Report No. 29396-LK Authorized Sri Lanka Disclosure Development Policy Review Public December 8, 2004 Poverty Reduction and Economic Management Sector Unit South Asia Region Authorized Disclosure Public Authorized Disclosure Public Authorized Disclosure Public Document of the World Bank Sri Lanka Development Policy Review TABLE OF CONTENTS ExecutiveSummary 1 I. DevelopmentOutcomesand UnderlyingCauses ......................................................................... 1 1. Social and Economic Outcomes ..................................................... 2. Underlying Causes and Trends ...................................... 4 A. The FirstThree Decades (1948-77): A Capsule History.................................... B. Post 1977 Reforms andtheir Impact ...... C. The Decades o f Civil Conflict, 1983-2001.......................................................................... 8 D. EmergingDevelopment Issues and Problems .................. The Fiscal Problem ....................................................... 11 Administrative Erosion ........................................... A Sluggish Rural Economy ....................................................... 13 An Infrastructure Backlog........................................ Increase inRegional Disparities Quality o f Education Constrained........................... E. Growthand Poverty, 1990-2002: An Ov 11. Ceasefire andUnfinishedReforms ............................................................................................. 24 A. Peaceand Reform Agendas: 2002-04 ............................................................................... 24 B. Early Gains from Reforms andPeace................................................................................ 30 111.TowardsFaster andMoreEquitableGrowth ............................................................................ 32 1. Managing Public Finances.......................................... 32 2. Tax Policies to Promote Growth and Fiscal Consolidation............................................... 38 3. External Sector Policies for Faster Growth ................................... 4. Reviving Growth o fNon-plantation A 5. Infrastructure Priorities for Growth, Connectivity and Equity...... 6. EducationReforms for Growth and Equity .................................... 57 7. Peace, Conflict and Development ............................................................................ 8. Development Risks.................................... References........................................................................................................................................... 65 Annexes ............................................................................................................................................. 73 Sri Lanka Development Policy Review List of Tables Table 1.1 Growth of GDP andMajor Economic Sectors............ ............................................. 3 Table 1.2 Sri Lanka in an Asian Perspective:Per capita GNP re e to US (`33) .......................... 3 Table 1.3 Social Expenditureas Per Cent of GNP, 195011-1977.................................................... 5 Table 1.4 Indictors of Fiscal Stress (percent o f GDP)................................................................... 12 Table 1.5 Civil Service Staffing in Selected Countries................................................................. 13 Table 1.6 Table 1.7 Table 1.8 Table 1.9 GDP Shares by Province....... Table 1.10 Economic Activity by Regions............ ................................................................ 19 Table 1.11 Proportionof Primary Children Achieving Mastery o f Language Skills, 2003 ............20 Table 1.12 Proportionof Primary Children Achieving Masteryo f NumeracySkills, 2003 ...........20 Table 1.13 Table 1.14 Table 1.15 The headcount indicesby distri Table 1.16 Table 1.17 Table 2.1 .............................................................. 30 Table 3.1 Sri Lanka Summary o f Gove erations................................................. 33 Table 3.2 BaselineMacroeconomicAssumptions ............................................,....................,......34 Table 3.3 Target Primary Balance and SelectedLong-Run Variables.......................................... 35 Table 3.4 Medium Term Revenue Scenario.................................... Table 3.5 Rural HouseholdAccess to Infrastructureand Financial Table 3.6 InternationalComparisono f Sri Lanka's RoadTransport Table 3.7 EducationExpenditurein Sri Lanka and Other Countries............................................ 57 Listof Boxes Box 1.1 Sri Lanka and the Millennium Development Box 1.2 The HumanitarianEffects ofthe Conflict ....... Box 1.3 The Economic Costs ofthe Civil Conflict Box 1.4 Economyof the Norht and East Box 2.1 Summary of SelectedReforms, 2002-03 ...................................................................... 25 Box 2.2 The FiscalManagement(Responsibility) Act ....................,.,.,......................................26 Box 2.3 Financial Burdenof SOEs............................................................................................. 27 Box 2.4 Box 3.1 Box 3.2 US Sri Lanka FTA .......................................... 43 Box 3.3 Box 3.4 The Board of Investment ................................................................. 47 Box 3.5 Summaryof Key Govern Box 3.6 Rice inthe Sri Lanka Rural Economy........................................................................... 51 Box 3.7 Long Term GenerationExpansionPlan......................................................................... 56 Box 3.8 Realizing the PeaceDividend ......... .................................... .........61 Listof Figures Figure 1.1 SocialDevelopmentIndicators; ComparativeRankings................................I................2 Figure 1.2 Trade Indices 1950-2000(1990=100) ............................................................................. 4 Figure 1.3 Trends inExport Share (1950-2000)............................................................ Figure 1.4 Trade Balance, CurrentAccount Balance& Remittanceinflow as percent Figure 1.5 Gross Domestic Saving, GrossNational Saving & Investmentas percento Figure 1.6 Tourist Arrivals & Foreign Direct Investment.......... Figure 1.7 Evolution of Public Debt and FiscalDeficit.................................................................. 12 Figure 1.8 RoadDensity by Province, 2002................................................................................... 16 Figure 1.9 Rehabilitation Funding Requirementsfor Provincial and Local Roads, 2004-2008.....16 Sri Lanka Development Policy Review Figure 1.10 Comparisonof Electricity Tariffs of ASEAN Countries............................................... 17 Figure 1.11 Percentageof Electrification by Region....................................................................... 18 Figure 1.12 Poverty Headcountin Provinces: 1995-96.................................................................... 22 Figure 1.13 Growth-Inequality Decompositionbetween90-91 and 2002........................................ 23 Figure 2.1 Real Growth of Sri LankaExports and World Trade 31 Figure 3.1 InterestPayment to Tax Ratio andthe Nominal Interest Rate on Debt......................... 32 Figure3.2 Primary Balance and Interest PaymentsConsistent with FMRA Targets ..................... 34 Figure3.3 InterestPaymentsas a Percentageof Total RevenuesReturnto Pre-2002Policiesvs. Full Compliancewith FMRA ........................................................................................ 36 Figure 3.4 Recent Steps Towards a Low Tariff Regime................................................................. 40 Sri Lanka Development Policy Review ACKNOWLEDGEMENTS This report was prepared by Shankar Acharya (External Consultant) and Rocio Castro with advice from Ijaz Nabi. The peer reviewers were Farmkh Iqbal , Martin Rama, and Paul Isennman (OECD). The authors acknowledge the contributions from Prema-Chandra Athukorala (External Consultant) on external sector policy issues and the World Bank team who collaborated in the preparation o f the report including Dina Umali-Deininger on agriculture and rural development; Ismail Radwan on peace, conflict and development; Tara Vishwanath, Ambar Narayan and Nobuo Yoshida on poverty trends and welfare reforms; Isabel Chatterton and Amali Rajapakse on infrastructure; Mario Cuevas on fiscal sustainability and growth trends; Cresenta Fernando on recent policy reforms; and Harsha Aturupane on education. Valuable inputs were providedby Shantayanan Devarajan, Princess Ventura, Sriyani Hulugalle, Nihal Fernando, and Terrence Abeysekera. Oxana Bricha assisted with the production and formatting o f the report with the help o f Malathi Ratnayaka. Chulanganie D e Silva assisted with the cover design and printingcoordination o f the version published in Sri Lanka. EXECUTIVE SUMMARY 1. This report provides an integrated view of Sri Lanka's long term development challenges for sustainable growth and poverty reduction. The report, which draws on existing analytical work, i s intended to have some durability and i s designed to contribute to the national debate on these issues. Similar reports have been completed for India, Pakistan, Bangladesh, andNepal. I. DevelopmentOutcomesandUnderlyingCauses Social and EconomicOutcomes 2. Sri Lanka's substantial achievements in human development are well known. In several dimensions-such as universal primary enrollment, gender equality, infant and maternal mortality-the country i s well positioned to meet the MillenniumDevelopment Goals (MDGs). In addition, housing conditions have substantially improved relative to the early 1980s, in particular with respect to housing materials and access to electricity, safe water and sanitation facilities. Despite improved overall access to basic services, however, large disparities remain inthe access to and quality o f most o f these services. SelectedMillenniumDevelopmentGoals I Sri Lanka'sPosition Enroll all children inprimary school by 2015 I N e t primary enrollment currently at 96 percent 0 Eliminating gender disparity inprimary and Gender equality achieved at all levels of secondary schools by 2015 education Reduce infant andchild mortality ratesby two Between 1975 and2001, infant mortality thirdsbetween 1990 and2015 fell from 45 to 12 per 1,000 live births; child mortality fell from 100 to 17 Reduce maternal mortality ratesby three Maternal mortality (23 per 100,000 live quarters between 1990 and 2015 births),onpar with middle-income countries 0 Reducepoverty incidence by halfbetween 0 Poverty fell from 26.1 % 1990-91 to 1990 and 2015 22.7% 2002 3. O f particular concern i s the fact that povertyreduction has been slow while income inequalityhas risen in recent years. At 22.7 percent, the national poverty headcount ratio remains high for a country with US$ 900 per capita GDP.' Furthermore, the rate o f decline in this ratio has been modest despite sustained per capita annual GDP growth o f over 3 percent per year over the last two decades. Between 1990-91 and 2002 the national poverty headcount ratio fell by only 3 percentage points. This modest decline underlies sharply unequal poverty trends across sectors and regions. Poverty incidence in urban areas was halved while rural poverty ratios declinedby less than 5 percentage points and poverty in the estate sector actually increased by 50 percent. Similarly, differences in poverty ratios across provinces have been pronounced: the poverty headcount ratio was 11 percent in the Western Province and around 35 percent in Sabaragamuwa and Uva. PovertyTrends 1990-91 to 2002 (YO) Per Capita Growthby EconomicSector (YO) 1990-91 1995-96 2002 1991-96 1996-02 1990-02 Urban 16.3 14.0 7.9 Agriculture -0.0 0.4 0.2 Rural 29.4 30.9 24.7 Industry 5.7 3.0 4.1 Estate 20.5 38.4 30.0 Services 4.4 3.3 3.8 I National 26.1 28.8 22.7 I I GDP 3.9 2.5 3.2 I Source: Departmentof Census Statistics and Central BankAnnual Reports 'ExcludesNorth andEast. i 4. Sri Lanka's highly uneven poverty record over the period reflects several factors, including a stagnant agriculture and uneven development across regions (with a heavy bias in favor o f Colombo and neighboring districts). Notably, agncultural value added barely increased in real per-capita terms while that in services and industry grew by over 3 percent annually. Moreover, the share o f the Western province innominal GDP increased from 40 to 48 percentbetween 1990 and 2002. I GDP Per CapitaandPowrtyHeadcount,2002 GDP Shares by Province Westem 40.2 43.7 48.1 N.Westem 1 1 . 1 11.3 I O I Central 12.1 10.0 9.4 Southem 9.5 9.0 9 1 Sabaragamuwa 8.1 9.0 6.9 Eastem 4.2 4.8 4.9 Uva 8.1 5.1 4.3 N. Central 4.8 4.6 3.9 Nonhem 4.4 2.4 2.6 Western Southern Uva Central Sabragamma North North Central National GDP 100.0 100.0 IO00 Western S u m !:?O -95 h i z ~ t o lnk p i o i \ a : x a l P l m x g Drpcncci -Poverty Headcount tGDP Per Capit 5. The disappointing trend in national poverty incidence also reflects a long-term growth performance significantly below the country's potential. Despite its early lead in social development, Sri Lanka has been left far behind the high-performing East Asian countries like Korea, Malaysia, and Thailand. This can be largely attributed to: first, the pursuit (from the mid-50s to the mid-70s) o f policies favoring widespread public sector controls over economic activity and resistance to intemational market forces, while the East Asian comparators pursued policies liberalization; and second, the 20-year civil conflict which afflicted the country since 1983. The conflict retarded economic performance in two ways: (i) directly by reducing economic growth (by an estimate o f 2-3 percent annually) and the average income per capita (by about 40 percent); and (ii) indirectlyby divertingpublic resources and the attention of policy makers away from economic priorities and reforms. 6. The core theme o f this report i s thus the following: Sri Lanka must achieve a higher growth rate, but do so in a manner that poor people and the poor regions of the country can more fully participate in this growth. EarlyReformsand their Impact 7. Following inward-looking policies during the 1960s and part o f the 1970s, Sri Lanka liberalized its trade and investment regime -including the establishment o f Free Trade Zones (FTZs)-beginning with the `first wave' o f reforms in the late1970s. These reforms unshackled the economy from stringent controls and wrought a transformation in the country's trade and industrial structure-fiom land- intensive, plantation exports to labor-intensive manufacturing.2 The impact o f the early reforms was amplified by a number o f favorable external factors, including: (i) the massive foreign assistance in support o f the reforms; which, in turn, boosted the confidence o f foreign and domestic investors; (ii) the institution o f the quota system under the Multi-Fiber Arrangement (MFA) by industrialized countries which helped Sri Lanka's fledgling garment export industry to take-off (and grow from nil to US$2 billion in two decades); and (iii) upswing in world tourist trends which led to a doubling o f tourist the arrivals between 1978 and 1982. * Th e transition to a more liberalizedregime was painful for several import-substituting f i r m s which were wiped out in the process,without appropriate safety nets. 11 8. Sri Lanka suffereda devastatingblow in 1983 when the civil conflict broke out. It i s remarkable, however, that despite a raging armed conflict, Sri Lanka continued to register economic growth rates o f 4- 5 percent per annum. This resilience i s owed largely to the continuity o f the liberalization reforms initiated inthe late 1970s. It also shows the highpotential o f the economy if the risk o f the conflict were eliminated and hrther reforms to liberalize the economy were to be implemented. 9. The success o f liberalizationpolicies over time won the support o f the two main political parties -the United National Party (UNP)and the Sri Lanka Freedom Party (SLFP)-to a more open, outward- looking economy with a growing private sector. Thus, when the SLFP-led coalition replaced the UNP government in 1994, there was no significant change in the broad direction o f economic policies. Indeed, privatization efforts were intensifiedduringthis period. 10. However, despite the increased role o f the private sector, little progress was made in fundamentally redefining the role o f the state, partfy owing to Sri Lanka's strong attachment to a large public sector. The state continued to dominate important economic activities (public utilities, transport, and financial sector), it maintainedsignificant controls on key agncultural commodity and factor markets (agncultural input subsidies, land, labor, and financial services) and remained, in some ways, the employer o f first resort (absorbing about 14 percent of the labor force). At the same time, the budget continued to reflect a preference for public consumption (including subsidies) over much needed investment ineconomic infrastructure. 11. An important message o f this report is that much o f Sri Lanka's skewed growth record and ensuing increased income inequality, spanning through alternate governments, i s a reflection o f the unfinished reform agenda. More liberal trade and industrial policies spurred higher growth in industry and services, while pervasive controls and weak infrastructure stifled the dynamism o f agriculture. A corollary o f this uneven economic management also i s that income growth has become heavily skewed in favor o f Colombo and neighboring districts while poverty persists inrural areas where 90 percent o f the poor live. The economy's development was also constrained by the severe war damage to the economy o f the North East and the negative impact o f conflict on foreign investment and tourism. Importantly, the conflict contributed to growing fiscal deficits and rising interest payments, declining public investment (and maintenance) in roads and power infrastructure, falling quality o f education, and weakening o f administration and governance as the state system diverted its resources and energies to coping with the ongoing civil conflict. 11. Cease-Fire andUnfinishedReforms 12. In2001 the Sri Lankan economy was hit by a number of adverse shocks (escalation of the conflict, floods and droughts, a major attack on the international airport by the LiberationTigers o f Tamil Eelm (LTTE), and declining demand for Sri Lanka's exports following September 11). These factors, combined with the country's structural weaknesses, brought a sharp deterioration in the macroeconomic situation, includinga 1.4 percent contraction inreal GDP-the first since Independence in 1948. Against this backdrop, a new government took over in December 2001. The cease fire agreement, which was swiftly negotiated inFebruary 2002 and still holds today, paved the way for economic recovery and the rapid implementation o f wide-ranging reforms encompassing fiscal consolidation, tax system, labor laws, land markets, banlung, electric power, privatization, public sector reform and the welfare system. Many o f these were second generation reforms that built on reforms initiated and developed under the preceding administration. Some of the early gains, mostly o f a macroeconomic nature, from implementing these reforms and the continuation o f the cease fire include: 0 Economic growth resumedwith real GDP growth reaching 4 percent in2002 and 5.9 percent in2003. Growth in2003 was broadbased as the textile-led industrial sector recovered from ... 111 the 2001 downturn, agriculture benefited fi-om good weather, and economic activities in the North East resumed. Notably, tourist arrivals hit a historicalrecordo f half a million. 0 The inflation rate fell from 14.2 percent in 2001 to 6.3 percent in2003; nominal interest rates were halved; and private credit increased. Unemployment moderated from an average o f 8.8 percent in 2002 to 8.1 percent in 2003, although it remained above the 7.9 percent recorded in2001.3 0 Foreign direct investment, which had slumped to US$ 82 million in 2001, rebounded to US$ 242 million in 2002, but having reachedUS$ 170 million infirst half o f 2003 closed the year at US$ 229 million, amidst increased political uncertainties. 13. While noteworthy and welcome, these gains remain fragile as the peace and reform agendas are still unfinished. On the peace front, while the cease fire i s holding, negotiations have been stalled since April 2003. Inthe economic arena, initial gains in fiscal consolidation-which contributed to declining inflation and interest rates-have been eroded as a result o f weak revenue performance and mounting fiscal pressures that began in the run-up to the April 2004 Parliamentary elections (including large subsidies on petroleum and fertilizer against rising world prices). Moreover, most o f the structural reforms (much o f them legislative) that were launched during the period have yet to be implemented. Against this backdrop, it i s rather disquieting that aggregate investment continued to be at a level (22 percent o f GDP in 2003) significantly below the levels seen in the late 1990s (27-28 percent). This happens at a time when export growth i s showing signs o f deceleration, raising concerns as to the durability o f the sector as an engine o f growth. 14. It is also noteworthy that the gains made over the period did little to impress important segments o f the population, particularly the rural poor, and the last elections turned power over to the United People's Freedom Alliance (UPFA)-comprising the SLFP, the Janatha Vimuktthi Peramuna (JVP), and other political parties. While the details o f the new government's economic reform program are still being worked out, it is clear fi-om their manifesto that there will be areas o f continuity as well as of departure from previous regimes. Notably, the UPFA will rely on restructuring rather than privatization as a means o f improving the performance o f remaining strategic state-owned enterprises. The new government has also pledged to give greater emphasis to addressing regional inequalities and fostering rural development. The sections that follow address some o f these difficult challenges andraise a number o f key issues that would need to be considered in defining a more pro-poor growth strategy for Sri Lanka. It is noted from the outset that such a strategy must combine investment, i.e., in the transport and power sectors, as well as policy reforms to increase farmers' incentives and rural productivity. 111. Towards Faster andMoreEquitableGrowth 15. Sri Lanka's recent poverty trends underscore the need for faster and more broad-based economic growth. This will be a challenge in light o f the country's structural weaknesses, including: high fiscal deficits and public debt that constrain private sector development, limited export diversification (against the backdrop o f the imminent abolition o f the MFA from January 2005), continued stagnation o f agnculture, serious infrastructure bottlenecks and declining quality o f human capital due to falling quality o f education. 16. Addressing the structural weaknesses will require containing the burgeoning fiscal deficits, diversifying the export base, realizing the growth potential o f non-plantation agriculture, strengthening the road and power infrastructure and revamping the education system. To do all will require puttingin place a more conducive public investment program and encouraging foreign and domestic investment. In 3 Excludingthe NorthEast iv addition, critical for both growth acceleration and poverty reduction will be the success in attaining a durable peace settlement. It will be also essential that in managing the economy, policy reversals be avoided and that gains from past reform efforts be sustained and hrther strengthened. Managing Public Finances 17. A sustainable fiscal stance remains the cornerstone o f any viable growth strategy. Despite progress in fiscal consolidation during 2002-03 and the 2002 enactment o f the Fiscal Management Responsibility Act (FMRA),the fiscal situation remains under substantial stress. In2003 the fiscal deficit still exceeded 8 percent of GDP and the public debt stood at over 105 percent o f GDP, a long way from the FMRA targets for 2006 (of 5 and 85 percent o f GDP, respectively). The extent o f the problem i s illustrated by the fact that interest payments, defense, and civilian wages account for almost all o f tax revenues (13 percent o f GDP in2003)-which have been decreasing since 1990. Key priorities include: 18. Reducing public debt to manageable levels. Given the heavy public debt burden- absorbing over half o f tax revenue-it i s imperative to bring the public debt to manageable levels by at least eliminating the primary deficit and by limitingnon-concessionalborrowing Civil Service Staffing inSelected Countries(per 100 population) Non-Central Total Country and Region Gov't. Civilian Emp'oyees Employees Gov't.* South Asia Region India 0.4 0.4 1.2 Bangladesh 0.4 0.6 Pakistan 0.4 0.6 1.5 S r i Lanka 2.3 1.6 3.9 East Asia China 0.1 1.6 2.8 Indonesia 0.7 0.3 2.1 Korea 0.6 0.7 2.2 Malaysia 2.3 1.1 4.5 Source: World Bank (1998), ResearchWorking Paper "Govemment Employment and Pay: The Global Regional Evidence."*Total civilian government includesemployeesfrom education andhealth sectors. 19. Reforming wage and recruitmentpolicies. Sri Lanka has one o f the largest bureaucracies inthe region, with a ratio o f 3.9 civil servants per 100 population. Although the size o f the (civilian) wage bill i s not unmanageable at present (around 3 percent o f GDP), the trends are worrisome. Between 1990 and 2001, public sector employment grew at 3.6 percent annually, outpacing growth inpopulation and labor force. While keepingthe wage bill in check, strong political commitment will be needed to address well- known constraints to public service delivery. These include overstaffing (particularly at the lower grades); excessive salary compression (8: 1) which limits the public sector's ability to attract skilled stafc administrative fragmentation, duplication, and wastage (partly exacerbated by the ineffective devolution o f functions); and outdated processes andprocedures. 20. Rationalizing public spending and linking it to the poverty reduction strategy. The scope for expenditure rationalization i s significant, not least because o f the need to address the duplication and overstaffing problems o f the public administration. In addition, there i s a serious imbalance in public spending, with the bulk of the budget being directed to fund recurrent costs (i.e., interest payments, wages, and subsidies) and very little to investment. The Ministry o f Finance i s developing a medium term budget framework (MTBF) which could potentially combine macro (i.e., attaining fiscal sustainability) and micro objectives (such as enhancing the development impact o f public spending). Following the recent introduction o f budget ceilings and o f a three-year planning horizon, there i s a need V to strengthen the links between priorities, resources, and outputs/outcomes. This will require a more strategic and consultative budget formulation process and the identification o f trade-offs within and across sectors. For instance, there i s an urgent need to increase the effectiveness o f Sri Lanka's welfare system, notably the Sumurdhi program which while covering almost half o f the population, misses a significant proportion o f the Improving the targeting o f existing welfare programs to better reach the poor, through objective eligibility criteria, and making transfers more progressive would go a long way toward improving the lives o f the poor. 21. Improving theperformance of state-owned enterprises (SOEs). The financial burden o f SOEs5 i s high and service delivery i s poor. Most are overstaffed, incur operating loses, and have large debts (mostly to the state-owned banks). Direct subsidies alone absorb 3 percent o f GDP annually, which i s above the entire education budget. A program o f fiscal consolidation and improved service delivery requires that these companies be substantially restructured (through privatization or other methods), be allowed to operate on a commercial basis with no political interference, and be subject to a hard budget constraint. The success o f current plans to restructure twelve strategic SOEs, under the oversight o f the recently established Strategic Enterprise Management Agency (SEMA), hinges on improving their autonomy and accountability. 22. Efforts need to focus on reversingthe decline inrevenue by strengthening tax administration and expanding the tax base for major taxes (VAT and income tax). Greater realism in projecting future revenue and a more stable tax policy framework are also essential to support fiscal consolidation. 23. Tax policy for growth and fiscal consolidation.6 The challenge in designing tax policies to promote growth and fiscal consolidation i s to reverse the massive decline in the tax-to-GDP ratio (from 19 percent in 1990 to 13 percent in 2003) and to do so in a manner which i s supportive of growth and efficiency. The challenge i s daunting given the sharp drop in trade taxes (from 6 to 2 percent o f GDP since 1990); the stagnation o f income taxes at around 2.5 percent o f GDP, which i s very weak by international standards; and the incomplete transition from a complex system o f tumover taxes and special levies to a Value Added Tax (VAT) which has recently lost at least one percent o f GDP ( to 5.4 percent in2003).' Increasedrevenue inthe order of 2-3 percent o f GDP may be achievable through: I Tax Revenues (As % of GDP) 20 1 15 12 I lo ~1985'1990~1995~1996~1997'1998'1999'2000'2001'2002 2003'20042005'2006' 1 .. R.' (Budget) " , -Actual Tax Revenues Estimated Tax Revenues 1 Source: CentralBank of Sri Lankaand staff estimates. Specifically, Samurdhimisses40 percentof householdsinthe two poorest quintiles,while almost 44 of its budget is spent on householdsfromthe top three quintiles. 'Includespublic institutionsand corporations. As this report was beingcompleted, the 2005 Budgetwas presentedincludinga number ofrevenuemeasureswhich are not discussedhere. 'TheVAT introducedin2002 to replacethe Goodsand ServicesTax (GST) and the National SecurityLevyconsistedoftwo rates(10 and 20 percent). v1 '3 Strengthening VAT. While the 2004 unification of the two VAT rates into a single rate o f 15 percent facilitated administration and reduced leakages, its immediate impact was revenue- reducing. (The introduction o f a three-tier VAT system in the 2005 Budget will likely complicate revenue collection, but its overall impact i s yet to be seen). Going forward, additional revenue will be possible if the VAT coverage i s extended to retail trade and exemptions are reduced. '3 Raising the income tax yield. Underlying reasons for low collection include: the provision o f long term tax holidays for the rapidly growing Board o f Investment (BOI) sector, a long history o f investment allowances, concessionary rates, generous depreciation, and exclusion of most public sector employees from personal income tax. The scope for substantially raising the income tax yield is limited inthe short term, but significant gains could be made if a moratorium on tax holidays were to be gradually phased in and existing exclusions and concessions are pared down. 9 Excise taxes. Onbothrevenue raising and equity grounds, there i s a good case for expanding the coverage o f excise taxation to a number o f income-elastic, luxury consumer products (e.g., air conditioners, refrigerators, washing machines and televisions). 9 Improving tax administration. For several reasons, including the coexistence o f parallel systems (for customs and BO1firms), Sri Lanka's tax administration has limited experience inadministering moderntaxes (e.g., VAT andincome tax) that require verification, audit and risk monitoring. A well integrated revenue administration, along the lines o f the originally envisaged Revenue Authority, needs to be established as soon as possible. Complementary measures include the establishment o f a separate tax audit unit and malung the large tax payer unitmoreproactive. Policies for Faster Export Growth 24. As noted earlier, liberal external sector policies (trade, foreign investment, and emigration) have underpinned Sri Lanka's export-led strategy and explain the resilience o f the economy in the face of prolonged stress and occasional shocks. Inparticular, these policies have contributed to bringmg down the unemployment rate to single digits (from close to 20 percent inthe 1970s). Notably, talung advantage of the opportunities opened by the MFA and a largely liberal trade regime, garment exports have become one o f Sri Lanka's major industries, currently accounting for half of all exports and employng 5-6 percent o f the labor force (mostly female) while providing indirect employment to about one million workers (15 percent of the labor force). Similarly, liberal emigration policies have contributed enormously to the generation o f foreign exchange (about 10 percent o f GDP) and of employment opportunities for about one millionworkers (also mostly female), helping to more than halve the female unemployment (from over 30 percent inthe 1970s). 25. Although the medium term prospects for Sri Lanka's garment industryinthe post-MFAera seem positive-as larger firms have been able to move up to higher value segments o f the market-there could be significant disruptions among smaller exporters who account for about two-thirds o f all exporters and around 35 percent o f employment in the sector.8 There is, therefore, an urgent need to strengthen Sri Lanka's ability to compete in a changing world environment, especially after the MFA ends (January 2005), by improving existing external sector policies and addressing remaining `behind the border' * Whilesmaller firms account for 5 percent oftotal export value, it is speculatedthat about 10percentofjobs (or about 30,000) might belost. vii constraints that undermine the competitiveness o f Sri Lankan firms. Some key priority issues are discussed below: 26. Avoiding tradepolicy reversals. The pursuit o f a competitive real exchange rate, which has been central to Sri Lanka's export-led strategy to date, will need to be continued. In addition, further efforts are needed to meet the medium term goal o f a single low, uniform tariff. While the country has low average tariffs, modest effective rates o f protection, and a declining incentive bias against exports, the tariff regime has become less tidy since 2001 due to a new tariff surcharge, reintroduction o f specific duties for agncultural products, a new tariff band o f minimumduty, some proliferation o f ad-hoc duty exemptions, case-by-case duty adjustments and the introduction o fpreferential tariffs under several trade agreements. In the short run, significant progress towards a low, uniform tariff can be achieved by increasing the minimum rate to 5 percent, eliminating ad hoc exemptions and rates, and gradually reducing agricultural tariffs. Key Featuresofthe Sri LankanGarment Industry Strengths Weaknesses - good reputation for quality - reliability indelivery --highleadtimes poor road and transport infrastructure ------- low wages - low labor productivity highcompliance withintemationallabor norms - highly restrictivelabor laws duty fkeeprovisionfor import ofyarnandfabrics prestigious customer base ----- limitedlocalavailability ofraw materials narrow export market, mostly US and EU close proximity to India for fabrics unreliable andhigh cost o fpower competitive local exchangerate high cost financing especially for SMEs low clearance charges lengthy clearancetimes 27. Preferential trade liberalization. Sri Lanka has concluded the India Sri Lanka Free Trade Agreement (FTA) and has initiated FTA negotiations with the U S and several other countries. While there are substantial economic and political benefits to be reaped from FTAs with India and the US, a proliferation o f suchbilateral agreements could distract from Sri Lanka's successful program o f unilateral trade liberalization, incur sizable administrative and economic costs o f complex rules o f origin o fmultiple FTAs, and complicate the tariff structure. In the post-MFA era, an FTA with the US would help Sri Lankan garment exports to compete since half o f U S imports o f these products occur under various preferential arrangements. These benefits would have to be balanced against the costs o f possible requirements to free the capital account and to source inputs from the US. It i s also important to note that preferential access to markets cannot substitute for needed improvements inlabor productivity, lead time, and cost effectiveness. 28. Sustaining a liberal emigration policy. With workers remittances accounting for about 10 percent o f GDP, Sri Lanka has reaped enormous benefits from following liberal labor emigration policies. Aside from its contribution in strengthening the external position, labor migration has played a key role inraising the incomes o fthe poor, as migrant workers (mainly female) come primarily from low-income families. These positive outcomes have been greatly helped by easy access to bank facilities and a market-responsive exchange rate policy, which needto continue. 29. Refocusing investment policy. A liberal investment regme has effectively complemented Sri Lanka's successful export-oriented policies, with foreign direct investment (FDI) driving the growth o f the garment sector and other labor-intensive activities. However, a major policy issue relates to the provision o f generous direct tax incentives to companies operating under the BOI. Inaddition to having an adverse fiscal impact, these incentives are hghly variable (across products, sectors and scale o f investment) and subject to substantial discretion. Also, contrary to original plans, many firms have become perpetual beneficiaries o f tax holidays. Despite these generous incentives, however, the overall ... V l l l level o f foreign direct investment (FDI) to Sri Lanka has been modest (averaging less than one percent o f GDP annually) by regional and international standards. Recent legislation to h n g fiscal incentives into line with InlandRevenue laws and away from the BO1needs to be implemented. Inaddition, given fiscal considerations, some form o f phasing in o f a moratorium on direct tax incentives needs to be seriously considered. 30. Inadditiontothe conflict, several `behindtheborder' constraints to investment may account for low FDIin Sri Lanka. Indeed, according to a recent investment climate assessment (ICA), top constraints include poor infrastructure (electricity and transport), cost of and access to finance, and rigid labor regulations (for urban firms).g There is therefore a need to refocus investment policy towards removing these constraints while reducing excessive reliance on overly generous tax incentives. Inparticular, to assist garment firms in adjusting to a more competitive environment in the post-MFA period, policies should encourage consolidation, facilitate restructuring o f credit to small exporters, improve infrastructure and trade facilitation, and follow throughwith labor reformsthat would facilitate enterprise restructuring. Regulations for Labor Redundancy for SelectedAsian Countries $itat~tt~ry e ~ ~ ~ ~ d a ~ e ~ Payment Per Year of service Sri Lanka Yes, ifover 15 enqtloyees Not fixed, ease by case India Yes, ifover 100eiiqtioyees. Not I S days applicable to managerial and a d I n ~ i s ~ r employees, a ~ ~ ~ e Palcistan Yes, ifclosing dotm or retrenching 20 days Inorethanhalfo f workers Malaysia No 10-20days Thailand N O Approximately 30 days; cappedat 180 days total Vietnam K O 2weeks Source: World Bank (October 2004) InvestmentClimateAssessment. Reviving Non-Plantation Agriculture" 31. The slow growth o f non-plantation agnculture i s a primary factor explaining the persistence o f poverty in Sri Lanka. The modest decline in rural poverty between 1990 and 2002, i s consistent with a barely positive growth trend in per capita agricultural value added over the period. Heavy public sector interventions inagricultural commodity and inpuvfactor markets-primarily geared toward the attainment o f self-sufficiency inpaddy production-have hampered growth in the sector. Although many of these interventions (e.g., subsidies, land, technology, trade, irrigation policies) have sought to protect the interest o f paddy farmers, their unintended outcome has been to squeeze the returns from agricultural production, limit productivity and income-enhancing investments, hold back diversification to higher value activities, and"pushed" many into low-paying, casual non-agricultural wage labor. 32. More rapid agricultural productivity growth and rural development will require stronger commitment to removing existing policy and regulatory constraints which stand in the way of a more diversifiedproductionbase. Some o fthe issues that the report discusses are: 'See Sri Lanka: Improving the Rural and Urban Investment Climate (Draft October 2004, to bepublished),jointly preparedby the World Bank andthe Asian DevelopmentBank. lo Issuesrelated to plantationagriculture were beyond the scope ofthis report. However, inlight o fthe poverty statistics for the plantation sector, a special study o f the issues is being launched. i x Average Annual UDB growth rmta (%) Rubber -1.2 -2.3 Source: Central Bankof Sri Lanka and staff estimates. 33. Allowing full and transferable ownership rights to land. The transfer of state-owned land to smallholder farmers through grants and permits, initiated in the 1930s, succeeded in promoting greater equity inland distributionand food security. However, restrictions on the transferability and use o fthese grantdpermits have led to the (informal) fragmentation o f land into non-economical `miniature' holdings," with adverse equity and efficiency implications. The conversion o f grandpermits into title and the ensuing development o f land markets, i s essential to enhance farmers' ability to expand and consolidate land into economically viable holdings as well as to access credit (using land as a collateral) for much needed productivity-enhancing investments. Notably, lack o f access to credit (which i s linked to the lack of land collateral) i s regarded as a top constraint for the development o f rural firms. To address concerns that many f m e r s may become landless if private land ownership is allowed, considerationmightbe givento developing appropriate safety nets and impact assessments. 34. Facilitating farmer access to improved technologies. Currently farmers are faced with restrictive seed and phyto-sanitary regulations and limited extension"and research services (which have primarily focused on paddy for the domestic market). Farmers' access to improved technologies could be facilitated by adopting more liberal seed and phyto-sanitary regulations, as envisaged under National Seed Policy and 1999 Plantation Act, as well as by expanding the scope o fresearch and the supply o f extension services through increasedprivate sector participation, including o fNGOs. 35. Phasing out trade distortions. Hightariff protection is still gven to selected commodities (i.e., rice, potatoes, chillies and onions) and tariffs are frequently changed to off-set domestic market fluctuations. These policies lead to price distortions and create considerable uncertainty for farmers, consumers and local firms, discouraging private investment (e.g., in storage). The correction o f ensuing inefficiencies in the allocation o f resources will require a more stable trade regime and a gradual reduction intariffprotection. 36. Ensuring the sustainable use of water. Substantial public resources have been invested in imgation infrastructure (including the Mahaweli program launched in the early 198Os).l3 However, the sustainability o f large past investments i s threatened by the lack o f maintenance, leading to the rapid deterioration o f canal systems and costly rehabilitation requirements as well as poor quality of services. Also, the provision of free water has reduced the incentive for farmers to save and use water more efficiently. Building consensus on a national water policy-which effectively addresses maintenance, rehabilitation, and water management issues to deal with the growing competition for water-will be essential for the sustainable development and management o f water resources. This will need to be supported by suitable legal andinstitutional arrangements. Over 40 percent of small holdings have less than one acre and over 60 percent below two acres. 12According to the Sri Lanka Integrated Survey (SLIS),199912000, only about 13 percent of agricultural householdsreported receiving technical assistance from a govemment extension agent (15 percent from all sources). l3Between 1980 and 1997, about Rs. 215 billion (constant 1996rupees) in public funds were spent in irrigation. X 37. The removal o f policy and regulatory constraints will need to be accompanied by increased investments in basic rural infrastructure. According to the rural ICA, transport constitutes the single most important constraint to rural firms, followed by cost and access to credit and elect~icity.'~ Given scarce public resources, public spending needs to focus on designing andpromoting a regionally equitable strategy for rural infrastructure development; fimding maintenance and rehabilitation that cannot be covered by communities (which are likely to yield hgher returns than input subsidies, for example); and encouraging private sector participation and investments. The following section deals with some o f these issues. Infrastructure Priorities 38. For rapidly approaching middle income level countries like Sri Lanka, reliable infrastructure i s paramount for further economic development, especially o f export-oriented activities including manufacturing and tourism. These activities require reliable power supply at internationally competitive prices, adequate port and air facilities, world-class telecommunications capability, and a transport system that works in a seamless fashion. Although Sri Lanka has made progress in some o f these areas, substantial challenges remain to bringthem up to world standards and to reduce inequities inthe access and quality o f these services, particularly inthe provision o froads and electricity. As mentioned earlier, transport and electricity represent a barrier to entry and a significant operational costs to both urban and rural firms. 39. Roads. Although Sri Lank's road density (1.4k"2) i s relatively highby regonal standards, only 10 percent i s in good condition because o f decades o f neglecting maintenance. Funding is inadequate, with public spending in maintenance and rehabilitation covering only 10-15 percent o f estimatedrequirements. Inaddition, few investments have been made over the past 40 years inupgrading trunk roads and constructing new highways to provide rapid access to urban centers. This situation results in high costs for road users, high unit cost for road provision, low road utilization and low road safety compared to other countries inthe region. Moreover, there exist deep disparities inroad density and road condition across provinces, with the Western Province showing the best indicators and conflict- affected provinces in the North and East showing the worst. To address these inadequacies, measures needto be taken to: 40. Increase the efficiency of road utilization. Given the huge maintenance backlog, funding for road maintenance and rehabilitation (estimated at US$870 million over the 2004-08 period) needs to be substantially increased, with a focus on improving the national highways which carry 70 percent o f road traffic. Inparticular, a functioning Road Fund to cover maintenance expenditures should be in place, financed with petroleum taxes. There i s also a need to strengthen the capacity o f public sector agencies responsible for formulating and implementingtransport policies. 41. Improve the economic andfinancial sustainability of the road network. Inaddition to carrying out timely road maintenance, this could be achieved by increasing private sector participationintransport infrastructure and service provision; and making public sector agencies more market sensitive through restructuring pricing and financing systems. 42. Improve equity and environmental sustainability. Increased efforts are needed to integrate the needs o f the poor in transport strategy and policy setting; increase participation o f stakeholders (particularly the rural poor) in the planning and implementation o f investments; and improve the management o f congestion. l4World Bank (2004), "Sri Lanka: Improvingthe Ruraland UrbanInvestment Climate" (to bepublished). xi 43. Electricity. While access to electricity has increased dramatically since the early 1980s (from 16 percent to 74 percent o f households), generation capacity has not kept up with growing demand. Increased reliance on expensive emergency supply as well as large technical and financial losses by the state-owned public utility, the Ceylon Electricity Board (CEB) have led to relatively high cost of electricity in Sri Lanka relative to other Asian countries. Disparities in access across provinces have likely exacerbated poverty in underserved areas, especially in the North and East where electrification rates are the lowest. Efforts needto concentrate on: 44. Expanding capacity generation. Just keeping up with historical demand will require adding 200MW per year. Therefore, efforts needto focus on expanding generation capacity in accordance with an updated least-cost expansion plan that takes into account the full range o f capacity options, including for private provision. In the long term, the independent transmission company should be allowed to procure new plantswithout political interference. 45. Raising the eficiency of electricity provision. This will require moving forward with the restructuring o f the sector. Key measures include: the unbundling o f CEB into independent companies operating on a commercial basis; the effective operation o f the Public Utilities Commission o f Sri Lanka (PUCSL) as an independentregulatory body; and phasing out cross subsidies to residential consumers. Comparison o f Electricity Tariffs of A S E A N Countries I SriLanka Malayda Singapore maland Philipme8 lndonesa LaOF ResidentiaClow Residential- High 0 Comnercial- low OComrcMl HQh- Industrial- low Industrial Hgh - 1 Source: ASEANwebsite 46. Addressing inequities. Priorities to address inequalities in the provision o f power include: providing off-gnd connections to rural households (e.g., through more effective utilization o f existing facilities such as the IDA-supported Renewable Energy and Rural Economic Project); and granting direct subsidies incases when rural electrificationcannot be provided on a commercialbasis. Educationfor Growth and Equity 47. Sri Lanka has maintained high levels o f educational attainment, although with severe disruption to the school systems in the conflict-affected areas. However, disappointing leaming results point to weaknesses inthe quality of education. Cognitive achievement tests among primary school children, for instance, show substantial shortfalls in mastery of basic language (especially o f English) and numeracy xii Skill Sri Lanka Urban Sector Rural Sector Sinhala or Tamil Language Skills 37 51 34 I Numeracy EnglishLanguage Skills 10 7 Skills I 3 8 I 23 52 I 3 5 Source: National Education Researchand EvaluationCenter. University of Colombo. 48. Increasing cost-effectiveness. Considerable savings could be generated by rationalizing schools, especially o f small uneconomic schools; raising the student-teacher ratio (currently 21:l) at all levels o f education; and rationalizing administration and staffing. 49. Increasing private sector participation. The volume o f resources devoted to education and quality couldbe increasedthroughprivate sector provision, particularly at the tertiary level which tends to benefit students from upper income families. Main recommendations include: developing a new education act and repealing the 1960s legislation which acts as a barrier for the establishment o f private schools; and increasing the understanding among the public at large that public funding o f tertiary education i s regressive and costly, and that international trends are strongly infavor o f private investment intertiary education. 50. Shifiing public resources within the sector. Public spending on tertiary education should be contained in favor of expanding allocations for primary and secondary education. Access by the poor to tertiary education could be secured through a voucher system. Over time, the balance o f recurrent and capital spending needs to shift toward capital investments, andwithin investments toward technology and science equipment rather than construction o fbuildings. 51. Decentralizing education management. International experience shows that the quality o f service delivery could be greatly enhanced by decentralizing education management to schools. This principle could also be applied to other education institutions such as the National Colleges o f Education and Teachers' Centers. 52. Increasing the economic relevance of the education system. To improve the linkage o f education to the world o f work, several initiatives have been adopted, including increasing the use of technology, especially information technology (IT), and allowing the use o f English as a medium of instruction. These measures could be strengthen by reallocating resources toward technology and equipment and permitting schools to offer Englishmedium education from grade 1onwards. IV. Peace, Conflict andDevelopment 53. The cease fire since February 2002 has already brought substantial gains to the Sri Lankan economy. Growth has rebounded, inflation and unemployment have declined, and foreign investment and tourism have recovered. Defense expenditure has fallen, reducing the pressure on public finances. There has been significant recovery in the conflict-depressed economy o f the North East and some reintegration with the rest o f the Sri Lankan economy. Thousands o f war displaced refugees have "The elimination of English as a mediumof instructionin the 1950suntil2001 accounts for poor learningoutcomes in the English language. xi11 ... returnedto their old homes. Much demining o f land has occurred. Above all, the absence o f a shooting war has become the accepted norm. Put another way, the interests vested inpeace have grown. 54. Although these gains are real, they have been constrained by the absence of durable peace o f the kmd entailed in a formal peace accord. Uncertainty still weighs heavy on both domestic and foreign investors. More generally, international experience suggests that ending a long drawn-out civil conflict i s a necessary but not sufficient condition for realizing the full peace dividend. Typically, private investment and fiscal revenue are slow to recover and security expenditures cannot be reduced quickly for a variety o f reasons. While it takes time and resources to restore damaged or neglected economic infrastructure, the damage to institutional infrastructure i s even harder to overcome. The transition to peaceposes fresh problems and challenges that have to be managed. 55. However significant these challenges and costs may be for managing the transition to peace, they are obviously to be preferred over any resumption o f conflict. Sri Lanka's twenty year history o f civil war bears eloquent testimony to the direct social and economic costs of conflict and the possibly even more substantial (though less tangible) indirect costs of missed opportunities and distracted policies. Furthermore, there i s little chance that Sri Lanka could sustain the 5 percent growth o f the 1990s if there i s a resumption o f conflict. Incontrast to the early 1990s, the fiscal situation today is muchmore fragile, infrastructure i s suffering from decades of neglect, agriculture i s inthe doldrums, trade and industry face novel external challenges andexternal assistance has dwindled (and will not increase if conflict flares up). 56. Inbrief, for SriLanka's economyto achievesustainedgrowthat 6percentplus(let alone the 8to 10 percent aspirations sometimes voiced), durable peace i s a prerequisite. Key economic reforms will be also needed to resume the path o f rapid development since peace i s a necessary but not a sufficient condition. The country's economic history supports this assessment: the only five year period when economic growth averaged above 6 percent was in 1977-82, when economic reforms held center stage in policy-making andthe civil war hadnot yet begun. V. DevelopmentRisks 57. Sri Lanka i s at a cross-roads in its development path. As noted above, any resumption o f the conflict poses a major risk to the country's development prospects, especially in a world which since 9/11 is muchmore sensitive to conflict and terrorism. 58. Being a small open economy, Sri Lanka is also vulnerable to external shocks. The best insurance against unforeseen external events and trends lies inbuilding a more diversified export base, a flexible economic structure (with easy conditions for entry and exit), and sound investment in education and skills. The policy reforms suggested in this report should help Sri Lanka strengthen its economic flexibility. 59. Finally, a lingering risk to sustained economic and social development i s the country's legacy o f civil strife and complex politics. It is very clear that economic reforms in Sri Lanka (like everywhere else) are deeply political. The challenge lies inbuildingthe requisite support in favor o f reforms which can weather the instability caused by short election cycles. Sri Lanka's own experience clearly shows that market-fhendly policies have succeeded in bringing substantial economic growth and poverty reduction in urban areas, while heavy state intervention in agriculture has stifled the rural economy with seriously adverse implications for the poor. Therefore, there i s enough evidence inSri Lanka to show that there is no trade-off between market-driven growth and poverty reduction. However, a well functioning market economy i s usually associated with a well functioning public sector. Indeed, the state has a very important role to play on several fronts: in establishing an appropriate regulatory framework; addressing market failures; providingpublic goods and services; andprotectingthe poor. xiv Sri Lanka Development Policy Review Part I.Development Outcomes and IhderlyingCauses 1. Social andEconomicOutcomes I. SriLankaisjustlyfamousforitsearlycommitmenttohumanresourcedevelopment. Itwas 1 one o f the first developing countries to provide free and universal health and education services to its citizens. Together with a commitment to gender equality, social mobilization and an extensive food subsidy, they constituted the pillars o f the early Sri Lankan welfare state, which had strong roots in colonial era laws and programs. As a result o f these policies, Sri Lanka had achieved, by the mid- 1970s, rates o f school enrolment, literacy, infant mortality and life expectancy far ahead o f other low income countries and better than manymiddle income ones (Figure 1.1). 1.2 This early impetus for strong social development policies retained substantial momentum in later decades (despite civil conflict and strained public finances). Thus, in most dimensions, Sri Lanka's performance with respect to the Millennium Development Goals has been very good (Box 1.1). povertyby halfhetween 1990and 2015. *percent behveeii 1990-9Iaid 2002. A m ISPohalve erWi" pclvatyby 2010 - - h a l l all cliiidreii 1x1yriiimary scliool hy 201s. 1he nct priiiiary ciuollnirrir rawctrnmitly stands ai 96 pmcent Make prrsges towards gender equality and Cierida equality achieved in primary, seccnidary, and even cntpowerinring woinen by eliminating pnder disparihes tertiary levels. angmnrnaryandsecondary schoctlby 2015 Reduce irifiiitt and cliild iiirrtaltty ratcrr by iwo tliirds Between the mid 1970sand 2001, inlkntmonatity fell t h t hduecii I990and2915 45 IO I 2 per I.OfHf live births, under-tiretitortality fellf b ~ i 10010 17 Iinptement natioiial strategts for wistaii~able Sitice 1994, two successive national envtronmental action development by XMs lo reverse the loss of plans have been dc\elopd and are presently under ~ V I I orrnienral resouicesby 2019 rmplenimation. 1.3 Moreover, housing conditions and access to household amenities have substantially improved during the last two decades (SA Table A31).' Notably, the share o fhouses with permanent structures has tripled from 26.2 percent in 1980/81 to 75.3 percent in2003/04, access to electricity has increased almost five-fold from around 15.8 percent o f households to 73.9 percent, access to a telephone line has gone up from virtually nil to 23.9 percent, and the proportion o f households with a water seal latrine has more than doubled from 31.9 percent to around 78.4 percent See CBSL, Annual report 2003, Box 1. HaveLivingConditionsImprovedin Sri Lanka?. 1 2 Sri Lanka Development Policy Review Table 1.1: Growth of GDP and Major Economic Sectors 19Sf-t977 1978-2002 IO81-1990 1`)91-20uO 2001 2002 2003 Agrtculture 2 8 2 2 2.9 1.9 -3.4 2.5 -~ Iridustry 4.0 4.6 4.2 6.8 -2.1 I.o -- Services 4.1 5 3 4 7 5 7 -0 5 6.0 -I Note: Colunma~tna simple periodaverages o f annual p ~ rrates (mclustve ofstarhiig and ending yeam) ale h ?%us 1951-1977 is the siniple average of27 year-wise annual gruwthrates Frguresfor 2003 are estimates. Source Central Hank of Sri Ldnka Aiinual Report,(various issues) and SraBcalwIations 1.4 In one critical area, performance has fallen well short of potential-income poverty. Surprisingly, for a country with a per capita GDP o f about US$ 900, the head count ratio of poverty remainsquite highat 22.7 percent. Furthermore, the rate of decline inthis ratio hasbeenmodest and uneven over the last two decades despite sustainedper capita GDP growth o f over 3 percent per year (Table 1.1). Inpart this reflects a skewed distribution of economic growth across sectors andregions. It also results from a long-term growth performance significantly below the country's potential. Despite its early lead insocial development, Sri Lanka has lost out inthe growth stakes compared to the hgh-performing East Asian countries like Korea, Malaysia, and Thailand. Having being at comparable levels o f per capita income inthe mid-1950s (or even 1960), these countries now boast per capita US $ GDP levels some three to ten times higher than Sri Lanka's.2 Nor does the comparison improvewhen conductedinPPPterms (Table 1.2). Table 1.2: Sri Lanka in an Asian Perspective: Per capita GNP relative to the US (Ye) 6 5 7 4 6 0 5 7 8 2 Singapore NA 16.6 24.2 4.13 85 4 Note: NA (Data not Available) Source: National Bureau o fEconomic Research: PennWorld Tables (web page), quotedinAthukorala (2001). 1.5 What constrained Sri Lanka's growth performance in the past? Have the constraints been released? What newproblems have emergedto challenge the country's development prospects?These issues are explored inthe ensuingsection. 'O f course, Sri Lanka's percapita growth o f 3.2 percent since 1980compares favorably with other South Asian countries and many Latin American countries. ButSri Lankan authorities prefer to benchmark their country's growth performance against highgrowth East Asian economies (GoSL, 2003). 3 Sri Lanka Development Policy Review 2. UnderlyingCauses and Trends A. The FirstThreeDecades(1948-77): A Capsule History 1.6 At the time ofIndependencein 1948, Sri Lanka was almost a textbook example of a dualistic export economy with a well-established and relatively modem plantation export sector, coexisting with a traditional, subsistence small-holder agriculture ~ector.~Tea, coconut and rubber generated over 95 percent of the country's export earnings and provided the tax base to sustain the 'welfare state' of free education andhealthand a generous subsidy on the consumer staple, rice. This colonial legacy of a welfarist, plantation economy with open trade andpayments policies continued for nearly a decade. There were ups and downs caused largelyby swings in world demand and prices for Sri Lanka's three major plantationcrops. 1.7 From the mid-1950s to the mid-1970s Sri Lanka suffered a massive and sustained decline in its external commodity terms o f trade, with the index (1990=100) dropping from around 500 to about 150 (Figure 1.2). Over the same period, the export volume index hardly increasedby 20 percent. In common with other South Asian (and indeed, many developing) countries, Sri Lanka embarked on defensive trade and payments controls partly as a technocratic response to the worsening foreign exchange situation and partly influenced by the contemporary development parahgm favoring import-substituting industrialization and strong state interventi~n.~ As the foreign exchange crisis deepened in 1961-64, trade and payments controls intensified. At the same time, industrial controls, establishment of state enterprises and nationalizations gathered steam. Unsurprisingly, economic growth in 1960-65 droppedto averageonly 3.6 percent. Figure 1.2: Trade Indices 1950-2000 (i990-'100) l.ll" ll__._l_..ll._l _I 1.8 Following a change in government, there was a briefand half-hearted attempt at liberalization of trade and industrial controls in 1965-70, along with renewed emphasis on non-plantation agriculture'. As a result, economic growth recovered somewhat to average 5.2 percent. However, the ensuing seven years under another government witnessed intense and extensive policies of state intervention in all economic arenas and the greatest delinking from international commerce in Sri Lanka's history (Figure 1.3). By 1977, plantationlands hadbeen nationalized and limits imposed on landownership, over sixtypercent o fmanufacturing was inthe public sector andaccountedfor halfo f all manufacturing employment (Athukorala and Rajapatirana, 2000). State trading corporations were establishedandnurtured. By 1976public sector entities accountedfor nearly 90 percent o f all imports and 30 percent of exports. Part o f the penchant for avowedly socialist policies was a responseto the first JVP-led youth insurrection o f 1971. That, in turn, may have been fed by the cumulative failure Beginninginthe 1930s, land was transferred to small holder farmers through various land settlement programs. This dirigiste paradigmwas especiallyinappropriate for Sri Lankaas a small island economy. For excellent recent assessments ofthe early period see Kelegama (2000) and Abeyratne and Rodrigo(2002). Cuthbertson and Athukorala (1990). 4 Sri Lnnkn Development Policy Review of the economy to create enough job opportunities for a rapidly growing, young labor force with rising levels o f education, skdls and expectations. This was also the period when Sri Lanka (like all other oil-importing nations) was buffeted by the first `oil shock' o f 1973/4. Despite growing fiscal pressures, food subsidies were continued at anunsustainable4 percent of GDP (Table 1.3). Table 1.3: SocialExpenditureas Per Cent of GNP, 1950/1-1977 0.4 10.6 0.5 7 2 1965 6 4 7 2. I 3.6 tf 6 I t 0 Source: Kelegama(2000) 1.9 By the mid-l970s, the intensified strategy of inward-looking import substitution and increasing state intervention was beset with serious problems. Economic growth had plummetedto average below 3 percent in 1970-77, with per capita growth amounting to a little over 1 percent a year. The unemployment rate had risen above 20 percent (the highest recorded in Sri Lanka's post- independence history), with high incidence among educated youth (Statistical Annex Table AI2). Incomes o f the poorest 40 percent showed little growth (Papanek, 2003). External assistance had dwindled to a trickle. The country's export basket reflected the failure of trade policy to promote meaningfbl diversification, with the three plantation crops still accounting for nearly 80 percent of merchandise exports in 1977. Clearly, the chosen development strategy had led the Sri Lankan economy into something o f a blind alley. The gravity o f the economic difficulties of the mid-1970s easedthe way for the far-reaching reforms that followed. B. Post 1977Reformsandtheir Impact 1.10 The 1977-79 reforms had profound (largely beneficial) effects on the Sri Lankan economy. These reforms (bolstered by the `second wave' inthe first halfof the 1990s) unshackled the economy from stringent controls andwrought a remarkable transformation of the country's trade and industrial structure. The initial reforms included phasing out o f quantitative import controls, compression and reduction o f the import duty structure, opening up to foreign investment, establishment of an efficient free trade zone (FTZ) scheme, freeing o f credit markets, institution o f a market-responsive, unified exchange rate, substantial retreat from government ownership and control of productive enterprises, withdrawal of state tradmg monopolies, lifting o f price controls and the replacement o f the highly subsidized, rice ration system by food stamps targeted to lower income groups. The reforms o f the early 1990s carried forward reform o f trade policies, realigned the exchange rate, freed up exchange control on current transactions, gave a major impetus to privatization and strengthened the policy framework for foreign direct andportfolio investment. 1.11 The far-reaching impact o f these reforms can be readily gaugedby the following summary o f consequences: Economic growth recovered smartly to average 6.2 percent in 1977-82, the only five year period in which growth has averaged above 6 percent in Sri Lanka. This meant a tripling o f the percapita GDP growth experiencedin 1970-77. Manufactured exports (especially garments) took off, growing at 32 percent per annum between 1978 and 1995 incurrent US dollars (Athukorala and Rajapatirana, 2000). The share ofindustrial exports intotal exports increaseddramatically from 14percent in 1977 to 72 percent in 1992andfurther to 78 percent in2000 (Figure 1.3). 5 Sri LankaDevelopment Poky Review The share o f exports in GDP rose steadily from a nadir o f 13percent in 1972 to 34 percent in 2000 (a ratio last attained in 1956). Sri Lanka had once again become an open economy, engaging successfully inworld commerce. The factor content o f exports changed hugely from less than 3 percent in labor-intensive manufactures in 1962-77 to nearly 60 percent in 1990-95 (Athukorala and Rajapatirana, 2000). This reflects a remarkable transformation from land-intensive, plantation exports to labor-intensive manufacturing. The `export coefficient' (the ratio of exports to gross manufacturing output) increasedfrom a mere 3 percent in the late 1970sto over 50 percent by the late 1980sand over 70 percent by the late 1990s. The prolonged (1955-75) terms of trade shock was overcome by this successful diversification o fthe export structure (Figure 1.2). The share o f manufacturing inGDP rose from around 12 percent inthe period 1950-80to 17 percent inthe late 1990s. The share o f state enterprises in manufacturing value-added fell steadily from 66 percent in 1974to 3 percent in2000. The share of factory industry in manufactured production increasedfrom about 50 percent in 1975 to over 80 percent in 1995, while that o fprocessingexport crops fell from 37 percent to 11percentover the sameperiod (Abeyratne andRodrigo, 2002). Employment inthe organized manufacturing sector more than tripled from about 140,000 in 1978to over 500,000 inthe mid-1990s (Statistical Annex Table A14). The new FTZsprovided an enormous fillip to employment (mostly inmanufacturing), adding over 400,000 jobs in 25 years. Since the bulk o f these jobs went to women workers in garments, it had apotent influence on women's economic andsocial empowerment.6 Foreign direct investment played a big role in the transformation o f Sri Lanka's trade and industry. The share o f foreign firms inmanufactured exports increasedfrom about 25 percent in 1977to nearly 80percent inthe mid-90s. Between 1985and 1995,85 percent ofthe total increment in manufactured exports came from foreign firms (Athukorala and Rajapatirana, 2000). It would be hard to find a more convincing case of trade and industrial transformation o f a small island economy through market-friendly policy reforms. At the same time, analysts (such as Kelegama (2000)) correctly point to several external factors which may have amplified and assisted the impact ofreforms inthe initialyears. First, there was massive foreign assistance insupport o f the early reforms, which supported a bigboost to public investment and financed large fiscal and extemal account deficits. The jump in foreign aid must have also boosted confidence of foreign private investors. Second, the rapid increase in garment exports from nil to over US$ 2 billion in less than two decades may have been helped by the institution o f the quota system under the Multi-Fibre Arrangement (MFA)as a protectionist device by industrial countries. Without this, it is a moot point whether Sri Lanka's fledgling garment export industry would have taken off in the manner it did. Third, the doubling of tourist arrivals between 1978 and 1982 (which boosted employment and earnings) mayhavehad as muchto do with world tourism trends as Sri Lankanpolicies. 1.13 Perhaps the most valuable, though intangible, contribution o f the reforms and their impact was that, over time, they won the support o f both major political parties (UNP and SLFP) to a common economicpolicy paradigm o f an open, outward-loolung economy with a growing role for the private sector. Thus, when the UNP government was replaced in 1994 by an SLFP-led coalition, there was no significant change in the broad direction o f economic policies. Indeed, there was some intensification ofthe privatizationpolicies. The share of female workers inmanufacturing employmentnearly doubledto 60 percentbetweenthe early 1980s and mid- 1990s (Athukorala and Rajapatirana,2000). 6 Sri Lanka DevelopmentPolicy RLview 1.14 One other important feature of Sri Lanka's economic landscape dates back to the early reforms era. This is the importantrole played by large scale external migration and the return flows o f remittances. Prior to 1977 both extemal migration and remittances were negligible. Both grew rapidly (though not steadily) in the next 25 years. By 2002, about one million Sri Lankans were working abroad and remitting nearly US$1.3 billion through official channels. That constituted 18 percent of all cunent external receipts and amounted to 9 percent o f GDP or nearly triple the level o f net foreign capital inflow (Figure 1.4 and Statistical Annex Table A15). Several factors contributed to this upsurge in migration and remittances: the opening up o f the Sri Lankan economy and general relaxation o f controls; the economic boom in the labor-scarce oil economies o f the Mid-East; Sri Lanka's market-responsive exchange rate policy; the push-factor of the prolonged ethnic conflict and the slow growtho fthe rural economy. Flpure 1.4: Trade Balance, Current Account Balance B Remlttance Inflow (As Percent of GDP) 176 1981 1888 1981 1996 2001 Year Source: Central Bank of Sri Lanka and staff estimates. 1.15 The economic consequences o f this rising migration were substantialand generally beneficial for the domestic economy. First, the highlevel o f remittances helped finance Sri Lanka's large trade deficits (Figure 1.4). Second, by adding to domestic savings the transfers allowed higher levels o f domestic investment. This was important against a background o f persistently low levels o f domestic savings inthe economy (Figure 1S). Third, the large outflows o fmigrants (cumulatively estimated at nearly half the increment inthe country's labor force between 1978 and 2002) eased domestic labor market conditions andhelped reduce unemployment especially among the female labor force. Fourth, since the migrants came predominantly from low income families, the poor benefited f?om relatively well-paid (often temporary) overseasjobs and their remittances home may have added as much as 20 percent to household incomes back home. Fifth, Papanek (2003) stresses that "migration disproportionately benefited poor women: more than two-thirds of all migrants since 1989 have been women" o f whom "88-95 percent were either housemaids or other unshlled women, overwhelmingly drawn from the poorer families". 7 ' Sri Lanka Development Policy Review Figure 1.5 : Gross Domestic Saving, Gross National Saving 8 Investment as percent of G D P , (1960-2000) 40 35 => 30 25 4 v, 2 0 Z (3 15 03- n 10 ( 3 5 00 \Q Year -GDS - - - G N S -Inv 1 Source: Central Bankof Sri Lanka and staff estimates. C. The Decades of Civil Conflict, 1983-2001 1.16 Inthe early eighties the Sri Lankaneconomy was on a strong upswing. Growth was robust, investment was at a peak, exports o f manufactures were booming, tourist arrivals were soaring and foreign investment was risingfast (Figure 1.6). The prospects for continued buoyancy lookedbright. Unfortunately, in 1983, the long-festering problem o f ethnic discord flared up and plunged the country into a twenty year civil conflict, which remains unresolved despite a cease-fire since February 2002. The roots o f the conflict go far back into colonial times and are largely political in nature. Some analysts (eg. Abeyratne, 2003) have stressed some o f the economic causes. The basic argument i s that the relatively slow growth o f output and employment (1956-77) could not meet the rising expectations o f the growing numbers o f freshly educated entrants to the labor force, spawnedby the largely successful policies o f free and universal education. Unemployment was particularly high among the young and school leavers. Furthermore, during this period, the state's control over formal sector employment increased greatly (through nationalizations, new state enterprises and heavy government recruitment) and employment patronage was often dispensed on ethnic lines. The benefits o f economic liberalization and heavy public investment (Mahaweli imgation project) in 1977-82 were also unevenly distributed across communities and regions. Coupled to the discriminationembeddedinlanguage, education and social policies after 1955, it breda volatile mix, which contributed to both the Sinhalese-Tamil ethnic conflict and the 1987-89 resurgence o f JVP militancy. 1.17 While there may be considerable debate and speculation on the economic causes o f the civil war, there is little doubt about the adverse consequences o f the conflict on the economy. The humanitarian costs have been nothing short of disastrous (see Box 1.2). At the macro-level, the Central Bank's estimate of lost GDP growth of 2-3 percent per year (after 1983) seems as reasonable as any (CBSL, 1998). But such a summary statistic masks the multi-dimensional andpervasive costs ofthe war. These include: The massivegrowthinmilitaryexpendituresfrom about halfapercentofGDP inthe 1970s to around 6 percent o f GDP in2000, as the size o f the armed forces increasedfrom less than 10,000 to around 300,000. The share o f security expenditures in total government expenditure rose from about 3 percent in the early 1980s to well over 20 percent in the late 8 Sri Lanka DevelopmentPolicy Rtview 1990s. Sri Lanka spent more on the militarythan any other country inSouth Asia (Statistical The persistence o f high fiscal deficits, typically around 10 percent o f GDP, throughout the period, leading to the usual problems o f crowding out o f private investment and a serious public debt problem. By 2001, despite large andprolonged flows o f concessional assistance, Sri Lanka's government debt exceeded 105 percent o f GDP and interest payments consumed about 7 percent o f GDP. Of course, war-related expenditures were only one part o f a larger fiscal problem. The sharp rise in military expendtures constrained other civil expen&ture programs. Thus, expenditure on education and health remained at around 4 percent of GDP through most o f the period, well below the levels o f 5-6 percent allocated for security inthe 1990s. The rapid increase in foreign investment (1978-82) leveled off for the remainder o f the 1980s as security became a major concem for potential investors. In particular, large electronics multinationals (eg. Motorala, Harris, Sony and Sanyo) which had been exploring major commitments in 1982, shifted their investment plans to South East Asia. In later years several initiatives for private infrastructure projects foundered on the excessively highretums sought by foreign investors to compensate for war risks. There was a similar leveling off intourist arrivals after the initialboom o f 1978-1982. Let alone growth, the 400,000 arrivals level attained in 1982 was only touched in a couple o f years in the 1990s. Over the same period, other nearby destinations such as Maldives, Mauritius and Thailand which had started form a similar base, went on to clock several million tourists per year. Sri Lanka's ambitions for becoming a financial hub andan aviationhubwere frustrated by the continuing uncertainties o f the conflict (Kelegama, 2002). There was massive destruction of economic and social infrastructure facilities inthe war- tom North East (and sporadic bombing o f targets inColombo and environs). Traditional economic activities (agriculture, fishing) in the North-East suffered serious decline as much o f the area became an active war zone (Box 1.4). Outside the North East, the economic utility of interconnecting infrastructure (such as highways) was lost due to closure o f such links because o f the conflict. As a result o f war-strained public finances and continuing focus on social and welfare policies, public investment remained low (around 5 percent o f GDP) throughout the period, leadingto the major shortages inkey infrastructure sectors o fpower and roads. The conflict exacted a substantial toll in the efficiency and integrity of institutions o f economic and social governance. 9 Sri Lunku DevelopmentPolicy Review SelectedHumanDevelopmentIndicators 2001 North srl L , m k 9.u 29 486 30 12 81 2% 6Si.3k1 YO I% Schmldropout 15% 4% Per-cqita tnronte (ZOW) K 37,206 R 63,OOo Batticaloa distnct only Source WHO and DCS. IndirectCos& incomelost fi~snfitregonepblic i ~ ~ ~ ~ ~ 136e n ~ 5 I, ,, reducedburrstarrivals 17.0 ,, t, reducedforeigninvestment 71 2 ,, *~ dtsplacnnent ofpeople 5 3 ,~,, humaicapitalofdeadandinjured 2.5 Oittpt foregonein NorthernPwvince 1.? 168.0 Onthe samebasis, e ~ ~ ~to include the cast5 iwunedin $997-2001,would yield a EWE tn excess ofwrce Sn t a ~ ~ n ~ Laiika's CDP (1996) 1.18 Perhaps the most pervasive, though intangible, economic cost of the conflict was that it diverted the energy and attention o f the policy-makers from the arena of economic reforms and policies. This was certainly true in the period 1983-89, when the government leadership was preoccupied with the "two wars" (against the LTTE inthe north and, later, the JVP inthe south) and there was very little progress in economic reforms. A similar distraction from economic priorities is discernible inthe period 1995-2001, when there was an intensification ofthe civil conflict. 10 Sn LankaDevelopmentPolicy Review Box 1.4: Eennnnty of the North and East Since the f'ghting d ~ n the war was isolated ti? the Fitctrph and E&, these provinces have suffered the rnt)a, g including a deathtoll estimatedat o w 76,000 anda halving ofthe prosincia1w~3n0ayduringthe 19%~. Agrrculrured o m ~ nthe~pre-ware~onom~ a ~ ~ ofthe North Eastaccrwiitrngfor 33 percentofnufquts and 80 perrent ~ f e m ~ l o Dunngthenwar, ~arris underctiltivath inthe NorthEast fekl by between S(d and 80 percent, while y ~ ~ ~ ~ the share of ttte ~sorth~ a s11%p t ~p ry~ d udropped~fi~mninf /3"" in 1980iojust 3 percent in~ 0 1 3 .Pmiers and ~ ~ ~ ~ fi~hemenof the T+srr;k East *ere re111ceantto invest in capital assets ~ t longterm ~~~~~~~s " Irrigation systems we= destroyed and e l ~ ~ ~ ~IS hawt. ~~~~r~r hlargescale industry was newr a major ~ ~ ~ n t rprior tuf ~ ~ ~ r , o u ~ h ~ ~ f thewar, theregtonp r ~ ~ i ccement, caustic sada, chlwine-and~ ~ c h lacid~amongstother materials e d o c The longkcm impai3ofthe war IS refI&%afIJ ~the s w a l development ~ ~ofthe Northand East where child ~ ~ d ~ ~ ~ m a l n u ~ r IS~ ~44~ ~ ) ~ preent, litwiacy 613 percent ampard to national averages of 29 pe~centand 87 percent respwtiwdy. Teachers and drxtors are i ~ r ~ d ~ ~unwitlirrp~to~serve y conflict areas, multrng in severe a ~ i b ~ iii shortages and poor outconim. The must affmtd are vitlnerablegroups such as those forced to drop out of scRncil (nmarethan 100 schools have hem closed down in the North and East), war widows and their families and the dtsltbled. Even ina ~ heatlh nmalns fmited and exposure to malana, diarrhea, res^^^^^^? inkclions and H1'6'lAbt)s increase ~ ~situation, mortality mtes~wilt n ~ ~ ~ continueto rm~ c ~ ~ ~as public slmding on ~ ~ i ~ d r ~ ~ a ~~ ~n ~ ~the ~~ y~ ~ ltunaan a ~ region's r ~ d e~v e~~ ~npindicators will take h i e and effort and m~~ ~ \nyillrequire donor suppm%in the short %em.Careful attention needs to be paid to the provision of basic s w a t smites and socialprotection For vulnerablegroups asthey form fertile rmmitinggminds Forguerilla ~ o v e ~ ~ ~ I i g s The war hashena facwr.behinda h " a millionSn Lanhs~ e a ~ ~isfmd, Many confinuetoremit savings n g ~ e to their relativesinthe Northan$ h t , wurw of triceme that protect local pngulawic-)ns&om the worst ravagesof war" The h'orph East has received snrall amounts of aid sflice the conflict began. ~ u m a assistance~has~been ~ ~ n deliveredby a number of NCWsthat were activethroug~~u~ the eorrfllct, Mwerrrreidy the projtxts supported by the WtwId Bank and ADB resprively have pm4ed fest'mrces ta rebuild vrffage ~ ) r ~ ~ ~~ ah ~a b~ ~~mfi~as&mcrureeand provide lwal ~ ~ ~ o ~ y~ o~~ u nni~~~~~i e fse, r organizational, ~~ n~sa +~ e s r I I ~ ~ matiagerial atad technrcal skills to lac& communities, Last year, the World Bank also launched its lvdorrh Easi ~nier~ency ReconstructirmPr(3jecE(NF?RP)which will helprestweprmary healthcare, urbanwater scliemesand prov~dereturningfandies with income and ~ n ~~ ~ p o r lt i andi the~~North~East ~ ~ ~o ~ t i ~ ~ MousingProject was ~ just approvedlate this year, which will helppoor f ~ ~ ~ i ~ r their honies. e ~ u i l d Source: Staravananthan, Muttuknslma. "An ~ ~ ~to therconflict time economy~of thei Ntnlhand~East Provinceof Sn ~ ~ ~ o ~ L,anklt". ICESworkingpaperNo. 1 May 2003 D. EmergingDevelopmentIssuesandProblems 1.19 Against this background, a number o f development issues andproblems emergedas threats to Sri Lanka's long-term development potential. TheFiscal Problem 1.20 Sri Lanka's public finances have been subject to virtually continuous strain over the past two decades. Large fiscal deficits averaging 10 percent o f GDP (andprimary deficits around 3-4 percent of GDP) have led to high levels of public indebtedness and debt burden. Several factors have contributed to this situation. First, the tax-to-GDP ratio declined from around 19 percent in mid 1980sto about 14percent in2002, with virtually all the decline occurring after 1990. This, in turn, was due to the decliningrole o f foreign trade taxes, the rapid growth of a virtually tax-free BOVFTZ sector:the stagnation o frevenues from income taxation, proliferation of concessions and exemptions (Stat. Annex Table A20), a complex transition from turnover and excise taxes to a system o f VAT with a few select excises, and persistent weaknesses in the administration of the InlandRevenue and Customs (partly exacerbated by the prevalence o f the parallel revenue concessiongranting authority The exportincomefrom the BO1sector accounts for about 80 percent ofthe country's manufacturedexports.The provisionsoflongtermtax holidays and duty and tax-free status for imports havekept this importantanddynamic segment ofthe economy virtually outside the tax net. 11 Sri Lanka Development Policy Review o f the Board o f Investment). Second, during these decades, defense/secwity expenditures rose from around 1 percent o f GDP to around 6 percent. Third, government (civilian) salaries and pensions consistently absorbed 5-6 percent o f GDP, reflecting a bloated and (at lower levels) overpaid bureaucracy. Repeated efforts to control the size o f the government salariat and its compensation levels have produced little progress. Fourth, throughout the 1990s the role o f concessional extemal financing o f the fiscal deficit has declined steadily, with a corresponding increase in recourse to market-based domestic funding, resultingina risingratioo f interest payments to GDP. Finally, inthe face o f declining public resources, priority was given to funding social and welfare policies rather thaninvestingineconomic infrastructure. Figure 1.7: Evolution of PublicDebt and FiscalDeficit 1985 1990 1995 1998 1999 ZOO0 2001 2002 PublicDebt --t OveranFiscalDefkit Source: Central Bankof Sri Lanka and staff estimates. 1.21 The consequences o f these long-termpressures on the fiscal system may be seen indeclining ratios o f public investment, constrained budgets for education and health, persistently high fiscal deficits andpublic debt-to-GDP ratios averaging more than95 percent inthe 1990s. The vulnerability of such highratios was heightenedinthe severe economic downturn o f 2001, which took the debt-to- GDP ratio above 105percent andbrought home the issue o f fiscal/debt sustainability. InterestPayments 4.6 6.4 5.7 6.7 Source :CBSL Annual Reports Administrative Erosion 1.22 The past two decades have witnessed substantial erosion o f institutional capacity and economic governance in the nation's adrmnistrative system. With 14 percent o f the labor force employed in government (18 percent ifthe semi-government sector is included), Sri Lanka has one o f the largest bureaucracies per capita inthe region (Table 1.5) with a ratio of 3.9 civil servants per 100 population. This bureaucracy is large, costly and suffers fkom low effectiveness and efficiency. The incentive system is perverse. Government workers with the lowest levels o f education on average earn almost double the wages o f their private sector counterparts. Top policy-makers and professionals are insufficiently compensated. As a result, Sri Lanka's compression ratio about (8: 1) is 12 Sri Lanka Lkvdopment Policy Review the lowest in South Asia. Controls on public recruitment have weakened over time. Despite the growing role o f the private sector and increased privatization, government employees have risen in both absolute numbers and as a proportiono f labor force. Since 1990 the number o f employees have increasedby almost 50 percent (to over 900,000) andtheir share inthe labor force has risen from less than 1Ipercent to about 14 percent. Regular large scale recruitment waves have occurred, typically linked to the electoral cycle. For example, in May 1999, 10,000 unemployed university graduates were recruited without their attachment to any existingprofessional scheme o f service (World Bank, 2000). Politicization o f public recruitment has weakened the norms prevailing in the administration, bloated itsnumbers andreducedits effectiveness.* Pakistan SriLanka EastAsia China *Total CivilianGovernment includes employees from the education and healthsectors. Source: World Bank (1998), Research Working Paper "Government Employment and Pay: The Global and Regional Evidence." Staffestimates. 1.23 Rising govemment recruitment has gone handinhand with the proliferation of government institutions inthe past 20 years. The numberof central ministries has fluctuated around a rising trend (from 28 in 1990 to 51 in 2003). There are over 40 provincial ministries and some 150 statutory boards and public corporations. The administrative apparatus has often been duplicated at the provincial level, resulting insubstantial confusion and ambiguity regardmgroles and responsibilities of different levels o f govemment and between elected and'administrative arms. Contrary to expectations o f a decentralized system, the central government has not withdrawn from devolved fimctions and, in some instances, has increased centralization. Moreover, Sri Lanka's public admmistration continues to rely on outdatedmanagementmodels andprocedures. A SluggishRural Economy 1.24 Unlikeindustry, the recordofreforms inagriculture and rural development was quite limited duringthese conflict decades (leaving aside the privatization andrehabilitation of plantations). This was so despite the fact that (even inthe mid-nineties) 80 percent o f the population and 90 percent of the c0unt.r~~~poor lived in rural areas. A broad range of government policies had the unintended effect o f reducing the capacity o f farmers to enhance their productivity and competitiveness and respond to local and intemational market opportunities. These included unpredictable trade policies that increased domestic price uncertainty; highly restrictive seed and quarantine regulations which limited access to improved technologies; commodity price interventions that discouraged private sector participation; complex and restrictive landpolicies that hindered the efficient use o f land(e.g., the fragmentation of land into non-economic holdings o f less than one acre); and poorly functioning I t may have had more pernicious effects. The Presidential Commission on Youth, established after the violent youth insurgency inthe South of 1987-89, reportedthat there was a strong consensus inthe country that politicization and abuse of power and injustice were the main causes ofyouth unrest in the country (World Bank, 2000). It singledout rampant politicization inthe systems ofpublic recruitment as a source of injustice and dissatisfactionand recommended strongly in favor of restoringthe independence of the Public Service Commission. 13 Sri Lanka helopment Policy Review water delivery systems. The combined effect of these measures was the creation of a complex structure of obstacles and disincentives, which damped agricultural productivity growth, impeded diversification, discouraged investments and reduced opportunities for raising agricultural and rural incomes. A continuing emphasis on input subsidies (fertilizer, water, seeds) diverted public resources away from creating critical rural infrastructure (roads, electricity, research and extension). This i s despite the clear need for investments in nual infrastructure. According to the rural part o f the recently completed investment climate assessment (ICA) on Sri Lanka, transport constitutes the single most important constraint to ruralfirms, followed by cost and access to credit and electricity? Table 1.6: Agriculture, Forestry and FisheriesGDP, Growth and Structureof Value Added Subsector Average Annual GIlP growth rate(YO) Share inAgrlcultum, Forestry and E3skerier WW (%) 1982-90 1991-00 1998-02 1982-84 1990-92 2000-02 Agrirultitre, Fcxestry I.7 I.9 0.9 and Fisheries to0 1 lx, 1CN, Agriculture 2.8 1.h 0-4 85 83 80 `Iea 2.Y 2.4 2.5 I? 8 ti -3.8 1.5 -3.2 x 6 Rice -0.3 -0.7 0.6 21 19 14 kher 6.3 I.9 0 x 37 40 52 Forestry 6.1 2.3 2.8 5 8 8 I.ishmcs 0.7 0 1 2 2 9 9 1.3 Source: Mission estimates. 1.25 Agriculture (including forestry and fisheries) grew at barely 2 percent per year during this period. Moreover, excluding forestry and fisheries, there has been a noticeable deceleration in agricultural growth inthe 1990s (Table 1.6) down to 1.6 percent. As a result o fthese trends the share of agriculture in GDP declined from over 30 percent in early 1980s to 20 percent in 2000, while agriculture's share intotal employment fell from46 percent to 36 percent (Table 1.7). The growth o f the rice sub-sector, a traditional mainstay in the country, has been negligible or negative over these decades despite high protection. Growth o f tea and rubber improved over the 1990s (perhaps reflecting privatization o f plantations), but the performance o f other non-traditional crops (which had grown at 6 percent inthe eighties) weakened substantially. The fisheries sub-sector showed the best growth inthe 199Os, talung its share inagricultural value-added to 13 percent by2000-02, comparable to the share o frice. Table 1.7:Employment and Value Added by Sector Year Unemplo ShareinTntd Employment Shareof katneAdded in CDP Agriculture Indnsaq. Senires Agricnltnre lndmtr) Services World Bank (2004), "Sn Lanka: Improving the Rural and UrbanInvestmentClimate" (Draft to bepublished) 14 SriLankaDevelopment Policy Rcview 1.26 The slow growth of agriculture was a primary factor explainingthe persistenceofhighlevels of rural poverty and its substantial variation across provinces. In the decade between 1985/86 and 1995/96 agricultural households accounted for about 40 percent o f Sri Lanka's poor. More recent estimates (using the Sri Lanka Integrated Survey-SLIS, 1999/2000) indicate that agricultural households(excluding the estate sector) comprise almost half o f the poorest rural households, defined as the bottom 40 percent o f the expenditure distribution." More importantly, rural poverty rates are highest inprovinces with the highest proportion o f agricultural households. Income from agriculture was critical (as a share o f total household income) in some provinces, particularly North Eastern (67%), Sabaragamuwa(60%), Uva (59%), Southern(48%) and Central (47%). Theseprovinces have some o f Sri Lanka's highest ruralpovertyrates. 1.27 At the same time, the possibilities for better agricultural performance were indicated by the changing composition o f food expenditures resulting from rising per capita incomes, increasing urbanization, changing prices andpreferences, and growth of exports and tourist consumption (Table 1.8). Despite policy obstacles, there was some diversification inproduction in response to changing demand. The production o f maize, fruits, vegetables, spices and livestock (especially poultry) have increasedinrecent years. Agricultural exports grew at over 4 percent annually inthe 1990s, reaching about US$950 million inthe triennium ending 2002 (or almost 20 percent of total exports). Ths was primarily drivenbytea and some non-traditional exports of fruits, flowers, vegetables, fish and spices. 2002 1988181 Rice I`J.3 22.4 3I.S Wheat flour & bread 5 5 5.6 7.1 Pulses 7.5 4.1 2.4 Meat 8ad fish (id.&I&) lh.7 12.2 9.9 MilkLMilkproducts 7.9 4.7 3 3 Sugar 3 3 b.2 7.8 Tyreport. An Infrastructure Backlog 1.28 The heavy build-upof security expenditures and continuing focus on consumption spending diverted resources from public investment in economic infrastructure (power, water, transport, communications), which fell from around 14percent of GDP in 1985 to 6-7 percent by 2000. Against this backdrop, the recently completed ICA reports that inadequate transport and electricity services constitute top constraints indoing businessin Sri Lanka (World Bank, 2004). 1.29 Although the c0unt1-y'~road network is dense (1.5 Km/Km2 of area) by regional standards, only 10% of the paved road network is ingood condition due to lack o f maintenance. This situation results in increasedroad user costs, poor quality and frequency of transport services, and low road safety. Sri Lanka has the highest road accident fatality rate in Asia. The National Transport Policy estimatedin2000 that the annual cost o froad accidents is about 0.4% of GDP ($50 millionper year). Inaddition, despitea substantial increaseintraffic demandover the past four decades, there hasbeen very little investment inimproving trunk roads, upgrading (e.g. widening) of roads and construction o f new national highways to provide rapid access to urban centers. Although new public investment Io Agriculturalhouseholdsare those involvedinfarmproduction, raisinglivestockor casual agriculturallabor. 15 Sri Lanka Development Policy R n i m has concentrated on small, rural access roads, there remain serious regional disparities both in road density and condition of roads. The Western Province has the highest density of roads inthe country (about 4 km/Sq. km) but due to its populations levels, it has the lowest number of kmper inhabitant. The North Eastern Province, having the lowest road density inthe country, even low by international standards, is second to the Western Province in terms of km per inhabitant. In contrast, the North Western Province has similar road densityto Central, Southernand Sabaragamuwa Provinces,but has the highest number o f km per inhabitant in the country (see Figure 1.8). This regional disparity in terms of access is particularly worrying, as it is the utilization of assets rather then their existence which determines the contribution of infrastructure assets to economic growth". Roads in good maintainable condition with highutilization rates, such in the Western Province, yield higher returns to the economy. Incontrast, maintenancebacklogs for rural roads, are so severe inall other provinces (see Figure 1.9) that rendersthemimpassableduringmost of the year. As a result, Sri Lankans living outside the Western Province do not enjoy the connectivity that others have. The situation is even worse inthe conflict-affected areasinthe NorthandEast Provinces. Figure 1.8: Road Density by Province, 2002 WP CP SP N E P W N C P up SAP Notes: All national, provincial andlocal authority roads are included. Sources: RDA for national roads and GOSL' concept paper for Road Sector, December 2003 Figure 1.9: Rehabilitation FundingRequirements for Provincial and Local Roads, 2004-2008 WP CP SP NEP NWP NCP UP SAP I Provincial Roads (US$ m) Local Authority Roads (US$) 1 Sources: GOSL' concept paper for Road Sector, December 2003 `IIthas beendocumentedthat road density does not have a clear association with per capitaincomeor growth rates inAsia. Source: J.M. Antle, 1983, "Infrastructure and Aggregate Agricultural Productivity" International Evidence", Economic Developmentand Cultural Change, vol31, pp.609-619. 16 Sri Lanka LWelopment Policy Review 1.30 The unreliability and cost o f electricityis another major constraint to economic development. Despite an impressive expansion inaccess to electricity over the past two decades (from 15 percent to 74 percent o f households), capacity constraints in power generation threaten to limit future growth. With a installed power generating capacity o f 2002 MW (of which 59% is hydro and about 20% is privately owned), Sri Lanka has been experiencing a chronic shortage o fpower since the mid-l990s, stemming from frequent extreme drought conditions and rapidly growing demand, averaging 8 percent per annum duringthe last decade. As a result, the sector has been forced to rely on expensive emergency power generation. While generation capacity could be met by the private sector (which is allowed and has recently added 350 MW), the Government lacks an appropriate planning framework to forecast future demand and the decision makingto procure new plants is extremely ineffective. Figure1.10; ComparisonofElectricityTariffsofASEAN Countries Srilsnka Malayia Singapre Thailand PhilipinneB Indonesia LWS - - ~IResidentiaClow Residential- t-figh Comrcial low 0Comnercial- High hdustrial-low hdustrial Hgh I Source :ASEAN website 1.31 The use o f expensive emergency generation coupled with past (uneconomic) pricing policies by the state-owned Ceylon Electricity Board (CEB) has led to large financial losses in the sector, delayed investments innew transmission and distribution facilities, and often neglected maintenance o f existent infrastructure. To address these problems, the industry is undergoing a process o f restructuring which involves the unbundling o f the generation, transmission and distribution o f the CEB and the Lanka Electricity Company (LEC0)12 as well as the establishment o f an appropriate regulatory framework through the effective operation o f an independent regulatory body. 1.32 These constraints result in high cost o f electricity for commercial and industrial use by international standards (see Figure 1.10). Indeed, due to increased reliance on expensive thermal emergency generation, large cross-subsidies to residential consumers, and CEB's financial and technical losses in the transmission and distribution o f electricity, commercial establishments face costly and unreliable electricity supply which reduces their competitiveness inthe market place. For example, about 75 percent of urbanmanufacturing firms own a generator inSri Lanka compared with lower levels incompetitor countries (e.g., 27 percent inChina). 1.33 In addition, regional inequities in access to electricity across the country have likely exacerbated poverty inunder-served areas. The diagram below shows that access to electricity among households varies widely between regions - the Colombo district is also shown. As expected, the percentage o f electrification in the Western Northprovince i s 95 percent o f households compared to l2LECO handles 15% percent of customers inthe Negambo to Galle coastal strip andurban areas around Colombo 17 Sri Lanka Development Policy Review 41 percent in the North Central province. North and the East being conflict areas indicate 9 percent and 36percent respectively. Figure 1.11: Percentage of Electrificationby Region I I 90 I 80 I 70 I YO 60 Electrificatlon 50 40 30 ~ 20 10 0 Regions Source: CEB. "Medium Voltage DistributionDevelopmentPlan 1999-2007". Basedon 1998householddata. Increase in RegionalDisparities 1.34 Economic activity in Sri Lanka has, for many decades, been concentrated in and around the capital city and intemational shippingair gateway o f Colombo. In that sense, the economic dominance o f Western Province i s not new. However, new estimates of provincial GDP since 1990 points to a continuing rise inthis dominance and raises concerns about increasing regional disparities (Table 1.9). The share of Westem Province rose from around 40 percent o f the country's GDP to almost 50 percent by 2000, even though the province accounted for only 29 percent o f the nation's population. Over the same period, the share ofNorthem Province halved from an already low level o f 4.4 percent to 2.2 percent, largely reflecting the direct impact of prolonged and intense civil conflict. Morebroadly, the share of the four low-activity provinces (Eastern, Uva, North Central andNorthem) fell from 22 percent o f national GDP in 1990 to 15 percent in 2000, although about a quarter o f the country's populationlive inthese province^.'^ Table 1.9: GDP Shares hv Prnvince -~ _. .._ --.I- - - ----- Pro, ince/year 1990 1995 1996 I997 tY98 1990 2000 2001 westem(M'1') 411.2 42.3 43.7 44.3 45.3 48.7 49.6 47.3 N WesterritNWP) 1 1 I 9 6 1 1 3 12.1 12.0 10.4 1\14 10.9 (`mtral( ('PI 11.1 11.7 10.0 10.5 9.8 9.2 9.4 9.8 StjutherII( SP) 9 s 9.5 9.0 8.8 9.3 9.6 9 4 9 9 SahaPaptiiuwa(Sal') 8.1 7.7 9.0 7.b 6.7 h.4 6.7 6.3 r~astrl~lr IV) 4.2 4 6 4.8 s.0 5 5 5 0 4 5 S .o tjva( til') 8.1 7.7 5. I 5.o 4.9 4.1 3.9 4.5 K ('eimal15~'l') 4 8 0 3 4 (I 4 U .\h 4 1 .\ 9 3 8 Nu&iwn( N1') 4.4 3.1 2.4 2.8 2.9 2 s 2.2 2.5 N\`atiunalril)I' 100.1) 1 00.0 100I) 1OU 0 101) I) 10I) 0 IOU 0 1OI),O Sources: 19900-OSdata from Dept. ofNational Planning(Sawanantham, 2003). Data from 1996-2001 from CBSL (2002) 2000 is preferredover 2001becausethe exceptionaldecline inreal GDP was concentratedonthe Westem Provinceand biasestrends. 18 Sri Lanka DevelopmentPolicy Review 1.35 Grouping the provinces into three regions, Highactivity (Western), Medium (the next four in relative shares: NWP, CP,CP and Sap) and Low (EP, UP, NCP and NP) brings out some further dimensions o fthe problem o fregional disparity in Sri Lanka (Table 1.10). First, as expected, the High region (WP) accounted for 60 percent o f the country's industry in the late nineties, while the Low regionprovinces had only 9 percent o fnational industrial value added. Second, there i s a similar skew in favor of the Highregion in the dstribution of services value added, whle only 11 percent o f the country's agriculture is to be found in the High region. Third, the High region enjoyed a per capita GDP twice as high as that in the Medium region and more than two and a half times the level prevailing in the Low region. Fourth, and perhaps most disquieting, the growth o f nominal GDP was far higher in the Highregion as compared to the other two. The estimates available are for current price provincialhegional GDP. However, as noted by the Central Bank, the average GDP deflator " during this period was about 8 percent. Thus, a simple estimation o f real growth indicates that region Hgrew at over 8percent, regionMat around2percent andregionLatless than 1percent.. ." (CBSL, 2002). These are hugely disparate rates o f growth inreal regional GDP and they highlightthe problem o f growing disparities acrossregions and provinces. Source: CBSL AnnualReport, 2002. 1.36 These regional disparities are also reflected in the available data on household poverty (see Figure 1.12 in Section E below). In 2002, the incidence o f poverty was 11 percent in the Western Province, while it was at 25 percent or higher inall other provinces for which data was available (with the sole exception of Central province, where poverty incidence was 21percent). 1.37 A number of factors have contributed to the problem of regional dsparities, including the sluggish growth o f agriculture and related activities, the concentration o f the FTZs in the Western Province, regional disparities in education facilities and the persisting weaknesses in infrastructure noted earlier, especially the connectivity o f lagging provinces. The underlying trends have been aggravated by the civil conflict, which has impeded economic development in several o f the poorest provinces. However, to cite the Central Bank again, "early attention to this issue i s critical, as widening regional disparities not only affect the achievement o f a sustainable highoverall economic growth directly, but also indirectly, through their potential impact on social and political stability in the country." Quality of Education Constrained 1.38 Throughout the decades of conflict Sri Lanka maintained the high levels o f educational attainment achieved by the 1970s, although there was severe disruption to school systems in the conflict-affected areas of the North-East. For the country as whole, by 2002, the net enrolment rate at grade 1 was close to 100 percent and net primary completion was around 95 percent. Of students enrolling in grade 1, about 83 percent completed grade 9 or basic education. The 17 percent o f students who failed to complete the compulsory education cycle were mainly children inwar affected areas, children living inremote rural and estate regions, and those inpoor inner-city areas. 19 Sri Lanka Development Policy Review Table 1.11: Proportion of Primary Children Achieving Mastery of LanguageSkills, 2003 Skill First Imguage (Sinhair or 'l'amil) English1,anguage Sti I m k a 1,rbaii Il\Iriii SI1 1.allkiI I~lbdll Rlll.11 sL:c.ll~I scztt)I Sect01 SeC1l)I 0 0 9.0 0 , ) O O Q o 9 0 Vtn.alwl3r) 70 81 68 35 48 30 c 'c~iiiprel~rnsii~ii45 5h 43 Ih I\ 1 3 Syntax $0 44 27 20 34 17 Wtltltle - 18 40 2s 1 4 1 'Total d7 S I 34 10 21 7 Source. National EducationResearchand Evaluatlon Center, Umversityof Colombo. Table 1.12: Proportion of Primary Children Achieving Mastery of Numeracy Skills, 2003 Skill Sri t,rinka llrbra Sector Rural Sector % Yo YV Concepts 45 42 Source: NationalEducationResearchand Evaluation Center, Universityof Colombo. 1.39 In contrast to the high levels of education attainment at the primary and junior secondary levels, disappointing leaming results point to weaknesses in the quality o f education. Cognitive achievement tests among primary school children, for instance, show substantial shortfalls inmastery of basic language and numeracy skills, with mastery o f English being particularly low (Tables 1.11 and 1.12). The low level o f cognitive achievement among primary students i s especially worrying, from a policy perspective, as primary education forms the foundation upon which higher levels of education and various types o f skills training are built. It is also a concern that primary students in rural schools do relatively worse than those in urban schools in all subjects. Shortcomings in the quality of education are also evident at the secondary level. For instance, over the 1998-2002 period only 37-40 percent o f students passedthe GCE O/L examination and only 47-56 percent passed the GCE A/L examination. A substantial proportion of students appear to struggle with subjects such as Mathematics, English, Science and Social Studies. Ths is also o f serious concern since the successful completion of these examinations i s necessary for a variety of skills training courses or access to tertiary educationprograms, or for entrance into arangeo flabor market occupations. E. Growth andPoverty, 1990-2002:An Overview 1.40 As mentioned earlier, Sri Lankahas lagged behindinreducing income poverty. The national poverty headcount showed a modest decline over the decade -from 26.1 percent in 1990-91 to 22.7 percentin2002 (Table 1.13) with unevenpovertyreduction across sectors and province^.'^ 1.41 Inthe intervening period, nationalpoverty increasedby almost 3 percentagepoints between 1990-91 to 1995-96, followed by a decline o f more than 6 percentagepoints between 95-96 to 2002. Other measures o f poverty, namely depth and severity o f poverty, show similar trends. Evidence presented later will show that even during the period ofpovertyreduction between 1995-96 and 2002, growth continued to be unevenly distributed, resulting in increasing inequality co-existing with gains inaveragelevels ofconsumption. l4 Excludesthe Northemand Eastemprovinces. 20 Sri LankaDevelopmentPolicy Review 1.42 Duringthe decadefrom 1990-91to 2002, the Table 1.13: Poverty headcountsfor Sri Lanka'' poverty gap between the urban sector and the rest of 90-91 95-96 2002 the country widened. Urban poverty halved during the period, while rural poverty declined by less than 5 percentage points. Poverty in the estate sector, on the other hand, increased by about 50 percent over 29.4 30.9 24.7 the decade, making this sector now the poorest.'6 20.5 38.4* 30.0 The trend in rural poverty is consistent with positive Source: HIES data for relevant years, using official poverty lines trend inper-capita agricultural production during the *;Comparability o f estate headcount for 95-96 with that later part of the decade (as the sector recovered from for other years may be affectedby the fact that HIES in the severe drought of 1996), while per-capita growth 95-96 was sampled differently for the estate sector in the preceding years was negative. A decrease in income earners per estate household during the period i s likely to explain part o f the increase in poverty inthe sector." 1.43 In interpreting the urban and rural estimates above, it is important to note one caveat. The classification of urban and rural areas changed between the 1990/91 1995/96 HIES surveys. Town councils were considered part of the urban sector in the HIES 1990/91 survey, but were Source: HIES for relevant years, using official poverty lines then reclassified as part of the rural sector inthe HIES 1995/96 and HIES 2002 surveys." As a result, the estimated proportion of urban population reduced by around 7 percentage points from the HIES 1990/91 to HIES 1995/96 and stayed almost constant since then. This suggest caution in interpreting the absolute magnitude of changes inrural and urban poverty from 1990/91 to later years. The implication o f this change could result in an overestimation o f the decline inpoverty in the rural sector and an underestimation of the decline in poverty in the urban sector. However, this does not change the main message, which is urban poverty has reduced dramatically while rural poverty has been stagnant during the period 1990191 to 2002. The estimates reported below (Table 1.15) also support this conclusion. Poverty in the Colombo district, which constitutes a large part of the urban sector, more than halved between 1990/91 and 2002. While, poverty reduction inpredominantly rural districts has been minimal during the same period. Regional Differences inPoverty Incidence 1.44 Sri Lanka i s also characterized by sharp regional differences inpoverty incidence. As Figure 1.12 shows, poverty incidence was by far the lowest inthe Western Province in2002, with a poverty headcount of 11 percent. There is a wide gap even between the headcount rate of the Western Province and that o f North Central, w h c h has the next lowest poverty incidence o f 25 percent. In contrast, the poorest provinces of Sabaragamuwa and Uva have headcount poverty rates of around 35 percent. IsBased on official poverty lines (Rs. 1423, Rs. 833 and Rs. 475 for 2002, 95-96 and 90-91) respectively. The official poverty line is derived using a "cost o f basic needs" method on 2002 HIES data, and deflated by Colombo CPI to obtain nominal poverty lines for the other years. l6Note that becausethe estate sector comprises o f a relatively small part o f the population, the HIES sample (about 4 and 7 percent o fthe weighted sample in95-6 and 90-1 respectively) yields poverty estimates with a higher degree of "error." I'Preliminary findings o fthe Central Bank`s Consumer Finance and Socio Economic Survey (CFS 2003/04) indicate that income receivers per householdinthe estate sector has declinedsignificantly from 2.3 in 1986187to 1.7 in2003/04 while the national average stayed constant at 1.6. '*This implies that the definition o f `urban' in 1990/91 included urban, municipal, and town councils, whereas that in 1995- 96 and 2002 included only urban and municipal councils. 21 Sri Lanka Development Policy Review 1.45 The low poverty incidence in the Fig. 1.12: Poverty Headcount(YO) Protinccs in 2002 for Western Province is largely dueto the [ 40 ......... ....37...... .. ... ., ^ , location of Colombo where most of Sri Lanka's economic activity is concentrated, accounting for half of the c 0 u n ~ 'GDP. Since only 29 percentof s the population reside in the province, it is no surprise that poverty incidence is much lower. The hgh incidence of poverty in some of the other provinces areconsistent WP CP SP NWP NCP UP Sap withthe evidence on wide disparity in Source: HIES 2002. provincial incomes and share of GDP. The numbers underscore how poverty is concentrated geographicallyin Sri Lank%and also seem to suggestthat overthe past decade, there has been a tendency towards wider regional disparity. 1.46 Disaggregating poverty headcounts by Table 1.15: The headcountindices by districts (%) districts shows even wider differences in poverty incidence (Table 1.15). The district Province District 90-91 95-9G 3002 with the lowest poverty incidence, Colombo, Colombo 16 12 G has only 6 percent of its population who are Western Gampaha 15 14 11 poor. In contrast, 37 percent of the population 32 Kalutara 29 20 in the districts of Badulla and Monaragala are poor --- districts with the hghest poverty Kandy 36 37 25 Central Matale 29 42 30 incidence. The speed of poverty reduction also varies markedly among districts. Between 90- Nuwara Eliya 20 32 23 91 and 2002, gains in poverty reduction have Galle 30 32 26 been substantial inthe Western Province and Southern Matm 29 35 27 other urban areas like Kandy, whle poverty Hambantota 32 31 32 incidencehas been largely stagnantor increased Kurunegala 27 26 25 in the rest of the country.District poverty North West Puttalam 22 31 31 trends also show the vulnerability of certain North districts to shocks such as droughts. In 1995- Anuradhapura 24 27 20 96, there have been large increases in poverty Polonnaruwa 24 24 20 incidence for districts that experienced severe Uva Badulla 31 41 37 drought. Monaragala 34 56 37 Sabara- Ratnapura 31 46 34 1.47 Poverty trend over the decadewas also gamuwa Kegalle 31 36 32 associated with increasing inequality in Source:HIES 90-91,95-96,and 2002 consumption, along with an increase in average levels of consumption. Table 1.16 shows that while mean consumption increased by 50 percent for the top quintile and 25 percent for the 4" quintile, it increased by only 2 and 6 percent for the lstand 2"d quintiles, respectively. Increasing inequality over the decade is also reflected by theGini coefficients (Table 1.17). The gini increased between 1990-91and 2002 -by almost 24 percent for the country as a whole, including an increase of 19percent for the urban sector and 30 percent for the rural se~tor.'~ 19TheDCS (2004) reports that Ginis did not change much since 1990-91 (see TableHI). TheGinisreportedby DCS are howeverdifferent from theonesreportedhere in 2 ways: the DCSGinis are basedonhousehold Olcome and not consumption, andtheDCSmeasureofincome does not adjust forspatialpricedifferences. In thecontextofpoverty analysis,sincepoverty is definedinterms of consumption,gini of consumptionshouldbeconsidered morerelevant. Moreover, household incomes are difficult to constructand therefore tend to be less reliable than consumption measures. 22 Sri Lanh DevelopmentPolicy Review 1.48 An analysis of pro-poor growth, using the well-known Table 1.16. Mean realper capita technique o f Growth-Incidence Curves (GICs) provides more consn. (at 2002 prices) by quintile! insights. While growth in levels o f consumption had a poverty- Consn. quintiles 90-91 2002 reducing effect between 1990-91 and 2002, the benefits accrued Q1 1045 1068 disproportionately among the better-off." During the decade, all 42 1499 1596 segments o f the urban distribution experienced gains inper capita 43 1909 2168 consumption, but the extent o f gain was higher for those at the 44 2489 3117 upper end o f the distribution. Similarly, inrural areas, growth in QS 4871 7325 consumption was higher for the relatively better-off. Between Total 2363 3055 1990-1 and 1995-96, consumption growth was negative for the Source: HIES 90-91, 95-96, 2002 ani lower 50 percent o f the rural consumption distribution - which CCPI can explain the increase in rural poverty. Similarly, the GICs between 1995-6 and 2001-02 can account for the fall inpoverty headcount and widening o f the rich-poor gap inall sectors. between poverty reduction, growth and inequality in Table 1.17: Inequality: Gini coefficient of consumption can also be explored through the exercise per capita expenditure* of growth-inequality decomposition o f changes in 90-91 2002 poverty headcount. During the last decade the growth- Urban 0.37 0.42 inequality decomposition shows pronounced growth and Rural 0.29 0.39 redistribution effects. If inequality had not increased Estate 0.22 0.26 during the decade, we would have observed a National 0.32 0.40 significantly greater reduction in poverty. With no Source: HIES 90-91,95-96,2002 andCCPI to create a comprehensive safety net system Urban Rural Nattonal The GIC maps the average annual rate ofgrowthofreal per capita consumptionbetweenthe relevant years for all centiles (1 percent quantile) of the consumption distribution (see Ravallion and Chen, 2003 for details). 23 Sri Lanka Development Policy Review Part 11. Ceasefire and UnfinishedReforms 2.1 The year 2001 may have reflected the old adage o f it being "darkest before dawn". The emerging long term problems, outlined earlier, were severely exacerbated by several unforeseen shocks. There was an escalation o f the civil conflict, including the massive surprise attack by the LTTE on Colombo Airport. The fiscal situation worsened with the deficit mounting to nearly 11 percent o f GDP and governhent debt exceeding 100 percent o f GDP. Agriculture was hit by major floods and later drought, with real value added falling by over 3 percent. The power sector suffered from brown-outs, with consequent disruptions for industries and services, both o f which also registered negative growth. An already difficult situation was severely compounded by the events o f September 11, heightenedpolitical uncertainty and sluggish demand for Sri Lanka's exports. Foreign investment and tourism plummeted. Aggregate investment dropped sharply from 28 percent o f GDP in2000 to 22 percent in2001. Overall GDP fell by 1.4 percent, the first contraction inthe economy since Independence in 1948. Inflation accelerated sharply to 14 percent because o f drought, production shortfalls and substantial depreciation o f the Rupee. Defense spending and fiscal laxity grew inthe runupto the December 2001 general elections. 2.2 In the elections, the UNP-led coalition returned to power after seven years with a slender parliamentary majority and a mandate for negotiating peace andundertakingeconomic reforms. The new government swiftly negotiated a cease-fire with the LTTE inFebruary 2002 andinitiateda series o f peace talks aimed at securing lasting peace. On the economic front, it launched a wide ranging program o f fiscal consolidation andstructural reforms aimed at puttingSri Lanka on a faster trajectory o f growth and poverty reduction. The program was articulated in the "Regaining Sri Lanka" (RSL) document which was published in December 2002 following a preparation o f four years.*' The principal thrusts o f RSL were : (i) securing lasting peace; (ii) buildinga supportive macroeconomic environment; (iii) accelerating economic growth, with private sector leading the effort; (iv) investing in people and creating opportunities for the poor; and (v)strengthening "governance" in the public sector to enhance the development impact o fpublic resources. 2.3 The RSL received the broad approval o f the international community at the 2003 Tokyo Donor Conference where donors pledged US$4.5 billion to support the Government's post-conflict reconstruction efforts over a period o f four years. It thus provided the basis for increased support by the World Bank through its FirstPoverty Reduction Support Credit (PRSC I) by the IMF through and a new Poverty Reduction and Growth Facility/Extended Fund Facility (PRGFEFF). Other major donors providing increased assistance includedthe Asian Development Bank (ADB) andthe Japanese Bank for International Cooperation (JBIC). A. Peace andReformAgendas: 2002-04 2.4 The range and depth o f the economic reforms undertaken and initiated in 2002 and 2003 warrant the description "third wave" (Box 2.1). Much o f this agenda involved second generation reforms built on initiatives launched and developed under the preceding administration. A brief description o f some o fthe major elements o freform follow. Fiscal Consolidation 2.5 The 2002 budget signaled a commitment to fiscal consolidation, structural reforms and economic growth. Selective expenditure cuts were introduced, including a public sector hiringfreeze. A dual rate VAT system replaced the Goods and Services Tax andthe National Security Levy. The budget deficit target was set at 8.5 percent o f GDP (outcome was close to 8.9 percent) reflecting over 2 percent o f GDP deficit reduction in one year. The 2003 budget reflected a continuity o f these 21 PartIo fthe strategy laid out the "Vision for Growth"while Part I1articulatedthe strategy for connectinggrowthto poverty reduction. The latter partwas compiledmostly by the precedingSLPF-ledadministration. 24 Sri Lanka DevelopmentPolicy Review initiatives. To ensure that fiscal consolidation continues, the Fiscal Management (Responsibility) Act (FMRA) was passedinJanuary 2003. The Act sets a strict limit on borrowing from Central Bank, and fixes the deficit target for 2006 at below 5 percent o fGDP andgovernment debt at a target of 85 percent o f GDP (Box 2.2). It requires regular mid-year and end fiscal year fiscal position reports to parliament and also mandates a ''Re-election Budgetary Position" report to discourage pre-election profligacy. B@x2.1: Summary of Yeleetad Refornis,2002-03 Publie~ ~ ~ e ~andAdministration d ~ ~ n r a ineffect since O r t o k 2082. Pensaonscheme for new publicsector workersenactediFekruary 20U3). Vofilnkary ~ ~ ~ r Scheme~(VRS)n ~ e n ~ brgetedto tvduce over-staaffingin the lower and nndddlegradesannounced (Novemba 2003). Salary Review ~ o n t establishedto ~ ~ n ~ ~ ~ review wage scalesand ratiottalimtinn of g o v e ~ ~ ~agenczes (March 2003). e n t New, i n ~Public Service ~ ~ ~ ~ n ~ C'ornmissm appointedCZ00Z). Medium'ram Bdgttt Pm~ewnrk(MEBF) initiated(2004 budget). FinancialSedor Financialadviserhiredto assessoptions for the c o n ~ n ~ ~ c i a ~ iofPeople's ~ a r ~ o n Bank.(suspended) l3ank's capitaladequacyratio increasedto 109%(2003). , ~ ~ ~ o ~~ n~ d~Actato~~enhance ~n i~ g ~ ~ ~ CBSL's pnidcnrial o wsight of banking systempiis& (Dee 2002). Srt Lanka Insurance Chporatian prxvatid (April 2003).. Source: WorldRankKe1mrl.s 2.6 To contain the growth of government wages and pensions, several initiatives were launched, including a contributory pension scheme for new public sector workers and a Voluntary Retirement Scheme (VRS) to reduce overstaffing inlower and middle grades. Work was initiated on a Medium 25 Sri Lanka Development Policy Review Term Budget Framework (MTBF) aimed at improving public expen&ture management and a Salary Review Committee was established to look at salary scales and rationalization o f government agencies. 2.7 Despite these initial improvements in fiscal management, tax revenues continued to fall in 2003, partly reflecting the transition to a new (two-tier) VAT system. The slippage relative to the budget was about 1.5 percent o f GDP. With expenditure cuts partly offsetting revenue shortfalls, the budget deficit closed at 8.0 percent o f GDP in 2003, down from 2002 but higher than the budget target (7.5 percent o f GDP). State Owned Enterprises 2.8 The provision o f key utilities and economic services by state-owned enterprises (SOEs) has resulted in low quality o f service delivery, politicized pricing and management, and employment decisions. This has adversely affected the competitiveness o f the entire economy and created additional burdens on public finances (see Box 2.3). 2.9 Improvements in the performance o f SOEs was sought through increased private sector participation and the establishment o f autonomous regulatory bodies. To this effect, the Public Utilities Commission o f Sri Lanka (PUCSL) was established in October 2002 to act as a multi-sector regulatory agency initially covering electricity and water and infuture petroleum and railways. This followed the model o f the Telecom Regulatory Commission that is successfully overseeing an unprecedented boom inthe privatized telecom industry. However, since the PUCSL had not assumed regulatory functions, prices remained subject to political manipulation. 2.10 Inthe petroleumsector, the monopoly ofthe Ceylon Petroleum Corporation(CPC) inimport and distribution was ended by opening the market to two additional players while two thirds o fCPC's retail network was privatized. The competition resulted in visible improvements in the quality o f retail services. Pending PUCSL's assumption o f regulatory functions over the petroleum sector, prices were adjusted according to an automatic formula tied to world prices, but the pass-through o f oil prices increases was stopped in late 2003 inthe run-upto the April 2004 elections. Inthe power sector, the unbundling o f the generation, transmission and distribution functions o f the state-owned CEB into independent companies was started, but is yet to be completed. In transport and communications, the extemal gateway monopoly o f Sri Lanka Telecom (SLT) was ended and additional shares divested, bringing about a 60 percent decline inlong distance tariffs. Under the Sri Lanka Railway Authority Act, the loss-making Railways Department was moved out o f the consolidated budget, albeit amidst strong political opposition. Attempts to privatize the Sri Lanka 26 Sri Lanka Development Policy Review Transport Board through the stock exchange failed to attract interest from credible buyers due to the inflexibility o f sale conditions. AgriculturalLand 2.11 The incomplete transfer o fproperty rights o f state lands alienated to farmers and Sri Lanka's inefficient land administration system are major impediments to promoting investments in the rural areas and improving productivity in agriculture. Thus, the thrust o f recent land reforms has been to: (b) remove restrictions for sale, mortgage, lease and transfer o f land permits and grants alienated to farmers under the Land Development Ordinance (LDO), involving about 1.4 million hectares (35 percent o f arable land); and (b) improve the cost efficiency, public confidence and transparency inthe land administration system. 2.12 A Land Ownership Bill -to convert restricted land grants to full private ownershp-was presentedto Parliament inNovember 2003, but was challenged inthe Supreme Court who ruled that the Billviolated a provision o fthe 13thamendment to the Constitution." Inview o fthis, work started to amend the LDO upholding the legal mandate o f the Provincial Councils (PCs) in land administration. In addition, draft amendments to the 1988 Registration of Title Act (RTA) were prepared and activities initiated to develop a policy/institutional framework for the combined implementation o f RTA and the conversion o f grants to free hold title. Efforts also aimed at consolidating all agencies dealing with land administration 23under a single Ministry to reduce the administrative burden. An important first step inthis direction was the appointment o f the Registrar General o f Lands as the Registrar General of Title and the enactment o f the Survey Act, which takes into consideration landtitling requirements. 22According to this, the Provincial Councils have the authority to deal with land allocation matters. 23The Survey Department, the LandSettlement Department, the Registrar General's Department, and the Commissioner of Lands Department. 27 Sri LankaDevelopment Policy Review Labor Market 2.13 Sri Lanka's labor market has been constrainedby costly processes involved inresolvinglabor disputes and retrenchment. Under the 1953 Termination of Employment of Workmen's Act (TEWA) -amended in 1971-firms with more than 15 workers are required to obtain the written consent o f the employee and/or the permission of the Labor Commissioner for the termination of the employee on non-disciplinary grounds. By 2001, labor disputes took on average 440 days to resolve and labor tribunals had accumulated a backlog of 18,000 cases. Sri Lanka's politicized unions and often entrenched management styles have strained industrial relations, further reducing chances o f quick resolutions. Employers have responded by limiting expansions and job creation, suppressingbase wages, and resorting to outsourcing and subcontracting. Uncertainty and high retrenchment costs make it difficult for f m s to plan exit costs and have diminishedSri Lanka's ability to attract foreign investment outside the boundaries of the FTZs which, as explained elsewhere, have been subject to a less restrictive labor regime?4 2.14 To begin addressinglabor market rigidities, TEWA was amendedto require the application of a compensation formula (effective January 2004) for job termination cases. The Industrial Disputes Act was also amended, setting timelines for tribunal cases, arbitration cases and termination applications. Although severance costs under the reformed TEWA remained very highby world and regional standards, they were made more predictable (and somewhat lower) than in the previous regime. However, the systemremained subject to arbitrariness as the labor commissioner's approval of retrenchment applications was still required. Shortly after assuming office, the UPFA's government suspendedthe formula which is now beingrevised. Financial Sector 2.15 The financial sector improved considerably during the past two decades reflecting a strengthened regulatory and legal h e w o r k and a declining share o f state-owned financial institutions. Nevertheless, two state-owned banks still dominate the bankmg system, the Bank o f Ceylon (BC) and People's Bank (PB), accounting for 46 percent of the banking assets (74 percent in 1980). After two recapitalizations inthe 199Os, the PB is insolvent andbothbanks' continue to have highnon-performing loans (15-20 percent). Highintermediation costs (associatedto overstaffing and loss-making branches) have contributed to high cost of funds. Political interference, including in lending decisions, account for these financial weaknesses. In view of this, options were initially developedto commercialize the PB (andthus strengthen its governance structure), including through a sale of amajority equity staketo a strategic investor. However, the UPFA government has ruledout the privatization option and plans to rely on restructuring to deal with the banks' problems. Meanwhile, the financial position o f the banks i s again under considerable strain owing to increased lendingexposureto the state-owned CPC and CEB which, due to delayedprice adjustments (for fuel and electricity) are incurring substantial losses. Welfare System 2.16 The persistence of poverty underscores the importance o f well-targeted safety nets in mitigating poverty and vulnerability. Although Sri Lanka has a long standing welfare system with a mandate to play this key role, its impact on reducing poverty and vulnerability has been below potential owing to a number o f problems, including overlapping objectives (and coverage) and mis- targeting. Notably, due to a highdegree o f politicization, the Samurdhi program (the largest o f the safety net costing 0.5 percent o f GDP in2003), covers almost half of the population but misses about 24 For instance, CalPers, the largest USpension fund decided not to invest in Sri Lanka in2002 citing labor market rigidities as a bigger concem than the political instability. Sri Lanka has generated large employment opportunities and attracted FDIinthe FTZs where labor unions have had a limited presence. 28 Sri Lanka Development Policy Rwiew 40 percent o f households from the poorest quintile, while almost 44 percent of its budget is spent on householdsfrom the top 3 q~intiles.*~ 2.17 The Welfare Benefit Act enactedin October 2002 provided the legal framework for reforms aimed at rationalizing the selection o fbeneficiaries underall transfer programs inthe country, starting with Samurdhi. The Welfare Benefits Board (WBB) establishedunderthe Act-has completed field testing the application form, finalized the formula for determining eligibility o f Samurdhi beneficiaries, andinitiateda process of consulting stakeholders. Next steps include the completion o f the enrolment phase o f the pilot, followed by preparations for the island-wide application and targeting process. This would need to be complemented by a widespread publicity campaign to generate awareness about the rationale and rules o f the new system, before it becomes operational. The full implementation o fthe Welfare Benefit Act are expectedto: a) limit the Samurdhi programto the bottom 30 percent of the population, b) improve the selection and identification o f the poor to be coveredbywelfare programs through aproxy means test formula, c) make transfers more progressive, i.e. giving larger amounts to poorer households, and d) remove overlap by other public welfare programs. Ifimplemented as planned, this reformwill have direct positive implications inthe lives of poor families, especially o fthose currently excluded fromthe welfare system. Box 2,4: luipreving the 'I'rrgeting dWelfare Pmgrams An c>bjeC%ivecriteria f i &e selection of beneficiaries under the Samwdhi transfer prograni has been developed, involvrng memas tha%assigns "scores" to householdsdependingon certain ~~~~~~~~~~~~ easily ohvable ch tics, like h o t ~ ~ h ~ o~ ~~~ p h iownership of assets and durables, and d o c s , quality ufhousing. In ~~~~~~r~~a ~ ~ ~ p u ~ r wilt~be~usedto inmitor e l ~ ~ ~ bentryl ~andyexit database i d ~ t ~ andcrosscheck for benefitsreceivedfronr multiplesources, toniininiite ~up~ica~ions p m p " across To test the p r u x ~ n ~testedn~sp ~ l ~ cand~selectionpr(mssfbr beneficiariesot`the Samurdht program, ~ ~ a ~ o n a pilot covering 4 $ , householdsin 114 Gram N~fadhmdivisicms was conducted, and its results were ~ ~ ~ discussedat a ~ oorga&&~ by thekWElB in~November2003,~wlrl?~ ~ a r ~%mi all~stakeholder~ ~ ~ ~ ~ i o ~ ~ ministries. The workshop concluded with Bnumber of opepntiond r e c ~ ~ r ~including limiting ~ ~ ~ ~ , ~ ~ d a ~ ~ e l t ~ i b j ltotthe poorest 25-28 percent of the population ~ ~ I in comparisonto the current coverage ufnearly 50 percent of the population; and a payment scheme that combines a fixed aniount for all eligible households with an anzourit per vulnerable member of the h~?use~u~(~ (chifdren, disabled, elderly) Sepaating &e ~~~u~~ mnsfers &.om the:mmpulsary savings, inswame and lottery ~hemes,was also ~ ~ ( ~ ~ ~ t ~ e ~ d e ~ . Siniulation exerci 2.18 Other major reforms initiated during the period include the areas o f banking, insurance, public administration and expenditure management.(refer to Box 2.1). Towards SecuringPeace 2.19 Although peacenegotiations have stalled since April 2003, the cease fire agreementhas held for over two years now and progress has also been made in other peace-related fronts. Notably, Parliament passeda law inNovember 2003 that conferred citizenship to about 170,000 EstateTamils, thus putting an end to fifty years of statelessness. In addition, an action plan was developed to address cases of childrecruitment for military activity. In2002, the road to Jaffna was reopenedafter 12 years, the economic embargo on LTTE-heldareas lifted, and large tracts of land de-mined. This has facilitated the returnto their homes o f some 345,000 (Internally Displaced Persons (IDPs) (about 47 percent o f the internally displaced population). Donor support (about US$430 million) was 25 See Glinskaya (2000) for a detailedevaluation ofthe Samurdhi program 29 Sri LankaLkvelopment Policy Review coordinated to finance post-conflict assistance and reconstruction in the North East. Damaged schools (148) and health care centers (115) have either been repaired or are in the process of being repaired. School drop-outs (20,000) have returned to school inthe NorthEast and catch up education is provided inall districts. Landunderpaddy cultivation inthe Northincreasedby 35 percent leadmg to five-fold increase in the winter harvest o f 2003. The total fish catch in the North East almost doubled inthe same period. B. Early Gains from Reformsand Peace 2.20 There have been some clear indications of initial gains from the implementation of the strong economic reformprogam andcontinuation o fthe cease-fire. 0 Economic growth resumedwith real GDP growth reaching 4 percent in 2002 and 5.9 percent in 2003. Growth in 2003 was broad based as the textile-led industrial sector recovered from the 2001 downturn, agriculture benefited from good weather, and economic activities in the North East resumed. Notably, tourist arrivals hita hstoricalrecordo fhalf amillion. 0 The inflation rate fell from 14.2 percent in 2001 to 6.3 percent in 2003; nominal interest rates were halved; andprivatecredit increased. Unemployment moderated from an average o f 8.8 percent in 2002 to 8.1 percent in 2003, although it remainedabovethe 7.9 percent recorded in2001.26 0 Foreign direct investment, which had slumped to US$ 82 million in2001, rebounded to US$242 million in 2002, but having reachedUS$ 170 million in first half of 2003 closed the year at US$ 229 million, amidst increasedpolitical uncertainties. 2.21 While noteworthy and welcome, these gains remain fragile as the peace and reform agendas are still unfinished. Onthe peacefront, while the cease fire is holding, negotiations have beenstalled since April 2003. Inthe economic arena, initial gains in fiscal consolidation-which contributed to declining inflation and interest rates-have been erodedas a result of weak revenue performance and other additional fiscal pressures in the run-up to the April 2004 Parliamentary elections (including large subsidies on petroleum and fertilizer against rising world prices). Moreover, most o f the structural reforms (much o f them legislative) that were launched during the period have yet to be implemented. Against this backdrop, it is rather disquieting that aggregate investment continued to be at a level (22 percent o f GDP in 2003) significantly below the levels seen in the late 1990s (27-28 percent). Private investment has not rebounded back from the trough o f 2001. In this context, the political uncertainties sinceNovember 2003 may have exacteda significant toll on the confidence and intentions o f both domestic and foreign investors. In case o f the latter, the BO1 has instances of severalmultinational firms whichhaveplacedtheir investment plans for Sri Lanka onhold. Table 2.1: RecentTrendsin Investment (Percent of GDP) Investment Source :CBSL (2003) 2.22 Ofparticular concern is that this downturn inthe private investment to GDP ratio occurs at a time when export growth is showing signs o f deceleration. As shown inFigure2.1, during the early 199Os, Sri Lankanexports were growing more rapidly than world trade. However, having peaked at an annual growth rate o f over 10percent (on a five-year averagebasis), export growth has been on a downward trend since the mid-1990s and by 2002 it was approaching an annual rate of 3 percent, below world trade growth, indicating that the country has been losing market share. Being a small 26 Excludingthe NorthEast 30 Sri Larzka Devclopnient Policy Review openeconomy, Sri Lanka's growthprospectsdependcritically onthe country's ability to sustainrapid export growth. Therefore, while some of the recent deceleration of export growth may reflect temporary lingering effects from the 2001 shock, the downward growth trend suggest the presence of longer-term factors undalyingsuch decline. Figure2.1: RealGrowthof Sri LankaExportsand World Trade +World Trade(m realterms) +Ekports o fG&S (inrealterms) 2.23 It is also noteworthy that the gains made over the period did little to impress important segments of the population, particularly the rural poor, and the last elections turned power over to the UnitedPeople's Freedom Alliance (UPFA)-comprising the SLFP, the JanathaVimukthi Peramuna (JVP), and other political parties. Inadequate communication and consultation mechanisms in connection to economic reforms and the peace process, may have also been a factor in the election outcome. While the details of the new government's economic reformprogram are still being worked out, it is clear from their manifesto that there will be areas o f continuity as well as of departure from previous regimes. Notably, the UPFA will rely on restructuring rather than privatization as a means of improving the performance o f remaining strategic state-owned enterprises. The new government has also pledged to give greater emphasis to addressing regional inequalities and fostering rural development. The sections that follow address some o f these difficult challenges and raise a number o f key issues that would need to be considered in defining a more pro-poor growth strategy for Sri Lanka. It i s noted from the outset that such a strategy must combine investment, i.e., inthe transport and power sectors, as well as policy reforms to increase farmers' incentives and rural productivity. 31 Sri LankaLkvelopment Policy Review III.TowardsFasterandMoreEquitableGrowth 3.1 Democracy and competitive politics frequently lead to a change o f government. The Sri Lankan general elections o f April 2004 have demonstratedthat again. The UnitedPeople's Freedom Alliance (UPFA) composed of the SLFP and its new ally the Marxist-oriented JVP leads the new government o f the country. The underlyingpolitical situation remains complex, with the recent split in the LTTE and the increasedparliamentary strength of the JVP and the Tamil National Alliance constituting important new factors in the equation. Whichever government i s in power, the basic agenda o f peace, rapid development and poverty reduction are shared goals for the people o f Sri Lanka. Sustaining the cease fire and securing durable peace are quintessentially political challenges on which this report must remain silent. The discussion inthis section focuses on a selection of key medium-term development issues and challenges that confront the new government (indeed, any government) of Sri Lanka as it strives to meet the aspirations of the people for faster economic growth, morejobs andlesspoverty. 3.2 Achieving faster growth and poverty reduction will be a challenge given the counby's structural weaknesses, including: high fiscal deficits and public debt, limited export diversification (against the backdrop o f the imminent abolition o f the MFA from January 2005), stagnating agriculture, serious infrastructure bottlenecks, and declining quality o f human capital due to falling quality o f education. Addressing the structural weaknesses will require containing the burgeoning fiscal deficits, diversifying the export base, realizing the growth potential o f non-plantation agriculture, strengthening the road andpower infrastructure, and revamping the education system. It will be also essentialthat inmanaging the economy, policy reversalsbe avoided and that gains from pastreform efforts be sustainedandfurther strengthened. 1. ManagingPublicFinances 3.3 Despite significant gains in fiscal consolidation between 2001 and 2003 and the milestone enactment of the FMRA, the fiscal situation remains under substantial stress. In 2003 the fiscal deficit was 8 percent o f GDP and the public debt stood at almost 106 percent of GDP, a long way from the FMRA targets for 2006 (of 5 percent and 85 percent o f GDP, respectively). Interest payments still accountedfor 7 percent o f GDP, absorbing more than halfo ftotal tax revenue andwell in excess of the expenditure on education and health (4 percent of GDP) or public investment (less than 5 percent of GDP). Highinterest payments reflectedhighpublic debt andrising dependence on relatively more expensive domestic financing as compared to concessional external finance, particularly duringthe 1998-2002period. O fpossibly even greater concern was the poor performance of tax revenue, which dropped to a low o f 13.2 percent o f GDP, despite havingbeenbudgeted at 14.8 percent o fGDP and despitethe strong recovery o fthe economy in2003. Figure 3.1: Interest Payment to Tax ratio ant he Nominal Interest Rate on Debt , Interest Paymentsto Tax Revenues 10 I----- i 7 4 1 I , 1998 1999 2000 2001 2002 2003 .,", 1998 1999 2000 2001 2002 2003 +Nominal Rate +Domestic Borrowing 32 Sri LankaDevelopment Policy Review 3.4 The Budget 2004 had targeted a fiscal deficit o f 6.8 percent o f GDP and tax revenues o f 14.5 percent o f GDP. ThePre-Election Budgetary Position Report released inMarch 2004 (the first issued under the FMRA obligations) revised the deficit upwards to 7.3 percent o f GDP, mainly on account o f a weaker primary balance due to reduced revenues. The implications o f these adverse trends for achieving FMRA targets for 2006 and beyond for the budget deficit and govemment debt to GDP ratio are explored below. . . /*Budget as revisedper Pre-ElectionReport A. FiscalandDebtSustainabilityAnalysis 3.5 The fiscavdebt sustainability analysis for Sri Lanka shows that achieving the FMRAmedium and long term targets will be very challenging, especially given the recent increase on fiscal imbalances. Such targets include: (a) a fiscal deficit not to exceed 5 percent o f GDP by 2006; (b) total public debt o f 85 percent o f GDP by 2006; and (c) government debt not to exceed 60 percent o f GDP by 2013. Basedon the estimated fiscal position in2004, the model used for this purpose projects the path for the primary balance that would be required to meet the FMRA targets, using baseline macroeconomic assumptions and a consistency framework that allows for sensitivity analysis using altemative assumptions on growth and interest rates. 33 Sri Lanka Developmenr Policy Review Table 3.2: BaselineMacroeconomicAssumptions 2004 2005 2006 2007 2008-2013 GDP growth(%) 6.0 6.3 6.5 6.3 6.0 Change inGDP deflator (%) 7.0 6.0 5.0 5.0 5.0 Nominalinterestrate on public debt* (%) 5.9 6.0 6.0 6.0 6.0 *Externalfinancing (as % of GDP)** 0.4 2.8 1.8 1.8 1.8 Average interest rate that applies to total govemment debt, expressed innominal terms. **Net Extemal financing, including grants. Note: The real exchange rate is assumed to appreciate mildly during 2004-2006 andto remain constant thereafter. 3.6 As shown inFigure 3.2, a zero primary balancebetween 2005-2013, will be required to bring the fiscal deficit down to 5 percent o f GDP by 2006 and public debt to 60 percent by 2013. (Under this scenario, public debt is projected at 88 percent o f GDP in 2006, somewhat above the FMRA target). This is based on an estimatedprimary deficit o f 1.3 percent of GDP in2004, as envisaged in the pre-election report, but it should be noted the deficit i s currently estimated at about 3 percent, implyingthat a M e r adjustment wouldbeneededto remainwithinthe FMRAtargets. Primary Balance I Pub'icDebt Interest Payments Overalldeficit 3.7 The projected paths o f the primary balance and debt ratios are consistent with lower fiscal deficits and interest payments on public debt. Starting with interest payments on public debt equivalent to 6 percent o f GDP in2004, fiscal savings equivalent to 1percent of GDP would accrue by 2006, and 2.6 percent of GDP by 2012-2013. Total interest payments on public debt would stabilize at around 3.5 percent o f GDP by 2012. Further, the overall fiscal deficit in this scenario complies with the 2006 deficit ceiling set by the FMRA o f 5 percent o f GDP. Moreover, the overall deficit would fall by nearly 4 percentage points o f GDP between 2004 and the end o f the projection period. 3.8 Contingent liabilities. These calculations do not take into account the impact o f explicit and implicit contingent liabilities." Inthe case o f Sri Lanka, such liabilities could include liabilities of state-owned enterprises (SOEs) and obligations associated with the pension system. Over the past two years, some measures have been taken to improve the financial position of SOEs, but a number o f *'Explicit liabilities include state guarantees, legal disputes, and uncalled capital. Implicit liabilities include the risk of default, liabilities of entities being privatized. 34 Sri Lunku DevelopmentPolicy Review enterprises continue to incur operating losses, have large debts to the financial sector, and carry substantial labor liabilities.28 On pensions, the recently introduced contributory pension fund, may reduce the immediate fiscal burden, but long term sustainability is not guaranteed. Also, the cost o f the non-contributory pension scheme may still increase by 0.5-1 percent o f GDP, from about 2 percent o f GDP, over the next ten years. Under the circumstances, it would be prudent to begin providing for the fiscal risks posed by contingent.liabilities by targeting a primary surplus o f around 1.O percent o fGDP. 3.9 Sensitivity Analysis. The extent o f the fiscal adjustment i s highly sensitive to changes inkey macroeconomic assumptions (e.g., real GDP growth, the nominal interest rate, and the exchange rate. For instance, the baseline scenario assumes that annual real GDP growth stabilizes at 6 percent inthe long-run. However, ifreal GDP growth remained at its historical level of 5 percent, cetevisparibus, a primary surplus closer to 0.5 percent of GDP (as opposed to a small deficit) would be required to reach a debt ratio o f 60 percent in 2013. Similarly, the base line scenario assumes an average nominal interest rate o f 6.0 percent, but with a one percent increase in the interest rate, a primary surplus equivalent to 0.5 percent o f GDP on a permanent basis, would be required to bring the debt ratio down to 60 percent by 2013. Table 3.3: Target Primary Balance and SelectedLong-Run Variables RealGDP Growth(2007-2013) Annual Percentaxe Rate - 4.0 5.0 6.0* 7.0 8.0 Target Primary Balance** 0.5 -0.5 -0.9 As percentage of GDP 0.9 0.0 Nominal Interest Rate(2007-2013) PercentageRate 1 4.0 5.0 6.0* 7.0 8.0 Target Primary Balance** -0.9 -0.5 0.0 0.5 0.9 As percentage of GDP *Baseline macroeconomic scenario. 3.10 TheImpact of Fiscal Slippage. The Pre-election Report revisions to the 2004 Budget already represented a deviation from the medium term fiscal objectives set out under the FMRA. Unfortunately, additional fiscal pressures @oth on the revenue and expenditure side) inthe run-up to the elections have ledto a M e r deterioration o f the fiscal position and the gains made inthe last two years have been eroded. For instance, with primary deficit returning to levels seeing over the past decade, averaging 3.1 percent o f GDP, it is unlikely that the economy could grow above the historical of 5 percent annually and foreign external financing of the budget would probably decline inresponse to the fiscal slippage. Under these circumstances, the interest-to-GDP ratio will remain highat 6-8 percent over the next ten years, absorbing over 40 percent o f total revenue, even assuming a sustained increase inthe tax-GDP ratio o f 2-3 percent o f GDP (see Figure 3.3 below). 3.11 The above analysis illustrates the vulnerability o f the fiscal situation to several factors, including macroeconomic assumptions and policy shocks. Such vulnerability suggest the need to target a primary surplus inthe order o f one percent o f GDP inthe short to medium term, if Sri Lanka i s to acheve a sustainable fiscal framework. 28 Reportedly,all public enterprises inSri Lanka are over-staffed. 35 Si-i Lanka Development Policy Review Figure 3.3: Interest Payments as a Percent of Total Revenues Returnto Pre-2002 Policiesvs. FullCompliancewith FMRA* 15: 10 ! 1 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 *FMRA -cPre-2002 *The "FMRA" adjustment scenarioleads to compliance with the 2006fiscal deficit target (5 percent of GDP) andthe 2013 debt/GDP target (85 percent). B. PriorityFiscalReforms 3.12 A sustainable fiscal stance remains the cornerstone o f any viable growth strategy. Despite recent gains, further fiscal consolidation will be neededto keep inflation and interest rates down, and ensure greater resource availability for private investment, which as mentioned earlier, has been slow to recover. Inturn, the stimulus to growth o f an environment of increased fiscal discipline, will help narrow fiscal imbalances. The need to tighten fiscal policy in the medium term, against mounting needs for investment in economic and social infrastructure, makes it compelling that scarce public resources be usedmore effectively and efficiently thanhitherto. Key priorities include: Reducing the PublicDebtBurden 3.13 As discussed earlier, a key challenge for Sri Lanka is to bring the public debt to sustainable levels and thus reduce the crowding out o f other (more productive) spending by the public and private sectors. This will require a steady reduction in fiscal imbalances and in domestic financing o f the budget deficit. Inthis context, medium term limits on public debt and the budget deficit, such as those providedby FMRA, needto be an integralpart o f any meaningll fiscal framework. ReformingPayand EmploymentPolicies 3.14 Although the size o f the civil service wage bill is not at present an unmanageable burden for government, the trends are worrisome. Between 1990 and 2001, public sector employment grew at 3.6 percent annually, outpacing growth in population, labor force and real GDP (Statistical Annex Table A29). At the same time, the share o f the wage bill in total revenue has increased (from 26 percent to 34 percent), suggesting that the wage bill is crowding out other government expenditure. Moreover, rapid growth inpublic employment (particularly at lower grades) has strained the ability o f government to offer competitive salary scales for skilled, well-performing employees, with adverse consequences on the quality o f services. Given overstaffing, low salary compression and serious admmistrative duplication and fragmentation, there is clearly ample scope for wage and employment rationalization without unduly increasing the wage bill burden. However, although these problems (and their solutions) have been known for some little has been done to address them. Instead, successive governments have been unable (or unwilling) to de-politicize wage and employment 29 Indeed, earlier reports, notably the Davendra Commission, had identified the need to reduce the overall size o f the public service, but also to improve public sector salaries. 36 Sri LankaDevelopmentPolicy Review policies, particularly in the context o f short election cycles and budget horizons. 30 It is therefore essential that governments explicitly consider the medium term implications o f alternative pay and employment policy options and adopt a set o f consistent measures that could be sustained overtime despite changes inleadership. ImprovingExpenditureManagement 3.15 Like many countries elsewhere, meeting debt sustainability andother macro/ micro objectives requires adopting a medium term approach to budgeting andplanning, where the cost andbenefits o f particular policies can be properly assessed. For instance, provided appropriate strategies, a longer term horizon would permit a reallocation o f funds from salaries and debt service (which appear fixed in the short term), toward other expenditure needed to attain desired development outcomes. The predictability o f the resource constraint is a pre-requisite for effective budgets, especially when a primarypolicy objective is to achieve a given deficit target. This, inturn, requires increased realism inrevenue estimates, including external donor financing, anda stable tax policy framework. 3.16 Sri Lanka's one-year budget system where allocations are done incrementally @revious year's budget plus or minus) provides little scope for policy makers to link the budget with government's strategies/policies. However, initial steps have been taken recently to address these shortcomings. These include (i) the adoption o f budget ceilings since 2003 (as opposed to bottom-up budgeting) consistent with overall resource availability and the (ii) introduction o f a Medium Term Budget Framework (MTBF), starting with the 2004 budget. The next step is the introduction o f a strategic phase in the budget formulation process, involving greater consultation among government agencies, including Cabinet, on the allocation o fbudget ceilings. 3.17 Intra and Inter-Sectoral Resource Allocation. Within the parameters set at the macro level, the allocation o f public expenditures (including humanresources) needs to be closely linked to sector policies and objectives. However, the current administrative fragmentation (and duplication) o f government agencies and functions makes it diEcult to identify policy priorities, let alone public service outcomes, associated with public spending. Moreover, there are often inconsistencies between stated policiedstrategies and/or serious misalignments between priorities and funding levels. In addition to the establishment o f a more strategic and consultative budget foxmulation process, expected to underpinthe MTBF, a shift toward sector ceilings (as opposedto a multiplicity o f ceilings for individual spending units), could help increase the consistency between policies, resources and public service outcomes withm sectors. It would also permit policy makers to consider more explicitly the potential trade-offs from shiftingresources within and across sectors. Managing SOEs and Contingent Liabilities 3.18 The fiscal burden o f SOEs in Sri Lanka is highand service delivery is poor. For instance, at nearly 3 percent of GDP, direct budgetary transfers to SOEs are in excess o f the entire allocation to the education sector. There are also large contingent liabilities which are increasing as a result of political interference (e.g., delays in the adjustment o f fuel and electricity prices) and slow restructuring. Further efforts are needed to improve the financial viability o f SOEs, particularly o f enterprises with large labor and financial liabilities (such as CEB, CPC, the Railways Department, and the state-owned commercial banks), by imposing a hard budget constraint, reducing political interference, and allowing these firms to operate on a commercial basis. The Government also needs to develop regular institutional mechanisms to monitor and record explicit and implicit contingent liabilities. Pensions' policy is also an area that would need close monitoring, especially ensuring the long-term sustainability o fthe recently introduced contributory pension fund. 30For example, the newly electedgovernment has pledgedto provide employment to 30,000 new graduates within three months o ftaking office. Inaddition, there are risks that wage increases granted to health sector laborers early inthe year could be extended across the civil service inresponse to union pressure. The latter will increase the wage bill by over 1 percent o f GDP and further reduce the salary compression ratio. 37 Sri Lanka DevelopmentPolicy Review Strengthening Revenue 3.19 A key priority for Sri Lanka is to increase tax revenue within a stable tax policy framework, so that revenue estimates becomemore realistic. Muchwill dependonthe stability and consistency o f tax policies, including measures to raise the yield o f major taxes such as the VAT and income tax. These issues are discussed inmore detail inthe following section. 2. Tax Policiesto Promote Growth and Fiscal Consolidation 3.20 The main challenge in crafting tax policies in Sri Lanka i s to reverse the massive decline in the tax-GDP ratio (fiom 19 percent in 1990to 13 percent in2003) and to do so ina manner which is supportive o f growth and efficiency. The challenge is daunting, given the sharp drop in trade taxes from 6 to 2 percent o f GDP since 1990, stagnation o f income taxes at around 2.5 percent o f GDP and the incomplete transition from a complex system o fturnover taxes and special levies to a consumption VAT whichhas lost at least one percent o fGDP (Table 3.4). 3.21 Early projections indicate a needto raise the tax-GDP ratio to the about 16-17 percent o f GDP by 2006 in order to meet the FMR goals. Inprinciple, such revenue increase could be achievable within a broad commitment to the goal o f a simple, broad based and transparent tax system, which includes: (a) a simplified income tax on a wider base, keeping concessions to a minimum, (b) a uniform, low customs tariff duty, (c) a well-focused set o f excise taxes and (d) a l l l y consolidated, single rate retail level VAT on the widest possible base. Clearly, tax policies and their administration need greater attention ifthe tax-GDP ratio is to be raised to at least 16 percent o f GDP by 2006. The broad direction o f policy changes is indicated in the ensuing brief discussion, organized by major taxes. 38 Sri Lanka LkvelopntentPolicy Review Income Tax 3.22 Sri Lanka's income tax performance, at less than 2.5 percent o f GDP, is quite weak by international standards. Comparable figures for recent years are 4.7 percent for Korea, 6.9 percent for Malaysia, 6.2 percent for Philippines, 4.8 percent for Thailand and 3.4 percent for Pakistan. The reasons include the provision o f long term tax holidays to companies under the Board o f Investment (BOI) regime (see Box 3.4), a long history o f investment allowances, concessionary rates, generous depreciation allowances, as well as the exclusion o f most public sector employees from the personal income tax net andweaknesses intax administration. 3.23 The economic service charge (ESC) recently imposed on all entities with a turnover inexcess o f Rs.20 million or total assets inexcess o f Rs.10 million, should bringa large number o f commercial entities into the tax net. Another important proposal is to require all companies and partnerships, or any other business registered with the Registrar o f Companies or the Registrar o f Business Names, to register and file annual returns with the Departmento f InlandRevenue regardless as to whether they are liable to pay income tax or not. All persons liable to deduct withholding tax must also required to register with the Department o f Inland Revenue and h s h monthly returns. These measures, together with the new EX, are likely to substantially broadenthe base for business taxation. 3.24 In2003, the BO1Law was amended to limit tax concessions accordedto BO1companies to those stipulated in the Inland Revenue Act. This is a step in the right direction, provided the amendment is enacted. A more radical option would be to declare a moratorium on tax holidays, including for new foreign investment. Arguments against such a proposal are: first, the revenue yield i s likely to be low since transnational companies can manage their accounts to show low profits in their Sri Lankan operations; and second, the disincentive effects o f such a measure could be substantial since it would bring BO1 enterprises into direct and regular interface with the revenue administration. On the other hand, it would be difficult to achieve a sustained improvement in the country's direct tax to GDP ratio if the large and rapidly growing BO1sector remains indefinitely out o f the tax net. 3.25 The present rate structure for business taxation i s complex. There is a strong case for unifyingthe company tax rate at 30 percent, with a single concessional rate o f 15 percent for export income. 3.26 With respect to taxation o f personal income, there is really little justification for excluding incomes o fpublic sector employees. This &stortion was introduced in 1978 andpartially corrected in 1997by bringingemployees incommercial enterprises into the tax net. Value Added Tax 3.27 The VAT was introduced in 2002 inplace o f the Goods and Services Tax (at 12.5 percent) and the National Security Levy (at 7.5 percent and on a wider base than the GST), with the objective o f reducing the cascading effect o f the sales tax. However, the creation o f two rates o f 10 and 20 percent contributed to complexities in tax administration and led to substantial revenue leakages, which explain in part the poor revenue performance in 2003. The unification o f the two rates into a single 15 percent rate in the 2004 budget was therefore a positive step. In addition to reducing exemptions, the next major step should be to extend the coverage o fVAT to pharmaceutical products, life insurance policies, supply of healthcare services and, most importantly, to retail trade. Excise Taxes 3.28 Over time the scope o f excise taxation has been narrowed down to a handful o f commodities such as petroleum products, liquor, cigarettes and motor vehicles. On both revenue and equity grounds there is a sound argument for expanding the coverage o f excise taxation to a number o f 39 Sri Lanka DevelopmentPolicyReview income-elastic, luxury consumer products such as air conAtioners, refrigerators, washing machines, and televisions. CustomsDuties 3.29 Since 1977, and especially after 1990, the Sri Lankan tariff structure has undergone a major reduction o f levels and compression o f tariff bands (Figure 3.4), reflecting the country's sustained commitment to trade liberalization. By 2000, tariffs were grouped into four bands o f zero, five, ten and 25 percent, except for some agricultural items kept outside the standard structure at 35 percent. Driven largely by revenue considerations, the policy direction changed somewhat since 2001. A temporary surcharge o f 40 percent (reduced to 20 percent in 2002). An additional rate band o f 2 percent was introduced in2002 and more than 1500 line items were shifted fromzero to this 2 percent rate. Some items were raised from 10 to 20 percent and others from 5 to 10 percent. Furthermore specific duties were reintroduced inplace of ad-valorem duties for selected agricultural and industrial products. The 2004 budget reduced the import surcharge down to 10percent, increased the minimum positive tariff from 2 to 2.5 percent andplaced a few items inthe 20 percent rate. Figure 3.4 :RecentSteps Towards a Low Tariff Regime 3.30 As a result o f sustained tariff reductions (at least until 2001), the average import duty collection rate fell from 19 percent in the mid 1980s (accounting for about 30 percent o f total tax revenues) to 4.6 percent in 2002 (11 percent o f total tax revenues). Another factor was the rapid growth o f the duty-free BO1sector and the associated decline inthe share o f dutiable imports (intotal imports) from nearly 70 percent inthe mid 1980sto below 30 percent by 2000. 3.3 1 Given the imperatives o f trade liberalization and the medium term objective o f a low, uniform tariff, a rise inthe ratio import taxes to GDP (2 percent o f GDP) i s unlikely inthe future. On the other hand, revenue considerations require no further decline in &us already low ratio. A near term tactic could be to increase the minimumpositive tariff rate to 5 percent to compensate for the abolition o f the remaining 10percent import surcharge. Ingeneral, the pursuit o f a uniform tariff objective should be consistent with revenue neutrality, at least for the next three or four years. Administrative measures to prevent misuse o f duty free concessions should also be deployed. 40 Sri Lank Development Policy Review ExciseTaxation OtherTaxes 0.3 0.0 Tax Administration 3.32 Sri Lanka's tax administration has been weak for several reasons including: the coexistence o f parallel regime (particularly for customs) and the existence o f legal provisions enabling the BO1to override InlandRevenue and Customs laws in granting tax concessions, the limited experiences with modern taxes such as VAT and Income Taxation which require verification, auditing and risk monitoring and lack o f modernization in terms of humanresources as well as technology. Reported arrears at end 2002 amounted to Rs. 69bn (7.1% o f GDP), consisting o f Rs.29 bnfrom Income Taxes, Rs.19 bn from GST and Rs.17 bn from Turnover Tax and National Security Levy. Although such amounts may not be collectable now consequent to the Inland Revenue (Special Provisions) Act o f 2003 which granted a tax amnesty, the recurrence o fthe default needs to be addressed. 3.33 Proposed administrative measures include: First, to establish a well integrated revenue administration along the lines o f the previously envisaged Revenue Authority, to realize the informational and administrative synergies from combining the two existing tax departments. Second, within income tax, there is a need to set up a separate tax audit unit for carrying out tax audits 41 Sri Lanka DevelopmentPolicy Review on non-performing tax files and issuing regular tax assessments. Third, the large taxpayer unit o f the Inland Revenue Department could be more proactive inpursuing potential revenue; only about 400 o f the 1100 or so files appear to be active currently. Fourth, there is substantial scope for much greater use o f National Identity numbers for identification purposes in significant commercial transactions andtaxation. 3.34 The above tax policy recommendations are summarized in Box (3.1). Taken together, they could yield a tax-GDP ratio o f almost 16percent in2006 and 16.5 percent by 2008 (Table 3.4). 3. External Sector Policies for Faster Growth 3.35 Sri Lanka's external sector policies since 1977 (trade, exchange rate, foreign investment, emigration) have been a major success. The rewards to the (broadly) consistent and sustained liberalization o f these policies have been very substantial. The returns are likely to have been much greater if not for the protracted civil conflict, which has weakened the investment climate and macroeconomic stability. On the other hand, there is little doubt that the demonstrated resilience o f the economy (outside the North East) inthe face o fprolonged stress andoccasional shocks has largely been due to the greater economic flexibility impartedby a liberal trade andinvestment policy regime. 3.36 Quite obviously, the most important single "policy" for injecting more dynamism in trade, investment and growth is the achievement o f durable and lasting peace in the country. That said, there is scope for improvement in existing external sector policies that can contribute to Sri Lanka's development and its ability to cope with a changing external economic environment, especially as the country prepares to face the abolition of MFA inJanuary 2005. Trade and Tariff Polices 3.37 Sri Lanka's average (unweighted) applied tariff halved between 1995 and 2002, from 20 to 9.4 percent. Over the same period, the collection rate fell from 7.5 to 4.6 percent. These numbers are the lowest in South Asia by substantial margins, reflecting the much greater distance that Sri Lanka has traveled on the path o f trade liberalization (World Bank, 2003). Indeed, the average tariffrate in Sri Lanka in 2002 was comparable with that o f Indonesia, Malaysia and Philippines and somewhat lower than that o f China, Thailand and Vietnam. However, the degree o f dispersion o f tariff rates was greater in Sri Lanka, reflecting the larger number o f tariff bands (after 2001) and continued higher tariffs on agricultural products. 3.38 Official estimates indicate that the effective rate o f protection (ERP) for import-competing manufacturing declined from 70 percent in 1991 to 56 percent in2002 (SA Table A27). The decline inERPwas much smaller thanthe fall innominal protection because o f an increase inthe degree of tariff escalation over these years. Although there are no official estimates o f ERP for export oriented production, rough estimates suggest that the ERP for export production rose from 30 percent in 1991 to 37 percent in 2003 (Athukorala, 2004). These trends point to a significant reduction in the incentive bias against exporting during these dozen years. 3.39 Since 2001, the tariff structure has become somewhat more untidy because o f the new tariff surcharge, reintroduction o f specific duties for agricultural products, a new tariff band o f minimum duty, some proliferation o f ad-hoc duty exemptions and case-by-case duty adjustments and the introduction o f preferential tariffs under the South Asian Preferential Trade Agreement (SAPTA) and the India Sri LankaFree Trade Agreement (ISLFTA). 3.40 The government needs to recommit its tariff policies towards a medium term goal o f a single low, uniform tariff. Inthe interim, the efforts should focus on clustering existing tariff bands towards a central tendency, eliminating ad-hoc exemptions and rates and gradually reducing agricultural tariffs. This program will have to contend with the problems posed by a growing number of preferential trading arrangements. 42 SriLanka Development Policy Review PreferentialTrading Arrangements 3.41 Sri Lanka appears to have embraced the newfound global fondness for regional / bilateral economic cooperation. A bilateral free trade agreement was signed between India and Sri Lanka in December 1998 and it became operational in March 2000. In May 2003, Sri Lanka and India agreed to enter into a Comprehensive Partnership Agreement aimed at extending economic cooperation beyond trade into services and investment. A Trade and Investment Framework Agreement was signed between Sri Lanka and the US in July 2002 as part o f the on-going negotiations for forming a FTAbetween the two countries. Negotiations are also underway for entering into FTAs with Egypt, Bangladesh, Pakistan and a number o f other countries. Sri Lanka has also been a participant of three regional trading agreements: the Bangkok Agreement, the South Asian Preferential Trading Agreement (SAPTA), and the Bangladesh, India, Myanmar, Sri Lanka, Thailand Economic Cooperation (BIMSTEC 3.42 )agreement. b m 3.2: US Yri Lanka ETA The LIS has hen the largest sin& marker for Sri Lannka exports mer the past two decades. In 2002, the US acccfunted for 42% of total mmhandtseexpcwts with gmen?salone accounting for over f ~ ) ~of~eoaal ~ ~ I ~ s exports, On &6he impoia side, the US WBS the fifth largest source of irnpom.acco~~#~~ng for about.5% of tobl irnpmts. The mde balancc:between the two countries has been in Srt h k a ' s favor [in 2002: exports $1758 million, imports$218 million). It is believed that a FI'A with the U S is crucial ftw n ~ ~ ~ ngowth~ dynamism uf ~ x ~ ~ ~ ~ o r i e n t ~ d r ~ n ~ ~ ~ g n ~ ~ the eomWygi.A point~often naadeIS ~ in ~ ~ ~ Chat Yr?iLankaiis likely xu face s ~ ~ ~ ~ c oinmthe tJS~ i o n ~ ~ ~ market after abotitictr?ofthe MPA because over 50% of clothing iniportsto the country ctiine kom developing counrrses which enjoys concessionsunder FT'As or similar armgements, There ISstrongsupport for the p~opnseda ~ ~from~Sri rLankan exporters tnfgarrrrentsand other light n r ~ ~ ~ ~goods At ~the ~ aa~ l l e~ ~ ~ ~ r time, there is no oppo$it~o~~ i ~ ~ ( ~ ~ ~ ~ c odonieslic~~n~ga n u f a given~thesc ~ ~ ~ p ~ sof rmprts from n i p e t ~ u ~ o ~ ~ i o n from the US, there is no ptmthlthreat of import competition. Revenue inlplicartonsare also wtikety to be ~ nbecausemost of~die importsare duty free~(otherthantohaccct). ~ ~ n As in other ~ c e n t ~ y ~ ~ ~FI'As,o the~maineemphastsof8re 13sgovernment 1s on openingthe smices trade e ~ ~ a ~ d The agreement will providean o ~ ~ ~ to strengthenthe financial systems by bringingabout imprtwernents i ~ i t y into the financia1 ~ & ~ t ~ rInepreparation fix the negotiations, Sri Lank has already ltfied reartctions on . foreign s h ~ e h o ~ d ~the~financialsewices and insurance sectors For Sri Lank, the most diflcuis denwidto tn n meetisFQ kee capitaltrarx&rsinthe context ofthe investment regimeenvisagedunder the R A . Judging fmrn the ne~o~~ations ofpreviousPI'As, the US govemnenemaynot bewilling to concedeexcepttonsto the principle of &et?capital transfer Wlwhr the recent shift in intem;lrional cpgimon onthe issueof capital aceount opening indevelopingcountries(followng the financial crises in emergingmarketecrsnomies)would I&to achange in 3.43 The India Sri Lanka FTA came into effect in March 2000 and its scope was expanded in March 2003, when the number of items on the concession list tripled. Early results indicate substantial expansion o fbilateral trade (especially Sri Lanka's exports to India), although from a low base (Box 3.3). The FTA with U S is still under discussion (Box 3.2). Sri Lanka's espousal o f these two FTAs is drivenby bothpolitical and economic considerations. Closer economic cooperation with India i s considered essential for consolidating political relations between the two countries and thereby helping to resolve the civil conflict inSri Lanka. Close economic and politicalrelations with U S are also important for improving the international environment for the peace process and for locking in both preferential access to the huge US market as well as Sri Lanka's open trade and investment policy regime. 43 Sri LankaDevelopment Policy Review 3.44 So far, Sri Lanka's gains from participationinthe three regional trading agreements(SAPTA, Bangkok Agreement and BIMSTEC) have been negligible. By the later 1990s only 6.2% o f all tariff lines (at the HS six digit level) o f Sri Lanka's foreign trade had come under regional tariff concessions under the three agreements. The share o f intra-regional trade covered by these concessions in country's total trade was evenlower (less 3 percent o fimports andless than 1percentso f exports). tJntil the ISX,PYA cme in to force Sri Lankan imports outstripped exports by 20 times, but betwen 2000 and 2002 expcvrcs &om Sri Imka to India virtually q ~ d r u ~~'~ef~ren~jai ~ ~ . imports act-mintedFor 68% of totiat impoitsfrom lndiain 2002, upfiom 21% in 2001 f Kelegama,2003). Importshave inereas& at a slower rate, by about 40% pzcertt between these two years. This is tinderstandablebixawe ehe average prefrrencemargin (the dtfTmrnce between MFN and p&kren:nce &rimunder the ISX,ETA on the import side has been mai it, given &e very low MPNtarif& in Sri L,mka. Despiterapid expnt &rub%h,the balance oftrdde is still very much in India's favor, Srl l,anka fndia I'rade 1990-2882 - 3.45 While there may be substantial justification for the US and India FTAs, there seems to be much less rational for the other FTAs under discussion and negotiation. Nor is there much evidence to suggest that Sri Lanka can gain much through active participationin existing or proposed regional trade agreements. Indeed, there are good reasons to re-assess the apparent enthusiasm for proliferatingFTAs andregionaltrade agreements: 0 The PTA approach to trade liberalization i s becoming a major distraction to the process o f unilateral liberalizationwhich has servedSri Lanka so well inthe past 25 years. 0 Proliferation o f PTAs generates a serious problem o f multiple"rules of origin", which entail significant costs for both government andprivate economic agents. They complicate customs o f administration and weaken efficiency improvements inthe custom system. 44 Sri Lonkn I>rvcloprncnr Policy Review With multiple PTAs, Sri Lanka's tariff structure will become highly diffe~entiated,depending on the country of origin, thus compromisingthe goal of uniformity in tariffs and giving rise to new inefficiencies in resource allocationand specialization. Restrictions imposed on sourcing of inputs by rules of origin embodied in FTAs can have adverse implications for export competitiveness (for example, if Sri Lanka is required to source fabrics for garment exports from US). The MFA Challenge 3.46 By the end of 2004 the long standing Multi-Fiber Arrangement (MFA),which for decades has governed most of international trade in textiles and garments, is scheduled to come to an end. This could have massive and highly uncertain implications for production and exports of these products in many developing countries. Sri Lanka is no exception. Textiles and clothing (over 90 percent being the later) account for about half of Sri Lanka's exports, almost the same share of industrial production directly employs around 300,000 workers (5-6 percent of the labor force) and indirectly about one million workers (IPS, 2004). The industry employs mostly female workers (about 90 percent) and is concentrated in the Western Province (72 percent of the garment factories),which may partly explain the high concentration of growth and low poverty incidence in the area. 3.47 Among the many factors which will determine outcomes in the post-MFA world, will be the underlying competitiveness of the textileslclothng sectors of exporting countries and the extent of preferential (tariff) access to major markets, notably the US and the EC. The latter goal drives Sri Lanka's strong desire to conclude an FTA with the US, especially since over half of US imports of these products already occur under various preferential arrangements with other countries. 3.48 As regards the hdamental question of competitiveness, the outlook is somewhat mixed. On the onehand, there are several factors attestingto Sri Lanka's competitivenessin this key sector: The concurrent liberalization of Sri Lanka's trade and FDI regime seems to have provided Sri Lanka with an environment conducive for improvements in product quality, marketing slulls and rapid penetration in the brand-name dominated high end of global markets. Unlike other South Asian countries (where substantial segments of the domestic textiles / clothing markets are highly protected) the Sri Lankan industry has grown in a largely ffee trade regime for the past 25 years. Aside from the dominance of the FTZ system throughout, tariffs on textiles and clothing were mostly eliminated in 1998. Sri Lanka has been relatively successful in moving up to higher value segments of the market as reflected by increases in both volume and unit price (StatisticalAnnex Table A28). The share of non-quota markets in total Sri Lankan garment exports has risen appreciably (to about 40 percent) in recent years. In large part this reflects removal of quota restrictions in the EU market in 2001 (lPS, 2003). In addition, only one third of US quotas are still binding. Sri Lankan manufacturers have a good reputation abroad for adhering to internationally accepted labor norms as evident by Sri Lanka successfully obtaining additional GSP (20 percent in addition to standard 20 percent concession) from the EU based on labor compliance. Only Sri Lanka and Moldova have successfully obtained additional GSP concession to the EU based on labor compliance to date. There has been substantial consolidation already in the Sri Lankan garment industry, with about 10 percent of some 860 exporters accounting for about 70 percent of exports (IPS, 2003). 3.49 On the other hand, wages in Sri Lanka are somewhat higher than some South Asian competitors (e.g., Pakistan and Bangladesh) and labor productivity is lower than more efficient producers in East Asia and China. Moreover, several 'behind the border problems' such as poor inffastructure (power and transport), high cost of and low access to finance, and rigid labor regulations, adversely affect the cost of production and the overall competitiveness of the Sri Lankan Sri Lanka Development Policy Review garment industry. Electricity charges in Sri Lanka are the highest in Asia, with the exemption of Hong Kong. Leadtimes (time it takes from when an order is placed to when it is delivered) are also lengthy --around 100days inSri Lanka, whereasit is about 60 days inmore competitive countries. 3.50 While the medium term prospects for Sri Lanka's garment industry remain positive, there could be significant disruption in initial stages, especially among smaller exporters (who while representing 5 percent o f the value o f exports account for about two thirds o f all exporters and about 35 percent of employment inthe ~ector).~'To assistwith this painful adjustment, government policies should encourage consolidation, facilitate restructuring of credit to small exporters, improve infrastructure and trade facilitation and follow through with ongoing labor reforms that would facilitate enterprise restructuring. Inparallel, efforts need to continue to reach a favorable FTA with the U S andthereby improvemarket access conditions for Sri Lankan exports. Foreign Investment Policies 3.51 A liberal foreign investment regime has been an integral part of Sri Lanka's economic policies since 1977 and especially after 1990. The country's experience highlights the complementary role o f investment liberalization for reaping the potential gains from trade liberalization. O f the BO1 companies which started operations during 1978-2002, over two thuds involved foreign direct investment (FDI), mostly through joint ventures with local firms (80 percent of companies). Initially, there was a heavy concentration o f foreign investment in the garment industry,but since 1990therehasbeen anoticeable increaseofFDIinother labor intensive activities such as footwear, travel goods, plastic products, gems and jewelry, rubber-based products and ceramics. There are many cases of the local partner talung over the entire production operation and continuingto thrive inanexport business after an initialjoint venture state. 3.52 The BO1 policy regime (see Box 3.4) has sometimes been criticized as an "enclave arrangement", breeding dualism inthe Sri Lankaneconomy. This characterization may be somewhat harsh. Unlike many other countries which have implemented FTZ schemes as an appendage to a highly restrictive trade regime, in Sri Lanka, BO1privilegesare available to both foreign and local investors who meet the approval criteria (the prime one being export orientation). Furthermore, an increasing number of BO1 approved f m s now operate outside the FTZs. Their geographic spread would have been greater if infrastructure and connectivity had been better outside the Western Province. 3.53 A major policy issue relates to the provision of generous fiscal incentives to BO1 firms. Many studies support the view that tax incentives have little impact on foreign investors' location decision. However, when a clear distinction is drawn between "market-seeking" (tariffjumping) and "efficiency-seeking'' (export oriented) investments, a number o f studies find that fiscal incentives do matter, provided other determinants such as political stability, favorable geographical location and infrastructure are favorable (Wells 1985, Morisset and Prinia 2001, Guisinger and Associates 1985, Weigand 1983). Inlight o f the latter, it might be risky to suddenly terminate direct tax incentives for all future investment. However, it should be noted that in addition to adversely impacting tax revenue, fiscal incentives offered to BO1 firms are lllghly variable (across products and sectors and scale of investment) and subject to substantial discretion and against the original intention, many firms have become perpetual beneficiaries o f tax holidays. Moreover, the experience in Sri Lanka shows that despite generous fiscal incentives, the overall level o f FDI (including flows linked to privatization) has been modest (representing around one percent o f GDP annually) by regional and international standards. There is therefore a case for rationalizing incentives presently offered to BO1 firms. Gwen the fiscally-driven urgency for raising tax revenues, some form o f phasing in o f a moratorium on direct tax incentives needs to be seriously considered. In the meantime, recent legislationto bringsuch incentives into line with InlandRevenuelaws is to be welcomed. 31It is speculated that 10percent ofjobs (or about 28,000) mightbe lost. 46 Sri Lunku Development Policy Review Box 3.4: The Baits&of Invmtmerrt@Sl) The-B w dof`fnvestmmt(`I311t)basits origins111the CrearerColoraibo Economic~ ~ ~ n ~~ ~$ s~s a~m~~ ~) n~ s h ~ 1998 to bylrdss the various laws (tat, customs, &fr&n exchange) atrd ~ d ~ n ~ n ~ s ~ r a t i ~ (tax and customs) Systems e h ~ n d ~foreign~~ r i i ~ n ~ p ~~~ ~~c u~ne~ ~ley~ ~n - o~ racwties.~ Whiled thme laws and systems rethiam ~l ~~~ r t ~ probleihia~~~~ regmehasgrow inscope to cover hth foreigii and ~ ~ ~ ~ l l yRnns)kjthnin~andout of the BO1 - ~ w Free Tide Zones (I"I"Zs) WQI approvedprqwtsaccount for MYi tlfexpwts, 75% of industrialc n ~ and~ ~ o ~ ~ a h " 1XfO% c#Sri Lanka's vital garments industry. kn ddiriun to providingdirect tax ~nce~~tives cxempt~ota and froin ttmign ~~c~~~~~cnntmls, the BOE probides a separate lepl and adm ciastants(both ~ ~;andplrpicallogistics). Thoughi Sri 1,ank;t's rgidlab u ~ ~ ~ ~systems ~for taxation arrd are ofPicially ineffect tor BO1firms, ~ ahasbemfa$strrict. ~ f ~ ~ ~ ~ ~ ~ ~ I I ~ i bdepends~ony type of ans~ivitywith ~ ~ I ~ ~ ~~ n~ i vand~export~~ t~ u~ ~ r~ ~While~foreign equity ~ u ~ n~ e - ~ ~ s , was never a r ~ u ~the~c ~~ ~ ~~&att~ f~i,mis have at least foreign IO~TISbeen relaxed. So o i ~ has were restrictiortson Iiw&oa, Morethaa 80% of BO1firm cpmte aiitside FWs. Large non-BOI sectors of the e ~ ~ most not&te SMEsEs,~ ~ ~ ~ u ~ continue ~ face a diFFiculr business ~ y TQ v ~ ~ ~ n ~ ~ t . 3.54 Inaddition to the adverse impact of the conflict, several `behind the border' constraints to investment may account for low FDI in Sri Lanka. Indeed, according to a recently completed investment climate assessment (World Bank, 2004), top constraints to doing business include poor infkastructure (electricity and transport), cost of and access tofinance, and rigid labor regulations (for urban firms). Addressing these constraints will go a long way in encouraging overall private investment andinimproving Sri Lanka's ability to attract FDI. Currency Policy and Convertibility 3.55 Since 1978 Sri Lanka has operated with an unified exchange rate for the rupee withm a system o f "managed float". The degree of Central Bank intervention has varied over time, reflecting the exchange rate goals o f the authorities. FromJanuary 2001, currency policy ostensibly shiftedto a "free float". Since then, there has been substantial nominal depreciation of the rupee, more than compensating for Sri Lanka's higher rate o f inflation relative to trading partners and thus leading to modest real depreciation. By 2003 the real effective exchangerate was back to the levels of the early 1990s. After mid 2002, the Central Bank has intervened actively (despite the terminology o f "free float") to build up reserves and moderate the nominal appreciation that would have otherwise occurred because o f the US dollar's depreciation against major currencies. Given the sizable trade deficits in the balance o f payments and the limited prospect for reducing the fiscal deficit, such policiesto avoid real appreciation ofthe rupee seem quite appropriate. 3.56 Sri Lanka moved to full current account convertibility in 1993. The movement towards convertibility on capital account has been measured, although the abolition o f the foreign exchange surrender requirement for exporters amounts to a defacto relaxation, to some degree, o f controls on capital outflows. Especially since the 1997-98 Asian currency crisis, Sri Lankan authorities have adopted a cautious approach towards further liberalization o f the capital account. Such prudence i s warranted until there i s more success with fiscal consolidation, reforms o f the domestic financial sector and greatersurety o flastingpeace. Labor Migration and Remittances 3.57 As noted earlier, since 1978 Sri Lanka has followed liberal policies towards labor outmigration. As annual outflows reached significant levels, the Department o f Labor, through its newly established Sri Lanka Foreign Employment Bureau (SLFEB), developed an institutional capacity to monitor the process and to protect migrant workers from malpractices o f recruitment agents and also to provide some institutional support for Sri Lankans in foreign employment. More 47 Sri Lanka Development Policy Review recently, SLFEB has launched a number o f schemes to promote foreign employment, including an universal insurance scheme for migrant workers and a bank credit scheme for returning migrants. 3.58 With inward remittances accounting for 18 percent o f total current receipts and 9 percent of GDP in 2002, this liberal policy appears to have reaped rich dividends. The outcomes have been greatly helped by easy access to banking facilities, the pursuit o f market-responsive exchange rate policies and the virtual absence o f a black market for a foreign exchange. 4. Reviving Growth of Non-plantation Agriculture 3.59 More rapid agricultural productivity growth is hdamentalfor reducing poverty inSri Lanka, since nearly 90 percent o f the poor live in the rural sector. This would require strong commitment to removing existing policy and regulatory constraints which have stifled growth in the agricultural sector. Short to medium term priorities include adopting policies to facilitate farmer access to improved technologies, creating a more transparent and stable trade policy regime, allowing full and transferable ownership rights to land, and ensuring the sustainable use o f water. Adopting policies to speed up currently lagging private sector participation and investments would also be critical to promoting growth inboth the agricultural and non-farm sectors. These include rationalizingcurrently restrictive labor regulations and promoting a regionally equitable development strategy for rural infrastructure and services development, with increased emphasis on operation and maintenance o f physical assets to ensure their longer term performance. Although several o f these reforms were part o f the Agricultural Policy Recommendations formulated by the National Development Council in 1996, implementation has lagged behind (see Box 3.5). Consequently, the anticipated supply response and much desired diversification o f agriculture has been slow to materialize. ~ ~ societiesa ~ ~ v ~ ~ffarwardcoratrwfs Source: NationalDevelopmentCouncil, 1996, "Agricultural Policy Reconxnendations," Report ofthe National Development CouncilWorlung Group, Volume 1, Colombo, Sn Lanka.; Weliwita and Epaarachchi2002, CentralBank Annual Report, various issues. 48 Si+ Lanka Development Policy Review Agricultural Technology Policy 3.60 Access to productivity-enhancing technologies by farmers has been constrainedby restrictive seed and phyto-sanitary policies. Fortunately, there is increasing consensus that excessive regulation serves more as abarrier to entry than as an environmental filter. Many requirements are outdated and inadequateto meet the rapid advances inresearchand technology occurring worldwide, and subject to costly permit and inspection procedures. New seed and phyto-sanitary regulations have been developed, inconsultations with stakeholders. Their approval will remove uncertainty and encourage greater participation o f the private sector, producer organizations and other agencies in distributing improved agricultural technologies to farmers. Modernizing and streamlining the national seed certificationand plant quarantineservicewill alsobe important for the effective implementation o fthe new seedandplant quarantineregulations. 3.61 Agricultural research is performed by a large number o f government institutes in Sri La~&a.~'The government's researcheffort has beenrelatively successfulinraisingthe productivity o f rice, but less successll for other crops. There is very limited private sector involvement in agricultural researchowing inpart to the absence o f intellectualproperty rights protection, restrictive seed and phyto-sanitary regulations and procedures, and subsidized sale o f planting materials by government agencies. Improving the effectiveness o f the agricultural research system, in large part elaborated in the National Agricultural Research Policy (2003), will require fostering a pluralistic national agricultural research system, including the government, private sector, NGOs and other agencies; and institutional reforms o f existing public agricultural research system to strengthen its demand orientation and improve the quality o f its research activities, including through greater participationo ffarmers andother stakeholders inprogramgovernance,priority setting and evaluation. 3.62 Agricultural extension services have been severely weakened since it was devolved to Provincial Councils inthe early 1990sunder the 13" Amendment to the Constitution. Only about 13 percent o f agricultural households reported receiving technical assistance from a government extension agent (15 percent fkom all sources)33. Improving the effectiveness o f the agricultural extension system in Sri Lanka would necessitate increased focus on: (i) expanding the supply o f extension services by fostering greater participation o f private firms, NGOs, and producer organizations, including through government sub-contracting o f services; (ii)improving the effectiveness o f the public extension service - strengthening client orientation through adoption o f participatory approaches inplanningand implementation; (iii) linkingproviders to multiple sources o f innovation (research and others); and (iv) expandmg use o f new information and communications technologies to deliver a wider array o f information o f value to farmers through new, innovative channels. Agricultural LandPolicy 3.63 A critical feature of Sri Lanka's land ownership pattemis Government's ownershp of a large share o f land in the counlq-1.72 million o f the Sri Lanka total 2.79 million ha is state-owned. (World Bank 1996). These lands were transferred to farmers through various land settlement programs beginninginthe 1930sunder a number of legislations, one o f the most important o f which i s the Land Development Ordinance (LDO).34 %le these legislations succeeded in promoting greater equity in land distribution, their highly restrictive nature hurt farmers in several ways. For LDO land, which totals about 1.4 million ha, restrictions on mortgaging preclude its use as collateral to access credit. The small size o fthe holdings (as over 40 percent o fholdings have less than one acre ''These include the Departmento f Agriculture, plantation (tea, coconut, rubber)researchinstitutes, several national institutes reportingto the Council for Agricultural ResearchPolicy anduniversities. 33 SLIS data (1999/2000). 34 Other key legislations included the Sale o f State Lands (Special Provision) Law 1973, Land Development (Amendment Act) 1981, Agrarian Services Act 1979, Land Reform Law 1972 and 1975, Land Reform(Special Provisions) Act 1981, and the Agrarian Services Development Act No. 46 2000 49 Sri Lanka Development Policy Review and 60 percent less than two acres), the lack o f secure property rights, and legal restrictions on acquiring or leasing land, constrain efficiency in land-use and reduce incentives for productivity- enhancing investments. Those interested in shifting out o f agriculture into non-farm activities, or merely moving to another location have to leave without compensation. In addition to fostering a large cadre o f part-time farmers, these legal provisions limit the efficient allocation o f land resources (World Bank, 2003). 3.64 Following a decision in2002 to grant 111ownership rights to farmers cultivating LDO land, a draft Land Ownership Bill (providing the legal framework to implement the policy) was sent to Parliament in November 2003, but as noted elsewhere, was withdrawn due to court challenges. To move the process forward, the LDO is being amended instead. As part o f this work, due care should be taken o f procedures inefficiencies embedded inthe Bill, including the absence o f clear criteria for approving or disapproving application^.^^ The prompt processing of the amendments to the LDO, including addressing these procedural inefficiencies, would be essential to enhance the ability of farmers to make the best use o f agricultural land. LandAdministration 3.65 For private landowners, the poorly h c t i o n i n g land administration system hinders the operation o f land markets. Most private land records take the form o f deed registrations that record transactions and serve as evidence in support or proving title to land. The insufficiency o f deed registration in guaranteeing ownership o f land, combined with the pervasive co-ownership o f however, increases the transaction cost in the land and credit markets. Furthermore, the deeds may not agree with the actual land specifications, hence disputes are common. In case o f a dispute, settlement times for land cases in courts usually take more than ten years, and lack o f affordability limits the access o f poor landholders to dispute resolution. Recently, the government initiated a shift to a title registration system. A registration of title system i s one where the parcel has been registered, with transactions thereafter registered against the parcel. A registration of title system facilitates boundary dispute resolution, transparency o f records and efficient land administration. 3.66 The Registration o f Title Act (RTA) 1998,provides for unencumbered and clear title to every parcel o f land inthe country, establishes a system o fregistration o f landparcels and ownership rights, and creates a new system o f land-transaction regi~tration.3~Amendments to the Act currently being drafted aim to: (i) eliminate restrictions on the registration o f co-owned land; (ii) the issuance o f allow a title to people without ownership documents on the basis o f occupancy; and (iii) protect interests that are difficult to register formally. AgriculturalTrade Policy 3.67 Agricultural tariffs in Sri Lanka are subject to unpredictable and frequent change. The government intermittently lowers the tariffs for major agricultural imports through duty waivers and controls import volumes through licensing during months when domestic prices rise. These frequent unpredictable changes create considerable uncertainty, heightening price risks for farmers, consumers and local entrepreneurs, and greatly dampen the incentive for private sector investments instorage.38 35 For example, people obtaining new land allocations under the LDO will still be required to go through the process o f receiving a pennit first. 36 Every time a parcel o f land is sold, transferred or used as collateral, an extensive search-- asfar back as 30 years -- o f all past deeds associated with the landparcel has to be conducted to confirm ownership rights. The problem o f co-ownership arisesbecauseo ftradition involving dividing land among all children. 37 This includes records containing information onparcelboundaries, name o f owners and rights that thirdparties have over the land such as mortgages, leases and easements. 38 For instance, inOctober-December 1999, the import duty on rice was cut from 35% to 10%to dampen the seasonal price rise. Many millers who bought and stored large quantities o f paddy duringthe previous harvest found it difficult to offload their stocks, because o fcompetition from cheaper imports. Having suffered large storage losses, this reduced their incentive to purchase paddy during the 2000 Maha harvest, which ledto a sharper fall inproducerprices, promptingthe govemment to ban rice imports from mid- July and to undertake procurement operations to prop up prices. However, with the drop in 50 Sri Lanka Development Policy Review 3.68 To induce private investments in agriculture, ago-industry and storage, reduce within-year price fluctuations, and improve food security, the government will need to resist the pressure to seasonal altering o f tariffs. Over the medium term, tariff protection for various agricultural commodities would need to be gradually phased out to reduce the bias in favor o f partmdar crops (e.g. rice, potatoes, chillies, onions) and thus allow improved domestic resource allocation; and to reduce the taxation o f consumers who are forced to pay above-world-market prices. With the removal o f price distortions, cropping patterns could adjust to changing economic incentives, including shifhng from low-value and low-productivity activities (such as rice production) toward commercial production o f altemative higher-value crops. The phased reduction intariffprotection will need to be accompanied by parallel measures to lift the constraints on domestic, commodity and factor (land, seeds, technology and water) markets and to improve rural infrastructure. These complementary actions will help ensure that farmers have the freedom and the capacity to alter their resource-use decisions to meet the changing needs o fthe market. Rice is the hasic staple fir the large nnajorrty of Ore ~pn~ationPaddy p ~ u c t ~ o(un-nxlled nce) is the miit i wcupataonandsoui~eofmcoineande n ~ l o ~ nforna~largeuuntbmofpeople inpupalareas, especiallythe poor s Policy approach: The ~ h ~ e vofe~ ~ ~ i ~ ~~ tn nce prcdluction has underpinned Sri~Lanka's a ~~ ~ ~ ~ ~ l ~ u ~ ) ~ # ~ ~ ~ policy tn the ~ s ~ ~ i n d e p e ~perm3c ePolicies to increase pmduccionhave comprised public a"&ural ~ e n research and extension, government p r ~ u ~ t I oand i n subsidieed sale o f paddy seeds, large inw.tlllenb in zmgatmn, costly Fertilizer aMi cmdsF substdtes, a land law pmvi p i ~ c u r e m ~ofn ~paddy output, and II@I Pice Ciowmnsntsoldsulsidizedncethrough Caap Chtcomes. Srt Lanka bas achieved nearty s e l f ~ s u ~ c I eI ~Z nce ~ ~ d ~ ~fwlrh ~lagher proiiii~ti~i~y orher ~ y c t o n than countpies in $lierqgoii), but ab B 11tgh cost. Paddy pmdu~~~ongwwe hg.only around 2 % atmually over the past has three decades and growth in wlue-added has been at best negligble. Moreover, these policies are havingadverse conseqnmces rin farmers' welfare Forcing farmem to continue gPnwing paddy through the paddy land's provision unthwts their ticomeIwtentialandpments file use ofhighly productive[in tmstcasesinigated) lands for other more lucrative imn-paddy p ~ m w ~ Pertiliter subsidies for urea, which p~nlrtnlybrnefit rice farmers (which account for ahout '75% of urea ~ n s u i ~ ~emourage}their ovefzise, raistng co11cem about inq~mpersod nlaiiageinent. ~ i o ~ ~ Ironrcatly, while fertilizer applicaeton rates amng rice pr.ciut;em niom than douhld 111the 199i1s, average yields rncreltsed only by 8% Fertilizer subsidies itre pnmarityberafitingncher rice Famm about 51% of total rice area is cultivated by houselioldsinthe top 40% ofthe mralexpenditurequinttles, comparedto only about 25% in the case of housshoids inthe two porest quinPiles. Wigh imp%tariffs*to the extent that they keep rice piices high fiz producers increaseheir cost for consumers, espeerall>hurtingthe poor, Future Prospects: As FnLanka enten the 21" century, several developmentspintto the needco reassessthe strategy for the nce sector. h t , nsing incomes an9 changnig consunier preferencesare reductng householddemand fur Betweerr )980/8t anti 2002, aceiage coirsurrqpllondeclinedby 23% ( h m 47 to 36 IrghomehoIrVmonth) andthis trend is expectedto continue(aC3 2002) Second, Sri Lankahas acluewd near rice seif-suficintcy arid the renin)of paddy Ian& U, production111 the NorthEast IS likely result L I ~sniglus pdnction in the neat future ?'%iid, even bitl~slight a supply shortfall, inbenlatmnel tmde offen an irrtptwtant alternatiw safety valve for sourring doniestic rice ne&. Pmally, donmttc arid enprt deniarid fox higher vdue products (e.g hortieuhre and Iivestwk) is rapidly rncreming, offering afternativehighha iticomegeneratingopportunities fiw fmiers CommodityMarketing 3.69 The public sector has gradually moved away from agricultural commodity marketing, including through the recent liquidatiodprivatization o f the Cooperative Wholesale Establishment (CWE). However, in 1980 a 20-year monopoly was granted to PRIMA, a Singapore-based company, for the fmto mill and supply flour for distribution through various government outlets. The contract was extended by another five years in 2000. PRIMA benefits from duty-free imports o f wheat grain and mill-related equipment as well as income-tax exemptions, while private imports o f wheat flour and wheat are made uncompetitive by import tariffs. Since Sri Lanka does not produce wheat, the producer prices, farmers had neither incentive nor resources to increase cultivation for the next harvest. According to official estimates, the area of land cultivated by farmers for the 2001 Maha harvestfell by about 30% from the 2000 harvest. 51 Sri Lanka Development Policy Review high import tariff on wheat flour and wheat does not protect farmers and only raises prices for consumers. Phasing out the wheat monopoly upon expiration o f existing contract in 2006 will contribute considerably to improving the efficiency o fwheat marketing andpricing in Sri Lanka. Water ResourcesManagement and Irrigation 3.70 Duringthe last five decades, irrigation development was as a major pillar o fthe government's rural development strategy. Between 1980 and 1997, about Rs 215 billion (constant 1996 rupees) in public funds were spent on irrigation infrastructure development. Most o f these investments focused on the construction o f new dams for power generation and surface irrigation systems,. The most important program was the Mahaweli Project, initiated in the late 1970s and managed by the Mahaweli Authority o f Sri Lanka (MASL). 3.71 The long term sustainability o f past huge investments in expanding surface irrigation infrastructure is threatened on several fronts. Inadequate priority and funding for operations and maintenance led to the rapid deterioration o f canal systems and to poor quality o f services. This explains the need for repetitive (often every 5 years) and costly rehabilitation. Institutional weaknesses in the water-agencies combined with minimal involvement o f farmers impeded greater improvement in the quality of and "user-orientation" in service delivery. Poor reliability o f water delivery and frequent lack o f access to water by tail-enders, combined with lack o f access to agricultural extension and improved technologies, contributed to low crop yields. Inadequate farmers' involvement in decisions on water delivery (both in terms o f quantity and timing), has constrained their ability to diversify to higher value crops or alternative crops besides paddy.39 Providing water for free also reducedthe incentive for farmers to save anduse water efficiently. 3.72 The Government is now faced with tightening inter-sectoral competition for water among various users (agriculture, drinking water, industry, etc). The fragmentation o f functional responsibilities among about 40 water agencies, makes coordination difficult. Although the National Water Resource Council (NWRC) has been established to manage water resources at national level, not much has been achieved in the absence o f a national water policy. The outgoing Government began drafting a National Water Policy and National Water Bill. Finalizing the Bill, however, has been complicated and delayed by the social and political sensitivity o f water issues inthe country and overlapping jurisdictions between the Central and Provincial governments. Buildingconsensus and approval o f the National Water Bill will be essential to any strategic vision for the sustainable development, management o f water resources. Infinalizing the Bill, priority would need to given to: (i) fromsupply-drivengoalstocomprehensiveplanning,allocationandmanagementwithina shifting river-basin framework; (ii)formulating an appropriate regulatory framework and reprioritizing expenditures (from the creation o f new assets to demand-driven maintenance and rehabilitation investments) to support such a shift; and (iii) reforming institutional structures and procedures, building on increased participatory management o f systems, to improve the management o f water resources in Sri Lanka. Rural Infrastructureand Services 3.73 Improving rural infrastructure beyond the Western Province is essential for rural development. Limited access by rural households to essential basic infrastructure and services (e.g. telecommunications, roads, electricity, banks), contribute to increasing transaction costs in the marketing o f agricultural produce and in sourcing inputs, thus significantly affecting the competitiveness andprofitability o f Sri Lankan agriculture. Regressionanalysis usingthe SLIS 1999- 2000 data also show that access to rural infrastructure in addition to land owned, the size o f the household, and education, are significant determinants o f household participation in rural non-farm activities. It shows that lack o f access to roads (the distance from the nearest mainroad) is negatively correlated with the probability o f participating innon-farm employment. According to the rural part 39Currentwater deliveryschedules are still designed for paddycultivation. 52 Sri Lanka DevelopmentPolicy Review o f the Investment Climate Assessment (ICA) on Sri Lanka (World Bank 2004), transport constitutes the single most important constraint to rural firms, followed by cost and access to credit, demand and electricity. Improved access and quality o f rural infrastructure would contribute not only to raising the quality o f rural life, but also to the successful implementation o f govemment development plans for the modernization o f agriculture and improving the investment climate for rural industries and services. Participatory planning and implementation o f rural infrastructure projects, which involves government and targeted users, would be valuable to ensure the appropriateness o f investments undertaken. Source: SLIS 1999-2000, Gunewardena2000,2002 HIES. 5. InfrastructurePrioritiesfor Growth, ConnectivityandEquity 3.74 For a country like Sri Lanka, approachingrapidly middle income levels, reliable infrastructure i s paramount for further economic development, especially o f export-oriented activities such as manufacturing and tourism. These services require reliablepower supply at international prices, port and air facilities, world-class telecommunications capability, and a transport system that works in a seamless fashion. Although Sri Lanka has made progress in some o f these areas, substantial challenges remain to bringthem up to world standards and to reduce regional inequities inthe access and quality o f these services. Inparticular, as reported inthe I C A (World Bank 2004), shortcomings intheprovisionofadequatepower andelectricity services representmajor impedimentsto investment inSri Lanka. RoadsSector Challenges 3.75 The transport sector needs to correct years o f misallocation o f public resources that have left Sri Lanka's relatively extensive road network seriously under-maintained, and cater for additional demands arising from a growing economy and from the integration o f the northeast to the rest o f the country. Against these challenges, the government needs to (i) balance carefully the urgent needs for repair and maintenance against proposals for new investment; (ii) concentrate scarce public resources exclusively on those activities where the private sector cannot operate andor invest effectively; and (ii) oncontinuingtheeffortsregardingtransportcommercializationandrestructuring. focus 3.76 According to official estimates, about US$870 millionwill be required during 2004-08 for the rehabilitation and maintenance o f Sri Lanka's road network and a further US$655 million for new planned expressways. Against these requirements, RDA spent only US$13 million for rehabilitation o f national roads in 200340and US$ 16 million for periodic and routine maintenance, representing about 10-15 percent o f estimatedneeds. 40Excluding$15 millionfor landacquisition for the proposedColombo-Katunayakehighway. 53 Sri Lanh Development Policy Review 3.77 Efficiency of Road Utilization. Despite Sri Lanka's relatively highroad density, network utilization is inadequate. As Table 3.6 shows, Sri Lanka has less than halfthe number o fvehicles per Kmo froadthanthe average for the SouthAsia Regionandless than2per cent o fthe average inLatin America. Also, although the number o fvehicles per capita is highcompared to the South Asia region, it is low compared to all other regions of the world (except East Asia). Increasing the efficiency o f utilization o f the existing road infrastructure will require substantially improving the condition o f the roads, through regular maintenance and rehabilitation o f deteriorated road networks, as well as buildingexpress corridors connectingmajor cities. Priorities to achieve these goals include: 0 Provision of adequate fundmg for road maintenance and rehabilitation. l b s would require raising funding two-fold (from current budgetary allocations) by increasing the revenues o f the recently established Road Fund4' and mobilizing further funds from other sources including user charges and fees. Inthe medium term, Sri Lanka would need to develop an overall pricing policy for the transport sector.42 0 Improving the balance between recurrent and capital investments, with a focus on addressing the maintenance backlog and improving the condition o f the national highways which cany 70 percent o f roadtraffic. 0 Strengthening the capacity of agencies responsible for formulating and implementingpolicies inthe transport sector. Table 3.6: InternationalComparisonof Sri Lanka's RoadTransport I Source: World Bank estimations usingWorld DevelopmentIndicators, 2002. Most data i s for 2000. I t I I I I 3.78 Economic and Financial Sustainability. The unit cost o f road provision is very highin Sri Lanka as a result o f relatively higher rehabilitation due to lack of periodic maintenance and cost overruns and other inefficiencies inthe provision o f road works. While the cost o f routine and periodic maintenance for national roads is about US$2,000 per km, rehabilitation can amount to as much as US$400-500 thousand per km44. Priorities to ensure the economic and financial sustainability o fthe system, especially to reduce currently highunit costs o froadprovision, include: 0 Increasing private sector participation in transport infrastructure and service provision, at all stages, including policy setting. This is particularly important given the winding up o f the Road Construction Development Corporation (RCDC). The Government should create and adopt appropriate structures to ensure the efficient functioning o fpublic-private partnerships. 0 Raising the efficiency o f public sector agencies by introducing competition, making public sector management more-market sensitive, and restructuring pricing and financing systems. 41 The RoadFund is currently raisingrevenues from a national fuel cess introduced in2003 at IRp/liter for petrol and OSRp/liter for diesel. 42 This would involve a comprehensive assessment o f all sector-related charges, fees, taxes, cess, etc where pricing normally reflects true economic costs. 43 After twelve and eighteen years, rehabilitation costs are 2-3 times the costs to maintain a road ifithas not been periodically maintained. 44 Source: SL, Concept Paper: Road Sector (Expressways, National Highways, Provincial and Local Authority Roads), December 2003. 54 Sri Lanka Development Policy Review 3.79 Equity and Environmental Sustainability.Roads have been maintained relatively better in the Westem Province, creating inter-province inequality. Reversing this bias and improving connectivity in other provinces could substantially ameliorate the growing problem o f regional economic inequality. Mainpriorities include: 0 Integrating the needs o f the poor, (particularly in rural areas) for access and mobility in transport strategy and policy setting. To this end, mechanisms to ensure the participation o f rural communities inthe setting o fpolicy need to be created. So far, it has been assumed that the mobility needs o fthe poor can all be resolvedby improving transport networks andpublic transport services inrural areas. Given that transport services do not penetrate all rural areas, and as a result problems o f mobility and access are acute in the most disadvantaged areas, attention needs to focus on promoting non-motorized transport modes and providing appropriate localrural roads for such transport. 0 Increasing the participation o f stakeholders, including the rural poor, inthe planning and implementation o f investments inthe sector. Where poverty alleviationprograms exist and unemployment levels are high,participation o frural labor to improve roads can provide a least cost option for carrying out such works. Improving the management o f congestion, pollution, andsafety. The degree o f congestion andassociated cost inSri Lanka canbe as highUS$18 As for a 6 percent increase inGDP, the demandfor roadspacehasbeenestimatedto increasesby 8 percent46,itis paramount that the measures outlinedinthe CleanAir 2000 are implemented inorder to control the effects o f congestion. Power Sector Challenges 3.80 Sri Lanka will have to expand power generation capacity to keep up with growing demand, against serious inefficiencies in the sector. Unfortunately, progress with the on-going industry restructuring, which could begin to address the existing inefficiencies, i s slow and the results uncertain. The vesting o f the new companies, initially planned for October 2003, has been delayed pending the conclusion o f negotiations with trade unions on the terms and conditions o f workers' transfers. Inaddition, the independent Public Utilities Commission o f Sri Lanka (PUCSL), created in 2002 to regulatethe sector, is not yet fully operational. 3.81 Adequate Generation Capacity. Just to keep up with the historical growth in electricity demand (of 8 percent annually), Sri Lanka will need to add to the system approximately 200 MW each year. Despite the existence o f a Long Term Generation Expansion Plan (LTGEP), prepared by the CEB, decisions on the procurement o f scheduled plants has been subject to long delays, while plants not included in the plan have been implemented. Efforts need to focus on expanding generation capacity in accordance with an updated least-cost expansion plan that takes into account the full range o f capacity options, including for privateprovision. Mainpriorities include: 0 Accelerate the decision making process for implementing the updated least-cost expansion plan. Inthe event o f delays, early plans should be made to address resulting shortages. 0 In the longer term, allow the single buyer, the Transmission Company, the independent procurement o f plants without political interference while avoiding short term uneconomic plants. 45The cost ofcongestioninthe Western Province (Colombo MetropolitanRegion) was estimatedas aroundRs. 550 million (US$ 9 million) in 1995. Source: World Bank, 1997, Sri L a n k Transport Sector Strategy Study, Vol 1:MainReport, January (Washington DC). 46Source: Ibid. 55 Sri Lanka Development Policy Review 0 Implement the loss reduction program and other energy efficiency and demand side management initiatives. This would contribute to the optimization o f the existing generation capacity. 3.82 Efficiency of Electricity Provision. Technical and economic inefficiencies in the system have resulted in consumers paying higher electricity tariffs than inEast Asian countries. Sri Lankan residential consumers using more than 300 kWWm pay 16 U S cents per kwh compared to about I O US cents per kw charged to Filipino and Malaysianconsumers. Commercialand industrial consumers are equally at a disadvantage. These tariffs reflect, amongst other things: (i) dependency on expensive thermal sources for the generation and purchase o f power47; (ii) operating inefficiencies in the CEB that result in high technical and financial losses; and (iii) lack o f a transparent and competitive environment to minimize the cost o f electricity and to pass efficiency gains to the consumers. 330%3.7: CEB's LongTerm GenersttionExpansion Plm* According Isthe base case of&@L`I"Gli3P 12003to 2017) the fallowingplanrsareplanned: A 200 MW diesel power plant ~ ~ ~ awarded)d tfiybe ~ o J I I ~ ~ s s ~ itr~earlyJ . ~ a o ~ ~ 2005. A 300 MW ~ombmedcycle p w e r plant ta be c a ~ ~ ~ s $m20%.e Studiesby ~ o n ~ the Asian ~ ~ ~ l Bank (pW w~2003) indicate that p w a supply shorrages o e ~ ~ wouldoccw ifthisplantisdelayedanty beyonddanitary 2007, A 300 MW Coal powerPlant The planehasbeenon the cards for L5 years. Xfk ~ [ ~ ~ ~of ~this~pfant(or alfematave ieast-c-.ostoption) is delayedbeyond s ~ o n i ~ ~ * early 2005, it wauld result.in a deAcit of about 640 GWhby 2OW, A 150 MW hydropower plant at Upper Ko~hmJe,beingdeveloped by the TXB, * to bec~mmissior~~ in 2009 A second Coal planr for which ~I~~ITIIJI~ commenceby 2011 needsto I XuurGe. CEB 3.83 Despite the relatively hightariffs by regional standards, the CEB is still unable to pass on to the consumers the 111cost of electricity due to political interference. A moderate tariff increase in June 2000 followed by a 61 percent increase April 2002 raised CEB's cost recovery to approximately 90%. CEB's inefficiencies coupled with high losses have contributed to the high electricity costs. CEB's technical and non technical losses in dstribution add to around 17.2%, with non-technical losses are estimated at 8% reflecting tampering o f meters, illicit connections and defective (lack of) metering.48Inthis context, mainpriorities include: 0 Accelerating the restructuring o f the sector, including the effective operation o f the PUCSL as an independent regulatory body andensuring that successor companies operate based on commercial principles. 0 Enable the PUCSL to: (i) the successor companies to promote competition and regulate efficiency improvements; (ii) phase out cross subsidies from General purpose and Industrial consumers to residential consumers over a reasonable period o f time and implementation o f tariff regulations that enable full recovery o f costs49; and (iii) overseeingthe implementation o f generating capacity. 3.84 Access and Regional Equity. Access to electricity varies greatly from 95 percent in the Western Province to 9 percent inthe conflict-affected are inthe North Province. This access should 47 Presentlythe average cost of thermal generation is Rs. 6.79 KWhcompared to Hydro of Rs. 1.81kWh. The average cost ofpurchase from IPPs is Rs. 6.74 kWh, while the cost of emergency power ranges Rs 9.79.kWh to Rs 10.6l.kWh - calculated at US$28/bbl.. 48 Incontrast, LECO has adistribution loss of around 6% today, reduced from 28% since 1994. 49 Dueto the progressive rate structure o fthe block type residential tariff, residential consumerSover 300 kWWmpay double the rate ofthe small andmediumsized residentialconsumers. 56 Sri Lanka Development Policy Review be expanded in the most economically efficient manner, includmg connection to the main grid and, where this i s not feasible, off-grid services at the village or household level5'. Mainpriorities include: 0 Utilizing more efficiently existing programs to provide off-grid connections, eg the DA- fimded Renewable Energy for Rural Economic Development Project (RERED). This facility has been underutilized because o f governments' politically-motivatedtendency to offer connections to the CEB grid, for example under the Deepaloka program, which in the enddo not materiali~e.~'Moreover, REREDcould facilitate the coordmation between the CEB and the Solar Industry Association for the identification o f areas for rural electrification. 0 To the extent that rural electrification cannot be undertaken on a commercial basis, Government assistancethrough the provision o f dlrect subsidies may be considered. 6. EducationReformsfor GrowthandEquity 3.85 The government faces several key challenges to substantially increase the quality o f education, enhance equity o f public spending, strengthen service delivery within the system, and improve the economic and social relevance o f education institutions at all levels. These challenges exist at a time when public education spending has been declining in real terms and the state experiences a large fiscal deficit which compels it to adopt a conservative fiscal policy. Table 3.7: Education Expenditure in Sri Lanka and Other Countries C'ountry As P "/eof As a YOof Hec. Expenditure AverageTeacher National Gorernnient per Student as a Warier as a Income Expenditure of Nationallneomo 'Ye of National per capita larome per capita Sources: Sri Lanka, calculations from CBSL Annual Reports, various issues; Other Countries, World Development Indicaton (World Bank) and UNESCO Statistics, vanous issues. The information above for Sri Lanka is computed from 2002data. Other countries andregions are fromthe closest availableyear inthe late 1990's and early 2000's. 3.86 Sri Lanka's Public spending in education i s relatively low by international standards, particularly interms o f recurrent expenditures per student and teachers' salaries, even incomparison with other South Asian countries (Table 3.7). Given the tight budget situation, efforts need to focus on improving the efficiency andequity o f public spending by correcting current misallocations within the sector and increasing cost-effectiveness. In this context, the government can consider the following set o fmeasures. Consistent with the current National Rural Electrification Strategy. An example, is the unrealisticprogramto connect 288,000 in290 days under the Deepalokaprogram 57 Sri Lanka Development Policy Review Increasing private sector investment and participation 3.87 Increasing private sector participation in education, particularly in tertiary, could produce important benefits. First, it would increasethe overall volume o fresources invested inthe education sector. Second, since the students attending private schools andtertiary education institutions tend to be drawn from upper income families, it would releasemore public resources, on a per student basis, for students from poorer families. Third, it would stimulate economic activity in a sector where investment has been artificially restricted and contribute to higher growth. Fourth, it would provide an alternative mode o f service delivery, which would be compelled to offer high quality services to competewith free public education institutions. 3.88 Creating a favorable environment for private investment in school education would require developing a new education act andrepealing the legislation, passedinthe early 1960's, which act as a barrier to the establishment o f private schools. The National Education Commission (NEC 2003 ), recognizing the importance o f private sector participation in education recommends three types of private schools: (a) fee-levying private schools, which finance their entire expenses, teachthe national curriculum and prepare students for national examinations; (b) fee-levying international schools, which finance all their expenses, teach foreign curricula and prepare students for overseas examinations, within an accreditation framework set by the education authorities; and (c) non-fee levying assisted schools, which teach the national curriculum and prepare students for national examinations, but where there is cost-sharing between the state andthe schools, with the government typically paying teacher salaries andthe schoolsbearing capital expenditures. 3.89 While past attempts at establishing private universities have met with strong resistance, including student violence, it would be important to open the debate with the public at large and ensure sufficient understanding that public funding o f tertiary education i s not only very costly but also regressive(see below). Inthe meantime, options to expandprivate participation intertiary level non-university education need to be further explored, especially in professional and technical fields where the demandfor labor, bothwithinand outside the country, is strong. Enhancing the equity ofpublic education spending 3.90 The pattern o f student enrolment over major grade cycles, and the unit costs o f education by grade level, show that investment in primary education and secondary education are relatively progressiveandbenefit students from low andmiddle income households, while investment intertiary education tends to benefit students from upper-income households. Inconsequence, there is a strong case for allocating the major share of any increase in public resources for the education sector to primary and secondary schooling, while carefully controlling the expansion of the public university systemand allowing enrolment expansion intertiary education take place mainly inthe private sector. Opportunities for poorer students to access tertiary education could be expanded through schemes such as studentvouchers andstudentloans, rather thanthrough direct provision. Increasing cost-effectiveness of the education system 3.91 The tight resource constraint faced by the government makes it extremely important that limitedpublic resources, includmg those for education, be used ina cost-effective manner that may generate internal savings. Inthis context, three important policy measures are available. i.Schoolrationalization. Thecost-effectivenessofoperatingtheschoolsystemcanbe enhanced by consolidating and amalgamating the large number o f small, uneconomical schools, where this can be accomplished without reducing access to schooling and affecting enrolment and attendance. This i s feasible given that a sizable proportion of small schools are located close to other, larger government schools. MHRECA and the provincial councils designed and implemented a successful school rationalization program, during 1998-2002, but the program was temporarily suspended due to intense 58 Sri Lanka Development Policy Review political pressure. The government could consider re-opening this school rationalization program, but with careful safeguards to protect access for poor and vulnerable groups. ii.Increasingthestudent-teacherratio. Thestudentteacherratioinschoolsis21:1,which is low compared with countries with outstanding education systems andfar higher income levels, such as South Korea, Singapore and Hong Kong. The current teacher employment and deployment policy is based on a target student-teacher ratio o f 22:l for secondary grades and 26:1 for primary grade. The government could consider increasing the target student-teacher ratio to at least 23:1 for secondary grades and 27: 1 for primary grades in the fist stage o f reform. In addition, the student-teacher ratio in the public universities which tends to be very low, at 14:1, could be raised to about 18:1, so that cost savings can be generated within the system. iii.Rationalizationof administrative staffing. Theroles, functions andresponsibilities of staff within the multi-tiered education administrative system tends to be rather unclear. A careful analysis o f work loads, and de jure and de facto roles, functions and responsibilities o f staff in the various education ministries, provincial councils, zonal education offices, division education offices and tertiary education institutions i s likely to identifybothduplicationofroles andfunctions, and gaps and shortages o f staff. Overall, there may be an excess o f staff, which a carellly designed voluntary retirement scheme could decrease, reducing pressure on the budget and releasing more resources for investment. 3.92 In order to provide incentives for the implementation o f the cost saving measures proposed above, it is vitally important that the Ministry o f Finance permits funds saved to be channeled back into the education system andbe re-oriented, for instance, to meet quality-enhancing objectives. Intra-budget shifts inresource allocation 3.93 Over time, the balance o f capital andrecurrent spendmg within the education budget needs to shift to allocate a greater share o f resources to capital expenditure. Within the capital budget, a higher proportion o f resources needs to be invested in qualityinputs such as IT centers, science laboratories, libraries, activity rooms, multi-purpose rooms, equipment, technology and tools. The construction and expansion o f buildings, which absorbs the highest share o f the capital budget, needs to be rationalized, prioritizing the urban school system which i s facing rising demand. Withm the recurrent budget, too, resources need to be reallocated from salaries and administrative expenses to quality processes, such as teacher education and training, management and academic training o f school principals, andthe purchaseo f teaching material andleaming resources. Protectingand graduallyincreasingpublic investmentineducation 3.94 Improving the quality o f education would necessitate measures and resources to modernize the curricula, enhance teacher training and salaries, strengthen school leadership by principals, increase the use o f equipment andtechnology at schools and other institutions. Higher public funding levels will not necessarily guarantee improved quality, unless current misallocations and inefficiencies in the system are addressed. However, given the relatively low level of public education spending, and the major challenges Sri Lanka faces inimproving the quality and economic and social relevance o f the education system, it is important that the government preserve the current level o f real expenditure inthe short-term and increase public investment in education gradually over the medium-term, provided that progress is also made in improving the allocation o f existing resources.. The possibility o f at least maintaining the real value o f the education budget in2005, and gradually raising expenditures from 2006 onwards, needs to be considered inthe context o f competing requirementsby other sectors and the target budget deficits. 59 Sn' Lanka Development Policy Review Increasing the quality of servicedelivery: decentralizing of education management 3.95 Key policy proposals to improve service delivery within the school system include the establishment o f school boards and decentralization o f education management to schools. The objective o f this policy initiative i s to empower principals, section heads andteachers, enable schools to forge links with local communities to improve resource mobilization and public accountability, increase the efficiency o f decision making by decreasing administrative layers, and improve service delivery. The school board model is to be pilot tested in about 80 schools around the country. A proper evaluation of this pilot program would be extremely helpful for future policy making on the structure o f school govemance. The government could also consider applying the principle of decentralization to other education institutions, such as National Colleges of Education (NCOEs) and Teacher Centers (TCs). Increasing the managerial autonomy o f these institutions (on curriculum design, recruitment of academic staff, administration etc.) would greatly enhance their operational efficiency and quality o f service delivery. Increasing the EconomicRelevanceofthe Education System 3.96 Several measures have been adopted to improve the linkages o f the education system to the world of work. These include: (i) introducing student centeredand activity basedlearningapproaches that promote work ethics and team building; (ii) increasing the use o f technology, especially IT, in education; (iii) expanding and strengthening the teachmg and leaming o f English, including allowing Englishto be used as a medium of instruction; and (iv) making provision for career guidance and counseling in schools and universities. In addition, efforts have been made to strengthen the labor market relevance o f vocational training and technical education by increasing reliance on the private sector for skills development activities. These are positive initiatives that need to be further strengthened and developed. Key initiatives to strengthen these policy measures would be: (a) reallocating resources through cost savings in the education sector to investment in equipment and technology; and (b) permitting schools to offer English medium education from grade 1 onwards, insteado fjust the GCE A/L cycle, as at present. 7. Peace, Conflict and Development 3.97 The cease fire since February 2002 has already brought substantial gains to the Sri Lankan economy. Growth has rebounded. Inflation and unemployment have declined. Foreign investment and tourism have recovered quickly. Defense expenditure has fallen, reducing the pressure on public finances. There has been significant recovery inthe conflict-depressed economy o fthe North andEast and some reintegration with the rest o f the Sri Lankan economy. Thousands o f war displaced refugees have returned to their old homes. Much demining o f land has occurred. Above all, the absence o f a shooting war has become the accepted norm. Put another way, the interests vested in peace have grown. 3.98 Although these gains are real, they have been constrainedby the absence o f durable peace, o f the kind entailed in a formal peace accord. Uncertainty still weighs heavy on both domestic and foreign investors, accentuatedby the politicaltwists and turns since November 2003. More generally, international experience suggests that ending a long-drawn civil conflict is a necessary but not sufficient condition for realizing the fullpeace dividend (Box 3.8). Typically, private investment and government revenue are slow to recover and security expenditures cannot be reduced quickly for various reasons. It takes time and resourcesto restore damaged or neglected economic infrastructure. The damage to institutional infiastructure is even harder to overcome. The transition to peace poses fresh problems and challenges that have to bemanaged. 3.99 What if the current quasi-peace continues for several years ? Can the economy recover close to its full potential? The answer has to be: very unlikely. The single greatest obstacle will be the uncertainty inherent in such a situation. High levels o f uncertainty are antithetical to the high and sustained investment that the country needs to accelerate growth, expand employment and reduce 60 Sri Larika Development Policy Review poverty. Furthermore, an uncertain peace will entail high levels o f security expenditures, continued large fiscal deficits and their associated adverse consequences. Perhaps as important, the political management o f a quasi-peace would continue to distract government andpolicy makers from focusing on the agenda for economic reforms necessary to realize the progressivevision inSri Lanka's poverty reduction strategy. BQX3-8: RealizingPbe PeaceDividend I9ie "pace dividenJ" refersto the tnrproved&~onomcand social outccsmmthat are expectedto materialize i nrhc transition to peace In particular, a ~ r a n ~ ~ a m toaa~peace-time monomy would he expemd to ~ ~ o n rmuh kn reduced militasy spendingand reCtuctd bud%&deficits wtth a cc~rrespondiiigreduction 111interest ratesand inilation ~ o w ~ m can ? t ~ r once again fozus on its develop~entagenda and welfare s ~ ~ e nas~well~asp r o ~ oe~onorntc: d ~ g ~ ~ n ~ growth The cessationofha 11sherin an era of growth resultingfrom increasedtourism, PD1, trade and exports I% h likely to resuit is a i nincreaseddonor fun& for ~ e c ~ n s ~ ~inetheiNorthEasr and paberryaflevration~ h r o u ~theucountry ~ o n o ~ 1heendofthewar shnutdalsomuttinthere-integrationoftheecommyoftheNorthandEastwiththerest ofthe country. Th~swill resulttncountc~cyclical agriculturalproduction(gwen the c o n ~ p ~ e n s ofthe ~ r i ~ ~ ~ ? ~ agricultural seasons m the North and South), and recovery of the fishing industry. Mom of the internally displacedandsome refugeeswill beabte to retaimto their hc~niesaridtake uptxon~m~c activities Children will be able to raumto their schwls and doctors wilf no ionger fear working in former conflict zonests, thus social indicatorswill beginto rtxower. Efoweber, endingthe war is a ntxmsary but not sufficient condirion for realiring the peace dividend. The benefitsofthe peacedividend will not materialize autn-natrcallyandthey are iudo"sly diRcult to achreve inthe short term. Pnvate i n v e ~ ~ andgove~n~en~ e n ~ revenue are slow to recoverand military e ~ ~ ~ d r ~ ~ i r e IS not easily reduced as soldiers cannot be d ~ i n o b i leasily and govemment will haveea continueto make i ~ debt payments for military hardware purehamiduring the conflict. kn this context, it wotdd be useful to exaniine alternative uses fap Fhe military that mild contribute to eeonantic revival and improved munty Some options worth e~)ns~de~ng include allowing the niilitary to take part m peace-keeping missions itverseas that would e m valuable fnreign currency and playing a role in domestic: de\elvpment hirgades that couldunde~~r~k~e o I ~projectssuch~as de-mining, rebuildinghousrngand wfmstrueture. ~ ~ ~ ~ a n Realtsingthe peacedividend will also require a conscious efYort to identify and address the vested interests in pralongingthe war. Several grouys wialiin Sri Lanhn wcicty are currently "benefiting" from the *ar arid may feet threatened, were the peace ptaeessto progress heyonda certain point. 'These groups include: soldiers 111the Sri tankan m y receiving a wage preniiu~n,1,TTE cadres receiving substantial funds from overseas remittances; and d hoc tax revenues, and profiteers d~strib~t~ng gocxis and services during the economx embargo onthe North, 10 end a violent conf9i& it is vital to ~~~ders~atid"and change, the incentives that have made violence a \iahle stmtegy dtiiirtgthe conflid, ?"hetransition to peacein Sri Imka shouldbe seen as an op~rt~niry fo dtscussand resolve~ong-s~and~ng rha haw ledto conflict and have the parentla1eo leadto conflict in issues the fiiture. For ~ n sthis may n a~good time to consider a new devolution ~ ~ ~ he ~ ~ ~ that might be ~ ~ ~ g reyurred to restoreand secure pace It may atw be apprtspnate to rmr?nsidera larger role fir the use of English as a rnediuin of ~ ~ s (dropped in ~the late 195Us),n given its rmpai~anceas a link language ~ c ~ ~ between ~ t h groups as weft as its importance for 11' related develo~m~nt n ~ and ~ n t e ~ ~ a tbaisiraess ~ j ~ ~ n a hnkagcs, Throughout this process, it is tmpowtant that goteminent c ~ ) n i n ~ uthese ~ ~n i ~messages cleiarly to the relevantpopulationsaidmanagesthe e ~ ~ ~ t aoftall stakeholders ~ o ~ ~ s Sources: 4zam et, al , Breakingifie Conflict rmp :Civil War and Development Policy, Wwld Bank Policy Research Paner7007 3.100 However significant are these challenges and costs for managing the transition to peace, they are obviously to be preferred over any resumption o f conflict. Sri Lanka's twenty year history o f civil war bears eloquent testimony to the direct social and economic costs o f conflict and the possibly even more substantial (though less tangible) indirect costs o f missed opportunities and distracted policies. 61 Sri Lanka Development Poli91Review Furthermore, there is little chance that Sri Lanka could sustain the 5 percent growth o f the 1990s if there is a resumptiono f conflict. Incontrast to the early 1990s, the fiscal situation today is much more fragile, infrastructure is suffering from decades o f neglect, agriculture is in doldrums, trade and industry face novel external challenges and external assistance has dwindled (andwill not increase if conflict flares up). 3.101 Inbrief, for Sri Lanka's economy to achieve sustained growth at 6 percentplus (let alone the 8 to 10 percent aspirations sometimes voiced), durable peace is a prerequisite. The country's economic history supports this assessment: the only five year period when economic growth averaged above 6 percent was in 1977-82, when economic reforms held center stage inpolicy-making and the civil war had not begun. 8. Development Risks Resurgence of Conflict 3.102 Today, more than ever before, Sri Lanka is at a crossroads on its development path. As discussed in the earlier section, any resumption o f conflict poses a major risk to Sri Lanka's development prospects. Indeed, since the events o f September 11, 2001, the world is much more sensitive to conflict and terrorism. Even ifthe cease fire holds, without a durable peace settlement the threat o f a resurgence o f conflict will continue to linger, weakening the investment climate and diverting limitedpublic resources away from development needs. External Vulnerabilities 3.103 Being a small economy, Sri Lanka's growth prospects hinge critically on its ability to remain competitive in a rapidly changing global economy, especially in the post-MFA era. The external environment offers opportunities but also poses risks. Proximate examples o f the latter are the economic downturn after 9/11, the recent surge in oil prices and the challenge o f the post-MFA trading environment. No country enjoys immunity from such adverse developments in the international economy. .History has shown, decisively, that closing off from the world economy is not a viable alternative for Sri Lanka, or indeed, for any country. As a small open economy, Sri Lanka's best insurance against unforeseen extemal events and trends lies in building a more diversified export base and a flexible economic structure (with easy conditions for entry and exit), as well as sound investment ineducation and skills. The policy reforms suggested inthis report should help Sri Lanka strengthen its economic flexibility. Political Economy of Reforms 3.104 Perhaps the greatest risk to realizing Sri Lanka's development aspirations is the country's legacy o f civil strife and complex politics. It is very clear that economic reforms in Sri Lanka (like everywhere else) are deeply political. The challenge lies inbuildingthe requisite support in favor o f reforms among the political leadership and the public in general. This needs to be accompanied by a genuine effort to implementing poverty-reducing strategies, i.e., revamping the rural economy and improving the effectiveness o f safety nets to protect the poor. 3.105 Short Election Cycles and Perspectives. Unfortunately, the current volatile political environment, characterized by a fragile balance o f power among the main political forces and the ensuing short election cycles, poses serious hurdles to sensible economic policies. In this situation, national economic interest can often be compromised by actions taken for short term political gains. The short election cycle spawns populist measures, which undermine fiscal health and feed unrealistic expectations. As long as the expectations o f ordinary citizens continue to be unmet inthe short term, there is a real risk o f even worse political scenarios --such as in Latin America-- where opportunistic political figures win favor amonga population disenchanted with traditional political groups. 62 Sri Lunku Developmenr Policy Review 3.106 Welfare State vs. Market. Beyond the differences in policy stance and the struggle for power of the two leading political parties, a major challenge is to manage the relatively high expectations of Sri Lanka's population which, by and large, continues to regard the state as the main provider of economic welfare. This expectation, which is deeply rooted in Sri Lanka's socialist tradition, i s clearly depicted in Joan Robinson's commentary in 1958 on the constraining effect of labor relations on the country's effort to design a national development strategy to absorb a rapidly growing labor force: Ceylon has imported from the advanced capitalist countries (along with a modernized death-rate) ...the ideals of the werfare state, and her trade union movement has imported the conception that belong to unions in a developed economy,whose business is to keep profits in check and secure an acceptable share of national output for the workers. Theseadvanced capitalist economies were under-developed once, and they have behind them a long histo9 of entelprise and accumulation (not to mention ruthless exploitation of labor). They did not evolve their present relatively humane and relatively egalitarian philosophy until after the industry was well rooted and until after birth rates had come down to manageable levels. Ceylon has tasted thefruit before she hasplanted the tree. Her trade unions are anxious to share in profits but the energetic, enteiprising and thrijiy capitalistfor them to share have not yet appeared (Robinson 1958, pp 40-41) (emphasis added). 3.107 Interestingly, despite policy reforms over the years, Robinson's characterization of Sri Lanka's labor market situation still rings true. This view o f the state as a prime benefactor seems to underlie the wide-spread resistanceto privatization, the expectation that the state shouldbe employer o f first resort, and explains why in some areas, such as university education, people refuse to even consider private sector participation. Related deep-rooted beliefs include the paternalistic approach toward farmers (who are not really trusted to make sound decisions on the use o f their land) and the view that `self-sufficiency`, and not higher incomes, is the best way to ensure food security. Seemingly well-intentioned policies may have backfired. The agricultural subsidies leading to concentration in low-value paddy production is a case in point. Another may be the policy o f Swabasha, making Sinhala and Tamil the medium o f instruction. Originally introduced to correct the perceived unfkr advantage accruing to upper-income groups sending children to English-medium schools, the policy appears to have contributed to declines in quality o f education without reaping significant improvement inequity. 3.108 Against thls backdrop, Sri Lanka's own experience clearly shows that market-oriented policies have succeededinbringingsubstantial economic growth andpoverty reductioninurban areas while heavy state intervention in agriculture has stifled the rural economy with seriously adverse implications for the poor. Therefore, there is enough evidence in Sri Lanka to show that there i s no trade-off between market-driven growth and poverty reduction. However, a well functioning market economy is usually associated with a well functioning public sector. Indeed, the state has a very important role to play on several fronts: in establishing an appropriate regulatory framework; addressingmarket failures; providing public goods and services; and, very importantly, inprotecting the poor. 63 Sri Lanka Development Policy Review REFERENCES Abeyratne, Sirimal: "Economic Roots o fPolitical Conflict: The Case of Sri Lanka", forthcoming. Ahmed, S. and Ranjan, P. (1995) Promoting Growth in Sri Lanka: Lessons from East Asia, World Bank, Policy ResearchWorking PaperNo. 1478. Ahmed, S.M. and Ansari, M.I. (1998) Financial Sector Development and Economic Growth: The South-Asian Experience, Journal o fAsian Economics, 9(3): 503-517. Anderson, Kym (2002), `Agricultural Development and Trade Reforms: Their Contribution to Economic Growth, Poverty Alleviation and Food Security in Sri Lanka' (Background paper prepared for a World Bank Sector Mission to Sri Lanka), School o f Economics and Centre for International Economic Studies, Universityo fAdelaide. Arunatilake et. al. "The Economic Cost ofthe War in Sri Lanka, IPS January 2000." Arunatilake, N., Jayasuriya, S. andKelegama, S. (2001) The Economic Cost o fthe War in Sri Lanka, World Development, 29(9): 1483-1500. Asian Development Bank (1995), Industry Sector Policy Review (TA. N o 2161-SRI), Manila: ADB. Asian Development Bank (2003) Country Strategy and Program, 2004-2008: Sri Lanka (CSP SRI 2003-19), Manila: Asian Development Bank (august 2003). Athukorala, P. (1996) Labour Productivity inthe Manufacturing Sector in Sri Lanka, Department of National Planning, Ministryo fFinance andPlanning, Colombo. Athukorala, P. and Jayasuriya, S. (1994) Macroeconomic Policies, Crisis, and Growth in Sri Lanka, 1969-90,Washington, DC, World Bank. Athukorala, P. and Rajapathirana, S. (1993) Liberalization o f the Domestic Financial Market: Theoretical Issues with Evidence from Sri Lanka, International Economic Journal, 7(4): 17-32. Athukorala, Prema-chandra (1990), `International Contract Migration andthe Reintegration o fReturn Migrants: The Experience o f Sri Lanka', InternationalMigration Review, 2(2), 322-346. Athukorala, Prema-chandra and Sisira Jayasuriya (1996), Macroeconomic Policies and Export Competitivenessin Sri Lanka, Colombo: The National Development Council. Athukorala, Prema-chandra (1995), `Foreign Direct Investment and Manufactured Export Expansion: the Case o f Sri Lanka', WorldEconomy, 18(4), 543-64. Athukorala, Prema-chandra (1992), ` The Use o f Migrant Remittances in Development: Lessons from the Asian Experience', Journal of InternationalDevelopment, 4(5), 511-541. Athukorala, Prema-chandra and Sisira Jayasuriya (1994), Macroeconomic Policies, Crises, and Growth in Sri Lanka, 1969-90,Washington, D.C.: World Bank. Athukorala, Prema-chandraand Bambang H. Santosa (1996), `Gains from Indonesian Export Growth: Do Linkages Matter?', BuZZetin of Indonesian Economic Studies,33(2), 73-96. Athukorala, Prema-chandra and Saman Kelegama (1998), "The Political Economy o f Agricultural Trade Policy: Sri Lanka inthe Uruguay Round', Contemporary SouthAsia, 7(1): 7-26. 65 Sri Lanh DevelopmentPolicy Review Athukorala, Prema-chandra and Sarath Rajapatirana (2000), Liberalization and Industrial Transformation: Sri Lanka in International Perspective, Delhi andOxford: Oxford University Press. Azam, Jean-Paul, David Bevan, Paul Collier, Stefan Dercon, Jan Gunning and Sanjay Pradhan. "Some Economic Consequences of the Transition for Civil War to Peace". World Bank Policy ResearchWorking Paper 1994. Bandara, J. (1995) Dutch Disease in a Developing Country: The Case of Foreign Capital Inflows to Sri Lanka, SeoulJournal o fEconomics, 8(3): 311-329. Bhalla, Sujit S. (1991), "The Political Economy o f Agricultural Pricing Policies in Sri Lanka" in Krueger, Anne, M. Schiff and A. Valdes (eds.), The Political Economy of Agricultural Pricing Policy, Volume 2, Ballimore: Johns HopkinsUniversity Press. Board of Investment o f Sri Lanka (2003) Industrial Factor Costs: A Guide to Investment Tariffs and Costs inSri Lanka, Colombo . Boardof Investmento f Sri Lanka (2003) Textile & Apparel Sector: An Overview, Colombo. BoardofInvestment o f Sri Lanka (2003) The IT Industry, Colombo. Boardof Investment o f Sri Lanka Annual Report (2001), Colombo. Breakingthe Conflict Trap: Civil War andDevelopment Policy. World Bank Policy ResearchPaper 2003. Bruton, H. (1992) The Political Economy o f Poverty, Equity and Growth: Sri Lanka and Malaysia, New York, OxfordUniversity Press. Bulir, Ales and Timothy Lane. "Managing the Fiscal Impact of Aid". IMF Working Paper, December2002, Volume 39, Number4. Burges, Robin and Nicholos Stem (1993), `Taxation and development', Journal of Economic Literature, 31(2), 762-783. Campos, Jose E. and Hilton L. Root (1996), The Key to the Asian Miracle: Making Shared Growth Credible, Washington D.C.: Brookings Institution. Central Bank o fSri Lanka. Annual Report, various years. Collier, Paul et. al. "Some Economic Consequences of the Transition from Civil War to Peace". Policy ResearchWorking Paper 1392. World Bank 1994. Corbo, Vittorio (1997), `Trade Reform and Uniform Import Tariffs: The Chilean Experience', American Economic Review (papers andproceedings, 87: 73-77. Corden W. M. (2002), Too Sensational: On the Choice of Exchange Rate Regimes, Cambridge, Mass.: MIT Press. Cuthberson, S. (2003) Tariff and Trade Policy Framework for Sri Lanka in 2003, Economic and Social Commission for Asia and the Pacific-TariffAdvisory Council, Colombo (draft). Cuthbertson, Sandy and Prema-chandra Athukorala (1990) `Sri Lanka,', Part 3 in D Papageorgiou, M Michaely and A MChoksi (eds) Liberalising Foreign Trade: Indonesia, Pakistan and Sri Lanka, Oxford: Basil Blackwell, pp283-411. 66 Sri Lanka Development Policy Review Deraniyagala, S. (2001) Adaptive Technology Strategies andTechnical Efficiency: Evidencefrom the Sri LankanAgricultural Machinery Industry, Journal of International Development, 13: 59-71. Deraniyagala, S. (2001) The Impact o f Technology Accumulation on Technical Efficiency: An Analysis of the Sri Lankan Clothing and Agricultural Machinery Industries, Oxford Development Studies, 29(1): 101-114. Dunham, D. and Kelegama, S. (1995) Economic Reform and Govemance: The Second Wave of Liberalization in Sri Lanka 1989-93, Institute o f Policy Studies, ResearchStudies Govemance Series No. 2. Dutz, M. and Suthiwart-Narueput, S. (1995) Competition Issues Beyond Trade Liberalization: Distribution and Domestic Market Access, Chapter 4 in "Regulatory Policies and Reform: A Comparative Perspective", World Bank, Washington, DC. Easterly, W. and Levine, R. (2001) It's Not Factor Accumulation: Stylized Facts and Growth Models, The World Bank Economic Review, 15(2): 177-219. ESCAP (2003), Tariff and Trade Policy Framework for Sri Lanka in 2003, Bangkok: ESCAP (processed). Fink, C., et a1(2002) Trade inIntemationalMaritime Services: How MuchDoesPolicy Matter?, The World Bank Economic Review, 16(1): 81-108. Fry, M. (1995) Money, Interest and Banking in Economic Development, Johns Hopkins University Press, Baltimore. Ghosh, S. and Kraay, A. (2000) Measuring Growth in Total Factor Productivity, PREM Notes Economic Policy, 42. Girijasankar, M. and Chowdhury, A. (2001) Inflation and Economic Growth: Evidence from Four South Asian Countries, Asia-Pacific Development Journal, 8(1): 123-134. Guisinger, Stephen and Associates (1985), Investment Incentives and Performance Requirements, New York: Praeger. Gunatilaka, R. and Kelegama, S. (1996) Study on the Impact of Labour Legislation on Labour Demandin Sri Lanka, DepartmentofNationalPlanning, Ministry of Finance and Planning, Colombo. Haughton, Jonathan. "The Reconstruction o f War-Tom Economies". Consulting Assistance on Economic ReformI1DiscussionPaperNo. 23 June 1998. Heltberg, Rasmus. "Labor Markets to Support Growth and Poverty Alleviation in Sri Lanka". Unpublished backgroundpaper. World Bank, November 2003. Hettiarachchi, W. (1998) Globalization-Liberalizing the Capital Account in Sri Lanka, Institute of Policy Studies, ResearchStudies Macroeconomic SeriesNo. 11, Colombo. Humphrey, J. and Schmitz, H. (1995) Principles for PromotingClusters & Networks o f SMEs, United Nations IndustrialDevelopment Organization, Discussion PaperNo. 1. Intemational Monetary Fund. GFS, various years. InstituteofPolicy Studies, Sri Lanka: State ofthe Economy (various years), Colombo. 67 Sri Lanka Development Policy Review Jayasundera, P.B. (2003), `Designing a Tax Structure to Support Economic Growth and Fiscal Consolidation inSri Lanka.' Colombo. Jayasuriya, Sisira (2003), `Exchange Rate Issues and Debates in Sri Lanka', paper prepared for a Festschrift volume inhonour o f Gamani Corea, Departmento f Economics, University of Melbourne (processed). Jayasuriya, Sisira and Dushni Weerakoon (2001), `India-Sri Lanka Trade and Investment Links', in Parthasarathi Shome (ed.), India and Economic Cooperation in South Asia, New Delhi: Indian Council For International Economic Relations, 111-139. Jayawardena, A.S. (2002) Economic Performance o f Sri Lanka inthe Nineties, South Asia Network of Economic ResearchInstitutes, New Delhi(conference paper). Kaufinann, D.and Kraay, A. (2002) Growth Without Governance,World Bank. Kaufinann, D., et a1 (2002) Governance Matters 11: Updated Indicators for 2000/01, Policy Research Working PaperNo. 2772, World Bank, Washington, DC. Keen, D. "The Economic Functions o fViolence inCivil Wars". Adelphi PaperNo 320 1998. Keesing, Donald B. (1967), `Outward Looking Policies and Economic Development', Economic Journal,~77(306):3030-320. Kelegam, Saman (2003), `Sri Lankan Exports to India: Impact of Free Trade Agreement', Economic and Political Weekly,July 26,3 153-54. Kelegama, S. (1999) Indo-Sri Lanka Trade and the Bilateral Free Trade Agreement: A Sri Lankan Perspective, Asia-Pacific Development Journal, 6(2): 87-106. Kelegama, S. (2000) Development in Independent Sri Lanka: What Went Wrong?, Economic and Political Weekly, April 22,2000. Kotler, P., et a1(1997) The Marketing o fNations: A Strategic Approach to BuildingNational Wealth, The Free Press, New York. Krugman, P. (1994) Competitiveness: A DangerousObsession, ForeignAffairs, 73(2) Lal, Sanjaya and Ganesh Wignaraja (1992), `Foreign Involvement by European Firms and Garment Exports by Developing Countries', Asian-Pacific DevelopmentJoumal, 1(2), 21-28. Laporta, R. (1996) et al, Law and Finance, National Bureau of Economic Research Working Paper No. 5661. L e Billion P with Joanna Macrae, Nick Leader and Roger East. "The Political Economy of War: What ReliefAgenciesNeedto Know". Humanitarian PracticeNetworkPaper 33, July 2000. Levine, R., et a1 (2000) Financial Intermediation and Growth: Causality and Causes, Joumal of Monetary Economics, August 2000,46 (1): 31-77. Little, I.M.D. (1981), `The Experience and Causes o f rapid Labour-Intensive Development in Korea, Taiwan Province, Hong Kong, and Singapore, and Possibilities for Emulation', Edy Lee (ed) (1981), Export-led Industrialization and Development, ILO: Geneva) (Reproduced I.M.D. Little (1999) Collection and Recollections: Economic Papers and their Provenance, Oxford: Clarendon Press, 213- 40.). 68 Sri Lanka DevelopmentPolicy Review Mahajan, S., "TFP Toolkit"; World Bank; 2001. Michaeley, Michael, Demetis PapageorgiouandArmeane Choksi (1991), LiberalizingForeign Trade: Lessons of Experience in theDeveloping World,Oxford: basil Blackwell. Morgan, Theodore H. (1998), Foreign Direct Investment and Development: The new Policy Agenda for Developing Countries, Washington DC: Institute for International Economics. Morisset Jaques and Neda Pirnia (2001), `How Tax Policy and Incentives Affect Foreign Direct Investment': A Review', . Foreign Investment Advisory Service Occasional Paper 15, Washington DC: International Finance Corporation andWorld Bank, pp.6-108. Ofstad, A. (2000) Economic Growth, Employment and Decentralized Development in Sri Lanka, MichelsenInstitute, Working PaperNo. 11. Papanek, G. (2003) Growth and Poverty in Sri Lanka: From Controls to Market, Development Alternatives Inc.-Boston Institute for Developing Economies, Pro-Poor Economic Growth Research Studies, No. 10. PCTT (Presidential Commission on Tariff and Trade)( 1994) Report of the Presidential Commission on Tarifland Trade- 1994,Colombo: Government PublicationBureau. Peiris G. H. and K.M. de Silva. "Managing Group Grievances and Internal Conflict: Sri Lanka Country Report". Working paper 13, Netherlands Institute of International Relations Clingendael, June 2003. Radwan, Ismail. "Afghanistan : Post Conflict Reconstruction and the Role of the Private Sector". World Bank internalpaperNovember 2001. Rajapatirana, S. (1988) Foreign Trade and Economic Development: Sri Lanka's Experience, World Development, 16(10): 1143-1156. Riedel, J. (1974) `Factor Proportions, Linkages and the Open Developing Economy', Review of Economics and Statistics, 75(4), 487-94. Robinson, Joan (1958), ` Economic Possibilities o f Ceylon' in Planning Secretariat, Papers by Visiting Economists, Colombo: Government Publication Unit, 39-45 (Reproduced in Prema-chandra Athukorala (ed), The Economic Development of South Asia, Volume 1, Cheltenham, UK: Edward Elgar, 76-82. Rodngo, Chandra (1983), A National Wage Policy for Sri Lanka, Colombo: Sri Lanka Institute o f Development Administration. Salaries Commission (2002). Tissa Devendra Salaries Commission Report. Government of Sri Lanka. Saravananthan, Muttukrishna. "An Introductionto the Conflict Time Economy of the North and East Province of Sri Lanka". ICES working paperNo. 1May 2003. Sarvananthan, Muttukrishna. "What ImpedesEconomic Revival inthe N&EProvince of Sri Lanka?' ICES working paperNo 2. Shanmugaratnam, N. "Peace Dividend, Development and the Distributional Problem in Sri Lanka". SouthAsia Citizens Web, April 13,2003. 69 Sri Lanka Development Policy Review Singh, A. (1997) Asia Pacific Tourism Industry: Current Trends and Future Outlook, Asia-Pacific Journal o f Tourism Research. SLBFE (Sri Lanka Bureau of Foreign Employment) (2003), Statistical Handbook on Migration - 2002, Colombo: SLBFE. Snodgrass, Donald R. (1998), "The Economic Development o f Sri Lanka: A Tale of Missed Opportunities", inRobert I.Rotberg (ed.), Creating Peace in Sri lanka: Civil War and Reconciliation, Washington DC: Brookings Institution Press, 89-107. Soto, H., (2000) The Mystery o f Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else, Basic Books, New York. Sri Lanka: Assessment of the Needs inthe Conflict Affected Areas. UN, ADB and World Bank May 2003. Stewart, Francis andValpy Fitzgerald. "Assessing the Economic Costs of War." Subramanian, Arvind (1994), The Case for Low Uniform Tariffs, Working Paper, Washington DC: International Monetary Fund. Subramanian, Arvind and Devesh Roy (2003), `Who Can Explain the Mauritian Miracle`? Mead, Romer, Sachs, or Rodrik?', in Dani Rodrik (ed.), In Search o f Prosperity: Analytical Narratives on Economic Growth, Princeton and Oxford Princeton University Press, 205-243. UNCTAD (2003), Investment Policy Review: Sri Lanka, November, Geneva: UNCTAD (draft report). U1Haq, M. (2001). Human Development inSouth Asia. Velde, D.W. (2001) Policies Towards Foreign Direct Investment inDeveloping Countries: Emerging Best Practices and Outstanding Issues, Overseas Development Institute, London. Venugopal, Rajesh. "The Global Dimensions o fConflict inSri Lanka". QEH Working paper number 99 February 2003. Weigand, Robert (1983), `International Investments: Weighing the Incentives', Harvard Business Review, 61(6), 146-152. Weliwita, A., et a1 (1999) Real Exchange Rate Volatility and Sri Lanka's Exports to the Developed Countries 1978-96,Journal o f Economic Development, 24(1): 147-165. Wells, Louis T. Jr, and Nancy J. Allen (2001), `Using Tax Incentives to Compete for Foreign Investment: Are They Worth the Cost?', Foreign Investment Advisory Service Occasional Paper 15, Washington DC: International Finance Corporation and World Bank,pp. 1-67. Wells, Luice T. Jr. (1986), `Investment Incentives: An Unnecessary Debate', n e CTC Reporter,22: 58-60. World Bank (1988) Sri Lanka: A Break with the Past-The 1987-90 Program o f Economic Reforms and Adjustment, Report No. 7220-CE. World Bank (1990) Sri Lanka: Sustaining the Adjustment Process, Report No. 8951-CE. 70 SriLnrika Development Policy REview World Bank (1992) Sri Lanka: Strengthened Adjustment for Growth and Poverty Reduction, Report NO.10079-CE. World Bank (1995), Sri Lanka: Non-PlantationCrop Sector Policy Alternatives, Workingpaper I-7, Agriculture andNaturalResourceDivision, South AsianRegions, WashingtonDC: World Bank. World Bank (1998), Research Working Paper "Government Employment and Pay: The Global and RegionalEvidence," Washmgton, D.C. World.Bank (1999). Sri Lanka:A FreshLook at Unemployment, ReportNo. 19609-CE. World Bank (2000). Sri Lanka:RecapturingMissed Opportunities,ReportNo. 20430-CE. World Bank (2002) Sri Lanka: PovertyAssessment, ReportNo. 22535-CE. World Bank (2003), Trade Policies in South Asia: An Overview, Washington DC: World Bank. brocessed). World Bank (2003) Sri Lanka: PromotingAgricultural andNon-Fanning Sector Growth, Report No. 25387-CE. World Bank (2004) Sri Lanka: Improving the Rural and Urban Investment Climate (Draft, to be published) WTO (WorldTrade Organization) (1996), Trade Policy Review: Sri Lanka- 1995, Geneva: WTO. 71 Statistical Annexes Box A 1:A List of PolicyOptionsfor Consideration vlanagingPublic Finances ' Publicdebtto bereducedto amanageablelevelby reducingtheprimary fiscaldeficit andlimitingreliance on non-concessionaryborrowing. 1 Publicexpendituresto berationalizedby establishingapolicy driven, outputbasedandconsultativebudget formulation systemunder amediumtermbudgetframework. 1 Financialburdenof state-ownedenterprisesonthe budgetto bereducedby limiting transfers andcontingent liabilities andsubjectingthemto ahardbudgetconstraint. 1 Effectivenessofexistingwerfareprograms, includingSamurdhi, to be increasedby improvingtargetingto better reachthe poorandmakingpaymentsmoreprogressive. 1 Public sector wage andrecruitment policies to be de-politicizedaddressingoverstaffing, and correcting inappropriate salary scales and administrative fragmentation 1 Tax administrationto bestrengthen by establishingawell integratedrevenueadministration, establishinga separatetax auditunit, andmakingthe largetax payer unit moreproactive. v Incometax yield toberaisedbyphasinginamoratoriumon existingtax holidays, exclusions, and concessions. 1 Extendingexcisetaxes to income-elasticluxury consumer goods to be explored 1 VAT coverage to retail trade andreduceexemptionsto beconsidered. rrade Policy A competitiverealexchangerateto bemaintained 1 Tariff reversalsespecially inthe agriculturalsector to be avoidedandmovemadetowards a single, low, uniform tariff. D InbilateralFreeTradeAgreements(FTA) the cost oftrade diversions, capitalaccountconcessions, and distractionsto the unilateraltrade liberalizationagendathat bilateralFTAs entailto bereviewed. [nvestment Policy D Foreigninvestmentpolicy regimeto bemadeattractiveby makingeffectiverecentlylegislatedlawsandbringing fiscalincentivesinlinewith InlandRevenueLaws. A liberalemigrationpolicy to bepursuedandmigrantworker householdsprovidedeasy access to bank facilities and amarketresponsiveexchangeratepolicy. Recent labor reforms to be implemented, alongwith encouragingconsolidation, improving infrastructure, and facilitatingcreditto small garmentexportersto assistwith the postMFA adjustment. Nonplantation Agriculture Full andtransferableownershiprightsof landbe allowedthrough appropriatelegislationandinstitutionalsetting. More liberalseedandphyto-sanitaryregulationsto be allowed Scopeofresearchandextension servicesthroughprivatesector participation,includingNGOs, to bepursued. Sustainableuse ofwater throughthe enactment ofan appropriateNationalWater Billto be ensured. Infrastructure Fundingto beincreasedfor roadmaintenanceand rehabilitationfocusedon improvingnationalhighways. Public institutionsintransport sector to bestrengthenedandprivatesector participationintransport infrastructure and serviceprovisionincreased. Participationofstakeholders(rural poor andenvironmentgroups) informulatingtransport strategyto be increased. CEBto berestructuredandregulatoryfunctionstransferredto independentPUCSL. Electricitygenerationcapacityto beexpandedinaccordanceto anupdatedleast-cost expansionplanandwithout political interference. Provisionofoff-grid connectionsto ruralhouseholdsto beaccelerated (e.g., IDA financed REREDProject). Education Allocationfor primaryand secondary educationto be increasedbyreallocatingfunds fromtertiaryeducation. Privatesector participationineducation, especiallyintertiary education,to be encouragedby developinganew educationact and securingaccess by the poorto tertiaryeducationthrough avoucher system. Cost effectivenessofeducationto be increasedby rationalizingschools, administrationandstaffing; andraising the student-teacher ratio. Servicedeliveryto be improvedby decentralizingeducationmanagementto schools. Relevanceofeducationto beincreasedwith afocus on ITandallowingEnglishas mediumof instruction. 73 StatisticalAnnexes Table A1:Social Indicators Latestsingleyear Same regionlincomegroup Lower- South middle- 1970-75 1980-85 1995-03 Asia income POPULATION Total population, mid-year (millions) 13.5 15.8 19.4 1,355.1 2,047.6 Growth rate ("hannual average for period) 1.5 1.4 1.3 1.9 1.1 Urbanpopulation ("ho fpopulation) 22.0 21.1 23.6 28.4 42.0 Total fertility rate (births per woman) 3.9 2.9 2.1 3.3 2.1 POVERTY (?? ofpopulation) National headcount index 22.7 Urbanheadcount index 7.9 Ruralheadcount index 24.7 INCOME GNIper capita (US$) 310 380 930 440 1,130 Consumer price index (1995=100) 12 34 154 142 146 Foodprice index (1995=100) 34 159 INCOME/CONSUMPTION DISTRIBUTION Giniindex 34.4 Lowest quintile ("ho f income or consumption) 8.0 Highest quintile (%o f income or consumption) 42.8 SOCIAL INDICATORS Publicexpenditure Health (% o fGDP) 1.7 0.9 2.3 Education ("ho f GDP) 2.7 2.6 3.4 3.0 4.6 Social security andwelfare ("ho f GDP) 6.1 3.0 2.9 Gross primaryschoolenrollment rate PAof agegroup) Total 110 91 Male 111 91 Female 110 91 Access to an improvedwater source (?Aofpopulation) Total 83 87 80 Urban 91 92 95 Rural 80 85 69 Immunizationrate (% under 12 months) Measles 20 99 53 89 DPT 70 98 57 89 Child malnutrition (% under 5 years) 33 49 11 Lifeexpectancy at birth hears) Total 66 69 74 62 69 Male 65 67 72 62 67 Female 68 71 76 63 72 Mortality Infant(per 1,000 livebirths) 45 24 12 73 33 Under 5 (per 1,000 live births) 100 48 17 96 41 Adult (15-59) Male (per 1,000 population) 214 200 244 227 192 Female (per 1,000 population) 196 152 124 212 125 Maternal (per 100,000 live births) 23 Birthsattendedbvskilled health staff (%) 97 \ I CAS Annex B5. This table was produced from the C M U LDB system. Note: 0 or 0.0 meanszero or lessthanhalfthe unit shown. 74 Statistical Annexes Table A2: Key Economic Indicators Actual Estimate Indicator 1998 1999 2000 2001 2002 2003 Nationalaccounts(as YOof GDP) Gross domestic producta 100.0 100.0 100.0 100.0 100.0 100.0 Agriculture 21.1 20.7 19.5 19.5 19.8 19.0 lndustry 27.5 27.3 27.5 26.7 26.6 26.5 Services 51.4 52.1 53.0 53.8 53.6 54.5 Total consumption 80.9 80.5 82.6 84.7 85.5 84.3 Gross domestic fixed investment 25.1 27.3 28.0 22.1 21.3 22.3 Government investment 3.3 3.2 3.3 3.0 2.0 2.3 Private investment 21.8 24.1 24.7 19.0 19.3 20.0 Exports (GNFS)b 36.2 35.5 39.7 37.0 36.1 35.8 Imports (GNFS) 42.2 43.3 50.5 43.7 42.9 42.4 Gross domestic savings 19.1 19.5 17.4 15.3 14.5 15.7 Gross national savings 23.4 23.5 21.4 19.5 19.5 21.3 Memorandumitems Gross domestic product 15795 15657 16305 15669 16545 18237 (US$ million at current prices) GNIper capita (US$, Atlas method) 810 860 890 840 850 930 Real annualgrowth rates (%) Gross domestic product at market prices 4.7 4.3 6.0 -1.4 4.0 5.9 Gross domestic income 9.6 3.6 3.6 -0.3 4.1 6.4 Real annualper capita growthrates ("A) Gross domestic product at marketprices 3.5 2.8 4.3 -2.9 Total consumption 5.7 3.8 3.3 -1.1 Private consumption 5.7 3.8 2.7 16.9 Balanceof Payments (US$ millions) ~xports(GNFS)~ 5712 5578 6475 6183 5967 6539 Merchandise FOB 4798 4610 5522 4817 4699 5131 Imports(GNFS)b 6659 6800 8235 7135 7080 7681 Merchandise FOB 5889 5980 7320 5974 6106 6669 Resourcebalance -947 -1222 -1760 -952 -1111 -1143 Net current transfers 900 912 998 960 1128 1234 Current accountbalance -227 -564 -1066 -272 -236 -101 Netprivate foreign direct investment 193 177 176 172 181 171 Long-term loans(net) 398 397 297 354 91 515 Official 376 155 154 184 112 548 Private 22 242 143 170 -21 -33 Other capital(net, incl. errors & omissions) -135 -351 -26 16 302 -83 Change inreserves` -229 340 619 -270 -338 -502 Memorandum items Resourcebalance ("ho fGDP) -6.0 -7.8 -10.8 -6.1 -8.5 -8.4 Real annualgrowthrates Merchandise exports (FOB) 13.2 4.5 9.8 2.4 3.4 7.4 Merchandise imports (CIF) 9.9 3.7 13.7 1.3 11.0 11.7 75 StatisticalAnnexes Table A2: Key Economic Indicators Actual Estimate Projected Indicator 1998 1999 2000 2001 2002 2003 Publicfinance(as % of GDP at marketprices)d Current revenues 17.2 17.7 16.8 16.5 16.5 15.7 Current expenditures 19.6 18.7 20.2 21.4 20.8 19.0 Current account surplus (+) or deficit (-) -2.4 -1.o -3.4 4.9 -4.3 -3.3 Capital expenditure 5.3 5.5 5.4 4.8 4.6 4.9 Foreign financing 1.7 0.7 0.7 1.6 0.6 2.9 Monetaryindicators M2/GDP 37.1 38.7 38.5 39.2 40.1 40.2 Growth o fM2 (%) 13.2 13.4 12.8 13.6 13.4 15.3 Priceindices(YR82 =loo) Merchandise export price index 91.4 84.0 91.7 78.1 79.0 84.0 Merchandise import price index 94.7 97.0 108.2 87.1 85.7 84.9 Merchandise terms oftrade index 96.4 86.6 84.7 89.6 92.0 99.0 Real exchange rate (US$LCU)" 126.2 117.4 115.5 116.4 116.6 114.2 Consumer price index (% change) 9.4 4.7 6.2 14.2 9.6 6.3 GDP deflator (%change) 9.2 4.2 6.2 13.2 8.4 5.1 a. GDP at factor cost. b. "GNFS" denotes "goods andnonfactor services." c. Includes use ofIMFresources. d. Consolidatedcentral government. e. "LCU" denotes "local currency units." An increase inUSWLCUdenotes appreciation. 76 StatisticalAnnexes Table A3: Gross DomesticProductat CurrentFactor Cost Prices In Rs. Million Year GDP Agriculture Industry Services GNPat Current Factor Costs 1978 40,479 12,332 11,030 17,l 17 40,547 1979 49,782 13,412 14,047 22,323 49,880 1980 62,246 17,151 18,450 26,645 62,229 1981 79,337 21,977 22,206 35,154 77,913 1982 94,679 24,964 24,887 44,828 92,925 1983 113,878 32,180 29,992 51,706 110,758 1984 140,039 40,138 36,856 63,045 136,32 1 1985 148,321 41,069 38,859 68,393 144311 1986 163,713 44,355 43,548 75,810 160,204 1987 177,73 1 47,923 48,763 81,045 174,106 1988 203,516 53,600 54,300 95,616 199,107 1989 228,138 58,462 61,041 108,635 223,280 1990 290,615 76,488 75,555 138,572 284,907 1991 337,399 90,257 86,284 160,858 330,667 1992 386,999 100,080 99,2 18 187,701 379,834 1993 453,092 111,659 116,007 225,426 447,954 1994 523,300 124,370 137,005 261,925 515,574 I995 598,327 137,678 158,764 301,885 592,310 1996 695,934 156,108 184,056 355,770 684,676 1997 803,698 175,774 216,177 411,747 794,120 1998 912,839 192,665 251,401 468,773 901,283 1999 994,730 205,599 271,388 517,743 976,899 2000 1,125,259 223,926 306,977 594,356 1,102,177 2001 1,245,598 249,790 333864 661,944 1,221,768 2002 1,403,309 287,840 368695 746,774 1,378,151 2003 1,560,806 296,5 10 410082 854,214 1,541,985 Source: Central Bankof Sri LankaAnnual Report (multiple issues) 77 Statistical Annexes Table A4: Gross Domestic Product at 1996 Prices I n Rs. Million Year GDP GNP RealGDP Growth (%) 1978 302,426 302,931 8.2 1979 321,543 322,179 6.3 1980 340,209 340,114 5.8 1981 359,862 353,402 5.8 1982 378,113 371,109 5.1 1983 396,867 385,994 5.O 1984 416,915 405,846 5.1 1985 437,582 427,227 5.0 1986 456,3 16 446,535 4.3 1987 462,950 453,508 1.5 1988 475,442 465,143 2.7 1989 486,141 475,789 2.3 1990 516,153 506,016 6.2 1991 539,955 529,182 4.6 1992 563,062 552,637 4.3 1993 602,172 595,343 6.9 1994 636,06 1 626,670 5.6 1995 670,742 663,997 5.5 1996 695,934 684,676 3.8 1997 739,763 730,947 6.3 1998 774,796 764,908 4.7 1999 808,340 794,340 4.3 2000 857,03 5 840,200 6.0 2001 843,794 829,056 -1.5 2002 877,248 862,712 4.0 2003 929,038 918,261 5.9 Source: Central Bank o f Sri Lanka Annual Report (multiple issues) 78 Statistical Annexes Table A5: GrossDomesticProduct at Current Market Prices In Rr MiIlion Gross Domestic Net NetFactor Net Domestic National Year Productat Importsof Income Private Public Private Public Market GNFS Abroad Transfers From Private Consumption Consumption Investment Investment SavingsRatio SavingsRatio (YoofGDP) (Yo of GDP) Prices 1978 42,665 2,037 -246 342 32,105 4,043 5,83 1 2,723 15.3 15.5 1979 52,387 6,309 -219 754 40,371 4,798 9,783 3,744 13.8 14.8 1980 66627 15,022 -389 2,260 53,399 5,685 16,776 5,689 11.2 14.0 1981 85,005 13,666 -1,706 3,918 68,75 1 6,310 19,604 4,006 11.7 14.3 1982 99,238 18,757 -1,981 5,494 79,226 8,242 25,583 4,944 11.9 15.4 1983 121,601 18,365 -3,265 6,441 94,945 9,889 29,234 5,898 13.8 16.4 1984 153,746 9,132 -3,475 7,03 1 111,235 11,935 32,693 7,015 19.9 22.2 1985 162,375 19,409 -3,428 7,212 126,503 16,599 30,890 7,792 11.9 14.2 1986 179,474 20,839 -3,583 7,983 139,370 18,480 32,879 9,584 12.0 14.5 1987 196,723 20,664 -4,363 9,226 151,949 19,538 34,726 11,174 12.8 15.3 1988 221,982 23,886 -5,342 10,187 173,457 21,849 37,437 13,125 12.0 14.2 1989 251,891 23,921 -5,865 11,840 194,680 26,4 10 40,33 1 14,391 12.2 14.6 1990 321,784 25,364 -6,685 14,518 244,288 31,405 58,790 12,665 14.3 16.8 1991 372,345 37,658 -7,367 16,623 288,214 36,633 69,273 15,883 12.8 15.2 1992 425,283 39,394 -7,821 20,253 320,466 40,972 89,557 13,682 15.0 17.9 1993 499,565 47,686 -5,979 27,090 373,785 45,791 106,920 20,755 16.0 20.2 1994 579,084 68,361 -8,310 30,989 434,933 56,002 138,849 17,661 15.2 19.1 1995 667,772 69,714 -6,958 34,820 489,057 76,604 148,180 23,645 15.3 19.5 1996 768,128 68,573 -11,258 39,242 569,4 16 81,021 162,76 1 23,503 15.3 19.0 1997 890,272 62,866 -9,409 46,472 643,839 92,196 187,150 29,953 17.3 21.5 1998 1,017,986 61,154 -11,556 54,785 723,506 99,745 221,879 34,010 19.1 23.4 1999 1,105,963 86,090 -17,831 62,438 790,379 99,851 266,593 35,230 19.5 23.5 2000 1,257,636 133,373 -23,083 73,810 906,188 132,189 311,480 41,152 17.4 21.5 2001 1,407,398 87,768 -23,830 87,902 1,041,041 144,441 267,3 18 42,366 15.8 20.3 2002 1,582,655 108,555 -25,159 104,938 1,214,117 139,311 305,860 31,922 14.5 19.5 2003 1,760,280 115,824 -18,821 116,307 1,343,896 139,268 352,174 40,766 15.7 21.3 Source: CentralBankofSri LankaAnnual Report(multiple issues) 79 Statistical Annexes Table A6: Index ofRealValue Added in Industry 1978=100 Wood& Paper& Chemicals, Non-Metallic Metal, Year Food & Textiles & Wood Petroleum, Basic Metal Machinery Beverage Apparel Paper Mineral Total Products Products Rubber and Products & Transport Plastic Products Equipment 1978 100 100 100 100 100 100 100 100 100 1979 90 80 112 98 113 98 132 80 95 1980 96 114 192 117 174 114 106 74 114 1981 84 128 163 84 242 92 88 84 117 1982 91 110 117 74 202 122 14 149 119 1983 247 1014 399 227 1089 181 265 186 436 1984 260 1275 418 195 1121 193 149 205 467 1985 313 1588 456 253 1016 194 91 222 503 1986 342 1 909 386 260 812 203 197 232 509 1987 342 2 277 387 259 923 199 201 247 557 1988 195 635 237 127 235 108 42 168 217 1989 219 694 232 115 198 92 41 199 226 1990 217 721 189 131 113 299 87 231 242 1991 244 715 201 135 135 278 100 246 268 1992 275 1206 216 147 153 305 96 250 323 1993 292 1430 242 184 165 337 93 227 356 1994 300 1312 288 202 184 400 109 250 371 1995 329 1582 290 200 203 387 98 241 402 1996 333 1765 214 187 236 374 113 228 420 1997 331 2 117 254 175 260 379 138 258 454 1998 348 2 284 245 158 319 375 152 272 484 1999 366 2 547 248 157 296 398 159 284 514 2000 378 3 001 260 155 356 385 184 219 563 2001 373 2 741 244 152 343 385 224 281 537 2002 390 2 174 241 149 363 393 230 286 553 2003 415 2 809 252 148 392 407 215 307 575 Source: Central Bank of Sn LankaAnnual Report (multiple issues) and Staffcalculations. 80 Statistical Annexes Table A7: Selected Tourism Statistics Year Tourist Arrivals in Thousands Number o f H o t e l Rooms 1978 196 5,347 1979 250 5,599 1980 322 6,042 1981 371 6,891 1982 407 7,539 1983 338 8,852 1984 318 9,627 1985 257 9,826 1986 230 9,794 1987 183 9,92 1 1988 183 9,977 1989 185 9,459 1990 298 9,556 1991 318 9,679 1992 394 10,214 1993 392 10,365 1994 408 10,742 1995 403 11,255 1996 302 11,600 1997 366 12,370 1998 381 12,772 1999 436 12,918 2000 400 13,311 2001 337 13,626 2002 393 13,818 2003 501 _ _ Source: Central Bank o f Sri Lanka Annual Report (multiple issues) 81 Statistical Annexes Electricity Fuels Year Installed Power Consumption (MT '1000s) Capacity Generation (MW) (GWB) Petrol Diesel Kerosene 1978 402 1,385 130 309 245 1979 402 1,525 115 349 230 1980 422 1,668 108 398 188 1981 522 1,872 109 421 168 1982 562 2,066 114 465 174 1983 592 2,114 117 464 162 1984 812 2,26 1 119 481 151 1985 1,016 2,464 122 489 154 1986 1,010 2,653 131 489 154 1987 1,138 2,708 140 497 156 1988 1,208 2,800 177 499 164 1989 1,240 2,858 189 484 160 1990 1,289 3,150 179 513 167 1991 1,289 3,376 160 540 173 1992 1,409 3,540 165 606 189 1993 1,409 3,979 173 666 192 1994 1,409 4,364 184 728 207 1995 1,409 4,783 190 728 222 1996 1,409 4,530 198 1,048 228 1997 1,575 5,145 193 1,295 225 1998 1,636 5,683 204 1,224 236 1999 1,691 6,184 213 1,377 243 2000 1,779 6,644 220 1,715 229 2001 1,901 6,627 244 1,675 228 2002 1,930 6,95 1 278 1,752 229 2003 2.223 7.612 _- -- -- Source: Central Bank o f Sri Lanka Annual Report (multiple issues) 82 Statistical Annexes RailTransport Road Transport Port Services Year Operated Public Sector PrivateSector Locomotives Kilometerage Operated Passenger Passenger *Vehicle Stock No. of Vessels Cargo Handled (millions) Kilometerage Kilometerage Kilometerage (thousands) Arrived MT (thousands) (millions) (millions) (millions) 2003 __ 8.08 __ _- _- 2,074 4,032 30,500 Source: CenmlBankof SriLankaAnnual Report(multiple issues) 83 Statistical Annexes Table A10: Selected Telecommunications Statistics Year No. of fixed Cellular Wireless Telephone Lines Phones Phones 1978 58,987 _ _ -_ 1979 59,617 _- -- 1980 60,070 -- -- 1981 62,193 -- -- 1982 68,201 -- -- 1983 71,112 -- -- 1984 80,770 -- 1985 87,686 ---- -- 1986 92,065 -- -- 1987 97,832 -- -- 1988 104,108 -- -- 1989 105,007 -- -- 1990 121,388 1,010 -- 1991 125,834 1,973 -- 1992 135,504 4,000 -- 1993 157,774 6,242 _- 1994 180,724 29,182 -- 1995 204,3 50 51,316 -- 1996 254,500 71,028 527 1997 315,241 114,888 26,381 1998 455,598 174,202 67,93 1 1999 580,199 256,655 91,717 2000 653,144 430,202 114,267 2001 708,200 667,662 121,082 2002 768,620 932,000 114,488 2003 823,000 1,393,000 116,000 Source: Central B a n k o f Sri Lanka Annual Report (multiple issues) 84 StatisticalAnnexes Table A11: Key EmploymentStatistics In Thousands Year Source Total ByIndushy Agriculture Manufacturing Mining Construction Services Consumer Finance and Socio- 4,641 2,415 582 56 234 1,360 1978/79Economic Survey Census o f 1981Population 4,119 1,864 417 38 125 1,675 Consumer Finance and Socio- 4,673 2,361 574 78 238 1,422 1981/82 Economic Survey Labour Force & Socio-Economic 5,132 2,53 1 648 67 227 1,659 1985186 Survey Consumer Finance and Socio- 5,27 1 2,5 15 706 100 300 1,650 1986/87Economic Survey Quarterly Labour 1990Force Survey 5,047 2,361 669 80 197 1,740 Quarterly Labour 1991Force Survey (a) 5,015 2,130 751 56 237 1,842 Quarterly Labour 1992Force Survey (a) 4,962 2,089 650 80 238 1,906 Quarterly Labour 1993Force Survey (a) 5,201 2,159 684 81 227 2,05 1 Quarterly Labour 1994Force Survey (a) 5,281 2,085 156 43 216 2,181 Quarterly Labour 1995 Force Survey (a) 5,357 1,967 I88 86 285 2,230 Quarterly Labour 1996Force Survey (a) 5,537 2,072 807 87 297 2,274 Quarterly Labour 1997Force Survey (a) 5,608 2,032 920 92 312 2,274 Quarterly Labour 1998 Force Survey (a) 6,005 2,436 856 74 296 2,342 Quarterly Labour 1999Force Survey (a) 6,083 2,205 902 76 322 2,578 Quarterly Labour 2000 Force Survey (a) 6,3 10 2,274 1,045 67 348 2,545 Quarterly Labour 2001 Force Survey (a) 6,236 2,033 1,057 110 324 2,711 Quarterly Labour _ _ 2002 Force Survey (a) 6,467 2,204 1,084 279 2,899 Quarterly Labour -- 2003 Force Survey (b) 6,946 2,412 1,117 374 3,043 Source: Central Bank of Sn Lanka Annual Report(multiple issues) (a) ExcludingNorth& East @) Excludesthe North, butincludestheEast. Average of first three quarters. 85 Statistical Annexes TableA12: SelectedUnemployment Statistics As %of Labour By Age Group (years) By Education(`000) Year Source Total Unemployment Forceby sex (`OOO) (`ow ("/.) Male Female 15-24 25-55 >55 rextension) 24.2 for foreign invest~nent~ settmngupofinttusrr5es intax free exportprocessing arid zones. 1979 Exemptionofpuhlii.sector enployaes frctin inconxetax andreductionincxport 19.7 taxes. 1986-7 Abolition ofEstate Duty and Gif2Tax. ~ r ~ ~ l e ~ n ~oftliear te~c~ ~n ~ n ~ ~ e n17.4~- 17.8 ~ n t ~ i o of &e second Yresidentral TariffC'oiixnission of 1985, 1980-91 Expaiisivnof HOI incentives and separate ('ustoins arrangmienlsio 1301 18.9 18.3 inciustricsto ppottwtethe appunrl industry inbackwardareas. 19% Abolition ofWealth Tax aind ExportsDuty on tea, rubber and coconut. 18.0 ~ ~ n n ~ i ~of~theephasedreductionof Stamp Dutyon Letters of Ckedit, ~ e n t TurnoverTax onbanking andfinancial iiistmnentsandgovenmlanf debt ~ n s ~ u n ~ e n ~ 1993-4 Irnpl,lenmtationot'tlte rrooni~rte~ufittions clt'tbe third I'rrsidential 'I'aiifl' 17 2 C'onimssionof 1992. IKr tariiTsmicliire which had rates ringingfrom I O 5(#) o/o and niore lhlin 19 taritl'bartds was rcdiicedto 13 bands in 1990 and 4 bands in 1991. 1995 Iamease ofthe tax free threshoId fbr income taxation andthe introduction o f a 17.8 threehand tariWstructure with IO, 20 and JQ%O rate bands. 1996 t -- IDA IMF G Short-term -- Netflows 72 115 55 212 Interestpayments 4 15 13 14 Nettransfers 67 100 42 I98 DevelopmentEconomics 6/28/04 110