Report No. 30737 Bhutan Country Assistance Evaluation November 23, 2004 Operations Evaluation Department Document of the World Bank Contents ................................................................................................................................ i Preface Summary........................................................................................................................... ... 111 1. Background ........................................................................................................... 1 Country Characteristics ............................................................................................ 1 Overview o f Economic and Social Development .................................................... 3 2 . The Bank’s Role in Bhutan ................................................................................... 8 Context: The Bank as a Small Player in a Small Country ...................................... 8 Bank Assistance Strategy ....................................................................................... 10 3 . Implementationof the Bank’s Program ............................................................ 12 Analytical and Advisory Activities (AAA) ........................................................... 12 Lending .................................................................................................................. 15 Education ................................................................................................... 16 Forestry and natural resource management .............................................. 18 Infrastructure ............................................................................................. 20 Strengths and weaknesses o f lending ......................................................... 21 Donor cooperation ..................................................................................... 23 4 . Evaluation o f the Bank’s Assistance .................................................................. 24 Outcome o f the Bank’s Assistance ........................................................................ 25 Sustainability of Positive Outcomes ...................................................................... 26 Institutional Development Impact.......................................................................... 26 Bank Performance .................................................................................................. 26 5 . Lessons and Recommendations .......................................................................... 27 Tables 1.1: Bhutan . Selected Macroeconomic Indicators ...................................................... 4 1.2: Bhutan Social Indicators ........................................................................................ 5 1.3: Bhutan - Discrepancies-between U N and R G O B Estimates of Social Indicators ............................................................................................... 6 1.4: Bhutan - Poverty Incidence in 2000 ...................................................................... 7 2.1: Bank Group Lending to Bhutan 1983-2003 .......................................................... 9 3.1 : OED Review Ratings ........................................................................................... 16 Contents (continued) Figures 1.1: Bhutan: N e t Capital Inflows from India and Power Export Revenues.................2 1.2: Composition o f GDP at Factor Cost in 1980 Prices. by Sector ............................. 4 2.1 : Comparison o f Bank Lending and Foreign Aid Flows to Small and Low-income Countries ........................................................................ 9 Text Boxes 3.1: Bhutan’s PRSP..................................................................................................... 14 3.2: Trust Fund for Environmental Conservation ....................................................... 19 Annexes Annex Foreign Assistance to Bhutan by Donor and Sector ............................ 29 A: Annex Forestry Pricing and Marketing Policies .............................................. 31 B: Annex List o f People Interviewed .................................................................... C: 33 Annex Guide to OED’s Country Evaluation Rating Methodology .................35 D: Annex E: Comments from the Bank’s Country Management Unit. South Asia Region ................................................................................ 39 Annex F: Comments from the Government o f Bhutan......................................... 43 Annex G: Management Action Record (MAR) .................................................... 46 Annex H: Chairman’s Summary Committee on Development Effectiveness (CODE) ................................................................................................ 49 Statistical Tables .............................................................................................................. 53 i Preface This evaluation provides an independent assessment o f the role o f World Bank assistance to Bhutan during the period 1993-2003. The Country Assistance Evaluation (CAE) examines whether: (a) the objectives o f Bank/IDA assistance were relevant; (b) the Bank’s assistance program was effectively designed and consistent with i t s objectives; and (c) the Bank’s program achieved i t s objectives and had a substantial impact o n the country’s development during this period. Examining these questions allows the C A E to draw lessons and recommendations for future Bank assistance. Annex D describes the methodological approach. The basis for the C A E consists o f OED project assessments, sectoral reviews, and interviews with Government officials, Bhutanese c i v i l society, other donors, and Bank staff. A l i s t o f those interviewed i s shown in Annex C. An OED mission visited Bhutan during M a y 2003. The Regional Management o f the Bank provided comments which are found in Annex E. The report was also sent to the Bhutanese authorities whose comments are in Annex F. O E D has taken account o f factual and editorial comments in this draft. This Country Assistance Evaluation was written by Gene Tidrick (Consultant) with the assistance o f Daniel Ritchie (Consultant). Tatsuhide Kanenari (Consultant) provided support o n the education sector as well as overall research assistance. This evaluation also benefited from comments o f three peer reviewers: Mr. Sudhir Shetty (PRMPR); Mr. Asita D e Silva (OEDCR); and Mr. Ronald Parker (OEDST). Danuta Danilova provided additional research assistance. Betty Casely-Hayford and Janice Joshi provided administrative and editorial support. ... 111 Summary 1. Bhutan has a well-deserved reputation for good economic management and environmental stewardship. The national development strategy o f maximizing “Gross National Happiness” seeks to balance the pursuit o f conventional GDP growth with preservation o f the country’s cultural and environmental heritage. I t has broadly succeeded o n both fronts. Helped by exports o f hydropower and substantial foreign aid, output per capita has grown at an annual average rate o f 4 percent over the past two decades, social indicators have improved significantly, and yet Bhutanhas avoided many o f the problems o f social dislocation and environmental degradation o f other countries. Nevertheless, Bhutan i s s t i l l a poor country with tremendous geographical barriers to development; its environment remains fragile and subject to pressure from a growing population; employment is growing too slowly to absorb the increasing number o f secondary school graduates; and there i s a s t i l l only partly-resolved dispute over citizenship claims o f rehgees o f Nepalese origin. 2. IDA i s a small player in this heavily-aided country, ranking as the 8th largest donor and providing less than 2 percent o f Bhutan’s foreign assistance during the 1990s. The Bank’s assistance strategy has recognized IDA’Slimited role and has given priority to analytical and advisory activities (AAA) while engaging in selective lending for education, forestry, and (more recently) infrastructure. Bank strategy shifted in the late 1990s, however, as the Bank actively sought opportunities for engagement and government willingness to borrow increased. Two new projects in infrastructure (urban development and rural access roads) were approved, and the Bank expressed a willingness to provide economy-wide program lending. At the same time, the Bank stated that its comparative advantage was in transferring knowledge and ideas rather than finance, and i t proposed a large program o f AAA. 3. The shift in strategy articulated in the 2000 CAS has not been implemented. In the event the govemment was not prepared for program lending and Bank involvement in many o f the proposed studies did not materialize, primarily because alternative support from Asian Development Bank (ADB) was available and because multiple donor involvement i s burdensome. M o r e generally, the government i s wary o f taking on too many studies which, even if free, have a high opportunity cost in terms o f scarce senior official involvement. The Bank has provided some useful analytical work and policy advice over the years, however. The last Country Economic Memorandum was in 1996, but in the last five years, the Bank carried out (in collaboration with the government and other donors) a Private Sector Survey, a country financial accountability assessment, an informal analysis o f the macroeconomic impact o f the hydropower boom, and informal sector work related to Bank projects. N o formal sector work has been done. I iv 4. Bank lending in Bhutan has been a basic program o f infrastructure development-roads, schools, water supply and sanitation and forest management-reflective and respectful o f the needs o f a traditional society s t i l l in the early stages o f i t s development process. Each project has included basic capacity building, usually funded by bilateral partners with grant assistance, while the focus o f the Bank’s investment has been o n building basic physical infrastructure. The lending program has been client-driven, carried out in partnership with bilateral agencies and reflective o f the country’s holistic approach to development (including its cultural and traditional values). In a sense, it has been a model o f the comprehensive development framework long before i t found i t s way into the Bank lexicon. 5. The Bank lending program has been largely effective. Five o f six completed projects (six o f seven counting a Bank-executed Global Environmental Facility project) had a satisfactory outcome. One o f the strengths o f the lending program has been the use of co-financing for technical assistance and other “soft” activities to complement the Bank’s funding for construction and other “hard” components. There have been weaknesses as well: insufficient sector work to underpin sector strategy; inadequate preparation o f two infrastructure projects; weaknesses in supervision; lack o f staff continuity; and occasional insensitivity to the special requirements o f a small country. 6. The Bank has made an important contribution to improving access and quality o f education in Bhutan. In forestry, the outcome o f i t s assistance has been mixed, with the most recent forestry project rated moderately unsatisfactory. The overall outcome o f Bank assistance to Bhutan i s rated moderately satisfactory. Sustainability i s rated likely and institutional development impact is rated substantial. 7. Some o f the lessons arising from the evaluation o f Bank assistance to Bhutan may be relevant in other small countries, or in other situations in which the Bank plays an unaccustomed small role. The Bank was effective in Bhutan when i t listened to what the government wanted rather than trying to figure out what the government needed and then trying to persuade the government to accept i t s proposals. Proposals for the Bank to serve primarily as a knowledge bank have not been successful as most o f the Bank offers o f knowledge services were not taken up by the government. Other lessons include: co- financing i s an especially useful instrument in a small country; staffing continuity i s critical for effectiveness; and, above all, i t i s essential to remain selective and focused. I t i s a mistake to try to be a “full service bank” in a small country unless the Bank i s the dominant lender. N o r i s i t easy to be a knowledge bank in a small country. The Bank needs to invest resources in order to provide useful advice based o n local conditions and to inform a selective lending program, while the government i s constrained by the limited time o f senior officials to work with the Bank V 8. The main recommendation i s that the Bank should return to the more selective approach o f i t s earlier assistance strategy, but make it more effective through prior sector work. The Bank also needs to develop ways to compensate for i t s lack o f field presence. I t could usefully follow the example o f the ADB in having annual rolling program consultations and making better use o f United Nations Development Program (UNDP) facilities to improve coordination with other donors based in Bhutan. Finally, the Bank should increase i t s sensitivity to the constraints o f a small country by reviewing its procedures o n procurement, reporting, and core economic studies. Gregory K. Ingram Director-General Operations Evaluation 1 1. Background Country Characteristics 1.1 Bhutan i s a small, isolated Buddhist kingdom with a distinctive development strategy. Because o f i t s strategy, i t s location, and i t s development constraints, Bhutan has attracted considerable outside interest and support. 1.2 Isolation. Bhutan is isolated by geography and, historically, by policy. Located in the Eastem Himalayas between India and the Tibet region o f China, Bhutan has an estimated population o f about 750,000’ within an area about the size o f - but more rugged than - Switzerland. Within fewer than 100 km the altitude varies from less than 200 meters in the South to over 7,000 meters in the North. More than half the population lives at least h a l f a day’s trek from the nearest road. Historically, Bhutanese policy has reinforced the isolation imposed by geography in an effort to preserve political and cultural independence. Until 1959 Bhutan remained inaccessible by road, and access by road and air remains limited, unreliable, and costly. 1.3 Dependence on India and hydro power exports . Bhutan’s budget, trade, and production are heavily dependent on the much larger Indian economy (900 times as large as Bhutan’s). Although smaller than in earlier years, Indian grant aid s t i l l covers nearly one quarter o f Bhutan’s government expenditure and about 10 percent o f GDP.2 Bhutan’s trade is integrated with the Indian economy by transport l i n k s (India built the first road into Bhutan and s t i l l directly maintains 30 percent o f Bhutan’s main road network) and institutional arrangements (a free trade agreement and currency links).3 India accounts for 95 percent o f Bhutan’s exports and 78 percent o f imports (three-year average). Finally, Bhutan has become increasingly dependent on the power sector. In recent years power has contributed about 11 percent o f GDP, 40 percent o f export earnings, and 40 percent o f govemment revenue. The power sector, in turn, i s also dependent o n India. India has funded several major hydro power projects through a combination o f grants and soft loans, and electricity exports go exclusively to India. ’ There i s a wide range o f population estimates. F o r 1998 the govemment estimated total population at 636,500 while the United Nations (UN) estimate was 1,948,000. The W o r l d Bank estimate for the same year was 759,150. The United Nations (UN) continues to project population f r o m the higher estimate used by the govemment at entry t o the UN in 197 1. * After Indian independence, India took over Britishbudget support to Bhutan (arising f r o m a war in the 19* century in which Bhutan was defeated by the British and ceded lowland territory to colonial India). The f i r s t two five-year plans in the 1960s were entirely financed by India. Although Bhutan has traditional cultural and trade ties with the Tibet region o f China, the border i s officially closed to trade and the potential i s limited in any case by the absence o f roads and difficult terrain. 2 - Figure 1.1: Bhutan N e t Capital Inflows from India and Power Export Revenues (as percentage o f GDP) 50 40 et capital flows from India ” 30 20 Power export revenue 10 0 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: RMA Annual Reports (various issues). l i Consists o f concessional loans and grants. 1.4 Bhutan ’s development strategy. Bhutan’s approach to development i s often defined as maximizing Gross National Happiness (GNA).4 The concept o f GNH, introduced by the King in 1980, implies an attempt to balance material progress with environmental sustainability and preservation o f Bhutan’s national independence and religious and cultural heritage. The country’s leadership wishes to preserve Bhutan’s traditional way o f life as much as possible and avoid the mistakes o f other countries, particularly deforestation and uncontrolled urbanization. In line with these objectives, Bhutan has vowed to maintain forest cover for at least 60 percent o f i t s area, not only for preservation o f i t s o w n environment but in “trust for mankind”. Similarly, Bhutan’s guarded approach to opening up to outside cultural and economic forces i s reflected in i t s restrictions on direct foreign investment, exports o f logs, and mass tourism. Nevertheless, Bhutan has gradually opened up the country and the economy in the past two decades and has enjoyed a high rate o f growth o f Gross Domestic Product (GDP), the common international measure o f material output. 1.5 Political and social issues. Bhutan i s in the process o f drafting a written constitution to establish parliamentary democracy. The country enjoys a reputation for good government, including l o w corruption and efficient administration. Land reform in the 1960s abolished serfdom and created a relatively equal distribution o f arable land. See, for example, Royal Government o f Bhutan (RGOB) Planning Commission, Bhutan 2020: A Vision for Peace, Prosperity and Happiness (1999), and Centre for Bhutan Studies, Gross National Happiness (1999). 3 Women have equal status in l a w and by tradition, because o f matrilineal land inheritance among the largest ethnic group. But there are also tensions arising f r o m an exodus in the early 1990s o f people o f Nepalese origin f r o m Bhutan, many o f w h o m ended up in refugee camps in Nepal. An agreement reached in 2003 provides for a settlement, but i t s t i l l remains to b e implemented.’ In addition, revolutionaries seeking to gain independence for Assam f r o m India operated for some years f r o m bases in southeastem Bhutan. The Bhutanese army expelled the rebel groups f r o m camps in its territory in December 2003. Overview of Economic and Social Development 1.6 Growth and macro management. Bhutan has had sustained high growth o f GDP averaging 6.6 percent per annum (4.0 percent per capita) over the past 20 years (Table 1.1). G r o w t h has been driven by power sector development and supported by a high level o f external assistance. Revenue f r o m hydro power averaged over 10 percent o f GDP during the 1990s. Aid averaged about 40 percent o f GDP during the 1980s and 27.5 percent during the 1 9 9 0 d U n l i k e some countries with large natural resource exports and foreign aid inflows, Bhutan has skillfully managed its foreign exchange and government revenue bonanza in three important respects. First, it has maintained a high rate o f investment, averaging w e l l over 40 percent in the 1990s. Second, a large share o f public expenditure has been directed toward education, rural development, and other activities which help share the benefits o f power and aid resources widely. Third, macro management has been prudent. Domestic borrowing has been l o w during most o f the past two decades, inflation has remained in single digits except during the early 1990s, and Bhutan has steadily accumulated rupee and convertible foreign exchange reserve^.^ 1.7 High investment rates and spending for social development have helped counter the tendency for the power sector t o become an enclave. Although the power sector was one o f the largest contributors to growth, other sectors have contributed more and a l l sectors have grown (see Figure 1.2). Reflecting the government’s success in promoting broadbased growth and good macro management, the Bank’s internal Country Policy and Institutional Assessment (CPIA) ratings have placed Bhutan in the upper quintile o f IDA countries in most years8 Around 100,000 people have been living in refugee camps j o i n t l y run by UNHCR and the government o f Nepal. The RGOB maintains that many o f those in refugee camps were recent illegal immigrants or may never have lived in Bhutan at all. See the Government’s comments in Annex F. These figures are based o n RGOB data. WDI data show lower, though still substantial, shares o f aid (14.3 percent o f G D P during the 1980s and 22.5 percent during the 1990s). The discrepancy arises because the WDI figures exclude Indian aid. ’ The overall budget deficit including grants rose to 11 percent o f GDP during 2001 because o f lumpy capital expenditures and delays in foreign aid receipts, but the deficit f e l l again in 2002 and 2003. Ratings were recalibrated in the last round to reflect that, while outcomes in some areas such as quality o f budgetary and financial management were high, institutional development lagged. Despite t h i s recalibration, Bhutan’s rating remained in the top quintile. 4 T a b l e 1.1: Bhutan - Selected Macroeconomic Indicators Indicator Period Averages Unit 1981-85 1986-90 1991-95 1996-00 2001-02" GDP per capita constant 1995 US$ 272.4 358.4 411.9 492.9 566.9 GDP per capita growth annual % 4.7 6.0 2.7 3.8 4.4 Exports o f goods and services current million US$ 24.9 71.3 85.9 126.8 122.2* Exports of goods and services % o f GDP 14.4 26.4 32.8 31.0 23.2* Power exports % of GDP 0.0 7.8 6.4 9.1 Imports o f goods and services % o f GDP 45.3 44.9 44.8 45.6 43.4* Aid (WDI definition) % o f GDP 10.8 17.7 28.0 17.0 12.6* Total Net Capital Inflow % o f GDP 52.0 30.3 24.7 31.5 Gross domestic savings % o f GDP 9.4 16.2 36.1 26.5 27.9* Gross capital formation % o f GDP 40.3 34.7 48.1 41.1 48.1* Overall budget balance, including % o f GDP -1.9 -5.3 -0.2 3.8 -1o.o* grants Current revenue, excluding grants % of GDP 10.6 19.6 22.7 22.7 Current revenue, including current % o f GDP 36.0 39.8 42.4 39.7 grants Inflation, consumer prices annual % 9.4 9.0 11.2 8.2 2.8 * 200 1 data only. F i g u r e 1.2: Composition o f GDP at F a c t o r Cost in 1980 Prices, by Sector 4500 €7Electricity 4000 Ei Other* 3500 0 Community, social & personal m i c 3000 (Government administration and - h defense) E 2 NMining, quarrying and manufacturing 2500 z * 0 ln 5 2000 0 Forestry - .- - .- 3 1500 IIIAgriculture and livestock 1000 *includes: construction; wholesale and retail trade, restaurants and hotels; 500 transport, storage and communications; and financing, insurance, real estate and business services. 0 5 1.8 Despite overall good management and performance, Bhutan s t i l l faces a number o f challenges. Foremost among these i s the challenge o f providing jobs for a growing labor force. Neither the power sector nor associated power-intensive industries provide many jobs. The government sector has also ceased to expand employment just as large numbers o f secondary school graduates with high expectations are coming into the labor market. A second, related challenge i s to create a policy environment for private sector development. Non-power exports have declined in recent years. This appears to have more to do with restrictions on exports o f unprocessed timber than with exchange rate overvaluation (“Dutch disease”), but i t i s a concern nonetheless. Finally, grant aid i s likely to decline in future. Although this i s a problem arising from Bhutan’s success, the projected decline poses a challenge for debt and public expenditure management. 1.9 Social development and poverty reduction. Bhutan’s social indicators have improved significantly over the past 20-25 years (see Table 1.2). Between 1977 and 2000 l i f e expectancy rose by 43 percent, child mortality was nearly cut in half, and the adult literacy rate tripled. Improvements reflected both economic growth and government programs in health, education, and road-building to improve access to economic opportunities and social services. The advances are all the more notable because o f the scattered population and difficult terrain. For example, for most children in rural areas secondary schooling requires boarding. The government provides free room and board as well as free tuition at all level^.^ Basic health services are free and are provided by 454 outreach clinics (in 2000) as well as 160 basic health units and 28 hospitals. The biggest increases in social indicators came in the earlier years, but indicators have continued to improve and the government has stated that i t hopes to meet most Millennium Development Goals ahead o f time (before 2015). Table 1.2: Bhutan Social Indicators 1977 1990 1994 2000 Life expectancy at birth (years) 46.1 66.1 66.1 Population growth rate (%) 2.3 3.1 2.5 Total fertility rate (birthsiwomen) n.a 5.6 4.7 Infant mortality rate n.a 90 70.7 60.5 Under 5 mortality rate 162 123 96.9 84 Immunization coverage (%) 84 90 Access to clean water 31 45 57 77.8 Number o f basic health units 31 71 71 160 Number o f primary schools 92 250 T o t a l enrollment in schools (000) 14.6 70.3 70.3 109.9 Gross primary enrollment rate 55 72 Adult literacy rate (%) 18 48 54 Source: Royal Govemment o f Bhutan, Bhutan 2020: A Vision for Peace. Prosperity and HapDiness and RGOB, Bhutan 2003: People at the Center o f Development, Eighth Round Table Meeting, February 2003, Geneva. But as the Bhutan 2020 document cautions: “Although free basic education and health care are today reaching these pockets o f disadvantaged and vulnerable groups, the need t o contribute to the construction and maintenance o f buildings, the cost o f travel, contributions t o school welfare funds, the cost o f school and so forth, while modest, are sufficient t o place the services beyond the reach o f the poorest farmers and their families.” 6 1.10 M a n y of these estimates are subject to a wide range o f error because o f uncertainty over total population size (see Table 1.3). The U ” s estimate o f total population i s three times higher than that o f the government. Depending o n whether indicators are based o n samples or o n gross numbers, differences range from relatively minor (less than 10 percent for life expectancy) to enormous (a six-fold difference for enrollment rates). Amidst these confusing estimates, two conclusions stand out. First, there i s less disagreement about the trend than the level o f social indicators. All sources agree that social indicators have improved significantly over time. In most cases, even the levels are not that different. For example, UNDP’s 2003 Human Development Report places Bhutan among the group o f countries with a medium human development index (HDI = .5 11) even though Bhutan is s t i l l a low-income country. Second, however, the conflicting estimates will not be cleared up until there i s a population census and other improvements in data. Table 1.3: Bhutan-Discrepancies between UN and RGOB Estimates o f Social Indicators 199 7 Estimates UN RGOB L i f e expectancy a birth (years) 60.7 66.1 Adult literacy (%) 44.2 54 Combined enrollment ratio 12 72 Real GDP per capita (PPP$) 1,467 1,534 H u m a n Development Index 0.459 0.581 Source: Bhutan -National Human Development Report 2000. Note: As noted in the report, the differences in estimates arise from the fact that RGOB uses estimates based on the most recent demographic survey whereas the U N ’ s estimates are based on demographic projections. Related to this fact i s the difference in population estimates: UN estimates the population to be 1.9 million whereas RGOB estimates i t to be 0.64 million in 1998. 1.11 The level and trend o f poverty are more difficult to assess. A Bank report succinctly voiced the view that has long been the conventional wisdom o n poverty in Bhutan: “few Bhutanese are hungry, landless or homeless.”” But while few Bhutanese face severe deprivation, a 2001 Asian Development Bank pilot household survey found a poverty rate ranging from 25.3 percent to 36.3 percent o f the population (Table 1.4).” Poverty i s largely a rural phenomenon and i s associated with limited cultivable land holdings, l o w productivity o f land, and poor access to markets and economic opportunities. Poverty also differs by region, with a higher share o f poor people in the far North and in the Southeast. 1.12 Environment. Bhutan’s environmental management has effects far beyond i t s borders. I t i s a major repository o f bio-diversity for the w o r l d as a whole and i t s seven river valleys feed the Brahmaputra system. Bhutan’s watershed management thus affects lo Bhutan: Development Planning in a Unique Environment ( W o r l d Bank, Report No. PuB7189, 0 1/3 1/89>. ’I The poverty lines were calculated by grossing up a f o o d poverty l i n e based o n the cost o f a basket o f 32 items w h i c h would provide 2124 calories per person. The f o o d poverty l i n e was Nu. 458.9 per capita per month, using current prices in the capital, Thimphu. The “lower poverty line” was Nu. 612.1 per month (equivalent to about PPP $1.10 per day) and the “upper poverty line” was Nu. 748.1 per m o n t h (PPP $1.35 per day). See RGOB, Millennium Development Goals Progress Report 2002. PPP equivalents estimated using WDI 2002 data for the year 2000. 7 not only the sustainability o f i t s own hydropower industry, but also helps protect downstream users in India and Bangladesh from erosion and flooding. - Table 1.4: Bhutan Poverty Incidence in 2000 Category Sub-category Units 2000 Note Poverty Incidence Lower Poverty Line (Nu612.limonth) 25.3 Upper Poverty L i n e (Nu748.Vmonth) 36.3 % o f Poor to Total Population Lower Poverty Line (Urban) % 2.9 Lower Poverty Line Rural % 29 Upper Poverty L i n e (Urban) O Y 6.4 Upper Poverty Line (Rural) % 41.3 Poverty Gap (% o f GDP) Lower Poverty Line % 7.0 Upper Poverty Line O Y 11.0 Source: ADB Country Strategy and Program Update (2003-2005). Bhutan. July 2002. Table Al.2. 1.13 Bhutan has earned widespread recognition for i t s conservation efforts. Under the Bhutan Trust Fund for Environmental Conservation (see B o x 3.2 in section 3 below), one-quarter o f the country’s area was set aside for conservation, and the government made a recent “Gift to the Earth” o f another 9 percent for wildlife corridors to prevent habitat fragmentation. In the past decade forest cover has increased from 65 percent to 72 percent o f the national territory, well beyond the national target o f 60 percent. The emphasis placed upon conservation o f the environment i s illustrated by the following statement given in the Bhutanese Environmental Assessment Sectoral Guidelines, produced in1997 by the National Environment Commission: Over the past two decades there has been an increasing recognition by the RGOB that i f forests were not to go through a process o f deterioration similar to some o f Bhutan’s neighbors, a systematic forest management program would have to be put in place. The management program should be based on a balancing o f conservation and economic development goals through long-term, sustainable, multi-purpose forest management. This has been the focus o f a revised forest policy drafted in 199 1. The primary focus o f the new policy i s to ensure conservation o f the environment, and only thereafter aim at derivation o f economic benefits from the forest as a rationally managed resource. 1.14 Despite i t s strong conservation record, Bhutan faces growing pressures o n the environment. Overall population density i s low, but Bhutan’s rural population per unit o f arable land i s higher than that o f India and Pakistan, and Bhutan has the highest rate o f population growth in the region.’* Especially in Eastern Bhutan farmers are bringing steep, marginal, non-arable land under cultivation. Urbanization i s also contributing to ecological pressures. With very little land suitable for urban growth and infrastructure-even finding a suitable site for a landfill for a small town i s a major challenge-urban growth i s largely at the expense o f forested slopes, agricultural land, or wetlands. Water erosion now affects 10 percent o f agricultural land and localized water shortages have been increasing due to growing sediment loads in rivers. Bhutan also faces the unusual environmental r i s k o f glacier lake outburst floods. The most recent o f these (in 1994) killed 22 people and affected hundreds o f others. A recent United l2 The comparisons are based upon WDI figures, but they would s t i l l be true using the government’s lowe1 figures for population total and growth. 8 Nations Environmental Program (UNEP) sponsored study identified 24 glacial lakes in Bhutan (and 20 in Nepal) as potentially dangerous. 1.15 Bhutan has a good record o f environmental management, but the fiagility o f i t s environment requires special vigilance and poses difficult choices for development strategy. 2. The Bank's Role in Bhutan Context: The Bank as a Small Player in a Small Country 2.1 The World Bank i s a small player in a heavily aided country. During the 1990s IDA was only the 8th largest donor in Bhutan. The Bank's small role partly reflected the government's preference for using the grant aid which most bilateral donors provided. Overall, more than 90 percent o f assistance to Bhutan during the 1990s was on grant terms. But IDA also disbursed less than 60 percent as much as the Asian Development Bank, the other major source o f concessional 10ans.l~ 2.2 The overall level o f aid to Bhutan was about $100 per capita (give or take 12 percent)-high compared to neighboring South Asian countries but similar to other small countries (see Figure 2.1).14 IDA disbursements to Bhutan during the 1990s (averaging $1.80 per capita per annum) were lower than to comparable small low-income countries, and lower even than per capita Bank assistance to much larger low-income countries such as Bangladesh and Pakistan. However, from 2001-2003 IDA disbursements to Bhutan increased nearly four-fold to an average o f $6.80 per capita per annum. The government has expressed interest in continuing to borrow more from IDA than in the past because o f uncertainty over the future level o f bilateral grant aid.15 2.3 IDA has provided ten credits totaling $95 million to Bhutan in the past twenty years, for education, forestry, industry, rural roads, urban development and technical assistance (Table 2.1). The Bank was also executing agency for a $10 million GEF grant.I6 Lending has been episodic, coming in batches o f closely-spaced loans (within an 18 month period): three in the early 80s, two in the late 8Os, and three in the late 90s, with several years o f no new lending in between. The most recent education project (Education Development Project in FY04) i s expected to be followed by projects for l3Based o n Royal Monetary Authority, Annual Report 2001/2002, for the years 1992193-2001102. Other m a i n donors were India, Denmark, Japan, Asian Development Bank (ADB), Austria, Switzerland (Helvetas and Swiss Development Corporation), Netherlands, UNDP, and EEC. On a grant element basis, IDA was the 12" largest donor. See Annex A for details. l4According to WDI figures, aid per capita t o Bhutan averaged $89.8 per annum during the 1990s, but government data show an average o f $1 12.1 for the same period. The discrepancy arises because government figures have a more complete coverage o f Indian aid. l5Denmark plans to phase out i t s grant aid to Bhutan because o f the increasing availability o f revenue f r o m hydropower exports. Bhutan also considers the future level o f Indian aid uncertain. l6IFC also approved an investment loan in December 2003 o f up t o $10 m i l l i o n in Bhutan Resorts Corporation Limited. Bhutan became a member o f IFC in 2003. 9 HIV/AIDS, decentralized rural development, and private sector development, plus a new GEF grant for land management. Figure 2.1: Comparison of Bank Lending and Foreign Aid Flows to Small and Low-income EJ Aid Per Capita2 IDMBRD Net Disb. Per Capita zw.0 180 0 160.0 140 0 1200 100 0 80 0 60 0 40 0 20 0 00 ” R G O B data. All other data are from World Bank Development Indicators. R G O B data show a higher level of aid because they have a more complete coverage o f Indian aid. GNIF’C 2002 (Gross National Income Per Capita). Table 2.1: Bank Group Lending to Bhutan 1983-2003 Project Amount ($M) Approval Date Status Technical Asst. 3 .O 10118/83 Closed Forestry I 5.5 4/17/84 Closed Calcium Carbide 9.0 5116/85 Closed Primary Education 4.2 4119/88 Closed Forestry I1 1.1 4/19/88 Closed Third Forestry Devel. Project 5.4 7116/93 Closed Education I1 13.7 3/3/98 Active Urban Development 10.8 12/21/99 Active Rural Access 11.6 12121/99 Active Educ. Devel. Project 31.0 8/21/03 Active Memo: Trust Fund for Env. Cons. 10.0 (GEF grant) Closed 10 Bank Assistance Strategy 2.4 The Bank’s assistance strategy for Bhutan, articulated first in 1993 and again in 2000, i s aligned with the government’s development goals. l7 Bank strategy statements have consistently stressed three themes: the Bank’s limited role; priority for analytical and advisory activities; and selective lending. Practice has not always followed theory, however. 2.5 Limited role. The Bank has always acknowledged that it has to adapt i t s strategy to fit IDA’s role as one o f Bhutan’s smallest foreign aid donors.’* During most o f the 1980s and 1990s this meant responding to government requests for projects rather than actively promoting lending. Because o f the availability o f grants, the government did not even take up the IDA allocation to which i t would have been entitled on grounds o f income and performance. The Bank was also content to let UNDP and ADB take the lead among multilateral institutions operating in Bhutan.” This changed in the late 1990s, however, with the change in regional management. Although Bank strategy statements s t i l l stressed IDA’Slimited role, the Bank actively sought opportunities for engagement, while government willingness to borrow increased. As a consequence, the IDA lending program was ramped up significantly. T w o projects (Urban Development and Rural Access) were presented along with the CAS in FYOO, and the third education project, the Educational Development Project (EDP), was approved by the Board in FY04. 2.6 Priority for analytical and advisory activities ( M A ) . Analytical work has always figured prominently in the Bank’s assistance strategy for Bhutan. In the first decade o f i t s relationship with Bhutan, the Bank wrote two economic reports and advised the government o n sector strategies in forestry and basic education. Economic policy advice was one o f three proposed priority areas in the Bank’s 1993 strategy (the other two were natural resource management and primary education). Because o f limited resources, the Bank proposed focusing AAA o n budget analysis. The 2000 assistance strategy was even more explicit o n the priority o f AAA. IDA’s comparative advantage was seen as transfemng knowledge and ideas rather than finance. Emphasis on analytical work also made sense in terms o f filling the Bank’s information and knowledge gaps resulting from limited past involvement in Bhutan. The proposed program to implement the 2000 strategy included potential AAA o n a vastly expanded range o f topics.20 ” The 2000 CAS, prepared in 1999, stated that I D A ’ s mission i s “to help Bhutan manage its considerable natural, institutional, cultural and environmental assets in ways that can translate economic growth into rapid and inclusive social transformation and improvements i n l i v i n g standards.” l8 The 1993 country assistance strategy envisioned a limited and selected role for IDA in Bhutan’s development. The 2000 CAS said that the Bank has a highly selective mission to fit IDA’s continuing role as one o f Bhutan’s smallest donors. l 9 According to a former senior manager in the W o r l d Bank’s South Asia Region, there was n o formal understanding that ADB would be the lead multilateral lending agency, but the W o r l d Bank was happy to n the smaller countries o f South Asia. p l a y a secondary role i 2o The Region has commented that the proposed AAA program reflected government requests and was indicative. See footnote 1 in Annex B 4 o f the C A S document and the Region’s comments in Annex E. 11 2.7 Selective Lending. The bulk o f Bank lending has been for education, forestry, and infrastructure. The 1993 strategy limited lending to education and forestry. The strategy introduced in 2000 stressed that the Bank would continue to be selective and to be, in effect, the lender o f last resort--IDA lending would be limited to sectors and activities for which grant fimding was not readily available. But the scope o f project lending could be expanded to cover infrastructure, e.g. the two new projects for urban development and rural access roads, as w e l l as education and forestry. Moreover, the Bank expressed a willingness to provide economy-wide program lending, provided this was preceded by a public expenditure review, baseline poverty assessment, macroeconomic monitoring in collaboration with the IMF, and a Poverty Reduction Strategy Paper (PRSP). 2.8 Gap between theory andpractice. There was a marked break in the Bank’s assistance strategy for Bhutan in the late 1990s. The rhetoric o f selectivity and the Bank’s limited role remained the same as in the early 1 9 9 0 but ~ ~ in practice the Bank embarked on a much more ambitious program. The overall level o f lending increased sharply; the proposed program o f AAA expanded enormously; and the Bank moved into new areas o f lending, and even raised the prospect o f program lending, while at the same time saying that lending would remain a secondary activity. 2.9 The strategy articulated in 2000 has not been implemented. Although the Bank has expanded into lending for infrastructure and has other projects planned, nothing came o f the proposed program lending. Moreover, the Bank’s preparation o f the two infrastructure projects in FY2000 was hurried.21Finally, the proposal to expand E S W and effectively act as a knowledge bank was never taken up by the government. 2.10 In some respects, the Bank’s position in Bhutan i s similar to i t s position in many middle income countries. The Borrower has alternative sources o f finance and a clear sense o f what i t wants from the Bank.22 The Bank i s eager to be responsive, but risks being marginalized. One response to l o w demand for borrowing i s to pursue a knowledge bank role. But it i s difficult to construct an assistance strategy based entirely on a role as a knowledge bank in any country, and even more difficult in a low-income country such as Bhutan for several reasons. First, Bhutan lacks the capacity to interact with the Bank on a wide range o f AAA (see below). Second, relevant knowledge o n most issues depends o n familiarity with local conditions as w e l l as international experience. Except in sectors o f active lending, the Bank lacks this knowledge in Bhutan. Third, the ADB with i t s larger involvement and i t s annual allocation o f $1.5 m i l l i o n o f technical assistance grants to Bhutan i s much better placed to provide a wide range o f AAA and TA than the World Bank. 2’ Some interviewees in the Bank saw this as an effort t o increase the amount and scope o f lending. The Region commented that the two projects were prepared rapidly to respond t o the government’s request. See Annex E for further details and for Regional Management’s comment o n program lending. 22 See the Region’s comments in Annex E. As in most countries, line agencies m a y have their o w n agendas. In preparation for the CAS, some agencies were asking the B a n k for projects or studies w h i c h were subsequently turned down by the Ministry o f Finance. The central authorities, too, sometimes change their views, but compared t o most countries, the government has a clear idea o f what it wants f r o m the Bank. 12 2.11 The effort to increase the relevance o f the Bank by expanding i t s role has not worked out. The government has, in effect, asked the Bank to continue its historical role of lending to selected sectors and providing AAA o n a few issues. This is a realistic assistance strategy for the Bank in Bhutan and one which the Bank has performed reasonably well. The outcome o f the Bank’s evolving strategy i s evaluated in the following two sections. 3. Implementation o f the Bank’s Program 3.1 Four areas o f operational emphasis have dominated the Bank’s program in Bhutan over the past decade: economic policy advice through analytical and advisory activities (AAA), and lending in education, forestry, and (more recently) infrastructure. This section will evaluate progress in implementing the Bank’s strategy through AAA, lending, and cooperation with other donors. Analytical and Advisory Activities (AAA) 3.2 The Bank has produced four country reports o n Bhutan: in 1983, 1989, 1994, and 1996.23 Since 1996 only one formal report has been produced: Bhutan: Private Sector Survey (published June 14,2002). The Bank also produced a country financial accountability assessment (CFAA) in 2002 as part o f i t s fiduciary safeguard work. There has been n o formal sector work. 3.3 The economic reports provided a good overview o f Bhutan’s development progress and issues. The 1983 report i s especially noteworthy for i t s highly informative tour deforce, rich in institutional and sector detail. Several themes recurred in the reports, with varying degrees o f emphasis: Human resource development, not finance, i s the main constraint on Bhutan’s development. Considerable unexploited potential exists for forestry development. This view was especially prominent in early reports, less so later. Fiscal management i s the main macroeconomic constraint. Increased revenues are vital for funding recurrent expenditure domestically. Expenditure planning needs to take better account o f the recurrent expenditure implications o f capital investments. The government needs to encourage private sector development to diversify the economy and provide jobs for a growing labor force. This theme became increasingly prominent, culminating in the Private Sector Survey. Poverty reduction i s a matter o f growth and public expenditure for social services. As the 1994 C E M put it: “There appears to be relatively little scope for reducing poverty through redistributive measures. This suggests the 23 The first two were published: Bhutan: Development in a Himalayan Kingdom (PUB4307 01/31/84), and Bhutan: Development Planning in a Unique Environment (PUB7189 01/31/89). The 1994 and 1996 reports are unpublished: Bhutan: Country Economic Memorandum (Report No. 13495, 11/29/94), and Bhutan: Country Economic Memorandum (Report No. 16113, 12120196). 13 solution to poverty has to come through economic growth and development o f the human capital base. The latter will require prudent fiscal policies to ensure that resources are available to properly fund social services for the poor.” 3.4 Bhutan’s achievements in fiscal management and provision o f social services have been substantial. (Achievements in education and forestry are discussed below). Domestic tax revenue has risen from 7.2 percent o f GDP in 1993 to 9.9 percent in 2001, though total domestic revenue (excluding grants, but including power royalties) declined marginally from 23.8 percent o f GDP to 22.9 percent over the same period. I nmost years there has been a surplus o f domestic revenue over recurrent expenditure. Education and health account for 35 percent o f recurrent spending and 25 percent o f overall budget outlays. 3.5 What i s unclear i s how much influence the Bank had o n economic policy. The 1993 assistance strategy foresaw the Bank complementing IMF macroeconomic advice through periodic review o f budget documents to assist the government and other donors in planning and fiscal management. In the event, Bank advice in this area seems to have been marginal. The Bank offered advice o n recurrent costs and budgetary methods which prompted some discussion and probably brought some benefit. The authorities asked for advice o n two issues: the exchange rate regime and management o f the budget deficit. O n the first issue, the Region’s Chief Economist gave advice which was well-received. On budget management, the Bank responded by trying to develop a RMSMX model for projections but this came to little because o f data problems. 3.6 The 2000 assistance strategy stressed the Bank’s comparative advantage in transferring knowledge and ideas rather than finance. The CAS stated that the government had asked for consultations o n macro policy, financial sector strategy, long- t e r m health and education strategies, and accounting and auditing standards, and for assistance in poverty generation and analysis. In addition, the Bank proposed possible analytical and advisory work o n private sector development, foreign direct investment, trade and investment reform, public expenditure management, employment analysis, school curriculums, rural road capacity development, and policy frameworks for economic and sector development. 3.7 The Bank did very little o f this work. Although one government agency had expressed interest in financial sector work from the Bank, another had asked the ADB to do a financial sector study. In the end, ADB was asked to do the study and the World Bank gave informal comments. ADB also did the poverty survey, which has proved to be very influential in changing perceptions about the nature and extent o f poverty. The Bank did carry out a private sector survey in collaboration with the Ministry o f Trade and Industry, the Bhutan Chamber o f Commerce and Industry (BCCI), and UNDP. This study provides useful advice o n a number o f important issues and was very well received by both B C C I and the ministry, though some officials in the Ministry o f Finance felt that the report was more a u s e f i l summary o f what was known about private sector development than a source o f new ideas. The Bank also did an informal study o f the macroeconomic impact o f the hydropower export boom in 2002. This highly relevant 14 and carefully-done study concluded that Bhutan has avoided the “Dutch disease’’ phenomenon in which a natural resource boom (or large foreign aid inflow) discourages production and investment in other tradables. However, this i s an issue which bears revisiting as better data o n prices become available, especially in view o f recent export trends and urban property values. Finally, the Bank conducted a country financial accountability assessment in 2002 with the support o f the Dutch government. Although there i s a culture o f fiscal conservatism and c i v i l service probity in Bhutan, i t i s recognized that institutional procedures need to be developed to assure accountability and transparency in a more complex economy. The report provides several recommendations to that end, and has been followed by an IDF grant for institutional strengthening. 3.8 In brief, the Bank has provided some useful economic policy advice to the government, but the scope o f activity has been far less than anticipated. Why was the World Bank not involved in the finance and poverty studies? Primarily, it seems, because the government does not want donors separately involved in studies. It i s too costly and burdensome for a small country with a small c i v i l service. Since Bhutan has a long relationship with the bigger donor (ADB), the government opted for ADB support. More generally, the government i s wary o f taking o n too many studies which, even if free, have a high opportunity cost in terms o f scarce senior official involvement. The Bank now seems to have recognized the danger o f study overload. I t has limited its proposed economic work to public expenditure analysis, which i s an area in which the Bank i s experienced and could provide useful support in helping integrate recurrent and capital budgets into a medium term expenditure framework. Moreover, the Bank has offered to adopt a low-key, informal approach-more like TA than AAA-which recognizes capacity constraint^.^^ Capacity constraints also have implications for the PRSP process (Box 3.1). Box 3.1. Bhutan’s PRSP For s m a l l countries with capacity constraints the PRSP process presents a special challenge. In Bhutan, it also posed a risk o f strategy fatigue. The Vision 2020 exercise supported by U N D P had already developed sector strategies and an overall strategy o f growth and equitable development. These underpinned the government’s own 9* Five Year Plan, which was a highly inclusive process o f bottom-up decentralized planning. The government was understandably hesitant, therefore, to embark o n a PRSP. But the Bank and Fund agreed with the government that the 9FYP was, in effect, the government’s PRSP, with the proviso that the government prepare a cover note placing the Plan within a macro framework and outlining a monitoring and evaluation system. This agreement has encouraged Bhutan t o go forward with a PRSP. Bhutan i s using the PRSP t o round o f f and complete the strategic planning process started in Vision 2020 and the 9FYP at minimum cost. This is a sensible approach, and i t may be worth considering using a similar approach more pro-actively in future. This would have higher costs, but also higher benefits. The cost would be the opportunity cost o f doing other activities. Potential benefits over the medium t e r m could be: (a) sharpening the focus o f the Five Year Plans to relate these t o reform priorities, including the allocation and management o f a l l public spending, both investment and recurrent; (b) focusing attention o n better systems for monitoring income and non-income poverty trends; and (c) helping to shape donor programs and align them with the government’s priorities. This hybrid approach could serve as both a Plan and a PRSP. Bhutan will have t o decide h o w m u c h effort it wants t o put into the planningPRSP process t o maximize net benefits. 24 The government has recently asked for B a n k TA for development o f multi-year budgeting capacity. 15 3.9 The Bank’s AAA, especially i t s economic and sector work (ESW), has objectives other than to influence government policy, notably to inform Bank management, other donors and the outside world about development issues and progress, and to lay the foundation for the Bank’s strategy and lending. On most o f these other dimensions the Bank’s ESW in Bhutan has been effective for a small country program. Economic reports have usually been tabled as background documents for Donor Roundtable meetings. Other donors found these valuable. Some o f the Bank’s informal ESW, such as the study o f Dutch disease and the work o n forestry, has also proven useful to other donors. 3.10 The main shortcoming o f the Bank’s AAA in Bhutanhas been inadequate sector work to provide a grounding for the Bank’s lending and policy dialogue.25 Over the course o f involvement in education and forestry, the Bank has produced a useful body o f analytic and strategic work, but it has tended to be too little, too late as well as too scattered and inaccessible to provide an adequate basis for institutional memory.26 There have been signs o f improvements recently. With the Private Sector Survey and an informal transport sector note, the Bank seems to have improved the link between analysis and lending. Lending 3.1 1 Bank lending in the two decades o f involvement in Bhutan has been concentrated o n education, forestry and infrastructure. Apart from an early technical assistance operation to build local government capacity for development and one foray into industry (a calcium carbide operation with a large forestry component), all ten Bank-financed projects have been to help build schools, manage forest resources or build local roads and urban infrastructure. 3.12 This concentration reflects the government’s preference for using grants wherever possible. IDA credits are sought primarily for infrastructure investment such as school construction because grant funding i s less readily available for infrastructure. This has also often led to a division o f labor within projects in which IDA finances construction and other “hard” elements while a co-financier funds technical assistance (TA) and other “soft” elements with grants. 25 The Region has commented (see Annex E) that “the Bank i s faulted for pushing AAA, but i t i s recommended that sector work should be increased upfront (thus implicitly in a ‘supply-driven’ m od) e- although traditionally the government has not been interested in this type o f Bank assistance.” However, as noted above, ESW has multiple purposes and audiences. The B a n k needs sector work to formulate a strategy and analytical framework for projects. I t i s as m u c h a part o f the way the B a n k does business as appraisal reports and safeguard procedures. 26 The Bank dropped a proposed “Natural Resources Management Policy Review” after i t became clear that the government is not interested in further Bank lending. This should b e reconsidered. Bhutan i s a small country, but an important one for forestry issues. The Bank could fulfill its knowledge bank role by trying to distill lessons for Bhutan, other countries, and Bank sector assistance strategy. Since a considerable amount o f informal work has already been done, the marginal cost o f doing such a study should be relatively low. 16 3.13 Table 3.1 gives OED’s ratings for completed projects (including the GEF project). The overall impact o f the program has been positive with outcome ratings above the Bank average for the past decade. Table 3.1: OED Review Ratings Project Outcome Institution a1 Development Sustain ability Technical Assistance Satisfactory Substantial Likely Forestry I Satisfactory Substantial Uncertain Forestry I1 Satisfactory Substantial Uncertain Education I Satisfactory Modest Likely Calcium Carbide Satisfactory Modest Uncertain Third Forestry Moderately Unsat. Modest Non-evaluable Memo: Trust Fund for Env. Satisfactory Substantial Likely Education 3.14 Access to education has expanded enormously since the introduction o f a modem educational system 40 years ago. Gross primary enrollment has increased to 82 percent and secondary enrollment (grades 7-10) to 49 percent;27 the proportion o f girls has risen to 47 percent in primary and 45 percent in secondary; and the quality o f education i s high by regional standards. Challenges remain, however. Enrollment remains below the South Asian average because o f the difficulty o f serving a remote and scattered population. I n addition, only 7 1 percent o f children entering primary school complete the full cycle; annual repetition rates are 12-13 percent; and performance o n public examinations has declined somewhat in recent years. These problems reflect the lag in supply o f qualified teachers in the face o f rapidly growing enrollment. They also suggest that the education authorities are trying to maintain high standards, but they raise concems nonetheless. Finally, there i s a growing problem o f unemployment among secondary school leavers despite a shortage o f skilled labor. This reflects some combination o f unrealistic expectations and a shortcoming o f schools in providing needed skills.28 3.15 The Bank has been one o f the largest donors in Bhutan’s education sector.29 H a l f the Bank’s lending commitments in Bhutan have been for education ($48.9 m i l l i o n o f $95.3 million total commitments). The third and most recent project (Education 27 F r o m the latest government figures using a population base o f 600,000. Using population figures f r o m the 2001 Labor Force Survey, the estimates are about 5 percent lower. Education at a l l levels i s free. The medium o f instruction i s English. 28 The 2003/2004 budget speech notes: “The perception o f an emerging unemployment crisis arises largely f r o m the mismatch between the s l u l l s o f the school leavers and the jobs available in the economy. W h i l e most school leavers aspire for white-collar desk jobs in the c i v i l service, the government cannot absorb a l l j o b seekers, as i t would lead to an increase in the bureaucracy and additional costs o n the economy.” 29Bankassistance i s estimated t o have been 15-20 percent o f external aid to education and i t s share will increase in the next five years. Other donors active in education include Switzerland, India, Denmark, Canada, Australia, Japan, UNICEF, and U N E S C O (see Annex A, Table A.4 for activities by donor). ADB has been a large donor in education, with a program similar in size t o that o f IDA, but the government has decided not to borrow for education f r o m ADB in future. 17 Development Project for $3 1 million approved in June, 2003) alone accounts for nearly one-third o f IDA’Scommitments in Bhutan. 3.16 The basic objectives o f all three projects has been to increase capacity at the primary and junior high school level to meet the demand o f a burgeoning school-age population (school enrollment i s growing at about 8 percent per annum) and improve education quality through enhanced teacher preparation, improved learning processes and assessment and stronger management. The first two projects were co-financed with the Swiss Development Corporation (SDC), who financed teacher training and other “software” elements. Bank finds have mainly been used for construction o f a total o f 23 primary and junior high schools, accommodating over 14,000 students or more than 13 percent o f the total number o f students in grades K-12 in the country. Over 80 percent o f the recently-approved EDP will also be devoted to school construction. The project has no external co-financing, though the Bank i s coordinating with Switzerland, Denmark, and other donors who are financing complementary projects. 3.17 The contribution o f funding school construction should not be disparaged. The government finds i t difficult to obtain grant funding for construction and i s very appreciative o f the Bank’s support. OED’s recent Ghana education impact evaluation found that building quality had a significant impact o n education outcome^.^' The Bank has shown attention to quality and respect for Bhutanese tradition in i t s building designs. The schools built with Bank finding under the first two projects were constructed in traditional stone and wood designs, some with elaborate paintings o n both facades and interior spaces. The additional costs compared to “standard” designs were not estimated, but may have added 15-20 percent to the basic cost. 3.18 Although the bulk o f IDA lending goes for construction, the Bank has been engaged in all aspects o f educational development. The Bank has done n o formal sector work in education, but it has accumulated considerable sector knowledge through project- financed studies and i t played a key advisory role in helpingthe government develop i t s education strategy in 2000. Bank projects have supported education improvements including continuous assessments, in-service training, examinations and local language textbooks. The decentralization o f education administration i s also being supported. There has been no separate project implementation unit for Bank projects, which remained the responsibility o f several departments within the Ministry o f Education. This has helped promote an integrated approach to education development issues. 3.19 The main shortcoming o f Bank support for education has been the lack o f formal, up-front sector work. There have also been weaknesses in project implementation. Monitoring and evaluation o f education outcomes have been weak, despite attempts to address M&E systems in Bank projects. Construction delays have also slowed the implementation o f Bank projects.31 The first operation took nine and a h a l f years to be completed, including an 18 month extension o f the original closing date. The second ~ 30 OED (2004). “Books, Buildings and Learning Outcomes: An Impact Evaluation o f World Bank Support to Basic Education in Ghana”. 31 There i s a shortage o f local stone masons to construct the traditional stone and wood buildings. The government has been reluctant to give work permits for Indian workmen to accelerate construction. 18 project has been extended two years beyond i t s original closing date to June 30,2005, in order t o complete all the schools. 3.20 The main strength o f Bank support for education has been i t s continuous involvement in the sector. The two completed education projects have largely met their objectives and the sustainability o f their reforms i s likely, due to the ownership o f the program, clarity o f goals, and the strong partnership among donors and the Ministry. The new EDP i s designed to address the weaknesses found in the first two projects. The project will introduce a new prefabricated building design and technique to speed construction while preserving the traditional appearance o f school buildings. The project will also help redesign the mathematics curriculum, improve M&E, and conduct tracer and other studies to evaluate the relevance o f the curriculum for the local labor market. Although it i s too early to evaluate i t s impact, the project design appears relevant and responsive to difficulties identified in earlier projects. Forestry and natural resource management 3.21 Forests play a central role in Bhutan’s socio-economic development. Forestry activities account for about 11 percent o f GDP, 3 percent o f government revenue through product sales and royalties, and 20 percent o f export revenue, second only to electricity. Rural households are dependent o n forests for fuel (Bhutan has the highest per capita consumption o f firewood in the world), building material, and cattle grazing. Indirectly, forests are essential for protection o f soil and water regimes as well as to hydropower generation. 3.22 The government gives high priority to protecting the environment and has therefore followed a cautious policy o f commercial exploitation. The area covered by forest has actually increased over the past decade and i s currently estimated at 72 percent o f total land area. The Department o f Forestry Services (DFS) i s responsible for forestry management and regulation. A semi-autonomous public entity, the Forestry Development Corporation (FDC), i s responsible for commercial exploitation, though some logging operations are contracted out to private operators. At the same time, the government sees forests as a development asset and wants to promote wood industries, primarily through the private sector. 3.23 A critical issue for development strategy i s whether Bhutan has the balance between conservation and economic exploitation right. Or to put i t more restrictively, within the constraints o f sound conservation, i s Bhutan effectively exploiting the sustainable economic potential o f i t s natural resources? 3.24 The Bank has generally argued that Bhutan has not effectively exploited the sustainable economic potential o f i t s forests. The first two economic reports argued that potential was far greater than production, and early projects sought to promote plantation forestry. In retrospect, it seems clear that the Bank has a history o f over-estimating the potential for expanding the country’s commercial forestry production by failing to 19 recognize significant logistical and manpower constraint^.^^ Beginningin the early 199Os, the Bank began to give more emphasis to using forest resources to improve rural livelihoods through community-based natural resource management (social forestry), though it continued to support improved forest management practices and development of wood-based industries. 3.25 Although the Bank’s economic reports addressed forestry issues, there was no formal sector work. Informal sector work addressed pricing and marketing issues (see Annex B) as well as overall sector strategy, but the main instrument o f Bank strategy was projects. The Bank has made three loans for forestry, in FY84, FY88 and FY94 for a total o f $12.0 million. The last two projects were co-financed with the Swiss ~~ Development C ~ r p o r a t i o n .In addition, the Bank was executing agency for a Global Environmental Facility grant o f $10 million for the Trust Fund for Environmental Project (see Box 3.2); supported a large forestry component under an early industrial sector project (the Calcium Carbide Project in 1985); and provided an IDF grant (Forestry Institutional Development Initiative) for $500,000. Box 3.2: Trust Fund for EnvironmentalConservation An innovative Trust Fund financed by a Global Environmental Facility (GEF) grant o f $10 m i l l i o n in 1992 has drawn world-wide attention to Bhutan’s conservation efforts. This was the f i r s t such Trust Fund in the world. Co-financing and earnings o n investments enabled the Fund’s assets t o grow to $36 m i l l i o n before market reverses in 2001-2002. Eamings f r o m the Trust Fund were used t o strengthen and extend Bhutan’s protected area system which, with the recent separate “Gift to the World” allocation o f connecting corridors, n o w covers 36 percent o f Bhutan’s land area. The Fund has become the primary source o f finance for the recurrent costs o f protected areas. The Trust Fund has become a frequently-cited model for financing bio-diversity conservation. I t has helped reinforce Bhutan’s already-high commitment to conservation. I t has also contributed t o Bhutan’s potential for eco-tourism, b o t h by enhancing the country’s intemational reputation and by alerting the country’s leadership t o the possibility o f generating revenue f r o m forests through conservation and tourism as w e l l as (or instead of) commercial logging. OED rated the outcome o f the Trust Fund project Satisfactory in 1998. Institutional development was rated Substantial and sustainability Likely. Outcomes o n the ground are difficult to assess because o f time lags in monitoring species and habitats, but the Trust Fund’s governing board has recognized the need for outcome monitoring to improve effectiveness o f conservation investments and t o promote accountability. 3.26 The overall impact o f the Bank in the forest sector has been mixed. The three projects tested new approaches to plantations, reforestation, environmental management and social forestry. Physical targets under the First Project were met or exceeded, but reforestation and plantation techniques introduced under the project have not been widely replicated in Bhutan. Their impact has been very modest. The Second Project helped train staff in techniques for dealing with pest management and improved processes for salvage o f overage and diseased trees. But it left unresolved a number o f questions 32 This i s the conclusion of an intemal draft paper by the South Asia Region which reviewed forest sector cooperation between Bhutan and the W o r l d Bank between 1983- 2002. 33 Other donors active in the sector include Germany and ADB. See Annex A, Table A.4. 20 regarding the profitability and role o f the Bhutan Logging Corporation (BLC) and emerging issues o f forest product pricing and marketing. 3.27 The Third Forestry Project in FY94 focused o n Eastern Bhutan and sustainable forest management practices, including scientific management and extraction o f natural forests, introduction o f social forestry for community-based forest management, trial reforestation techniques and institutional strengthening, especially for decentralization o f the DFS. Although not specified as such, the project was probably the first effort in the South Asia Region to introduce “reduced impact logging,’’ an innovative approach. The project also introduced more systematic and professional approaches to forest management and formed the basis for a shift in thinking about community engagement in forest protection and exploitation. 3.28 On the other hand, stated objectives o f introducing financially self-sustaining commercial forestry into Eastern Bhutan and rationalizing the economic and financial incentive framework through changes in price policy and forest legislation were not achieved. Decisions to ban the export o f unprocessed logs (1999) and semi-finished products (2000) changed market conditions radically and only 84 percent o f the logs harvested during the project period were ever sold. The Bank might have had more impact if i t had done more systematic sector work at an earlier stage (see Annex B) and enforced project conditionality more 3.29 OED rated outcome o f the Third Forestry Project Moderately Unsatisfactory. The social forestry and reforestation components o f the project were unsuccessful, due in part to the reluctance o f the DFS to cede responsibilities to local communities, as well as the lack o f a clear policy o n grazing that has impeded the natural regeneration o f forests in the Region. Implementation was adversely affected by frequent turnover o f Bank staff and inadequate supervision (see para. 3.39 below). Given the regional nature o f the project, the Bank had limited influence o n national forest policies. 3.30 The Swiss Development Corporation agreed that the l o w rating o f project performance was justified. The Bhutan Government has decided not to borrow again for forestry, despite the intense interest o f the DFS. This suggests that both the borrower and funding partners found the Bank’s visible impact in the forestry sector to have been rather limited. Infrastructure 3.3 1 Urban Development Project. In December 1999, the Bank made a loan o f $10.8 million to support development o f urban infrastructure in 10 provincial towns.35 The loan supports investments in water supply and solid waste management in a l l ten towns, and selective investments in street improvement, footpaths, markets and river protection. It 34 The government resisted implementing covenanted pricing reforms. F o r this and other reasons, a supervision mission proposed rating the project unsatisfactory, but this was overruled by regional management. This sent mixed signals and may have delayed reform. Shortly thereafter, an internal assessment review rated the Bank’s supervision o f the project unsatisfactory (see below). 35 This i s the f i r s t Bank project in urban development in Bhutan. Other agencies involved in the sector include DANIDA, ADB, and UNDP. 21 also strengthens service delivery capacity. The basic objective i s to improve the quality o f l i f e in these communities to slow migration to Bhutan’s two major cities. 3.32 The project was approved too quickly before preparation had been completed, with a resulting delay in implementation o f almost two years and an unsatisfactory rating for implementation progress (since upgraded). Implementation has since picked up and the project i s likely to be completed by the scheduled closing date o f December 31, 2005. 3.33 Supervision missions have expressed concern over the lack o f adequate supervision locally or from Thimphu. The OED mission noted considerable variability in the apparent competence o f local engineering staff. Staff at the Dzongkhag (provincial) level did not yet seem to feel “empowered” by the decentralization process. Investments in some localities appear to be based more on sentiment than economic analysis (such as the $1 million for Trongsa with a population o f 2,500 and no room to expand). 3.34 Rural Access Project. Also in December 1999, the Bank approved an $1 1.6 million loan to finance 122 km o f feeder roads in three eastern provinces o f Bhutan. The investment represents over one-third o f the road length to be built under the 8th Five Year Plan (1997-2002). Poverty i s concentrated in remote villages more than a day’s walk from the nearest road. The project i s designed to provide access and reduce transportation Road construction i s based o n the concept o f Environmentally-Friendly Construction (EFC) methods, an approach that was developed under the project and reflects Bhutan’s commitment to protection o f the environment in i t s development effort. Co-financing from the Netherlands provides technical a s ~ i s t a n c e . ~ ~ 3.35 The Rural Access program suffered the same teething problems as the Urban Development Project. I t was rushed through the Board without adequate preparation. The EFC techniques were new to Bhutan and i t took time to understand and build competence in using EFC. However, the EFC approach has now been adopted on all feeder roads in Bhutan-a significant impact o f the project. Strengths and weaknesses of lending 3.36 Bank lending in Bhutan has been a “no-frills” program o f infrastructure development-roads, schools, water supply and sanitation and forest management, reflective and respectful o f the needs o f a traditional society s t i l l in the early stages o f i t s development process. Each project has included basic capacity building, usually funded by bilateral partners with grant assistance, while the focus o f the Bank’s investment has been on building basic physical infrastructure. 3.37 The lending program has been client-driven, carried out in partnership with bilateral agencies and reflective o f the country’s holistic approach to development (including i t s cultural and traditional values). In a sense, it has been a model o f the 36 The project i s estimated to bring over 25,000 people living up t o two days’ walk f r o m the nearest road t o within h a l f a day’s walk o r less. A study in one o f the project districts showed that t h i s degree o f improved access i s associated with a doubling o f b o t h school enrollment and household income. 37 Other donors in the sector include India, Japan, Switzerland, and ADB. 22 comprehensive development framework long before it found i t s way into the Bank lexicon. The lending program has been: 0 Responsive. Bhutanhas borrowed only when grant funding has not been available. The government has decided the sectors and overall size o f the lending operations. Education, forestry and infrastructure have been, and remain, high priority sectors for the development o f the economy. Project designs have respected Bhutan’s cultural and environmental traditions, even at somewhat higher cost. 0 Effective. As indicated in table 3.1 above, five o f the six completed projects have been judged to have had satisfactory outcomes, although sustainability has been considered uncertain in three operation^.^' Institutional development impact has been mixed, with three o f the projects rated substantial (four including GEF) and three modest. 0 Hardware-orientedbut with associated technical assistance. Bhutan has used IDA funds primarily to finance the physical aspects o f projects. However, all projects have had significant technical assistance funded by grants from bilateral partners. 3.38 At the same time, the lending program has had some shortcomings: 0 Sometimes inadequately prepared. Two recent projects, for Urban Development and Rural Access, were prepared in a great hurry and submitted together with the Bhutan CAS without sufficient preparation. The result was a loss o f up to two years in implementation, and an unsatisfactory IP rating for the Urban Project in 2002. 0 Sometimes indifferently supervised. Supervision has also been problematic. Projects in remote areas were rarely visited by financial or procurement staff. An internal assessment review rated the Region’s supervision o f the Third Forestry Project as unsatisfactory in 1998. The review and the subsequent I C R criticized supervision missions for this project for being poorly prepared and spending too l i t t l e time in the field. Management oversight and follow-up actions were also criticized (see footnote 29), the review concluded that the Borrower had little commitment to the operation. The Bank appears to have responded to these criticisms, however, because a follow-up review in 1999 rated supervision o f the same project satisfactory. 0 Lack of continuity. Staff turnover has been a problem. Forestry I11had six task managers, which may partly account for its poor outcome. Continuity was exemplary for many years in the education sector, but EDP has had three TMs, lost a year in preparation and frustrated the principal partner o f the first two 38The Calcium Carbide project has proven t o be sustainable, though it was rated uncertain shortly after completion. 23 projects-Helvetas o f Switzerland-into going i t s own way. Urban Development i s o n i t s third T M in three years. 0 Insensitive to size. The overall Bank lending program has respected Bhutan’s approach to development-the concept o f Gross National Happiness-and i s w e l l regarded by borrower representatives. However, project requirements have not reflected the extremely small size o f the country. With a c i v i l service o f less than 15,000, including teachers but excluding the military, Bhutan lacks the staff capacity to conform easily to Bank requirements for reporting, financial management and procurement. The Urban and Rural Access Projects were required to adopt the “ L A C Y approach to financial reporting with i t s Project Management Report (PMR)-based disbursements which proved beyond the capability o f many larger borrowers (and has since been simplified). Procurement rules such as pre-qualification requirements for bidders have not made accommodation for the unique circumstances o f a nascent contracting industry. Donor cooperation 3.39 The government i s the main aid coordinator at the economy-wide and sector level. The Bank formally co-chairs (with UNDP) the periodic Round Table meetings between the government and donors, and usually provides an economic report or other background document on the economy, but the main estimates o f aid requirements come from the FYP process. Similarly, at the sector level, the government decides which donors will be involved in which sectors and in which kind o f activities. Donors share information at the sector level, though India tends to operate outside that network and deal directly with the government. 3.40 For the Bank, the main form o f donor cooperation has been at the project level, through co-financing. Co-financing has been a critical element in most IDA projects because o f the government’s preference for using grants wherever available. This typically leads to a division o f labor in which IDA mainly funds infrastructure and other “hard” elements and the co-financier funds technical assistance and other “soft” elements. When the Bank has funded a project without co-financing it has sensibly sought assurances from the government that complementary TA will be provided by others if Bhutan does not want to borrow from IDA for this purpose. Coordination needs to work well to ensure project success and, for the most part, it has. Even when the project outcome has been less than satisfactory the Third Forestry Development Project (TFDP) co-financing was a strength. Cooperation with the Swiss Development Corporation (SDC) ensured many positive technical outcomes, provided flexibility on financial procedures and topping up which the Bank did not have, and provided project continuity to help offset the high turnover o f Bank task managers. 3.41 Nevertheless, there have been criticisms, including from co-financiers, o f the Bank’s cooperation with other donors. SDC stated that i t s contribution t o TFDP was inadequately acknowledged, that i t s concerns and proposals during supervision missions were not taken seriously enough by the Bank, and that n o one from Bank management 24 had ever visited the project area.39 Field representatives o f other donors said that visiting Bank missions did not always brief other donors active in the sector on their activities, in contrast to ADB which regularly briefed Bhutan-based sector donor groups convened by UNDP. And, as noted earlier, delays in Bank decisions led Helvetas to withdraw from co-financing EDP.40 3.42 These criticisms should be kept in perspective. The critics stressed that their experience o f working with the Bank had been largely positive, and that some o f the difficulties reflected the normal strains o f working together. Delays and conflicting procedures have arisen o n both sides, or were sometimes due to government decisions (or indecision). Critics also recognized that the absence o f local representation complicated Bank coordination. Working relations also varied from task manager to task manager, and some o f the problems arose because o f staff turnover. 3.43 What can the Bank do to improve cooperation with other donors? One thing i t should not do i s avoid co-financing simply to avoid the complications o f coordination. There are too many benefits to the Bank and to Bhutan o f intemalizing coordination at the project level. The most obvious positive thing the Bank can do is to provide better continuity o f staffing. The Bank could also use the UNDP office to help liaise with other donors. Visiting missions should make a point o f discussing mission plans and findings through briefings o f other donors at the beginning and end o f missions. At the programming level the Bank should coordinate with ADB o n i t s proposed E S W to avoid duplication. 4. Evaluation o f the Bank’s Assistance 4.1 Bhutanhas successfblly used a large aid program and foreign exchange earnings from hydropower exports to promote growth and provide widespread social benefits to the population. Inthis, Bhutan provides a rare example o f successful management o f windfall resources for broad-based social and economic development in the face o f severe development constraint^.^^ This section will evaluate the extent to which the World Bank’s small assistance program contributed to this success in the areas in which the Bank sought to help, and the impact o f Bank assistance in helping address Bhutan’s development. The section will rate the outcome, sustainability, and institutional development impact o f the Bank’s assistance, and examine Bank performance. 4.2 The Bank’s early strategy was to provide economic policy advice and lending assistance to selected sectors. I npursuing this limited role, the Bank recognized i t s position as a relatively minor player. In 2000, the Bank broadened the proposed scope o f i t s activity considerably. W h i l e recognizing that it couldn’t do everything, i t was prepared in principle to consider doing anything the government wanted. Moreover, the 39 These statements were made in the donor comment annex o f the project ICR. 40 Although separate Danish and Swiss projects will proceed in parallel with EDP, the absence o f formal co-financing l i n k s m a y be a handicap for the Bank. 41 Botswana i s the leading successful example among small economies. 25 Bank sought t o increase i t s relevance to Bhutan by offering i t s services as a knowledge bank in line with i t s presumed comparative advantage. Outcome o f the Bank’s Assistance 4.3 The overall rating o f the outcome o f Bank assistance depends mainly o n the relevance o f the Bank’s assistance strategy and the efficacy and efficiency o f i t s products and services in the main operational areas o f Bank support. Outcome depends not only o n the Bank’s performance (discussed separately below) but also o n the performance o f the government and other donors and on extemal events. Annex D elaborates the definitions and methodology used by OED in i t s ratings. 4.4 Relevance o f Bank assistance. The Bank’s assistance strategy was broadly relevant during the past decade. The three main objectives o f Bank assistance during most o f the period-providing economic policy advice, supporting education development, and improving forestry management-were all important development issues, and the way the Bank provided assistance followed government preferences and complemented the assistance o f other donors. 4.5 In education the bank moved steadily from helping expand primary schooling opportunities to expanding secondary schools and, more recently, to trying to link school curriculum to the practical needs o f the labor market, while promoting private sector development to increase provision ofjobs. In forestry the Bank initially had unrealistic expectations o f the scope for economic and sustainable commercial exploitation. But in the past decade, the Bank enhanced the relevance o f i t s assistance by scaling back earlier expectations o f forestry development potential and putting greater emphasis o n reduced impact logging, pricing issues, and social forestry. With respect to economic policy advice, proposed assistance o n fiscal management and later, private sector development, were both relevant, but the ambitious expansion o f proposed AAA was unrealistic and insufficiently selective. More generally, the proposal to act primarily as a knowledge bank in Bhutan was inconsistent with what the Bank was actually doing (expanding lending) and not that relevant to Bhutan’s needs or the Bank’s (or Bhutan’s) capacity. I n practice, however, the areas o f assistance accepted by the government continued to be relevant. 4.6 Efficacy ofBank assistance. Although the Bank generally set relevant objectives and had a realistic strategy for achieving them, efficacy in implementing these objectives has been mixed. Five o f six project outcomes (six o f seven counting the Bank-executed GEF project) have been rated satisfactory, but the exception-the largest, most important, and most recent forestry project-was an important one. The AAA which the Bank has provided has been sound, though the impact o f recent studies remains to be seen. Moreover, the absence o f sector work before lending was a serious shortcoming. 4.7 Efficiency of Bank assistance. I t is difficult to obtain a meaningful estimate o f the efficiency o f Bank assistance to Bhutan because o f the special circumstances o f the country. The average cost o f lending, supervision, and ESW per approved project for the period FY93-FY02 was $1.157 million-significantly below the average for the South Asian Region ($1.627 million) and for the Bank ($1.456 million)-but this comparator i s 26 unreliable because o f the lumpiness o f the Bhutan lending program. Different periods give substantially different results. Supervision costs per project during the late 1990s (FY96-FY99) were about 16 percent higher than the Bank average, but given Bhutan’s difficult terrain and the location o f the TFDP in Eastern Bhutan, i t i s doubtful whether the Bank allocated enough funds for proper supervision during this period. O n balance, there i s n o reason t o believe that administration o f the Bank’s program in Bhutan was inefficient. 4.8 Overall outcome. Based on the outcome o f individual objectives and the relevance, efficacy, and efficiency o f the Bank’s products and services, overall outcome o f Bank assistance to Bhutan i s rated moderately satisfactory. Sustainability of Positive Outcomes 4.9 Using a simple project count, only two Bank projects (three counting GEF) were rated as likely to be sustainable. Taking into account ongoing as w e l l as completed projects, sustainability o f development impact i s likely in the education sector and in rural road construction (where EFC has already been adopted as the standard throughout the country). In forestry, the outlook i s mixed; many technical achievements appear sustainable, but innovations in social forestry appear unlikely to be sustained at this stage, and the future o f the pricing and marketing regime i s unclear. The sector rating for sustainability in forestry, l i k e that for the latest project, i s non-evaluable. Similarly, it i s too soon to evaluate the Bank’s program in urban development. O n balance, overall sustainability i s rated likely. InstitutionalDevelopment Impact 4.10 Institutional development impact (IDI) was rated substantial for three o f the Bank’s six completed projects (four o f seven counting GEF) and in the remaining three projects ID1was rated modest. A separate IDF grant in forestry has had a substantial impact on ID1both within DFS and in helping create an Association o f Wood-Based Industries. The Bank has also brought several Bhutanese officials to Washington to work as interns within the Bank for several months. This has reportedly had a positive impact. Overall, the Bank’s small program has had a substantial institutional development impact. Bank Performance 4.11 Bank performance has been very mixed. On the one hand, the Bank has been very responsive to government wishes, sensitive to Bhutan’s cultural and environmental traditions, and effective in using partnerships with other donors to complement Bank contributions to projects. On the other hand, the lack o f sector work, inadequate preparation and supervision for some projects, and occasional frictions in cooperation with other donors have limited the effectiveness o f Bank assistance. Bank performance has been affected by continuity in staffing and attention. Excessive and sudden attention can be as unproductive as inattention in a small country because i t i s overwhelming. Continuity was exemplary at the country officer level and in education, especially during 27 the 1990s, but was seriously lacking in forestry and in the aftermath o f Bank reorganization. Recently the Bank seems to have improved o n this issue. 5. Lessons and Recommendations 5.1 Some o f the following lessons and recommendations may be applicable to the Bank’s assistance strategy in other small countries, or in other situations in which the Bank plays an unaccustomed small role. 5.2 Listening enhances effectiveness. The Bank was effective in Bhutan when it listened to what the government wanted rather than trying to figure out what the government needs and then trying to persuade the government to accept the proposal. Thus, providing funding for basic infrastructure has generally been successful because this i s what the government wanted from the Bank. Proposals in the 2000 CAS for the Bank to serve as a knowledge Bank have not been successful because, in the event, the govemment did not request the Bank to move in this direction.42 5.3 Co-financing is an especially useful instrument in a small country. I nBhutan co- financing gave the Bank financial flexibility and invaluable local and specialized technical knowledge. I t also provided continuity in the face o f Bank staff turnover and helped make up for the Bank’s lack o f field presence. 5.4 Need for stafing continuity. This i s particularly difficult in a small country with episodic Bank involvement, but i t i s important nonetheless. One reason for the better outcomes in education than in forestry was high staff turnover in the latter sector. The lack o f field presence puts an even higher premium on continuity o f staffing in Washington. 5.5 The importance o f being selective and focused. I t i s a mistake to try to be a “full service bank” in a small country unless the Bank i s a dominant lender. I t i s better t o be involved in depth and over time in a few areas in order to have an impact. I t i s too costly for the Bank and for the country to have the Bank involved in a large number o f activities. Also, from Bhutan’s point o f view, a wide-ranging program increases the risk o f “donor tourism” in which many managers and staff visit Bhutan once or twice but contribute l i t t l e to the country. This i s a concern because o f the burden it imposes o n Bhutan’s scarce administrative resources.43 5.6 I t is necessary to be selective in AAA as well as lending because o f capacity constraints o n both sides. I t i s hard to be a knowledge bank in a small country. The Bank needs to invest resources in order to provide useful advice based on local conditions. The government i s constrained by the limited time o f senior officials to work with the Bank or 42 See the Region’s comments in Annex E. See also footnote 21 above. 43 See the Region’s comments in Annex E. Management and staff visits are necessary for any agency to conduct business and can be beneficial for both the program and the country. But the burden this places o n a s m a l l country should be recognized by a l l donors. 28 discuss the issues. A stand-alone non-lending services program i s likely to be o f limited value. 5.7 More upfront sector work is needed. Selectivity in lending and in AAA should be joined. The Bank should return to the more selective approach o f i t s earlier country assistance strategy, but make it more effective through prior sector work. Indeed, ESW can help support the selection process, as has been the case for private sector development. 5.8 The Bank needs to develop ways to compensatefor its lack offield presence. The Bank has no field office in Bhutan. I t would be expensive to provide an office for a small program and the government does not want it. N o r does the government wish to be part o f country unit based in a regional hub. T o help serve Bhutan better with these constraints, the Bank could adopt some o f the procedures o f the ADB which also coordinates i t s Bhutan program from its headquarters in Manila. The most important o f these are annual rolling program consultations and use o f UNDP facilities by visiting missions to liaise with other donors based in Bhutan. 5.9 Sensitivity to constraints of a small country. Many Bank procedures (procurement and reporting) are poorly adapted to a small country with a small c i v i l service. The high and growing demands o f the Bank and other agencies for reporting o n a standard set o f issues (fiduciary safeguards, environment, etc.) and for formulating strategies to meet national and international targets (PRSP, Vision 2020) while each valuable individually, add up to a formidable burden for small countries. The Bank should adapt i t s procedures to adjust to the special circumstances o f small countries. 29 Annex A: Foreign Assistance by Donor and Sector Table A.l: Grant Assistance by Donor, 1992/93-2001/02 (millions o f $US using FY average exchange rates) Total Share of Total (92/93-01/02) (%) India 322.9 51.1 DANIDA 65.6 10.4 JICA 45.7 7.2 Netherland 27.4 4.3 UNDP 17.8 2.8 Austria 17.7 2.8 Helvetas 17.9 2.8 SDC 16.4 2.6 EEC 15.2 2.4 UNCDF 13.5 2.1 Others* 71.2 11.3 Total 631.3 100.0 Source: R M A A R 01/02, p.140. *These are grants going directly to the central govemment budget under revenue, and do not include grant. Table A.2: Concessional Loans Disbursed by Donor, 1981/82-2001/02 Total Million US$ %o f Total India* 215.5 56.7 ADB 62.5 16.4 WB 36 9.5 KFAED 27.7 7.3 Austria 20.4 5.4 IFAD 15.7 4.1 Mitsui & Co. 1.7 0.4 EFIC Australia 0.8 0.2 Total 380.3 100.0 Source: RMAAR 01/02, Table 24, p.83 *FY average exchange rates were used to convert annual Rupee loans post 1995196, and simple average o f FY average exchange rates were used. Table A.3: Grant Assistance by Sector, 1990/91-2001/02(millions $US) Total Share of Total (90/91-01/02) (%) Agriculture 119.5 16.9 Communications 85.2 12.0 Culture and Y o u t h 13.7 1.9 Health and Education 106.6 15.0 Power 83.0 11.7 Trade and Industry 4.5 0.6 Other Government Agencies' 132.2 18.7 Autonomous 5.4 0.8 Others' 158.8 22.4 Total 708.9 100.0 Source: RMAAR 01/02, p.141 'Capacity building o f government agencies. Program Grants from India. Annex A fcontinued) 30 I ~ C c .z I 6 V s B Q --tt 31 Annex B: Forestry Pricing and Marketing Policies 1. Until 1969 use o f forestry was free to all. For many years thereafter differential l o g prices were set for four categories o f users: rural, urban, industrial, and export. Export prices were market-based and domestic prices were administratively set at below-market rates with highest subsidies for rural users. In 1991 a pricing reform eliminated subsidies to urban consumers, and subsidies to rural users have gradually been reduced and the category o f eligible rural consumers has been narrowed. 2. In 1999 the government introduced a reform that established public timber auctions for all sales and banned exports of round-wood, sawn wood, and firewood. The new policy had multiple objectives: to promote more efficient use o f forest resources, to strengthen incentives for domestic wood industries, and to make timber available for domestic consumers at affordable prices. T h e export ban was not total. Surpluses in the domestic market are sold through special auctions open t o international buyers. After the new system was introduced, prices rose markedly. The government suspected leakage through export o f partially-processed products and collusion among sawmill owners. It responded by extending the export ban to semi-finished products and by setting up state-owned sawmills to compete with the private sector. 3. The Bank had encouraged the government to simplify the pricing structure, reduce and improve targeting o f subsidies, and introduce an auction system, but it recommended against an export ban. I na letter to the government before the policy was introduced, the Bank pointed out that international experience suggested an export ban i s an inefficient instrument for protecting a local wood-processing industry. Restricting l o g exports tends to lower domestic l o g prices, thus encouraging inefficient processing and environmentally unfriendly harvesting practices. Financial retums from inefficient local processing may end up being lower than those from direct export o f raw logs. An export tax would be a more efficient instrument for this purpose, though in view o f India’s reluctance to accept an export tax it could not be used. The Bank noted that the proposed system o f exemptions to the ban might work as a second-best expedient but this should be monitored closely to avoid negative outcomes. 4. After the 1999 reform was introduced, the Bank and the Swiss Development Corporation did a study o f the timber market which found n o evidence o f collusion among sawmillers. The study concluded that other factors (inelastic public sector demand, fragmented markets) accounted for pricing patterns and trends, and it recommended against public investment in sawmills. The government went ahead, however. More recently, the Private Sector Survey reported that the auction system had several shortcomings (such as auctioning only mixed lots o f wood) that did not meet buyer preferences. This led to wastage, l o w prices for some high-value varieties, and high costs for local producers. Under an IDF grant (Forestry Institutional Development Initiative) the Bank supported a working group that brought together the private sector and forestry officials and catalyzed formation o f an Association o f Wood-Based Industries that successfully lobbied the ministry t o rationalize guidelines for export o f semi-finished products. 5. The Bank has offered good analysis and advice on forestry pricing and marketing and has been persistent, but i t s overall impact o n policy has been limited. The Bank might have had more impact if had done more systematic sector work at an earlier stage and enforced project conditionality more strictly. 33 Annex C : List o f People Interviewed Ministry o f Finance Aum Lhaden Pem Project Manager Lyonpo Yeshey Zimba Minister of Finance Ministry o f Trade and Industry Dasho Wangdi Norbu Secretary for Finance Mr. Achyut Bhandari Director General, Department o f Trade Department of Aid & Debt Management Mr. Sonam Yangley Director, Department o f Industry Aum Y a n h T. Wangchuk, Director General Mr. Sonam P. Wangdi Head, PPD, Mr. Lepo Ministry o f Trade and Industry Senior Program Officer Ms. Karma Mr. Ugyen Norbu PPD, MTI Program Officer Mr. Sonam Wangchuk Head, Foreign Trade Division, M T I Department o f Education Mr. Pema Thinley Dzongkhag Administration, Bumthang Director General Daso N i m a Tshering Ms. Pema Lhazom Dzongdag Project Manager Mr. Sangay Wangchuk Mr. Thinley Rinzin Deputy Dzongdag Project Manager for SEP Mr. Namgay District Education Officer Rural Access Project Mr. Tempa Gyaltshen Mr. Kunzang Wangdi District Engineer Project Manager Mr. P.L. Subba Mr. Niralal Rai Project Engineer Assistant Project Manager Wangchholing. Junior High School Urban Development Project Ms. Karma Choden Mr. Tshering D o r j i Headmistress Director General Annex C (continued) 34 Chume High School Government o f India, N e w D e l h i Mr. Pema Chophel Mr. Ghanshyam Principal Director, Nepal and Bhutan Ministry o f External Affairs DzongkhaP Administration, Trongsa World Bank Mr. Pema Thinley District Engineer Sadiq Ahmed, Sector Director Simon C. Bell, Sector Manager Mr. T.R. Gurung Guan Zhe Chen, Sector Manager Project Engineer Christian Eigen-Zucchi, Economist Subramanian V. Iyer, Lead Financial Analyst DeDartment o f Forestrv Services Peter Jipp, Sr. Forestry Specialist Toshiaki Keicho, Sr. Urban Environment Mr.Ugyen Thinley Specialist Director M a r k Laprairie, Education Specialist Chingboon Lee, Lead Education Specialist Mr. Gopal Alastair J. Mckechnie, Country Director Joint Director Peter Nicholas, Lead Country Officer Enrique Pantoja, Country Officer Daniel M. Sellen, Sr. Agric. Economist S N V , Thimphu Frederick Swartzendruber, Consultant Marilou Jane Uy, Sector Director Mr. Peter Newsum Tara Vishwanath, Lead Economist Acting Country Director John W. Wall, Country Director Mr. Henrick Visser Former W o r l d Bank Staff Engineer, EFRC, S N V Chris K e i l M i e k o Nishimizu HelvetasKDC Coordination Office Manuel Penalver Grant Sinclair Mr. Erwin Koenig Joseph W o o d Resident Coordinator UNDP Ms. Reneta L o k Dessalien Resident Representative Danish Liaison Office, Thimphu Mr. Torben Bellers Resident Coordinator 35 Annex D: Guide to OED’s Country Assistance Evaluation Methodology 1. This methodological note describes the k e y elements o f OED’s country assistance evaluation (CAE) m e t h ~ d o l o g y . ~ ~ CAEs rate the outcomes of Bank assistanceprograms, not Clients’ overall development progress 2. An assistance program needs to be assessed o n h o w well it met i t s particular objectives, which are typically a sub-set o f the Client’s development objectives. If an assistance program i s large in relation to the Client’s total development effort, the program outcome will be similar to the Client’s overall development progress. However, most Bank assistance programs provide only a fraction o f the total resources devoted to a Client’s development by donors, stakeholders, and the govemment itself. In CAEs, OED rates only the outcome o f the Bank’s program, not the Client’s overall development outcome, although the latter i s clearly relevant for judging the program’s outcome. 3. The experience gained in CAEs confirms that program outcomes sometimes diverge significantly from the Client’s overall development progress. CAEs have identified assistance programs which had: satisfactory outcomes matched by good Client development; unsatisfactory outcomes in Clients which achieved good overall development results, notwithstanding the weak Bank program; and, satisfactory outcomes in Clients which did not achieve satisfactory overall results during the period o f program implementation. Assessments of assistanceprogram outcome and Bank performance are not the same 4. By the same token, an unsatisfactory assistance program outcome does not always mean that Bank performance was also unsatisfactory, and vice-versa. This becomes clearer once we consider that the Bank’s contribution to the outcome o f i t s assistance program i s only part o f the story. The assistance program’s outcome i s determined by the joint impact o f four agents: (a) the Client; (b) the Bank; (c) partners and other stakeholders; and (d) exogenous forces (e.g., events o f nature, international economic shocks, etc.). Under the right circumstances, a negative contribution from any one agent might overwhelm the positive contributions from the other three, and lead to an unsatisfactory outcome. 5. OED measures Bank perfonnance primarily o n the basis o f contributory actions the Bank directly controlled. Judgments regarding Bank perfonnance typically consider the relevance and implementation o f the strategy, the design and supervision o f the 44In this note, assistanceprogram refers to products and services generated in support o f the economic development o f a Client country over a specified period o f time, and client refers to the country that receives the benefits o f that program. Annex D (continued) 36 Bank’s lending interventions, the scope, quality and follow-up o f diagnostic work and other AAA activities, the consistency o f Bank’s lending with i t s non-lending work and with i t s safeguard policies, and the Bank’s partnership activities. Evaluation in Three Dimensions 6. As a check upon the inherent subjectivity o f ratings, OED examines a number o f elements that contribute to assistance program outcomes. The consistency o f ratings i s further tested by examining the country assistance program across three dimensions: (a) a Products and Services Dimension, involving a “bottom-up” analysis o f major program inputs -- loans, AAA, and aid coordination; (b) a Development Impact Dimension, involving a “top-down” analysis o f the principal program objectives for relevance, efficacy, outcome, sustainability, and institutional impact; and, (c) an Attribution Dimension, in which the evaluator assigns responsibility for the program outcome to the four categories o f actors (see paragraph 4. above). Rating Assistance Program Outcome 7. Inrating the outcome (expected development impact) o f an assistance program, OED gauges the extent to which major strategic objectives were relevant and achieved, without any shortcomings. Programs typically express their goals in terms o f higher-order objectives, such as poverty reduction. The country assistance strategy (CAS) m a y also establish intermediate goals, such as improved targeting o f social services or promotion o f integrated rural development, and specify h o w they are expected to contribute toward achieving the higher-order objective. OED’s task i s then to validate whether the intermediate objectives produced satisfactory net benefits, and whether the results chain specified in the CAS was valid. Where causal linkages were not fully specified in the CAS, i t i s the evaluator’s task to reconstruct this causal chain from the available evidence, and assess relevance, efficacy, and outcome with reference to the intermediate and higher-order objectives. 8. Evaluators also assess the degree o f Client ownership o f international development priorities, such as the Millennium Development Goals, and Bank corporate advocacy priorities, such as safeguards. Ideally, any differences o n dealing with these issues would be identified and resolved by the CAS, enabling the evaluator to focus o n whether the trade-offs adopted were appropriate. However, in other instances, the strategy may be found to have glossed over certain conflicts, or avoided addressing k e y Client development constraints. In either case, the consequences could include a diminution o f program relevance, a loss o f Client ownership, and/or unwelcome side- effects, such as safeguard violations, all o f which must be taken into account injudging program outcome. 37 Annex D (continued) Ratings Scale 9. OED utilizes six rating categories for outcome, ranging from highly satisfactory to highly unsatisfactory: Highly Satisfactory: The assistance program achieved at least acceptable progress toward all major relevant objectives, had best practice development impact o n one or more o f them. N o major shortcomings were identified. Satisfactory: The assistance program achieved acceptable progress toward all major relevant objectives. N o best practice achievements or major shortcomings were identified. Moderately Satisfactory: The assistance program achieved acceptable progress toward most o f i t s major relevant objectives. N o major shortcomings were identified. Moderately Unsatisfactory: The assistance program did not make acceptable progress toward most o f i t s major relevant objectives, or made acceptable progress on all o f them, but either (a) did not take into adequate account a key development constraint or (b) produced a major shortcoming, such as a safeguard violation. Unsatisfactory: The assistance program did not make acceptable progress toward most o f i t s major relevant objectives, and either (a) did not take into adequate account a key development constraint or (b) produced a major shortcoming, such as a safeguard violation. HighIy Unsatisfactory: The assistance program did not make acceptable progress toward any o f i t s major relevant objectives and did not take into adequate account a key development constraint, while also producing at least one major shortcoming, such as a safeguard violation. 10. The institutional development impact (IDI) can be rated as: high, substantial, modest, or negligible. ID1measures the extent to which the program bolstered the Client’s ability to make more efficient, equitable and sustainable use o f its human, financial, and natural resources. Examples o f areas included injudging the institutional development impact o f the program are: e the soundness o f economic management; the structure o f the public sector, and, in particular, the c i v i l service; the institutional soundness o f the financial sector; the soundness o f legal, regulatory, and judicial systems; the extent o f monitoring and evaluation systems; the effectiveness o f aid coordination; the degree o f financial accountability; the extent o f building NGO capacity; and, the level o f social and environmental capital. Annex D (continued) 38 11. Sustainability can be rated as highly likely, likely, unlikely, highly unlikely, or, if available information i s insufficient, non-evaluable. Sustainability measures the resilience t o risk o f the development benefits o f the country assistance program over time, taking into account eight factors: technical resilience; financial resilience (including policies on cost recovery); economic resilience; social support (including conditions subject to safeguard policies); environmental resilience; ownership by governments and other key stakeholders; institutional support (including a supportive legalh-egulatory framework, and organizational and management effectiveness); and, resilience to exogenous effects, such as international economic shocks or changes in the political and security environments. 39 Annex E: Bhutan Country Assistance Evaluation Comments from the Bank’s Country Management Unit, South Asia Region Overall, the CAE team i s to be commended for preparing a very comprehensive and w e l l written evaluation report on Bank engagement in Bhutan. The team have pulled together a substantive amount o f relevant information from a wide range o f sources and perspectives, and present the conclusions in a constructive and independent fashion. The report will be very a valuable input to the new CAS for Bhutan, currently planned for FY05. While w e agree with the report in general, we would l i k e to raise the following issues for your consideration before finalizing the report: The CAE still unfairly conveys the notion that the CAS was prepared without consulting RGOB. This notion i s exemplified in the statement offered in the Memorandum to the EDs and the President: “The government was notprepared for program lending and did not agree to Bank involvement in many o f the proposed studies, primarily because alternative support from ADB was available and because multiple donor involvement i s burdensome”(emphasis added). This type o f statements run throughout the report, and culminate in Section 5 (Lessons and Recommendations) with the somewhat unbalanced statement that “Proposals in the 2000 CAS for the Bank to serve as a knowledge Bank have not been successful because the government had not asked the Bank to move in this direction.” (Para 5.2, p. 27) As we explained before, when the CAS was prepared in 1999, the government was in full agreement with the role proposed for the Bank, Le., increasingly placing i t s knowledge resources at the disposal o f the government without tying them to IDA credits. As stated in the CAS (p. iii), at the time, “[tlhe government ha[d] already requested such knowledge based consultations o n macroeconomic policy, a financial sector strategy, long-term health and education strategies, and accounting and auditing standards.” In contrast, the CAE gives the impression that the Bank pushed aggressively and unilaterally for AAA. Examples o f statements that may mislead the reader are: in the Memorandum (p.2), “Proposals for the Bank to serve as a knowledge bank have not been successful because in the event most o f the Bank offers o f knowledge services were not taken up by the government.” I nthe main text, “The proposed program t o implement the 2000 strategy included AAA o n a vastly expanded range o f topics” (para 2.6). A look at Annex B 4 o f the CAS document, shows that the proposed AAA program reflected government requests, while footnote 1 specifically shows that the program was indicative, more so since “AAA assistance [was] seen as open-ended in that i t will respond to the needs o f the government and when needed, and will therefore continue while i t i s justified in terms of knowledge transfer needs o f the government.” Interestingly, the CAE’s recommendation in this regard has an implicit contradiction: the Bank i s faulted for pushing AAA, but i t i s recommended that sector work should be Annex E (continued) 40 increased upfi-ont (thus implicitly in a “supply-driven’’ mode) - although traditionally the government has not been interested in this type o f Bank assistance. At the same time, the CAE unwittingly gives the impression that the Bank was also trying to increase lending volumes during CAS preparation. Contrary to what i s assumed in the CAE, the two projects that went to the Board with the CAS were prepared rapidly in part to respond to the government’s request, not mainly to “increase the amount and scope o f lending.” (CAE, para 2.9, p.11). For the record, the lending volumes in Bhutan have traditionally been so small that they do not have any impact o n regional or Bank wide levels. I n this respect, the CAE needs to reflect better the context of CAS preparation and implementation. The government policy in 1999 was to rely as much as possible o n grants, as there were serious concerns with Dutch disease symptoms, debt management, and expectations o f continuous grant flows. Thus, R G O B had informed the Bank that i t did not intend to borrow from IDA, particularly after the two projects that were being prepared and that went to the Board with the CAS. The Government, however, wanted the Bank to continue providing non-lending services as needed. For your information, RGOB s t i l l prefers to borrow IDA for capital investments, use grants for TA and finance recurrent expenditures with domestic revenue. What the Bank should be perhaps faulted with i s not having prepared a new CAS sooner, particularly as i t became evident that (i) the government was not seeking as much Bank assistance in non-lending services as initially envisaged; and that (ii) Bhutan was increasingly interested in borrowing from IDA as grant flows declined and the financing gap for capital investments in the NinthFive Year Development Plan (2002-2007) could not be fully bridged. The report also misrepresents the CAS statement regarding program lending. The government has expressed interest during CAS consultations, as mentioned in the CAS, “in greater harmonization o f donor support and in discussing the merits o f more program- oriented funding and ha[d] asked for IDA’Shelp in thinking through the issues.(p.l4)” The CAS did not assert that the government was prepared for program lending, or that this was a specific deliverable for the CAS period. In line with the government’s thinking, the CAS mentions program lending as a possibility, while explaining that before this could take place, firther capacity building was needed in areas such as public expenditure management, financial management and auditing and accounting. I n relation to above, the CAE should also make more explicit that priorities of RGOB changed rapidly (and often after agreeing with the Bank on specific tasks such as the Financial Sector Report) O n the other hand, w e should recognize that the Bank has not done enough to educate i t s Bhutanese counterparts o n what the Bank can offer in terms o f knowledge services. This trend along with the related fact that the government has not always been clear o n the assistance needed, indicates that the country has not fully benefited from the knowledge resources available at the Bank. But t o be fair to the 41 Annex E (continued) Bank, the C A E should not so easily assume that “the Borrower has.. .a clear sense o f what it wants from the Bank.” (Para 2.10, p. 11). Additional Suwestions / Corrections: Para 5.5, p.27: notion o f “donor tourism” i s somewhat harsh. For a small country with a small budget and without a country office with lower cost national staff, it i s perhaps inevitable that managers and some staff visits are infrequent. Even if a person visits only once, the information gained i s highly valuable for a contribution from Washington or management oversight. The key point i s in para. 5.4, the need for staff continuity. 43 Annex F: Comments from the Government o f Bhutan Annex F (continued) 44 45 Annex F (continued) 47 Annex G: Bhutan Country Assistance Evaluation Management Action Record Recommendations Management Response 1. Selectivity. The Bank should return to the selectivity i s indeed critical for Bank assistance in Bhutan, more selective approach o f i t s earlier $veri limited IDA resources and relative high cost o f assistance strategy. T o improve the nanaging the country program. Accordingly, selectivity effectiveness o f i t s lending and policy advice, w i l l be applied in the new CAS that will be prepared in the Bank should pursue a more integrated TYO5, especially considering the country’s increasing approach with upfront sector work and ieed to borrow from IDA and the results-based orientation development o f a clear sector strategy. i f the Bank strategy. Selectivity will also require that the 3ank better inform i t s government counterparts about ivhat we have to offer in terms o f lending and non-lending services, while establishing periodic program reviews. 2. Compensatingfor the lack offield presence. Xecognizing that Bank engagement in Bhutan has been The Bank should compensate for i t s lack o f iffected by the lack o f field presence, we are discussing field presence through annual program with UNPD on future arrangements t o have access to their consultations with the government, malung facilities, logistical support for Bank missions, and better use o f UNDP facilities to improve increased coordination with other development partners coordination with other donors based in misting Bhutan. Bhutan, and ensuring better staffing continuity. Ensuring staff continuity has been a challenge due to the small size o f the portfolio in each sector, which results in task team leaders being reassigned t o larger programs when work pressures demand it. In a broader sense, it i s difficult to build a Bhutan country team given that the various TTLs and team members are fully engaged in the larger counties and for the most part have only one activity in Bhutan. 3. Sensitivity to constraints of a small country. This i s a problem not only in Bhutan, but in general in The Bank should review i t s procedures o n IDA counties with limited technical and institutional procurement, reporting, and core economic capacity. In small countries, however, the problem i s even studies to adapt to Bhutan’s capacity more acute, and we are considering ways to take full constraints. advantage o f current Bankwide efforts t o use country systems, and to approach the mandated diagnostic work, i e . , PER, CFAA, etc,. in a flexible manner, e.g. by consolidating tasks together, or by seeking exemption when justified o n substance and small country grounds. 49 Annex H: Chairman’s Summary COMMITTEE ON DEVELOPMENT EFFECTIVENESS (CODE) Bhutan Country Assistance Evaluation Meeting of September 8,2004 1. The Informal Subcommittee (SC) o f the Committee o n Development Effectiveness (CODE) met on September 8 to discuss the Bhutan Country Assistance Evaluation (CAE) prepared by the Operations Evaluation Department (OED). 2. Background. According to the OED report, the Bank’s lending program in Bhutan has been client-driven, carried out in partnership with bilateral agencies, and sensitive to Bhutan’s cultural and traditional values. OED’s evaluation noted that IDA is a relatively small player in Bhutan, and i t s strategy has been to give priority to analytical and advisory services, while engaging in selective lending for education, forestry and infrastructure. The Bank has provided some useful analytical work and policy advice over the years, but no formal sector work has been done. The report noted that while the Bank has made important contribution to improving quality and access to education in the country, the outcome o f i t s assistance in forestry has been mixed. According to the OED report, the main lessons from the Bank’s experience in Bhutan are: (i) usefulness o f co-financing in a small country context; (ii) importance o f staffing continuity for effectiveness; and ( iii) selectivity and focus are crucial. The report pointed out the difficulty o f being a full-service and a knowledge Bank in a small country, and stressed the need to provide advice based on local conditions and to inform a selective lending program, while recognizing that the Government i s constrained by the limited time o f senior officials to work with the Bank. The O E D evaluation concluded that some o f lessons from Bhutan experience might be relevant in other small countries or situations when the Bank plays an unaccustomed small role. 3. Management commended OED’s sound evaluation o f the Bank’s engagement in Bhutan and noted that i t s recommendations will be taken into account during preparation of the results-based country assistance strategy for Bhutan in the next few months. Management noted that despite the overall emphasis o n analytical and advisory services, Bhutan has not benefited as i t should have from the knowledge resources available at the Bank, mainly due to l o w implementation capacity, lack o f knowledge o f what the Bank can provide and availability o f alternative resources. Management added that, in response to that challenge, it i s going to approach diagnostic work more flexibly to better respond to the needs and capacity o f the country, and that i t has intensified dialogue with the government to increase its awareness o f Bank products and identify emerging opportunities to rely more o n country systems. 4. The Chair representing Bhutan welcomed the discussion and thanked OED for a candid and thought-provoking report that contained lessons not only for Bhutan, but also for other small landlocked poor countries, especially o n infrastructure development and the knowledge-bank role o f the World Bank. H e noted that despite recent achievements, the main challenges for Bhutan remain providing employment for growing population and enhancing productivity o f the country’s limited arable land. The important lesson to be drawn by the Bank i s that i t i s doing business in an increasingly competitive Annex H (continued) 50 environment where clients tend to prefer less cumbersome and more cost- and time- effective partnerships. H e stressed that the Bank will have to increasingly tailor its product to the needs o f the client. 5. Main Conclusions and Next Steps. The Subcommittee welcomed the CAE and agreed with the OED rating o f the outcome o f Bank assistance as moderately satisfactory. They encouraged. management to consider the CAE’s recommendations when formulating the next country assistance strategy and stressed the importance o f drawing lessons from the Bank’s experience in Bhutan for future involvement in small countries where the Bank i s not the primary lender. The following points were raised. 6. Lessons learned. Members noted that the Bank has operated in a different-than- usual environment in Bhutan - being a marginal lender in a very small country, and stressed that i t should adapt better to the country circumstances and show more flexibility, focus and client orientation. M a n y members emphasized the importance o f lessons learned and asked about the implications for Bank assistance in similar situations - being a non-primary lender in a small country, constrained by shortage o f staff on the Bank’s side and limited administrative capacity o f the client. OED replied that it will be preparing a brief “lessons note” o n the Bank’s work in small countries, looking at: (i) selectivity, especially in the context o f coordination with other donor; (ii) types o f knowledge products to be delivered by the Bank and the modes o f their delivery; and ( iii) staff continuity in the resource-constrained environment. Management welcomed the OED initiative and added that in the past, the Small States Task Force has identified the high transaction costs and the limited capacity among the main constraints, and that recent work aimed at reducing transactions costs i s covered in a status report which serves as a background document for the upcoming Small States Forum to be held in tandem with the World Bank-IMF Annual Meetings. 7. Donor coordination. M a n y members stressed the importance o f better coordination with other donors in order not to crowd out the country’s scarce administrative resources and take advantage o f the knowledge accumulated by other institutions. Some members were interested in seeing a more explicit approach to co- financing strategy. Questions were raised about the reasons o f Bank’s secondary role in Bhutan vis-&-vis the ADB. OED explained that this i s a result o f Bank’s conscious strategy in the past o f letting other multilateral donors take the lead in small countries. OED added that other major factors are: physical proximity o f some o f the donors (ADB) and abundance of grant assistance from India. Management concurred with the OED and noted that i t i s likely that the Bank will remain a small player in the country, relative to other donors. 8. Bank’s program. Some members were interested in the reasons for the limited role o f the Bank in Bhutan - whether i t was the large share o f grants f r o m other donors, or the availability o f more attractive alternative sources. In that context, some members asked about the future o f the Bank’s program, noting that the Bank does not necessarily have to play the leading role, but be a better complementary player. Management replied 51 Annex H (continued) that the situation has changed in terms o f Bhutan’s demand for IDA funds, and the authorities have recently expressed interest in the fbll envelope o f IDA loans. Management added that Bank’s program would focus on education and infrastructure, with a limited number o f AAA products. Several members stressed the need to include statistical capacity building. Management replied that while the Bank i s not engaged in a fbll-fledged capacity building activity in Bhutan, it i s providing necessary assistance on an ad hoc basis in selected areas. A speaker asked whether the comprehensive development framework for Bhutan should be oriented to reducing the ratio o f gross capital formation to GDP growth in order to stimulate growth and increase employment. OED replied that the relatively high ratio reflects the capital intensity o f the hydropower sector, and the lumpiness o f investment, as well as the fact that the revenues from the power sector are often used for long-term social investments. 9. AAA. Members generally agreed with OED o n the need to improve the selectivity o f the Bank’s analytical products and noted the importance o f raising client awareness o f their usefulness and potential benefits. They noted the mixed results o f AAA work in Bhutan and asked whether the evaluation o f the AAA products could have been more quantitative. Some members felt that the overall conclusion o f the OED assessment o f inadequacy o f sector work vis-&vis lending program could have been substantiated by more analysis, and brought up the example o f the successfbl program in education that was not preceded by analytical work. OED replied that many o f the problems in other sectors do reflect inadequate sector work, and noted that because the education projects were directed at expanding physical capacity, the lack o f prior sector work was less significant. Chander Mohan Vasudev Chairman 53 STATISTICAL TABLES 55 Bhutan at a glance 9130104 POVERTY and SOCIAL South LOW- Bhutan Asia income i Development diamond* 2003 Population, mid-year (millions) GNI per capita (Atlas method, US$) 0.87 670 1,425 510 2,310 450 1 Life expectancy GNi (Atlas method, US$ billions) 0.59 726 1,038 Average annual growth, 1997-03 Population (%) 2.8 1.8 1.9 2.6 2.3 2.3 GNI Labor force (%) ~ per Most recent estimate (latest year available, 1997-03) capita Poverty (% of population below nationalpoverty line) Urban population (% of total population) 9 28 30 I Life expectancy at birth (years) 63 63 58 - Infant mortality (per 1,000 live bilths) 74 68 82 Chiid malnutrition (% of children under 5) 19 48 44 Access to improved water source Access to an improved water source (% ofpopulation) Illiteracy (% ofpopulation age 15+) Gross primary enrollment (% of school-age population) 62 84 41 95 75 39 92 - Bhutan Male 103 99 Low-income group Female 88 85 KEY ECONOMIC RATIOS and LONG-TERMTRENDS 1983 1993 2002 2003 Economic ratios* GDP (US$ billions) 0.18 0.24 0.60 0.71 Gross domestic investmentlGDP 39.8 46.1 40.0 Trade Exports of goods and serviceslGDP 12.7 31.5 28.4 Gross domestic savingslGDP 8.4 33.6 24.6 - T Gross national savings1GDP 26.9 Current account balancelGDP -42.6 -29.8 -6.9 -3.0 Domestic interest payments1GDP 0.0 1.o 0.3 0.3 savings Investment Total debtlGDP 1.o 41.0 62.6 59.5 Total debt servicelexports 0.0 7.8 4.6 4.7 Present value of debtlGDP 57.6 Present value of debtlexports 250.6 indebtedness 1983-93 1993-03 2002 2003 2003-07 (average annual growth) __ GDP 6.5 6.7 6.7 6.5 ---Bhutan GDP per capita 4.1 3.6 3.8 3.7 Low-income group Exports of goods and services 18.0 4.5 31.6 I STRUCTURE of the ECONOMY 1983 1993 2002 2003 Growth of Investment and GDP (Oh) (ss of GDP) Agriculture 53.3 40.0 33.7 33.2 Industry 25.5 39.7 39.4 39.5 Manufacturing 5.5 10.7 7.3 7.7 Services 21.2 20.3 26.9 27.3 Private consumption 66.8 49.2 54.4 General government consumption Imports of goods and services 24.8 44.1 17.3 44.0 21 .o 43.7 I -GDI "C>'GDP I ' I 1983-93 1993-03 2002 2003 Growth of exports and imports (%) (average annual growth) Agriculture 3.2 4.0 2.5 4.2 45 T Industry 10.4 8.2 11.1 7.3 Manufacturing 14.9 3.7 4.9 2.1 Services 8.2 7.8 4.5 7.9 ' I Private consumption 3.4 9.9 5.5 General government consumption 2.8 8.2 9.7 -30 1 Gross domestic investment 7.5 6.1 -10.7 Exports *Imports Imports of goods and services 6.7 8.1 8.3 Note: 2003 data are preliminary estimates. The diamonds show four key indicators in the country (in bold) compared with its income-group average. If data are missing, the diamond will be incomplete. 56 Bhutan PRICES and GOVERNMENT FINANCE 1983 1993 2002 2003 Inflation (%) II Domestic prices 2o T (% change) 13.6 2.5 1.6 15 Consumer prices Implicit GDP deflator 8.8 6.7 7.8 7.8 10 5 Government finance (% of GDP, includes current grants) 0 Current revenue 23.0 30.3 25.2 98 99 00 01 02 Current budget balance 5.7 15.0 10.6 -GDP deflator *CPl 03 Overall surplusldeficit -12.8 -4.2 -8.3 I TRADE 1983 1993 2002 2003 Export and import levels (US$ mill.) (US$ millions) Total exports (fob) 16 66 98 113 250 T Agricultural products 12 14 15 Electricity 19 49 55 Manufactures Total imports (cif) 66 125 188 189 Food 31 53 72 Fuel and energy 10 25 36 I Capital goods 97 98 99 00 01 02 03 Export price index (1995=100) 38 82 177 Import price index (1995=100) 38 82 177 Exports Imports Terms of trade (1995-100) 100 100 100 BALANCE of PAYMENTS I 1983 1993 2002 2003 I Current account balance to GDP (%) (US$ millions) Exports of goods and services 31 92 130 139 Imports of goods and services 107 162 22% 231 Resource balance -76 -70 -99 -92 Net income 0 0 -25 -13 Net current transfers 0 0 82 84 Current account balance -76 -70 -42 -21 Financing items (net) 74 84 62 81 Changes in net reserves Memo: 2 -14 -20 -59 -10 1 I Reserves including gold (US$ millions) 94 317 376 Conversion rate (DEC, local/US$) 10.1 30.5 48.2 47.1 EXTERNAL DEBT and RESOURCE FLOWS 1983 1993 2002 2003 (US$ millions) Composition of 2003 debt (US$ mill.) Total debt outstanding and disbursed 2 97 378 422 iBRD 0 0 0 0 0 19 39 49 G:7 B:49 IDA Total debt service 0 7 6 7 IBRD 0 0 0 0 IDA 0 0 1 1 Composition of net resource flows Official grants 5 35 32 Official creditors 1 9 93 22 Private creditors 0 -2 0 0 Foreign direct investment 0 0 0 Portfolio equity 0 0 0 World Bank program Commitments 0 5 0 31 1 A-IBRD E. Bilateral Disbursements 0 1 6 7 . B IDA D Other multilateral . F Private G - Short-term ~ Principal repayments 0 0 0 0 C-IMF Net flows 0 1 5 7 Interest payments 0 0 0 0 Net transfers 0 1 5 6 The World Bank Group: This table was prepared by country unit staff figures may differ from other World Bank published data. 9/30/04 M E \9\9 3 3 3 3 i d 2 s d c 60 Annex Table 5: Bhutan - OED and QAG Ratings Source: Business Warehouse as o f March 23,2004 Country Number of Net Number of YOAt Risk Commitment YO Projects Commitment Projects at At Risk Commitment Amt Risk at Risk Bhutan I _ _ 4_ _ _ _ 67 _ _ ~ . 0 0 0 0 Bangladesh 24 2,186 5 21 585 27 India 65 12,693 7 11 1,118 9 Nepal 8 333 2 _ 25 130 39 _ SAR 126 17,127 22 17 2.927 17 lThe World Bank 1,354 94,444 267 20 17,722 19 61 Annex Table 6: Bhutan - Millennium Development Goals 1990 1995 2001 2m 1. Eradicate extreme p o v e r t y and hunger farget = halve 1990 $I a day poverty and malnutrition Population below $ 1 a day (%) Poverty gap at $ 1 a day (“Yo) Percentage share o f income or consumption held b y poorest 20% Prevalence o f child malnutrition (% o f children under 5) 37.9 18.7 Population below minimum level o f dietary energy consumption (%) 2. Achieve universal p r i m a r y education 2015 target = net enrollment to 100 N e t primary enrollment ratio (% o f relevant age group) Percentage o f cohort reaching grade 5 (%) 82 1 90 4 Y o u t h literacy rate (“Yo ages 15-24) 3. Promote gender equality 2005 target = education ratio to 100 Ratio o f girls to boys in primary and secondary education (%) Ratio o f young literate females to males (% ages 15-24) Share of women employed in the nonagricultural sector (%) Proportion o f seats held b y women in national parliament (%) 2 2 9 4. Reduce c h i l d m o r t a l i t y 115 target = reduce 1990 under 5 mortality by hvo-thin Under 5 mortality rate (per 1,000) 166 133 95 Infant mortality rate (per 1,000 live births) 75 67.8 74 Immunization, measles (% o f children under 12 months) 93 85 78 5. I m p r o v e maternal h e a l t h 5 target = reduce 1990 maternal mortality by three-fou Matemal mortality ratio (modeled estimate, per 100,000 l i v e births) 500 Births attended b y skilled health staff (“Yo o f total) I49 6. C o m b a t H I V / A I D S , m a l a r i a and other diseases 2015 target = halt, and begin to reverse, AIDS, etc. Prevalence o f HIV, female (%ages 15-24) Contraceptive prevalence rate (% o f women ages 15-49) 18.8 30 7 Number o f children orphaned b y H I V I A I D S Incidence o f tuberculosis (per 100,000 people) 136 5 Tuberculosis cases detected under D O T S (“Yo) 29 27 I . Ensure environmental sustainability 2015 target = various (see notes) Forest area (“Yo o f total land area) 64 2 64 2 Nationally protected areas (% o f total land area) 21.3 21 2 2 GDP per unit o f energy use (PPP $ per k g o i l equivalent) CO2 emissions (metric tons per capita) 02 04 05 Access to an improved water source ?Yo o f population) 62 Access to improved sanitation (“Yo o f population) 70 Access to secure tenure (% of population) 8. Develop a G l o b a l Partnership f o r Development Y o u t h unemployment rate (% o f t o t a l labor force ages 15-24) F i x e d line and mobile telephones (per 1,000 people) Personal computers (per 1,000 people) General indicators Population 600 1 thousand 695 1 thousand 828 0 thousand 850 8 thousand Gross national income ($) 292 3 million 265.1 million 460 3 million 505 2 million GNI per capita ($) 490 380 560 590 Adult literacy rate (“Yo ofpeople ages 15 and over) Total fertility rate (births per woman) 5.9 58 52 5 1 L i f e expectancy at birth (years) 57 7 59 5 62 7 63.2 Aid (?Yo o f GNI) 17.6 26 6 12 6 Extemal debt (% o f GNI) 31 4 39 2 56 3 Investment (“Yo o f GDP) 36 1 48.7 48 1 Trade (% o f GDP) 60 5 79.3 66 5 Source: World Development Indicators database, April 2002. Note: In some cases the data are for earlier or later years than those stated Goal 1targets: Halve, between 1990 and 2015, the proportion of people whose income is less than one dollar a day. Halve, between 1990 and 2015, the proportion of people whosuffer from hunger. Goal 2 target: Ensure that. by 2015, chi!dren e v e w e r e . boys a?d girls alike,, will be able,to complete a full course of pnmary schooling 2015. Goal 4 target: Reduce by two-thirds, between 1990 and 2015, the under-five mortality rate. Goal 5 target: Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio. Goal 6 targets: Have halted by 2015, and begun to reverse, the spread of HIV/AIDS. Have halted by 2015, and begun to reverse, the incidence of malaria and other major diseases. Goal 7 targets: Integrate the principles of sustainable development into country policies and programs and reverse the loss of environmental resources. Halve, by 2015, the proportion of people without sustainable access to safe drinking water. By 2020, to have achieved a significant improvement in the lives of at least 100 million slum dwellers. Goal 8 targets: Develop further an open. rule-based, predictable, non-discriminatory trading and financial system. Address the Special Needs of the Least Developed Countries. Address the Special Needs of landlocked countries and small island developing states Deal comprehensively with the debt problems of developing countries through national and international measures in order to make debt sustainable in the long term In cooperation with developing countnes, develop and implement strategies for decent and productive work for youth. In cooperation with pharmaceutical companies, provide access to affordable. essential drugs in developing countries. In cooperation with the private sector, make available the benefits of new technologies, especially information and communications.