Report No. 30188-LA Lao PDR Country Economic Memorandum Realizing the Development Potential of Lao PDR (In Two Volumes) Volume II: Main Report December 23, 2004 Poverty Reduction and Economic Management Unit East Asia and Pacific Region Document of the World Bank MAF Ministryo fAgriculture andForestry MA1 Mean Annual Increment MDGs Millennium Development Goals MEPAs MiningExplorationandProductionAgreements MFA Multi-Fiber Agreement MFN Most FavoredNation MICS Multiple Indicator Cluster Survey MIH MinistryofIndustryandHandicrafts MOU Memoranda o f Understanding MW Megawatt NAFES National Agriculture and Forestry Extension Service NAFRI National Agriculture and Forestry ResearchInstitute NDF National Development Framework NGO N o n Government Organization NPEP National Poverty Eradication Programme NPL Non-Performing Loan NSC National Statistics Center N T 2 Nam Theun 2 NTFPs Non-Timber Forest Products OLS Ordinary Least Squares PAFO Provincial Agriculture and Forestry Office PCC Per-Capita Consumption PER Public Expenditure Review POP Population POV Poverty Headcount PPA Public Performance Auditing PPP Purchasing Power Parity PRGF Poverty Reduction and Growth Facility QR Quantitative Restriction SME Small and MediumEnterprise SOEs State-Owned Enterprises STEA Science, Technology and Environment Agency TEL Temporary Exclusion List THPC Theun-Hinboun Power Company UNIDO UnitedNations IndustrialDevelopment Organization VAT Value Added Tax VFAs Village Forestry Associations VFM Village Forestry Management WDI World Development Indicators Vice President: Jemal-ud-din Kassum Country Director: Ian C. Porter Sector Director: Homi Kharas Lead Economist Kazi M.Matin Task Team Leader: Sethaput Suthiwart-Narueput (EASPR) Task Team: Alessandro Magnoli, Amanda Carlier, Kaspar Richter and Tanatat Puttanasuwan (EASPR); Ana Revenga (EASHD); Craig Andrews (COCPD); Morten Larsen (EACVF); Stephen Mink (EASRD); Barry Trembath (EASEG); Graham Scott (consultant) ... 111 Table of Contents Chapter 1. Challenge facing Lao PDR 1 1 1Backgroundandcontext . Development ......................................................................................... ............................................................ 1.2 Development vision................................................................................................... 1 9 1. 3 Organizing framework............................................................................................. . . 15 Chapter 2 - Sources of growth for Lao PDR ......................................................................... 17 2. 1Sectoral analyses ..................................................................................................... 17 2. 2 Sources o f Growth................................................................................................... 43 Chapter 3 - Improvingpoverty and social outcomes ............................................................ 48 3. 1 Demand-side: Growth. poverty. and social outcomes............................................. 48 3 2 Supply side: Public interventions andpoverty and social outcomes...................... . 53 Chapter 4 - Realizing the development potential of Lao PDR ............................................ 66 4. 1Elements o f a development strategy reform roadmap for Lao PDR....................... 66 4.2 Improvingthe enabling environment ...................................................................... 67 4. 3 Improvingpublic sector governance ....................................................................... 70 4. 4 Priority actions......................................................................................................... . . 77 4. 5 Concluding remarks................................................................................................. 80 Boxes: Box 1: Lao PDR Country Policy and Institutional Assessment for 2003 .................................... 8 Box 2: NGPES . 9 Box 3: Constraints on enterprise development ........................................................................... Some outstanding issues .................................................................................. 26 Box 4: The value o f keeping a low profile ................................................................................. 28 29 Box 6: Landlocked or landlinked? Regional linkages and Lao PDR......................................... Box 5: Off-farm and non-farm activities inLao PDR................................................................ 31 Box 7: Economically productive uses o f electricity................................................................... 38 Box 8: A simple macroeconomic model for Lao PDR............................................................... 44 Box 9: Off-farm opportunities inLao PDR: The case o f sericulture ......................................... 60 70 Box 11: Service delivery inLao PDR: Lessons from the 2004 WDR ....................................... Box 10: How to avoid the "natural resource curse" inLao PDR............................................... 74 Box 12: Highlevel outcomes and milestones for public sector and financial management in Lao PDR....................................................................................................................... 75 Box 13: Basic steps required inintergovernmental reform inLao PDR.................................... 78 Box 14: Improving transparency: priority disclosure inLao PDR............................................. 80 Box 15: Three remaining gaps inanalytical work on Lao PDR................................................. 81 iv Figure 1:Real GDP Lao Growth Rate . Average ............................................................ Period 3 Figure 2: Govemance indicators for Lao PDR............................................................................. 7 Figure 3: Progress TowardsMDGs. selected indicators ............................................................ 10 Figure 4: Growth andpoverty reduction year 1990-2000 .......................................................... 11 Figure5: GDPper capita andinfantmortality ........................................................................... 12 Figure 6: Increasingpublic spending alone is not enough ......................................................... 13 Figure 8:Voice andaccountability............................................................................................. Figure7: Changes inspending andoutcomes ............................................................................ . . 14 15 Figure 9: Framework for growth, poverty reduction, and social outcomes ............................... 15 Figure 10: Agricultural value addedper worker (constant 1995 USD) ..................................... 21 Figure 11:Labor costs andvalue addedper employee inmanufacturing(1999) ...................... 25 28 Figure 13: Market margins are large .......................................................................................... Figure 12: Starting a business inLao PDRv. elsewhere ........................................................... 29 Figure 14: GDP flows from naturalresource sectors ................................................................. 46 Figure 15: Estimated Impact of additional 1% GDP growthon poverty headcountratio in 2006 by source of growth .......................................................................................... 50 Figure 16: Simulated Impact of Growthon Social Outcomes for Bottom Quintile................... 52 53 Figure 18: Naturalresource andtotal revenues -Base case plus............................................... Figure 17: Natural ressources revenue -Base case plus............................................................ 55 Figure 19: Recurrent expenditure inprimary educationper child (age group 5-14) andrecurrent expenditure inhealthper capita-Base case plus .................................................... 57 Figure 20: Impact onpoverty from selected scenarios (percent change inpoverty) ..................58 Figure 21: Marginal Impact on probabilityo f completing school ............................................. 62 Figure23: Summary overview o f elements ofreform roadmap ................................................ Figure 22: Increase inprobability o f seeking treatment relative to bottom quintile .................. 64 68 Figure 24: Accountability framework inservice delivery.......................................................... 74 Figure 25: Priority actions for Lao PDR.................................................................................... 79 Tables: Table 1: Lao PDR: General and MillenniumDevelopment Goal Indicators ............................... 1 Table 2: Povertyby District Category.......................................................................................... 4 Table 3: Elements of2020 Development Vision ....................................................................... 10 Table 4: Lao PDR Real GDP by Sector, PeriodAverages ......................................................... 19 Table 5: Lao PDR agriculturalplanted area andyields.............................................................. Table 6: Summary sectoral characteristics ................................................................................. 20 43 Table 7: Average contribution to GDP growth by sector (percent) ........................................... 45 Table 9: Additional Scenarios .................................................................................................... Table 8: Poverty, Growth and Inequality Scenarios................................................................... 49 50 Table 11: Simulations o f % Changes inConsumption and Poverty .......................................... Table 10: Estimatedrecurrent expenditures inprimary educationper child.............................. 56 Table 12: Summary o f key determinants o fpoverty and social outcomes inLao PDR ............61 65 V ANNEXTABLES: .......................... 83 Annex Table 2: OLS Regressions on LogPer CapitaConsumption.......................................... Annex Table 1:Detailed reformroadmaps ...................................................... 91 Annex Table 3: Core Social Indicators byProvince .................................................................. 92 Annex Table 4: Marginal effects on the probability o f starting/attending school for ages 6 to 14............................................................................................................... 93 Annex Table 5: Marginal effects on the probability of startingattending school for ages Annex Table 6: Determinantso f Household Medical Expenditures per Capita, LECS 97/98 ..95 6 to 14............................................................................................................... 94 Annex Table 7: Probability of Seeking Treatment by Quintile andEthnicity ........................... 96 Annex Table 8: Impact on poverty and Inequality o fAdditional 1%GDP Growthby Sector.. 97 vi ACKNOWLEDGEMENTS This report was prepared by a team consisting o f Sethaput Suthiwart-Narueput (task team leader); Ana Revenga (social outcomes); Kaspar Richter boverty); Graham Scott (consultant; public sector and public financial management); Craig Andrews (mining); Stephen Mink (agriculture); Amanda Carlier and Tanatat Puttanasuwan (manufacturing and services); Barr Trembath and Morten Larsen (hydropower); and Alessandro Magnoli (macroeconomics) ry Many thanks also go to Linda Schneider, Eric Sidgwick, and Hans Luther for sharing their insights and perspective. Pera Nantachai and Wallada Atsavasirilert provided excellent research assistance. Angkanee Luangpenthong, Poonyanuch Chockanapitaksa and Ruangrong Thongampai processed the report. Somneuk Davading participated in discussions with the Government and contributed to the translation o f the report. The report was prepared with the cooperation o f a Lao PDRcounterpart team comprising o f the National Economic Research Institute (NERI), the Committee for Planning and Cooperation (CPC); the Ministry o f Industry and Handicrafts, Lao National Chamber o f Commerce and Industry,the Ministry o fAgriculture andForestry, the Lao Tourism Authority, andthe Ministry o f Finance. The team would also like to thank the numerous other counterparts from other Government agencies who met with the Bank team members on various visits during the preparation o f the report. The revised report was presented to the CPC and the MOF in May 2004, and presentations of findings were also made to senior officials of MOF, CPC, MOC, MAFE, BOL, MCTPC, and MIHas well as to the macro/PSD donors working group inJune 2004. A first draft o f the report was discussed with the Lao PDR counterpart team in October 2003 and a revised draft together with a stand-alone Summary was sent to the Government inMarch 2004. The findings o f this revised report were presented to MOF and CPC inMay 2004, to the senior officials o f MOF, CPC, MOC, MAFF, BOL, MCTPC and MM at the FMAC- completion retreat, as well as the macro-psd working group o f donors in July 2004. Both the text o f the summary andthe mainreport were discussed with a high-level inter-ministerial team led by Vice President CPI on two different occasions, and this final draft was prepared incorporatingthe suggestions and comments o f that team. The report benefited tremendously from a Quality Enhancement Review by two external reviewers, Christopher Hermans and William McCleary, and organized by Vera Wilhelm. The team gratefully acknowledges the comments and suggestions provided by peer reviewers Jeffrey Hammer, Stefan Koeberle, and Mark Sundberg. Helpful comments were also provided by Ian C. Porter, Homi Kharas, Kazi M. Matin, Enrique Crousillat, Adrian Fozzard, Kirk Hamilton; and several others. Selected background papers are available at www. worldbank.or.th. vii Chapter 1-Development Challenge FacingLao PDR Lao PDR faces an enormous development challenge. IfLao PDR is to make signijicant progress towards realizing its development vision, concerted efforts will be needed to improve growth, poverty, and social outcomes. 1.1Background and context 1. With high poverty, poor social indicators, and weak governance, the development challenge facing Lao PDR is enormous. Lao PDR's economic and social indicators fall below the average for low-income countries and are among the poorest inthe region (see Table 1). With GDP per capita o f around 310 USD, Lao PDR is one o f the poorest countries inthe world.2 Two-thirds o f the households have no electricity, and every other householdi s without safe water access. Life expectancy i s low, child malnutrition is high, child and maternal mortality rates are three-to-four times those o f other East Asian countries and the percentage of children completing primary schooling i s low. Compared to the averages for the region and for low income countries, governance inLao PDRremains weak. Table 1: Lao PDR: Generaland MillenniumDevelopmentGoalIndicators Lao PDR MDGs Averages-MRY 1990 zoo0 2015 E.Asia SSA SAsia Low income GNP per capita 210 310* 1,000 500 440 410 Adult literacyrate (%) 36 49 85 60 53 40 Total Fertility Rate 6.3 5.4 2.1 5.4 3.4 3.1 Life expectancy 50 59 69 50 62 60 % Populationbelownationalpovertyline 45 39** Povertyand Malnutrition PopulationbelowPPP 1$a day (%) 26 15 14 47 37 % Childrenunderweight (under 5 years old) 44 40 22 22 33 51 Educationand Gender Indicators Net enrollment rate (ages 6-14)** 61 72 100 99 77 77 76 % Cohort reachinggrade 5** 53 50 100 93 60 Youth literacyrate (ages 15-24) 55 81 100 93 68 85 Ratioof girls to boys inprimary and secondary ed 74 77 100 93 85 71 Ratio of literatefemales to males 52 69 100 93 89 77 Health Indicators Infant Mortality Rate (per 1,000 live births) 108 82 45 59 92 75 68 Under 5 mortality rate (per 1,000) 170 106 55 47 151 89 107 ; Matemalmortality ratio (per 10,000 live births) 650 530 163 Prevalenceof HIV/AIDS (%) 0.1 % Population access to safe water 52 80 001. Updatedusing World Bank, 2003 and UNDP, Source: Lao PDR HD Strategy and Work Program iscussionI te. World nk mimeo 2001 Note:PPP-basedGNP per capita is $1660. 2002;* 1997198;*** AdministrativeData * Basedon WDIdata. LaoPDRNSC data cited inNGPES placesGDP per capitaat 331USD. 1 Lao PDR: Background and recent history Lao PDR is a small, landlocked country. It has a population o f 5.5 million, about 1/15 and 1/11 o f the population o fneighbouring Vietnam and Thailand, and i s experiencing rapid annual population growth o f about 2.3%. I t has a land area o f 236,800 sq. km. Urbanization i s low at about 20%, and agriculture accounts for over half o f GDP. Gross National Income (GNI) per capita stands at around 310 USD. Reliance on external support remains high. Donor fundedprograms accounted for nearly 40% o f total public expenditures and over 60% o fthe capital budget. Lao PDR was one of the earlier countries to attempt the move to a market transformation. In 1975, Lao PDR was proclaimed, and the Supreme People's Assembly was tasked with drafting a new Constitution. A policy o f "accelerated socialization" was pursued, including extensive agricultural collectivization. However, by 1982, the use of "market forces" in the development o f Lao PDR is promoted at the 3rd Congress of the Lao People's Revolutionary Party (LPRP). Many o f the priorities highlighted by the 3rd Congress relate to investment climate, governance, and human resource development issues which still remain relevant today. In 1986, the Government officially abandoned a central planning system and introduced the New Economic Mechanism (NEM). In 1991, the collapse of the Soviet Union and the loss of assistance from that country, created a needfor reforms and a re- orientation towards Asian and Western markets to promote growth. Multilateral and bilateral donors provided financing to support ongoing reforms. Following this first opening, however, reforms proceeded sporadically. In 1995, at the 6th Congress, three major SOEs (DAFI, BPKP, ADF) were to play a large role inthe development o f each region(Central, North, and South) o f Lao PDR. During the Asian crisis, ambitious irrigation investments were implemented without appropriate financing, resulting in triple digit inflation in 1998. Following a strong stabilization program by the Government, the IMF resumed its program in Lao PDR in 2001 and the World Bank provided structural adjustment support in2002. The new Constitution was approved in 1991. Under Article 3 o f the Constitution, the rights o f the people are "exercised and ensured through the functioning o f the political system, with the Lao People's Revolutionary Party (LPRP) as its leading nucleus." The LPRP i s headedby the Party Central Committee, andparty committees ineach province and district support the local authorities inthe implementation of the development plan at each level as definedinthe Constitution. The Head of State i s the President o f Lao PDR, elected by the National Assembly or a five year term and responsible for appointing individuals to key posts, including the Prime Minister, Deputy PM, and provincial governors. The National Assembly i s the representative legislative body o f the country, composed o f 109 deputies and elected every five years. The Judiciary i s composed o f the Supreme Court at the central level and provincial and district courts at the local level. The State Audit Authority was set up in 1998, but i s under the Prime Minister's Office, and not under the aegis o fparliament as inmany other countries. There are currently 13 ministries, 3 ministry-equivalent organizations, and 164 departments and department-level organizations. Provincial governors are responsible for a wide range o f government services. With the NEMin 1986, authority was devolved and local offices o f central ministries placed under the authority o f provincial governors and district heads. However, as a result, provinces became virtually autonomous, with significant fiscal impact for the central government. With the promulgation o f the Constitution in 1991, fiscal, administrative and legal powers were to return to the central government. In2001, P M Decree 01 had the objective o f "making the province the strategic unit, the district the planning unit, and the village the implementing unit." However, central- local relations remain difficult. Sources: Background note by Schneider (2003); Lao PDR Background Paper on Governance (2003) 2 Growth 2. L a o PDR has recorded reasonable economic growth. Real GDP growth rates averaged 6.1% duringthe first halfo f the 1990s and reached 6.6% inthe latter half o fthe 1990s (see Figure 1). Manufacturing and services were the main growth drivers during the early 199Os, while agriculture was the main driver during the latter half of the decade. Natural resources-including forestry, electricity (hydropower), and mining-also made a significant contribution to growth during the early 1990s, but the growth contribution o f the sector has dropped off as a result o f a decline inforestry. 3. M o r e recently, however, growth dropped off to less than 6%. An increase in the growth contribution o f manufacturing-notably in food and beverages-was not sufficient to offset the drop inthe growth contribution o f agriculture-especially in crops-from over 3 to less than 2 percentage points o f GDP. Exports, which grew very rapidly during the first half o f the 1990s, considerably slowed down duringthe secondhalfo f the 1990sas well as more recently.' Sustaining real GDP growth rates o f 6-7% going forward will pose a major challenge and require reforms across all sectors. Figure 1:Real GDP Lao Growth Rate Period Average - 7.0% 1 6.0% 5.0% 4.0% 3.0% 2.0% 1.O% 0.0% 7 Lao PDRexports are highly concentrated ina few product categories: garments (34% o f 2002 total exports); electricity (33%); wood andwood products(21%); andcoffee (6%). Collectively,these four categories accountedfor well over 90% of total exports. Inthe first half of 2003, garment exportsdeclinedby 19% on aYOY basis, while coffeeexports increasedby 38%. Exportsare largely to ASEAN (especiallyThailandandVietnam) andthe EU. 3 Poverty 4. As a result of this economic growth, Lao PDR has made impressive progress in reducing poverty. The poverty headcount ratio-the percentage o f the population whose per capita consumption is less than the national poverty line-fell from 46% in 1992/93 to 39% in 1997/98, representing 1.9 millionpoor. Based onpreliminary LECS I11data as discussed under the NGPES, the poverty headcount ratio fell to 32%, representing 1.67 millionpoor in2002/03. 5. Despite this recent progress, Lao PDR still faces an enormous challenges in respect of poverty. There are huge differences in poverty across regions. In the 47 (25) districts identified as poorest (poor) under the NGPES, the poverty headcount i s as highas 65% (42%). However, it should be noted that while the incidence of poverty i s higher in these districts, the majority o fthe poor are still locatedinthe non-poor districts. Table 2: Poverty by District Category District-type: Poorest 47 Poor 25 Non-Poor 70 Mean SE Mean SE Mean SE Headcount 0.651 0.023 0.419 0.041 0.301 0.016 Gap 0.213 0.012 0.105 0.014 0.067 0.005 Severity 0,091 0.007 0.040 0.007 0.022 0.002 Memorandum items: Poor population 0.371 0.031 0.108 0.019 0.521 0.033 Population 0.223 0.021 0.100 0.016 0.677 0.024 Source: Backgroundpaper by Richter (2003) Socialindicators 6. Lao PDR's social indicators are below the average for low-income countries (see Table 1). While infant and maternal mortality, life expectancy, and literacy have improved since 1990, completion rates for primary school have not improved, and net enrollment rates have increased very slowly. The aggregate indicators mask very significant differences in outcomes by gender, geography, and wealth.' 7. Existing weaknesses in the educational and health systems will continue to hamper improvement in the social indicators. The educational system i s characterized by low enrollment and completion rates andlow quality and efficiency. Multiple Indicator Cluster Survey(MICS 11) data indicate that while 82% of males over 15 are literate, only 59% of females are. Primary completionrates vary from a low of only 27% inPhongsalyto a high of 85% inVientiane municipality. Over 90% of house holds in Champasak or Xaynaboury have access to some form of health care, compared to only 33% inLuang Namtha. (See Annex Table 3). 4 e According to the MICS 11, nearly a third o f children between the ages of 6 and 14 were not enrolled in school. Enrollment was only 68%, and vanes significantly by gender, income, andprovince.6 While starting school on time is instrumental inhelping students-especially girls-to complete their studies, most children are starting school late, especially among poorer households. In the lowest quintile, the percentage o f children who have started school by age 6 i s only 19%, compared to 42% for the top quintile. e Only about one-half o f students who start school in Lao PDR reach Grade 5. Completion rates are highly influenced by drop out and repetition rates, and both are high in Lao PDR. About 10 percent o f students in grades 1-5 drop out every year, and between a fifth and a quarter o f studentsrepeat grades. Low completion rates arepartly a result of the low quality of sewices offered. Inadequate textbooks and learning materials, poor school maintenance, and in adequate teacher preparation all contribute to low instruction quality. Approximately only 80% o f secondary teachers have completed lower secondary school (grade 8) plus the three years o f required teacher training. Teacher wages remain very low and subject to delayed payments.' e Unequal access remains a problem. While 85% o f all villages report having a primary school, only 43% have a complete primary school offering all five grades. Between 30-40% o f schools offer only grades 1and 2, while villages inremote areas lack schools altogether. Accessibility for the disabled is also o f concern.* e Public expenditures are inadequate and inappropriately allocated. Overall public expenditure on education remains low at 8.8% o f the budget and 1.9% o f GDP in2001. In real terms, expenditures is only half the level it was in 1996. Of particular concern i s the drop in recurrent expenditure from over 7% o f the total public expenditure budget in 1996/97 to less than 4% in 2001/2001, resulting in new schools with insufficient teachers andmaintenance. 8. The health system is characterized by high mortality and morbidity, low health care utilization, poor quality o f services, and inefficient public expenditures. Improvements inhealth status seem to havebeen, inlarge part, overestimated.'O Actual enrollment rates using survey data are lower than those inadministrative data due to absenteeism and drop-outs during the Year. Enrollment was only 64% for girls compared to 71% for boys. For the wealthiest households inthe highest quintile o f assets, the enrollment rises to 78%. By contrast, enrollment falls to 47% for households in the lowest asset quintile (41% for girls and 53% for boys). Enrollment is over 90% inVientiane municipality, but are only between 45-50% inthe Northem 'While provinces of Oudomxay, Phonsaly, Luangnamtha, and Bokeo, with significant differences between boys and girls. the GOL granted a 40% pay increase to teachers inOctober 2000, salaries had fallen by a third inthe three years The UNestimates that only 5% o f disabled children attend school indeveloping countries, while about 10% have a disability. The investment: recurrent expenditure ratio is unsustainably high. With nearly 70% o f investment financed by donors, counterpart financing requirements divert resources away from the recurrent budget. There are large inconsistencies between Government data and those collected by multilateral agencies which throw some doubt as to the magnitude o f the actual improvement inhealth status. For example, the Lao Reproductive Health Survey 2000 estimated the under-5 mortality rate at 107 per 1,000 live births; for the same year, the World Health Report suggests a rate o f 150per 1,000 live births. Under the first estimate, there has been a markedreduction inunder-5 mortality since 1990, when it stood at 170per 1,000 births. Under the second, there has been very little improvement over the decade. 5 Mortality and morbidity rates remain high and are well above the averagefor the region. Factors contributing to high mortality include high fertility, in complete immunization, malnutrition, low education and literacy, and poor quality health services." Highmorbidity is prevalent among the rural population, with malaria-at over one million reported cases-the most common cause of death. Utilization of health care services is low, reflecting both demand and supply sidefactors as poverty and low coverage and quality. Poverty and a highreliance on user fees effectively prevents the poor from accessing health care. Since Decree No. 52 ended the policy o f free public health care, cost recovery has mushroomed and now accounts for about 60% o f total operating revenues o f central and provincial hospitals. Geographical coverage i s limited. about 38% o f m a l villages are located at least 4 km from the nearest health center. Inthe North, the average distance to the nearest hospital is around 26 km. Service quality is poor. The quality o f curative care remains low. District level hospitals are generally very small (5-30 beds), can only provide basic treatment, and lack supplies and trained personnel. There are fewer than 100 doctors per 100,000 people, one o f the lowest ratios in East Asia, and only 0.9 trained personnel per bed compared to a regional average o f 1.5. As in education, public expenditures in health are low and inefjciently allocated. The budget allocation to health i s less than 1% o f GDP, which largely goes toward capital investment in facilities. Ofparticular concem is the sharp contraction inrecurrent spending from around 60% to less than 40% o f total spending. While a focus on preventive measures such as health education, immunization, prenatal care, family planning could have had a greater impact at lower cost, much o f the resources have instead been directed at curative care. Governance 9. Widespread weaknesses in governance will constrain the ability of Lao PDR to significantly improve its development outcomes. There i s international evidence that good governance is critical to improving development outcomes. Based on a cross-country database o f governance indicators, Lao PDR rates poorly against the average for low-income countries.I2 Ineach area of governance, the upper bar depicts the percentage o f countries worldwide that rate below Lao PDR, while the lower bar depicts the average percentile rank for low-income countries. Lao PDR performs well on political stability, but rather poorly on voice and accountability, and on control o f corruption. Malaria, diarrhea, and acute respiratory infections are the main causes o f infantmortality. Child malnutrition remains widespread with 40% and 13% classify as underweight and severely underweight. l2 Based on Kaufmann, Kraay, and Mastruzzi (2003), "Governance Matters 111: Governance Indicators for 1996-2002." Indicators reflect the statistical compilation o f responses on the quality of governance based on a number o f survey 6 Figure 2: Governance indicators for L a o PDR LROS (2002) Governnent Effectiveness Regulatory Quality Rule of Lau Control of Corruption e 25 50 75 Conparison with incone category average {Lou Incone) {lower bar) Country's Percentile Rank (0-180) Source: D. KauCrmnn, R. Kraar and I(. Ilastruzzir 2003: Governance Hatters Ill:Governance Indicators Cor 1996-2002 l h t t p : / ~ , - r l d b a n k . ~ r g ~ i / ~ ~ u e r n a n c e / p u b s / g ~ v m 3 t t e r + 3 . h t m l l 10. These findings are consistent with the results of the World Bank's 2003 Country Policy and Institutional Assessment (CPIA) for Lao PDR, which identified significant weaknesses in several critical areas, including in the transparency and accountability o f the public sector and inthe quality of budgetary and financial management. (See Box 1). Without significant improvements in the overall policy and institutional environment, Lao PDR will find it difficult to make significant progress towards realizingits development vision. 7 Box 1:Lao PDR CountryPolicyand InstitutionalAssessmentfor 2003 Each year the World Bank assesses the quality o f each borrower's policy performance with regard to fosterin) poverty reduction, sustainable growth, and the effective use o f development assistance. The assessment cover twenty items grouped into four categories, including economic management, structural policies, policies for socia inclusion and equity, andpublic sector management and institutions. Lao PDR ratings were low in the categories o f structural policies (including policies for financial stability an( environmental sustainability) and public sector management and institutions (including the quality o f publil financial management and transparency and accountability inthe public sector). Selected CPIA areas Assessment Quality of budgetary and financial Weaknesses inpublic financial management management 0 Expenditures are not controlled to budget limits, leading to a significant arrears problem Ineffective cash management. There are multiple bank accounts, some outside the control o f Treasury. Some provinces are short o f cash while others hoard central government revenue. The manner inwhich fiscal decentralization is occurring is undermining fiscal control Transparency and accountability in Need to improve transparency of decision making the public sector processes, including between the centre and the provinces and inthe implementation o f sectoral policies New accountability organizations - e.g., National Audit Office, State Inspection Authority - are under-resourced andneedto be strengthened for task at hand Financial sector depth and efficiency Small banking sector with a large share o f credit flows from state owned banks becoming non-performing loans Poor access to financial services for private enterprises, consumers, and the poor Financial stability Sector remains highly concentrated with two insolvent state banks accounting for 80% o f system Inadequate regulatory, accounting and disclosure standards with weak enforcement Weak capacity in commercial banks and central bank supervision Policies and institutions for Deforestation remains the most important environmental environmental sustainability problem While laws to protect and manage environmental resources have been passed, implementation i s weak due to a lack of enabling regulations, institutional capacity, expenditures, and community participation 8 1.2 Developmentvision 11. As part of its development vision, Lao PDR seeks to graduate permanently from least developed country status by 2020. l3 Lao PDR estimates that this would require per capita GDP between USD 1,200-1,500 and average annual GDP growth o f 7%. The National Development Framework (NDF) sets out the main longer-term development indicators and objectives that underpin the National Growth and Poverty Eradication Strategy (NGPES) recently prepared by the Government. Some o f the outstanding issues revolving around the NGPES are discussed inBox 2. 12. Lao PDR will also face the challenge of meeting its Millennium Development Goals (MDGs) by 2015. These include, for example, halving the proportion o f people living in poverty compared to 1990; achieving universal primary education; promoting gender equality; reducing child mortality by two-thirds compared to 1990; and ensuring environmental sustainability, including halving the proportion o fpeople without access to safe drinkingwater. Box 2:NGPES- Some outstandingissues The NGPES report covers a wide range o f areas and recognizes the importance o f creating an enabling environment for growth and development-encompassing macroeconomic stability, private sector development, regional integration, and natural resource development-as well as the need to improve public sector governance. The NGPES states that its operational framework comprises four main sectors (agricultureiforestry, education, health, and infrastructure) and several other supporting or potential growth sectors. Rural development and a poor district focus are also central elements o f the NGPES. Intheirjoint comments onthe drafi NGPES, IMFandWorldBank staffnotedwhile the NGPESprovides auseful integration o f the poverty reduction strategy with the Government's overall national objectives and socioeconomic programs, additional structural and sectoral reforms are needed as part o f the poverty reduction strategy. Key policy issues flagged included the need to link the development o f natural resources to poverty reduction, greater revenue effort, public sector reform, and district targeting. To combat poverty, the NGPES has a strategy of poor area development and foresees channeling priority investments to 47 districts classified as "poorest. Based on the findings o f a poverty monitoring system questionnaire fielded by the Government in 2001102, 72 (out o f 142) districts-covering about one third o f the population-were classified as "poor." O f these 72 districts, 47 districts were classified as "poorest," in which at least half o f the households are poor according to household poverty criteria. Priority investments will be channeled to these 47 districts over the period o f the plan, while the remaining 25 poor districts will be targeted during 2006-2010. While the target districts o f the NGPES are on average much poorer than the rest o f the country, the majority of the poor reside outside the targeted districts. In 199718, the "poorest" 47 districts had a poverty incidence of 65 percent, compared to only 42 percent in the 25 "poor" districts and 30 percent in the 70 "non-poor" districts. The pattern i s confirmed when we look at other indicators o f well-being such as education, assets, village infrastructure. However, only 37 percent o f the poor reside inthe poorest districts, and over half o f the poor reside populated, and account for only a fifthHouseholdspopulation. (See Table 3). in districts classified as "non-poor." in the poorest 47 districfz are also more rural, less densely o f the total The NGPES strategy o f targeting poor districts should therefore be accompanied by measures to ensure that the allocation mechanisms within the districts are also functioning well, Le., that the non-poor inthe poorest districts are not receiving a disproportionate share o f resources. Sources: NGPES; poverty backzround paper by Richter (2003) "6thPartyCongress l4It will be importantto update these numberswith the forthcoming LECS 111and further refine the poverty reporting methodologyassociated with the monitoringsystem. 9 Table 3: Elements of 2020 Development Vision Indicators 2000 2005 2010 2020 Total Population(million) 5.2 5.9 6.6 8.2 Populationgrowth rate ("h) 2.6 2.5 2.3 2.2 GDP growth rate (%) 5.8 7.0 7.0 7.0 -Agriculture (%) 4.9 4.5 4.2 3.8 -Industry("A) 8.5 10 10 8.5 -Service f%) , , 5 10 10 9 Life expectancy at birth (years) 59 63 67 70 Adult literacv over 15 years (%) 70 78 84 90 Infant mortality under 5 (per 1,000 live births) 82 62 40 20 Maternal mortality (per 100,000 live births) 530 350 250 130 Access to clean water (% o fpopulation) 52 57 100 100 13. Concerted efforts will be needed if Lao PDR is to make significant progress towards realizing its development vision. Achieving many o f the MDGs by 2015 or the goals o f the NDF by 2020 will demand extraordinary efforts to reduce poverty and improve the health and education status o f the population. While there has been improvement in some key social indicators since 1990-including under-5 mortality rates, life expectancy, and literacy, there are other indicators that have not improved. While Lao PDR will realize the MDG for poverty, on current trends it will not be able to achieve the MDGs for universal completion o f primary school, maternal mortality, and child malnutrition by 2015 (see Figure 3). Figure 3: Progress Towards MDGs, selected indicators PrimaryCompletion Rates I Under-5 Mortality Rate % 1207 -- - #I2007 -Actual - - Trend to reach 0 --Trend to 50 reach MDGs I 1990 1995 2000 2005 2010 2015 Maternal Mortality Child Malnutrition 4 2 50 -Actual 40 -Actual = m 2 30 - - Trend to reach 5 'f20 -- Trend to E: reach 0 t 10 I z o O " " " ' i 1990 1995 2000 2005 2010 2015 2020 10 Figure 4: Growth and poverty reductionyear 1990-2000 0 eo O ~ O l 1 , 1 59% e 0 20 Middle EaetBNorth Africa . - 3 0 0 -200 - I O 0 0.00 1 0 0 e ' 2 0 0 300 400 5.00 800 7 0 0 Source: World Development Indicators 14. What will be needed for Lao PDR to make significant progress? Several lessons from cross-country experience are particularly relevant for Lao PDR and are worth highlighting. 15. First, economic growth is essential to realizing such improved development outcomes. Countries which grew faster achieved more poverty reduction. As indicated in Figure 4, the more rapid growth in East Asia and Pacific countries resulted in a much sharper reduction in poverty than inother regions, with consequent improvements inother social indicators. 16. Second, having more resources is no guarantee o f improved economic growth. Economic growth i s not mainly the result o f capital accumulation or additional resources. Empirical research on growth demonstrates quite clearly that productivity growth and not capital accumulation explains much o f the differences in economic growth across countries and over time." Resources may be important for some countries at particular junctures. The research on aid effectiveness, for example, demonstrates that additional aid resources can contribute to good development outcomes, but that depends critically upon an appropriate policy and institutional environment.16 But resources are not a "magic bullet." Indeed, recent empirical research find that some natural resources-oil and minerals in particular-actually have had a negative impact on growth through their damaging effect on the incentives to improve institutional quality." I5See, for example, Easterly and Levine (2001), "It's not factor accumulation: stylized facts and growth models." l6World Bank (1998), Assessing Aid: What Works, What Doesn't and Why " Recent econometric researchby the IMFfinds that natural resources have a strong negative impact on growth by impairing institutional quality. However, once the effect o f institutions i s controlled for, there i s either little or even a positive effect o f natural resources on growth. Readily appropriable or "point source" natural resources like fuel and minerals also have a stronger negative impact than do food or agricultural resources. (IMF Working Paper, Sala-i-Martin and Subramanian 2003). 11 17. The world i s strewn with examples o f countries with huge natural resources that have failed to grow. Nigeria provides a particularly stark cautionary tale. In 1965, per capita oil revenues were USD 33, and per capita GDP in Nigeria was about USD 245. By 2000, per capita oil revenues had increased to USD 325, but per capita GDP remained at the same level as in 1965. Since 1965, after deducting payments to foreign oil companies, Nigeria obtained cumulative revenues from oil o f around USD 350 billion measured at 1995 prices. Nigeria's hugeoil revenues havenot improvedthe standard o flivinginNigeria at all.l* 18. Third, while growth is essential, growth alone is not sufjcient for rapid improvement in social outcomes. Countries differ in terms o f how effectively their economic growth has translated into improved development outcomes. Vietnam, for example, managed to achieve many social indicators comparable to Thailand at a much lower level o f per capita GDP. For example, as indicated in Figure 5, Vietnam's infant mortality rate declined steadily and was lower than that o f Thailand in 2000 although its GDP per capita was only about one-eighth o f 'Thailand's. That growth alone i s not enough to improve social outcomes sufficiently i s also true at the global level. At current trends, for example, while the MDG for poverty i s likely to be achieved in all regions except for Africa, the under-5 mortality rate is unlikely to be achieved inany region o f the world by 2015.I9 Figure 5: GDPper capita and infant mortality 3,500 60 3,000 50 5i 2mm v) 2,500 '40 7 VI CI 4 5 2,000 .-b Ys .30 =$ 2P 0 0 1,500 0 7 8tn a z - 20 p 1,000 P (3 - 10 500 0 - 0 1985 ---- Thailand,MortalityRate Infant(LHS) --Vietnam, 1987 1990 1992 1995 1997 1999 i 2000 * ThailandGDP per capita(RHS) * - VietnamGDP percapita(RHS) MortalityRate Infant(LHS) I* Figures are from Sala-i-MartinandSubramanian(2003). l9World Bank, World DevelopmentReport2004. 12 19. Fourth, increasing public spending alone-without complementay improvements in service delivey-is also not enough. Increasing public spending without additional improvements in the efficiency and targeting o f that spending i s not likely to translate into better outcomes. Cross-country data, for example, typically indicate a weak relationship between public spending and health and education outcomes once the effect o f per capita income i s controlled for.2o 20. Evaluating changes in public spending over time yields a similar message. A comparison o f Thailand and Venezuela, for example, indicates that while Thailand increased public spending on primary schooling much more than did Venezuela, there was little difference incompletion rates for primary educationbetweenthe two. (See Figure 7). Figure 6: Increasingpublicspending alone is not enough Under-five mortalitv rate*. 2000 -150 -100 -50 0 50 100 150 Public spending per capita on * health, 1990s average" Public spending and child mortality are given as the percent deviation from rate predicted by GDP per capita. Source: World Bank, World Development Report 2004. Spending and GDPfrom World Development Indicators database; Under-5 mortalivfrom UNICEF. Because richer countries typically have better outcomes and spend more on health and education, a simple scatterplot o f spending and outcomes will show a line with a significant slope. Once the effect o f per capita income i s controlled for, however, the effect is much less marked. 13 21. As discussed above, previous studies have noted that recurrent expenditures on health and education in Lao PDR are below what is needed. But additional spending must be accompanied by improvements in service delivery if outcomes are to improve sufficiently, especially for the poor. The 2004 World Development Report identifies four main reasons why public spending on services have typically failed to benefit the poor: (i) spending on public services tends to benefit the rich more than the poor; (ii) fails to reach the frontline money service providers such as schools and clinics; (iii) service quality i s particularly low for poor people as incentives for effective delivery remain weak; and (iv) there i s insufficient demand from the poor due to other factors. Without also resolving these problems, there are likely to be limitedbenefits from additional public spending. Figure 7: Changes inspending and outcomes Publicspending on primary Primary completion rate education per child ,Thailand 350 1 loo 1 300 - 90 - 80 - 70 - Venemela 60 - 50 4 50 1980s 1990 1990s 1999 1980s 1990 1990s 1999 Source: Spendingdata from WDI database; primary completion from Brunset. al. (2003) 22. Fifth, good governance is critical to sustained growth. Cross-country data indicate strong positive relationship between various indicators o f governance-such as voice and accountability, control o f corruption, and rule o f law-and per capita incomes (See Figure 8). Interestingly, recent research indicates that the this relationship i s due to the strong positive effect o f governance on income, and not the other way around. 21 The research suggests that a strategy o f waiting for governance to automatically improve as the country gets richer is not likely to succeed, and that there is a need for systematic interventions to improve govemance where it is weak. 21Kaufmannand Kraay(2003), "Governanceand Growth: Causality which way? Evidence for the World, inBrief." 14 Figure 8: Voice and accountability 1.3 Organizing framework 23. Lao PDR's development vision is appropriately multi-faceted and involves not just growth but improved welfare. This report therefore looks not only at the sources o f growth, but of improved development outcomes more broadly. Figure 9 provides the analytical framework used to explore the linkages between growth, poverty reduction, and social outcomes. Growth leads to poverty reduction and improved household incomes @. Increased householdincomes ~ lead to greater demand for education and health services, resulting in improved social outcomes such as improved enrollment and mortality rates ~. At the same time, growth also results in greater public revenues and expenditures outputs as more teachers, schools, and hospitals 1,which 8. These expenditures translate into such can contribute to improved outcomes such as reduced poverty and again to improved social outcomes 8. These improved social ~ outcomes feed back to higher growth and lower poverty Figure 9: Frameworkfor growth, poverty reduction, and socialoutcomes Growth 15 24. Figure 9 does identify the important necessary linkages needed for growth to translate into reduced poverty and improved social outcomes. Improvements to the broader policy and institutional setting are also required. Growth may result in increased resources and expenditures, but ifpublic financial management and service delivery are weak then these may not translate into much improvement indevelopment outcomes. The 2004 World Development Report provides a useful framework o f accountability relationships between three sets o f actors-poor people, policy makers, and providers-in thinking about whether and how public services can be made to work for the poor, which we will revisit in the context o f Lao PDR. (See Box 11). The framework in Figure 9 also highlights the importance o f demand-side as well as supply-side factors inimproving social outcomes. Public schools, for example, may be built, but if households do not send or keep their children in those schools-e.g., because the returns to education are perceivedto be low-educational outcomes will not improve. 25. Figure 9 also provides the basic, organizing framework for this report. Chapter 2 discusses the sources o f growth for Lao PDR 1 through various sectoral studies (section 2.1) and a simple macroeconomic model (section 2.2). Chapter 3 discusses the various links by which growth can get translated into improved poverty and social outcomes. Section 3.1 focuses on the top-half o f the diagram and the demand-side impact. The impact o f growth on household incomes and poverty 2 is explored through simulations using POVSTAT, a methodology developed by the World Bank (section 3.1.1). H o w higher household incomes translate into demand-side improvements insocial outcomes 3 is also estimated (section 3.1.2). 26. The bottom half o f the diagram i s explored in section 3.2, which focuses on supply-side interventions, including how growth translates into an indicative resource envelope for increased public expenditures 4 i s addressed (section 3.2.1). How efficiently these increased public expenditures get translated into outputs 5 such as improved infrastructure or more schools and hospitals i s not explicitly covered inthis report. " How ever, the impact that these outputs have on poverty reduction 6 and improved social outcomes 7 is discussed in the analysis o f the determinants o f poverty and social outcomes. As will be discussed, however, the relationship between public expenditures and better outcomes i s a complex one which requires major improvements inaccountability in several links along the service delivery chain. Improvements in social outcomes 8 feedback into reducing poverty and improving growth. The analysis o f poverty determinants, for example, provides clear empirical evidence o f the relationship between better education and lower poverty. While not explicitly addressed inthis report, there is also widespread international evidence o f the relationship between better social indicators such as improved human capital and long runproductivity and economic growth. 23 27. Based on the results from the preceding chapters, chapter 4 discusses elements o f a strategy for realizing Lao PDR'sdevelopment potential, as well as some o f the key steps needed to implement such a strategy, including the importance o f improving the enabling environment for the private sector and public sector governance, through better information and transparency. ''The ~~ Lao PDR expenditure tracking survey will address some o f these issues inthe context of the education sector. 23 One way to address this feedback link is to conduct growth accounting exercises which take into account the effect of improvements inthe quality o f human and physical capital on aggregate GDP growth. However, as is discussed later inthe report, data constraints preclude any meaningful growth accounting exercise inLao PDR at present. 16 Chapter 2 - Sources of Growth for Lao PDR While Lao PDR has had reasonable economic growth, sustaining and improving upon this performance will require signijkant reforms across all sectors. It will be important to maintain agricultural performance to reduce poverty and to develop manufacturing and services as a long-term growth enginefor Lao PDR while improving the growth contribution of natural resources. 28. This chapter discusses some o f the sources o f growth for the Lao PDR economy. Section 2.1 provides summary sectoral analyses covering key sectors such as agriculture; manufacturing and services; andnatural resources, including mining, forestry, and hydropower. Based on the preceding analysis, what can Lao PDRreasonably expect by way o f growth going forward? Section 2.2 provides some indicative figures from a simple macroeconomic model. Household Incomes and Poverly Social Growth Outcomes I /-----./ \ Revenuesand Expenditures 2. 1Sectoral analyses 29. Brief sectoral analyses o f agriculture, manufacturing and services, and natural resources-including forestry, mining, and hydropower-are provided. Each sectoral summary describes the current status and key issues faced by sector, including steps needed to increase the growth and development contribution o f the sector. The share o f GDP, growth rate, and contribution to overall GDP growth o f each sector are provided inTable 4. 2.2.1 Agriculture 30. While agriculture is the most critical sector for improving social and development outcomes in Lao PDR, the sector is largely characterized by subsistence ricefarming and low productivity. Growth will come from a transition to greater market orientation, including regional specialization characterized by rice intensi9cation in favorable zones and diversifcation away from rice in other areas. However, particular attention will need to be paid to managing risks associated with localfood security and resource degradation. 17 31. Agriculture i s the most critical sector for improving development outcomes in Lao PDR. O f the 798,000 households in Lao PDR in 1999, 668,000 were agricultural households. With the overwhelming majority o f the poor located in rural areas, progress in poverty reduction depends critically upon performance in the sector. As noted in Table 4, even excluding forestry, agriculture (crops, livestock and fisheries) accounts for nearly half o f GDP. The sector also contributed nearly half o fthe GDP growth recorded duringthe latter halfo fthe 1990s, although this has dropped to about a thirdmore recently. 32. The sector has performed quite well. Agricultural sector growth rates between 1991- 2000 have been among the highest in the region at 4.7 percent, compared to Vietnam (4.6%); Thailand (2.4%); and Cambodia (2.0%). Rice production and yields grew strongly during the 1990sby an impressive 53 and 43%, respectively. 33. A significant increaseinyields underliesthis performance. As indicated inTable 5, total planted area increased by less than 3% during the past decade. However, rice yields, especially for season andirrigated rice, increased significantly. There has also been a shift inplanted area away hom upland rice and maize towards irrigated rice and higher value products such as vegetables and If accurate, the above data, which i s from a NSC report based on MAFF data, indicates a significant and surprising drop in planted area from 1990 to 1995. 34. Despite its strong performance, the sector is still characterized by a high subsistenceorientation. Subsistence rice farming on small holdingsremains the predominant activity. 94% o f the agricultural households in Lao PDR produce mainly for their own consumption, and 92% grow rice.25 Only 35% o f households sold some o f their produce. Holdingsare small andthe distributionofland appears reasonably equitable. 24Between 1996-1999,the averageprice per kg for salad and longbeans at markets of five cities was 15% and30% higher, respectively, than the priceper kg for glutinousrice (data reportedby NSC). 25Lao Agricultural Census 1998-1999. 18 Table 4: Lao PDRReal GDP by Sector, PeriodAverages LAOPDRREALGDP PERIODAVERAGES - SHAREOF GDP GROWTHRATES CONIRIBLITIONTOGROWTH Sector 1991-1995 1996-2000 2001-2002 1991-1995 1996-2000 2001-2002 1991-1995 1996-2000 2001-2002 52.5% 47.5% 47.5% 2.5% 6.8% 3.7% 1.3% 3.2% 1.8% 29.5% 28.0% 30.1% 4.2% 10.0% 3.8% 4.3% 2.7% 1.2% 19.3% 17.8% 18.2% 0.0% 7.7% 3.6% -0.3% 1.3% 0.7% 1.2% 1.6% 2.4% 12.3% 23.0% 17.3% 0.1% 0.3% 0.4% 9.0% 8.6% 9.6% -1.2% 13.2% 1.Ph 4.1% 1.1% 0.1% 23.0% 19.5% 17.4% 7.1% 25% 3.5% 1.6% 0.5% 0.6% 11.6% 9.9% 8.7% 8.0% 2.2% 2.7% 0.9% 0.2% 0.2% 8.8% 7.3% 6.4% 6.3% 2.3% 2.4% 0.5% 0.2% 0.2% 1.7% 1.6% 1.6% 7.5% 3.2% 12.9% 0.1% 0.1% 0.2% 0.8% 0.8% 0.7% 4.5% 5.7% 5.5% 0.0% 0.0% 0.0% 16.3% 19.4% 20.9% 13.0% 8.4% 11.5% 1.9% 1.6% 2.3% 13.1% 16.4% 18.5% 14.3% 10.3% 12.5% 1.7% 1.6% 2.2% 10.1% 12.8% 14.6% 12.1% 12.8% 115% 1.O% 1.5% 1.6% 0.7% 0.6% 0.5% 12.7% -3.1% 16.4% 0.1% 0.0% 0.1% 0.7% 1.O% 1.1% 43.4% 3.7% 17.7% 0.2% 0.0% 0.2% 0.8% 0.6% 0.6% 43.3% 24% -1.3% 0.2% 0.0% 0.0% 0.1% 0.1% 0.0% 4.3% -11.3% -0.6% 0.0% 0.0% 0.0% 0.Ph 0.4% 0.6% 37.8% 7.1% 68.7% 0.1% 0.0% 0.3% 0.2% 0.3% 0.3% 6.9% 16.1% a.5% 0.0% 0.0% 0.1% 0.3% 0.7% 0.9% 73.2% 20.6% 9.6% 0.1% 0.1% 0.1% 3.1% 3.0% 2.3% 9.3% -1.3% 3.3% 0.3% 0.0% 0.1% 0.1% 0.1% 0.1% 0.6% 14.4% 18.5% 0.0% 0.0% 0.0% 24.6% 25.4% 25.1% 6.8% 6.6% 5.7% 1.7% 1.7% 1.4% 5.1% 5.7% 6.0% 6.8% 8.3% 8.5% 0.3% 0.5% 0.5% 4.8% 5.0% 5.1% 5.3% 7.5% 5.5% 0.2% 0.4% 0.3% 0.0% 0.1% 0.1% 7.9% 17.4% 3.1% 0.0% 0.0% 0.0% 0.3% 0.6% 0.8% 42.8% 20.5% 34.3% 0.1% 0.1% 0.2% 8.0% 9.2% 9.7% 10.5% 8.8% 8.2% 0.8% 0.8% 0.8% 1.1% 1.29h 0.8% 13.1% 4.9% 8.7% 0.1% -0.1% 0.1% 3.7% 3.2% 2.8% 3.7% 3.2% 2.5% 0.1% 0.1% 0.1% 4.3% 3.0% 2.8% -4.9% 3.9% 1.7% -0.2% 0.1% 0.0% 1.5% 0.9% 0.6% 16.3% -2.2% -2.7% 0.2% 0.0% 0.0% 0.9% 1.9% 2.1% 76.3% 15.4% 0.2% 0.3% 0.3% 0.0% 0.1% 0.2h 0.2% 31.6% 12.8% 3.6% 0.0% 0.0% 0.0% 6.6% 7.7% 6.5% 20.7% 1.1% 48% 1.2% 0.1% 0.3% 1.3% 2.0% 2.8% 6.3% 27.8% 2.8% 0.1% 0.5% 0.1% 5.1% 5.3% 3 2 h 40.7% -8.8% 6.6% 1.1% -0.5% 0.2h 0.2% 0.4% 0.5% 15.1% 27.7% 5.7% 0.0% 0.1% 0.0% 0.0% 0.2% 0.3% 67.0% 11.9% 0.1% 0.0% 0.1% 0.1% 0.0% 10.6% 9.9% -26.4% 0.0% 0.0% 0.0% 0.1% 0.2% 0.2% 13.1% 15.7% 7.5% 0.0% 0.0% 0.0% 100.0% 100.0% 100.0% 6.1% 6.6% 5.8% 6.1% 6.6% 5.8% 'Muding naturalresourcesedcfs. Agricultureexcludesforestry; industryexdudeselectricity,mining,andquafrying sourcet calcu/afedbasedon fiaur-esSUDD!%~bv LaoPDRNational SfafisticdCenter 19 Table 5: Lao PDRagricultural plantedarea andyields YOchange 1990 1995 2000 1990-2000 Rice 664 560 690 4.0% Seasonrice 392 367 500 27.4% Irrigatedrice 11 14 1I O 900.0% Uplandrice 260 179 80 -69.3% Starchy roots 20 14 8 -59.6% Maize 48 28 I O -79.2% Vegetablesandbeans 8 9 40 406.3% Rawmatls for manuf 58 58 72 23.7% Total 797 669 820 2.8% Rice Seasonrice 2.77 2.92 3.27 17.9% Irrigated rice 3.55 3.71 4.23 18.9% Uplandrice 1.46 1.65 1.63 11.3% Starchy roots 8.22 7.09 6.49 -21.O% Maize 1.70 1.71 2.36 38.6% Vegetables andbeans 7.68 6.48 6.38 -17.0% Rawmatlsfor manuf 3.41 2.10 3.45 1.1% Source: MAFF data cited by NSC productivity remains relatively low (see Figure lo). 35. While agriculture value added per worker improved by about a third over the 1 9 9 0 ~ ~ Factors that have contributed to low productivity include the very low use of chemical fertilizer, the small proportion o f irrigated cropland, and the near absence o f high yielding cereal varieties.26 While overall fertilizer use quadrupled over the 1990s, it remains low on an arable hectare basis at 10 kg/ha. in 2000, compared to 121 kg/ha for Indonesia and 364 kgha for Vietnam. In 1998/99, only 28% of farm households in Lao PDR used mineral fertilizer, and only 29% used farm machinery. Because o f mountainous terrain, the proportion of land suitable for irrigation is limited. The proportion of irrigated cropland in Lao PDR in 1994-96 was only 18%, compared to 28% for low-income 26Lao PDRagricultural production systems canbe grouped into four major types. The lowlandrainfed system is characterizedby a single paddy crop during the wet season. The lowland irrigated system supports both a rice crop inthe wet seasonand a dry seasoncrop. Upland systems are characterized by shifting cultivation. Plateau farming involves mostly coffee cultivation. 27However, the area served by irrigation schemes has increased significantly recently from217,000 ha in 1997198 to 295,000 ha in 1999/2000 as a result o f heavy investment inirrigation schemes and diesel pumps beginning in 1997. 20 36. Giventhe prevalence o f subsistence farming, agriculture contributes only a small share o f total exports, unlike in most low-income Asian countries.28 The openness o f the sector i s very low, with the ratio o f total agricultural imports plus exports to agricultural GDP among the lowest in the region.29 Substantial growth of exports o f commodities i s not likely inthe near- term.30 Main export opportunities are likely to be cross-border and sub-regional, as inthe case of livestock (pig) production inHouapanhprovince for Hanoi, but needto be facilitated. Figure 10: Agricultural value added per worker (constant 1995 USD) .... -..__ ~ ---......_ Source: FAOSTAT 37. Long-term growth will come from a transition to greater market orientation, including diversification away from rice, intensification in favorable areas, and increased regional specialization. An agriculture potential assessment study by JICA seems to indicate substantial scope for expanding agriculture from the standpoint o f suitable land resources and estimates the potential total area for annual and perennial crops to be as high as 3.1 million ha, compared to the 1.1million ha. currently under c~ltivation.~~Coupled with an expanding rural labor force due to population growth o f nearly 2.5% per year, such additional land and labor inputs will drive agricultural expansion. Sustained growth over the long term, however, will need to be driven by productivity growth and efficiency improvements from increased diversification, intensification, and specialization. '*Coffee alone accounts for over half o f official agricultural exports. It should be noted, however, that unofficial and unrecorded trade, e.g., in live cattle, is also likely to be very significant. 29 The ratio for Lao PDR was only 0.09, much lower than inCambodia (0.18); Indonesia (0.30); Vietnam (0.42); and Thailand (0.57). 30Reasons include (i) transportation infrastructure and services; (ii) poor relatively low purchasing power innearby economies; (iii) similarity between Lao and neighbouring countries agricultural products; and (iv) risk aversion o f rough small-holder production base and weak links to export opportunities. Furthermore, farmers interested inexport markets will therefore have to distinguish between production for own or domestic consumption with that for world markets. While glutinous rice accounts for about 96 percent o f total rice consumption inLao PDR, world markets are predominantly for ordinary rice. 31Based on physical parameters such as landuse, soil and climatic conditions, topography, and forest cover (Ministry o f Agriculture and Forestry (MAF) Master Plan (2001). 21 38. Diversification from rice i s likely to increase as markets are better integrated and new areas are brought into cultivation. Under rain-fed conditions, other crops are better suited to cultivation indrought-prone areas. Much o f the potential area expansion cited above will be in hilly areas rather than lowlands and more suited to tree crops and livestock rather than food crops. 39. Regional specialization has been low and considerable scope exists for intensification o f rice production in favorable areas and increased regional specialization. Intensification would be based on more widespread use o f improved varieties, increased fertilizer use, better crop protection practices, and improvedpost-harvest processing and management 40. Better domestic market integration will be needed to support the above process. Rice markets remain fragmented as evidenced by wide differences in provincial and regional prices.32 Improved market integration would help households manage the tradeoff between lowering risk through local self-sufficiency inrice with increasing incomes through responding to market opportunities incash crops. 41. Improving security of tenure to agricultural land is important for farmers' incentives to make productivity investments, including in perennial crops. While all land in Laos is vested in the State, land use rights are informally recognized and a process of administering and recording land transactions with these rights is being developed. Since the early 1990s the Govemment has moved to formalize the land administration system, with the 1991 Constitution affirming the rights o f individuals to use land and subsequent decrees addressing land management and admini~tration.~'Expanding the land registration system to encompass rural areas and rationalizing the Government's program o f allocating agricultural and forest land to eligible beneficiaries will be important challenges to be met. The framework to enable assignment o f rights to groups also needs development and piloting, as this will be more appropriate than individual rights systems in many areas where swidden agriculture currently dominates. 42. Particular attention will need to be paid to managing risks associated with local food security and resource degradation. Given inadequate rural infrastructure, off-farm opportunities, and formal social safety nets, attention to local food security and i s a prudent transitional stance. Households are still subject to significant climatic risks as indicated by the saw-tooth pattern o f agricultural growth. Achieving sustainable, higher rates o f growth will also require attention to land capability, watershed and river bank protection, and appropriate use o f arable, pasture, and forest land." Land and water resources are under considerable demographic pressure. Data from the 1998/99 agricultural census indicate that as much as 29% o f agricultural land was cleared within the last 3 years. Coupled with the surprising drop in planted area from1990-1995 cited earlier, this could indicate the possibility o f huge land degradation andweak landtenure.35 Cited inBourdet (2000) The Land Law o f 1997 identifies types o f land rights that can be granted by the State to citizens, land classification and landmanagement(including maximum size limitsfor various types o fagricultural land), establishes provisions for land registration and issuance o f landcertificates, and defines administrative responsibilities inthe landmanagement system. A LandTilting project has been implemented since 1997 as the first phase of a long-term strategy to complete registration o f all rights inland and strengthen landmanagement. The area under forests has been decliningwith a loss o f about 1 million ha. or 8 percent over the past two decades. At current rates o f deforestation, remaining forest area will decline from 45% o ftotal land area to below 35% by 2020. Ifbothdataare correct, thiswould indicate that previouslyclearedland(drop inharvestedarea from 1990to 1995) was so useless or degraded that it was more worthwhile to clear new landthan use previously cleared land, and that land tenure is sufficiently weak to allow this to happen on such a scale. 22 2.2.2 Manufacturing and services 43. W%ilethe manufacturing sector has performed well, itfaces signiJicant challenges and risks in core sub-sectors. Furthermore, the sector is overwhelmingly dominated by very small enterprises, suggesting possible constraints to the ability of the sector to expand exports and employment. Sustaining the performance of the sector will require improvements in key building blocks such as the environment for doing business, infrastmcture, and market institutions, as well asfacilitating regional linkages. 44. The manufacturing and services sectors have performed well. Over the last decade, the manufacturing sector has enjoyed growth o f over 10% on average, and accounts for over 18% o f GDP. Most o f the growth contribution has come from food and beverages, which alone accounts for over 78% o f value added in manufacturing and 15% o f total GDP. Despite very rapid growth rates during the early 1990s, garments and wood processing still account for a relatively small share of total value added in manufacturing, only about 6% and 3%, respectively. It should be noted that UNIDO survey data yield quite different subsectoral breakdowns inmanufacturing compared to NSC data used for estimating value added inGDP.36 K e y reforms areas inagriculture Sustaining the growth performance o f the sector will require measures to facilitate diversification while maintaining food security, including Promoting market integration and efficiency. This includes (i)improving transport infrastructure, especially rural roads and (ii) removing policy distortions which contribute to fragmentation o f domestic markets and affect farm gate prices. Improving rural roads is a high priority as fewer than 2% o f rural roads are paved, most are impassable in the rainy season, and alternative transport options are limited.37 Price controls are still practiced, with rice and meat prices set within ranges determined by the central government with actual prices set by the provinces. Transportation o f rice and livestock between provinces is strictly controlled. Strengthening agricultural services. Core research and extension agencies-National Agriculture and Forestry Extension Service (NAFES) and National Agricultural and Forestry ResearchInstitute (NAFRI)- are o f recent origin and need to be strengthened given the need for priority research o n diversified cropping under irrigation inthe dry season. Effectiveness o f such programs will be enhanced by village organizations with whom Government can partner. Progress o n developing sustainable agricultural credit and mico-finance i s also needed, including a move away from reliance o n subsidized lending through APB. Strengthening land administration to improve security o f land tenure and provide incentives for investments inimprovingproductivity, including inperennial crops. j6 Lao PDRhas norailway, limiteduse o f inlandwaterways, and problematicthird countrytransit arrangementsand access to seaports. j7These figures are basedon GDP figuresprovidedby the NSC. There are wide discrepanciesbetweenthe subsectoral breakdownof manufacturingbetweenNSCdata and figures from the UNIDO manufacturingsurvey. The WID0 survey estimatedamuchlower share of value added inmanufacturingfrom food andbeverages (29%), andamuchhigher share from tobacco(10%); garments(25%); and wood andwood products(10%). Since the analysis inthis report focuseson contributionsto aggregate GDP growth, for consistency,we rely on the NSC data. The UNIDO data also do not provide subsectoral data over time to allow estimation o f subsectoral growth rates. 23 45. Services account for an even greater share o f GDP (over 25%), but performance has been less strong, with growth around 6% on average. The bulk o f the growth contribution has come from wholesale andretail trade, which accounts for over 38% o f services value added and nearly 10% o f total GDP, and has grown at over 8% on average. The second most important subsector is transport, which accounts for about 20% o f services value added and 5% o f GDP. Tourism is widely cited as an area o f high growth potential in services for Lao PDR, but the present, poor information base precludes any detailed analysis.38 Using World Development Indicator figures for cross-country comparisons, however, it is clear that Lao PDR attracts only a tiny fraction o f the tourists received by its neighbors (about 3% and 14% o f Thailand and Vietnam, respectively). Improving tourist infrastructure and service quality could allow Lao PDRto tap into these large regionaltourist flows as part o f a multi-country visit. 46. While the manufacturing sector has performed well, it faces several significant challenges going forward. First, competitiveness is limited. While labor costs are low, labor productivity-as proxied by value added per worker-is correspondingly low. For example, while Lao PDR's labor costs are only 44% (50%) o f Indonesia (Sri Lanka), its value added per worker i s only 33% (46%)." 47. Not surprisingly, given its landlocked status, logistics and transport costs are much higher than for other countries. Estimated transport costs for a 20 ft. container for general commercial cargo from Vientiane to Los Angeles is over 1.4 times that from Bangkok or Beijing. The costs o f shipping a container from Vientiane to Bangkok i s estimated to be 1.4 times the cost from Beijing to Bangkok.4o The cost o f importing and exporting through Thailand is high due to high trucking costs and time consuming customs procedures. Despite these high costs, Thailand remains the preferred transit route over Vietnam due to even more cumbersome procedures with the latter. These higher logistics and transport costs are not offset by significantly lower costs inother areas. The comparative cost o f doing business inLao PDR i s also high. While Lao PDR has cheaper labor costs, it has average electricity, telecommunication, andrental costs compared to other East Asian countrie~.~~ 38 Basedon Lao PDRdata, Lao PDRreceived674,000 touristsin2001, of which 56% came from Thailand, down from about 750,000 in2000. World DevelopmentIndicators(WDI) data, by contrast, recordthat Lao PDRattracted300,000 touristsin 2000. While there is no estimatefor the presentGDP contributionof tourism, the hotel andrestaurant sector currently accounts for about 2% of GDP. 39 UNIDO IntemationalYearbookof IndustrialStatistics; MIHSurveyof Industrial Establishments 1999. 40 Basedonquotationsof freight forwarding companies. Most goods shippedout of LaoPDR are shippedunbundledin trucks then loadedto containers inThailand. 41 Basedon estimatesfor Lao PDR comparedto regionalinformationcompiledby JETRO. 24 Figure 11: Labor costs and value added per employee inmanufacturing (1999) 60.00% 50.00% 50.00% 1 n46.30% 44.40% 40.00% + 32.80% 30.00% 27.10% a 20.00% 13.30% - 10.70% 10.00% 0.00% 1 Philippines Thailand Indonesia Si lanka Labor costs inLao PDR relative to those incountry Value added per employee inLao PDRrelative to those incountry source : UNIDO InternationalYearbook of industrial Statistics ;MIHSurvey of Industrial Establishments 48. Second, several important subsectors face specific risks. Garments and wood and wood products together accounted for over 50% o f total exports in 2002. In the garments sector, the phasing out o f export quotas under the Multi-fiber Agreement (MFA) in 2005 represents a significant risk for the Lao PDR garment sector as the main commercial justification for many o f the garment factories currently located in Lao PDR is to take advantage o f the country's quota allocation. The competitiveness o f the garment industryinLao PDRrelative to those inChina and Vietnam remains questionable. Already, foreign investment inthe sector has slowed down andhas fallen from USD 4.8 millionin2000 to USD 0.4 million in 2001. Garment exports fell by nearly 20% during the first half of 2003 on a year-on-year basis. Wood and wood processing is characterized by low value-added and efficiency, large SOE presence, non-transparent allocations o f logging quotas, and capacity utilization o f only 20-30%. To operate at full capacity, the wood processing industry would require an equivalent to 3-5 times the sustainable annual allowable cut. As noted in the discussion o f the forestry sector, on present trends and without significant reforms, the sector will be unable to contribute meaningfully to the economy inthe not too distant future. 4 25 49. Inservices, wholesale and retail trade face infrastructure and other constraints. The private sector plays the dominant role as state trading monopolies have been largely abolished. While there is a dearth o f reliable statistics on the subsector, anecdotal evidence suggests that larger firms are often engaged in some form o f wholesale trade o f imported products, and that the informal and household sector dominate retail trade. Constraints faced by the trade subsector include the high cost o f business operations, poor transport and logistics infrastructure, and the small size o f the domestic market. Tourism faces additional constraints of underdeveloped tourist destinations and service skills.4z 50. The Lao PDR manufacturing sector is overwhelmingly dominated by small, informal private enterprises. Medium and large enterprises (those with more than 10 employees) accounted for only 30-40% o f total manufacturing output and value added, with the balance accounted for by small enterprises. Only around 540 firms employed 10 or more employees, o f which half are located in Vientiane municipality and province, with foreign owned or controlled garment manufacturers by far the largest firms. The private sector plays the dominant role inmanufacturing, with SOEs contributing only 20% of value added. 51. Formal manufacturing and services firms account for limited employment. The formal manufacturing and services sector i s estimated to employ about 100,000 persons and 22,000 persons, respectively, or only around 5% o f the total labor force o f about [2.4] million. Small manufacturing establishments and the approximately 150 large manufacturing enterprises with over 100 employees each accounted for about 45% o f formal employment. Box 3: Constraints on enterprise development A survey of 111mediumand large private companies inLao PDR by the Mekong Project Development Facility in 1997 identified fluctuations in exchange rates; access to foreign currency; finance; poor quality of roads; and entering foreign markets as the five main constraints in operating their businesses. More recently, focus group meetings inPakse and Savannakhet inJanuary 2002 on Lao enterprise development issues organized by the World Bank identified similar issues. Finance and trade facilitation were particularly highlighted. Participants noted the difficulty o f accessing finance and finding working capital. Interest rates were high, with SOEs and businesses affiliated with Government received more favorable interest rates. Poor trade facilitation remains a clear constraint. Interms o f production, participants cited lack o f availability o f good quality domestic inputs and the needto import appropriate inputs. However, without the right contacts and network, the cost of importing is high to rent-seeking behaviour by authorities. Exchange rate volatility also made i t difficult to manage costs and production. Transport logistics are poor and high cost. Participants noted that transport logistics are largely dominated by a few Government controlled firms, with smaller family-owned firms unable to transport large volumes at lower cost and delay. Source: 1997MPDFsurvey; 2002focus groups 42Lao PDRtourismnumbersare only a fraction of its neighbours, suggestingsignificantupside potentialifit cantap into regionaltourismflows as part of a multi-country visit. Basedon World DevelopmentIndicator figures for cross-country comparisonpurposes, Lao PDR attracted only 300,000 touristsin2000 comparedto 2 and9.5 million inVietnam and Thailand, respectively. (Lao PDR official figures recorded674,000 tourists in2001, of which 56% were fromThailand.) 26 52. The very small size of these firms indicates possible barriers constraining the growth of firms and limits on the ability of the sector to expand exports and employment over the near term. A 1999 UNIDO survey o f small manufacturing establishments recorded a total o f 21,800 establishments with an average size o f about 2 persons. The "missingmiddle" and lack o f medium-sized firms suggests that there are a number o f barriers blocking the growth o f small scale firms.43 While there has been limited formal, analytical work done on the constraints to firm growth, anecdotal evidence indicates that government interference may be an important reason constraining firms from growing (See Box 4). Surveys and focus groups have also identified access to finance, inputs, and logistics as important constraints on enterprise development in Lao PDR (See Box 3). Given its central importance for long-term growth inLao PDR, a formal analysis of such constraints needs to be undertaken. 53. The inability to grow such medium-sized firms will place significant limits on the ability o f the sector to adopt new technology and expand employment. Coupled with the small, scattered population, low urbanization, poor transport links, the lack o f medium and large scale enterprises also suggests limits on the ability o f Lao PDR to rely heavily on a strategy o f low cost, labor-intensive manufacturedexports to major world markets, at least untiltransport costs and these other constraints are reduced. 54. Nevertheless, the sector has a critical role to play in improving development outcomes by providing off-farm employment opportunities. Three-quarters o f the enterprises in the UNIDO survey were micro-grain rice mills, with an average number of only 1.7 persons.44 Such small, rural enterprises in both manufacturing and services will play an important part increasing household incomes and improving poverty and social outcomes (See Box 5). However, if enterprises-both rural and urban-are unable to grow, then the scope for off-farm employment will be correspondingly limited. ~~~~ 43 A formal analysis o f such constraints needs to be undertaken given its importance. A survey o f 111mediumand large private companies inLao PDRby the Mekong Project Development Facility in 1997 identifiedfluctuations inexchange rates; access to foreign currency; finance; poor quality o f roads; and entering foreign markets as the five mainconstraints in operating their businesses. Almost all o f the establishments were runas individual proprietorships as opposed to cooperatives or partnerships. These findings are consistent with those o f an earlier 1996 GTZ study o f 2,799 SMEs which found the averagenumber o fworkers per firm to be only 1.9 workers, including the owner. 27 Box 4: The value of keepinga low profile "Two unrelatedfirms...recently began attending handicraft trade shows in Europe and the US. They began to make a profit from their new orders. Followingpublicity about the trade shows' successes, local authorities began to charge thesefirms higher taxes and higher 'unoficial' export duties for finished goods. It also began to take more than a monthfor theirfinished goods to clear customsproceduresfor export. Higher amounts of money and time were needed to manage export and tax authorities. This anecdote illustrates a commonpractice and the well- known 'tallpoppy 'syndrome: the tallestflowers are most vulnerable to the mower. " Source: Mekong Project Development Facility (1998), A Survey ofMedium and Large Companies in Lao PDR Government interference and cumbersome regulations may be important reasons which contribute to the large number o f small, informal firms. Based o n the recent "Doing Business" database of the World Bank, for example, while the number o f procedures involved is roughlycomparable to other countries, the number o f days involved in starting up a business inLao PDR is three times the regional average. Figure 12: Starting a businessinLao PDRv. elsewhere 198 I OECD Average RegionalAverage LaoPDR Source :WorldBank, Doing business Data 28 I Box 5: Off-farm and non-farm activities inLao PDR I Rural industries inLao PDR typically consist o f small-scale operations and is an additional activity undertakenby households inaddition to fanning. Typical off-farm and non-farm employment include such activities as weaving, sugar cane processing, bamboo handicraft, and fishing and fish processing. A recent small survey of 240 householdenterprises in these four areas found that (i) workers are family members with only a very small most proportion o f hired workers; (ii) acquired their skills through self-study and leaming from family, friends, most and relatives rather than from formal schooling or training; (iii) there was a reliance o n traditional tools with almost no mechanisation (with the exception o f fishing); and (iv) the share o f household income from these activities was between about 60-70%(but was much lower with sugar cane processing). Importantly, the study found huge gaps between the prices received by producers and market prices even during the dry season, reflecting poor transport and market links, and suggesting large profits by traders and intermediaries. As indicated in Figure 13 producer prices in handicrafts, for example, were less than 60% o f market prices. The study identified the main constraints on the development o f rural off-farm and non-farm activities inLao PDR to be (i) poor transport infrastructure and communications; (ii) underdeveloped markets and lack o f market information; (iii)lack o f skilled labor; (iv) underdeveloped rural savings and credit systems. Figure 13: Market margins are large Producer prices as % o f market prices 82 8 86 8 Handicraft- Weaving-Silk Sugar cane Fishing-Smaked Small sticky rice skirt processing- and dried fish basket Brown sugar Source: Keo-y et.al. (2003) "Off-farm and Non-farmE?mployment and Activities in the Lao PDR" inOff-farmand Non-farmemployment inSoutheast Asian Transitional Economies and Thailand. (2003) 29 55. Sustaining the performance of the sector will require progress on improving human resources and infrastructure, developing market institutions, establishing a predictable business environment, and facilitating regional linkages. Addressing infrastructure needs i s an obvious priority. Based on WDI data, the percentage o f paved roads in Lao PDR (14%) is among the lowest in the region and considerably lower than its neighbours Vietnam (25%) and Thailand (98%). The difficult challenge o f improving human resources has beendiscussed ina preceding section. 56. As may be expected for a transition economy, Lao PDR still needs to develop institutions required for an effective market system. Of central importance are institutions to enforce the rule o f law and ensure contract enforcement. Their absence may result inexcessive reliance upon "personal relationships" with decision makers, which contributes to an unpredictable or uneven playing field for business. Inadequate o f transparency o f decision making processes in government exacerbate this problem. Privileges o f SOEs through regulatory and administrative procedures create biases against the private sector. Inthe forestry sector, for example, allocation o f logging quotas generally favor SOEs. The range o f products that can be imported by private sector companies is more restricted than that for SOEs as a result o f Ministry o f Commerce regulations. SOEs are particularly dominant at the provincial level in Lao PDR, and provincial governments are often more supportive o f them than private firms. 57. While improving humanresources and infrastructurewill take muchtime andresources, Lao PDR could do much to remove the constraints faced by private manufacturing and service firms simply by "letting go." InVietnam, for example, the enactment o f the Enterprise Law in 2000 reflected a change in official mindset from a business license system-where the government perceived its role interms o f restraining and redirecting private sector activities-to a business registration system, where the govemment set the ground rules and let the private sector operate freely within prescribed limits. The response was dramatic, with about 80,000 new enterprises registered between 2000 and 2003. As noted in Box 4, freedom from government interference and less cumbersome regulations could help reduce the "tall poppy" syndrome and encourage more private firms to become bigger and to move into the formal sector inLao PDR. 58. Regional linkages are likely to be more important for Lao PDR than they were for its neighbours. As noted earlier, exports o f low cost, labor-intensive manufactured goods to major world markets are unlikely to be as important a source o f growth for Lao PDR as it was for its neighbours given its land-locked situation. By contrast, its higher share of ASEAN trade, and location amongst rapidly growing economies highlight the relative importance o f its regional trade and the need to undertake measures to facilitate such linkages. 59. The 2002 Lao PDR Industrialization and Modernization Strategy (2001-2020) notes the country's location in the region as a possible source o f advantage and identifies a three-phase process for developing transit service infrastructure, encompassing not just road but water, air, and rail links. As noted in Box 6, however, while infrastructure is undoubtedly important, it must also be accompanied by increased openness and policy and administrative measures to facilitate trade. Otherwise, expensive infrastructure will be underutilized. 30 Box 6: Landlockedor landlinked?Regionallinkagesand Lao PDR Lao PDR is a small, mountainous , landlocked country surrounded by five other countries - China, Vietnam, Cambodia, Thailand and Myanmar. Owing to a sparse population and a rough topography, improvements in transportation are difficult and costly. Compounding the problem i s widespread deterioration caused by a neglect o f maintenance and war. During the past decade, some improvements have been achieved in rehabilitating the national road system. Major improvements have been achieved inRoute 13, the North-South route connecting the Chinese, Thai and Cambodianborders and passing through the country's major urbanized areas. Major investments inRoutes 8 and 9 have connected Lao PDR with the Vietnamese ports of Vinh and Danang. LongstandingThai road investments have meant that Vientiane has long been well connected with Bangkok and the Thai ports. This progress has permitted substantial national and international access for the capital and a number o f provincial capitals to important centers o f economic activity. The major problem lies with provincial and district roads and is likely to remain serious for some time to come since not enough resources are available to rehabilitate or provide appropriate maintenance for them. As a result, a substantial part of the Lao population will remain without adequatetransport links for the foreseeable future. Geography has an important impact on economic growth. One study finds that long coastlines and the ratio of coastline to total land area make a positive contribution to growth Bycontrast, landlocked countries face a number ofproblems including poorer terms oftrade, lower capital goods imports, andlower export-oriented FDI. Since higher transport costs limit the range o f trading partners, the fortunes o f landlocked countries are more closely linked with those o f nearby countries. The neighborhood they are in matters a lot. Lao PDR's own experience appears to bear this out. Growth in GNP per capita for a sample o f 20 landlocked countries in Sub- Saharan Africa, South Asia, and East Asia averaged only 0.2% during 1990-2001, compared to 1.4% for lo^ income developing countries.46 By comparison, Lao PDR's per capita GNP growth o f 3.8% was the highest for a1 landlocked countries, suggesting that Lao PDR may have been able to reap some advantages from being in a fas growing neighborhood. Improving transportation and trade links with its regional neighbours should allom itto build on these advantages. Improving transportation Among landlocked countries, Lao PDR already stands out for the amount o f resources it devotes to the transport Inadditionto the internationalroutes citedabove, it will soonbepossible to make the 2000 km.trip from Kunming in Yunnan, China to Bangkok in less than 20 hours owing to the completion o f the North-South Corridor. With appropriate complementary feeder road investments, this has the potential o f connecting millions o f Yunnan and Lao small farmers andbusinesseswith nearby countries and other international markets. (Cont.) 45 Sachs, J. ,Radelet, S. and Lee, J-W "Economic Growth inAsia" (draft paperpreparedfor ADB study on Emerging Asia:Changesand Challenges; July 1997). 46Of course, this does not control for other factors. 47A studyo f 30 landlockedLDCs foundthat Lao PDR devotedthe largest share of GNP to transport (4.9%), alongwith Mongolia (3.5%) andthe CentralAfrican Republic (2.7%), while 13 others devotedonly 1-2% andthe remaining 14 less than 1%. See Chawdhury,A.K., "EmergingPriorities for Transport"(Wor1d BankTransport Forum2003; WashingtonD.C.; January21,2003) 31 (Cont.) I Improving trade and the trade regime During the past decade, tariffs have beensubstantially lowered with a major reduction in 1995 when a multiple rate system with a 150% maximum rate was reduced to only 6 rates ranging from 540%. Exemptions are numerous, with about 80% o f tariffs foregone.48 While import and export licensing were previously pervasive, a number o f liberalizing changes were introduced in2001, including abolishing export licenses (except for logs and sawn timber) and dropping license requirementsfor those who submit and get annual business plans approved by relevant agencies. However, there still appears to be no publicly available list showing which imports do and do not require licenses. Going forward, it will be important for Lao PDR to improve trade facilitation and make progress on its commitments under the ASEAN Free Trade Area (AFTA) to improve its regional linkages. Complicated administrative requirements, non-harmonized procedures and documents all hinder trade. The ASEAN Framework Agreement on the Facilitation o f Goods in Transit is intended to simplify and harmonize transport, trade and customs regulations and requirements for goods in transit, but is not yet effective. Lao PDR's preferences and commitments to AFTA are likely to be critical. Unlike the original six ASEAN (ASEAN-6) members whose trade is predominantly with non-ASEAN members and whose ASEAN trade is primarily with Singapore, a near free trade economy, ASEAN takes one thirdo f Lao PDR's exports and supplies two thirds o f its imports, figures that would be even larger were Vietnam included. AFTA membership carries obligations to liberalize trade with other members on set schedules and according to potential imports classified on four lists, including (i) Inclusion List (IL) o f goods immediately eligible for preferential treatment under certain an conditions; and (ii) a Temporary ExclusionList (TEL) o f goods which must be moved to the ILaccording to a set ~chedule.~'Each member specifies a Common Effective Preferential Tariff (CEPT) for each IL and TEL item, and IL items canbe imported at this rate-exempt from all non-tariffbarriers-under three conditions.50 The four new ASEAN (ASEAN-4) members are committed to moving all TEL items to the IL and setting all IL rates to zero by 2008. By May 2001, the ASEAN-6 had included 98% o f all tariff lines inCEPT arrangements and 93% o f these had tariff rates inthe 0-5% range, with zero to be achieved by 2003. This creates a large incentive foi Lao PDR move items onto the IL with rates o f 20 percent or less to take advantage o f preferential rates from the ASEAN-6.51 Annex Table 1discusses some o f the changes inthe trade regime needed for Lao PDR to take full advantage of integration into regional and world markets. I 48Dutiesas ashare oftotal importsis only 3%, comparedto an import-weightedaverage tarqf o f 14.7%. 49The other two lists are sensitivegoods (unprocessedagriculturalgoods) andthe generalexceptionlist (GEL) coveredby Article XX of the GATT. These are (i) of value added must be fromwithin ASEAN; (ii) itemmust also be on the exportingcountry's IL; and (iii) 40% the exportingcountry'sCEPTmust bebelow 20% ifthe importingcountry's CEPT is below20%. By May 2001, the ASEAN-4 had includedonly 54% oftheir tariff lines on the IL,but had43% percent on the TEL. 32 Key reform areas in manufacturing and services 1. Improve business environment by creating a predictable, transparent and favorable policy environment and level playing field, above all by "letting go," including (i) moving away from "picking winners;" (ii) removing biases in administrative procedures and regulations that favor SOEs over domestic or foreign private firms; (iii) identifying and removing unnecessary regulatory and administrative complexities; and (iv) strengthening market institutions, including dispute resolution and contract enforcement. 2. Improve supporting infiastructure, including (i) transportation and logistic infrastructure to access regional and international markets; (ii) access to financial resources and services; and (iii) resources to increase human productivity through education and training. These are also the areas identified as constraining the development o f the off-farm and non-farm sector inLao PDR. 3. Facilitate regional linkages, including carrying out commitments under AFTA (speeding up move of Temporary Exclusion List (TEL) items to the Inclusion List (IL), setting Common Extemal Preferential Tariff (CEPT) rates below 20 percent to take advantage of preferential access to ASEAN-6) and extending CEPT rates on a MFNbasis. I 2.2.3 Forestry 60. While the forestry sector has the potential to contribute muchmore significantly to Lao PDR's development, the sector remains in disarray. Improved forest management, including greater community participation, i s critical to improving the development contribution o f the sector. 61. The forestry sector has the potential to contribute much more significantly to Lao PDR'sdevelopment both by increasing revenues and improving livelihoods. While forests cover nearly halfo f Lao PDR, it contributes only around 3% of GDP in2001-2002, down from over 5% duringthe 1990s, but still contributes significantly to exports.52 While the sector grew at over 40% on average duringthe first halfo f the 199Os, it contracted bynearly 9% on average duringthe second halfo f the decade before growing again in2001 and 2002. The contribution of log royalties to government revenues has decreased from 17% o f total revenues in the mid 1990s to around 9% in 1998/99." The main reasons include low royalties, waste, irregularities, and low collection rates.54Between FY1994/95-1998/99, the Government received USD 114 million in royalty payments, while the market value o f the logs i s estimated at USD 340 million. As a result, the Government realized only about one-third of the estimated market value of the timber harvested. Forestry can also contribute directly to the livelihood ofthe rural poor by providing food, fuel, fiber and shelter. Such non-timber forest products (NTFPs) are estimated to contribute 40% o ftotal rural family income. li2Woodand wood productsaccountedfor 21% of2002 exports. 53 Payments for logs are comprised o f royalty, reforestation fee, and provincialAoca1development taxes and fees. The first two are payable to the national treasury while the third are based on local level decisions that vary from province to province with considerable variation. li4A comparison ofroyalty rates with intemational referenceprices appears to indicate that Lao prices were 67-93% of international prices. In-forest waste is high as logs are not recorded untildelivery to the second landing. 33 62. However, the sector is endangered by current practices. Deforestation and forest degradation are accelerating at an alarming rate. Forest area was estimated at more than 64% in the 1960s, and at 47% in 1988/89. Between 1993-1997, the deforestation rate was estimated at nearly 0.6% per year. At this rate, forest cover will fall below 35% by 2020. Of equal importance i s the degradation o f forest quality through uncontrolled logging o f select, commercially valuable species, leading to questions regarding ~ustainability.~~ 63. Poor forestry policies and practices-characterized by inadequate transparency, predictability, and accountability-are contributing to this situation. Procedures are not clear and criteria are not disclosed. Agencies involved can seldom provide a clear description o f the processes involved in setting quotas, issuing permits, and establishing royalties. Inadequate transparency, predictability and accountability all contribute to poor forestry practices and problems. The Forestry law covers forest management but does not clearly address village forestry, forest plantations, sustainable management practices. The Agriculture Law and Land Law overlap with forests and forest lands. Three state-owned enterprises with special access dominate the forestry sector. 64. Policies to increase the share o f value added products by encouraging exports o f plywood and furniture have also appeared to have had questionable While they have received emphasis by the Government as a future source o f timber to replace natural forests, timber plantations are not a substitute for the natural forest, constitute only a very small portion o fthe resource base, require substantial investment, andhave low produ~tivity.~~ 65. Business as usualwill resultin depletionof this resourcebasewithout contributing to national development. Lao PDR's forests are a tremendous resource which conserves biodiversity, soil and water values and could contribute substantially to poverty alleviation. However, the lack o f appropriate forest management, weak legal and administrative settings, and an incentive structure which favors short-term earnings over longer-term sustainability are resultinginoverexploitation. These conditions have to be reversed and legislation improved if production forestry is to remain as a major economic sector. 55 Total extraction(recordedcommerciallogging andestimatedhouseholduse) fromproductionforests is estimatedto be on the order of 1.3 million m3 andthe interim annualallowable cut (AAC) estimatedat 0.6-0.7 million m3. 56 Analysis o fnet economicrevenue indicates that on averagethe benefitto the Lao economywas USD41 higherper m3 of log ifthe logswere exportedunprocessedwith all royaltiesandtaxes recoveredthan ifthey hadbeenprocessedlocally and then exported. 57Plantationsestablishedduringthe 1990s accountedfor only 2.3 % oftotal forest area. Productivity of establishedplantations measuredby privateinvestorsis low. The MeanAnnual Increment (MAI) for Eucalyptuscamaldulensis was 10-30 3/ha/year, which compares poorly with improvedhybrid stock which managedwith appropriate silviculture andsite-species matchingcan achieveMA1in excessof 100m3hdyear. 34 Key reform areas in forestry Realizing the development contribution o f the sector will require substantial reforms to improve forest management andutilization; expand community participation inforestry management; and control illegal logging. Improvingforest management and utilization. The regulatory apparatus needs to be strengthened. The line o f command between the Department o f Forestry, provincial authorities and Provincial Agriculture and Forestry Office (PAFO) foresters needs to be clarified and an effective regulatory body capable o f overseeing forestry operations needs to be established. Market-based procedures inawarding logging and management rights and transparent bidding andpricing o f log sales would increase public revenues. Expanding community participation. Appropriate village communities should be eligible to receive management rights to forests and the commercial autonomy of village forestry associations (WAS)assured. Participatory forestry is relevant to 54% o f the potential production forest area with a beneficiary population o f 1.S million people. Control illegal logging. Unclear and overlapping jurisdictions and responsibilities between different government authorities and the lack o f enforcement capacity and compliance compound the difficulty of defining and controlling what constitutes illegal logging. Virtually every logging activity has beenauthorized by someonewho has or claims to have authority. 2.2.4 Hydropower 66. While Lao PDR has impressive hydropower potential, many upfront steps are required before the development contribution of the sector can be realized, including improving the enabling environment for private sector participation. Developing hydropower is not without signi$cant risks. Ultimately, however, the development contribution of the sector will depend critically on how the revenue streams are managed. 67. Hydropower presently contributes crucially to power generation and exports. The electricity sector in GDP has grown extremely rapidly, especially during the latter half o f the 199Os, raising its share of GDP from slightly more than 1% o f GDP during the first half of the 1990s to nearly 3% o f GDP presently. Over 97% o f total electricity generating capacity inLao PDR is based on hydropower resources. Total generation capacity is 643 MW, while peak domestic demand at the end o f 2002 was about 210 MW. Surplus power is exported and rep resents an important source o f foreign exchange for Lao PDR.58 With the addition o f exports by Theun-Hinboun (THPC) and Houay Ho power companies (HHPC) to those by Electricite du Laos (EdL), overall power exports increased to over 100 million USD as o f 2000, representing over a third o f national exports.59 The Government has agreements-in-principle with Thailand and Vietnam for the export of 3,300 MW and 2,000 MW of power over the decade, respectively.6o '.3Power trade inthe regionhas been facilitated with the signin of an Inter ovemmentalAgreement for PowerTrade inthe GreaterMekon Subregion(GMS) inNovember2002, whicirvides tfe authority and institutional frameworkfor the developmentofa regionalpower market. The GMS includes ambodia, Lao PDR, Myanmar, Thailand, Vietnam, and 59 Yunnan Province, China. Lao PDR also importspower, and electricity imports havebeen around USD7 million inrecent years. 6o This includes the proposed 1080MW NamTheun 2 project. 35 68. Only a small proportion of Lao PDR's substantial hydropower potential has been developed. The total exploitable hydropower potential has been estimated at 23,OO MW. However, the upper limit which could currently be regarded as economic ranges from 3,000- 5,000 MW. The Ministry o f Industry and Handicrafts (MIH) is currently employing consultants to carry out a power system planning study which considers domestic power requirements and export potential. The study i s expected to provide substantial guidance on the selection and sequencing o f hydropower development. EdL's current Power Development Plan foresees the development o f up to 13 smaller projects (each less than 100 MW) to be commissioned over the 2005-2020 period. 69. Many issues, however, need to be resolved before this potential can be realized. First, the economic viability of the proposed, smaller projects needs to be assessed. While hydropower costs are site specific, economies o f scale are important and the economic advantage o f hydropower typically increases with installed capacity. The smaller projects under consideration by EdL may therefore be economically unattractive with the possibility o f meeting domestic demands by taking a share o f larger, more competitive export projects or alternatively, import o f power from neighbouringcountries. 70. Second, mechanisms for involving the private sector and alternative financing models will therefore have to be developed. While larger hydropower projects are more likely to be economically viable, the Government i s unlikely to be able to develop such projects without private sector assistance. Financing o f EdL's system expansionhas traditionally been provided by multilateral and bilateral agencies through soft loans. However with annual growth rates coming o f f a higher base, capital requirements are increasing at a time when competition for concessionary financing i s becoming increasingly fierce. This dilemma can be met ina number o f ways. First, public-private partnerships models with adequate sharing o f risks are seen as one o f the solutions to overcome the financing constraints inherent in the public sector. Second, non-traditional lenders such as China and India are playing an increasing role through the provision o f export credits for projects where goods and construction services are sourced through their respective countries. Notwithstanding, Lao PDRwill only realize the full gains o f such alternative financing models if there i s competitive bidding o f construction contracts. ADB-supported Theun Hinboun (THPC) project provides a good example where significant cost-savings were realized through competitive bidding, whereas other examples confirm the linkage betweenlack o f competition and excessive costs. 71. Third, better sector planning and site-specijk preparatory studies are needed. While Lao PDR was a pioneer during the 1990s in developing policies to encourage private development o f hydropower for export, experience has shown that the procedures for awarding mandates to hydropower developers were inadequate because the mandates were awarded with very few performance requirements included. Particular weaknesses include legally binding Memoranda o f Understanding (MOUs) which enable a developer to tie up an attractive site indefinitely without developing it. While Lao PDR's policies led to the signing o f numerous MOUs in the 1990s, most have become inactive. Efforts are currently being undertaken to terminate inactive MOUs, but the process will be time consuming and, if not carefully implemented, are not without risk to the country's reputation as a recipient o f FDI. Insufficient transparency in negotiating royalties and negotiating such royalties in advance o f power 36 purchase agreements are also undesirable features. For example, construction on the Houay H o project (HHPC) was well underway before final agreements were in place. More comprehensive preparatory studies will allow the Government to adopt more transparent solicitation procedures in order to attract reputable developers and ensure a proper benefit sharing in terms o f royalties. Improvements to the legal, regulatory and institutional environment will also be needed (see below). 72. Developing hydropower is not without risks. The economic attractiveness o f hydropower exports depends significantly upon gas prices in Thailand. The analysis o f the Thailand Power Scenario, for example, noted that the savings to Thailand from including NT2 in its generation expansion plan increase substantially with higher gas prices.6i Alternative sources o f supply inthe region could also pose a threat.62 73. Current power exchange tariffs between EdL and the Electricity Generating Authority o f Thailand (EGAT) are also outstanding issues. Export tariffs have fallen inrecent years as a consequence o f surplus generation in Thailand following the Asian crisis combined with EdL's inability to guarantee power supplies to EGAT. The structure o f the tariff agreement is also important in order for Lao PDR to reap the full benefits o f its hydropower resources. Currently, the tariff agreement does not provide EdL with the correct incentive in determining the installed capacity o f new hydropower plants.63 A uniform, periodically updated tariff level and structure covering all power exchanges between Lao PDR and Thailand is important to facilitate the optimal development o f hydropower inLao PDR. 74. While hydropower does have the potential to contribute significant revenues to Lao PDR, projects need to be carefully selected, structured, and managed if such potential is to be realized. While IPPs can be major export revenue earners, their export revenues are offset by debt repayments, dividends to foreign shareholders, and foreign operating costs. THPC i s estimated to provide annual royalties on the order o f USD 3 million; annual tax revenue of around USD 4-6 million, and annual dividends o f U S D 15-20 million to EdL. However, HHPC, which is not expected to generate net profit inthe foreseeable future, provides a sobering example. While HHPC provides annual royalties o f aroundUSD 1million, the Government i s incurring a net outflow on the servicing o f the debt incurred to purchase an equity stake inHHPC. 61 Vemstrom(2003), ThailandPower Scenario Study, p. 54. 62 In2002, ThailandsignedaMOUwith Myanmarfor thejoint developmentofhydropotentialonthe SalweenRiver nearthe Thai-Myanmarborder. However,no formal studies or estimateshavebeen undertaken, andmost information regardingits potentialis uncertainandspeculative. 63 Inamixedthermal-hydropowersystem, it is economic to increasethe installedcapacity ofhydropowerplantssuchthat they are ableto deliver all availableenergy during the expensivedaily peak periodandnot during inexpensiveoff-peak hours. 37 I Box 7:Economically productive uses of electricity I Inthe village ofTapen, where 58 housesare suppliedby a2.5 kWvillage hydro system, families areusing electric light to work inthe evenings to make baskets, adding about $20 extra income after costs per month. The Tapen villagers sell miniaturized baskets to tourists passing nearby. The tourists also consume cooled drinks. An icemaker was constructed by a local engineer and driven from the turbine by belt during daytime hours, producing enough ice to satisfy both the tourist demand andthe demand from a neighboring village. One villager i s using the electricity to power a fridge that she uses to make sweets for local sale, as well as to charge batteries for customers from nearby villages who pay a fee for this service. Another lady is using the electric light to extend her sewing business into the evening hours, adding about $5 a month to her net income - much more than her monthly expenditure on electricity. It was also found that a carpenter was operating power tools for his furniture and wood preparationbusiness. In Tapen various ideas and plans for using electricity productively are discussed with enthusiasm. For example, the village electricity manager i s planning to install a 7kW hydro-driven generator that would power a rice mill mechanically (this application is universal to most villages with enough hydro potential and inalmost all cases the amortization cost o f increased capacity i s more than covered by the revenues o f the milling operation). But even with the current capacity, Mr. Sinh i s planning to incubate poultry, and a weaving co-op is under discussion that would benefit from light inevening hours. To satisfy strong demandby tourists for local paper, a papermaking co- operative is planned which would use electrically or mechanically drivenpulpers. Source: Project Concept Notefor Southern Provinces Rural Electrijkation II,Annex I I 75. Because the sector will not generate substantial employment in the long run, the development contribution o f hydropower will come largely from increased rural electrification and effective use o f government revenues from hydro-power. The electrification rate in Lao PDR stood at 35% at the end o f 2001. Improvements in rural electrification can contribute significantly to facilitating off-farm opportunities (See Box 7). As with other natural resource sectors, the development contribution o f hydropower will depend crucially upon how the revenues generated are used and shared and what impact that has on the economy.M Participants at a 2002 power sector workshop identified several priorities for the successful development o f hydropower in Lao PDR, including reforms to improve governance arrangements and the enabling environment for private sector participation inthe 2.2.5 Mining 76. While the mining sector currently makes a marginal contribution to the economy, Lao PDR has excellent geological potential to develop a vibrant mining industry. Developing the mineral wealth of the country will require addressing signijkant weaknesses in the existing legal, regulatory, and tax regime. As with other natural resource sectors, the development contribution of the sector will ultimately depend on how the revenue and benefit streams are managed. While EdL doesnot payroyaltiesto GoL inrelationto its hydropowergeneration, EdL i s requiredto subsidize rural consumers. Rentsflow to users rather than host, local communities directly. 65 "Options for Power Sector ReforminLao PDRI organized inNovember 2002. 38 Key reformareasinhydropower 1.Clarifi governance arrangements in the sector, including (i) amendments to the Electricity and other relevant laws and preparation o f implementing regulations; and (ii) commercialization and regulation o f EdL, including the separation o f EdL's role as an operator and investor, establishment o f an appropriate tariff structure for EdL and mechanisms for regulating tariffs on a regular basis, establishment o f transparent cross-subsidy mechanism, and complete separation o f EdL and G o L finances. 2.Facilitate private participation in IPPs by (i) single agency responsibility for issuing IPP licenses; (ii) define resolve issues o f Government shareholding in IPP projects; (iii) rationalize inactive M O U s and mandates; and (iv) develope models o f project agreements such as concessions and PPAs to use as starting points in negotiations with developers. 3.Improve sector planning and site-spec@ preparatory studies, Current planning and implementation system at EdLandMIHare projectrather thanprogrambased, resulting ininconsistent studies anda piecemeal approach inpartbecause it depends onother countries decisions.66More comprehensivepreparatory studies would allow more transparent solicitation procedures and proper benefit sharing. 77. At present, the miningsector is little developedand contributes only marginallyto the nationaleconomy. Miningand quarrying are currently estimated to account for less than 0.5% o f GDP.As o f 2001, it was estimated that the market value o f all mineral produced inthe formal sector in amounted to approximately USD 8 million, with exports accounting for about halfo f the productionvalue67;tax revenues and employment thus remain very limited.6* 78. However, this situation changed in 2003. The new Sepon gold mine began operations. Financed and operated by Oxiana, a private Australian company, it will produce an estimated 140,000 ounces o f gold, with an estimated gross market value o f approximately USD 43 million per annum. In 2005, plans also call for exploiting copper resources and producing an estimated 60,000 tonnes worth about USD 100 million per annum. Production o f the gold and copper miningphases is expected to last 20 years. Other mines in the formal sector produce other metals, industrial and construction materials, but are small, operations in intemational 66 With amore programmatic approach, investment projects could be formulatedto support an optimum development program. At present, resource inventories are done inapiecemeal fashion and gridextensionplanningonly occursduring preparation of a proposedproject. 67Operating costs and value added are not available. 68 Basedon declaredproduction figures, tax revenues are estimated at approximately USD 120,00O/year. No reliable figures exist for employment, but formal sector employment is estimated at less than5,000, with between 8,000-15,000 persons engagedin informal mining. 69 State owned enterprises are few innumberbut account for approximately half o f production. Inthe informal sector, gemstones-primarily sapphires-andplacer gold are exploited inrural areas by artisanalminers. Quarry materials are exploitedby small scale miners along the Mekongnear urbanandperi-urbanareas. 39 79. Lao PDR has excellent geological potential to develop mineral resources further. Based on geological data available, it i s estimated that Lao PDR is comparable to other countries inthe region such as the Philippines, Papua New Guinea, and Indonesia, all o f which have large mining industries. Geological survey work and other investigations carried out to date have identified several potential mineral deposits which could be developed. These include the PhuBia gold-copper deposit; the Xanakham gold-copper deposit; sapphire inHouei Say area; Tonghap potashdeposit; and gold anomalies incentral Lao PDR.'O 80. Developing this mineral potential will require attracting domestic and international capital as the amount of investment required to develop mineral deposits and the risks of failure are very large. A United Nations-World Bank survey o f major international mining companies identified criteria used in evaluating potential investment in emerging economies. These criteria are (i)good geological potential; (ii) legislation mining which provides clear mining rights, title, and ownership and control o f operations; (iii) tax regime which provides competitive fiscal conditions; (iv) political stability and appropriate govemance; (v) availability o f infrastructure. Recent experience has indicated that governments that address these issues can and do attract significant amounts o f inve~tment.~'InArgentina, for example, exploration value rose dramatically and the value o f minerals production increased five-fold duringthe 1992-2001period. 81. Lao PDR will need to address remaining legal, regulatory, and tax weaknesses in the mining policy regime to attract such investment and develop the sector. There are several weaknesses inthe legal and regulatory regime. First, there i s need for greater clarity on which legislation-the 1997 Mining Law or MiningExploration and Production Agreements (MEPA)apply to new mininginvestments under what circumstances. The 1997 MiningLaw is viewed as less attractive by any private companies and little, if any, new exploration has taken place under its terms as MEPAs can still be negotiated. Second, the same rules should apply to all investors in a transparent manner. If MEPAs are used, then the basic terms and conditions should bepublishedand they shall apply to all investors. 82. The 1997 Mining Law contains several weaknesses which could be addressed by applying intemationally accepted principles such as the "first come, first served" principle, finders rights, and clear rules for title issuance. Liquidity o f mining investments should be improved by making explicit the right to transfer, pledge, or mortgage the mining title.72 Security o f tenure for mining titles could be improved by strengthening the obligations o f government to issue a title in the event o f discovery as well as specifying the conditions under which a title may be revoked. Companies will not invest inexploration unless they are granted 'OThe Phu-Biagold-copperdeposit is located90 kmnortheast of Vientianeand is estimatedat 1.02 million oz of gold. Pan Australianmining is currentlyoperatingunder aMEPA concessionagreement andconductingfeasibility studies for bringing the deposit into production.Drilling andpreliminary mappingindicateup to 50 billion tones ofpotashinthe Tonghap potashdeposit. Across the borderin Thailandundervery similar geologicalconditionsis the UdonThani Potash projectwith estimated earningsof USD4.3 billion over an estimatedmine life of 22 years. 71These includeArgentina, Chile, Mexico, Peru, Ghana, Mali, andTanzania. Measuresundertakentypically includeupdating mineralpolicy; rewriting mining legislation;updatingthe taxationregime; improving govemanceandreinforcing government supervisory institutions; andimproving the scientific database. 72Inreality,smallcompaniesexplore, but largercompaniesdevelopandoperatemines. Companieswhich beginexploration maynot have the financialresourcesto continueexplorationor developthe deposit. While the 1997Mining Law statesthe principle of transferability,the requiredproceduresare vague. Miningrights acquiredunder a MEPA may only be transferreduponconsent ofthe Government. 40 exclusive rights within the title area, and most countries provide for an automatic right to develop the deposit in the event o f commercial discovery. In Lao PDR, neither the Mining Law nor the MEPAprovide for such automatic 83. Institutional arrangements and the governance and regulatory regime around mining need to be clarified. There is a need to mandate a single government institution to deal with investors, a mining title registry and cadastre office to grant mining titles, and an inspectorate to monitor commitments and financial obligations o f the title holder. At present, the division o f responsibilities between the Ministry o f Industry and Handicrafts (MIH) and the Department for Foreign Cooperation (DDFI) within the Committee for Planning and Cooperation (CPC) i s not clearly laid out inpublishedadministrative guidelines. This results in discretionary interpretation by the Direction o f Geology and Mines (DGM) and uncertainty for investors. State ownership o f enterprises operating in the sector means that the role o f the government as shareholder/operator and as regulator needs to be clarified and vested in separate legal entities. 84. Effective monitoring and oversight are required. Environmental permits and monitoring needs to be strengthened. Environmental regulations and guidelines specific to the mining sector need to be completed and synchronized with the main environmental laws. Weak capacity in provincial authorities to effectively monitor mine operations i s a key issue that must be addressed given their significant role in mining oversight. Royalties from the Sepon gold and copper mine, for example, are paid directly to Savannakhet province, and a trust fund for socio-economic enhancement o f the area isjointly managed by district authorities and the Sepon mine. 85. The relationship between the central and local govemment inthe ownership and control o f mineral resources and mining revenues needs to be clarified. International experience, notably inAfrica, has indicated problems fiom conferring ownership o f mineral rights on sub- national entities (e.g., provinces or state enterprises) or private concessionaire^.^^ Ultimately, whether mineral resources accrue in the first instance to the centre or to the province differ among countries. What is important i s that there be a transparent mechanism for revenue assignment and a match between revenue and expenditure responsibilities as part o f a clear intergovernmental framework which i s also enforcedinpractice. 73Instead, they provide that the govemment will "consider the application for a mining license" subject to a number o f factors. 74Because mineral resources are an exhaustible national asset, a significant share o f the benefits from their exploitation should accrue to the nation for reinvestment inother productive activity. This is particularly true ifthe Govemment has invested large public resources indeveloping the infrastructure required by mining, or ifthe mining products are high value-low bulk items (such as gold) which can be readily transported or smuggled out o f the country. 41 86. The mining tax regime should provide a transparent and level playing field for all investors, be stable over time, and be internationally competitive. The Lao PDR tax regime for mininghas several weaknesses inthis regard. Because the Foreign Investment Law overrides the general taxation laws, the tax regime for a miningproject-notably profit taxes and royalties- i s negotiated on a case-by-case basis. The principal product o f these negotiations i s the MEPA, which contains different tax treatment for different taxpayers, without reference to published, standard guidelines or tax rules. Furthermore, because MEPAs are considered confidential, transparency of tax rules i s not achieved. Establishing and publishinga standard template for the taxation terms o fMEPAs would improve transparency. 87. Ultimately, as with the other natural resource sectors, the development contribution o f the sector will depend largely upon how the government revenues that are generated (i.e. royalties, taxes, dividends) are managed and used. The effectiveness and efficiency o f their use will determine the final impact o f mining sector's expansion on economic growth, poverty reduction and social outcomes. Key reform areas inmining 1. Improve governance arrangements in sector by (i) avoiding conflicts o f interest inrole of the State as "owner/ operator" or state owned mining enterprises and the role of the State as "regulator" o f the sector; (ii) ensuring clear mandates for government oversight institutions to avoid potential delays which may be caused by unnecessary division o f labor between MIHDGMand the DDFI; and (iii) completing the mining regulations and sector specific environmental regulations. ~ 2. Improve enabling environment for private sector participation by (i) establishing a track record o f consistent and fair dealing with investors (both local and foreign) through appropriate regulations, guidelines and procedures which will facilitate the expeditious issuance of mining titles; (ii) putting in place an internationally competitive taxation package which encourages investment while, at the same time, maximizes govemment taxation revenues; and (iii) establishing a standard set o f investment conditions or work contracts for the sector which would apply to a11investors and avoid the need for "case-by-case" negotiations. I 3. Build capacity bytraining and logistical support for the Direction de la Geologie et des Mines (DGM). 42 Table 6: Summary sectoralcharacteristics Sector Characteristics Messages Agriculture Subsistencerice farming mainly; 1 Critical for povertyreduction Limitednon-rice crops for sale; Growthwill come from reformscreatingproper enabling environment for increaseddiversification, commercialization productivitygrowth 1 Developoff farm opportunitiesto diversify riskandfacilitate commercialization 1 Manage environmentalrisks Manufacturing Very small firms with low 1 Critical source o foff-farm opportunities andgrowth and services skilled workers andlow 1 Create enablingenvironmentby establishinglevel, predictable productivity playingfield for businessby "letting go", creatingandkey market, human, and physicalinfrastructure 1 Facilitateregionalandcross border opportunities Naturalresources Hydro, developingwith high 1 Developmentwill require improvedgovernanceframework, potential especially givenneedto attract foreign participationgiven high financingrequirements 1 Development contributionwill stem from electrification benefits anduse ofgovernmentrevenuesgenerated Mining Largelyundevelopedwith high 1 Developmentwill requireimprovedgovernanceframework, - potential especially givenneedto attract foreign participationgiven high financingrequirements Developmentcontributionwill dependuponuse of government revenuesgenerated Forestry Overexploited,depleting 1 Improveforestrymanagementpracticesby increasing resource communityparticipation 1 Development contributionwill stemfrom provision of off- farm opportunitiesanduse ofgovernment revenuesgenerated 2.2 Sources of Growth 88. Taken together, what does all this imply for growth, poverty reduction, and social outcomes? Inthis section, we construct a simple macroeconomic model of Lao PDRto provide indicative, order of magnitude for the growth figures. They are not meant to provide detailed and precise projections. Given the simplicity o f the model, the figures are intended to be illustrative rather than predictive. Inthe next chapter, we provide some indicative simulations o f the impact o f such growth on poverty, social outcomes, and public revenues. The structure o fthe model and the basic assumptions employed are described inBox 8. 43 I Box 8: A simple macroeconomic model for Lao PDR I The model combines top-down estimates o f sectoral growth rates for agriculture, industry and services with bottom-up estimates for the GDP and public revenue contribution o f selected natural resource sectors including hydropower, mining and forestry under different scenarios based o n estimated project Estimates for the natural resource sectors are bottomup inthat they are based on estimates o f value added andrevenue from project- specific flows. For example, estimates are built up from flows for EdL, THPC, HHPC, and NT2 for hydropower and from the Sepon gold-copper mine for mining. T o avoid double counting with the natural resource sector figures, the figures for the industry sector exclude mining and electricity, gas and water while the agriculture figure excludes forestry. Bottom-up estimates o f revenue contributions from natural resources are combined with top-down revenue estimates calculated from estimated tax and non-tax elasticities to generate an overall envelope for expenditures. Expenditure shares (based on 2001/02 actual outturns published in the April 2003 Official Gazette) are then used to generate sectoral expenditures, which are benchmarked to projected population figures. Because the bottom-up estimates for the value added or GDP contribution o f the natural resource sectors are not calculated in the same way as the National Statistics Center (NSC), adding them together with the top-down estimates for the contribution o f the other sectors from the NSC leads to differences inthe levels for baseline GDP, e.g., when comparing the resulting estimate for the level o f GDP in 2003 with the historical 2002 GDP figure calculated by NSC. However, the analyses conducted inthe report are driven by changes and growth rates, rather than the levels themselves. For example, the revenue and expenditure estimates-as well as the poverty and social outcome impact-are driven by changes inGDP and not their level. ,Several scenarios are run. Different sectoral growth rates are assumed for agriculture, industry, and services, and different scenarios run for the natural resource sectors (e.g., with N T 2 in hydropower, additional gold-copper mine, improved forestry management). Other assumptions include (i) tax and non-tax revenue elasticity are estimated at 1.12 and 1.13, respectively; (ii) grants, net foreign and net domestic financing are assumed to be 2%, 3% and 0% o f GDP in total, respectively; (iii) deflator declines to 5% after 2005; and (iv) exchange rate GDP depreciates at 4%-5% assuming U S inflation o f 1% and relative purchasing power parity (PPP). 89. Three growth scenariodcases are run. For the natural resource sectors, the contributions are estimated on a "bottom up" basis. Inhydropower, the base case assumes no new investments in hydropower, but incorporates contributions from EdL along with revenue from the existing IPPs: THPC and HHPC. Inmining, the base case includes estimates from the ongoing Sepon gold-copper mine. In forestry, the base case reflects declining production undera "business as usual" scenario as modeledinthe 2001 World Bank Forestry Sector Study. Underhe scenarios modeled inthe forestry sector study, the sector declines under "business as usual" with production declining on average by -0.9%. By comparison, the actual average growth rate inforestry between 1995-2000 was -2.6%. 90. The base case growth rate assumptions for agriculture excluding forestry (3%), for industry excluding natural resources (8%), and for services (6%) are based on recent subsector growth rates (averages for 2001 and 2002) adjusted for factors described in the sectoral analyses, and rounded to the nearest percentage point. In agriculture, for example, paddy i s assumed to increase more slowly than in the past because o f expected diversification towards other crops (which is assumed to increase more rapidly) and because the past paddy increase 75It shouldnot be noted that electricity, gas and water is overwhelmingly dominatedby electricity, over 97% of which consists of hydropower (in terms o f generation capacity). Water, for example, is estimatedto contribute only about 0.1% o f GDP. 44 reflected increases in irrigated rice due to new irrigation investments. In manufacturing, garments-which grew by nearly 18%-are now assumed to decrease going forward due to increased competition from China's accession to the WTO. The decline inwood processing (-1.3%) i s also expected to accelerate due to subsector problems identified in the sectoral analysis. In services, growth in wholesale and retail trade i s assumed to decelerate from the over 8% recorded recently, due to questions about sustainability. 91. Because the model is not intended to be predictive, scenarios are run to illustrate their implications. The base case-plus scenario uses the base case parameters but assumes Nam Theun 2 (NT 2) i s constructed in hydropower. The high case assumes higher growth rates based on historical 1995-2002 average growth rates for agriculture (4%); industry (10%); and services (7%). For the natural resource sectors, the highcase assumes that Nam Theun2 (NT2) i s constructed along with 13 smaller hydro projects under the EdL power expansion plan; improved village forestry management (VFM) in forestry; and the exploitation o f an additional gold-copper mine inthe mining sector. 92. The results of the model with point-estimates are presented below inTable 7. Aggregate real GDP grows around 4-5% inthe base case; around 5-6% inthe base plus case; and between 6-8% under the highcase. Evenexcluding miningand electricity, industry makes an increasing contribution to growth. Given its large share in GDP, agriculture continues to make an important but declining contribution to growth, and i s eventually surpassed by services interms o f growth contribution. Table 7: Average contributionto GDP growth by sector (percent) Source: Model simulations 45 Figure 14: GDP flows from naturalresource sectors (billions of kip) 4000 3500 3000 2500 2000 1500 1000 500 0 +- Forestry -+- Mining Hydropower 1 * 93. The direct contribution of natural resources (hydropower, mining, and forestry) to GDP growth is relatively small. Under all scenarios, the contribution to growth from the natural resource sectors reaches a peak during 2006-2010 (e.g., less than 2% under the base plus case scenario) and then drops off. Even under the high case, the contribution to growth from natural resources increases from around 0.4% in 2004-05 to 2.3% during 2006-10-which i s the same as the contribution from the industry sector-then drops o f f to 1.5%, which i s less than the contribution to growth from agriculture (1.6%). To a large degree, this limited growth contribution reflects the nature o f the natural resource sub-sectors. When developed, the sectors provide a steady and significant flow contribution to GDP which raises the level o f GDP, but-absent productivity or price improvements-does not directly contribute continuously to increased GDP growth over time. 94. It should also be emphasized that these figures reflect only the direct contribution o f the sectors based on estimates o f sectoral value-added based on project-level data (in the case o f hydropower and mining) and do not take into account the additional impact on GDP growth from multiplier effects due to natural resource development (e.g., from additional rural electrification) or from growth-effects o f higher public spending or additional private investments that these natural resource projects generate. As such, the figures provided should be considered as a roughbaseline. 95. Lao PDR should take steps to increase the, direct growth contribution of the natural resource sectors. As noted above, forestry can provide significant opportunities for off-farm employment and income. Hydropower can provide electrification benefits, and miningthe possibility o f backward and forward linkages. For more enclave type sectors such as mining and hydropower, increasing the opportunities for the involvement o f local construction firms and employment during the construction stages o f projects will be important in increasing the direct growth contribution of the sector. Ultimately, however, the 46 development contribution o f natural resources will come in large part from how the public revenues generated are used. Without better revenue and expenditure management, the development contribution o f the natural resource sectors will be limited. 96. Growth in other sectors-agriculture, manufacturing, and services-will thus continue to be needed. The lack o f diversification and high concentration o f economic activity in Lao PDR also indicate that growth inthe near term will continue to depend in large part upon the performance of a few key subsectors: rice (nearly 40% o f agricultural value added); food and beverages (around 80% o f manufacturing value added); and wholesale and retail trade (nearly 40% o f services).76 Collectively, these three key subsectors alone accounted for around 42% o f GDP. Significant reforms such as those identified above in the sectoral analyses, however, will be needed to allow these sectors to continue to generate the growth needed by Lao PDR. 97. The 2002 Lao PDR Industrialization and Modernization Strategy (2001-2020) identifies five priority industry sectors: electricity; agriculture and forestry processing; tourism; mining; and construction materials. While such industries appear to have good prospects, it is far more important for Lao PDR to improve the overall enabling environment and governance for private investments and exports rather than to focus on "picking winners" interms o f sectors. Without such improvements, it will be difficult for Lao PDR to develop the diversified economic base that it needs going forward. 76For consistency, agricultural value added excludes forestry which is included innatural resources. 47 Chapter 3 -ImprovingPovertyand Social Outcomes Growth alone is not enough. Complementary improvements in public financial management and sewice delivery are also required to improve adequatelypoverty and social outcomes in 98. The previous chapter discussed the sources o f growth for Lao PDR. Growth, o f course, i s not an end in itself. In this chapter, we explore the demand and supply side sources o f improved poverty and social outcomes in Lao PDR. Growth can contribute to improved poverty and social outcomes through at least two channels: (i) directly increasinghousehold by incomes and the demand for health and education; and (ii) by increasing public revenues and expenditures. Section 3.1 focuses on the first channel and uses the indicative sectoral growth rates derived in the preceding chapter to assess their direct (demand-side) impact on poverty reduction and improved social outcomes. This corresponds to the top halfo fthe diagram inour organizing framework. Section 3.2 then focuses on the second channel and explores how growth can contribute to increased revenues and expenditures on critical public (supply-side) interventions to improve poverty and social outcomes. This corresponds to the bottom half o f the diagram. U 3. 1Demand-side: Growth, poverty, and social outcomes 3.1.1 Poverty reduction 99. In this section, we use various growth rates to estimate the impact on the entire distribution o f household income in Lao PDR, from which we derive the impact on poverty reduction ~. We assess the impact o f macroeconomic growth on poverty reduction under several different scenarios using POVSTAT, a poverty simulation tool developed by the East Asia Region o fthe World Bank." 77 POVSTAT is an Excel-based pro am designed to simulate overty measures under altemative growth scenarios and forecast overty measures over,a&ure projection horizon. ioverty ro`ections are enerated using country-s ecific househob survey data and projection parameters for that country. &e kasic methoiology underlyin POVSfAT i s that head. &e Datt and Walker$5002),at POVSTAT 2.1: A Poverty Projection Toolkit.Projection Toolkit. the rate and sectoral pattem ofow growth determine how poverty measures evolve over time. Householfper ca ita consum tion is assumedto the same rate as that of per capita output inthe sector o f employment o ft\e household 48 100. The results are reported in Table 8. Under the base and base case plus scenarios, aggregate GDP i s expected to grow by an average of 5%-6% during 2005-2010, before tapering off. Under these macroeconomic scenarios shown above, we estimate that poverty should be reduced to around 26% in 2005, 24% in 2010, and 23% in 2015. The difference in poverty impact between the base and base-plus scenarios does not capture the impact of the effective use of government revenue generated by natural resource projects that are part o f base plus case. This finding thus highlights the fact that exploiting natural resources will only have a better impact on poverty-reduction ifthe revenue is spent by government in a way that benefits the poor. 101. Underthe NGPESmedium-termmacroeconomic framework, GDP is projected to grow at around 7% on average over the plan period. With population growth o f around 2.5%, this implies an average per capita growth rate o f 4.5%. Industry also expands the most and agriculture the least. We project that the NGPES macroeconomic environment would reduce poverty to just over 22% by 2006 and 16% by 2010. While this would be about 3% above the NGPES target, it would nevertheless represent remarkable progress with a decline in the number o f poor by above half a million compared to 1997/98, despite an increase inpopulation o f about 630,000. By comparison, the MDG-Goal is to halve poverty over 25 years. Taking the 46% incidence o f poverty in 1992/3 as a benchmark, this would imply a poverty headcount o f 23% by 2017/8. Ifthe projections are realized, however, this target would have beenreached by 2006 inonly 8 years instead of 25. Table 8: Poverty, Growth and Inequality Scenarios Actual Base Base Plus High NPEP Poverty Gap Poverty Seventy 1372 1108 Coefficient) 363 406 42.6 31.7 40.3 Memorandumitem Real Per Capita GDP Growth 4.3 3.2 4 6 6 4 4 6 4.6 4 9 Note: *Povertyand inequalitypredicatedbasedon actualmacroeconomicgowth rates. 102. Additional scenarios are presented in Table 9 which highlight three risks that could threaten such progress. First, increases in inequality can limit the reduction in poverty. Second, lifting people out of poverty i s more difficult in poor areas than in other parts of the country. Poverty estimates for the 46 poorest districts are provided in the lower half o f (see Table 9). More than one in two persons in these districts remain poor, compared to a target level o f one in three. With population growth, the number o f poor hardly changes. These findings confirm the need for particular attention to be paid on poor rural areas as proposed under the NGPES, but indicate that this needs to be adequately supported by appropriate growth. Third, without agricultural growth, progress in poverty reduction is much more difficult. 49 103. Growth in agriculture is critical to reducing poverty. This i s not surprisinggiven the reliance of most poor households on subsistence agriculture. Ifthe agricultural sector grows at only 2.5%-as it has in3 out of the last 10 years-instead o f4.5%, then the percentage ofthe poor drops by only 1% over the next 3 years, and the number o f poor remains roughly unchanged at 1.6 million. Table 9: Additional Scenarios (1) (2) Actual Slow Agriculture Growth Slow Growth andRisingWithin Sector Inequality Poverty Headcount Poverty Gap Poverty Seventy Number of Poor (`000) Inequality(Giiii Coefficient) Memorandum item: Real Per Capita GDP Growth 4.3 4.2 Note: `Poverty and inequality predicatedbasedon actualmacroeconomicgrowthrates. **The same macroecononiicandpopulation growthrates are assumedfor the poorest47 districts andotherdistricts 104. The source of growth matters as much as the overall rate of growth in reducing poverty. The poverty impact o f an additional percentage point in growth depends critically upon where the growth comes from, and again underlines the overwhelming importance o f agricultural growth. An additional increase inthe GDP growth rate by 1%between2003-2006 relative to the NGPES macroeconomic scenario reduces the poverty headcount only slightly to over 21% by 2006 if the additional growth contribution i s from industry (compared to around 22% under the NGPES macroeconomic scenario), but reduces it to below 18% if the growth contribution i s instead from agriculture (See Figure 15 below and Annex Table 8). Figure 15: EstimatedImpact of additional 1% GDP growth on poverty headcountratio in 2006 by source of growth 21.50% 21.40% 20 00% 15.00% 10 00% 5 00% 0.00% NPEP Macro Industry Services Agriculture Scenario \---' With Additional 1%GDP Growth Contribution Source: POVSTATsimulations 50 3.1.2 Social outcomes 105. As noted above, economic growth contributes to improving social outcomes through at least two channels: (i)raising household incomes and (ii)increasing public revenue and expenditures. Inthis section, we focus on the former. As households get richer, their demand for health and education services increases, improving their social outcomes 3 106. To obtain some very rough indications o fthe direct impact o f growth on different social outcome variables for poor households, we first ask (i) what impact growth has on their household income; then (ii) multiplythis with our estimate o f how social outcomes improve as household incomes increa~e.~'The results are presented in Figure 16 and indicate that, even under the highcase, while growth does have an impact on social outcomes over time, the direct impact remains relatively modest over the medium-term. Even after eight years o f sustained growth in per capita consumption for the bottom quintile of around 2.5% per year, Lao PDR would still be quite far from achieving the MDG target o f universal primary enrollment and completion rates. 107. This result has two important implications. First, growth alone is not enough. Growth needs to benefit the poor to improve Lao PDR's social indicators. Although projected average growth rates for the economy overall are highunder the NGPES scenario (on the order of 6%), the projectedgrowth rates for the bottom quintile are actually quite low. " We explore the link between household income and social indicators more formally using econometric analyses later in the chapter. 79More formally, we use estimates o f the elasticity o f outcomes (it., enrollments, completion rates, and health care utilization) with respect to per capita consumption and combine these with projections o f growth inper capita consumption (PCC) for the bottom quintile obtained fromthe POVSTATsimulations to predict how the probability of enrollment will change for households inthe bottomquintile over time. We obtain the implicit elasticities by converting the probit coefficients into comparable linear probability coefficients, and then using average per capita consumption for each quintile to calculate the elasticity for each quintile "step," e.g., from 42 to Q1. 51 _. 108. Second, constraints other than low incomes need to be addressed and the quality and coverage of public services improved. Given the relatively low impact of higher incomes on social outcomes, the scope for relying solely on demand-side effects from rising incomes to improve social outcomes dramatically over the near term remains limited. While additional resources could help, they would need to be matched by corresponding supply-side improvements in service delivery. Assessing the impact o f increased expenditures on health and education outcomes requires understanding (i) spending translates into outputs (such how as textbooks, teachers, schools and health facilities) and (ii) how such outputs translate into improved health and education outcomes, a process which depends critically upon the quality o f service delivery. At present, there i s inadequate information to address these two critical issues. While some information on the first will be collected as part o fthe expenditure tracking survey now underway, information on the second would require additional work.*O Figure 16: SimulatedImpact of Growth on Social Outcomes for Bottom Quintile (Ql) I 50.0 1 45.0 40.0 35.0 I 30.0 C 2 25.0 a, Q 20.0 15.0 10.0 5.0 Enrollmentrate PrimaryCompleionrate Health utilization(if ill) 0Average rateforQ1 Ed Projectedrate-5years Projectedrate-1Oyears Source : Calculationsfrom LECS 97/98 data and regressions. For consistency, starting outcome measures a- based on LECS suney For education, this can be done inpart by usingtesting and other measures o f student achievement, incombination with detailed measures o f classroom and school inputs. For health, this would require much better information on expenditures and flow of funds inthe sector, as well as a survey o f facilities to collect data on treatment protocols, costs and results. 52 3.2 Supply side: Public interventions and poverty and social outcomes 3.2.1 Growth, revenues and expenditures 109. While growth alone i s not enough to directly improve social outcomes s,uffciently, growth can also contribute to increased public revenues and expenditures. This section explores how growth can contribute to increased revenues and expenditures on critical public (supply-side) interventions to improve poverty and social outcomes. We use the macroeconomic model o f the previous chapter to obtain an overall resource envelope based on bottom-up estimates of revenue contributions from natural resources and top-down revenue estimates calculated from estimated tax and non-tax elasticities 4. Expenditure shares are then usedto generate sectoral expenditures, which are benchmarked to projectedpopulation figures. U Figure 17: Natural ressourcesrevenue Base case plus - Billionsof kip 1,200 1,000 800 600 400 200 I 2004-05 2006-10 2011-15 1 W Forestry 0 Mining Hydropower 1 53 110. While its direct contribution to GDP growth may be limited, natural resource sectors can make a substantial contribution to public revenues. Under the base case plus scenario, for example, average annual natural resource revenues are slated to grow from 2004-05 levels by nearly 50% by 2006-2010 to nearly double by 2011-2015. Not all o f this i s driven by NT2 revenues. Average annual revenue flows from mining-the Sepon gold-copper mine-are expected to grow sharply from negligible levels to nearly 300 billion kip after 2011. As a point o f reference, this is greater than total expenditure on education in 2000/01 (289 billion kip). (Allfigures are real2003 kip.) 111. It must be emphasized that not all these revenues may accrue to the central government. The figures presented in Figure 17 include hydropower revenues (taxes, royalties, dividends and other payments to GOL from EdL, THPC, HHPC, and NT2); forestry revenues (royalties and contribution to GOL); and mining revenues (royalties, taxes and other contributions from Sepon). As noted previously, in the case o f Sepon, the mining royalties are paid at the provincial level. Comments from counterparts indicated that while miningroyalties are paid at the provincial level, in principle they accrue to the central government and are subsequently shared with provinces. Given past problems in remitting revenues from provincial to central governments, at the very minimum, greater transparency and accountability in central-local sharing o f such royalty revenues i s needed. 112. Managing these natural resource flows will be a challenge and will require substantial improvements in governance as well as monetary and fiscal discipline. Successful large scale natural resource investments can lead to substantial balance o f payments surpluses and a build up o f liquiditywhich ifpoorly managed can result inovervalued real exchange rates and a loss o f competitiveness (Le., "Dutch disease"). Onthe fiscal side, Lao PDRwill need to be ware o f "fiscal ratcheting," increasing expenditures substantially inresponse to higher revenues without adequate provision for retrenchment when revenues decrease. Lao PDR has both renewable as well as non-renewable natural resources. While flows from renewable resources can be considered as income, flows from non-renewable resources should be considered as liquidating assets. The general rule o f thumb is that liquidating assets should be offset by investments in other assets, e.g. in developing human resources.81 However, because Lao PDR already has a serious imbalance between recurrent and investment expenditures, it should avoid any presumption that resources should only be for investment purposes unless adequate provision for recurrent expenditures i s made. The implications and lessons for Lao PDR on avoiding the "natural resource curse" are discussed further inthe last chapter. 113. However, relying on natural resource revenues alone will not be adequate, and Lao PDR will have to take additional revenue measures and increase tax effort. As indicated in Figure 18, even under the base case plus scenario where N T 2 i s constructed, the ratio o f total revenue:GDP does not appear to increase that dramatically, although the share o f natural resource revenues intotal revenues does increase.82 As discussed further in [Box - Avoiding natural resource curse], for example, Botswana set up statutory development funds for this purpose. Even under high case scenario, the revenue: GDP ratio increases to only around 16% for 2011-2015. 54 Figure 18: Naturalresource and total revenues-Base case plus As share oftotal revenues and GDP 1- 2004-05 2006-10 2011-15 Natl res: total rev. (Base +) +Total revGDP (Base +) 1 114. What does all this mean for the population in terms o f actual expenditures which can contribute towards improved outcomes? We focus on recurrent expenditures in health and education which were identified in the 2002 PER as seriously underfunded and critical to improvingoutcomes. Based on existing expenditure shares, Figure 19 estimates the implied recurrent expenditures in primary education and health benchmarked to the approximate relevant population group under the base case plus ~cenario.~'Inhealth, per capita.expenditures hardly increase until2015. Primary recurrent expenditures per child (population aged 5-14) are estimated to increase from around 50,000 kip in 2001 to around 80,000 and 100,000 kip by 2005 and 2010, re~pectively.~~ education, recent increases in the share o f expenditures In improves the picture somewhat given the very low base in2001. Nevertheless, given the very low qualityo f services beingdelivered at present, such increases are not likely to be sufficient. 115. Lao PDR will need to tackle this problem directly by increasing the share of recurrent expenditures in the social sectors. Table 10 presents the impact of different scenarios on recurrent primary expenditures per child. While necessary, relying upon higher growth, natural resource revenues, and even improved tax effort to deliver the required increases inrecurrent expenditures is not sufficient. These have to done in conjunction with a direct increase inthe share o f budgetary resources going to recurrent expenditures. Recent data on 2001/02 actual budget outturns indicate that Ministries with large expenditure shares include Finance (19.5%); Ministry o f Communication, Transport, Post and Construction (16%); Agriculture (11%); and Defense (9%), compared to sectoral expenditures on education and health o f 11%and 5%, respectively. 85 ''Expendituredata classifiedby function are limited inLao PDR. The share of educationandhealthintotal expenditureswas 7.4% and3.7% in2000/01, respectively, andwas increasedonly recentlyto 10.7% and 5% inthe 2001102 budget outtums. Basedon the PER, the share of recurrent expendituresineducationwas 52%in2000, ofwhich 48% was inprimary education. Inhealth, the share of recurrent expenditures has fallen to 40% or less inrecent years. 84As noted inthe discussion o fsocial outcomes, childreninLao PDRbeginand finish primary school late. The 5-14 population subgroup i s therefore reasonably close to the relevant one for primary education. 8sItremains to be seenwhether recent increasesineducation expenditure willbe sustained. While actual education expenditures were 10.7% in2001102, the budgetedshare for 2001102 envisages a decrease inshare to 10%. 55 Table 10: Estimatedrecurrentexpendituresinprimary education per child Base case 51 54 65 Base case plus 51 78 103 Base case plus 20% increase intax effort (elasticity) + 51 81 114 Source: Model simulations 3.2.2 Determinantsof povertyand social outcomesin Lao PDR 116. What kind o f expenditures and interventions will have the most significant impact on improving poverty and social outcomes I? Identifying such interventions requires first ~ understanding the determinants o f poverty and social outcomes inLao PDR, which i s the topic of this section. Determinantsof povertyoutcomes 117. Designing an appropriate poverty reduction strategy requires understandingthe determinants of poverty. We follow the standard approach and start with a multivariate regression analysis o f the determinants o f living standardsag6The regression results indicate the importance of demography, household head characteristics such as education and occupation, ownership o f productive assets, and village infrastructure in contributing to higher per capita consumption. (The results are reported in Annex Table 1). We then use these regression results to simulate the impact o f changes in these explanatory variables on consumption and poverty. 87 86 *'Thissimulationsare approachfollows Chaudhuri(2000), Datt andJolliffe (2001); IFPRI (1998) and Ravallion (1996) The done inthree steps. First the estimatedparametersfrom the regressionresultsare usedto generate predictedconsumption. Second, basedon predictedconsumption, we derivethe probability of ahouseholdto bepoor. Third, we comparethe predictedprobability ofbeingpoor under observedpopulationcharacteristics (our reference point) with the one obtainedwith key determinants of living standards change in value. The difference is the impact onpoverty 56 Figure 19: Recurrent expenditureinprimary educationper child (age group 5-14) and recurrent expenditurein healthper capita-Base case plus Thousandkip 8 120 - I05 - 103 $, .- . . ... 90 5 75 .~- -- ... . .... 2m - 60 . 4 s . v 3 0 - $ a I S -- - 5 ' - I Prmaryrecurrent educationExpenditureper child age group 5-14 years 0 Recurre5bjh Expenditurep "iw 2010 118. These simulations have several advantages over just regression results when assessing the determinants o f poverty. While regressions focus on the statistical significance o f the variables, the simulations indicate their economic significance. The overall impact o f a change ina (policy) variable onpoverty outcomes derives from (i) size ofthe relationship betweenthe the variable and consumption and poverty (the regression coefficient); (ii) the size o f the change in the variable, and (iii) the share o f the population affected by the change. Certain variables may have a statistically significant effect on increasing per capita consumption, but have a negligible impact on aggregate poverty outcomes because the coefficient or the population affected may be small. 119. Improving education, village infrastructure, and off-farm employment opportunities all have a significant impact on national poverty reduction (See Figure 20). We consider thirteen simulations over four groups o f variables: demography, education, agriculture and economy, and village infrastructure. Results are estimated for both the total and "affected" population only, i.e. only those households affected by the variable being changed. For each, results are separately estimated for the impact nationwide, within the 47 poorest districts, and within other districts. 57 Figure 20: Impact on poverty from selected scenarios (percent change in poverty) -12 Increase Provide all Provide all Retrain farmers to education o f villages with villages with being self- household head piped water electricity employed by 1year 120. The results are reported in Table ll.@ in the table refer to percentage The figures changes o f both real per-capita consumption (PCC) and poverty headcount (POV) from a baseline o f 100. Under simulation 6 (retrain farmers to be self-employed), for example, the nationwide effect on the entire population is to reduce the poverty headcount by 9 percent (from a baseline o f 100 down to 91), and increase per capita consumption (PCC) by 5 percent (up to 105). In the poorest 47 districts, poverty is reduced by 10 percent and per capita consumption increased by 7 percent. Since not everyone i s a farmer, not everyone will be affected by this change. The affectedpopulation columns therefore estimate the impact only on those affected by this change. Among the affected population, the nationwide effect i s to reduce poverty by 13 percent (down to 87) and increase consumption by 9 percent (up to 109). 121. Demographics. Household size, composition, and gender o f the head matter for poverty. Reducing householdsize by one for all households with more than one member lowers poverty by 5 percent n a t i o n ~ i d e .Changing household composition by replacing one child up ~ ~ to age 6 by one prime-aged adult reduces poverty by about 9 percent. Finally, female headed households have 14 percent higher poverty. Overall, families in the poorest districts are less affected by changes in household size and composition, and gender o f the head, than those in other districts. While demographic characteristics evolve gradually over generations, this information can still be used for targeting public assistance or investment programs. 88Several important caveats are worth noting. First, the simulations are conducted holding everything else constant and ignore general equilibriumeffects. This may be appropriate for incremental changes, but is less defensible for large policy changes. Second, the impact o f the changes i s likely to differ across households withinas well as across regions. Third, the timing o f the impact o f the interventions will differ. Fourth, the simulations focus on the benefits interms o f poverty reduction o f different interventions, but ignore cost differences across various interventions. 89Less than 0.1 percent o f the population live in single-member households. 58 122. Education. Building human capital of household heads and spouses leads to lower poverty. This is confirmed in Simulations 4 and 5, where we look at the impact o f increasing the education o f all heads, and all spouses, by one year o f schooling. This is a relatively large reform as it affects all households, but it i s only a modest improvement in educational standards. Poverty drops by about 3 percent nationwide. Further, remarkably, the reduction in poverty from the increase in the years o f education o f heads and spouses is similar for the affected pop~lation.~~Finally, we find a slightly higher effect o f head's education outside the poorest districts. One possible explanation i s that returns to education from are higher in more urban areas. 123. Agriculture and economy. Off-farm activity, irrigation, productive assets and livestock are three main exits from rural poverty, as shown in Simulations 6 to 9. Again, we illustrate the impact o f large reforms, so our main focus is on the direction rather than the point estimate o f the changes. Switching heads from being subsistence fanners, accounting for about three fifth of the population, to self-employed triggers a drop in poverty o f 9 percent (Simulation 6). For the affected population, the benefits are somewhat higher in the non- poorest areas, where activities like trading are more profitable. Box 9 provides an example o f how off-farm opportunities can raise household incomes in significant ways. Irrigating all dryland reduces poverty nationwide by around 3 percent. Again, among the affected population, households outside the poorest districts reap greater benefits from these reforms. 124. Village infrastructure. Sanitation and electricity are important to improve living standards. Providing piped water and electricity to all villages lowers poverty by 8 percent among the newly covered families and by 5 percent and 7 percent nationwide, respectively. With the exception o f electricity, the impact on poverty is larger inthe poorest districts that in other districts among affected households. Determinants of social outcomes 125. To analyze the main determinants o f the poor social outcomes observed, we also use multivariate statistical techniques. The analysis uses survey data from the 1997/98 Lao Expenditure and Consumption Survey I1(LECS 11) and the 2000 Multiple Indicator Cluster Survey (MICS). Given data constraints, the analysis focuses on (i) primary enrollment and completion rates ineducation; and (ii) health care utilization rates. 90However, the spouse education coefficients are less significant than the head ones. 59 Box 9: Off-farmopportunitiesinLao PDR:The case of sericulture "The Lao Economic Accelerated Project for the Silk Sector (LEASS) o fthe Consortium Lao [a NGO] is currently working closely with local people...to increase their incomes by planting new kinds o f mulberry trees and teaching thembetter ways o fraising silkworms to produce silk. "Mr.Mankeo, headoftheproject inBanTha, Xieng Khuongprovince, saidthat since the project hadstartedithad created morejobs and incomes for local farmers.. .Theroads inthis area were not good, but this should not hinder the transportation o f the silk to cities because silk was so light. 'The product may be light, but the price is over 100,000 kipper kg.,' addedMankeo... "Mrs. Khammone Vilaphane, 44, is head of the project unit in Sop Hau village, Huaphanprovince.. .She and her family had been involved in low-level silk production for some time, but have now taken the Consortium's advice and improved their methods. Mrs. Khammone now earns more money from her silk production than her rice fields.. . 'Ihavesixmembersinmyfamily,sheexplained. 'BeforeIusedtofarmricetwiceayear...andthatprovidedus with enoughto feed the family, but didnot let us save money. So, at the same time, Iusedto plant mulberrytrees to raise silkworms so as to produce silk in the old way...Ihad 1,200 m2 planted with mulberrytrees, but Ionly used to produce 4-5 grams per month...but after Iwas trained by the Consortium project Icould produce a minimumof one kg. ofsilkpermonth. AndIearn 120,000 kipperkg.' "Mrs.Khammone not only produces her own silk, she also weaves it into garments. 'My daughter andIweave our s i l k to earn more money. Ican make 10 scarves with one k g o f silk and Imake a minimum o f 25,000 kip- depending on quality and size-for each scarf and Ican make two a day,' she said.. .Now they are involved in silk production, Mrs.Khammone's family can earn more than 500,000 kip per month." Source: "Villagers learn to love their silkworms"; "Silk becomes an economic saviour: new ways of sericulture will boos1 farmer's incomes" in Vientiane Times, 16-19May 2003 ~ Educational outcomes 126. To analyze the determinants o f educational outcomes, we estimate regressions o f the relevant educational outcome measure for each child in the sample between ages 6 and 14 (starting and enrollment rates) and for each youth between ages 15 and 19 (completion rates). Explanatory variables include household, child, village and location characteristics. The results are presented in Annex Table 4 and Annex Table 5 which report estimates from the LECS I1 and from the MICS 2000, re~pectively.~~ 9L FollowingFilmer andPritchett(1998a, 1998b), wealthquintiles were derivedfrom an asset index constructedbasedon each household's ownership of consumer durables, characteristics o fhousehold'sdwelling, andownership of assets such as land, animals, etc. 60 Table 11: Simulationsof % Changesin ConsumptionandPoverty Entire Population Affected Population District-Category District-Category Number Description Indicator National Poorest Others National Poorest Others Demography 1 Reduce by one the number of PCC 104 103 104 104 103 104 householdmembers POV 95 97 94 95 97 94 POP 100 100 100 100 100 100 2 Replaceone child aged 0-6 by PCC 105 107 104 105 107 104 one adult aged 15-49 POV 91 91 91 92 92 92 POP 100 100 100 77 84 75 3 Move headshipfrom female to PCC 101 100 101 110 103 111 male POV 99 100 99 86 97 83 POP 100 100 100 5 3 5 Education 4 Increasehead'seducation by one PCC 102 102 102 102 102 102 year POV 97 98 96 97 98 96 POP 100 100 100 100 100 100 5 Increasespouse'seducation by PCC 102 102 101 102 102 101 one year POV 97 97 97 97 97 97 POP 100 100 100 95 97 94 Agriculture & Economy 6 Retrainfarmersto being self- PCC 105 107 104 109 110 109 employed POV 91 90 91 87 88 86 POP 100 100 100 60 77 56 7 Irrigateall dryland PCC 268 1263 103 325 1471 105 POV 97 97 97 96 97 96 POP 100 100 100 79 87 77 Villagehfrastructure 8 Providingall villages with PCC 103 105 103 105 106 105 electricity POV 93 93 93 92 93 91 POP 100 100 100 69 92 64 9 Providing all villages with piped PCC 102 106 102 105 108 104 water access POV 95 92 97 92 90 94 POP 100 100 100 50 75 44 Note: PCC stands for real per-capita consumption,POV for poverty headcount,and POPfor population. 127. Household income, the quality o f village infrastructure and services, and access have a key impact on educational outcomes. As indicated in Figure 21,for example, having textbooks available increases the probability o f completing primary schoolby over 40%. 128. Household characteristics. Household wealth or income is extremely important for educational outcomes. Controlling for other factors, a child from a household inthe top wealth quintile i s nearly 50% more likely to start school, 30% more likely to attend school, and over 80% more likely to complete five years o f school than a child from the bottom quintile. These effects are ordered across quintiles, with each quintile increasing the probability o f completion by at least 10%. Having the head o f household be a farmer has a negative impact on starting and especially on completion rates, although no statistically significant impact on enrollments. Age o f the household head and having a female head both have a positive impact on starting and completion rates. Ethnicity does not appear to have a large independent effect once other characteristics are controlled for. 61 Figure 21: MarginalImpact on probability of completingschool 41-3% In top wealth quintile Village has lower Text books amilable Village accessible in secondary school rainy season *Selected non-demographic factors Source :Regression-analysis on LECS 97/98 Data by Asset quintiles 129. Child characteristics. Gender has an important impact on educational outcomes. Controlling for other factors, girls are 22% less likely to be enrolled in school and 41% less likely to complete school than boys. 130. Village infrastructure. Having a primary school in the village helps to get children into school, while having a secondary school increases the chance that they complete school. A primary school inthe village increases the probability that a child will start and be enrolled in school by over 30%. A lower secondary school in the village increases their probability of completing primary school by 20%, while the presence of an upper secondary school increases the probability by another 14%. This i s consistent with research for other countries which shows that the expectation o f being able to continue on to secondary has a positive impact on completion o f primary." 131. The quality of schooling also matters. The existence o f textbooks has a large and positive impact on all educational outcome measures and increases the probability o f starting and completing school by nearly 30% and over 40%, respectively. The presence o f a permanent teacher also has a positive, but smaller, impact on the probability of completing school. 92See Cuadra and Moreno (2003). 62 132. Access affects educationaloutcomes. Living ina village that is accessible during the rainy season has a large and positive impact and increases the probability o f starting, enrolling, and completing primary school by between 20-35%. An urban location also has a strong positive impact on all three measures o f educational outcomes, and increases the probability o f starting and enrolling inschool by around 20-30%. 133. Provinceeffects. Unexplained, residualprovince-specific effects-which capture all the factors that are specific to each province and are not controlled by the other variables included in the regressions-are large. Vientiane municipality is a positive outlier, while Phongsaly, Luang Namtha, Sekong and Saravane are the largest negative outliers. These province effects could reflect several possibilities, including differences inprovincial income, spending, and the quality and returns to education. However, while rank correlations indicate an association between the magnitude o f these province-specific effects and provincial poverty, there is little correlation with budgetary pend ding.^? This result is consistent with extensive research which shows little correlation between levels o f spending and educational outcomes.94These province effects could also reflect differences inreturns to education across provinces, which affect the value that households place on the investment in education. A recent paper on Vietnam, for example, found that the returns to education were much lower for ethnic and geographically isolatedminorities than for the majority population, resulting inlower demand for education.95 Healthoutcomes 134. We focus on health care utilization rates as a measure o fhealth outcomes usingboththe LECS I1 and MICS 2000. The LECS 97/98 survey data contain information on medical expenses which can be used as a rough proxy for health care ~tilization.~~We proceed as before and use regression analysis to look at such determinants o f per capita health expenditures as household income and characteristics, demographic characteristics, village characteristics, and location. The results are presented inAnnex Table 6. 135. Income andphysicalaccess are key determinantsof healthcare spending. There is a strong link between health care spending and income. All else equal, households from the top two quintiles (net o f medical spending) spend on average between nearly 100-140% more on medical care than do those from the bottom quintile. This relationship probably reflects greater access to better quality services among the non-poor. Physical access is also important. Households in villages that remain accessible during the rainy season report higher medical expenditures than those in non-accessible villages." Medical expenditures seem to go down 93The rankcorrelationsbetweenthe province effects andper capita total and recurrent spendingwas only 0.25 and0.19, respectively. 94Ifpublic spendinghadahighimpactonoutcomes, onewould expectthattheunexplainedprovinceeffects (whichcaptures all effects not controlledfor inregressionswould behighly correlatedwith spending.) 95See, for example, Filmer andPritchett (1998a). 96van de Walle and Gunerwadena (2000) 97Surprisingly, the dummyvariable for urban is negativeand significant,suggestingthat once all other factors are controlled for, rural households actually spendmore on healthcare than do urbanones. It i s not possiblewith these datato say whether this is becauseruralhouseholds are more likely to have spells of illness; more serious illnesses; or whether they are payingmore for services. 63 with distance to the nearest hospital, and the presence in the village o f a pharmacy or trained nurse i s also associated with higher spending.98 136. Occupation, education, and demographics also have an impact. Households headed by farmers show lower levels o fhealth care spending than other groups. Education has a small but positive impact. Demographic characteristics have a small but statistically significant impact. Households with elderly and children have higher per capita expenditures, while those with greater numbers o f women have lower expenditures, reflecting a possible gender bias in allocating health expenditures. Unfortunately, the data from the LECS 97/98 do not allow us to examine health care utilization conditional on being ill.The above findings therefore do not take into account the patterns and burden o f illness among the population. To examine health care utilization conditional on illness, we turn to the MICS 2000 survey.99The MICS 2000 survey results indicate that only about 50% of those suffering episodes o f illness receive any treatment at all. While the main reason cited for not receiving any treatment was because the illness was mild, 17% cited lack o f transport or difficulties in physical access, while another 15% cited the expense o f the treatment.Io0Regression results using the MICS 2000 provide further confirmation of the critical importance o f wealth for utilization, with those in higher quintiles far more likely to seek treatment compared to those inthe bottom quintile (see Figure 22). Figure 22: Increase in probability of seeking treatment relative to bottom quintile 1 3m g 80.0% , 73.8% 70 0% 3 600% -m - 50 0% 400% 0 300% nm g 200% .r Q 100% -- I g 00% L C Second Quintile Top Quintile Professional healthservices B# Any treatment (incl. traditional) Source : Regression analysis on MICS 2000 data 98 By contrast, having a community health worker inthe village does not appear to result inhigher medical expenditures. 99 The downside o f the MICS data is that (i) number o f respondents who report an episode o f illness i s relatively low (only the 959 responses out of 6,600 households); and (ii) data do not contain informationon household income or consumption the and requires the use ofthe assetiwealth index constructed for the educationregressions. looOfthose seeking treatment, 14% went to traditional healers or relatives, 37% went to a pharmacy, with the remainder going to private doctors, health clinics, or hospitals. 64 137. Table 12 summarizes the main findings from the econometric analyses of poverty and social outcomes. Key outcome drivers included household incomes, education, off-farm activities, village infrastructure and services, access, and location. While these results suggest some implications for policy change, care needs to be exercised intranslating these results into specific recommendations for expenditure allocations.Io1 The analyses, for example, identified a clear need to improve the quality and coverage of key public services ineducation, health and infrastructure. However, given the lack o f information regarding the link between public spending and outputs cited earlier, increasing spending alone may not result in improved services. For this reason, it is also vitally important to devote more resources and attention to monitoring outcomes and understanding their links to policy variables such as public expenditures as part o f an overall effort to improve budget management and governance inthe public sector. Table 12: Summary of key determinantsof poverty and socialoutcomes inLao PDR Key drivers Povertyreduction Improvementin educationand health (povertyheadcount) outcomes (primary enrollment andcompletion; health expenditures and utilization) Householdincomes 1 Needgrowthwhich translates 1 Top quintile 80% more likely to into higher incomes for lowest complete primary school than quintile bottomquintile 1 Top quintile 70% more likely to get some-form o f treatment when ill- Education 1 Education o f household head 1 Education o f householdhead has increases incomes and reduces impact on health expenditures poverty by 2% and 3%, respectively Off-farm activities 11 Switching farmers to being self-employed triggers a diop inpoverty of9 percent Quality of village 1 Providing all villages with 1 Presence o f primary and secondary infrastructureand electricity anduiuedwater schools invillage services reduce poverty-bi 7% and 5%, 1 Textbooks respectively 1 Presenceo fDharmacv. trained nurse invillage I, 1 Accessibility o f village during rainy Improvedoutcomes innon- seasonimproves education poor v. poorest districts outcomes 1 Positive urban effect 1 Unexplained province effects 1 Distance to hospital . .. Implicationsfor publicintervention Focus on growth which benefits lower income households Increase quality and coverage of rural infrastructure, especially roads, and health and education services Focus on measuring outcomes and understanding their links to expenditures and other policies '"In terms ofour organizingframework, increasedgrowthprovidesgreaterresourcesfor public spending.Absent growth, however, additionalpublic spendingwould come out of householdincomes(via current or futuretaxes), which reduces householddemandfor social sector services. 65 Chapter 4 -Realizingthe DevelopmentPotentialof Lao PDR An appropriate strategy for improving development outcomes in Lao PDR should include maintaining agricultural performance and off-farm opportunities for poverty reduction; developing manufacturing and services as a long-term growth engine; increasing the development contribution of natural resources; and improving public financial management and service delivery. Implementation of the above strategy, however, will require substantial improvements in the overall enabling environmentfor theprivate sector and governance in the public sector. Although there has been an abundance of strategies and roadmaps, implementation and high level commitment have been in far shorter supply. Without strong commitment to reform and implementation of reforms, Lao PDR's signijkant development potential will continue to go unrealized. L 4.1 Elementsof a developmentstrategy reformroadmapfor Lao PDR 138. We identify the key elements o f an appropriate development strategy for Lao PDR based on a recap o f our key findings from previous chapters using our basic framework. 17 Pavenuesand SupplySide Maintain agricultural pevformance and facilitate offfarm opportunities for poverty reduction. The poverty analyses highlighted the central role o f agricultural growth in reducing poverty, and also identified off-farm opportunities as a key determinant o f higher rural household incomes 1. The analysis o f social outcomes highlighted the importance o f household income as a driver o f improved social indicators 1.Since the overwhelming majority o f lower income households engaged in agriculture, the scope for increasing household incomes will therefore continue to depend critically upon how the agricultural sector performs. 66 Develop manufacturing and services as the long-term growth enginefor Lao PDR. Given the limited growth contribution o f the agriculture and natural resources sectors, manufacturing and services will have to play a central role in generating the long-term growth and employment required by Lao PDR 1.However, the hugepreponderance o ftiny firms and absence of medium-sized firms suggest that there are significant constraints preventing firms from growing, which will place limits on employment and export growth. Improve the development contribution of natural resources. Natural resources are not a "magic bullet" for Lao PDR. Their direct contribution to long-term GDP growth and their contribution to public revenues-while substantial-is not so large as to preclude the need for agricultural and manufacturing growth. Thus additional reforms for developing those sectors and for revenue and expenditure management will be key. Ifnatural resource revenues are poorly managed, then the deployment contribution o f the natural resource sector will be limited. Improve public financial management and service delivery. Because growth alone is not enough to improve social outcomes, supply-side improvements in key public services (education, health, and infrastructure) are also required ~. However, without significantly improved public financial management and service delivery, additional growth and resources are unlikely to translate into adequate improvements in outcomes unless critical accountability relationships are strengthened (See Box 11). 139. Implementing the above strategy will require substantial improvements in the overall enabling environment and in public sector governance. Without policy and institutional reforms to improve the enabling environment for the private sector and trade, agricultural performance will be inadequate to deliver the required improvements in social and poverty outcomes, and manufacturing and services will be unable to provide the requisite growth. It will also be difficult to attract the private investment required to fully develop key natural resources such as mining and hydropower. Without improved governance inthe public sector, however, it will be difficult to sustain this performance or translate it into improved poverty and social outcomes. The reform roadmap for Lao PDRwill therefore need to cover a wide range o f measures to bolster these two critical areas. Details o f these reforms are discussed inthe next two sections and summarized inAnnex Table 1. 4.2 Improvingthe enablingenvironment 140. Without a better enabling environment for the private sector, the potential of key economic sectors-including natural resources-is likely to go unrealized. Improving the enabling environment will require a wide range o f sectoral and cross-sectoral reforms. Sector- specific reforms were discussed in each o f the sectoral analyses, but several themes recur which fall under the general rubric o f putting in place a better enabling environment for the private sector, including letting the markets work and establishing a transparent and level playing field, e.g., by clarifying the legal and regulatory framework and separating out the role o f the state as owner/operator andregulator. 67 Figure 23: Summary overview of elements of reform roadmap Area I Reformelements Improveenablingenvironment Sector-specificreformsin agriculture, manufacturing for privatesector andservices, and naturalresourcesas discussed in sectoral summaries Cross-sectoralreformsinareas identifiedas weak in 2003 CPIA, including trade; human resources; environment; financial sector 0 Publicfinancial andpublic sector managementreforms Improvepublicsector as described inB o x 12. governance Decentralizationreformsas described inBox 13. I Note: Details provided inAnnex Table 1 141. The critical role o f the private sector is well-recognized in the NGPES, which cites private sector development along with macroeconomic stability, regional integration, and natural resource development as key elements o f an enabling environment for growth and development. The private sector, FDI, and trade are cited in the NGPES as prime factors in driving the economy. Based on findings from the First Foreign Investment Forum o f May 2002, specific measures noted in the NGPES to promote private sector development include better information and consultation; legal, regulatory, and administrative streamlining (e.g., business registration and licensing, import and export approvals); macroeconomic reforms; and better coordination and management capacity (e.g., enforcement o f bankruptcy, debt recovery, tax and other laws). 142. The NGPES also states that "as a matter o f principle, the Government o f the Lao PDR has increasingly involved the private sector in areas that until recently have been public responsibilities or areas o f state enterprises." Specific examples cited include the private involvement in the provision o f transport services and infrastructure construction and maintenance. Such greater private sector involvement can help to alleviate the limited financial andmanagerialresources o fthe Government. 143. The NGPES also recognizes that regional integration is "imperative" for Lao PDR, through ASEAN, AFTA, GMS, and other fora. Regional linkages will be important for Lao PDR to overcome drawbacks from its landlocked status and take advantage o f its central location in such a rapidly growing region. This will require Lao PDR to adopt appropriate trade policies to facilitate such linkages, including implementing the CEPT under AFTA. The dominant role o f ASEAN in Lao PDR trade cited earlier highlights the importance o f such measures. 68 144. Equally important i s the need to improve trade facilitation by reducing licensing and other administrative barriers to trade. Concerns about the current account balance and the balance o f payments are warranted, but should be addressed through macroeconomic measures rather than by licensing and other import administrative measures.1a2Forthcoming studies under the Integrated Framework (IF) which will provide technical assistance to improve the trade- related capacity o f Lao PDR should help map out how Lao PDR can more fully take advantage o fregional and global linkages, beyondAFTA. 145. The 2003 CPIA exercise also identified other cross-sectoral policy and institutional areas where Lao PDR was particularly weak, including environmental management, human resources, financial sector management, and public financial management. Inenvironmental management, for example, key priority measures flagged include improving enabling regulations for environmental management; strengthening compliance and enforcement; and improving delivery o f key services such as solid waste collection. By threatening agricultural Performance, environmental degradation poses risks to Lao PDR's progress on improving development outcomes. Swidden agriculture, attended by shortening fallows under population pressure, risk declining soil fertility as well as soil erosion that has downstream impacts on the hydrological behavior o f watersheds and silting o f irrigation infrastructure. Capture fisheries appear to be reaching biological constraints, with greater fishing effort not matched by correspondingincreases inproduction. Inthe human resources area, overwhelming priority i s placed on improving public resource allocation by increasing recurrent expenditures in health and education and incentives for better service delivery. 146. At present, the financial sector in Lao PDR is functioning neither competitively nor efficiently. Suggested reforms in the financial sector include reducing the domination o f the two state-owned banks and APB and improving their transparency, reducing the legal and regulatory gap between domestic and foreign banks, and greater transparency in subsidized credit schemes. 147. Itwill also be important for Lao PDRto manage foreign exchange inflows from natural resource development to avoid the dangers o f the "Dutch disease" and an appreciation o f the real exchange rate which could threaten the performance o f other key sectors such as agriculture. This will require both careful macroeconomic management, structural reforms to reduce costs as well as a well-functioning financial sector. Experience in other resource-rich, low income countries has underlined the important role which domestic financial institutions play in absorbing and recycling increased export earnings within the economy if they function efficiently and competitively. Otherwise, if not properly managed, a significant build-up in liquiditywithinthe financial sector canhave harmful consequences on the economy. '02The NGPES states that the Ministry of Commerce's policy stance is better described as import "administration" rather than "controls," and as price "surveillance" rather than "controls," but ifLao PDRwants to reap fully the benefits from regional integration, it should allow the private sector to develop inresponse to competitive market signals. 69 4.3 Improvingpublic sector governance 148. Successfully implementing any strategy to improve development outcomes will also require Lao PDR to substantially improve its governance. A strategy o f developing natural resources, for example, i s not without risks for Lao PDR and hinges upon establishing appropriate govemance. As noted earlier, recent research indicates that large natural resource revenues have typically had such a deleterious impact on institutions that their overall impact on growth is negative. "Point-source" natural resource revenues-e.g., oil, minerals-are found to have had a particularly negative impact. These findings are o f particular relevance for Lao PDR because (i) o f its prospective natural resource revenues-hydropower, minerals-are much also "point-source;'' and (ii) its governance setting remains weak. Box 10: How to avoidthe "natural resource curse" inLao PDR Governments in countries with non-renewable resources face uniquely difficult challenges. A common characteristic o f resource-led economies is that a large share of the benefits o f development accrue to the state in the form o f mining royalties, taxes, and possibly income distribution and dividends. The key to success i s how well the governments channel these rents into the creation o f productive capacity or the provision o f important social services, including health and education. O f the 15 or so African countries generally regarded as havingthe most propitious physical endowments, only three -Botswana, Namibia, and South Africa - have experienced sustained development and significant annual increases in per capita incomes. While average annual per capita income in 12 o f the best endowed African countries remains below USD 500, annual per capita incomes are now around USD 3,000 inthe three countries. What were the key elements in their success and the main reasons for the economic stagnation which has characterized virtually all o f the other African countries? Most of the elements fall under the rubric of good governance. First, concepts of representative government and public accountability were deeply entrenched. In all cases, the independence o f the judiciary and public service; the role o f the auditor general's office; basic human rights; and freedom o f speechand association were constitutionally enshrined. The rule o f law and legislative processes were respected. Laws affecting taxation, labor relations and the granting o f mineral concessions were transparent and fairly enforced. Second, early measures wereput in place early on to strengthen the domestic institutional capabilities necessay to ensure that thepublic sectorfunctioned efficiently. The included developing core governmental functions such as economic planning and public service management capabilities; budget system; effective tax, geological survey and miningdepartments; central statistical bureaus; and the ability to control, support and regulate natural resource development. A major factor in Botswana's success, for example, was the avoidance o f disputes within the government through the centralization o f responsibility for planning and coordinating national development, with the support o f the President. The Ministryo f Finance and Development Planning was also responsible for leading the negotiationswith external donors as well as the mining companies; coordinating and monitoring the provision o f government infrastructure; and ensuring that the entire nation knew what was going on. Third is the integrity of public polices. Central to those polices has been the commitment o f the national leadership that the benefits from the utilization o f non-renewable national resources should accrue to the entire nation, and for the benefit o f future as well as current generations. There was a strong resolve not to dissipate mineral rents on non-productive activities or for purely short-term political pu~poses.~'~Recognizing that large IO3That resolvewas reflectedindifferent ways ineach of the three countries. InSouthAfrica, emphasiswas placedon the centralizationto provincial governmentsto give local communitiesalarger say and stake. InBotswanaandNamibia,with their smallerpopulationsandweaker sub nationalgovemments, emphasiswas placed onmajor nationalpublic investment programsto unitethe countries andto extendthe benefitsof developmentto all communities. 70 mining developments couldresult indualistic economies, the governments o f all three successful African countries took important steps to increase participation in, and distribute the benefits accruing from mineral development. While private operation o f mines was encouraged on grounds o f efficiency andto ensure access to adequatecapital and modem mining technologies, ownership o f the mineral resources was effectively brought under the control o f the state through negotiation or by legislation.IM In addition to these institutional and sector-specific measures, sound macro-economic policies also contributed importantly to the success of these countries. Central to this success was the recognition that non-renewable resources should be used to finance productive investment, especially inhuman resources. Fiscaldiscipline was also critical. Revenues which could not be utilized for immediate productive purposes were deliberately set aside for the benefit o f future generations. Various different fiscal and monetary mechanisms were used to accomplish that objective. In Botswana, two special statutory funds were established, with the full approval o f Parliament, prior to the advent o f mineral revenues.'O' A statutory Development Fund was also created, to which surplus or unspent revenues were automatically credited each year. The legislation establishing these three funds providedthat they could only be used for designated purposes. Any annual revenues accruing in excess o f the requirements o f the recurrent budget and the requirements o f the other two statutory funds, were credited to the Development Fund. Funds could only be released from the Development Fundfor projects listed in approved five or six year national development plans and specifically expropriated by resolution o f Parliament. On the expenditure side, in Namibia and Botswana mechanisms were introduced to relate budget allocations to established national priorities, and control was exercised to reign in demands for public spending. In all three countries, statutory Public Accounts Committees, served by the offices o f independent Auditors General, were empowered to hold senior officials personally accountable for over expenditure o f budgets or for mismanagement o fpublic assets. Subsidies were restrictedto socially desirable programs, and cost recovery applied. Monetary polices are equally important. Successful large scale natural resource investments tend to generate substantial balance o f payments surpluses which can distort the domestic economy. Overvalued exchange rates reduced export competitiveness and undermined the prospects o f agricultural expansion in several African countries, notably Nigeria, and skewed activity towards the urban elite, underminingthe rural economy. Large scale investments can also generate a rapid build-up o f excess liquidity in the domestic financial system. In all three African countries, steps were taken from an early stage to ensure that excess liquidity in the domestic banking system due to natural resource revenues did not create inflationary pressure or cause interest rates tc become negative inreal terms, discouraging local savings.'06 IO4InBotswana,the tribeswere persuadedto cedetheir mineralrightsto the state. Legislationwas enactedto tax privately ownedmineralrights at marginalrateswhich encouragedtheir efficient privateexploitation, discouragedspeculation, and providedfor the voluntary surrender ofmineralrightsto the state. Although the governmentsofBotswana,Namibia and SouthAfrica have significant equity stakes inmajormines intheir countries, mineralrights havenot beennationalizedin any of them 'OsThe PublicDebt Service Fund and the RevenueStabilizationFundwere earmarkedexclusivelyto service future debt and smoothrevenuefluctuations. IO6InSouthAfrica, the emergence of sizabletreasurybill, bondandequitymarkets ratedamongthe twentylargest capital marketsinthe world - helpedto absorb and recyclemuchof the excess liquidity. InBotswana, the centralbank issued - short-termcertificateswhich servedthe dual purposeof soaking up excess liquidity andproviding an effectiveinstrument for managingdomestic interestrates. Inall three countries, proactivefiscal andmonetary policiescontributedto stable exchange ratesandmoderatedinflation. 71 149. Countries which have successfully translated natural resources into improved development outcomes have managedto develop good governance. Box 10 discusses some o f the lessons for Lao PDR on how to avoid the "natural resource curse'' which have afflicted other resource-rich developing countries. Most o f the elements which characterized the experience of the successful countries fall under the rubric o f good govemance and include clear public accountability; effective public sector institutions; and the integrity o f public . policies, including monetary and fiscal discipline. The challenges which Lao PDR faces are daunting. There is no single policy or set of policies which will permit the country's natural resource endowment to be harnessed efficiently to generate sustained development. Equal weight has to be attached to each o f the measures taken by successful countries ifbroad-based and equitable development i s occur. Fiscal discipline can so easily be negatedby lack o fpolitical commitment. Absent a strong independent central bank, excess liquidity can lead to inflation. Without strenuous efforts to build institutional capacity ingovernment, even the best policies are unlikely to be efficiently implemented.. 150. At present, however, governance in Lao PDR remains weak. As noted earlier, results from a 2002 cross country govemance database indicate significant weaknesses in such areas as voice and accountability; regulatory quality; control o f corruption; rule o f law and government effectiveness relative to other low-income countries. Many o f the factors cited as critical to the experience o f those countries which successfully translated natural resources into improved development outcomes-e.g., high public sector effectiveness, transparency, and accountability-are currently quite weak inLao PDR. Both monetary and fiscal discipline have also been fragile. 151. While there is no single set of policies which can guarantee that resources in Lao PDR-natural or otherwise-get translated into good development outcomes, there are two clear implications for Lao PDR from Box 10. First, failure to improve governance substantially upfront-before resources start to flow-all but guarantees that the resources will not translate into good development outcomes. Second, improving governance will require substantial capacity buildinginthe public sector. 152. Public sector reform will have to play a central part in improving governance in Lao PDR and in delivering upon any reform strategy. Improving governance inthose areas where Lao PDRwas identifiedto be weak-accountability; regulatory quality; corruption control and public sector effectiveness-will be difficult without reforming the public sector including public administration. Improving development outcomes in Lao PDR also requires a public sector that is capable o f designing and implementing appropriate policies and programs; raising adequate resources and allocating them to sensible priorities; and ensuring that those resources get translated into useful outputs that improve social and poverty outcomes. This inturnwould require reasonably well-functioning public financial management and public administration at both the central and local levels. The draft NGPES also identifies public sector management as a critical area in strengthening the overall framework for sustainable growth and development. 72 153. However, widespread weaknesses in the public have been well-documented in recent reports, including the 2003 Lao CPIA, the joint World Bank-ADB-IMF 2002 Public Expenditure Review, and the 2003 IMF Fiscal Affairs Department (FAD) report on Public Expenditure Management Reform. The IMF FAD report, for example, noted that Lao PDR's public expenditure management system was weaker than many low income countries. Particularly noteworthy were the very low ratings on the reliability o f the budget as a guide to actual budget outturns; the lack o f an adequate functional classification o f expenditures; the inability to identify poverty-reducing expenditures; and the absence o f timely audited financial information. 154. Because the gap between what exists and what i s required is wide, making substantial progress in this area i s therefore critical for Lao PDR to have a reasonable expectation that additional resources-be it in the form of development assistance or natural resource revenues- get translated into improved development outcomes. The critical steps and high level outcomes required for the public sector to be able to deliver improved services and development outcomes are discussed below and summarized in Box l2.Ia7 The 2004 World Development Report also identifies the budget, public administration, and decentralization as the three critical institutional structures most likely to influence service delivery.lo8 Achieving the milestones described inBox 12 inthese areas will therefore be important. 155. First and foremost, there i s a clear and overwhelming need to improve accounting and financial control systems as described inthe IMF FADreport and addressed inthe IMF PRGF program1ogwhich should precede the development of a more effective system o f budget formulation and policy prioritization. Otherwise, efforts in the latter may produce a nice design that i s overwhelmedby the realities o f the current system. 156. Policy control o f expenditures would be enhanced by developing linkages between policy programs and financial data. Initially, this can only realistically be done on an input basis. Grouping inputs into key programs that have common goals but still reflect the structure o f ministries would be a useful interim step. The initial focus could be on priority programs with high fiscal risks or poverty impact. A next step could involve developing strategic plans, program goals, and performance indicators. A final step would be to develop outputs based on cost data, though this is likely to be a long way off. Together with enhancing the capability to monitor and report on the effectiveness o f expenditures, efforts should be made to improve service delivery by increasing choice and participation o f the poor in service delivery to enhance their "client power". Given existing weaknesses in "voice" and in the accountability of providers to policy makers, experimenting with different mechanisms to strengthen the accountability o fproviders directly to the poor will be needed to improve service outcomes for the poor (see Box 11). The recent draft instruction to identify administrative units within the accounting nomenclature i s a welcome step and provides a useful basis for improving accountability at the service provider level. I O 7This sectiondraws heavilyuponbackgroundwork by GrahamScott. log See chapter 10("Public sectorunderpinningsof service reform") inWorld Bank, World DevelopmentReport 2004. I O 9The IMFPRGFprogramincludesan actionplan for improving public expendituremanagementwhich covers actions to be undertakenby the Treasury (regulations,fiscal reporting, government banking, and arrears); Accounting (budget classification);andBudget Departments(expenditurecontrols)as well as the audit agencies. 73 I Box 11: Service delivery inLao PDR: Lessons from the 2004 WDR The 2004 World Development Report (WDR) on "Making Services Work for Poor People" provides a useful framework for understanding the service delivery chain. The framework identifies three key sets o f actors and accountability relationships. (See Figure 24 - Framework o f accountability relationships.) Poor people can try to influence public action and policy makers by exercising "voice" through the ballot box or other means. Policy makers inturncan try to get providers to deliver services through an implicit or explicit "compact." Figure 24: Accountability framework in service delivery "Voice' "Compact" "Client Source:World Bank, World DevelopmentReport2004 However, weaknesses in the above two links-the "long route" o f accountability-are observed all too frequently, resulting infailures inservice deliveryto the poor. Weaknesses inthe electoral system or lack o f informationmay undermine the influence o f the poor inthe formulation o f public policy. Even when the poor can influence policy makers and politics are sufficiently "pro-poor," policy makers may have a difficult time to ensure that providers deliver services to the poor due to inadequate performance management, incentives, and monitoring. Greater recourse to the "short route" o f accountability, by directly increasing the "client power" o f the poor over providers can improve service outcomes. InLao PDR, clear weaknesses inthe "long route" of accountability suggest the needto improve the participation and choice o f the poor in service delivery to enhance client power and the "short route" o f accountability. As noted earlier, governance indicators for Lao PDR indicate substantial weaknesses in voice and accountability. Extensive weaknesses inpublic financial management indicate fundamental gaps inthe ability o f policy makers to manage providers. The inability to control spending within budget limits, especially at the provincial level, for example, highlight that basic financial accountability-much less accountability for performance-is still lacking. As noted inthe 2004 WDR, there are several ways to enhance the "client power" o f the poor in service delivery, including use o f contracts, self-monitoring providers such as NGOs, community control, and vouchers. Experience from other countries, for example, suggest that enhancing community participation and involvement in service delivery can significantly improve outcomes. In El Salvador, for example, the Educo program significantly improved enrollments without any decline in learning quality by allowing Community Education Associations, an elected committee made up primarily o f students' parents, to operate schools according to a one- year renewable agreement with the Ministry o f Education which allowed them select, hire, monitor, and retain or dismiss teachers. In Ceara, one o f the poorest states in Brazil, infant mortality was dramatically reduced during the 1990s by strengthening community leverage and voice over health providers and getting municipalities, local communities, health workers, and families to work together. Lao PDR will have to explore what improvements to existing service delivery arrangements will make the most sense inthe context o f Lao PDR. But several points are worth emphasizing. First, the analysis o f social outcomes in Lao PDR clearly indicates that outcomes are substantially worse for the poor and that public services are failing the poor. Second, because increases ingrowth and spending are not enough, substantial improvements in service delivery will be requiredto improve outcomes among the poor. Third, increasing "client power" o f the poor-through increased participation or other means-will be essential to improving service delivery in Lao PDR. Lastly, as emphasized in the WDR, what underlies virtually every successful experience in improving service delivery is better information and transparency. Without such information, policy makers cannot monitor performance and the poor cannot exercise their client nower 74 Box 12: Highlevel outcomes and milestones for public sector and financial managementinLao PDR High level outcomes 1. An improved systemo f financial accounting andcontrol. 2. A system o f assuranceand audit that helpsto strengthen the accounting system. 3. Building on top o f this, there must be a budget process that has the capability to appraise and forecast the resources o f the government and to allocate them through a proper system o f political accountability and transparency. 4. A classification of expenditures that has sufficient meaning in terms of policy goals and programs for resources to be allocated at the margin to priorities that are decided on by the appropriate political authorities. 5. A capability to monitor and report on the delivery against policy intentions in a wider sense than financial audit and to beginto assess the effectiveness o f expenditure interms o fpolicy goals inselected areas. 6. A programto improve service delivery by enhancing the "client power" o f the poor through increased choice andparticipation inservice delivery. 7. A general system of collection and execution ofpublic sector revenues capable of exercising sufficient control for additional revenues to be allocated to new policy initiatives and not diverted into other "unavoidable" expendituressuch as "out-year" payments on capital projects which would otherwise have to be stopped. 8. A program of SOE reform to staunch the fiscal demands of inefficient and unproductive commercial and quasi- fiscal activities that could similarly cause irresistible demands on the public revenues. 9. A reform oflocalgovernment that has clear divisionofroles andlines of accountability and makes substantial progress inclosing the opportunities for local administrators to use central government fiscal resources. 10. A program o f administrative reform and capability building without which none o f the above critical requirements will be met. Milestones ....Treasury i s reformed to ensure application o f single account system New budget cycle is inplace and working reasonably smoothly Programstructure for budget inplace inchosen policy areas Strategic plans for policy outcomes andtargets are inplace * Intergovernmental reform has proceeded to the point where there i s reasonable confidence that the lines of . fiscal accountability and roles are clear across levels of government and within strategic plans for priority areas. Basic capability for policy implementation at local level has beenbuilt and participation methods are working reasonably well which allow greater accountability o f service providers to the poor. 75 158. The existing tax base i s narrow and relies heavily on tumover and trade taxes, which are collected by provinces and at times held at the provincial level beyond what i s intendedby fiscal decentralization. The low level o f tax administration capacity and the large informal economy, however, means that it will be some time before a comprehensive, modem tax system can realistically be expected and that there will continue to be reliance on taxes which- while distortionary-can be imposed and collected by existing administration capacity. The proposed move to a value added tax (VAT) i s desirable in principle, but the introduction o f a VAT is highly demanding in terms o f administrative capacity even among very developed economies. If intemational experience i s any guide, introducing a VAT in Lao PDR will require careful planning and a very long lead time."' Inthe interim, Lao PDR could work on steadily improving the administration o f the trade and tumover taxes by (i) reducing the number and level o f rates; (ii)extending coverage; and (iii) aligning sales and income tax systems to harmonize administration and improve compliance. Immediate priorities include reviewing the administration o fresource royalties and improving forecasting o f tax revenues. 159. The transition to a market economy requires a new conception o f how a state owned enterprise (SOE) should be governed and managed. Unless fundamentally transformed, they are likely to be a serious fiscal drain through the need for subsidies and contingent liabilities which are not properly accounted for inthe government's financial management system. While the principles underlying SOE reform are not difficult to define, the record o f failure inmany countries highlights the great difficulty in implementing reform inthis area."' As a first step, annual financial plans and accounting reports o f SOEs need to brought under the control o f the Ministry o fFinance. 160. Progress on administrative reforms i s an urgent priority which will have an important bearing on progress o f other reforms. The agenda in administrative reform i s huge and i s addressed in the Lao PDR Govemance report, which provides frank observations on the challenges faced and the measures required."* Delivering on this agenda will require not just technical reforms andprocesses, but addressing the weakness inthe center. 161. A program of intergovernmental and local government reform is also absolutely critical as many of the observed weaknesses in Lao PDR are symptoms of underlying problems caused by poorly-functioning central-provincial government relations. Weaknesses inpublic financial management, for example, stem inlarge part from the inability o f central agencies such as the Ministry of Finance to exercise sufficient control over revenues collected and expenditures undertaken by provinces. The results o f the 1998-99 audit o f Khammouan province provides a sobering example, where total spending exceeded the planby loo%, with overruns o f 260% in non-social sectors of 260% and 180% in investment and construction expenditures. 'lo For example, when New Zealand introduced the simplest version of a VAT with a single rate and no exemptions, one thousand extra staff were employed to administer it. Ina country where even small businesses were already accustomed to regularly keeping adequate business records and filling in sales tax and income tax returns, the introductionof the VAT took two years and required a substantial team o ftax specialists and a major public education program. This includes, for example, developing a comprehensive policy statement concerning the principles for governance, management and regulation o f SOEs and implementingthese principles ina well-planned program over time. 'IzSorting out roles, structures and functions can be done inparallel with the work on local government. Other elements of administrative reform such as strengthening the merit principle, ethics, culture, promotion and remuneration can proceed at the central level. 76 162. The practice o f ministry officials working at local levels being under the control o f and paid by provincial governors i s one factor that is making it difficult for the center to maintain administrative control over expenditures. As noted inthe IMF FAD report, expenditure arrears are significantly higher among provinces than line ministries. As noted previously, these government arrears are also contributing to non-performing loan (NPL) problems in the banking sector.113 Inthe natural resources sectors, substantial revenues-e.g., large royalties in the case o f the Sepon mine-apparently accrue significantly to the province inthe first instance. 163. Many o f the services critical for poverty reduction have been devolved to local governments with insufficient capacity to deliver those services. While local governments in principle can better tailor services to meet local circumstances, the UNDP provincial pilot project in Luang Prabhang shows inpractical terms the challenges o f raising capability in the local governments to levels needed to implement better delivery o f public services. Unless a reliable and transparent decentralized system is put in place, issues o f efficiency and equity across local governments will remain outstanding. 164. The basic steps required to address centre-local relations contributing to the above problems are discussed in Box 13. It should be emphasized that intergovernmental relations are ultimately a political decision, and that this report is not recommending what the appropriate intergovernmental framework for Lao PDR should be. What is important is that whatever framework i s decided upon provides transparency and accountability, matches expenditure and revenue responsibilities, and does not undermine macro controls. 165. Many o f the governance reform areas flagged above are consistent with those identified by the Government. Priority governance initiatives identified in the 2003 Lao PDR Governance Report, for example, cover four categories: (i) service improvement; (ii) public people's participation; (iii)rule o f law and access to justice; and (iv) sound financial management. Implementation, however, remains the critical challenge. 4.4 Priority actions 166. Many o f the reforms described in Annex Table 1 will require significant capacity buildingand a long lead time for implementation. However, there are several priority actions which Lao PDR should undertake as they are absolutely necessary for any sustained improvement indevelopment outcomes. These are described below inFigure 24. 167. The above priority actions will require time and resources. They all also require significant capacity building to implement. However, there i s one important thing that Lao PDR can immediately do which will help improve development outcomes but which does not require much by the way o f additional resources: improve transparency. As noted in the 2004 WDR and elsewhere, better information can have a surprisingly powerful effect on incentives and accountability. Box 14 lists some information which Lao PDR already has and could quite readily disclose. Such disclosure would also send an excellent signal that the government is serious about improving governance inLao PDR. 'I3 Contractorsborrow from banks on the basis of contracts from ministries and provinces for which there are inadequate budget provisions. 77 - Box 13: Basic steps required in intergovernmentalreform inLao PDR "Locally based treasury officials find it difficult to follow centrally mandated.. .financial management practices in the face o f conflicting pressures from the Governor. A clear example o f this control problem i s the persistent failure o f provincial authorities to remit taxes and duties collected at the provincial level to the central treasury as requiredunderthe law." Source: 2002 World Bank, IMF,ADB Public Expenditure Review Decentralization is rarely an easy process. However, in Lao PDR-where the central government has been described as formally strong and with wide authority, but weak in terms o f power the way in which decentralization is occurring is worsening fiscal While decentralization will necessarily evolve according to country realities, the following basic steps are needed to ensure that fiscal control is strengthened while building local government capacity. '" 1. Review the constitution and subsidiary laws that establish the central government and local government and describe their roles, governance, functions, accountabilities and sources o f finance. Identify weaknesses in terms o f ambiguity or peculiarity by reference to general principles o f effective governance at both levels.'L6 2. Review the revenue raising powers o f the central, provincial and district levels o f government and assess their real revenue raising capacity inpractical terms. 3. Consider whether there i s a major imbalance between the costs o f the roles o f the provincial and district governments and their revenue raising capability undertheir assigned taxes. 4. Review and advise on the design and implementation of provincial fiscal transfers or reallocation o f roles to address any serious imbalances. 5. Review the constitutional, administrative and financial arrangements under which provincial and local government provide services as an agent o f the higher levels o f government rather than as an exercise o f their own discretion. 6. Review the relationships between territorial local governments and special purpose bodies or higher levels of government that operate within their territorial boundaries.'I' 7. Designgovemance and financial arrangements that put an endto the present unsatisfactory outcomes under which local government appropriates tax revenues that belong with the central government. 'I4 UNDP GPARreport While it is difficult to get very far with local government capacitybuilding andparticipationreform, if there is insufficient clarity about theroles, functions, andrevenuesources o fthe levelsof government, the effort to improvelocal capacity cannotwait untilthere is completeblueprint for local government reform. Inreality, the two will developtogether. 'I6 Accordingto the Matrix of MainPriority GovemanceInitiatives inthe LaoPDR Governancereport, thisprocessis currently underway. This is emergingas apotentialproblemin Thailandwith respect to regionalhealthandeducation organizationsof central government, for example. 78 Figure 25: Priority actions for Lao PDR Strategy to improve development outcomes inLao PDR 1 Maintain agricultural performance and facilitate off-farm opportunities for poverty reduction. 1 Develop manufacturing and services as the long-term growth engine for Lao PDR. a Improve the development contribution o f natural resources. 1 Improve public financial management and service delivery. Successful implementation o f above strategy will require improvements in enabling environment for private sector trade and as well as governance inpublic sector as described indetailed reformroadmap Four priority actions in reform roadmap to focus on Rationale 1. Establish a clear and transparent framework Needed to send proper signals to encourage entry and and level playing field for the private sector, growth o f firms, to facilitate private participation and to includingforeign direct investment, and promote exports. The framework for foreign direct trade investment inkey sectors such as natural resources will also need to be complemented by safeguards relating to environmental and social impacts. 2. Improve quality and coverage of critical Critical drivers o f poverty and social outcomes. Rural rural infrastructure, health and education infrastructure, especially rural roads, needed for market services by increasing recurrent integration to facilitate commercialization and off-farm expenditures and participation o f poor in opportunities. Unless accompanied by increases in service delivery recurrent expenditures as part o f an improved budgetary process, other reforms in social sectors are not likely to work. Simply increasing inputs without reforming existing arrangements, however, will not lead to sustainable improvements. Given weaknesses in accountability, will also need to enhance "client power" o f poor for better service delivery outcomes. 3. Establish financial accounting, control and Without basic control over expenditures and codtments, audit as part of overall program to additional public financial management or public sector strengthen expenditure management reforms not likely to work. 4. Develop detailed decentralization Many observed weaknesses in public sector due to poorly implementation plan to clarify functioning central-local relationships, which risk intergovernmental framework undermining fiscal management and service delivery. 79 Box 14: Improvingtransparency: priority disclosureinLao PDR I Providing more information can have a powerful development impact by affecting incentives and improving accountability. Several examples o f information which would be helpful to disclose are listed below. In most instances, these do not require Lao PDR to gather new information, but simply to disclose informationwhich it alreadyhas. I Publish the audited financial statementsand audit reports o f the StateAuditAuthority (SAA) infull. I Disclose expenditure allocations to administrative units, e.g., schools and hospitals, at disaggregated level to local communities, inline with recent Draft Instruction on Use o fAccounting Code for Administrative Units. I Make public annual audits o f state banks. I Disclose financial statements o f SOEs. I Publicize informationon negotiated royalties inall natural resource sectors. I Disclose procedures, processes, and criteria in setting quotas and issuing permits in forestry. Disclose information onbidding andpricing o f log sales. I Establish and publicize a template for the taxation terms o f Mineral Exploration and Production Agreements (MEPA) in mining to provide level playing field, consistent tax treatment, and avoid suspicions o f "special I Develop and publicize uniform models o f project agreements such as concessions and power purchase agreements (PPA) to use as starting points innegotiations with developers inhydropower. Make all regulations and laws affecting business community readily available. 4.5 Concludingremarks 168. This report reviewed the development challenge facing Lao PDR; analyzed the sources o f development potential; and discussed possible strategies and roadmaps for realizing that development potential. The preceding section laid out some steps for what needs to be done. While implementing these priority actions and reforms will be a challenge, the risks o f not taking action are also high. Ifthe enabling environment for the private sector is not improved and trade opened up further, growth will be lower than in the 1990s. If govemance in the public sector i s not improved, then resources will be ill-used and service delivery inadequate to improve development outcomes. 'I8 Indonesia, for example, has signed many hundreds o f contracts ofwork (COW) since 1967, but there have been only seven "generations" of COW with the terms o f each COW consistent with the published template for that generation. 80 169. The strategies and roadmaps presented are indicative. There is no single blueprint for improving development outcomes, and Lao PDR will have to develop its own roadmap which responds to its own particular circumstances. Otherwise, technically elegant action plans will founder when they cut across underlying, hardpolitical realities, as they have done inthe past. 170. As noted above, many o f the recommended reforms are consistent with those identified in previous government reports such as the Lao PDR NGPES and the Govemance Report. What has been lacking i s implementation o f those reforms. Without high level commitment to reforms, strategies and roadmaps will not get implemented, and Lao PDR's significant development potential will go unrealized. Box 15: Three remaininggaps in analyticalwork on Lao PDR ~ Despite the significant amount o f reports previously written on Lao PDR-including this one-there are still several gaps remaining in terms o f our understanding o f how growth, poverty, and social outcomes can be improved in Lao PDR. This box therefore briefly reviews previous reports and identifies at least three main analytical gaps outstanding inLao PDR which should be filled by future analytical work. First, firm and farm-level growth and productivity studies. What has been constraining the growth of small firms? Previous manufacturing firm surveys have identified a huge number o f tiny firms but a lack o f medium- sized firms ("missing middle"), suggesting that there are significant constraints preventing such frms from growing. Previous work in this area has tended to focus on (larger) firms in formal manufacturing sector (e.g., MPDF Survey o f Medium and Large Firms in Lao PDR, and even to some extent the MIHAJNIDO survey). An earlier GTZ survey focused on smaller firms inthe informal sector, but limited formal, quantitative analysis was conducted. A similar point applies to a previous small, informal survey o f off-farm and non-farm employment in Lao PDR. Given the need to develop manufacturing and services as a long-term growth engine for Lao PDR, a better understanding o f what has been constraining the growth o f small firms in both the manufacturing and services sector is needed, especially inrural areas where they provide off-farm employment. Similar analyses are required at the farm level inthe agriculture sector. Further growth in the agriculture sector i s critical to reducing poverty and is likely to come from commercialization and diversification. It would be useful to explore this question more formally using farm-level data to identify needed policy interventions. What has been driving ana constraining such commercialization and diversijication by smallfarms? Second, detailed analysis of the relationship between expenditures, outputs, and service outcomes. While the analysis in this report identified some o f the relationships between various outputs (e.g., schools, textbooks, clinics) and poverty and social outcomes, the relationship between expenditures and outputs was not explored, What is Lao PDR getting by way of outputs (e.g., schools, textbooks, clinics) for its expenditures in these sector? An expenditure tracking survey inkey social sectors could help to answer this question and identify how muchof the funds are reaching front-line service providers in schools and clinics. Additional work is also needed tc understand how service delivery can be improved for the poor. mat new mechanismsfor increasing choice anLi participation of thepoor in service delivery might be suitablefor Lao PDR? Third, analysis of how decentralization is affecting public financial management and service delivery and what should be done. Many o f the observed weaknesses inpublic financial management identified inprevious reports (e.g., WBiIMFiADB Public Expenditure Review and IMF FAD technical assistance report) stem from 81 underlying problems in central-local relations, but these were not directly addressed in these reports. Previous UNDP GPAR reports identified significant capacity building needs at the local level and the absence o f an adequate legal and regulatory framework (UNDP GPAR Pilot Luang Prabhang Province report and Decentralization report). There is an urgent need for greater clarity and specificity in Instruction OlIPM-which provided as a broad, general concept "building the province as the strategic unit, the district as the budget planning unit, and the village as the implementation unit"-and to move towards a detailed decentralizationimplementation plan supported by a clear legal and regulatory framework. While the latter i s a political decision, analytical work inthis areacancontribute to the development ofsuchan implementationplan. Whatspecific weaknesses inpublic financial management and service delivery are occurring as a result of the unclear decentralizationprocess and what are the corrective measures required? While there are undoubtedly other analytical gaps, the above three are particularly important for understanding and improving growth, poverty, and social outcomes in Lao PDR. Without understanding and addressing the micro firm and farm-level underpinnings of growth, it will be difficult to get adequate growth and poverty reduction at the aggregate macro level. Without a better understanding o f what one i s getting by way o f outputs and service outcomes from additional expenditures, it makes little sense to throw additional resources at those sectors. Lastly, it makes little sense to focus on providing technical and management solutions to problems which have their hndamental causes elsewhere due to the underlying political arrangement. 82 Annex Table 1:Detailedreform roadmaps I.IMPROVINGT EENABLINGENVIRONMENT A. SECTOR-SPECIFICREFORMS Agriculture Sustaining the growth performance o f the sector will require measures to facilitate diversification and intensification while maintaining food security, including 1. Promoting market integration and efficiency. This includes (i)improving transport infrastructure, especially rural roads and (ii) removing policy distortions which contribute to fragmentation o f domestic markets and affect farm gate prices. Improving rural roads is a highpriority as fewer than 2% o f rural roads are paved, most are impassable in the rainy season, and altemative transport options are limited.Il9 Price controls are still practiced, with rice and meat prices set within ranges determined by the central government with actual prices set by the provinces. Transport o f rice and livestock between provinces .is strictly controlled. 2. Strengthening agricultural services. Core research and extension agencies- National Agriculture and Forestry Extension Service (NAFES) and National Agricultural and Forestry Research Institute (NAFRI)-are o f recent origin and need to be strengthened given the need for priority research o n diversified cropping under irrigation in the dry season. Effectiveness o f such programs will be enhanced by village organizations with whom Government can partner. Progress on developing sustainable agricultural credit and micro-finance i s also needed, including a move away from reliance o n subsidized lending through APB. 3. Strengthening land administration to improve security o f land tenure and provide incentives for investments inimproving productivity, including inperennial crops. 4. Strengthen MAF capability in preparing, implementing, and monitoring and evaluating ambitious medium-term public investment program. 5. Manage risks posed by environmental constraints to agricultural growth, including exploring alternatives to swidden agriculture, development of aquaculture ' I 9Lao PDRhas no railway, limiteduse of inland waterways, and problematic thirdcountry transit arrangements and access to seaports. 83 Annex Table 1:- Detailed r e f o r m roadmaps (Cont.) I.IMPROVINGT [EENABLINGENVIRONMENT A. SECTOR-SPECIFICREFORMS Manufacturing and Manufacturing and Services services and off-farm opportunities 1. Improve business environment by creating a predictable, transparent and favorable policy environment and level playing field, above all by "letting go," including (i)moving away from "picking winners;" (ii)removing biases in administrative procedures and regulations that favor SOEs over domestic or foreign private firms; and (iii) identifying and removing unnecessary regulatory and administrative complexities 2. Strengthen market institutions, including dispute resolution and contract enforcement. 3. Improve transportation and logistic infrastructure to access regional and international markets 4. Facilitate access to financial resources and services [see below - financial sector management] 5. Improving human resources to increase productivity through education and training Of- farm opportunities Facilitate off-fannemployment to improve householdincomes, diversification, andjob creation by (i) improving physical infrastructure (roads, electricity, communications); (ii)improve legal framework for establishing businesses (e.g., per Enterprise Law in Vietnam); (iii)improve microfinance to provide credit to growing, successful enterprises; (iv) improve property rights over key assets such as land to be used as collateral in accessing credit; and (v) crafting facilitation and extension programs that are sensitive to local conditions and cultural differences. Mining 1. Establish a track record of consistent and fair dealing with investors (both local and foreign) through appropriate regulations, guidelines and procedures which will facilitate the expeditious issuanceofmining titles. 2. Establish a standard set o f investment conditions or work contracts for the sector which would apply to all investors and avoid the need for "case-by-case'' negotiations. 3 . Put into place an internationally competitive taxation package which encourages investment while, at the same time, maximizes government taxation revenues. 4. Complete the mining regulations and sector specific environmental regulations. 5. Enhance funding for training and logistical support for the Direction de l a Geologie et des Mines (DGM) to buildcapacity inthe relevant functions. 6. Ensure clear mandates for government oversight institutions to avoid potential delays which may be causedby unnecessary division o f labor betweenMIHiDGM and the DDFI. 7. Avoid conflicts o f interest inrole o f the State as "ownerJoperator" or state owned miningenterprises and the role ofthe State as "regulator" of the sector. 84 Annex Table 1:-Detailedreformroadmaps(Cont.) [.IMPROVINGTI[EENABLINGENVIRONMENT A. SECTOR-SPECIFICREFORMS Hydropower 1. Complete commercialization (including full corporate independence) o f EdL and assign regulatory powers to a specially designated body with a particular view to separating EdL's commercial activities from social obligations, and potentially create a separateholding company for EdLshareholding inIIPs. 2. Define a rural electrification plan including zoning of areas to be grid-connected and off-grid areas. Review the regulatory framework to facilitate and create incentives for non-utility electrification. 3. Review export and import tariffs to value peak versus off-peak power with a view to optimizing the advantages o f hydropower as peak power. Potentially, create a national single-buyer agency to negotiate cross-border tariffs and coordinate dispatch. 4. Develop a power sector financing strategy and power development plan in order to address the pressing financing needs and explore public-private partnership models. 5. Initiate pre-feasibility studies o f prioritised sites in order to ensure optimum government returns (through royalties, taxes and possibly dividendsirents). 6. Strengthen the framework for private participation including: (i) rationalize inactive MOUs; (ii) streamline the legal and regulatory framework and establish model contract documents; and (iii)designate a single agency with licensing powers. Forestry 1. Improve forest management and utilization by (i)defining the nature and arrangements for management o f production forest areas; (ii) complete, clarify, and simplify legal and regulatory framework; (iii)modify timber sales, log allocation and quota system; (iii)disclose audits and discontinue preferential contracts for SOEs 2. Expand community participation in forestry by (i) endorsing village forestrq association (VFA) management and set targets for expansion; (ii) c o n f m timbei marketing terms for existing and future VFA 3 . Control illegal logging by ceasing approvals o fnew primary processing plants anc establishing illegal logging monitoringcapacity. 85 Annex Table 1: Detailedreformroadmaps(Cont.) - I.IMPROVING'I[EENABLINGENVIRONMENT B. CROSS-SECTORAL REFORMS Trade-related 1. Remove foreign exchange controls andbring official andparallel markets together reforms to improve transparency and efficiency. 2. Ensure that the duty drawback system for exporters works as well inother sectors as ingarments 3. Carry out commitments under Asean Free Trade Area (AFTA) and speedup move o f Temporary Exclusion List (TEL) items to the Inclusion List (IL) and set Common External Preferential Tariff (CEPT) rates below 20 percent to take advantage o f the preferential rates that will be available from the original six ASEAN members as they move to low and zero rates. Set CEPT rates to zero by 2008. 4. In addition to eliminating Quantitative Restrictions (QRs) on IL items by 2005, replace all QRs with tariffs. 5. ExtendCEPT rates underAFTA to all nations on an Most FavoredNation(MFN) basis. 6. End the extensive duty exemptions which appear to apply to 80 percent of imports, possibly by setting a uniform low tariff in the order o f 5 percent which would be consistent with medium term AFTA obligations. HumanResources 1. Overwhelming priority i s to increase recurrent expenditures in health and (Health and education Education) 2. Improve resource allocation in sectors by addressing imbalances between recurrent and capital expenditures and across regions. Ensure that large majority o f public resources i s spent on primary rather than vocational and tertiary education and towards preventive rather than curative care inhealth. 3. Improve incentives for service delivery, including higher salaries and increased local accountability through community participation 4. Increase scope for partnership with private sector to encourage private provision, especially incurative health care and technical and vocational education. 86 Annex Table 1: -Detailedreformroadmaps(Cont.) B. CROSS-SECTORALREFORMS Environmental Policy and institutionalframework management 1. Improve enabling regulations for environmental management 2. Strengthen capability o f supporting organizations. The management o f the key institutions, STEA and MAF, needs to be strengthened and their budgetary resources increased. 3. Improve participatory processes, by providing guidelines and opportunities for community-led programs as inforestry management 4. Improve capacity of Provincial govemments in area of environmental management. Instruments 1. As an immediate priority, put inplace transparent mechanisms for the allocation o f forest, mineral and other concessions and quotas 2. Strengthen compliance mechanism by improving technical capacity and demonstrating political commitment to proper enforcement 3. Improvereliability o f environmental impact assessment system. 4. Put in place adequate environmental monitoring and disclosure, including developing reliable baseline environmental data 5 . Improve delivery o f key public services, including coverage o f solid waste collection 87 Annex Table 1:-Detailed reform roadmaps (Cont.) B. CROSS-SECTORALREFORMS Financial Sector Comuetition aolicv Management 1. Reduce domikation of two state owned banks and Agricultural Production Bank (APB), including their restructuring andrecapitalization 2. Establish consensus on a medium-term strategy and implementation plan for the financial sector, including participation by foreign and non-bank financial institutions 3. Reduce barriers to entry and expansion for private and foreign commercial banks in Lao PDR, including increased ownership in domestic commercial banks. Amend Banking Act and regulations to reduce the legal and regulatory gap between domestic and foreign banks, including lowering the unremunerated statutory requirement [check]. 4. Develop enabling legal and regulatory environment to encourage non-bank financial institutions, including micro-finance, insurance, leasing, factoring, mortgage lending, consumer finance. 5. Planfor the privatization o f one o f the two state banks. Legal regime 1. Amend Banking Act to meet BIS norms, including appointment o f independent directors to govem state banks. 2. Establish predictable regime for resolution of default situations to attract new lending and investment, including (i) implementing voluntary, out o f court debt resolution framework, especially for large non-performing loans (NPLs) to SOEs; (ii) streamlining civil procedures and amending foreclosure and bankruptcy laws to strengthen debt collection and restructuring; (iii) accelerate land titling and registration; (iv) enact secured transactions law; and (v) enhance capacity o f judiciary and courts to adjudicate commercial cases expenditiously 3. Enact anti-money laundering act, establish financial investigative unit, train banks to report suspicious transactions and empower B O L to examine and banks to report suspicious transactions Regulatory and supervisory regime 1. Improve transuarencv of state banks and APB to provide accurate monitoring of institutions and their portfolios, including making public annual audits. Strengthen key regulation pertaining to loan classification and provisioning, B O L 98, inline with BIS and ongoing technical assistance. 2. Upgrade information technology systems o f Bank o f Lao PDR (BOL) and strengthen supervisory capacity o f Bank and Financial SupervisionDepartment 3. Conduct Base1 Core Principles Assessment to benchmark regulatory and supervisory regime and identifykey needed steps Creditpolicy 1. Allow banks to set interest rates and manage portfolios without interference as nominally providedfor under existing regulations 2. SOEs to borrow from private banks instead o f allocated that lending from state banks 3. Develop altemative sources o f funding for public sector borrowing 4. Fund subsidized credit schemes transparently through the budget and gradually phase out over time. 88 Annex Table 1:-Detailedreformroadmaps(Cont.) B. CROSS-SECTORALREFORMS Public financial and 1. A basic system o f financial accounting and control. sector management 2. A systemof assuranceandaudit that helpsto strengthen the accounting system. 3. Building on top o f this, there must be a budget process that has the capability to appraise and forecast the resources o f the government and to allocate them through a proper system o fpolitical accountability andtransparency. 4. A classification of expenditures that has sufficient meaning in terms of policy goals andprograms for resources to be allocated at the margin to priorities that are decided onby the appropriate political authorities. 5. A capability to monitor and report on the delivery against policy intentions in a wider sense than financial audit and to begin to assess the effectiveness o f expenditureinterms ofpolicy goals inselected areas. 6. A program to improve service delivery by enhancing the "client power" of the poor through increased choice and participationinservice delivery. 7. A general system o f collection and execution o f public sector revenues capable o f exercising sufficient control for additional revenues to be allocated to new policy initiatives and not diverted into other "unavoidable" expenditures such as "out- year" payments on capital projects which would otherwise have to be stopped. 8. A program o f SOE reform to staunch the fiscal demands o f inefficient and unproductive commercial and quasi-fiscal activities that could similarly cause irresistible demands on the public revenues. 9. A reform o f local government that has clear division o f roles and lines of accountability and makes substantial progress inclosing the opportunities for local administrators to drain central government fiscal resources away. 10 A program of administrative reform and capability building without which none o fthe above critical requirements will be met. Decentralization 1. Review the constitution and subsidiary laws that establish the central government and local government and describe their roles, governance, functions, accountabilities and sources o f finance. Identify weaknesses interms o f ambiguity or peculiarity by reference to general principles o f effective governance at both levels. 2. Review the revenue raising powers o f the central, provincial and district levels o f government and assesstheir real revenue raising capacity inpractical terms. 3. Consider whether there is a major imbalance between the costs o f the roles o f the provincial and district governments and their revenue raising capability under their assigned taxes. 4. Review and advise on the design and implementation o f intergovernmental fiscal transfers or reallocation o f roles to address any serious imbalances. 89 Annex Table 1:-Detailedreformroadmaps(Cont.) E.IMPROVINGPUBLICSECTORGOVERNANCE B. CROSS-SECTORALREFORMS I I 5. Review the constitutional, administrative and financial arrangements under nhich I provincial and local government provide services as an agent o f the higher levels ofgovernment rather than as an exercise o f their own discretion. 6. Review the relationships between territorial local governments and special purpose bodies or higher levels of government that operate within their territorial boundaries. lZ1 7. Design governance and financial arrangements that put an end to the present unsatisfactory outcomes under which local government appropriates tax revenues that belong with the central govement. lZoAccording to the Matrix of Main Priority Govemance Initiatives inthe Lao PDR Governance report, this process i s currently underway. 121This is emerging as a potential problem inThailand with respect to regional health and education organizations o f central government, for example. 90 Annex Table 2: OLSRegressionsonLogPer CapitaConsumption Vientiane Other Urban Rural South Rural Center Rural North Variables Coef t-stats Coef t-stats Coef t-stats Coef t-stats Coef t-stats Household Demographics - . Size (#) -0.056 -2.9 -0.052 -4.6 -0.015 -1.7 -0.028 -3.4 -0.029 -3.6 Age Groups: # 15-49 # 0-6 -0.036 -1.0 -0.104 -5.9 -0.055 -3.8 -0.060 -5.0 -0.077 -6.1 # 7-14 -0.049 -1.7 -0.071 -4.7 -0.055 -4.0 -0.033 -3.0 -0.039 -3.5 # 5oDlUS -0.088 -1.7 -0.056 -2.2 -0.034 -1.7 0.011 0.6 -0.029 -1.9 Head Male -0.032 -0.2 0.170 1.8 0.045 0.6 0.035 0.5 0.041 0.6 Age (Years) -0.046 -2.3 -0.010 -1.0 0.007 1.2 0.010 1.9 0.004 0.8 Age Square 4.E-04 2.2 l.E-04 1.0 -6.E-05 -0.9 -1.E-04 -2.4 -7.E-05 -1.1 Education (Years) 0.024 3.7 0.029 7.0 0.018 4.1 0.021 6.0 0.018 4.6 Occupation: Subsistence farmer Employee 0.007 0.1 -0.001 0.0 0.030 0.7 -0.013 -0.3 0.117 2.7 Employer 0.127 0.5 0.591 6.0 0.076 0.2 0.224 1.4 0.593 3.8 Self-employed 0.149 1.6 0.078 1.8 0.153 3.9 0.105 3.1 0.140 4.3 Unpaidfamily worker -0.110 -0.6 0.238 2.4 0.249 1.o 0.002 0.0 0.311 5.5 Other -0.018 -0.2 0.020 0.4 -0.012 -0.3 0.044 1.2 -0.002 -0.1 Spouse Spouse -1.665 -4.1 -0.432 -2.2 -0.171 -1.2 -0.134 -1.3 0.021 0.2 Age (Years) 0.061 3.6 0,005 0.6 0.008 1.4 0.001 0.2 -0.001 -0.3 Age Squared -0.001 -3.1 0.000 -0.2 0.000 -1.2 0.000 0.7 0.000 1.o Education (Years) 0.014 1.7 0.013 2.7 0.010 1.8 0.009 2.2 0.021 4.2 Agriculture & Assets Car 0.325 4.3 0.394 7.1 0.389 3.7 0.345 4.8 0.479 4.9 Tractor 0.041 0.4 0.111 2.1 0.556 5.9 0.289 7.7 0.202 5.0 o x 0.057 0.2 0.043 0.6 0.090 1.9 0.023 0.5 0.106 2.0 Boat -0.187 -1.9 0.018 0.4 0.071 2.3 0.122 4.8 0.185 5.8 Fishing net -0.006 -0.1 -0.050 -1.7 0.014 0.5 -0.025 -1.1 0.006 0.3 Rice husking 0.275 1.1 0.082 1.4 0.162 3.2 -0.001 0.0 0.034 1.1 Dry land (hectare per capita) 0.290 1.1 -0.084 -0.6 0.880 5.8 0.745 4.8 0.011 0.4 Irrigated land (hectare per capita) 0.179 0.2 0.094 0.8 0.323 0.3 1.457 4.2 0.760 3.5 Cows (number per capita) 0.070 1.6 0.080 4.5 0.034 2.2 0.033 3.2 0.075 5.8 Housetenancy -0.056 -0.7 0.158 2.1 -0.166 -2.8 -0.138 -1.4 0.018 0.2 Distranceto main road (km) 0.000 0.2 0.000 -0.7 0.000 -0.2 -0.001 -3.2 -0.001 -2.1 Village infrastructure Access to road during rainy season? na na 0.017 0.2 0.104 3.0 0.093 3.9 -0.018 -0.7 Access to public transport? 0.226 2.2 0.112 2.3 0.088 2.4 -0.014 -0.6 0,044 1.6 Access to electricity? na na 0.065 1.8 -0.032 -0.6 0.025 0.9 0.015 0.4 Access to piped water? -0.018 -0.2 0.020 0.6 -0.067 -2.5 0.093 4.0 0.109 4.7 Developmentproject -0.102 -1.8 0.099 3.3 0.062 2.5 0.019 0.9 0.102 4.4 Social infrastructure index -0.090 -0.9 -0.038 -0.7 0.196 4.9 0.119 2.7 0.059 1.7 Cultivated land (hectare per capita) -0.004 -0.8 0.000 -0.2 0.000 0.8 0.000 0.1 0.000 0.9 Irrigated land (hectare per capita) 0.016 1.1 -0.008 -0.8 -0.007 -1.6 0.002 0.8 0.009 2.1 Constant 12.1 22.7 10.5 40.4 9.8 57.2 10.0 59.5 9.9 65.8 Observations 500 1628 1583 2338 2831 R-Square 0.3213 0.3122 0.3287 0.3131 0.3145 91 Annex Table 3: Core Social Indicators by Province Province Enrolment Primary Female % HH %HH %HH rate ages completi literacy access to access to hospital Poverty 6-14 onrate rate health safe 8+hours Rate 15+ care* water** away ientiane 90.7 84.7 84 78 89 0 12.2 iunicipality hongsaly 54.4 27.2 33 50 27 12 64.2 uangNamtha 49.5 47.6 19 33 27 9 57.5 luodomxay 44.6 31.3 36 87 32 28 73.2 okeo 50.2 29.6 33 43 45 17 37.4 ouang 65.1 43.8 50 54 37 3 49.4 rabang oua Phanh 68.9 50.7 38 79 23 36 74.6 aygnaboury 76.3 79.7 68 91 33 0 21.2 iengKhoang 77.5 66.1 58 66 37 8 34.9 orikhamxay 71.3 47.5 63 69 65 0 25.8 hammuane 60.5 44.2 60 68 38 20 41.6 avannakhet 70.8 58.8 50 86 66 4 37.1 aysomboom 75.4 72.2 43 69 39 0 55.0 ientiane 89.0 77.9 69 86 61 0 24.3 rovince aravanh 50.1 34.9 41 72 39 10 39.6 ekong 60.2 32.4 37 60 43 23 45.7 hampasak 76.2 66.4 57 92 58 0 35.6 ttapeu 62.7 43.O 57 71 30 18 45.3 * Access to skilled medicalpersonnel(medical practitioner, trainednurse, community healthworker or **dispensary) Piped water or protected well Source: LECS 97/98;MICS 2000; Lao PDR PER; Kakwani et a1(2002) as cited inWorld Bank, PER, 2002. 92 Annex Table 4: Marginal effects on the probabilityof startingattending schoolfor ages 6 to 14 LECS 97/98 Data -Asset Quintiles (9.30) (5.59) (4.50) (7.44) (4.61) (3.76) Province fixed effects Yes Yes Yes Yes Yes Yes Number o f Obs. 15,729 15,729 5.906 12,143 12,143 4,411 Notes: Dependent variable i s probability o f starting or attending school for ages 6 to 14; or probability o f having completed at least 5 years o f school for ages 15-19. 93 Annex Table 5: Marginaleffects on the probability of startingattending school for ages 6 to 14 MICS 2000 Data-All Laos (Asset Quintiles) 1.114 1.16 1.41 1.42 1.92 1.69 (16.35) (16.55) (22.14) (22.60) (3.94) (3.43) Age-Sq -0.047 -0.047 -0.067 -0.067 -0.055 -0.048 (-13.20) (-13.33) (-20.44) (-20.78) (-3.81) (-3.32) Sex -0.116 -0.120 -0.220 -0.222 -0.363 -0.234 (1=female) (-2.36) (-2.43) (-5.51) (-5.57) (-6.54) (-4.14) Age HH 0.007 0.006 0.005 0.005 0.010 0.010 (2.29) (2.35) (2.24) (2.27) (3.11) (2.97) Sex HH 0.072 0.081 0.113 0.131 0.133 0.182 (1=female) (0.44) (0.52) (0.85) (1.03) (0.94) (1.37) Education HH 0.199 0.157 0.175 0.146 0.167 0.182 (4.79) (3.86) (4.61 ) (3.87) (2.39) (2.67) Urban 0.551 0.517 0.550 0.527 0.825 0.594 ______ (5.86) Lao -- 0.237 __ 0.153 _- ----+-- 0.3 11 (1.92) (1.46) (2.10) Phoutai -- 0.115 -- 0.043 -- -0.092 (0.50) (0.20 (-0.35) Kammu -- -0.308 _- -- -0.186 (-1.83) (-1.34) (-1.04) Hmong __ 0.011 _- -0.055 _- -0.203 (0.05 ) (-0.24) (-0.90) Leu _ _ 0.475 -- 0.359 _- 0.597 (1.97) (1.76) (2.69) Katang -- -0.218 __ -0.275 -_ 0.183 (-0.89 ) (-1.35) (0.67) Province fixed effects Yes Yes Yes Yes Yes Yes Number o f Obs. 6,716 6,716 6,741 6,741 2,616 2,6 16 94 Annex Table 6: Determinantsof HouseholdMedicalExpenditures per Capita,LECS 97/98 (dependent variable= log(medica1 expenditures per capita) All Laos Ruralonly (1) (2) (1) (2) Ouintile 2 . 0.317 I1 0.318 0.304 I1 0.306 (4.96) (4.99) (4.41) (4.43) Quintile 3 0.609 0.614 0.639 0.646 (9.68) (9.82) (9.33) (9.49) Quintile 1III0.914 II1I0.919 1III0.905 IIII0.912 (14.40) (14.57) (12.81) (13.OO) Quintile 5 1.388 1.393 1.362 1.373 (20.56) (20.78) (17.59) (17.91) Sex HH -0.076 -0.069 0.067 0.070 (1=female) (-0.91) (-0.83) (0.65) (0.69) Age HH 0.001 0.002 0.001 0.001 (0.79) (0.97) (0.39) (0.56) EducationHH IIII0.005 II1I0.006 II1I0.004 IIII0.007 (2.42) (3.42) (1.31) (2.48) No. of elderlyinHH 0.031 0.0288 0.038 0.036 (2.30) (2.11) (2.52) (2.38) No. children6 y r s inHH II1I0.024 II1I0.023 11I10.036 IIII0.036 (3.87) (3.71) (5.30) (5.21) No. women inHH -0.024 -0.022 -0.007 -0.005 (-3.27) (-3.1 1) (-0.82) (-0.61) HHis a farmer -0.006 -0.006 -0.009 -0.011 Urban I -0.109 I -0.048 _ _ -- Source: own calculations from LECS 97198 primary data. 95 Annex Table7: Probabilityof SeekingTreatmentby Quintileand Ethnicity Age - -0.007 -0.0005 0.003 0.002 (-0.39) (-0.27) (1.41) (1.36) Sex -0.054 -0.51 -0.045 -0.045 (-0.66) (-0.62) (-0.50) (-0.49) Wealthquintile 2 0.395 0.369 0.182 0.193 (3.14) (2.81) (1.24) (1.26 Wealthauintile 3 0.474 0.440 0.374 0.399 (3.67) (3.24) (2.56) (2.59) Wealthquintile 4 0.685 0.649 0.583 0.614 (5.62) (5.07) (4.3 1) (4.28) Wealthquintile 5 0.738 0.671 0.483 0.517 (5.50) (4.70) (3.25) (3.26) Constant -0.380 -0.330 -1.061 -1.097 (-3.39) (-1.98) (-592) (-5.83) Phoutai -- -0.041 -- 0.241 (-0.21) (1.17) Kammu -- -0.184 -- 0.096 (-1.23) (0.58) Hmong -- -1.598 -- 0.093 (-0.65) (0.34) Leu -_ 0.549 -- 0.047 (1.97) (0.16) Katang - -- -0.328 -- -0.156 (-1.80) (-0.76) Provinceeffects Yes No Yes No N.Obs. 959 959 959 959 Source: own calculations, MICS 2000 data 96 Annex Table 8: Impact on poverty and Inequality of Additional 1%GDP Growth by Sector Impact on poverty and inequality of additional 1% GDP Growth by Sector 2003 2004 2005 2006 NPEP-Baseline Poverty Headcount 27.3 25.5 23.6 22.1 Poverty Gap 6.5 5.9 5.4 5.O Poverty Severity 2.31 2.1 1 1.92 1.76 Number of Poor ('000) 179 185 193 201 Inequality (Gini Coefficient) 36.2 36.6 37.1 37.7 AgricuIture Poverty Headcount 26.3 23.7 20.7 17.9 Poverty Gap 6.1 5.3 4.6 3.9 Poverty Severity 2.18 1.87 1.60 1.36 Number of Poor ('000) 1516 1402 1252 1113 Inequality (Gini Coefficient) 36.0 36.2 36.5 36.9 Industry Poverty Headcount 27.1 25.1 23.1 21.5 Poverty Gap 6.4 5.8 5.3 4.9 Poverty Severity 2.29 2.08 1.89 1.72 Number of Poor ('000) 1566 1486 1397 1335 Inequality (Gini Coefficient) 36.0 36.2 36.5 36.9 Services Poverty Headcount 27.1 25.0 22.9 21.4 Poverty Gap 6.4 5.8 5.3 4.8 Poverty Severity 2.28 2.06 1.87 1.71 Number of Poor ('000) 1562 1476 1386 1327 lnequality (Gini Coefficient) 36.0 36.2 36.5 36.9 97 References Bourdet, Yves. 2000. "Emerging Rice Market in Laos?" Country Economic Report 2000:7, SwedishInternational Development Authority, Stockholm. Brenning, M. and Davading, S. 2003. Prospects for Further Integration of Lao PDR's Manufacturing Sector into ASEAN. Vientiane, UNIDO IntegratedProgramme for Lao P.D.R Brenning, M.,Davading, S., andDissmann, B. 2002. Composition and Evolution of Lao PDR 's External Trade. Vientiane, UNIDO Integrated Programme for Lao P.D.R. Chaudhuri, Shubham. 2000. "Empirical Methods for Assessing Household Vulnerability to Poverty.", mimeo, Department of Economics and School o f International and Public Affairs, Columbia University. Chawdhury, A.K, "Emerging Priorities for Transport" World Bank Transport Forum 2003; Washington, D.C.; January 21,2003. Cooperation and Planning Committee. 2002. Industrialization and Modernization Strategy (2001-2020). Vientiane, Cooperation andPlanning Committee. Datt, G. and T. Walker. 2002. PovStat 2.12, A Poverty Projection Toolkit, User`s Manual, World Bank,Washington, D.C. Processed. Datt, Gaurav, and Dean, Jolliffe. 2001. "Poverty in Egypt: Modeling and Policy Simulations.", mimeo. Development Analysis Network. 2003. Off-farm and Non-farm Employment in South-east Asian Transitional Economies and Thailand. Easterly, W., and Levine, R. 2001. "It's not factor accumulation: stylized facts and growth models." Filmer, Deon, and Lant Pritchett. 1998a. "Estimating Wealth Effects without Income or Expenditure Data- or Tears: An Application to Educational Enrollment inIndia", World Bank Policy ResearchWorking PaperNo. 1994.Washington, D.C. Filmer, Deon, and Lant Pritchett. 1998b. "Educational Enrollment and Attainment in India: Household Wealth, Gender, Village and State Effects", South Asia Region IDP-173, The World Bank. Grant, Evans. 2002. "A Short History o f Laos, The LandinBetween." Australia. Gunewardena, Dileni, van de Walle, Dominique. 2000. "Sources o f Ethnic Inequality in Vietnam." World Bank Working Paper 2297. Washington, D.C. 98 IFPRI (International Food Policy Research Institute), Ministry of Planning and Finance, and Eduardo Mondlane University. 1998. UnderstandingPoverty of Well-Being in Mozambique: TheFirst National Assessment (1996-1997). Kaufmann, D. and Kraay, A. 2003. "Governance and Growth: Causality which way? - Evidence for the World, inbrief' Lao PDR, Steering Committee for the Agricultural Census, Agricultural Census Office. "Lao Agricultural Census, 1998/99, Highlights." Vientiane, February 2000. Mekong Project Development Facility. 1998. A Survey of Medium and Large Private Companies in Lao P.D.R.. National Statistics Center. 2001. Statistical Yearbook 2001, Vientiane, Committee for Planning andCooperation. Ravallion, Martin. 1996. "Issues in Measuring and Modeling Poverty.", The Economic Journey, 106, 1328-1343. Sachs, J., Radelet, S. and Lee, J-W. "Economic Growth inAsia" (draft paper preparedfor ADB study onEmerging Asia: Changes andChallenges; July 1997). Sala-I-Martin and Subramanian. 2003, "Addressing the Natural Resource Curse : An Illustrationfrom Nigeria", IMFWorking Paper. Scott, Gramham, "A Long Term Roadmap for Public Sector Financial Management and GovernanceReforminthe Lao PDR'. State Planning Committee National Statistical Centre. 2000. Basic Statistics of the Lao P.D.R. Vientiane, State Planning Committee National Statistical Centre. State Planning Committee National Statistical Centre. 1997. Results from the Population Census 1995. Vientiane, State PlanningCommittee National Statistical Centre. Steering Committee for the Agricultural Census, Agricultural Census Office. 2000. Lao Agricultural Census, 1998/99Highlights. Vientiane, Agricultural Census Office. The National Poverty Eradication Programme (NPEP). 2003. A Comprehensive Approach to Growth and Development, Eight Round TableMeeting. Vientiane. UNDP (United Nations Development Programme). 2003. Governance and Public Administration Reform. U N D O (United Nations Industrial Development Organization). 2003. The International Yearbook of Industrial Statistics. 99 Upadhyaya, Shyam, Yossri, Ittichai. 1999. "Survey of IndustrialEstablishments, 1999, Final Report." UnitedNations IndustrialDevelopmentOrganisation. Vemstrom(2003), ThailandPowerScenario Study, p.54. World Bank. 2003.Lao P.D.R.Economic Monitor. Vientiane: WorldBank. World Bank, Aid Effectiveness Research. Assessing Aid: What works, F'hat Doesn 't, and Why. www.worldbank.org/research/aid/ . 2003. WorldDevelopment Indicator. Washington, D.C.: WorldBank, 100