Report No. 20211-NEP Nepal Public Expenditure Review (In Five Volumes) Volume 1: PER Overview-The Main Report April 11, 2000 Poverty Reduction and Economic Management Unit South Asia Region Document of the World Bank CURRENCY Currencv NRsJUS$ 1980 12.00 1981 12.34 1982 13.24 1983 14.55 1984 16.46 1985 18.25 1986 21.23 1987 21.82 1988 23.29 1989 27.19 1990 29.37 1991 37.26 1992 42.72 1993 48.61 1994 49.40 1995 51.89 1996 56.69 1997 58.01 1998 65.97 1999 68.25 Note: The Nepali fiscal year runs from July 16 through July 15. Source: IMF, International Finance Statistics (IFS), line "rf" (period average). a Starting November 30, 1985, the Nepali Rupee was linked to a basket of currencies. Subsequently in early 1992, a two-tier exchange rate system with partial convertibility was introduced. In February 1993, the dual exchange rate system was replaced by a unified market-determined exchange rate system providing de facto convertibility for all current account transactions. Vice President: Mieko Nishimizu Country Director: Hans M. Rothenbuhler Sector Director: N. Roberto Zagha Task and Team Leader: Para Suriyaarachchi TABLE OF CONTENTS Currency Abbreviations and Acronyms Acknowledgements Economic Development Data Social Indicators Executive Summary CHAPTER 1 Macroeconomic Setting and the Need for Improvement in Public Expenditure Policies ........ 1 A Introduction ........................................................................1 B. The Scope of the Report ........................................................................3 CHAPTER 2 Why is Public Resource Management Weak in Nepal? Key Cross Cutting Issues ...................5 A. Introduction and Key Issues .5 B. Why has public spending in Nepal been ineffective? .6 C. Fiscal Structure and Its Implications .7 D. Deficiencies in Resource Allocation and Management .11 E. The Fungibility of Aid .13 F. Project Implementation Problems .14 G. Corruption .20 H. Performance of Public Enterprises .22 CHAPTER 3 Sectoral Expenditure Programs and Key Issues in Allocative and Technical Efficiency ....... 25 A. Introduction .25 B. Inter Sectoral Expenditure Pattern. 25 C. Education .28 D. Health .35 E. Decentralized Expenditure Programs .44 F. Agriculture, Irrigation and Forestry .49 Agriculture .51 Irrigation .52 Forestry .53 G. Transport .55 H. Roads .55 I. Air Transport .60 J. Power .62 CHAPTER 4 Improving Public Resource Management: The Reform Agenda ............................................. 69 A. Nepal's Development Strategy .69 B. The Public Resource Management Reform Agenda .71 C. Improvements in Budget Allocations .82 D. Role of External Assistance .84 E. Phasing of Reforms .85 ANNEX 1 Fiscal Structure and Its Implications ANNEX 2 The Ninth Plan and Its Macroeconomic Projections The Ninth Plan Macroeconomic Projections ABBREVIATIONS AND ACRONYMS AIC Agriculture Input Corporation APP Agriculture Perspective Plan BPEP Basic and Primary Education Project CAAN Civil Aviation Authority of Nepal CIAA Commission and Inspection of Abuse of Authority DALY Disability Adjusted Life Year DCA Department of Civil Aviation DDC District Development Committee DOLIDAR Department of Local Infrastructure Development and Agricultural Roads DOR Department of Roads FCGO Financial Comptroller General's Office GTZ Technical Cooperation of the Federal Republic of Germany IDA International Development Association LACI Loan Administration Change Initiative MLD Ministry of Local Development MRCU Maintenance and Rehabilitation Coordination Unit NEA Nepal Electricity Authority NFC Nepal Food Corporation NGO Non Governmental Organization NIC National Insurance Corporation NLSS Nepal Living Standard Survey NTC Nepal Telecommunications Corporation NWSC Nepal Water Supply Corporation PDDP Participatory District Development Program PEs Public Enterprises PFHP Population and Family Health Project PIP Priority Investment Program RBB Rastriya Banijya Bank RNAC Royal Nepal Airlines Corporation SDC Swiss Agency for Development and Cooperation SLC School Leaving Certificate SNV Netherlands Development Organization SOEs Statements of Expenditure STOL Short Take-Off and Landing TIA Tribhuvan International Airport TYREP Three Year Rolling Expenditure Program VDC Village District Committee ACKNOWLEDGEMENTS This report was prepared by a team comprising Para Suriyaarachchi (Team Leader, SASPR), Roshan Bajracharya (Consultant), Shakti Shrestha (SASPR), Neena Shrestha (SASPR), Jeeva Perumalpillai-Essex (SASRD), Douglas M. Addison (AFIM4), James Keith Hinchliffe, (SASED), Ian Morris (SASHP), Brajesh Pant (SASED), Guan Zhe Chen (SASIN), Juan Gaviria (SASIN), Tjaarda Storm Van Leeuwen (SASEN), Takeshi Kikukawa (AFTG1) and Tashi Tenzing (SASIN). The Kathmandu office of UK Department for International Development (DFID) also made an important contribution to the institutional analysis in the report. Roberto Zagha (Sector Director, SASPR) and Hans M. Rothenbiihler (Country Director, Nepal) provided valuable guidance and helped to improve the focus and clarity of the report. The report also benefited from comments from Sadiq Ahmed (Chief Economist, South Asia Region), Shantayanan Devarajan (Peer Reviewer, DECRG), Richard Westin (Peer Reviewer, MNCEH), Lawrence DeMilner, (Resident Representative, LvIF), Barend Frielink (Program Officer/Economist Asian Development Bank), Nicholas Highton (Economic Advisor, DFID), Kenichi Ohashi (SACNP), Bigyan Pradhan (SARFM) and Tirtha Rana (SASHP). Naomi Dass (SASPR) and Lin Wang Chin (SACPK) graciously assisted with the arrangements for printing the report. The report draws heavily on various documents and background papers prepared by the donor community and the government for an upcoming Nepal Development Forum (NDF) meeting. The report also benefited from an interactive process of discussions with government authorities, representatives of the donor community and members of civil society held in Kathmandu during February and March 2000, as part of the preparatory work for the NDF meeting. We gratefully acknowledge the cooperation extended by government officials, especially in the Ministry of Finance, the National Planning Commission and the Nepal Rastra Bank, and by the representatives of the donor community in Kathmandu. Economic Development Data GNP Per Capita (US$, 1998): 210 a! Gross Domestic Product (FY99) b/ FY99 Annual Growth Rate (% p. a., constant prices) US$ Million % of GDP FY86-90 FY91-94 FY95-99 GDP at Factor Cost 4754 96.5 4.8 5.6 3.9 GDP at Producers' Prices 4926 100.0 4.6 5.6 3.9 Gross Domestic Investment 956 19.4 19.8/c 21.9/c 22.8/c Gross National Saving 700 14.2 12.2/c 14.2/c 14.2/c Current Account Balance -256 -5.2 -7.6/c -7.6/c -8.6/c Output, Employment and Productivity Value Added, FY 99b/ Labor Force d/ V.A. per Worker US$ Million % of Total Million % of Total US$ % of Average Agriculture 1906 40.1 7.2 76.1 265 52.7 Industry 998 21.0 0.9 9.8 1072 213.5 Services 1849 38.9 1.3 14.0 1391 277.0 Total 4754 100.0 9.5 100.0 502 100.0 Govermnent Finance e/ NRs. billion % of GDP As % of GDP FY99 FY86-90 FY91-94 FY95-99 Revenue Receipts 37.2 11.1 9.0 9.2 11.1 Total Expenditures 63.2 18.9 18.8 17.4 18.4 Regular Expenditures 31.8 9.5 6.4 6.5 9.0 Development Expenditures 31.4 9.4 12.4 10.9 9.5 Overall Deficit -24.7/f -7.4/f -9.8 -8.2 -7.4 External Assistaice 18.7 5.6 7.4 6.2 5.5 Domestic Borrowings 6.0 1.8 2.4 2.0 1.9 Money, Credit, and Prices gl FY94 FY95 FY96 FY97 FY98 FY99 NRs. Billion outstanding, end of period Broad Money (M2) h/ 69.8 81.0 92.7 103.7 126.5 152.9 Bank Credit to Government (net) 23.5 25.2 27.5 29.2 31.8 35.5 Bank Credit to Private Sector 32.3 44.9 57.7 64.7 76.8 90.8 Percentage or index number Broad Money as % of GDP 35.0 36.9 37.2 37.0 42.6 45.7 Consumer Price Index (1983/84=100) 284 305 330 356 370 417 Annual Percentage Changes in: Consumer Price Index 9.0 7.6 8.1 7.8 4.0 12.7 Broad Money (M2) 19.6 16.1 14.4 11.9 21.9 20.9 Bank Credit to Government (net) 0.2 7.3 9.3 6.2 8.6 11.7 Bank Credit to Private Sector 35.1 38.8 28.7 12.0 18.8 18.2 a/ World Development Report 2000, The World Bank. b/ National Accounts of Nepal 1999, Central Bureau of Statistics. c/ Data for Gross Domestic Investment, Gross Domestic Savings and Current Account d/ Nepal Labor Force Survey 1998/99, Central Bureau of Statistics. Balance are as percentage of GDP. e/ Economic Survey FY99, Ministry of Finance. f/ After "non-budgetary receipts" of Rs. 1.3 billion. g/ Nepal Rastra Bank. h/ Includes money supply (Ml) and time deposits. Balance of Payments (US$ Million) Merchandise Exports (Average FY95 - FY99) FY97 FY98 FY99 US$ Million % of Total Exports of Goods & NFS 1295.0 1108.3 1061.9 Woolen Carpets 147.9 35.5 Merchandise, fob 397.4 444.6 525.3 Readymade Garments 112.1 26.9 Imports of Goods & NFS 1854.7 1645.7 1495.9 Others 156.5 37.6 Merchandise, cif 1642.3 1439.1 1290.6 Total 416.4 100.0 of which POL products 126.5 154.0 128.6 Trade Balance -1244.9 -994.6 -765.3 Total External Debt,1998 a/ Non Factor Services (net) 685.2 457.2 331.3 US$ Million External Debt 2646.0 Resource Balance -559.7 537.4 -434.0 Net Factor Income 5.0 13.0 25.0 Debt Service Ratio for 1998 a/ Net Transfers 94.5 103.0 152.9 % of current Receipts Debt Service 7.1 Current Account Balance -460.3 -421.4 -256.1 Official Grants 170.8 176.3 171.4 IDA Lending, January 2000 / Medium & Long Term Capital US$ Million Gross Inflows 197.0 229.8 195.2 Outstanding & Disbursed 1211.1 Principal Repayments 49.8 62.4 61.3 Undisbursed 253.6 Miscellaneous Capital Flows 198.5 254.8 95.5 Overall Balance (-1ncrease) -56.1 -177.0 -144.8 Gross Reserves (end year) 865 969 1128 Rate of Exchange Period Average Rate (US$/NRs.) 57.03 61.95 67.95 i/the World Bank. Nepal Social Indicators Latest single year Same reglon/lnconm group South Low- 1970-75 1980485 1992-97 Asia income POPULATION Total population, mid-year (millions) 12.8 16.5 22.3 1,281.3 2,035.6 Growth rate (% annual average) 2.4 2.6 2.5 1.5 1.7 Urban population (% of popubation) 5.0 7.8 10.9 27.0 28.4 Total fertility rate (births per woman) 6.3 6.1 4.4 3.5 4.0 POVERTY (% of population) National headcount index .. .. 42.0 Urban headcount index .. .. 23.0 Rural headcount index .. .. 44.0 INCOME GNP per capita (USS) 120 170 220 380 350 Consumerpriceindex(1995=100) 17 35 114 117 122 Food price index (1995=100) .. 33 111 INCOMEICONSUMPTION DISTRIBUTION Share of income or consumption Gini index .. 36.7 Lowest quintile (% of income or consum .. 7.6 Highest quintile (% of income or consurr .. 44.8 SOCIAL INDICATORS Public expenditure Health (% of GDP) .. .. 1.2 0.8 1.0 Education (% of GNP) 1.5 2.6 2.8 3.0 Social security and welfare (% of GDi 0.1 0.1 Net primary school enrollment rate (% of age group) Total 58 58 Male .. 79 Female .. 36 Access to safe water (% of population) Total 8 24 59 81 69 Urban 85 78 61 84 80 Rural 5 20 59 80 66 Immunization rate (% under 12 months) Measles .. 34 85 81 74 DPT .. 32 78 87 76 Child malnutrition (% under 5 years) 69 .. 47 53 Life expectancy at birth (years) Total 43 49 57 62 59 Male 44 50 58 62 58 Female 43 48 57 63 60 Mortality Infant (per thousand live births) 160 125 83 77 82 Under5(perthousandlivebirths) 234 180 117 100 118 Adult (15-59) Male (per 1,000 population) 482 376 274 219 274 Female (per 1,000 population) 476 395 314 212 255 Matemal (per 100,000 live births) .. .. 1999 World Development Indicators CD-ROM, World Bank. I EXECUTIVE SUMMARY 1. Nepal, with a per-capita income of only $210 p.a., is the ninth poorest country in the world. Nearly half of its population is below the poverty line; and her social indicators, though improving over time, are comparable to those of poorer African countries. Much of this backwardness is attributable to centuries of political and economic isolation, a land-locked mountainous terrain, and a late start in development-Nepal embarked on economic development very late, only in the nineteen fifties. Since then, considerable public resources, supported by external aid of the order of 6% of GDP p.a., have been spent on promoting development, with noteworthy progress in many areas. Government spending, on average, has been nearly one fifth of overall GDP; and its share of monetized GDP has been significantly higher. Nevertheless, there is considerable evidence that a large share of these resources have been misspent. With the incidence of poverty around 42%, unchanged over the last 25 years, particularly in rural areas where about 90% of the population live and work, the country can ill- afford waste of its scarce resources. The main objective of this report is to identify the incentives and institutional factors that lead to this waste and propose remedies. 2. Nepal's macroeconomy is reasonably sound, with an improving balance of payments, adequate foreign exchange reserves (equivalent to about nine months' imports), relatively modest inflation, fiscal deficits financed largely by external aid and relatively low levels of domestic borrowing. However, it is stuck in a low level equilibrium, with declining investment, modest economic growth, high population growth (2.5% p.a.) and, as noted, widespread poverty. In addition to poor use of public resources, Nepal has also been unable to implement a policy framework conducive to high levels of economic growth. Although a brief period of economic reforms led to a significant acceleration of economic growth (from an average of 4.6% p.a. in the second half the eighties) to about 5.6% p.a. in the early nineties, since then the growth rate has decelerated to 3.9% p.a. in the mid-to-late nineties and to only 2.8% p.a. in the last two years. Moreover, even the faster growth in the early nineties largely benefited the urban areas, and the rural economy has seen little improvement. Fiscal management has been particularly weak; and fiscal stability is being maintained largely at the expense of development activity. Although public spending continues at a rate which should be sufficient to provide a reasonable level of basic public services in most countries, in Nepal there is little to show for it, particularly in rural areas, reinforcing the widespread perceptions of misuse and waste of scarce public resources. 3. There are no simple explanations for the apparent ineffectiveness of public spending in Nepal. A complex web of systemic factors, which cut across virtually all sectors and projects/programs, has consistently contributed to poor implementation and development results. Among these, deficiencies in the budget planning, resource allocation and expenditure management process have been a major factor contributing to low productivity. These are evident at many points in the budget cycle: (i) Budget preparation is bifurcated, with the Ministry of Finance (MOF) preparing the "regular budget" and the National Planning Commission (NPC) preparing the "development budget." These two budgets are not well integrated; and resources are often allocated on an incremental basis (based on previous year's utilization and negotiating strengths of implementing agencies) rather than on resource needs of projects/programs and well-considered priorities. (ii) The budget (particularly its development component) is heavily over-programmed. Because of political pressures to accommodate new projects, resource estimates (for both revenues and external aid) have been consistently over-optimistic, permitting - 11 - the size of the budget to be set at levels which are not consistent with the actual availability of resources and institutional capacity. The consequences of this approach have been predictable: too many projects chasing too little resources, and limited resources spread too thinly among too many projects. Thus, the number of projects/programs have risen from around 400 in FY94 to over 700 in FY98; and budget allocations per project have fallen to about half the level in the mjd nineties, in real terms. (iii) There is little prioritization of the budget. Although there are ambitious five year plans, the link between these plans and the annual budgets is weak. Nepal still does not have a rolling three year public expenditure program which can serve as a bridge between its five year plans and annual budgets, or clearly defined sector strategies which can help define project/program priorities and guide budget allocations. (iv) Despite several years of technical assistance, institutional mechanisms for project screening and expenditure management are still weak. A rigorous process for screening new projects (particularly for fully government financed projects) or a tightly defined core program which can be used as a guide for fund release decisions has not been operative in many years. Similarly, mechanisms for monitoring actual implementation and physical progress of projects remain unsatisfactory. Although there has been considerable improvement over time in financial reporting by implementing agencies due to stronger financial monitoring by the Financial Comptroller General's Office (FCGO), this alone is not sufficient to ensure that funds have been effectively utilized to achieve expected outputs and outcomes. (v) The fund release process favors quick spenders rather than priority projects/programs, as funds are released largely on the basis of whether previous releases have been spent and accounted for by implementing agencies. However, there are instances, for example in the case of local development programs, where the FCGO cannot withhold fund releases to projects/programs even if expenditure reporting is poor or not made on a timely basis, because of political pressures. 4. The lack of ownership of projects and programs by key stakeholders, notably the government and local communities, is another important factor contributing to poor periformance. This is evident at several levels. At one level, many donor-financed projects which constitute about 70% of the development budget are perceived to be "donor-driven ", (i.e., initiated and prepared by donors, with little involvement of and consultation with local stakeholders, and implemented sometimes directly by donors or under special implementation arrangements), rather than owned and implemented by the government. At another level, fully government financed projects also suffer from similar weaknesses. The government budget preparation process is widely perceived as a top-down process, which is unresponsive to local needs and initiatives. Moreover, many HMG projects are centrally administered and managed through line ministries and departments and their district offices, with little involvement of local beneficiary/community groups. An inevitable corollary of weak ownership is a lack of commitment to such projects, manifested in insufficient budget allocations and fund releases, long delays in fulfilling agreed effectiveness conditions and long completion delays, inadequate beneficiary involvement and poor maintenance. 5. Institutional weaknesses remain probably the most critical set offactors affecting project implementation and the effectiveness of public spending across sectors in Nepal. Institutions at all levels-NPC, line ministries, local governments, village committees etc.-are weak. The requisite technical skills are lacking at many points, while motivation is a systemic issue. An important reason for these deficiencies is the weaknesses in the public administration sysitem itself: over-staffing, low salaries, low compression ratios, political interference in appointments and transfers, inadequate recognition of efficiency and performance etc. These in turn affect public resource management directly. As noted, the capacity to monitor the implementation of - iii - the public expenditure program is weak at all levels. Inadequate supervision, poor financial management, dilatory government procedures and lack of co-ordination among government entities, are some other indicators, all of which lead to poor development results. The recent experience of local development programs provides a good example of how these institutional weakness affect the quality and effectiveness of public spending. Although successive governments have strongly emphasized decentralization and channeled considerable funds to district and village councils/committees, the experience so far has been mixed. In a majority of cases, technical capacity to prepare and manage projects is low; financial reporting and accountability is poor; and irregularities and misuse of funds have been numerous. However, there-are also many examples where, with appropriate technical assistance for building capacity and promoting beneficiary involvement and transparency at the local level, there have been encouraging results in terms of better allocation and use of resources and greater accountability. 6. Governance issues have been a major factor undermining the effectiveness of public spending, increasingly so in the last 4-5 years. These have taken many forms, notably (i) excessive political interference at key points in the project cycle; and (ii) increasing corruption. As noted, short-lived governments have been more concerned with starting new projects and securing access to resources rather than with effective implementation of projects/programs and development results. The most common examples have included excessive political interference in the appointments and (frequent) transfers of civil servants and project officials, political involvement even in simple procurement decisions, misuse/diversion of public funds (and project vehicles) for unauthorized purposes, and cabinet decisions serving parochial interests. This has undermined the effectiveness of public resource management in several ways: (iii) such interference has led to increased politicization of the civil service and insecurity among public officials who, to protect themselves in a uncertain political environment, pass on even routine matters to higher levels; (iv) It has also created an environment which is conducive to leakages and lack of accountability. For example, on the revenue side, it has encouraged gold smuggling as well as collusion between tax officials and tax payers in regard to customs valuation, tax assessment etc; and on the expenditure side, increased incidence of financial irregularities at various levels; and (v) However, the mechanisms/legal safeguards for addressing these deficiencies remain weak. For example, although the Auditor General's recent reports have reported numerous financial irregularities, there are no effective legal or institutional arrangements to follow up on these reports and check such abuses. Similarly, although the Commission for Investigation of Abuse of Authority (CIAA) has filed several cases against various individuals in recent years, there has not been even a single case of conviction so far. 7. Apart from the systemic factors discussed above, the pattern of resource allocation and use has also contributed to poor performance: (i) As noted, development budget resources have been scattered thinly among too many projects, with the result that in several cases, allocations are not sufficient to undertake any meaningful development work, in turn leading to completion delays and poor results. (ii) Government-financed projects account for only about one-third of the development budget allocations, but in practice receive the larger share of scarce local currency resources. However, given inadequate screening, monitoring and accountability of such projects, they tend to be even less productive than aided projects. A good example is local development, which receives the largest share of local currency resources, but its performance has been mixed so far. (iii) There is also considerable disconnect between the government's development objectives as articulated in Five Year Plans and annual budget allocations. In recent years, the favored sectors in terms of increased budget allocations have been power, local development and health. The combined share of these three sectors in development budget - iv - allocations has risen from 22% in FY94 to 41% in FY99. This has been largely at the expense of agriculture, irrigation and forestry. Despite their critical importance for agricultural/rural development and poverty reduction, and the government's declared commitment to the Agricultural Perspective Plan (APP), resource allocations for these three sectors combined has fallen from 27% of the development budget in FY94 to 17% in FY99. This disconnect has been even more pronounced with regard to actual expenditures, since the combined share of power, local development and health rose from 18% of development spending in FY94 to nearly 42% in FY99, while that of agriculture, irrigation and forestry fell from 31% to 17% over the same period. In this context, it is worth noting that the relative stagnation of the rural economy has been probably the most important factor contribution to the lack of progress in poverty reduction during the past decade; and the neglect of the agricultural sector in terms of budget allocations and their poor utilization (see below) in turn contributed to poor agricultural performance. On the other hand, power, though very important for non-agricultural sector growth, directly benefits only about 15% of the population mainly in urban areas; expenditures on local development, though having considerable potential for transforming the rural economy, have not been effectively utilized, as noted above; and increased expenditures on health have been largely channeled to support urban hospitals and construction activities without leading to any tangible improvement in rural health care delivery so far. 8. The reasons for poor performance/development results however go beyond sectoral allocation and expenditure patterns. Even where resources are allocated and provided, there are considerable intra-sectoral misallocations and inefficiencies in the use of resources. Some notable examples include: agriculture, where a large part of the development budget has been used until recently to subsidize inefficient operations of the Agricultural Inputs Corporation, and where the extension service has failed to deliver relevant crop research and improved cultural practices to rural farmers; transport, where a significant proportion of resources has been channeled to unplanned construction of uneconomic (political) road projects; primary education where, despite substantially increased funding, the drop-out rates are high and quality of education remains low; and health, where construction of rural health centers is not supported by adequate trained staff and basic drugs, medicines and equipment. Also, across sectors, funding for operations and maintenance activities has been inadequate, contributing directly to the poor quality of outputs and services, as well as to increased rehabilitation/reinvestment needs which cost several times more than the foregone maintenance expenditures. These inefficiencies are reflected across sectors, sometimes in increased unit costs, but primarily through failure to deliver public services of acceptable quality and regularity. Thus, although budgeted funds are spent, service delivery particularly to rural communities remains poor. 9. In addition, in a number of instances, the benefits of public spending accrue largely to the relatively better-off segments of the community and cannot be defended on equity grounds. For example, a significant proportion of increased spending on health has largely benefited the relatively well-off urban population; and in education, a substantial part of the sectoral budget goes for higher education which has primarily benefited the upper income groups. Similarly in power, as discussed above, the benefits accrue largely to urban communities; and the rural areas where the majority of the population live are badly under-served. 10. Finally, the public enterprise sector has been a major user of public resources; but its performance both in terms of financial results and efficient delivery of goods and services has been poor. Many public enterprises (PEs) are poorly managed. A recent analysis of public enterprises carried out by the Ministry of Finance shows that: (i) 28 out of 43 PEs incurred operating losses during FY98-the latest year for which full year data are available; (ii) The total operating profits of the entire PE sector (including operating subsidies provided by HMG to PEs) was only Rs1.3 billion in FY98, equivalent to 1.6% of net capital employed (i.e., share capital, government loans plus other long term borrowing). Since 4 PEs-namely Nepal Telecommunications Corporation (NTC), Nepal Electricity Authority (NEA), National Insurance Corporation (NIC) and the Rastriya Banijya Bank (RBB)-reported a combined operating profit of Rs.3.4 billion (some of which may be open to question too, if proper accounting and auditing standards were applied), the other 39 PEs together made a net loss of Rs.2.1 billion, equivalent to 10. 7% of their net capital employed in FY98. (iii) Government subsidies and transfers to these PEs as equity and long term loans for financing capital investment have averaged around Rs.2-3 billion a year in the past few years, and nearly Rs.60 billion on a cumulative basis. However, the dividends paid by the PEs to the government in FY98 was a paltry Rs.24 million. Moreover, most of the public enterprises (apart from NTC and NEA) have not been paying their debt service obligations to the government, which has to take over itself the responsibility of paying external lenders. Although the government does not directly guarantee the financial commitments of most public enterprises, nevertheless, the government may be exposed to substantial contingent liabilities, particularly in the case of the two largest (government- dominated) banks. 11. Nepal is at a major cross roads in its economic and political development. As noted, widespread poverty and pervasive public perceptions of "failed development" have led to considerable disenchantment among the public, particularly among the rural communities, with development policies, politicians and the Kathmandu/urban elites. There are also new demands from rural communities for greater involvement and participation in decision-making and access to public resources. These pressures are to some extent an inevitable corollary of a social awakening unleashed by the democratization process; unless effectively addressed, they constitute a potential threat to the political, social and economic fabric and even to the democratic process itself. 12. The government's development strategy, as broadly articulated in its Ninth Plan, emphasizes that poverty eradication is its primary development objective. It recognizes as key elements of this strategy, the need to accelerate economic growth (to 6.0% p.a. over the Plan period and to over 7% in the following decade and a half) and slow down population growth, increase employment opportunities, promote social development with fewer regional, gender and ethnic disparities and improve access to basic infrastructure. The Plan's emphasis on harnessing the growth potential in agriculture, forestry, hydropower and tourism; undertaking much-needed investments in infrastructure and human resource development; enlisting the private sector as the key agent in development, supported by complementary role for the public sector and market- oriented economic policies, is highly appropriate. The Plan however suffers from a number of weaknesses: (i) Its key targets are overly ambitious. (ii) It tries to do too many things in many sectors without recognizing resource limitations and priorities. (iii) It is also doubtful whether the Plan can be effectively implemented without a significant improvement in the political environment. Indeed, in the first two years of the Plan period, the Plan's implementation performance and sectoral achievements have fallen significantly below the targets. 13. Alternative scenarios show that under more realistic assumptions, resource availabilities and sustainable levels of public investment would be significantly less than anticipated in The Ninth Plan. Nevertheless, by adopting appropriate public resource management policies and economic reforms, it would still be possible for Nepal to achieve economic growth rates of the - vi - order of around 5.0% p.a. and improve the pattern of its economic growth so as to benefit the largely rural poor, while making modest improvements in national savings and the balance of payments. Given the past record of public resource management, this however will be a major challenge. 14. This report takes the view that traditional budget reforms, as important as they are, alone will not be sufficient to improve the effectiveness of public resource management in Nepal at this time. To make such reforms work, several important changes, in the political environment, behavior and attitudes and institutional capacity, will be needed. 15. Efforts to improve public resource management in Nepal must begin with a re- examination of the role of the public sector in the economy. (i) Even under optimistic assumptions regarding revenue mobilization, the government's resource position will improve only modestly over the next few years; and this will severely limit what the state can do in terms of public spending. (ii) Second, given the poor record of the public sector in terms of delivering outputs and services to the public efficiently, there is little assurance that more public spending will lead to better performance. (iii) Third, the experience of many countries, including those in South Asia, clearly demonstrates that less direct government involvement, rather than more, is compatible with and conducive to faster economic growth and better provision of goods and services to consumers. Countries which rely on market forces and effectively harness the creative energies of their private sectors and civil society through liberal economic policies have a much stronger record of performance than those who do not. 16. This report takes the view that, given its serious fiscal constraints, the government will need to focus its limited resources on key priority areas where such resources can be put to most effective use. (i) Wherever the private sector can undertake economic activities and/or deliver goods and services competitively, there is little justification for public involvement in such areas. (ii) Public involvement should be limited to those areas and activities which the private sector will not find attractive, i.e. where social benefits exceed private costs, but are not adequately factored in by private providers. (iii) In terms of resource allocation, on a broad sectoral basis, public resources should be channeled to such areas as human resource development (education, health, drinking water), rural infrastructure, farm-to-market roads and irrigation priorities in the APP, poverty alleviation programs, and backward area development, among others. Additional resources are also needed for 0 & M activities virtually across the board. But areas such as telecommunications, power generation and tourism should be left to the private sector. (iv) Even within the priority areas identified above for public investment, there is considerable scope for private involvement within specific sub-sectors. Some notable examples include: fertilizer and seed supply and distribution, higher education, urban hospitals and air transport. (v) There is also little justification for continued government involvement in many public enterprises, particularly where these involve commercial activities. Where there might be justifiable social reasons for public involvement in some enterprises, the social costs of such involvement should be quantified and separated from enterprises' losses. (vi) Finally, even where the government needs to remain involved, alternative ways of carrying out these responsibilities and delivering outputs and services more efficiently to the public need to be examined. 17. Ensuring good governance and transparency is the most critical factor which can have a relatively quick and decisive impact on improving the effectiveness of public spending over the short to medium term. For this purpose, the governance issue needs to be interpreted broadly (i) - vii - While controlling corruption and misuse of public resources is very important, good governance also needs to include broader and more effective political commitment to development by the government, major political parties and civil society at large. For any technical solutions to work effectively, the behavior of political leadership and political parties needs to change significantly. The key challenge in this regard is to ensure constructive involvement of key political parties and other constituencies to facilitate the development process and public resource management: (ii) There is an urgent need for a national consensus among major political parties on key development issues and solutions and how to go about implementing them. For example, with regard to key economic reforms, such as VAT, privatization, price adjustments etc., such a consensus is lacking. Political parties support reforms when they are in power, but oppose them when they are in the opposition. Where genuine differences of views exist on such issues, it is essential to let the government take necessary decisions and bear the responsibility for them. (iii) Engaging political leadership and political parties early on in the budget preparation process is essential in order to make them understand the realities of the resource situation and the trade-offs that have to be made. (iv) It is also necessary to minimize political interference in the public resource management process, for example, in staff transfers and procurement decisions; and to (v) Re-channel the energies of politicians from advocacy for new projects to greater involvement in monitoring and implementation of those already under execution. 18. Similarly, decisive action to formulate an anti-corruption agenda, for example, to reduce revenue leakages and the misuse and waste of public resources, and to strengthen legal and institutional mechanisms to ensure greater accountability can have an important impact in terms of improving the effectiveness of resource use even in the short term. In addressing this issue, the government undoubtedly needs to take the lead; but it is ultimately the responsibility of Nepalese civil society, including major political parties, authorities at different levels, the private sector and NGOs. It is also important to recognize that whether the governance situation in Nepal is better or worse than in other developing countries is a moot point. As one of the poorest countries in the world, Nepal can least afford the misuse and waste of its limited development resources through continued governance problems; and it needs to address these issues quickly without further semantic debate. 19. To improve the effectiveness of public spending, it is necessary to promote greater local ownership of the public expenditure program. This will require actions both from the government and external donors. For example, the government will need to take the lead in designing, financing, and implementing the development program and, even more importantly aid co-ordination itself. It will need to decide what its own development priorities and programs are, and ask the donors to support such programs. Where donor aid does not fit its priorities and program objectives, the Government will need to reject such "donor-driven" aid. This will of course require building up its institutional capacity and a greater degree of self-financing/self- reliance through stronger efforts to mobilize revenues, which may take considerable time. External donors will need to support this process, as discussed in para 28 below. Clearly such a strategy will entail risks for both donors and the government. But, if it works, it should help ensure greater sustainability and effectiveness of Nepal's development efforts. 20. Similar efforts are needed in the case of fully government-financed programs also. - To improve effectiveness, it is necessary to create greater synergies between institutions at the center and those at local levels and to build partnerships involving the private sector, NGOs and other stakeholders. This can take place at many levels, for example through: (i) greater - viii - devolution of responsibilities by NPC to line ministries for program formulation (within sectoral guidelines and policies agreed with the National Planning Commission), monitoring and supervision; (ii) increased involvement of local bodies and beneficiary groups in determining priorities and designing projects to beiter reflect local needs and implementing such projects; (iii) developing alternative approaches to (currently centrally managed) project implementation and service delivery, for example, involving the private sector as an alternative to traditional government mechanisms, contracting out to NGOs specific tasks such as supervision, technical services or program implementation on behalf of community groups; and (iv) strengthening the decentralization process (see below). 21. Another element which is vital for improving public resource management is strengthening the institutional capacity for carrying out development activities. Despite considerable technical assistance, institutional capacity remains weak at virtually all levels; but improvements in two areas are particularly important: (i) civil service reform and (ii) decentralization. 22. Civil Service Reform needs to be geared to the type of development administration that is needed to deliver public services more effectively in the future and to the severe local resource limitations hampering Nepal. First, with increased emphasis on transferring responsibility for development activities to local levels through decentralization, the role of the center, line ministries and supervisory agencies will need to change over time. For example, the NPC may need to transform itself into a planning agency responsible for setting overall development goals and priorities as well as sector strategies, while the MOF takes the lead in formulating and managing well integrated annual budgets consistent with these priorities. Similarly, the line ministries themselves will need to change from actual implementers of programs to facilitators, while taking on broader responsibilities for sectoral program formulation, provision of technical support to local bodies, and supervising and monitoring project/program implementation. Such a transition will obviously take considerable time, but will have important implications for skill needs, staffing levels and personnel management. Second, to provide adequate incentives, the overall compensation levels may need to increase significantly, together with a widening of salary differentials between lower and higher grades (i.e. decompressing the salary structure) and the introduction of a merit- based, performance-oriented rewardlpromotion system. How to pay for civil service reforms thus will be a major issue. Third, different alternatives therefore will have to be considered. Enhancing government revenue through tax reforms is one such option. This will however present its own challenges. Moreover, additional revenues are also needed for increased 0 & M and other development activities. Other alternatives therefore will also have to be looked at, including downsizing, as well as donor funding for civil service reforms on a relatively short term basis. External assistance however is likely to be forthcoming only if a credible reform program is developed and initiated. 23. Decentralization is an important alternative to the presently highly centralized system for improving service delivery to local communities and for enhancing effectiveness of public spending. It is however not a complete substitute for the role of the central government. Even as more functions and responsibilities are transferred to local levels, the line ministries and government departments will have an important role to play in facilitating, providing technical support to, and supervising and monitoring the activities of, local bodies. It is also important to note that decentralization is not an easy solution; and several things need to be done to develop an efficient management and delivery system at the local levels: (i) First, it is necessary to remove the political and bureaucratic impediments which currently hamper the decentralization - ix - initiative. (ii) Second, it is necessary to improve the technical capacity of local bodies (DDCs and VDCs) for identifying their needs and priorities; resource mapping and planning; project design, preparation and implementation; accounting, record-keeping and expenditure reporting; etc. Although some efforts have already been made by the government and external donors to build capacity at district and village levels, much more remains to be done. (iii) Currently substantial resources are already allocated to local bodies, in addition to several parallel programs implemented by central government ministries and departments, (for example in health, education, drinking water, rural roads etc., among others). What is needed is to streamline these centrally managed programs to ensure that they do not overlap with those that are locally managed, so as to avoid duplication and waste of resources, together with clearer definition of responsibilities of both local bodies and line ministries for these programs. However, as noted above, the needs of deprived and backward areas may require special consideration. (iv) The most effective way of promoting better management and accountability at local levels is by enhancing the participation of beneficiary groups in the decision-making and implementation process. Greater transparency (for example, through an "open-book system") in regard to the availability of budget funds, the nature of projects being financed and public access to financial records have permitted greater accountability and better use of resources. This would call for greater efforts to promote community mobilization and empowerment, particularly of the poorer and under-privileged groups in the communities. 24. These behavioral and institutional changes need to be complemented by technical reforms to improve fiscal management. To improve expenditure management, (i) It is desirable to integrate the regular and development budgets. Instead of the current practice of allocating resources on an incremental basis, it is necessary to look at the resource needs for 0 & M, other recurrent activities and for ongoing development projects/programs in an integrated manner, and to provide the required resources for them in terms of priorities for growth, poverty reduction and service delivery objectives. (ii) To control the development budget and minimize the waste of resources associated with over-programming, it is essential to ensure greater realism in resource forecasting. Projections of revenue, aid inflows and the size of the development budget would need to be in line with recent performance and reasonable expectations. (iii) Such estimates, prepared for three years at a time, should provide a medium term framework for the development budget. Such a framework in turn would help bring out the key trade-offs that would have to be made, including the need for prioritization and the (limited) scope for new project starts, and help persuade the political leadership about the need to rationalize or downsize projects/programs which are not working well in order to bring the development program to a more manageable level. (iv) This needs to be combined with strengthening project screening and appraisal capacity in the National Planning Commission and the line ministries (by providing adequate staff and budgetary resources) so that greater discipline could be exercised over the process of initiating new projects into the development budget. (v) The budget process also needs to be made more responsive - less "top-down" and more "bottom-up" - in terms of accommodating programs (so long as they are consistent with sectoral strategies and priorities and the available resource envelope) proposed by local level constituents and line ministries. 25. Once formulated, the expenditure program would need to be better managed, without resorting to ad-hoc expenditure cuts which disrupt project implementation. To strengthen expenditure management and control, a number of steps can be taken: (i) The mid year budget review process could be used to reassess resource outlook during the year to make expenditure adjustments in a more systematic manner. (ii) To protect priority projects from any resource shortfalls, a core program of priority projects would need to be formulated and fund releases to projects/programs based on the core program. (iii) Expenditure reporting by spending units need to be improved, both in terms of timeliness and quality of reporting. (iv) The capacity for monitoring expenditures as well as physical progress of projects/programs needs to be strengthened at various levels of the government, particularly the NPC and line ministries. For this purpose, the government may want to consider re-instituting the practice of holding periodic progress/implementation reviews by the Prime Minister and the Cabinet with active involvement of parliamentarians in this process, so as to send out strong signals regarding the importance of speedier implementation. 26. Recent experience of major donors suggests that the key to improving project implementation is to clean up the portfolio, focus attention and resources on a smaller number of projects/programs and to monitor their implementation closely. In addition to measures outlined above, other steps can be taken to address problems which currently hamper project implementation: (i) To reduce procurement delays, standard bidding documents have been recently adopted by the Government. This system needs to be made effective by familiarizing government staff with the new procedures through training. (ii) Even more importantly, political interference in procurement decisions need to be eliminated. Clearly defining the approval limits for procurement decisions that can be taken at various administrative levels, together with clarification of procedures to be followed and mandatory punishments for transgressions (through an anti-corruption program) may be needed for this purpose. (iii) To encourage project managers to make project-related decisions and to hold them accountable, an atmosphere conducive to such decision-making needs to be created by reducing insecurities associated with frequent political changes. Strict implementation of the recently amended civil service rules regarding staff transfers would help reduce such insecurities. 27. If such actions are taken to improve the effectiveness of public spending, then additional revenue mobilization by the government would be warranted. Increased revenues could help finance anticipated increases in current expenditures (including those for civil service reforms), provide for much needed 0 & M expenditures and counterpart funding for development projects, and help promote greater self-reliance for and ownership of the development program. Without making the improvements on the expenditure side, however, it will be difficult to justify tax increases, since there would be little assurance that increased resources will not be misused or wasted. 28. The longer term solution to Nepal's complex fiscal management problems would be for the government to take charge of its overall expenditure program, integrate the regular and development budgets and prioritize it to truly reflect its development/poverty reduction needs and priorities; and for the donors to support such a program with the additional resources that are needed without imposing their own project and technical assistance priorities. While this is a desirable longer term objective, the transition to it is likely to require a more gradual process. To support such a transition, important behavioral changes will be required from external donors as well as the government. While the government takes the lead in, and ownership for, introducing critical fiscal reforms, donors' role also will need to change from providers of "supply-driven" aid, implementers of projects, and purveyors of expensive technical assistance to facilitators who will provide financial and technical support that fit the country's needs and priorities in a way that fosters local ownership, institution-building, and longer tern sustainability. Towards this end, donors could support the government's fiscal reform process in several ways: (i) The government's technical capacity to undertake budget reforms is limited at present; and donor assistance will be needed to help carry out this work (for example, to help the government to - xi - formulate a prioritized three year expenditure program, and sectoral strategies and investment programs consistent with those priorities), in the short term, as well as for strengthening institutional capacity at various levels over the longer term. (ii) Donors will need to accept the consequences of expenditure rationalization (for example, downsizing/canceling poor projects, limiting new project starts for some time etc.); and adjust their own aid programs to support only those projects/programs which fit the country's needs. (iii) It will also require stronger efforts among the donors themselves, within the context of agreed sectoral investment programs, to co- ordinate their support without competing with each other. (iv) It may also call for changes in the way donors prepare and manage projects in order to promote greater local ownership of project preparation, implementation and management. For example, in regard to technical assistance, instead of extensive reliance on external consultants, they need to look at alternative more cost- effective ways of providing technical support (for example, using teams of local consultants under the direction and supervision of expatriate advisors who could be hired for short periods of time). Also, technical assistance needs to be carefully tailored to the country's needs, rather than being a mechanism for developing new follow-up projects, as too often is the case now. (v) Similarly, donors need to examine whether quick disbursing cash support is warranted under present circumstances, given the fungibility of such assistance. Such assistance can be potentially useful, but only when it does not substitute for revenue generation by the government and where the public expenditure program is reasonably well prioritized, so that it actually supports such priorities and sound sectoral expenditure programs. Such an approach will entail risks for both the government and the donors, since project preparation could take more time, projects may have to be smaller and more simple, and aid utilization and commitments may decline initially. Even so, such risks are worth taking if they lead to greater ownership, better development results and increased sustainability over the longer term. Chapter MACROECONOMIC SETTING AND THE NEED FOR IMPROVEMENT IN PUBLIC EXPENDITURE POLICIES A. INTRODUCTION 1.1 This report focuses on improving public resource management and aid effectiveness in Nepal in order to reduce its widespread poverty. 1.2 Nepal, with a per-capita income of only $210 p.a is the ninth poorest country in the World.' Isolated for centuries because of its inward-looking regimes and land-locked mountainous terrain, Nepal embarked on the process of economic development very late, only in the nineteen fifties. At that time Nepal virtually did not have any roads, schools (except for the privileged ruling class in Kathmandu), basic health care or modem amenities. In subsequent years, some progress has been achieved in many areas, including greatly increased access to primary education, a country-wide network of health posts, a basic road network in the low-land Terai and parts of the hills, substantially expanded irrigation and community forestry, and a basic electricity and telephone system. Nevertheless, Nepal still lags behind other South Asian countries in most aspects of development; and her social indicators are comparable to those of poorer African countries rather than to those of her South Asian counterparts (Table 1.1). Table 1.1: Social Indicators in the South Asia Region/a Indicator Nepal Bangiadesh India Pakistan Sri Lanka SAR/b SSAc Poverty Ratio (%), 1991-92 42/d 42.7 40.9 34.0 40.6 1995-96 42.0 35.6 35.0 .. 35.3 GDP Per Capita ($) 210 260 380 480 740 380 490 Gini (Inequality) Index 35.0 28.3 29.7 31.2 30.1 Population Growth Rate (%), 1980-96 2.5 2.1 2.0 3.0 1.4 2.1 2.8 Total Fertility Rate 5.0 3.4 3.1 5.1 2.3 3.4 5.6 Contraceptive Prevalence Rate ? 45.0 43.0 14.0 n.a Life Expectancy (years) Male 57 57 62 62 71 61 51 Female 57 59 63 65 75 63 54 Infant Mortality 85 77 65 88 15 73 91 (per 1000 live births) Maternal Mortality 1500 850 437 340 30 (per 100000 live births) Adult Illiteracy % - Male 59 51 35 50 7 38 34 Female 86 74 62 76 13 64 53 Gross Primary Enrollment (%) 100 92 100 74 113 99 75 Net Primary Enrollment (%) Access to Safe Water (%) 48 79 81 60 n.a. 78 45 a/ Except where stated, the data are for 1996. Source: World Development Indicators. bI South Asia Region. cd Sub-Saharan Africa. di Poverty ratio for Nepal for 1991-92 is not available. Estimate for 1984.85 was 42%. Using similar methodology comparable figures for 1996 range from 42% to 50%. 'Only 8 others - all low income African countries - rank below Nepal in per capita income. - 2 - 1.3 Early efforts to accelerate economic development through increased public spending and foreign aid helped to accelerate economic growth in the mid eighties and early nineties and expand the rudimentary social and economic infrastructure, although as discussed below, the results achieved have not been commensurate with the resources spent. Reflecting these initiatives, total government spending as a percentage of GDP rose from 9% in FY75 to 19% in FY99; while foreign aid disbursements also increased correspondingly from 2.3% of GDP to 6% over the same period. However, Nepal has not been able to sustain sound economic policies conducive to rapid economic growth and development. Thus, the pace of progress slowed down markedly in the last 5-6 years, with economic growth decelerating from an average of 5.6% p.a. in the early nineties to 3.9% p.a. in the second half of the nineties and to only 2.8% p.a. in the last two years. Since the population has been growing at an annual rate of about of 2.5% p.a, there has been little increase in per capita incomes over this period. To a large extent, this has been due to continued political instability-there have been nine different governments during this period. Although successive governments have maintained macro-economic stability, there has been little action to accelerate economic development. Political instability, along with increasing politicization of the civil service and corruption, has also undermined development administration and the effectiveness of public spending. In addition, continued easy availability of fungible aid, against a background of weak budget planning and management, has enabled Nepal to undertake projects and programs of doubtful economic justification (Chapter 2). 1.4 Recent research-a Nepal Living Standards Survey (NLSS) was completed in 1996- indicates that, despite sustained increases in public spending and aid inflows over the past two decades, there has been little or no improvement in the poverty situation in Nepal. Although differences in survey definitions and coverage make strict comparisons of NLSS findings with those of earlier surveys difficult, one statistic is truly staggering: Nearly half of Nepal's population remains below the poverty line, as was the case in the mid-seventies. This is a reflection of weak unbalanced economic growth over many years and sustained high population growth (averaging 2.5% p.a.) which have combined to reduce the rate of per-capita income growth to the lowest in the region. Thus, while neighboring countries have made some headway against poverty, Nepal has at best stood still. Population dynamics in the meantime has increased the absolute number of Nepalese living in poverty by more than 60%-from about 5.7 million in 1976 to an estimated 9.2 million twenty years later. Moreover, the national averages for poverty statistics mask dramatic regional and gender disparities. For example, in the urban Kathmandu valley the incidence of poverty is only 4% and illiteracy is about 24%, where as in the rest of the country poverty is ten times as high and illiteracy is almost three times as high. Poverty is also more widespread and deeper in the more remote areas-the mid and far western development regions and the mountain belt. There are also significant differences between men and women across socio-economic strata, among others, in terms of literacy, life expectancy, legal status and access to resources. For example, illiteracy is as high as 80% among women, while it is less than 60% for men. 1.5 Despite these recent setbacks, Nepal's macroeconomic situation remains reasonably sound. Its broad macroeconomic indicators are comparable and in some respects better than those of other South Asian countries. For example, its balance of payments situation -has strengthened, gross foreign exchange reserves have increased steadily to the equivalent of nearly 9 months' imports, monetary growth has been moderate, and government's domestic borrowing to finance its fiscal deficit remains below the South Asian average. Thus, Nepal is not a country - 3 - which is about to fall apart macroeconomically because of fiscal profligacy. Its main problem has been a persistent failure to sustain development activity and utilize substantial public expenditure and donor aid programs effectively to mitigate its widespread poverty (see below). 1.6 The slow progress in reducing poverty and accelerating Table 1.2: Sustainability of Projects economic development has given Countryt Projects Proets at CommIt. Rma:.m rise to considerable disenchantment ReWone Rpk t%) at risk 39 in29x Nepal 1 4 50 39 29 in recent years in Nepal both with Bangladesh 22 18 7 100 regard to public expenditure policies India 72 14 10 100 Pakistan 29 52 63 67 and foreign aid. Many agree that Sn Lanka 14 14 18 100 public resources are not effectively South Asia 154 25 18 74 Bank Wide 1574 26 22 67 used to provide essential * Bhutan and Maldives together have a total of 3 projects and are infrastructure and public services therefore not Included in the table. especially in rural areas. As noted, the quality and coverage of most public services remain unsatisfactory. Projects in Nepal generally take much longer to complete as compared to other Asian countries, resulting in cost overruns and time delays, which reduce anticipated benefits. Moreover, even when completed, most projects are not properly maintained (for lack of adequate 0 & M and beneficiary involvement), and are not sustainable. A recent study of the World Bank-financed projects completed in the 1990s (undertaken by the Operations Evaluation Department of the World Bank) indicates that in the case of Nepal, implementation performance was satisfactory in only 38% of the projects (compared to an average of 70% for the South Asia Region); borrower preparation was satisfactory in just 40% of projects (81% in South Asia Region) and compliance with loan covenants was satisfactory in 52% (70% in SAR). While there has been some improvement in portfolio management in recent years, this progress has not been sustained. Thus, currently 50% of the bank's portfolio is classified as "At Risk" as compared to 30% two years ago, the South Asia Region average of 25% and Bank-wide average of 26%. Only Pakistan (52% of the portfolio at risk) appears to be worse than Nepal in this regard. While other multilateral and bilateral donors claim slightly better results, the experience of many of them with regard to implementation performance and sustainability of projects has been broadly similar. B. THE SCOPE OF THE REPORT 1.7 Against this background, this report reviews Nepal's public expenditure program and associated issues with a view to helping the Government to allocate and utilize its limited resources more efficiently in order to support development/poverty reduction objectives articulated in its Ninth Plan. The report starts with an overview of recent fiscal trends and cross- sectoral issues, highlighting the pervasive weaknesses of Nepal's fiscal system and how they affect development management and performance (Chapter 2). Chapter 3 reviews the public expenditure program in greater detail, and examines its relevance and effectiveness. Chapter 4 then discusses the public expenditure reforrn agenda, covering a range of key areas/issues (such as governance, ownership, institution building, decentralization, civil service reform, and inter- and intra-sectoral budget allocations, among others), which need to be addressed in order to improve development results of both public spending and donor aid. Chapter WHY IS PUBLIC RESOURCE MANAGEMENT WEAK IN NEPAL? KEY CROSS CUTTING ISSUES A. INTRODUCTIONAND KEY ISSUES 2.1 The nature of the problem. Nepal is not a country which is facing imminent fiscal collapse. To the contrary, Nepal's fiscal situation remains quite stable. Public spending continues to be managed in a way which is consistent with the maintenance of macroeconomic stability. Although Nepal's overall budget deficit, equivalent to 7-8% of GDP p.a. before grants, is somewhat higher than those of neighboring countries (see below), this deficit is financed by foreign grants and highly concessional loans. Consequently, Nepal has managed to keep gross domestic borrowing generally below 2% of GDP p.a.-well below the levels of most of her neighbors. Thus, Nepal's external debt as a ratio of its GNP is about 50% and domestic debt ratio is only about 15%. 2.2 Despite this apparent stability, however, Nepal is facing a serious public resource management crisis. This crisis is about the lack of effectiveness of public spending in one of the poorest countries in the world, which can ill-afford waste and inefficiency. These inefficiencies are reflected most often in terms of poor quality and irregular delivery of public services, especially to rural communities. Even though public resources continue to be spent as budgeted, very little of the planned benefits of public spending trickle down to the intended beneficiaries. Notable examples include, among others: an education system where the quality of public schooling is very low, and where dropout and wastage rates are high (it takes on average 8-10 years to complete a five year schooling cycle at both primary and secondary levels, and only about 40% of students who enroll actually complete schooling); in health where most rural communities do not have access to basic health care (clinics without staff and drugs and medicines); a dysfunctional urban water supply system and even more limited coverage in rural areas (water taps without water); in agriculture, an ineffective research and extension system which does not respond to farmers' needs; in irrigation where, despite heavy past investments in surface irrigation, only about 15% of cultivable land has year-round irrigation; and public enterprises which have consistently failed to provide basic inputs and services to clients on a timely basis (fertilizer and seeds to farmers, basic food and medicines to communities in remote areas who are perennially affected by famine and epidemics) etc. This "failure of fiscal policy" is seen as a major factor contributing to widespread poverty (para. 1.4). This in turn has given rise to concerns about the appropriateness of public spending and foreign aid itself' and the need for change in development policies. 2.3 As noted below, the aggregate level of public spending is not a major public expenditure issue in Nepal. Total public spending is broadly comparable to the South Asian average (Annex 1). However, Nepal has been able to maintain such a level of public spending (well beyond what is permitted by its own revenue effort) only because foreign aid finances a large share of total public expenditures in Nepal. But, given the widespread perceptions of inefficiency and waste of public resources, there are already indications of "aid fatigue" not only among the Nepalese, but 1 See e.g. "Nepal's Failed Development -- Reflections on the Mission and the Maladies," Devendra Raj Pandey, April 1999, as a leading example of many similar views. -6 - also among the donors themselves. Donors therefore are beginning to look at ways of linking the availability of aid directly or indirectly to country performance. This in turn will mean that, in order to maintain (let alone to increase) aid flows and public spending at present levels in relation to its GDP, Nepal will need to take effective steps to improve the quality and efficiency of public resource management, and particularly aid resources. This will be necessary in any case if Nepal is to effectively address its central development challenge of reducing poverty. B. WHY HAS PUBLIC SPENDING IN NEPAL BEEN INEFFECTIVE? 2.4 There is no simple answer to this question since the factors contributing to weaknesses in public resource management are numerous and complex. Nevertheless, these can be grouped under a few broad headings: (i) weaknesses in Nepal's fiscal structure; (ii) deficiencies in the way resources are allocated and managed; (iii) limited institutional and implementation capacity at virtually all levels of government; (iv) the pattern, quality and incidence of specific sectoral/individual programs and projects themselves; and (v) systemic implementation problems. Among these broad headings, a number of specific problems are singled out for more detailed discussion in this report: (i) Nepal's fiscal structure is relatively weak. The government's recurrent expenditures have been rising steadily, preempting virtually all of its limited revenue. However, most of these expenditures consist of wages and salaries and debt service payments; and essential 0 & M activities are under-funded, affecting the efficiency/quality of outputs and services. (ii) The planning and budgeting system is also weak; and budget formulation has been unrealistic, especially in the last few years. There is little prioritization of the expenditure program. Due to continued addition of new projects and programs, resources are spread too thinly among too many projects/programs, leading to under-funding and implementation delays. Thus, few projects are completed on time. (iii) Limited institutional capacity is a major hindrance in several ways: A weak, underpaid and unmotivated civil service is a serious bottleneck. Also, the Government's expenditure reporting and monitoring systems are underdeveloped. As a result, the links between budget allocations, actual spending and associated outputs/outcomes remain tenuous. (iv) Lack of ownership (especially of donor-assisted projects and programs) is an important contributory factor to poor performance. Project/program formulation has generally been a top-down process, with little involvement/participation of key stakeholders, particularly at lower levels. At one extreme, most donor-financed projects are seen by the government as being "donor-driven." At the other extreme, there has been very little involvement of beneficiaries in the design and implementation of projects/programs and in maintaining and operating publicly created assets, although there has been some recent efforts to change this situation. (v) Several other factors, directly or indirectly contribute to delays in project implementation. The more important among these include: excessive political interference in the appointments and transfers of key staff and in the procurement process, delays in decision-making at various levels; and dilatory government processes, as well as the lack of uniformity in the procedures adopted by the government and various donors, etc. (vi) These difficulties have been compounded in recent years by serious governance issues. Given the absence of effective institutional arrangements to enforce accountability and the shortness of governments which limits the time they have take advantage of their tenure, there are indications that corruption has increased significantly, leading to diversion of public funds to unauthorized uses, and to misuse/leakages of resources. (vii) An inefficient public enterprise sector continues to be a major drain on public resources. Nepal still has about 43 enterprises in the public sector; and among these, examples of unproductive use of public resources abound. Annual budget transfers to such enterprises for providing both operational subsidies and investment funds are considerable-about 1% of GDP p.a.-in turn depriving other legitimate development activities of much needed funds. 2.5 This chapter discusses these broad cross-cutting issues which undermine the effectiveness of the public expenditure program in general, while sector-specific issues which hamper the allocative and technical efficiency of sectoral expenditure programs are discussed in Chapter 3. C. FISCAL STRUCTURE AND ITS IMPLICATIONS 2.6 Weaknesses in Fiscal Structure. From one perspective, a major factor constraining public resource management is the inherent weaknesses in Nepal's fiscal structure itself. In recent years, rising current expenditures have preempted much of the meager revenue growth. This has had two important consequences: (i) Public savings have been generally low or negative; and much of development activity has been directly dependent on foreign aid; and (ii) since current expenditures largely consist of wages and debt service payments, essential 0 & M activities have been under-funded. 2.7 The first argument however needs to be placed in its proper context. The reliance on foreign aid by itself does not contribute to public expenditure problems. Indeed, the increased availability of aid has enabled Nepal to maintain higher levels of public spending than would have been possible on the basis of its revenue performance alone. However, the easy availability and fungibility of aid, against a background of weak budget planning and management and political pressures, have enabled Nepal to undertake projects and programs of doubtful economic justification (para 2.22). It has also created a mindset of dependence and lack of accountability. In other words, while low public savings are undoubtedly a serious impediment, much of the responsibility for ineffectiveness of public spending can be traced to poor public expenditure policies and management and other related, particularly institutional, deficiencies (see below). 2.8 Overall trends. Table 2.1 summarizes the recent trends in Nepal's key fiscal aggregates: * Nepal's total expenditures have fluctuated around 18-19% of its GDP. While regular expenditures have risen steadily from 6-7% of GDP in the early nineties to 9.5% in FY99, (largely due to rising compensation and debt service payments-para 2.11), development expenditures have been falling as a percentage of GDP. * Nepal's revenue/GDP ratio has stagnated around at 11% of GDP. In the early 1990s, as part of its economic reform program, Nepal made strong efforts to raise its revenue ratio from about 9% of GDP in FY91 to 11% in PY95. However, these efforts have weakened in -the past few years. - 8 - . The revenue surplus (i.e. revenue minus regular expenditure) which represents the government's own contribution towards financing the development budget, has gradually declined over time. Moreover, when the recurrent component of the development budget is added to these regular expenditures, public savings (i.e. government revenue minus current expenditures) have been either negligible or negative. * As a result, the development budget is largely financed by foreign aid. As noted above, this should not be a reason for poor performance of public spending. However, it has led to a lack of ownership of aid-financed projects in Nepal, undermining their effectiveness (para 2.37). 2.9 The changes in key parameters shown in Table 2.1 can be largely explained by changes in the government's fiscal stance in the nineties. From FY91 to FY95, Nepal went through a period of fiscal consolidation as an integral part of an economic liberalization and reform program at that time. These reforms included Table 2.1: Government Expenditure and Its Financing, FY91-FY99 a significant (In percent GOP consolidation and Etmt u reduction in public Total Expenditure 19.6 16.9 17.8 18.9 18.8 20.2 Regular 6.3/a 6.2/a 8.8 9.1 9.5 9.2 spending by about 2% Development 13.3 10.6 9.0 9.8 9.3 11.0 of GDP between FY91 Govemment Revenue 8.9 9.8 11.2 11.1 11.1 11.7 Overall Deficit -10.7 -7.1 -6.6 -7.8 5.7 -8.6 and FY95, while strong Foreign financing 7.0 5.8 5.1 5.5 5.6 7.2 efforts to increase Grants 1.8 1.2 1.8 1.8 1.8 2.3 revenues led to a Loans 5.2 4.6 3.3 3.7 3.8 4.9 Domestic Borrowing 3.7 1.3 1.5 2.3 1.9 1.4 similar improvement Memo Items: (by over 2% of GDP) Revenue Surplus 2.6 2.6 2.4 2.0 1.6 2.5 in the revenue/GDP Foreign Aid/Dev. Exp 52.6 54.7 56.7 56.1 60.2 65.0 /a In FY95, salaries of teachers and health workers were transferred from the development ratio. Thus, the overall budget to the regular budget. fiscal deficit was Source: Ministry of Finance. reduced sharply by about four percentage points of GDP, with both foreign financing and domestic borrowing declining by the equivalent of two percentage points each of GDP. In the last four years, however, these trends have been reversed. Expansionary policies adopted by a succession of short-lived governments have increased public spending again to about 20% of GDP, while revenue performance has stagnated. Moreover, this increase in spending has been accompanied by a significant deterioration in the quality of public spending. 2.10 Rapid growth of current expenditures. Nepal's budget data is typically presented in the form of (i) regular expenditures and (ii) development expenditures. Virtually all of regular expenditures are current expenditures, while another 20% or so of the development budget is also recurrent expenditures in the form of salaries, subsidies and transfer payments to public enterprises. The increase in regular expenditures in recent years noted in Table 2.1 has had two important consequences: (i) It has preempted much of the limited growth in government revenue, leaving only a small surplus for financing development activity. (ii) Secondly, 85-90% of these regular expenditures consist of wages and salaries and debt service payments. In effect, allocations for operations and maintenance activities in the regular budget have been highly inadequate. While the development budget also provides for some 0 & M allocations, (for example for maintenance of roads and irrigation systems), these are often diverted to other purposes (for example, in the transport budget for construction of new roads). Thus many of the existing facilities are under-utilized, in part because of the lack of funds for 0 & M. The health system is a classic example. While hundreds of new rural health centers have been built, - 9 - virtually all of them suffer from chronic shortages of equipment, drugs and medicines and poor maintenance, mainly because of inadequate 0 & M funding (Chapter 3). 2.11 Table 2.2 brings out this problem clearly. When teachers' salaries (which comprise over 80% of the "grants and subsidies" in line 2 of Table 2.2) are added to salaries, allowances, medical benefits and pensions of public servants, Table 2.2: Composition of Regular Expenditures, FY96 - FY2000 (line 1 of the same table), (Rs. Million) they account for over half FY96 FY98 FY99 FY2000 (54%) of the regular budget. Budget Another 30% of regular Salary, allowances, retirement- 8,217 10,767 11,652 12,089 benefits etc. expenditures comprises Grants and subsidies/a 5,074 6,130 7,167 7,532 interest and amortization Debt Service Payments 7,225 8,170 9,109 11,526 2 Contingency expenses (unallocated) 233 391 1,152 1,471 payments on external and All others (for operatonal activities) 2,073 2,525 2,872 2,768 domestic debt. While the Total Budgeted 22,822 27,983 31,952 35,386 Actual Expenditures 21,562 27,694 31,896 n.a. budget also provides for Memo Items: "contingencies," these are Current Expenditure 26,558 33,002 38,352 n.a. Govemment Revenue 27,893 32,799 37,400 n.a. mainly for exceptional, Public Savings 1,335 -203 -952 n.a. rather than for routine Public Savings (In % of GDP) 0.5 -0.1 -0.3 n.a. operational expenses. Thus, /a Nearly 90% of this item represents grants to educational institutions at primary, secondary and tertiary levels. The bulk of the education sector expenditures in the resources remaining turn constitute teachers' salaries. within the regular budget for Source: Ministry of Finance "'all other" activities-mainly for essential operation and maintenance-are very limited. Moreover, whenever there is a shortage of funds (as often happens), it is this component of the regular budget which is tightly controlled or squeezed, (since it is not possible for the government to hold back salaries, or debt service payments), so that funding for operational activities get further reduced during the year. 2.12 This inevitably raises the question: Is Nepal's wage bill too high? (i) A quick answer is that compensation payments as a proportion of total expenditures (30% in Nepal) do not seem to be high in comparison to other South Asian Countries. (ii) Second, the average wage of government employees directly supported by the budget, (i.e. excluding employees of public enterprises) itself is quite low in Nepal. For example, the average compensation payments per government employees is only Rs. 55,000, equivalent to $800 dollar per year, compared to a per capita income of $210 in Nepal. Taking into account the average family size of about 5.5 in Nepal, this would mean that a family of a government employee who is dependent purely on his/her job-related earnings alone, would have a lower per capita income than the national average! This has important implications for both the quality and efficiency of the public service and the effectiveness of the public expenditure program in Nepal. It is an important factor contributing to poor morale, absenteeism, lack of commitment and low productivity of the public service and to searching for salary supplementation through corruption and other abuses (para 2.40). (iii) Third, there are no reliable statistics on government employment currently in Nepal to conclusively prove or refute whether there is significant "overstaffing." Counting defense and security (army and police), teachers, health workers and staff of other project implementing entities, the total number of personnel supported by the budget is estimated to range from 400,000 or more, i.e. about 4% of the total labor force. This however does not mean that there is 2 From a public finance perspective, amortization payments should not be considered a charge against current revenue. In Nepal, however, the government's budget presentation typically includes amortization payments as a part of its regular/current expenditure. The government's classification has been retained in the PER, for ease of comparison for government users. - 10- no overstaffing. To the contrary, anecdotal evidence suggests that due to political pressures to accommodate party supporters by successive governments, public employment, particularly of temporary staff, may have increased in recent years (see the recent Budget Speech). Even more important, the overstaffing issue needs to be looked at both in terms of the level of services provided by the Government and the skill mix needed to provide such services. From this perspective, overstaffing certainly exists at many levels of the government. For example, while there are shortages of staff with necessary skills at higher and intermediate levels, there are far too many unskilled persons, or with poorly matching skills, at lower levels of the government, in turn contributing to low levels of efficiency. Overall, there are too many on the payroll, both in relation to the services provided by the government and in relation to the numbers that can be compensated adequately to provide them with a degree of financial security, incentives and motivation to ensure good performance. 2.13 Is Nepal getting into a debt trap? Nepal's increasing dependence on external aid and its rising debt service payments have given rise to widespread perceptions in Nepal that the country is falling into an external debt trap. Available evidence suggests that the popular perception that Nepal is falling into an (external) debt trap is incorrect, at least at this juncture. 2.14 Nepal's total debt outstanding as a proportion of GNP has risen steadily from 15% to about 65% over the past two decades. Of the total debt stock, external debt is by far the largest component, so that the latter is now about 50% of GNP. And, service payments on external debt account for about 60% of Nepal's total debt service. Second, measured by conventional criteria, Nepal's service payments on external debt are equivalent to 7-8% of its earnings from exports of goods and services. This is a very low ratio which suggests that Nepal is unlikely to get into debt servicing problems in the near future, if it continues to maintain a prudent stance on short-term and commercial borrowing. The main reasons for this relatively low external debt service ratio are: (i) Nepal's external debt has been contracted on highly concessional terms - mainly from multilateral lenders at 3/4% - 1% interest with 30-40 year repayment periods; and (ii) its non- concessional commercial borrowing (including publicly guaranteed debt) has been negligible so far. Consequently, according to World Bank estimates, the present value of Nepal's external debt service is about 118 percent of its exports and 31 percent of its GNP. By both measures, Nepal would be classified as a less indebted developing country. Nevertheless, it should be noted that, if Nepal were to resort to significant commercial financing of infrastructure, for example, for financing telecommunications and hydropower development in the future, it will have important implications for external debt management; and that the presently favorable position noted above could change rapidly. Moreover, any significant depreciation of the Nepalese Rupee will have important budgetary implications, given the low revenue base noted earlier. 2.15 Finally a brief explanation may be in order on the stagnation of revenues, although this is not directly related to the expenditure management issue. There are many reasons for this stagnation, some of which are structural in nature: Nepal is heavily dependent on import taxation, since it has so far not been able to develop an efficient system of domestic sales, consumption and income taxes. A Value Added Tax has been introduced only recently. Moreover, trade policy reforms in the early nineties led to a significant lowering of the average incidence of import duties which, together with the recent deceleration of economic growth and imports, contributed to a weakening of import duty collections. At the same time, government commitment to find alternative ways of mobilizing revenues through tax reforms and tariff adjustments has been lacking. In the last few years, weak governments have generally steered - I1 - away from tax reforrns, while revenue leakages have increased due to a weak tax administration, political instability and corruption. D. DEFICIENCIES IN RESOURCE ALLOCATIONAND MANAGEMENT 2.16 These structural weaknesses are only one part of the explanation for Nepal's unsatisfactory fiscal performance. Even more important, there have been persistent weaknesses in the planning and budgeting system. Chief among these have been the lack of effective mechanisms to ensure realistic budgeting, prioritize the public expenditure program and screen new projects before they are included in the development budget. As a result, government agencies responsible for budget preparation have not been able to withstand both political and donor pressures, with adverse effects on resource allocation and management. 2.17 Over-programming of the Budget -- Too many projects chasing too little money. By and large, Nepal's budgeting process has been highly unrealistic in recent years. In each of the last 4-5 years, the budget targets have been set at unduly high levels, particularly for government revenue and foreign aid. This over-estimation of resources in turn has enabled the government to set similar unrealistic targets for the development budget and to accommodate too many new projects. However, the actual budget outcome has fallen significantly short of these optimistic expectations every year (Table 2.3). And, since there is little scope for cutting back regular expenditures as noted above, the brunt of the fiscal adjustment has been made through cutbacks in development spending. The implications of this process for development management are discussed below. Table 2.3: Budget Projections and Actual Outcomes, FY96-FY2000 (Percentage Increase Over Previous Year) FYN FY97 FY98 FY98 FY2000 B A a A B A a A a Total Expenditures 32.6 18.9 23.3 9.0 21.2 10.6 20.7 12.3 22.6 Regular 16.6 11.9 13.0 12.0 13.2 12.4 15.4 17.3 10.9 Development 48.9 26.3 32.6 6.0 28.7 9.1 25.7 7.5 34.6 Financing Total Revenue 28.6 13.4 21.3 9.0 18.7 8.2 20.4 13.0 19.7 Foreign Aid 49.6 27.6 36.8 4.9 38.7 10.0 36.6 13.2 46.1 Domestic Borrowing 23.8 37.5 36.4 18.2 13.3 7.7 38.5 6.7 -8.3 BzBudget; A=Actual Source: Ministry of Finance. 2.18 This accommodative approach to budgeting has enabled the Government to include in the development budget too many projects for which sufficient financing is in practice not available. Thus, several new projects have been taken on board every year; and the total number of projects in the development budget almost doubled from 423 in FY94 to 770 in FY99 (Table 2.4). As a result, total expenditures per projectlbudget line fell from Rs. 50 million to Rs. 41 million in nominal terms over the same period. Since domestic inflation averaged about 10% p.a., this has meant that expenditures per project in real terms has fallen by about 42% below FY94 levels. In - other words, in FY99 expenditures per project in real terms amounted to little more than half the level of such expenditures five'years ago. If the two largest projects/programs (i.e. the Kali Gandaki power project and the village development program) are excluded, the average allocation per project in FY99 was only about Rs. 35 million. In many cases, especially road projects, allocations have been miniscule -- a few million rupees per project -- so that they would -12- require 40-50 years to complete! In some cases, allocations have been barely sufficient to pay wages and overheads, leaving little for capital investmentlconstruction work. An inevitable corollary of this fragmentation and frittering away of resources is that few projects are completed on time, leading to substantial cost over-runs; and benefits from development spending are not being realized. Table 2.4: Development Budget, FY94 - FY2000 Development Budget (Rs. Billion) 21.2 19.8 25.0 28.9 31.5 41.9 Number of Projects 423 560 524 720 770 681 Exps. per project (Rs. Million) 50.1 35.3 47.7 40.2 40.9 61.5 Memo Item: Development BudgetIGDP(%) 10.6 9.0 10.0 9.8 9.3 11.0 Source: Ministry of Finance. 2.19 How has this happened? The over-programming of the development budget, among others, has been due to three main factors: (i) institutional weaknesses in planning and budgeting; (ii) political pressures, and (iii) easy availability of fungible cash aid. 2.20 Institutional weaknesses. Despite extensive technical assistance provided by donors over the past decade, Nepal's planning and budgeting system suffers from several weaknesses: (i) The responsibility for budget preparation is bifurcated, with the Ministry of Finance (MOF) preparing the regular budget and the National Planning Commission (NPC) taking the lead on the Development Budget. There is little integration between the two budgets. (ii) The institutional goals of the two institutions are somewhat different, creating considerable tensions in the budget preparation process. Since the NPC is responsible for development planning and the integration of such plans into the annual budget cycle, it is more predisposed to accommodating new projects and programs. MOF, on the other hand, is saddled with the responsibility of cash management and ensuring fiscal stability; and is therefore more concerned with limiting its budgetary commitments.. (iii) Although there are well-established processes for ensuring institutional co-ordination, these do not work well in practice. For example, although there is a Resource Committee for forecasting resource availability (which is then used as a basis for indicative budget allocations to line ministries), such resource estimates are overly optimistic, as noted above. The line ministries themselves do not adhere to these indicative ceilings, and pad up their budget requests to give themselves room for maneuvering in pre-budget negotiations with NPC and MOF. (iv) Although national development plans are prepared at five year intervals, the link between these plans and the annual budget is very tenuous. Only a few sectors have well- articulated sectoral plans with clearly defined priorities. Consequently, there is little forward planning of the resource needs of ongoing projects and programs, or prioritization and ranking of projects in terms of their potential for achieving sectoral and national development goals. Thus, budget allocations are generally made on an - 13 - incremental basis, largely depending on the prior year's expenditures, the negotiating strengths of various line ministries and projects, and political pressures. (v) These problems are exacerbated by the absence of a rigorous project evaluation and screening process both at the level of the NPC and line ministries. New projects are often "approved" for inclusion in the budget on the basis of scanty information, often without any serious economic evaluation. But, once these projects are included in the development budget, they are difficult to get rid of. Even if the Finance Ministry were able to withhold funding for a short period, it has been generally unable to deny funding for more than a year or two. 2.21 Political instability has had a major adverse impact on budget planning, allocation and the implementation process over the past few years. First, in Nepal, political leadership has not generally been involved in budget formulation at early stages (for example in reviewing resource estimates, providing guidance and leadership on development priorities etc). This happens somewhat perfunctorily very late in the budget cycle when the Cabinet is expected to approve the draft budget just before its presentation to the Parliament. As a result, there is little interaction on, and understanding of, the realities of the budget-resource constraints and their implications for the size of the budget and new projects, project and program priorities, and the choices and trade-offs that ought to be made at sector and project levels etc. Partly as a result, and also because of politicians' desire for new projects, there are enormous pressures to include large numbers of new projects (see below) and expand the size of the development budget every year. Second, frequent political changes have been a major impediment; and short-lived governments have been more interested in short term political and economic gains rather than with longer term development. At one level, this has exacerbated the prevailing predilections in Nepal to demonstrate political commitment to development by announcing larger and larger (development) budgets and securing more aid commitments from donors. At another level, the emphasis has been on starting new projects and getting funds released rather than on ensuring satisfactory implementation and effective delivery of outputs and outcomes. Third, every change of government has involved transfers of public servants in key administrative positions and projects on a significant scale, leading to insecurity of tenure, dependence of civil servants on politicians for job security, promotions and patronage, and increased politicization of the public service. This in turn has undermined the independence and morale of the civil service and its capacity to withstand political pressures to expand the development budget. E. THE FUNGIBILITY OF AID 2.22 While political pressures have provided the main impetus for over-programming of the development budget, the easy availabiliy of aid has facilitated this process. With foreign aid financing around two-thirds of the development budget, it has enabled the government both to divert its own resources to its preferred programs (so called "fully government-financed projects") and to undertake new projects and programs which it would otherwise not have been able to do. What is particularly significant in this context is that several of these programs are of doubtful economic justification, with low economic rates of return. Given the absence of an effective project- screening mechanism, these projects are not subject to careful scrutiny and appraisal before they are included in the development budget. While they receive the larger share of available local currency resources (para. 2.35), they are subject to less rigorous monitoring and supervision, as well as expenditure reporting and accountability standards (para. 2.34) as compared to "donor-financed projects". Thus, in many instances they contribute to poor use and waste of resources, as in the case of small road projects and local development programs. - 14 - 2.23 The experience with cash/program loans provide a good illustration of how this process works in practice. Such assistance in the form of program loans, debt relief grants and commodity assistance (e.g. for fertilizer) have been extended by well-meaning donors to help generate counterpart funds to support ongoing projects as well as to provide incentives to the Government for undertaking much-needed sectoral reforms. However, given the weaknesses in the budgeting process and the fungibility of such assistance, they have been used in practice to accommodate a large number of new initiatives. 2.24 Table 2.5 provides some interesting insights into this phenomenon. Government data in the top half of the table show that, (i) "fully government-financed (HMG) projects" account for much of the recent increase in the number of projects in the development budget. Thus, the number of fully government-financed projects rose from 143 in FY95 to 409 in FY99, while "donor-assisted projects" also rose, but less rapidly, from 280 to 361 over the same period. The table also suggests that (ii) the bulk (about 80 - 85%) of the budget allocations were earmarked for donor-assisted projects, while the amounts allocated to fully government-financed projects, despite the huge increase in the number of such projects, remained more or less constant, or even declined slightly, in nominal terms. A more detailed analysis in the second part of the Table, however, tells a somewhat different story: Over the past few years the government has received increasing amounts of cash assistance (as debt relief grants, program loans and commodity assistance) from donors; and it has primarily used such assistance for financing projects which it would otherwise have had to finance out of its own resources (i.e. "HNMG projects"). Thus, in FY99 such assistance has been used to provide part-financing for another 147 projects, (totaling over Rs.5 billion) which are then classified as "donor assisted" projects. When such projects are properly reclassified as fully government-financed projects, (iii) the number of "HMG projects" would quadruple from 143 to 556 over the FY95-FY99 period! (iv) Similarly, budget allocations for such projects would then double from Rs. 4-6 billion in mid nineties to about Rs. 11 billion in FY99. (v) Conversely, the number of genuinely donor-assisted projects would then fall to around 214 in FY99, and account for only about 70% (and not 85%) of the development budget. Table 2.5: Fragmentation of the Development Budget, FY95 - FY99 (Amount In Rupees Biliion) OFMsFM _ _i_____ ____ _ FYSO _99A 00# q<'10000 .0=ii # Anst W n # Aimount Total Projects (Budget) 423 22.90 524 28.82 568 32.58 720 34.04 770 37.73 HMG/a 143 4.20 262 6.96 297 7.08 435 6.83 409 5.58 Donor-Assisted 280 18.75 262 21.86 371 25.51 285 27.21 361 32.15 (of which Cash Aidedyb n.a n.a. n.a. n.a. (34) (1.99) (71) (4.33) (147 (5.33) Total Projects (Reclassified) 423 22.90 524 28.82 568 32.58 720 34.04 770 37.73 HMG 143 4.20 262 6.96 331 9.07 506 11.16 556 10.91 Donor-Assisted 280 18.75 262 21.86 237 23.52 214 22.88 214 26.82 /a Financed out of government's own resources. /b Projects which are fully or partly funded by Japanese debt relief grants, program loans from ADB and in-kind loans (e.g. fertilizer), but excluding other forns of aid. Source: Ministry of Finance.. F. PROJECT IMPLEMENTATION PROBLEMS 2.25 Deficiencies in budgeting and resource allocation process is only one set of factors contributing to the poor performance of public expenditures in Nepal. Persistent problems contributing to project implementation delays have been even more important in undermining the effectiveness of public spending. - 15 - 2.26 Given the land-locked and remote location and difficult terrain, project implementation is inherently more costly, risky and time-consuming in Nepal. These disadvantages apart, a series of factors have been identified as contributing to implementation delays. Over time, some of these weaknesses (for example, deficiencies in accounting and auditing, differences between government's and donors' procedures etc.) have been partially addressed. But, there has also been less progress or even deterioration in some areas (for example, excessive political interference, corruption etc.). The more important of these impediments are briefly discussed below. 2.27 Lack of counterpart funds or delays in their availability has been a perennial problem, especially for donor-assisted projects. If at all, this has become even worse in recent years, given the over-programming of the development budget and recurring revenue shortfalls (see paras 2.34 - 2.36). 2.28 Political interference. As noted, in recent years increased political interference has had a negative impact on project implementation in several ways: (i) Political involvement in procurement decisions has contributed significantly to slowing down the procurement process. Ministers are often involved in relatively small procurement decisions which should be taken at lower levels in the administration. (ii) Frequent transfers of civil servants and project officials, motivated in part by rent-seeking as well as rewarding party and personnel loyalties, have had a particularly disruptive effect on implementation. In some ministries and projects, senior civil servants and project officials have been transferred several times during the year. This in turn has led to significant implementation delays as new officials need time to familiarize themselves with project objectives and activities, donor procedures etc. And, no sooner than they become functional they have been again replaced. Moreover it has created considerable insecurity in the public service. To protect themselves in a period of frequent political change, public officials have generally pushed decision-making even on simple routine matters to higher levels. This in turn has slowed down the decision-making process at many levels. 2.29 For a variety of reasons, procurement problems (related to construction work, goods and services) have been a major bottleneck which cuts across most sectors and projects. Until recently differences between donors' procurement procedures and those of the government had been a major source of delay, since government staff implementing the projects tended to follow government procedures rather than donor procedures. Attempts to resolve this problem have been made recently by developing country-specific bidding documents. But, as discussed above lack of familiarity at the project level with the revised bidding documents and procedures has been a persistent problem. Secondly, as noted, political interference in the procurement process and slow decision-making in the government have also contributed to delays. The procurement process is seen by many as a prime example of commissioneering, involving manipulation of the bidding process and protracted negotiations. Sometimes, bid evaluation committees have been re-constituted several times; and in some cases, contracts have been awarded 11/2 years after the date of bid opening when they should have been awarded within four months. 2.30 Institutional weaknesses remain probably the most critical set of factors affecting project implementation and the effectiveness of public expenditures across sectors in Nepal. While these cover a broad range of problems, the key elements include the following: Weak public administration: Institutions at all levels -- NPC, line ministries, local governments, village committees and so on -- are weak. The requisite skills are lacking, with significant mismatches (for example, few skilled staff at higher levels and too many unskilled - 16 - employees at lower levels) in many places; while lack of motivation is a serious issue. This is mainly a reflection of the weaknesses in the public administration system itself. As noted, over-staffing, low salaries, political interference in appointments and transfers, a promotion system which in the past did not provide adequate recognition for efficiency and performance have been key contributory factors in this regard. Some efforts have been made in recent years to address these problems, but without a significant positive impact so far. * Lack of capacity to effectively monitor and supervise the public expenditure program. The monitoring of public expenditures has so far focussed largely on the tracking financial flows by the Financial Comptroller General -- the agency within the Finance Ministry responsible for cash management. As noted below, these efforts have mainly involved regulating the release of funds to ensure that money is spent for approved purposes (para 2.34). But there is no effective system to monitor physical progress and outputs, as against financial progress. Thus, even though money is spent for approved purposes, there is no reliable basis to find out what is being achieved through such spending. The NPC, together with the line ministries, is in theory responsible for monitoring implementation of projects, including physical progress; but their internal capacity in terms of staffing and resources and the attention devoted to this task have been highly inadequate. This in turn is a reflection of the lack of emphasis on monitoring project implementation and results orientation at higher levels of the government. For example, the efforts to involve the Prime Minister's Office and the Cabinet in monitoring project implementation have been at best sporadic and inconsistent. While monitoring efforts have been recently renewed at the level of the planning commission, they need to be strongly supported by higher levels. * These and other factors have generally contributed to inadequate supervision of the public expenditure program. Government staff who are responsible for overseeing project activities seldom conduct field supervision or visit project sites in rural areas on their own. This is in part due to the fact that the daily allowances provided by the Government are insufficient to cover basic accommodation and subsistence costs. The government's capacity to enhance the travel/daily allowances to realistic levels is hampered by its local currency shortages and the over-commitment of the development budget. Whenever some donor agencies provide higher payments directly to project staff, supervision is more regular. This is, however, not an option which can be pursued for all donor-financed projects. It is even more difficult to induce public servants to live and work in rural areas where the basic social and economic infrastructure is rudimentary, in comparison to urban centers. Not surprisingly, many public facilities in rural areas (for example health units) suffer from understaffing, absenteeism, lack of essential drugs and maintenance; and are dysfunctional in terms of providing basic services to the public. 2.31 Poor financial management within projects and in line ministries has consistently contributed to unsatisfactory portfolio performance. This has taken many forms: not conforming to specified budget categories in carrying out project/program activities, and spending money on unbudgeted tasks; delays in submitting expenditure statements on a timely basis; delays in providing audit reports and so on. In the case of aid-financed projects, a frequent problem has been the failure to submit reimbursement claims and audit reports on a timely basis. These problems in turn have been due to several factors, such as lack of trained/qualified accounting staff; lack of familiarity with donor procedures, in part because of frequent transfers of project staff; improper expenditures etc. Some of the consequences of these weakness are: (i) reimbursements cannot be claimed from donors even though the pre-financing provided by the government for carrying out project activities have been fully spent, tying up the government's - 17 - scarce local currency resources and slowing down project implementation for lack of funds. (ii) actual expenditures incurred in the past for many projects cannot be readily verified, so that it is difficult to both match financial and physical progress and to control cost over-runs. (iii) It also leads to significant/widespread financial irregularities. The latest Auditor General's report (for FY98) has commented that such irregularities in one form or another have cumulatively amounted to over Rs. 20 billion at the end of FY98, equivalent to 60% of development budget in that year. 2.32 Some actions have been taken recently to address these problems, including enhancing the capacity of the Auditor General's Office for carrying out regular audits, requiring project offices to accelerate accounts finalization and audit report preparation, establishing consistent budget codes and reporting formats, tightening fund release procedures by linking them to expenditure reporting by spending units etc. The Government has also agreed to implement the LACI for new World Bank financed projects appraised after July 1, 1998, linking disbursements directly with more comprehensive expenditure reporting, including indicators of physical progress of projects. While this is encouraging, there has also been significant back-tracking in some areas which has putfinancial management within the government across the board at risk. First, the Ministry of General Administration has placed all accountants of project agencies, government departments and line ministries (who were previously under the control of the Financial Comptroller General) under its own control. This change seriously weakens the capacity of the FCGO to ensure proper and timely expenditure reporting, since the latter no longer has control or authority over the accountants. Second, although the Auditor General has painstakingly put together voluminous reports on financial irregularities, little action has been taken by the government so far in following up on these issues. Indeed, the Public Accounts Committee of the Parliament which would normally review the Auditor General's report, has not even reviewed the last two Auditor General's reports -- last year because of government changes and impending elections, and this fiscal year because the Public Accounts Committee was re- constituted by the new Parliament only in late December 1999, after a six months' delay. 2.33 Dilatory government procedures contribute to avoidable delays and poor implementation quality. A good example is the government's expenditure authorization process. Although the annual budget is presented to the Parliament in early July (the fiscal year begins on July 16), it is usually approved by the government only in mid-to-late September, after prolonged debate. Thereafter, in Nepal budget implementation requires two further steps: (i) finalization of work programs consistent with the ratified budget for each spending unit by the respective line ministries and their approval by the NPC, and (ii) formal delegation (through official notification) of authority to spend from the Secretary of Finance to line ministries and project managers. These intermediate steps, which could be taken early in the budget cycle, unnecessarily hold up implementation for several weeks, so that very little work is carried out in the first trimester of the fiscal year. Similarly, given its generally tight cash position and limited capacity for expenditure monitoring, the Finance Ministry often issues instructions to spending units to "freeze" spending, a month or two before the end of the fiscal year. In effect, project implementation activities can generally take place at full capacity only during seven to eight months of the fiscal year! A second factor which affects the timeliness and quality of work particularly for multi-sector projects or those involving more than one agency, is the lack of co- ordination among government entities. A good example is urban roads in Kathmandu, where the Department of Roads (DOR) routinely carries out its maintenance work, while the Nepal Telecommunication Corporation (NTC) and Nepal Water Supply Corporation (NWSC) dig up the same roads for operational activities on their own schedules, so that urban roads are in a permanent state of disrepair, while considerable public money is wasted. - 18 - 2.34 Perhaps the most significant procedural factor affecting project implementation is the fund release process. Delays in fund releases are the most frequently cited reason for slow project implementation consistently over time and across sectors and projects. This is however more a symptom/consequence of the systemic deficiencies in the budget process-over- programming, lack of prioritization, local currency shortages and weak monitoring and expenditure control mechanisms. Given these systemic deficiencies, the Ministry of Finance (MOF) uses the fund release process (rather inefficiently) to achieve multiple objectives: to provide funding for development projects, to ration limited local currency, to minimize misuse and leakages of funds, to enforce compliance with expenditure reporting requirements and to ensure submission of reimbursement claims to donors. Over time, numerous variations of the fund release process have been tried to reconcile these conflicting objectives, with varying degrees of success. The current practice involves the initial release of the equivalent of two months' allocation to all projects and programs, and tying subsequent releases both to the amount of unutilized funds held by the spending units and the submission by them of statements of expenditures (SOEs) for previous releases. In the case of donor-assisted projects (where pre-financing is initially provided by the government to carry out planned project activities), further releases are often withheld to force project entities to submit reimbursement claims on time. This system could work reasonably well if it is applied even-handedly and linked to development priorities. But this does not happen in practice, and leads to unforeseen consequences: (i) In the absence of well-defined expenditure priorities (for example, a core program of a manageable size identifying priority projects/programs which should receive the first claim on limited local funds does not exist at present), the fund release mechanism generally benefits quick spenders, rather than development priorities. For example, infrastructure projects (which can utilize funds more quickly) generally benefit at the expense of social sector programs. (ii) Projects/programs which command greater political patronage receive more funding, as compared to donor-assisted projects. This is because the FCGO is often unable to withhold funding from politically favored programs (for example "local development") even though expenditure statements are not provided on time and the quality of expenditure reporting is very poor, whereas in the case of donor-assisted projects (where such political pressures are insignificant), the FCGO is able to effectively withhold funding. (In this context, the recent government decision to transfer the responsibility for project accountants from the FCGO to the Ministry of General Administration will further undermine the former's ability to withstand political pressures and to enforce accountability and timely expenditure reporting). This in turn creates serious misallocations of scarce local currency and sub-optimal use of such funds (see below). (iii) Given the over-programming and limited availability of local currency, most projects/programs are thus affected by funding shortages and resulting implementation delays. 2.35 Table 2.6 shows how these conflicting pressures work to distort the fund release process in practice. It shows that according to the budgets announced at the beginning of the fiscal year, fully government-financed (HMG) projects are expected to receive only a small share of local currency resources, while the major share is allocated to "donor assisted" projects. In reality, however, actual local currency availability falls far short of these expectations every year, due to the overestimation of the revenues and foreign aid noted earlier. The table also shows that, in rationing this reduced amount of cash/local currency: (i) the fund release process favors fully HMG projects, which generally receive close to, or even exceed, the initial budget allocations; while (ii) donor-assisted projects bear the brunt of the funding shortfalls. The Table also separates the truly "donor-assisted" projects from those government projects which receive part financing from program loans provided by donors as defined in para 2.24 above. When thus reclassified, Table 2.6 clearly shows that in practice much of the local currency is spent on HMG projects, and not on genuinely donor-assisted projects. - 19 - Table 2.6: Development Budget and Local Currency Utilization, FY94-FY99 FY94 FY9T FY98 FY99 8 A B A B A B A Development Budget 20.86 21.19 32.58 26.55 34.04 28.94 37.74 31.48 HMG Projects 4.16 5.74 7.08 6.28 6.83 6.85 5.59 6.16 Donor-Assisted 16.70 15.45 25.50 20.27 27.21 22.09 32.15 25.32 Sources of Local Currency 13.79 14.82 21.40 16.74 24.14 19.30 24.24 20.47 Revenue Surplus 5.20 7.17 9.23 6.19 9.07 5.10 7.52 5.25 Domestic Borrowing 1.82 2.46 3.00 5.23 3.40 7.38 4.71 7.56 Cash Aid 6.79 5.19 9.17 5.32 11.67 6.82 12.00 7.66 Uses of Local Currency (Govt. Format) 13.79 14.82 21.40 16.74 24.14 19.30 24.24 20.47 HMG Projects 4.16 5.74 7.08 6.28 6.83 6.85 5.59 6.16 Donor-Assisted 9.63 9.08 14.32 10.46 17.31 12.45 18.65 14.31 (through program aid/a) - - (1.99) (1.80) (4.33) (3.79) (5.33) (4.59) Uses of Local Currency (Reclassified) HMG Projects (including program aid) 4.16 5.74 9.07 8.08 11.16 10.64 10.92 10.75 Donor Projects (excluding program aid) 9.63 9.08 12.33 8.66 12.98 8.66 13.32 9.72 B = Budget Allocation. A = Actual Expenditure. /a Debt relief grants and program aid. Source: Ministry of Finance; FCGO. 2.36 What this shows is that in practice there are two kinds of projects/programs within the development budget: (i) those that the government itself has initiated and is owning, managing and financing; and (ii) those that the donors are largely financing through direct payments, reimbursements and project imports. The government's contribution to the latter usually consists of counterpart funds as agreed (generally varying from 10-20%), as well as pre-financing/cash advances to carry out project activities which are subsequently reimbursed by donors. The counterpart funds are usually provided, though there are frequent delays related to both the pace of implementation of projects and the fund release process as noted above. The provision of cash advances, on the other hand, is usually subject to significant delays. The net result is that actual cash releases for aided projects are somewhat less than budgeted, where as fully government-financed projects get the lion's share of cash/local currency resources. Thus, much of the cash resources go for programs which are weakly monitored and poorly accounted for, while the bulk of the donor assistance goes for programs which have weak government ownership and are generally slowed down by various implementation bottlenecks. It is worth noting that donor-assisted projects which have reasonably satisfactory implementation records are those that have either a high degree of government/agency ownership (such as telecommunications, Kali Gandaki A power generation), or turnkey infrastructure projects which some donors have built themselves (such as emergency bridge rehabilitation by Japan). The latter however is not an approach that can promote ownership and sustainability (see below). 2.37 Lack of ownership is another important factor contributed to the weak implementation and ineffectiveness of the public expenditure program in Nepal. At one level, donor-assisted programs (which constitute 60-70% of the development budget) are perceived to be "donor- driven" or "donor projects," rather than owned, managed and implemented by the government with financial assistance and technical inputs from donors. This perception is reinforced by the fact that, given the limited project preparation capacity in the government, many projects are initiated, prepared and pushed by donors; separate arrangements (for example project implementation units) are set up for their implementation, which are then monitored and supervised by donors, or sometimes the projects themselves are directly managed by the donors themselves; and financed largely by donors with little counterpart funding by the government. Although such projects invariably require the concurrence of the respective line ministries, the -20 - latter are seen as unequal partners. There are also frequent questions, among others, of inadequate consultation of local stakeholders; relevance and appropriateness; design standards; and the degree of involvement of local partners, especially consultants and contractors. While these problems in part reflect weaknesses in institutional capacity and processes for budgeting, project preparation and approval within the country, as well as the impatience/over-eagerness of donors themselves to get on with things, they have nevertheless led to weak ownership of donor- assisted projects and concommitant lack of serious commitment to their implementation. The differential treatment of donor-assisted and HMG projects in the release of scarce local currency resources is a result of this mindset. 2.38 The lack of ownership is not confined only to donor-assisted projects. At another level, it also affects all projects, including "HMG projects". This is because the government's budget process is highly centralized-centering primarily around the NPC, Finance Ministry and the line ministries-and is widely perceived to be unresponsive to local needs and initiatives. Very few projects and programs originating below are perceived to be incorporated in the budget. Similarly the budget implementation is seen as a top-down process, involving the line ministries and their district level agencies. Although significant efforts have been made in recent years to devolve expenditure programs to district and village levels, the degree of involvement of local community groups, NGOs and beneficiaries in project preparation and implementation is still small. This is a major weakness in the current system which contributes to lack of accountability, leakages and waste of resources at many levels. G. CORRUPTION 2.39 No discussion of factors contributing to the lack of effectiveness of public expenditures is complete without reference to the role of corruption. As outlined in a recent paper on Governance, there are many manifestations of this problem, including weak institutions and procedures, lack of accountability and mismanagement of resources, failure to provide effective delivery of public services including law and order, and the absence of a well-functioning judicial system. Despite frequent assertions of intent by political leaders, Nepal has yet to develop an effective anti-corruption agenda. All of these deficiencies, particularly the absence of effective institutional arrangements to enforce accountability, together with the incentives for short-lived governments to maximize personal gains as quickly as possible have created a fertile environment for corruption. Although there is little statistical evidence to conclusively prove whether corruption in Nepal has increased or not, there are enough examples to indicate the frequency of its incidence and how widespread the problem is. 2.40 Corruption takes place at many levels: petty theft and diversion of resources at lower levels, salary supplementation in a variety of ways, more significant revenue leakages, commissioneering and manipulation of the procurement process and blatant use of public enterprises for personal gain. Some of the examples include: * The poor revenue performance in recent years has been due in part to extensive leakages, for example through collusion between tax payers and tax officials in regard to tax assessments, customs valuation, gold smuggling etc. * On the expenditure side, through commissioning and manipulation of the procurement process. This is also a major inducement for starting new projects. The misuse of project vehicles for personal purposes is a widespread problem. - 21 - * Diversion of funds to unauthorized uses, and lack of proper accounting for money spent. Every year roughly 20-25% of pre-financing provided by the government to donor-assisted projects cannot be claimed, because of improper expenditures or accounting. As noted, this in turn has led to denial of additional pre-financing to projects and consequent implementation delays. * Conversely, project officials at times make double claims for the same expenditures from the donors and from the government. For example, where donors are expected to reimburse a specified percentage (say 80%) of eligible foreign expenditures, project officials have sometimes sought 100% reimbursement from some donors, while subsequently claiming the residual (20%) from the government. * Frequent transfers of staff noted earlier also have strong pecuniary motives; officials in financially advantageous positions sometimes have to buy their appointments or stave off transfers through cash payments (for example in revenue departments, public works and donor-financed projects). * Examples of questionable doals involving public procurement of works, goods and services, often involving public enterprises and sometimes even higher political levels abound. Examples include Cabinet decisions to allow the importation of duty free cars and jeeps by politicians and senior bureaucrats (involving over Rs. 200 million in lost revenues), which are then resold in the domestic market at lucrative prices; directives conferring monopoly rights to the Timber Corporation and authorizing it to sell existing timber stocks at 25 - 50% of prevailing market prices (again involving large revenue loses to the exchequer); Cabinet directives to the Food Corporation to sell large quantities of rice to the (largely privately- owned) Salt Trading Company at prices equivalent to 75% of the ongoing market price and on the basis of deferred payments etc. Similar deals have been widespread-single source procurement of fertilizer by Agriculture Input Corporation at above market prices on short notice; short-term aircraft leases by Royal Nepal Airlines Corporation allegedly involving sizeable commission payments; questionable loans by publicly owned banks to fictitious parties violating the banks' own internal procedures; letters of credit scandals; the Red Passports scandal involving Parliamentarians; and so on. 2.41 These are only some of the examples of financial abuses within the public sector, which have significant implications for both revenue mobilization and efficiency of public spending. Many of those decisions directly or indirectly affect the budget, since they lead to revenue losses, additional expenditures to cover enterprises' deficits, or unproductive use or waste of public resources. What is disconcerting is that despite the frequency of these occurrences, virtually no steps have been taken to punish or discourage such transgressions, which alone speaks volumes about the pervasiveness of the problem. For example, as noted above, the Auditor General's reports have recorded numerous such transgressions involving large sums of money. But there is no effective legal or institutional arrangements in place to correct such abuses. While the Commission for Investigation of Abuse of Authority (CIAA) has filed several cases against various individuals in recent years, there has not been even a single case of conviction against any significant offenders. This in turn erodes the credibility of the CIAA and encourages further wrongdoing as miscreants get away without any penalties whatsoever. -22 - H. PERFORMANCE OF PUBLIC ENTERPRISES 2.42 Finally, the poor performance of public enterprises is an important factor contributing to the lack of effectiveness of public spending. The public enterprise sector has been a major user of public resources; but its perfonnance both in terms of financial results and efficient delivery of goods and services has been poor. Moreover, many of the public enterprises are poorly managed; and are a prime source of leakages and corruption. 2.43 A recent analysis of the performance of 43 public enterprises (PEs) carried out by the Ministry of Finance (summarized in Table 2.7) shows that: * The majority of PEs (28 out of 43) incurred operating losses during FY98 - the latest year for which full year data are available. Generally utilities and financial enterprises made profits, while those engaged in manufacturing, trading and services made significant losses. * According to MOF data, the total operating profits of the entire PE sector (including operating subsidies provided by HMG to PEs) was only Rs. 1.3 billion in FY98. 4 out of the 43 PEs, namely Nepal Telecommunications Corporation (NTC), Nepal Electricity Authority (NEA), National Insurance Corporation (NIC) and the Rastriya Banijya Bank (RBB),3 made a combined operating profit of Rs. 3.4 billion. Thus, the other 39 PEs together made a net loss of Rs. 2.1 billion. * Operating profits of all PEs as a percentage of net capital employed (i.e. share capital, government loans plus other long term borrowing) amounted to only 1.6% in F98. Virtually all of this represented operating profits of NTC (which is a highly profitable monopoly so far), NEA, NIC and RBB. The other 39 enterprises had a combined operating loss amounting to 10. 7% of their net capital employed in FY98. * To keep these enterprises afloat, the government has had to provide operating subsidies and capital infusions in the form of additional equity and long term loans. In addition, much of the capital investment of these enterprises have been provided through external loans on-lent by the government. As of end FY98, assistance directly provided by the government as equity and loans to the PE sector (i.e. excluding government-guaranteed bank loans), was nearly Rs. 60 billion. Government assistance to the PE sector (including operating subsidies) have averaged around Rs. 2-3 billion a year in the past few years.4 However, the dividends paid by the PEs to the government in FY98 was a paltry Rs. 24 million. Moreover, most of the public enterprises (apart from NTC and NEA) have not been paying their debt service (principal and interest) obligations to the government, which has had to take over itself the responsibility of paying external lenders. This is an important element of the government's rising debt service burden noted above. 2.44 The poor financial results of PEs are in turn mirrored by their unsatisfactory physical performance in terms of delivery of goods and services and service standards/efficiency levels. Some notable examples, among others, include the following: 3 RBB reportedly made a profit of Rs. 400 million in FY98. However, it has made large cumulative losses over the present decade and its net worth is estimated to be substantially negative. Some of the reported profits may also be open to question if proper accounting and auditing standards were applied. 4 The government has also provided financial assistance to Gorkhali Rubber (a private sector recipient of government-guaranteed loans) -- Rs. 235 million in FY98 and a budget allocation of Rs. 560 million in FY2000. - 23 - * The Agricultural Inputs Corporation (AIC), which until recently had a monopoly of fertilizer import and distribution, had been consistently unable to deliver adequate fertilizer to farmers on a timely basis, contributing to low productivity and poor agricultural output growth. * Nepal Food Corporation (NFC) has been frequently unable to deliver food supplies to food deficit areas in the far west, where famine, malnutrition and lack of basic health care have led to hundreds of deaths each year. * RNAC's aircraft leasing policy (allegedly an important source of "commissioneering") has led to frequent disruptions in the scheduling and quality of its air services, in turn adversely affecting the tourism industry. * RBB's loan losses, particularly its high level of non-performing assets, has been an important factor contributing to the large spreads and the high real lending rates charged by Nepalese banks and the lack of competitiveness in the banking sector. * The perennial problem of shortages of water supply in the Kathmandu Valley has been as much the result of the Nepal Water Supply Corporation's (NWSC) failure to improve and upgrade the urban water supply distribution system, as of overall supply constraints. Table 2.7: Selected Indicators of Financial Performance of Public Enterprises, FY95- FY98 (Rs. Million) FY95 FY96 FY97' FY96 Operating Profit ILoss All PEs 1719 1345 737 1319 NTC, NEA, NIC, RBB 2303 2285 2766 3397 All Others -584 -940 -2029 -2078 Operating Profit + Net Capital Employed(%) All PEs 2.9 2.2 1.0 1.6 NTC, NEA, NIC, RBB 6.5 6.3 6.0 5.5 All Others -2.5 -3.6 -8.1 -10.7 Budgetary Support to PEs 1430 2279 1582 2827 Operating Subsidies 571 726 714 988 Share Investments 859 1553 868 1839 Memo Item: Budget SupportDev. Budget (%) 7.2 6.2 6.0 9.9 2.45 The continuing poor performance of PEs can be ascribed to a number of factors including, among others; lack of commercial orientation, poor management, inadequate autonomy, political interference, poor accounting and accountability, over-staffing, inappropriate skill mix and increasing corruption. Management of PEs has become increasingly politicized, undennining the institutions' autonomy. This has increased their vulnerability to questionable practices, which in turn has been facilitated by the lack of proper accounts, auditing and accountability. The Finance Ministry's report cited above indicates that in July 1999 only 13 out of 43 PEs had been able to complete their final audited accounts for FY98 (ending in mid-July 1998). Of the remaining 30 enterprises, 12 had finalized their audits for FY97, 8 for FY96, 5 for FY95, 4 for FY94 and 1 for FY93. Politicization has also led to pressures for new hiring; increased unionization of the labor force along party lines, in turn undermining labor discipline; and the erosion of the management's willingness to take necessary pricing decisions on a timely basis, further weakening the fragile financial position of many enterprises. -24 - 2.46 Conclusions. All these factors have contributed in varying degrees to the lack of effectiveness of public spending in Nepal. There is little doubt that institutional factors (including deficiencies in the planning, budgeting and expenditure monitoring process, as well as weaknesses in institutions, particularly the civil service/administration), have played a key role in the over- programming of the budget, its lack of focus and prioritization, and implementation problems. The eagerness of external donors to help has also encouraged over-programming and the lack of prioritization of the public expenditure program. The lack of ownership of projects/programs at various levels and the absence of accountability have also undermined the quality and effectiveness of public spending. And, increasing politicization and corruption have significantly increased the potential for leakages and wastage of resources - both domestic and external. Finally, and what is perhaps even more important, there has been a persistent lack of corrective actions in the face of numerous transgressions; such inaction and tacit acceptance of transgressions has bred an attitude- "ke garne"-which in turn involuntarily encourages and perpetuates further misuse and corruption. Chapter SECTORAL EXPENDITURE PROGRAMS AND KEY ISSUES IN ALLOCATIVE AND TECHNICAL EFFICIENCY 3 A. INTRODUCTION 3.1 Apart from systemic factors which have contributed significantly to poor expenditure performance in Nepal, the effectiveness of public spending also depends on how resources are allocated among and within sectors (inter-and intra-sectoral allocations), and how well the allocated resources are used in these sectors. This chapter begins with an overview of the inter- sectoral pattern of public resource allocations. It then briefly reviews intra-sectoral priorities and the effectiveness of use of resources within key sectors-social sectors (education, health and drinking water); decentralized local development; agriculture, irrigation and forestry; and Infrastructure (transport and power)-keeping in mind the re-orientation of the role of the public sector recommended in Chapter 4. It attempts, wherever possible, to evaluate the existing expenditure pattern from the viewpoint of whether the government is providing public goods efficiently, particularly where there are considerable externalities and market failures (for example, where the private sector is not capturing potential social benefits or where it is unable to provide services adequately at present), and whether the government's equity objectives are being met. B. INTER SECTORAL EXPENDITURE PATTERN 3.2 The broad inter sectoral pattern of government budget allocations is summarized on a functional basis in Tables 3.1 and 3.2. These tables show that: The major priorities of the government in terms of budget allocations and actual expenditures have been decentralized local development, health and power. Since the first two sectors (along with education, drinking water and "other social services") are classified as "social services," the combined share of the latter has risen from about 25% of total allocations/expenditures in FY94 to about 30% in FY99. As discussed below, successive governments have made conscious efforts to promote decentralization in recent years by increasing budget allocations for district and village level programs. This is seen in Table 3.2-development budget allocations and expenditures-where the share of local development nearly trebled, and those of health and power doubled between FY94 and FY99. . The allocations for administrative functions (general administration, pensions and social security, contingencies etc), defense and internal security, and debt servicing have remained around 30% of total allocations/expenditures. * Corresponding to the increased allocations/expenditures on local development, health and power, the share of economic services has fallen significantly over the FY94-FY99 period. This is particularly evident in the case of agriculture, irrigation and forestry whose combined share of total budget allocations/expenditures fell from 18-20% in FY94 to around 10-11 in - 26 - FY99, and of development budget allocations/expenditures from 27-30% to around 17 over the same period. Table 3.1: Sectoral Pattern of Total Budget Allocations and Expenditures, FY94-FY99 (In Percent) Budget Allocations Social Services 25.9 31.7 31.6 30.4 Education 12.8 14.1 13.1 12.5 Health 4.1 5.0 6.2 5.7 Drinking Water 3.9 3.5 3.6 3.8 Local Development 2.9 7.6 6.8 6.3 Economic Services 21.5 18.6 14.5 14.6 Agriculture 7.9 5.3 3.9 3.5 Irrigation 6.8 6.2 5.3 5.2 Forestry 2.8 2.1 1.7 1.9 Industry 1.7 1.6 1.3 0.8 Infrastructure 21.7 21.3 22.7 23.6 Transportafion 9.7 9.4 9.1 9.5 Communication 3.3 3.2 2.4 2.3 Power 8.7 8.6 11.2 11.8 Others 30.9 28.5 31.2 31.4 Defense and Securty 9.5 8.2 8.3 8.5 Debt Service 13.7 14.3 13.2 13.1 Others/a 7.7 6.0 9.7 9.8 Total 100.0 100.0 100.0 100.0 Actual Expenditures Social Services 25.2 27.9 30.9 29.8 Education 13.6 13.2 13.9 13.1 Health 3.2 3.7 5.6 5.2 Dnnking Water 3.2 2.6 3.0 2.9 Local Development 3.0 7.2 6.6 6.4 Economic Services 23.5 16.3 13.6 13.9 Agriculture 7.0 4.9 4.0 3.6 Irrigation 9.6 6.4 4.5 4.7 Forestry 2.9 1.8 1.7 1.8 Industry 2.0 1.1 1.2 0.8 Infrastructure 19.5 23.5 22.0 21.0 Transportation 10.5 13.3 10.4 8.0 Communication 2.1 3.3 3.1 1.9 Power 6.9 6.9 8.4 11.1 Others 31.8 32.3 33.5 35.3 Defense and Security 10.0 8.7 9.0 9.2 Debt Service 14.5 14.4 13.7 14.4 Others/a 7.3 9.2 10.8 11.7 Total 100.0 100.0 100.0 100.0 ia General administration, constitutional bodies, pensions and social security, contingencies etc. Source: Ministry of Finance. 3.3 As will be discussed in greater detail in Chapter 4, the recent sectoral pattern of development spending has important implications for the overall growth and poverty reduction in the Nepalese economy and for the effectiveness of the public expenditure program: The priority accorded to the social sectors-mainly health and local development-is encouraging. But, as discussed later, recent initiatives to promote local development have not been very effective so far; and there has been considerable wastage and leakages of resources. Also within the social -27 - sectors, the internal efficiency of expenditures in both education and health has been low; and much could be done to improve the effectiveness and poverty impact of these expenditures (sections C and D below). While the development of Nepal's hydropower resources is important both for promoting economic growth and exploiting the potential for longer term development of the economy, investments in power will benefit only a small proportion of the households over the next several years; and there are legitimate questions whether a large share of limited public resources should continue to be spent on power generation (para. 4.28). Despite their significant short term potential for poverty reduction through improving incomes, consumption and living standards in the rural economy (where the majority of Nepal's poor live), agricultural development (including irrigation and forestry) has received only lip service until recently. Moreover, the ineffective use of resources in this sector has been an important factor contributing to the stagnation in the rural economy and lack of progress in poverty reduction in recent years (para. 1.4). Table 3.2: Development Budget Allocations and Expenditures, FY94 - FY99 (In Percent) FY94 FY96 FY96 FY99 Budget Allocations Social Services 34.5 35.2 37.6 35.3 Education' 16.7 8.5 7.8 6.1 Health 4.1 5.5 8.0 7.2 Drinking Water 6.1 6.2 6.5 7.0 Local Development 4.4 13.3 12.4 11.6 Economic Services 32.9 28.8 22.7 22.8 Agriculture 12.2 9.0 6.9 6.2 Irrigation 10.6 10.4 9.3 9.2 Forestry 4.3 1.9 1.5 1.5 Industry 2.6 2.1 1.7 1.0 Infrastructure 31.8 34.9 38.7 41.0 Transportation 14.5 15.6 15.7 16.9 Communication 3.6 4.1 2.6 2.5 Power 13.7 15.1 20.4 21.7 Others 0.8 1.1 0.9 0.9 Total 100.0 100.0 100.0 100.0 Development Expenditures Social Services 33.5 30.5 35.7 35.0 Education/a 18.0 7.2 7.0 6.1 Health 2.6 3.7 7.2 6.6 Drinking Water 5.1 4.8 5.8 5.8 Local Development 4.8 13.4 12.7 12.8 Economic Services 36.5 26.6 22.1 23.1 Agriculture 10.9 8.9 7.4 6.8 Irrigation 15.3 11.5 8.4 9.0 Forestry 4.6 1.5 1.4 1.3 Industry 3.1 1.2 1.6 0.9 Infrastructure 28.9 41.4 39.8 39.2 Transportation 15.9 23.9 19.4 15.4 Communication 2.1 4.6 4.1 1.7 Power 10.9 12.9 16.3 22.1 Others 1.1 1.5 2.4 2.6 Total 100.0 100.0 100.0 100.0 /a The decline in education's share after FY94 is due to the transfer ot teachers' salaries from the development budget to the regular budget in FY95. Source: Ministry of Finance. -28 - C. EDUCATION 3.4 Introduction. The government in recent years has given high priority to the education sector and, within that, to providing over time universal coverage in basic and primary education. Given that very high rate of illiteracy in the country, substantially increasing adult literacy has been another declared objective. The government also committed itself in the early nineties to expand support for secondary education by absorbing in a phased manner the salaries of teachers in private secondary schools. 3.5 The Level and Composition of Education Expenditures. Reflecting these priorities the education sector's share of total government expenditures rose from 9% in the early 1980s to around 11% in the early 1990s and to 14% in the mid 1990s, though falling back slightly to 13.1% in FY99 (Table 3.3). Since the mid 1990s, the education sector's share of the regular budget (largely comprising teachers' salaries and recurrent spending for operating government and government- supported primary and secondary schools) has remained around 20% while its share of the development budget (for new, as opposed to ongoing activities) increased to a peak of 8.9% in FY97 before falling back to 6.1% in FY99. Compared to other countries in the region, Nepal allocates a slightly above average share of total government expenditures to education, and as a share of GDP (2.8%) expenditures are above those in both Bangladesh (2.3%) and Pakistan (2.2%), though lower than in India (3.9%). Despite these relatively favorable trends and comparisons, three factors need to be considered: (i) education expenditure as a share of total government expenditure is lower in Nepal than in 15 out of 17 countries in sub-Saharan Africa for which data is available for 1995, (ii) given the poor coverage of education at present and rapid population growth, the absolute allocation for education is not adequate in relation to the needs; and (iii) more resources are required initially for consolidation, quality improvements and efficiency gains. These investments, if made carefully, are likely to yield high returns, and also lay the foundation for greater resource mobilization to sustain the system in the future. Table 3.3: Share of Education in Total Government Expenditures, FY91 - FY99 1 (In Percent) Regular 4.8 6.0 20.2 20.0 21.2 20.0 Development 10.9 18.0' 7.2 8.9 7.0 6.1 Total 8.9 13.6 13.2 14.2 13.9 13.1 /a The sharp drop in development spending and the corresponding increase in regular expenditure after FY94 reflects the transfer of teachers' salaries from the development budget to the regular budget in FY95. Source: Ministry of Finance. 3.6 The composition of total education expenditures has changed significantly over the past two decades. In FY81, only 36% of total education spending was on primary education, while 38% went to tertiary and 25% to secondary levels (Table 3.4). Since then, primary education's share has risen steadily to 53% of sectoral expenditures in FY98, reflecting the high priority accorded to basic and primary education (BPE). Conversely, tertiary's share has fallen more or less steadily to 22% in FY98. Other significant changes during the 1990s include: (i) increased spending on secondary education (at the expense of tertiary) reflecting the government's commitment to "free secondary education" policy and the gradual absorption of the full cost of teachers' salaries of public - 29 - secondary schools in the government payroll'; and (ii) a conscious decision to encourage Tribhuvan University to increases its level of cost recovery and private financing. These changes have been highly appropriate. Indeed, the important question is whether they have gone far enough, given the considerable inequities and inefficiencies in the present system and the need to optimize/ensure cost effectiveness of limited public funding available to the sector (see below). Table 3.4: Composition of Education Expenditures by Level, FY81 - FY98 R ~~~~~~~~(In Pe,rcent) FY81 FY86 FY91 FY95 FY97 FY98 Primary 36.3 41.6 55.1 53.5 55.6 53.2 Non Formal 0.8 0.3 0.4 1.0 1.4 1.5 Secondary 25.2 17.7 16.0 17.1 22.4 21.89 TechNoc. 0.0 1.4 1.8 4.8 3.2 1.7 Tertiary 37.6 38.9 26.6 23.6 17.3 21.8 Total 100.0 100.0 100.0 100.0 100.0 100.0 Source Source: Ministry of Finance 3.7 The pattern of development expenditures (which roughly accounts for about a third of the total education sector expenditures) is summarized in Table 3.5. The table shows expenditure allocations, since actual expenditure data on a project by project basis and their actual financing pattem are not available. The Table nevertheless indicates the current priorities in the development budget: - primary education receives the largest share-on average, about 50%-of the development budget. - the share of secondary education is rising, but remains relatively small at under 10%. - the tertiary sub-sector continues to receive over a quarter of the development budget (through the University Grants Commission). * the share of technical education has been falling rapidly, as projects are gradually scaled down. Table 3.5: Composition of Development Expenditures, FY96 - FY99 (In percent of development expenditure) FY90 FY97 FY98 FY99 Basic and Primary 56.7 43.5 51.4 35.9" Secondary 6.6 7.3 7.1 8.5 Higher/Tertiary 22.3 22.5 31.2 29.0 VocationallTechnical 11.6 8.1 6.7 4.7 Others 2.9 18.6 3.6 21.8 Total 100.0 100.0 100.0 100.0 /a The sharp decline in FY99 reflects completion of Basic and Primary Education I project. Source: Ministry of Finance. The practice in Nepal is that communities start schools (called proposed schools) and once they meet the criteria (mainly number of students), the Government absorbs the salary cost. In the case of secondary education, the government absorbed the full salary cost of private schools that were receiving partial salary support where the posts were approved by the Government. -30 - 3.8 Appropriateness and efficiency of public spending. The actual achievements (literacy rate, participation and cycle completion rate) have always fallen short of targets. However, more inputs (number of teachers and schools/institutions) have been provided than targeted. This is an indication of unplanned growth and lack of strategic focus. The policy of wider coverage has been achieved at the cost of quality, and this strategy has begun to take its toll. Appalling rates of internal inefficiency combined with rapid growth in private schooling, and huge disparity between examination pass rates in public and private schools are indications of erosion in the quality of public schools. This calls for rethinking of the ways in which public education is organized and delivered and the urgent need to focus on consolidation. 3.9 Issues in Technical Efficiency. Although the level of public education spending is commensurate with that across the region as a whole, there are important questions about the coverage quality and effectiveness of education spending in Nepal. These have important implications for both the present pattern of resource allocation and future policies. The key issues are discussed below. 3.10 Starting from a situation in the fifties when there was virtually no schooling except for privileged few, Nepal has made considerable progress in the past 3-4 decades in terms of school enrollments, literacy levels, access to primary education, geographical coverage of schooling etc. Notwithstanding these achievements, the current status of educational achievements is unsatisfactory in many respects: 3.11 Internal efficiency is low at virtually all levels of the education system, as evidenced by several key indicators: * Despite the high priority accorded to education, the overall literacy rate is only 36% (andfor women only 19%) in Nepal, compared to 90% in Sri Lanka, 52% in India and 39% in Bangladesh. Only Pakistan has a lower rate (35%). * At the primary level, although the gross enrollment ratio is officially reported at 117, the net enrollment rate is only 69 (57% for girls). At the lower secondary level (grades 6-8), gross enrollment ratio is 54% (44% for girls),and at the secondary level (grades 9 and 10) around 31% (only 26% for girls). Moreover there are large variations between urban and rural areas and between geographical regions with regard to literacy achievements, enrollment ratios and gender balance at all levels of the education system. * Dropout and repetition rates are high. At the primary level, less than 18% of those who enroll complete the primary cycle on time. Only just over two out of every five Nepali children eventually complete primary schooling and those who do complete take an average of seven years to cover the five years cycle. At the secondary level, out of every 100 students entering the lower secondary, 25% drop out by the end of the second year; and only 40% complete the five year cycle. * The quality of education particularly in government schools is low, and pass rates are low throughout the education system. At the primary level, achievement scores in public schools (in grade HI in 1997) averaged between 44 - 50% compared with an average of over 75% for private schools. At the secondary level, pass rates for the School Leaving Certificate (SLC) examination taken at the end of grade 10 have ranged between only 25 - 35%; while at the - 31 - tertiary level pass rates in the Proficiency Level Certificate (taken after grade 12) averaged 24% in 1997 and 39% in 1998. The SLC pass rates in private schools have been over 70% with top scorers coming mostly from these schools. Pass rates at degree level vary according to subjects, ranging from only 17% to 25% in 1998 for students in Education, Management and Humanities. Such low pass rates are indicative of an inefficient and low quality system in crisis. * The low levels of achievement in turn reflect a variety of problems, including among others, large numbers of untrained teachers (roughly 60% are untrained at the primary level and about 50% at the secondary level), teachers' absenteeism, inadequate female teachers, overcrowding of classrooms particularly in the Terai, insufficient teaching/learning materials (in part reflecting the lack of 0 & M noted earlier), and inadequate supervision. Teacher- student ratios are around 46:1 at the lower secondary level, and 25:1 at the secondary level, while only one in eight teachers is female. 3.12 Equity Effects. From a social equity perspective, the current distribution of education expenditures does not benefit the poor/lower income groups as much as it potentially could. There are two main reasons for this: (i) a significant part of the benefits from the current pattern of education expenditures accrue to the upper income groups (i.e. the non-poor); and (ii) the poor are primarily dependent on the public education system; and the low quality of the latter (as compared with private education) effectively penalizes the poor (see below). While there are many reasons for the relatively low participation of the poor in education in Nepal as in some other countries, these need to be effectively addressed (para. 3.22). 3.13 The recent Nepal Living Standards Survey (NLSS) provides some interesting insights on who benefits from the current education expenditure pattern. Table 3.6 below (which is derived from NLSS data) shows that the share of all children attending public primary schools is broadly similar for all household income quintiles, though it is slightly lower for the bottom 20%. Thus, of every Rs. 100 spent by the government on primary education, Rs. 17 is spent on children from the poorest 20% of households. And, since consumption expenditures in Nepal are generally co- related with income levels across households, similar shares of expenditures on primary schooling accruing to each income group helps reduce the inequalities in consumption. This is not the case with higher education. According to NLSS, only 6% of the Table 3.6: Participation in Public Schools by Household students come from the poorest 60% Income Groups of households and only half of 1% (In Percent) from the poorest 20% of households. ncomn Quftle Prinary Secondary ertiary The representation of females from Bottom 20% 17 5 0.4 the poorest 40% of households is 21 -40% 22 16 5 negligible. Publicly provided higher 61 - 80% 23 24 13 education in Nepal is accessed 81-100% 19 44 81 almost entirely by children from Source: Nepal Living Standard Survey, 1996. higher income families. 3.14 Overall, although education expenditures accruing to the lower income groups are a higher share of their total expenditure than is the case for better off groups, in absolute terms they are lower. Thus, the current expenditure pattern as it is allocated across the various levels of education cannot be justified on the grounds that it increases equity. From an equity perspective, - 32 - more public spending on the lower levels of education, particularly primary, will have a greater equity impact. In contrast, expenditures on higher secondary and tertiary education accrue predominantly to the upper income groups and have regressive effects on income/consumption distribution. These expenditures, apart from the relatively small share which accrue to the poor, cannot be justified on purely equity grounds. 3.15 Private schooling is an increasingly important alternative to public schooling at higher levels of education, particularly for the upper income groups. Thus, at lower secondary levels, enrollment in private schools rises to 20%, (i.e. publicly funded schools 80%); while at higher secondary levels the ratio rises to 28% (i.e. public 72%),. However, given the higher costs of private schooling, the poorer households cannot afford it; and have to rely on public schooling. And since the quality and educational achievement standards differ significantly between public and private schools, children from lower income groups have substantially lower probabilities of getting a quality education as they move up the ladder. 3.16 Cost Effectiveness. An analysis of the cost effectiveness of various levels of the public education system brings out some interesting issues relating to the equity and sustainability of the present education expenditure pattern. Unit costs have been rising at all levels through the nineties, reflecting both increases in real teacher salaries and improvement in student: teachers ratios. In FY97, average cost of education per student at primary level stood at Rs. 1270, a real increase of 65% over FY91 level; and at the secondary level Rs. 1644, representing a 54% real increase over FY91. In tertiary education, unit costs vary significantly by field of study, ranging from Rs. 1390 per student in Management and Rs. 2124 in Law to Rs. 38,700 in Agriculture and Rs. 46,000 in Medicine. In FY96, the average cost for all students at the tertiary level was Rs. 13,329. However, if the costs of the four professional institutes (agriculture, engineering, medicine and veterinary sciences) which enroll just 4.6% of tertiary students are excluded, the average expenditure per tertiary student was Rs. 7420. On that basis, unit cost ratios for primary, secondary and tertiary education were roughly 1.0: 1.3: 5.8. 3.17 The unit costs at primary and secondary levels, and indeed in some fields of study at tertiary levels, can be compared. While unit costs at the tertiary level can reasonably be expected to be somewhat higher, in Nepal, the costs of tertiary education are driven up by the fact that the latter includes grades 11 and 12, which are normally a part of higher secondary education in other countries. (Thus, the tertiary level in Nepal includes two years of upper secondary grades 11 and 12, a two-year bachelor's degree and post-graduate studies). In FY96, out of a total enrolment of 140,000 students at the tertiary level, only 44,500 were enrolled in degree courses, i.e. the large majority were grades 11 and 12 students. This preponderance of upper secondary students, combined with the low pass rates and high repetition rates at the tertiary level noted earlier, imposes a heavyfinancial burden, driving up unit costs throughout the tertiary level. This is an expensive way to teach grades 11 and 12 which needs to be redressed. Moreover, there is very little cost-recovery at the tertiary level. As a result of tertiary education reforms and associated increases in fees initiated in the early nineties, internally generated revenue as a share of University expenditure rose to about 20% in 1993 (from only 9% in 1985) but has again fallen back to 11-12%. Given the fact that the large majority of tertiary student are from upper income families and that the benefits of tertiary education accrue primarily to the participating individuals, (unlike in the case of primary and even secondary education where there are significant social benefits), there is little justification for this situation. - 33 - 3.18 It is important to interpret the unit costs cautiously. Circumstances in the Technical Institutes and much of the rest of the University differ considerably. The Technical Institutes (engineering, medicine, agriculture and forestry, and animal sciences) have much higher unit costs than the other faculties but they are also able to enroll much better qualified students and the examination results are much better. Conversely, when the unit costs of the Management Faculty (which are around NRs.1,500) are reviewed, it is unlikely that the quality of education meets the standard of higher education. When much of the education at the University is an extension of secondary education, little provision is made for research/project work and.the environment is far from what one would expect in a typical higher education institution, it is not surprising that the unit costs in the non-technical faculties are low. However, the unit costs will rise sharply once PCL programs are phased-out of the university system and the bachelor and masters programs are upgraded to improve the overall quality of higher education. This will lead to increased unit costs of higher education. When this occurs, additional resources will need to be mobilized from students and from other means while freezing or reducing the Government funding. 3.19 The free secondary education policy has led to greater inequity in the system because the schools that were already starved for funds for quality improvement prior to the introduction of this policy are left with even fewer resources. This is because, under the free education policy, the Government bears only the teachers' salaries and very little of other important non-salary components. The schools that were able to raise funds from student fees prior to this policy can only raise limited amounts which are grossly inadequate to meet the quality improvement needs. Against this background, it is not surprising that many students are migrating to private schools. 3.20 Table 3.7 provides some useful information relating to this issue. The table is an aggregation of estimated expenditures (drawing on budgetary data, NLSS and information on aid flows) on education during FY95 by the government, external donors and Nepali households. It shows that (i) the government and donors together provided roughly 59% of the total education expenditures, while households contributed about 41% of the expenditures; (ii) by level, 45% of the total expenditures were on primary and non-formal, 32% on secondary and 16% on tertiary education; (iii) By and large, primary education was financed by the government and donors (64%) with households financing the remaining 36%; while the situation was reversed at the secondary level, with households financing 72% of expenditures and the government 27%; (iv) Out of the households' expenditures, roughly two thirds of the expenditures at secondary level (or 14 percentage points out of 22.8) Table 3.7 National Education Accounts by Sources and uses of Funds, FY95 were spent on (In Percent) private schooling; Use of Funds f7urGes Total while the balance as Goverment Donorsia Households well as the bulk of Primary 21.6 5.9 15.7 43.2 the expenditures at Non-Formal 0.3 1.8 .. 2.1 primar leve were Secondary 8.4 0.3 22.8 31.6 primary level were Tech.Noc 0.4 5.6 .2 6.0 spent on covering Tertiary 7.0 6.4 2.9 16.3 the costs of children Other 0.5 0.3 .. 0.8 Total 38.2 20.4 41.4 100.0 attendin1g pubic .. Negligible schools: mainly /a Most of the donor expenditure in the education sector is channeled through the budget, registration fees, through there are also direct providers. text books, clothes, Source: Hotchkiss et.aI. 1997. -34 - school supplies and private tuition; and (v) In contrast, the costs of tertiary education were borne predominantly by the government and the donors (82%), with households contributing a meager 18% of total expenditures. Looked at another way, for every Rupee the Government and the donors spent on primary, secondary and tertiary education, households spent Rs. 0.57, 2.62, and 0.22 respectively. 3.21 Conclusions and Key Issues. The foregoing analysis suggests that, although the present system emphasizes priority to basic and primary education and a considerable share of education expenditures is allocated and spent on the primary level, there are major inequities and inefficiencies in the current pattern of education sector expenditures. The quality of education in public schools is consistently low at all levels and access for the poor is limited at the secondary (and tertiary) levels. The poor are thus disadvantaged both by the low quality of education and limited access, which are by themselves important deterrents, among other factors, contributing to low participation by the lower income groups. While the share of education expenditure on tertiary education has fallen over time it is still substantial and remains iniquitous, with almost all of the benefits accruing to the higher income groups and low levels of cost recovery. 3.22 Significant improvements in equity and cost-effectiveness can be achieved through reforms in the current structure and pattern of education expenditure. This is very important because, as discussed in Chapter 4, public resources for the education sector (as well as to other sectors) are likely to be limited in the foreseeable future. To achieve greater allocative and technical efficiency in education is therefore both desirable and necessary. The key reforms include: (i) Improving the quality of education at both primary and secondary levels-reducing dropouts and repetition rates, improving achievement levels/pass rates and thereby shortening the abnormally long time taken at present to complete the primary and secondary schooling cycles. In addition, efforts are needed to improve the current incentive structure on both the demand and supply sides. For example, enhancing the quality of teacher training and supervision, depoliticization, and increased involvement of beneficiary/parents' associations and local governments (as discussed below) are important for improving the quality of education. While progress in quality improvements itself is an important incentive, other steps (such as energetic public campaigns to encourage the poor to send their children to school and measures to enhance the enrollment of girls etc.) should also be considered. These in turn will require significant improvements in key inputs and funding. (ii) Expansion of coverage is necessary especially at the secondary level, particularly as efforts to improve quality and retention rates at primary levels bear fruit, in order to absorb the increasing numbers who are expected to graduate from the primary level. As noted, the access of the poor to public schooling at secondary level is particularly limited. However, these interventions will need to be carefully designed so as to not to undermine or replace private initiatives which currently finance secondary education and are expected to play an even more important role in providing secondary education in future. (iii) Major reforms are needed at both upper secondary and tertiary levels in order to reduce government funding at these levels and make them more equitable and cost-effective. - 35 - These include (a) the separation of grades 11 and 12 from the tertiary level and their accommodation in the upper secondary program; and (b) sharply curtailing government subsidies presently provided to the higher income groups at the tertiary level. Although the relocation of grades 11 and 12 will initially entail additional expenditures on upper secondary education, over the medium term they are likely to generate substantial cost savings at the tertiary level. Similarly, there is little justification to provide large subsidies to upper income groups at the tertiary level; and these should be redirected to accommodate poor students (for example, through scholarships and student loans), while raising the fees at the university levels to increase cost recovery. (iv) Seriously explore the quantity vs. quality tradeoff. The education system will not become sustainable unless quality improves. Also, free education is not affordable at all levels. Primary education should be free as far as possible, but cost sharing should increase at higher levels. Cost-sharing particularly at post elementary levels should be pursued with a strong mechanism to target support to students from poor families. Such targeting is feasible through decentralized management based on clearly defined criteria and procedures. (v) Use education to lead the decentralization reforn. It is important to increase the accountability of the education system, depoliticize the system which is heavily politicized at present, and develop criteria and procedures to provide block grants to schools, including supplementary funds based on performance. Such reforms can provide an interesting example of community involvement in the management of local level programs, and spur similar initiatives in other areas. 3.23 It is important to explore the possibilities for strengthening the role of the private sector in providing educational facilities and improving the quality of education at all levels. As noted, there is considerable scope for increasing private involvement at both secondary and tertiary levels. The present system of financing teacher salaries at the secondary level appears to be the worst of both worlds-it adds to public costs, without providing any mechanism for ensuring the effectiveness of use. Alternative mechanisms of public funding of education to ensure greater effectiveness (for example, paying a flat rate for every student enrolled together with a premium based on performance at tests, among others) need to be explored instead of direct public provision. D. HEALTH 3.24 Overview. In recent years the share of total government expenditure devoted to health services has increased, but per capita allocations remain among the lowest in the world. Both the quality and coverage of services are low, and indicators of health status are unsatisfactory, even for the region. Public resources from both government and donors, while increasing, are likely to remain low relative to the huge needs until the economy records much higher, and sustained, rates of economic growth. In these circumstances, the overall challenges facing the government is to ensure that the relatively small amount of public resources is distributed across health service activities in a way which maximizes the government's equity objective and its responsibility to fund and often to provide those critical services which others cannot or will not. Within this overall objective there is a need to: (a) ensure adequate programs focused on infectious diseases (including HIV/AIDS), maternal and perinatal problems, and micro-nutritional disorders (which -36 - currently account for 70 percent of the burden of disease and are highly unlikely to be adequately funded or provided by the private sector without high or total subsidies); (b) ensure that the poor are not denied access to essential services because of their poverty or the absence of the required facility; (c) encourage private expenditures for publicly provided services from those who are able to pay; (d) develop an environment which encourages the provision of health services by the private and non government sectors; (e) allocate expenditures across activities in a way which accurately reflects the government's stated objectives and (f) ensure that the publicly provided inputs are used more efficiently. 3.25 The Level and Composition of Health Sector Public Expenditures. The health sector receives a very significant share of total public sector resources-apart from education, power, local development and transport it receives the largest share of the national budget. Further, a considerable proportion of estimated public sector health resources are not reflected in the health sector budget (or in some cases in the national budget at all). For example, much of the activities of external donors and INGOs have not been reported until recently in government records. Further, spending by Ministry of Education (on medical education and related programs), Finance (on medical assistance to public sector employees) and Defense (on medical care for its employees) are not included in the health budget data. The National Health Accounts for Nepal (Hotchkiss et.al. 1999) suggest that in FY95 the share of the public sector in overall health expenditures was relatively small. It is estimated that 76% of all expenditures for providing or accessing healthcare were made by households, 13% by donors (both directly and through the government) and 11% by the government through internally generated resources. However, two- thirds of these household expenditures are estimated to have been associated with attending a government facility, while only 25% covered the costs of services provided by private facilities and another 10% services provided by NGOs and by donors directly. Thus, despite the government's relatively small direct share in health expenditures, the public sector remains by far the major provider of health services in Nepal. For the poor and especially those living in relatively remote areas (given the focus of both private and NGO service provision in the Kathmandu valley), the government is usually the only modern health service provider. 3.26 Total public expenditures on health have increased steadily, though from a low base, over the present decade. Health sector expenditures have risen on average by about 14% p.a. in real terms between FY91 and FY98; and their share of total government expenditures have risen from 2.8% to 5.6% over the same period. Nevertheless, public expenditures on health, in per capita terms is very low in Nepal - after adjusting for under-reporting, $3.10 per head in FY98. Only a very few developing countries, virtually all of them in Sub-Saharan Africa, spend less on health in per-capita terms than Nepal. This is about one-fourth the level of resources estimated by the World Bank in 1993 to be needed to provide a package of essential health services in a developing country. This inadequate level of spending is compounded by a range of other factors highlighted in this review. Further compounding this, a very high population growth rate averaging 2.5% p.a. - at this rate, Nepal's population would double in about 29 years - has rapidly increased the demand for public health services, while the institutional capacity to plan and implement health sector strategies and programs remains weak. 3.27 Allocative Efficiency of Public Spending in Health. Estimates of the burden of disease (measured through Disability Adjusted Life Years-DALYs). made by the World Bank in a recent study on Nepal "(Operational Issues and Prioritization of Resources in the Health Sector, - 37 - 1999)" provide a useful basis for analyzing the allocative efficiency of overall health expenditures in Nepal. The study shows that: * Of the total burden of disease, a very high proportion (69%) is due to Group 1 disorders (infectious diseases, maternal and perinatal ailments and nutritional deficiencies), while Group II (degenerative and non-communicable) diseases account for 23% and Group III (injuries and accidents) another 9%. * More than half the disease burden in Nepal (51%) is due to premature mortality of children under five, the main causes being diarrhea, measles respiratory infections and perinatal conditions; while another 25% reflects deaths of 15-44 year olds. * In both age groups, females bear a higher burden of disease indicating both discrimination against females in health services and perhaps in the distribution of food and high levels of maternal morbidity and mortality. * What is also striking is that Nepal's incidence of the disease burden from Group I conditions is five times as large as China's and 50% more than in neighboring India. The latter moreover may soon witness an epidemiological transition, as Group II (non-communicable) diseases may overtake Group I conditions as the primary cause of the disease burden. Projections for Nepal, on the other hand, show that this transitional stage may be delayed for several years, with the Group II share of the total burden rising from 23% in 1996 to only 29% by FY2011, i.e. indicating only a marginal decline in Group 1 diseases. 3.28 The level and composition of the burden of disease estimates for Nepal clearly demonstrate health outcomes are far less than desirable and reflect a country still in a "pre epidemiological transition" where infectious and nutrition related diseases and maternal and perinatal health problems (Group I disorders) significantly predominate over degenerative and non-communicable diseases (Group II disorders) and injuries and accidents (Group III disorders). 3.29 In general, burden of diseases estimates also give a good overall approximation of the total (public and private) priorities for total spending on health. They do not, however, necessarily reflect the priority for public (including donor) expenditure priorities. Public expenditures (including on health) should pass both economic efficiency and equity tests. Public resources (including subsidies) are justified on efficiency grounds when private markets are prone to failures which prevent socially optimal levels of services being provided. This happens when there are large externalities, public goods, non-competitive markets and an absence of well-informed consumers or producers. Subsidies can also be justified on equity grounds when reliance on private markets results in inadequate provision for and consumption by the poor. 3.30 Across the full range of health services, private markets fail to work to different degrees. Disease control measures such as spraying are generally regarded as being close to pure public goods, and are unlikely to be provided by the private sector. Infectious disease control and certain immunization programs result in large externalities. In general, preventative care is often undervalued by individuals because of inadequate information. In all such cases, provision determined solely by private valuations would almost certainly result in levels lower than are socially desirable. Curative health services are not public goods, and apart from for infectious diseases, are less likely to result in externalities. A functioning market for such services is more likely. Even in this case, however, the imperfect information of consumers (patients) often resulting in the need to trust the judgment of the sellers (health service providers) often fails to ensure appropriate care at the lowest prices. In addition, the high cost of treating particular illnesses -38 - together with a lack of insurance markets results in a fear of exposure to great financial risk leading to large welfare losses. 3.31 In Nepal, the conditions which contribute most to the burden of disease require interventions (e.g., Expanded Program of Immunization (EPI), safe motherhood, micro-nutritional, tuberculosis, and leprosy programs) which are broadly the same as those which can be justified by the tests of efficiency and equity. Thus, such programs should have been the prime concern of public sector investments. However, Nepal's public expenditure composition in the health sector appears to have been increasingly skewed the wrong way over much of the past decade. Indeed, Table 3.8 (a) shows that at the beginning of the Table 3.8: Allocation of Total Health Expenditures by Major nineties, over three fourths of Programs, FY92 - FY98 the budget allocations in (In Percent) health were for primary care, F ... FY5 F.. .... Primary Care/a 76.8 63.1 57.2 earmarked for hospitals Service Delivery 34.8 29.2 24.9 (central, regional and zonal), Support Services/b 6.3 8.4 7.6 and 6% and 3% respectively National Health Programs 35.7 25.5 24.7 Health Policy & Managementic 5.7 3.8 2.5 for health policy and Hospitals/d 14.6 30.3 37.5 management, and traditional Traditional Medicine/e 2.9 2.9 2.8 medicine. However, over the Total 100.0 100.0 100.0 rest of the current decade, /a. Primary care service delivery includes all expenditures (regular & development) up to district rest of the current decade, ~~level. most of the increase in health /b. Support services include training, IEC, MIS, laboratory services, epidemiology etc. sector nding appears to {c. Health policy and management includes MOH, DHS, Regional Health Directorates. sector fun ing appears o /d. Central hospitals: Speciality institutions, central, regional and zonal hospitals. have been channeled away /e. Traditonal mediane inrdude Departments of Ayurved, Unani, Naturopathy, clinics and from primary care and related hospitals run by traditional systems. activities to hospitals. This Source: Budget documents for various years, Ministry of Finance. has led to a sharp reduction in the share of budget allocations for primary care, which fell to only 57% of the total in FY98, while the share of hospitals has risen sharply to 38%. Within primary care, funding for various national health programs and for service delivery has been seriously constrained, with their relative shares falling by 30% each. Although traditional medicine has roughly maintained its share, a major casualty of this allocation pattern has also been health policy and management, whose share fell by nearly 60%. The latter is particularly significant in view of the weak institutional and management capacity in the health sector noted below. 3.32 A review of the development budget between FY96 and FY98 (Table 3.9) brings out this pattern starkly. The Table 3.9 Allocations in the Development Budget, FY96 - FY98 development budget increased (Rs. Million) rapidly over these three years, F7Y977 6 % FY97 w FY mainly due to an expansion in Hospitals 290.0 767.8 1,225.8 donor support which, in FY98, Health Services 585.2 968.4 1,433.1 was responsible for two thirds Central 346.7 728.4 1,112.1 District 238.5 240.0 321.0 of the total. The main area of Other 37.1 50.8 52.2 expansion has been support Total Development 915.5 1,787.0 2,711.2 for hospitals, which received Central (29) 670.1 1,538.0 2,380.1 45% of the development District (18) 245.4 249.0 331.0 - Note: Numbers in parenthesis are numbers of individual projects. Programs are also budget in that year. Of the divided into central and district according to who administers them. eight major hospitals funded Source: Ministry of Finance. - 39 - within the development budget, three received large amounts of donor funding. In addition, one of the hospitals directly financed by Government received almost 30 percent of the total development budget. The implications of this upsurge in support for hospitals are not limited to development expenditures. They are likely to lead to increased future demands for recurrent funding from the regular budget, when their development phase is completed. 3.33 The Health Services component of the development budget currently includes 47 separate projects. Of the 29 centrally run projects, 19 are supported by donors and one of these-the IDA-supported Population and Family Health Project (PFHP)-absorbs almost 40 percent of the total. (Indeed, the significant increase in the health service component in absolute terms during FY97 and FY98 was largely due to the implementation of this project. If not for this, the share of health services and indeed of primary care shown in Table 3.8 would have fallen even more sharply). Within health services, total expenditures on disease control projects in FY98 were only Rs. 367 million, just 13.5 percent of the health sector development budget. Apart from the PFHP, the largest single allocation within the development budget is for "medicine and instruments supply". However, as a proportion of development expenditures, it amounted to less than 5 percent in FY98; and this proportion has fallen steadily during the decade. 3.34 The rapid growth of the development budget in health in recent years needs some explanation. In part this reflects increased emphasis by the government on health. It also reflects several other factors: (i) general tendency to over-program the budget discussed in Chapter 2 above; (ii) efforts by the Government to correct the extensive under-reporting of donor funding in the health sector; (iii) given the local currency constraints discussed earlier, predilections on the part of the government to limit the growth of the regular budget and to include many recurrent activities in the development budget, in order to attract donor financing; and (iv) increased involvement of donors themselves in development activities in the health sector. As indicated in Table 3.10, regular expenditures comprise only 30% of health sector allocations, which are normally fully borne by the government; Table 3.10: Allocations in and Sources of Financing for the Health Budget, FY98 while the (Rs. Million) development budget Total HMG Donors Percent Share (%) constitutes about _ 70% of total Hospitals 1446 606 840 41.9 58.1 Regular 221 221 - 100.0 - spending, mainly Development 1225 385 840 31.5 68.5 supported by donors Health Services 2280 1321 958 58.0 42.0 Regular 847 847 - 100.0 - (to the extent of Development 1433 474 958 33.1 66.9 two-thirds). Indeed, Total Budget 3850 2022 1828 52.5 47.5 donor support for Regular 1138 1138 - 100.0 - Development 2711 883 1828 32.5 67.5 the development Source: Ministry of Finance. budget has risen rapidly in recent years. Donors also appear to have supported in equal measure both hospitals and health services, though in reality much of the support for the latter reflects the PFBP, as noted above. Since the share of hospitals in the development budget has risen rapidly, this also suggests that the government has substituted its own funding (released by increased donor support for health services) for supporting hospitals. -40 - 3.35 Clearly, the change in emphasis - particularly in the mid 1990's - towards urban hospitals and away from rural facilities and public health programs is cause for concern and runs contrary to the pattern of allocation which would be implied by the general considerations of the justification of public expenditure. However, publically funded hospitals often need to play a role in providing facilities beyond primary care for the rural poor, and also for the Nepali middle class in cases of rare but expensive clinical care. Emergency obstetric care, a core component of the safe motherhood agenda, is a case in point. This clearly requires strategically located facilities across the country able to perform this function. It also requires a fully functioning primary health care system with an effective referral system in place - still a major challenge for Nepal. More generally, public funding of hospitals can be justified on the grounds of "catastrophic risk" for everyone (at least where private insurance markets are not operational) and on equity grounds for the poor. High levels of cost recovery from those who can afford it should (as discussed below) be a core feature of hospital funding arrangements. 3.36 Technical Efficiency. Apart from the pattern of budget allocations for health, the effectiveness with which they are used is also relevant. One indicator is particularly troubling. Over the past decade a significant proportion of health budget allocations (particularly from the development budget) have not been spent. Since FY91, actual expenditures as a proportion of allocations in the development budget have averaged only 68 percent. While the proportion has been increasing in recent years, it remains lower than in other ministries, indicating major inefficiencies in the sector. 3.37 A major factor hampering the effective use of allocated resources in the sector is the recurring shortages of basic inputs and human resources which are needed to operate facilities constructed, especially in rural areas. The Government has made considerable efforts in the past few years to expand the coverage of rural health facilities. For example, sub health posts increased from 200 in 1992 to 2597 in 1996. However, a Health Facilities Infrastructure Status survey undertaken in 1966 for the World Bank-led health sector work found that none of the health posts in the Far West Mountain region and only a quarter of those in the Far West Hills has water supply. Lack of electricity, toilets and staff quarters were common, while basic equipment such as thermometers were absent in many facilities. Budget allocations for medical instruments and medicines, as noted, have decreased significantly as shares of total expenditure. The survey also indicated that while vaccines and drugs for malaria, leprosy and tuberculosis were available in sufficient quantities, many other essential drugs such as antibiotics were not available for up to six months of the year. Shortages were particularly severe in the primary health centers. Staff vacancies and absenteeism are acute in many rural health facilities. The Long Term National Health Plan (1998) reported that a third or more of the sanctioned posts of health assistants, auxiliary nurses, midwives and village health workers were vacant; and that over a quarter of appointed staff were absent. Reasons for under-staffing include under-funding for posts and training, reluctance of individuals to fill positions in remote areas, and management weaknesses in staff deployment and supervision. Increased demand for staff from new and expanded hospitals may also help explain staff shortages in rural areas. In all these cases of inadequate inputs and staff, there are major regional differences particularly between the Central region and the Terai, and between the former and the Mid and far West regions. Unfortunately, it is in these latter regions that non-government services are least available and where the dependence on the public sector is greatest. -41 - 3.38 Most of these problems, as discussed in Chapter 2, are systemic, common to virtually all the sectors. The key question therefore is why is absorption capacity particularly weak in the health sector? While it is difficult to provide a conclusive answer to this question, there are some important relevant factors. First, institutional and management capacity has been traditionally weak in the health sector. Second, both health and education sectors, given the large numbers of health workers and teachers, are far more labor-intensive, requiring considerable management, monitoring and supervision capacity. These demands on management skills are intensified by the fact that in the case of health nearly 90% of the development budget consists of centrally administered projects and programs. (Table 3.9). However, there is very little effective supervision of outreach health workers and very little monitoring and accountability. One important indication of these weaknesses is that actual expenditures on past activities are hard to come by in the health ministry; (and much of the current analysis is therefore based on budget allocations rather than actual expenditures). This in turn gives rise to problems of accountability, delays in fund releases by FCGO and slow implementation as noted in Chapter 2. 3.39 The Facilities Survey described above provides interesting information on the cost patterns at different levels of facility. It was found that the lower the level of the facility, higher is the share of salaries. For example, in the regional and zonal hospitals salaries absorb just over half of the total expenditures, while the share is 80% in the district hospitals; and it is as high as 90 percent in the primary health centers and health posts. One consequence is that in the health centers and health posts which primarily serve the rural areas, there is very little left in the budget for anything else. On average, only about two percent of overall costs are spent on medicines, compared to 17 percent in the regional and zonal hospitals. 3.40 Equity. The foregoing analysis shows major inefficiencies in the health system. The allocation pattern is increasingly biased in favor of urban areas as against rural, hospitals as against primary care, and curative as against preventive interventions. Technical inefficiencies impede the effectiveness of service delivery. In effect, the more remote areas which are most in need of and dependent on public facilities are under-served and receive poor quality of service as well. Although there is a significant network of private providers offering a wide variety of services, recent surveys indicate that the "modern" private system is geographically concentrated in the Central Region and, even there is limited to the high income groups. Given the levels of poverty and consequent inability to pay, there is currently little incentive for private providers to venture outside the Central Region into more remote areas. Overall, the private and non government sectors are responsible for around one quarter of the provision of health care. 3.41 According to NLSS, household expenditures spent on gaining access to publicly provided health services vary substantially by income groups - from Rs. 470 a year for the poorest quartile to Rs. 5016 for the wealthiest quartile. These expenditure as a share of their respective total incomes are less than half for the highest income group as compared to the lowest income group. The average cost of seeking treatment is Rs. 367 (varying from Rs. 183 in primary health centers to Rs. 637 in a hospital), which is equivalent to two weeks of per capita consumption. While part of the reason for the poor not accessing public facilities may be also due to the poor service orientation, (including the derogatory way in which they are sometimes treated in such facilities), these data strongly suggest that the poor simply do not have the resources required to purchase necessary amounts of health care, or even to afford the expenses such as for travel, to access them. NLSS analysis also demonstrated that along with distance to a health facility, -42 - household economic status is a significant determinant of whether or not to seek treatment even at the current level of subsidization. Without a significant degree of subsidization the poor would not have access to health services. 3.42 The incidence of poverty in Nepal is significantly lower in urban areas (23 percent) as compared to rural areas (44 percent). According to the NLSS, twenty-eight percent of cases treated in urban areas over the previous twelve months were in public hospitals. The corresponding figure for rural areas was 16 percent. Conversely, 40 percent of rural cases have been treated at health centers and health posts compared to just 13 percent of the urban cases. Other data suggest that over a twelve-month period, almost five percent of the wealthiest quartile used public health facilities while less than three percent of the poorest quartile used them. Overall, even though the urban population made twice as much use of private facilities as the rural population, they benefited most from public healthcare subsidies because they sought medical treatment more often and used higher cost public facilities more than the rural population did. The available data suggest strongly that the subsidization of higher level facilities such as hospitals benefits the higher income population disproportionately. If subsidization of health services is to be based on the rationale of equity, its pattern must be one which actually benefits the lowest income groups more than proportionately. For this to occur there will need to be greater targeting in the location of facilities with a greater focus on providing facilities in rural areas rather than spending a large proportion of public money on urban hospitals. In addition there is, as discussed, a need for means-based cost recovery for hospital-based treatment. 3.43 Cost recovery. While Nepal does have an embryonic system of cost recovery in public health facilities it is generally low and highly uneven across individual institutions. This in part reflects decentralized efforts in some districts - often pilot projects with donor support - to increase the level of cost recovery for health services and drug supplies. According to information provided by the Nepal Living Standards Survey (1996), in some zonal hospitals and even primary health centers, cost recovery was almost 50 percent of the recurring expenditures for outpatient treatment, while in others the rate was less than 10 percent. In the health posts covered by the survey, patients' contribution averaged Rs. 7, equal to around 15 percent of the total cost of treatment. However, at those health posts operating drug schemes, payments of around Rs. 25 were made by patients. In turn, these health posts spent around 20 percent of their total expenditures on drugs compared to around two-to-three percent in those health posts not operating such schemes and dependent for supplementary income solely on outpatient registration fees. This implies that higher levels of cost recovery are possible but require a higher quality of service. Not surprisingly, individuals' expenditures increase at higher levels of facility. On average, these are twice as high in zonal hospitals as in district hospitals. 3.44 Developing a deeper understanding of both public and private sector health expenditures and financing by level of care, income group, rural-urban location and input mix would greatly assist policy development and assessment of options to refocus public sector health financing priorities in Nepal. Two immediate steps are suggested. First, there is a need to develop means- based instruments for cost recovery for hospital-based treatment (which can also set upper-limits on total fees paid by individual families to guard against catastrophic risks). Second, feesg for curative care in primary health care facilities should be set by village health committees (with some guidance from the Ministry) in a manner which discriminates in favor of the poor. It would also be desirable for these fees to be collected by individual facilities with (at least) an agreed -43 - share to be retained for service improvements at the facility level. There is quite a level of rich experimentation in Nepal which could be used to inform implementation of cost recovery policies at the primary health care facility level. 3.45 Priorities for public involvement. Formal health sector policies and strategies in Nepal provide for improved collaboration between the public and private and NGO sectors. Adopted policies are intended to create a more conducive environment - particularly for NGOs - but implementation has lagged and the potential for closer relationships and partnerships are yet to be fully realized. Given the range of administrative constraints to the effective delivery of preventive and basic primary health care services in many parts of the country within the current public service structure discussed above, one serious possibility is for government to consider (and to immediately pilot) options to contract out a "package of essential health care services" in different districts to NGOs (or other private providers). This option underscores the distinction between the rationale for public financing of preventive health services and subsidies for the poor for curative care and the management and delivery of those services. The case for public financing of health services does not imply the public sector need deliver those services. 3.46 The burden of disease analysis in conjunction with considerations of equity and efficiency strongly suggest that the central theme of the Governments Second Long Term Health Plan - which provides for an "essential health care package" for Nepal should be vigorously implemented. This plan sets out a long term vision of the health sector to be implemented in phases. Initially, the core programmatic focus of the plan should provide for: (i) reproductive and child health needs (including family planning, safe motherhood, EPI and integrated management of childhood illnesses (IMCI); (ii) Communicable diseases including HIV/AIDS; (iii) nutrition; (iv) health promotion and information; and (v) strengthening intersectoral collaboration with agencies and activities which can have a major positive impact on the underlying causes of health problems (e.g., education can help change health behaviors and rural water supplies and sanitation services can radically reduce disease). In order to make progress on these core activities, the institutional capacity of the Ministry of Health to deliver core health and population programs should be strengthened through: (a) public-private partnerships; (b) decentralization of the planning, budgeting and management of key aspects of health and population programs to the district level (consistent with the Decentralization ACT); (c) strengthening of district health care systems; (d) enhancement of the technical capacity of the Ministry to support quality health and population programs; and (e) human resource capacity development and training. 3.47 Conclusion. Similar to all of the social services in Nepal, the development of a national health service is very recent. Over the past forty years much has been achieved by the government, donors and by a variety of other providers. In some ways this delayed start and the variety of experiences which the range of providers has been able to bring has been advantageous, and there are many examples in the country of health provision which is innovative. However, there are areas in which further attention is now required. First, there is the need when planning public interventions to have a full understanding of the current provision, future plans and the potential of the private and non government sectors. These currently provide one quarter of all health services. Second, the government needs to give further consideration to the limited objectives of publicly financing health care. Generally these should be to provide for the poorest groups in general and for services which have wide societal benefits such as preventing and treating infectious diseases. In addition, though lower in importance, a -44 - case can be made for financing treatments of conditions which are rare and expensive to treat, particularly in countries such as Nepal where insurance markets are thin. The economic and social circumstances of the population in Nepal, the evidence from the burden of disease and the very restricted amounts of public resources strongly suggest that the bulk of expenditure should be devoted to conditions which disproportionately affect the poor and have wider societal costs. And the remaining resources need to be focused on preventing and treating the non communicable illnesses of the poor. Recent trends in expenditure patterns do not correspond as closely as they should to these principles. The use of government resources to provide treatment for non communicable illnesses of those who can afford to pay needs to be restricted at this time. This does not imply that government should not provide services for the whole population; but where it does so, the level of subsidy should vary across income groups so as to target it to those who can least afford to pay. Finally, the government may need to consider alternative ways of delivering health services, for example by financing services/activities which can be delivered by private and non government bodies if they are more cost-effective and efficient than direct public provision. E. DECENTRALIZED EXPENDITURE PROGRAMS 3.48 Overview. Decentralized expenditure programs have been the main instrument of successive governments in promoting economic development and a measure of autonomy/decentralization at local levels - i.e. at district and village levels. Originally started on a small scale in the early nineties as an initiative to involve District Development Committees (DDCs) in development activities, these programs have been expanded rapidly since 1995. However, as discussed below, the results of these initiatives have been mixed so far; and there are important lessons to be learnt from them for strengthening the decentralization process in Nepal. The main problems have been the following, among others: First, successive governments have also used these programs as a means of building political support at local levels. Thus some governments have sought to channel funds through District Development Committees (DDCs), while others have favored the strengthening of Village Development Committees (VDCs). This has led to a degree of ambivalence in promoting decentralization and the implementation of these programs. Second, these programs have largely represented a top- down effort to promote decentralization; and the programs have been pushed through without adequate capacity building at various levels-VDCs, DDCs and central government agencies providing technical support for, and supervision and monitoring of these programs. Third, the involvement and empowerment of beneficiary groups has been limited (although there have been stronger efforts recently to promote such involvement), contributing to the poor use of resources and weak management of these programs; and in many instances, the benefits have been largely captured by local elite groups. The rest of this section looks at the effectiveness of these programs in greater detail. 3.49 Budget allocations. As noted in Section B above, apart from power, local development has been clearly the most favored sector in recent years. The sector's performance is important for several reasons. (i) Unlike other sectors, virtually all of the sector's budget represents development spending (Table 3.11). Starting from a low base of around 4% in FY94,- the sector's share has risen sharply to nearly 12% of development budget allocations and about 13% of actual development expenditures by FY99; (ii) It is also the most local currency-intensive sector in the economy. In FY99, roughly 80% of the budget allocations for the sector comprised -45 - HMG's own resources (i.e. 20% donor- Table 3.11: Budget Allocations and Expenditures on Local Development, FY94-FY99 financing), as compared to an average local currency 4 F F9 F allocation of 32% (i.e. Total Allocation 1015.3 2647.5 3848.6 4261.8 4277.1 Regular 11.6 11.6 123.8 29.3 34.1 68% donor-financing) for Development 1003.7 2635.9 3834.8 4232.5 4243.0 the entire development Ata xedtr the entire devt Actual Expenditure 1016.5 2427.5 3357.2 3705.8 4004.3 budget. Moreover, local Regular 9.6 11.4 11.9 27.0 31.8 development accounted Development 1006.9 2416.1 3345.3 3678.8 3972.5 for nearly half of the Memo Item (%) budget allocations for Implementation Rate for Local Dev. 100.3 91.7 87.2 86.9 93.6 fully government financed Implementation Rate for Dev. Budget 93.6 86.4 86.7 85.0 83.4 /a Local development: actual expenditures in percent of development allocation for the sector. projects indicated in Table lb Total development budget total development expenditures in percent of total dev. budget 2.1 above; and (iii) And, cof Fnance. its implementation rate has been significantly higher than other sectors' i.e. it has utilized its budget allocation more rapidly (Table 3.11). Thus, the sector's recent perfornance in terms of the quality of its spending has a crucial bearing on the how well (or badly) scarce local currency resources have been used within the development budget. 3.50 All of the local development programs, except for two projects (Gulmi IRDP managed jointly by MOF and MLD, and Production Program for Rural Women by MOF) are managed by the Ministry of Local Development. MLD's involvement largely consists of (i) administering the grant programs, and (ii) carrying out infrastructure, area development and upliftment programs. The bulk of budget allocations and actual expenditures (over 70%) is accounted for by four programs-the Village Development program (currently Rs. 500,000 per VDC); District Development grants; municipal grants and constituency development programs (currently Rs. 400,000 provided to each member of Parliament). Since then, a number of other initiatives-area development, poverty programs, local infrastructure and road building, women's empowerment, upliftment programs for the under-privileged etc.-have been added to these. (Table 3.12). Table 3.12: Composition of Development Expenditures on Local Development, FY98 - FY99 (Rs. Million) Program FY98 FY99 FY99 Actual Budget Rvised % of Exps. Central Level (15) 285.4 116.6 121.9 3.1 District Level (28) 3393.4 4126.4 3850.6 96.9 Village Dev. Grant 1918.7 1960.0 1960.0 49.3 District Dev. Grant 646.4 650.0 650.0 16.4 Constituency Dev. Grant 102.2 106.0 106.0 2.7 Municipal Grant 139.9 145.0 145.0 3.6 IRDP/Area Dev. Programs (5) 89.0 109.4 113.7 2.9 Rural Infra./RoadslConstruction (7) 386.5 953.1 730.1 18.4 Poverty/Women Dev. Upliftment(7) 95.8 126.4 109.5 2.7 Other (5) 14.9 76.5 36.3 0.9 Total Development 3678.8 4243.0 3972.5 100.0 Note: Number of projects shown in parenthesis. Source: Ministry of Finance. -46 - 3.51 Allocative and Technical Effilciency: By and large, the experience with MLD's grant programs at both the DDC and VDC levels4 has been mixed. A distinction has to be made between the performance of local bodies which have made significant progress in participatory management and of those who have not. From the experience of the former, interesting lessons can be drawn for strengthening the decentralization process (para 3.54) But in the case of the majority of local bodies, there are a number of weaknesses and problems which need to be addressed, including the following: * Detailed procedures and guidelines about the projects to be financed and the allocation criteria and processes at both DDC and VDC levels have been prescribed by the Ministry of Local Development (MLD). But these are not followed in practice. Although at least 25% of the funds are expected to be allocated to human resource development activities, this is not adhered to. The allocation of funds is often made based on the political interests of the VDC chairperson rather than on development needs expressed by the population. In a recent survey of 32 VDCs in 8 districts of Nepal, it was found that 35 percent of VDC development funds are typically spent on rural roads, 38 percent on buildings such as VDC offices or schools. The remaining funds go to drinking water and small irrigation schemes and grants for schools. There tends to be little spending on the maintenance of existing infrastructule, on social needs or on human resource development. * Generally, project selection is not based on rational criteria or analysis of costs and benefits. There is little awareness of the need for targeting funds to the poor or under-privileged groups, although this is one of the main specifications in MLD guidelines. * In many cases, fund allocation at both DDC and VDC levels is not closely linked to community needs. Generally funds are divided equally among the VDCs, and within VDCs, more or less equally among the wards. There is little effort to link fund allocation to performance, or as a means of mobilizing additional resources at local levels. * Typically, decision-making power for the allocation of VDCfiunds is usually concentrated in the hands of the VDC chairperson and the VDC secretary. Participation of other members of the VDC committee, the VDC council, local institutions and the population is marginal in many cases. However, in some better functioning DDCs and VDCs, where user groups have been formed and are functioning reasonably well, there is greater participation and beneficiary influence on the allocation process, as well as implementation (para 3.46). This is however often related to the economic and social status of participating groups, degree of activism, awareness of the availability of funds and the objectives of the local development programs, and their ability to influence decision-makers. 3.52 Technical capacity is limited at virtually all levels for project preparation, design, implementation and monitoring, and ensuring effective use of funds and accountability. Although MLD has tried to strengthen technical capacity at the village level by providing one technical assistant for each VDC, this has not materially improved the situation. This in turn has led to several negative and undesirable consequences: 4 This discussion draws on the findings of three recent studies on district and village level programs: "The Impact of Decentralization on the Poor," Human Resource Development Center, Kathmandu September 1998; "A Review of the Effectiveness of Government Funded Programs," Dr. D.R. Khanal, December 1998; and "Combating Rural Public Works Corruption: Food for Work Programs in Nepal" Meagher," Upadhyay and Wilkinson; may 1998. -47 - * Financial record-keeping is poor in most DDCs/VDCs. In many cases, DDC/VDC chairmen do not report to the community or to other members of their respective committees on financial or physical progress of their programs. MLD's monitoring and supervision capacity is equally weak. There is little supervision, and virtually no monitoring of physical progress and outputs. * There is little auditing. Audit reports are non-existent in many cases. Although statement of expenditures are expected to be submitted as a pre-condition for release of funds, this is not observed in practice, both because of improper record keeping at DDC/VDC levels, poor monitoring capacity of MLD and political pressures which effectively prevent the FCGO from withholding fund releases to transgressors. * As a result, leakages and misuse of resources are considerable. Irregularities in the selection of contractors for infrastructure projects and kickbacks are common. Cases of VDC chairpersons using their influence to illegally obtain large personal advances from their VDC funds are numerous. Accusations against local officials for misappropriating funds are common, particularly in the Terai. 3.53 Program implementation has also suffered from limited capacity of the overseeing line ministry-MILD. Apart form its grant programs, MLD has displayed a consistent pattern of under spending in other areas, notably on rural infrastructure, roads and women's development and upliftment projects. The average implementation rate for these has been generally below 80% of the allocations. This supports the view that MLD does not have the capacity at present to properly manage sector projects. A related concern is that there are considerable overlaps at the district and local levels between programs run by MLD and those of other ministries, such as education, health and transport. These are not well coordinated both in terms of budget allocations and implementation modalities, leading to duplication and potential waste of resources. It may therefore be more efficient to let MLD focus exclusively on grant programs, streamline other overlapping programs and transfer them as appropriate either to the relevant sectoral ministries, or departments, or to (strengthened) local bodies themselves. 3.54 Equity Effects. According to the recent studies cited above, in many cases it is the relatively better-off population which has benefited most from the DDC/VDC programs. (i) The better-off have used most of the funds because they have better access to resources, they belong to the upper social and economic groups, are more influential, more literate and more aware of the availability of funds; (ii) In contrast, only a small proportion of funds has gone to the poorer and disadvantaged sections of the community. Women as a group have been even more disadvantaged in the fund allocation and utilization process. In the VDCs surveyed, roughly 80 percent of women and 60 percent of the members of the worse-off groups were not aware of the fund position of the VDCs; and (iii) Third, there is a problem of targeting most of the typical infrastructure development projects to the perceived needs of the poor. The rich tend to benefit more than the poor from these investments. The poor and the women generally find it difficult to relate most of these projects to their immediate or long-term needs. For example, the road/trail projects were, in most cases, rehabilitation or maintenance of existing ones, in which the poor saw no additional benefits. Similarly, the construction of buildings (mainly VDC buildings and school buildings) has taken up one-third of the VDC funds and it is difficult to relate most of these projects to the perceived needs of the poor. However, as noted above, there are also examples of local bodies where technical support and participatory management have made a significant difference in the way they manage -48 - and use resources; and these provide a model for strengthening the decentralization process and resource management by local bodies. 3.55 Recent Reforms. To address these shortcomings of local bodies, several programs are being implemented in Nepal to strengthen their management and technical capability. The largest of these are UNDP's Participatory District Development Project (PDDP), and DANIDA's local government training program. Germany (GTZ), Switzerland (SDC) and the Netherlands (SNV) are also supporting smaller programs. Some interesting lessons have been learnt from the more successful experiments. For example, increasing awareness and access to information help to increase participation at the village level, particularly of the poor and underprivileged groups. Equally important, greater transparency and disclosure with regard to the availability of resources and the purposes for which they are to be utilized help to promote better use of resources and greater accountability. Where there is greater participation, there is also evidence of community groups transcending political divisions to work together to achieve common objectives in the use and allocations of resources at local levels. Also, NGOs and other community organizations can play a supporting role in this process by fostering community involvement, infornation sharing and empowerment and by providing technical support to local communities. While these developments are encouraging, they have not made a significant difference in the majority of VDCs so far. 3.56 The second important change is the recent enactment of the Local Autonomy Bill, which devolves additional development, financial and judicial powers and functions to local bodies. According to the provisions of the new law, the VDC's main development functions include the procurement of projects from ward committees and NGOs, the formulation of annual plans, the management of local development funds, the preparation of resource maps, for their areas, the explanation of projects to local people, the training of user groups for the implementation of their projects, and the formal coordination of NGOs. The main development functions of the DDC include the formulation of annual district development plans, management of district development funds, the coordination of NGO activities, the supervision of line agencies, municipal and VDC development activities, and the explanation of projects to local people. The new legislation also empowers local bodies to levy local taxes. However, there has been little progress so far in implementing the new Act. The administrative regulations which are necessary for its implementation have been formulated only recently (late December); and as discussed in Chapter 4, there are important political and bureaucratic impediments which need to be removed (para 4.21). 3.57 Conclusion. The rapid increase in resources allocated to local development since early to mid nineties reflects the strong political support for decentralization that exists in Nepal. Increased powers and duties have been devolved to locally elected bodies. However, the experience so far with the implementation of most of these programs has been mixed. The main issues at stake in the decentralization process are (i) the need for improved governance and transparency; and (ii) for enhancing management/technical capability. The first priority should be to increase transparency and accountability of local governments to their constituencies. In parallel, the institutional and management capacity to formulate and carry out projects and programs by local governments, as well as effective supervision and monitoring mechanisms need to be strengthened. Promoting greater participation of local level beneficiaries and community groups is essential for strengthening -49 - accountability and transparency. At the same time, as discussed in Chapter IV, the roles of central government agencies need to change to provide technical support and monitoring. F. AGRICULTURE, IRRIGATIONAND FORESTRY 3.58 Introduction. Agriculture, broadly defined to include irrigation and forestry, is a key sector in the economy in terms of the allocation and use of public resources. It is particularly important from a development and poverty reduction perspective since nearly 90% of the population and the large majority of the poor live in rural areas and depend on agriculture for most of their incomes and employment. The recent Poverty Assessment provides some important information on the linkages between pervasive poverty and the poor performance of the agricultural sector in Nepal: (i) agriculture is the main source of employment for 83% of the labor force; 86% of households cultivate some land; 80% have some livestock; and two-thirds of the incomes of all households are derived from self-employment in agriculture or agriculture- related wages; (ii) Poverty is also more widespread and deeper in rural areas. In 1996, 44% of the rural population lived below the poverty line, as compared to 23% in urban areas and a national average of 42%. In terms of income differentials among the various sectors of the economy, according to the FY99 Labor Force Survey, value added per worker in agriculture was only $265 per annum, i.e. the equivalent of 53% of the national average, and only 25% and 19% of average value added by those working in industry and services sectors respectively; (iii) The poor performance of the agricultural sector in terms of production/income growth has been a major factor contributing to the lack of progress in reducing rural poverty in Nepal. The rate of growth in the agriculture sector in real terms was only 2.6% p.a. over the past two decades, compared to an average population growth rate of 2.5% p.a. Indeed in the last decade, the production of food crops declined between FY91 and FY95 and of forestry products stagnated between FY94 and FY97. The overall agricultural growth rate remained positive during this period only because the output of livestock products and cash crops grew by about 4% p.a. 3.59 The poor overall performance of agriculture can be explained in terms of a number of inter-related factors: low yields/low productivity; population pressure on the land contributing to fragmentation, small land holdings and cultivation of marginal land; lack of irrigation facilities and modern inputs; low levels of technology; and limited access to markets, among others. Nepal so far has not witnessed the green revolution which benefited other countries in the region over the past two decades. Only 35% of the cultivable land in Nepal is irrigated, and only about 40% of that is under year-round irrigation. Similarly, fertilizer usage in Nepal is only about 33 kg per hectare, which is only about one-third the level in Bangladesh and Pakistan. 3.60 Agriculture sector policies. Past policies aimed at accelerating agricultural development mainly consisted of efforts to expand infrastructure and the provision of key inputs through public interventions-for example, the creation of an Agricultural Inputs Corporation for the import and distribution of fertilizer and seeds; agricultural research, extension and marketing support; surface irrigation schemes; and rural credit largely undertaken by, or provided through, government departments and public sector agencies-and integrated rural development projects financed by the government and the donors. These efforts however met with limited success for a variety of reasons, including poor project design and implementation, institutional weaknesses, - 50 - poor delivery mechanisms, and inefficiencies of public sector agencies involved, among others (see below). 3.61 In order to reverse this trend, the Agricultural Perspective Plan (APP) - a twenty year growth framework-was developed and incorporated in the Ninth Plan. The main objectives of the APP are to: (i) accelerate growth in the agriculture sector by increasing productivity; (ii) diversify agriculture to high value crops; (iii) ensure a policy environment more conducive for private sector/community participation; and (iv) reduce poverty through increased agricultural growth and employment opportunities. The emphasis of the APP is on harnessing the considerable growth potential in the Terai, while encouraging livestock development and the development of high value crops on a selective basis in the hills. The APP's implementation strategy focuses on: (i) expansion of year-round irrigation-mainly shallow tubewells-with farmer participation; (ii) increasing the supply of fertilizer with private sector participation; (iii) streamlining agricultural research and extension and adapting them to farmers' needs; and (iv) provision of adequate rural infrastructure, especially rural roads. The emphasis of the APP is to increase investments by private sector, community organizations and the public sector, facilitated by appropriate policy and institutional reforms and the increased availability of credit. 3.62 Though conceived in the mid nineties, the implementation of the APP strategy was delayed until FY98, and hence its impact on agricultural growth has not been noticeable so far. However, there has been some significant progress on the policy front with elimination of all fertilizer subsidies, reduction of irrigation subsidies and institutional reform of the Agriculture Inputs Corporation (AIC) and Nepal Food Corporation (NFC). Also, budget allocations in the agriculture sector in the past 1-2 years have been increasingly adjusted towards meeting APP needs, although much more remains to be done. The discussion below focuses on the historical pattern of sectoral spending since the mid-nineties and its effectiveness, changes underway in terms of institutional and policy reforms, and expenditure prioritization/reallocation that is needed to implement effectively the APP strategy. 3.63 Allocation and Efficiency of Resource Use. The broad trends in the allocation and use of public resources within the agricultural sector are summarized in Table 3.13. The table shows that: * Agriculture sector's share of budget allocations and expenditures have fallen steadily over the past few years from about 15% of the government budget in FY95 to around 10% in FY99. This in part was due to government decisions to give priority to local development, health and power generation noted earlier, and also due to conscious decisions to reduce funding to some areas within agriculture as discussed below. * Within the overall agriculture sector, irrigation and forestry both received some significant increases in nominal terms which allowed them to more or less maintain their expenditure levels in real terms. However, allocations for the agriculture sub-sector were reduced even in nominal terms, leading to significant cutbacks in real terms. - 51 - Table 3.13: Budget Allocations and Expenditures for Agriculture, FY95 - FY99 (Rs. Million) FY95 FY98 FY97 FY98 FY99 Budget Allocations Agriculture 6187 6948 6944 6782 7394 Agriculture 2843 2758 2653 2446 2438 Irrigation 2427 3117 3290 3263 3598 Forestry 917 1074 1000 1073 1358 Actual Expenditures Agrculture 6142 6141 5767 5716 6346 Agnculture 2703 2292 1965 2225 2241 Irrigation 2649 2989 2826 2541 2945 Forestry 790 860 977 950 1160 Memo Items: Agriculture's share of govemment budget (%) Budget Allocations 14.5 13.8 12.1 10.9 10.6 Actual Expenditures 15.7 13.2 11.4 10.2 10.1 Source: Ministry of Finance. AGRICULTURE 3.64 The main focus of programs managed by the Ministry of Agriculture has been on (i) fertilizer and other supplies (20-30% of the allocation); (ii) research and extension (20-24%); and (iii) livestock development (18-24%) (Table 3.14). In addition, within the agriculture sector there have also been substantial agricultural credit programs which have been channeled through the Ministry of Finance. 3.65 Actual spending within the sub-sector declined in nominal terms from an average of 99 percent of the total allocation in FY95 to about 83% in FY97. After adjusting for inflation, real spending fell even more to only 61% of the level in FY95 (Table 3.14). The main reasons for the poor implementation rate have been: (i) the credit program (mainly funded by donor assistance and channeled through the Agriculture Development Bank of Nepal) has been gradually curtailed as donor funding dried up given the poor performance of ADBN; and (ii) fertilizer subsidies have been gradually reduced as a matter of policy (see below). (iii) However, the livestock development program has been expanded because of the introduction of three donor-assisted programs. 3.66 A major reason for poor yields and agricultural productivity in Nepal has been the low application rate of fertilizers, the lowest in South Asia. The fertilizer subsidy has been a large part of the budget and yet inadequate to meet the needs of farmers. Hence, as part of an Agricultural Sector Reform Program, the subsidy has been eliminated since November 1999. The private sector has been given the responsibility to import fertilizers and the extent of their involvement is yet to be seen, though there is evidence to show that fertilizer supply has increased significantly. The role of the Agricultural Inputs Corporation has been reduced, but to a limited extent it is competing with the private sector. - 52 - 3.67 The MOA has failed to provide farmers with the research and extension services they need. The Nepal's Agriculture Research Council (NARC) has the responsibility to carry out agricultural research, but most of its work has not been well connected to the relative importance of different commodities. The performance of the Department of Agriculture, which is responsible for agricultural extension, has been poor. Its staff are spread thinly across the country and they have little incentive to work directly with farmers. MOA has been advocating reforms of the research and extension systems through the "pocket" approach but there has been very little evidence on the ground. 3.68 MOA's role in directly providing research and extension services should be phased out, focussing instead on the formulation of overall policies and the monitoring and the evaluation of private sector service providers. To begin phasing out its direct provision of services, the MOA should consider innovative alternatives, which have already been successful in other South Asian nations, such as sub-contracting research and extension to specialized NGOs and firms. 3.69 The Government has been distributing food in 38 remote districts by providing a subsidy on transportation through Nepal Food Corporation (NFC). Expenditures have been around NRs. 225-300 million annually for the distribution of 15,000-17,000 metric tons of food. NFC has been incurring losses (NRs. 225 million in FY98) and reforms were introduced recently (January 2000) to restructure NFC and target 27 remote districts that do not have a road link nor production potential. The number of staff has also been reduced significantly and a ration card system has been introduced to target the most vulnerable. IRRIGATION 3.70 Budget allocations and expenditures. Most of the governments' irrigation programs are managed by the Ministry of Water Resources (MOWR). There is also a credit program managed by the ADBN which is not covered in this review. The composition of expenditures indicates that the focus of the ministry is on direct physical investments in infrastructure. Irrigation works consumed 87 percent of FY97 allocations to the irrigation sector. Within the latter, the largest investments were for surface schemes, with tubewell programs the next largest. Only 3.1 percent was allocatedfor overheads - mostly for operations and maintenance. 3.71 There has been some shift in the intra-sector allocations recently from large surface water projects and deep tubewell projects to shallow tubewells, small and medium scale surface schemes and flood control and emergency repair programs. These changes are consistent with the strategy laid out in the APP (Table 3.15). Nevertheless, the share of large scale surface irrigation continues to be high. Because of the increase in nominal budget allocations for irrigation, most of the programs have been able to maintain their expenditure levels in real terms, except for operations and maintenance expenditures, lift irrigation and flood control and emergency repair. The latter however is now receiving higher priority and increased allocations under the APP in the FY99 budget. 3.72 The main issue in the sector continues to be how to promote irrigation development in the most cost-effective manner to promote agricultural development. In the past, the public sector focussed mainly on large scale surface irrigation which, in the absence of effective cost recovery, - 53 - has been highly subsidized. Table 3.16: Budget Allocations and Expenditure Trends in Forestry, FY95- Concurrently, the FY98 government also (In Percent) subsidized farmers FY95 FY96 FY97 FY- 8 for private and Budget Allocation community Forest Development 48 47 45 44 development of Soil & Water Conservation 13 10 13 14 shallow and deep National Park and Wildlife 22 27 26 25 Protection tubewells at varying Research and Extension 8 6 6 4 rates-40% for Administration 10 10 9 11 individual shallow Total 100 100 100 100 tubewells and 80% Actual Expenditures (Indices in for community Constant Prices) Forest Development 100 104 95 - shallow tubewells Soil & Water Conservation 100 70 130 and 90% for deep National Park and Wildlife 100 118 109 Protection tubewells. However, Research and Extension 100 67 74 the subsidy budget Administration 100 132 112 has been inadequate Total 100 103 103 Source: Ministry of Finance and staff estimates. to meet demand. Last year the subsidy for individual shallow tubewells was eliminated, while those for community tubewells and deep tubewells were reduced to 60% and 84% respectively. Recently, the government decided to further reduce capital subsidies on community shallow tubewells to 30% and to eliminate the subsidy by July FY2000. This has however created an anomalous situation since government continues to finance surface irrigation schemes without adequate cost recovery. Under the IDA financed Irrigation Sector Project (ISP) a study is being carried out to assess the subsidy issues in irrigation, which is expected to be completed by the end of 2000. 3.73 Credit programs for farmer-owned surface schemes is subsidized through a variety of donor projects as well as the government and are disbursed to farmers through the Agriculture Development Bank. These programs are also under review to determine their efficacy and the rationale for public involvement. 3.74 The government levies a tariff for water use for the large public surface schemes, but the present rate is supposed to cover only 40% of the estimated costs. Moreover, the collection rate is very poor, averaging only 30%, making the effective cost recovery very low and the implicit subsidy quite large. As a consequence, real expenditures on operations and maintenance are sub- optimal and undermine the performance of the public irrigation systems. The overall implicit subsides in surface irrigation schemes are being reviewed by the same study. FORESTRY 3.75 The forestry sector program provides four inter-related services: forestry management and development, parks and wildlife management, soil and water conservation and research and extension. In terms of budget allocations, the largest programs are soil and watershed conservation and reforestation, and protection of the forests and national parks. In terms of - 54 - programs, expenditures in the forestry sector are for national, community, leasehold and private forests. National forests are owned and managed by the government. Community forests are owned by the government but managed by Forest User Groups. Leasehold forests are owned by the government but leased to individuals or organizations for commercial purposes. MOFSC expenditures for all categories of forestry make up 44 percent of its FY97 budget. (See Table 3.16) Government resources are spent almost exclusively on community forestry, leasehold forestry and the District Forest Offices. The MOFSC does contribute towards private forestry but mainly through the free distribution of seedlings and limited technical assistance. The District Forest Offices (DFOs) accounted for 21 percent of the FY97 budget. Real spending on DFO activities increased slightly in FY96 and FY97. The increase has been attributed to management support to communities in the early years of transfer and also monitoring and evaluation activities. 3.76 Considerable progress has been made in implementing the community forestry program under which District Forestry Officers (DFOs) are empowered to transfer any portion of a national forest to qualified Forest User Groups. User groups are allowed to use and freely sell forest products within general guidelines that promote good management. However, a recent directive (November 1999) by the Ministry has restricted the sale of forest products by the communities. This will negatively impact on the successful community owned and operated forestry program. Additionally, the transfer of forests has gone smoothly in the hill districts but lags far behind in the Terai, where, only 5 districts out of 21 have actively handed over portions of the national forest to user groups. Part of the problem is due to unresolved issues related to the conversion of forest to agricultural land in response to heavy population pressure. Hence the Government will need to develop a suitable policy framework to encourage community forestry taking into account the trade-offs with agriculture so as to protect the forests. 3.77 The leasehold program has been a significant source of revenue. In FY96, total revenue earned was NRs. 442 million (about half of the forestry budget). The Government may need to consider developing this program to encourage communities rather than individuals to take over leasehold land. Additionally the leasehold program may be an effective and profitable way for the government to reassert control over illegal logging. 3.78 A related issue is the monopoly conferred to the Timber Corporation of Nepal (TCN) in the marketing of forest products. Apart from the fact that TCN is an unviable institution making significant losses every year, this arrangement is a source of corruption and leakages. In the interest of promoting a commercially viable forestry sector on a competitive and transparent basis, there is a need for de-monopolizing the marketing of timber. 3.79 Affordability of Rural Development Strategy. The Agriculture Perspective Plan (APP) has been adopted by the government to accelerate agricultural growth and rural development. The first phase of the APP was originally intended to cover the years 1995 through 1999. The necessary investments represented a major increase in public spending compared to previous levels. But, given the lack of resources, the government was unable to make a serious commitment to APP implementation; and public expenditures on projects consistent with APP objectives represented roughly around 70% of the APP target levels in its first three years. In some sectors, such as livestock or horticulture, it was far less. - 55 - 3.80 To effectively implement the APP will be a major change. Resource requirements for the APP for the next few years, without counting substantial investments in farm-to-market roads and rural electrification, will be over Rs. 6 billion a year. To meet this need, over 200 heads of expenditures in the rural sector will need to be reviewed by the government and the donors. Projects and programs that are not consistent with the objectives of APP objectives would have to be restructured or eliminated. Second, even if most non-APP activities were eliminated, the total funding consistent with the APP objectives may still fall short of APP requirements. Hence it would be necessary to rethink the role of government in the agricultural sector, with a view to minimizing public involvement and catalyzing private investment to the maximum possible extent. For example, as noted earlier, the role of the government in input provision, extension and research and production-oriented activities (agricultural farms, commercial forestry etc.) may need to be phased out. Similarly, the governments' present strategy of irrigation development (i.e. heavily subsidizing large scale surface irrigation, while removing subsidies to private ground water development) may need to be reviewed. Public investments will need to focus on areas which are still critical to agricultural development, but where private investment has little scope i.e. investments in farm-to-market roads, which are presently heavily under-funded. Even then, it is likely that the APP will not be fully affordable, and its targets may need to be scaled down. G. TRANSPORT 3.81 Introduction. Transport is a key sector in the economy, accounting for roughly 9 -10% of the government's development budget allocations and about 10% of actual development expenditures in the nineties. While sharing many of the weaknesses in public resource management common to all sectors, the transport sector has made significant progress in recent years in terms of expenditure planning, institutional development and capacity building. The sector largely consists of two sub sectors-roads and civil aviation-which account for roughly 83% and 17% respectively of sectoral expenditures in recent years. Sections H and I of Chapter 3 review the recent performance of the roads and civil aviation sub sectors respectively and highlight their significant achievements, as well as the key issues which need to be addressed in order to enhance the efficiency of resource use within the transport sector. H. ROADS 3.82 Recent Performance. Isolated from the rest of the world for centuries, Nepal had virtually no road system until the early 1960s. Since then, considerable progress has been achieved in creating a core network of roads and trails (as well as an internal air transport system) with assistance from external donors, who have continued to finance over 60% of development activities within the sector. Much of the emphasis in the early decades had been on creating and expanding the road network without due attention to its maintenance, and have been characterized by typical weaknesses hampering effectiveness of public spending in Nepal, including such factors as poor planning, implementation and management and waste of resources. Since the early nineties, however, significant efforts have been made to address these weaknesses with encouraging results, including the following: * The road network has expanded from a total length of about 2000 km of basic roads in the 1960s to 11867 km in 1997, consisting of 3660 km of black-top roads, 3098 km of gravel - 56 - roads and 5109 of earthen roads. In addition, there are about 7000 km of main trails. However, the network is still modest in relation to the country's needs, particularly for farm- to-market/agricultural roads; and 17 out of Nepal's 75 districts remain unconnected by roads. * Substantial improvement in the overall condition of the strategic network. According to a survey carried out by the Department of roads (DOR) in 1992 only 48% of the strategic network was in fair to good condition. Subsequent surveys show that 80% of the network has been upgraded to fair-to-good condition by 1996 and to almost 90% by 1997. This improvement however has largely resulted from major road rehabilitation programs financed by external donors throughout the nineties. While the strategic network has now been brought up to a maintainable condition, continued government commitment to regular maintenance in order to ensure sustainability is now a key issue. * Significant improvements in institutional capacity within the sector for expenditure planning and resource allocation, project implementation, maintenance management, monitoring and ensuring compliance with environment guidelines and standards. The key initiatives in this regard include: - The formulation of a Priority Investment Program (PIP) for the road sector within a reasonably tight budget constraint; - The establishment of maintenance planning and programming capacity within DOR by setting up a Maintenance and Rehabilitation Coordination Unit (MRCU) in 1992; - The establishment of a Road Sector Skills Development Unit to address human resource constraints within DOR; - Establishment of an effective field program of strengthened maintenance units on a regional (6) and divisional (25) basis to program and carry out maintenance at the local level, - Streamlining DOR's functions and responsibilities in conjunction with the devolution of responsibility for construction and maintenance of district/village level roads to local authorities, and the establishment of a Department of Local Infrastructure Development and Agricultural Roads (DOLIDAR) within the Ministry of Local Development (MLD) to provide technical support and guidance to local authorities for such activities. * New initiatives to involve the private sector and local community groups in road maintenance; mobilize additional resources for, and to enhance the effectiveness of such resources in, road maintenance activities through the recent establishment of a Road Fund. This initiative is now considered a model for the South Asia region and other developing countries. 3.83 Allocation and Efficiency of Resource Use. Recent trends in the pattern of resource allocation and financing in the roads sub sector are summarized in Tables 3.17 and 3.18. These show that: (i) Over 95% of the roads sub sector allocations comprise development spending, which includes virtually all of the operations and maintenance activities; and (ii) donors finance roughly 70-75% of development spending; (iii) In regard to the composition of budget allocations, roughly 60-65% has been allocated for rehabilitation and maintenance in recent years, with another 10% earmarked for the construction and maintenance of bridges. Within the allocation for rehabilitation and maintenance, the lion's share represents rehabilitation -57 - activities; while allocation for Table 3.17: Trends in the Roads Sector Budget Allocations and Financing, maintenance - FY92 - FY99 roughly Rs. 420 (In Million Rupees) million in FY99 - FY92 FY94 FY96 FY98 FY99 are much below Total Roads Budget 2561 2751 4061 4840 5328 requirements. On Regular 161 118 174 178 184 the other had, Development 2400 2634 23887 4662 5145 Financing 100.0 100.0 100.0 100.0 100.0 allocations for Donor Contribution 51.3 66.0 57.2 64.6 67.5 new construction HMG Contribution 48.7 34.0 42.8 35.4 32.5 Memo item 7.7 are considerable - Roads as % of National Budget 9.7 8.2 8.7 8.4 7.7 about 25% of the Source: DOR, Mlnistry of Finance. development budget for the roads sub sector annually; (iv) It is also worth pointing out that the bulk - about 75% on average - of the rehabilitation activities noted above are financed by external donors; (v) the government earmarks a substantial part of its own resources for new construction. The latter includes new roads linking district headquarters (which are a high priority for the government), as well as several (in some years over a hundred) new road projects every year. Many of these "political projects" have little economic justification, and generally receive a few million rupees each. However, collectively they add up to a few hundred million rupees, representing a significant misallocation of scarce local currency resources; and (vi) With the implementation of decentralization policy, from FY93 onwards DOR is no longer directly responsible for the construction and maintenance of rural roads. These activities have been passed on to local authorities and municipalities; and are now financed through grants to DDCs, VDCs and municipalities, as well through various infrastructure programs financed through MLD. The DOR's roads sub sector budget includes only a few activities under rural/local under roads - such as pilot labor-based district roads rehabilitation, the postal roads program and the suspension bridges project, etc. - for which about Rs. 200 million a year has been channeled through DOR. 3.84 In regard to effectiveness of public spending, the recent emphasis on rehabilitation and maintenance has been highly appropriate and effective. But, as noted, there is considerable wastage and frittering away of resources associated with the new construction activities and the multiplicity of small and uneconomic projects noted above. Moreover, as discussed elsewhere (Chapter Im Section E), although substantial resources are provided for and spent on local road construction and maintenance activities, much of these resources are poorly utilized or misused. There is thus considerable scope for improving the effectiveness of expenditure programs in the roads sub-sector. 3.85 Although budget allocations for the roads sub-sector have been increasing at an average rate of about 11% p.a. in nominal terms during the nineties, actual expenditures have been consistently below allocations. The utilization ratio, though improving over time, has averaged only about 72% of the allocations during the nineties. The wide gap between the budget allocations and actual expenditures reflect a number of constraints identified in chapter lI above, such as over- programming of the budget, delays in fund releases, limited absorptive capacity and various implementation problems, including procurement delays, frequent transfers of project staff, poor contract-management, diversion of resources to political priorities, political interference and -58 - corruption. However, the various initiatives undertaken by DOR since the early nineties to strengthen its institutional capacity have led to significant improvements over time. Table 3.18: Composition of Roads Sector Budget Allocation and Financing, FY98 and FY99 (Rupees Million) Regular 178 0 178 184 0 184 Development 1536 3126 46612 1550 3595 5145 New Construction 625 565 1190 505 796 1301 (52.6) (47.4) (25.5) (38.9) (61.1) (25) Rehabilitation & Maintenance 728 2106 2834 833 2495 3328 (25.7) (74.3 (60.8) (25.0) (75.0) (64) Bridges 129 376 505 156 273 429 (25.5) (74.5) (10.8) (36.4) (63.6) (8) Miscellaneous 53 48.7 34.0 56 31 87 (39.9) (60.1) (2.9) (64.4) (35.6) 7.7 Total Allocation for Roads 1714 3126 4840 1734 3595 5329 Note: Figures in parenthesis are percentage shares of sub-totals. Source: DOR and Ministry of Finance. 3.86 Key Issues. Notwithstanding the substantial progress achieved in recent years in the roads sub-sector, a number of key issues need to be addressed in order to improve the efficiency of resource use in the sector: (i) The major need in the sector is to ensure sustainability of the progress that has been achieved so far and to build upon it. In particular, the strategic road network has been brought upto a maintainable condition through extensive rehabilitation programs largely financed by donors. The past practice of allowing the road system to deteriorate through lack of timely and adequate maintenance and then rehabilitate it through donor assistance may have made sense from the government's perspective in the past as it freed up its own resources for undertaking new construction and spending on other areas (such as local development). However, from an overall resource management perspective, it is a highly inefficient way of using scarce donor (as well as government) resources. Indeed, recent studies show that the annualized cost of road construction and rehabilitation has been 3-4 times as high as that of foregone maintenance, underlining the need for giving topmost priority to maintenance. Current estimates show that Nepal needs to spend at least Rs. 1.0 billion a year to maintain the road network in good condition, as against Rs. 400 - 500 million currently allocated for such activities each year. Thus, a better balance between maintenance and new construction is necessary to ensure the sustainability of the road network. (ii) The formulation of a Priority Investment Program (PIP) in recent years and its endorsement by the government is an important innovation in improving resource management in the sector. However, in actual implementation, the PIP has not been adhered to in practice, as is evident from the continuing high allocations for new roads and the introduction of several new small roads projects each year. Indeed, during the Eighth Plan (FY93 - FY97), 1092 km of planned roads were completed against a planned target of 1778 km; but an additional 1771 km of unplanned roads were built in that period indicating both - 59 - ad-hoc allocation of resources in response to political pressures, as well as diversion of funds to new construction from plan priorities, including maintenance. The challenge in the coming years is the execution of the public expenditure program in accordance with the PIP. (iii) A quick review of the Ninth Plan (FY98 - FY2002) shows why redefining the PIP and adhering to it strictly during implementation are likely to remain important issues over the next few years. By and large, the Ninth Plan's objectives seem appropriate from a long term planning perspective. The Plan states that its major priorities in the transport sector include among others, maintenance of national highways and north-south feeder roads; completion of strategic roads in the west, as well as access roads for nationally important projects for tourism development; extension of agricultural and rural roads; and developing capacity for transport management, contract administration and strengthening the domestic construction industry. In the roads sub-sector it emphasizes giving priority in the next 20 years to construction of roads connecting district headquarters, agricultural and farm-to-market roads, tourism routes and social roads contributing to poverty alleviation; initiation of environmentally friendly and low-cost road system; development of an integrated transport network plan; and adopting measures to reduce vehicle operating costs through proper road maintenance. The Plan then envisages construction of additional 600 km of district roads connecting district HQs per year (i.e. 12,000 km over the next 20 years). However, these plans do not take adequate account of resource availability in the foreseeable future and appear to be as unrealistic as in the past. (iv) Although significant efforts have been made recently to strengthen institutional capacity, project implementation continues to be a problem in the transport sector as elsewhere. There are still weaknesses at many levels which need to be addressed, particularly political interference in staffing and procurement decisions, frequent staff transfers, deficiencies in procurement, contractor management, project monitoring, accountability and corruption. This is a particularly serious problem at the level of local institutions - DDCs, VDCs and municipalities - where the main issue at present is not the lack of resources, but institutional capacity as well as effective technical support and supervisory arrangements to ensure that resources are deployed and effectively used in a manner that promotes efficiency and accountability (para 4.21). (v) The recent initiatives in the roads sector to ensure greater commitment to maintenance activities, to mobilize resources for the purpose and to closely involve the private sector and beneficiary groups in such activities through the establishment of a road fund is very encouraging. But is important to recognize that in Nepal in many sectors good initiatives have been started in the past, but these have not been effectively followed through or produced the expected results mainly due to lack of commitment to their implementation. The real test of this latest initiative is whether strong government action to implement it forcefully would be forthcoming, particularly since it would involve the devolution of considerable resources and their management from the government to a fund largely managed by the private sector and beneficiaries. - 60 - L. AIR TRANSPORT 3.87 Introduction. Apart from roads, air transport is the only other modem means of transport of any significance in Nepal. Indeed, for many remote towns and villages in the mountainous areas of the country, it remains the only practical transportation option. On average a little less than a fifth of the transport sector budget has been allocated to this sub sector. 3.88 The air transport sub sector development began only recently in Nepal. The Seventh Plan (FY98-FY92) mainly focussed on the improvement of existing airfields, rather than construction of new ones. Since then, there have been some important changes: (i) The Ninth Plan (FY98- FY2002)while giving special attention to the development of the Tribhuvan International Airport (TIA) in Kathmandu; (ii) also gave increased attention to the promotion of air traffic in the regional hubs as well as STOL (short take-off and landing) airports, with a view to encouraging tourism development and facilitating access to mountainous and remote areas in all development regions; (iii) towards this end, the government embarked on the liberalization of civil aviation policy, opening up domestic air services to private airlines in 1992; and (iv) More recently, as a logical corollary to rapid growth in domestic air services, the government also embarked on transforning the Department of Civil Aviation (DCA) which was then responsible for regulation and supervision of civil aviation operations, into an autonomous Civil Aviation Authority of Nepal (CAAN) in 1997. As a prelude to this institutional change, a comprehensive national Civil Aviation Policy was introduced in 1993, covering a wide range of provisions for ensuring the safe and efficient development of air transport, in keeping with the liberal sky policy and deregulated environment. 3.89 Budget allocations and utilization. There has not been any consistent pattern in the budget allocations and their utilization in the air transport sector. This in part reflects the fact that expenditure allocations have been largely driven by the availability of external aid for key development projects in the sub sector. (As Table 3.19 shows, regular expenditures consist of roughly 10-15% of sectoral expenditures, the rest represents development activities). (i) Thus, total budget allocations have fluctuated from a low of Rs. 474 million in FY95 to a high of Rs. 1329 million in FY99, with the development budget allocations as well as actual expenditures largely mirroring these fluctuations; (ii) Much of the development budget in turn has reflected upgrading and improvements of TIA, with the development of other hubs and STOL airports receiving only Table 3.19: Air Transport: Budget Allocation and Expenditure, FY94-FY97 (Rupees Million) 9~~~~~~ Total Allocation ? ? ? Regular ? ? ? Development 625 490 764 Total Expenditure 565.9 419.2 379.4 Regular 58.0 69.8 80.0 Development 507.9 349.4 299.4 Total Development Expenditure + Allocation (%) 81.3 71.3 39.2 Memo Items: Revenues from Airports 354.4 532.7 563.4 Revenue/Expenditures (%) 62.6 127.1 148.5 /a Tribhuban Intemational Airport. Source: Budget Documents; Ministry of Finance. -61 - small proportion of the allocations on average; (iii) However, there has been a large increase in budget allocations in FY99 and again in FY2000, reflecting the establishment of the CAAN and government commitment to providing its initial resource requirements; (iv) In regard to actual utilization, during the Eighth Plan period (FY93-97), annual expenditures averaged Rs. 538 million, in FY93 prices (i.e. about 17% of the transport sector expenditures and about 3.1% of all development expenditures under the Eighth Plan; (v) As in other sectors, actual expenditures have been significantly less -only about 76% budget allocations-in the FY94 - FY96 period, reflecting a variety of factors such as over-programming, fund release problems, project implementation problems etc.; and (vi) Unlike other sectors, however, the air transport sector has been consistently mobilizing substantial resources, which in recent years have generally exceeded its expenditures. However, it should be noted that these expenditures do not include debt servicing charges for past investments which have been borne directly by the government up to now, and which the CAAN is now expected to take over. 3.90 Achievements. Recent policy changes and reforms have led to some notable improvement in the air transport sector: * There is now an extensive but rudimentary network of airports supplementing the road transport system - 44 airports, which include TIA as the central hub, five other regional airports and STOL and helipads. * Following the opening up of domestic air services in 1992, air services to remote areas have seen considerable growth. During the Eighth Plan there was a substantial growth in aircraft fleet strength by 70% and passenger traffic by 57%. Also, passenger traffic movement through airports in the Terai has increased by over 100% since FY94. * The reliability and efficiency of domestic air services has improved considerably. There are now over 10 domestic airlines, with two providing international services. Effectively the monopoly of the Royal Nepal Airlines Corporation (RNAC) over domestic air services has been broken. * The institutional framework and arrangements for regulation and supervision of the sector have been strengthened with the creation of CAAN. 3.91 Key Issues. Notwithstanding these improvements, significant problems remain to be addressed in order to improve the efficiency and effectiveness of public spending in the air transport sector: * Despite substantial growth, the condition of most airfields remain rudimentary. The majority of the small airports including the 25 STOL airfields have only a minimal level of services and lack basic physical infrastructure and navigation aids; and safety standard are typically low. As a result of their rapid growth and inadequate maintenance, the sustainability of airport facilities have increasing become a major concern. * Several private airplanes crashes in the last two years underline the critical importance of effective implementation of safety standards on a regular basis by CAAN. The failure to do so would have a serious impact on the fledging industry, as most of the private airlines would not be able to cope financially with a recurrence of such mishaps. -62 - While the establishment of CAAN as government-owned, but autonomous commercial entity is an important step forward, the financial viability of CAAN is a serious issue, particularly if it has to build, operate and maintain unprofitable regional airports. Presently all regional airports and small airfields are operating at a loss. CAAN also currently lacks the appropriate financial, management and supervisory skills to effectively and profitably deliver a high level of airport and regulatory services to the industry. * The government has also made strong efforts to improve international air services into and out of Nepal which are critical to promoting tourism development. But the effectiveness of this policy has been hampered by the continuing mismanagement and inefficiency of RNAC - the national flag carrier. RNAC continues to make substantial operating losses when its debt service payment and other financial obligations are included (as they properly should be); and is often cited as one of the major examples of mismanagement and corruption among public enterprises in Nepal. Moreover the inefficiencies of RNAC and the continued government protection provided to it have contributed to the poor quality and irregularity of its international air services, adversely affected the growth of tourism and hampered the entry of private operators into, and the evolution of a more competitive environment in, international air services. J. POWER 3.92 Introduction. In terns of longer term development, hydropower is one of the most important potential sources of growth in the Nepalese economy. As discussed below, the rationale for continued heavy public investment in the power sector is weak both because of efficiency and equity considerations and the need for channeling scarce public resources to those areas which have higher social benefits. The major issues in the sector include limiting public spending on power while increasing the efficiency of such investments, and catalyzing private investment to facilitate future development of the sector in a sustainable manner. 3.93 Until recently power development in Nepal has been undertaken primarily by the public sector - mainly through the Nepal Electricity Table 3.20: Power - Composition of Capital Investment, Authority (NEA). This FY96- FY99 policy was modified in the (Rs. Million) early nineties, with the adoption of a revised hydropower development Generation 772.6 1080.0 4600.3 4932.7 5060.5 policy which sought to (33%) (29%) (60%) (79%) (62%) increase the participation Transmission 155.3 391.6 334.2 185.2 2207.1 of the private sector (by (7%) (11%) (4%) (3%) (27%) opening up the power Distribution 720.6 1862.3 1894.2 641.5 330.2 opening up the power ~~~~~~(30%) (50%) (250%) (10%) (4%) sector to local and foreign Miscellaneous 704.4 392.4 858.6 462.6 609.5 private investment, and (30%) (10%) (11%) (8%) (7%) bringing about changes in Total 2352.9 3726.4 7687.3 6222.0 8207,3 electricity legislation), (100%) (100%) (100%) (100%) (100%)i while making public source: Nepal Electricity Authority. institutions operating in - 63 - the sector efficient and creditworthy. This new policy and regulatory framework have helped to attract significant private (including foreign) investment; two medium-sized projects have already been concluded so far. In mid-1996, the 60 MW Khimti Khola project reached financial closure after four years of preparatory work; and in early 1998, agreements for the 36 MW Upper Bhote Khoshi power plant were signed. Nevertheless, public spending still continues to be dominant in the power sector, with the Nepal Electricity Authority (NEA) accounting for the bulk of such spending. Other public expenditure in the sector is insignificant and mainly comprises studies for various projects and some support for isolated small hydro schemes (Table 3.20). 3.94 Level and composition of capital expenditure. Annual capital expenditure on power has grown from Rs. 2.4 billion in FY95 to Rs. 8.3 billion in FY99. Table 3.16 shows that most of the funds were spent on hydro generation projects. The large increase in generation in the last few years mainly reflects expenditures on the Kali Gandaki A project, which is expected to be completed by mid 2001. Expenditures on transmission have been minimal in recent years, but are expected to increase in order to bring transmission capacity into line with increased generation. 3.95 Most of NEA's capital expenditure program over the last five years-on average about 50% and as high as 70% in some years-has been funded from grants and concessional loans from bi-lateral and multilateral sources, passed on either as equity or loans to NEA (Table 3.21). NEA does not receive direct operating subsidies from HMGN. Conversely NEA's self financing of its capital expenditures has been about 26% over the past five years. Funds provided by foreign donors either as grant or loan to HMGN are onlent to NEA as local currency loans with an average interest of 10.25% p.a., while the government bears the foreign exchange risk. Table 3.21: Financing of Power Sector Investment, FY95 - FY99 (In Rupees Million) FY95 FY9 FY97 FY98 FY9 Foreign Grants and Loans 1077 46% 1729 46% 2659 35% 3649 59% 5697 69% Govemment Equity 1326 56% 1109 30% 1721 22% 1371 22% 552 7% NEA Contribution (incl. bond sales) (50) -2% 888 24% 3307 43% 1202 19% 1958 24% Total Investment 2353 1 100% 3726 100% 7687 100% 6222 1 100% 8207 100% Source: Nepal Electricity Authority. 3.96 Technical Efficiency. Despite the relatively high public investment in power so far, domestic power supplies are still insufficient to meet local demand. Only 260 MW out of a potential estimated hydropower generating capacity of 43000MW has been developed so far; and total installed capacity including other forms of generation is only 319 MW. (The Eighth Plan period target of 347MW of additional generating capacity was not met). Per capita electricity consumption is only about 42 kWh, among the lowest in the world, and a large proportion of the population remains without access to electricity. The quality of supply is poor; and dry season generation capacity is inadequate. Shortages are met through load-shedding, especially during the evening peak demand periods, and through imports. At present, Nepal is a net importer of power from India. In 1998, purchases from India stood at roughly 210 GWh (about 15% of total - 64 - available energy); provisional estimates for 1999 show an increase in imports from India by about 10% to 232 GWh. However the supply situation is expected to change over the next few years, with the commissioning of seven (7) projects (including more than 100 MW being built by private investors) which will virtually double the installed capacity of the system in the next three years. Indeed, when these new projects come on stream, (particularly Kali Gandaki - A 144 MW in 2001-02), a short-term power surplus is likely to emerge. But, a recent NEA report shows that, based on the expected growth in demand, the power deficit will again re-emerge in 2003-04 and worsen beyond that unless new capacity is added shortly thereafter. Moreover, the peaking power shortage situation during winter months is not expected to be fully resolved. 3.97 NEA's investment program is guided by system planning studies .vhich are regularly updated by NEA staff with the help of consultants. Some of the imbalances in the transmission system have been recently corrected and the capacity to trade power with India has been increased to about 30 MW. However, investments in the distribution system rehabilitation and expansion have not been adequate, and transmission and distribution losses still average about 25%. In part, this has been due to NEA's weak financial position and its managerial inefficiencies which have hampered the effective implementation of its investment and loss reduction programs. Bill collection from government agencies in particular remains problematic. With a ratio of about 80 customers to staff, NEA is still considered rather heavily overstaffed; and its accounting and billing systems leave much to be desired. 3.98 Equity Effects. The main consumers of electricity in Nepal are households. In 1998, households accounted for 91% of electricity consumption, while transport (3%), industry (3%) and other sectors such as commercial establishments and agriculture (3%) accounted for the remaining 9%. However, only about 13% of the population country-wide has access to electricity. In rural areas, where the bulk of the population (nearly 90%) live, the ratio is only 5%, and implicitly the coverage in urban areas is around 85%. Thus, the coverage of services at present is heavily biased in favor of the urban population, which is also significantly better off in terms of per capita incomes and living standards as compared to their rural counterparts. (According to the 1996 NLSS, the poverty ratio was only 23% in urban areas, compared to 44% in rural areas). Accordingly, from an equity perspective, the commitment of a significant share of limited public resources to power generation makes little sense, particularly where there are competing claims for such funds for other socially more productive purposes. Table 3.22: NEA's Financial Summary, FY95 - FY99 Electricity Sales (GWh) 825 937 1,011 1,057 1113 Ave. Tariff (Rs/kWh) 3.98 4.15 4.96 4.96 4.98 Total Op. Revenue (Rs. Million) 3,464 4,012 5,084 5,433 5,770 Self Financing Ratio 26.3% 12.5% 40.6% 22.4% 10.1% ROR 1.5% 1.8% 1.7% 0.6% 0.3% Debt Service Coverage Ratio 1.3 2.7 2.0 1.7 1.5 Source: Nepal Electricity Authority. 3.99 Financial Management. A series of tariff increase (60% in November 1991, 25% in March 1993, 38% in March 1994 and 20% in June 1996) enabled NEA to substantially improve - 65 - its financial position by the mid nineties. Since then, however, NEA's finances have weakened due to several factors: (i) The government did not allow any tariff adjustments until November 1999 when it approved a 28% increase. (ii) NEA's system losses remains high at 25% - a major source of inefficiency and financial weakness. (iii) Increasing costs, partly due to the devaluation of the Rupee, and (iv) a rapidly growing generation investment program have contributed to financing needs. As a result, NEA's rate of return on its assets which had improved to 1.8% in FY96 has fallen steadily to 0.3% in FY99 - which is an unacceptably low level (Table 2.18). Although NEA is able to fully service its long term debt and meet its royalty payments for water and income tax liabilities to the government, its contribution to its rising investment program (as measured by its self-financing ratio) has been inadequate (Table 3.22). 3.100 As noted, most of NEA's capital expenditure are financed from on-lending by the government of loans and grants provided by bilateral and multilateral donors. In the mid-nineties a large proportion of NEA's capital expenditures was provided as Government equity, but as shown in table 3.23, this proportion has been declining. Because NEA, apart from its debt service, is also paying royalties for the use of water and income taxes, the net effect of NEA's activities on the budget is marginally positive. The situation is expected to improve in future years with increased sales volume and improved efficiency. Table 3.23: Net Impact on the Government Budget, FY95 - FY99 (Rs. Million) FY95 FY96 FY97 FY8 Y9 Cash Outflow from the Budget Equity contribution - Local 1,326 1,109 1,721 1371 552 Debt serv. to Donors - Principal 591 196 156 202 331 Debt serv. to Donors - Interest 54 86 237 303 312 Subtotal 1971 1391 2,114 1,876 1,195 Cash inflow from NEA Debt serv. from NEA - Principal 591 196 156 202 331 Debt serv. from NEA - Interest 659 1,362 933 1,208 1,192 Royalty 0 260 380 393 403 IncomeTax 37 225 294 115 81 Subtotal 1,287 2043 1,763 1,920 2,009 Net Impact on Budget (684) 652 (351) 44 814 Source: Ministry of Finance and Nepal Electricity Authority. Key issues 3.101 The current average power tariff in Nepal (Rupees 5.9 or about 8.7 US cents/kWh) is one of the highest in South Asia. For example, the average tariff in Nepal is more than double that in India, although a part of this difference reflects the subsidized electricity provided to agriculture and residential consumers in India. Another important reason is the relatively high cost of power in Nepal, mainly attributable to the nature of its largely hydro - based (about 82%) generation mix, which is inherently more capital intensive than thermal generation. For example, in Nepal hydro plants cost in the range of $2000-$3000/kilowatt (kW), as compared to $400-$700/kW for thermal generation based on combustion turbines. However, a combination of other institutional, technical and economic factors also contribute to the high cost of power in Nepal including: - 66 - small size of generation units (which makes the exploitation of scale economies of hydro plants difficult); lack of interconnections; a mismatch between consumption and what is produced, a high degree of spillage; and last but not least, inefficiencies in the National Electricity Authority (NEA) system, which result in high transmission and distribution losses (of around 25% on average, and as high as 50% in some areas). 3.102 HMGN has made the development of Nepal's hydro electric potential to serve the energy needs of its people as one of its key developmental objectives. For example, The Ninth Plan seeks to: (a) produce sufficient hydro-electricity at cheaper cost to meet the demand for agriculture, industry, transport, domestic, commercial and miscellaneous sectors at an affordable price and to export power at a competitive price; (b) link up rural electrification with rural economic activities to support the growth of the rural economy; reduce rural-urban disparity in electricity supply as well as to maintain a regional balance; and (c) pay special attention to the environmental aspects of hydro-electricity development and to develop hydro-electricity to reduce adverse effects on the environment to a minimum level. However, to achieve these objectives, some key issues need to be addressed: (i) The investment needs of developing Nepal's power sector are large and appear well beyond the financial capacity of both NEA and HMGN. For example, even on the basis of conservative estimates of projected domestic demand for electricity (7.8% p.a.), generation investments for meeting domestic demand alone are estimated at about US$830 million over the next 10 years. Similarly, transmission investments are estimated at US$940 million, i.e. a total of US$1.77 billion over the next 10 years, without counting the funding requirements of export-oriented projects. By way of comparison, NEA's capital expenditures over the FY91-98 period totaled US$730 million (the bulk of which was financed by external loans and grants and government equity); and it seems doubtful whether, even under optimistic assumptions with regard to improved self-financing by NEA and assistance from bilateral and multilateral donors, such assistance could finance more than a fraction of future resource needs for power development. Moreover, as noted above, it is difficult to justify the commitment of limited aid and public resources to the power sector on equity grounds, given the competing social needs in other areas of the economy. Increasingly therefore investment resources for power development, including particularly for export-oriented projects, will need to be mobilized through the private sector. Similarly, public enterprises such as NEA will also have to rely increasingly on internal resources and private capital markets, which in turn require NEA to remain creditworthy. (ii) A second challenge facing Nepal is to develop its power potential in a least cost way that makes power affordable to domestic users and allows its exports to remain competitive. As noted, the high cost of power in Nepal can be attributed not only to institutional factors such as inefficiencies in the NEA system, but also to Nepal's inability to mobilize investments for relatively large export-oriented projects which could benefit from economies of scale. A policy to reduce costs should be oriented towards an institutional structure that promotes efficiency, creditworthiness and trade since ultimately expanded trade, and increased efficiency would help lower costs of production. In this context, while the government's development objectives for the power sector as outlined in the Ninth Plan are commendable, further fundamental changes in policies, regulations and institutions which govern and operate in the power sector are necessary if the -67 - government's stated objectives are to be met. The priorities for reform should include, among others, the finalization of the national water resources strategy and new hydropower policy; rationalization of the roles of policy-making bodies and the power system planning process; improving the framework for attracting private investment including the introduction of competitive bidding for new generation; improving licensing procedures and the regulatory framework; creating a Power Development Fund to catalyze private investment; increasing the autonomy of NEA; and creating a consensus and ownership for restructuring of NEA. (iii) It is also important that the financial implications of private financing of power development are assessed carefully. The government has been providing certain guarantees, including converting Rupee payments made by NEA into foreign exchange, to some private IPPs. This could lead to sizable contingent liabilities for the government and therefore, the sustainability of developing the power sector under such arrangements with private investors will need further review. I Chapter IMPROVING PUBLIC RESOURCE MANAGEMENT: 4 THE REFORM AGENDA A. NEPAL'S DEVELOPMENT STRATEGY 4.1 Introduction. Nepal is at a major cross roads in its economic and political development. Poverty is widespread; and pervasive public perceptions of "failed development" have led to considerable disenchantment with, and a crisis of confidence among the public, particularly among the rural communities, in development policies of the government, politicians and the Kathmandu/urban elites. There are also new demands from rural communities for greater involvement and participation in decision-making and access to public resources. These pressures are to some extent an inevitable corollary of a social awakening unleashed by the democratization process. Unless effectively addressed, they also constitute a potential threat to the political, social and economic fabric and even to the democratic process. 4.2 This chapter discusses how Nepal's public expenditure policies/programs can be restructured to meet this challenge. Its main message is that, in Nepal the government has an important role to play in fostering economic and social development, but this requires a major reorientation of its present role. As discussed in Chapter 3, from a development perspective, the government in the past has committed important errors of both commission and ommission. It has involved itself directly and indirectly in producing goods and services which can be better done by others (for example, owning and managing banks and the airline, among others), while failing to provide effectively basic social goods and services which cannot be left to others (such as basic education of an acceptable quality, basic health and preventive care, promoting awareness of behaviour that will lead to improvements in health and nutritional status etc). Such a reorientation of its role will help the government to focus its attention on core priority areas that would maximize social benefits, help accelerate economic growth and increase economic opportunities for the poor through such growth. The rest of this chapter briefly reviews Nepal's development strategy and the Ninth Plan's priorities and its realism, and then discusses the key technical reforms which are necessary to make public spending more effective, including reorinting the role of the government. Finally, the chapter briefly reviews the implications of such reforms for external assistance. 4.3 Nepal's development strategy and the role of public resource management. The reduction of widespread poverty is undoubtedly Nepal's central development challenge. This is fully recognized by all the key actors involved in Nepal's development process - the government, beneficiaries, NGOs and INGOS, civil society and external donors. There is also broad agreement on the key elements of the development strategy for reducing poverty: (i) Rapid economic growth is the key to poverty reduction in Nepal. It also needs to be more balanced and equitable so that it will also benefit the rural economy where the poor predominantly live, and not just the modern urban sector of the economy; (ii) Increased and more equitable access to basic services, such as social services, drinking water, basic infrastructure, etc., which help to improve living standards as well as opportunities for income generation and human development; and (iii) special interventions and programs to reach the most poor and vulnerable groups and those living in the backward regions of the country. There is also agreement on the - 70 - broad sectoral implications of this strategy. Agricultural and rural development provides significant scope for increasing rural incomes in the near term; but to achieve high levels of growth and development on a sustainable basis, Nepal also needs to develop its considerable potential in water resources and tourism, its as yet underdeveloped and uneducated human resources, and its services and exports sectors. It is also recognized that the effective implementation of this development strategy will require the nurturing and increased involvement of a nascent private sector, which will need to be the engine of growth; a pro-active government which would primarily act as a regulator and facilitator; an efficient and well- functioning civil service and public insitutions; and a strong commitment to pursuing sound economic policies and refonns which are conducive to enhancing efficiency and creating a competitive environment for both the public and private sectors. In this context, public resource management is perhaps the most important instrument available to the government to implement its overall development strategy, keeping in mind the limited scope for public interventions in Nepal (see below). 4.4 The Ninth Plan. This strategy has been broadly articulated in the government's Ninth Plan (covering the FY98-FY2000 period) which emphasizes that eradication of Nepal's widespread poverty is its primary development objective. The Ninth Plan recognizes the need to accelerate economic growth (from an average of 4.5% p.a. in the past decade to 6.0% p.a. over the Plan period and to over 7% in the following decade and a half) and slow down population growth (2.5% p.a. currently) to a rate consistent with reducing total fertility to the replacement level by FY 2017. The Plan also stresses the need for increasing employment opportunities; accelerating social development and improving access to basic infrastructure; while reducing regional, gender and ethnic disparities. To accelerate economic growth, the Plan aims at harnessing growth potential in agriculture, forestry, hydropower and tourism; increasing investments in infrastructure and human resource development; and promoting industrialization through expanding trade and investment oportunities with neighboring countries. The Plan highlights the enhanced role of the private sector in economic development, the need for a complementary role for the public sector as a facilitator and provider of social and economic infrastructure, and the importance of liberal and market-oriented economic policies. 4.5 Although its broad objectives are appropriate, the Plan suffers from a number of weaknesses: (i) Its key targets are ambitious, unless insitutional impediments which have hampered progress in the past can be effectively addressed. For example, the Plan envisages eliminating illiteracy (currently 52 percent for those over 15), reducing the child mortality rate (currently 118 per thousand) to 63 per thousand, and increasing average life expectancy by 12 years - all by FY 2017; (ii) Its macro-economic framework (which calls for rapid improvements in national savings, revenue mobilzation and aid disbursements) is unlikely to be fully realized; also, (iii) The Plan tries to do too many things in many sectors without recognzing resource limitations, and is therefore not sufficiently prioritized; (iv) The Plan also lacks a concrete implementation strategy to achieve its stated policies and goals, for example, withdrawal of the government from sectors where the private sector could be more effective (such as banking, power, telecommunications and the airline); or to remedy the dysfunctional system governing the allocation of public resources. Thus, this Plan runs the same risks as past plans which have lacked a proactive govenrnent commitment to their implementation. - 71 - B. THE PUBLIC RESOURCE MANAGEMENT REFORM AGENDA1 4.6 Alternative scenarios (summarized in Annex 2) show that under more realistic assumptions, resource availabilities and sustainable levels of public and private investment would be significantly less than anticipated in The Ninth Plan. Nevertheless, by adopting appropriate policies and economic reforms, it would still be possible for Nepal to achieve economic growth rates of the order of around 5.0 % p.a. or more and improve the pattern of its economic growth, while making modest improvements in national savings and the balance of payments. Improvements in public resource management are an important part of such reforms. 4.7 Given the past record of public resource management, achieving such development results will be a major challenge. It will require major improvements in the budget planning, resource allocation and expenditure management process to ensure that limited public resources are channeled to, and are effectively utilized in those areas which can have a significant impact on poverty and development. This report also takes the view that traditional budget reforms, as important as they are, alone will not be sufficient to improve the effectiveness of public resource management in Nepal at this time. To make such reforms work, several other important changes in the political environment, behavior and attitudes and institutional capacity will be needed. The more important of these include: (i) Improved governance and transparency. Traditional fiscal reforms will have little impact by themselves unless the political environment in which they operate improves significantly and there is genuine commitment to development and consensus among major political parties about key development priorities and the reform agenda; (ii) There needs to be greater local ownership of and participation in the public resource management process. For development projects and programs to work better, they need to be owned by the Nepalese themselves; and there needs to be greater involvement of beneficiaries at the local level in the design and implementation of development programs, together with greater transparency and better use of resources at the local level. (iii) Strengthening institutional capacity and decentralization would be necessary to ensure that projects and programs are efficiently implemented and there is greater accountability for spending decisions. The rest of this chapter discusses these key issues. It begins with a discussion of the role of the public sector in the economy against the backdrop of fiscal constraints, and then reviews the governance, ownership and institutional development issues. It concludes with a discussion of technical improvements that would be needed in the budget planning and expenditure management process and how budget allocations could be improved to support the poverty reduction and economic development objectives. It needs to be emphasized that in all of this, it is the Nepalese themselves who will need to make up their minds about the public sector reform agenda, its specific content and its phasing and timing. It is hoped that this report would help clarify issues in this regard, and help develop an appropriate reform agenda. 4.8 Redefining the role of the public sector. Efforts to improve public resource management in Nepal should begin with a re-examination of the role of the public sector in the economy. This is still a matter of considerable debate and ambivalence in Nepal. Some argue for continued broad government involvement in the economy on various grounds: for stabilizing prices; directly undertaking investments over a broad range of areas, given the considerable i This chapter has benefited from collaborative work undertaken recently by DFID -- 'Ynstitutional Issues in Public Expenditure Management in Nepal" Andrew Lawson, et. al, June 1999. -72 - unmet social and infrastructure needs in the country; market failures and the need to protect consumers, especially the large numbers of the poor; and so on. These are also used as arguments for government support for public enterprises, and the continued involvement of the latter in the provision of goods and services, often under monopolistic arrangements. 4.9 While some of these arguments have some validity, they need to be re-examined on the basis of ground realities in Nepal and the accumulated evidence of international experience in other countries. The most compelling reason for doing so is the fact that, (i) as noted in Annex 2 below, the government's resource position for financing development programs will improve only modestly over the next few years, even under optimistic assumptions regarding revenue mobilization. This will severely limit what the state can do in terms of public spending. (ii) Secondly, the past record of the public sector in terms of providing goods and services to the public on a competitive and efficient basis has been particularly poor. While there are many reasons for this poor performance, institutional weaknesses, ineffective leadership, lack of timely decision-making and poor management have been among the major factors. Unless and until these deficiencies are effectively addressed, there is little assurance that more public spending will necessarily lead to better performance. (iii) Third, the experience of many countries, including those in South Asia, clearly demonstrates that less direct government involvement, rather than more, is compatible with and conducive to faster economic growth and better provision of goods and services to consumers.2 Countries which rely on market forces and effectively harness the creative energies of their private sectors and civil society through open and liberal economic policies have a much stronger record of performance than those who do not. 4.10 This report takes the view that the government will need to focus its limited resources on key priority areas where its limited resources can be put to most effective use and leave the rest as far as possible to the private sector, as well as sub-national authorities and other stakeholders. From this broad proposition, some important conclusions can be drawn for the public sector's role and the allocation of public resources: (i) Wherever the private sector can undertake economic activities and/or deliver goods and services competitively, there is little justification for public involvement in such areas and activities, unless there are compelling reasons to justify public interventions on an exceptional basis; (ii) Public involvement should be limited to these areas and activities which the private sector will not find attractive, i.e. where social benefits exceed private costs, but are not adequately factored in by private providers. (iii) In terms of resource allocation, on a broad sectoral basis, public resources should be channeled to such areas as human resource development (education, health, drinking water), rural infrastructure, farm-to-market roads and irrigation priorities in the APP, poverty alleviation programs, and backward area development, among others; but areas such as telecommunications, power generation and most of tourism should be left to the private sector. (iv) Even within the priority areas identified above for public investment, there is considerable scope for private involvement within specific sub-sectors, implying that public investment should focus mainly on other activities within these sub-sectors. While these allocation choices are discussed in greater detail below, some notable examples of sub-sectors where the private sector can effectively take the lead can be pointed out: fertilizer and seed supply and distribution (within agriculture), higher education (in education), urban hospitals (in health) and air transport (in civil aviation). (v) There is also little justification for continued government involvement in most public 2See World Development Report 1997. - 73 - enterprises, particularly where these involve commercial activities and need to be run on commercial lines. Where there might be justifiable social reasons for public involvement in some enterprises, the social costs of such involvement should be quantified and separated from enterprises' losses, instead of the present practice of justifying all government involvement in such enterprises and their losses on the basis of social imperatives. Also, alternative ways of meeting these social needs should be examined closely. Thus, the rationale for government involvement in public enterprises need to be established on a case by case basis, if the current hemorrhaging of scarce public resources through PE losses is to be stopped. (vi) Finally, even where the government needs to remain involved, alternative ways of carrying out these responsibilities and delivering outputs and services more efficiently to the public need to be examined. The role of the private sector in all of this is, of course, premised on the government playing an effective role as regulator, setting rules and enforcing them. 4.11 Ensuring good governance and transparency. This is the most critical factor which can have a relatively quick and decisive impact on improving the effectiveness of the public expenditure program in Nepal over the short to medium term. Given the erosion of public confidence in the political and development process, it would also need to be the central element of any government strategy to combat threats to the democratic process from within. 4.12 For improving public resource management, the governance issue needs to be interpreted broadly. While (i) controlling corruption and leakages and misuse of public resources for personal gain is an important element, (ii) good governance also needs to include broader political commitment to development not only by the government but also by all the major political parties, civil service and other public sector institutions, the private sector and civil society at large. Indeed, to be effective, good governance needs to be an essentially domestically-driven and owned agenda. While external donors can play a role in supporting such initiatives, the leadership clearly needs to come from within. This section of the report does not seek to put together an anti-corruption agenda for Nepal which is clearly lacking upto now; but it points out some key areas where improvements are needed from a public resource management perspective. 4.13 Role of political parties and government leadership. The key message that comes out of a review of public expenditure management in Nepal over the past few years is that, for any technical solutions to work effectively, the behavior of political leadership and political parties needs to change significantly. Indeed, there is considerable political involvement at virtually all levels of the public resource management process. This involvement however has had a largely negative impact so far. The key challenge is how to ensure constructive involvement of key political parties and other constituencies to facilitate the development process and public resource management in particular. There is an urgent need for a national consensus among major political parties on key development issues and solutions and how to go about implementing them. There is already broad agreement among political parties about Nepal's longer term development objectives and strategy and even on the Ninth Plan itself. But when it comes to implementing key reforms which are essential to operationalize this development strategy, a consensus is sadly lacking. Political parties who strongly support such reforms when they are in power, denounce and oppose them when they are in the opposition, for example, with regard to VAT, privatization, price adjustments etc., The recent nation-wide campaign by opposition -74 - parties to roll back recent price adjustments for power and petroleum products is a case in point. Such politicization is a main reason why economic reforms and the development process have stalled in Nepal in the last few years; and without changing this behavior pattern, it would be difficult for any party in power to implement the reforms that are needed to improve public resource management. * Engaging political leadership and political parties early on in the budget decision-making process. At present, the involvement of political leadership in making key decisions with regard to the size of the budget, its expenditure priorities and portfolio decisions is superficial and comes very late in the budget preparation cycle. There is little understanding about the realities of the resource situation and the trade-offs that have to be made. This is a major reason for over-programming of the development budget. In order to introduce more realistic budgets, focus limited resources on highest priority areas and streamline the project portfolio, as well as to adopt necessary measures to mobilize additional revenues all of which will require difficult political decisions, the early involvement of the political leadership (including the Cabinet as well as opposition leaders) will be essential. * Minimize political interference in the public resource management process, for example, in staff transfers and procurement decisions. This will require strict implementation of recently amended civil service rules which mandate civil servants to stay in a post for a minimum specified period before they can be transferred, as well as of recently promulgated procurement guidelines, respectively. * Re-channel the energies of politicians from advocacy for new projects to greater involvement in monitoring and implementation of those already under execution in their particular constituencies. At present, the involvement of politicians in monitoring implementation of projects is minimal, in part because the monitoring process largely does not exist within the government itself. A re-invigorated monitoring system will need to involve the political leadership; but it can work only if there is a consensus on development issues and priorities and rethinking on how political parties can support the development process. 4.14 Controlling corruption and leakages. As in many countries, corruption is increasingly a pervasive problem in Nepal at many levels. It involves elements of political leadership and parties, civil service, local governments and public sector institutions, the private sector and members of civil society, and also external suppliers as bribe givers. Predictably, the solutions to the problem are not easy; and they need to be tailored to country-specific circumstances. In addressing the issue, the Government undoubtedly needs to take the lead; but it is ultimately the responsibility of the Nepalese civil society including major political parties, authorities at different levels, the private sector and NGOs. In Nepal's case, any efforts to address the issue will need to incorporate the following elements, among others: * To be effective, any anti-corruption program needs strong political commitment from the top to its implementation. There needs to be unequivocal signals that corrupt practices, big or small, will not be sanctioned. * To be credible, such a program will need to start at the top of the political ladder. The political leadership will need to set the example to the rest of the country and civil society. Key actions could include declarations of personal (and corporate) wealth and accounting for -75 - notable gains made in relation to previous declarations; settirg up standards/codes of behavior for politicians, with transgressions punishable by Parliamentary or party discipline and/or appropriate legal action; public accountability for decisions (including cabinet decisions) which benefit individuals or party interests etc. * Strengthening the legal and judicial machinery so as provide effective sanctions against corruption. For example, the CIAA is woefully weak and under-staffed. It does not have the resources or the authority and the standing to conduct investigations as needed, or to ensure compliance by government agencies (for example, the Cabinet Secretariat) with its requests for information. Possible solutions, among others, could include: strengthening CIAA (by providing it with more resources and independence, for example by ensuring that it is headed by a reputable independent person and reports to the Parliamnent rather than to the Executive) and Auditor General's office; revising existing legislation to make it mandatory for government agencies and public office holders to comply with CIAA's requests for information; better monitoring and accounting of public expenditure, together with greater Parliamentary oversight; and special procedures and tribunals to deal expeditiously with anti- corruption cases, given the dilatory legal procedures in the country. * Limiting the motives and opportunities for corruption. This could take several forms for example, limiting the scope for politicians' involvement in staff transfers and procurement activities by enforcing new civil service rules and procurement guidelines; streamlining/downsizing the civil service and providing competitive wage levels to the extent possible to minimize incentives for corruption; reforming tax laws and simplifying tax rates to reduce discretionary powers of tax officials and to minimize contact between tax payers and tax officials, and so on. - Involving civil society -- the press, NGOs beneficiary groups, local communities etc -- in the anti-corruption campaign. As noted earlier, greater transparency and public accountability already are having some positive effects at the local levels. 4.15 Promoting ownership and participation. As discussed, a major factor contributing to the poor expenditure performance of (especially donor-financed) projects' performance is the lack of ownership of projects and programs by major stakeholders - notably the government, as well as potential local beneficiaries. Many projects are seen as being donor-driven, donor- financed and donor-executed under special implementation arrangements. In the case of donor- assisted projects (which comprise roughly two-thirds of the development budget), changing this mindset seems to be necessary to improve their implementation performance. 4.16 The surest way of promoting local ownership is for the government to take the lead in designing, financing, and implementing the development program itself. The government will need to decide what its own development priorities and programs are, and ask the donors to support such programs. Where donor aid does not fit its priorities and program objectives, the Government will need to be able to reject such donor-driven aid. This will also require, apart from building up its institutional capacity, greater degree of self financing/self reliance through stronger efforts to mobilize revenues over time. But this might not happen quickly; and at least over the medium term, the development budget will continue to be significantly dependent on donor-financing. The challenge in this regard is how donors can continue to provide financing and still foster a sense of greater ownership by the government. For this to happen, donors will -76 - need to step back and avoid pressurizing the government to take on board new projects. Indeed, a major problem is the fact that donors are competing with each other to find useful outlets for their funding in part because some of the aid programs are expanding; and this translates into supply driven pressures for new projects and technical assistance programs. To foster greater government ownership, donors will need to be more patient, and help the government build up its own capacity for project formulation and preparation. This should include assistance for developing sector policies and priorities, for identifying and designing projects/programs to achieve agreed sectoral goals, as well as for strengthening institutional capacity to implement those projects on a sustainable basis. In other words, to promote government ownership, the role of donors will need to change from doers and implementers to facilitators. Clearly such a strategy will entail risks for both donors and the government. In the short term it is quite likely that project preparation and implementation could take more time as local capacity is built up gradually, the average size of donor projects could become smaller, and the volume of donor assistance may fall. However, the upside is that, if it works, it may help to ensure greater sustainability and effectiveness of Nepal's development efforts. 4.17 Similar efforts are needed in the case of fully government-financed programs also. As noted above, budget formulation in Nepal is largely a "top-down" process, with little involvement of local bodies and beneficiary groups. Moreover, most government programs are centrally administered and managed. To improve effectiveness, it is necessary to create greater synergies between institutions at the center and those at local levels and to build partnerships involving the private sector, NGOs and other stakeholders. This can take place at many levels for example through: (i) greater devolution of responsibilities to line ministries for program formulation (within sectoral guidelines and policies agreed with the National Planning Commission) and monitoring and supervision; (ii) increased involvement of local bodies and beneficiary groups in determining priorities and designing projects to better reflect local needs and implementing such projects; and (iii) developing alternative approaches to (currently centrally managed) project implementation and service delivery. For example, the creation of Boards involving the private sector as an alternative to traditional government mechanisms or contracting out to NGOs specific tasks such as supervision, technical services or program implementation on behalf of community groups are some of the possibilities. This will have to be a gradual process, given the institutional weaknesses and the need for strengthening technical capacity at many levels (as discussed below). For example, it is hard to see how a program budgeting approach could be adopted in Nepal at this time, without first improving the budget formulation and management process and without developing sectoral priorities and strengthening the capacity of line ministries. This is however not a compelling reason for delaying the adoption of new initiatives. Indeed because such changes would take time to bring about results, it is essential that they are introduced as quickly as possible; and as discussed below, their potential benefits in terms of greater relevance of projects to community needs, improved implementation and monitoring, community participation and greater transparency could be considerable. 4.18 Institutional Development. A third critical element which is vital for improving public resource management is strengthening the institutional capacity for carrying out development activities, at all levels of the government. As noted elsewhere, a weak and unmotivated administration, together with over-centralization of development activities and lack of delegation, have been among the major impediments to effective implementation of projects and programs. Institutional capacity remains weak at virtually all levels-at the center, line - 77 - ministries, local bodies, oversight agencies, public enterprises and, as noted, the judiciary. Increased politicization has over time exacerbated these institutional weaknesses. Over the past few years, efforts have been made through technical assistance to strengthen the capacity of institutions at various levels and to promote decentralization; but as noted, progress so far has been limited. This section focuses on two areas: (i) civil service reform; and (ii) decentralization. This report takes the view that decentralization is important in term of getting resources down to local levels, and for building capacity at district and village levels to manage development projects in order to deliver services to the beneficiaries more efficiently; but strengthening the capacity of institutions at the center, line ministries and oversight agencies (particularly for supervision and monitoring) within the context of a revised framework of functions and responsibilities is also essential. Civil Service Reform is important in this context because of its pervasive impact on the functioning of the development administration at all levels. 4.19 Civil Service Reform needs to be structured in the context of both the type of development administration that is needed to deliver public services more effectively and the severe local resource limitations hampering Nepal. First, with increased emphasis on transferring responsibility for development activities to local levels through decentralization, the role of the center, line ministries and supervisory agencies will need to change over time. For example, the NPC may need to be transformed into a planning agency responsible for setting overall development goals and priorities as well as orchestrating sectoral strategies and less concerned with the actual implementation of the development program, while the role of the line ministries themselves well need to change from actual implementers of programs to facilitators, while taking on broader responsibilities for sectoral program formulation, provision of technical support to local bodies, and supervising and monitoring project/program implementation. Similar changes may also be required of the Finance Ministry and its oversight agencies (such as the FCGO). Once an integrated budget is finalized by MOF, its emphasis may need to change from orchestrating project implementation and priorities through the fund release process to providing more regulatory and oversight functions. Such a transition will obviously take considerable time, but will have important implications for skill needs, staffing levels and personnel management which need to be take into account in planing civil service reforms. Second, there are major concerns about current incentive levels, efficiency, relevance and the size of the civil service. To provide adequate incentives, the overall compensation levels may need to increase significantly, together with a widening of salary differentials between lower and higher grades (i.e. decompressing the salary structure) and the introduction of a merit- based, performance-oriented reward/promotion system. This will, however, increase the salary bill which is already large in relation to the government's limited revenue generation and its current expenditures. How to pay for civil service reforms thus will be a major issue. Third, feasible alternatives therefore will have to be considered. Enhancing government revenue through tax reforms is one such option. While this will present its own challenges, it is also important to recognize that increased revenue will also be needed to support essential operations and maintenance (which is heavily underfunded at present) as well as for financing development activities in a number of important areas. Other alternatives therefore will need to be looked at, including downsizing, as well as donor funding for civil service reforms on a relatively short term basis. External assistance however is likely to be forthcoming only if a credible reform program is developed; and over the longer term, a substantial increase in resource mobilization will be needed to finance these expenditures on a sustainable basis. - 78 - 4.20 Decentralization is seen by many stakeholders as an important alternative to the presently highly centralized system for improving service delivery to local communities and more generally for enhancing effectiveness of public spending. As a result, some important initiatives have been taken in recent years, including the formal devolution of greater spending and taxing powers to local bodies, increased budget allocations, and more recently the promulgation of a Local Autonomy Bill. However, as discussed, the results of these initiatives so far have been quite mixed. There are many instances of corruption, misuse and misallocation of resources at the district and village levels as well. But there are also some examples of better use of resources linked to community needs and increased accountability where there has been greater community involvement and transparency in the allocation and use of resources, for example in PDDP-assisted VDCs. 4.21 It is important to recognize that decentralization is not a complete substitute for the role of the central government. Even as more service delivery functions are transferred to local levels, the line ministries and government departments will have an important role to play in facilitating, providing technical support to, and supervising and monitoring the activities of local bodies. It is equally important to note that decentralization is not an easy process, as experience in other countries demonstrate; and that a phased approach is necessary to its implementation. Nevertheless given the ineffectiveness of the centrally-managed system and the considerable potential of decentralization, every effort should be made to make the decentralization process work better and more efficiently. Moreover, since it would take considerable time and effort to develop an efficient management and delivery system at the local levels (though perhaps less than what it would take to reform the present centrally-driven system through civil service reforms etc.), it is essential to try to build up capacity at the local level as quickly as possible. Several things need to be done to achieve this objective: (i) First, it is necessary to remove the political and bureaucratic impediments which currently hamper the effective implementation of decentralization initiatives. There are two main problems in this regard. One is the fact that, although the major political parties are equally committed to decentralization, the divergent self-interests of the two major political parties (the Nepali Congress now controls the central government, while the UML controls the majority of local bodies) have so far prevented a coalescence of commitment to its rapid implementation. Another is bureaucratic reticence to decentralization, often motivated by concerns about the lack of accountability and misuse of resources by some local bodies, as well as a reluctance to relinquish the power, influence and privileges associated with the central control of resources. However, both these will need to change if the decentralization process is to be accelerated. (ii) Second, of utmost importance is the need to improve the technical capacity of local bodies (DDCs and VDCs) for identifying their needs and priorities; resource mapping and planning; project design, preparation and implementation; accounting, record- keeping and expenditure reporting; resource mobilization through local taxation, etc. As noted, some efforts have already been made by the government and external donors to build capacity at district and village levels. But, to adequately equip local bodies to carry out their expected functions and obligations, much more remains to be done. This will require concerted efforts to provide the necessary technical assistance and support in critical areas identified above, but also a redefinition of the role of government departments and line ministries (see below). - 79 - (iii) Currently, substantial resources are already allocated to local bodies, in addition to several parallel programs implemented by central government ministries and departments, which are not effectively co-ordinated. A major effort is required to co- ordinate these central/district programs (for example in health, education, drinking water, rural roads etc., among others) with those presently managed by local bodies in order to avoid duplication and waste of resources. In addition, the needs of deprived and backward areas may require special consideration. (iv) In parallel, clearer definition of responsibilities of both local bodies and line ministries for these programs will also be required. For example, the role of line ministries and departments will have to change from that of implementers to being facilitators, for providing technical advice and support on a continuing basis, and for monitoring auditing and oversight as a second line of defense for ensuring good program management, and quality and accountability at the local level. Until local capacity for implementing these programs is built up, line ministries and departments may also have to continue to carry out some of these functions on behalf of local bodies, while providing technical support. This is being done in some sectors, for example in road transport, where the responsibility for rural roads has been transferred to local bodies, and a Department of Local Infrastructure Development and Agricultural Roads (DOLIDAR) has been set up (under MLD) for providing technical support to local bodies. However for the reasons noted above, these efforts have been half-hearted. It may also be necessary to contract out particular activities (for example, technical support and monitoring, and even service delivery on behalf of local communities) to NGOs and community-based organizations as a way of getting around the capacity constraints of both the line ministries and local bodies. (v) The most effective way of promoting better management and accountability, however, is by enhancing the participation of beneficiary groups in the decision-making and implementation process. As noted elsewhere, greater transparency (through an "open- book system") in regard to the availability of budget funds, the nature of projects being financed and public access to financial records have permitted greater accountability and better use of resources. Indeed, public accountability for the allocation and use of resources is clearly the best safeguard against corruption and leakages and for ensuring that public spending actually leads to provision of outputs and services demanded by the community at not only local but also at central levels. To ensure this, however, the obstacles created by some stakeholder groups who benefit from the status quo will have to be overcome. At the local level, this would call for greater efforts to promote community mobilization and empowerment, particularly of the poorer and under- privileged groups in the community. 4.22 Technical Reforms to Improve Public Resource Management. These behavioral and institutional changes need to be complemented by technical reforms aimed at improving fiscal management. In this context, revenue reforms are potentially an important part of the measures needed to improve public resource management in Nepal. Increased revenue mobilization is needed for several reasons: (i) to reduce the dependence on external aid and enhance self- reliance, which would also help to create greater ownership as discussed above; (ii) for financing anticipated increases in current expenditures. As noted above, 0 & M activities are currently under-funded and, given the inadequacy of revenues, many recurrent activities are included in - 80 - the development budget in order to attract donor funding; and (iii) Increased development spending on the social sectors will over time generate increased demands for operating schools, health centers etc, as will civil service reforms, as noted above. Even if external assistance could be found to support such reforms in the short term, Nepal will need to finance its increased current expenditures out of its own resources over the longer term. However, as noted below, additional revenue mobilization is warranted only if actions are taken to improve the effectiveness of public spending. 4.23 This report focuses mainly on the reforms that are needed on the expenditure side. This is because, (i) as noted throughout this report, there are significant inefficiencies and waste in the use of public resources; and it would be possible to make relatively quick gains from reducing such waste and inefficiencies. (ii) Without making such improvements, it will be hard to justify additional tax increases, since there would be little assurance that increased resources will not be misused or wasted; and (iii) Tax reforms themselves will not be any easier than expenditure reforms in the present context in Nepal, since resistance from interest groups across the political and social spectrum who benefit from the status quo will have to be overcome. To build support for tax reforms, it will be necessary to convince tax payers that the additional revenues will be effectively used. The following paragraphs discuss the needed expenditure reforms in some detail. 4.24 To improve the budget process: The foremost need is to integrate the regular and development budgets. At present, these are prepared as two separate budgets by the MOF and NPC. There is little linkage between the two; and resources are generally allocated on an incremental basis. As a result, the relationship between budget expenditures and expected outcomes in terms of project implementation and service delivery remains weak; 0 & M and other priority activities are often under-funded, while many low priority activities are retained within the development budget. It is now necessary to take a close and integrated look at the budget and its projects/programs, decide what the priorities are in terms of growth, poverty reduction and service delivery objectives, and provide the necessary resources for these activities. 4.25 The key to getting control of the development budget and minimize the waste of resources associated with over-programming is to ensure greater realism in resource forecasting. (i) Projections of revenue, aid inflows and the size of the development budget would need to be in line with recent performance and reasonable expectations. (ii) Such projections, prepared for three years at a time, would provide a medium term framework (for a Three Year Rolling Program) for the development budget. (iii) Such a framework would help to bring out the key trade-offs that would have to be made within the development budget including the need for prioritization and the limited scope for new project starts. This information is critical if the political leadership is to be properly informed about the fiscal constraints and to convince them about the need for making critical choices and deciding priorities. Needless to say that such trade-offs will not become apparent if the resource projections themselves paint an incorrectly rosy picture, as has been the case in recent years. (iv) But, if made realistically, such a framework should help to persuade the political leadership about the need to drop or downsize projects/programs which are not working well in order to bring the development program to a more manageable level from the 681 projects/programs at present. It would also have-the additional advantage that, given the increased realism in budgeting, the availability of resources to line ministries and implementing agencies would become more predictable; and therefore MOF can hold the latter more accountable for expected outputs. (v) In parallel, project - 81 - screening and appraisal capacity in the National Planning Commission, Ministry of Finance and the line ministries would need to be strengthened (by providing adequate staff and budgetary resources) so that greater discipline could be exercised over the process of initiating new projects into the development budget. (vi) The budget process also needs to be made more responsive -- less "top-down" and more "bottom-up" -- in terms of accommodating programs proposed by local level constituents and line ministries within the available resource envelope. For this purpose, the budget preparation cycle needs to be advanced from its present schedule in order to provide more time for such interactions. 4.26 Once formulated, the expenditure program would need to be better managed, without resorting to ad-hoc expenditure cuts which disrupt project implementation. To strengthen expenditure management and control, a number of steps can be taken: (i) The mid year budget review process could be used to reassess resource outlook during the year and to make expenditure adjustments in a more systematic manner. Currently, this review is undertaken too late in the fiscal year to be of any practical usefulness; (ii) To protect priority projects from any resource shortfalls, a core program of priority projects could be formulated and fund releases to projects/programs would need to be based on the core program. If expenditure cuts are needed, fund releases to low priority projects would need to be accordingly reduced; (iii) Expenditure reporting by spending units need to be improved, both in terms of timeliness and quality of reporting. To facilitate such reporting, a uniform budget classification system was introduced in FY97, with technical assistance from DFID, and it needs to be made fully operational; (iv) The capacity for monitoring expenditures as well as physical progress of projects/programs need to be strengthened at various levels of the government in order to ensure reliability of reporting by implementing units, improve accountability and prevent the misuse of resources. For example, the government will need to consider re-instituting the practice of holding periodic progress/implementation reviews by the Prime Minister and the Cabinet, and actively involving parliamentarians in this process. To facilitate and service such a review process, the technical capacity of the National Planning Commission and the line ministries for project monitoring need to be significantly enhanced. This is also needed for the Financial Controller General's office which is responsible for the release of funds and for the Auditor General's office which is responsible for conducting final audits of government accounts and for reporting irregularities to the Parliament. In this context, various initiatives to improve monitoring and accountability have been started recently, including the introduction of Loan Administration Change Initiative (LACI) for IDA - financed projects, and the creation of a District Treasury System by the government. (Under this system check issuing as well as expenditure reporting and monitoring responsibilities would be transferred to a smaller number of Cost Centers, as compared to nearly 3000 paying offices/implementing units at present). Strong government commitment to implementing these initiatives would go a long way in checking irregularities which are widespread at present. Also, the government's accounting system may need to be reviewed in light of the need to link physical progress with financial progress. 4.27 Recent experience of major donors suggests that the key to improving project implementation is to clean up the portfolio, focus attention and resources on a smaller number of projects/programs and to monitor their implementation closely. In addition to measures outlined above, other steps can be taken to address problems which currently hamper implementation: (i) To reduce procurement delays, standard bidding documents have been recently adopted by the Government. This system needs to be made effective by familiarizing government staff with the new procedures through training. Even more importantly, political interference in procurement - 82 - decisions need to be eliminated. Clearly defining the approval limits for procurement decisions that can be taken at various administrative levels together with clarification of procedures to be followed and mandatory punishments for transgressions (through an anti-corruption program) may be needed for this purpose. (ii) To encourage project managers to make project-related decisions and to hold them accountable, an atmosphere conducive to such decision-making needs to be created by reducing insecurities associated with frequent political changes. Continuity of tenure of project managers, senior government officials etc. is essential in this regard. Strict implementation of the recent civil service rules regarding staff transfers would help ensure such continuity. C. IMPROVEMENTS IN BUDGET ALLOCATIONS 4.28 Technical improvements in budget planning and management and procedures should help improve the framework for better allocation and use of resources. But, the real substance of budget reforms involves actual changes in the existing pattern of budget allocations and their utilization to better reflect the government's development objectives and priorities. To bring about such changes, considerable preparatory work by the government would be needed. This report seeks to provide some suggestions about how the budget allocation pattern can be improved: (i) As discussed, the starting point of such an exercise is to prepare (a) a realistic assessment of resources available for the budget, (b) a detailed assessment of resource needs for all ongoing activities in the budget; and (c) Decisions are then required on how to allocate resources, i.e. where they should and should not go. This should be guided by both the perception of the role the public sector should play (discussed in para. 4.10) and the key development objectives (para. 4.3). (ii) Using this criteria, it can be argued that scarce public resources (domestic as well as aid) should not go into areas where they can potentially displace private capital and investment. Some clearest examples include telecommunications, industry, power generation and tourism, where the private costs can be fully captured by appropriately charging private beneficiaries and where the associated social benefits are also likely to be marginal. Similar arguments can also be made for limiting public involvement in most public enterprises, for example, airlines, manufacturing activities, input distribution and services. In some activities, for example providing basic food supplies in remote areas to prevent famines etc., some public involvement can be justified, but not all of the marketing, procurement and price stabilization functions currently performed (rather badly) by the Food Corporation. There are also many sub sectors where public funding should be reduced (see below). In these areas, however, the government still has an important role to play as a facilitator, regulator and protector of public/consumer interests. 4.29 Similarly, there are a number of areas where increased allocation of public resources is warranted: (i) It is necessary to provide adequate resources for operations and maintenance and other recurrent expenditures which are essential to make better use of existing assets/past investments and to provide better delivery of services (where such activities cannot be privatized). This includes a variety of tasks in both the regular and development budgets, such as operating expenses to provide basic government services (such as electricity and - 83 - telephone bills, travel and supervision, adequate per-diems); essential maintenance (for example of roads, irrigation schemes, buildings and vehicles); and basic inputs needed to deliver services, (such as drugs and medicines and equipment). It is to assess these needs realistically that an integrated look at the regular and development budgets is necessary. (ii) In accordance with the development priorities in para 4.3, on a broad sectoral basis, development budget allocations for a number of sectors should be increased: (a) Farm-to- market roads and small-scale surface irrigation which are identified as high priorities in the Agriculture Perspective Plan (APP) for accelerating agricultural growth and reducing rural poverty; human resource developmentlsocial services (such as education, health and drinking water) which are essential for improving living standards and poverty reduction in the short term, as well as social and economic development in the longer term; and rural infrastructure which is necessary for both developing the agricultural potential as well as for revitalizing rural communities through decentralization initiatives. Given the large numbers of the poor who are well below the poverty line, a case can also be made for selective poverty interventions, particularly for those in disadvantaged groups and backward regions. In this context, it may be desirable to give special attention to backward areas in social sector and rural infrastructure programs, as well as through expanded and more effective area development programs, for both social and political reasons. However, given the limited public resources and weak administrative capacity, such interventions will have to be limited in scope and well-targeted. (iii) Within certain sectors, increased resources can be warranted for priority activities, for example, primary education, preventive services in health as well as rural health care, food security and nutrition particularly for vulnerable groups, road maintenance and rehabilitation, improving airfields in more remote areas and improving safety and regulatory standards in civil aviation and so on. In the case of decentralized local development, which already receives considerable resources through a variety of programs, the major need at present is for co-ordination of these programs and better utilization of available resources. Additional resources for these activities can be considered, as greater progress in capacity building is achieved and utilization performance improves. (iv) A strong case can also be made for improving the capacity of institutions at virtually all levels which will require considerable public (and donor) funding over time. Some of the immediate priorities include assistance for strengthening the technical capacity for decentralization efforts at district and village levels; strengthening the judicial and tax administrations (to check corruption and leakages and improve governance); and strengthening the monitoring and supervision capacity of line ministries, NPC, MOF, the Auditor General's Office and the FCGO. Civil Service Reform will demand even greater resources; and, as noted, will require considerable donor support as well political consensus for it. (v) However, not all areas within the priority sectors mentioned above should receive increased public funding, since greater private involvement is feasible and desirable in many activities. For example, in agriculture, subsidies are already being reduced in order to provide a level playing field for promoting private involvement in fertilizer-and seed distribution. The role of the public sector in research and extension needs to be monitored and evaluated; and if the performance of current public sector programs does not improve significantly, privatization of these activities should be considered. - 84 - Similarly in forestry, while more resources would be needed to promote community forestry in the Terai, the role of the Timber Corporation -- a prime source of corruption and leakages as well as mismanagement of forestry resources - needs to be phased out. Similar examples are common in almost every sector, for example in education, where public support for higher education should be phased out, while increasing funding to lower levels both to improve the quality of primary education and to support reforms at secondary levels. D. ROLE OF EXTERNAL ASSISTANCE 4.30 Role of external aid. External donors have played an important role in supporting Nepal's development efforts so far. Although there are questions about the role and effectiveness of aid both among the Nepalese and the donor community, there is little doubt that Nepal will continue to need donor support in the foreseeable future. However, in an environment of increasing competition among developing countries for limited donor aid, the availability of such assistance is likely to depend on country perfornance, particularly the effectiveness of aid utilization, rather than need. 4.31 Despite Nepal's need for its external assistance, a number of criticisms have been leveled against the past pattern of aid. The more important of these, among others, include the following: (i) It is "supply-driven" and "donor-driven." Donor-financed projects are perceived to be often pushed by donors, and sometimes inappropriately designed and unsuited to the country's needs and priorities, and therefore lack local ownership; (ii) Donor projects are often implemented either directly by donors or through special arrangements (such as project implementation units), and undermine domestic institution building; (iii) Technical assistance provided by donors is too expensive and does not sufficiently use available local capacity; moreover, there is little transfer of knowledge and skills. It has therefore not contributed to sustainable institutional development. (iv) Donors often compete with each other and do not provide consistent advice to the government; (v) Donors are "too easy" on the government, thereby encouraging misuse and waste of resources; and (vi) They attach too many conditions which infringe on the country's sovereignty and/or are inappropriate. While some of these criticisms are debatable in the way they are often presented, they represent important concerns which need to be addressed by donors. 4.32 To increase the effectiveness of aid and public spending in Nepal, important behavioral changes will be required from external donors as well as the government. As noted, the government will need to be more assertive and take the lead in, and ownership for, introducing critical fiscal reforms. Correspondingly, the donors' role also will need to change from providers of "supply-driven" or "donor-driven" aid, implementers and doers of projects, and purveyors of expensive technical assistance to facilitators who will provide financial and technical support that fit the country's needs and priorities in way that fosters local ownership, institution-building, and longer term sustainability. This indeed will be a major challenge for donors, and will have important implications for the projects they support and the type of aid they provide. 4.33 The longer term solution to Nepal's complex fiscal management problems would be for the government to take charge of its overall expenditure program, integrate the regular and - 85 - development budgets and prioritize it to truly reflect its development/poverty reduction needs and priorities; and for the donors to support such a program with the additional resources that are needed without imposing their own project and technical assistance priorities. While this is a desirable longer term objective, the transition to it is likely to require a more gradual process. To support such a transition, important behavioral changes will be required from external donors as well as the government. While the government takes the lead in, and ownership for, introducing critical fiscal reforms, donors' role also will need to change from providers of "supply-driven" aid, implementers of projects, and purveyors of expensive technical assistance to facilitators who will provide financial and technical support that fit the country's needs and priorities in a way that fosters local ownership, institution-building, and longer term sustainability. Towards this end, donors could support the government's fiscal reform process in several ways: (i) The government's technical capacity to undertake budget reforms is limited at present; and donor assistance will be needed to help carry out this work (for example, to help the government to formulate a prioritized three year expenditure program, and sectoral strategies and investment programs consistent with those priorities), in the short term, as well as for strengthening institutional capacity at various levels over the longer term. (ii) Donors will need to accept the consequences of expenditure rationalization (for example, downsizing/canceling poor projects, limiting new project starts for some time etc.); and adjust their own aid programs to support only those projects/programs which fit the country's needs. (iii) It will also require stronger efforts among the donors themselves, within the context of agreed sectoral investment programs, to co- ordinate their support without competing with each other. (iv) It may also call for changes in the way donors prepare and manage projects in order to promote greater local ownership of project preparation, implementation and management. For example, in regard to technical assistance, instead of extensive reliance on external consultants, they need to look at alternative more cost- effective ways of providing technical support (for example, using teams of local consultants under the direction and supervision of expatriate advisors who could be hired for short periods of time). Also, technical assistance needs to be carefully tailored to the country's needs, rather than being a mechanism for developing new follow-up projects, as too often is the case now. (v) Similarly, donors need to examine whether quick disbursing cash support is warranted under present circumstances, given the fungibility of such assistance. Such assistance can be potentially useful, but only when it does not substitute for revenue generation by the government and where the public expenditure program is reasonably well prioritized, so that it actually supports such priorities and sound sectoral expenditure programs. Such an approach will entail risks for both the government and the donors, since project preparation could take more time, projects may have to be smaller and more simple, and aid utilization and commitments may decline initially. Even so, such risks are worth taking if they lead to greater ownership, better development results and increased sustainability over the longer term. E. PHASING OF REFORMS 4.34 The foregoing is an imposing agenda, and its implementation will be a challenge to a committed government. The phasing of these reforms is therefore important, and will need to be considered both in terms of practical expediency as well as the urgent need to get on with the reform agenda. As discussed, reforms in governance, institutional strengthening and building ownership are complex and time-consuming; and technical reforms should not wait for their completion. Indeed, technical reforms are the easiest to start with, provided a consensus for - 86 - them can be developed; but, work will need to begin in parallel on other areas, recognizing the interdependence of reforms. Thus, it may be necessary (i) to start a dialogue involving the political leaders and civil society to discuss and develop a consensus on the reform agenda. (ii) Initiate the (FY2001) budget preparation process very early, with the objective of presenting the Budget to the Parliament earlier than usual (say sometime in May), and finalizing procedural requirements for its early implementation (for example, finalization of work programs by line ministries and the delegation of spending authority to them by the Finance Ministry before the beginning of the new fiscal year). This would help to jump start budget implementation which now does not start till the second quarter because of procedural delays. (iii) Prepare realistic estimates of resource availabilities, (recognizing that it is more prudent to be conservative than over-optimistic in this regard), as well as estimates of resource needs of ongoing activities and projects. (iv) This work in turn should enable the early presentation to the Cabinet and political leaders of clear proposals about the size of a feasible budget, resource needs, the trade-offs involved and how to deal with them. Such proposals could be used to emphasize the need to complete projects quickly and to provide better service delivery, and seek guidance on such issues as sectoral priorities, how to dealt with low priority projects, limiting new starts etc. This will not be an easy process, but it would help involve the political leaders in a constructive process of committing themselves to difficult budget decisions early on. (v) Once the budget is finalized, it would need to be backed up by a core program, so that the key priority projects could be protected in case of resource shortfalls. To be operationally useful, the core program needs to be defined in terms of local currency needs of projects/programs, (sine the latter is the major constraint to be addressed), and be limited to the highest priority projects. (vi) Concurrently, work would need to proceed on a Three Year Rolling Expenditure Program (TYREP) covering the FY2001 - 2003 period, which would provide the framework for expenditure reform for the medium term. As a part of such an exercise, the government will also need to work in parallel with donors and other stakeholders on sectoral programs and priorities which could be supported by donors through project/program and technical assistance. Once the TYREP is formulated, the annual budget for FY2001 could be modified (through the mid-year review) to fit within the TYREP framework. 4.35 Conclusion. Improving the effectiveness of public spending and external aid is a major challenge for both Nepal and the donor community. Without such improvements, Nepal cannot achieve its development objectives of reducing poverty and attaining sustainable growth in incomes and living standards in the foreseeable future. A continuation of present practices and performance will perpetuate waste and inefficiency which, in the absence of tangible progress in poverty reduction and development results, may further exacerbate social and political tensions. It will also pose a major dilemma for external donors who recognize Nepal's pressing development needs, but will nevertheless find it difficult to justify increased (or even maintenance of present) aid levels without a significant improvement in performance. This report's view is that with serious commitment to development by the government and political parties, behavioral changes, institutional strengthening and appropriate technical reforms, Nepal can significantly improve development results from its public spending. The donor community can play an important supporting role in this process. In doing so, they need to re- examine how they can better support the government's development priorities, what forms aid should take, and how it can be provided/leveraged to build local ownership and institutional capacity. It is a strategy which is fraught with difficulties and risks for both the government and the donors. But it is a risk worth taking because the upside benefits could be considerable, while the downside consequences of inaction could be very damaging. Annex 1 Page 1 of 2 FISCAL STRUCTURE AND IT'S IMPLICATIONS Is Nepal's fiscal structure weaker than those of comparable low-income countries and her South Asian neighbors? 1. Nepal's fiscal structure is broadly similar to those of comparable low income countries in Graph 1:Total Public Ekenditure/GMP Africa and Asia, but with notable differences in 40 some respects, (Table 1): In 1996, Nepal's 30 expenditure/GDP ratio (18.7% of GDP) was Q2. about one percentage point higher than those of o comparable low-income countries, while her 10 revenue/GDP ratio was similar to the o I i I I Year comparators' average. As a result, Nepal's overall fiscal deficit is also about 1-1½/2 Government Revenue/GDP percentage points larger than those of her 30 comparators. Since its domestic borrowing is 2 20 about the same as the comparators' average, 2 15 Nepal relies on foreign aid far more, (nearly 2% 10 of GDP more), to finance her fiscal deficit, 5 compared to other low-income countries. 0 I i i Year 2. A comparison with other South Asian countries (Graph 1) also shows some important 14 OverallFiscalDeficit/GOP differences: (i) the level of total government 12 spending in Nepal is significantly higher than in 110o- Bangladesh, but lower than those of her higher 8 income neighbors-India, Pakistan and Sri 6 - Lanka. In the case of India, her fiscal data 2 include substantial state level revenues and 0- i I I Year expenditures which are non-existent in Nepal. Also, both Pakistan and Sri Lanka spend the Foreign Aid Financing=GDP equivalent of 5-6% of their GDP on defense, 8 compared to 0.7% in Nepal. Excluding defense, : 6 the ratio of other expenditures to GDP in Nepal 4 is about the same as in Pakistan and Sri Lanka 92 2 -countries with much higher per capita Yew incomes. (ii) On the other hand, Nepal's revenue/GDP ratio, though comparable to Domestic Borrowin/GDP Bangladesh's, is significantly lower than those 15 in other South Asian countries. Pakistan and Sri r1 Lanka, for example, have revenue ratios around - 16-18% of GDP, while Nepal's and 5 Bangladesh's are around 10-11% of GDP. __ _ _ These differences in regard to revenue are not OFY91 FY92 FY93 Y94 FY95 FY96 FY97 FY98 surprising because revenue performance is N a Bw*adesh closely related to income levels; and countries __ta -- Pahstan higher up in the per capita income ladder are SrLanka able to mobilize more domestic resources Annex 1 Page 2 of 2 tosustain higher spending levels. (iii) Given its relatively high expenditure level and its low revenue generation, Nepal's overall fiscal deficit is much larger (by 2-3 percentage points of GDP) than Bangladesh's and Pakistan's and also, (in the last two years) Sri Lanka's. (It should be noted that unlike Nepal, Sri Lanka is fighting a major civil war which has added considerably to its public expenditures and the fiscal deficit). (iv) Nepal is far more dependent on foreign aid than other South Asian countries. In the case of Nepal, aid financing of the fiscal deficit is about 6% of GDP, while the ratio is less than 2.0% of GDP for other South Asian countries. Thus, in Nepal, foreign aid finances on average a third of total public spending (and over 60% of the development budget); whereas in other South Asian countries this ratio is much lower-19% for Bangladesh, 6% for Pakistan and 8% for Sri Lanka. (v) On the positive side, however, Nepal and Bangladesh have resorted least to domestic financing among the South Asian countries. Domestic financing of the fiscal deficit is two to five times larger in Pakistan, India and Sri Lanka, as compared to Nepal and Bangladesh. 3. These ratios by themselves indicate that Nepal's fiscal situation is significantly weaker than those of her neighbors. The low level of revenues combined with relatively high expenditures have led to largerfiscal deficits and heavy aid dependence in Nepal. Even more significant is the fact that over the past few years, all the other countries in the region have taken strong measures to strengthen their fiscal position, particularly through expenditure adjustments, thereby significantly reducing their overall fiscal deficits and aid dependence. Nepal, on the other hand, has moved in the opposite direction. In the last few years, the expenditure/GDP ratio has risen by about 2 percentage points (from 16.8% in FY94 to 19% in FY99), largely reflecting pressures by a succession of unstable coalition governments to add new projects and programs for political reasons. As a result, Nepal's overall fiscal deficit has risen by about 2 percentage points of GDP, while aid dependence and domestic borrowing have also increased. Annex I Table 1: South Asia - Key Macroeconomic Indicators, FY91 - FY98 Nepal India Bangladesh Pakistan Sri Lanka FY91-95 FY96-98 FY91-95 FY96-98 FY91-95 FY96-98 FY91-95 FY91-95 FY96-98 Real GDP Growth (p.a.) 5.2 3.7 6.2 6.6 4.4 5.5 4.8 5.4 5.1 Per Capita GDP Growth 2.7 1.2 4.0 4.9 2.8 3.9 1.9 4.1 4.0 Inflation rate 11.3 6.6 11.6 8.8 5.5 5.4 11.5 10.3 9.9 Gross Investment/GDP 22.2 21.6 28.5 25.5 19.1 20.9 16.1 25.1 24.7 Domestic Services/GDP 12.2 9.6 26.2 21.7 13.8 14.4 17.1 14.9 17.0 National Services GDP 13.4 10.2 26.2 23.8 17.2 15.4 n.a. 18.4 21.1 Government RevenueslGDP 9.6 11.1 29.5 23.9 8.7 9.3 17.6 20.0 18.4 Total Expenditure/GDP 18.0 18.7 43.0 33.0 13.5 13.6 24.8 29.4 25.6 Current Regular 10.2 10.4 29.0 23.0 6.7 7.1 19.1 21.8 20.9 Development 7.8 8.2 14.0 10.0 6.9 6.6 5.7 7.6 4.7 Overall Deficit/GDP -8.4 -7.6 -13.5 -9.1 -4.9 -4.3 -7.2 -9.4 -7.2 Foreign Aid/GDP 6.2 5.7 0.9 1.9 4.0 2.6 0.0 4.5 1.5 Domestic Financing 2.2 1.9 12.6 8.8 0.8 1.7 5.4 4.9 3.2 Exports of Goods 8.6 8.5 9.0 9.4 7.3 10.7 13.2 26.6 30.8 Imports of Goods 23.7 31.1 11.6 13.4 12.7 17.3 18.0 37.4 39.6 Services, Net 5.2 10.7 0.4 0.2 0.1 0.0 -1.6 1.0 1.1 Current Account Balance (excl. -7.7 -9.9 -2.9 -2.2 -2.2 -2.6 -3.6 -5.2 -3.5 grants) Foreign Aid 6.3 6.6 2.5 -1.9 4.2 1.8 3.1 4.5 n.a. Debt Service Ratio (% of exports of 10.5 7.7 3.7 3.4 14.8 11.6 4.8 4.0 4.2 GNFS) External Debt/GDP - - . - - - - - Gross Reserves: US$ Million 679 878 14053 25824 2088 1836 2146 1980 2667 Months of Imports of Goods & 7.2 6.2 5.2 5.6 5.2 2.8 2.2 4.9 5.1 Services Average Exchange Rate (Rs. per 43.98 58.07 30.17 35.37 38.6 43.0 26.8 47.0 60.0 US$) Broad Money (M2) Growth % p.a. 20.8 14.7 24.5 11.9 13.6 9.7 22.1 20.4 11.5 Domestic Credit Growth % p.a. 19.6 16.6 17.3 15.5 7.8 16.8 18.3 14.3 13.4 Source. World Bank Uve DataBase (LDB). Annex 2 Page I of 3 THE NINTH PLAN AND MACROECONOMIC PROJECTIONS The Ninth Plan 1. The Ninth Plan, finalized in July 1998 as well as the FY2000 budget emphasize that eradication of poverty is the government's major development objective. The Ninth Plan's goal is to reduce poverty from an estimated 45 percent to 32 percent by FY2002 and to 10 percent in twenty years - by FY2017. This is to be achieved by: (i) Accelerating economic growth rate to an average of 6.5 percent p.a over the Ninth Plan period and to 7.2 percent p.a in later years. (ii) Bringing rural population into the mainstream of development by giving priority to agriculture, water resources, tourism, rural infrastructure and agro-based industries (to increase employment opportunities); and promoting human resource development, involving local communities closely in this process. (iii) Reducing economic disparities between regions and communities by increasing access to basic services and developing special programs for the most vulnerable and underprivileged groups (for the latter purpose, a poverty alleviation fund and an employment generation fund are to be created out of the government's own resources on an experimental basis). and (iv) Emphasizing decentralization of programs (especially human resource development and rural infrastructure) as a mechanism for involving local communities in the development process. Despite its good intentions, however, the Plan is over ambitious both in regard to its development targets and its macroeconomic framework. For example, the plan seeks to eliminate illiteracy (currently 73 percent) by FY2005, increase electricity generation capacity (currently 290 mw) to 20,000 mw, reduce child mortality rate (currently 79 per Annex 2 Table 1: Ninth Plan - Key Macroeconomic Indicators, FY97 - FY2002 ________________________________ (At FY97 prices) FY97 FY2002 Average Marginal Macro Balances - In percent of GDP Real GDP 100.0 100.0 6.5 100.0 Total Consumption 87.6 85.9 6.1 81.2 Total Investment 24.7 24.3 6.1 23.0 NaUional Savings 14.0 15.8 8.5 19.5 Imports of Goods & Non factor Services 38.8 42.7 8.5 53.2 Exports of Goods & Non factor Services 26.4 32.5 11.0 49.0 Current Account Balance -10.7 -8.5 -1.7 -2.5 FY9T Five Year TOtaW Annual Grwfth Rate Y - - Averae Dercent p.a Financing the Government Budget -Rs. Total Expenditures 50.7 337.3 67.5 10.7 Regular 24.2 147.7 29.5 7.5 Development 26.5 189.6 37.9 13.3 Total Revenue 30.4 210.8 42.2 12.1 Foreign Aid (Grants and Loans) 15.1 111.6 22.3 14.3 Domestic Borrowing 5.2 14.9 3.0 -11.9 Memo Item Revenue Surplus/b - Rs. billion 6.2 63.1 12.6 25.1 Domestic Inflation rate - percent p.a 7.5 6.5 a Indicate the proportion of the incremental GDP between FY97 and FY2002 that is assumed to be consumed, invested, b Government revenue minus regular expenditures. Source: Ninth Plan Approach Paper, National Planning Commission. Annex 2 Page 2 of 3 thousand) to 25 per thousand, and increase average life expectancy for males (currently 55.9 years) to 72 and for women (currently 53.4 years) to 75-all by FY2017. (i) As in the case of recent budgets, its resource projections are unrealistic (Annex 2 Table 1). Total revenues are projected to increase by 15 percent p.a and aid disbursements by 22 percent p.a in real terms- equivalent to nominal increases of 23 percent p.a and 30 percent p.a respectively; (ii) Expenditure targets, based on these projections in turn, are unlikely to be attainable. Development spending is projected to increase by 20 percent p.a and total expenditures by 15 percent p.a in real terms, equivalent to nominal increases of 28 percent p.a and 22 percent p.a respectively; (iii) To finance its overall investment levels, national savings are envisaged to rise from 12 percent of GDP (as estimated in the Approach Paper) in FY97 to nearly 20 percent of GDP by FY2002, with a corresponding reduction in foreign savings (i.e. improvement in the current account deficit) from 13 percent of GDP (estimated in the Approach Paper) to 6 percent of GDP. The achievement of these targets would require that 40 percent of the incremental GDP between FY97 and FY2002 should be saved and the equivalent of 65 percent of incremental GDP should be exported over the same period. The Plan thus runs the same risks as recent budgets which have failed to achieve their development targets because they are overcommitted in too many areas, without adequate prioritization and without due regard to resource availabilities. Macroeconomic Projections 2. Alternative scenarios show that under more realistic assumptions with regard to resource availabilities and sustainable levels of public (and private) spending, the Nepalese economy can still grow at a more moderate rate and still make some reasonable progress in poverty reduction and development effectiveness. Such a scenario is shown for illustrative purposes in Annex 2 Table 2 below. The scenario assumes that the Government will implement policy reforms and adhere to macroeconomic targets for revenues, expenditures, domestic borrowing, money growth and foreign exchange reserves which are necessary in order to maintain a sustainable macro-fiscal framework. Thus, government revenues are projected to grow by around 0.5 percent p.a from 10.9 percent of GDP in FY97 to 12.5 percent of GDP by FY2000; current expenditures are projected to remain stable at 9.0 percent of GDP, gross domestic borrowing is assumed to be limited to less than 1.0 percent of GDP p.a (equivalent to 0.5 percent of GDP p.a in net terms); and foreign financing of the budget is expected to remain of the order of 6.3 percent of GDP over the next three years. This would permit development spending to rise by about 1.8 percent of GDP (from 9.2 percent in FY97 to about 11.0 percent in FY2000. Since a significant part of this increase and foreign financing (about 1 percent of GDP) would represent investments related to the Kali Gandaki Power Project, other development spending could increase only modestly-by less than 1 percent of GDP over FY97 levels. Nevertheless, by focusing pubic spending on critical areas and by improving the business environment for the private sector, it may still be possible to achieve an average economic growth rate of the order of 4.8 percent p.a, and improve the pattern of growth (for example by accelerating agricultural growth) to benefit the rural areas; while making modest improvements in national savings and the balance of payments. 3. Slippage from this scenario in terms of lower revenues and/or higher regular expenditures would affect development expenditures directly, as lack of counterpart funds would slow down project implementation, and with it aid utilization. Slippage in terms of higher government Annex 2 Page 3 of 3 borrowing would lead to higher interest rates and a crowding-out of private investment. Such developments would have a negative impact on GDP growth, which could slow to 31/2-4 percent. Conversely, particularly good revenue performance might allow for an expansion of public investment if regular expenditures are contained, and sound credit management policies could lead to higher private investment, thus accelerating growth to five percent or more. Annex 2 Table 2: Macroeconomic Projections, FY95 - FY2003 FY95- FY99 fY2000 FY200I FY2D02 FY2003 Estimate Proj.cfions Real GDP Growth at producers' price 4.0 3.4 5.5 5.0 5.0 5.0 Gross Investment/GDP 23.6 19.2 19.8 20.3 20.6 21.0 National Savings/GDP 14.2 14.0 15.0 15.5 15.9 16.3 Inflation 6.3 10.0 8.0 7.0 6.0 6.0 Broad Money Growth (%) 16.1 18.1 15.0 13.0 13.0 13.0 Exports Value Growth 4.8 -3.4 11.4 10.9 11.3 9.3 Import Value Growth" 7.2 -9.3 9.4 10.2 10.6 9.9 Current Account Balance/GDP -9.4 -5.2 -4.8 4.7 4.6 -4.6 Gross Reserves US$ Million 876 1131 1125 1146 1228 1322 Months of Importse 6.2 8.8 8.0 7.4 7.2 7.0 Government Revenue/GDP 11.1 11.1 11.5 12.0 12.5 13.0 Total Expenditures/GDP 18.4 18.6 19.1 19.4 19.6 19.9 Regular Expenditures 8.9 9.3 9.3 9.3 9.3 9.3 Development Expenditures 9.5 9.3 9.8 10.1 10.3 10.6 Overall Deficit/GDP -7.3 -7.5 -7.6 -7.4 -7.1 -6.9 Foreign Financing of the budget 5.4 5.6 5.8 5.7 5.5 5.4 Domestic Borrowing (Gross) 1.9 1.9 1.8 1.7 1.6 1.5 /a Goods and Non-factors services in US$. /b Imports of goods and services. Source: World Bank Staff Projections. 4. The base case macroeconomic scenario helps to illustrate the importance of economic reforms in order to effectively implement the development/poverty reduction agenda. For example, without a significant improvement in the government's revenue performance and strong efforts to curtail the growth of salaries and debt service payments, it would not be possible to expand the development budget even modestly. Similarly, given the limited availability of public resources, better public resource management is necessary to ensure that such resources would be effectively used to maximize their development impact. Economic reforms are also needed in several other areas to help improve the business environment for the private sector, (for example through financial sector reforms, tax reform and privatization of public enterprises) and improve the efficiency of resource use in the economy. Thus, economic reforms are an important means of improving the effectiveness of development programs - public and private - at both macro and sectoral levels.