure and service i s pivotal inpromoting trade and economic growth. Fink et al. (2002) estimate that a 10 percent decrease in the bilateral price o f phone calls is associated with an 8 percent increase in bilateral trade. InAfrica, a telephone growth rate o f 10 percent instead o f 5 percent (and growth in electricity generation o f 6 percent instead o f 2 percent) could generate an increase inAfrica's growth by at least 0.9 percentage points. The information infrastructure in a country consists of telecommunications networks, strategic information systems, policy and legal frameworks affecting their deployment, as well as skilled humanresourcesneededto develop and use it. Inthe ICT domain, Tanzania i s still at a very nascent stage o f ICT application and use. Progress has, however, been made under the 1997 National Telecommunications Policy, and this trend is expected to continue. The mobile telephone market for one i s fully competitive in Tanzania. Significant liberalization has taken place in various segments: private operators are now providing basic, mobile, data, paging, internet, pay-phone and other value-added services. The mobile-telephone market involves a number o f operators, and i s growing rapidly. Table 12 shows that overall teledensity-mainlines plus mobile phones-has increased to 24 per 1,000 in 2003. Anecdotal evidence suggests that mobile phones are increasingly being used in Tanzania to get business-related information and to reduce transaction costs. For example, traders inDar es Salaam now can place orders with producers ofbananas throughout the country-thus linkingdemand and supply inrealtime and enhancing the efficiency o fmarkets. Sustaining and SharingGrowthinTanzania 36 Table 12. ICT Indicators(per 1,000 people),2002 1996-2003 2002 2000 2003 Mobile Personal Internet Internet Mainlines telephones Radios Televisions Computers Hosts Users Sub-Saharan Africa average 15 3 1 198 69 11.90 3.10 6,233 Source: World Bank.African Development Indicators2005. Despiteincreasingcompetition,tariffs remainhighandteledensity is one ofthe lowestinthe SADC region. This is mainly due to the poor interconnection framework, lack o f regulatory independence, and other issues such as lack o f infrastructure sharing. Thus, in general, the country's postal and telecommunications services are weak, and the provision o f telephone lines (fixed lines) has been meager. An inadequate regulatory framework persists, and competition has been hampered by various issues, such as inadequate interconnection agreements/directives, high level o f fees and royalties levied by the TCRA, and absence, or non-transparency o f regulatory oversight. Recommendations. Inorder to strengthenits information infrastructure, the following issues are critical: Finalizing and adopting the new electronic communications bill, which i s key to defining the ground rules for sector development (including rural areas). Implementingthe new converged licensing framework, which will ensure further liberalization o f the market. Reviewing and modernizing telecommunications policies and regulations to generate fair competition andreduce highcommunication and operational costs. Building capacity to undertake such reforms, including through the establishment of systems and processes to review the performance o f the regulatory institutions. For example, given the great demands and expectations placed on the regulator (TCRA) by telecommunications sector reforms, the Swedish government, through the Swedish International Development Cooperation Agency, i s helping TCRA create capacity to meet its existing and hture challenges and learn from its experiences inoperatingina more competitive market. Supporting the development rural telecommunications infrastructure, such as by developing universal access schemes. Rural areas lack telecommunications services or have only limited access in areas adjacent to main towns and on major trunk roads. This effort requires developing content inlocal languages (such as Swahili). Enhancing technical and business-related skills development among the population using ICTs through technical institutes and vocational centers. For example, the University o f Dar es Salaam i s offering ITtraining inits computer center to the public. Sustaining and Sharing GrowthinTanzania 37 0 Continuing to use global experiences to enhance the efficiency o f the telecommunications sector. In many areas o f telecommunications reform, Tanzania has benefited by adopting best practices from both industrial and developing countries. The functions and roles o f the national regulator (TCRA) are the best example. Further benefits from global experience and best practices depend on the capacity o f TCRA and other institutions to learn from the experiences o f other countries. Sustaining and Sharing Growth inTanzania 38 3. MAKINGGROWTHPRO-POOR I n order to ensure that Tanzania's growthprocess includes rather than marginalizes the poor, the report highlights four areas that requireparticular attention. First are reforms inthe agriculture sector which are likely to have a direct impact on the poor, as most ofthem derive their livelihoods from this sector and their transition to other sources of income is likely to be gradual and slow. Second are micro and small enterprises, most of which are presently inthe informal sector. This part o f the economy has been an important path out of poverty inthe past and is likely to continue to do so infuture. Thrd i s a focus on Tanzania's wealth o f natural resources, which if properly managed can provide an important source o f income and employment for the poor, but also of revenue for government for the financingof Tanzania's Strategy for Growth and Reduction o f Poverty. Finally, and probably most important, i s a focus on strengthening the capacity o f the poor to contribute to and benefit from economic growth. This includes not only access o f the poor to education, nutrition, and health services, but also issues related to vulnerability andriskas well as populationpolicies. SUPPORTING AGRICULTURE SECTOR DEVELOPMENT As primary agriculture contributes more than 40 percent of GDP and employs 63 percent of the labor force, growth of this sector is central to Tanzania's overall growth performance and its success in reducing poverty. Recent agricultural growth of 4.8 percent annually between 2000 and 2005 is moderately highby regional standards, but needs to increase inthe future to have a significant impact on poverty reduction, especially inrural areas. This growth has occurred in a very challenging external price environment. As prices o f traditional exports fell, farmers quickly increasedproductiono f exportable food crops such as maize, beans, pulses and, horticulture, and successfully increased production of import competing goods such as milk and dairy products, sugar, rice, and oilseeds. The yield trend has been positive for rice, some horticulture, potatoes, bananas, and probably for milk and dairy, but disappointing for other agricultural products. Livestock growth was slower, but shows significant potential due to increasing demand for meat and related products at home andinAsia andthe MiddleEast. Tanzania's past growth has depended on the expansion of area cultivated, and, as is the case in much of Africa, has been accompanied by only modest increases in labor productivity. As shown in Figure 15, labor productivity inAfrica has trended up only since 1994. Tanzania's average annual increasein labor productivity o f 1.1 percent is close to the normfor its neighbors. The figure also highlights the emerging disparity inlaborproductivity between Tanzania and Asian countries. The gap in labor productivity levels increased from less than 10percent in 1990to more than 40 percent in 2002. Sustaining and Sharing Growth inTanzania 39 Figure 15. Labor productivitylevels in Tanzania and comparators AgriculturalLabor Productivity (ag value added per ag worker) 380 , *Asia - - - Developing -Sub-Saharan Africa -Tanzania t Notes: (9countries with incomplete data forseries excluded;@)AsiaDeveloping includes India&China; & (3)Sub-SaharanAfrica excludes South Africa. Source: Author's calculations basedon data from World Bank Development Data Platform With very modest increasesin labor productivityand yields almost stagnant, Tanzania's growth derived largely from land expansion by farms usingtraditional technology, with little evidence of improved technical inputs and management practices. Examples of land expansionthrough labor saving innovation have been localized (for example, expansion via animal traction in the Rukwa region), and undoubtedly account for much of the land expansioninexcess of labor force. But for the sector as awhole, the growthpatternof the past replicates poverty, rather than reducing it, since households do not experience income growth. Moreover, land expansion using existing techniques carries environmental costs, as forests are encroached upon and increasingly marginal land comes into cultivation. The 2002/2003 Agricultural Sample Survey suggests that there has been no further land expansionsince theprevioussurvey carriedout in 1998/1999. Table 13. Land use and potentialfor land expansion(mid 1990s) Urban 65 0.1% Rural protected land protectedforest/woodland 13,838 1 5% otherprotected(wildlife, nat.park) 13,291 14% 1 Agricultural land currently used (10.8 million hatotal) temporary crops 3,700 4% pasture 6,150 6.5% permanentcrops 950 1.5% Rural unprotectedland available but not used - unprotectedforestlwoodland 26,321 28% suitable for croppingbut unused 7,000 7% grasslandbushlandnot suitable for cropping; may be suitable for grazing" (may includesome water area) 23,221 25% Total land area 94,536 100% ~ Source:FA0 and World Bank 2001. Table 13, based on aerial surveys done in the 199Os, suggests that Tanzania still has significant potential to expand agricultural production through land expansion. However, in order to improve agricultural incomes, land expansion needs to go hand inhand Sustaining and Sharing GrowthinTanzania 40 with improvements in agricultural productivity through greater investment in land improvements, both on existing and newly cultivated areas, by adopting yield increasing technologes, and by shifting the composition of output toward products with higher value added. Empirical analysis of two representativesurveys of farm householdsinKilimanjaro and Ruvuma - two cash crop growing regionsinTanzania shows that poorer householdsdo - not only possess fewer assets, but are also much less productive." Agricultural productivity directly affects household consumptionand hence overall poverty and welfare. Stochastic production frontier analysis indicates that many farmers are farming well below best practice inthe region. Holding inputs constant, they attain on average only 60 percent of the output obtained by their most productive counterparts. Analysis of allocative efficiency suggests that family labor is substantially over utilized, a sign of considerable excess labor supply. Use of intermediate inputs on the other hand is well below what is commensurate with'the estimated value of their marginalproductivities. An important reason for low input use i s lack of credit to purchase inputs, but difficult access to the inputs themselves andbeing connected to the economy more broadly are also important impediments. Easy access to credit is positively associated with being a member of a savings association or being in a contractualarrangement with a cooperative or fm. Irrigation infrastructurefacilitates access to credit. Together these findings support a continuing emphasis on increasing agricultural productivity in Tanzania's poverty reduction efforts. Better agronomic practices and increasedinput use will be crucial inthis strategy. Better access to inputs and improvedroads and transport services will further help boost input application. Financialconstraints might be relieved through fostering institutional arrangements facilitating contract enforcement (e.g. contract farming, marketing cooperatives) and institutions that facilitate saving by the households themselves. They may also be relieved by the provision of more adequate consumption safety nets. The overall results suggest that a pro-poor rural development strategy needsto be anchored aroundimprovements inagricultural productivity. Owingto the many farm and marketlevelconstraintson smallholder producers, there is a vital, positive role for national government and local institutions in enabling agricultural growth and rural poverty reduction. Removal of constraints on agricultural marketing,processing, andfannproductivity will require: 0 Improvedimplementationoflandtenure reforms; 0 Expansionof agriculturalresearch effort, and continuedresearch and extension focus on client responsiveness and engagement of farmers in the research process, and strong emphasis on sustainableuse of landandwater resources; 0 Irrigation improvements; and 0 Support for improved functioning of output and input markets, and for associated rural services, including finance. Improved implementation of land tenure reforms. Although land appears to be ample, rural areas suffer from a frequency of disputes over land and insecurity of tenure that one would expect to see only in a country with higher density of settlement and hence greater competitionfor land. To address the growingproblems, the country passedthe nationalland policy (NLP) of 1995 and subsequently revised the land legislation, to provide more expeditious access, enhanced security of tenure, and better management of land as a natural resource. A plan of actions and investments to implement the laws was drawn up in 2005. Appropriate funding and implementationof thisplanwill be critical to its successs. Technological change to foster growth: generation, dissemination, and adoption. Three elements must be present in the technology system to support increased productivity. Researchers must have developed new varieties that can be grown with success and profit Sustaining and Sharing GrowthinTanzania 41 under conditions that smallholders face; and efforts must be made to improve the environment for investment in private input supply networks. Producers need information about the availability o f the varieties and guidance on how to use them. And money must be available to cover the costs o f early adoption o f the new varieties and techniques, while ensuring that a broad range o f technical options, including low cost ones, is available to farmers. All elements o f the technology system; Le., the knowledge, and the financing must be linked ina way that provides smallholders access. The contribution of agricultural research, extension, and empowerment of farmers in the 1990s and early 2000's was positive, but limitedby severalfactors. Services generally focused on increasing production through short-term technical packages, without paying attention to farmers' circumstances, markets, and sustainability. Despite various attempts to strengthen them, the linkages between research, extension and training were weak, and collaboration between public and private partners limited. Disproportionate emphasis was placed on generation and dissemination o f technology, and less on empowerment (both intellectually and financially) o f farmers to adopt the technology. As a consequence o f weak linkages within the system, research did not always address priorities o f greatest potential impact to the production systems. Technical breakthroughs did not yield good economic returns. And promising technologies remained on the shelf due to lack o f knowledge or financing for adoption. Ruptured links inthe technology chain reduced returns to investments in each o f the elements. Moreover, the system was underfinanced, but given the somewhat depressed returns resulting from the institutional deficiencies, underfinancing was an appropriate response. Under the Agriculture Sector Development Program (ASDP) which was launched in 2003, a major reform of the institutionalstructure of the agriculturaltechnology system has been designed and is under implementation. This provides a foundation for hture growth that merits increased public financial support (Box 5). Box 5. Underlyingprinciplesfor the designof an agriculturaltechnology system Demand-driven and market-led technology development and adaptation: Farmers select, test, compare and adapt appropriatetechnological,service and marketoptions. Pluralism of providers of services and approaches: diverse methodologies, processes and funding, as well as service providers are supported. Public funding for the system remains important, but services can be providedby public extensionworkers, NGOs, andprivate advisors. Subsidiarity: A constructivedivision of labor betweennational, district, and local levels is maintained. At the national level, the extension service feeds knowledge into the system, though training, identification of new approaches and technologies, and preparationof materials. Service standards are also defined and enforced at the national level. At the local level, organizations are able to contract the most suitable service providers,both public andprivate. Focus on economics and natural resource management, HIV/AIDS and malaria, as well as technical solutions: Agricultural service providers assist on issues of economic decision making and managementof soil, forests, and water. Transparency and accountability: Accountability is built in through performancecontracts, monitoring of performance, and ability of farmers to choose their providers. Farmers' feedback on services is integratedinto the periodic evaluationof service providers. Irrigation to raise productivity and incomes. Tanzania is well endowed with water, both on the surface and below ground, but suffers nevertheless from water shortages due to insufficient capacity to store and access it. Cumulatively, the lakes, wetlands and aquifers provide huge natural storage capacity. The country also has 2.7 million hectares o f wetlands Sustaining and Sharing Growth inTanzania 42 (Usangu, Malagarasi). The total renewable water resource in Tanzania is estimated to be around 80 cubic Myear, ofwhich 30 cubic M y e a r are groundwater (FAO, 2004). Tanzania's ample water is matched by ample land suitable for irrigation. Of the total 44 million hectares suitable for agricultural production in Tanzania, only 10 million ha i s under cultivation and out of this only 200,000 ha i s irrigated. This represents a mere 2 percent of total cultivated area in the country. The estimated irrigation potential is up to 2 million hectares. Approximately three quarters of the presently irrigated area is farmed by smallholders in about six hundred small-scale irrigation schemes, usually using diversions and/or furrows in one of the nine major river basins. Very little irrigation is at present based on abstraction of ground water, which providesa promising area for future development with direct and affordable benefitsto the poor. Rice is by far the most important crop irrigated in Tanzania, but sugarcane is also irrigated. A suitably designed ground water irrigation system (open wells, borehole tubewells, rainwater harvesting) could reduce the reliance on large bodies of water, including rivers and lakes, and promote more sustainable use of locally sourced and managed irrigation systems. Since the surface water available varies with rainfall, open wells and borehole or tube-wells can be constructed to spread the availability of water throughout the growing season. Compared with large surface irrigation schemes, the design of which is driven by topography and hydraulics, ground water development is often much more amenable to povertytargeting and is generally less capital intensive. Ground water irrigation can complement that based on surface water. Integratingground water abstraction with rain water harvestingand watershed management, alongwith efficient water distribution systems, will leadto reliable, cost-effectiveirrigation systems. Irrigation at present is constrained by the high cost of investments required and by limitedprofitabilityof their use. Equipment, even the relatively modest implementsneeded for localized access to ground water, is more expensive in Tanzania than in, for example, India, by a factor of about three (FAO, 1997). To compound the adverse impact of high initial costs, producers face difficulties accessinghighyielding varieties and moving products to market. Irrigation and agricultural productivity are clearly intimately linked, and neither can advance substantiallyindependentlyfrom the other. Policy reforms in several key areas are needed to underpin growth in irrigation. Administrative regulations and restrictionson marketing and trading of irrigation equipment need to be simplified or removed. Proceduresfor the import of irrigation equipment such as drilling machines, pumps, etc. need to be simplified. Instruments such as micro-finance lending, matching grants, and joint ventures should include among eligible projects the smallholder groundwater irrigation projects (micro finance, joint venture financing). Development partners and government could join forces with NGOs such as International Development Enterprises (IDE) to enhance the supply of low-cost irrigation equipment (e.g. treadle pumps, drip irrigation). Finally, actions that affect the profitability of irrigated agriculture, suchas the ongoing adjustment inthe real exchange rate, continuedliberalization of trade, and reforms of crop boards, will support an expansionof the area under irrigation in Tanzania.. International experience also highlights the importance of an appropriate institutional and incentive framework for the sustainable development of smallholder irrigation (Table 14). Sustaining and Sharing Growth inTanzania 43 Table 14. InstitutionalFrameworkfor Sustainable Developmentof SmallholderIrrigation Systems 1 r- reliance 1 Farmersknowledgeableabout water managementand 0 & M I 1 . Appropriate choice oftechnology Attention to environmentalissues Financial Self- i Rationalization ofthe tax regime that small farmers face reliance Better access to financial services, especiallysavings I Active private firms in supplying equipment and implements Institutional and 4 Clarity on roles andresponsibility of public servants at district andnational levels organizational IrrigationSection, ZIUs and LGAs support --------- Strengtheningheformof Legal attentionto landtenure, water rights, ownership andresponsibility of irrigation infrastructure --- Investmentclimateto Improved access to advisory services Capacityto collect water fees and pay 0 & Mcost support growing constellationof small firms manufacturing Agricultural marketing and producer prices. Agricultural marketing and intermediary costs have decreased over time as policy reforms during the 1990sreduced price interventions and eliminated monopoly purchases by government bodies, thereby allowing greater scope for private sector trade and investment. However, a recent study (Nyange, 2005) suggests that marketing margins are still significant in relation to final prices, accounting for 30 to 50 percent o f the border value (fob). The components o f marketing margins vary across crops and locations (Figure 16). Large items can include taxes and fees, the trader's margin, the processor's margin, transport, finance, or packaging materials. I s there scope for reduction in these? In some cases, perhaps not: for example, packaging material costs may simply reflect the 111import cost o f the required materials. But for other items it may be possible to reduce costs via public sector investments (intransport, or inpower and water, which are particularly important for processing) or policy (tax and regulatory). Clearly, the public sector role in improving prices for farmers via cost reducing measures, and the scope for such reductions, depends on local conditions and the specific crop under discussion. Inmany cases, the specific remedy will require knowledge o f the local cost constraint and o f tradeoffs between removal o f the constraint and other objectives (e.g. tax revenue needs). Thus, there is an important role for district level growth strategies in identifying marketing constraints and possible solutions. Insome cases, local bodies will be able to address the problems on their own. Inmany others, they will need to collaborate with outside service providers, national or regional government bodies, and private sector groups. Sustaining and Sharing Growth inTanzania 44 gure 16. Cost components of marketingmargins vary significantly with crop and location 4rablca coffee Robustacoffee Tobacco Cashews Klllmanjaro Kagera Tabora 1 Coast 1 Mtwara 1 Shz:::ga 100% 10 Traders'Margin 90% Taxes 80% 70% 0 Feesand commissions 60% Finance 50% 40% I Processing 30% Packaging 20% materiaI 10% ITransport 0% Source: Nyange 2005. Producer margins are also crucially affected by labor costs - especially harvesting and weeding. Producer margins are sometimes very high; but as with traders there may be non- pecuniary marketing costs that are hard to observe (e.g. search for a buyer). Furthermore, since the farmer typically travels a considerable distance to sell at market, transport costs between farm and market are an important unaccounted-for component o f the producer's margin. Crop boards exert a significant influence on the environment for production of and investment in Tanzania's key agriculturalexports. A recent study (Kolavalli and Beddies 2004) covering coffee, tea, cotton, and cashew nuts, highlights a number o f problems attributed to the current activities o f the boards, including disruption o f marketing and exports, costs in excess o f services received, and interference in growth o f the private sector inmarketing. At the sametime, the boards are recognizedto offer services that mightnot be provided by the private sector, particularly input supply for smallholders, and regulation of monopsonistic behavior by private marketing agents. Reform of crop boards is critical to enhance the export competitiveness of Tanzania's agriculture. Key priorities for reform are a clear separation o f the public and private functions o f the crop boards and the appropriate alignment of financing, in order to ensure accountability to stakeholders. Key recommendations with regard to coffee are to make the coffee auction voluntary (including reviewing the relevance of the 'one license rule'), reduce license fees, and discontinue the input voucher scheme. Other reforms include opening o f coffee cherry buying to those who invest in common pulping units; ensuring accurate weights and Sustaining and Sharing GrowthinTanzania 45 measures; and ensuring an effective legal framework for branding and appellation. Incotton there is a needto deal with coordination to overcome problems associated with quality as well as supply o f inputs and credit more effectively. Innovative pilots are needed in the cashew sector as well to learn more about institutional arrangements conducive to quality improvement. The forced saving programs that seem to be critical to makinginputs available to producers (e.g. the passbook program) can be made more effective by enhancing stakeholder ownership and control over these programs. The public sector also has an important role in supporting agriculturaldevelopment. It not only needs to spend enough public money, but must also spend it well. This requires developing responsive local government and national institutions that support local growth initiatives; improving local government technical capacity and finance to planfor growth; and increasing accountability and transparency. In order to enhance the efficacy o f public spending in agriculture, it is necessary to focus on core functions o f public institutions inthe agriculture sector, carry out results-based assessments o f public expenditure, with expenditure increases focused on areas where there is a clear labor productivity payoff for smallholders. There is also a need to learn about the impact on technology adoption and smallholder marketing o f recent "smart subsidy" pilots, as for example, the matching grants scheme under the District Agricultural Development Grant. Spending on agriculture is increasing, and this is a positive indicationof willingness to rebalance the budget in favor of competitiveness. At the same time, the efficiency o f expenditure in the current allocation relative to goals o f growth and competitiveness appears to below. Too little is spent on research, extension, irrigation, livestock services, roads, land administration, and energy. Too much is spent on input subsidies and probably on the strategic grain reserve. Expenditures on "Crop Development" inthe budget o f the Ministryo f Agriculture and Food Security, and Cooperatives warrant careful evaluation, since they are quite large. The consistency o f present expenditures with ASDP understandings is not readily apparent in the budget as it is now presented. This hinders assessment o f the consistency of public expenditures with growth and poverty reduction aims. Alignment of expenditure around core and complementary activities that support the ASDP would contribute to efficiency of expenditure. The core items would be those production and research items currently included under the MAFC mandate, livestock under the Ministry o f Livestock Development, and marketing and finance under the Ministry of Industry, Trade and Marketing. Activities supported under these mandates presently take place at both national and at the local levels, and spending at the local level will increasingly be channeled through the Prime Minister's Office-Regional Administration and Local Governments (PMO-RALG). As local governments increase their capacity and as District Agricultural Development Grants becomes operational, public institutionalresponsibilities for agricultural development will not be defined by traditional ministerial prerogatives, but rather by local capacity, i.e., that national level institutions only undertake activities that districts and local bodies cannot handle. By 2007/08 three quarters o f expenditures to support the ASDP are expectedto flow through district governments. As participants in the political process become more familiar with the ASDP and the implicationsfor organization of public support, a reconfiguration of responsibility at the national level may be appropriate and feasible. Such a reconfiguration would imply merger o f the departments and entities that presently have responsibility for elements o f the ASDP at the national level into one ministry, with close coordination with the ministries responsible for complementary activities (roads, energy, and water), particularly at the planning stage. Such a reconfiguration o f responsibility at the national level and improved coordination with both local and complementary counterparts would facilitate adequate funding o fexciting elements o fthe ASDP agenda, such as the following: Sustaining and Sharing GrowthinTanzania 46 Share costs of adopting improved technologies, rather than subsidizing a single input. Tanzania is already using a public-private cost-sharing mechanism (through a matching grant) for adopting new technology under the Participatory Agricultural Development and Empowerment Project withinthe ASDP. The selected technology is one that is profitable in the longer run, and the matching grant makes fertilizer, seeds, seedlings, plant protection agents, implements, and any other needed inputs affordable during the period o f adoption (inmost cases, two years). Inputsare purchased from the private sector, and farmers save increased earnings while receiving the matching grant so that they can continue to purchase inputs on their own. This approach canbe scaled up through the DADPs. Continued reform of taxes. The government has already undertaken a number o f commendable reforms to improve the tax regime for agriculture, including setting limits on local taxes for traded agricultural commodities and reducing customs fees and value added tax. Enforcement o fthose measures should be pursued and their effects monitored, because they can improve profitability if well implemented. The value added tax on port charges andtransport costs addsto the cost offertilizer andcouldbereevaluated. Reduce cumbersomeimportation proceduresfor fertilizer. Such procedures include double inspection o f consignments, preshipment inspection by Cotecna at 1.2 percent o f the free- on-board price, and the Tanzania Bureau o f Standards (TBS) quality inspection at 0.25 percent o f the cost, insurance, and freight price. Any delay by the TBS means port charges accumulate, thereby increasing the cost o f fertilizers to farmers. In addition, importers are required to produce certificates o f quality from manufacturers. Abolishing inspection requirements andretaining the certificate o f quality from manufacturers, together with spot- checks by the TBS, would significantly reduce the cost o f fertilizers to farmers. Improve the road network. ASmuch as 40 percent o f the cost o f grains inthe major urban markets is attributed to transportation costs. Reducing this cost would increase farm profitability. Develop and disseminate more profitable technologies. This effort includes ensuring that agricultural research and extension focuses on developing more profitable technologies and varieties more responsive to the application o f inputs. It also means ensuring that farmers have access to information and recommendations specific to the technologies (rather than blanket prescriptions) and that profitability is part o f the calculus in deciding which technologies are recommended. Ensure consistency in thepolicy environment.The costs o f inputs decline as the volume o f transactions increases. The number o f private input dealers anddistributors is more likely to expand when the policy environment is consistent and the government stays away from direct intervention. Improve the opportunities for viable systems tofinance technology and inputs. Vertically integrated private systems o f extension, credit, and inputs linked to output markets (as for tobacco) are hctioning in some areas. They can be encouraged by reducing inefficiencies in the banking systems (reducing interest on borrowing), strengthening institutions for contract enforcement, improving taxation (as noted above), and improving the flow o f information. For example, small farmers in Msowero and Sonjo villages were assisted under the PASS (Private Agricultural Sector Support) p r o g r e s u p p o r t e d by DANIDA (the Danish International Development AgencyFin forming groups and accessing loans that were repayable over a three-year period. Initial reports on the programare positive. Reduce risks. Periodic weather shocks and external price shifts, together with household- level production risk, significantly affect technology adoption, profitability, and incomes. New methods to manage external shocks, such as weather insurance, forward contracts, and Sustaining and Sharing Growth inTanzania 47 options for price risk, are being piloted and should be evaluated carefully to determine the scope for scaling them up. Improve market access and product quality. Investment in, awareness of, and compliance with sanitary and phytosanitary standards for high-value products and stronger linkage o f groups o f smallholders with supermarket chains will improve marketability and profitability. The public sector can play a very constructive role here, even though the market transactions are between private parties. For example, through an agreement with Technoserve, the US.Agency for International Development (USAID) has helped coffee growers respond to October 2003 changes in the policy regime, whereby Tanzanian premium coffee producers and specialty roasters can export highquality coffee directly. This change inthe marketing regulations (the contribution o f the public sector) creates new opportunities for Tanzanian farmers to supply roasters with specialty coffee for premium prices. TechnoServe's conservative estimate is that farmers' incomes will be boosted by US$20 million over the next 10 years as a result o f the combined effect o f direct export o f premium coffee and a new value added tax reclamationpolicy. Concerted pursuit of those steps and others through adequate funding of activities under the ASDP will provide a reasonable likelihood that the growthtargets can be met. Five percent annual growth is ambitious but not impossible. Meeting this target, however, requires moving ahead decisively on the agendas o f policy reform and alignment o f public expenditure. Half-measures and compromises to accommodate special interests require sober reassessment o f the growth targets or admission from the outset that they will not be met. HARNESSING NATURAL RESOURCESFOR SUSTAINABLEGROWTH Tanzania is rich in natural resources, but their contribution to growth and poverty reduction could be significantly improvedthrough better governance and management. Gemstones, gold, fish, forests, and wildlife are among Tanzania's most important natural resources. Their exploitation and use provides a tremendous potential for the generation o f income and employment as well as revenue for government. Contribution of natural resources to growth. During the past decade, mining, fisheries, and tourism have been the most dynamic sectors in the economy. The fishing industry i s currently concentrated around Lake Victoria, which is already displaying signs o f depletion o f the fish stock. However, Tanzania's other lake resources and coastal areasprovide significant potential for expansion o f the fish industry. Most currently known mineral deposits are being already tapped which could slow down the growth in this sector in the near future. Tourism expanded rapidly during the 1990s; however, the growth o f the sector slowed down in recent years, particularly after the terrorist attacks inNairobi and Dar es Salaam in 1998. Forestry, wildlife and marine fisheries resources, though declining, are still relatively abundant, rendering largelyuntapped growthpotentials. Although these natural resources contribute to the economy and subsistence base o f the rural population, their value and potential is underestimated. This underestimation is partly based on missingmarkets in the case o f public goods, imperfect competition in the case o f distorting government interventions, as well as pricing o f natural resources below market value. Results o f all these market failures lead to sub-optimal economic decision making and loss o f income to the country. Due to weak governance and capacity in revenue generating sectors, resources are offered below market price to the benefitof a few powerful winners, and at the loss of the majority of the rural population. Yet, these natural resources provide substantive potential for income to communities in rural areas. The weaknesses in governance regimes in all three sectors, forestry, wildlife and fisheries, include primarily lack o f Sustaining and Sharing GrowthinTanzania 48 transparency and accountability in issuing rights o f resource extraction and revenues accrued thereof, lack o f equitable sharing o f benefits with communities, as well as weak monitoring and surveillance o f stocks. In all four principal sectors providing natural capital to the Tanzanian growth equation, that is, forestry, wildlife, fisheries andmining, royalties are set arbitrarily and do not reflect scarcity. Royalties are hence not used as a policy instrument o f inter-temporal resource pricing and sustained yield management. As long as these weaknesses are not addressed, a substantial base o f economic growth will be slowly eroded andpoverty reduction objectives are unlikely to be achieved. Accounted and unaccounted for growth contributions. In general, a great share o f the economic contributions o f forestry, wildlife and fisheries, does not enter the GDP and export statistics, and is hence not taken into account in the analysis o f growth. Availability and quality o f data is a problem that needs to be tackled urgently. Forestry provided over US$5 million in government revenue between 2003/04 and 2004/05. It contributes officially 2-3 percent to GDP and a 10-15 percent share o f export earnings. Taking unaccounted services and non-industrial forestry into account, estimates suggest that the value added in forestry may be as much as 10-15 percent o f GDP. Forests provide around 75 percent o f building materials and 100 percent o f indigenous medicinal plants and supplementary foodproducts. Inaddition, forests provide an important component o f value added to national income through their eco-system wide service functions, by providing for industrial anddomestic water and energy supply. Wildlife resources generate significant income through game viewing by international tourists. In 2001, Tanzania's national parks drew over 100,000 international visitors. This generated receipts o f almost 5 percent o f GDP, equivalent to about US$400 million. About US$30 million are generated from tourist hunting and an additional US$9 million generated by the private companies leasing hunting concessions from the government. In addition, wildlife provides substantial unaccounted for subsistence values. Well over two-thirds o f Tanzanians eat wild meat, with up to 95 percent o f the rural population claiming it as their most important meat protein source. Tanzania's fisheries sector has grown at a rate of 6 to 7 percent annually since 2000. In terms o f export earnings, fisheries contributed 10 percent o f total exports in 2003, which equaled US$130 million, the export value o f Nile Perch being US$lOO million. A great share o f the marine catch does not enter GDP and export statistics but plays an important role in livelihood support. The official number o f artisanal fishermen has doubled since 1995 and was close to 120,000 in2003. Although its contributionto GDP is still small at 3.5 percent, mining is the single most important earner of foreign exchange for Tanzania. About 50 percent o f export earnings accrue from minerals, predominantly from gold mining by large-scale foreign owned operators. Inaddition, mineral resources are o f importance to the artisanal miningsector. Untapped growth potential. With regard to game viewing tourism, untapped potential exists in the southern parks o f Tanzania. While the northern tourism circuit has supposedly reached maximum carrying capacity in terms o f numbers o f visitors, places like Ruaha and Katavi National Parks in the south are still fairly unknown. Shifting marketing and infrastructure development to these areas would provide new growth potential for Tanzania's growing tourism sector. Marine fisheries have recorded a sharp revenue increase due to increased license revenuesfrom foreign vessels in the ExclusiveEconomicZone (EEZ). There are estimates that the presently earned revenue does not reflect the total amount that the government could earn, and that real catch is much higher than what has been assumed as the basis to set the Sustaining and Sharing GrowthinTanzania 49 license fees. Notably, there is no catch based license or fee, and the vessels are allowed unlimited catch once they are in possession o f a valid license. Resource rent estimates fi-om marine fisheries captured by Private Fisheries Agreements (PFAs) show that license fees as a percentage o f revenue generate a gross resource rent o f approximately 2.2 percent. This is slightly less than half o f what might be expected in a western industrial fishery. Calculations for tuna seiners vary somewhat more depending on different catch scenarios. Although the current license fee arrangements o f PFAs inthe EEZ generate substantial revenue, the level is too low to capture a reasonable resource rent (>5 to 7 percent o f gross revenue) for the coastal state. The PFAs as currently offered to foreign fishing enterprises are hence considered untenable. With regardto freshwater fisheries, past growth rates are mainly based on Nile Perch exports from Lake Victoria. Other lakes, such as Lakes Tanganyika and Nyasa, as well as harvesting o f other species have so far been commercially underdeveloped. Diversification could also be sought in terms o f exploring additional export markets, as risks and vulnerability increase in a situation where export earnings in a sector are entirely dependent on a single market. This is the case o f fisheries exports from Lake Victoria, which are mainly destined for the EU. Commercial fisheries present an important emerging revenue source for the country and the sector. If the sector is well managed, commercial fisheries can have a positive impact on economic growth and poverty reduction at the same time. Principles o f management would need to include retention and re-investment o f revenue into the sector and putting up safeguards for artisanal fisheries to protect their rights and access to the resource. Potential for local spin-off effects. Local spin-off effects are presently missing in the context o f marine fisheries inthe EEZ. While new fisheries agreements are being negotiated with foreign countries, no fish is expected to be landed ashore and few supplies will be sourced from within Tanzania. If no such spin-off effects are created, the net impact o f commercial fisheries on poverty reduction may be negative, given increased competition with artisanal fisheries over the same resource. The fact that Tanzania is a net importer o f forest products is a sign o f lost opportunities for income generation for the local economy. Similarly, the mining sector seems to have had limited influence on poverty reduction in the local economy. Employment in the large-scale mining sector is limited, although younger employees may receive significant salaries. The majority o f those employed in the mining sector operate as self-employed, small-scale sector, artisanal miners. Their activities generate very low returns, especially when one considers the hardship associated with this kind o f employment. Inaddition, an increasing income disparity seems to be emerging between those employed in the small-scale and large-scale mining sector. To the extent that those recruited by the large-scale mining sector are recruited outside the local community, the local community is thus restricted to opt for poorly paid employment opportunities in the small- scale sector. Large-scale mining may have positive effects for local communities through the improvement of basicinfrastructure.There is, however, no indication that the expansion o f the mining sector triggers significant growth in the local economy, since mining operations generally are detached fi-om local supply chains and therefore create employment opportunities primarily inthe services sector. Potential for poverty reduction. In addition to their potential for government revenue generation, wildlife, fisheries and forestry resources provide the non-agricultural subsistence base for rural communities in remote locations. Increased emphasis on natural resources related enterprises has the potential to create additional income opportunities for the rural population. This study highlights several examples o f successful participation by villages in wildlife tourism, hunting, and forestry. However, despite the conducive policy framework in both the wildlife and forestry sectors, weak governance systems at both central and local Sustaining and Sharing GrowthinTanzania 50 levels have so far limited the realization o f the poverty reduction potential through community based natural resources management. The main focus within community wildlife management has been on institutions and distribution o f benefits rather than enterprise opportunities at household level. Sustainability of growth. Inthe context o f sustainability o f natural resource based growth, the following are the major constraints that can lead to revenue loss and possible deceleration o f growth inthe long term: 0 Under-pricing o f resources not allowing capturing o f resource rents; 0 Weak environmental governance systems; and 0 Limited knowledge o f stocks, their values and changes over time. Sustainable growth based on renewable resources requires that the cost of extracting a resource and the notionalcost of replacing a unit of the resource, commonly known as 'resource rent', is evaluated so that the wealth base is not eroded. While royalties are the most important source o f government revenue in the forestry (83 percent), wildlife (96 percent hunting licenses) and fisheries (84 percent royalties, 15 percent export licenses) sectors, they are set arbitrarily and capture neither market values nor resource rents. Similarly, in the mining sector, licenses to foreign investors do not take the `capital component' into account. Tax incentives have been generous towards foreign investors to attract capital investment and to open up the market at the expense o f sustainability principles. In such a scenario, acceleration o f growth comes at the expense o f pricing resources below market value, which leads to loss o f income, erosion o f critical stocks and an associated deceleration o f growth inthe long-term. The under-pricing of natural resources is compounded by weak environmental governance, including inefficiencies in revenue collection, limited transparency and accountability, as well as outright corruption. This leads to a loss o f income and livelihoods for rural communities. Finally, limited knowledge o f resource stock values and stock changes implies that limits o f extraction and quota associated with licenses can only be set arbitrarily and cannot be based on sound ecological calculations and realistic projections. Externalities. Natural resource extraction has potentially important externalities such as the impact o f commercial fishing and large scale miningon atisanal fishing and mining. Unless such externalities are properly addressed, natural resource extraction may have negative economic, social, and environmental impacts that may interfere with the poverty reduction objectives o f the NSGRP. Summary. This review has generally argued that the natural resource sector in Tanzania is dynamic and a potentially important contributor to overall economic growth and poverty alleviation. It does, however, remain fragile and the use o f well-placed interventions can serve to secure its role as an engine o f growth. In particular, much o f sectoral output remains hidden, either because it is not recorded in national statistics, or because potential economic values are either destroyed or squandered as a result o f various weaknesses fall into the categories o f poor governance; inefficient management; and, vulnerability. There is substantial difference among the sectors considered and efforts are underway by government inall cases to correct some existingweaknesses. Poor governance is reflected in persistent corruption and lack of transparency in how resources are administered by government. It can arise from the lack o f a responsible institution, or fiom poor record keeping and blatant illegal activities. Governance issues are least serious in freshwater fisheries, tourism, and large-scale mining, and most serious in marine fisheries, forestry, wildlife management, and artisanal mining. Trade in carbon Sustaining and Sharing GrowthinTanzania 51 resources suffers from the newness o f the activity and the lack o f government capacity inthe area, but there i s no evidence ofpoor governance through corruption or illegal activity. Inefficient management relates to situations where market or policy failures have reduced the value of the resource base or reduced the rent that can be captured effectively by existing mechanisms. The most common inefficiencies arise from unclear property rights, incorrect resource pricing, tax system distortions, and inequitable benefit sharing (which creates conflicts and reduces management effectiveness). Losses associated with inefficient management are currently most significant with marine fisheries and wildlife resources because o f resource pricing and property rights issues; similar inefficiencies persist in artisanal mining but these are compounded by social and environmental externalities relatingto humanhealth and environmental impacts. Vulnerability is a system-wide effect that speaks to the ability of the resourcebase, and of those dependent on the resource, to withstand unforeseen shocks. These shocks are becoming more frequent, and are related both to internal factors (such as general poverty or dependence on single resources) as well as external factors (such as global security, climate change, and international commodity prices). All sectors are vulnerable, in various ways, to such shocks. Recommendations and Priority Actions. Addressing these weaknesses requires an integrated approach that addresses the different classeso f weakness. The basic foundation (or first pillar) is associated with reforms that promote good governance. The elimination o f corruption and improved transparency is the foundation for any type o f subsequent improvement in resource management. An ideal rent collection system or an optimal fiscal regime is o f no use if the overall system is corrupt or lacks basic administrative capacity to monitor and assess activities. While Tanzania has made progress in overall governance, some additional interventions are appropriate. Reforms that could be considered relate to the establishment o f responsible authorities, training and human resource development, andoverall transparency and accountability. The second pillar involves management interventions that improve efficiency of resource development and exploitation. These presuppose the existence o f a regime and administrative structure that reflects good governance. Interventions in this area generally include the creation or enforcement o f property rights, correct resource pricing, elimination o f distorting taxes and subsidies, and improvement inbenefit sharing. The third pillar relates to the protection of the resource base through safety nets and is intended to reduce vulnerability of the resource base and the populations that depend on the resource. The instruments that could be considered include revenue retention schemes, adaptive co-management models, education and awareness, vulnerability monitoring systems, and diversification. All o f these tactics, however, must be coordinated by an over-arching strategy o f vulnerability management that - at this stage - does not yet exist. A key recommendation is thus to define and institutionalize such a strategy within a central coordinating mechanism inthe President's Office or Vice President's Office. A summary o f the recommendations is provided in Table 15 which shows the key actions within each o f these three pillars. The table is presented as a simplified log-frame or results matrix, including some potential indicative indicators for tracking results andimpacts. Sustaining and Sharing Growth inTanzania 52 Table 15. Summary of Objectives,PriorityRecommendedActions and Indicative ResultsIndicators for Enhancingthe Contribution of the NaturalResourcesSector to EconomicGrowth Reform: Management: Protection: Promoting Good Governance ImprovingEfficiency Providing Safety Nets Objective: Objective: Objective: Eliminatecorruptionand improve Eliminatemarket andpolicy Reducevulnerabilityofnatural transparencyof governanceinthe distortions,improverent capture, and resources to shocks, andenhancethe sector. achieve a more equitable distribution resilience ofpopulationsdependent o f benefits. on these resources. Priority Reform Actionsfor Priority Management Actionsfor Priority Coordinating Specific Sectors: Specific Sectors: Mechanism: Operationalize Deep Sea Strengthencommunityproperty Establisha high level FishingAuthority. rights inwildlife and forestry coordinatingunit to address Create and Operationalize throughimplementingmore vulnerability issues at a ExecutiveAgency for Forestry. communitymanagedareas. nationallevel, reportingto the Reviewinstitutional Establishmentof acommunity Presidentor Vice President. arrangements affectingWildlife territorial sea along coast. Development of vulnerability Division andTANAPA, with a Operationalize cadastralsystem profilesbasedon social and view to establishing anew for artisanalmining. naturalresourcemapping executiveagency inthe wildlife Establishalegalframework for (building onpovertymapping). sector. ownershipof carbonresources. Priority Protection Actions: Reviewcapacity andtraining Reformandharmonizelicensing Increaseuse ofrevenue needswithin MEMto improve system for offshorefisheries, retentionschemeswith modalitiesofartisanal mining with net increases infee preferencefor those involving governance. structures. Consider uncorrelatedrisks andpooling Trainingand humanresource implementationofbidding o f revenues (suchas the Marine development throughuse of system. LegacyFund). anti-corruptiontraining Implement an open tendering Increase reliance on adaptive programs for leadagencies inall scheme for wildlife licensing co-managementarrangements sectors (fisheries, forestry, with transparentmechanismsfor with localcommunities,private wildlife, tourism, mining, an appropriatelevel ofdomestic sector, andgovernment. carbon). preference. Development ofnorms and Traininginanti-corruption Reviewoftaxationinstruments standards, and support for methodsat District level and to encourage effectivevalue- educationandawareness sensitization for implementing addedprocessinginmarine aroundvulnerable resources. partners inNGOs andprivate fisheries andwildlife products. Implementationof vulnerability sector, particularlyinfisheries, Adopt uniformtaxation monitoringsystems, including forestry, andwildlife. provisionsfor mining sector in earlywarning systems for Enhancements to accounting, lieu of individually negotiated knownthreats. monitoringandreporting agreements. Implementdiversificationand systemsinall sectors. Increasedirect involvementof protectionmechanismsfor Acceleratedimplementationof privatesector inmanagement of vulnerable groups, focusingon MCS databasefor offshore fisheries, wildlife, tourismand protectionof critical fisheries. mining. ecosystemsand guarantied Acceleratedestablishment of a access to traditionaluses of GPS/GISmineralcadastral resources. system and database for the mining sector. Key Reform Indicators: Key Management Indicators: Key Protection Indicators: Operational agencieswith Increasedproportionof resource An emergencyresponse transparentmonitoring and subject to managedaccess (as networkandplansinplace reportingsystems inplace. opposedto openor unmanaged nationally. Reducedleakages from access). Povertyandvulnerability corruptionand illegal activities. Increasedrent generationand mappingbeingusedat District Improvedawareness regarding capture. level. good governanceamong Increasedparticipationby local D Monitoring systems inplace government staffandother communitiesinresource and deliveringtimely reports. stakeholders. management,with increased Demonstratedpositiveoutcomes in share ofbenefits. aftermathofunanticipatedshocks. Reducedresourceuse conflicts. Sustainingand SharingGrowthinTanzania 53 SUPPORTINGTHEINFORMAL SECTOR AND MICRO, SMALL, AND MEDIUM SIZE ENTERPRISES The informaleconomy plays an important role inincomegeneration, employmentandskills development and often serves as an entry point for widespread participationin the private sector. International estimates (Schneider 2004) -which include informal activities in agriculture - suggest that Tanzania's informal economy accounts for about 60 percent o f the Tanzanian gross national income (GNI). The informal sector i s thus relatively large, both in regional and international comparison (Figure 17). Data from the Tanzanian Integrated Labour Force Survey (ILFS) 2000/01 (URT 2001) -which exclude informal sector activities in rural agriculture- suggest that the informal sector employs about 16 percent o f the total labor force. The Institute for Liberty and Democracy (ED 2005) found that about 98 percent o f economic activities in Tanzania were within extralegal boundaries inthe informal economy, and 89 percent o f all proprieties are held extra-legally. According to the ILD, the Tanzanian informal economy has assets worth US$29 billion. Figure17. Size of the informaleconomy for selected countries(as YOof GNI) 80 70 3 60 50 -E 40 30 20 10 n I Source: Schneider 2004. Informal economy operations can be found in most sectors in Tanzania. According to the ILFS 2000/01, the industries that employ most people whose informal economy activity i s their main activity are retail trade o f agricultural products, meat and chicken (20.7 percent); stationary, photography and general retail (18.8 percent); retail trade o f processed food (10.5 percent); and restaurants and hotels (12.4 percent)." The Tanzanian informal economy thrives because it provides opportunities of income generation to the poor and unemployed, and because it offers a low cost ground for experimentation with business ideas that can lead to growth and formal enterprises. According to the ILFS 2000/01, one o f three households was active in the informal economy in 2000/2001, as opposed to one o f four households in 1990/1991. The survey shows that the number o f households with informal economy activities grew during the 90s from 42 percent o f Sustaining and Sharing GrowthinTanzania 54 all households in urban areas to 61 percent. Inrural areas, 27 percent out o f all households had informal economy activities in 2000/01 as compared to 21 percent in 1990/91. Also, interms o f employment, the informal economy holds a bigger share in urban areas as compared to rural areas. InDar es Salaam, for instance, 36 percent o f the labor force i s employed in the informal economy. Generally, for households engaged ininformal economic activities in urban areas, this tends to be their main activity, while it i s more likely to be the secondary activity inrural areas. Crop growing i s the sector employing the majority (94.4 percent) o f the people whose informal activity i s a secondary activity. The informalsector-especiallyurbanself employment-has beenanimportant pathout of poverty for many Tanzanians. Not only did the share o f self-employed persons in the adult population increase from 4.8 percent to 8 percent,12 but expenditures by households headed by a self-employed individuals grew by 18 percent between 1991/92 and 2000/01, compared with a much more modest growth o f expenditure o f agricultural households, which grew by only 7 percent during that period. In Dar es Salaam, the growth o f expenditures by households headed by a self-employed person was even more dramatic at 65 percent, which was evenhigher than the average growth in expenditures o f 60 percent experienced by households headed by a person in paid employment. Many small enterprises in Tanzania operate under a semi-formal legal status without the necessity of registration with state authorities. Semi-formal operators appear on a list of operators at the local authority. They pay taxes that are collected by local authorities. H o w long informal operators may remain at any level o f informal or formal status differs widely in Tanzania. An operator may progress via a semi-formal stage, or move directly from informal status to the official registration (Table 15). Table 16. Typologyof forms of enterpriseinTanzania Formalstatus Legalform Descriptionandsubcategories Illicit None An enterprise is illicit when there is no legally permitted licensed or registered counterpart. Informal None An activity (for which there is a formal counterpart) that does not comply with requirements o f the regulatory system regarding licenses, permits, certificates, notification or registration o f the activity i s licit but informal. Semi-formal Local An activity carried on by an operator who appears on a local authority authority list o f licensed operators o f enterprises but i s not licensed registered with the state registrar i s semi-formal. Existing enterprise forms: (i)Hawkers license, (ii)Business license, and (iii) Market stalls. Formal State Person(s) registered to conduct business activities under a registered registered business name; for example, enumerated on a state enterprise register as a sole ownership enterprise, a private limited company, or a joint stock company open to public subscription shareholdings. Source: Nelson and de Burijn2005. Informal operators state a number of constraints to the growth of their businesses. Most importantly, they are often not in a position to afford a permanent premise for their business. However, there are a number o f success stories o f informal operators formalizing their business after the municipal authorities provided or assisted inthe access to permanent premises. Second, SustainingandSharingGrowthinTanzania 55 they lack access to credit. A lack o f business management skills as well as very limited access to new technology i s detrimental to the growth o f informal businesses. There i s also concern about reported harassment o f informal entrepreneurs by local authorities - some businesses are demolished, property i s taken away and inthe worst cases, entrepreneurs face charges. Although the situation i s reported as having improved considerably since 2003, continued attention to removing obstacles to economic activities-both in the formal and informal sectors - and facilitating the transition from informal to formal status, i s important. The cost of starting and operating a formal business are high inTanzania. According to the World Bank's Doing Business Survey (World Bank, 2006a), entrepreneurs can expect to go through 13 steps to launch a business over 35 days on average, at a cost equal to 161.3 percent o f the $330 per capita income, compared to 10.8 percent, 48.2 percent and 117.8 percent o f per capita income in Botswana, Kenya and Uganda respectively or 6.5 percent in OECD countries. However, the minimum capital required to obtain a business registration number in Tanzania (6 percent o f income per capita) is considerably lower than the regional average (297.2 percent) or the OECD average (28.9 percent). The ILD found costs and burdens o f the procedure to legally incorporate a private Bureau o f Change in Mbeya include 10 stages, 103 steps, 379 days and US$5,506, the respective figures for Dar es Salaam are 95 steps, 283 days and US$3,816. The ILO/UNIDO/UNDP Roadmap Study calculated that the cost o f coping with regulatory and non- regulatory constraints would amount to as much as 75 percent o f monthly sales o f informal operators in the case o f the firewood and charcoal sector, which explains the decision to remain inf0rma1.l~ In order to register a business, entrepreneurs have to travel to Dar es Salaam, since the Business Registration and Licensing Agency (BRELA) is only located there. The ILO/UNIDO/UNDP study shows that business licensing is equally cumbersome, because it requires various procedures in different offices at the regional and district level. Both processes, licensing and registration, are required and often involve bribing officials further increasing the cost o f starting a formal business. Costs o f formalization are often further increasedbecause some officers are reported to be unhelpful, obstructive and ~ncaring.'~The Business Activities Registration Act, which was passed by Parliament in early 2007, contains important provisions for the decentralization o f BRELA and the streamlining o f registration o f registration and licensing requirements and procedures. Implementation o f this new act i s thus an important step towards facilitating the establishment o f new businesses and the transition o f businesses from the informal to the formal sector. Registering property is an expensive and time-consuming undertakingin Tanzania. It takes 12 steps and 61 days to register property compared to 4 procedures and 33 days in OECD counties. The cost o fregistering property amounts to 12.2 percent o f the property value, which is considerably higher than in OECD countries (4.7 percent). The ILD found that the procedure to allocate land for urban purposes and to obtain a building permit takes 13 stages including 68 steps, which take 8 years and cost US$2,252. D e Soto (2001) explains that a major problem the informal economy faces i s its inability to convert what he refers to as "dead capital" in form o f untitled assets into capital that can be usedas for example collateral for loans. Also, enforcing formal contracts costs on average 35.3 percent of the debt, compared to the OECD average of 10.9 percent. According to the ILD, resolving claims for a debt at the Commercial Court Division take 9 stages, 96 steps and 390 days and cost US$11,964. Likewise, it takes 1,286 days and costs US$1,022 to collect a debt byexecuting a court decree. Inaddition it i s comparatively difficult to hire or fire employees in Tanzania. However, the Sustaining and Sharing GrowthinTanzania 56 ILO/UNIDO/UNDP Roadmap Study o f the informal economy in Mainland Tanzania did not identify employment regulations as a severe constraint, since they are not enforced. Tax rates,which are perceivedas beingtoo high, often act as a disincentiveto formalize. A recent study (FIAS 2006) shows a number o f anomalies with the tax regime for small businesses, inparticularthat the system is regressive for non-record keepers. Moreover, the study states that the "tax environment encourages expansion o f the informal economy," as "small businesses face a proportionatelyhigher time and financial costs to comply with administrative requirements and therefore may not see any benefit o f joining the tax net." Furthermore, newly registered businesses have to pay taxes up-front, which further increase the required startingcapital. The marginal effective tax rate is higher for small businesses than for large firms. Local governments still levy a large number o f taxes, fees, licenses and charges. Local taxation i s seen as a major constraint to formalization. In addition, bribes are estimated to add up to more than 5 percent o f total sales. Comparative investment climate survey data sets collected by the World Bank suggest, however, that the share o f revenue that Tanzanian businesses deliberately fail to report for tax purposes i s more than 30 percent, compared with around or less than 20 percent in Kenya, Uganda and Zambia. It should, however, be mentioned that informal entrepreneurs do pay taxes to local authorities. For example, in Dar es Salaam the local tax i s TSh. 100per day, which i s collected from every informal operator. Benefits of Increasing Formalization. Formalization creates the basis for formal transactions with other entities, including financial intermediaries for access to credit, the public sector for the provision o f public services, and clients and suppliers for contract based transactions. Given the rigidities, costs and attitudes o f formal sector regulators and service providers, the access o f informal entrepreneurs to important public goods such as utilities like electricity or water, and other inputs and services i s limited. From the perspective o f business operators, formalization o f their businesses increases the trustworthiness for customers-a benefit that has been mentioned by several entrepreneurs who formalizedtheir business. Furthermore, operatorsinthe informaleconomy cannot convert "dead capital" inthe form of untitledassets into "productive economic currency" such as collateral for loans to start or expand a business. Moreover, their access to the capital market i s very limited. Formal registration i s generally a prerequisite for access to credit and small business loans (US$15,000- US$30,000). Therefore, informal sector operators are deemed to remain small without being able to exploit potential economies o f scale. As a result, productivity tends to be low. Ingeneral, the labor productivity i s lower in Tanzania than in many other African countries-the median value added per worker i s US$2,061. In the case o f micro enterprises (of which most are informal), labor productivity i s even lower (Figure 18). Sustaining and Sharing; GrowthinTanzania 57 Figure 18. Median value added per worker MedianValue Addedper Horkerio US Qllars Mediand u e addedpr Horkerbysize (in Us Qllars) 4,397 $4,500 $4,011 $3,000 $1,500 $0 Tanzania Uganda Zamia Kenya China 100) ource: World Bank 2004b The most obvious benefit to the government of increasing formalization i s higher tax revenue. Furthermore, the government's ability to implement policies and the effectiveness o f government programs aimed at the private sector will rise since informal enterprises operate outside of the government system o f regulation. Box 6. Examples of voluntary formalization Alan Mungo experimented with informal wine production, insurance agency, clothes shop, and eventually progressed to wine production and safari tour companies. Having developed a marketable product and identified market opportunities for expansion, he decided that it would be in his interest to legitimize his enterprise. However, after preliminary enquires he found that the premises he used, being a motorcar garage on the grounds o f his dwelling house, would not be approved by health inspectors, which meant he would not be able to apply for an operator's license. At that time he attended a meeting between representatives o f the Association of Food Processors (TAFOPA), the local trade licensing officer and a representative o f the Ministry for Agriculture and Food Security. There it was agreed that small-scale food processors who could not comply with regulations would not be forced to cease their operations, instead they could continue to operate without licenses, but still be subject to supervision by health inspectors. This incidentally implied the advantage of being o f f the database of operators liable to pay formal taxes. The acknowledged reason for the Ministry's decision was that Tanzania's food processing industry was important, relied mainly on small-scale producers, most whom were not licensed and would be forced to close ifregulations were enforced. Mungo was able to continue operating and licensing costs were deferred untilafter his premisesreachedthe standard about one year later. At this point, his formalization costs were very low. Dan Himba formalized his business after an activity period of 29 years. H e conducted informal enterprise activities as a supplement to his salaried employment. Duringthis period he enjoyed a long runo f starting informal enterprises, maximizing profits while he could and abandoning them when they showed signs o f having runtheir most profitable life cycle. When the conditions o f his employment deteriorated, a fill-time entrepreneurial career became more attractive to him and increased the stimulus to formalize. He subsequently abandoned hisjob for a career inprivate enterprise. Source: Nelson and de Burijn2005. In sum, the informal economy is large in Tanzania since for micro enterprises the benefits of formality are dwarfed by its costs. "Whether in monetary terms (direct cost or income foregone), or interms of time and energy, the cost o f compliance turns out to be too highfor most startingbusinesses, who are therefore obliged to start inf~rmally."'~Government is thus aiming to establish an institutional environment that i s conducive to investment in the formalization o f informal enterprises, since according to ILO/UNIDO/LTNDP(2002) the "current regulatory set up a) fails to meet objective of ensuring quality control for the majority; and b) traps the Sustainingand Sharing Growth inTanzania 58 entrepreneurs in low quality settings, puts their upgrading and growth too far out o f reach and limitsthe contribution ofthe subsector to povertyreductionandnational growth." In order to allow for a low-cost arena for business experimentation, as well as a means of income generation for poor and unemployedpeople in the absence of state-provided safety net such as unemployment benefits, small scale informal economic activities should be tolerated and supported. Forced formalization risks could damage fragile enterprises and livelihoods for very little benefits, and suppressing business experimentation and development. The decision o f a small scale informal operator to formalize should be a voluntary one. In addition, it would be important to provide for an adequate institutional framework that i s conducive to and provides incentivesfor voluntary formalization of informalbusinesses. Possible incentives are the abolishment o f up-front payment o f taxes or free training on procedures o f formalization. Moreover, local governments could provide more permanent premises at low rents (such as for example the market stalls in Ilala, Dar es Salaam) for informal entrepreneurs. The Small and Medium Enterprise Development Policy o f 2002 addresses the problem o f infrastructural requirements and calls upon local authorities to allocate and develop land for SMEs, to develop industrial clusters and trade centers, and to identify and allocate underutilized public buildings to SMEs. A higher degree o f realization o f this policy would be very beneficial. The focus of government has been primarily on encouraging formalization of businesses, while less attention has been paid to supporting small scale, informal sector activities. In addition to the review o f laws and regulations that may unduly impinge on small scale informal activities, this requires also a change in mind set o f public officials recognizing the value and importance o f informal sector activities for Tanzania's economic development. Inparticular local authorities, who deal with the informal sector on a daily basis, would benefit from capacity buildinginthis area. Access to credit posessevere constraintsto informalentrepreneurs. Often, even microfinance schemes are hard to comply with for informal businesses. Alternative credit schemes are for example rotating savings and credit societies, such as SACCOS, savings and credit cooperatives. Developing and improving occupational SACCOS can be an effective alternative to formal banks. In the long run SACCOS could function as a way to connect informal savings with the formal financial sector. Government has already started to implement measures that aim at reducing the cost of doing business and facilitating the formalization of businesses and property. In October 2004, the Property and Business Formalisation Programme was launched. The objective o f the program i s to provide legal titles for assets and guarantee property rights. Concluding the Diagnosis Phase o f the program, the ILD presented a comprehensive report in October 2005 commenting on legislation governing property ownership and compiling data on the patterns o f movable and fixed asset ownership. Formalization reforms will be designed on the basis o f the findings.16 Government has also taken important measuresto reducethe burden of businesslicensing and registration and a business activities registration bill was passed by the Tanzanian Parliament in 2007. The goal is to create a business licensing system which is transparent, and efficient, aiming at registration rather than revenue collection and control in place. Furthermore, the license fee for small businesses was abolished in 2004/05. Further to simplifying and harmonizing legislation and streamlining regulations o f business and property registration, Sustainingand SharingGrowthinTanzania 59 reducing the costs o f formalization i s important to facilitate the transition from informal to formal status. For example, the opening o f branches throughout the country by BRELA, would reduce travel expenses o f entrepreneurs who wish to license a business. Officers of local and nationalgovernment agenciesneedto bewell informed and trained to assist and help potential licensed business owners, They should pass on correct and useful information in order to facilitate business licensing. Inparticular, officers should be aware o f legislativechanges such as the new taxation schedule and comply with it. In addition to government officers, potential and operating informal entrepreneurs should have a fair understandingof the formalization processas well as benefits and opportunities of running a formal business. Informal workers should be encouraged to form cooperatives. Training for informal business operators in managerial skills through seminars and workshops i s o f utmost importance and seen as very beneficial to fostering private sector development. Furthermore, there is a needof advocacyfor the informaleconomy. Currently, there are three associations representing the interests o f informal economy operators: VIBINDO, an umbrella organization; TASISO (Tanzania Small Industrialists Society); and TAFOPA (Tanzania Food Processors Association). However, sectoral associations are still weak. The ILO SYNDICOOP project facilitated the formation o f a national steering committee that includes the Trade Union Congress o f Tanzania, Tanzania Federation o f Cooperatives, Savings and Credit Cooperative union, the government and individual informal economy groups. STRENGTHENINGTHE CAPACITY OF THE POORTO PARTICIPATEAND CONTRIBUTE TO ECONOMIC GROWTH To achieveits poverty reductionobjectives, Tanzania needsto devoteparticularattention to providing the poor with the capacity to participate and contribute to the growth process. Interventions that would enhance the capacity o f the poor to participate in economic growth opportunities include the upgrading o f their human resources (health, nutrition, and education), facilitating the accumulation o fproductive assets, and mitigating the effects o f risks. To benefit fromgrowth,the human capitalbaseof the poor must be strengthened. There is a strong relation between the level o f education and poverty showing the importance o f investing in well educated individuals. The productivity of labor is also constrained by malnutrition and disease. The household budget survey revealed that more than 25 percent o f the population was unable to work for at least one week inthe four weeks prior to the survey because o f illness. Except for education, access to services and human welfare outcomes has not improved in the past years. Major progress has been made in access to primary education, though concerns about quality o f education remain. Greater access to secondary education remains important. Poor households also did not benefit from improvements in access to clean water (only non-poor household did). Healthremains a key problem, as most health outcomes did not improve over the 1990s. An exception may hold for infant and child mortality for which census data -but not DHS data, suggests a decline. Twenty nine percent of those aged 15 and above are illiterate(HBS 2001) and accordingto the 2002 census the average years of educationofthe workingpopulation(those aged20-64) i s 5.1 years. Not only i s this one o f the factors contributing to Tanzania being one o f the poorest economies inthe world, but within Tanzania differences ineducation are strongly associated with income levels. Inrural areas, where poverty i s highest, the average level o f education o f heads o f Sustaining and SharingGrowthinTanzania 60 household i s 4.3 years as opposed to 7.8 years in Dar es Salaam where poverty incidence i s lowest (HBS 2001). Consumption regressions show that individuals living in Dar es Salaam in households headed by someone who completed secondary education have a per capita income that i s 49 to 68 percent higher than that of households headed by someone with no education. In rural areas the difference i s even larger: 70 to 84 percent (Table 17). Table 17. Increasein per capita consumptionrelative to householdsheadedby individuals with no- education Headofhousehold has,. . Dar es Salaam Other urban areas Rural Some primary education 25%-36% 19%-23% 17%-22% Completedprimary education 57%-80% 30%-38% 42%-48% Some secondary education 36%-52% 42%-57% 48%-68% Completedsecondary education 49%-68% 52%-67% 70%-84% Post secondary education 90%-134% 73%-93% 90%-112% Adult education only 43%-58% -1%-2% -2%-3% Note: Increases in consumption are determined from consumption regressions (Tables 62-68). The first numberinthe range is from a regressionthat excludes consumptioninformation, the second number from a regression that includes this information Source:Staff calculations based on Fox and Simonsen 2005 Education is associated not only with income but also with non-income dimensions of poverty. Low levels o f education lead to higher total fertility, lower levels of child nutrition (REPOA 2005), higher child mortality (Rafalimanana & Westoff 2001) and an intergenerational transfer o f poverty as children from poor households are less likely to attend school themselves. The total fertility rate o f women aged 40-49 i s 6.5 if they do not have any education but drops to 4.9 if they completed at least primary education (TRCHS 1999). Evidence from Kagera shows that children with educated parents have better nutritional outcomes (Alderman et al. 2005). And according to the HBS, 52 percent o f Tanzanian children aged 7-10 years attended school, but only 44 percent from the first two quintiles did. The Primary Education Development Program has not only resulted in a significant increase in enrollment, but evidence from Kilimajaro and Ruvuma suggests that inequalitiesin access to primary educationthat existedpreviously have disappeared. Figure 19 presents, for rural Kilimanjaro, concentration curves for 2001 and 2003. The concentration curve for 2001 shows how pre-PEDP access to primary education was unequally distributed: children from wealthier households attended school relatively more often than children from poorer households. The distribution was as unequal as the distribution o f consumption, here represented by the Lorenz curve. With PEDP inequalities in access to education disappeared and the concentration curve coincides with the 45-degree line. Though not shown in the Figure, results for Ruvuma are comparable. Sustaining and Sharing GrowthinTanzania 61 Figure 19. Changesin the distribution of access to education, rural Kilimanjaro CancentrationCUM Access to Pmary EdKalion 2M3 Cum distnbution of population Source: Staff calculations basedon HBS2000l01and Sarris, Karfakis and Christiaensen, 2004. Despite major progress in enrollment and addressing aspects of the gender gap, there remainregions, particularlythe poorer and more isolated ones, where the implementation of PEDP lags behind. This i s evident from net enrollment figures. It i s 90 percent in 2004 for the country as a whole but in Tabora it was only 68 percent and inKigoma 77.2 percent whereas inDar es Salaam it was 93.1 percent. Interestingly, some poor regions do particularly well, e.g. Ruvuma has a net enrollment o f 99.3 percent so that poverty i s not the only explanatory factor for the divergence inperformance. In secondaryeducationthere havebeenimprovementsin recentyears, but the poor remain largely excluded. The number of primary school leavers entering secondary school increased from a low o f 3.4 percent in the mid eighties to 22 percent at the start o f the millennium. Inthe past years this has increased further and in 2006, 40 percent o f those that finish primary school were able to obtain a place in a government secondary school. The probability that a child aged between 14 and 18 attends secondary school i s 2 percent if the child i s from a household in the first consumption quintile and 13 percent ifthe child comes from a household in the top quintile. The prevalence of illness in Tanzania remains high. According to the 1999 TRCHS, 35 percent o f children under 5 reported being affected by fever in the past 2 weeks, 12 percent were affected by diarrhea and 14 percent experienced acute respiratory infections. The HBS 2000/01 reports that 27 percent o f its respondents (adults and children) indicated to have experienced illness in the preceding four weeks. Malaria/fever and diarrhea are, again, the most common causes o f illness. Such high levels of illness have economic consequences. Almost one in four missed at least one week of school or work a consequence of illness that i s evenly distributed across consumption quintiles (Table 18). Sustaining and Sharing GrowthinTanzania 62 Table 18. Number of days misseddue to illness,by quintile Consumption Quintile Days missed Lowest Second Middle Fourth Highest Average None 0.30 0.3 1 0.29 0.30 0.32 0.30 Week or less 0.45 0.46 0.46 0.46 0.43 0.45 1-2 weeks 0.13 0.11 0.11 0.12 0.13 0.12 More than two weeks 0.13 0.11 0.11 0.12 0.13 0.12 Source: Staff calculations based on HBS 2001. In recent years, progress has been made in improving the health of Tanzanians. Life expectancy increased from 44 years in 1978 to 49 years in 1988 to 54 years for males and 56 years for females in 2002 (Census 2002). Infant mortality dropped in all regions and nationally fi-om 115 per 1000 in 1988 to 95 per 1000 in 2002." Inaddition, child malnutrition, which remained unchanged over the course o f the 1990s, declined rapidly between 1999 and 2004. Other health indicators saw less progress or even a deterioration. Maternal mortality which was 529 per 100,000 births in 1996 didnot decline, and may even have increased though the increase to 578 in 2004 i s not statistically significant. The share o f blood donors infected by HIV/AIDS increased from 7 percent in 1994 to 12 percent in 2003 for women and from 5 percent to 8 percent for men." Though there are few differences betweenpoor and non-poor householdsin their exposure to major diseaseslike malaria, diarrhea or respiratory infectionsor the share of budgetary expenses on health services, health outcomes differ considerably by wealth status (Table 18). Infant and child mortality is 15-20 percent higher amongst the poor than amongst those in the top quintile. The difference for nutritional indicators i s even larger: 30-50 percent. The one indicator in which poor households do substantially better than non-poor households i s in HIV prevalence, which i s 3.4 percent amongst those from the poorest quintile and 10.5 percent amongst individuals from the wealthiest quintile. Part o f these differential health outcomes may be explained by differences in coverage o f preventive health services. Poor children, for instance, are less reached by vaccination services and women from poor families and are almost three times less likely to have their birthattended by trained medical personnel than non-poor women. Table 19. Differencesinhealth outcomesby quintile Quintile Indicator Lowest Second Middle Fourth Highest Average Infant mortality rate, death per 1000 114.8 107.5 115.4 106.8 91.9 107.8 Under-five mortality rate, death per 1000 160.0 159.3 192.7 155.0 135.2 161.1 % of stunted children under 5 49.5 52.5 45.0 36.6 23.4 42.7 % ofunderweight children under 5 32.2 35.1 28.8 23.9 21.7 28.8 HIVprevalence 3.4 4.5 5.6 9.4 10.5 7.0 Source:Gwatkin et al. 2004 from 1999 TDHS data ;HIV prevalence, THIS 2004. Differences in behavior or economic circumstance also explain difference in health outcomes between poor and non-poor households. For instance, poor individuals have less access to clean water and are less well educated. Children from the poorest quintile are five times less likely to sleep under a bed net than children from the top quintile. Individuals living in poor households are less likely to consume iodized salt, and women in the poorest quintile are almost three times more likely to have experienced female genital cutting than those from the wealthiest Sustaining and SharingGrowthinTanzania 63 quintile. Inacknowledging that non-health factors play an important role in determining health outcomes, the role o f the health sector in improving the health o f the poor (and the population at large) is put in perspective. Improving the health outcomes o f Tanzanians requires a broadly shared effort across different sectors, including health, ed~cation'~and water. There are several challenges that need to be addressed in the health sector. The first i s the funding situation. Though funding has improved -the total per capita allocation o f public expenditure to health increased from TShs 5,100 in 2001 to Tshs 7,374 in 2004 (MoH, 2005)*' it remains extremely low. And, with rising costs o f drugs the observed increase inthe budget (of 45 percent) overstates the possibility to provide additional care. Drugresistance to antimalarials and increasingly to TB treatment demands new, expensive drugs, inflating the cost of health care, without offering new services. Another challenge i s the high cost o f treatment for HIV/AIDS. Apart from the pressure this will create on the overall budget, due to the pattern of HIV/AIDS which affects non-poor households much more than poor households, increasing the share o f financing that supports treatment will make the healthbudget less pro-poor..2' The health sector is affected by a serious human resource crisis. Only one third o f the positions for medical officers (going by the staffing norms) i s filled, and only 23 percent o f the assistant medical officer and public health nurse positions are filled (MoH 2005). This human resource crisis goes back to the mid 1990's when the total health workforce decreased from around 67,000 to 49,000 in2002, with the population increasing at the same time from 25 million to 33 million inhabitants. This affects people living inpoor areas especially as, in the absence o f additional incentives, it i s hard to motivate medical personnel to take up positions in areas with lack o f houses, no good schools, and low quality medical facilities. Malnutrition is high (over 40 percent) and is closely associated to a reduced ability for learning, earning an income and bad health. Improved nutrition i s an important goal in itself which, like education, will contribute to higher future growth. High income growth i s not sufficient to reduce malnutrition to acceptable levels but cost effective interventions do exist. To sustainably attack malnutrition, greater policy focus on nutrition i s needed not only in the health sector, but also in other sectors, especially in agriculture and education, requiring institutional reforms. Nutrition rates are worst amongst the poor. According to the 1999 TRCHS 50 percent (34 percent) of children inthe bottom two quintiles are stunted (respectively underweight) as opposed to 23 percent (22 percent) o f children from the top quintile. Tanzanians are not only affected by protein-energy malnutrition, but many also suffer from micro-nutrient deficiencies such as iodine deficiency, iron deficiency and vitamin A deficiency. According to the 2004 TDHS only 43 percent of households use adequately iodized salt and 46 percent o f children aged 6-59 months benefited from vitamin A supplementation in the six months preceding the survey. Approximately two thirdof children and 43 percent of women are anemic. An active population policy could support Tanzania's efforts to reduce poverty. Fertility is high and is not declining in Tanzania. Incombination with the decline in infant mortality, this suggests that population growth is increasing. While the average population growth rate between 1988 and 2001 was 2.9 percent, projections carried out by the National Bureau o f Statistics (URT 2006) show a population growth rate o f 3.3 percent in 2003 (with a population o f 34,876,231) which i s projected to decline to 2.9 percent in 2025 (with a population of 68,794,180), without HIV/AIDS assumptions. Taking into account HIV/AIDS assumptions, the projections indicate that Tanzania population growth rate will increase at a rate o f 2.9 percent in 2003 (with total Sustaining and Sharing GrowthinTanzania 64 population of 34,719,999) while in 2025 the increase will be 2.6 percent (with population o f 63,516,735). Highfertility leads to highdependency ratios which inhibitsthe ability o f households to save and invest. Children from large families are worse educated and are less well nourished. Highfertility contributes to (gender) inequalities and puts a budgetary strain on the ability o f the public sector to deliver quality services inkey sectors such as health and education. To strengthen the capacity of the poor to participate in growth, exposure to risk must also be reduced. Risks are pervasive and these disproportionately affect the poor. Risk is a structural determinant o f poverty and reduces growth, because it inhibits people to take up the most remunerative activities, and because its presence reduces the long run value o f the capital stock. Much can be done to reduce risk through sector interventions, including more effective health services, preventive care, agricultural research into drought resistance varieties, financial market and rural infrastructure development and enhanced access to clean water. Vulnerability is best addressed through a mix of economic growth, attention to risk reduction and a selective use of safety nets. Vulnerability results from a combination o f low income and exposure to shocks and reducing vulnerability requires income growth and less exposure to risk. Safety nets should be relied upon selectively and need to be well-targeted and based on a sound understanding o f the problem at hand, including the distortions safety nets may cause to the business environment. An understanding o f poverty traps would help to identify whether safety nets are needed or whether bottlenecks, once addressed, would allow vulnerable groups to participate in growth. The quality of, and access to, data for monitoring and evaluation of poverty and growth need to be improved. The debate around poverty and growth is hindered by a lack o f data, uneven data quality, and a restrictive provision o f data to researchers and other interested parties. Entire surveys have been lost, regional GDP data appear to be inconsistent, inadequate quality control contributes to unreliable survey data, and the long time span between data collection and reporting slows the policy debate. Knowledge about causes for poverty and growth or effective interventions will remain restricted as long as leading research institutions in Tanzania do not have access to complete data sets like the HBS. Sustaining and Sharing GrowthinTanzania 65 4. MANAGINGPOLICIESAND RESOURCES FOR SUSTAINED SHARED GROWTH STRENGTHENINGINSTITUTIONSTO STEER THE DEVELOPMENT IMPLEMENTATION AND OFTANZANIA'S STRATEGYFORGROWTH AND POVERTYREDUCTION The focus of economic policy during the past decadehasbeenprimarily onmacro-economic stabilization and reformof the public sector and more recently on poverty reduction with relatively little emphasis on the quality of economic growth and structural transformation. Inaddition, economic management has been fragmented among a variety of institutions which has resulted in a lack o f adequate coordination o f policy formulation, implementation and monitoring. A successful growth process requires institutions that ensure that government policy and its implementation i s fully aligned with its growth objectives and that its implementation i s monitored accordingly. In addition to coordination within government, a regular dialogue between the public and the private sector i s essential for the identification o f growth opportunities and concerted action by the public and private sector to exploit these opportunities. There are six key challenges that need to be addressed urgently in order to ensure that economic policy making i s able to support and sustain higheconomic growth and to react appropriately to the evolution o f the domestic and international economy. Complement economic stabilization with greater focus on economic growth and align institutional arrangements accordingly. The National Strategy for Growth and Reduction o f Poverty rightly puts greater emphasis on economic growth as a key mechanism contributing to reducing poverty and suggests a more proactive role o f government in the pursuit o f economic growth. This will require a strengthening o f institutions related to the earlier narrower scope o f the PRSP and may also require evolution inthe respective roles o f VPO, MPEE and other central agencies in the NSGRP coordination and implementation. The government's ongoing crosscutting reforms provide an opportunity for aligning the goals and objectives o f these reforms (public service, public finance, local government, and legal/judicial reforms) to support the growth agenda. Strengthen coordination of economic policy formulation and implementation. In addition, while the Ministry o f Finance and the Bank o f Tanzania are doing a commendable job in managing the Tanzanian economy for stability, coordinating and formulating the broader growth agenda exceeds their institutional mandates and capacities and would dilute their focus on their core responsibilities. In discussions with a variety o f stakeholders, there was a clearly perceived lack o f effective policy coordination. The main responsibility for the coordination o f the formulation, implementation, and monitoring o f a growth strategy that would rely on contributions by all parts o f government needs to be located at the centre o f government. Clarifying the respective roles o f a number o f ministries, departments and agencies (MDAs) and committees with the aim o f strengthening the policy managementprocess inTanzania needs to be a priority. This will also include greater attention to ensuring adequate capacity to effectively carry out the coordination function for improved policy implementationinTanzania. Sustainingand SharingGrowth inTanzania 66 Create a platformfor a NationalDialogueon Growth relatedissues. The process andcontent o f policy formulation and coordination clearly needs to be an open one, which encourages dialogue and participationby stakeholders. Invarious areas, Tanzania has been very successful in establishing processes that allow for continuous dialogue between the government and stakeholders and which draw on resources outside o f government for the formulation o f plans and strategies. Such processes include the Public Expenditure ReviewNTEF process, where the Ministry of Finance has opened up the budget process while at the same time establishing a framework in which analytic work by various parties i s coordinated and brought to bear on government processes. Similarly, the Vice President's Office has complemented the broad consultative processes on the PRSP that take place duringthe preparation of the PRSP or progress report with the establishment o f the Research and Analysis Working Group (RAWG) which on the one hand serves as a platform for an ongoing dialogue between government and stakeholders on poverty related issues, but which also serves as an instrument to coordinate analytic work in this area. In the area o f economic growth and structural transformation such a platform is currently not available. This results ina situation where a variety o f initiatives proceed inparallel with minimal interaction or coordination and potential overlap and duplication. A similar set-up to the PER or the RAWG processes could be envisaged in the area o f economic growth, with a focus on establishing a platform for the regular dialogue on growth related issues among the various stakeholders, including government, the private sector, civil society and development partners. Ensure adequate governance arrangements for growth enhancing government interventions. Inaddition to government's responsibility for creating an enabling environment for private sector activities, specific market failures need to be addressed through targeted government interventions. Inthe past, limited government capacity, rent seeking, bureaucratic inertia, or corruption have often resulted in ineffective programs and the waste o f resources. Thus to increase the likelihood o f success, it will be important to carefully select and design such interventions, and to put in place appropriate monitoring and governance arrangements that would allow the evaluation o f whether a program i s indeedreaching its objectives. Inparticular, programs should have clear criteria for success and failure, contain a sunset clause to prevent that ineffective interventions persist, focus primarily on new activities to foster new areas o f comparative advantage, target activities rather than sectors, and support activities with the potential for positive spill-overs and demonstration effects. Inconsidering such interventions, an important question has to be whether the capacity to monitor and implement such a program i s indeed available. The implementation should be assigned to autonomous agencies with demonstrated capacity and clearly defined accountability to a principal who has a clear stake in the outcomes and political authority at the highest level. Redefine government-private sector relationship. Recent research on economic growth (e.g., Hausmann and Rodrik 2004) highlights the importance o f public-private sector interactions in finding appropriate approaches that will result in higher growth, both at the micro- and at the macro-economic level. The annual Investor Round Table, which serves as a high level platform for dialogue between the government and the private sector i s a laudable initiative which should be continued and strengthened. This needs to be supported by revisiting the private-public sector interface that takes place on an ongoing basis through central government institutions such as the Tanzania Investment Center, the Tanzania Revenue Authority, the Ministry for Industry, Trade and Marketing or the agriculture sector ministries with a view towards redefining the interaction from a purely regulatory or administrative interaction to a problem solving interaction. Strengthen private sector institutions. For the public-private sector relationship to be productive, the private sector needs to develop appropriate institutions to make its views heard. A Sustaining and Sharing GrowthinTanzania 67 variety o f such institutions exists, including the TPSF, Confederation o f Tanzania Industries (CTI), Tanzania Bankers Association (TBA), Tanzania Chamber o f Mines (TCM), Tourism Council o f Tanzania (TCT), Tanzania Oil Marketing Companies (TOMC), Tanzania Chambers o f Commerce, Industry and Agriculture (TCCIA), Tanzania Association o f Consultants (TACO), Tanzania Chamber o f Agriculture and Livestock (TCAL). An important issue would be to survey these institutions with respect to their role in the government-private sector dialogue, their satisfaction with the dialogue, and their capacity to effectively represent their constituents. Strengthen the capacity of institutions at the regional and district levelto play a greater role in promoting growth at the local level. The study highlights the regional diversity of Tanzania's economy with respect to potential sources o f growth, access to infrastructure, and natural resource endowment. A shared-growth strategy thus requires not only strong institutions at the national level, but also at the local level. Tanzania's ongoing decentralization process touches on important elements o f the growth agenda. In particular, responsibility for service delivery in agriculture and infrastructure (district roads, water) i s being shifted to the local authorities, while the role o f the central government inthese areas i s limited to policy formulation and monitoring. Funding for these activities i s primarily through earmarked transfers from the central government. A formula based system for budgetary transfers to the districts has been adopted, based on demographic and social indicators. In order to ensure that local authorities play a supportive role for regional growth, it will be important that the use o f these resources i s guided by regional growth strategies developed in partnership between local authorities, the private sector, and other stakeholders. Strengthening o f accountability arrangements at the local level needs to accompany increased resource flows to the local authorities. Regional or district level growth strategies combined with strong accountability arrangements form also the basis for a successful switch from conditional to unconditionaltransfers, which i s envisagedinthe medium term. Unconditional transfers will provide local authorities with greater scope to implement a growth strategy that i s tailored to the specifics o f the district or region. In this context, the division of revenue sources between central government and local governments i s of importance. At present, the revenue sources for local authorities are limited to a closed list and typically provide only 10 to 20 percent o f a district's revenue. The current system provides scope for redistribution o f resources across districts. However, a system where resource availability at the local level i s more closely linked to revenue generation at the local level mightprovide more incentives for a greater focus on economic growth at the local level. SCALINGUPPUBLICSPENDINGFORTHE IMPLEMENTATIONOF THE NSGW Public expenditures play a critical role in Tanzania's efforts to achieve sustained growth and to reduce poverty. Central government expenditures have increased from 16 percent o f GDP in 1998/99 to 28 percent in 2005/06. The allocation o f expenditures was guided by Tanzania's Poverty Reduction Strategy and growth related and social expenditures were the primary contributors to the overall increase in public expenditures. The annual Public Expenditure Review Process serves as an important platform for the open dialogue on public expenditure issues and the Annual Public Expenditure and Financial Assessment report provide detailed analysis o f aggregate fiscal discipline and allocative and operational efficiency. Estimates of the cost of achieving the MDGs suggest that these by far exceed the currently available resources. T h i s calls for developing an appropriate strategy for mobilizing and managing the resources required to finance the implementation o f Tanzania's strategy for growth and poverty reduction. Sustaining and Sharing GrowthinTanzania 68 Potential sources of financing of the NSGRP (aside from contributions by households,the private sector, and NGOs) are taxation, seignorage, domestic and foreign borrowing, and foreign aid. Each source o f financing has different macro-economic cost. Intheory, as long as the marginal benefits o f government spending are higher than the marginal cost o f funds, it would be advisable to expand spending. The marginal benefit o f additional government spending i s likely to be declining while the marginal cost o f finance is increasing. InTanzania, absorptive constraints such as the shortage o f qualified nurses, secondary school teachers, or road engineers are likely to result inrapidly declining marginal benefits o f government spending. Regardingthe marginal cost o f funds, policy making inTanzania i s typically based on the assumption that these are relatively low for foreign aid inflows (but constrained by the amount donors are willing to provide) and domestic revenue. Implicitly, the cost o f domestic borrowing (including the crowding out effect on private credit) and foreign, non-concessional borrowing are considered to be exceedingly high and thus excluded from the financing package. The following sections examine the potential for expanding domestic resources and aid inflows. Domestic revenue currently stands at 14 percent of GDP and is the main source of government financing. Tanzania has made substantial progress in strengthening its tax administration, which underlies most o f the recent increase inrevenue mobilization. Overall, the Tanzanian tax policy i s sound and not inimical to growth. (FIAS, 2006) However, there are several issues which could enhance the contribution o fthe tax system to fostering shared growth. First, there remains an "urban bias" in the tax system, where effective tax rates are higher for farmers as for - mostly urban -businesses. Inparticular, thecropcesscollectedbylocal authorities imposes a relatively heavy tax burden on agriculture. A recent study (FIAS, 2006) estimates the average effective tax rate for small farmers as about 5 percent, while that for small urban businesses i s estimated at around 1. 4 percent. Similarly, large businesses benefit, by paying 7.5 percent average effective tax compared to 12.5 percent for large farmers. Second, the presumptivetax regime for small businesses is one of the more sophisticatedly designed in the region. Different groups o f taxpayers are categorized in the four bands depending on estimates o f average turn over for particular trades or professions. The system i s based on generally low tax rates and provides flexibility to small businesses o f different record- keeping abilities. Itprovides incentives to keep records, which i s good management practice, and i s rewarded by a lower average tax burden. However, it i s regressive for small businesses that do not keep records.22 As non-record-keepers are more likely to lack the capacity to keep proper accounts and not understand the benefits o f maintaining records. While the system i s thus designed to provide an incentive to keep records, its success hinges on proper outreach and education so that small businesses can better understand the benefits o f keeping records and graduating to the record-keepers system. The system i s inconsistent and inequitable with respect to Personal Income Tax, which has an initial tax-exempt band for total income less than TSH 960,000. Under the presumptive regime, businesses in the bottom band (turnover below TSH 3 million) that do not maintain records are required to pay TSH 35,000 irrespective o f their actual turnover. Third, there are significant weaknesses in the taxation of natural resources, discussed elsewhere in this report. These weaknesses in collecting natural resource rents result both in distortions to the sustainable exploitation o f natural resources and suboptimal collection o f revenue for the use o f natural resources. Further reforms of tax policy need to carefully consider their impact on economic growth, s investment and saving. This is particularly important in the presence o f large aid inflows and Sustaining and SharingGrowthinTanzania 69 possible absorptive constraints on the overall level o f government spending. Here it i s important to recognize that in the medium to long run, the primary driver o f enhanced revenue is an expanding revenue base through a combination o f economic growth and structural transformation, which would expand and broaden the revenue base. For example, per capita economic growth o f four percent would more than double per capita revenue by 2015. Foreign aid finances more than 40 percent of government expenditure and commitmentsat the G 8 summit inGleneaglesholdthe prospect of a significant scalingup of aid. Inaddition to the absorptive constraints mentioned before, the macro-economic effects o f aid need to be monitored closely. Our analysis o f the sources o f economic growth suggests that aid financed expenditures have provided an important demand side stimulus to the economy inrecent years in Tanzania. Inthe medium to long run, aid financed investments in human and physical capital are intended to strengthen the supply side o f the economy as the basis for sustained growth. However, higher aid flows do not necessarily lead to higher economic growth and the achievement o f targeted development objectives, with many examples o f countries that despite high inflows o f foreign aid saw economic stagnation or declineOz3International experience suggests that it is critical to manage two separate, but related sets o f problems that could undermine the positive impact o f aid: 0 Weakening o f institutions; and 0 Weakening o f competitiveness through Dutch disease effects. A weakening of institutions may occur if aid undermines accountability to domestic stakeholders, distorts incentivesfor public sector performance,or removes the pressure for an efficient revenue collectionsystem. It is thus important, that scaling up o f aid goes hand in hand with reforms that improve governance and domestic accountability. In addition, the design o f aid delivery mechanism should support domestic accountability rather than replacing it with accountability to the donor agencies. In this respect, processes such as the Tanzania Assistance Strategy and the Joint Assistance Strategy have an important role to play inproposing appropriate accountability mechanisms. Efforts to fully integrate foreign aid into Tanzania's budget system and thus make them subject to the same accountability process as domestic resources are also important. The impact on competitivenessoccurs through a spending effect and a resource movement effect, which are commonly labeled "Dutch disease" effects. A "spending effect" arises from the fact that unless all inflows are spent on imports or exportable goods, the real exchange rate will appreciate which implies a deterioration o f a country's competitiveness. Inaddition, there i s a "resource movement effect," as resources would move into the booming sector (Le. the government and development bureaucracy) and the non-traded sector. Analysis o f real exchange rate movements and developments in the tradables sector do not suggest a negative impact of ODA flows inTanzania. Simulations using a computable general equilibrium model o f Tanzania also suggest that the impact o f aid-inflows on the real exchange rate i s likely to be modest and more than offset by aid inducedproductivity increases. Nonetheless, to ensure that increased aid inflows enhance rather than reduce Tanzania's international competitiveness, scaling up o f aid should be accompanied by the following set o fmeasures (Foster et. al., 2005): 0 ensuring that increased aid-financed spending i s accompanied by increased absorption o f the foreign exchange, which (assuming that government spending continues to have a high local content) will probably require acceptance o f exchange rate appreciation. spending the aid without absorbing the foreign exchange does nothing to increase the real Sustaining and Sharing GrowthinTanzania 70 resources available to the economy, but makes it likely that government will crowd out the private sector; further liberalizing imports to help increase absorption and reduce the need for real exchange appreciation or reserve accumulation; focusing on expenditures that will quickly release supply constraints and have a higher import content, including transport investments; continuing to improve the efficiency o f the banking sector to ameliorate the need for high real interest rates; considering adopting a more relaxedmonetarypolicy; and exchange rate, monetary, and fiscal policy need to be coordinated with the implications o f aid inflows inmind. Followingthe implementationof the EnhancedHighly IndebtedPoor Countries(HIPC) and the Multilateral Debt Relief (MDRI) initiatives, Tanzania's debt sustainability indicators are well below the debt sustainabilitythresholds. The net present value o f debt-to-export ratio declined to 64 percent in 2006. Panel (a) o f Figure 20 shows the baseline projections o f the net present value o f exports-to-GDP ratio. The ratio i s projectedto increase initially to 71 percent by 2010, but decline subsequently to 63 percent by 2026, assuming average export growth o f 8.5 percent, and annual growth o f concessional credit disbursements (primarily from World Bank and African Development Bank credits) by 2 percent. A doubling o fmultilateral credit disbursements in2011and subsequent annual growthby2 percentwouldresultina gradually increasingNPV of debt-to-export ratio which would reach 101percent by 2021 and decline thereafter. Contingent on strong GDP and export growth, significant increases in concessional borrowing seem thus to be feasible, without leadingto unsustainable levels o f debt. Figure20. Multilateral credit disbursements and Debt Sustainability: Net PresentValue of Debt-to- Export Ratio, 2006-26 (a) Growthof multilateralcredit disbursementsby 2 YO MLTdisbursements (US$ nllion) NW of debt-to-exportratio (YO) (b) Doublingof multilateralcredit disbursements in2011 Sustainingand SharingGrowthinTanzania 71 1600 120 + 1400 100 1200 -E -E 1000 80 EBx 800 60 -x 8 3 600 40 8 400 c 0 200 2o = i? 0 o NW of debt-to-export ratio (%) Source: IMF and World B a n k staff estimates Given Tanzania's low per capita income, aid flows that are large in relation to Tanzania's own resourcesare likely to remain a reality for the foreseeable future. The key to reduced aid-dependence i s sustained economic growth. Changes inthe revenue to GDP ratio are unlikely to significantly reduce the degree o f aid dependence. However, efforts to increase efficiency and reduce transaction cost through donor harmonization require continued attention in order to maximize the impact o f aid and to sustain the confidence o f the donor community in supporting Tanzania's growth and reformagenda. The structure of public expenditure and revenue should preserve enough flexibilityto deal with fluctuations in foreign aid. This implies avoiding a situation where spending that is difficult to scale back, such as wages and salaries, dominate. Similarly, the recurrent cost implication o f donor financed projects need to be evaluated carefully. Outsourcing and public private partnerships may also be useful instruments to facilitate expenditure side adjustments, in case there i s a shortfall in aid. In addition, this would be a means for developing the private sector in Tanzania by using aid inflows to provide business opportunities for the private sector. Onthe revenue side, an appropriate strategy may be not to fully exploit the revenue potential and to focus on a strengthening o f tax administration inorder to be able to scale up revenue collection to compensate for a permanent decline inaid. Finally, it should be mentioned that the implementation of the NSGRP is not only constrained by the lack of resources, but also by absorptive constraints. Absorptive constraints in Tanzania relate to the availability o f trained manpower, the quality o f public administration, the capacity to manage an expanded expenditure program, or the quality o f the institutionaland policy frameworks at the sectoral level. The limitedsupply of specializedskills and unattractivecompensationschemes inthe public sector are a serious absorptive constraint. Even at current levels o f financing, a shortage of nurses, qualified teachers, road engineers, and the like, constrains the implementation o f the NGSRP. In order to preserve the competitiveness o f the private sector, bidding away human resources from the private sector through more attractive compensation and benefit packages should be avoided. Instead, emphasis needs to be placed on cooperation with the private sector and expansion o f the skill pool through training. Sustaining and Sharing GrowthinTanzania 72 As a low income developing economy, the policy and institutionalcapacity of Tanzania is modest. While insome o f the several sectors the institutional capacity is reasonable, inothers it remains weak. Sector development programs are an important instrument for puttingappropriate policy and institutional frameworks in place. Sectors that already have operational sector development programs inplace, such as the education and health sectors, are able to use available resources reasonably effectively, and could absorb additional resources. Inother sectors such as roads, power, or water, major increases in budgetary funding will only be effective as policy, institutional, and capacity constraints are addressed. With the change inthe role of government in many sectors from implementer to regulator, the capacity of the privatesector is a factor. Inthe water androads sectors, for example, there are not enough private contractors with adequate expertise and necessary equipment who can handle tasks or deliver services on time. The weakness i s again acute in rural areas. However, experience from other countries indicates that the private sector i s quick to develop adequate capacities once business opportunities become available. While public expenditure and financial management at the central government level is fairly well developed in Tanzania, capacity in local authorities is generally much weaker and shows much variation across districts. At present, most government expenditure, including spending by local authorities funded through central government transfers, i s under fairly tight control o f the central government with only limited scope for leakages at the district level. The public finance management reform program as well as the local government reform program are the principal mechanisms for strengthening financial management systems and capacities at the local level to ensure that as decentralization progresses adequate capacities are beingput inplace. Tanzania's public sector reform program has contributed significantly to the rationalization and strengtheningof the public administration. Nonetheless, various factors still constrain the quality o f public administration, including low pay levels and a generally low level o f education inthe population which i s reflected inthe level o f education found among civil servants. Capacity constraints are generally larger at the district than at the central government level. The Local Government Reform Program i s intended to strengthen the capacity o f local authorities to plan and execute programs as under the fiscal decentralization program responsibilities are shifted to them. Increasingpay levels inthe public sector is seen as one of the pre-conditions for creating an environment in which other reforms to enhance public sector performance can become effective. Government is implementing a Medium Term Pay Policy that seeks to increase and decompress government wages. Significant wage increases were granted in 2006, raising the government wage bill to 5.9 percent o f GDP. A comprehensive review o f the MTPP has been undertaken by the Presidential Commission on Public Service Pay, with a view to examine options to accelerate the implementationo f the pay policy by implementing the policy inrelation to the employment levels for each ministry, department, and agency and their budgetary implications. Sustainingand Sharing Growth inTanzania 73 References Anand S., and J. Morduch. 1996. Poverty and the `Population Problem'. The Harvard Center for Populationand Development Studies and Department o f Economics, HarvardUniversity. Bank of Tanzania. 2004. TanzaniaInvestment Report 2004. Dar es Salaam. Barro R. 1991. Economic Growth in a Cross Section of Countries. Quarterly Journal of Economics 106(2): 407-441. Behnnan J., H. Alderman and J. Hoddinott 2004. Hunger and Malnutrition. Copenhagen Consensus: Challenges and Opportunities. Bosworth, Barry, and Susan Collins. 2003. The Empirics of Growth: An Update. Brookings Institution. Washington. D.C. Christiaensen L., V. Hofmann and A. Sarris. 2004. Coffee Price Risk in Perspective: Vulnerability Among Small Holder Coffee Growers in Tanzania. UnpublishedManuscript. World Bank. WashingtonD.C. Christiaensen, L.and Lionel Demery. 2006. TheRole of Agriculture in Poverty Reduction in Sub Saharan Africa Revisited. Processed.May, 2006. Collins, Susan and Barry Bosworth. 1996. Economic Growth in East Asia: Accumulation versus Assimilation. Brookings Paperson Economic Activity, 2:1996. D e Soto, Hemando. 2001. The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else. Basic Books. Deaton A. and S. Zaidi 2002. Guidelinesfor Constructing Consumption Aggregatesfor Welfare Analysis. World Bank. LSMSWorking Paper No. 135. Dercon S. 1996. Risk, Crop Choice, and Savings: Evidence from Tanzania. Economic Development and Cultural Change 44(3): 485-513. Dercon S. 1998. Wealth, Risk and Activity Choice: Cattle in Western Tanzania. Journal of Development Economics 55(1): 1-42. Dercon S., T. Bold, J. de Weerdt and A. Pankhurst. 2004. Extending Insurance? Funeral Associations in Ethiopia and Tanzania. UnpublishedManuscript.Mimeo. Economic and Social Research Foundation. 2005, Vulnerability and Resilience to Poverty in Tanzania: Causes, Consequences and Policy Implications. Dar es Salaam. Elbers C., J.W. Gunning and B.Kinsey. 2003. Growth and Risk. Tinbergen Institute Discussion Paper. TI 2003-068/2. SustainingandSharingGrowth inTanzania 74 Foreign Investment Advisory Service (FIAS). 2004. Free Zones: Performance, Lessons Learnt, and Implicationsfor Zone Development. FinalDraft. November 2004. Foreign Investment Advisory Service (FIAS). 2006. Sector Study of the Effective Tax Burden - Tanzania.Processed. April 2006. Fink, Carsten, Aaditya Mattoo and Ileana Cristina Neagu. 2002. Assessing the Role of Communications Costs in International Trade. Working Paper #2929, 2002. The World Bank: Washington, D.C. Food and Agriculture Organization (FAO). 1997. Irrigation Technology in Support of Food Security. Proceedings of a RegionalWorkshop, Rome, 1997. Foodand Agriculture Organization (FAO). 2004. AQUASTAT. FAO's global information system on water and agriculture. FAO. Rome. httd/www.fao,oralAG/anl/ag;lwlaauastat/water redindexstm Food and Agriculture Organization (FAO) and World Bank. 2001. Tanzania Agriculture: Reaping the Benefits of Reform. Foster, M. et al. 2005. TheMacroeconomicImpact of Scaling up Aid Flows: The Case Study of Tanzania. Final report prepared for the OD1-London. Mimeo. September 9,2005. Harding, A., G. Kahyarara, and Neil Rankin. 2002. Firm Growth, Productivity and Earnings in Tanzania 1992-1999. Center for the Study of African Economies (CSAE). Universityof Oxford. Hausmann, Ricardo, DaniRodrik, Andreas Velasco. 2004. GrowthDiagnostics. September 2004. Hausmann, Ricardo, and Dani Rodrick. 2005. Self-Discovery in a Development Strategy for El Salvador. Economia Journal of the Latin American and Caribbean Economic Association 6(1): - 43-101. Hemming, Richard, Michael Kell, and SelmaMahfouz. 2002. TheEffectiveness of Fiscal Policy in StimulatingEconomic Activity - A review of theLiterature. IMFWorking Paper WP/02/08. December2002. Horton S. 1999. Opportunities for Investment in Low Income Asia. Asian Development Review 17(1,2): 246-273. Horton S. and J. Ross. 2003. TheEconomics oflron Efficiency. FoodPolicy 28(1):51-75. Huppi, Monica, and Martin Ravallion. 1991. The Sectoral Structure of Poverty in Indonesia during an Adjustment Period: Evidencefor Indonesia in the Mid-1980s. World Development 19 (12): 1653-78. IMF.2004. Tanzania-SelectedIssues and Statistical Appendix. July 22,2004. Kessy F. 2004. Rural Income Dynamics in Kagera Region, Tanzania. ESRF. Dar es Salaam. Background paper to CEM. Sustaining and Sharing Growth inTanzania 75 Kolavalli, S. and S. Beddies.2004. Report on Phase One: Institutional Mapping of Coffee and Cotton, Reform of the Tanzania Cotton, Coffee, Cashew and TeaBoards: Options and Implicationsfor Other Crop Boards. World Bank, Washington, DC. Kweka JosaphatP. and Oliver Morrissey. 1999. Government Spending and Economic Growth Empirical Evidencefrom Tanzania (I965-1996).Draft preparedfor DSAAnnual Conference. UniversityofBath. September 12-14, 1999. Lanjouw P and M. Ravallion.(l995). Poverty and Household Size. The Economic Joumal, 105 (November), 1415-1434. LorenzN.and C. Mpemba. 2005. Review of the State of Health in Tanzania 2004. Mimeo. Klasen S. 2003. I n search of the holy grail: how to achieve pro-poor growth? Annual World Bank Conference on Development Economics -Europe. Toward Pro-Poor Policies: Aid Institutions and Globalization. Editedby B. Tungodden, N. Stem and I.Kolstad. World Bank / Oxford UniversityPress: New York. pp: 63-94. Larson D. and F. Plessmann. 2002. Do Farmers Choose to be Ineflcient? Evidencefrom Bicol, Philippines. Policy ResearchWorking Paper 2787. World Bank, WashingtonD.C. Lindeboom W. and B. Kilama. 2005. Trends and Determinants of Malnutrition in Tanzania. PaperPresentedat the 10* Annual REPOA ResearchWorkshop. Mankiw N.G., J. Roemer and P. Weil 1992.A Contribution to the Empirics of Economic Growth. Quarterly Joumal of Economics 107(2): 407-37. Ministry of Education and Culture, 2004 & 2005. Basic Statistics in Education 1999-2003. National Data. Ministryof Educationand Culture, Dar es Salaam. Ministry of Health. 2004. National Aids Control Program. HIV/MDS/STI Surveillance Report. January-December 2003. National Bureau of Statistics. 2002. Household Budget Survey 2000/01. National Bureau of Statistics. Dares Salaam. Nelson, Edwin G. and Erik J. de Burijn. 2005. The Voluntary Formalization of Enterprises in a Developing Economy - The Case of Tanzania. Journal of International Development, Issue 17, 2005. Nyange, David. 2005. Survey of Agricultural Marketing Logistics Costs in Rural Tanzania. background paper for the Tanzania DTIS. Owens, Trudy, and Francis Teal. 2005. Paths out ofpoverty in Ghana and Tanzania in the 1990s. Mimeo. I Rafalimanana H. and C. Westoff 2001. Gap Between Preferred and Actual Birth Intervals in Sub-Saharan Africa: Implicationsfor Fertility and Child Health. DHS Analytical Studies No. 2. CalvertonMaryland: ORC Macro. Sustainingand Sharing GrowthinTanzania 76 Rajan, Raghuram G., and Arvind Subramanian. 2005. What Undermines Aid's Impact on Growth? IMFWorking Paper 05/126. InternationalMonetary Fund.Washington, D.C. Rodrik, Dani. 1998. Saving Transitions. Draft. July 1998, Rosenzweig M.R.and K.Wolpin 1993. Credit Market Constraints, Consumption Smoothing and the Accumulation of Durable Production Assets in Low Income Countries: Investments in Bullocks in India. Journal of Political Economy lOl(2): 223-224. Sarris, A., Sara Savastano, and Luc Chstiaensen. 2006. The role of Agriculture in Reducing Poverty in Tanzania :A Household Perspective from Rural Kilimanjaro and Ruvuma. Mimeo. May 2006. Schclarek, Alfredo. 2005. Fiscal Policy and Private Consumption in Industrial and Developing Countries. Draft. September30, 2005. Schneider, Friedrich. 2004. The Size of the Shadow Economies of 145 Countries all over the World: First Results over the Period 1999 to 2003. E A Discussion Paper Series. E A DP No. 1431. December 2004. Soderbom, Mans, Francis Teal, Anthony Wambugu and Godius Kahyarar. 2004. TheDynamics of Returns to Education in Kenyan and Tanzanian Manufacturing. Centre for the Study of African Economies (CSAE), Universityof Oxford, CSAE WPS/2003-17, January 2004. UN Millennium Project. 2005. Investing in Development: A Practical Plan to Achieve the Millennium Development Goals. Earthscan.London. UNICEF. 1990. Womenand Children in Tanzania: An Overview. UNICEF, Dar es Salaam. UnitedRepublic ofTanzania. 2004. TanzaniaMini-Tiger Plan 2020. December2004. UnitedRepublic of Tanzania. 2006. 2002 Population and Housing Census: Projections. Volume VI. National Bureau of Statistics. Ministry of Planning, Economy, and Empowerment. Dar es Salaam. World Bank. 2001. Social Protection Sector Strategy: from Safety Net to Springboard. Sector Strategy Paper. The World Bank. Washington D.C. World Bank. 2004a. World Development Report 2005. A Better Investment Climate for Everyone. Oxford UniversityPress. New York. World Bank. 2004b. Investment Climate Assessment: Improving Enterprise Performance and Growth in Tanzania. Washington, DC: World Bank. World Bank. 2005a. Tanzania - Diagnostic Trade Integration Study. June 2005. Washington, DC: World Bank. World Bank. 2005b. GlobalMonitoring Report 2005. Washington, DC: World Bank. World Bank. UnitedRepublic of Tanzania -Public Expenditure Review FY03. June 2003. Sustaining and SharingGrowth inTanzania 77 World Bank. 2006a. DoingBusiness 2007: How to Reform. Washington D.C. World Bank. 2006b. TanzaniaPilot Rural Investment ClimateAssessment. Stimulating Non-farm Microenterprise Growth.Mimeo. June 27, 2006. SustainingandSharingGrowthinTanzania 78 ANNEX 1: LIST OFBACKGROUNDSTUDIES Alderman, Harold, J. G. M. Hoogeveen, and Mariacristina Rossi. 2005. "Reducing Child Malnutrition in Tanzania: Combined Effects of Income Growth and Program Interventions." World Bank, Washington, DC. Aubert, Jean-Eric, and Godwill Wanga. 2005. "Innovation in Tanzania: Insights, Issues, and Policies." World Bank, Washington, DC. Chandra, Vandana, Pooja Kacker, and Ying Li. 2005. "Tanzania: Growth , Exports, and Employment inthe ManufacturingSector." World Bank, Washington, DC. Christiaensen Luc, Vivian Hofinann, and Alexander Sanis. 2004. "Coffee Price Risk in Perspective: Vulnerability among Small Holder Coffee Growers in Tanzania." World Bank, Washington, DC. COWL 2005. "Natural Resource Based Growth: Summary Paper." World Bank, Washington, DC, Demombynes, Gabriel, and J. G. M. Hoogeveen. 2004. "Growth, Inequality, and Simulated Poverty Paths for Tanzania, 1992-2002." Policy ResearchWorking Paper 3432, World Bank, Washington, DC. Hoogeveen, J. G. M.2004. "The DistributionalImpact of the PEDP inRural Kilimanjaro." World Bank, Washington, DC. .2005. "Risk, Growth, andTransfers: Prioritizing Policies ina Low-Income Environment with Risk-The Case ofTanzania." World Bank, Washington, DC. infoDev. 2005. Improving Competitiveness in Tanzania: The Role of Information and Communication Technologies. October 2005. Kessy, Flora. 2004. "Rural IncomeDynamics inKagera Region, Tanzania." Economic and Social ResearchFoundation, Dar es Salaam. Kessy, Flora, and Oswald Mashindano. 2005. "Moving Out of Poverty: Understanding Growth and Democracy from the Bottom Up-The Case o f Ruvuma Region, Tanzania." Economic and Social ResearchFoundation, Dar es Salaam. Kilama, Blandina, and Wietze Lindenboom. 2004. "Trends in Malnutrition in Tanzania." Researchon Poverty Alleviation, Dar es Salaam. Kopicki, Ronald. 2004. "Supply Chain Development inTanzania. World Bank.Washington D.C. Mahamba, Robert, and Jorgen Levin. 2005. "Economic Growth, Sectoral Linkages, and Poverty ReductioninTanzania" World Bank, Washington, DC. Makki Shiva S., IJsbrand de Jong, and Henry Mahoo. 2005. Smallholder GroundWater Irrigation inTanzania. WorldBank.Washington D.C. Mkenda, Adolf. 2004. "The Benefits of Malnutrition Interventions: Empirical Evidence and Lessonsto Tanzania." University of Dar es Salaam, Dar es Salaam. .2005. PopulationGrowth, Economic Growth and Welfare Distribution: An Overview of Theory, Empirical Evidence and Implications to Tanzania. University of Dar es Salaam, Dar es Salaam. Mpango, Philip. 2005. "Spatial Dimensions of Growth and Poverty Reduction in Tanzania Mainland." World Bank, Washington, DC. Simonsen, Marianne., and Louise. Fox. 2005. "A Profile of Poverty in Tanzania." World Bank, Washington, DC. Sustainingand SharingGrowthinTanzania 79 Skof, Annabella. 2006. "Constraints to Technology Access in Tanzanian Horticulture: A Case Study of Barriers to the Introduction of Improved Seed and Pest Control Technologies." World Bank, Washington, DC. Tanzania Food and Nutrition Centre. 2004. "Causes of Malnutrition and Tanzania's Nutrition Programs: Past and Present." World Bank, Washington, DC. Utz, Anuja. 2004. "Fostering Innovation, Productivity, and Technological Change: Tanzania in the Knowledge Economy." Knowledge for Development Program, World Bank Institute, Washington, DC. Utz, Robert. 2005. "Review of Growth Performance and Prospects." World Bank, Washington, DC. van Dijk, Meine Pieter. 2006. Urban Rural Dynamics in Tanzania, through informal redistributionmechanisms. Mimeo. 10/15/2006. Sustaining and Sharing Growth inTanzania 80 Notes 1To the extent that increased government spending fell on imports, the effect on growth i s reduced. 2 Shares represent the average distribution o f FDI by Sector, 1999-2001 (Bank o f Tanzania. 2004) The KAM (http:l/www.worldbank.org/kam) includes 80 quantitative and qualitative variables that help to benchmark a country's position, vis a vis 128 countries, on key elements o f the four pillars o f the KE framework (economic and institutional regime, education, innovation, and ICTs). 4Regional GDP data for Tanzania could be overstating regional income disparities partly because production in the regions which contribute little to national GDP i s largely for subsistence and i s not fully captured inmarket-based GDP numbers. The purchasing power o f the shillingalso tendsto be higher inthe poorer regions. See Huppiand Ravallion (1991). 6Britain, Canada, Netherlands, Sweden. 7The median wait was considerably shorter than the average typical wait because several enterprises reported exceptionally long delays. For example, nine enterprises reported typical delays o f over 50 days. United Republic o f Tanzania, Legal Sector Reform Programme MediumTerm Strategy FYs 200516-200718. 'As much as 25% o f potentially suitable additional pasture land is affected by tse tse fly and cannot be used for cattle at present. "Christiaensen, L. and Lionel Demery. 2006. The Role o f Agriculture in Poverty Reduction inSub Saharan Africa Revisited. Processed. May, 2006. "Thesearefigurescalculatedonthebasisofthenationaldefinitionofemployment. l2Datafromthe IntegratedLabourForce Survey (National BureauofStatistics2001) and the Household Budget Survey (National Bureau of Statistics 2002) provide different measures o f informal sector activities, but the magnitude and trends provided by these two surveys present broadly similar pictures. l3 Regulatory constraints are complicated, lengthy and unpredictable procedures, inadequate institutionalarrangements, rent-seeking civil servants, unreasonable specifications and standards or a multiplicity o f taxes and levies etc. Non-regulatory constraints are for example poor clients, lack o f access to finance, poor infrastructure, unfair competition. The 75 percent is the highest average cost for coping with constraints that occurred in the firewood and charcoal sector. Inthe cloth making sector, for example, the highest average cost calculated was 5.8% of monthly sales. Sustainingand SharingGrowthinTanzania 81 l4This has been shown-among others -by the ILO/LTNIDO/UNDP study. l5ILO/UNIDO/UNDP (2002): "Roadmap Study o f the Informal Sector in Mainland Tanzania", Dar es Salaam, April 2002, p. 3. l6Instituto Libertad y Democracia (2005): "Program to Formalize the Assets o f the Poor o fTanzania and Strengthen the Rule o f Law: Final Diagnosis Report." Lima, September 2005. l7The decline in infant mortality i s contested. The DHS and TRSCH surveys carried out during the 1990s show a stagnation o f mortality rates up to 1996 and slight increase thereafter. Data from surveillance sites, on the other hand, supports the decline ininfant and child mortality. 18There i s some evidence o f a decline inthe infection rate amongst blood donors. It i s not clear whether this i s a statistical artifact or a reflection o f a decline inthe infectionrate. l9Apart from learning students proper care practices, one low-cost nutrition intervention to reduce anemia would be deworming o f all children at first day o f school. *'In constant 2001 TShs. The nominal amount for 2004 i s TShs 8,815. The 2001-2004 price index i s calculated from the Economic Survey, 2004, Table 11. 21This is not to say that the use of ARVs should not be promoted, or that there are no pro-poor elements to ARV treatment. Apart from the life saving benefits, HIV/AIDS and the prolonged illness that precedes it creates a large economic burden. And because many people infected decide to go home (poor) rural households carry a disproportionate share o f this burden (de Waal et al. 2004). 22About 70 percent o f presumptive tax payers registered with TRA do not keep proper records, despite the fact that this results ina higher average effective tax rate. 23Rajan and Subramanian (2005) provide a concise summary o f the debate.