Report No. 16094-LA Lao PDR Public Expenditure Review Improving Efficiency and Equity in Spending Priorities Background Papers February 28, 1997 Country Operations Division Country Department I East Asia and Pacific Regional Office 4.. Documnt of the Wori Om* -.-'''' .' ' '-' CURRENCY EQUIVALENTS (D,-b.wer 1996) C-mna Ur-. - kip K) kep IdO U50W 105 LISS: 00 K9!4 _ L1ST OF ABBREVIATIONS ADB - Asian Deve-opmen- Rank AFTA ASE.A.N Free Trae Area AIT - Ai-an 1acrare ofTe-varlogy in T-s.ad) APB Aginculini P-F tcion Bank ARF Acute neolmi -ry Lnfecdoe ASEiN - Aesonian-r af SasSh Ease Asian Nainny AaaASA - Asarr=iia Asency far aR -LnCretI Develparner BoL - Bank of Lao FDR C.%fEA Counsl far MIctal Economic As -sacce CPC - C-reinee for PFanning and Caardioanae (-aow SPC) CPf . Caor,smer [mae Lndex CFPR Couery Pofri.aj Perfovmeaoe Reratw DAFSO - Diumsc AgnscIune and Fonvery Service Office DARCS - Deypn,renr c [Lacc AdmiXZc -n and Le Ciil Service DLVS ODepe -fene afLi-veerack and 1ean-nac- Ser-iaea DPOP - DOp,a-ent of Public Lnaces-eac Pla EdL - E deseicie Su Laes EGAT - Eleceiairy Gene-eta Aia-ry fThledd ENAG - - N-anol School Admia=snoa md .Mademeiamt EPTL PFs- & Teiecammuaica=nae ESAF ' e5-ced Sar.acnnl Adjuan.eat Facciry FACD - Foreigm Creeny Affrs Dqean-me FAIMC - F-eig. Aid.Manareo- t C.-ai.ee FDI - F-reiga Duecs I-veeatnim FLMC - F-reigp Lanvere,t.M0agemer- Ca-nene FTU - Fede.-sonn afTnde Unhrea GDP - Gaos Onerevc Product GFS - overmenr Fi-iac,e S.aica HRD - Human Resourcea Dep,,sne t IBRD - Tnceaai,ade Bsnik far Reanae.. r. s n and Denelamena ICSID - Lieersaes Cee far Sea ae-e=eialonvaeran DCianaee LDA - Iisremnraial D-elnepmer Aaaaniatioa fOF - lrrseivuueel De-ceaprmee Fcnd IFC )rerrrsraslFinanceCarpankan LLO - Inrevastdanal Lunean Orgsc-.nna PMF . Irrerraiecl rlaocerasy Fund [MR - efant Maonairy Pace [RD - lcepg-nd Rurl De-v)apmeee 13Rf - uniernational Rice Ra-eal lanisenr JICA - Japane-e lecematmnac Coopmao,n Agency LECS - La.. Eapediecre and Censpane SWines LFBSS - Lao Fe-rcry and Binh Sp=ing Survey LFS - L.. SBa Fuel Company LSFP - La - Swedieh Farey Pr-ec LSIS - L-n Saccal Iadi-aar Scr-ey MAP F Mieiscy If Asr.calnc nd Foreeny MCH - M -enr.al and Child ear. MCTFC -I Minisy a) Co-mmu-nicrno Tiannypn. Post and Caoenmcnian MIGA - aIvluleral l-ceaavest Gcunanree Agency MLSW l4miay afLabv, and Sccia WeWfane Moo - Mi-ery afOvfte MaF - Mmntvcy a) Finance MoH - Minisry of Heitlh MoPH - Mimney ofFblic Hearlh MOU . Memarmdum,a)Cdevoandag NCs - Nadairal Canepefeive Biddig NEM - Nea Ecacaeic Meccr ncmv NCO - Naa-Gacerreeatul Offaniatnoc NRRP - Nascaul Ric: Reseamh Pgttn Nfl - Nun Thean 2 (hydropawev peejec,) ODA - Official D-Oelape,nt Asuieraate O&M - Operstia and Mam-ceemae OPFC - Org of inaici oPaentleam Eaponmg Coaiee PAFSO - P-vi.rail Ag=ciueee and Femay Service ffice PAYG - Pay As Yoa Ga (fEn-ciel syseeme) PDG - fraje,cc D-elop-ent Gnpup PDR - Peaplee ensn ,ciac Rvprnblic PER - Publc EBpc,dinen Review PFP - Paiccy Fvamecnak Fapec PHRD - Policy and Huaiun R..atce. Decelaopmeet Fund PFLMS - Pblic Ioneaaer- n Moautriug System PIP * fublia Lan .IvevsPln /Prognm PfPR P Freeee Lmplemieaag RPdes and Re,glansan PfU - Pfrjecm Insple-menesic uCts PMRSI - Pric-altacon and Mmaqeaent af RLoad Senmr Lasadnaim PPA - Prnjeat Ptfa-onoce Audit RAD - Road Admieaflwn DOivinir RMSM - Revised Minimum Standand Model SLDA - Swediea Lcteeaeoel D-velap.eev Autbnriry SEP - Snntonl Ianeimeet Frogsm SOCB - Suae-Owned Commecial Bank SOE - Stae-Owned Ecterprian SPC - Stae P!ancnin Coowmnee (foemeely CfC) TA Technical Aaaiaranee TVET Tec-hical and Vordonal Educdan U5MR - Uncer, -Fie F - eaY Mraseary Rumi t,NCDF - Unrvtd NIatean Cupiul De-e!opienv Faed OODP - u.ecd Nations De elap-mc FPrnmmree UNESCO (Jired Niutos Succaiavaf Scevntdc and Cuvanal OCgeirntt UNICEF - Ocred Macave Chiidnens Fand WDR - Warld Cnelopment Repan WHO - World vl9lca OrEnecacic WUG - War,Ocn- r Graup V ne PFe,dent; lnn-Xlccr dcerna Orveciar: lavad Kha1lladrr-Shionis Oic,eoun ChcieLMuanagr: 5un,lDar (noelniene Soil Nlvrner, nu S. Zanvvvebe - v - TABLE OF CONTENTS BACKGROUND PAPERS PART I. PUBLIC EXPENDITURE ANALYSIS IN KEY SECTORS CHAPTER 1. EXPENDITURE AND PROJECT PRIORITIZA TION ......................................1 CHAPTER 2. TRANSPORT ................................................... 9 A. SECTOR OVERVIEW ................................................... 9 B. GOVERNMENT OBJECTIVES FOR TRANSPORT ................................................... 10 C. FUNDING MECHANISMS ....................................................11 D. EXPENDITURE TRENDS ...................................... 14 E. SCOPE FOR PRIVATE SECTOR PARTICIPATION ...................................... 16 F. EFFICIENCY OF PUBLIC SPENDING ON TRANSPORT ...................................... 18 G. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM ...................................... 22 H. RECOMMENDATIONS ...................................... 26 APPENDIX: EXTERNAL ASSISTANCE -- TRANSPORT AND RELATED SECTORS .. 31 CHAPTER 3. AGRICULTURE ............................................................ 35 A. SECTOR OVERVIEW ............................................................ 35 B. GOVERNMENT OBJECTIVES FOR AGRICULTURE ........................................................... 37 C. FUNDING MECHANISMS ........................................................... 42 D. EXPENDITURE TRENDS ........................................................... 43 E. EFFICIENCY OF PUBLIC SPENDING ON AGRICULTURE ................................................... 46 F. SCOPE FOR PRIVATE SECTOR PARTICIPATION ........................................................... 49 Q. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM .................................................... 50 H. RECOMMENDATIONS ........................................................... 52 CHAPTER 4. HEALTH ........................................................... 57 A. SECTOR OVERVIEW ........................................................... 57 B. GOVERNMENT OBJECTIVES FOR HEALTH ........................................................... 59 C. FUNDING MECHANISMS ........................................................... 60 D. EXPENDITURE TRENDS ........................................................... 62 E. EFFICIENCY OF PUBLIC SPENDING ON HEALTH ............................................................ 64 F. SCOPE FOR PRIVATE SECTOR PARTICIPATION AND COST RECOVERY ............................ 66 G. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM .................................................... 68 H. RECOMMENDATIONS ............................................................ 71 - vi - CHAPTER 5. ED UCA TION ......................................................... 79 A. SECTOR OVERVIEW ........................................................ 79 B. GOVERNMENT OBJECTIVES FOR EDUCATION ........................................................ 80 C. FUNDING MECHANISMS ........................................................ 80 D. EXPENDITURE TRENDS ......................................................... 87 E. EFFICIENCY OF PUBLIC SPENDING ON EDUCATION ....................................................... 92 F. SCOPE FOR PRIVATE SECTOR PARTICIPATION ........................................................ 94 G. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM .................................................... 95 H. RECOMMENDATIONS ........................................................ 96 PART II. BUDGETARY INSTITUTIONS CHAPTER 6. BUDGETARYARRANGEMENTSINLAOPDR ............................................. 99 A. PREVIOUS BUDGETARY ARRANGEMENTS ........................................................ 99 B. CURRENT BUDGETARY ARRANGEMENTS ........................................................ 100 C. DESCRIPTION OF THE BUDGET PROCESS ........................................................ 100 D. THE PUBLIC INVESTMENT PROGRAM ........................................................ 102 E. THE REVENUE SIDE OF THE BUDGET ...............................103.......................... . . 103 APPENDIX: A FRAMEWORK FOR ANALYZING THE LAO BUDGET SYSTEM .................... 105 CHAPTER 7. PUBLIC INVESTMENT PLAN (PIP): METHODOLOGICAL PROBLEMS ........................................................ 109 CHAPTER 8. THE REGIONAL DIMENSION: CENTER-PROVINCE RELA TIONSHIPS ........................................................ 115 CHAPTER 9. ISSUES IN PROJECT IMPLEMENTA TION ................................................ 123 A. PROJECT IMPLEMENTATION: INSTITUTIONAL CONSTRAINTS ...................................... 123 B. RECOMMENDATIONS ........................ . . . . . .............. 129 CHAPTER 10. AID COORDINATION ............................... 133 A. OVERVIEW OF AID MANAGEMENT .............................................................................. 133 B. KEY ISSUES IN AID COORDINATION AND MANAGEMENT ............................................ 134 C. RECOMMENDATIONS .......................................................... 137 PART III. REVENUE MOBILIZATION CHAPTER 11. REFORMS OF THE TAXSYSTEMAND TAXADMINISTRATION ... I41 A. CONSTRAINTS OF TAX REFORM ............................................................... 141 B. THE FIRST REFORM YEARS, 1986-88 ............................................................... 141 C. TAXREFORMS SINCE 1988 ............................................................... 142 D. THE NEW TAX LAW OF 1995 ............................................................... 142 E. IMPROVED SYSTEM OF REVENUE COLLECTION ........................................................... 143 F. ELEMENTS OF FURTHER TAX REFORM ............................................................... 144 - vii - CHAPTER 12. HYDROPOWER AND PUBLIC EXPENDITURE . ....................................... 145 A. PUBLIC FINANCE ISSUES IN POWER DEVELOPMENT .................................................... 145 B. DOMESTIC POWER SUPPLY ...................................................... 146 C. SCOPE OF POWER EXPORTS ...................................................... 147 D. OPPORTUNITIES AND RISKS OF HYDROPOWER ...................................................... 148 E. FINANCING HYDROPOWER DEVELOPMENT ...................................................... 149 F. REVENUES FROM POWER EXPORTS ....................................................... 152 G. CONCLUSIONS: NEED FOR A STRATEGY ....................................................... 153 CHAPTER 13. TIMBER AND TIMBER ROYALTIES - ISSUESAND PERSPECTIVES ...................................................... 155 A. REFORMS IN THE TIMBER TAXATION SYSTEM ....................................................... 155 B. NEED FOR REFORM IN TIMBER TAXATION ...................................................... 156 C. ELEMENTS OF FUTURE REFORM IN TIMBER TAXATION ............................................... 158 PART IV. PUBLIC EXPENDITURE MANAGEMENT C.HAPTER 14. STA TE-O WNED ENTERPRISES ...................................................... 161 A. THE PRIVATIZATION PROGRAM ...................................................... 161 B. IMPROVING THE PERFORMANCE OF THE REMAINING STATE-OWNED ENTERPRISES ...................................................... 166 CfHAPTER 15. CIVIL SER VICE REFORM ....................................................... 171 A. REFORM PROGRESS TO DATE ...................................................... 171 B. OPTIONS FOR FURTHER REFORM ...................................................... 176 CHAPTER 16. SOCLAL SECURITY .................................................. 185 A. REVIEW OF EXISTING SOCIAL PROTECTION PROVISIONS ............................................. 185 B. OPTIONS FOR REFORM AND DEVELOPMENT ...................................................... 193 C. SUMMARY OF RECOMMENDATIONS ...................................................... 205 STATISTICAL APPENDIX ...................................................... 209 - viii - TEXT TABLES CHAPTER 2 TABLE 2.1: COMMUNICATIONS AND TRANSPORT SECTOR SHARE IN 1991-95 PIP .......... 14 TABLE 2.2: LAO PDR TRANSPORT INVESTMENT PROGRAM, 1991-95 ............................ 15 TABLE 2.3: FOURTH FIVE-YEAR PLAN TRANSPORT INVESTMENT PROPOSAL ................. 23 TABLE 2.4: SUGGESTED EXPENDITURE PROGRAM FOR TRANSPORT, 1996-2000 ............ 26 TABLE A2. 1: EXTERNAL ASSISTANCE FOR TRANSPORT SECTOR .32 TABLE A2.2: PROVINCIAL PUBLIC INVESTMENT PROGRAM, 1995/96 ................................ 33 CHAPTER 3 TABLE 3.1: AGRICULTURAL PROGRAMS RELATIVE TO SUBSECTOR GDP, 1995/96 ........ 46 TABLE 3.2: RESEARCH AND EXTENSION IN ASIA RELATIVE TO AGRICULTURAL GDP .................................................. 46 TABLE 3.3: INTERNATIONAL COMPARISONS - FOOD PRODUCTION GROWTH, ARABLE LAND, AND FERTILIZER USE IN ASIAN COUNTRIES ........................ 47 TABLE 3.4: GOVERNMENT PIP FOR AGRICULTURE AND FORESTRY, 1995/96-1999/2000 .................................................. 51 TABLE 3.5: SUGGESTED EXPENDITURE PLAN FOR AGRICULTURE, 1995/96-1999/2000 .................................................. 52 TABLE 3.6: ALLOCATION OF PUBLIC INVESTMENT IN AGRICULTURE AND FORESTRY 1990/91-1994/95 .................................................. 55 CHAPTER 4 TABLE 4.1: HOUSEHOLD CONSUMPTION, BY ECONOMIC STATUS AND URBAN/RURAL .................................................. 61 TABLE 4.2: ECONOMIC COMPOSITION OF PUBLIC HEALTH EXPENDITURES .................... 63 TABLE 4.3: HEALTH SPENDING AND HEALTH INDICATORS, SELECTED COUNTRIES ........ 66 TABLE 4.4: EXTERNAL FINANCE, 1995/96 PIP .................................................. 69 TABLE 4.5: PUBLIC INVESTMENT PLAN FOR HEALTH, 1995/96 - 1999/2000 .................. 70 TABLE 4.6: GOVERNMENT EXPENDITURE ON HEALTH, 1986-95 ..................................... 74 TABLE 4.7: MoH EXPENDITURE, 1995/96-1999/2000 .................................................. 75 TABLE 4.8: HEALTH SPENDING AND INDICATORS BY REGIONS ....................................... 75 TABLE 4.9: PLANNED AND ACTUAL HEALTH SPENDING AT THE PROVINCIAL LEVEL, 1994/95 .................................................. 76 TABLE 4.10: ACTUAL GOVERNMENT PROVINCIAL HEALTH EXPENDITURE .77 CH1APTER 5 TABLE 5.1: COMPARATIVE DATA ON PUBLIC SPENDING ON EDUCATION AS A SHARE OF GDP, AND AVERAGE DURATION OF SCHOOLING .83 TABLE 5.2: EDUCATION BUDGET, SHARE IN GDP AND OVERALL BUDGET, 1990-96 ..... 83 TABLE 5.3: THE MAGNITUDE OF EXTERNAL FINANCING FOR EDUCATION ...................... 84 TABLE 5.4: SOURCES OF SCHOOL REVENUE ................................................. 85 TABLE 5.5: BUDGET PROJECTIONS, 1995/96 - 1999/2000 .............................................. 86 TABLE 5.6: DISTRIBUTION OF RECURRENT EXPENDITURES BY LEVEL OF EDUCATION ................................................. 88 TABLE 5.7: AVERAGE UNIT RECURRENT COST ESTIMATES, 1994-95 ............................. 88 TABLE 5.8: UNIT COSTS OF EDUCATION BY LEVEL ................................................. 89 - ix- TABLE 5.9: DISTRIBUTION OF UNIT COSTS BY TYPE OF EXPENDITURE AND LEVEL OF EDUCATION ................................................................ 90 TABLE 5.10: CHARACTERISTICS OF PRIMARY SCHOOLS ................................................... 92 TABLE 5.11: ENROLLMENT IN PRIVATE EDUCATION, 1989-95 ......................................... 94 TABLE 5.12: PUBLIC INVESTMENT PLAN FOR EDUCATION, 1995/96-1999-2000 .............. 95 CHAPTER 6 TABLE A6. 1: KEY INSTITUTIONAL ARRANGEMENTS AND EXPENDITURE OUTCOMES ...... 105 CHAPTER 8 TABLE 8.1: PLANNED NATIONAL REVENUE BY PROVINCES, 1995/96 .121 CHAPTER 9 TABLE 9.1: PUBLIC INVESTMENT PROGRAM, 1995/96-1999/2000 .126 CHAPTER 10 TABLE 10.1: EXTERNAL ASSISTANCE, 1990-94 .133 TABLE 10.2: EXTERNAL ASSISTANCE BY SOURCES OF FUNDS, 1991-95 .134 CHAPTER 12 TABLE 12.1: EDL - FINANCIAL STATEMENTS AND PROJECTIONS ................................... 150 TABLE 12.2: MOST PROMISING HYDROPROJECTS IN LAO PDR ...................................... 151 TABLE 12.3: PROPOSED FINANCING STRUCTURE FOR THE NAM THEUN 2 HYDROELECTRIC PROJECT ........................................................ 152 CHAPTER 13 TABLE 13.1: TIMBER LOGGING QUOTAS AND TIMBER ROYALTIES, 1994/95 .156 TABLE 13.2: PLANNED AND ACTUAL TIMBER REVENUE, 1991-95 .157 TABLE 13.3: LOGGING AND TIMBER ROYALTIES, 1994/95-1995/96 .158 CHAPTER 14 TABLE 14.1: STATE-OWNED ENTERPRISES CLASSIFIED AS "STRATEGIC" .......................... 166 CHAPTER 15 TABLE 15.1: PRE- AND POST-REFORM SALARY STRUCTURE .......................................... 173 TABLE 15.2: AGE DISTRIBUTION OF CIVIL SERVANTS .................................................... 177 TABLE 15.3: PERSONNEL AND SALARIES IN THE PUBLIC SECTOR, 1988-95 .................... 182 TABLE 15.4: TOTAL NUMBER OF CIVIL SERVANTS BY CATEGORIES ............................... 183 TABLE 15.5: LIST OF CIVIL SERVANTS AT THE CENTRAL AND PROVINCIAL LEVELS ....... 184 CHIAPTER 16 TABLE 16.1: ESTIMATE OF PAYMENTS IN 1994-95 AND DISTRIBUTION BY TYPE OF BENEFIT .186 TEXT FIGURES CHAPTER I FIGURE 1.1: INDICATIVE PRIVATE/PUBLIC SECTOR ROLES .3 CHIAPTER 3 FIGURE 3.1: GROWTH OF REAL VALUE ADDED .35 FIGURE 3.2: ANNUAL GROWTH RATES OF RICE PRODUCTION .35 FIGURE 3.3: RICE SUPPLY/DEMAND BALANCE .38 CHAPTER 5 FIGURE 5.1: EDUCATION BUDGET, 1990-96 .81 - x - FIGuRE 5.2: EDUCATION'S SHARE OF TOTAL GOVERNMENT BUDGET ............................ 82 FIGURE 5.3: AVERAGE ANNUAL COST OF KEEPING A CHILD IN SCHOOL ........................ 85 FIGURE 5.4: DISTRIBUTION OF EDUCATION RECURRENT BUDGET, BY TYPE OF EXPENDITURE, 1994/95 ................................................ 91 FIGURE 5.5: PER CAPITA EDUCATION EXPENDITURE, BY INCOME QUINTILE AND RESIDENCY ................................................ 93 CHAPTER 6 FIGURE 6.1: MAJOR CATEGORIES OF NATIONAL REVENUE, 1994/95 ............................ 104 FIGURE 6.2: PLANNED VERSUS ACTUAL REVENUES, 1992/93-1994/95 ........................ 104 FIGURE A6. 1: BASIC BUDGETARY OBJECTIVES . .107 CHAPTER 8 FIGuRE 8.1: PROVINCIAL INCOME AND POPULATION ............................................... 116 FIGURE 8.2: PROVINCIAL LEVEL PUBLIC INVESTMENT PROGRAM ................................ 118 FIGURE 8.3: REVENUE AND EXPENDITURE AT CENTRAL AND PROVINCIAL LEVELS ......1 19 FIGURE 8.4: REVENUE AND EXPENDITURE BY PROVINCES, 1995/96 ............................. 119 FIGURE 8.5: REVENUE AND EXPENDITURE PER CAPITA BY PROVINCES, 1995/96 ......... 120 CHAPTER 9 FIGURE 9.1: FINANCING OF PUBLIC INVESTMENT PROGRAM, 1991-95 ......................... 126 FIGURE 9.2: PROJECT DISBURSEMENT PERFORMANCE, 1990-95 .................................. 128 CHAPTER 14 FIGURE 14.1: VALUE OF SOES TO BE PRIVATIZED .164 CHAPTER 15 FIGURE 15.1: COMPOSITION OF PERSONNEL BY SALARY CATEGORIES .173 FIGURE 15.2: LAO CIVIL SERVANTS BY MINISTRIES AS OF JANUARY 1996 ...................... 175 FIGURE 15.3: CIVIL SERVANTS BY PROVINCE AS OF JANUARY 1996 ................................ 175 TEXT BOXES CHAPTER 3 BOX 3.1: THE AGRICULTURAL PROMOTION BANK .......................................... 41 CHAPTER 9 BOX 9.1: TECHNICAL ASSISTANCE FOR PROJECT IMPLEMENTATION ........................ 129 CHAPTER 10 BOX 10.1: THE BROAD SECTOR APPROACH TO INVESTMENT LENDING ....................... 139 CHAPTER 14 BOX 14.1: STATE ENTERPRISE REFORM IN VIETNAM .......................................... 161 BOX 14.2: HAVE THE CHINESE SOES BEEN SUCCESSFULLY REFORMED? .................. 162 BOX 14.3: THE TRANSFER OF SOES TO LOCAL COMMUNITIES - AN ALTERNATIVE TO LEASING ................. 165 BOX 14.4: LAO STATE FUEL CO ................. 167 BOX 14.5: NAM PAPA ................. 169 CHAPTER 15 BOX 15.1: NATIONAL SCHOOL OF ADMINISTRATION AND MANAGEMENT (ENAG/NOSPA) ............................................. 172 BOX 15.2: TECHNICAL ASSISTANCE FOR PUBLIC ADMINISTRATION REFORM ............ 172 - xi - BOX 15.3: TARGETED EARLY RETIREMENT ....................................... 177 BOX 15.4: COST OF EARLY RETIREMENT FOR EX-REVOLUTIONARIES ........................ 178 BOX 15.5: COST OF EARLY RETIREMENT FOR "ANCIEN RąGIME" CIVIL SERVANTS .179 MAP LAO PEOPLE'S DEMOCRATIC REPUBLIC PUBLIC EXPENDITURE REVIEW .29 -1- PART I. PUBLIC EXPENDITURE ANALYSIS IN KEY SECTORS CHAPTER 1: EXPENDITURE AND PROJECT PRIORITIZATION 1. In selecting specific projects and determining expenditure priorities for the public investment program, the Government faces cross-cutting issues of strategic importance. Public investment decisions should be guided by key criteria of economic efficiency. Public expenditure analysis aims to highlight to what extent intrasectoral outlays are internally and externally efficient: * Internal efficiency asks whether the mix of outlays within sectors is adequate. For instance, is the mix of school construction, teacher education, teacher salaries, and educational materials appropriate? The mix of capital and recurrent spending must ensure that infrastructure such as roads and bridges can be maintained properly and that adequate funds for materials and supplies are budgeted in the social sectors. * External efficiency is measured by the extent to which sectoral goals are accomplished, as shown by the relationship between inputs (for example, spending for public health) and their outcomes (for example, infant mortality rates). Efficiency considerations suggest that intrasectoral outlays be concentrated on activities with the highest social rates of return, such as primary education and community health. Activities with high private returns and lower levels of externalities should be left to the private sector or should be subject to cost recovery. Equity considerations imply a focus on programs that benefit the poor, especially in rural areas. 2. Before decisions are made on spending scarce public resources on a particular project, be it a road or a hospital, some key questions should thus be asked: * Government's Role and the Private-Public Mix. Does the Government have a legitimate role in this project or should it be left to the private sector? * Cost-Benefit of Project Inputs and Outcomes. Is this project justified in terms of its rates of return? * Absorptive Capacity. Can additional projects be effectively absorbed and implemented? * Operations and Maintenance. Is new construction really preferable to maintaining and repairing existing facilities? * Cost Recovery Policies. Should part or all of the project costs be imposed on the actual users and beneficiaries through user charges? * Impact on the Poor. Will project benefits accrue to the poor? -2- 3. (i) Government's Role and the Private-Public Mixm Public investment should make increased private investment possible rather than crowding it out. To determine cross-sectoral expenditure priorities, the Government needs a clear perspective regarding the respective roles of the private and the public sector. Lao PDR has already gone a long way in reducing the role of the state in areas which are best left to the private sector, such as agriculture and industry. The country has been following the path of other transition economies, where disappointing results of public ownership and management of productive assets have prompted substantial privatizations of industrial activities as well as infrastructure. At the same time, the state is likely to remain dominant in many sectors of the Lao economy despite the growing private sector, owing to the large flow of external financing driving the public investment program. This will require a careful evaluation of proper state intervention, considering the experience of successful East Asian economies, which suggests that the state can play an important role in promoting economic growth.' Moreover, it is clear that public investment is needed to meet urgent social needs. For example, building a hospital in a remote rural area might have a low financial rate of return and would thus not attract private investment. 4. For Lao PDR, the lessons of its successful neighbors indicate that the country is well advised to proceed with its strategy of reducing the role of the public sector in the production of goods while enhancing the quality of government services in areas for which the private sector is currently not well suited, such as education, health, utilities, and public safety. Thus, the state will have a different role across different sectors, as suggested in the categorization of the mix between public and private sector participation in Figure 1.1. While there is no unique state role, this classification by the degree of government involvement in a market-oriented economy can provide broad guidance regarding the relative shares of different sectors in total government expenditures that Lao PDR should aim to achieve. On the one hand, Lao PDR will want to enhance its core government functions, of providing public goods and general administration (Box I, in Figure 1.1). On the other hand, economic liberalization and privatizations have allowed the state to largely withdraw from providing goods and services that can now be left to the emerging private sector (Box IV). Given Lao PDR's stage of development and the nascent nature of the private sector, there are numerous areas in which the state can expect to maintain a majority with private sector participation whenever possible (Box II) or where the private sector will be predominant (Box III). In practice, of course, the scope for private sector involvement needs to be carefully evaluated at the level of programs or even projects rather than sectors, as suggested in Chapter 3 of the Main Report. For Lao PDR, the key issue will be to carefully maintain an appropriate balance to maximize the complementarities between public and private investment in each sector. In particular, the state is required to grow into a role of safeguarding private sector activity by creating a market-enhancing legal, institutional, and macroeconomic policy environment. There is a growing body of literature on the East Asian Miracle, see, for example, World Bank, "The East Asian Miracle: Economic Growth and Public Policy," 1993, and Wade, "Goveming the Market," 1990, for diverging views on the role of the state. -3- FIGURE 1.1: INDICATIVE PRIVATE/PUBLIC SECTOR ROLES L Exclusively Public Sector II Public Sector Majority with Private Sector Participation Environment - management and regulation Culture Government administration Education ' Law and order Energy Judiciary Electricity distribution Public safety and defense Health Highways Natural resource management (forest, minerals, water) Transport infrastructure b (airports, ports) Water supply and sewerage III. Predominantly Private Sector IV. Exclusively Private Sector Communication Agriculture Electricity generation Commerce and trade Finance (banking and insurance) Construction Hydropower Fuel and petrol Mining Manufacturing Transport superstructure b Natural resource use (logging, mining, fishing) Sports Sales and service Telecommunications Tourism Trucking and shipping E Education includes normal academic education, plus other forms of education, such as physical education, vocational education, adult education, and agricultural extension. b Infrastructure includes basic facilities such as landing strips, highways, and piers and channels, while superstructure refers to ancillary facilities such as terminals, cold stores, general warehouses, specialized cargo handling equipment, and services for passengers. 5. (ii) Cost-Benefit of Project Inputs and Outcomes. In addition to the order of magnitude of government capital expenditures, their quality is of at least equal importance for the sustainability of economic growth. Individually approved capital expenditures should thus be well screened, cost effective, and likely to lead to private or public sector productivity gains. The best-known measures of economic efficiency are various forms of discounted cash flow analysis, such as rates of return. However, the information needed for discounted cash flow analysis generally does not exist when the public investment plan is being formulated. Consequently, for selecting projects within sectors, strategic concerns and priorities such as maintenance and repair, energy efficiency, environmental impacts, completion of ongoing projects, and future estimated operating costs must be considered for the initial screening of projects to be included in the public investment program (PIP). Nevertheless, before projects are actually included in the annual capital budget, a detailed rate of return analysis should be carried out to compute financial and economic rates of return, comparing the rate of return on the marginal project with the marginal costs of funds. This analysis -- at least for important projects -- will help assure that only economically acceptable public investment is identified and financed, and that optimal technical designs are chosen. If rate of return analysis is not possible because benefits cannot be quantified, alternative types of analysis, such as cross-over discount rates and least cost analysis, should be used. Least cost analysis might be appropriate, for example, in the case of schools or clinics where the education and health services are not sold at competitive market 4- prices that cover all costs. This would allow a cost-effective targeting of social services to be offered at discounted prices to the poorest. 6. (iii) Absorptive Capacity. On economic efficiency grounds, expenditure decisions also need to take account of the capacity of a sector to effectively absorb and implement additional projects. Project selection should thus focus on the complementarity of public investment to institutional constraints by limiting expenditures in areas in which project implementation capacity is limited. Instead, emphasis should be placed on strengthening the efficiency of existing projects, improving the quality of services, and attracting, training, and motivating qualified staff. 7. (iv) Operations and Maintenance (O&M). Building new projects is almost always more appealing than fixing up old ones. However, experience from around the world indicates that investments in deferred maintenance and rehabilitation of otherwise viable facilities that have been allowed to deteriorate can have far higher rates of return than the construction of new projects. In fact, unless the existing facility has suffered serious structural damage or would need to undergo major modifications because of fundamental changes in its intended use, it can generally be assumed that rehabilitation will be more cost effective than new construction. This is particularly true in a country like Lao PDR, where a substantial inflow of foreign aid has been directed to building new structures while domestic resources have not been made available to maintain and repair existing facilities. Aid dependency creates a dilemma: donors do not fund O&M for existing assets, and the country cannot afford to refuse a project but does not have the money for maintenance expenditures. 8. Expenditure decisions on O&M may affect the quality of a wide range of public expenditure programs since the physical condition of the capital stock may be the key factor in determining the quality of service derived from it. However, inadequate coordination of the capital and current planning procedures in Lao PDR leads to a lack of information on implied recurrent costs at the project preparation stage. As a result, future O&M outlays stemming from the PIP are not adequately reflected in forward planning of the budget. Although donor-financed capital projects usually include recurrent costs during the project, no O&M expenditures are generally budgeted for the time after project financing is finished (see Chapter 7). In the transportation sector, for instance, roads eventually need to be rehabilitated at a multiple of what appropriate maintenance would have cost. There is thus a real risk that donor aid not only finances a large share of capital expenditure but also substitutes for current expenditure needs (such as rehabilitating road projects). 9. Recommendations. Using existing infrastructure more efficiently should precede major investments to supplement infrastructure. There are two ways to do this: (i) with better demand management; and (ii) with efficient non-wage O&M policies. Appropriate pricing policies are essential to using infrastructure efficiently and can also improve investment decisions and increase the efficiency of public provision by raising revenue for O&M. A sound O&M strategy is also required to ensure that scarce public resources are used efficiently. The budgeting of maintenance expenditures in Lao PDR can best be improved in the context of an overall expenditure policy and planning strategy that includes the following actions: => Priority should generally be given to rehabilitation and repair projects, rather than to new projects, when framing the PIP. The public investment programming and budgeting process needs to include calculations of ongoing O&M expenditure in a systematic way. Forward budget estimates for recurrent items should be developed as part of the annual budgeting process. This would include extending the PIP into a medium-term budgeting framework which would include both capital outlays and forward estimates of recurrent outlays that are currently included in the -5- annual budget. Specifically, officials should regularly undertake a technical assessment of the O&M implications of the PIP, using "r coefficients" (the ratio of net recurrent expenditure requirements to total investment cost) and drawing on World Bank data or technical assistance. = The Government should assign to a single agency the primary responsibility for overseeing O&M expenditure needs and prograns. The Budget Department in the Ministry of Finance should be given the necessary official mandate (and technical expertise) to establish a consistent forward-looking multi-year O&M strategy to guide public investment decisions, carefully balancing clearly identified expenditure requirements with realistic revenue projections. = Under the auspices of the State Property Department, spending ministries and agencies, in conjunction with the relevant provincial authorities, should compile sectoral data bases on public assets and their maintenance requirements and then define sectoral O&M objectives, programs, costs, and other requirements and establish ongoing monitoring and evaluation of O&M expenditure programs. This would help prevent the premature deterioration of state property owing to shortfalls in maintenance, and should reduce costly rehabilitation and make the need for adequate O&M budgeting more transparent. - Government budget accounting systems should incorporate line items and coding structures that identify actual O&M expenditures within the normal Government Finance Statistics (GFS) economic classifications. The new budget nomenclature to be adopted for the 1996/97 budget will go some way toward greater clarity but should be revised as needed to highlight O&M requirements, particularly those stemming from donor-funded projects. 10. (v) Cost Recovery Policies. Cost recovery policies for goods and services provided by the public sector involve imposing part or all of their costs on the actual users and beneficiaries through user charges, rather than on the general public through the tax system. Such policies have not yet been widely introduced for government services in Lao PDR, but they deserve thorough consideration and wider application in the future for several reasons. First, partial or full cost recovery could help reduce dependence on tax revenues and could lower the fiscal deficit. Second, cost recovery encourages a more efficient allocation of resources. As public goods and services are priced closer to cost, consumers (including other public sector users) are encouraged to ration their demand for them and use them more economically and efficiently. Third, the resulting commercialization of public sector activities makes them similar to private sector activities and, over time, may help generate private sector competition. Fourth, cost recovery policies could stimulate a more consumer-oriented approach to service provision within the public sector. This would complement other efforts (such as the public sector pay and employment policies proposed in Chapter 15) to raise the overall quality and efficiency of public sector activities. 11. Whether cost recovery policies are useful depends to a large extent on the nature of the public good that is being provided and the government's objective in providing it. For economic efficiency, wherever the beneficiaries can be identified and goods and services can be provided in a commercial manner, the target should in principle be full cost recovery by appropriate user charges. In general, cost recovery policies need to take into account consumers' ability to pay. Reduced prices, limits on total annual charges paid, or direct income transfers could provide an effective means of social protection for the poor. The existence of privately provided equivalents, or near substitutes, may assist in identifying an appropriate market pricing structure. However, where current expenditure is provided on equity grounds, or where the beneficiaries are difficult to identify, financing expenditures from tax revenues is likely to be more efficient. -6- 12. Recommendations. In practical terms, there is a wide range of opportunities for extending user charges and cost recovery principles in Lao PDR. Obvious examples include public utilities (for example, gas, electricity, and water); transport and communication services (roads, airports, and telephones); and social services (health and education). The role that user charges could play in some of these areas is considered in the sectoral chapters of this PER. However, the concept has potentially much wider application. The Government could proceed in a systematic way to examine the following wider opportunities for the introduction of a cost recovery policy: > Have direct spending ministries and agencies prepare comprehensive lists of the goods and services provided to public and private sector enterprises, or the general public (e.g., Lao Aviation planes carry mail without cost recovery, SOEs run up arrears on their electricity bills, etc.) > Identify standard costs (for example, per unit prices for public utility services such as water or electricity, or average hourly rates for specific tasks) for each spending agency, taking into account the current operational costs (salaries, consumables, etc.), depreciation allowances, and any interest or financing costs Determine from this list the potentially chargeable goods and services; identify important social factors (targeting services to the poor) and externalities (such as environmental effects) that might influence pricing and charging decisions Identify and compare standard cost parameters with market prices of similar or equivalent goods or services provided by the private sector - Determine initial cost recovery targets for each chargeable activity or operational government unit (for example, 50% within two years) Determine and implement appropriate pricing and charging policies -- Adjust the budgets of spending agencies to maintain initial budget neutrality in relation to inter- agency charging. 13. Earmarking. When trying to introduce new charging policies, line ministries tend to argue in favor of earmarking cost recovery revenues for use in specific expenditure programs. (This issue is being discussed particularly with regard to road user charges and fuel taxes for road construction and maintenance.) Two frequently cited advantages of earnarking are that it can: (i) increase consumers' willingness to pay for services and thus improve revenue collection; and (ii) alleviate the crowding out of necessary maintenance and other expenditures (such as training) that are of high priority over the medium term but that -- owing to a general myopia of the policymaking and budget process -- often tend to be cut disproportionately during stringent fiscal periods. As is discussed in Chapter 4 of the Main Report, however, earmarking not only potentially safeguards long-term interests but also reduces the flexibility of budget authorities to change priorities according to changing policy and economic priorities, and it can lead to a misallocation of resources by removing such revenues from the normal budgeting (that is, priority-setting) process. Therefore, despite its potential attractions, earmarking of revenues should as a general rule be avoided. 14. User charges can also be an effective way of financing depreciation and interest costs. Other sources of funding for maintenance include earmarked special taxes or revenues and counterpart -7- funds from external assistance and foreign borrowing. However, none of these alternatives fully offsets the need for appropriate budgeting policies that assess the benefits of improving or maintaining the capital stock against alternative uses for current expenditure. 15. (vi) Inmact on the Poor. Cost-benefit analysis merely tends to sum up the benefits and costs of a project without due consideration of particular groups, particularly the poor. Poverty concerns are usually not part of individual project evaluations, owing to analytical, capacity, and empirical difficulties in determining distribution weights. However, it may be feasible to evaluate access to the poor of some broad expenditure programs rather than evaluating each project in this respect.2 This will require an evaluation of the benefits incidence of broad expenditure programs to ensure that expenditures meet the objective of poverty alleviation. Benefit incidence examines the ways in which public expenditure allocations benefit or accrue to different income groups, including the poor. 2 See Pradhan (1996): "Evaluating Broad Allocations of Public Expenditure," World Bank. - 9 - CHAPTER 2: TRANSPORT A. SECTOR OVERVIEW 1. Lao PDR's transport system is rudimentary, dilapidated, and unreliable except for some recently restored sections. This has made internal and border crossing transport extremely difficult and costly. The Government realizes that the present lamentable state of the transport system is a key impediment to economic growth, with particularly negative and disruptive effects on the economic and social opportunities of the widely dispersed rural population. This chapter covers all modes of transport, but focuses mainly on the road system, since it has absorbed more than half of the total public investment program over the last five years. Of particular concern are the systems for planning, budgeting, and implementing maintenance works on the road system. 2. A number of country specific characteristics can explain the system's underdevelopment, and they are at the root of its failure. Lao PDR is a very mountainous, landlocked and thinly populated country about two-thirds the size of Vietnam (237,000 sq m), stretching 1,700 km from north to south, between China in the north, Vietnam in the east, Thailand and Myanmar in the west, and Cambodia in the south. Most of the 4.4 million people (about 6% of the population of Vietnam), are farmers and live in relatively isolated settlements, dispersed throughout the country, with long distances between settlements. Moreover, the country's small economy severely limits the resources to provide the population with an efficient, high quality transport system and services. In addition, years of hostilities, lack of commitment to preserve the country's transport assets, conflicts about the roles of central and provincial authorities in system management and service provision, and weak institutions responsible for transport infrastructure have contributed to the decline of existing assets. 3. Official records indicate that the road network comprises about 14,800 km and is classified into about 4,100 km of national roads, 5,800 km of provincial roads, and 4,900 km of district roads. However, these statistics are misleading because of inconsistencies in the classification system and even in the definition of what constitutes a road. In theory, the national roads should be the backbone of the network; provincial roads should provide the main links with the national road network within each province; and district roads should facilitate inter-village communications, farm to (local) market access, and natural resource development (e.g., logging). 4. In practice, the network has fallen into severe disrepair and many roads cannot be used during the rainy season, except for some recent donor-financed improvements to the national roads. Moreover, the design standards on most roads are inadequate for heavy commercial vehicles now in use, which contributes to rapid road deterioration, and only about 40% of national roads and 7% of provincial roads are paved: the remainder are either gravel or earth. Even paved sections are in such bad condition that they are unmaintainable and should be reclassified. A sample survey of national roads carried out in 1993 found, for example, that only 10% were in good condition. Many key sections in the road network are so degraded that they have to be reconstructed and, even more worrisome, as a result of inadequate maintenance half of the roads completed since 1980 have deteriorated to such an extent that they are now classed in only fair condition. 5. River transport has shrunk in relative modal share since the improvement of the road network, according to operators. There are about 1,190 km of sections of the Mekong River which are navigable, depending on the season and other factors. Seasonal constraints in depth, difficult - 10- channels, and lack of navigation aids prevent the efficient use of larger cargo vessels. Strong currents and extreme fluctuations in water levels make construction and operation of river ports difficult. For many small communities, however, river transport remains the only means of access for goods and people. 6. Air transport consists of one international airport in Vientiane and about 16 small airports and airfields distributed throughout the country. Their efficiency is low at present and they are in need of rehabilitation and modernization, as is the air traffic control system. 7. There is no railway at present. However, survey and design studies have been carried out for a possible connection of Vientiane to the Thai rail system at Nong Khai, using the Mittaphab Bridge. ADB Subregional Cooperation studies, as well as Chinese and Thai interests, have also considered the eventual linking of Chinese and Thai rail systems through Lao PDR. B. GOVERNMENT OBJECTIVES FOR TRANSPORT 8. During the Third Five-Year Plan (1990-95), the main emphasis in the Plan and the transport sector was on improving the primary national road system. Road improvement priorities were to improve the main Road No. 13, which is the north-south spine of the basic national network, and to provide access to the southern agricultural areas such as the Bolavens Plateau. The upgrading of Road No. 13 has experienced delays and is now expected to be completed by the year 2000. 9. Lao PDR has also sought to improve east-west linkages to Thailand and Vietnam, including the construction of the Mittaphab Bridge across the Mekong near Vientiane and Roads No. 9, 8, and 20 as "access to the sea" and in the service and development of transit trade between China and Thailand. Interprovincial roads and the linking of provincial capitals to the national road network have also been priorities. 10. Parallel with the development and restoration of transport infrastructure, a beginning has been made to reform and simplify restrictions and regulations on trade and travel. The Government has begun to simplify regulations for goods transport, to commercialize and privatize state enterprises, and to encourage the private sector to participate in the investment process. However, domestic and international traffic regulations still remain cumbersome. As pointed out in numerous ESCAP reports, compared with other Pacific Rim countries, the entire region has lagged in facilitating trade and developing international conventions for the movement of goods and vehicles including appropriate pricing and vehicle control measures. 11. During the Fourth Five-Year Plan (1996-2000), the Government's transport development objectives seek to continue the previous efforts of network development and to deepen institutional and regulatory improvements. They support the Government's broader policy goals of completing the transition to a market based economy, attaining regional balance, achieving integration into the international economy, and developing the human resource base. => Important specific targets for institutional improvement in transport are: * To reform administration and processes, including improving traffic safety; reforming the road classification system; devolving responsibility for certain road classes to local authorities; establishing design standards for specific classes of roads; improving the - 11 - Government's capability to manage local contractors and consultants; strengthening capabilities to formulate and implement construction projects; and improving road maintenance * To accelerate economic reforms relating to transport -- privatization, financial discipline, and accountability * To pursue regulation reforn: to adopt a Road Traffic and Transport Act and to facilitate cross-border movements and vehicle control. => The targets for the development ofphysical infrastructure include the following: * To continue the construction/rehabilitation of the trunk road system * To invest in international transit corridors to diversify access to seaports and promote the country's potential role as a land bridge, including investing in roads facilitating the trade of border provinces with neighboring countries * To (re-)construct interprovincial and intraprovincial roads to improve access to rural and remote regions * To maintain and preserve the transport network. 12. Although the program seems very ambitious, the scope and balance of the institutional improvements appear reasonable, especially since a number of different government units are likely to formulate and/or implement them (assuming they are supported by the decision makers). The draft Road Traffic and Road Transport Act, under review by both the MCTPC and the Ministry of Justice, will simplify the incoherent mix of decrees, regulations, and directives which at present make consistent administration difficult. An implementation strategy design is needed. However, the existing institutional, human resource and project management and local funding weaknesses are unlikely to be overcome rapidly enough to deliver the desired physical outcomes. Details are discussed in the review of the medium-term investment program (Chapter 7). C. FUNDING MECHANISMS 13. Domestic Funding. Three major sources fund capital and recurrent costs in the transport sector: official external assistance from bilateral or multilateral donors (ODA); local funding, and private sector finance. In addition, there are contributions in kind, especially in the rural areas, but their extent is not known. As regards local funding, the issue of cost recovery from users will also be reviewed. 14. Foreign Funding. Foreign aid has been crucial in financing road construction with international donor programs and commitments amounting to about US$500 million. As detailed in Appendix Table A2.1 at the end of this chapter, foreign funds come from many sources. Among multilateral institutions, ADB has been the largest contributor, with road projects totaling over US$250 million in transport, urban, and rural development sectors. IDA is next with planned or committed projects in the order of US$150 million. Among bilateral donors, Australia, Japan, and Sweden have the largest programs. 15. Over the years, external funding has taken on increasing importance and has replaced local funding. For example, during the early to mid- 1980s, foreign funding amounted to 65 to 70% of the - 12- road investmnent program. The influx of Western donors and multilateral financing institutions in the late 1980s increased the donor share to an average of 80% of capital expenditure and over 50% of spot improvements and emergency repairs included under maintenance expenditure. 16. Looking to the future, it appears that the volume of financing through bilateral and multi- lateral institutions will decrease. In addition, two major lenders, ADB and SIDA, have both indicated that future road sector programs may focus more on rural access and feeder roads, with a winding down of trunk road programs and maintenance support. Thus, the total amount of foreign funding for national and provincial roads will probably be substantially less than was hoped for in the strategic and long-term plans of the Ministry of Communication, Transport, Post and Construction (MCTPC) and somewhat below the SPC proposals. Should this be the case, the shortfall could probably not be entirely compensated for by local funding. 17. While the high share of official development assistance (ODA) in funding capital expenditure is not unusual for a low income country such as Lao PDR, it is of concern that ODA seems to have crowded out local funding. On the one hand, this increases dependence on foreign funding, and on the other, it tends to weaken local ownership and accountability. Another concern is that, as grace periods expire in the next decade, debt repayments will mount. For example, it has been estimated that by the year 2000 road sector loan repayments will amount to about US$6 million, and will double again by 2008. 18. Local Funding. The decrease in local funding is the mirror image of the increase in ODA. In its review of the implementation of the 1991-95 PIP, the Government identified insufficient provision of localfunds as one of the major shortcomings of the program, causing delays in project implementation and payments to contractors. In some instances, even when local funds had been budgeted, payment was often delayed owing to stifling bureaucratic procedures, or was not fully paid out at all and instead was used for other purposes. The problem seems to have been particularly acute with the funding of recurrent costs, since they are part of the annual budget managed by the Ministry of Finance (MoF) and thus, rather than being based on actual maintenance requirements, they are subject to short-term fiscal considerations and funding pressures typical of this process. Also, in some cases, the current year budgets contain large provisions for payment of arrears to contractors, which further reduces the work that can be carried out during the current budget year. 19. Private Sector Finance. To date, private sector finance for transport infrastructure has not been tapped as a funding source. However, there are plans to increasingly resort to this instrument, particularly for international transit corridors and roads used primarily for the extraction of natural resources (e.g., logging). The scope, and limitations, of private sector finance are discussed in detail in Section E. 20. Cost Recovery. Revenue from road users is raised from (i) taxes on fuel used on roads; (ii) import taxes on vehicles; and (iii) annual vehicle license fees. For 1994/95, sector tax yield has been estimated at US$15 - 19 million, or roughly 15% of total tax revenue. This is road users' contribution to the costs of operations and maintenance (O&M) of the roads, although from a fiscal point of view it is part of the Government's general revenue. 21. A more difficult question is whether this amount compensates for the wear and tear road users inflict on the roads. Economic theory suggests that optimal road use (and thus resources use) is - 13 - assured if road users pay for the incremental road damage they cause.' This means that they should pay at least for the variable cost of road maintenance and, under certain conditions, make a contribution to the fixed cost of road maintenance as well. It could also be argued that road users should pay for spot improvements (although this activity cannot strictly be considered maintenance). Estimates of the costs of road maintenance (routine and periodic) and of spot repairs, taking into account a feasible reconstruction program, amount on average to about US$3.6 million and US$1.2 million, respectively, per year. In addition, about US$1 million per year should be added for bridge maintenance. Comparing this with the total yield from road user specific charges of US$15-19 million per year, it is evident that from the point of view of economic policy there is no need to increase user charges at this time. 22. However, the answer for Lao PDR is complicated by the fact that the majority of the road network is dysfunctional or has disappeared. Consequently, the accumulated backlog of reconstruction/rehabilitation is enormous, exceeding US$1 billion by some estimates. If road users were made to finance this task as well, massive increases in user charges would be required. This may be expedient from a fiscal point of view. It should be noted, however, that the likely long-term economic distortions in factor and product prices, and decisions about the location of economic activity, vehicle choice, and so on, resulting from massive increases would far outweigh the short- term fiscal advantages. It is therefore recommended that construction or reconstruction/rehabilitation be funded through long-term borrowing and general revenue (including any surplus from economically appropriately determined road user charges) rather than through earmarking. 23. The crux of the problem seems to be that the Ministry of Transport is reluctant to regularly fund systematic road maintenance and instead prefers new (donor-financed) investment and reconstruction of existing roads. Earmarking of levies collected from road users for maintenance purposes is not efficient, since it restricts budgetary flexibility and the likely revenues would be insufficient to cover maintenance requirements.2 The MoF should thus continue to retain the revenue from road users for general budget purposes. As is discussed in Chapter I of the Main Report, the MoF should also increase its efforts to improve the efficiency of tax collection to raise additional revenue and create more equitable taxation. Measures to increase the tax yield from the road sector include: * Enforcing imports controls through gateways by adding trained staff and comprehensive data collection/monitoring systems and tools for Customs and Taxation Departments. * Rationalizing the vehicle taxlfee regime to reduce exemptions, evasion, and avoidance by charging fees that more closely reflect relative road damage responsibility and ability to pay. * Establishing a national vehicle registration data base and adopting an annualfee system for active vehicles which show paid status with a clearly visible decal. * Eliminating distortions in the structure of road user charges. For example, heavy vehicles do not appear to pay charges commensurate with the damage they inflict on See A. A. Walters, "The Economics of Road User Charges," World Bank Staff Occasional Papers, No. 5, Baltimore, 1970. 2 For a discussion of cost recovery and earmarking, see Chapter I of these Background Papers. - 14- roads. Another fiscal issue is that fuel prices in Lao PDR should be similar to those of neighboring countries to discourage smuggling and tax evasion. Thus, there is considerable room to increase the yield and improve the structure of user charges without increasing overall rates. D. EXPENDITURE TRENDS 24. Improving the transport system has been the Government's first priority in the 1991-1994/95 Third Five-Year Public Investment Program (PIP). The sector PIP share for the communications and transport sector increased from 4.1% of GDP in 1991 to 5.2% in 1994/95. It peaked in 1993/94 with 7.1% of GDP, or 76 billion kip (Table 2.1). This is high if compared with other countries in Asia (e.g., Indonesia, Korea, Vietnam) at a similar stage of development when capital spending varied between 2.2% and 3.3% per year. However, slightly higher spending levels of, say 3.5-4% of GDP would probably be justified in view of the urgency to provide the country with at least a core transportation system covering basic transport needs. 25. Transport programs absorbed nearly half of the Government's capital expenditure of 500 billion kip over the 1991-1994/95 period. The rate of growth in public spending for communications and transport was on average 22% per annum in local currency terms and 17.5% in U.S. dollar terms during the period, which is only marginally higher than the rate of growth of the remainder of the PIP (excluding transport). This growth compares with an annual growth of about 14% of GDP in local currency terms and 9.5% in U.S. dollar terms (current prices) for the same period. 26. In addition, the PIP provides an incomplete picture ofpublic sector expenditure, since not all spending on transport infrastructure has been included. For example, the US$30 million Friendship Bridge between Thailand and Lao PDR, crossing the Mekong in Vientiane, is not part of the PIP. TABLE 2.1: COMMUNICATIONS AND TRANSPORT SECTOR SHARE IN 1991-95 PIP (BILLION KIP; CURRENT PRICES) 1991 1992 1/ 92/93 93/94 94/95 Total Total Investment 69.1 82.2 65.6 131.8 150.9 499.6 Comm. and Transport 29.4 37.4 32.2 76.0 68.3 243.3 C&T as % of Total 42.5 45.5 49.2 57.7 45.3 48.7 GDP 722 848 942 1,069 1,323 4,904 C&T as % of GDP 4.1 4.4 3.4 7.1 5.2 5.0 27. Composition and Magnitude of Major Program Expenditures. Road improvement and reconstruction has taken up the largest share of the MCTPC investment budget with more than 95% of the sector investment since 1990. This share includes expenditure in the maintenance category. National roads absorbed more than 80% of the allocated funds, and provincial roads the remainder. In the 1995 and 1996 allocations, provincial road construction and maintenance amounted to 20% and more. Expenditure for river transport and aviation absorbed about 1% and 3% of sector investment, respectively (Table 2.2). - 15 - TABLE 2.2: LAO PDR TRANSPORT INVESTMENT PROGRAM, 1991-95 (US$ MILLION) All National National Road Inter- River Civil Total provincial Trans- Roads Road Maint. Roal Transr Aviation Road (*) port Expenditure 317.1 305.1 234.8 23.6 46.7 2.8 9.2 % of Total 96.2 74.1 7.4 14.7 0.8 2.9 (*) Including maintenance. 28. Road improvement priorities were concentrated on upgrading main Road No. 13 which is the north-south spine of the basic network and provides access to the southern agricultural areas such as the Bolavens Plateau. Although originally due for completion before 1996, Road No. 13, is now not expected to be completed before the year 2000 because of delays and cost overruns, which required that the Government seek supplementary funding. In addition to Road No. 13, Lao PDR has also sought to improve roads linking provincial capitals to the national road network and to develop east- west linkages to Thailand and Vietnam, including Roads No. 9, 8, and 20, as "access to the sea," and to facilitate transit traffic between China and Thailand. 29. Economic Composition of Spending. As is discussed above, the bulk of spending during the period of the PIP was devoted to the reconstruction of the trunk network and, likewise, the bulk of spending in the maintenance category was directed toward ad hoc emergency operations designed to prevent segments of the dilapidated network from total collapse. Given the state of the network and the fact that systematic road maintenance (i.e., scheduled routine and periodic maintenance) is possible on only about 1,000 km of road, the heavy emphasis on reconstruction appears justified for the time being. Of course, once more roads have been reconstructed and need systematic maintenance by about the year 2000, the balance should shift somewhat more toward the maintenance side. 30. As regards wages and salaries, the amount spent on administration and supervision is relatively small in comparison with the number of projects being managed. For example, the administrative budget for MCTPC amounts to only 2% of total expenditures, in addition to personnel expenditures included in project budgets. While the tight restraint on administration expense is admirable, the composition of this expenditure--particularly for critically needed skills and staff support--should be examined closely by the MCTPC's Executive and Manpower Development Division. More skilled people are needed in the MCTPC in view of the work load of senior management in the Ministry and the lack of technicians and young professionals to process data, carry out analyses, and prepare staff work for policy, planning, and administration. 31. Allocations by Provinces. Within the transport and communications sector investment budget, about 15 to 18% of the total PIP has been allocated to provincial governments in recent years. The 1995/96 PIP budget for the transport and communications sector at the provincial level amounts to about 14 billion kip. Most of the funds are allocated to road construction and upgrading. Minor sums are also allocated for river port improvement and airport construction. - 16 - 32. Allocations by province vary considerably (Appendix Table A2.2 to this chapter). Data available for 1995/96 show, for example, that almost a quarter of the provincial investment program was allocated to Vientiane Prefecture and Province (4.1 billion kip). Another quarter was allocated to the northern provinces of Phongsaly and Luangnamtha, and Savannakhet received 1.3 billion kip. The remainder was shared by the other provinces. In general it appears that equity among provinces is a consideration in the allocation of funds. However, other factors seem to play a role as well, since, for instance, such provinces as Huaphanh, Luangprabang, and Oudomxay received a smaller investment per capita than is indicated by their relative needs. 33. Unit Costs and Sources of Variation. Considerable variation has occurred among unit costs and contract prices for various projects undertaken during the 1990-95 period. The range in type of terrain and the standard of construction/reconstruction explain part of this variation. Other differences can be traced to the geographical isolation of some civil works, the low level of competition for local contracts, the lack of a level playing field between state and private enterprises, unrealistically low unit prices set by the state, and low bids in ICB competitions which has resulted in substantial cost overruns. E. SCOPE FOR PRIVATE SECTOR PARTICIPATION 34. There are various ways for the private sector to participate in the development of the transport sector. In transport infrastructure, the private sector can finance, contract, and operate; in the transport industry it can own and operate transport enterprises. Although state policies express the desire to continue the privatization process for non-strategic enterprises and to increase the efficiency of those enterprises kept in the in the public sector, progress in the implementation of this policy in the transport and construction sector has been slow and uneven. For public enterprises in various stages of transition to private ownership and management, the development of commercial, arm's length relationships with MCTPC or government customers has not yet matured. The Plan acknowledges that the privatization process has not proceeded according to the policies set out by the Party--that progress has been slow and corruption has occurred. As is stated in the MCTPC 1995- 2000 Plan, "Privatization is still one complicated and confusing problem." 35. Private Sector Financing of Transport Infrastructure. With the development of Yunnan Province to the north, the booming economy of Thailand to the west, and the prospect of future trade growth in the Vietnam and Myanmar economies, Lao PDR's potential as a land bridge has spawned public and private investor proposals for transit corridors across the country. Common to many of the projects are the vagueness of the agreements and their apparent handling outside the PIP and budget system. One example is the upgrading of Road 1 between Houayxai and Boten to provide a transit corridor between China and Thailand, with a Thai-Lao joint venture sharing toll revenue and income from development projects along the corridor. Other examples include the construction of Road 8B as part of the Nam Theun Dam project, and the building of a connector between Pakxon and Road 16 as part of a dam project in Champasak. The latter is a joint venture for both hydro development and logging with the private loan to be repaid through profits. Development roads are also being constructed/upgraded by private and state enterprises to exploit other logging concessions. The recently initiated six roads program for Vientiane Prefecture entailing the major upgrading/new construction of 63.4 km of arterial roads in the city has also been announced -- reportedly with commercial loans mobilized by bidders. It is understood that these loans are to be repaid by the Government but the terms and conditions are not known. Another project being considered by ADB - 17 - is the co-financing of the East-West Corridor Project, which would include a bridge across the Mekong at Mukdahan, and connections from Savannakhet via Road No. 9 to Vietnam and the Port of Dong Ha. ADB reportedly is considering a 60/40 public/private sector financing with cost recovery through tolls. 36. Private sector participation in the funding of transport infrastructure can be beneficial, provided the economic benefits to society exceed the financial and economic costs (including the exploitation of natural resources and the costs of environmental degradation) of private provision of transport infrastructure. It is doubtful whether the low volume roads now being planned and the agreement being reached in Lao PDR can be included in this category, since it is economically inefficient to use private sector finance on low volume roads. Moreover, the Government should exercise extreme care with agreements where construction and performance standards have not been defined. First, restricting access to underutilized roads would be a waste of scarce resources (assuming roads are tolled) and second, the potential toll income is unlikely to cover road development and/or operating costs. Hence, the Government has to provide substantial additional incentives, such as timber and/or mineral concessions and other subsidies. Third, the Government might incur contingent liabilities if concessionaires were to default on their contractual obligations, or the Government might incur unplanned expenditure for the maintenance and operation of a road when the concession agreement ends. 37. The Private Sector as Contractor in Construction and Maintenance. There are 17 state enterprises capable of carrying out road construction and maintenance: 13 are under the MCTPC; another is under the Ministry of Defense; 2 are irrigation enterprises; and I is in a joint venture with the Swedish contractor on Road No. 13S. According to the ADB-sponsored 1994 study, Privatization and Management of Road Sector Institutions (PMRSI), the four largest companies were estimated to have an annual turnover of 1 billion kip and could, if combined, have the capacity to construct 100 km of road per year. The remainder combined could construct 100 km of gravel road per year, including small bridges. At the provincial level, most of the enterprises are considered privatized, but operations are small, with annual turnover in the range of 50-200 million kip. Annual total capacity of the provincial enterprises is estimated to be in the order of 2 billion kip of work, equivalent to the reconstruction of about 100 km of provincial and local road. 38. The prospects of a domestic contracting industry in a fair and competitive environment are remote at present. Difficulties arise because of the Government's lack of experience in contract management and supervision, the uncompleted transition of state companies to the private sector (which tends to distort fair competition), the lack of competition in remote parts of the country, company cash flow problems because of late payments, unrealistic unit prices set by the state, and, on the side of the companies, a lack of management, financial, and administrative skills, poor equipment, and undercapitalization. 39. The Ministry and the contracting industry are slowly moving toward a system in which measurable work (which can be itemized in a bill of quantities) should be let under competitive bidding conditions. This assumes that the system of contract procurement is equal for qualified public and private companies. Competitive bidding for locally funded projects is being introduced with the stipulation that negotiations should be carried out with at least three companies before contract award (Directive No. 1103/SPC, 14 December 1994). While this appears to be feasible for larger contracts, particularly if contractors are free to work throughout the country, minor projects on provincial roads do not attract enough interest for competitive bidding. - 18 - 40. Because of the supervisory and structural difficulties experienced in contracting out routine and emergency maintenance, the Road Administration Division (RAD) has set up "technical contract units" which have a status similar to state enterprises. These units will maintain only part of the national roads, and the rest will be maintained by negotiated contract. 41. The experience gained to date shows that a transitional period of several years will be required to develop contractor skills in financial and project management. The ADB and IDA supported programs to build up a wider and deeper domestic construction capacity will require intensive coaching and a long-term commitment to technical assistance. Establishing a competent, financially strong local contracting industry may take much longer than was anticipated. 42. The Road Transport Industry. Privatization of the road transport industry has been a success. With the New Economic Mechanism and the removal of many restrictions on private carriage of freight and passengers, the road transport industry is now dominated by the private sector: roughly 80% of goods and passenger vehicles for hire are privately owned. Fairness in taxation between state and private enterprises and large and small operators has been a concern, with the need to adjust turnover tax rates and profit taxes to ensure broader and fairer coverage. 43. In freight haulage, more than 40% of vehicles are independently owned; another 40% are owned by members of associations or private companies, and the remainder are state owned. Of the state-owned fleet, many vehicles have been leased or sold on credit terms or put into joint ventures. Only strategic transport activities such as fuel delivery and essential service to remote areas are intended to stay in state hands. 44. Passenger transport is carried out by a mixture of private associations and companies, some of which were former state companies. Many of these are now owner-operator companies and family concerns. In rural areas, truck-buses are common because of the state of the roads and because of the freight carried by passengers. F. EFFICIENCY OF PUBLIC SPENDING ON TRANSPORT Internal Eff ciency 45. Key issues in delivering efficient transport service to the user and in fulfilling the MCTPC's mission of access at reasonable cost are: (i) planning; (ii) management and budgeting; (iii) staffing and human resources; and (iv) maintenance of roads and bridges. 46. (i) Planning. While the sizable planned transport investment seeks to support national and regional development goals, three issues are urgent: => Evaluating trade-offs between programs competing for funds (including the concurrent development of international, regional, and local linkages) = Assessing and comparing relative returns and priorities among projects > Estimating required complementary investment and potential expanded benefits through network connectivity. Projects should not be proposed in isolation. - 19- 47. As is discussed in Chapter I of these Background Papers, projects must be ranked or screened in the context of a coherent sector strategy to avoid dissipating managerial and financial resources across a diffuse spectrum of wants and needs. This "wish list" approach gives the lenders, who may have their own priorities, de facto screening power. The donor-driven approach, however, risks loss of ownership and accountability on the part of the Government and is ultimately counterproductive for the donor community as well. Experience has shown that many donor- financed projects have not been properly maintained. In this context, it should also be mentioned that the entire pool of resources available to the Government should be subject to the same rigorous tests of project selection advocated here. Therefore, all projects, irrespective of the method of funding (e.g., grants), should be included in the Govemment's investment program. 48. Projects proposed in isolation often underperform, experience reduced benefits until tied in with the rest of the network hierarchy, and jeopardize complementary network improvements and their potential retums to the national economy--which hampers, rather than promotes, greater regional integration. 49. (ii) Management and Budgeting. When the MCTPC was reorganized, it was anticipated that authority would be divided and delegated between policy and operations at the central level; in practice this has not occurred. Most decisions are still passed to the senior level, which increases the work load and hampers organizational efficiency. A direct outcome of this practice is the inadequate preparation and vetting of long-term policy and priorities. In addition, planning cycle time is increased and prompt decisions required for seasonal works are delayed. 50. Because rights and responsibilities are unclear between the center and the provincial authorities for planning, programming, budgeting, and other matters (e.g., senior appointments), there is confusion, wasteful spending, loss of accountability, and friction among decision-makers. Decision 21/PP regulates the sharing of responsibility between the MCTPC and provincial administrations. In theory, the MCTPC holds responsibility for overall policy direction, quality and budget control. In practice, however, the DCTPC appears to take its direction from provincial administrations, with limited reporting to the MCTPC. The lack of control over budget planning and implementation is also reflected in the poor and uncertain reporting offoreign and local funding commitments. Backlog payments to state enterprises is another area in which no accurate or comprehensive record has been maintained. 51. In addition, provincial governments have much influence and little expertise in the annual and long-term budget preparation process, which results in low quality project proposals. Moreover, since project funds come from the central govemment, the provinces have a strong incentive to propose as many projects as possible. The lack of responsibility on the part of the provinces for funding also leads to a lack of budget discipline in commitments which exceed budget allocations. 52. (iii) Staffing and Human Resources. Technical capabilities at the central and provincial levels are insufficient to manage and control the number and scale of projects under way in the transport sector or the complex requirements of intemational financial institutions. Too much effort is expended in bureaucratic, time-consuming procedures. A better coordinated program of technical assistance is required to meet the need for skilled and experienced technical personnel. 53. The shortage of trained staff at the central MCTPC and provincial offices has hindered the development and maintenance of data bases for assessing traffic growth pattems, as well as the - 20 - establishment of inventories and condition surveys of the national and provincial road system. For example, traffic count and origin/destination survey data collected in 1994/95 had only been partially processed by 1996. Timely information for feasibility studies has not been available and planning and decision-making, are delayed or based on guesswork. 54. The recently established Manpower Development Division has started to draft a long-termn plan and policy to improve the managerial and technical expertise of professional staff, but staff development will take time. To fill immediate and medium-term needs, shortfalls could be made up by contract staff from elsewhere in the region. However, long-term staff development plans with domestic and foreign training components need to be implemented and commitments made to increase the stock of human capital within the organization. The human resources development plan should be put into place as soon as possible to coordinate training and education components and tailor inputs to specific needs. 55. (iv) Maintenance. Systematic road maintenance is critical because the economic returns on this activity are generally much higher than new construction in terms of both reduced user costs and the extended life of the road assets. Unfortunately, despite receiving substantial technical assistance, equipment, and funding, the results in road maintenance have not been encouraging. Maintenance suffers from a mix of systems and practices, none of which appears to be effective. Works executed often do not correspond to the endorsed program. Technical skills are deficient and, as a consequence, the use of inappropriate materials and methods prevails. Even with donor-supported technical assistance on maintenance works, results have been disappointing. These problems have both an institutional structure and a financial dimension. 56. Institutional Structure Problems. There is confusion among different levels of government regarding the responsibilityfor maintenance. Maintenance on national roads is unclear in several cases; and the roles of the central and regional offices of RAD vis-a-vis provincial DCTPC offices are still in flux. Spot improvement and heavy maintenance programs are poorly coordinated, as is the phased graduation of stabilized road segments to normal recurrent maintenance. There is no systematically updated inventory of roads and bridges; information is not organized and is not exchanged within the Ministry or between headquarters and field. The lack of control over budget planning and implementation is also reflected in the poor and uncertain reporting of foreign and local funding commitments. Backlog payments to state enterprises is another area in which no accurate or comprehensive record has been maintained. These problems combined have resulted in the substandard quality of works planning and budget preparation. 57. The MCTPC Development Plan acknowledges the need to improve planning and implementation capabilities, but a strategy and a detailed plan are required to plot the development of a system for the efficient delivery of maintenance. RAD was set up in the 1993 reorganization of the Ministry, and several policy shifts have had impacts on organization and staff responsibilities. The new field organization, established in 1994 to plan, supervise, and inspect maintenance contracting for national roads, has been largely put into place; it has established maintenance regions in the south, the central region, and the north. 58. However, the 1995 decree devolving the responsibility for all roads to the provincial DCTPCs has again changed the role of RAD and its maintenance engineers.3 Under the new policy, 3 Further elaborating on the Decree No. 163/PM on the Establishment of Foreign Aid Management Committee, Temporary Regulation No. 201 (30 September 1995) discusses the division of responsibility between central and local authorities. - 21 - the MCTPC is responsible for technical designs and standards and the RAD appears to be responsible for planning, budget preparation, and supervision for national roads. The provincial governors are responsible for political and administrative matters, and maintenance on national and provincial highways will be the responsibility of provincial DCTPC offices. 59. This devolution of responsibility to provincial departments threatens progress and decreases system effectiveness because of the generally sparse provincial capacity to plan, supervise, and inspect maintenance works. In theory, delegation of responsibility to the lowest level has appeal, and, in the longer term, the more wealthy and sophisticated jurisdictions could assume much of the responsibility, but to expect all 16 provinces -- most of which have minimal staffs and/or physical resources -- to assume this task within the current planning horizon is unrealistic. 60. RAD has continuing responsibility for long-term planning, fund allocation, and broad supervision; it must also actively shepherd the DCTPCs to plan and supervise contracts and works on the ground. The regional offices of RAD will need to be strengthened for this purpose and for planning and supervising contracts for heavier work and periodic maintenance for a more appropriate regional structure. 61. At the local level, the phasing in of labor-based routine maintenance through contracts with local villages is being tested, but the extension of this system throughout the country will take careful planning and management. As demonstrated in Korea, this system can be a good means of furthering local ownership, but to avoid failure and disillusionment the approach requires technical support and close supervision as well as materials, tools, and backup equipment when works exceed the capacity of hand labor. 62. Financial Problems. Maintenance funding problems are a vicious circle, starting with (i) inadequate preparation of short-term and longer-term work plans and leading to (ii) soft justification of the annual budget request, which in turn invites (iii) subsequent cutbacks in allocations approved by the Ministry of Finance (e.g., by more than half in the last two years). 63. The funding problem must be addressed, and control and approval systems must be streamlined to alleviate the problems of late payments and cumbersome administrative procedures for approvals and disbursements. To remain viable, enterprises have to take on outside work, and as a consequence scheduled works are not carried out to established standards of quality, time, and cost. 64. To break this cycle, planning, budgeting, supervision, and inspection must be an integrated system from headquarters down to the regional, provincial, and district levels. Furthermore, forward budgeting or multiple-year entitlements that are keyed to RAD's rolling maintenance works plan should be worked out between the MCTPC and the MoF to secure the annual flow of funds required to sustain consistent maintenance. Financial and technical audits with reported measures of performance against plan targets should monitor the quality and cost-effectiveness of work. These audits should be developed as an integral part of the system. External Efficiency 65. Transport For transport infrastructure, targets set in the 1991-95 five-year transport investment program have been met. During this period the bridge between Lao PDR and Thailand, crossing the Mekong near Vientiane, was completed, and about 25% of national roads (1,000 km) were reconstructed for rehabilitation. Another 25% of national roads are currently under - 22 - construction. Only about 1,000 km of the remaining network can be used during the dry season and the remainder of about 1,000 km can normally only be used by special purpose vehicles. The majority of the reconstruction projects are donor financed and are managed by expatriate technical assistance. However, the institutional problems mentioned in the previous section have hampered project execution. Thus, for example, the planned 1996 completion of the all-important north-south trunk Road No. 13, between Pakmon and Pakse, has been delayed until at least the year 2000. 66. As regards airports, the Government is now implementing the Civil Aviation master plan with funding from several countries and technical assistance and coordination from ADB. Civil works are underway at Vientiane, with Japanese assistance for items such as the control tower and terminal and with French assistance for navigation aids. Construction of new facilities is proceeding at Luangprabang with OPEC, NDF, Thai and French funding, and improvement of service to other regional airports such as Pakse is also in the master plan. An airport project at Oudomxay is proceeding with local funds. 67. Traffic For traffic, the statistics are unreliable and do not appear to adequately reflect the growth in underlying economic activity. For example, official traffic statistics show a decline in freight and passenger traffic over the period 1993-95 of more than 30% for freight and more than 20% for passenger transport, despite a healthy GDP growth of 6-8% during that period. The principal causes of the deficient statistics are reported to lie in methodological and institutional weaknesses in data gathering and processing as well as widespread underreporting (traffic statistics are used as a taxation base). Nevertheless, there seems to be a substantial amount of pent up demand, since road traffic on restored sections of the national road (which is more accurately recorded) has grown by up to 25%, albeit from a low base. G. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM 68. Future Expenditure Trends. Future expenditure trends relative to the Fourth Five-Year Plan (1996-2000) are not yet clear, since there are two starkly different expenditure proposals for transport: one from the MCTPC and the other from the SPC (see Tables 2.3 and 2.4). The MCTPC proposal for from the Plan period projects capital and recurrent expenditure of US$805 million, which would amount to about 8.4% of cumulated GDP, up from about 5.5% of cumulated GDP in the previous Plan. In contrast, the SPC proposal projects capital and recurrent expenditure of US$444 million, down to about 4.7% of cumulated GDP. In terms of sector share to total PIP, the MCTPC program would absorb about 53% of total planned capital and recurrent expenditure and would therefore increase slightly, if compared with the high spending level of the previous Plan. The SPC proposal would reduce the transport share to about 26% of the total PIP, which implies a rebalancing of priorities in support of other sectors of the economy. 69. The MCTPC approach is based on the goal of developing, during the Fourth Five-Year Plan, an all-season road network between the center and the provinces, and linkages between districts, while at the same time constructing transit corridors and border crossings to facilitate international and bilateral trade. In contrast, the SPC proposes a strategy of incremental and selective system improvements. The SPC proposal would complete the north-south trunk road and improve access to northern, eastern and southern districts that currently have poor linkages between isolated and developed areas; it would also begin the construction of international transit links and border crossings. In addition, there is a rural road program included in agricultural projects; its size is not - 23 - known. Grant-financed projects do not seem to be included in the Plan. Table 2.3 compares the MCTPC proposal with the SPC draft and also shows the extent to which funds are already available or committed for the various programs. TABLE 2.3: FOURTH FIVE-YEAR PLAN (1996-2000) - TRANSPORT INVESTMENT PROPOSAL OF MCTPC AND SPC (US$ MILLION) Program 1996/2000 MCTPC Total PIP (SPC) Difference Ongoing and Total (1 minus 2) Committed -1I - -2- -3- -4- 1. China - Cambodia 125 86 39 118 2. National Roads 63 69 - 6 39 (Rural Areas) 3. Road No. I (*) 30 -30 0 3. Intl. Transit Links 127 38 89 32 4. Provincial Border 182 38 144 70 Crossings 5. Interprovincial R. 39 34 5 12 6. Intraprovincial R. 118 68 50 0 7. Maintenance (**) 82 25 57 0 8. River Transport 7 7 0 1 9. Aviation 49 49 0 39 Total 805 444 361 311 (*) Included in interprovincial roads? (**) Roads and bridges. Sources: MCTPC Proposal and draft PIP (SPC) 1996-2000. 70. Evaluation and Issues. Both proposals reflect the weaknesses in planning, programming and budgeting already noted in the previous Plan as systemic problems. There is no indication that projects have been ranked or screened using project selection criteria and likely budget constraints, or showing trade-offs and relative returns. While MCTPC's vision certainly could be considered as a long-term target, the implementation time frame seems unrealistic. As regards the SPC proposal, it is not clear on which basis the proposed projects were chosen and how they rank in terms of priority. In particular, the following issues should be considered: * Funding Constraints. The Government should reassess its transport proposals and focus on a coherent, high priority program of capital and recurrent expenditure, bearing in mind the likely funding constraints. The tendency observed in the past, to start projects without fully secured financing, is a wasteful practice and should be avoided. To date, the lion's share of total transport expenditure has been funded by donors and the share of local funding has tended to decrease since the mid-1980s. This trend is likely to reverse itself. Donors are redirecting support into other sectors of the economy, and local contractors are increasing their participation in project implementation and maintenance. The SPC assumes that local funding will increase gradually to about one-third of project cost. However, past problems with local funds - 24 - are not easily overcome. The MoF has re-allocated its project funds for other purposes, and delays in the timely release of local funds have caused delays in project executionas well. * Size of the Transport PIP. Compared with the suggested benchmark of 3.5-4% of GDP, the SPC proposal is slightly higher and the MCTPC proposal is twice as high. If realized, the latter would crowd out priorities in other sectors, especially if GDP projections turn out to be optimistic, as has been noted elsewhere in this report. It is also noteworthy that two-thirds of the funding required for the SPC proposal is already committed. Thus, redirecting fresh funding into other priority areas such as health and education would be much easier to accomplish. Moreover, if investment and maintenance in transport and communications were kept at 3.5% of GDP, new funding requirements could be reduced by about US$137 million (130 billion kip) to about US$403.6 million (383.5 billion kip). * Composition of Planned Expenditure. There are also sizable differences between the two proposals in the composition of planned expenditure. In the MTCPC plan, much greater emphasis appears to be placed on creating maximum year-round access within and between provinces and across international borders. In this context, it is not clear whether the multitude of proposed transit corridors and border crossings can be justified economically. The Government should clarify to what extent it expects private finance or participation. In addition, the substantial increase proposed under the heading "'maintenance" seems high at this stage, since the majority of the road network has fallen into such disrepair that it has become unmaintainable and needs to be reconstructed. It appears that the bulk of these funds is slated for "emergency" and spot repairs designed to keep as much of the network operable as possible. This strategy can be counterproductive because it is an ineffective and costly palliative that temporarily masks the underlying problem of structurally unsound roads. * Costs. Both proposals contain substantial "hidden" expenditures in the sense of underestimated costs. In many cases, cost estimates appear low for future projects in view of recent unit cost experiences, particularly for the work to be carried out and nature of the terrain. This seems to imply either a lower design standard or underestimation of costs to induce commitment to particular projects. Examples include the completion of the China-Cambodia trunk road which is likely to cost on the order of US$173 million (139.4 billion kip) -- rather than the US$125 million (77.1 billion kip) estimated by the MCTPC or the US$86 million estimated by the SPC. Recent experience also suggests that other estimates, such as unit costs of US$80 thousand/km for the improvement of Road No. 2, Oudomxay-Pakbeng, with new construction west to Hongsa, or the reconstruction of Road No. 9 for US$50 thousand/km, are too low. In both cases, the projects are likely to cost twice this amount or more if a satisfactory standard of work is to be achieved. The ongoing review of rates and costs for different types of works should provide more insight into current costs, bid rates, and regional variations. Consequently, project cost estimates should be revised accordingly. * Absorptive Capacity. It is unlikely that the existing institutional constraints at the level of the implementing agency and the level of local contractors can be overcome in a short time. For example, the MCTPC plan would be a fourfold increase of works per - 25 - year (from 200 km to about 800 km). Thus, the meager institutional and human resources would be stretched beyond the breaking point. The MCTPC and its regional offices have to date demonstrated limited capacity to administer large volumes of projects. At the same time, the use of foreign consultants should stay within reasonable limits to avoid eroding ownership and accountability. Although the MCTPC is planning to improve its project management capabilities and to strengthen the private sector, this process is slow and it would be unwise to strain its capacity. 71. Expenditure Options. The size and scope of the MCTPC proposal appear unrealistic for the given time frame. The proposal should also be justified and prioritized in line with the country's absorption capacity, and therefore a longer time frame -- for example, 10 to 15 years -- might be more appropriate. Even the SPC proposal (when revised) is likely to strain the sector's absorptive capacity. The options are (i) to decrease public transport expenditure to a manageable level; (ii) to seek private sources of funding; or (iii) to resort to a combination of both. 72. This report suggests selecting a few priorities that (i) best fit the Government's development strategy, (ii) are economically justified, and (iii) can realistically be completed within the stated time frame. First, it is important to define a coherent "economic " core network (commensurate with the existing funding and implementation capabilities) of, for example, a total of 2,000-3,000 km for an effective road maintenance and reconstruction effort. Such a program is proposed below and is estimated to require investments of about 257.6 billion kip (Table 2.4). The core network could be expanded if it is economically justified and is in line with funding and implementation capacities. Public funding could be supplemented by private sector finance. It would be important to involve the donor community in this approach, including donor projects funded by grants. The core program is based on the assumption that: (i) projects already ongoing and committed will be completed; (ii) additional projects will be included only if they are in an advanced state of preparation and are economically justified; (iii) no other sources of funding (e.g., private) are available for these projects; and (iv) cost estimates have been updated to the extent possible. 73. As is the case with capital expenditure, recurrent expenditure for routine and periodic maintenance should also be limited to a core maintainable network and should be funded as the highest priority ( i.e., before funding new construction), to keep the core network operable at all times at the designed standards. The advantage of this approach is that it would increase the credibility of requests for maintenance funding and would stay within the limits of the sector's absorptive capacity. In addition, some funding should be provided for spot improvements, emergency repairs, and the like, for other parts of the network where a minimum level of network connectivity and accessibility must be provided. Table 2.4 presents the suggested alternative transport expenditure program for the period 1996-2000 and compares it with the SPC proposal. - 26 - TABLE 2.4: SUGGESTED EXPENDITURE PROGRAM FOR TRANSPORT, 1996-2000 (BILLION KIP) Gov't. PIP PER Proposed Change Comments Bn kip % Bn kip % Bn kip China-Cambodia 77.1 19.9 93.0 36.1 16.0 Costs underestimated, stretch implementation National Rural Roads 58.3 15.0 33.9 13.1 -24.4 Stretch implementation Route No. 1 26.3 6.8 - - -26.3 Consider later Intl. Transit Links 31.2 8.1 8.9 3.4 -22.4 Only high econ. priority; seek privatefunds Prov. Border Crossing 31.9 8.2 32.2 12.5 0.3 Interprovincial Roads 28.7 7.4 9.7 3.8 -19.0 Stretch Implementation Intraprovincial Roads 63.7 16.4 - - -63.7 Consider later Maintenance 22.0 5.7 31.4 12.2 9.4 Adaptedfrom Road Maintenance Study Routine 15.3 5.9 15.3 Periodic 4.8 1.9 4.8 Bridges 4.0 1.6 4.0 Spot Improvement 7.3 2.8 7.3 River Transport 6.6 1.7 6.6 2.6 0.0 Aviation 41.9 10.8 41.9 16.2 0.0 Total Transport PIP 387.6 100.0 257.6 100.0 -130.0 PER proposed - 130 billion kips less Telecom. & others 125.9 125.9 0.0 Incl. telecom. and other infrastructures Total TransjTelecom. 513.5 383.5 -130.0 Note: Amounts shown in MCTPC strategic plan to 2000 do not appear to reflect requirements to complete Savannakhet- Pakse plus additional on Pakse-Vuenkham (ADB7); Road No. 1, Bolen-Nateuy-Pakmong (unfunded at present) is not included. Source: World Bankstaffestimates based on informationfrom MCTPC SPC consultants, and donors. H. RECOMMENDATIONS 74. Lao PDR is a landlocked country with a scattered, diverse population; it requires an efficient transport system to tie the country together and open markets for trade. In many parts of the country the terrain is difficult, and the rudimentary road system has been degraded by war damage and maintenance neglect. Restoring and improving the basic road network between regions and providing access to isolated provinces and districts will be a high investment priority for years to come. 75. The proposed program must be rebalanced and a "maintainable " network must be identified given the scarce financial resources and the limited absorptive capacity. Resources must be concentrated and scheduled to hasten the development of an integrated hierarchy of roads. Benefits to users and the economy from completed projects should begin to flow as soon as possible, which entails avoiding incomplete projects and isolated segments. The successful implementation of even a scaled down and refocused program will depend crucially on strengthening central and provincial sector institutions. Moreover, rights and responsibilities between the center and the provinces must be clarified and each level must be held accountable. The recommendations below identify five areas that need to be addressed to successfully implement the Fourth Five-Year Plan and to increase the efficiency of the transport sector. - 27 - Focus transport expenditure on a core program of about 257.6 billion kip. During the period of the Fourth Five-Year Plan the Government should aim at an overall level of capital and recurrent spending for transport and communications of about 3.5-4% of GDP. Within that resource envelope the Government should formulate a program of priority capital and recurrent road expenditure based on the concept of a "maintainable" core network of about 2,000 to 3,000 km. If additional resources should become available and absorption capacity could be expanded, the core network could be expanded with non-core projects, provided they are economically justified and fit into the overall program. The private sector could be invited to participate in the funding of priority roads provided the overall economic benefits to the country exceeded the costs, including the costs of environmental degradation. A core program comprising investments of about 257.6 billion kip is proposed for the Government's consideration. => Concentrate on systematic maintenance of the core network. Implementing the core maintainable network would greatly facilitate efforts to establish an efficient and (financially and operationally) sustainable system of routine and periodic maintenance (of roads and bridges). Available financial and human resources should be directed to routine and periodic maintenance of the core system and should be funded as the highest priority. Funding for spot improvements and emergency repairs should provide a minimum level of network connectivity and accessibility. To secure the steady flow of funds required for systematic maintenance, the Government should consider implementing a system offorward budgeting linked to the rolling maintenance work plan along the lines suggested in Chapter 4 of the Main Report. It is estimated that a maintenance program based on this concept would require about 31.4 billion kip for the period of the Plan. Improve organizational effectiveness. The administrative and process reforms envisaged for implementation during the Fourth Five-Year Plan would help resolve many of the internal efficiency issues mentioned in this report. In this context, particular attention should be directed to the following: (i) streamlining and simplifying the decision-making process within the MCTCP by clarifying responsibilities, delegating authority to the lowest appropriate level, and establishing appropriate monitoring and auditing systems; (ii) clearly defining the relationship between the center and the provinces and defining each level's rights and obligations; (iii) improving financial reporting and payment systems; and (iv) developing arm's length relationships with contractors, updating the unit price system, and ensuring contractor mobility among provinces. Cost recovery: Improve collection effwiency and eliminate distortions in the structure of user charges. The total yield from road user specific charges appears to cover the incremental road damage they cause on that part of the road system that is operable. For reasons of economic efficiency it is recommended that the accumulated backlog of road reconstruction should be funded from a combination of long-term borrowing and general revenue. While there is no need to increase user charges to fund road maintenance, it is recommended that the efficiency of tax collection should be improved and distortions in the structure of road user charges should be eliminated. The additional revenue collected in this manner should be more than sufficient to help fund routine and periodic maintenance of a core network. = Promote private sector participation selectively. Private sector funding of transport infrastructure can be beneficial if economic benefits to society exceed the financial and economic costs to the government of private provision. It is doubtful whether the projects now - 28 - being planned and agreements being reached with private investors in Lao PDR can be included in this category. Therefore, the Government is advised to exercise extreme care with such projects and evaluate the financial and economic costs to society before concluding any such agreements. By contrast, it is recommended that the Government vigorously pursue the involvement of the private sector as contractor in construction and maintenance and eliminate, as soon as possible, any remaining obstacles. - 29 - - ~CHINA i i LAO PEOPLE'S DEMOCRATIC REPUBLIC mg / 1 r, \ & f \ VIETNAMr | 7 K*errso \~h VIENqTIANE \ \ KH+OU E i # s !' ..E<1 X -k~~~~~~~~~ 4t.j k E Z 8&s,, 7§ ~~~~~~To Quarig Tn | Airposh | 9t\ ~~~~~~~~~~~~~SAVANNAKHET _ L~~~~~~~~~~~~~~~ S o loed Tovns Fiat=htai\\< | s + W 13~~~~~~~~~~( Proyince Hecadqucarters \J \SLVAN | _ - ~~~~~~~~~~~~~~Province Boundori" t) do | + _ I~~~~~~~~~~~~~~~nternaobonol goundories \ i E2N | R;J5~~~~~~~~~~~~~~~~~~~~~HAMPASAK | ~~~~~KILOMETER5tS 5,0 Igo liO gp/"t' MP-ES O 5 '0 1~00JI*<1 F vS CAM80DIA < 2 ~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~ToPhniorn Penrh,% ? j ts4- Iq \~~~~~~~~~~~~~ ~~~~~~~~~~Hoh MkhhCity ( ALMt Resuc Maty 1996 - 31 - APPENDIX: EXTERNAL ASSISTANCE -- TRANSPORT AND RELATED SECTORS 1. ADB has been the largest donor to this sector, with US$250 million committed or planned in the transport and communications sector as well as road components in urban and rural sectors. ADB is also considering the possible financing of a new Mekong Bridge near Savannakhet. The next largest donor is IDA, with three road projects and two provincial/uplands development projects with road components, together totaling over US$90 million. With the inclusion of technical assistance and maintenance programs, the ongoing IDA program would amount to over US$100 million. IDA 3, now under consideration, may add another US$40-50 million to the portfolio. Both ADB and IDA programs are funded with soft loans with 10-year grace periods and annual service fees on drawn funds of about 1%. 2. UNDP has provided technical assistance funding totaling US$2.7 million in conjunction with the ADB 2 and 3 programs on Road 13. The UN Capital Development Fund (UNCDF) with UNDP as project administrator, is funding improvements to Road IA between Xiang Ngeun and Paklay. UNCDF and other agencies such as IDA and OPEC have also funded local roads as part of rural development projects administered by other ministries. 3. Sweden's program through SIDA has provided about US$35 million for the upgrading of Road 13. The Nordic Development Fund, also administered through SIDA, participated in funding bridges on Road 13. Australia, through AusAID, provided US$32 million for construction of the Mittaphab Bridge and another US$14 million for bridges on Road 13. Germany, through KfW, is financing improvements to Road 6 with a US$10 million grant. In addition, BMZ, with ILO acting as an executing agency, is financing US$1.2 million for the development of feeder roads with labor- intensive methods. 4. Japan, through JICA, has donated funds for the purchase of equipment for Road 8, and bridges on Road 13, as well as carrying out a feasibility study of another crossing of the Mekong at Pakse. It is anticipated that this bridge project will proceed in 1996/97 with Japanese grant aid. Japan is also a large contributor in the air sector, providing funds for a new terminal, control tower, and equipment at Vientiane International Airport (Wattay). While the draft paper prepared for the Donor Coordination Meeting by SIDA consultants listed air sector contribution at US$17 million, the total contribution is reported at US$45 million by the Embassy of Japan. 5. France is also participating in the air sector, with grants for the purchase of air navigation equipment. UNDP has provided technical assistance for civil aviation and development of the Aviation Master Plan. ADB is acting as coordinator for implementation of this Master Plan (see Tables A2. 1 and A2.2). - 32 - TABLE A2.1: EXTERNAL ASSISTANCE FOR TRANSPORT SECTOR a/ (US$ MILLION) Approval Planned Total Cost Remark Year CompL Year (IncL local) b/ Project TtI. Program ROAD SUBSECTOR ADB I 19S3 1988 8.0 8.0 8.0 Rds. Vte. prov. ADB 2 1986 1995 12.0 12.0 16.2 Rds. Bolovens Plateau ADB 3 1987 1994 19.0 23.8 23.8 Rte. 13 Vte-Vangviang ADB 4 1989 1996 33.0 39.0 42.8 Rte. 13 Vangviang-Luangprabang ADB 5 1991 1997 34.0 34.0 41.6 Rte. 13 Luangprabang-Pakmong ADB 6 1993 1999 26.0 26.5 33.0 Rds. Bolovens Plateau ADB 7 1995 1999 40.0 40.7 51.0 Rds. Champasak ADB 8 cJ 1997/98 30.0 30.0 43.6 Rds. Xiengkhoang ADB Urban Infra. 1996 20.0 Vte. Pref. urban infra. incl. rds. ADB E/W Transp. d/ 1997 125.0 Co-finance Mekong Br., Rte. 9 ADB Rural Access d/ 1999 20.0 Rural Access rds. Subtotal ADB 242.0 IDA So. Transp. Proj. 1987 1994 14.1 14.1 15.5 Rte. 13 Sav.-Pakse IDA I e/ 1991 1996 45.0 50.0 52.6 Rte. 13 Pakkading-Sav. IDA2 1994 2000 18.7 30.0 33.7 Rte. 13Sav.-Pakse IDA 3 c/ 1997 2000 40.0 60.0 65.0 IDA Upland Dev. 1993 10.0 Irrigation & 300 km of rral rds IDALuangnamtha 1993 2002 4.1 7.4 10.2 Ruralaccess&watersupplies Subtotal IDA 131.9 SIDA I 1990 1994 23.6 23.6 Rte. 13 Vte-Pakxan SIDA2 1993 1997 11.9 11.9 18.9 Rte. 13 Vte-Pakkading AUSAID 1 1991 1994 32.0 32.0 39.0 Mittaphab Br. AUSAID 2 1995 1997 14.3 14.3 14.3 Rte. 13S.3 Br. JICA I 1993 1994 7.5 7.5 12.4 Rie. 8, equipment JICA 2 1994 1995 12.6 12.6 12.6 Rte. 13S, 13 Br. JICA 3 1995 Feasibility. Pakse Br. KIW 1994 1998 10.0 11.0 Rte. 6 INDP 1987 1993 1.1 TA forRte. 13 &20 (ADB 2) UNDP 1988 1993 1.6 TAforRte. 13N (ADB 3) UNDP/UNCDF 1992 1996 5.8 Rte. IA to Paklay NDFf/ 1994 1996 7.0 7.0 Rte. 13S, Br. BMZ/ILO 1994 1996 1.2 1.2 FeederRds,2Prov. Subtotal Roads 502.5 WATER SUBSECTOR UNCDF 1984 1993 2.7 Repair/re-equip Thanleng port AIR SUBSECTOR UNDP 1990 1993 0.7 TA for Aviation Master Plan UNDP 1993 1994 0.1 TA for Civil Aviation France 1993 2.1 Air Navigation equipments Japan 1994 1998 17.0 Vhe. Intl Airport renovation Japan 1995 1998 28.0 Vte. terminal, tower, equipments Subtotal Air 47.9 TOTAL TRANSPORT SECTOR 553.1 Notes: a/ Does not include concessional funding from OPEC, China, Thailand, or some other external donor not tracked by MCTPC. b/ Does not reflect delays in implementation. c/ Includes maintenance funding and technical assistance. d/ Proposed projects. e/ Announced at Donor Coordination Meeting, February 6, 1996. f7 Total includes USS5 million from Nordic Development Fund. - 33 - TABLE A2.2: PROVINCIAL PUBLIC INVESTMENT PROGRAM, 1995/96 (BILLION KIP) Populaion PP/Cap ta Irans/Corn. Total PIP Transport/Communication PIP Population PIPC PIP/Capita ('000) (000 kip) ('000 kip) Bn. kip % of total Local Foreign Phongsaly 2.70 1.99 73.7% 1.99 0 154.2 17.5 12.9 Luangnamtha 3.13 2.31 73.8% 0.68 1.63 124.1 25.2 18.6 Oudomxay 1.59 0.57 35.8% 0.37 0.2 296.8 5.4 1.9 Bokeo 1.2 0.5 41.7% 0.5 0 68.3 17.6 7.3 Xayabury 2.94 0.35 11.90/% 0.35 0 211.1 13.9 1.7 Luangprabang 1.65 0.66 40.0% 0.66 0 343.9 4.8 1.9 Huaphanh 1.15 0.56 48.7% 0.56 0 226 5.1 2.5 Xienghuang 1.13 0.52 46.0% 0.52 0 219.6 5.1 2.4 Vientiane Province a/ 4.41 2.25 51.0% 0.41 1.84 317.9 13.9 7.1 Vientiane Prefecture b/ 3.05 1.77 58.1% 1.77 0 523.4 5.8 3.4 Borikhamxay 1.15 0.32 27.9% 0.32 0 189.5 6.1 1.7 Khammuane 1.75 0.62 35.4% 0.62 271.4 6.4 2.3 Savannakhet 10.03 1.32 13.2% 1.32 652.2 15.4 2.0 Champasak 3.82 0.71 18.6% 0.45 0.26 478.4 8.0 1.5 Saravane 1.1 0.4 36.4% 0.4 250.3 4.4 1.6 Sekong 1.15 0.65 56.5% 0.54 0.11 61.3 18.8 10.6 Attapeu 1.00 0.3 30.0% 0.3 81.1 12.3 3.7 Total: All Provinces 42.95 15.80 36.8% 11.76 4.04 4469.50 9.6 3.5 al Includes 1.56 billion kip for water supply. '/ Includes 1.28 billion for routine maintenance and bridge repair but no item for "Six Roads" project. Source: CPC Public Investment Program for Fiscal Year 1995/96, March, 1996. Transport and Communication Sector: Central MCTPC and Provincial PIP Budgets 1993/94 1994/95 1994/95 Avg. Annual iBn. kip 1% Share |Bn. kip 1% Share iBn. kip 1% Share | Change (%/6) Central MCTPC 41.8 81.8 71.4 84.7 71.6 83.3 35.5 Provinces 9.3 18.2 12.9 15.3 14.4 16.7 25.2 Total 51.1 100 84.3 100 86 100 33.5 - 35 - CHAPTER 3: AGRICULTURE A. SECTOR OVERVIEW 1. Agriculture is central to the lives of most Laotians. Although agriculture's contribution to GDP has been falling -- from 63% in 1987/88 to 54.3% in 1995 -- the sector remains the primary source of income for about 85% of the working population.' In recent years the growth of urban- based industry and services has outpaced rural sectors significantly (Figure 3.1), raising concerns about a possible deterioration in the distribution of income. While at present Lao PDR's income distribution is comparatively equal, it is estimated that 87% of Lao PDR's poor live in rural areas and 99% of the rural poor live in farm households. Any attempt to raise living standards and reduce poverty in Lao PDR will, therefore, need to focus on raising living standards in rural areas. 2 2. Given that the Government's resources are limited, it is clearly FIGURE 3.1: GROWTH OF REAL VALUE ADDED (I987=100) important to increase the effectiveness of public spending in 180 the rural sector, particularly in agriculture. And while Lao PDR has Other Sectors made considerable progress in the 140 120 .. . . . . . . last few years in defining priorities that address major problems, further loo Agrulture improvements can be made in both o l l l l the allocation of the Government 1987 1988 1989 1990 1991 1992 1993 expenditures within the sector and the performance of the Government agencies responsible for implementing policies and achieving national goals for agriculture. 3. The Government of Lao PDR is conceed abouemt the markedly FIGURE 3.2: ANNUAL GROWTH RATES OF RICE PRODUCr1ON is concerned about the markedly slower growth in the agricultural sector because it implies not only a 15 widening gap between rural and 10 urban incomes, but also increased o food security problems in the 0 1976.80 1980-86 1986.90 1990-95 countryside, and, if the trend continues, a potential long-term threat to social stability. Rice is the basic staple food in Lao PDR; it represents about 40% of agricultural GDP: rice production figures are closely monitored. Average annual growth rates Unless otherwise noted, "agriculture" includes forestry. A recent comprehensive review of the Government's involvement in agriculture was undertaken by the World Bank (1995), "Lao PDR Agricultural Sector Memorandum," Report No. 13675-LA. 2 World Bank (1995), "Lao PDR Social Development Assessment and Strategy," Report No. 13992-LA. - 36 - (exponential trends) for total rice output (Figure 3.2) are a cause for concern, because estimated population growth is 2.8% per year. 4. Relative to its major regional trading partners (Thailand, China, Vietnam) Lao PDR has abundant arable land per capita. It need not, and perhaps should not, become dep?endent on food imports. Lao PDR has the potential to become a substantial net exporter of food. To accomplish this, the country must transform the prevailing subsistence-oriented, low productivity, low input farming systems. This depends greatly on addressing key constraints to agricultural growth. Developing rural infrastructure is a key priority, in particular investments in rural roads to provide access to markets. Owing to the poor transport infrastructure, no rice is marketed in half of the villages, although rice is the main crop in nearly all villages and accounts for about 80% of total cropped area in Lao PDR. Agricultural productivity has also been held back by underdeveloped markets systems, declining fertility in the uplands, lack of skilled manpower, lack of a national system of research and extension, and undeveloped rural savings and credit systems.4 5. By any measure, livestock problems have not received the attention they deserve. Although livestock5contributes about 40% of agricultural GDP and could become a major export sector, diseases, parasites, and poor nutrition keep both productivity and exports well below potential. Foreign donors in particular showed little interest in livestock development prior to 1995. In the past, the Government has not been able to provide adequate, timely supplies of vaccines. A national network of village para-veterinarians exists, but it suffers from inadequate training and resources. 6. Between 250,000 and 300,000 families (40-50% of farm households) practice some form of shifting, slash-and-burn agriculture in upland areas, either exclusively or to supplement lowland paddy production. Owing to population growth, these farmers--primarily ethnic minorities--face the consequences of shortened fallow cycles, including greatly increased weeding requirements and lower yields. Weed problems increase because herbaceous species still dominate when fields are burned; less above-ground biomass fails to generate sufficient heat to kill the seeds and roots of competing species. In areas where fallow periods have fallen to four years or less, as in much of Luangprabang province, labor inputs for weeding can reach 300 days per hectare. 7. Deforestation caused by logging and shifting cultivation is a serious concern shared by the Government and donors. Deforestation in Lao PDR is leading to increased erosion problems (greater river sediment loads and localized flooding) and threatens the country's natural biodiversity, its watershed catchment areas, and the continued, traditional use of forests by villagers. Logging activities are very difficult to monitor or plan rationally, owing to significant unrecorded logging by firms operated by the Lao military and private road construction. As a result, planning is haphazard and ineffectual. 3 The forestry sector is currently the main source of export earnings, but commercial logging operations benefit relatively few people directly and risk causing major environmental damage. 4 See World Bank (1995), "Lao PDR Agricultural Sector Memorandum." 5 Ranked by value added, livestock in Lao PDR consists of pigs, buffalo, cattle, poultry, and fish. Recent growth is reported to be mainly in pigs and poultry, though this is not reflected in official statistics. 6 This paragraph draws on Bounthanh Keoboualapha, J. M. Schiller, and V. Manivong, "Prospects and Priorities for Upland Rice in the Lao PDR," paper prepared for Upland Rice Research Consortium Workshop, Padang, West Sumatra, Indonesia, January 1996. - 37 - 8. Management of forest resources requires improvement. Monitoring and planning forest exploitation is made more difficult by the poor documentation of logging operations. The notional oversight over three state-owned logging companies has passed from the Ministry of Defense to the Ministry of Finance, but repeated shortfalls in timber collection suggest that the relative autonomy of these regionally based SOEs opens the scope for diverting substantial off-budget revenues from processing and exporting timber products. The monitoring of the activities and finances of these companies--the responsibility of the State Asset Department in the MoF since July 1995--is weak. Because of the general lack of transparency, the party with the greatest financial stake in the sustainability of forestry revenues--the military--has little incentive to cooperate fully with experts trying to ensure the sustainable management of forests, which results in haphazard reforestation efforts and ineffective planning. Moreover, although the poor operational and financial performance of the SOEs has resulted in significant losses in timber royalties to the Treasury, the SOEs do not appear to have suffered any negative consequences for allowing large payments arrears to accrue. 9. Other important agricultural problems in Lao PDR include: (i) the vulnerability of agricultural communities to droughts and floods8; (ii) the lack of an effective national extension system; (iii) reliance on an underdeveloped and biased system for collecting basic agricultural data; and (iv) low rates of lending and investment in the sector. Moreover, formal credit is rationed and does not reach poorer farmers.9 Foreign private investment has been insignificant except in the forest sector. B. GOVERNMENT OBJECTIVES FOR AGRICULTURE 10. Since 1986, when the Government initiated its New Economic Mechanism (NEM), there has been a host of policy changes affecting revenues and expenditures, and the composition and focus of government programs in agriculture and forestry. The World Bank's Agricultural Sector Memorandum for Lao PDR (March 1995) deals at length with agricultural policy issues, sector by sector. What follows here is an overview of the policy environment for the sector and a description of policies and priorities for key areas such as food production, irrigation, shifting cultivation, and agricultural credit. 11. In its Socio-Economic Development plan for 1996-2000, the Government has identified eight development programs for the nation. These are intended to help form and implement priority programs within the line ministries. Four of the eight programs concern agriculture explicitly: (i) increasing food production; (ii) stabilizing/reducing slash-and-burn cultivation; (iii) promoting the production of commodities (commercial crops and livestock); and (iv) promoting integrated rural development. The first three are part of the six "priority programs" pursued by the Ministry of 7 In the first three quarters of 1994/95, out of US$35 million in royalties on removed timber, US$21.3 million was actually received, leaving more than one third, US$13.7 million, in payment arrears (according to World Bank and MoF data). As a percentage of total revenues, timber royalties have declined from 20% in 1992/93 to 16.5% in 1994/95. See Chapter 13 for details on timber and timber taxation. Only 2% of the area under rice has secure supplies of water, leaving 98% of production susceptible to drought. However, flooding is often the more serious problem, as in 1995 when floods destroyed 50-60,000 hectares of lowland rice (about 10 %/0). See Box 3.1 on the Agricultural Promotion Bank. - 38 - Agriculture and Forestry (MAF), together with human resource development, agricultural research, and water resources development, including watershed management and small-scale irrigation. 12. Food Production. The food production program primarily focuses on boosting rice production on the larger plains, with emphasis on the Vientiane Plain, to ensure adequate supplies of rice for the capital city. Historically, the North has had a food deficit (Figure 3.3) and a specific objective is to achieve food self-sufficiency by the year 2000.10 13. The goals of the food production FIGURE 3.3: RICE SUPPLY/DEMAND BALANCE program continue the Government's long- (relative to assumed consumption of 300 kg per capita) standing policy of promoting national self- sufficiency in rice.11 In practice, this policy has amounted to little more than setting rice production targets that are translated to province-specific and district-specific goals. These targets are frequently not met, as is evidenced by the declining output trend mentioned above. Despite this failure, the Government has set an ambitious annual growth target for agriculture of 5.5%, while overall GDP is targeted to grow at 8%. If these targets were met, agriculture would remain an important sector, initially accounting for over a third of Supply/Demand Balance aggregate GDP growth. E Deficit 14. Management Irrigation Systems. a Surplus Government policy on management irrigations systems underwent a crucial shift in 1991 (which was codified in official regulations in 1993) toward low cost farmer-managed irrigation Note: Based on 1988-92 average production. schemes as well as donor-supported programs to Source: Ministry ofAgriculture and Forestry. rehabilitate the old capital-intensive schemes and to hand them over to farmers to manage. The Government decided then to relinquish direct responsibility for operating and maintaining irrigation schemes, and financial and administrative responsibility for operations and maintenance 10 For reasons of history and geography, the country's 18 provinces are customarily grouped into three regions: North, Center, and South. Relative to the Center and South, the North is more mountainous, road links with the rest of the country are less developed, and shifting cultivation is more widespread: in the Northern provinces of Luangprabang, Oudomxay, Phongsaly and Huaphanh, upland rice occupies 80-90% of the total rice area. This dependence on relatively low-yielding upland rice leads to frequent rice deficits which are met either through imports or by substituting maize, cassava and foods gathered in the forest. However, recent evidence suggests that actual hunger and malnutrition may be more widespread in the South. (See World Bank, "Lao PDR: Social Development Assessment and Strategy," Report No. 13992-LA, 1995.) " The 1991-95 development plan omits the goal of self-sufficiency in rice, and instead allows that "...the distribution of factors of production and the composition of agricultural production will be left to market forces and farm household decision making". (Kaysone Phomvihan, "LPDR Political Report to the Fifth Party Congress," April 1991: quoted in W. Worner, "Agricultural Change and Development Policy in Laos," Lao Studies Review, vol. 1, 1993, p. 46). However, the Govemment has continued to set specific growth targets for rice, other crops, and livestock. - 39 - (O&M) was transferred to local Water User Groups (WJUGs) formed among the beneficiaries of particular irrigation schemes. This shift in policy underpins planned government investment. 15. In the past, the Government's irrigation policy reflected its emphasis on rice self-sufficiency. Farmers in government-sponsored irrigation schemes were often required to grow rice, which typically shows low economic returns relative to alternative crops. This policy helped to undermine the viability of many irrigation schemes, particularly capital-intensive ones. Now, farmers are ostensibly encouraged to use irrigation water for higher-value crops, at least during the dry season, although official emphasis on "boosting rice production" suggests a continuing bias in favor of rice. At least 70% of dry season irrigation and virtually all wet season irrigation is for glutinous rice. 16. Farmers participating in WUGs provide land (e.g., for canals), labor, material, and cash for purchased inputs. Complete recovery of project costs is required by regulation only for small-scale projects. For all projects, O&M costs are supposed to be covered by water user fees (cash or in kind) by the WUG after each harvest. According to the regulations, one half of the money collected, after deduction of any electricity charges, is for maintenance funds, the other half for "administrative costs, salaries for cadres, etc."12 Indicative water costs range from 3,500-5,000 kip/ha/season for gravity fed systems, to 25,000-50,000 kip/ha/season for diesel pump systems, with actual charges depending on the nature of the system, the region, and the opinion of the farmers. In fact, practice varies greatly among provinces and among similar schemes. In some schemes, a sufficient amount is collected to cover periodic maintenance and repairs. Others collect fees only for electricity charges. 17. Major repairs of "engineered" structures have been contracted out to a handful of irrigation construction companies, mostly provincial level enterprises, two of which remained under state ownership in early 1996.13 Under these arrangements, irrigation personnel in the provinces and districts concentrate on surveying and designing potential new irrigation projects, and providing technical expertise for the construction and repair of irrigation structures where needed. 18. In principle, the existing regulations provide a framework for sustainable and cost-effective management of irrigation systems, especially in cases where village involvement in planning, design and construction was high. However, in the past such involvement was lacking for almost all but the smallest schemes carried out by individual villages with unusually extensive experience managing their own temporary irrigation structures. For the medium-size and larger schemes, effective management is nearly always beyond the technical and organizational capacity of village-level authorities. A provision exists in the regulations for the Government to manage large or internationally bid projects "for a certain period of time in order to assist the farmers" before handing over responsibility to the WlUGs, but funds have not been provided to provincial authorities (or the Department of Irrigation) to adequately maintain irrigation assets.14 In consequence, most of the 12 Ministry of Agriculture and Forestry, Regulation No. 0598/MAG.DI, Part Five: Principles and regulations for the use of water, congratulations and penalties, Article 17 (unofficial translation reproduced in L. Chazee, Practical Handbookfor the Development ofRural Communities in Laos, UNCDF: Vientiane, 1994). 3 The July 1993 irrigation management regulations define four categories of irrigation projects according to construction cost and irrigated area. In terms of area, the categories are: 1. "popular" or "family" = less than 30 ha; 2. small scale = 30 to 100 ha; 3. medium scale = 100 to 1000 ha; 4. large scale = greater than 1000 ha. 14 Some financing of irrigation occurs through the parallel structure of Rural Development Committees. It seems that funding has come from the 30% share of tax revenues above the annual targets which provinces are allowed to retain (see Chapter 8). - 40 - larger irrigation schemes have been poorly maintained, are grossly underutilized, or have simply shut down. 19. Shifting Cultivation. Reducing shifting cultivation is one of the Lao Government's eight priority programs. It seeks to introduce sustainable farming practices, including agro-forestry; to promote alternative income-generating activities for upland villages; and to resettle part of the upland farming population in lowland areas. These efforts are focused on watersheds, reserve forests, and perennial forests near major towns and roads. 20. The ongoing national policy on upland agriculture is to stabilize and reduce slash-and-burn cultivation. This policy was clarified in 1995 in two key respects. First, the Government recognized the impracticality of eliminating all upland rice production. In areas where there is little available land suitable for lowland (i.e., flooded) rice, improved, sustainable methods of growing upland rice are to be developed using crop rotation and fertilizers. Second, the traditional rights of families to use plots of land in the uplands are to be recognized to encourage them to invest in environmentally sustainable land use, including planting tree crops. 21. The most controversial activities within the Government's shifting cultivation reduction program are efforts to "promote" the resettlement of entire villages--mostly Hmong--from the highlands to lowland areas. The Government should be aware that there are very few examples of successful resettlement schemes in other countries, and that many have been disastrous failures (e.g., "villagization" in Tanzania; resettlement from Java to Sumatra in Indonesia). In a one-party state such as Lao PDR, elements of coercion are difficult to avoid, even if they are unintended. Outsiders usually suspect coercion even when it does not exist. Given the sensitivity of donor countries to this issue, the Government-driven resettlement efforts could easily jeopardize foreign assistance inflows. 22. Forestry. In the forestry subsector, the Government's goals include reducing illegal log exports, encouraging replanting with fast-growing species, and promoting the domestic wood processing industry to increase value added before products are exported. 23. In October 1994, virtually all of the logging and much of the processing operations in the country were consolidated into three state resource development companies, one for each of the northern, central, and southern regions. The Ministry of Defense (MoD) controls these companies and it retains profits after paying timber royalties to the Treasury. The consolidation was intended to improve central government control over logging and increase the collection of timber royalties. However, it is difficult to monitor the companies' operations and finances, and the collection of royalties has actually declined. The combination of state ownership and the monopoly power enjoyed by the SOEs seems likely to work against improving the efficiency of their management in the future. 24. Crops and Livestock. Commercial crops and livestock, the "commodity production" program, seeks to boost farmers' incomes and diversify away from subsistence rice production by promoting commercial crop and livestock production, largely through improvements in market access (mainly road construction and repair). This raises the question of to what extent the public sector should be involved in directly supporting private sector commercial activities. 25. Agricultural Credit. The Government has made a clear policy choice to subsidize agricultural lending to alleviate rural poverty. Credit for agriculture is channeled at below-market interest rates, through the Agricultural Promotion Bank (APB), financed by the Bank of Lao PDR - 41 - (BoL) with donor assistance (see Box 3.1). However, the vast majority of Lao farm households in need of credit are too poor, or lack the right connections, to be reached by APB loans. Box 3.1. THE AGRICULTURAL PROMOTION BANK (APB) The APB was formed in June 1993 by consolidating the agricultural loan portfolios and related staff of the three state commercial banks providing agricultural credit. It has a de facto monopoly on agricultural lending since other banks are barred from establishing branches in rural areas. The APB is supposed to act as a catalyst for agricultural development by providing loans to intensify and diversify production. Agricultural officials reveal that the APB is being asked to assist with farmer-driven irrigation construction, resettlement of upland farmers to lowland areas, expansion and improvement of livestock herds, and promotion of technical packages for rice. APB funding has come from foreign and domestic sources. Foreign donors include US$4.4 million from IFAD, a US$1 million credit from the World Bank, and US$7 million from ADB, on the condition that APB establishes a new accounting system. Domestic funding includes transfers from the BoL and the government budget; transfers to the APB out of state-owned commercial banks' profits; and a legal requirement--partially enforced--that commercial banks either lend 10% of their deposits in the agricultural sector or transfer an equivalent amount to the APB. The size of the APB's outstanding loan portfolio is not clear, but lending from January through October 1995 was 15 billion kip. According to Irrigation Department officials, requests for loans in support of farmer-managed irrigation projects alone totaled 20 billion kip in 1995, for which the APB lent only 4 to 5 billion. It is not clear whether the APB fulfills its objective of poverty alleviation, since it effectively funnels resources from other sectors and donors to a fortunate minority of relatively well-off farmers. Its rate structure and lending policy are not geared toward maximizing the mobilization of resources for agricultural lending. First, the APB's loans are heavily subsidized: as of November 1995, the APB was charging 10%, 8%, and 7% for short, medium, and long term loans, compared to a range of 17.6% to 26.5% charged by commercial banks for kip loans. Second, the APB's lending policies are intended to benefit poorer farmers but do not reach them. For a sample of 1,164 households receiving loans in 1995, the average loan was about US$700, ranging from US$100 up to US$15,000, with only a handful below US$300. APB officials rightly explain that a portfolio consisting of many small loans to poor farmers would be too costly for a relatively centralized bank like the APB to administer. Finally, repayment conditions and enforcement are reported to be "soft," implying a further limiting of the pool of funds available for lending to farmers. Reflecting its commitment to agriculture--one that goes back well before the creation of the APB-- the Lao Government's decision to subsidize agricultural lending is a clear policy choice. However, the vast majority of Lao farm households in need of credit are too poor or lack the right connections to gain access to APB loans. There is thus a clear need for private rural credit institutions tailored to small- holders. Lao PDR could well benefit from appropriate technical asststance to examine rural credit options for the country, which would draw on the experience that donors have accumulated in helping develop rural credit organizations in other countries. In the meantime, greater recognition is needed of the trade-off between subsidizing agricultural credit and mobilizing additional funds for that purpose. Under current policies, agricultural credit is effectively rationed, imposing an unnecessary constraint on the pace of private investment in agriculture that can only hurt the majority of farmers. - 42 - C. FUNDING MECHANISMS 26. Administrative Structure. Overall responsibility for carrying out Government policies and programs in agriculture nominally rests with the MAF. In practice, however, translating policies into spending allocations involves other central level agencies (primarily the SPC, but including the National Assembly and the Ministries of Finance and Foreign Affairs), the donor community, and the provincial administrative authorities (under provincial governors). Furthermore, since spending proposals are initiated and projects are initiated and implemented within the MAF at the provincial level or below, the Provincial and District Agriculture and Forestry Service Offices (PAFSOs and DAFSOs) also play important roles. Finally, the APB controls agricultural lending, and with increasing frequency is the institution that decides whether local level projects go forward (Box 3.1). This is particularly true for new land development and small-to-medium scale irrigation project proposals coming from the districts or villages. The APB has altered its lending policies to expand its support of irrigation development, but funds are limited. 27. The most fundamental change in the formal administrative structure since the beginning of the NEM was the "centralization" of the line ministries in the early 1990s, beginning with the MAF. Since then, agricultural programs and projects managed by the provincial governments -- and funded from revenues retained by the provinces -- have come under central government control. Provinces no longer have independent budgeting authority and cannot collect revenues for their own use. The MAF submits an overall budget proposal that includes requests for provincial or lower level projects. Provincial governors are supposed to ensure that the central government's priorities and policies are implemented correctly within their jurisdictions. However, provinces differ when setting spending priorities. In practice, the more powerful provincial governors have exercised considerable discretion over the use of funds. Some provinces have retained irrigation O&M units and construction enterprises as an agreed arrangement among the central and provincial authorities to support project implementation and O&M at the provincial level in the absence of competent private companies. 28. Foreign Assistance. Agricultural spending is somewhat less dependent on foreign donors than other sectors in Lao PDR. However, there remains a strong tendency to allocate Government funds to areas in which donor interest is the greatest while neglecting other areas. In the irrigation subsector, for example, the Government focused much of its own resources on large, capital- intensive schemes when external funds were forthcoming. But donors have begun to pull back from supporting large schemes and the Government has signaled that its emphasis in the 1996-2000 development plan will be on developing small-scale irrigation works. These low-cost,16 farmer- managed systems continue to be supported by NGOs and other donors. It is likely that the Government's attempts to maximize foreign assistance to this sector led to inappropriate policies for the Lao context which turned out to be very wasteful. is Looking at the proposed 1996-2000 public investment in four sectors--agriculture, health, education, and transport and communications -- the share from foreign sources ranges from 66.6% in agriculture to 75.8% in the health sector. The lower share from donors in agriculture may reflect greater priority given to agricultural and rural development on the part of the Lao Government compared to foreign donors. t6 On a per irrigated hectare basis, some small-scale irrigation schemes have proved very costly, according to project documents. Unlike larger schemes, however, most can at least be managed and maintained by local Water User Groups. - 43 - 29. Coordinating donor activities in Lao PDR has been difficult in most sectors, but recently there are signs of improvement. At a macro level, the UNDP plays a key role through its annual roundtable meeting between the Government and donors. Donor conferences are also being organized for specific sectors or sub-sectors, and it is hoped that this trend will continue. Another useful approach to donor coordination is to create an "umbrella" organization within the appropriate ministry, with responsibility for overall planning and coordination of activities within a given sector. For example, NAMSAAT, within the Ministry of Public Health, was set up to plan and monitor water supply and sanitation activities. 30. Finally, on the donors' side, it is helpful to have a single external assistance agency designated "lead donor" to assist the Government to develop national level plans and programs.17 A good example is the Swiss-funded Lao-IRRI project, under which the International Rice Research Institute has helped the MAF to develop a National Rice Research Program. 31. The project planning process is also changing. In the past, provinces had independent budgeting authority and donors often approached them directly to discuss project proposals. But budget processes were centralized in 1992/93, and now the link between the Government and donors for planning and funding is at the central level, mainly via the SPC (the CPC until mid-1996) and, to a lesser extent, the departments and units of the MAF. 32. Investment planning is now intended to be a bottom-up process. Local officials are supposed to collect and assess the feasibility of proposals from villages. Following this, lists of possible capital spending initiatives are drawn up by units within the provincial and district offices of the Ministry (e.g., for irrigation, the survey and construction sections of the irrigation divisions). The PAFSO uses these lists for its initial budget proposal, which goes to the provincial departments of the SPC and to the provincial government for examination and approval, after which it is submitted to the MAF and SPC in Vientiane. At the central level, the SPC identifies sources of donor funding for the provinces' proposals and the MAF submits a separate request for local budget funds to the Ministry of Finance. 33. The planning system has changed rapidly in recent years in an apparently appropriate response to the real problems experienced in the past. There was a lack of community involvement, and consequently the users/beneficiaries lacked a sense of ownership and contributed little toward O&M. However, the process for most major initiatives involving foreign assistance remains top- down and donor-driven; donors negotiate project proposals with the SPC and MAF departments, and there is little input from lower levels. As the current planning approach has only been followed since 1995 it is too early to evaluate it. D. EXPENDITURE TRENDS 34. Total spending on agriculture and forestry averaged 4.3% of agricultural GDP over 1992-95, and has remained under 5% since 1989. This is less than half of the average proportion of spending to sector GDP during the 1980s found in a sample of 26 developing countries. I However, since 17 Sectoral Investment Programs would be another effective approach for improved donor coordination (Chapter 10). is Van Blarcom, B., 0. Knudsen and J. Nash., "The Reform of Public Expenditures for Agriculture." World Bank Discussion Papers, No. 216, 1993. The 26 countries mentioned are those which, like Lao PDR, had received adjustment loans. They spent, on average, 9.4% of agricultural GDP on agriculture during 1984-88. Another sample of 14 countries without adjustment loans spent an average of 15.7%. - 44 - agriculture is by far the largest sector in Lao PDR, agriculture expenditures form a large part of the government's budget compared to developing country averages (11% in Lao PDR over 1988-95, compared to roughly 7% for developing countries in the 1980s). In 1994/95, current and capital spending on agriculture accounted for 9.7% of the total budget, and fell to 7.8% in 1995/96 (see Table 3, Statistical Appendix, Main Report). 35. Current expenditures have been quite low, between 1.7 and 2.5% of total government current expenditures. It should be noted, however, that past reporting practices included O&M spending as capital investment if it was part of a foreign-funded project, reducing reported current expenditure figures. Spending on wages and salaries in the sector has also remained very low, averaging about 3% of the total government wage bill, and is low compared to reported capital expenditures. 36. Although past total O&M spending cannot be accurately determined, it is widely acknowledged to have been grossly inadequate. For example, the utilization rate for existing irrigation systems in 1993 was just 30% of designed capacity during the dry season. Looking at the use of government funds by the MAF and the provinces, about 20% of "investment" during 1991-95 was allocated to O&M spending (Table 3.6). However, such funds accounted for only 12% of public investment in the sector over 1993-95, while the bulk of investment came through foreign donors who generally refrain from paying for the recurring costs of development projects. 37. Inadequate provision for maintenance of capital investments is a problem common to all sectors in Lao PDR (see, for example, the discussion of road maintenance in Chapter 2 of these Background Papers). The Government's shift in the 1996-2000 development plan in favor of small- scale irrigation works maintained by local water user groups is one appropriate response to this problem. However, in terms of design capacity, about half of the existing irrigation resources can be considered capital-intensive systems. Maintenance of these has been turned over to user groups, or to local irrigation companies which are mainly former SOEs. Since rehabilitating large irrigation works is likely to be beyond the means of most of these organizations, the desirability of continuing with the present management arrangements needs to be assessed in the near future. This could be incorporated in a comprehensive evaluation of all irrigation investments, to build an information base for prioritizing further investments in rehabilitation and maintenance. 38. A large part of the Government's agricultural budget is spent directly in the provinces rather than in the central Ministry. The share of current expenditures going to the provinces increased from 50%, prior to 1992, to 71% in 1995. Most investment from the Government's own sources also goes to the provinces: 5.1 billion out of 7.4 billion kip in planned 1995/96, investment. For 1995/96, two provinces (Champasack and Savannakhet) and the Special Region receive 40% of total provincial- level spending. On a per capita basis, however, with the exception of Champasack, spending is greatest in the relatively poor provinces: Sekong and Attapeu in the South; Phongsaly and Bokeo in the North. Other relatively poor provinces (e.g., Huaphanh and Luangprabang) receive much smaller amounts per capita out of the domestic agricultural budget. 39. Turning to departmental programs within the MAF, public investment, excluding foreign sources, became increasingly concentrated on irrigation. The Irrigation Department's share of investment grew from 35% in 1991/92 to 64% in 1994/95, and only 3.7% of this was allocated to O&M (Table 3.6). In contrast, areas with high expected rates of return were clearly underfunded, including research and extension and livestock/veterinary services. In the MAF's investment plan for 1996-2000, irrigation absorbs up to two-thirds of local funds and comparatively minor amounts - 45 - go to other areas. This seems to contradict the PIP for 1996-2000, which includes funds from donors. 40. Data on public investment from foreign sources are sparse for earlier years, but project spending was also heavily focused on irrigation (Table 3.6). In 1993, ongoing irrigation projects totaled US$44.8 million out of a total of US$100 million. However, the MAF's public investment plan for 1995/96 shows a drastic reduction in irrigation investments in favor of the forestry subsector -- with 15% for irrigation compared to 46% for forestry. Investment expenditures on forestry had fallen from 28% of total in 1988 to 9% in 1992. The second largest share (19%) of total planned agricultural investment in 1995/96 is for "rural development" projects that do not fit into the more narrowly targeted departmental categories, including so-called integrated rural development (IRD) projects. Overall public investment in agriculture is currently planned to grow faster than anticipated inflation, with its share of total public investment decreasing slightly from 11.9% in 1995/96 to 11.5% in 1999/2000. 41. Within the irrigation sub-sector, past government investment was overwhelmingly in large- scale schemes with command areas greater than 500 hectares. Since the rate of return on these projects in Lao PDR has been extremely low or negative, the recent policy shift in favor of small- scale, farmer-managed irrigation is justified. However, experience in other countries suggests that efforts to rehabilitate and improve the operation of existing systems could have a high payoff and thus should not be neglected. For example, a 1991 review of World Bank assisted irrigation projects found that the average economic rates of return for new surface irrigation projects were 16.7%, 18%, and 20% for large, medium, and small projects, respectively. But projects to upgrade existing systems had an average rate of return of 25%, and projects to improve irrigation system operations had an average rate of return of 38%. In India, O&M expenditures have shown returns in the 30- 40% range, and in Indonesia they were estimated to have a return of over 100%.19 42. Table 3.1 shows that forestry and irrigation receive disproportionate shares of investment. Based on this comparison, even if actual forestry production were ten times the official estimate, 1995/96 investment in the sector would still appear large compared to other key areas with high expected rates of return such as research and extension and livestock and veterinary services. For agricultural research, rates of return in the 30-60% range are not unusual, according to crop-specific studies.20 Investment in irrigation or forest conservation and reforestation is valuable but at the margin some reallocation toward programs to raise the productivity of crops and livestock would be likely to yield greater benefits and would also be more in line with the Government's expressed priorities. 19 These figures are from World Bank, "India: Agriculture Public Expenditure Review," Report No. 12196-lN, May 1993. 20 World Bank, World Development Report 1991, June 1991, p. 74. - 46 - TABLE 3.1: AGRICULTURAL PROGRAMS RELATIVE TO SUBSECTOR GDP, 1995/96 Agriculture (research) & Extension Investment/ Crops GDP 0.4% Livestock & Veterinary Investment/ Livestock GDP 0.3% Forestry Investment/ Forestry GDP 11.7% Irrigation Investment/ Crops GDP 0.8% Total Investment Under MAF/Agricultural GDP 2.5% Source: State Planning Committee. 43. When public spending allocations within agriculture are compared with averages for Asian countries at a similar stage of development in 1970 and 1980, "investment" in agricultural research and extension in Lao PDR (which includes significant operating costs) appears to be extremely low (Table 3.2). A comparable figure for spending on research and extension in Lao PDR for 1995/96 is below 0.2%. TABLE 3.2: RESEARCH AND EXTENSION IN ASIA RELATIVE TO AGRICULTURAL GDP Public Sector Ag. Public Sector Ag. Research Extension Regions Expenditures/Ag. GDP Expenditures/Ag. GDP 1970 1980 1970 1980 East Asia 2.01 2.44 0.67 0.85 South Asia 0.19 0.43 0.23 0.20 Southeast Asia 0.28 0.52 0.37 0.36 Source: Evenson, R.E., in Schuh, E. (1986), "Technology, Human Capital, and the Food Problem, " University of Minnesota. E. EFFICIENCY OF PUBLIC SPENDING ON AGRICULTURE 44. Outcome Indicators. As Table 3.3 shows, food production per capita decreased between 1979/80 and 1992/93 and the use of modem inputs declined (represented by fertilizer use). 21 Several factors contributed to this poor performance, including an inadequate transport network, poorly maintained irrigation structures, poor animal vaccine coverage, and the absence of proven locally adapted technical packages for rice and other crops. There are signs that these constraints are diminishing; large investments are under way in transport, including rural feeder roads, and the national rice research system is beginning to generate promising results, including area-specific fertilizer recommendations. In the medium term, inadequate extension services could increasingly 21 Rates of fertilizer use are low in spite of heavily subsidized fertilizer sales by the Government. - 47 - become a major bottleneck as other constraints to increasing agricultural productivity are being addressed. 45. A precise evaluation of the effectiveness of public spending on different sub-programs is not yet possible, due to a lack of data on unit costs of government-provided goods and services. Nevertheless, it is clear that the record is poor in most cases. In addition to the dismal performance of capital-intensive irrigation projects, some of which have cost over US$4,000 per irrigated hectare, public investment has been hampered by severely limited human capital resources for planning and administration. The human resources problem is compounded by the proliferation of donor projects and the lack of strong mechanisms to coordinate them and share information and experiences.22 There are over 100 agriculture and forestry projects in operation. Consequently, some of the most competent staff at the central level are forced to spend much of their time and effort meeting the complex reporting requirements imposed by donors. This leaves little time for planning and monitoring. TABLE 3.3: INTERNATIONAL COMPARISONS - FOOD PRODUCTION GROWTH, ARABLE LAND, AND FERTILIZER USE IN ASIAN COUNTRIES Food production Persons per Fertilizer consumption Country per capita hectare of (kilograms per hectare (avg. ann.growth, %) arable land of arable land) 1979-93 1993 1979/80 1992/93 Malaysia 4.3 3.9 94.4 197.7 China 3.0 12.2 149.7 300.5 Vietnam 2.2 11.1 23.6 134.7 India 1.5 5.3 32.9 72.0 Nepal 1.2 7.8 9.7 39.1 Thailand 0.0 2.5 15.0 54.4 Lao PDR -0.2 2.4 * 5.8 4.2 Myanmar -1.3 4.4 10.0 6.9 Philippines -1.3 8.1 38.3 54.0 * Includes potential agricultural lands up to 5 % slope. Sources: World Bank (W'orld Developnment Report 1995 ) and Ministry of Agriculture and Forestry, Lao PDR, (Agricultural Statislics Yearbook 1994 ) 46. Areas in which agricultural spending has failed to produce satisfactory results include the livestock subsector (all aspects), agricultural data services, and extension. In the livestock sub- sector, national vaccination coverage ranges from zero to 30%, and consequently veterinary services have been ineffective in controlling diseases. Meanwhile, the agricultural data system yields production figures that have limited credibility. Extension workers lack the skills and incentives to contribute to the dissemination of useful technologies. Most of these failures are caused by lack of resources provided via the budget. 47. In the case of vaccines, cost recovery -- equivalent to 65% of production costs in 1993--was perhaps excessive, given the externalities inherent in disease containment. For agricultural data collection, the same cadre of village-level workers responsible for tax collection is relied on to perform this task. The cost of this approach is very limited, but the value of the result is also 22 Some notable exceptions include irrigation "Technical Advisory Group" meetings and the Government-organized meetings of forestry sector donors, both held quarterly. - 48 - minimal. Similarly, for extension, the Government's approach relies on the efforts of unpaid extension "volunteers" and the selection of a number of "model" farmers whose superior production practices will, it is hoped, inspire emulation in their neighbors. Current plans call for designating up to four extension volunteers per village, with each volunteer specializing in one or more areas such as crops, livestock, or irrigation. 48. Within the Government's portfolio of agriculture and forestry projects there are some important success stories. The Swiss-funded Lao-IRRI rice research program, for example, has established a promising national rice research system with branches in every province. In 1990, Lao- IRRI began developing a national rice research program (NRRP). With a network of over 60 full- time technical staff, the NRRP now includes extensive research in both lowland rain-fed and upland environments. It has developed some area-specific recommendations for fertilizers and has selected several well-adapted local and regional varieties for multiplication and distribution to farmers. 49. In contrast to many donor-funded projects which exist as enclaves outside of the governmental organizations, Lao-IRRI is well integrated into the national rice research program. The ratio of government personnel to expatriate technical assistance or long-term staff is extremely high, and the proŽect has received enthusiastic support from the highest levels of the MAF. For these reasons, the project is likely to be sustainable and to produce impressive long-term benefits for the country. Moreover, with an annual budget of only US$1 million, the cost of the program--now mostly for training--is very modest in comparison with the scale of the effort and the potential returns, making it well within the means of the Government to take over after 1999 when the current phase of the project ends. 50. In the forestry sector, the Lao-Swedish Forestry Project (LSFP) is also well regarded. Like Lao-IRRI, it is funded by a bilateral agency (SIDA) and emphasizes training government counterpart staff. Currently in its fourth five-year cycle, running from 1996 to 1999, the project's basic objectives are: strengthening institutions (particularly the Department of Forestry in the MAF), developing successful models of community-based forestry, and creating a forestry research capacity. The Department of Forestry is now considered by many to be the most technically competent of the MAF departments, largely owing to the long-term involvement of the LSFP. 51. Lao PDR's problems with donor coordination have improved in the last two or three years. Overall guidance is provided by the UNDP, and the trend is toward holding sector-specific donor conferences. Among these are the regular meetings of the Government and donors in the roads, water supply and sanitation, forestry, and irrigation sectors. Another increasing trend is that, increasingly, a single donor or external agency is given broad responsibility for developing a coherent sectoral program. To do this it collaborates with a national program or agency on the Government's side--and coordinates the efforts of NGOs and other donors active in the sector. 52. Finally, in assessing the Government's involvement in the agriculture sector, it should be stressed that the Government employees have been coping with a continuous series of changes in policies, regulations, and administrative structures since the late 1980s. Despite the rapid pace of change, both central and provincial level administrators generally possess a clear understanding of their professional mission and are able to articulate precisely the roles and responsibilities of different parts of the system. Moreover, since the capacity to fund, plan, and manage an ambitious program of Government involvement in agriculture was lacking internally, foreign donors, international agencies, and technical advisers were relied on to fill the gap. Spending priorities were - 49 - very much donor-driven. Consequently, if agricultural spending has failed to yield sufficient concrete benefits thus far, the donors must bear the greatest share of responsibility for this failure. F. SCOPE FOR PRIVATE SECTOR PARTICIPATION 53. Agriculture in Lao PDR essentially consists of private sector farming. At the same time, the Government concentrates on key public goods in the sector--research, collection and dissemination of information, market integration, control of epidemic diseases in livestock, and so forth. Greater private participation is feasible in construction and repair of irrigation structures, seed multiplication, agricultural lending, and veterinary services. 54. Private irrigation construction firms exist, but they are small and lack adequate equipment. Also, since they have experienced long and unpredictable delays in receiving payment for work done for the Government, they prefer either to subcontract through the remaining irrigation SOEs or to deal directly with donors. If the private irrigation construction sector is to be strengthened, then bidding and contracting regulations are required, including clear standards, payment schedules, and processes. 55. Seed multiplication is now carried out by state seed multiplication centers with the participation of one Thai company in Vientiane.24 Since farmers must have confidence in purchased seed quality, a government seed certification program would help create and expand private sector seed multiplication. Fertilizer imports are subsidized via Japanese "KR2" project funds. The Government effectively rations fertilizer at its "state" price, which is only half the going market price. A preferable approach would be to sell most of the donated fertilizer in open auction to private traders and use the proceeds to fund additional research on soils and soil fertility management. 56. Agricultural credit is another area in which greater private sector participation ought to be encouraged. As discussed above, if the ABP's monopoly on establishing branches in the provinces is removed, the demand for credit appears to be sufficient to support expanded lending for agriculture at commercial interest rates. The current policy hurts farmers by limiting opportunities to borrow for agricultural investments. Also, heavily subsidizing agricultural lending, whether through below- market interest rates or "soft" repayment conditions, concentrates the benefits of agricultural credit on a fortunate few and thus acts against the Government's stated desire to achieve broadly shared rural economic growth and development. 23 Of the two large irrigation SOEs at the central level, one has been transferred to the Saysombon Special Region and is under the jurisdiction of the Lao military. The other was privatized. The poor state of equipment made it necessary to consolidate 18 construction teams into 4 or 5 before the SOE was sold. Some large provinces (e.g., Savannakhet) retain provincial level irrigation enterprises involved in constructing large projects that the private companies are too small to undertake. Some provincial governments also have irrigation O&M "divisions." Mostly, these are vestiges of past projects and cover only a few districts each. 24 In another instance of private involvement in multiplication, farmers in Vientiane Province have established nurseries for growing teak seedlings. These are sold for about 200 kip each to other farmers for growing in upland fields. Although this activity appears to be effectively taken up by the private sector, the upland agriculture research station in Vientiane Province also produces teak seedlings and sells them for 150 kip. This is currently the research center's main activity. - 50- 57. Considering non-governmental, non-profit membership organizations (for example, the coffee producers' association) as part of the private seGtor means that a variety of important extension functions can be "privatized" for commercial crops and livestock/poultry. It is difficult to estimate the contributions made by voluntary associations of farmers, traders, and processors to the progress of agriculture in market economies. Private associations in Lao PDR could provide their members with information on market opportunities, new varieties of crops or breeds of livestock, and new production techniques. This is especially important in view of the lack of effective public extension services, but these private associations must complement the Government extension services, not substitute for them. Public extension programs are critical in rice production and community forestry, where private producers' associations will not emerge until the commercialization of agriculture is far more advanced. 58. Overall, the record of commercial investment in agriculture has been poor. In particular, foreign investment in agriculture has lagged substantially behind investment into the industrial and service sectors. Investment has been held back by the lack of formal land tenure security, by urban- based Lao investors and foreign investors who have no access to land through the customary land tenure system -- an issue which is being addressed by a World Bank supported Land Titling Project. A second key constraint that needs to be tackled to make agriculture attractive to investors is the official licensing procedures, which require provincial governments to approve foreign investments in agriculture over and above the licensing undertaken by the central government. A third issue is the bias in the structure of tax exemptions and tax holidays against investments with a long gestation period, which are typical for agricultural projects in tree crops, commercial forestry, livestock breeding, and research. Finally, the Government still intervenes excessively in the operation of markets, through government trading in agricultural inputs, restrictions on the movement of goods, and interventions in financial markets. G. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM 59. The Government's proposed expenditure program over the next five years goes some way toward addressing the constraints of the agricultural sector (Table 3.4). However, the balance among the different programs could be improved so that it follows the recommendations mentioned in this review. This is reflected in an alternative PIP proposal for the period 1995/96-1999/2000 in Table 3.5, which maintains the same projected aggregate spending levels used by the SPC to be consistent with the likely resource envelope (Chapter 1, Main Report). It proposes a reallocation in relative spending priorities among sectoral programs: providing higher spending for (i) livestock and (ii) research and extension; slower growth in spending on (iii) forestry and (iv) area development; and stabilizing spending on (v) irrigation. These proposed changes do not indicate a radical departure from the Government's proposed investment plans, but rather indicate a broad shift in emphasis toward higher priority areas. The reallocations would contribute to greater effectiveness in raising agricultural productivity, and could be implemented without posing an additional burden on the absorptive capacity of the sector and the Ministry. 60. The Government's spending plan provides a large and rapidly growing allocation to so- called "area development" projects, including, for instance, integrated rural development projects. Funds for this general category double from 3.6 billion kip in 1995/96 to 8.1 billion kip in 1997/98, and increase further to 15.5 billion kip in 1999/2000. Although the rural or community development projects under the MAF may be individually worthwhile, it is questionable whether they should - 51 - come to dominate the Ministry's portfolio. As these projects also absorb scarce government budget funds and human resources, the pace of expansion should be moderate enough to allow sufficient funding for other key programs and for O&M. This is reflected in the proposed PIP, while still maintaining the same growth in funding, from 23.4 million kip in 1995/96 to 44.0 million in 1999/2000. TABLE 3.4: GOVERNMENT PIP FOR AGRICULTURE AND FORESTRY, 1995/96 - 1999/2000 (BILLION KIP) 1995/96 1996/97 1997/98 1998/99 1999/00 Total '95/96- '99/2000 Research & extension 3.6 3.5 4.0 2.5 3.6 17.2 Ongoing & committed 3.6 2.8 2.7 - - 9.1 Research & extension projects 3.6 2.8 2.7 - - 9.1 New programs - 0.7 1.3 2.5 3.6 8.1 Research extension & crop promotion - 0.4 0.7 1.1 1.4 3.5 Programs Rural credit & cash cropping programs - 0.4 0.7 1.4 2.2 4.6 Irrigation 7.2 9.5 10.2 10.6 10.8 48.3 Ongoing & committed 2.2 1.7 1.1 0.7 0.7 6.4 Medium scale system 1.4 1.0 - - - 2.4 Small scale system 0.7 0.7 1.1 0.7 0.7 4.0 New programs 5.0 7.8 9.1 9.9 10.1 41.9 Support for small scale system - 0.7 1.0 1.4 1.4 4.6 Rehabilitation of existing system 1.8 2.8 2.7 2.8 2.9 13.0 Small and medium scale system 3.2 4.2 5.4 5.6 5.8 24.3 Forestry 8.3 9.9 9.4 11.3 11.5 50.4 Ongoing & committed 3.2 4.9 4.7 5.6 5.8 24.3 Re-afforestation 1.1 1.4 1.3 1.4 1.4 6.7 Conservation & Management 2.2 3.5 3.4 4.2 4.3 17.6 New programs 5.0 4.9 4.7 5.6 5.8 26.1 Re-afforestation 5.0 4.9 4.7 5.6 5.8 26.1 Livestock 0.7 1.3 1.4 1.4 4.9 Ongoing & committed - - - - - - New programs 0.7 1.3 1.4 1.4 4.9 Disease Control programs - 0.4 0.7 0.7 0.7 2.5 Breed - 0.4 0.7 0.7 0.7 2.5 Other Agriculture 4.3 5.3 9.1 12.7 16.6 48.0 Ongoing & committed 4.3 5.3 9.1 12.7 16.6 48.0 Area Developments 3.6 4.2 8.1 11.6 15.5 43.1 Narcotics reduction related development 0.7 1.1 1.0 1.1 1.1 4.9 Total Agriculture PIP 23.4 28.9 34.1 38.5 44.0 168.9 Current Expenditure 5.2 6.1 7.0 8.0 11.0 37.3 Total Agriculture Expediture 28.6 35.0 41.1 46.5 55.0 206.2 Total Agriculture Expenditure as % of GDP 1.8 1.8 1.9 1.9 2.0 1.9 Sources: Department of Public Investment Plan, CPC. - 52 - TABLE 3.5: SUGGESTED EXPENDITURE PLAN FOR AGRICULTURE AND FORESTRY, 1995/96 - 1999/2000 Gov't. PIP PER Proposed Change Comments Bn kip % Bn kip % Bn kip Research and Extension 17.2 10.2 37.0 21.9 +19.8 Continue data collection projects; conduct an agricultural census. Irrigation 48.3 28.6 33.0 19.5 -15.3 Focus on rehabiliting and improving operations of existing schemes; conduct national inventory/evaluation. Forestry 50.4 29.8 43.0 25.5 -7.4 Emphasize conservation; moderate growth in spending. Livestock 4.9 2.9 18.0 10.7 +13.1 Create national livestock and forage improvement program modeled on the National Rice Research Program. Area Development 43.1 25.5 33.0 19.5 -10.1 Moderate growth in spending. Narcotic Reduction-related 4.9 2.9 4.9 2.9 0 TOTAL Investment 168.9 100.0 168.9 100.0 0 Sources: State Planning Committee and World Bank staff estimates. H. RECOMMENDATIONS 61. The Government has made much progress in recent years in defining priorities and targeting its activities on areas requiring intervention. Nevertheless, substantial improvements remain to be made in public spending for agriculture. In terms of the aggregate allocation among subsectors, it is recommended that the Government: => Decrease planned spending on irrigation, forestry, and area development and increase funding for research and extension, livestock health services, and information services (statistics/surveys/publications). 62. Data. The poor quality of agricultural data continues to obstruct effective planning for agricultural development, because in the past the information function of the MAF was a low priority. It is therefore recommended that the Government: => Extend the initiatives begun by the FAO to improve the quality of basic agricultural data regardless of the availability of external assistance. In addition, funds permitting, the Government should undertake a comprehensive agricultural census in 1996/97. 63. Rural Credit. If the APB is the sole provider of agricultural credit, poor farmers are likely to be passed by. The APB is designed to handle medium-size loans beyond the means and requirements of smallholders and it cannot provide and monitor small loans in a cost-effective - 53 - manner. However, some donors have acquired considerable experience working in other countries to help develop local rural credit schemes for smallholders. Thus, the Government should: => Draw up plans for technical assistance to develop small-scale savings/credit institutions targeted at poorerfarmers. 64. Irrigation. Irrigation development is justifiably a major long-term priority for the agricultural sector. Despite shifting toward low-cost farmer-managed irrigation development, the Government investment plans continue to include new, capital-intensive projects. However, a more effective use of limited resources would be to concentrate on rehabilitating and maintaining existing schemes. Improving the quality information for planning and monitoring irrigation development is needed as well. The Government should: > Cease investing in new irrigation works. If new schemes are considered at all, they should be small enough to be managed and maintained by local water user groups (in most cases, schemes serving less than 100 hectares). => Train PAFSO/DAFSO irrigation staff to enable them to determine the demandfor irrigation water through surveys and simple analysis. > Prioritize investments in the rehabilitation of existing irrigation systems. Since a reliable information base is essential, a national irrigation inventory and evaluation survey should be conducted. (Given the limited availability of trained staff in the short run, two to three years will be required before this can be undertaken.) 65. Forestry. Forestry contributes roughly 5% of official agricultural GDP and is a major source of export earnings. Strong Government intervention is justified on both economic and ecological grounds. However, because of weak monitoring of the regional logging SOEs and private road construction, following a coherent strategy to manage forest resources is difficult at best. Greater transparency and accountability are essential. Furthermore, while the Government's stated priorities for the sector demonstrate an awareness of the important problems facing forestry in Lao, spending on reforestation and conservation should not exceed the capacity to use funds efficiently. To strengthen forestry management the Government needs to: = Limit the expansion of public investment in forestry over the next four years. Assess the cost- effectiveness of competing approaches to forest conservation and management to increase the efficiency of future investment. = Improve monitoring of the three regional resource development companies under the MOD. Greater transparency and accountability will allow rational forest planning and management. = Allocate a greater share of the sector's budgetedfunds for the operational expenses of long- duration forestry projects to demonstrate commitment to conservation and improved forest management. 66. Livestock. Livestock development (including fisheries) should be a high priority area for Government action. The sector accounts for about 40% of agricultural GDP, and livestock exports could become very significant if animal health conditions improve. At present, the Department of - 54- Livestock and Veterinary Services is especially weak, with only 15 or so staff at the central level. To effectively assist livestock, Government action should: > Strengthen the capacity of the Department of Livestock and Veterinary Services through training; consider increasing staff => Develop a national livestock and forage improvement program on the model of the National Rice Research Program. > Raise coverage rates by increasing the budget for vaccine procurement and reducing cost recovery; increase trainingfor village para-veterinarians. TABLE 3.6: ALLOCATION OF PUBLIC INVESTMENT IN AGRICULTURE AND FORESTRY, 1990/91-1994/95 (LOCAL RESOURCE COMPONENT, MILLION KIP) Cabinet Office (incL Agriculture & Livestock & Imigation Meteorology & Forestry Rural Personnel & Tech. Schools TOTAL buildings) Extension Veterinary Hydrology Development 1990/91 Capial 47 54 743 28 56 208 172 1,308 O&M 87 69 17 7 15 170 365 Total 87 116 70 743 35 71 208 342 1,673 1991/92 Capital 132 64 761 45 109 533 171 1,816 O&M 22 36 55 56 41 140 188 538 Total 22 168 119 817 86 249 533 359 2,354 1992/93 Capital 35 67 1,743 37 68 248 107 2,304 O&M 20 117 55 55 41 l05 184 576 Total 20 151 122 1,799 78 173 248 290 2,880 v, 1993/94 Capital 30 59 73 2,173 58 110 421 151 3,076 O&M 27 98 65 40 60 125 15 203 633 Total 57 157 138 2,213 118 235 436 354 3,709 1994/95 Capital 76 131 81 3,124 38 73 445 220 4,188 O&M 50 249 88 178 108 221 72 19 985 Total 126 381 169 3,302 146 294 517 239 5,173 1991-95 Capital 106 404 339 8,544 206 416 1,855 821 12,691 % 34 42 55 96 44 41 96 52 80 O&M 206 569 280 329 257 606 87 764 3,098 % 66 58 45 4 56 59 4 48 20 Total 312 973 619 8,873 463 1,022 1,942 1,585 15,789 Source: Ministry of Agriculture and Forestry. - 57- CHAPTER 4: HEALTH A. SECTOR OVERVIEW 1. The health conditions of the Lao people are among the poorest in the world. The Lao health sector is underutilized and underfinanced, and offers low quality care. Annual population growth is estimated at 2.4%, which strains health and social services, employment creation, and food availability. Government funds for health care have been inadequate, and this is compounded by inequitable distribution, fragmented organization, and a high dependence on external aid. 2. Recent increases in the health budget are an important step toward sharing the benefits of economic growth, but resource allocation toward preventive measures must be a priority. Health sector management must be improved by strengthening the national capacity in policy formulation and planning and program implementation. At the same time, improved coordination within the Ministry of Health (MoH) and between the MoH and foreign donors is required. 3. Health Status. The overall morbidity and mortality Major Health Problems in Lao Villages profile in Lao PDR is characterized by diseases that could (% of villages) be substantially reduced by better preventive care, Malaria 86.9% education, and improved access to good quality basic Diarrhea 80.2% health care. Malaria is the most serious public health Respiratory infection 559% problem in Lao PDR, yet until recently prevention has been Source: LECS/LSIS, 1992/93. seriously underfunded. A conservative estimate for 1994 was 1.4 million cases of malaria per year in a then total population of 4.4 million. Malaria is said to be a major health problem in nearly 87% of Lao villages.I Children and pregnant women are particularly at risk; malaria contributes to an under-five mortality rate almost four times higher in rural than urban areas, malaria also causes over one-quarter of antenatal maternal death. Health Indicators 4. Life expectancy at birth is about 50 years. IMR U5MR The infant mortality rate (IMR) has declined only Residence: urban (per '000) (per '000) slowly during the last 15 years, to the current rate rural 136 200 of 113-125 per 1,000 live births, compared to an Economic status: well-off 82 108 average of 71 for low-income economies.2 poor 148 250 Under-five mortality (U5MR) in Lao PDR is Ethnicity: lowlanders 115 161 mid/highlanders 147 230 about 176 deaths per 1000. As is demonstrated by Source: LSIS (1992/3). the table in this paragraph, health indicators vary widely between rural and urban areas and with different socioeconomic groups. Data from National Malaria Control Program (1995), and from Lao Expenditure and Consumption Survey (LECS), 1992/93, and Lao Social Indicator Survey (LSIS), 1992/93. The LECS/LSIS surveys were the first national surveys to collect information and consumption data. The sample size allows breakdowns by region (north, center, south) but not by province. 2 The 1992/93 LSIS estimated 125 deaths per 1,000, while the 1995 LFBSS estimated 113; the average figure of 71 for low-income economies is from the World Development Report 1993. - 58 - 5. Asia exceeded all other regions in improving female life expectancy and infant mortality over the 1965-80 period, but Lao PDR remains far behind its neighbors. Maternal, infant and child mortality rates remain unacceptably high in Lao PDR, and are key factors contributing to overall low life expectancy. They are also areas that can be targeted with knowledge, effective interventions such as health inputs for mothers before and during birth, immunizations, oral rehydration therapy, birth spacing, and nutrition supplements. 6. The major causes of child mortality and morbidity are communicable diseases, notably malaria, acute respiratory infections, measles, and diarrheal diseases. The incidence, severity and outcome of these diseases are aggravated by inadequate nutrition, lack of clean water, poor sanitation, and limited access to health care. These diseases can frequently be averted through effective preventive care. Maternal mortality is also very high, estimated at 660 deaths per 100,000 live births.3 Important risk factors include the high total fertility rate (currently 7.1 children per woman), early and late pregnancies, short intervals between births, and the very low proportion of attended births. Only about one-fifth of married women aged 15 to 49 report ever having used a modern method of birth spacing (LFBSS, 1995).4 7. Health Seeking Behavior. Surveys indicate that people are reluctant to use public Pattern of Health Provider Use health services and that only one of out ten Type Bottom 20% Top 20% Average Laotians visits modern facilities. Over half of NTcare 18 4 13 survey respondents self-medicate as an initial Traditional 25 10 17 response to illness; one-quarter sought a private Pharmacy 49 63 58 consultation. Patients do not shift to the public Modem 8 22 12 health sector if complaints persist. Only 4% of Source: LECS/LSIS, 1994. respondents chose to use sub-district services as an initial response, but even fewer chose it as their second choice of treatment.6 Overall, only 7% of women reported attending a hospital or clinic for childbirth; almost 40% chose to deliver in their own homes without assistance (see also Table 4.8). 8. Field surveys (LSIS; MoH/UNICEF) indicate that people are reluctant to use public services because of factors which generally reflect the poor quality of care available. These factors are particularly acute in rural areas and include distrust of, and dissatisfaction with, public health facilities; lack of adequate drugs and equipment; lack of basic infrastructure such as water supply in most facilities; geographical inaccessibility of facilities; and cultural and linguistic inaccessibility of public services to ethnic minorities. Utilization patterns also indicate that public hospitals were used for treatment in the later stages of illness, which is more costly and creates an inequitable distribution of resources. It is also likely to hinder successful treatment. 3 LSIS, 1992/93. 4 Recent data are available from the Lao Fertility and birth Spacing Survey (LFBSS), published in November 1995. 5 Lao Social Development Assessment and Strategy (LSDAS), 1995. 6 Vinard, 1994, in IDS, 1995. - 59 - B. GOVERNMENT OBJECTIVES FOR HEALTH 9. Background. The health care system in Lao PDR is relatively new. When the Lao People's Democratic Republic was proclaimed in 1975, the health infrastructure inherited from the French colonial regime consisted of urban and semi-urban hospitals concentrated mainly in Vientiane. The new regime developed an ambitious health plan focusing on rural areas and basic preventive services, resource constraints have impeded any meaningful implementation of these goals. 10. At present, the public health policy framework is the Government's Socio-Economic Development Plan 1993-2000. Within the Plan, Human Resources Development (HRD) is one of seven priorities, and within HRD, Public Health is one of nine programs. The only targets for improving specific health status indicators are for average life expectancy; the Plan states that it should be 56 years for men and 58 years for women by the year 2000. The Plan also states that morbidity and mortality rates for those diseases that are preventable through immunization should decline but no specific targets are set. Health sector activities are explicit in only three disassociated areas: immunization, cost recovery and payment schemes, and consumer safety. 11. The Government has recently adopted some health sector policies and action plans, including the National Drug Policy (1993); the National Malaria Control Program Plan (1994-2000); and HIV/AIDS Control. These plans are useful steps forward in policy development, including identifying the causes and patterns of disease, and developing properly costed action plans which can guide government and donor resource allocation. Nevertheless, policy development has been largely donor driven. The challenge for the MoH will be to integrate these initiatives into a coherent health services delivery strategy, to identify priorities for future policy development, and to make associated requests for donor support. 12. Translating Objectives into Resource Allocation. Some major donors -- WHO, UNICEF, and SIDA -- have been implementing health management development projects with the MoH. In 1993/94, the MoH with donor assistance began developing a system of program budgeting, in which both government and aid expenditures were divided into nine programs. In the 1995/96 fiscal year, the MoH began collapsing these into a six-program budgeting scheme: Health Protection and Promotion, Treatment and Rehabilitation (curative care), Consumer Protection, Human Resource Development, Medical Science Research, and Administration and Management (Table 4.7). Each program distinguishes investment and recurrent spending. Personnel costs, however, are not allocated by program but are aggregated under the Administration and Management category. 13. The new budgeting system should lead to a clearer picture of the relationship between health priorities and spending patterns. It is an important improvement over the previous incremental, line- item system. The budget planning meetings are also useful to provincial officials to discuss health priorities and share information. Nevertheless, the new system still will need to address other issues: => Clarify allocation of resources among provinces and between rural and urban areas. For the next budget (fiscal year 1996/97), the MoH is working with districts to examine how to divide the budget between hospitals and other programs. => Reduce the discrepancy between budgeted and actual spending. Table 4.9 shows that while 69% of the health resources had been budgeted for the provinces, only 61% of spending took - 60- place at that level, with the difference being spent at the central level.7 Given the higher incidence of disease in the rural areas, this budget reallocation is neither efficient nor equitable. => Set aside funds for recurrent costs, particularly as capital spending is projected to increase. > Reduce the fragmentation of the health system into vertical programs. This separation into parallel, rather than integrated, health departments reflects donor preferences, but does not encourage efficient health care provision, coordination among sector programs, and an adequate rural-urban balance. > Devolve decision-making to provincial health officials to strengthening health systems. Health sector administrators must be trained in financial planning and management. Public health sector administration in Lao PDR is largely conducted by physicians who have limited management skills. The relevance of policy formulation, standard setting, and management as the major role for the central government in the health sector is still to be recognized, and capacity at the provincial level is mixed. External assistance has played an important role in supporting provincial level efforts, and there seems to be a movement to support the development of provincial health systems instead of investing in isolated efforts.8 Over time, health systems could be strengthened by increasing the autonomy of provincial health officials, provided they have the right training. C. FUNDING MECHANISMS 14. Household spending is estimated to comprise over 50% of the total expenditures on health. The Government contributes about one-third,9 with the remaining funding coming from donors. 15. Government Spending. National revenues fund > . > . ' . . , ~~~~~~~~~Health Expenditures, by Source, 1994/95 four levels of care in the Lao health system: central, Tot. kip % Tot. Kip per provincial, district, and sub-district "health posts," (millions) capita "health centers," etc. The army and the police have Gov't 11,700 31.6 2,554 separate health care systems. During the 1986-91 period, Aid 6,152 16.6 1,343 the provincial health offices were the responsibility of the Household 19,118 51.7 4,173 provincial governor, and were financed directly from provincial resources. The recentralization in 1992 has meant that provincial offices now receive public resources from the MoH. 16. The budget process begins with provincial officials drawing up a draft budget, using program guidelines provided by the MoH. These guidelines are still being developed, and one aim is to make the funding that is directed to activities below the provincial level more transparent. Then 7 Some of the spending may have been allocated to central institutes, which then spent the funds at the provincial level. s At the provincial level, a few provinces have a health system reform under way or have just initiated the first steps toward establishing one. Among them are Xayabury, Phongsaly, Huaphanh, Savannakhet, Sekong, Saysomboun, Xiengkhuang, and Oudomxay. 9 This increases slightly when both recurrent and investment budgets are considered. Public health resources in the PIP are about an additional 10-15% of what is in the recurrent budget (for recent years and estimates for the 1996-2000 PIP). - 61 - the draft budgets are discussed and reviewed by central MoH staff and provincial health officials. The province then forwards the approved budget to the provincial SPC and Finance Offices, and the MoH. The MoH submits a budget request, which includes both central level expenditures and provincial expenditures, to the MoF. Once the final budget envelope is provided by the MoF, the MoH decides on allocations among provinces. Allocations are based on perceptions of provincial needs, taking into account the human resource capacity to carry out the proposed programs. The budget for pharmaceuticals is the only item calculated on a per capita basis. 17. Provinces do have some flexibility in reallocating budgets between different programs, although they must consult the respective departments at the MoH. Provinces can also request supplementary funds from individual departments if necessary, and the MoH has a line item "reserves" to meet such needs. 18. Household Spending. Of the total expenditures on health, household spending is estimated to comprise over 50%, which mainly consists of spending on pharmaceuticals. As is shown in Table 4. 1, there are significant differences in the total spending of poor and well-off households, although the differences between urban and rural households are not as marked. Even better-off and urban households, however, spend little on treatment. TABLE 4.1: HOuSEHOLD CONSUMPTION, BY ECONOMIC STATUS AND URBAN/RURAL Consumption Poor Better off Urban Rural item % kip % kip % kip % kip Food (total) 75 8,660 54 16,790 54 15,880 66 11,380 Medicines 2 240 2 700 2 530 3 430 Treatrnent 0 0 0 10 0 10 0 0 Source: LECSILSIS, 1992/93. 19. Foreign Aids External resources data for Sources of Aid health are neither precise nor comprehensive. For 1989 1992 example, in the table at the right, "aid" is only that Bilateral 330 12% 744 12% which is recorded by the SPC, and may significantly Multilat. 1,811 64% 3,141 50% underestimate the total by not fully capturing NGO NGO 686 24% 2,352 38% resources. This is expected to change because the Source: IDS, 1995. 6,237 100% International Cooperation Unit in the MoH is now compiling data, including registering NGO activities, to have more accurate information beginning in the 1995/96 fiscal year. The data will also be reported to the SPC. 20. Health sector resources have been substantially augmented by the World Bank's US$19.2 million Health System Reform and Malaria Control Project, and the ADB's health project. However, while this will increase the proportion of health funding by multilaterals, it is also clear that NGOs have accounted for a significant portion of extemal resources. There are about 54 NGOs active in health. In recent years the largest funders have included Save the Children (Australia and the United Kingdom), Medecins sans Frontieres (France and Belgium), CARE, and CIDSE. Many NGO projects are implemented at the provincial level and focus their activities on districts and villages -- mainly on primary health care, matemal and child care, and, increasingly, HIV/AIDS prevention. Aid resources are expected to finance most of the future investment in the health sector (see Public Investment Program [PIP], below). - 62 - D. EXPENDITURE TRENDS 21. Overall Health Expenditures. Government expenditures on health declined dramatically in the 1989-91 period, both as a percentage of total government expenditure, and as a percentage of GDP (see Table 4.6). Even after recovering in the 1992/93 fiscal year, government health spending reached only 0.7% in 1992/93, which compares to an average of 1.8% (in 1990) for Asia, excluding India and China. This proportion has remained relatively constant over the past three years, as GDP has also been increasing. As a proportion of total government expenditures, health received about 3% over the 1992-95 period; this is projected to increase to 3.8% in the 1995/96 budget. 22. In current prices, government health expenditure per capita was equivalent to less than US$1 per capita in 1990, rising to almost US$2 in 1992/93. This compares to 1990 estimates of US$1 in Vietnam, US$3 in Nepal, US$5 in India, and US$7 in China. In nominal terms, government per capita spending increased to about US$3.50 in fiscal years 1994/95 and 1995/96 (planned). 23. Per capita health expenditures will remain low even if the budget for health increases as indicated in Table 4.2. By the year 2000, nominal expenditures would reach about 11,400 kip per capita, which would be about US$12.35.10 However, in 1995/96 the MoH budget was only about 65% of that requested. Should a similar pattern occur in future years, per capita expenditures would gradually increase to only 7,400 kip per capita, or approximately US$8. 24. The old budget structure made it difficult to distinguish patterns of expenditures, but Table 4.2 groups line items in order to approximate such a breakdown. Salaries and allowances clearly accountfor a major part of the MoH budget. They increase from just over half in 1992/93 to about two-thirds of the total recurrent budget, proportions which are quite common in developing countries. A better breakdown of the data is necessary in two areas: - Salaries by types of health personnel (What is the proportion of professional to administrative staff?) - "Repair and new purchase" (How much is spent on supplies and how much on maintenance ?) 25. Pattern of Expenditures. The new program budgeting framework more explicitly distinguishes expenditures by types of service. About 40% of proposed MoH spending is budgeted for curative services (Table 4.7), mainly hospitals, and estimates indicate that this proportion also prevailed in the early 1990s. Administration accounts for over 40% in 1995/96, but is supposed to be trimmed to 31% by 2000. Construction makes up over 18% of MoH expenditure. Data are available on the share of hospitals (secondary and tertiary) in health budgets, but this is not strictly comparable with Lao PDR since the Lao data do not allocate staffing costs by facility. If this were done, Lao PDR would probably be close to the averages reported in other Asian countries. Hospitals are estimated to account for about 50-60% of budgets in Bangladesh, Indonesia, Nepal, and the Philippines, and over 70% in India and Sri Lanka. 10 Using an exchange rate of 925 kip to the US$, as indicated by IMF projections. - 63 - TABLE 4.2: ECONOMIC COMPOSITION OF PUBLIC HEALTH EXPENDITURES (MILLIONS OF CURRENT KIP) 1992/93 1993/94 actual 1994/95 actual 1995/96 approved m.kip % total m.kip n m.kip % total m.kip % total Salaries and allowances 2,214 51 3669 56 6,684 66 7,171 64 Drugs 900 21 800 12 1,105 11 1,555 14 of which: central* 371 (34) 655 (42) provincial 734 (66) 900 (58) Repair+newpurchase ** ** 758 12 769 8 781 7 Other recurrent*** 1,267 29 1,350 21 1,538 15 1,677 15 Total recurrent 4,381 100 6,577 100 10,096 1)0 11,185 100 Recurrent 4,381 72 6,577 85 10,096 86 11,185 81 Capital 1,727 28 1,155 15 1,605 14 2,636 19 * Note that some proportion of central allocations may be spent in the provinces by the Department of Hygiene. ** "Repair + new purchase" not given as separate line item. ** "Other recurrent" includes administration (include. electricity, water, etc.), training, and food subsidies. Capital costs are primarily construction. Sources: 1992/93 data from IDS (1995). Otheryears from MoH data. 26. Center-Province Distribution. About 35-40% of Distribution of Central Level the government health distribution is at the central level, Spending, 1994/95 and 55-60% goes to the provinces. The table at the right (million kip) shows that central spending consists primarily of budgets total / for hospitals, training centers, and research institutes. Cabinet 384.5 9.1 Some of the budget for the centers and institutes may Centers, institutes 565.7 13.4 University, regional actually be spent in the provinces, for example, on pilot schools 990.1 23.5 projects or in response to specific needs or requests. Central hospitals 1,951.4 46.3 However, as the numbers in the table show, these amounts Other + reserve 324.8 7.7 are usually small. Total 4,216.5 100.0 Source: MoH, 1994/95 budget. 27. Distribution by Provinces. Inequitable distribution by province appears to be increasing. The absence of provincial data makes it impossible to ascertain whether the existing wide differences in per capita expenditures among provinces reflect provincial needs (see Table 4.10). Further investigation would be needed to reveal the distribution of funds within provinces, particularly to rural areas. One concern, however, is that Vientiane Municipality receives far more than the average provincial per capita expenditure in 1994/95. This is in addition to expenditures at the central level, which largely benefit facilities in Vientiane Municipality (Table 4.9). If both provincial and central expenditures were examined, it is clear that there is a large and possibly growing bias toward Vientiane. 28. This allocative pattern in Lao PDR indicates inequities in the system, given that funds are mainly directed toward urban hospitals, particularly those in the Vientiane area. The disease profile in Lao PDR argues for increased allocation to rural areas. Government allocations to preventive care are expected to increase from 14% of the total budget in 1995, to approximately 20% in 2000. Providing services in lower-level facilities contains costs at more modest levels for minimal versions of the essential clinical package. - 64 - 29. Administrative Structure. The current central administrative structure for preventive medicine must be rationalized to deliver services at a level consistent with the morbidity rates in the country. According to the WDR 1993, the most sophisticated facility required to deliver the minimum elements of the essential clinical package is a district hospital. At the central level, the hygiene and prevention system includes as many as nine institutes and centers, most of them heavily dependent on external resources. In contrast, at the provincial level three of these institutes and centers are represented: the Mother and Child Care Section, the Malaria Section, and the Sanitation Section. The resources and administration at the center are diluted among nine institutes and centers, while the provincial level lacks the support for preventive services. 30. Composition of Donor Funding. Current data also do not allow an analysis of the distribution of donor resources in relation to need. Some indicative data available for malaria control (Table 4.8) show that donors do not always choose the provinces with the greatest need, owing to difficulties with security, transportation and communication facilities, institutional capacity of the province, and accessibility to Vientiane. Few data are available to show the functional composition of external support. An estimate from the MoH for 1992/93 suggests that total donor support is about equally divided among four categories: (i) experts, (ii) construction and vehicles, (iii) medical equipment and pharmaceuticals, and (iv) fellowships and study tours. E. EFFICIENCY OF PUBLIC SPENDING ON HEALTH 31. Internal Efficiency. Quantitative indicators of Lao health care provision compare favorably to other low-income countries but do not show a true picture of accessability of the health system. Each of the 18 provinces has a provincial hospital, and most of the country's 126 districts have a district hospital with 5-30 beds. Health centers below the district level, mainly in rural areas, increased significantly in the late 1980s, to reach 1,190 in 1989, though they decreased to 723 in 1993. This left an average of one health post per 6,188 people, which would be a reasonable ratio for a low-income country. The overall number of hospital beds available in 1993 was 4,904, implying an average population-to-bed ratio of 912. The average for Asia, excluding India and China, was 556 (WDR, 1993). The provision of higher level health professionals (mostly doctors) also compares well with regional averages: in 1993 there was an average of one higher level worker per 3,184 people, compared to the Asian average (excluding India and China) of 3,226. 32. As was previously indicated, the health system outside of the Vientiane area is underutilized, which indicates low internal efficiency. Considering the decreases in real health expenditure at the end of the 1980s, occupancy rates in central hospitals fell to about 50%, provincial hospitals to about 40%, and district hospitals to about 10%. Utilization of health posts is also very low, and many, perhaps most, of the posts provide little or no service. Health workers are also in great need of refresher training, and, as the real value of salaries fell in the late 1980s/early 1990s, more health workers turned toward second jobs. 33. The reduction in the number of health personnel Health Personnel, Total in the 1990s may have decreased the efficiency of the Level 1987 1992 health system. More of the lower level staff (nurses, Higher 901 8% 1,246 13% pharmacy technicians) have left, compared with the Middle 2,977 25% 2,758 29% higher level staff (doctors, diploma pharmacists), which Lower 7,883 67% 5,417 57% can probably be explained by the lower salaries for I Total 11 761 100% 9,421 100% lower and middle level staff. Moreover, staff reduction has taken place at the provincial level; over the 1988-92 period, staff at the central level actually increased. The result is that, as in many - 65 - countries, a significant proportion of the most qualified staff reside in the Vientiane area while the major health problems are in the rural areas, and this trend has been increasing in recent years. The MoH is presently planning to pilot the payment of additional benefits to staff in isolated areas, to try to encourage health staff to stay. 34. The number of children receiving a full three-dose DPT (diphtheria-pertussis-tetanus) immunization provides a reasonably good measure of the efficacy of immunization programs because administering a complete sequence represents a considerable motivational and logistical achievement for the health system. Lao PDR lags behind many countries on this indicator (see Table 4.3). 35. The health sector funding favors urban areas, which results in low quality services available in the rural areas. Low quality or non-existent care at lower-cost facilities also leads patients to bypass these facilities. The referral system is creating ineffective higher costs, as treatment costs are greater at higher level facilities. This is also inequitable, since it is the poor and rural population who are least able to afford the travel costs to higher level facilities. These inequities result in poorer health for rural populations. 36. External Eff iciency. " I The low quality and low productivity of resources in the health sector is one of the reasons for the poor health status of the Lao population. In rural areas, the low level of efficiency of public facilities is compounded by the few alternatives available. Only 38% of rural villages report access to a pharmacy within 3 km (compared to 92% of urban villages), and the corresponding figures for dispensaries are 30% and 85% (SDAS, 1995). 37. Total spending, however, has also been low (Table 4.3). In terms of cross-country comparisons, the variation in health outcomes stems partly from differences in economic circumstances, as well as in education and other variables, but health policies and financing choices also exert a significant influence. Other countries with fairly low population densities (e.g., Nepal, Uganda, Madagascar) display fairly similar outcomes for similar levels of per capita spending, although the priority placed on health may be inferred by their higher amount of public resources allocated to health. A densely populated country like Bangladesh shows somewhat better indicators with slightly higher spending. Nevertheless, very different outcomes are possible from similar levels of spending, and a good example is the achievement of Vietnam in health. 38. Thus, while the overall resource effort is important, Lao PDR will need to achieve a better impact per health dollar spent if it is to make any significant progress in the health status of its population. Government spending is planned to increase at about 29% annually until the year 2000. Nevertheless, assuming that the population growth rate has fallen to 2.4%, this will result in nominal per capita spending of about 11,428 kip, equivalent to (assuming an unchanged exchange rate of 925 kip to the dollar) approximately $12.35. If current patterns are maintained, it appears that Lao PDR will still be far from achieving the recommended type of government expenditures comprising the US$12 per capita recommended as a minimum public health package and essential clinical services (WDR, 1993). The ability to link health inputs to outcomes is limited by the paucity of disaggregated outcome data. The first national level survey to collect information and consumption and social conditions was conducted in 1992/93, but the sample size allows breakdowns by region and not by individual provinces, nor by urban-rural areas within provinces. - 66 - TABLE 4.3: HEALTH SPENDING AND HEALTH INDICATORS, SELECTED COUNTRIES, 1990 Health Health expenditure as % of Life Perinatal U5MR % children spending GDP expectancy mortality (per'000) immunized p.c. ($) at birth (per'000) total public private Lao PDR 5 2.5 1.0 1.5 50 85 171 22 Nepal 7 4.5 2.2 2.3 56 90 135 74 Bangladesh 7 3.2 1.4 1.8 56 75 137 87 Vietnam 2 2.1 1.1 1.0 67 40 46 85 Uganda 6 3.4 1.6 1.8 47 85 185 77 Madagascar 7 2.6 1.3 1.3 51 76 170 46 Developing 41 4.7 2.3 2.5 63 45 106 80 countries Note: Developing countries figures are weighted averages. All figures refer to 1990, except for children immunized (third dose of DPT) which is 1990-91. Source: WDR, 1993. The Lao PDR figure for perinatal mortality is far lower than the most recent estimate of 113 (over the 1989-94 period, LFBSS), however, it is used for comparative purposes, as other country data might show similar biases. 39. Redistributing public spending is also critical to achieving better outcomes. Government spending should ensure that the health system plays a positive role in assisting the poor to benefit from expected rapid economic growth. At present, the pattern of how different income groups use government health facilities is even more regressive than the distribution of household expenditures. No health care responses prevail as income declines. Among lower income groups, almost I in 5 individuals do not use any form of health care, not even traditional or home remedies, compared with less than 1 in 20 among higher income groups. 40. In the long term, a more effective integration of the various health programs will be critical to achieving better outcomes. Donors will also have to play a role in assisting the MoH in this regard, as they have largely funded the vertical health programs. In addition, this has solidified the situation of fragmented, specialized institutes at the central level, which continues to place a heavy burden on the government health budget. This process can begin by encouraging programs and pilot activities to improve integration of preventive health measures at the local level. F. SCOPE FOR PRIVATE SECTOR PARTICIPATION AND COST RECOVERY 41. The private sector is already very much in evidence in the health sector, mainly in drug retailing, but also in some medical services. The official reasons for allowing the development of a private health sector are (i) the inadequate salaries of public health workers and (ii) the inability of the state to finance the supply of medicines. In 1990 the Government gave permission for state employed doctors who had worked in the public service for at least seven years to run private clinics after working hours, and retired physicians were also allowed to practice privately. At present, it is still not legal for nurses to practice privately, although in fact this is not uncommon. Private clinics are increasing, though the majority are located in Vientiane. Data from 1993 indicate that there were 44 clinics in Vientiane and 18 in the rest of the country. 42. Most pharmaceuticals are produced in two state factories--about 90% by value of the drugs manufactured in Lao PDR (IDS, 1995). The factories have no budgetary implications for the MoH, as they are completely self-financing, and, moreover, hospitals can choose to buy drugs from either the factories or other suppliers. Hospitals negotiate directly with suppliers. Invoices are reimbursed - 67 - by the MoH, provided that the drugs fall within the accepted list and the hospital remains within its budget envelope. Prior to 1988, state factories had to sell exclusively to the Government, but since then it is estimated that purchases from private and donor clients make up some two-thirds of their revenues. 43. The commercial import of drugs is a growing business, and the number of private import- export companies dealing in drugs grew from 12 to 29 between 1990 and 1992 (IDS, 1995). There is also a sizable, but unquantifiable, unofficial import of drugs. 44. Private pharmacies are widespread. In 1994, there were about 1,700 registered pharnacies, but it is also known that there are a substantial number of unregistered outlets. Although in theory the MoH registers private pharmacies in three categories, the criteria for classification are not applied systematically and they are not well regulated. Hence, they raise both public safety (quality of drugs, poor prescribing and treatment) and equity issues. The Government has responded with the National Drugs Policy (MoH, 1993) to control and monitor the private phartnaceutical sector. The success of the policy remains unclear, and its implementation is a challenge to the capacity of the various levels of the public health system. 45. The other face of the private sector is traditional medicine. There is a well-established commitment to integrating traditional and modern medicine in Lao PDR, including support to the Research Institute on Medicinal Plants. There is one private traditional medicine hospital in Vientiane. Most village markets have small traditional medicine shops, and a large but unknown number of practitioners operate throughout the country. The popularity of traditional medicine was also revealed in the LSIS/LECS survey, with 17% of the population choosing visits to traditional practitioners when ill, a proportion which rises to 25% if the bottom quintile of the population is taken into account. 46. In 1992, the MoH introduced a more formal cost recovery policy in the public health sector, and while a variety of schemes exist, charges are made mainly for drugs. Since 1988, hospitals in Lao PDR have been allowed to charge for some consumable items, although these also remain part of the public health budget. 47. Household survey data indicate that spending on health mainly consists of drugs purchases from the private sector. Much of this is purchased without prior prescriptions. Another typical pattern of pharmaceutical use is for staff in public facilities to diagnose and prescribe medication, which patients will then buy at private pharmacies and, if necessary, return to the health provider for treatment. While some data indicate that cost recovery in the larger hospitals might be substantial,'2 this would be true of Vientiane but atypical in the provinces. 48. The development of private clinics signals a willingness to pay for services. This could raise expectations about the quality of care, and, if it encourages payment by those who can afford to pay, it could leave more resources free in the public sector. However, user fees could create a two-tier medical system because of the high incidence of poverty. The MoH does not have the capacity to 12 In 1993, Mahosot Hospital in Vientiane, the largest hospital in the city, recovered about 150 million kip in patient fees, of which it kept half and returned half to the MoF. The amount can be compared to the hospital's annual salary costs of about 300 million kip and its operating budget of about 100 million kip received from the Government in that year. (Vinard, 1993, cited in IDS, 1995). - 68 - target the remaining government subsidies to the neediest, which would be a necessary complement to a user fee strategy. 49. Revolving drug funds13 are also appearing, financed either by government or donor resources. However, a recent Prime Ministerial decree (No. 52/PM, dated June 25, 1995) requires the transfer of 20% of collected moneys to the Treasury. The exact application of this decree remains unclear but it may undermine the incentives of local hospitals and health centers for cost recovery.'4 If it is to apply to revolving drug funds in the public sector, it could act as a tax for drug purchasers, since the revolving fund must add this on top of drug costs if it is not to decapitalize its stock. To the extent that this transfer applies to all funds, it could penalize the poor and reduce the competitiveness of such funds with the private sector. Since drug supply is an important element in attracting patients to public facilities, extra costs could be a disincentive for patients to attend. G. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM 50. The Government's efforts to improve health conditions are reflected in increasing budget allocations to health. The 1996-2000 PIP plans to raise investment to 120.9 billion kip, or 8.1% of total PIP, and the share of current expenditures on health will be boosted from 2.6% of total government expenditures in 1995/96 to 5.1% in 1999/2000. Overall, health expenditures are supposed to increase from 1.5% of GDP to 2.5% of GDP by the year 2000 (Table 4, Statistical Appendix to the Main Report). Still, the 1991-95 PIP allocated only 18.7 billion kip (3.5% of the PIP) to public health. Most of the investment was targeted to disease prevention in programs for safe drinking water, maternal and child health (MCH), and the analyzing of food and pharmaceuticals. As discussed in Section A, investment in these areas is critical to improving health status. 51. The bulk of external resources -- almost 94% -- is allocated to program six, "Administration and Management" (see Table 4.7). 15 This amount comprises both donor resources going to the center, and project resources which are allocated and/or administered through the center. Hence, although the PIP format uses the same six-program framework developed under the new budgeting system, this presently gives very little indication of the functional direction of aid, particularly in terms of curative and preventive investments. Table 4.4 shows the breakdown of the foreign resources allocated to this category -- 12,332 million kip, or US$13.4 million -- by external donors.16 Allocating these amounts to each of the program categories would significantly change the percentages of foreign resource allocation suggested in Table 4.5. 52. The domestic funds allocated to other program categories are mainly for construction, and nearly all of this amount is for construction or rehabilitation of facilities in Vientiane. However, the absolute amount of resources is small, particularly since the total domestic allocation to the PIP is 13 The initial stock is usually provided by the public sector, financed either by govemment or donor resources. Drugs are then sold at cost plus some margin, which is sufficient to replenish the stock in an ongoing (or "revolving") fashion. Implementation of such schemes requires training for both establishing and managing the fund. 14 Article 14 specifies that "the remaining 80% of the funds are to be used as planned" and "will be supplemented by the Budget if they are not sufficient." 15 About 27% of domestic resources are allocated to program six. Of this, about 57% are allocated to spending at the main office of the Ministry, and the remainder is largely counterpart funds. 16 Funding from bilateral donors is shown as integrated into each of the six programs. - 69 - 1.2 billion kip, compared with 13.8 billion kip of domestic resources in the annual budget for 1995/96. TABLE 4.4: EXTERNAL FINANCE, 1995/96 PIP Donor Percent of Total Foreign Investment World Bank 36.6% UNICEF 26.2% WHO 14.8% ADB 5.6% Other UN 8.2% NGOs 8.7% Total kip (millions) 12,332.5 Source: Lao PDR Government data. 53. There are three major concerns regarding the dramatic increase in projected PIP investments and MoH expenditures to 1999/2000 in Table 4.7. In particular, it is questionable whether this is a realistic plan that can be implemented. As discussed in Chapter 1 of the Main Report, there is concern, first, that the projected GDP growth which underlies the revenue and expenditure planning in the PIP is too optimistic. The MoH should develop an alternative scenario, with more realistic assessments of the increase in resources, both domestic and donor funding, and forward looking estimates of total costs of investment programs. * The graph shows that the projected PIP increase over Health_PIP:_Projected the 1996-2000 period -- although in nominal figures -- Health PIP: Projected is very ambitious. The weak implementation capacity Billion kipv in the health sector raises some doubts about the level 3 30 of activity implied by the resource projection in the PIP (Table 4.5). 15 & . * The PIP also envisages a significant increase in 10 , . *....i domestic resources devoted to health investment 5 4 (Table 4.5). Domestic resources are expected to grow o between 37 and 46% each year in nominal terms, 1991 1993 19 1997 1999 tapering off to a 30% increase in FY1999/2000. est. However, past data show that compared with other . Tar9et -gActual sectors, the social sectors, and health in particular, have significantly underspent in relation to PIP targets. Over the 1991-95 period, total investment spending was only 62% of planned spending, and was well below 50% of planned spending in 1993 and 1994.'7 The SPC gave two reasons for the lower 1991-95 PIP disbursement: delays in the expansion of some public health facilities in certain provinces; and delays in soliciting assistance for malaria control and the MCH program, particularly for vaccination. 18 17 The figure for 1995 in the graph is an estimate. Is It appears that most of the underspending in health investment resulted from a shortfall of domestic spending. This in turn brought down donor spending if the Government was unable to provide counterpart funds. Other sectors such as - 70 - * Over three-quarters of health investment, totaling about US$100 million over the five- year period, is expected to come from donor resources which are expected to increase about 17% each year. This may be unrealistic, since donors are already heavily financing health. These substantial projected flows would increase dependency on donor resources. If minimum preventive and basic health care needs are to be met, the MoH must allocate resources even if donors do not. 54. The second concern is that the MoH must have a better understanding of the recurrent costs associated with construction, and with equipment purchase. Some sample studies would be useful for this and for better budgeting for future maintenance costs. Studies would also indicate whether the pace of construction is sustainable in terms of future calls on the budget, or whether resources would be better allocated to other cost-effective expenditures such as drugs and supplies. Donor spending of 91.7 billion kip is planned over the 1996-2000 PIP, and past estimates indicate that about one-quarter is allocated to construction. (Nearly one-fifth of the total government budget each year is allocated to construction, but the resulting recurrent costs have not been clearly estimated.) One positive development is that in 1995/96, the SPC allocated a specific amount to the MoH for counterpart funds, which also appears in the MoH annual budget. TABLE 4.5: PUBLIC INVESTMENT PLAN FOR HEALTH, 1995/96 - 1999/2000 (BILLION KIP) Total '95196- 1995/96 1996/97 1997/98 1998/99 1999/00 '99/2000 Ongoing & committed 8.0 1.9 0.9 1.0 1.0 12.8 Health & nutrkim program 3.4 - - - - 3.4 Basic heakh services Savannakhet, Sekong 0.7 0.6 0.7 0.7 0.8 3.5 Other projects 3.9 1.3 0.2 0.2 0.3 5.9 Newprograms 7.4 17.2 22.4 27.7 33.5 108.1 Malaria control (IDA) 3.4 5.4 7.8 8.4 7.7 32.8 Vaccination - 1.1 1.7 1.2 1.3 5.2 Driking water programs - 1.1 1.7 2.4 3.9 9.0 Sanitation - 1.1 1.7 2.4 3.9 9.0 Nutrkim 0.6 1.1 1.1 1.2 1.3 5.3 Birth spacing & maternal & chld heakh 0.6 1.1 1.1 1.2 1.3 5.3 Hosptal & curative services - 1.1 1.1 3.6 6.4 12.2 New rural upland programs 2.9 5.4 6.2 7.2 7.7 29.3 Total Health PIP, of which 15.4 19.1 23.3 28.6 34.5 120.9 Domestic Finance (/) 17.4 19.4 23.2 26.6 28.7 23.1 Foreign Finance (%/) 82.6 80.6 76.8 73.4 71.3 76.9 Current Expenditure on Health 10.1 15.0 20.0 26.0 34.0 105.1 Total Health Expenditure 25.5 34.1 43.3 54.6 68.5 226.0 Total Health Expenditure as % of GDP 1.6 1.8 2.0 2.2 2.5 2.1 Sources: Governnent authorities and World Bank staff estimates. communications have had less of a gap between planned and actual domestic resource allocation. Health underspending appears to result from changing Government priorities and delays in project implementation. - 71 - 55. The third concern is that, given the magnitude of the public health problems discussed in Section A, the percentage of the government budget being allocated to preventive health is inadequate. To some extent, this reflects the fact that the Government is inviting donors to fund the main preventive health programs. However, this represents some risk on the funding side, and is also a deterrent to the Government's eventual need to develop a more integrated system for planning, budgeting, and delivering preventive health services, particularly at the provincial and lower levels. 56. Although the balance between preventive and curative programs in the budget (20% versus 36% of total MoH budget by the year 2000) appears quite reasonable, improvements in national health indicators will require that expenditures within the preventive and curative programs be better targeted toward lower level facilities and away from central hospitals. At present, it appears that government resources for preventive programs are mainly directed to construction facilities in Vientiane. For both curative and preventive programs, health spending needs to be better directed to the village level. Finally, the PIP should break down the functional allocation of donor resources within program categories, in order to have a clearer picture of where the resources are directed. H. RECOMMENDATIONS 57. The Lao health system faces considerable challenges. Underutilization of facilities indicates an inefficient use of resources. Moreover, there is poor distribution of government resources, particularly at the central/provincial level, but also among provinces, that needs to be corrected through a more effective health planning and management system. In terms of equity, the better-off benefit more than the poor from public health spending. 58. In countries such as Lao PDR, where life expectancy is short, health improvements will be tied closely to reducing infant mortality and the mortality risk for women during their childbearing years; in addition, reducing infectious and parasitic diseases merits Government spending on the grounds of economic efficiency. Not only are these measures critical for reducing child morbidity and mortality, but they will also decrease death rates in all other age groups. 59. With the low level of spending in Lao PDR, the key issue for public financing of health is prioritizing expenditures. In the short term, it will be critical to address the level and composition of expenditures, cost recovery, and budget planning. In the longer term, the priorities should focus on the objectives of integrating vertical programs and rebuilding the referral system. = Increase total resources for health. Implement the projected MoH budget (Table 4.7), augmenting the figures annually for inflation, so as to ensure that the minimum recommended level of 11,400 kip (approximately US$12.15) per capita is reached in real terms by the year 2000. With appropriate reallocations within the MoH budget, this level of health spending will move Lao PDR closer to the recommended basic package of US$12 per capita in low-income countries. => Improve allocation of resources. (i) Ensure that sufficient resources are allocated to preventive care and basic clinical services. A basic package focusing resources on lower levels of care will be most cost-effective in addressing the current morbidity and mortality profile in Lao PDR. (ii) Prioritize sufficient funding within the preventive and curative programs to better targeted - 72 - services, toward facilities at the village level and away from central hospitals. This should focus on preventive care (including immunization, information and services for birth spacing, pregnancy-related care--particularly tetanus inoculations--and safe, attended delivery, and malaria prevention) and basic services (particularly care for the common serious illnesses of young children, (diarrheal disease, ARI, measles, and malaria). (iii) Increase investment expenditures toward provinces. The 1995/96 PIP transfers about 16% of domestic resources directly to the provinces. The MoH should consider increasing this proportion, as a means of ensuring that resources are directed toward needs in the provinces. Expand access to safe pharmaceuticals. The recent Treasury decree requiring 20% remittances could negatively affect the functioning and expansion of revolving drug funds, and such funds should be exempted for equity reasons. In the short term, expanded and safer access to pharmaceuticals is an area with potentially high cost-effectiveness, and drugs purchase already represents an important component of household spending on health. Implementation of the National Drug Policy is another means for the MoH to better ensure the quality and safety of drugs on the market. Strengthen program budgeting. Current efforts for program budgeting in the MoH should be encouraged, so as to indicate government priorities in the health sector more clearly. By integrating information on public and external aid resources by program, program budgeting should also in time provide essential information for coordinating and prioritizing external assistance in the health sector, and for ensuring that there is at least a minimum coverage of key diseases such as those identified in Section A. Particular attention should be paid to the following points. (i) Complete the first round of the new program budgeting to indicate surpluses and deficits of resources by programs and by provinces, and indicate where there are inequities in resource allocation among provinces. (ii) Undertake monitoring and analysis of expenditures, particularly their impact on services to rural areas, which is necessary if the intentions of program budgeting are to take effect. (iii) Strengthen capacity at the provincial level, and later at the lower levels, as one key to improve the management of resources, including those of donor activities. To achieve this, provinces will require financial and management training. (iv) Develop a better understanding of the recurrent costs associated with construction and equipment. The MoH should undertake some sample studies which would also allow better budgeting for future maintenance costs and would indicate whether the pace of construction is sustainable. (v) Clarify the PIP format to show the functional allocation of donor resources within program category. In the longer term, rationalize delivery systems. Achieving better outcomes will depend on a more effective rationalization of the current administrative structure for preventive medicine at the central level. Resources should be less diluted among the nine fragmented, specialized institutes at the central level. Reducing the heavy demands on the government health budget at the central level would make possible increased support to the provincial level for preventive services. Donors will have to assist the MoH, as they have largely funded the vertical health programs. This process can begin by encouraging programs and pilot activities to improve the integration of preventive health measures at the local level. In the longer term, rebuild the referral systenm The underutilization of the public health system indicates serious inefficiencies and inequities. Bypassing lower level services is both costly and inequitable, since it costs the poor proportionately more to access higher level care. The challenge in the longer term is to rebuild a functioning referral system, beginning with quality - 73 - service delivery at the lowest level of the health system. Restoring confidence in the public health sector will require that the components of better quality care, particularly pharmaceuticals, are present, and hence should guide the planning of both domestic and external resource use. The role of the MoH should be to ensure that no provinces are significantly left behind as these improvements take place. TABLE 4.6: GOVERNMENT EXPENDITURE ON HEALTH, 1986-95 (MILLION KIP) 1986 1987 1988 1989 1990 1991 1992/3(1) 1993/4 1994/5 1995/6(2) Ministry of Health 848 927 1,587 Provincial Level 409 576 645 Total Govt. Health Expenditures 1,257 1,503 2,232 1,960 2,080 2,020 6,110 7,732 11,700 13,821 Total (in COnstant 1987 kip) 1,440 1,503 1,800 980 770 660 1,810 As % of Total Govt. Expenditures 4.6 5.1 3.0 1.8 1.5 1.3 3.0 2.8 3.3 3.8 As % of GDP 0.7 0.8 1.0 0.5 0.4 0.3 0.7 0.8 0.8 Domestic Health Expenditures per Capita 341 396 571 488 499 473 1,397 1,726 2,548 2,993 (in dollars) 0.7 0.7 1.9 2.4 3.5 3.5 (1) In 1992 the Lao fiscal year changed from calendar year to October-September. > (2) Approved budget, not actual. Sources: World Bank 1990 (PHN Review); IDS 1995; MOPH data; World Bank staff data. - 75 - TABLE 4.7: MoH EXPENDITURE, 1995/96 TO 1990/2000, BY PROGRAM Program 1995-96 1996-97 1997-98 1998-99 1999-2000 Health Protection and Promotion 3,003 4,204 5,886 8,240 11,536 Treatment and Rehabilitation 7,344 9,547 12,411 16,134 20,974 Consumer Protection 536 696 905 1,177 1,530 Human Resource Development 1,809 2,442 3,296 4,450 6,008 Medical Science Research 42 53 66 83 103 Administration and Management 8,621 10,345 12,414 14,897 17,877 Total 21,354 27,287 34,978 44,980 58,027 of which: construction 3,927 5,047 6,507 8,413 10,907 Health Protection and Promotion 14.1 15.4 16.8 18.3 19.9 Treatment and Rehabilitation 34.4 35.0 35.5 35.9 36.1 Consumer Protection 2.5 2.6 2.6 2.6 2.6 Human Resource Development 8.5 8.9 9.4 9.9 10.4 Medical Science Research 0.2 0.2 0.2 0.2 0.2 Administration and Management 40.4 37.9 35.5 33.1 30.8 Total 100.0 100.0 100.0 100.0 100.0 of which: construction 18.4 18.5 18.6 18.7 18.8 TABLE 4.8: DATA ON HEALTH SPENDING AND INDICATORS BY REGIONS (MOST RECENT YEAR AVAILABLE) Year Total North Center South Estimated per capita spending, kip a/ 1994/95 1,683 1,531 1,417 Poverty, as % of total population b/ 1992/93 rural 53 53 47 66 urban 24 16 26 29 Population per health staff a/ 1995 340 551 845 Population per doctor a/ 1995 5,900 4,713 9,061 Type of delivery assistance (% of total): b/ 1993 none 38 39 48 4 healer/midwife 23 8 22 65 Infantmortality rate (IMR), per'000 live births c/ 1994 117 113 97* 107 Under five mortality rate (U5MR), per '000 live births c/ 1994 160 164 125*. 146 Estimated malaria cases per '000 population d/ 1994 308 279 308 354 Notes: a/ Source: World Bank, 1995. Figures are for "higher poverty line," calculated as income sufficient to buy 2,100 calories of food per person per day, and an allowance for non-food expenditures (total of 11,472 kip per person per month). b/ Source: MOPH data. c/ Source: National Statistical Center/UNFPA, 1995. Note that the central region figures do not include Savannakhet. Savannakhet IMR is 169, U5MR is 227. d/ MPE. National Malaria Control Programme Plan, 1994. - 76 - TABLE 4.9: PLANNED AND ACTUAL HEALTH SPENDING AT THE PROVINCIAL LEVEL, 1994/95 (MILLION KIP) Budgeted Actual as % Budgeted Actual Total . , Total Actual Total Total Total Admin. Admin. Total 11,700.0 11,700.0 100.0 8,370.0 8,369.01- Central 3,579.2 4,620.0 129.1 2,422.9 2,373.0 Provincial 8,120.8 7,079.0 87.2 5,947.1 5,995.0 I Phongsaly 248.8 194.0 78.0 152.8 139.0 2 Oudomxay 252.0 219.0 .86.9 153.5 164.0 f 3 Luangnamtha 209.3 156.0 74.5 158.8 146.0 4 Xayabury 372.1 325.( 87.4 297.5 296.0 5 Bokeo 326.8 288.0 88.1 192.8 191.0 6 Xienghon-Hongsa 126.0 99.0 78.5 52.0 46.0 7 Luangprabang 649.9 590.0 90.8 507.9 545.0 8 Huaphanh 410.8 346.0) 84.2 243.1 230.0 S 9 Xiengkhuang 356.8 329.0 92.2 249.6 265.0 10 Xaysomboun 124.1 0.0 102.1 11 Vientiane 523.0 584.0 111.7 383.0 495.0 12 Vte. Pref. 578.2 573.0 99.1 455.2 543.0 13 Borikhamxay 418.9 366.0 874 338.7 328.0 .: 14 Khammuane 616.1 535.0 86.8 413.6 414.0 15 Savannakhet 1,080.2 946.0 87.6 902.0 888.0 -1,r 16 Champasack 879.7 763.0 86.7 736.3 742.0 17 Sekong 176.3 135.0 76.6 108.3 95.0 18 Saravane 456.8 400.0 87.6 339.8 320.0 .. 19 Attapeu 276.0 231.0 83.7 160.0 148.0 - 77 - TABLE 4.10: ACTUAL GOVERNMENT PROVINCIAL HEALTH EXPENDITURE (CURRENT KIP PER CAPITA) Est. 1988 1992/93 1994/95 1994/95 PIP 1994/95 as3 Total 1,401 2,617 350 187 Central 165 825 1,583 192 Provincial I Phongsaly 69 1,147 1,301 360 113 2 Oudomxay 42 647 1,064 261 165 3 Luangnamntha 91 1,157 1,396 87 121 4 Xayabury 122 961 1,142 96 119 5 Bokeo 105 106 2,600 855 2,453 7 Luangprabang 114 893 1,655 123 185 8 Huaphanh 115 772 1,439 470 186 9 Xiengkhuang 122 575 1,685 321 293 10 Xaysomboun 814 * 11 Vientiane 223 817 2,092 311 256 12 Vte.Pref. 320 609 1,112 57 183 13 Borikbamxay 154 829 2,289 233 276 14 Khammuane 295 1,073 2,002 444 187 15 Savannakhet 123 634 1,443 87 228 16 Champasack 168 846 1,560 42 184 17 Sekong 181 1,841 2,167 627 118 18 Saravane 226 766 1,597 312 209 19 Attapeu 140 1,321 2,715 952 206 Standard deviation 73 352 492 263 Notes: Data for 1992/93 from LHSR draft; * "814" kip figure is for "special zones," here allocatedjust to Xaysomboun. The 'total" figure is from Table 16, LHSR. Figures for 1988 from PHN Sector Review, May 16, 1990. - 79 - CHAPTER 5: EDUCATION A. SECTOR OVERVIEW 1. Lao PDR is characterized by an economic duality which places unique demands on the education and training system. On the one hand, it is one of the poorest countries in the world, with a per capita income of US$320, illiteracy rates of about 36%, a national incidence of poverty of 46%, and an annual population growth rate of 2.4%. At the same time, Lao PDR is a rapidly growing and diversifying economy. Its favorable geographic location between the buoyant economies of China, Thailand, and Vietnam, its economic reform programs which have made possible a real GDP growth averaging 6.3% since 1990, and its natural resource potential in hydropower and timber give the country an economic potential far beyond its current socioeconomic indicators. 2. Poverty alleviation in Lao PDR requires broadening access to high quality basic education. Educational attainment is currently too low to enable the poor to take advantage of economic opportunities: the net enrollment rate in primary school is 73% and the cohort survival rate is about 30%; gross enrollment is 12% in upper secondary school, 2% in technical and vocational eduction (TVET), and 1% in higher education. At each of these levels, achievement needs to increase substantially, and regional and gender disparities must be reduced. 3. Conversion to a high growth, diversified, market economy requires changes in the skill mix being produced by the education system. Already in high demand are skills in accountancy, administration, law, engineering, computer technology, and foreign languages. Growth in non- traditional agriculture, industry, finance, and commerce sectors will be slowed if skill shortages persist. 4. The basic issue facing policymakers is how to orient the education and training policy framework to: (i) produce a flexible work force with strong general skills; and (ii) improve the ability of post-secondary education institutions to flexibly respond to emerging labor market trends and new skill demands. 5. The resource requirements for successfully pursuing this two-pronged education strategy will overwhelm the public sector unless a coordinated and sustainable investment program is developed. While external support has been readily available, it is fragmented and not integrated in a national framework. Nearly 75% of 1995/96 capital expenditures are provided by external sources, without clear criteria for prioritizing investments or assessing their overall recurrent cost implications. Households are contributing substantial resources to education, but further scope for household and community contribution is limited by the uniformly low levels of household incomes. The contribution of the private sector is sporadic and project specific. 6. Under these conditions education managers must (i) rationalize current expenditure patterns; (ii) prioritize new investments; and (iii) protect key sub-sectors and disadvantaged population areas or groups. Particular attention should be given to developing a coherent national education investment policy which will lay the groundwork for improved donor coordination. Savings in the current budget could be obtained from: (i) lowering the unit costs of post-primary education; (ii) investing in key quality inputs which would eventually improve efficiency; (iii) more - 80- efficiently and effectively distributing resources across schools; (iv) improving teacher utilization; and (v) developing more equitable financing arrangements for higher education. 7. This strategy demands strong managerial and institutional capacity. The Ministry of Education must build skills to develop and implement national education policies and to remove bottlenecks in the system. Linkages and responsibilities among key actors in the sector are unclear and still evolving; the administrative staff needs training; systematic data are scarce, and this is compounded by poor communication owing to rough terrain and inadequate infrastructure for many of the southern and northern provinces. B. GOVERNMENT OBJECTIVES FOR EDUCATION 8. Recognizing the large returns to education, the Government has placed a high priority on increasing the stock of human capital through education and training and has set ambitious goals for the sector through to the year 2000. Targets include: (i) expanding enrollment in primary education from 63% in 1990 to 80% in 2000; (ii) increasing retention rate from 27% to 63%; (iii) increasing literacy rates for the age group 15-40 from 60% to 80%; (iv) improving teacher quality; (v) providing the legal and regulatory framework for encouraging private education; and (vi) expanding and rationalizing higher education. These ambitious goals will be difficult to attain given the financial and human constraints outlined above. C. FUNDING MECHANISMS 9. The education budget has grown substantially during the previous six years; expenditures rose more than fourfold between 1990 and 1995/96 (see Figure 5.1). Most of the increase occurred in the investment budget, which grew in nominal terms from 0.5 million kip in 1990 to 19.4 million in 1995/96. The recurrent budget, in contrast, grew only two and one-half times over the same period -- from 11.3 million kip to 27.3 million kip.' Most of this growth funded large increases in teacher salaries and fellowships. Primary education teacher salaries, for example, went up from 106,500 kip in 1990 to 455,300 kip in 1995. During the 1990-1994/95 period, education expenditures as a percentage of GDP grew from 1.6 to 2.7; education expenditures as a percent of the government budget grew from 8.3 to 12.4 (see Table 5.2). These figures compare reasonably well with other Asian countries at a similar level of development. In real terms (constant 1990 kip) the recurrent budget rose from 11,310 million kip in 1990 to 15,328 million kip in 1995-96. The investment budget rose from 543 million to 10,394 million in the same period. - 81 - FIGURE 5.1: EDUCATION BUDGET, 1990-96 (MILLION KIP) 50,000 45,000 40,000 35,000 - 30,000 _ - 25,000_ _ 25 000 _E_ lnvestment 15,000 _______Recurrent 10,000 _ _ _ _ _ 5,000 _ _ 0 1990 1991 1992 1992-93 1993-94 1994-95 1995-96 Note: The year 1992 has a nine-month budget to allow for change from calendar year to fiscal year beginning in October. Source: Lao Government authorities. 10. Recurrent Cost Implications. The exact impact of the recent surge in donor financed investment on the future recurrent budget is unknown,2 but it will probably generate significant demands in areas of building maintenance, supplies, student stipends, and teacher salaries. Such costs would normally emerge in the budget process within a few years of completion of the investment. Experience from other countries shows that each dollar invested in the education sector generates as much as US$5-6 in recurrent costs. The comparable figure in Lao PDR is currently about US$3-4, but this figure is likely to increase progressively as qualitative reforms are implemented. 11. Therefore, continued high levels of investment will produce even more pressure on the recurrent budget. Disregarding classification problems between investment and recurrent costs, continued high levels of investment would appear to be unsustainable. Within Asia, capital investments in education in 1995 were 28% of total government expenditure,3 compared with 51% in Lao PDR's overall budget, and 41% in Lao PDR's education budget for 1995/96. This is a cause for concern since even "optimistic" projections for the government budget show relatively modest growth in total revenues and continued growth in the overall deficit (see Chapter 1, Main Report). 12. Sources of Funds. Funding for education comes mainly from government and international donors. Communities also make significant contributions for school construction and operation and for student support. In principle, domestic resources support the entire recurrent budget for education (see Figure 5.2). The average level of government recurrent spending on education over the past seven years was 13.3% of total recurrent spending (Table 5.2). Domestic resources supported only about 25% of the investment budget between 1992/93 and 1995/96. The sharp increase in investment spending since 1992 has largely been financed from international sources. 2 One needs to separate out "true" new investment, which generates incremental recurrent costs from other items, such as repair, reconstruction, repeated textbook production, and teacher training which may or may not generate significant new recurrent costs. 3 International Monetary Fund. Government Finance Statistics Yearbook, Washington, D.C.: 1995, pp. 40-41. - 82 - FIGURE 5.2: EDUCATION'S SHARE OF TOTAL GOVERNMENT BUDGET 18 161 14 12 10 8 6 4 2 0 1990 1991 1992 19243 1993-94 1994-95 1996-9 -4- Recurrent Budget _-Total investment budget |A+ Domestically financed investment Source: Lao Government authorities. 13. Evolution of Education Budget 1991-1995/96. The total education budget either as a share of GDP or of the total government budget is heavily loaded by the very substantial increase in investment expenditure mostly financed by external funding. The share of education expenditure was 7.5% of total government expenditure in 1991 but increased to 12.4% in 1994/95 -- as a share of GDP, it rose from about 1.6% to 2.7% in 1994/95 (Table 5.2). In comparative terms, and based on the 1992 data (the latest available comparative data), it appears that Lao PDR spends slightly less on education than comparable countries and achieves a reasonable level of quantitative development of its education system4 (Tables 5.1 and 5.2). 4 The dramatic increase in the investment budget since 1992 would increase the share of public spending on education as a proportion of GDP to 2.2% , making Lao PDR spending on education more comparable with that of other countries. - 83 - TABLE 5.1: COMPARATIVE DATA ON PUBLIC SPENDING ON EDUCATION AS A SHARE OF GDP, AND AVERAGE DURATION OF SCHOOLING GDP per capita (1992 Education as percent Average Duration of Country USD) of GDP schooling (years) Bangladesh 220 2.3 5.1 Bhutan 200 4.3 1.9 China 450 1.7 7.5 India 310 3.9 7.6 Lao PDR 230 1.9 7.1 Myanmar na 2.1 6.6 Nepal 180 2.0 6.1 Pakistan 420 2.7 4.1 Sri Lanka 540 3.3 11.5 Average 319 2.7 6.4 TABLE 5.2: EDUCATION BUDGET, SHARE IN GDP AND OVERALL BUDGET, 1990-96 (BILLION KIP) 1990 1991 1992 1992/93 1993/94 1994/95 _995/96 Education Budget 11.8 11.3 14.9 17.9 25.5 36.3 45.0 Recurrent 11.3 9.0 10.1 11.9 15.4 23.0 25.6 Capital 0.5 2.3 4.8 6.0 10.2 13.3 19.4 GDP at Current Price 615 722 848 942 1,069 1,323 1,630 Government Expenditure 143 151 175 171 259 294 362 Recurrent 70 82 92 105 127 143 165 Capital 74 69 82 66 132 151 197 Education as Percentage of GDP Education Total/GDP 1.93 1.57 1.76 1.90 2.39 2.74 2.76 Edu. Recurrent/GDP 1.84 1.25 1.19 1.26 1.44 1.74 1.57 Edu. Capital/GDP 0.09 0.32 0.57 0.63 0.95 1.01 1.19 Education as Percentage of Government Budget Edu. TotaVTotalGovt.Expd. 8.3 7.5 8.5 10.5 9.9 12.4 12.4 Edu.Recurrent/Govt.Recurrent 16.2 11.0 11.0 11.4 12.1 16.1 15.6 Edu. Capital/Govt. Capital 0.7 3.3 5.8 9.1 7.7 8.8 9.9 Notes: - The 1992 budget is a nine-month budget to allow for a change in the budget year calendar. - Budget does not include community contributions. Community contributions come from several sources, including families, enterprises, and community-based groups such as Parent Associations or School Support Organizations. Contributions are an important supplement to both the investment and the current budgets of schools and, most important, a critical source of support for individual students. Howewr, there appears little scope for increasing community contribution in the near term. Sources: Government authorities and World Bank staffestimates. - 84 - 14. External Resources. Over the last five-year period, Lao PDR received more than US$5 per capita in foreign assistance to education. The most important education projects are financed through bilateral aid (Australia, France, Germany, Japan, Norway, Switzerland) and the multilaterals (ADB, IDA, UNDP, UNESCO and UNICEF). It should be noted that the figures underestimate the actual resources for at least two reasons. First, they do not take into account small projects and fellowships for study abroad awarded by some countries under bilateral arrangements. Second, only official grants and loans, for which there is a formal signed agreement, are recorded in ODA statistics as external commitments. The magnitude of external spending for education is shown in Table 5.3. TABLE 5.3: THE MAGNITUDE OF EXTERNAL FINANCING FOR EDUCATION (US$MILLION AND BILLION KIP) 1990 1991 1992 1993 1994 1995 Total Grants (US$) 2.67 3.07 5.49 8.83 11.26 11.26 42.58 Loans (US$) 0.58 0.58 1.24 3.64 3.64 3.64 13.32 Total US$ 3.25 3.65 6.73 12.47 14.90 14.90 55.90 kip 2.34 2.63 4.85 8.98 10.88 13.86 43.54 Education budget (kip) Total 11.80 12.70 15.90 18.90 24.40 49.60 133.30 Investment 0.50 1.00 0.90 4.90 8.00 22.80 38.10 Grants and loans as a % of total 19.80 20.70 30.50 47.50 44.60 27.90 32.70 education budget Source: Mingat, p. iU. 15. Community Contributions. Community contributions to education (see Table 5.4) come from several sources, including families, enterprises, and community-based groups such as Parent Associations or School Support Organizations. Contributions appear to be an important supplement to both the investment and the recurrent budgets of schools and, most important, a source of support for individual students. However, there appears little scope for increasing levels of community contributions in the near term. 16. Government appears to provide most recurrent expenditures for schools (see Table 5.4), but in most schools this only includes teacher salaries and some pedagogical materials. Communities are often the sole source of revenue for books and materials, school maintenance, and other critical non- salary expenditures. In fact, non-government resources make up as much as 50% of the investment budget of a school. However, there is no consistent pattern across schools; some schools depend almost entirely on govermment resources, others depend almost entirely on community resources and donors. This suggests that a wide range of possible partnerships exists for constructing and maintaining a school. - 85 - TABLE 5.4: SOURCES OF SCHOOL REVENUE Government Donors School Enterprise Community Source enterprise Investment 50% 19% 0.3% 23% 8% MoE community survey Recurrent 96% 0% 0.6% 0% 4% MoE community survey Total budget 87% 5% 2% 1% 5% Microplanning study 17. Most community contributions to education are expenditures on individual students, rather than contributions to the school itself A recent survey of 383 parents shows that the cost of keeping a child in school is surprisingly high (see Figure 5.3). On average, parents spent 46,000-53,000 kip to keep one child in school for one year in 1995/96.5 Average costs increase by grade and level; a year of primary school costs approximately 33,500 kip; lower secondary, 63,500 kip; and upper secondary, 76,750 kip. The largest expenses are for uniforns, pocket money, books, and supplies. 18. In Lao PDR one often hears that parents do not enroll their children in school because they cannot afford it;6 the data would seem to support this conclusion. Indeed, the high cost of education and parents' limited ability to pay is a major source of inequity even among those children enrolled in school. Both the MoE survey and the LECS survey show that expenditures vary widely among households. Perhaps most disturbing for education policymakers is that when parents cannot afford the full price of schooling, they often cut expenditures on key inputs which enhance the quality of learning, such as textbooks, materials, and tutoring. FIGURE 5.3: AVERAGE ANNUAL COST OF KEEPING A CHILD IN SCHOOL, BY GRADE 100,000 90,000 80,000 70,000 60,000 50,000 go 4+-Kip 40,000 30,000 20,000 10,000 0 1 2 3 4 5 6 7 8 9 10 11 The low estimate includes expenditures on fees, tutoring, dress, books, supplies, pocket money, and transportation. The higher estimate adds in "other expenses," such as bonus gifts for teachers, miscellaneous contributions, and bicycles. The expenses are as reported by parents. Teachers were also asked what it costs to keep a child in their school. Their estimates ranged from a low average of 53,356 kip to a high (including other expenses) of 73,344 kip. 6 In addition to these cash outlays, parents must bear opportunity costs of enrolling their children. - 86 - 19. The structure of the education system itself might contribute to differences in the level of expenditure. First, the LECS survey shows that approximately 40% of expenditures occur at the beginning of the school year, between August and October, when farmers incomes are most stressed. Second, in rural areas the system for distributing and marketing educational materials is weak. Textbooks and learning materials are simply not within easy reach of many rural households. 20. Estimated Resource Envelope for Education 1995/96-1999/2000. Projections for growth in the government budget are presented in Table 5.5. They predict that education expenditures will double from 45 billion kip in 1995/96 to 89.4 billion kip in 1999/2000. However, in real terms (constant 1995-96 kip), the budget increase would be considerably less, from 45 billion kip to 72.9 billion kip. TABLE 5.5: BUDGET PROJECTIONS, 1995/96 - 1999/00 (BILLION KIP) 1995/96 1996/97 1997/98 1998/99 1999/00 Government budget (Current Price) Total 361.6 464.0 532.4 601.0 661.2 Recurrent 164.8 205.0 228.4 258.0 280.0 Investmnent 196.8 259.0 304.0 343.0 381.2 Education budget (Current Price) Total 45.0 57.6 67.7 78.3 89.4 Recurrent 25.6 34.0 39.0 44.0 50.0 Investnent 19.4 23.6 28.7 34.3 39.4 Education budget (1995/96 price) Deflators (1995/96=100) 100.0 107.5 112.3 117.4 122.7 Total 45.0 53.6 60.2 66.7 72.9 Recurrent 25.6 31.6 34.7 37.5 40.8 Investnent 19.4 22.0 25.5 29.3 32.1 Investment/recurrent 76% 69% 74% 78% 79% Sources: Ministry of Finance and World Bank staff estimates. 21. Recurrent Costs. What is the implication for the operational budget of such rapid growth in the investment program? Clearly, additional funding is needed for the operation of the programs or projects that are included in the capital investment budget. Unfortunately, it is here that two specific problems arise: (i) the problem of misclassification of expenditures; and (ii) the absence of any planning for the recurrent cost implications of capital investments. 22. There are genuine difficulties in apportioning some expenditures to investment or recurrent cost categories. For example, learning materials such as textbooks are often included under equipment and thereby classified as an investment item. However, the life of some learning materials is only slightly longer than one or two academic years with the concomitant need for a - 87 - regular annual textbook program. In such a case, it may be more legitimate to classify short-life learning materials as recurrent expenditure. Also, expenses related to in-service training of national staff are sometimes classified as investment expenditures because they are regarded as one-time investments in human capital rather than as training which is needed on a regular basis. Finally, although all externally funded projects are classified by the Lao Government as investments, it is common practice for some donors to consciously include in their projects non-negligible operational funds to ensure smooth project implementation. 23. The second and more serious problem is that the recurrent cost implications of foreign assistance to education are rarely quantified and allowed for in the operations budget. There are direct budgetary implications of external assistance programs, which include a domestic financing component plus the additional cost of operation and maintenance. However, while capital is often completely financed through external funds, many external donors have regulations against funding salaries or local operating costs of projects. Most multilateral organizations, including the regional development banks, target their funding on the foreign exchange component of a specific development project. Some external donors extend their support beyond the foreign exchange component of a project, and -- in exceptional circumstances -- bear a portion of the total project costs (thereby implicitly covering some of the local costs of the project). However, almost all funding agencies draw the line at financing directly the salaries of regular education system employees. Donors assume that the local salary costs of development projects are the responsibility of the recipient country. If salaries are funded by the external donor, it is usually as part of the project "start-up" costs and the financing is on a declining basis. D. EXPENDITURE TRENDS 24. As noted above, generating savings within the education sector will be the main tool that education managers have at hand to mobilize resources to sustain investments. This entails careful examination of education expenditures to (i) ensure that resources are allocated to priority areas; (ii) identify areas of inefficiency; and (iii) explore the potential for cost saving and cost sharing.7 25. Resource Distribution among Sub-sectors. At first glance, education expenditures in Lao PDR appear to be distributed in a balanced manner among sub-sectors (see Table 5.6) The Government's policy of providing broad access to basic education is reflected in the budget; of all recurrent expenditures, 46% is allocated to primary education and 16.9% to lower secondary education. Higher education consumes only 8.4% of recurrent expenditures, but this has grown from 5.8% in 1990 and is likely to grow further. 7 An important limitation of this section is that it deals only with the recurrent budget. Detailed information on allocation and utilization issues in the investment budget were not available at the time of preparing this report. This should be an area for intensive Government and donor review. - 88 - TABLE 5.6: DISTRIBUTION OF RECURRENT EXPENDITURES, BY LEVEL OF EDUCATION (PERCENT) 1990 1991 1992 1992/93 1993/94 1994/95 Pre-school 4.1 3.5 3.6 3.3 3.4 3.3 Primary 43.7 43.1 43.8 42.9 43.7 46.0 Lower secondary 18.5 18.5 18.7 18.2 18.0 16.9 Upper secondary 9.1 9.3 8.8 8.2 8.3 7.1 TVET 4.4 5.0 7.9 7.5 7.6 6.5 Teacher training 7.1 7.1 5.4 6.1 5.2 4.8 Higher education 5.8 5.9 5.0 6.4 6.3 8.4 Administration 7.3 7.5 6.8 7.5 7.4 7.2 Source: Lao PDR authorities. 26. Analysis of costs per student -- or unit costs -- however, paints a less balanced picture of education expenditure.8 While unit costs typically increase with the level of schooling, the costs of technical and vocational education (TVET), teacher training, and higher education are very high -- on a per student basis 16 to 18 times more expensive than primary education (see Table 5.7). TABLE 5.7: AVERAGE UNIT RECURRENT COST ESTIMATES, 1994-95 Level kip Per capita GDP Relative to primary Pre-school 31,335 0.120 1.76 Primary 17,841 0.068 1.00 Lower secondary 39,625 0.152 2.22 Upper secondary 43,319 0.166 2.43 TVET 286,872 1.101 16.08 Teacher training 311,546 1.196 17.46 Higher education 332,273 1.275 18.62 Source: Lao PDR authorities. 27. The unit cost of higher education in Lao PDR is also high by international standards. In comparison with other Asian countries, Lao PDR spends about 30% more on higher education. Only China and Pakistan -- two higher education systems in desperate need of reform -- spend more on higher education as compared to GDP per capita (see Table 5.8). 8 The unit cost (UC) is a function of five types of expenditures: (a) teachers; (b) non-teaching staff, (c) scholarships, (d) operation and maintenance, and (e) general administration (textbooks and learning materials are subsumed under the budgets for "scholarships" and "operations and maintenance" in Lao PDR's education accounting system). Breaking down the determinants of teacher and non-teacher expenses, unit costs can be expressed as a function of average teacher salary (TS), pupil-teacher ratio (PTR), average salary of non-teaching staff (NTS), pupil-non teaching staff ratio (PNTR), per student operation expenditure (OE), per-student scholarship received (SE), and per student administrative expenditure (AE): UC = (TS/PTR) + (NTS/PNTR) + OE +SE +AE. Teacher cost (TS/PTR) may alternatively be expressed as a function of average teacher salary (TS), pupil-teacher ratio (PTR), number of hours of learning of students (HLS), and hours of teaching duty (HTD): UC = (TS/CS * HLS/HTD) + (NTS/PNTR) +OE +SE +AE. - 89 - TABLE 5.8: UNIT COSTS OF EDUCATION BY LEVEL (AS A PERCENTAGE OF GDP PER CAPITA) Country Per Capita GDP Primary Secondary Higher Bangladesh 220 0.06 0.21 0.37 China 450 0.05 0.15 1.93 India 310 0.11 0.15 0.83 Lao PDR 250 0.07 0.16 1.28 Myanmar na 0.06 0.11 0.40 Nepal 180 0.10 0.14 na Pakistan 420 0.13 0.29 1.57 Sri Lanka 540 0.14 0.17 0.53 Average 336 0.09 0.17 0.99 28. These data do not imply that education spending in Lao PDR should conform to an international standard. The unit cost of education within a particular country is the outcome of a set of educational inputs and the economic and social context in which they are mobilized and utilized. Therefore, the message is that education managers must analyze what is driving unit costs, particularly at the higher, more expensive, levels of education, to decide on resource allocation. 29. There are substantial differences across sub-sectors in the relative importance of the five components of unit costs. Particularly striking is the high proportion of resources devoted to scholarships in post-secondary education (see Table 5.9). This ranges from 36% of the budget for teacher training to nearly half of the budget for TVET and higher education. These costly items should be reconsidered, especially in light of the planned establishment of a National University. Also relatively high are the costs of salaries in post-secondary education, largely because of very low pupil-teacher and pupil-non-teaching-staff ratios. 30. Increasing pupils per teacher to 15 in post-secondary education, and the pupil-per-non- teacher ratio to 45, would alone generate savings of approximately 1,097 million kip, or 3.9% of the total education bill. Additionally, cutting the scholarship bill in half (by some combination of cost recovery, reducing the overall amount of the scholarship, or targeting scholarships only to the neediest students) would generate an estimated savings of 885 million kip per year, or 3.1% of the total recurrent education budget. This money could, in turn, be reallocated to programs to meet the Government's stated priorities in expanding and improving basic education. 31. Resource Distribution across Expenditure Categories. Because of high dropout and repetition in Lao PDR, it currently takes an average of 9.5 years to produce a primary school graduate, 3.9 years to produce a lower secondary school graduate, and 4.2 years to produce a senior secondary school graduate. The scope for savings by improving educational flows is enormous. Repeaters are currently taking up 21% of the primary school places.9 9 An emerging body of research argues that in many developing countries there is little basis for the commonly held assumption that a trade-off exists between investments in quality and investment in quantity of education. Research in Brazil and Egypt concludes that investments in inputs which raise the quality of learning actually induce resource savings substantially larger than their original cost. The savings, these studies argue, come in the form of improved achievement, lower repetition, and higher completion rates. The savings effect can be exceptionally strong in low income and resource poor areas where one often finds unstable schooling environments, poorly trained teachers, and shortages of teaching and learning materials. In northeast - 90 - TABLE 5.9: DISTRIBUTION OF UNIT COSTS BY TYPE OF EXPENDITURE AND LEVEL OF EDUCATION Expenditure Pre-school Primary Lower Upper TVET Teacher Higher secondary secondary training education Administration 2,726 1,699 3,629 4,058 7,618 5,486 5,495 (8.0%) (8.70%) (8.4%/O) (8. 6%/) (2.6%) (1.7%) (1.6%) Operations/ 2,071 1,275 2,719 3,094 22,366 40,076 21,544 maintenance (6.1%) (6.5%) (6.3%) (6.5%) (7.6%) (12.6%) (6.4%) Scholarships 142,463 114,265 165,558 (48.4%) (36%) (49%) Non-teaching staff 5,142 1,610 4,967 4,103 39,708 46,517 31,810 (15.1%/O) (8.2%) (11.5%) (8.7%) (13.5%) (14.7%) (9.4%) PNTR 82.8 261.5 89.7 103.6 12.9 11.6 17.1 NTS 425,700 421,000 445,400 425,600 512,500 540,200 544,400 Teachers 24,122 14,955 31,938 36,122 82,334 110,687 113,360 (70.8%) (76.5%) (73.8%) (76.2%) (28.0%) (34.9%) (33.6%) PTR 17.6 29.5 15.3 16.0 6.5 5.9 6.4 TS 424,400 440,600 488,500 576,600 538,100 651,600 719,900 CS 28.1 27.5 35.3 39.7 21.7 HLS 30 30 HTD 20 18 % utilization 62 107 65 67 of teachers TOTAL 34,061 19,540 43,254 47,377 294,490 317.032 337,768 (I 00%) (I 00%) (I 00%rl) (I 000%) (100%) (100I%) (I 000%) Source: Lao PDR authorities. 32. To generate resources over the longer term, Lao PDR needs a sustained program of quality improvement. This should begin by identifying those inputs which would increase learning and promotion. In most developing countries, investments in curriculum, textbooks, teaching, and children's learning capacity (e.g., through improved health and nutrition) have improved learning achievement.10 33. Lao PDR currently allocates little to such investments. Eighty-four percent of the recurrent budget goes to salaries and administration (see Figure 5.4). Of the remaining 16%, 7% is for operations and maintenance (O&M) and 9% is for fellowships. Allocations for pedagogical materials are buried in the budget for O&M, and textbooks (along with other printed materials) are rather unconventionally under the heading "fellowships and subventions." Complete data are not available, but the best estimate for actual spending on textbooks and materials in 1994/95 is 429 million kip, or only 1.6% of the total education recurrent budget. Brazil, for example, it is estimated that each dollar invested in textbook and writing materials for primary schools saves between US$4.03 and US$6.95. One dollar spent on in-service teacher training appears to generate savings of US$1.88 to US$3.00. Investments in basic school furniture and facilities also appear to induce large savings. In Egypt, research concluded that the quality of education has a significant, independent impact on repetition and dropout. Although costing data were not available on quality inputs in Egypt, the magnitude of the impact of improved quality on repetition and dropout was so large that researchers concluded that "every effort should be made to improve quality." See: Harbison, Ralph W., and Eric A. Hanushek, Educational Performance of the Poor: Lessons from Rural Northeast Brazil. New York: Oxford University Press, 1992; Hanushek, Eric, and Victor Lavy, School Quality, Achievement Bias, and Dropout Behavior in Egypt. Washington: World Bank, LSMS Working Paper No. 107, 1996. to Lockheed and Verspoor. - 91 - 34. In effect, parents are responsible for providing critical inputs of books, writing materials, facilities maintenance, and even supplements to teachers' salaries. Parents contribute through their local school or Parent Organization, or by purchasing supplies for their children. This results in highly uneven provision since not all parents have equal ability to pay, equal access to markets for education supplies, or even the same set of spending priorities (for example, some parents may buy uniforms rather than books and writing supplies, or may prefer to educate sons rather than daughters). 35. The Government appears to allocate resources for teacher employment and training in an inefficient manner. First, the MoE employs many under-qualified teachers. Forty-three percent of primary school teachers have only five years of education and three of teacher training, while 30% have no training. The relatively flat salary structure and the cumbersome promotion procedures within the civil service provide few incentives for teachers to improve their own skills and talents. Finally, what training the MoE does offer is largely in the form of expensive pre-service training. There is no regular program of in-service training, although the Government is currently working with UNICEF and NGOs to expand existing pilot programs. 36. The MoE should invest in policy development, incentives, curriculum and materials, and managerial support structures to attract higher quality teachers. This should eventually translate into savings by improving overall system efficiency, but in the near term it will require larger budget allocations for key quality inputs. The money could come from (i) immediate reallocations within the current budget, or (ii) donor "bridge financing" until savings begin to accrue in salary and administration bills which can be transferred to critical non-salary expenditure items. FIGURE 5.4: DISTRIBUTION OF EDUCATION RECURRENT BUDGET, BY TYPE OF EXPENDITURE, 1994/95 Administration Fellowships 7% 9% Operation & Maintenance 7% Non-teacher Salares 10% - ~~~~~Teacher Salaries 67% - 92 - E. EFFICIENCY OF PUBLIC SPENDING ON EDUCATION 37. Resource Distribution across Schools. In Lao PDR there are many very small schools. This is probably unavoidable owing to the geographic and demographic conditions. Of approximately 7,200 primary schools, 3,043 (42%) have fewer than 40 students and 3,290 (46%) have only one teacher (Table 5.10). However, some of these schools should probably be amalgamated, where this would not hamper easy access for students. This would help managers to distribute managerial, personnel, and material resources rationally and economically. TABLE 5.10: CHARACTERISTICS OF PRIMARY SCHOOLS Number of pupils <20 21-40 41-60 61-100 101-200 >200 % of schools 13.1 29.1 15.1 14.7 16.5 11.5 Number of teachers 1 2 3-4 5-6 7-10 >10 % of schools 45.6 16.1 14.5 14.0 7.1 2.8 38. There is wide variation in pupil-teacher ratios, even among schools of the same size. This is true at the primary, lower secondary, and upper secondary levels. Variation tends to be greater in small schools. For example, primary schools of 80 students have pupil-teacher ratios ranging from 12:1 to 84:1. Ratios in lower secondary schools of 80 students range from 6: 1 to 30: 1. The variation could be caused by real need, by problematic deployment policies, or by a legacy of uneven funding across provinces. 39. Ideally, teachers should be redeployed from "overstaffed" to "understaffed" schools. Once opportunities for re-deployment are exhausted, teachers from overstaffed schools could be released. The MoE has staffing norms, but whether a school is overstaffed or understaffed needs to be settled on a case-by-case basis. 40. Teacher Utilization. In Lao PDR lower secondary teachers are estimated to teach only 13 hours per week on average, instead of the statutory 20 hours. Upper secondary teachers teach on average only 12.1 hours per week instead of 18 hours. These are very low. For example, relating the teacher utilization rate to the teacher salary bill for secondary education, one finds that, in effect, the MoE is paying 1,835 million kip (6.7% of the total MoE recurrent budget) for teaching services which it does not receive. 41. Substantial progress was made between 1993 and 1995 in improving teacher utilization in secondary schools. However, much room exists for further improvement. First, the MoE could require underutilized teachers to teach multiple related subjects -- for example, math and science -- or Lao language and history. The MoE could design a program of in-service training and self-paced study to encourage teachers to acquire the additional subject matter knowledge needed to broaden their skills. Second, the MoE could consider offering part-time employment to teachers that they do not expect to fully utilize. This would be all the more attractive to teachers in areas where other part- time employment, business, or educational opportunities exist. Third, the Government could consider simplifying the curriculum to combine or eliminate subjects. 42. Incidence of Public Expenditure on Education. Most poor households in Lao PDR have the following characteristics: (i) they are in rural areas; (ii) they earn most of their income from farming; (iii) they are headed by persons with low levels of education; and/or (iv) they are headed by - 93 - young people. The incidence of poverty is highest in the rural south, followed by the rural north and rural center. Isolated ethnic minority groups in those areas are at particular risk. 43. Public spending in Lao PDR is biased towards the better-off The most recent data available on education spending by income group are presented in the Lao PDR Social Development Assessment and Strategy (World Bank, 1995). This report highlights the fact that public education spending per capita is nearly three times higher for those in the highest income quintile than for those in the lowest income quintile (Figure 5.5). FP Ut RE 5.5: PER CAPITA EDUCATION EXPENDITURE, BY INCOME QUINTILE AND RESIDENCY 5,074 4,732 3,092 4 ,!5fi.:i ' - ___ __2,1 _ 2,032 * flfl 2,151 , ~ ~ ~ ~ ~ ~ ~~~~~ 4J _ e-C ;,C TU >rh-l Bank "Lao PDR Social Development Assessment and Strategy," 1995, p. 25. 44. This regressive pattern of expenditure is brought about though a combination of supply and demand tfctors. In short, higher levels of education cost more than lower levels and upper income gruiip,j have higher rates of enrollment in higher education. This can be seen in the cumulative share c4 per capita expenditures for quintiles at different levels of education. For example, the share of the b(tt: _ln quintile in lower secondary education is 7.5% compared with 34% for the top quintile. F( c irpper secondary education the shares are 3.5% and 48%, respectively. The most regressive patteii- is nigher education, where the bottom quintile does not receive any subsidy, since no enrollmnenti; are reported, and the top quintile receives 69% of the total.'1 , l ;id tiar,k, "Lao PDR Social Development Assessment and Strategy," p. 26. T - 94 - 45. What could the Government do to address these inequities? First, it must assess its higher education admissions policies to ensure that they do not inadvertently discriminate against rural and poor students. Over the longer term, the Government must improve the quality of education at the primary and secondary levels in poor areas to ensure that all students are equally well prepared to compete for positions in upper secondary school and higher education. Second, the Government may wish to raise additional income by applying a fee schedule on upper secondary and higher education, so that better off students receive less in subsidies and the poorest can be provided with assistance. Fees could also be used to finance lower secondary school construction and operation in areas where access is lowest. F. SCOPE FOR PRIVATE SECTOR PARTICIPATION 46. There is a small, but rapidly growing, private education system in Lao PDR. Presently, private schools are limited to the basic education level. Less than 3% of all students at the pre- school through lower secondary level are currently enrolled in private schools (Table 5.11). Private schools are concentrated in a few major urban areas, where the demand for education is relatively high, and where there is a critical mass of higher income families. TABLE 5.11: ENROLLMENT IN PRIVATE EDUCATION, 1989-95 1989/90 1990/91 1991/92 1992/93 1993/94 1994/95 % of total students 1994/95 Pre-school (700) (1,000) 1,655 2,324 3,242 (4,500) 13.4% Primary (2,000) (4,000) 6,886 10,415 14,409 (18,000) 2.5% Lowersecondary (0) (0) 165 713 996 (1,300) 1.1% Note: Figures in parentheses are estimates. Source: Lao PDR authorities. 47. The scope for expanding private education outside of existing urban areas is probably quite limited, owing to high rates of poverty and the limited profit margin. Generally, the private sector flourishes where there is substantial excess demand for education, a condition which does not yet exist in Lao PDR, except in selected areas of specialized training. 12 Nevertheless, the Government is encouraging the growth of private sector involvement in education by developing an enabling regulatory and incentive framework.'3 12 There are several training institutes in urban areas which offering specialized programs in subjects such as commerce, accountancy, computer technology, and English. In total, these institutes enrolled approximately 3,000 students in 1994. These schools could grow rapidly in the near future, as anecdotal information suggests a strong demand for skills which appear to relate well to the needs of private sector employers. 13 The ADB is providing long-term technical assistance to advise the Government on private education. - 95 - G. PROPOSED MEDIUM-TERM EXPENDITURE PROGRAM 48. The proposed balance of expenditures between programs in the education sector appears broadly reasonable (Table 5.12). However, there is the danger that rapid growth of post-secondary education by investing in the National University-- where expenses are very high in relation to basic education -- will place undue stress on future education budgets. In fact, a recent World Bank study on education finance and management concludes that the gap between budgeted and required current expenditures is likely to exceed 16,000 million kip by the year 2000. To close this gap, the Government has a choice of (i) shifting resources from the capital expenditure program to current expenditure requirements; (ii) increasing allocations to current education expenditures (and therefore cutting expenditures in other sectors to stay within the available resource envelope); (iii) compromising on its goals for expanding education; or (iv) preferably increasing the efficiency of available resource use by looking for ways to reduce the unit costs of post-secondary education in order to finance its growth. In an era of constrained resources, the Government must ensure that new investments are made according to a well-reasoned set of priorities which balance concerns for quality, growth, and equity. 49. There appears to be a considerable need to increase expenditure on inputs to improve the learning environment, including textbooks, teaching materials, in-service teacher training, principal training, and school maintenance. These categories are not normally financed by donors and will require re-examination of the current budget. The education system could certainly benefit from increased allocation for quality improvement programs, provided the capacity to manage such resources could also be developed. TABLE 5.12: PUBLIC INVESTMENT PLAN FOR EDUCATION, 1995/96 - 1999/2000 (BILLION KIP) 1995/96 1996/97 1997/98 1998199 1999/00 T '99/2000 Ongoing & committed 18.1 13.4 12.2 7.3 6.0 56.9 Prinary & secondary education 8.8 7.1 7.4 5.8 6.0 35.1 Technical & vocational training 4.1 3.1 2.4 1.5 - 11.0 National polytechnic 3.1 - - - - 3.1 Dong Dok tertiary institution 1.8 3.1 2.4 - - 7.2 Non-formal education 0.3 0.2 - - - 0.5 New programs 1.3 10.2 16.5 27.0 33.5 88.6 Primary & secondary education - 6.1 9.4 14.6 17.9 48.0 Technical & vocational training - - 1.8 4.4 7.4 13.6 Non-formal education - 1.0 1.8 2.9 3.0 8.7 Rural primary & vocational programs 1.3 3.1 3.5 5.1 5.2 18.3 Total Education PIP 19.4 23.6 28.7 34.3 39.4 145.5 Current Expenditure 27.3 34.0 39.0 44.0 50.0 194.3 Total Education Expenditure 46.6 57.6 67.7 78.3 89.4 339.7 Total Education Expenditure as % of GDP 2.9 3.0 3.1 3.2 3.3 3.1 Sources: Government authorities and World Bank staff estimates. - 96 - H. RECOMMENDATIONS 50. All levels of education in Lao PDR require improvement. The costs of achieving this goal are enormous, especially given the current patterns of resource management, allocation, and utilization. Increases in investment observed in recent years appear unsustainable. Growth in the recurrent budget will be limited, and there is little scope for increasing contributions from communities and parents. 51. The challenge, therefore, for Lao PDR 's education finance managers will be to economize, prioritize, and protect. First, they must ensure that available government and non-government resources are mobilized and used most effectively and efficiently. Second, they must ensure that new investments in the sector are made according to a well reasoned set of priorities. This should balance concerns for quality, growth, and equity. Particular attention should be given to developing a coherent national education investment policy which will lay the groundwork for improved donor coordination. Third, managers must develop planning and budgeting procedures to protect key categories of expenditures and vulnerable groups, such as basic education, provision of teaching and learning materials, and support for educational services in disadvantaged areas. Mobilize Additional Resources. The scope for increasing government revenues to the education sector is probably quite limited. One possibility would be for the Government to significantly increase its efforts at domestic resource mobilization, as discussed in Chapter 1 of the Main Report. A second possibility would be a deliberate policy decision to increase education's share of the recurrent budget by reallocating shares away from other sectors. The Government could also, theoretically, reallocate funds from investment to recurrent budgets. However as was pointed out above, the Government only finances about 25% of the investment budget, largely in the form of counterpart funds for donor-assisted projects. The most promising avenue is to achieve cost savings within the education sector itself by (i) lowering the unit costs of post- primary education; (ii) investing in key quality inputs to improve efficiency; (iii) distributing resources more efficiently and effectively across schools; (iv) improving teacher utilization; and (v) developing more equitable cost-sharing arrangements for higher education. 52. Very high rates of poverty and income instability14 limit the scope for additional community resources -- particularly cash expenditures at the lower levels of education. Only 18% of the of population can -- and do -- pay more for education (calculated from Social Assessment Table 1.3). The Government might consider ways to assist disadvantaged poorer population groups. That said, there may be ways to improve the current mobilization and utilization of community resources, by, for example, the following actions: * Encourage in-kind community contributions to school construction and maintenance. This implies construction models based on locally available materials and technologies. * Explore the use of simple, school or village-based "school savings" or "education lending" programs that would allow families with seasonal income variations to incur expenditures when their incomes allow. 4 The national incidence of poverty is estimated at 46%. Earnings from agriculture support about 83% of the income earning population. (World Bank, "Social Development Assessment and Strategy," 1995). Most of Lao PDR's agricultural sector is characterized by low-tech semi-subsistence activity; incomes are highly dependent on the vagaries of weather and poorly developed market mechanisms. - 97 - * Encourage a market for educational materials (specific steps to make textbooks, uniforms, supplies more widely available) * Disseminate lessons learned from case studies of communities or schools which have been particularly successful at raising supplementary resources for education. > Economize Education Resources * Introduce more equitable financing of post-secondary education by better targeting of fellowships and other subsidies. * Explore options for merging very small schools to form more economical units, taking into account geographic and demographic constraints. * Continue to decrease the proportion of non-teaching staff, particularly at higher levels of education. 3 Experiment with training, curriculum restructuring, or job restructuring in order to increase utilization of teachers. > Prioritize Education Expenditures * Develop a coherent, prioritized national investment policy in education and use this as a framework for donor coordination * Build analysis of recurrent cost implications of investments into the planning and budgeting processes. = Protect Core Education Program * Ensure adequate provision in the budget for textbooks, materials, in-service teacher training, and school maintenance, especially in poor communities where the capacity for resource generation is limited. * Monitor the growth of higher education. Use savings from efficiency gains in higher education as the principal source of growth for higher education. Maintain or increase the share of resources for primary and lower secondary education. - 99 - PART II BUDGETARY INSTITUTIONS CHAPTER 6: BUDGETARY ARRANGEMENTS IN LAO PDR 1. With the economic transition to a market-based economy, the benchmark of good public administration is now the effectiveness with which public sector institutions use resources otherwise available to the private sector and shape an enabling environment for the private sector. A. PREvIous BUDGETARY ARRANGEMENTS 2. Prior to 1992, public sector activity in Lao PDR was essentially managed through split-level budgeting arrangements involving separate budgets at both the provincial and national levels. Revenues were collected by the provincial authorities and divided between disbursement through the provincial budget and transmission to the central authorities. Provincial governments were not able to run deficits in their own right. Outlays of the central budget were largely executed through the return of funds to the provincial authorities, which disbursed them on behalf of the national authorities. 3. These two-tier budgeting arrangements had many of the characteristics of a weakly coordinated federal system and exhibited some of the drawbacks of shared fiscal responsibility. In particular, there were shortcomings in the monitoring and control of expenditures, since expenditure "envelopes" for national programs were centrally determined but provincially disbursed on specific program activities. Therefore, the center had difficulty determining program outlays owing to lack of information about the costs and benefits of incremental changes in program funding at the provincial level. 4. In 1991 a single national budgeting system replaced this split-level system, and now all public sector revenues and outlays are managed by the national Government. However, the provincial authorities (i) identify spending proposals and (ii) manage outlays. A National Treasury with provincial offices was established in 1992 to consolidate the financial operations of different government levels within a unified accounting framework. While outlays continue to be made at the provincial level -- utilizing credits made available by the provincial Treasury offices -- authorization of commitments and disbursements is required from Vientiane. 5. The effectiveness of the newly instituted national budgeting processes was further increased by several budgetary decrees issued during 1993. These included splitting the Ministry of Economy, Planning and Finance into the Ministry of Finance (MoF), which encompasses the National Treasury, for budget management, and the Committee for Planning and Cooperation (CPC) under the Office of the Prime Minister, to manage investment, foreign assistance, and the medium-term macroeconomic framework. Following a Government restructuring in 1996, the CPC was restructured as the State Planning Committee (SPC), while the responsibility for foreign aid was transferred to the Prime Minister's Office. - 100- B. CURRENT BUDGETARY ARRANGEMENTS 6. Reflecting these reforms, Lao PDR public sector resource management is current-ix 0anrltcd, coordinated, and implemented through the following: * A medium-term development plan containing overall directions and priorities, and an associated Public Investment Program (PIP) prepared by the Committee f6,r )l1anning and Cooperation and rolled forward annually * An annual State Budget for fiscal years beginning each October prepared by the Department of the Budget in the MoF, which reports on public expenditure outcomes for the previous fiscal year and authorizes commitments for the forthcoming year. 7. The annual State Budget helps reconcile the Government's medium-term develupment strategy with the resource envelope available year by year for public sector activity. IPrior to t 90, the State Budget financed public sector production and distribution of goods and services ky subsidized state enterprises. However, now most state enterprises are privatized and the role is to fund the implementation of the PIP, which accounts for almost half of budget outlays. Consequently: - A larger proportion of budget decisions are only decisions on whether to accept or reject a particular investment proposal with its associated financial commitments, rather than adjusting at the margin the level of ongoing funding for a state enterprise or sunkk capital - Budget institutions and processes need to effectively identify, prioritize, and imp4l'ment projects, rather than extracting surplus from public enterprises. C. DESCRIPTION OF THE BUDGET PROCESS 8. Budget processes in Lao PDR are somewhat opaque but rather effective. The main stages in budget preparation and execution involve identifying and prioritizing spending proposals, erisurilg that proposals are consistent with funding constraints, and financing arrangements fbi m- measur es approved in the budget. 9. Identifying and Prioritizing. The identification and prioritizing of arrangeenelits Ai the annual budget are relatively highly decentralized procedures. Early in the calendar year, line ministries submit to the SPC investment proposals for consideration in the current year s budget process. Other proposals are developed by the provincial offices of the line mninistries and coordinated by the provincial governor. Project proposals from the line ministries and provinces are prioritized by the Cabinet from March, based on the funds expected to be available for capital purposes, the funding requirements of projects already under way, and counterpart payments associated with new projects. These prioritizing decisions are made against the backgrouLnd or seven overall development priorities contained in the Five Year Plan (see Chapter 1 of the Main Report). 10. The investment component of the budget proposed by the SPC is then forwarded to the MoF for reconciliation with the latter's revenue projections for the budget year. Some 80% of l'IP pro jects are donor funded and most require some counterpart payments. Thus, the MoF must ensure tlhat the budget revenue can accommodate these commitments plus recurrent outlays. II. Consistency. The MoF then prepares a draft State Budgetframework around May, which includes total outlays, total revenues, and the resulting budget deficit/surplus for the forthcoming - 101 - budget year. The framework is submitted by the MoF to the Government, along with options for adjusting the bottom line deficitlsurplus. This budget framework is reviewed and finalized by the Government, which then transmits indicative credit ceilings for the budget year to individual line agencies, together with technical instructions on their activities for the forthcoming year. 12. On the basis of this advice from the Government, the line ministries formulate their detailed spending proposals in June/July, and send the proposals to the MoF. Finance consolidates the line ministry spending proposals into a detailed draft of the State Budget, by consulting with line ministries, to review and revise proposals. The detailed draft budget is then returned for review and endorsement to the Government, which arbitrates among different ministries and the provinces in a consensus-oriented approach. 13. When it has been finalized by the Government, the detailed budget is forwarded to the National Assembly for adoption prior to the budget year on October 1. The Lao PDR Constitution requires both the national budget and the Socio-Economic Development Plan to be approved by the National Assembly, which also reviews budget execution at midterm. 14. Financing Budget Outlays. Up to this stage, the budget process concerns planning budget outlays rather than actually financing them. The Department of Treasury in the MoF provides credits to the agencies, which disburse the appropriations approved by the National Assembly. All government revenues and outlays are accounted for by the Treasury which has the double responsibility of ensuring that: * Disbursements for activities do not exceed the amount appropriated for them * Sufficient funds (cash flow) are available to cover aggregate daily disbursements by the spending agencies. 15. Monitoring disbursements requires that the Department of Budget pre-audit requests by line agencies for credits to pay salaries, operations and maintenance (O&M), and capital outlays so that they are consistent with budget appropriations. The requests for credit are then passed on to the Treasury for action. To meet the second requirement, spending agencies submit proposed payment schedules to the Department of Budget which clears them against cash flow projections from the revenue side of the budget. Finance departments of the ministries in the provinces only have a limited authority for minor payments under review by the provincial budget office. 16. Some 85% of PIP funds included in the budget are disbursed through line ministries, and the balance is paid directly to the provincial administrations. However, management of line ministry outlays also retains a provincial structure, as does the collection of revenues. Within the central budgeting system, provincial branches of the MoF and the Treasury channel provincial revenue collections into payments to spending agencies in the individual province, reporting within-province cash flows to the central office on a monthly basis. These reports are consolidated into a monthly statement of Treasury operations for the whole country. Any excess of (authorized) outlays over revenues for an individual province is met by transfers from the MoF in Vientiane, which draws on surpluses from other provinces and receipts from foreign aid and borrowing. 17. These disbursement arrangements ensure that public sector activity is consistent with the budgetary intentions of the Government. However, the rigorous clearance arrangements imposed on spending agencies can also delay procurement activity and the implementation of projects. The result is a large gap between the value of the aid pledged by foreign donors and the aid actually - 102- utilized by the Lao Government. This raises the general question discussed in Chapter 4 of the Main Report of trading off strict budgetary control against streamlining budget processes to assist change management. D. THE PUBLIC INVESTMENT PROGRAM 18. The change management role of planning and budgeting institutions in Lao PDR is centered around a rolling five-year Public Investment Plan (PIP), which is supposed to reflect the priorities in the five-year development plan. In principle, the PIP is the main mechanism by which Lao PDR identifies projects that will require donor funding. 19. The design of the PIP influences the overall rate of economic growth in the country, the sectoral composition of growth (including the resources directed to social welfare), and the contribution of domestic and foreign resources to the growth process. It is therefore vital that the institutional processes for developing the PIP are consistent with effectively achieving both its macroeconomic role -- sustained growth consistent with stable prices and the balance of payments - and its compositional role -- the best possible choice of investment projects. 20. In its present form the PIP does not fulfill its central role as an instrument of expenditure planning. First, instead of focusing on programs and their total ongoing costs, the PIP places emphasis on a listing of investment projects that encourages the split among planning, policymaking, and budgeting. What is needed is to shift to a more integrated approach, where policy priorities drive resource allocation decisions within a realistic medium-term expenditure framework encompassing all expenditures. Such a policy and program approach to decision making, resource allocation, and aid management should be supported by forward expenditure estimates and sector investment programs (see Chapter 10). Second, there is no coherent methodology for project selection based on economic analysis or criteria for public sector involvement (along the lines discussed in Chapter I of these Background Papers). Finally, the usefulness of the PIP is undermined by methodological and presentational weaknesses (Chapter 7). 21. The PIP includes only those investment projects which will draw resources from the State Budget (either projects wholly funded by the Lao Government or, more generally, projects involving the provision by the Lao Government of counterpart funds for donor funding). The PIP therefore excludes projects which are fully donor funded, and projects involving government borrowing in a joint venture arrangement with private investors (such as hydroelectric projects). 22. The interaction between the annual budget and the five-year PIP is key to assessing the adequacy of institutional arrangements for managing government spending. Projects included in the PIP which are approved for funding in the forthcoming financial year are included in the national budget (for example, outlays on 3,172 PIP projects are included in the 1995-96 budget). Typically, the PIP components of the budget comprise around 50% of annual budget outlays, and the balance comprises recurrent spending on public administration and infrastructure maintenance below 10 million kip (recurrent spending on maintenance that is more than 10 million kip is included in the PIP). 23. The role of the forward planning process underpinning the PIP was strengthened in 1993 with the transfer of investment planning from the former Ministry of Economy, Planning and Finance to the Committee of Planning and Cooperation (CPC), which was restructured as the State Planning Committee (SPC) in 1996. The central role of the CPC in economic management was - 103 - further reinforced in 1994 when it took over the coordination of all grant aid and loan projects (previously decided in a relatively uncoordinated manner by individual line ministries and through a separate Ministry of External Economic Relations). 24. The Foreign Aid Management Committee (FAMC) was established in 1995 and is chaired by the Deputy Prime Minister and has inter-ministerial membership. The FAMC formulates development cooperation plans with donor countries and recommends to the Government priorities among projects proposed for foreign assistance (all proposals for foreign assistance must be approved by the FAMC). The FAMC can approve any project with outlays of less than US$1 million, while projects with higher outlays require clearance by the Prime Minister. Day-to-day aid coordination is undertaken by the Cooperation Department under the Prime Minister's Office (formerly under the CPC), which acts as the FAMC's secretariat. Since aid coordination is thus no longer housed under the same roof as the PIP department, it remains to be seen whether this system effectively integrates access to foreign aid with the Government's development planning priorities. E. THE REVENUE SIDE OF THE BUDGET 25. Some 80% of the PIP outlays are financed by foreign donors. However, this dominant role of donor financing does not mean that the budget can be formulated without reference to domestic resource constraints. Funds must be mobilized domestically to meet the counterpart payment obligations associated with donor finance, and to meet the recurrent component of the budget and the remaining capital outlays which are entirely financed from local sources. 26. Revenues are raised through profit and income taxes, turnover taxes, and taxes on intemational trade and on natural resources. Rate structures have been simplified and special exemptions minimized (Chapter 11). Revenue raising measures should be consistent with the objective of minimizing the distortion of private sector activity caused by the raising of public sector revenues. 27. The ratio of tax revenue to GDP has risen from 6.4% in 1989 to 9.7% in 1994/95, because the turnover tax was extended to services, income taxes were extended to individuals, and discretionary exemptions for customs duties were removed. However, realizing the full potential of broadening the tax base depends crucially on the quality of the institutional arrangements for revenue collection, involving the Tax Department and the Customs Department (see Chapter 1, Main Report). 28. Budgeting processes are compromised when inaccuracies in revenue forecasts for the budget year force mid-year revisions in budget outlays, or when some components of revenue collections are "quarantined" from the overall determination of budget outlays ("earmarking" of revenues). The Lao budget process appears to be largely free of both deficiencies: * Annual (but not necessarily medium-term) estimates of revenues are relatively robust (total revenue actually collected during the course of the budget year appears to closely reflect that estimated when the budget was promulgated a year earlier). This is illustrated in Figures 6.1 and 6.2, containing planned and actual revenues. - 104- FIGURE 6.1: MAJOR CATEGORIES OF NATIONAL REVENUE, 1994/95 30 Profit Ta Inom Tax Tunoe -mor Ta Timber . ,..5 .. ,˘> Tax Royalties 1;IGURE 6.2: PLANNED VERSUS ACTUAL REVENUES, 1992/93-1994/95 180 WW 40f X 20 :4 0 1992/93 62PLNEVEss1993/94 1994/95 |E lnned X Actual| * Most collections from the provinces are now credited to consolidated revenue rather than being partially withheld for provincial or other budgeting purposes. In addition, apparently no revenue is earmarked for particular uses outside the control of the annual budget. These arrangements give the Budget Department substantial freedom in using revenue to achieve the targets incorporated in annual budgets. At the same time, budgetary institutions must prevent revenue shortfalls from spilling into uncontrollable budget deficits and compromising macroeconomic stability. - 105 - APPENDIX: A FRAMEWORK FOR ANALYZING THE LAO BUDGET SYSTEM 1. The three basic objectives that any budgetary system aims to achieve are: (i) aggregate fiscal discipline; (ii) strategic prioritization; and (iii) technical efficiency in the use of budgeted resources. A budgetary system consists of institutional arrangements that assist in achieving one or more of these objectives. In Table A6.1, the key institutional arrangements are categorized according to the three basic objectives. Mechanisms for increasing transparency and accountability corresponding to each of the three categories are also presented. These mechanisms determine the credibility and effectiveness of the arrangements. To facilitate the discussion, the arrangements under each category are labeled A, B, etc. l TABLE A6.1: KEY INSTITUTIONAL ARRANGEMENTS AND EXPENDITURE OUTCOMES Institutional Arrangements Accountability Transparency L. Aggregate Fiscal Discipline A. Macro Framework and Targets Ex post Reconciliation Published B. Dominance of Central Ministries Sanctions Made Public C. Formal Constraints Openness of Financial Freedom of the Press Markets D. Hard Budget Constraints E. Comprehensiveness of Budget II. Prioritization A. Allocations linked to Strategic Outcomes Reporting on Outcomes Published B. Comprehensiveness of the Budget Ex post Evaluations Freedom of the Press C. Flexibility of Line Agencies Hard Budget Constraint Made Public D. Breadth of Consultations Technical Capacity of Comprehensible Parliament E. Use of Objective Criteria III. Technical Efficiency A. Civil Service Pay Performance-based Published B. Relative Autonomy of Line Agencies Chief Executive Tenure Made Public C. Accountability Financial Accounts, Audits Freedom of the Press Client Surveys 2. On the basis of a still evolving methodology first developed in Campos and Pradhan (1996), a graphical representation of the budgeting system of Lao PDR is presented in Figure A6.1 with the Indonesian system shown for comparison. The system is evaluated in terms of the three dimensions corresponding to the above-mentioned categories. In each dimension, each of the criteria labeled A, B, etc., corresponds to one of the institutional arrangements in the relevant category. For example, "A" under Aggregate Fiscal Discipline corresponds to "I-A" in Table A6.1 (Macro Framework and Targets). To evaluate the system in terms of individual criteria, several central and line agency senior officials were interviewed and requested to complete a diagnostic questionnaire. Their responses to the questionnaire were translated into index values, ranging from 0 to 1, for each For a detailed explanation of the arrangements and mechanisms and their relations, see Pradhan and Campos, "The Impact of Budgetary Institutions on Expenditure Outcomes," mimeo, World Bank, 1996. - 106- criterion/institutional arrangement. The shaded bars in each chart in Figure A6. 1 represent the index values associated with the corresponding institutional arrangement. They indicate how well the system can perform relative to an effectiveness norm.2 A value of I means that the system performs effectively; a value of less than I indicates some room for improvement relative to best practices. Given the index values, the relative performance of the budgeting system with respect to each of the three dimensions can be evaluated. For example, with respect to Aggregate Fiscal Discipline, it is observed that the system can be improved in terms of each of the five criteria (A through E). A measure of the degree of improvement is given by the relative slack coefficient -- in this case 0.538. The relative slack is nothing more than the area of the unshaded portion relative to the area of the whole "box".3 Although there is substantial room for improvement in the system in all three dimensions, there is much more slack in the dimensions of prioritization and technical efficiency. 2 This is not to be interpreted in the traditional welfare enhancing sense. The norms are established on the basis of best practices in the field, which of course have the potential for improvement. 3 The box has a total area of 5: a length of I is assigned to each criterion; the maximum height for each criterion is I (corresponding to the maximum index value). Since the methodology on which these results are based is being continuously revised, the results are rough and tentative. - 107 - FIGURE A6.1: BASIC BUDGETARY OBJECTIVES Lao - Aggregate Fiscal Discipline Indonesia - Aggregate Fiscal Discipline lI Relative Slaick = 0.507 Relative Slack = 0.538 R 0.5 0.5{ A B C D E A B C 0 E Lao - Prioritization Indonesia - Prioritization Relative Slack = 0.828 Relative Slack = 0.821 0.5 0.5 0 A B C D E A B C D E Lao - Technical Efficiency Indonesia - Technical Efficiency Relative Slack = 0.870 j Relative Slack = 0.856 0.5 0. 0 -,~0- A B C A BC - 109- CHAPTER 7: PUBLIC INVESTMENT PROGRAM (PIP): METHODOLOGICAL PROBLEMS 1. The Public Investment Plan (PIP) is the cornerstone of public expenditure planning in Lao PDR. It provides a medium-term prioritization of public investment decisions and indicates future budget obligations, including counterpart funding and the O&M implications of the capital expenditure program. The purpose of the PIP is to ensure that strategic investment priorities are selected, that project priorities are consistent with the policy actions for the sector, and that resources are mobilized to fund these projects. If used correctly, the PIP is an investment management tool that can maintain discipline over the level and allocation of domestic and external resources between and within sectors, consistent with development priorities and the available resource envelope. 2. As was mentioned in Chapter 6, the PIP at present is characterized by a number of flaws which undermnine its usefulness as an instrument of expenditure planning. These fall into five broad categories: (i) lack of comprehensiveness; (ii) unsystematic project selection and prioritization; (iii) incomplete project identification; (iv) inconsistent data; and (v) lack of a systematic presentation. The first two problems represent weaknesses in the analytical approach which require a profound overhaul of the present public planning process over the long term. The other three categories gravely weaken the accuracy of verifiable information regarding public expenditure plans (which this PER had to rely on) but could be addressed in the short term with a moderate effort to treat data in a more systematic fashion. 3. These methodological weaknesses in the current PIP fornat and process can be summarized as follows: (i) Lack of comprehensiveness: * The PIP is not integrated into a realistic medium-term expenditure framework encompassing all expenditures and focusing on programs and their total ongoing costs. Resource allocations are not driven by policy priorities in an integrated policy and program approach. Instead, planning, policymaking, and budgeting decisions are separated, with the PIP simply listing possible investment projects with little regard to the available resource envelope or forward expenditure estimates. * There is an absence of infornation regarding the estimated future recurrent expenditures implied by capital investrnents. Project-related operations and maintenance expenditures over 10 million kip are included in the PIP, but there are no O&M provisions after the end of the project. * There is inadequate project monitoring and evaluation of investment progress and project completion. - 110 - (ii) Unsystematic project selection and prioritization: * Projects are not prioritized in case of possible cuts in planned public investment. The PIP does not distinguish between a Core PIP (defined as a program of approved priority projects with assured funding for capital and current expenses) and a Reserve List (stand-by projects to be implemented when resources become available). * The PIP does not systematically discuss the public-private mix of service provisions and does not adequately examine the rationale of government intervention; * Expenditure allocations and project selection are not based on systematic criteria (along the lines of Chapter 1 of these Background Papers). The PIP provides no information on economic (cost-benefit) analysis or other principles that would guide prioritization across sectors and programs. * In identifying allocations within a sector the PIP does not look at the link between inputs (e.g., public spending on primary education) and outcomes (e.g., school enrollment at the primary level). (iii) Incomplete project identification: * Projects are often not clearly identified and data are incomplete. Especially in later years, projects at the concept stage are rather sketchy and consist of little more than a project name. Many fields are left blank owing to lack of information, yet subtotals are created and carried forward to summary schedules. When blank fields are encountered for a project, it is not possible for the user to tell whether the values are zero or whether the data could not be obtained. Many projects listed as foreign-financed do not identify a donor. * The PIP excludes major hydropower projects with large potential recurrent cost implications. Even large schemes for electricity generation with a sizable government share such as Theun Himboun or Nam Theun 2 are not presented, since they would otherwise dwarf the PIP share of other sectors. However, public funds may eventually be needed to staff, maintain, and repair these projects. * The PIP does include some capital investment programs in state-owned companies, such as water, electricity, and telecommunications, which are mostly donor financed (see Chapter 14). (iv) Inconsistent data: ' * A significant percentage of individual project line items fail one or more of two data consistency checks: (i) total actual project-to-date expenditures must be greater than or equal to the prior year's expenditures; and (ii) total estimated project costs must be greater than or equal to total project-to-date expenditures plus current year planned expenditures. The description of methodological problems in this chapter draws partially on a memorandum by Don Rudisuhle of PDP Australia, who worked on technical assistance for the Public Investment Monitoring System. - 111- * Throughout the PIP, different exchange rates are used to compute dollar and kip values, but these are not referenced externally, but instead, are buried within the equations themselves, making it impossible for the users of the data to verify or modify the calculations. Similarly, formulas used to distribute resource allocations among sources and time periods are also hard-coded into the spreadsheets. * The PIP does not discriminate properly between costs that are to be incurred in local currency and those that will require foreign exchange. In some cases, figures expressed in different units of currency and precision have been combined into a single column and totaled. * The handling of subtotals is not consistent throughout the PIP. In some cases, subtotals are listed only at the top of a section and are calculated using long, complex formulas that reference different areas of the spreadsheet. In other cases, subtotals are not supported by individual line items, but instead have simply been entered manually into the spreadsheet without any calculations based on individual supporting line items. (v) Lack of a systematic presentation: * There is no audit trail linking project line items to source documents such as project profiles or donor loan agreements. Short of an item-by-item 100% audit, there is no systematic way to validate the qualitative or quantitative information in the PIP. The lack of unique identifiers permits errors to creep into the data. Several large projects were duplicated within the PIP, possibly because they cut across under more than one economic sector. * The PIP was assembledfrom a large number of different, unlinked spreadsheets, which, although similar in appearance, often contained different internal structures. There is no methodology whereby the data cascade from the lowest level of detail to the global summary schedules. It was not possible to trace the amounts reflected in the summary schedules to the detailed sectoral project listings contained in the same document. There is no proper mechanism to ensure that changes made at lower levels of detail are carried all of the way through to the final summaries, which effectively renders the system incapable of reliable periodic updates and therefore useless for the purpose of monitoring projects. * Multiple versions of the same spreadsheets exist simultaneously on several computers and are not clearly identified as to what they represent and which one is final and definitive. At the other extreme, the entire data set for a sector was contained on a single floppy disk which was found to be corrupted and unreadable, forcing the manual recapture of all of the information. There is no systematic centralization and safekeeping, nor are there any appropriate file backup or anti-viral practices in effect. * Because the data are stored entirely in a large number of independent spreadsheets with a complex hierarchy of sub-categories, it is not possible to index the data on other keys, so as to produce quantitative data sorted and totaled by donor, status, or program. The separation of line ministry projects and provincial projects into two separate entities is an impediment to the production of meaningful analysis across these two areas. The - 112- user cannot view the global public investment picture according to common classifiers such as donor commitments. * There a lack of consistency in the naming of project classifiers. For example, one project might list the name of the donor country while another might list the name of that country's official aid agency. They are actually the same source, but would appear at first to be two different entities. * The PIP contains numerous small projects that should be consolidated into broader headings. The listing of these small projects detracts from important government priorities and should instead be presented by grouping bilateral, multilateral, NGO, and domestic sources of funding. * Throughout the PIP, there are a large number of sub-categories, often containing only one or two projects. This excessive disaggregation of information does not add anything to the analysis but, rather, contributes to clutter. * English labels of project names are often incomplete and awkward, frequently not conveying a proper meaning. 4. Extension of the PIP into a monitoring and evaluation system. In its current form, the PIP needs substantial enhancement before it can become a tool for project monitoring and evaluation. The Department of Public Investment Plan in the SPC has received substantial technical assistance to create a Public Investment Monitoring System (PIMS) (developed under ADB TA-2047 LAO) during 1995-96, but without much success. The suggested PIMS format has not been adopted and donors are evaluating how to salvage the project. As it stands, there has been little progress in overcoming the weaknesses noted above, which create serious obstacles to the conversion of the PIP data into a proper PIMS relational format. 5. Despite the lack of progress in building a PIMS, there is still a need for an effective monitoring system which would: (i) feed back the results of monitoring into the rolling annual investment plan; (ii) provide updated information from the line ministries and the Treasury to the Budget Department; (iii) coordinate monitoring functions among line ministries, the SPC, and PIUs; and (iv) show an up-to-date picture of actual project implementation and aid disbursements. 6. Recommendations. Further work is needed to strengthen the PIP as a useful instrument of expenditure planning. If there is any further technical assistance, it needs to determine whether the PIP Department's institutional capacity would be able to absorb additional advice and equipment effectively. It would also have to address a number of technical obstacles to the proper planning and monitoring of locally funded and foreign funded projects, namely: (i) Comprehensiveness: > Shift to a more integrated program and policy approach where policy priorities drive resource allocations. Decision making, resource allocation and aid management should be supported by forward expenditure estimates and sector investment programs (see Chapter 10). > Clearly identify and budget for longer-term recurrent costs implied by capital investments. - 113 - => Install a functioning system to monitor and evaluate public investment during budget execution and after project completion. (i) Project Prioritization: Distinguish high priority projects in a Core PIP from others on a Reserve List that can be postponed in case of funding shortfalls. > Use a coherent methodology for project selection (of at least the major projects) based on economic analysis and provide a rationale for public sector involvement (see Chapter 1 of these Background Papers). (iii) Project Identifi cation: > Fully identify projects and complete relevant project information. This would involve filling mere sector sub-categories with meaning by identifying actual programs and clarifying projects beyond the concept stage. Provide an alternative PIP presentation including all hydropower schemes with public sector participation. (iv) Data Management: = Create orderly management of equipment and records in the PIP Department. Implement formal operating procedures that will prevent data problems, in particular, archiving, backups, and virus prevention. f Train staff in methods of proper data collection, organization, and protection, stressing the need for accuracy, retrievability, and verifiability. Teach staff the fundamentals of spreadsheet and word-processing software. > See that actual and projected numbers agree with their respective sources in line ministries and in the MoF, and with foreign donors. Obtain feedback. Prepare reconciliations. = Transfer data into an effective Public Investment Monitoring System (PIMS) and issue reports for review and analysis. Train staff as part of the capture process. (v) Presentation: * Check for accurate and consistent labeling of all public investment project names. => Consolidate the hundreds of small projects into single line items on a sectoral and provincial level. = Write short abstracts for all major projects. = Classify all projects correctly regarding their status: "ongoing," "committed," "pipeline," or "concept." - 115- CHAPTER 8: THE REGIONAL DIMENSION: CENTER-PROVINCE RELATIONSHIPS 1. Historical and Geographic Legacy. Lao PDR is unique in its experience of fiscal re- centralization despite a worldwide trend toward devolution. While many countries had a tradition of centralization before evolving toward a more or less decentralized system, Lao PDR went through a phase of decentralization for about 15 years before deciding to recentralize power. Historically, the country's geography and ethnography have made it difficult to create a unified national economy: its landlocked position, mountainous terrain, and low population density left the country subject to continual foreign meddling in its domestic affairs and contributed to its fragmentation into a patchwork of often inaccessible local communities. An extremely poor communication network has traditionally contributed to a strong degree of regional autonomy and weak central authorities. 2. Regional Disparities. Economic growth in the last decade has sharpened the country's strong regional and ethnic disparities: increasing prosperity is concentrated in the urban centers, particularly Vientiane, while outlying provinces, where ethnic minorities dominate, risk being left behind. The center of the country (which includes Vientiane municipality and province), has benefited from growth and investment disproportionately to the population and is clearly more advanced than the north and the south in terms of poverty incidence, social service coverage, and infrastructure access. Per capita income by province is not readily available from official data, but an estimate for 1992/93 shows the highest levels in Vientiane municipality, whereas the lowest levels of GDP (and average consumption) per capita are found in the northern provinces (Luangnamtha, Phongsaly, Huaphanh) (Figure 8.1). Surprisingly, the poverty incidence in the South (60%) compares unfavorably with the North (46%) and the center (40%).1 However, the southern provinces (Champasack, Saravane, Attapeu, and Sekong) are also comparatively more prosperous (in terms of per capita GDP), which may indicate a pattern of growth that is not delivering benefits as equally as in the center or the more disadvantaged north. High disparities in average household consumption between villages with markets and those without markets suggest that a pattern of inequality has taken root in the south, with pockets of severe poverty, very low access to services, and the inability to nourish children adequately. 3. Toward Decentralization and Back. Changes in public administration have closely followed reversals in economic policy. After the Pathet Lao consolidated power after 1975, the old Constitution of the Monarchy was abolished and the administration of the state fell under the authority of the Party and became its executive agency. Political power was concentrated in the provinces which became quasi-autonomous. De facto, two levels of government existed, where central level line ministries had to obtain the authorization of the provincial governor to be able to cooperate with the corresponding technical departments in the provinces. Provinces had their own budget and were responsible for revenue collection as well as staff recruitment and compensation. See "Lao PDR Social Development Assessment and Strategy," World Bank (1995). - 116- Revenue surpluses did not accrue to the national budget but served to cover unexpected expenses. Apart from government subsidies for the poorest regions, there was no mechanism for fiscal revenue sharing among the provinces. This budget autonomy of the provinces just reinforced the regional disparities, where poor provinces without a strong revenue base could barely afford to provide essential health or education facilities.2 FIGURE 8.1: PROVINCIAL INCOME AND POPULATION 1992/93 GDP Per Capita 1995 Provincial Population 1992 US$ r > i Population 125 to 175 F 63,000 to 250,000 175 to 240 L 250,000 to 500,000 x 240 to 315 500,000 to 700,000 4. Center-Province Fiscal Relations. With the introduction of the New Economic Mechanism it became clear that the policy of decentralization had been premature, leading to a deterioration in public savings and difficulties in curtailing credit expansion. In 1991, the Constitution of Lao PDR was adopted which redressed the problem of excessive local autonomy without accountability by reconcentrating power in the central government.3 Decree No. 68/PM of August 28, 1991 established a system of centralized tax collection and budgetary functions in a single Ministry of Finance (MoF). Under the present system, all revenue collected in the provinces goes to the National Treasury, enters the central government budget, and is then allocated to the line ministries and provinces. Provincial administrations are nominally merely branches of the central ministries. 2 See "Lao PDR-Issues in Public Economics," World Bank (1990). 3 The Constitution also abolished the cantons (tassengs) as administrative units between the districts and villages, as well as the People's Assemblies that existed at the provincial, district, and canton levels. See P. Keulers, and S. Soukchaleun (1995), "L'Evolution Historique des Relations de Pouvoir entre les Autorites Centrales et les Autorites Locales dans la RDP Lao," Department of Public Administration and Civil Service/ UNDP. - 117- 5. Government Structure. Following the principle of "democratic centralism" (Article 5 of the Lao Constitution), the Lao People's Democratic Republic is composed of 16 provinces, the Vientiane prefecture, and the Saysomboun special zone, which in turn are organized into 134 districts and 11,371 villages. Provincial governors are nominated by the President at the suggestion of the Prime Minister. They have ministerial rank and report directly to the Prime Minister, who coordinates their activities. Governors are not only the formal heads of public administration in the provinces, but also generally hold the position of General Secretary of the Party at the provincial level. This dual mandate of important public administration officials blurs the distinction between the different executive levels of government and the corresponding Party organs, but limits the risk of creating a parallel administration. In the provinces, govemors are thus key figures at the political and administrative level, as they are in charge of supervising the provincial activities of both the Party and the public administration. Public expenditure planning is determined in an iterative process between the provinces and the central ministries, yet the leverage of individual provincial governors in influencing expenditure decisions during budget planning seems to vary with their political clout. 6. Provinces in the Budget Planning Process. The MoF, the SPC, and the line ministries have offices in the provinces which are all involved in the annual budget planning process. Provincial staff in each line ministry prepare investment proposals and budget requirements for the annual PIP. These are then discussed and endorsed by the provincial governor, usually after consultation with mass organizations. Rural Development Committees are influential in coordinating investment proposals at the local level. The endorsed provincial proposal is consolidated by provincial SPCs and finance officials who then submit it to the SPC and MoF in Vientiane. At the same time, provincial line ministries submit their requirements to the central line ministry where they are consolidated and prepared for submission to the SPC. The PIP Department in the SPC matches the project proposals with available resources. 7. Administration. While Lao PDR's centralized budget system has streamlined expenditure allocations and reduced misappropriations of public funds, it has not entirely eliminated confusion about the responsibilities of central and provincial administrations. For instance, it is not always clear which level of government is in charge of maintenance (e.g., for roads). Moreover, provincial administrations are poorly equipped and trained to handle some of their tasks. They have great influence but lack expertise in preparing the annual and long-term budgets, which can result in low quality project proposals. Moreover, since project funds come from the central government, provinces have a strong incentive to propose as many projects as possible. Provinces tend to lack the budget discipline to reduce commitments which exceed budget allocations. 8. Provincial Revenues. Provinces do not have their own sources of revenue but are rewarded for increased efforts at tax collection. If revenues exceed the annual targets, provinces are allowed to keep 30% of the surplus. Some of these funds are spent on improved equipment in the line ministries. Once the planning process is completed, there is little scope for provincial governors to change spending allocations during budget execution. However, during the 1995/96 budget ten provincial governors were rewarded for good performance by allowing relatively small discretionary funds for (emergency) spending without prior central clearance (but after consultation with the local Party). It remains to be seen whether this indicates a trend toward greater provincial autonomy in spending allocations. Figure 8.2 shows the 1995/96 provincial level PIP and ratio of revenue to PIP. - 118- FIGURE 8.2: PROVINCIAL LEVEL PUBLIC INVESTMENT PROGRAM (PIP) 1995/96 Provincial Level Ratio of Revenue to PIP, 1995/96 Revenue/PIP PIP per capita (kip) D- 3,900 to 5,000 4.0 to 8.6 * 5,000 to 7,000 * 7,000 to 26,500 9. There are substantial differences in revenues collected by each province (see Table 8.1 at the end of this chapter). Champasack, Savannakhet, Vientiane Province, and Khammuane perform best; their large net contribution to government revenues (Figure 8.4) reflects revenues from customs controls, logging and industry. This also suggests that the central govemment tends to collect higher tax and nontax revenues from provinces with higher per capita GDP. Overall, revenues collected by the provinces account for less than half of total state revenues. 10. Fiscal Sharing. The mechanism of fiscal resource transfers from richer to poorer provinces is designed to level regional income differences. However, the amount available for redistribution is relatively small, since only 15% of public expenditures is allocated directly to the provinces, while the bulk (85%) is budgeted for the central programs of the ministries (Figure 8.3). As a result, only a handful of provinces are net recipients of fiscal transfers, including Phongsaly, Oudomxay, Luangprabang, Huaphanh, Sekong, and Attapeu (Figure 8.4). Fiscal expenditure on a per capita basis varies widely across provinces (Figure 8.5). It appears that the most populous provinces (Vientiane and Champasak) receive less on a per capita basis than the smaller ones (Saysomboun, Sekong, and Phongsaly). Some poorer provinces in the north (Luangnamtha, Phongsaly), and particularly the Saysomboun special zone, receive higher transfers, but so do some richer provinces in the south (Sekong, Attapeu). Bokeo Province is a revenue outlier owing to its small population and high timber royalty. Unfortunately, this fiscal incidence is difficult to judge, since (i) central - 119- level line ministry expenditures often do not correspond to a particular province; and (ii) reliable per capita income figures are not available at the provincial level.4 FIGURE 8.3: REVENUE AND EXPENDITURE AT CENTRAL AND PROVINCIAL LEVELS Expenditures by Central and Revenues by Central and Provincial Levels, 1995/96 Provincial Levels, 1995/96 Provincial Provincial Level Level 15% 42% ~~~~~~~~~~~~~Level t= 58% Central Level 85% FIGURE 8.4: REVENUE AND EXPENDITURE BY PROVINCES, 1995/96 (MILLION KIP) 18,000 16,000- 14,000- 12,000- e 10,000 oExpenditure 0r 8,000 0 Revenue .__ 6,000- 4,000- - 2,000- 24.o°o°°fl a I 31. rh. rtml. rn ff. - i I 0 2 L } 4 i X g X I s | S g iL > This would also be misleading if provincial GDP is dominated by a single activity such as logging or hydropower (for which revenues are collected centrally). - 120- FIGURE 8.5: REVENUE AND EXPENDITURE PER CAPITA BY PROVINCES, 1995/96 (KIP/PERSON) 70,000 60,000 . 50,000 0 40,000 0 Ex.penditure :.30,000- 0 Revnu 20,000- - - 10,000_ 0L a. & E co E a * CDE g t X c f * C ~~ C. LO 11. Conclusions. Lao PDR's recentralization has created a system of fiscal sharing which seems to be performing reasonably well. As the economy grows and policymaking becomes more complex, it will probably be wise to start thinking about ways to adapt the role of provinces in public expenditure management. What will be needed is a stronger mechanism for leveling regional income disparities and addressing the particular needs of different regions, based on an automatic revenue sharing formula rather than political negotiations. Provinces could be given an even stronger incentive for revenue collection by allowing them to retain a larger share of revenue surplus. Similarly, more spending authority could be devolved to the provincial level to strengthen ownership in project planning and implementation. This will also make it necessary to clarify the administrative responsibility and strengthen the capacity of provincial authorities to prepare realistic project proposals. Provincial policymakers also frequently suggest a stronger focus of public expenditure planning on the country's three regions (north, south, center) to strengthen center- province relationships and local responsibility. Such a regional approach would have the advantage of drawing the central government's attention to similar issues (e.g., flooding, drought) faced by neighboring provinces; it may also reduce the administrative burden of bargaining among 18 different provinces and special regions. However, strengthening the regions may also create an additional layer of bureaucracy. Since the centers would be the three old royal capitals of former Lao kingdoms (Vientiane, Luangprabang, and Pakse), it may also be politically sensitive in a country concerned with national unity. TABLE 8.1: PLANNED NATIONAL REVENUE BY PROVINCES, 1995/96 (MILLION KIP, AS OF MAY 1996) ITEMS~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ mms g r = ~~~~~~~~1 * C s . ' if = j *f I TOWal 220,310 128,082 9,772 490 1,826 453 2,024 1,308 401 6,291 8,192 7,298 4,181 8,547 17,476 1,217 258 17,8t0 891 3,743 Ta.x *ad Custom Revenue 140,670 87,453 t,250 386 1,608 392 1,771 950 259 1,848 1,110 1,467 2,435 3,440 15,743 1,074 215 11,964 282 23 ProfitTnx(Companies) 24,350 21,901 15 9 25 50 75 102 23 98 58 94 250 103 803 81 68 591 4 bncomeTax 11,380 3,716 2,815 43 45 70 85 290 55 319 123 193 217 351 1,393 222 28 1,336 66 13 LandTax 3,100 490 69 23 48 35 95 29 197 39 184 90 110 743 255 18 643 32 Business LiUces 180 42 23 2 4 3 9 3 6 5 7 8 9 28 5 2 23 1 TunvxerTax 32,300 19,231 2,413 68 536 III 314 280 50 334 221 549 575 1,018 2,852 215 58 3,394 81 Excise Taxes 9,470 7,156 1 4 200 2 22 2 278 7 392 527 2 877 Nairnl Resous Duties 1,600 433 400 15 9 10 32 28 12 64 25 159 45 80 150 25 10 75 18 10 InpontTax 45,000 27,964 ItS 900 19 632 25 35 307 281 60 1,000 952 8,440 133 24 4,060 50 - ExportTax 6,150 5,150 19 20 62 30 1 3 365 4 108 29 4 129 3 216 7 1 Regslion Tax 1,740 476 824 16 4 6 9 24 8 23 10 17 10 26 140 44 2 96 5 OdierMisc. 5,400 1,384 1,270 29 43 8 356 94 19 133 66 96 204 395 538 89 5 653 18 Otber Revenues 79,640 40,629 1,522 104 218 61 253 358 142 4,443 7,082 5,831 1,746 5,107 1,733 143 43 5,896 609 3,720 Leasing 1,910 623 580 3 26 9 18 6 82 1 36 56 71 55 13 15 310 6 Concessions 320 152 100 2 3 3 8 5 1 2 1 30 2 10 I Finu/Penakies 3,540 2,966 20 1 37 5 16 20 2 55 26 11 28 30 197 17 1 100 8 Administutive Fees 900 578 3 7 10 5 6 3 48 42 5 123 18 26 11 15 DivideninsAmorization 3,250 2,940 20 15 37 37 10 70 80 6 35 Inteests Rtuurn on Loan 5,420 5,420 AssetSales 3,539 388 172 6 64 7 40 15 42 26 14 110 40 67 58 19 2,269 202 SOE Conversion 11,352 11,262 90 Capital Retum on Loan 6,909 6,800 25 6 71 7 Ovafli8ht 9,500 9,500 Explitaion of Lignite in Viengphgodka 0 utmberRoyakies 33,000 650 66 108 170 250 50 4,263 7,027 5554 1,500 5,000 1,130 3,146 381 3705 - 123 - CHAPTER 9: ISSUES IN PROJECT IMPLEMENTATION 1. The most appropriate sectoral allocation of public resources is meaningless unless projects are carried out effectively. In particular, the experience with donor-financed projects in Lao PDR has revealed numerous weaknesses in project implementation and lags in project disbursements. What are the typical causes of delays in implementing public projects financed by foreign aid? As with other developing countries around the world, there are at least four main problems that have contributed to delays in, or even failures of, projects in Lao PDR: * Inadequate project design and incorrect project selection * Lack of Government commitment * Insufficient donor coordination * Institutional constraints. 2. With hindsight, it is clear that, first of all, donors might in many cases have prevented delays in the first place with a better project design -- by being less ambitious, by choosing the implementing agency more carefully, by avoiding overloading project components, by taking account of the absorptive capacity of the recipient, and so forth. Failures of incorrectly selected projects underline the importance of selecting only projects that belong in the public sector and of screening these projects for viability and priority, along the lines discussed in Chapter 1 of these Background Papers. Second, project implementation has at times dragged because the Government was never fully committed, or because priorities changed for major project objectives for political or other reasons. As important as these factors may be, they are most effectively dealt with on a case- by-case basis rather than in a general report such as this. The third factor, insufficient donor coordination, has in some cases resulted in incoherent or contradictory policies and is discussed in more detail in Chapter 10. The present chapter addresses some of the key institutional constraints that typically lead to bottlenecks and unsatisfactory project performance in Lao PDR. A. PROJECT IMPLEMENTATION: INSTITUTIONAL CONSTRAINTS 3. Delays in project implementation have been a major problem of international donors. ADB and IDA, which together finance one-third of the foreign-funded projects in the country, have both experienced delays ranging from a few months to several years. 4. The Government has increased its efforts to enhance project implementation in several key areas, particularly procurement, accounting and auditing, disbursement, and training. However, more efforts are needed to overcome implementation constraints, especially since foreign assistance is so important for the PIP. Since it is critical for the Government to ensure that the proceeds of loans and grants are used economically and efficiently, sustained efforts should be made to improve the overall performance of the project portfolio by addressing implementation constraints in four critical areas: In response, IDA conducted a Country Portfolio Performance Review (CPPR) in May 1995 to address the major issues hindering the implementation of its investment program. ADB's 1994 Country Portfolio Review reported similar findings on implementation delays. By contrast, donor countries which rely almost exclusively on contractors and suppliers from their own country tend to face far fewer implementation problems (e.g., Japan). - 124- (i) Weakness of institutional capacity and severe shortage of qualified staff (ii) Insufficient and delayed release of counterpart funds (iii) Inefficient administrative review procedures (iv) Inadequate procurement and disbursement systems. 5. (i) Weakness of Institutional Capacity and Severe Shortage of Qualified Staff These areas are usually identified by donors as major constraints on project implementation and could be attributed to a number of factors, both at the central and the provincial level, for example: * Staff shortages * High staff turnover * Insufficient training and lack of skills * Weak local ownership of projects * Miscommunication resulting from cultural differences. 6. Staff Shortages. This limited institutional capacity is further stretched when a handful of qualified staff is expected to provide assistance to several donor-financed projects at once, or when such qualified staff members are reassigned to new projects. The increasing number of externally financed projects is severely taxing staff at the central level. In the main ministries, the increasing number of foreign assistance projects has multiplied the burden of a few qualified staff. Each donor- financed project requires managerial-level staff to consolidate, prepare, and coordinate implementation requirements set by different donors. At the same time, key personnel are sent on study tours or to donor-related meetings overseas, which bring business to a halt. 7. High Staff Turnover. High turnover, combined with staff shortages, has negatively affected project implementation. The Government cannot recruit and retain qualified staff without offering incentives to stay. In particular, this will require that civil service wages keep pace with the private sector and that patronage is replaced by a professional civil service recruitment process (see Chapter 15). 8. Insufficient Training and Lack of Skills. These two factors lie at the heart of human resource constraints in the public sector. In sectors such as health, agriculture, and forestry, skill shortages appear to be more pronounced at the provincial and district levels. 9. Weak Local Ownership of Projects. There is a stronger tendency to delegate responsibility to local project implementation units, but provinces and local authorities are often not adequately involved in the project design. Therefore, those responsible for carrying out projects have little opportunity to develop strong sense of ownership of a project and to adapt the project design to local needs. 10. Miscommunication Resulting from Cultural Differences. Frequent references to "weak institutional capacity" often disguise donors' lack of comprehension of Lao PDR's conditions. Feeble response by counterparts may indicate difficulty in communicating effectively in English on - 125- technical matters. It is remarkable that donors describe some agencies as "weak" although key staff often hold graduate degrees from foreign universities and speak fluent Russian, English, and French or other languages. In many cases, this seems to indicate a degree of cultural miscommunication between donors and Lao counterparts. In many cases, project delays may simply reflect a complicated, consensus-oriented decision-making process within the implementing agency, the Government, and the political party. Delays in processing technical assistance may simply mean that the donor has offered the equipment to an official who was not first in line or that the consultant was not acceptable to the Government. In other cases, however, this could indicate that the donor did not understand the polite refusal in the first place. 11. Technical assistance is usually the first response of donors to remedy a lack of institutional capacity. There is a clear need for capacity building and human resource development across all central and line ministries. IDA projects have included capacity building components, but there has been a traditional reluctance to borrow for technical assistance. In general, the experience with technical assistance in Lao PDR has been mixed; and there are probably as many examples of failed as there are of successful technical assistance. In several ministries foreign experts were hired to fill the gap for local qualified staff to help expedite project implementation. However, significant knowledge transfer or capacity building by experts has often been impaired by miscommunications ranging from simple language problems and lack of trained staff or adequate counterparts to cultural clashes and personality conflicts with counterparts. In many cases, experts have left frustrated, with little to show for their expensive assistance and even less that will endure after their departure. 12. (ii) Insufficient and Delayed Release of Counterpart Funds. Insufficient and delayed release of these funds have also contributed to delays in project implementation. Under the current budget process, counterpart obligations arising from donor-funded capital budgets contained in the PIP are estimated by the SPC, while the responsibility for setting domestic revenue targets for the annual Government expenditure budget lies with the Budget Department in the Ministry of Finance (MoF) Although consultations take place during the budget planning process, shortfalls in revenue collection have frequently resulted in substantial delays in releasing counterpart funds. 13. Comparing the actual values with the target values of the PIP reveals a consistent shortfall in the actual implementation of the five-year plan. This is clearly due to domestic shortfalls, since foreign donors have generally delivered on their promises (in fact, in 1995 foreign contributions even exceeded the target, owing to devaluation) (see Figure 9.1). By contrast, in aggregate terms, the Government has not been adequately fulfilling its requirements for domestic funding of the PIP. The causes of this shortfall can be found in the dual weakness of resource availability and implementation capacity. 14. Over the medium term, the Government intends to increase the domestic contribution to the total public investment as a way of addressing insufficient counterpart funds. The draft PIP includes estimated counterpart funds (domestic costs) along with the foreign investmnent costs as part of the budget planning process. The draft PIP 1995/96-1999/2000 estimates the total five-year investment program at an amount of 1,500 billion kip of which the bulk (73.5%) is financed out of foreign aid and concessional loans, while domestic funding only accounts for an average of 26.5% (Table 9.1). Given that the PIP represents about 59.7% of total budgeted expenditures, the Government is trying to increase domestic funding to support investment projects. As is discussed in Chapter 2, Main Report, substantial efforts to mobilize domestic resources will have to be made if domestic funds are to grow twice as rapidly as foreign funds, so as to finance almost a third of public investment by the year 2000. -126- FIGURE 9.1: FINANCING OF PUBLIC INVESTMENT PROGRAM, 1991-95 Domestic Financing Foreign Financing 140 0 2 0m j j _ 10 1991 1992 1993 1994 1995 1991 1992 1993 1994 1995 | *]InIActual | PlarI *ActuaI Source. Dra2t 199602000 PIP TABLE 9.1: PUBLIC INVESTMENT PROGRAM, 1995/96-1999/2000 (BILLION KIP) Total 1995- Sources of Funds 1995/96 1996/97 1997/98 1998/99 1999/2000 2000 Domestic 48.0 61.3 78.4 95.4 114.8 397.8 % over previous year 27.6 27.9 21.8 20.3 % of Total 22.6 23.6 25.8 27.8 30.1 26.5 Foreign 164.6 197.9 225.6 247.5 266.6 1,102.2 • over previous year 20.2 14.0 9.7 7.7 % of Total 77.4 76.4 74.2 72.2 69.9 73.5 TOTAL 212.6 259.2 304.0 342.9 381.4 1500.0 Source: Draft 1996-2000 PIP. 15. (iii) Inefficient Administrative Review Procedures. These inefficient procedures have contributed to delays in project implementation and can be said to result from: * Insufficient delegation of decision-making authority * Inadequate audit and accounting systems * Lack of monitoring. - 127- 16. The failure to delegate decision-making authority among Government agencies adds unnecessary layers to the approval process. The MoF has recently increased its efforts to reduce delays. Further streamlining of the approval process for withdrawal applications in line ministries is nonetheless necessary to speed up the payment process. In a typical World Bank project, the project implementation unit (PIU) prepares a standard withdrawal application. The document is then sent to the Finance Department of the line ministry, which generally takes one week; it is subsequently sent to the Cabinet of the Ministry, which takes several weeks for review prior to presenting it to the Minister for signature. After a total of about three to six weeks, the document is sent to the Foreign Currency Affairs Department in the MoF for approval and returned for submission to the World Bank for withdrawal. Eliminating just one step, such as the review by the line ministry prior to the review by the MoF, could save two to three weeks. This step unnecessarily duplicates work, particularly since the Foreign Currency Affairs Department still prepares the withdrawal applications. 17. In the absence of an appropriate auditing system, the MoF continues to review all expenditures in detail for each payment request regardless of the size of the withdrawal application. To ensure timely payments by contractors or suppliers, IDA generally provides Special Accounts for most IDA-funded projects. However, instead of relieving cash flow problems and expediting payments, the MoF insists on auditing and authorizing withdrawals before, rather than after, PIUs draw from the Special Accounts, thereby greatly undermining their effectiveness. The recent adoption of modernized accounting standards is an important step in delegating approval authority to the line ministries in the future. However, this would require further strengthening of the auditing system, greater autonomy for the implementing units, and post-disbursement audits. 18. (iv) Inadequate Procurement and Disbursement Systems. Such inadequacies render administrative procedures both time-consuming and cumbersome. A key problem remains the lack of absorptive capacity of Lao institutions. In several cases, the borrower did not manage to comply with loan conditions in time. However, several steps were undertaken during the past year to improve procurement and disbursement performance. 19. Public procurement policies and contracting services were highlighted by the CPPR and the Procurement Review of the Bank as major constraints hindering the effectiveness of public spending. As a condition for approval of the World Bank's Structural Adjustment Credit (SAC III), the Government recently created the basis of a public procurement system when Decree No. 95/PM, on the procurement of goods, consultants, and services, was issued on December 5, 1995. A modification of this decree to harmonize its provisions with the procurement practices of international donors is being discussed.2 20. Disbursement performance has been slow against increasing loan approval and this has gradually increased the undisbursed net loan balance. Although the disbursement ratio for IDA credit to Lao PDR had increased from 17.0% in fiscal year 1994 to 18.2% at the end of fiscal year 1995, it is still relatively low compared to the regional average of 23.9% in fiscal year 1995 and 24.5% in fiscal year 1994. While disbursement exceeded the original plan in fiscal year 1995 for transport sector projects, projects in agriculture and human resources have been characterized by 2 Upon review, IDA and ADB identified a number of inappropriate provisions, including (i) price negotiation with the seller; (ii) award preferences in favor of local firms and state-owned enterprises; (iii) a tie-breaking mechanismn requiring the submission of new proposals; and (iv) a weak waiver which conditions the applicability of multilateral agencies procedures to the above provisions. - 128- continued low disbursement ratios. In part, this reflects a certain reluctance on the part of the Lao Government to use IDA funds for non-infrastructure investments. Contrary to the experience of IDA, ADB has seen the overall performance of its projects decline significantly. The disbursement ratio on ADB's portfolio (excluding program loans) decreased from 18% in 1990 to 14% in fiscal year 1994.3 Delays in social infrastructure projects, in particular, have contributed to the erosion of the disbursement ratio. Figure 9.2 shows project disbursement performance for the period 1990-95. In Box 9.1, technical assistance programs intended to address project implementation are discussed. FIGURE 9.2: PROJECT DISBURSEMENT PERFORMANCE, 1990-95 35 2 5 . ....... ... .2 20 15_i _ W_ _ -> 3 _________i 15 10 1990 1991 1992 1993 1994 1995 FY !-+ - Country Profile Orginal Estimate *.A ..Actual Disb. Source: IDA actual disbursements. 21. At the MoF, corrective measures were implemented to reduce the elapsed time on disbursement applications from an average of two weeks to three to five days. Despite this progress, there is evidence that the internal processing time within some line ministries has increased from an average of one to two weeks to three to six weeks. Apparently, this further delay is due in part to the shift in responsibilities--the line ministries are now held responsible and accountable for submitting accurate and complete documents. To expedite the process, the Foreign Currency Affairs Department (FCAD) in the MoF issued general rules on "Foreign Loan and Grant Management" (Regulation No. 1369/MoF) in March 1996, together with an appendix, "Implementing Advice No. 366/MoF." As the line ministries familiarize themselves with these rules and regulations, the numbers of applications being returned by MoF because of insufficient documentation should diminish. 3 ADB, 1994 Country Projects Review Mission Report. - 129- Box 9.1: TECHNICAL ASSISTANCE FOR PROJECT IMPLEMENTATION Most major donors have experienced slow disbursement performance and, in response, have initiated a number of technical assistance programs to address project implementation issues, ranging from monitoring and procurement, to auditing and training. These programs, however, have had a limited impact on project performance to date. Moi.itoring: IDA has provided an IDF grant to assist the Government in establishing a disbursement monitoring system in the Foreign Currency Affairs Department (FCAD) in the Ministry of Finance (MoF). This could set the stage for the MoF to provide line ministries and PIUs with greater financial autonomy once the annual budget of a project has been approved. It would also assist the ministries to monitor actual against planned disbursements, and would make expenditure information available to the Budget Planning Committee and the Parliament in a more timely manner. UNDP also provided technical assistance to the SPC to develop a monitoring system to capture all ongoing project disbursement investment plans of all ministries. Input forms were developed and distributed to the line ministries so that they could provide updated information. Procurement: ADB is providing technical assistance to the FCAD to implement the recently approved Procurement Decree (No. 95/PM). Standard national competitive bidding (NCB) documents for civil works have been prepared and are currently being reviewed by ADB and other donors. ADB has also prepared the Procurement Implementing Rules and Regulations (PIRR). The objective of targeted training should now be to focus on specific project-related problems. Accounting andAuditing: The Government has adopted international accounting standards based on the French model. An IDF grant will assist the Govermnent in developing accounting and auditing standards. There is also a proposed technical assistance in project accounting financed by ADB which will train accounting staff. Technical assistance to improve budget accounting and auditing systems was also approved and recently implemented to handle ADB-financed projects and to develop a project accounting manual. And ISO/SWEDFOREST's finance accounting and standard loan system for Swedish-funded projects is also under way. Disbursement: The Bank will conduct comprehensive procurement, disbursement, and accounting training, and an auditing seminar during 1996 for the MoF and PIU. B. RECOMMENDATIONS 22. The Government has recently made good progress in streamlining its administrative procedures to improve portfolio performance. Further substantive efforts to facilitate project implementation will be required to build on the progress achieved to date. In the first place, this will require adequate project design and proper project selection, taking account of absorptive capacity, investment priorities, and the rationale for public sector involvement. Foreign assistance will continue to be the major source of funding for the public investment program in Lao PDR in the near future, which makes it critical task to ensure Government commitment and effective aid coordination. To address institutional constraints to effective project implementation, it will be crucial for the Government to pursue a coherent strategy to improve implementation in the following areas listed below. - 130- = Improve institutional capacity: * Increase qualifications ofpersonnel. The Government is aware that it needs to enhance the skill levels of public sector staff, particularly in key agencies dealing with aid programs and, in the future, with ASEAN. Human resource development should to focus on measures discussed in Chapter 15, on civil service reform: (i) provide opportunities and rewards for increasing short-term and long-term training of civil servants, particularly in the English language; (ii) establish a wage policy and incentive structure that will permit the civil service to attract and retain highly qualified professional and technical staff who may otherwise move to the private sector; (iii) improve the recruitment system through introducing a standardized examination and reducing political patronage; (iv) gradually replace redundant poorly qualified personnel, and (v) swiftly reassign staff between central and provincial offices to agencies facing skill and personnel shortages. In the same vein, the Government should increase its efforts to attract expatriate Laotians to provide expertise and investment funds to their native country. * Involve provinces and local authorities at the district and village level (including Rural Development Committees) at an earlier stage in the project design to increase local ownership. * Improve information technology. The Government should increase its use of computerized information systems, in particular, to strengthen the capacities to (i) link the exchange of updated project information among the SPC, the MoF, and line ministries and (ii) accurately monitor project implementation and disbursement to compare budgeted and actual expenditures. * Learn donor procedures and strengthen country dialogue. Further training is needed on implementation guidelines and project procedures so that each line ministry can comply with specific donor requirements. Project performance will only be satisfactory when implementing agencies become fully competent in standard procedures. To the extent possible, donors should collaborate to reduce the burden caused by diverging rules on procurement, disbursement, and so forth. Further efforts at improving the dialogue between donors and government agencies are crucial to clarifying policy objectives, ensuring local ownership, and making technical assistance relevant. > Improve availability of counterpartfunds: * Provide forward estimates of domestic funding requirements. The Government must realistically assess to what degree domestic funding targets arising from donor-financed public investment are attainable. As is discussed in Chapter 4 of the Main Report, this requires (i) strengthening the budget planning process by improving linkages between the PIP Department in the SPC and the Budget Department in the MoF; and (ii) providing reliable multi-year forward estimates. = Improve administrative review procedures: * Reduce the multi-layer review and approval process. The line ministries should streamline disbursement and replenishment decisions by eliminating lengthy reviews in the line ministries prior to the MoF's review. - 131 - Use post-review instead of pre-reviews. The MoF should simplify the procedures for releasing funds from special accounts so that funds are received in time to execute the project. Given the concern with staff qualifications, the MoF may, on a trial basis, choose to delegate the authority to withdraw funds directly from special accounts to those PIUs that maintain highly qualified staff who have proven to be able to efficiently undertake this function. The MoF could then perform an annual audit (post review) to ensure quality and efficiency in utilization of funds, as is done in other countries. This would be expanded when the financial management and accounting system become a requirement for the implementation of future projects. * Strengthen accounting capacity. To ensure quality control, the Government should formulate and implement an action plan to further strengthen accounting and auditing systems (using ongoing and future technical assistance). = Improve procurement and disbursement systems: * Adjust procurement decree No. 95/PM. This should take into account standard international procurement practices of major multilateral donors. With ongoing technical assistance, the Government should be able to rapidly promulgate the implementing regulations on public procurement. * Implement National Competitive Bidding (NCB). This should be done for civil works on all new projects, as prepared under ADB technical assistance. * Develop a financial management system for future ADB and IDA projects. The Government should provide for a finance management specialist to develop a system for monitoring the flow of funds at PIUs in the line ministries. The terms of reference should include strengthening the knowledge and expertise of local staff so that they can assume financial management of the project after a short training period. * Increase autonomy of PIUs. The Government should delegate the authority for loan withdrawal applications and fund replenishment to the PIUs. Although this may not be possible now, it would be desirable that, after the strengthening of national accounting and auditing systems, the MoF should release this function to PIUs to ensure internal efficiency. - 133 - CHAPTER 10: AID COORDINATION A. OVERVIEW OF AID MANAGEMENT 1. Lao PDR is among the highest aid receivers in East Asia. During 1990-94, annual per capita aid inflows were in the range of US$40-45, with multilateral and bilateral donors accounting for 52% and 45% of the total disbursed, respectively (see Table 10.1). This level of aid dependence is comparable only to Cambodia in the East Asia region, and to African countries such as Ghana (US$38), Tanzania (US$40), Niger (US$45), and Zimbabwe (US$46).1 External assistance has been the main source of financing for the government budget. Prior to 1993, 90% of government expenditure was donor-funded, and government revenues were insufficient to cover even current expenditures. This position, however, has been reversed in the last two years. External assistance now accounts for approximately two-thirds of government expenditure. Development grants and loans combined represent 10 to 11% of GDP. TABLE 10.1: EXTERNAL ASSISTANCE, 1990-94 (US$ MILLION) 1990 1991 1992 1993 1994 Total Disbursed 174.5 177.3 167.2 228.4 233.7 Multilateral 99.2 103.1 88.5 114.1 104.9 Bilateral 71.2 69.9 72.4 106.7 118.4 NGO 4.1 4.3 6.3 7.6 10.4 2. The bulk of the total external assistance received during 1992-94 has been directed toward (i) industry, energy, transport, and communications; (ii) economic management; and (iii) agriculture, forestry, and fisheries. IDA and ADB accounted for the largest share of multilateral funds disbursed during this period, while Japan was the largest bilateral donor (see Table 10.2), with a total of US$124.9, or 15% of the total disbursed. As of September 30, 1995, 36% of cumulative IDA lending was in the transport and communication sector. Apart from Bank-financed projects, there has been substantial external assistance to the road sub-sector from other agencies and bilateral donors. UNDP has provided technical assistance for implementation of road projects on a continuing basis, while Japan has concentrated on infrastructure and social development. Funds from Sweden have been directed mostly toward agriculture, forestry, and fisheries. Although the Australian Agency for International Development (AusAID) has concentrated on infrastructure in the past, following the inauguration of the Friendship Bridge in 1994 further commitments of Australia in the range of US$10 million annually would concentrate on education and health, control and prevention of AIDS/HIV, and family planning. I African Development Indicators, 1996. 2 See ADB- Lao PDR Country Profile, 1995, for a detailed breakdown of project and technical assistance lending. - 134- TABLE 10.2: EXTERNAL ASSISTANCE BY SOURCES OF FUNDS, 1991-95 (ACTUAL DISBURSEMENT IN US$ MILLION) TOTAL Donors 1991 1992 1993 1994 1991-1994 Dist. Multilateral (of which) 103.1 88.6 114.1 104.9 410.7 51% ADB 59.1 16.1 39.1 28.5 142.8 18% IDA 14.1 38.2 38.0 26.8 117.1 15% IMF 11.3 8.0 14.1 16.9 50.3 6% UNDP* 18.5 20.2 15.7 11.6 66 8% Bilateral (of which) 69.9 72.4 106.7 118.4 367.4 46% Australia 6.4 10.5 12.7 11.9 41.5 5% France 6.9 7.3 11.3 9.4 34.9 4% Germany 13.3 3 14.3 11.8 42.4 5% Japan 20.9 22.5 36.5 45.0 124.9 15% Sweden 14.9 22.2 17.1 13.3 67.5 8% Others 7.5 6.9 14.8 27 56.2 7% NGO 4.3 6.3 7.6 10.4 28.6 4% TOTAL 177.3 167.3 228.4 233.7 806.7 100% * Includes UNDP-administered fund. Source: UNDP Development Corporation Report, September 1995. 3. Lao PDR remains highly dependent on extemal financing for investment projects. Eighty percent of the Govermment's Public Investment Program (PIP) over the period 1991-95 was donor- financed. Given this large role of extemal assistance in Lao PDR's total resource availability, it is crucial for the Government of Lao PDR to coordinate aid effectively, to: (i) direct funds into the Government's priority areas; (ii) maximize the use of scarce foreign resources and minimize any overlaps and prevent gaps in programs; (iii) avoid conflicting goals; (iv) contain budgetary implications of projects due to requirements for counterpart funds and operations and maintenance expenditures; and (v) minimize the strain on weak domestic institutions of coping with donor requirements. B. KEY ISSUES IN AID COORDINATION AND MANAGEMENT 4. Institutional Framework for Aid Coordination. In principle, there is a clear institutional set-up for identifying the Government's development objectives, coordinating external assistance, and guiding donors. This consists essentially of three main institutions: * The 1995-2000 Socio-Economic Development Plan outlines the country's development objectives. The PIP sets out the Government's priorities and identifies the areas in which assistance is needed. Investmnent decisions are planned within this framework, and extemal assistance is then sought in support of the Govemrnment's priority programs. - 135 - * A Donors' Round Table Conference is held biennially in Geneva under the auspices of UNDP, in an effort to collectively structure the assistance program.3 As the development institution with the largest presence in the field, the UNDP is in charge of coordinating donor assistance in order to ensure that funds support the Government's priority areas. * The Foreign Aid Management Committee (FAMC) was set up in 1994 as an interministerial committee to coordinate external assistance operations and to ensure that donor projects reflect the development objectives set out in the Socio-Economic Development Plan. Acting as the FAMC's secretariat, day-to-day aid coordination was handled by the Cooperation Department of the defunct Committee for Planning and Cooperation (CPC), which also oversaw the PIP process. When the CPC was abolished in mid-1996, aid coordination (together with oversight of the foreign direct investment approval process) was placed under the Committee for Investments and International Economic Relations Planning under the Prime Minister's Office, while the newly created State Planning Committee (SPC) was charged with managing public investrnent. 5. Issues in Aid Management. In spite of these efforts to coordinate the large aid flows, every visitor to Lao PDR is confronted with the large number of foreign experts in most Government agencies, the stories of unsuccessful technical assistance, the lack of communication among development agencies, and the absence of a coherent agenda. Key Government officials often have to spend a large part of their time dealing with the different assistance strategies of numerous donors, but they find it difficult to translate broad development objectives into a specific plan of priority actions. Effective project implementation is often hindered by weak institutional capacity and other problems (discussed in more detail elsewhere in these Background Papers). All of the above would indicate that the central issue for an aid-dependent country such as Lao PDR is the absorptive capacity of its institutions. In particular, anecdotal evidence suggests that there are a number of key questions which need to be addressed with regard to the coordination and management of aid: (i) Are the Government's priorities clear, and are donor programs consistent with Government priorities? (ii) Is the coordination capacity (i.e., Round Table Coordination Department and FAMC) sufficiently effective? (iii) Are sectoral strategies resultingfrom donor assistance sufficiently coherent? (i) Government Priorities 6. Official development priorities are often not very clearly defined, nor are they well articulated in terms of how they translate into actual programs. Although the PIP and the Socio- Economic Plan are intended to identify sectoral investment priorities and areas where external funds are required, the Government's priorities within some sectors are not well expressed. Moreover, there are occasional conflicts between sectoral programs (e.g., rural roads are claimed by both the Agriculture and Transport Ministries). Where objectives are too broad or are incoherent, the development agenda is not led by the Government but is more likely to be a product of multiple 3 For the first time, an annual follow-up conference was held in Vientiane in 1995. - 136- assistance programns. In addition, projects have frequently been delayed because the authorities have changed their preferences on projects. 7. Donor-driven priorities are not always consistent with the Government's strategic priorities, or may frequently delay Government programs at the sectoral leveL The PIP also does not give a complete representation of donor assistance, since NGO-financed projects and some large hydropower projects are not included in the PIP. Unless methodological problems in the PIP are addressed and a Public Investment Monitoring System is developed, there will always be a risk that donors will prefer to finance new operations or technical assistance projects that suit their own preferences and agenda, rather than projects that are part of the PIP. There is a risk of divorcing the coordination of aid flows from effective public investment planning after the split of the CPC into two different functions. Although final approval for major projects has always ultimately rested with the Prime Minister's Office, the new institutional set-up may make it even more difficult to ensure that the bulk of public investment financed by foreign aid really meets the country's development objectives and takes account of recurrent expenditure implications. (ii) Coordination Capacity 8. The use of sector coordination groups is limited. The Round Table needs to be supplemented by more effective and more focused sectoral coordination of external assistance. The monthly ambassadors' meeting and the regular meeting of UN agencies provide a forum for donor discussions in Vientiane, but they are usually not focused on technical details at the sectoral level, do not include the Government or multilateral donors, or tend to be ineffective in addressing problems.4 Occasional sectoral conferences have already been held for certain sectors (health, transport, energy, forestry) but their effect in coordinating assistance has been limited. Building on this experience, the Government should take the initiative in establishing regular results-oriented meetings of sector coordination groups, representing all donors and key Government agencies. 9. The linkages among the aid management units in the ministries, the SPC, FAMC, and the donors--especially bilateral donors and NGOs--are weak, as some of these units operate independently. Bilateral donors and NGOs often bypass the SPC and deal directly with line agencies for individual projects. The PIP does not always clearly identify projects by donors or indicate whether financing is still needed. This causes difficulties in identifying priority projects in various sectors and places cumulative demands for additional counterpart funds and future demands on recurrent costs. Moreover, multiple bureaucratic requirements from different donors for procurement and disbursement pose a difficult burden for institutions which are already overstretched. Donors tend to overload the existing administrative capacities by trying to fulfill their individual mandates and following their own agendas. Frequent missions and strategy discussions by a multitude of donors places a heavy demand on the same handful of officials, limiting the time they have to do their day-to-day work. 4 For instance, there has been a long-standing and fruitless debate on persuading the Government to agree to issue a one-year (instead of six-month) visa for donor personnel. - 137- (iii) Sectoral Strategies 10. Aid coordination works well in some sectors but seems to be fragmented in others. As discussed in Chapter 4 of these Background Papers, recent experience suggests that donor coordination in the health sector, where 25% of the total spending comes from donors, has improved. Similarly, donors have been more successful in avoiding conflicting priorities in the transport sector. In the education sector, it remains unclear whether donor programs reflect the sectoral investment priorities outlined by the Government. Some of these coordination problems reflect a lack of effective communication among donors and the ministries, in addition to the complexity of the sector. 11. Lack of effective coordination leads to overlaps and conflicts. While it would be unlikely to find two donors financing the same national road, the same unfortunately cannot be said of some less visible technical assistance or analytical works. Although studies, surveys, sector reports, and other analytical research work often carry substantial costs, they frequently overlap or duplicate work done by other donors. More worrisome is the fact that the authorities may receive conflicting advice simply because donors have not shared their findings with each other. Since the Government is often not aware of the real costs of experts, it has little incentive to streamline supply-driven assistance and avoid duplication of this seemingly free good. C. RECOMMENDATIONS 12. Lao PDR is highly dependent on aid. In many respects it suffers from the typical problems with the implementation of development programs encountered in other aid-dependent countries: insufficient local ownership and commitment; the lack of a noticeable "trickle-down" effect from some individual projects; projects not maintained after initial implementation; confusion and dissipation of effort caused by different approaches pushed by different donors; excessive numbers of expatriate technical assistance personnel; weakening of government capacity by the creation of donor-financed project units; and, finally, inefficient allocations of public expenditures. Addressing these issues is a difficult process, but some key step can be undertaken to improve the coordination of aid flows and improve project implementation. = Involve donors at an earlier stage of the PIP process. Given the large role of external financing, aid donors are crucial to the implementation of the public expenditure program. The Government must take the lead in providing guidance on investment needs and ensuring that donor activities do not operate at cross-purposes. Establishing a dialogue between the Government and donors at an early stage of PIP preparation, prior to the presentation at the Round Table Conference, would be one way of ensuring that donors' own priorities and objectives are consistent with the projects included in the PIP. This would also require presenting the PIP in a more credible format as part of a medium-term expenditure framework (see Chapter 7). It will also be important to ensure an effective coordination between the management of external assistance and public investment planning despite the newly established institutional separation between the Prime Minister's Office and the SPC. => Improve the institutional framework for the coordination and management of aid Results- oriented Donor-Government Working Groups should be established, ideally including representatives of the MoF, the SPC, and the relevant provincial administration and line ministries. These sectoral working groups and sectoral conferences (already in practice for - 138- energy and forestry) should focus on a particular sector and involve donor representatives, including NGOs. The purpose of the sectoral working groups would be to coordinate technical exchanges between central and line ministries and the donors. Possibly under the guidance of one key donor for each sector, these coordination groups should plan sectoral strategies and each donor's segment of investment within the sector strategy. They could be complemented by a technical coordinating committee (under the guidance of UNDP, or a major donor) dealing with administrative aspects on a regular basis. These working groups should closely involve Government agencies and report to the FAMC to ensure Government ownership and avoid a donor-driven agenda. = Establish a computerized monitoring system. Government agencies involved in aid coordination should move toward establishing computerized links or a compatible information system. In particular, the Foreign Currencies Affairs Department in the MoF could become a focal point as the primary source of information on commitments and disbursements of loans and grants, which should be coordinated with an effective Project Monitoring System based in the PIP Department of the SPC (see Chapter 7). Consolidate administrative procedures among donors. A beginning has been made by a coordinated approach between the World Bank and ADB in assisting the Government to incorporate international procurement practices in the Procurement Decree. => Explore Sectoral Investment Programs (SIPs) as an operational instrumentfor implementing the broad sector approach to investment lending. The Government would be in a better position to influence the allocation of external resources within a sector if the Government and donors were to agree on an SIP for key sectors, based on a coherent, concrete, and operational sectoral strategy. Following the successful experience in other countries, especially in Africa, this could initially be established in a pilot sector.5 The World Bank's sector work is trying to place more emphasis on the whole sector approach through the Education Strategy and the Rural Infrastructure Strategy. The main objective of an SIP is to give the Government a tool for crafting coherent sector strategies, presenting them to donors, and enhancing the development impact of usually fragmented investment assistance from various donors. In other words, put the Government in the driver's seat (see Box 10.1). The conditions for giving SIPs more prominance are very favorable in Lao PDR since the country is politically stable and most macroeconomic distortions have been eliminated. For details on the preparation, financing, and monitoring of SIPs and SIP successes and failures, and for ongoing and planned Bank SIPs, see Peter Harrold and Associates, "The Broad Approach to Investment Lending -- Sector Investment Programs," World Bank Discussion Paper, No. 302, Africa Technical Department, 1995. - 139- Box 10.1: THE BROAD SECTOR APPROACH TO INVESTMENT LENDING Sectoral Investment Programs (SIPs) have been conceived in response to problems with the implementation of development programs in Africa and elsewhere. There are six essential principles of sound project development, which taken together appear to address most of the implementation issues: (i) an SIP is sector-wide in scope, where a sector is defined as a coherent set of activities which need to be looked at together to make a meaningful assessment, and it must cover all sector expenditures, both current and capital; (ii) an SIP has to be based on a clear sector strategy and policy framework; (iii) local stakeholders, meaning government, beneficiaries, and the private sector must be fully in charge; (iv) all main donors must sign on to the approach and participate in its financing; (v) implementation arrangements should to the extent possible be common to all financiers; and (vi) local capacity, rather than long-term technical assistance, should be relied upon for the project. Characteristics of the current donor assistance (without the whole sector approach): =: Absence of country ownership of projects, leading to problems of sustainability > Piecemeal and sometimes contradictory approach among donors on sector policies and institutional development issues => Proliferation of poorly coordinated programs of various donors =' Each donor uses own procurement, disbursement, auditing, and monitoring procedures => Setting up by donors of special project units to manage their projects, staffed with expatriates or with government officials lured from their posts by much higher salaries, which further weakens government implementation capacity. Characteristics of the proposed SIP strategy: => Greater country ownership in the preparation and implementation of projects, with assistance from donors and NGOs as needed => The starting point is agreement among the government and the donors on a coherent sector framework, covering all relevant policies, programs, projects, and institutional development issues => Much better coordination among donors on a jointly agreed program > Effort made to use identical procedures by all donors to avoid overstretching borrower's institutional capacity => Emphasis on local capacity building, use of existing or appropriately modified structures in ministries, with minimal use of long-term expatriate consultants. An SIP will neither be possible nor desirable in all sectors. In particular, it may not be successful in the presence of macroeconomic distortions. Also, donors will need to be content with the aggregate pattern of public expenditure. A difficult issue is whether good government implementation capacity is necessary. This should be addressed during project implementation, although some proven financial capacity is very desirable. Source: Peter Harrold and Associates, 1995. - 141 - PART III. REVENUE MOBILIZATION CHAPTER 11: REFORMS OF THE TAX SYSTEM AND TAX ADMINISTRATION A. CONSTRAINTS OF TAx REFORM 1. Mobilizing sufficient revenue to finance the large development agenda is a key policy challenge in Lao PDR, and it requires effective and equitable taxation. Considerable progress has been made in establishing the foundations of a sound system of taxation in Lao PDR despite the substantial constraints that policymakers face: * The Lao economy is dominated by a large subsistence sector that is likely to remain non- taxable for some time to come - The economy continues to rely heavily on external grants to finance development projects, and on foreign investment to finance growth, which means that imports and incomes associated with these projects and investments may have to be lightly taxed or even exempt in the short run * The country is landlocked and relies heavily on its neighbors for trade and investment, which implies that the tax system in Lao PDR cannot be widely divergent from that of its neighbors or others who provide the country with most of its foreign investment ' The limited tax and customs administration capacity, on the one hand, and the limited taxpayer bookkeeping and accounting, on the other hand, require that the tax system remain simple. B. THE FIRST REFORM YEARS, 1986-88 2. During the first years of the New Economic Mechanism (1986-88), fiscal policy was ineffective owing to a lack of accurate information on the budgetary process. In general, many expenditure components were severely underestimated, and correcting fiscal imbalances was nearly impossible because no timely information was available on the expenditure process. 3. Revenue performance was equally weak during those first years. As a percentage of GDP, revenue dropped from 14.2% in 1986 to 8.3% in 1989. The main source of government tax revenue was negotiated transfers from state-owned enterprises the (SOEs). When the financial position of SOEs worsened in 1986-87, government revenue dropped. Subsequent price liberalization improved the SOEs' financial position and increased tax revenue from this source. The drop in nontax revenue aggravated the situation in those initial reform years. The combination of deteriorating revenue performance and lack of control over expenditures made the budgetary current balance surplus disappear after 1986. - 142- C. TAx REFORMS SINCE 1988 4. Tax reforms since 1988 have been designed to meet the needs of a market economy by replacing tax "extraction" with a genuine tax system. This process has gone through stages of innovations and corrections. In 1988, the existing system of profit transfers from SOEs was replaced by standard profit and turnover taxes.' The tariff structure was rationalized and the maximum duty rate was lowered to 70% from the high level of 200% and more.2 Taxes on export profits were introduced at rates varying from 50% to 80%. 5. This tax reform boosted revenue, but the ratio of expenditure to GDP more than doubled in that year, mainly owing to large public sector wage increases in the wake of price liberalization (Table 1, Statstical Appendix to these Background Papers). In 1989 the current balance became a deficit. The deficit on the overall balance had already widened significantly in 1988, mainly due to the January 1988 exchange rate devaluation (the rate used to record government transactions) which had caused a sharp increase in foreign-financed capital expenditure. 6. The second stage introduced several new taxes and adjustments to existing taxes, to correct for their highly differentiated nature. These reforms, plus better control over expenditure and other budgetary reforms introduced in the early 1990s, put Lao PDR on the path to fiscal sustainability. The second stage came to a provisional end when the new tax law was adopted in early 1996. Taxes introduced since 1988 are: personal income tax on employment and non-employment incomes of Lao citizens with a progressive tax schedule (1989);3 export duties (1989); taxes on agricultural land and other natural resources (1989); excise duties on petroleum products and selected luxury items; a minimum profit tax of 1.5% of turnover (1991); and a business registration tax, ensuring that all taxpayers register at the Tax Department (1991). 7. In addition, existing taxes were gradually reformed; the rates of profit taxes were lowered and unified at 45%, except for a 60% rate on bank profits (1989); the base of turnover taxes was extended to cover a wide range of services (1989), and the turnover tax was rationalized from five into two rates--5% and 10% (1991); a minimum threshold for income tax of 150,000 kip in annual income was introduced; the tax on agricultural land was replaced by a land tax on both urban and rural land (1992); among the natural resource taxes -- introduced to promote environmental conservation--timber royalties became the most important; export taxes on most items were abolished in 1994 to promote exports. D. THE NEW TAX LAW OF 1995 8. In 1995, the existing provisions of the major domestic taxes were codified and consolidated into a new tax law. This new law also introduced several changes to the existing tax system, including the following. (i) The base of turnover taxes was widened. This tax now covers most imported goods (including raw materials and capital goods) and domestically produced goods. The I Because these SOEs were granted financial autonomy in 1988-89, the revenue system had to be adjusted. 2 A few exceptional rates, higher than 70%/o, were retained. 3 Before that date, personal income tax applied only to incomes of nationals employed by foreigners. The new rates ranged from 2% to 30%, with no minimum threshold for taxable income. - 143 - number of exemptions has been greatly reduced. There are now four rates: 3, 5, 10, and 15%. The lower rates apply mainly to agricultural products and foodstuff, while the 15% rate applies to luxury items. These modifications will have a beneficial impact on the buoyancy of this tax. (ii) The minimum threshold for income tax has been raised from an annual income of 150,000 kip to 360,000 kip and the tax rate for the lowest bracket has been raised from 2% to 10%. Since monthly wages in the private sector are estimated to range from 20,000 to 300,000 kip, the increase in the minimum threshold would involve a small revenue loss. (iii) The profit tax rate has been reduced and standardized at 35% for banks and other financial institutions, to minimize the impact on government revenue of the lower tax rate on banks. A flat tax for foreign investors has been introduced. State-owned commercial banks have been instructed to pay higher dividends toward the budget. The minimum profit tax (applicable to gross turnover of the previous year of enterprises subject to the extended or ordinary accounting system) has been lowered from 1% to 0.5%. (iv) The system of specific excises has been replaced by ad valorem duties for all items. The new rates are set in such a way that the effective rate of taxation on the individual items remains broadly unchanged. The switch to ad valorem duties will also allow greater buoyancy. 9. Several measures in the new tax law are likely to generate a lasting positive effect on revenue. The buoyancy of the system with respect to GDP will be increased by: ad valorem excise duties which replace specific rates; higher turnover tax rates and the widening of the base for this tax category; the introduction of a land tax; the use of the market exchange rate for valuation of imports for duty and tax purposes; and the phasing out of the "forfeit" system for assessing profit and turnover taxes and replacing it by a "pro rata" system. 10. The full effect of the 1995 tax reform will only be felt as the new measures are applied over the next few years. An IMF report had warned of a significant negative impact on revenue owing to the reduction in the rates of profit tax, exemptions of undistributed profits from profit taxation, and reductions in the rate of turnover tax for selected goods and services.4 However, during 1995/96, fiscal developments remained broadly on track, and the Government enacted a number of revenue raising measures (increasing the land tax and excise duties on petroleum products) as well as intensifying efforts to improve the collection of timber royalties. These favorable developments allowed a budget outcome in 1995/96 with total revenues reaching 13.3% of GDP. E. IMPROVED SYSTEM OF REVENUE COLLECTION 11. In 1992, a unified general budget and a Treasury in the Ministry of Finance were established to execute the budget (see Chapter 8). Before that time, provincial administrations had independent budgets, and the provincial branches of the Central Bank made disbursements at the local level. 4For a full review of the Lao tax system, see Ghandi et al. (1995), "Lao PDR: Scope for Tax Reform," IMF. - 144- 12. The general budget now encompasses all revenues and expenditures nationwide which eases implementation and control of the budgetary exercise. Since 1992, further steps have been taken to centralize tax collection and expenditure control systems, to publish the budget documents, and to implement legislation to establish an annual budget cycle. Under the new system, provincial authorities act as agents for the central government to collect revenues and execute expenditures. Provincial treasury offices have been established in the principal provinces and eventually all provinces will have one. Pending the adoption of new budget accounting and auditing standards, scheduled for 1996, the Treasury has been working with a simplified public accounting system. This system ensures prior control of all expenditure categories at both commitment and payment stages. Public procurement guidelines ensure competitive bidding and transparency. 13. Now, tax and nontax revenues are collected through five channels. The Customs Department collects duties on international trade at five regional customs posts; the Tax Department collects taxes on domestic activities and imports; the provincial authorities collect about 40% of the taxes that fall under the Tax Department's responsibility; the Land Registration Department (Cadastre) collects the land tax; and timber royalties are collected directly by the Budget Department through the provincial authorities. Finally, nontax revenues are the joint responsibility of the State Assets Department (leasing, dividends, and overflight) and the Budget Department (all other categories). F. ELEMENTS OF FURTHER TAX REFORM 14. Further reform of the tax system is needed to provide a sound basis for fiscal sustainability. A key objective will be to raise the ratio of revenue to GDP by expanding and fully capturing the present revenue base. This requires making the tax system more buoyant and stable by reducing the present heavy reliance on timber royalties, trade taxes, and non-tax revenues and expanding the scope of domestic direct and indirect taxes. To limit the need for constant change in the tax system or the imposition of newer taxes, improvements in the tax administration are urgently needed. Greater taxpayer compliance and improved tax equity should aim to relieve the burdens of taxation for the poorer segments of society, while the richer segments should be made to pay their fair share in taxes. Based on these principles, a successful strategy for tax reform on the part of the Government should incorporate at least four elements as follows: = Maintain the momentum of structural reforms and continue to maintain macroeconomic stability. Tax reform cannot work unless other economic policies provide the right enabling environment to savers and investors. An important element of a stable system of taxation is also consistency in applying tax rules instead of discouraging investors with unpredictable ad hoc changes. => Contain the growth of government spending to financeable levels, while the revenue productivity and buoyancy of the tax system is being enhanced. Excessive budgetary deficits fuel inflationary pressures, and this makes the adoption and implementation of desirable tax reform difficult. = Continue to modernize the system of taxation, in other words, make taxes on incomes, profits, consumption, and imports as broad-based and nondistortionary as possible. ' Enhance the capacity of the Tax and Customs Department through Government efforts and technical assistance. - 145 - CHAPTER 12: HYDROPOWER AND PUBLIC EXPENDITURE A. PUBLIC FINANCE ISSUES IN POWER DEVELOPMENT 1. Lao PDR's hydropower potential is one of the main resources of the country. Electricity production currently contributes a relatively small share (1.4%) of GDP. However, the country's hydropower resources are being developed, principally for export. The country has an estimated potential of some 18,000 MW from over 60 possible project sites on the Mekong tributaries, of which less than 2% has been developed. 2. The experience with power trade has been largely positive, beginning with the Nam Ngum Dam since the 1970s and following an encouraging start in recent times with the Theun Himboun venture. There is a clearly identified export demand for electricity in the short term in Thailand, and in the medium to long term in Vietnam or even Cambodia. The challenge is how to develop Lao PDR's hydro resources in the best manner, from an environmental, social, and economic point of view. In terms of public expenditure management, the key question will be to assure an adequate and reliable stream of hydropower revenues in the future. 3. Electricity exports will become an increasingly important source of export revenues as Lao PDR prepares to build new hydropower schemes over the next decade. The Government has thus far signed at least 23 Memoranda of Understanding (MOUs) with foreign developers to augment its hydropower potential, proposing an installed capacity of 7,000 MW. A protocol has been concluded with Thailand with an intent to purchase up to 1,500 MW from Lao PDR by the end of the decade plus an additional 1,500 MW after the year 2000. This would satisfy about 10 to 15% of Thailand's growing electric power demand. In addition, MOUs for power sales have been signed with Cambodia and Vietnam. 4. Large hydropower projects are likely to bring significant future revenues but also entail significant uncertainties for the public expenditure program. While the environmental and social risks tend to receive considerable attention in the public discussion, the Lao Government still has not designed a consistent hydropower strategy which outlines the economic and financial dimensions of its future plans. In particular, some of the following issues bearing on public expenditure will need to be clarified: * Is the Government's strategy of hydropower expansion compatible with macroeconomic stability? * How many of the proposed hydropower schemes are likely to be built in the next five to ten years? * Which financing arrangements will ensure a fair share of the revenues for the Lao Government? * What future revenues in the form of royalties, dividends, and taxes are likely to accrue to the budget? Is the Government realistic in its expectations regarding the size and - 146- timing of future revenues and the potential of hydropower to substitute for the present dependence on foreign aid? * How will the potential revenues flowing from the different realistic alternatives affect the Government's expenditure priorities? Will the Government really have substantial revenues that it can spend on schools, hospitals, etc.? B. DOMESTIC POWER SUPPLY 5. Government Objectives. Public investment in the power sector amounts to about US$30 million per year and is aimed at two different levels. One element of power sector development deals with the needs of the domestic sector and is concerned with the high cost of supplying power to remote rural communities; the other focuses on the exploitation of hydropower resources where the initial investment requirements are beyond the means of the Government's financial resources. The Government strategy for developing the country's energy potential is centered around the following objectives: (i) to create and expand industrial opportunities; (ii) to extend rural electrification; (iii) to replace dependence on imported fuels for electricity generation; and (iv) to provide a source of foreign exchange. To accomplish these goals, a two-pronged strategy will focus, on the one hand, on developing small and medium-size power projects through concessional financing and, on the other hand, on private sector participation, on an equity basis, in the construction and operation of large projects. 6. Domestic Power Supply Systems. With the completion in 1995 of the Northern 115 kV subtransmission lines connecting Luangprabang and Vientiane, the national electricity supply system now comprises three grids: (i) the Northem Grid supplied by the Nam Ngum I and the Nam Dong hydropower schemes; (ii) the Central Grid (Savannakhet and Thakhek) supplied from imports from Thailand; and (iii) the Southern Grid (Champasack and Saravane) supplied by the Xeset and Salambam hydroelectric schemes, with backup from Thailand during the dry season. The existing installed capacity is ample to satisfy domestic electricity consumption, which experiences a peak of about 80MW and an annual consumption of about 308 GWh (equivalent to 65 kWh/capita). Domestic demand is concentrated in the urban areas, with the greatest consumption (250 GWh in 1995) in Vientiane (population 464,000), where 70% of households are connected (70,000 consumers). Outside of Vientiane, the total number of consumers is about 70,000 with a consumption of 35 GWh in 1995. Taking into account rural electrification, domestic demand is estimated to grow at about 8% annually, reaching 1,088 GWh by 2010. 7. Investment planning in transmission and distribution has been tailored to the modest needs of small, scattered load centers, avoiding premature investment in long domestic transmission lines and seeking medium-term low cost options such as improved connections with grids of neighboring countries. Under the World Bank's Southern Provinces Electrification Project (Credit 1826 LA), completed in December 1995, and the ongoing Provincial Grid Integration Project (Credit 2425 LA), the 115 kV subtransmission and MV and LV distribution systems have been expanded in the central and southern provinces. These projects are intended to reinforce the existing supply arrangements with Thailand while expanding service to 500 new villages with about 20,000 new households, mostly located in rural areas. 8. A Rural Electrification Sector Review and Policy Formulation report for Lao PDR prepared by the Government with assistance from Norway, proposed a strategy to (i) increase food production, (ii) develop agro-industry and other types of industry, (iii) provide lighting to rural communities, and - 147- (iv) stimulate productive activities in general. It was also aimed at doubling present electrification by the year 2000, through the connection of 88,000 new households in 2,300 villages. The Government recognizes the ambitiousness of the targets and is aware of the technical and financial challenges they represent. 9. The country's electrification rate (16%) is very low compared to that of other countries with poor resources (e.g., Bangladesh and Indonesia at 40-50%). To date, the Government has taken a fairly responsible view toward investment in domestic energy programs (focusing on rural electrification), preferring to sell surplus power to meet budgetary needs rather than expand the rural electrification networks into unprofitable areas. However, with the greater publicity given to larger build-operate-transfer (BOT) projects, the Government may come under increasing pressure to accelerate its expensive rural electrification programs. Furthermore, there is considerable public pressure to share the perceived benefits of hydropower in the form of cheap electricity. However, the Government recognizes that subsidized electricity would be unsustainable and inefficient. C. SCOPE OF POWER EXPORTS 10. Present Power Exports. Electricity exports have been critical to Lao PDR for nearly 20 years. Since the early 1970s, power has been exported to Thailand from the Nam Ngum hydroelectric plant which is serving the Vientiane area. Until the mid-1980s, the amount of exported electricity increased continuously, and a formal sales agreement was signed between Electricite du Lao (EdL), the Lao Government-owned electricity utility, and the Electricity Generation Authority of Thailand (EGAT). The Lao and Thai power grids are interconnected at a number of places along the Mekong River, which allows some of the provinces south of Vientiane to take power from the EGAT system. At present, EdL exports about 655 of its output of about 900 GWh of energy to Thailand. In 1995, this accounted for about US$24.1 million in export revenues, or about 7% of all Lao exports. Electricity is an important source of tax revenues; in 1994, EdL accounted for nearly US$9 million in total taxes and duties. 11. Existing Hydro-power Plants. The experience with Nam Ngum has demonstrated the reliability of Lao PDR as a supplier and Thailand as a customer. The project, built in the late 1960s, was financed by the Nam Ngum Development Fund, composed of grants from a number of countries and administered by the Bank. Initially, it had an installed capacity of 30 MW, which was later increased to 150 MW by the addition of three 40 MW units, two financed by ADB and one financed by the Bank. In 1995, EdL as the owner of Nam Ngum received US$20 million in revenues from the sale of energy to Thailand. These revenues will gradually decline as Lao PDR draws increasingly on Nam Ngum to meet the domestic load. The Xeset Hydroelectric Project is part of the Southern Grid. During the flood season, surplus power from the Southern Grid is exported to Thailand, and power is imported in the dry season when the output of Xeset is low. Nam Ngum and Xeset are owned and operated by EdL, the agency responsible for generation, transmission, and distribution. 12. Private Sector Investment. Since 1992, Lao PDR has encouraged developers to seek concessions for the private development of hydropower projects to export energy to Thailand. The Government has been very successful in attracting private investment, both foreign and local, to operate and construct new hydro generation facilities. Agreements for power export projects for Theun Himboun and Nam Theun 2 have established a precedent for a reasonable sharing of costs, benefits, and risks among the Government, the developers, and the customers. Given the large and expanding power market in Thailand, this could lead to substantial net inflows of foreign exchange. - 148- Thailand has expressed a willingness to enter into agreements with developers in Lao PDR for a total of 3,000 MW, or some 10% of Thailand's demand by the time these projects may be in service. 13. Lately, the Government has also attracted developers to participate in EHV transmission development, which will facilitate optimal generation production and expand the export market in the Indochina region. Indeed the country's strategic location between major power importers and exporters offers the prospect of participation in power trade through the provision of transmission wheeling services to adjoining countries. These transactions could provide the means to pay for much of the estimated US$500 million transmission infrastructure requirements associated with the current plans for the development of hydropower. D. OPPORTUNITIES AND RISKS OF HYDROPOWER 14. Opportunities of Hydropower. Lao PDR has few options for driving economic growth other than its potential to exploit timber and export hydropower. In addition to hydropower, foreign direct investment has been concentrated in textile production and mining; but the country's manufacturing base remains weak. Rising labor costs and human resource constraints limit the ability of the manufacturing sector to increase export earnings. Timber has been a main export, but logging cannot be sustained without incurring serious ecological damage. Although the financial sector has been growing with the opening of foreign banks, it remains fragile, as domestic savings rates are far below regional averages. Confidence in the domestic currency, the kip, may increase over time, but a large proportion of domestic transactions are conducted in U.S. dollars and Thai baht. This renders effective monetary policy difficult and limits the chances for the country to become a regional financial center. Fiscal policy is still characterized by weak capacity for effective budgetary management and domestic resource mobilization. Apart from hydropower, Lao PDR's comparative advantage seems to lie in its geographical position as a transport corridor and crossroads of a high- growth region, and also in some limited tourism. 15. Risks of Hydropower. Hydropower development promises substantial benefits, but it is also risky, at both the project and the overall program levels. In addition to ecological and social considerations, risks stem from uncertainties related to technical factors (which are often unavoidable, regardless of the efforts made in project preparation), financial issues (financial arrangements remain uncertain until closing), and economic consequences (there may be low economic returns or adverse macroeconomic impacts). Hydropower is also a relatively new business for the private sector in the region. In the past three to four decades, almost all large hydropower investments were undertaken by the public sector, in recognition of the capacity of the state to absorb major risks and the fact that many benefits were not seen as marketable. (The benefits of dams include, for instance, flood control or irrigation, which tend to play only a minor role in Lao PDR's hydropower development.) Because of this, it is reasonable to expect that a fairly large number of these MOUs may not reach fruition. Therefore, the ratio between MOUs and the number of projects expected to be built in the next five to eight years (say, 23 to 6) may not be out of proportion. 16. Macroeconomic Risks of Hydropower. Lao PDR is pinning its hopes on future revenues from hydropower to reduce its dependency on foreign assistance and to finance its large future development requirements. In particular, the proposed Nam Theun 2 (NT2) project is supposed to provide the country with the necessary resources to focus on poverty alleviation and improving social conditions, but it also carries a number of macroeconomic risks: - 149- =: The dimension of NT2 is large relative to the size of the economy, with total costs representing over 70% of 1995 GDP. By virtue of its size, the project will affect Lao PDR's credit standing and debt service burden. Despite Lao PDR's impressive growth performance in recent years, the country's weak economic position would not allow it to meet the contractual obligation of a counter-guarantee, should it be called in the case of a default. => The complex requirements of an economy in transition have tended to overstretch the Government's administrative and budgetary capacity. Substantial revenues from hydropower may exacerbate the country's absorptive capacity constraints and increase the burden of macroeconomic policy management. => The local impact of the project in its vicinity, around Thakhek, is uncertain but is likely to be significant in terms of pressures on prices, urbanization, and migration of domestic and foreign workers. => The macroeconomic impact of the project will depend to a large extent on the Government's capacity for prudent fiscal policy (i.e., how will the Government spend substantial future revenues from the budget?), its exchange rate management (i.e., how can possible pressures on the real exchange rate [Dutch Disease] be alleviated?), and its ability to absorb the impact on the balance ofpayments (i.e., how can the Government ensure that revenues translate into effective development investment instead of increased imports?). 17. As a latecomer, Lao PDR is in a favorable position to learn how the impact of large dams on domestic markets for labor and goods and on macroeconomic stability was managed in other countries. For instance, the experience of Indonesia shows that the potentially disruptive effect of large windfall gains from resource-based revenues can be partially sterilized if resources are prudently invested overseas and the income stream is assigned as budgetary support. E. FINANCING HYDROPOWER DEVELOPMENT 18. Financing Strategies. The Government strategy of seeking direct private financing is to be commended. Given the scale of investments and risks of hydropower development, a major public investment is out of the question. But what should the Government's considerations be in deciding to how to participate in private schemes? To what extent should it opt for an equity share in the project so as to participate in revenues from dividends? The best way of financing hydropower should be assessed on a project to project basis, which will usually involve a combination of debt, equity, guarantees, logging concessions, etc. In all cases, the Government must balance the need to attract investors with adequate returns with the objective of maximizing public revenues and safeguarding sustainable development. Ideally, the Government should be able to participate with equity in the technically less risky projects. However, it needs to explore innovative ways for this equity contribution. For instance, following the example of other countries, the Government may want to negotiate for an equity share based on the country's contribution in kind (water, timber, land, and other natural resources), rather than borrowing for its share in the project. Another interesting option is through investment in the project's environmental component, which is likely to be significant in size and importance in almost all hydropower projects. This usually requires investment in environmental impact mitigation, resource management, etc. The environment appears to be a natural area for government investment, as well as a valid investment for consideration as an equity contribution. Further, it is likely that these investments could be financed by multilateral sources. - 150- 19. Financial Status of EdL. Over the medium term, the financial performance of EdL is likely to be unsatisfactory (Table 12.1). This is directly related to EdL's commercial borrowing program to finance the Government's equity contribution in forthcoming joint venture power companies. According to ADB projections, total debt service is likely to increase fivefold by the year 2000, to US$64 million from US$12.5 million in 1996. In fact, the debt service ratio might prove to be much higher, since the assumed 200% tariff increase by the year 2000 might not be realistic. This increased debt servicing load would lower the self-financing ratio from 21% in 1994 to -46% in 2000. The debt-equity ratio of EdL is forecast to decline from a conservative 57:43 in 1994 to a highly leveraged position of 80:20 by 2000. Dividends are normally paid from the net profit of a company. This is not the case with EdL, where dividends are calculated as a percentage of government equity and paid annually without regard for the net profit earned. To participate in private sector initiatives, a comprehensive financial plan will be needed to allow EdL to meet its self- financing requirements. The Government could either provide EdL with the equity equivalent to the loan amount for the Nam Theun 2 project, which would lower debt servicing requirements, or could lower EdL's corporate tax for the seven-year term of the commercial debt and allow EdL to retain the funds. TABLE 12.1: EDL - FINANCIAL STATEMENTS AND PROJECTIONS (US$ MILLION, UNLESS OTHERWISE INDICATED) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 stimated Total Power Sold (MWh) 797 16 705 860 1,135 980 1,125 1,128 1,132 1,301 1,647 Average Revenue (US$/kWh) 0.027 0.029 0.035 0.034 0.033 0.032 0.039 0.039 0.041 0.047 0.047 Operating Revenue 21.9 24.0 25.0 30.1 28.6 32.7 44.9 46.1 57.8 82.5 99.0 of which Income from Electricity Sales 21.6 23.4 24.6 28.9 37.2 31.2 43.4 44.5 45.9 60.9 76.9 Income from Investment Theun Himboun 4.3 13.6 13.8 Houay Ho 0.0 0.0 Nam Theun 2 0.0 0.0 OperatingExpenses 13.6 14.4 11.0 13.5 17.8 20.5 23.8 27.1 31.5 36.6 43.4 Operating Surplus 8.3 9.6 14.0 16.6 20.8 12.2 21.2 19.1 26.3 46.0 55.6 Total Remittances to Govermnent 4.9 5 6 11.7 11.8 5.7 10.7 9.9 11.4 15.7 18.5 Total Remittances incl. Interest 31.4 19.7 8.5 13.6 29.4 40.5 102.2 150.2 199.3 107.1 18.5 Self-Financing Ratio 30.7% 68.0% 28.3% 40.4% 20.7% 7.2% 8.6% 4.9% -0.3% -0.4% -45.5% Source: Asian Development Bank. 20. Proposed Hydropower Projects. It is difficult to predict how many of the two dozen proposed projects will ever be implemented. Given the complexity of hydropower development, the consensus seems to be that not more than six to eight of all projects are likely to be commissioned in the next decade. Electricity exports are likely to double by the late 1990s, when the Theun Himboun and the Houay Ho projects are completed and will provide a combined 350 MW of secondary energy sale to EGAT. Nam Theun 2 is being discussed as the largest of several possible alternatives for supplying the bulk of the 3,000 MW power export volume agreed with Thailand. Among the other proposed projects, Nam Mang 3, Nam Leuk, and an extension of Nam Ngum I may be the most likely to be carried out within the next decade (see Table 12.2). Projects with scheduled completion dates include Theun Himboun (by 1998), Houay Ho (by 1999), Nam Leuk (by 1999), the Nam Ngum extension (2000), and Nam Theun 2 (originally 2000, delayed until 2002). 21. The Theun Himboun project is presently under construction and has a capacity of 210 MW and an annual energy generation of about 1,350 GWh. The developer, operating under a 30-year license, is the Theun Himboun Power Company (THPC), owned by EdL (60%), MDX Ltd. of - 151 - Thailand (20%), and Nordic Hydropower ASB of Sweden (20%). Unless extended by mutual agreement, the project will be transferred free of charge to the Government at the end of the license period. A power purchase agreement (PPA) has been signed with EGAT that provides for time-of- day pricing and would average initially about 4 cents/kWh with provision for escalation. ADB is financing part of EdL's participation. Houay Ho is a Korean concession for 136 MW which will come on line in 1999. TABLE 12.2: MOST PROMISING HYDROPROJECTS IN LAO PDR Project Name Province Installed Developer Costs Status Capacity (million (MW) US$) 1. Nam Theun 2 Khammuane 681 PDG 1,200 under negotiation 2. Nam Theun Himboun Khammuane 210 ADB 280 under construction 3. Houay Ho Attapeu 136 Korean 192 under construction 4. Nam Ngum I extension Vientiane 60 World Bank approved 5. Nam Leuk Vientiane 60 ADB/OECF approved 6. Nam Mang 3 Borikhamxay 50 EDL approved 7. Nam Ngiep Borikhamxay 440 773 approved 8. Xe Kaman Attapeu 255 Australian 520 approved 22. Nam Theun 2, a much larger project with an installed capacity of 681 MW and an annual energy generation of about 5,000 GWh, is at an advanced stage of planning and design. In 1993 a group of developers consisting of Transfield (Australia), Electricitd de France, Italian-Thai Development PLC, Jasmine International, and Phatra Thanakit PLC, known as the Project Development Group (PDG), was given a mandate by the Government to develop the project on a BOT basis. The PDG together with Government agencies, forms the Nam Theun 2 Electricity Company (NTEC), which has made considerable progress in site investigations, designs, and environmental studies. 23. The total estimated cost of the NT2 project is US$1.2 billion, substantially more than the financing capabilities of the country, which had a GDP of about US$ 1.5 billion in 1995. The current financing plan (Table 12.3) presumes 30% equity and 70% debt. Financing for the Government's equity contribution in the order of US$89 million needs to be determined. Since private sector participation in this high-cost project is contingent on addressing the perceived risk of investment in Lao PDR, the Government has requested World Bank assistance for an IDA Social and Environmental Project and an IBRD "enclave" partial risk guarantee. The partial risk guarantee -- estimated at about US$94 million -- would cover instances of default resulting from sovereign risk and political force majeure. Several studies (on project alternatives, economic impact, and environmental assessment) are currently under way which will enable the World Bank to reach a decision on the extent of its involvement. - 152- TABLE 12.3: PROPOSED FINANCING STRUCTURE FOR THE NAM THEUN 2 HYDROELECTRIC PROJECT Sources US$ million % of Equity % of Total Financing Equity Government of Lao PDR 89 25 7.5 PDG 267 75 22.5 Total Equity 356 100 30.0 Debt % of Debt Export Credit Agencies 250 30 21.1 Thai Commercial Banks 387 46 32.6 International Commercial Banks/ IFC 194 24 16.3 Total Debt 831 100 70.0 Total Financing 1,187 100.0 F. REVENUES FROM POWER EXPORTS 24. Expected Revenues. In the longer term, power exports could be a major source of revenues, but their timing and magnitude are highly uncertain. EdL may be the recipient of projected net revenues increasing from the order of US$21 million at present to around US$55 million by 2000 (Table 12.1). However, revenue projections are notoriously unreliable even over the medium term and are therefore not presented here in detail. Much depends on when Nam Theun 2 and other projects now under investigation will come on line. A number of other private power projects are being investigated by developers, but to date none has reached the point at which it might generate revenue in the next five years. In terms of foreign exchange inflows to Lao PDR, the Nam Ngum, Xeset, and Nam Theun Himboun projects are the only ones likely to be contributors in the next five years, because neither NT2 nor other private projects are likely to come on line within that time frame. 25. Expected Revenues from NT2. Nam Theun 2 is projected to generate a substantial flow to the Government over the concession period from royalty fees, resource taxes, and dividends. Since financing arrangements are still subject to discussion and the beginning of operation has been delayed to at least the year 2002, alternative projections vary considerably. The formula for calculating the sales price in the recently expired original PPA would have resulted in a price of 5.4 U.S. cents per kWh by the time the project comes on line. However, the price is likely to be renegotiated in the context of a new PPA. According to preliminary estimates based on the original PPA, the proceeds from royalties, taxes, and dividends would start low in the first years of operation, from US$13 million when the project goes on line rising to about US$50 million after 10 years, rising to US$100 million per year (in constant 1995 U.S. dollars) at the end of the concession period. After 25 years, the entire project would revert to Lao PDR and would generate an annual gross revenue stream of around US$250 million (in 1995 prices) for the rest of the project's design life of 50 years. According to these estimates, revenues from the project were estimated to amount to about 25% of the Government's budget and would have to cover O&M and rehabilitation. 26. Public Finances and Hydropower Revenues. As far as public finances are concerned, the key concern will be to determine with a reasonable degree of certainty how much money will - 153- actually be accruing to the Government. To ensure that the national budget -- as opposed to the overall project -- will benefit from future hydropower development, the Government must carefully consider specific financing arrangements which depend on a number of key factors: * Price. The Government's bargaining position vis-a-vis EGAT as a monopsony purchaser of power supplies from Lao PDR is relatively weak. Will this bargaining power increase or decline as Lao power capacity develops? * Time Profile of Payments. The overall projected revenues (through royalties, dividends, or taxes) may look substantial over the 25-year lifetime of a project. However, expected revenues will of course be significantly less in net present value terms if they are backloaded. * Government Equity Share. If the Govemment participates as a shareholder, what form does its equity contribution take? * Other Financial Obligations. To what extent does the Government have to incur any additional costs that are not directly covered by the project, such as additional roads or other infrastructure, services for local personnel, training, or environmental degradation? 27. Use of Revenues. The key mechanism through which the macroeconomic effects of hydropower projects such as NT2 will be transmitted to the rest of the economy is likely to be the Government budget -- which raises the question of how the Government spends the revenues. To justify large-scale hydropower development, a critical issue is whether the Government will be able to direct the bulk of incremental revenues into developmental and poverty alleviation expenditures. The major challenge for the Government, therefore, will be to prioritize its tasks. If the Government relinquishes the duty to run enterprises to the private sector, reduces its role in the financial system to regulation, and enhances the efficiency of public resource management, increased hydropower revenues will enable it to focus on strengthening physical infrastructure and human capital to support private sector growth, alleviating poverty, and assisting those groups that have been excluded from the benefits of economic growth. G. CONCLUSIONS: NEED FOR A STRATEGY 28. Hydropower Development Strategy. There is a need to set out a consistent hydropower strategy which would clearly establish which of the proposed hydropower schemes are economically and financially justified; what future revenues in the form of royalties, dividends, and taxes would accrue to the budget; what financing arrangements would ensure a fair share of the revenues for the Government; how the potential adverse social, economic, and environmental consequences of different realistic alternatives could be mitigated; and what financing needs would arise from project-related financial obligations. Such a strategy for hydropower development should encompass the following key actions: > Clarify Financing Arrangements. The Government should be encouraged to participate with equity in less risky projects. Guarantees should be sought in key projects -- such as Nam Theun - 154- -- where a successful development could influence the overall hydro program. Otherwise, Government participation should be limited to its condition of resource owner, charging royalties. r Begin Strategic Planning. The Govemment should undertake a comprehensive planning exercise aimed at identifying the optimal hydro development for the local and export power markets. This study, proposed as a second stage of the Nam Theun Alternatives Study, should be useful in identifying domestic needs and further export opportunities, and in addressing investment requirements as well as policy and institutional issues. It should also warn of signing legal agreements with developers in excess of purchasing power commitments plus domestic needs. Hydropower planning should also be clearly informed by the international experience with similar ventures in other countries. => Introduce Competition. To obtain better conditions on royalties, financing arrangements, and prices, the Government should devise a strategy for introducing a more formal and transparent competition in the selection of developers and projects. There could be competition either in the selection of developers for a specific project site or between project developers. The latter case may appear more attractive, since the initial risks of project preparation would be borne by the developer. The interest shown by the international community in Lao PDR's hydropower potential is an indication that the market conditions are appropriate for such competition. - 155 - CHAPTER 13: TIMBER AND TIMBER ROYALTIES - ISSUES AND PERSPECTIVES A. REFORMS IN THE TIMBER TAXATION SYSTEM 1. Timber royalties have become a significant part of the Government's fiscal revenues. Revenue from timber exports is the largest single source of export earnings and tax revenue. Despite falling log and timber prices in the world markets and problems with tax collection, timber royalties made up over 16% of total fiscal revenue in 1994/95. Exports of timber and wood products in 1995 amounted to US$88 million, accounting for over a quarter of total export receipts. 2. Logging System before 1993/94. The logging industry, timber processing, and revenue from timber exports have all undergone drastic changes since the early 1990s.1 Prior to 1988, the Government had a policy of provincial self-sufficiency, whereby provincial authorities had complete control of logging and external trade. Following reports of overharvesting in several provinces and leakages in timber revenue collection at the central level, a policy of centralization was adopted. Starting in 1989, a new tariff code was applied to timber products, which implied a significant simplification of the previous export tax structure on timber products. In particular, only two rates were applied -- 3% on plywood exports and 30% on logs and sawn wood (mostly lumber and parquet). Logging permits -- including logging quotas-- were allocated by Provincial Committees to about 25 companies. Nine of these companies were state-owned, covering the entire country, and others operated at the provincial level. Three taxes were applied: (i) a specific rate stumpage fee (US$5-14) on all cut logs; (ii) a specific rate resource tax (US$20-400); and (iii) ad valorem export duties on semi-finished and finished wood exports. This pre-1993/94 system was plagued by a number of inefficiencies, including: (i) delays in payments to the logging companies, which impaired their profitability; (ii) arbitrary timber pricing policies, which were unresponsive to market conditions; (iii) poor administration and the absence of effective controls, which resulted in significant wastage and revenue leakages; and (iv) inefficient provincial mills that were being kept in the market by quota allocations at fixed prices. 3. Logging System Reforms after 1994. To overcome these problems the pricing policy was overhauled in 1993. A system of minimum prices, or fixed timber royalties, was introduced to establish a competitive log market and ensure that the Government maximized rent from this sector. The minimum price system replaced the stumpage fee and resource tax and incorporated the export tax on timber products. Logging companies were still subject to taxes (such as turnover and profit tax) applied to all other businesses. In principle, this system should have allowed provincial authorities to revise the centrally established minimum prices upward to reflect market developments and ensure predictable receipts to the Government. 4. Logging SOEs. However, the improvement in timber pricing policies did not result in the expected increase in timber revenues. With Prime Ministerial Decree No. 16, dated October 7, 1994, control of logging and log trading was taken away from private companies and given to three state- owned enterprises (SOEs) (AFD, Phoudoi, and DAFI) which were allowed to operate as a monopoly See also Chapter 3 of these Background Papers for a discussion of the forestry sector. For an overview, see Gandhi et al. (1995) "Lao PDR: Scope for Tax Reform" (IMF). - 156- in different parts of the country. AFD was assigned the seven northern provinces, Phoudoi (or BPKP, the largest), the two central provinces; and DAFI, the five southern provinces. These companies were initially managed by the Ministry of Defense, but the Ministry of Finance was formally put in charge in July 1995. In October 1995 the three enterprises were also given oversight of special development areas, such as hydropower and road projects. 5. Pricing. Prices of logs harvested by the SOEs -- for exports as logs or as processed wood -- are the result of negotiations between foreign buyers (primarily Thai merchants) and the SOEs. The SOEs pay the minimum price to the Government and retain the profits (the sale price minus their costs, including the royalties). The timber royalties, as minimum prices to be paid per cubic meter of harvested logs, are differentiated by timber categories (Table 13.1). The royalties, expressed in U.S. dollars, are based on volume and quality of wood instead of having ad valorem fees based on erratic provincial market sales prices, which tends to stabilize government receipts. TABLE 13.1: TIMBER LOGGING QUOTAS AND TIMBER ROYALTIES, 1994/95 (US$) Category of Wood Quota Royalty Quota Royalty Annual Projected (cubic (October 1993) (cubic (February 1995) Revenue meters) meters) Prohibited 21,300 225 11,300 180 2,034,000 Controlled 103,700 80 103,700 50 5,185,000 Branches 24,000 15 34,000 15 510,000 Roots 14,000 50 19,000 50 950,000 Nam Ngum 70,000 27 70,000 27 1,890,000 Nam Theun 2 300,000 70 300,000 65 19,500,000 Houay Ho 62,000 65 62,000 60 3,720,000 Cedar Concession 15,000 960 10,000 650 6,500,000 Total 610,000 610,000 40,289,000 Source: Ministry ofAgriculture, Department of Forestry. B. NEED FOR REFORM IN TIMBER TAXATION 6. The present system shows a number of weaknesses which call for reform to ensure sustainability in the long run. * Royalty collection has been encountering problems. Despite the changes in the system, logging enterprises have occasionally complained that minimum prices are too high given current world prices. The minimum prices were last adjusted in February 1995. Systematic improvements in tax collection, however, have not prevented substantial shortfalls in the collection of timber royalties in the past (Table 13.2). Shortfalls in timber royalties were related to the logging of lower quality wood, a drop in world prices, and lagging payments. - 157- TABLE 13.2: PLANNED AND ACTUAL TIMBER REVENUE, 1991-95 (BILLION KIP) 1991 1992 * 1992/93 1993/94 1994/95 1995/96 Tax Revenue Plan 74.9 70.9 82.3 103.6 180.2 Actual 54.3 63.5 61.2 85.7 99.8 175.9 Actual/ Plan 84.8% 86.3% 104.1% 96.3% 97.7% Non-tax Revenue ** Plan 18.5 20.0 28.3 32.9 34.6 Actual 20.3 26.9 25.9 29.1 36.5 41.4 Actual/Plan 145.4% 129.5% 102.8% 110.9% 119.5% Timber Royalties Plan 24.2 30.0 28.1 33.0 Actual 24.6 21.0 27.0 34.6 Actual/ Plan 101.7% 70.0% 96.1% 104.8% Total Plan 115.1 140.6 164.6 229.8 Actual 111.7 135.8 163.3 231.7 Actual/ Plan 97.0% 96.6% 99.2% 100.9% Nine months from January to September 1992. * For 1991 - 1993/94 excludes sale of assets; timber export revenue is included Source: Ministry of Finance. The performance of the three logging SOEs has been poor, both operationally and financially. Administrative problems in the collection process stem largely from inadequate operational supervision. There are also indications that actual logging exceeds the government quotas. Particularly in 1992/93, harvesting was well below the plan, payments arrears increased, and timber royalty payments were delayed. The State Assets Department of the Ministry of Finance does not have the capacity to effectively control the financial performance of theses SOEs, and at the same time management is hindered by the bureaucratic influence of too many public agencies.2 * The Government has not adjusted its minimum pricing policy in accordance with international market developments. The accumulation of large unsold timber stocks suggests that prices may have been too high for certain categories. Overpricing and excessive logging have resulted in wastage of large amounts of logs, which was estimated at about US$15 million in 1994/95. 7. Indications of the extent of the above problems in the 1994/95 budget are provided in Table 13.3. Actual logging was about 30% higher than the assigned logging quota. While budget revenues fell only slightly short of the budgeted amount, this amount itself was on the low side, since it represented only about 87% of potential royalties based on the assigned quota. However, this is only 2 The companies' directors were appointed by the Ministry of Defense; the annual logging plans are established by the Department of Forestry in the Ministry of Agriculture and ultimately approved by the Prime Minister's Office; the timber royalties are established by the Ministry of Agriculture and the Ministry of Commerce; and company investment plans are approved by an interagency central investment board. - 158- 67% of the potential royalties when actual logging is taken into account. Revenue performance irnproved in 1995/96, with collected timber royalties exceeding the plan. TABLE 13.3: LOGGING AND TIMBER ROYALTIES, 1994/95-1995/96 Budget 1994/95 Budget 1995/96 Original logging quota (cubic meters) 610,000 637,000 Actually removed logs (cubic meters) 791,000* -- Potential royalties (US$ million)** 40.3 40.4 Budgeted revenue from royalties (US$ million) 35.3 35.9 Actual budgetary revenue from royalties (US$ million) 34.9 36.5 * Estimated on the basis of actual removal during calendar years 1994 and 1995. * * Calculated at minimum prices of February 1995, applied to original logging quota. Note: Preliminary figures for 1995/96. Sources: Ministries of Trade, Forestry and Finance, and staffcalculations. 8. Problems with respect to royalty collection emerged again in 1995/96. Revenue from timber royalties during the first four months amounted only to 10% of the total originally planned for the year. However, renewed efforts since January 1996 to improve royalty collection and recovery of arrears from 1994/95 enabled the Government to increase revenue from this source to more than 34 billion kip during the 1995/96 fiscal year. C. ELEMENTS OF FUTURE REFORM IN TIMBER TAXATION 9. A key objective of timber taxation reform should be to maximize the appropriate rent for the budget from the use of this natural resource, while limiting logging to sustainable levels based on sound forest management practices (including logging control, reforestation, and environmental protection). This will also require maintaining a stable flow of fiscal revenues to the budget over the medium term and ensuring adequate profit margins for efficient operators in the industry. Over the medium term, a profound improvement in the structure of forestry management will require a decisive shift toward market determination of timber prices, coupled with the substantial disengagement of parastatal companies from logging and processing activities. Such reforms would encompass the following: = Reduce the present reliance on revenue from timber. The outlook for continued emphasis on timber royalties is not encouraging and should be kept in line with the authorities' environmental plans, which call for the conservation of the country's natural resources and a reforestation policy. This would mean that logging should be on a sustained basis at lower levels than in recent years. Annual logging plans should be established in accordance with the expected absorptive capacity of the market and processing facilities. The current levels of cutting -- over 600,000 cubic meters for 1995/96 -- are based on strip-logging of areas to be developed for hydropower projects (especially Nam Theun 2). Once these projects have advanced, the logging potential will decline sharply. Moreover, from a budgetary point of view, over-reliance on one - 159- source of revenue is not a sound practice, particularly if this (natural) source is not under the authorities' control. Introduce a market-based pricing systenL Proposals for further reform have been advanced by both the World Bank and the IMF. The pricing system could be further enhanced by adjusting minimum prices to reflect market conditions more closely. Prices could be determined in a nationwide auction system in which foreign buyers compete for logs with domestic processors. The auction system should be extended to include log allocations to mills, replacing the system of prices and quotas set by the Government. Collect arrears of unpaid timber royalties. This should be based on letters of credit issued by reputable banks. To discourage SOEs from delaying payments of timber royalties in the future, interest costs and penalties should be established on the remaining arrears. =. Monitor the performance of the three state-owned logging companies. In addition, financial and managerial discipline should be established through regular audits and management performance targets. Over the medium term, the three state logging enterprises could be converted into contracting firms with a controlling role to regulate the market, while their logging and timber processing activities would be subcontracted to the private sector. - 161 - PART IV. PUBLIC EXPENDITURE MANAGEMENT CHAPTER 14: STATE-OWNED ENTERPRISES 1. From 1988 to 1995, the state-owned enterprise sector in Lao PDR underwent major changes. A comprehensive privatization program dramatically reduced the number of enterprises owned by the central and local governments. The remaining state-owned enterprises (SOEs) were subject to hard budget constraints and their monopoly privileges were abolished, which significantly improved their financial performance. 2. However, the results of this impressive transformation are fragile because the divestiture process is increasingly called into question in view of the disappointing performance of a number of privatized enterprises; and the marginal profitability of some of the large SOEs, particularly the utilities, may lead to serious financial difficulties, primarily because of their massive investment program. A. THE PRIVATIZATION PROGRAM 3. Lao PDR has gone further than most other transition economies in divesting its SOEs (Boxes 14.1 and 14.2). After limited progress in the first two years following 1988, the privatization program gained momentum in the early 1990s. By 1995, most of the small and medium-size SOEs, and a good percentage of the larger ones, had been privatized. Recently, however, the program of divestiture has been increasingly called into question. The perceived failure of some of privatized enterprises is creating mounting pressure to: (i) slow down or, in some cases, reverse the process of divestiture from productive activities; (ii) protect local industry from the competition of imports and return to a policy of import substitution not grounded on economic analysis; and (iii) reactivate state- owned banking sector assistance to struggling enterprises. Box 14.1: STATE ENTERPRISE REFORM IN VIETNAM To a large extent, the budget constraint of SOEs in Vietnam has been hardened. SOEs have been weaned from direct operating subsidies, denied automatic access to central bank credits, and forced to buy inputs and sell outputs at market prices. The result of this increased autonomy and clear profit motive has been a significant improvement in the profitability and performance of SOEs, but not without painful adjustments, including the liquidation of about 2,000 enterprises and the merger of 3,000 others, resulting in the loss of about 800,000 jobs in state enterprises. By comparison with other transition economies, the contribution of SOEs to aggregate GDP is substantially lower (although industry is still important). What is surprising is that the SOE sector has experienced more rapid growth than the economy as a whole, while the number of SOEs, their employment, and their share in total domestic credit, have declined sharply. Moreover, the SOE sector has registered a steady increase in its net remittances of taxes and transfers to the state budget, which amounted to some 1 1% of GDP-- or about 50% of total budgetary revenues-- in 1993/94. Encouraging state-owned firms to become more efficient is a worthy cause, but only if it does not divert attention from the even more important cause of fostering the development of private enterprises in the industrial sector. - 162- 4. However, this shift in policy fails to take account of two observations. First, although some privatized enterprises are performing poorly, the robust growth of new private start-ups in the same sectors indicates the success of economic restructuring. Inefficient and uneconomic former SOEs are being eliminated. Second, the inferior performance of some privatized enterprises largely results from the privatization methods. The process of divestiture has not been transparent and has relied on leases rather than outright sales. This has not been conducive to the transfer of the productive assets to private operators with a stake in the long-term success of the privatized enterprises. Box 14.2. HAVE THE CHINESE SOES BEEN SUCCESSFULLY REFORMED? In terms of the contribution of the SOE sector to macroeconomic stability, the Chinese SOE reforms have not been very successful. Overall profitability of the SOE sector has been declining, and a large proportion of SOEs are still making losses a decade after the start of enterprise reforms. The SOEs may have become technically more efficient, but most of them have also become financially less viable. The easy availability of loans from the banks made it possible to increase labor costs and capital investment at the same time. The financial weakness of SOEs destabilized the economy through two channels. The first was through the state budget. Given the dependence of state revenue on income from the SOEs, the budget deficit widened from 2.1% of GDP in 1981 to 3.4% in 1991. The second channel was through the banking system. The result of the central bank's accommodation of the request for loans to cover losses and to finance investment is that the amount of reserve money growth that is unrelated to deficit financing is substantially greater than the deficit itself. There are some positive signs, however: the corporatization and management of enterprises through price signals generated by stock markets is becoming a widely accepted idea, and more and more state enterprises are being sold to non-state companies. Progress of Divestiture 5. In 1988, as part of its new economic liberalization strategy, the Government announced its commitment to open the economy to the private sector. When the reform began, there were around 600 SOEs, of which approximately 250 were in the industrial and manufacturing sector, accounting for about 10% of GDP at the time. Seventy percent of the SOEs were controlled by provincial authorities and the rest, mainly the larger ones, were directly supervised by central government ministries. Most SOEs were very small, with fewer than 30 employees. Only a few manufacturing plants and utilities were large (over 100 employees), although they were still fairly small by international standards. 6. The first stage in the economic reform of SOEs was to liquidate many of the very small uneconomic SOEs, while making the remaining SOEs more commercial. Employees of the liquidated enterprises either returned to their provincial authorities as civil servants, or moved on to the newly emerging private sector. Almost all the remaining SOEs lost their monopoly privileges Commercial Credit and were exposed for the first time to competition ' 150 from the new domestic private start-ups and from ' 100 imports. While still being retained under state - 500 - ownership, the SOEs are given autonomy and .5 0 directed to become self-financing. Although direct 1991 1992 1993 1994 1995 subsidies were quickly eliminated, a number of Credit to SOEs Total Credit SOEs continued to operate for a few more years under a soft budget constraint, accumulating - 163 - arrears with the state banks. This led to a major banking crisis in the late 1980s which led the Government and the Central Bank to implement much tighter credit policies and practices, particularly vis-a-vis SOEs. With the support of the ADB, a program of recapitalization and restructuring of the state banking sector is under way, including merger, liquidation, and possible privatization of the eight state banks. As a result, a noticeable degree of financial discipline is being restored in the SOE and state-banking sectors. Credit to SOEs now represents a small percentage of the total credit extended to the enterprise sector, and SOEs actually have fewer arrears (4% on average) than private borrowers (8%). 7. In parallel with much harder budget constraints in the SOE sector, the Government embarked on a wide-ranging privatization program in the early 1990s and took measures to open up most sectors of the economy to the private sector. The private sector, particularly new private start- ups, now dominates the insurance, manufacturing, transport, and distribution sectors, and it is strongly represented in banking and construction, where it competes with SOEs. In rare cases, monopoly privileges have been granted in the first phase of privatization to foreign investors buying large SOEs. However, the Government does not intend to renew the two remaining monopolies (Lao Brewery and Lao Soft Drink) when they expire in 1996 and 1997. 8. As a result of privatization and liquidation, the number of central and provincial SOEs decreased from over 600 in 1988 to 91 in 1995. There are now about 60 SOEs in the central government portfolio and 31 under provincial control. Of these, 58 SOEs (27 central and 31 provincial) are targeted for Number of SOEs privatization in 1996/97. This number includes the recently created Lao 500 Telecommunications, the tobacco factory, 400 all the road maintenance units, and all but 300 - 8 remaining commercial enterprises of 20 production and services. The 31 municipal and provincial SOEs slated for 00 privatization in 1996 comprise all but 5 of the SOEs that are still in public ownership 1990 1995 Expected at the local level. This would leave only 1997 32 "strategic" enterprises wholly state- owned. 9. The 58 SOEs included in the next privatization phase have a book value of over US$70 million (Figure 14.1). Most of these SOEs are small and medium-size, and would be fully privatized. For the time being, the Government intends to retain a majority ownership in the recently created telecommunications company; additional sale of shares is possible in the future. The proposed privatization of 35% of the company could generate over US$25 million in privatization proceeds, or roughly 40% of the total expected privatization revenues for all remaining SOEs. 10. Only about 12 billion kip in proceeds from the sale of SOEs has been included in the budget for fiscal year 1996-97, or less than 20% of the potential proceeds from the competitive sale of the 58 SOEs targeted for privatization. This estimate of privatization revenue takes into account a phased payment schedule for the shares in the telecommunications company, by far the largest source of revenue. But it also clearly indicates that the privatization process was already expected to - 164- FIGURE 14.1: VALUE OF SOES TO BE PRIVATIZED 25 25 20 -~2 0~~~~~~~~~~~~~~~~~~~~~~~2 2..~~~~~~~~~~~~~~~~~~....... :0 1, X S . 5~~~~~~~~~~~~~~~~~~~~~~~~~~ 0 ~~~~~~~~~~~~~~~~~~~~~~0 .01 $O.l-$O.5 $0.5-S1.0 S1.0-$3.0 $3.0-$l0.0 >Sl0.0 Company Size Range (US miillion) Total potential proceeds - Number of SOEs slow down significantly compared with the initial target set for 1996, when the budget was prepared in late 1995. 11. Methods of Privatization. The privatization process at the central level has been conducted less than openly. The Government usually relied on direct negotiations, rather than on open competitive bidding, to sell a majority ownership in the joint ventures and to define the conditions of the lease agreements. 12. For the larger, modern enterprises, the Government favored retaining a minority ownership in joint ventures with private, mostly foreign, investors. Most other SOEs at the central and provincial levels were leased, sometimes to outside entrepreneurs, often to the existing managers. The reliance on leases rather than outright sales reflected the reluctance of the Government, in the initial stages of the opening up of the economy, to relinquish control over state assets. It was also used to quickly transfer to private operators the responsibility for managing enterprises that the Government was reluctant to sell at a market price because, in most cases, they would have been significantly below book value. Indeed, except for a few modern entities in selected sectors (mainly beverages, tourism, and insurance), the Lao SOEs were usually far too small to be of interest to foreign investors. As for domestic entrepreneurs, they appeared to be more interested in creating in new ventures than in buying existing SOEs with a less than ideal location, dilapidated buildings and equipment, and a large and sometimes less than fully productive work force, once the Government opened up most sectors to competition. 13. The reassessment of the role of the Government in the economy, rather than the desire to generate maximum budget revenue, has been the driving force behind the Lao privatization program. In Lao PDR, as in many countries in transition, the methods used to estimate the value of state property usually lead to a gross over-estimation of the price at which the property should be offered for sale. While the privatization agencies rely on book valuation of the assets to set a floor price, - 165 - investors base their purchase offer on an estimate of the return they can make, not only on the initial investment required to purchase the company from the state but also on the new capital expenditures that are often required to operate the company profitably. Alternatively, they may base their purchase offer on the net worth of the company, which usually differs significantly from the value estimated by the privatization agencies, as SOEs often have a significant burden of liabilities. The sale of state property without reference to a floor price (i.e., for the highest bid received in a competitive process) is often a politically difficult decision. However, it is fully consistent with government policy to eliminate direct state involvement in commercial activities. 14. Recommendations. The following recommendations should guide the Government in the course of the privatization process: = Do not backtrack on the privatization progress. The Government should recognize that eliminating value-reducing activities is not a disappointing result of divestiture but a necessary economic process of creative destruction. Lao PDR cannot afford to divert scarce economic resources to artificially provide a life-support system for non-viable enterprises. Instead, the Government should focus its efforts and resources on improving the conditions for a sustainable growth of private activities. Replace leases with outright sales. The Government should eliminate leasing. In the future, SOEs should be privatized through outright sales to the highest bidder in a fully competitive process. Experience demonstrates that private operators that have invested their own financial resources as owners are far more likely to take an active interest in the long-term success of their property than operators such as lessees or management contractors, that simply use the assets of others at no financial risk to themselves (see also Box 14.3). Box 14.3. THE TRANSFER oF SOES To LoCAL COMMUNITIES - AN ALTERNATIVE TO LEASING? In China, SOEs and community-owned industrial enterprises (Township and Village Enterprises - TVEs) are fundamentally different, even though both are publicly owned in the legal sense and both are subject to government regulations. The first difference lies in the direct linkage between the local people's working efforts and their economic benefits, which not only reduces the cost of supervision but also improves the local owners' incentives to monitor management, exert pressures on managers to improve the business, and complain about managerial appropriation of resources when it occurs excessively. The most important institutional advantage of TVEs over SOEs is that the former cannot borrow forever to cover their losses. Since TVEs are non-state enterprises, the rescue of a bankrupt TVE is not the state's responsibility. SOEs, on the other hand, have repeatedly been able to force the state to print more money to bail them out. TVEs, unlike SOEs, can implement institutional innovations more rapidly, as they do not require central government approval. This freedom has enabled TVEs to move closer to modern international practice in corporate governance. In a nutshell, TVEs, unlike SOEs, live by the market and, contrary to leasing, a whole community has a vested interest in ensuring the long-term success of the venture. - 166- B. IMPROVING THE PERFORMANCE OF THE REMAINING STATE-OWNED ENTERPRISES 15. The SOEs that are currently excluded from the privatization process, far from being a drain on the budget, contribute significantly to government revenue. There is some concern, however, that the list of 32 "strategic" enterprises (see Table 14.1) includes a number of clearly commercial activities. Properly managing or supervising these enterprises imposes a burden that line ministries and the Ministry of Finance might not be adequately equipped to handle. Further, there is a risk that scarce public resources will be channeled to finance future investment programs required to maintain and develop these enterprises. This would jeopardize investments in other sectors, such as the social sectors, that are more clearly the responsibility of the Govemment, and for which no alternative (i.e., private) source of financing is readily available. TABLE 14.1: STATE-OWNED ENTERPRISES CLASSIFIED AS "STRATEGIC" Name of SOE Number of Employees Capital Utilities I Electricite du Laos 1,697 4,989,000 2 Nam Papa (Water) 340 1,833,640 3 Lao Post Enterprise 135 Quasi-Government Department 4 Lao National Tourism 112 279,602 5 Lao Employment Agency Commercial Enterprises 6 Lao Fuel Co. 470 10,502,601 7 National Printing Shop 138 791,664 8 Edition & Printing Enterprise (textbook) 45 453,822 9 Educational Material Production 10 Pharmaceutical Factory No. 2 100 811,430 11 Pharmaceutical Factory No. 3 112 1,068,106 12 Lao Import-Export Trading Co. 13 State Shop for Foreign Missions 14 Technical Material Supply Co. 15 Supply & Spare Parts Co. 36,851 16 Highland Development Co. 12,928,647 17 Development of Agro-Forestry Industry 200 1,362,558 18 Agro-Forestry & Services Co. 65 834,960 Banks 19-26 8 state banks (to be merged into 6) Miscellaneous 27 National Lottery Provincial Enterprises 28-32 5 printing, irrigation, and tourism - 167- 16. In addition, two major utilities have embarked on ambitious capital investment programs. It will require significant tariff increases and cost reduction to ensure that they continue operating without direct or indirect operating subsidies and generate enough cash flow to service their debt. 17. Reassessing the Role of the State in Productive Activities. The list of strategic enterprises established in July 1995 initially comprised 38 SOEs belonging to the central govemment, including the 8 state-owned banks, and 6 small enterprises belonging to the municipal and provincial governments. Between July and December 1995, the initial list of strategic enterprises was reduced to 32, following the decision to liquidate some and transfer others to the privatization program. 18. The remaining SOEs include the major utilities (electricity, water, and post), some SOEs that would more properly be classified as extensions of Government departments (such as the Employment Agency), the state-owned banks, and commercial enterprises. 19. All the commercial SOEs Box 14.4. LAo STATE FUEL CO. compete with imports and domestic private enterprises (or majority private Lao State Fuel Co. (LSF) lost its monopoly on fuel import and joint ventures) and operate at a profit distribution in 1989. Since then, CalTex, Shell, and several without direct or indirect subsidies. Thai companies have entered the retail market. In 1995, LSF The Government's objective in had an estimated 50% market share to the second largest keeping these enterprises in the public player, Shell, at 35%. sector is to prevent the creation of The Government uses LSF to avoid the creation of a private cartels or monopolies in such cartel and price fixing among private fuel distributors. LSF important sectors (see Box 14.4). In has a mandate or social obligation from the Government to particular, the Government considers provide fuel to all parts of the country at a reasonable and that the generic drug manufacturers steady price that enables it to cover its operating expenses. and textbook printing enterprises Profitability has been marginal since the devaluation of the support the Government's policy of Lao currency, and the operating profit probably does not providing affordable basic health and adequately cover for depreciation or allowance for capital education to the population. The investment. Government justifies public ownership The company has a fleet of 150 tanker trucks, as of logging enterprises by the need to well as fuel barges, and bulk storage and retail stations. An ensure the sustainable harvesting of emergency stock, about 10% of consumption, is maintained in natural resources and to control the each province. The LSF's contribution to government substantial export revenues they revenue, including import duty, consumption/excise tax, and generate (see Chapter 13). turnover and profit tax, amounted to 2.4 billion kip in 1992 and 5.3 billion in 1995. 20. While these are legitimate concerns, they could best be addressed through having an effective regulatory framework -- including legislation and enforcement capabilities -- and by implementing competitive procurement practices for state needs, rather than by maintaining state ownership. Full or partial divestiture would help mobilize private sources of capital to finance the expansion and modernization of the enterprises, and would free up public resources to fund activities and programs for which no alternative source of financing is readily available. It would also redirect ministries toward developing and enforcing a regulatory environment conducive to the sustainable growth of the concerned sectors, which is the proper role of government. 21. Government-SOE Financial Relations. The SOEs -- the commercial enterprises and the utilities -- in Lao PDR are significant sources of revenue for the national budget. They pay a 5% turnover tax, a 45% profit tax, and an "amortization tax" that is specific to SOEs (to contribute to - 168 - past investments by the state). In addition, the investment program of large state-owned utilities generates a significant amount of revenue for the budget through the on-lending by the Government to SOEs of the multilateral and bilateral credits and grants it receives at much higher rates and over shorter servicing periods. For instance, the IDA and ADB credits extended to the Government at less than 1% interest over 40 years, including 15 years of grace, to finance the investment programs of the electricity or water companies, are on-lent to the enterprises at between 6 and 8% over a period of 20 to 25 years, with only 5 years of grace. In addition, some grants extended to the Government by bilateral donors are on-lent to the utilities at interest rates of 6 to 8%. However, in the case of the water utility Nam Papa, the Government increasingly allows the company to receive all or part of the financing in the form of grants -- as an investment subsidy (see Box 14.5). Utilities in most other similar countries are huge financial burdens on the budget. However, the transfers from Lao SOEs paid to the Treasury are significant: more than 8 billion kip in taxes and 6 billion kip in repayment of interest and principal for the electricity company EdL in 1995, and around 200 million kip in taxes and 270 million kip in debt service for Nam Papa in the same year. The concessionary loans and grants can be considered windfall gains for the Treasury. The Government should be concerned with the opportunity cost of using such resources to finance capital expenditures that could be funded without public resources if the sectors were opened up to private participation. 22. As a consequence of the revenue they generate for the budget, there has been little incentive to date for the Government to strengthen its supervision over the SOEs. However, the financial situation of some SOEs, including the two major utilities (electricity and water), appears fragile. While all SOEs are making a profit on current operations, their margin of profitability is very thin; it does not allow them to generate anything approaching the amount of self-financing required to fund their investment program. 23. Smaller SOEs, such as the national printing shop, are receiving long-term loans for capital investment from state-owned banks and, in some cases, directly from the central bank (as was the case with the national printing shop). While the amounts to date are fairly small (a few hundred thousand dollars) and the enterprises are servicing their debt normally, it is not clear that the lending decisions or the lending terms and conditions have been based entirely on market principles. - 169- 24. Larger SOEs, such as the main utilities, benefit from donor-financed Box 14.5: NAM PAPA projects, but their investment program Nam Papa, the water utility, has a turnover of 3 billion kip of over the past four to five years has been which 80% is earned in Vientiane. It generates a small profit large. There is evidence that their gross of 3% of global revenue, with Vientiane's operations cross- margin will have to be improved to subsidizing the losses in most other provinces. The finance the local contribution to these enterprise does not receive any direct operating subsidy from investment projects (required by most Government, and paid around 200 million kip in taxes and donor agencies), and to service an 264 million kip in interest to the budget in 1995. donoreasiglenes) and deto servicetanc increasing level of debt. For instance, Nam Papa has recently embarked on an ambitious the debt service of EdL is expected to investment program to improve and extend water services in increase almost fivefold in the next five the country. This program, a total of US$76 million to date, years (see also Chapter 12). In at least is financed by a number of bilateral and multilateral donors. one instance, a major SOE could not Forty-three percent is received in the form of grants, while generate the counterpart funds required 57% is on-lent by the Government to Nam Papa at an by donors from its own revenue and had average rate of 6% over 15 years, with a 5 year grace period. to borrow from a state commercial bank, The local contribution of Nam Papa to this investment with the backing of the Central Bank program averages only 8% of the total cost. However, Nam and full government guarantee. The Papa's gross margin is proving insufficient to generate even utilities will achieve the necessary level this modest percentage of self-financing. In 1994, it had to borrow 1.3 billion kip from a state-owned bank to cover its of improvement to their operating contribution to a new donor-financed investment project. margin only through a combination of There is clearly some element of indirect subsidization in significant tariff increases and the this loan which was extended for an unusually long period of reduction of their operating costs. Tariff time (8 years with 3 years' grace, while the commercial increases, although inevitable in the banking sector hardly ever lends for more than I year), at a medium term, may prove socially and low rate (8% at a time when short-term rates averaged 18%), politically difficult. Although tariffs for and with full government guarantee. Improving the such essential services as water and profitability of Nam Papa's operations is essential in the face electricity are not high by regional of the company's rapidly rising debt service. standards, they nevertheless represent a significant burden on poor households. If the Government is reluctant to authorize the necessary tariff increases, the profitability of the utilities may well change dramatically, leading to a need for direct or indirect subsidization. 25. The prospect that significant tariff increases will be necessary to safeguard the operating profitability of the major utilities makes it all the more urgent to start implementing a convincing cost reduction program. To date, the Ministry of Finance has not been very proactive in supervising the results and performance of the SOEs and, in particular, the large utilities. Clearly, this is because the SOEs have been net contributors to the budget revenues. However, the situation may change. The Government would be well advised to implement effective governance mechanisms to increase the SOEs' financial performance. Appointing an effective Board of Directors and introducing performance targets could be the foundation of more balanced and forward-looking relations between the state as shareholder, represented by the Ministry of Finance and the line ministries, and the management of the SOEs. 26. Recommendations. The Government should therefore maintain sufficient managerial autonomy for the SOEs, but should carefully monitor their financial perfomance so that it can intervene in a timely manner. Even if the situation looks acceptable at present, there may nevertheless be nasty surprises. The Government must therefore remain committed to implementing a program of commercialization in five strategic SOEs, including the three utilities, in the context of - 170- the policy reform program supported by the World Bank's SAC III. Establishing institutional mechanisms is critical to properly monitoring the financial evolution of the utilities. The Government should swiftly take the following steps: => Appoint the members of the Board of Directors of the enterprises targeted for commercialization = Strengthen the capacity of the Ministry of Finance to move to active portfolio management from its present focus on control over the contribution of SOEs to the state budget, introducing technical and financial targets against which SOE performance and management will be measured > Over the medium term, divest the remaining comnmercially oriented SOEs and devolve them to the private sector. - 171 - CHAPTER 15: CIVIL SERVICE REFORM A. REFORM PROGRESS TO DATE 1. In 1986 the Government of Lao PDR embarked on a series of economic reforms designed to move the economy from central planning toward a market system under the New Economic Mechanism (NEM). Because it represented such a radical shift (for Lao PDR), the NEM necessitated reforms in the administrative apparatus. Administrative reforms have thus followed economic reforms closely. These have consisted of: (i) organizational change, (ii) reforms of the personnel management system governing the civil service, (iii) compensation, and (iv) retrenchment. Civil service reform in Lao PDR has generally meant reforms of the personnel management system, although more accurately it also encompasses organizational change. 2. (i) Organizational Change. The initial efforts at civil service reform involved organizational changes that were necessitated by economic reforms. The Foreign Investment Management Committee was established in 1989 to implement the foreign investment code. Jurisdiction over monetary policy was granted to the Central Bank in 1990. In the same year, customs and tax administration were separated. In 1991, a Privatization Unit was established in the then Ministry of Economy, Planning, and Finance. This ministry was split up later in 1993 into the Ministry of Finance and the Committee for Planning and Cooperation. In that same year, the Agricultural Promotion Bank and the Ministry of Labor and Social Welfare were also created, and in the previous year a National Treasury was established, primarily to handle all payments from the national budget. 3. (ii) Personnel Management Reform. Personnel management reform began in 1991 when the National School of Administration and Management (ENAG) was established (see Box 15.1). ENAG was created to address the pressing need to train civil servants and private sector managers in the skills needed to manage a market-based economy. 4. In December 1992, the Prime Minister issued a decree (no. 98/PM) establishing the Department for Public Administration and the Civil Service (DARCS) within the Office of the Prime Minister. The Department, which currently has only 25 full-time staff members, is responsible for a wide range of tasks. Since its inception, it has worked with ministries to reorganize by preparing position descriptions, enumerating the distribution and profile of central and provincial civil servants, replacing some retrenched civil servants with short-term contract employees, and directing a three-year Public Administration Reform Program funded by the UNDP (see Box 15.2). It should be noted that DARCS was an integral part of the Party apparatus and was set up separately by this decree partly in response to a conditionality of the IMF. - 172- Box 15.1: NATIONAL SCHOOL OF ADMINISTRATION AND MANAGEMENT (ENAG/NOSPA) Roughly speaking, ENAG is like a school of public administration and a business school rolled into one. Since 1995, ENAG has been formally merged with the Party School as the "National Institute for Politics and Administration" (NOSPA). It offers three types of programs: (i) short-term programs for professionals, primarily senior civil servants, who cannot leave their jobs except for very short periods (e.g., a five-day course in basic macroeconomics or small business management); (ii) a ten-month Certificate Program to retrain civil servants in the knowledge and skills needed to understand and service a market economy; and (iii) business administration (as opposed to public administration). The Diploma Program is roughly the equivalent of a master's degree in either public administration or business administration. ENAG has laudable objectives but at the moment cannot meet them adequately. The school is badly in need of professional staff. There does not appear to be a permanent teaching staff; most teaching is done by expatriates and by senior civil servants on a part-time basis. The recurrent costs of most of the teaching and curriculum development in the certificate and diploma programs are underwritten by a single donor, France, with limited assistance from Germany. There are some encov'raging developments such as a fornal cooperation agreement with the Asian Institute of Technology (AIT) in Thailand. The second year of the diploma program is now taught in AIT in Thailand in both French .nd English. 5. The Lao PDR civil service is governed by decrees which have been recommended by DARCS. Decree no. 171/PM establishes (i) the rules that define who is part of the civil service; (ii) the rules that govern appointment, confirmation, probationary status, promotion, and disciplinary measures; and (iii) the rights of civil servants. Decree no. 172/PM establishes the number of ranks and brackets in the civil service and classifies positions according to these ranks. There are six ranks: rank 6 is the highest and has 4 brackets; each of the other ranks has 15 brackets. Ranks are used to determine the entry level of an individual to the civil service. With some exceptions for ranks 5 and 6, higher ranks require higher levels of educational attainment.2 The brackets are used primarily for the purpose of promotion via seniority, much as in the French system. Individuals move up to the next bracket in their rank after two years at their current bracket and continue this way until they reach the fifteenth bracket. At that point, they must wait three years to move up to the next rank. There are exceptions to this but the rules are not clear. BOX 15.2: TECHNICAL ASSISTANCE FOR PUBLIC ADMINISTRATION REFORM The UNDP Public Administration Reform program began late in 1993 and has five stated objectives: (i) to help build the personnel management function for the entire civil service; (ii) to clarify the roles, relationships, authorities, and accountabilities between ministries and provincial governors (a pilot project with three ministries has been initiated precisely to investigate this issue); (iii) to help develop training programs for civil servants in accordance with new government policies and priorities; (iv) to create a national dialogue that would help make public administration decisions, operations, and services more transparent and accessible to the general public; and (v) to promote the participation of women and minorities in the country's public administration. No evaluation of the progress and effectiveness of the program is available at this time. At the same time, the World Bank is providing technical assistance to develop a monitoring system for civil service reforms based in the Ministry of Finance. 2 Ranks 5 and 6 are reserved for high-ranking government officials (e.g., ministers) who need not necessarily have better educational preparation than those in lower ranks. Unfortunately, there is not much information at this time about the educational composition of the civil service nor is there a clear specification of educational requirements for each rank. - 173 - 6. (Nii) Compensation. Decree no. 175/PM, promulgated in 1994, introduced a new salary structure to replace the remuneration system established in 1989. The decree increased basic salaries significantly, including the lower ranks where most of the personnel are concentrated (Figure 15.1 . Table 15.1 provides an approximation of the salary increases across the different ranks up to rank 5. FIGURE15.1: COMPOSITIONOFPERSONNELBYSALARY CATEGORIES Pre-Reform Post-Reform Grade 5 Grade I Grade 5 Grade 4 7.0% 10.0/ Grade74 0.3% Grade I 17.0% ~ ~ ~ ~ ~ ~ ~~1.% 40 2^ ~~~~Grade 2 Gae3f 32.0% Grade 3 35.0% Grade 3 ~~~~~~~~~~~~~40.0% 34.0% TABLE 15.1: PRE- AND POST-REFORM SALARY STRUCTURE Monthly Salary Rank Pre-Reform Post Reform % Change (in kip) (in kip) 1 15,218 29,200 91.9 2 16,968 31,000 82.7 3 18,821 37,900 101.4 4 20,800 42,200 102.9 5 25,800 50,500 95.7 Note: Salary figures are "mode" salaries, i.e., the salary most commonly awarded in each category. Sources: Lao Authorities and World BanklIMF staffestimates. 7. The rationale set forth for this increase was that civil service salaries were so low that it was mostly qualified officials (particularly at the higher levels) who were taking advantage of the severance schemes which the Government introduced to help downsize the civil service (more on this below). It is not clear how the increases were arrived at and, more specifically, whether they were calibrated with private sector salaries. There is a belief that these new salaries are still lower than those in the private sector but this needs to be verified. The compression ratio also appears to remain low. A rough approximation is given by the ratio of the highest allowable civil service salary, 104,000 kip, to an approximation of the lowest salary in the civil service, 23,000 kip, which is about 4.5. 8. (iv) Retrenchment. Since 1989 retrenchment has been at the core of civil service reform. The Govemment has made organizational changes and introduced new laws pertaining to the civil service, but the main thrust of the reform efforts to date has been to downsize the civil service. The 3 There is no information on the salary levels for brackets under rank 6. The highest salary at rank 6 is 104,000 kip. - 174- administrative structure has barely been replaced at all and much more needs to be done. The Government reduced the size of the (non-military) civil service from almost 95,000 in 1989 to about 76,000 in 1991 and further retrenched another 2,000 in 1992 (see Table 15.3 at the end of this chapter).4 Thus, over a three-year period, it reduced the size of the civil service by slightly more than 22%. Beginning in fiscal year 1993, however, the Government began to face greater difficulty in downsizing. As part of the IMF ESAF program, it had agreed to retrenching about 5,220 workers per year over the three-year period 1993 to 1995 -- for a total of 15,660 over the period. Net retrenchment for fiscal years 1993 and 1994 has been estimated at 5,045 (total for both years) and for 1995 at slightly over 1,000. The total over the period is more than 50% less than the agreed upon targets.5 9. Two observations stand out regarding the distribution of civil servants at present (a total of 70,534 as of January 1, 1996, as shown in Table 15.4 at the end of this chapter). First, a large number are concentrated in the Ministry of Education, indicating a large number of teachers in rural areas (Figure 15.2). However, these civil servants seem to be largely composed of ex-revolutionaries who are still on the Government payroll as a reward for their part in the country's pre-1975 struggle. Most of these "teachers" have few qualifications and apparently do little actual teaching. Second, the administration at the central (ministerial) level is actually quite lean compared with the provinces, where most of the staffing is concentrated (Figure 15.3). 4 While the numbers are unverifiable, some -- perhaps a good number -- may actually be "ghost" employees. Note also that at the moment there is no master list of civil servants on which to base movements in and out of the service. Any references to numbers are largely "guesstimates" based on conversations with various individuals. Unofficial sources indicated that there would be another 1,000 retrenched civil servants for fiscal year 1995. - 175- FIGURE 15.2: LAO CIVIL SERVANTS BY MINISTRIES AS OF JANUARY 1996 Pro. org. Prosecutn - - Health Education Labor Infor. Commerce ' Industry 'TIr< . Comm. Agric. Finance ..l CPC Justice Foreign Aff. Prime Min. _ - 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 FIGURE 15.3: CIVIL SERVANTS BY PROVINCE AS OF JANUARY 1996 Saysomboun Sp. Zone 1-- I - I - Attapeu | I i Champasack I Sekong _ ! ! Saravane I | i Savannakhet Khanmuane Borikaixay _ Vientiane ___ I ! Xiengkhuang Xayabury _ Huapharh - Luangprabang _ n Bokeo Oudomxay _ Luagnptlia m ti Pho=nealy _ _ i Vientiane Prefectur_ Ministerial - _ - 1,000 2,000 3,000 4,000 5,000 6,000 7.000 a,000 9,000 10.000 -176- B. OPTIONS FOR FURTHER REFORM 10. There are reasons to believe that not much more downsizing can be achieved. For example, the ratio of non-military civil servants to the total population may already be very low, at approximately 1.5% (UNDP, 1996).6 The ratio is higher in many other developing countries, including some of the highly successful East Asian economies, and in a few developed countries. The reform process should reconstitute the civil service to meet the demands of the emerging market-based economy. That is, change the skills mix and increase the effectiveness of the civil service as opposed to merely reducing its size. Further progress in civil service reform must focus on a range of measures. These are described below. 11. (i) Replace Redundant Civil Servants. To alter the skills mix of the civil service, individuals whose skills do not match current and future requirements must be replaced. These people are likely to fall under two categories: those who participated in the revolutionary war (1954- 75) and those who were civil servants under the "Ancien Regime" (pre-1975). The former were inducted into the civil service as a reward for their participation and sacrifices in the quest for independence. Most of these individuals lack the skills required of a civil service for a market-based economy. The civil servants from the previous regime are quite elderly and are less able to perform their tasks efficiently. Consequently, these groups exert a drain on the civil service. For the current Government to maintain its credibility (and thus its legitimacy), it must address the needs of these two groups in a fair way. 12. The actual number of individuals who fall under the two categories is not known. Unofficial estimates are around 6,000 to 7,000.7 Table 15.2 provides some support for this and indicates roughly how many might be in each of the two categories. According to DARCS, many of the ex- revolutionaries have served 30 or more years (including their military service under the revolutionary period). Assuming these individuals were 15 years of age at the time they joined the revolution, they would be approximately 45 years or older now. DARCS has also indicated that most of these individuals are below the age of 55. Hence, they would tend tc fall roughly between the ages of 46 and 55. Also, according to DARCS, the members of the "Ancien Regime" are likely to be 56 years or older. As indicated in Table 15.2, the total number of civil servants in the corresponding categories is 6,531; of these, approximately 5,459 are ex-revelutionaries and 1,072 are members of the "Ancien Regime." 6 In Lao PDR, civilian employees of the Ministry of Defense and the Ministry of the Interior are not counted as civil servants. However, teachers, health workers, employees of Party organizations, and also those in the provincial and district offices are so counted. 7 Prior to completion of the UNDP Report on Civil Service Statistics, estimates were placed at around 10,000. - 177- TABLE 15.2: AGE DISTRIBUTION OF CIVIL SERVANTS Age Category Number Percent Distribution Less than or equal to 30 33,292 47.2 Between 31 and 35 15,038 21.32 Between 36 and 40 9,783 13.87 Between 41 and 45 5,890 8.35 Between 46 and 50 3,590 5.09 Between 51 and 55 1,869 2.65 Between 56 and 60 853 1.21 Greaterthan60 219 3.1 70,534 100.00 Source: General Statistics on the Civil Service in Lao PDR, UNDP, February 1996, 13. Before any early retirement scheme (see Boxes 15.3 - 15.5) is introduced, it is important to evaluate how much it will cost the national Treasury. To estimate these costs, information about the distribution of civil servants according to both age and rank is needed but is not available at this time. However, a rough approximation of the costs shows that early retirement for ex- revolutionaries and civil servants from the "Ancien Regime" is affordable and will pay off in the long run. Box 15.3: TARGETED EARLY RETIREMENT The current pension law effectively discourages civil servants in both of the two groups from retiring even if they wanted to, because it requires them to serve additional years before they can collect their pensions. This law specifies two major eligibility conditions. Individuals must have served at least 25 years, and men must be at least 60 years old and women at least 55. But individuals in the first category (ex- revolutionaries) are required to serve 30 years and be at least 55 years of age for men and 50 years for women. And for those in the second category ('Ancien Regime" civil servants), the law specifies that three years of service under the "Ancien Regime" are equal to only one official year of service. So even if an individual served the full 21 years under the "Ancien Regime" and continued on as a civil servant after 1975, that person would legally have served only 22 years. Many civil servants in the first category have served at least 30 years but are considerably younger than 55 years; and those in the second category have effectively served no more than 20 years (owing to the "Ancien Regime" clause), though many are nearing 60 years of age or are older. DARCS has also indicated that many civil servants in either category would be willing to leave the service if they could get a pension. Given the political realities in Lao PDR, a feasible solution to the skills mix problem is likely to entail some type of early retirement scheme. DARCS has in fact hinted that it is considering recommending amending the current retirement decree to allow civil servants in these categories to take early retirement. - 178- 14. (ii) Reduce Entitlement Rate for Early Retirement. The above is based on the financial provisions of the current pension law, which does not address the issue of early retirement. As mentioned earlier, DARCS is contemplating legal changes which would allow the ex-revolutionaries to take early retirement. Given that early retirement makes pensions accrue at earlier dates, it makes sense to reduce the entitlement rate to less than 100%. The current law grants non-revolutionary pensioners a 75% to 90% rate. Hence, it may not be very difficult to obtain agreement on reducing the rate for the early retirement of ex-revolutionaries to 75%. If this was approved then the implied increase in the annual budget would fall to about 1.6 billion kip (or 2.3% of the current public sector wage bill). BOX 15.4: COST OF EARLY RETIREMENT FOR EX-REVOLUTIONARIES Individuals with only a general education, namely, with no professional skills (usually basic or general skills at the primary education level or below) are classified under rank 1, the lowest possible rank. The UNDP (1996) estimates that there are 5,691 civil servants classified under this rank and DARCS has indicated that the ex-revolutionaries are largely unskilled. If so, then they are likely to be classified under rank 1. It can be seen from Table 15.2 in this chapter that the number of individuals between the ages of 46 and 55 corresponds closely to the number of those with only a general education (5,459 versus 5,691), which is consistent with the contention that most ex-revolutionaries fall under rank 1. As shown in Table 15.1, the mode salary at rank I is 29,200 kip. The current pension law for public sector employees indicates that ex-revolutionaries are entitled to a monthly pension equivalent to 100% of their last basic monthly salary (entitlement rate). In addition, all retiring civil servants are entitled to child allowances, which currently range on average from 12,674 kip to 16,375 kip per year, and a lump sum payment equivalent to 15% of the last month's salary for each year of service (ILO/UNDP, 1996). Assuming that each ex-revolutionary's last month's salary at the time of retirement was at the mode monthly salary of 29,200 kip and each received child allowances of 16,375 kip, under the current law each would be entitled to an annual pension of 366,775 kip. Further, assuming the average age of ex- revolutionaries is 51 (the median in the category 46 to 55), and that each one lives, on average, to the age of 70, then given a total of 5,691 ex-revolutionaries, the Government would have to add, on average, around 2.09 billion kip each year for the next 20 years to the budget, which is about 3.1% of the last fiscal year's total public sector wage bill (of 68.3 billion kip), to cover the annual obligations to the ex-revolutionaries.' Moreover, assuming each revolutionary has had 30 years of service upon retirement, each would be entitled to an additional lump sum amount of 1,576,800 kip. The Government would thus have to make a one-time payment of approximately 8.9 billion kip to retire the ex-revolutionaries. Table 15.4 indicates that there are 5,691 civil servants in rank 1. According to the UNDP, the discrepancy of 1,877 may be attributable to delayed recording of skill upgrades (e.g., acquiring steno- typing skills) among those classified as having only a general education (i.e., no professional skills). This includes civil service wages and allowances as well as wages and allowances of military personnel and the police. - 179- Box 15.5: COST OF EARLY RETIREMENT FOR "ANCIEN RJGIME" CIVIL SERVANTS The civil servants from the "Ancien Regime" are likely to be past 55 years of age. There are at most 1,072 of these individuals and they are probably more skilled than the ex-revolutionaries, so they may be rank 3 or 4. At a 75% entitlement rate and a last month's salary equivalent to the mode salary for rank 4 (42,200 kip), each would receive a monthly pension of 31,650 kip. The implied increase in the annual budget would thus be 421 million kip for a period of ten years assuming these individuals are, on average, age 60 and live until age 70. As with the ex-revolutionaries, the Government would also have to make a lump sum payment in addition to the monthly pension. Given the peculiar service requirements of the pension law that apply to these individuals, each is likely to have legally accumulated only 20 years of service. Hence, this payment would amount to approximately 1.6 billion kip. If the entitlement rate were adjusted downward because of early retirement, then the annual increase in budgetary costs would also fall. At a 60% rate (75% of the suggested ex-revolutionary early retirement rate of 75%), the increase would only be 339 million kip. 15. (iii) Enhance Competence and Accountability. Civil service reform is a long-term process which requires that (i) personnel are competent, (ii) line agencies have the flexibility in deciding how to allocate resources across programs within their mandate and within programs to achieve least cost provision, and (iii) line agencies be held accountable for the results that they are expected to produce.8 The balancing of (ii) and (iii) is a difficult and delicate task, with which even the developed countries have had extreme difficulty. Moreover, success in such a balancing requires a certain degree of political openness. For example, client surveys are essential to the accountability of line agencies but they are unlikely to be entertained or accepted in countries with relatively closed political systems. And without accountability, flexibility is more likely to result in undesirable outcomes. Lacking accountability, line agencies will have little incentive to be efficient. This becomes especially problematic if they have considerable freedom in determining how to use their resources. This dilemma in fact helps explain why many developing country governments start out with highly centralized management systems: lacking the capacity to monitor the outputs of line agencies, they control the inputs. For this reason, Lao PDR cannot be expected to make much headway on this "balancing" requirement. A focus on building up competency may thus be more effective. 16. Given the lack of political openness in Lao PDR and the pressing need to move the Government civil service toward a greater degree of flexibility and accountability, it is worthwhile exploring the possibility of exposing senior Government officials to the experience of other countries with the merits of client surveys through workshops. Thereafter, assuming there is some buy-in into the merits, a pilot survey in one district could be proposed which could serve as a first step toward developing and promoting greater accountability in the civil service. 17. (iv) Establish Job Descriptions. Recruitment and compensation are directly linked to a transparent system of job descriptions and classifications. The Government must know what kinds of people to hire and where deficiencies and surpluses lie. It cannot do this without the descriptions 8 See World Bank 1993, Compos and Root, 1996. 9 For example, it has taken Australia and New Zealand more than ten years of continuous reform efforts, and several decades of experimentation, to establish a workable balance. (Pradhan and Campos, "The Impact of Budgetary Institutions on Expenditure Outcomes," mimeo, World Bank, 1996). - 180- and the classifications. The UNDP has had a two-year program to help the Government develop such a system. It may be worthwhile to explore the opportunities for cooperation in this area (e.g., linkages with compensation scales) to speed up the process of competency building. 18. (v) Narrow Salary Differentials between Public Service and Private Enterprise. Attracting competent personnel to the civil service requires compensation comparable to the private sector. The private sector in Lao PDR is still in a nascent stage, and thus compensation is unstable and is likely to be highly variable with high payoffs over short periods. Nonetheless, it would benefit the Government to obtain information about trends to reduce gaps between public and private compensation over the longer term. Large differentials will ultimately draw skilled people away from the civil service and will tend to reduce accountability for poor performance: in other words, "You get what you pay for." This makes it more difficult to achieve an environment conducive to promoting accountability. 19. (vi) De-Link Pension Adjustments. Pension adjustments are currently indexed directly and fully to increases in civil service salaries. This could pose a serious budgetary problem if the Government decides to increase civil service wages at rates sufficient to keep civil service compensation from lagging too far behind private sector compensation. An ILO/UNDP study (1996) has suggested that this requirement be changed to more accurately reflect the lifetime earnings pattern of a retiree. In particular, it recommends that pension adjustments be linked to corresponding civil service salaries averaged over a five-year period. This may be technically sounder, but it also requires more sophisticated data management (pensioners will be of different ranks and will have served a different number of years). A simpler, more manageable system is to link pension adjustment to the average increase in the CPI over the preceding three years with an upper limit of around 10%. The upper limit protects the Government from the heavy budgetary burden that sudden surges in inflation can impose. This system is much simpler to manage and more transparent, thus making it more likely for pensioners to receive the pension payments that maintain real values.'0 20. (vii) Reduce Political Patronage. The country has a single and very dominant political Party which hitherto has exercised considerable influence over appointments and promotions in the civil service. Unless this influence can be circumscribed, it would be very difficult to establish incentives that engender competence. This undoubtedly will take skill and patience on the part of the DARCS staff and its supporters in the Party. Introducing and institutionalizing objective criteria for selection is an important component of any strategy that aims to de-link civil service appointmnents from political considerations. This will not eliminate political patronage, but it will reduce the number of unqualified political appointees. Moreover, it would help signal a change in the direction of recruitment which could later be built upon to establish more effective de-linking mechanisms. 21. (viii) Standardize Recruitment. To justify the above early retirement scheme, the new skills mix that eventually emerges must offer sufficient gains in efficiency that will ultimately lead to lower costs and improved delivery of public services. Therefore, the Government must ascertain that the new recruits are better qualified and have the requisite skills to perform the jobs for which they are hired." A process of identifying qualified applicants with the requisite skills will need to be to Under SAC 111, the Govemment has agreed to follow this recommendation and link pension adjustrnents to an appropriate index such as the CPI. This may have additional budgetary implications since the educational levels of new recruits may be higher and thus their entry rank may be higher. Hence, it may actually be more costly to replace the early retirees. The difference in replacement cost would be an added budgetary burden. - 181 - established. In the more advanced economies in the region, this is generally accomplished through competitive entrance examinations (e.g., Singapore) or through recruitment on the basis of graduate degrees from established and reputable academic institutions (e.g., Thailand). DARCS may find it difficult at this time to launch and manage a full-scale entrance examination since this requires extensive preparation across different fields (e.g., law, economics). A simple aptitude type of qualifying examination may be more manageable.12 While this may not have a visible and immediate impact in the political and historical context within which the Lao PDR civil service has operated, it may in the long run represent a first step toward a more professionalized civil service. 12 The Civil Service Commission of the Philippines uses an aptitude examination for lower level staff that does not appear to be difficult to prepare or to administer. One purpose of the examination is to reduce the pressure on the Commission to satisfy the demands of politicians for patronage jobs in the public sector. TABLE 15.3: PERSONNEL AND SALARIES IN THE PUBLIC SECTOR, 1988 - 95 1988 1989 1990 1991 1992* 1992/93 1993/94 1994/95 1995/96 Change End Dec. End Dec. End Dec. End Dec. End Sept. Start Oct. Start Oct. Start Oct. Start Oct. 1989-96 Civil Servants (number of people)** 94,100 94,100 85,150 76,200 74,160 74,160 73,730 69,115 68,108 -25,992 % change 0.0 0.0 -9.5 -10.5 -2.7 0.0 -0.6 -6.3 -1.5 -27.6 Total Wage Bill (million kip)*** 11,250 19,590 35,683 36,725 37,334 44,055 56,102 68,313 78,410 %change 0.0 74.1 82.1 2.9 1.7 18.0 50.3 21.8 14.8 Monthly Average Basic Salaries (kip) 18,785 18,785 27,905 34,420 36,506 % change 0.0 48.5 23.3 6.1 Memorandum Items Inflation na 59.5 35.6 13.4 9.9 6.2 6.8 20.0 9.9 * January - September only. * Excludes military and police. *** Includes wages, salaries, and other compensation; includes military and police. Wage bill breakdown 55% for military and police and 45% for civil servants. Sources: Data provided by Lao PDR authorities; IMF and World Bank staffestimates. TABLE 15.4: TOTAL NUMBER OF CIVL SERVANTS BY CATEGORIES, AS OF JANUARY 1, 1996 Ministerial Provincial Mass Organization Total % Share Total Monthly Monthly Wage Bill Salary in kip (million kip) asofJan. 1996 Grade 6 53 19 38 110 0.16 Grade 5 34 23 9 66 0.09 50,500 3.3 Grade 4 9,804 575 206 10,585 15.01 42,200 446.7 Grade 3 23,823 1,477 116 25,416 36.03 37,900 963.3 Grade 2 26,911 1,651 104 28,666 40.64 31,000 888.6 Grade 1 4,921 725 45 5,691 8.07 29,200 166.2 Total 65,546 4,470 518 70,534 100 2,468.1 Source: Civil Service Administration. TABLE 15.5: LIST OF CIVIL SERVANTS AT THE CENTRAL AND PROVINCIAL LEVELS Total Prime Foreign Justice SPC Finance Agric. Comm. Industry Commerce Infor. Labor Education Health Prosecution Pro. org. Min. Aff. Ministerial 7,742 742 297 102 206 245 1,067 369 137 68 426 82 2,845 1,072 84 Vientiane Prefecture 6,202 52 294 215 64 28 42 77 42 4,594 478 44 272 Phongsaly 1,636 26 85 64 43 7 25 42 18 862 236 20 208 Luangnamtha 1,363 26 80 58 44 4 21 35 22 672 222 12 167 Oudomxay 1,678 27 17 77 79 44 14 29 53 24 788 279 26 221 Bokeo 1,373 29 90 68 30 10 24 36 22 684 237 8 135 Luangprabang 4,836 33 124 204 60 9 30 108 39 3,070 875 34 250 Huaphanh 3,007 27 86 102 49 9 24 48 27 2,037 328 20 250 Xayabury 3,587 35 143 200 48 7 26 49 21 2,372 464 17 205 Xiengkhuang 2,764 28 78 75 47 6 27 53 30 1,851 300 17 252 Vientiane 5,771 32 163 330 71 21 37 49 38 4,070 557 28 375 Borikhamxay 2,524 28 78 159 52 14 30 42 24 1,547 317 21 212 Khammuane 4,189 39 178 363 43 17 47 62 37 2,271 803 20 309 Savannakhet 9,823 49 333 659 129 63 60 100 42 6,749 1,208 31 400 Saravane 2,731 33 130 214 37 16 35 45 27 1,372 478 24 320 Sekong 958 26 54 91 35 4 19 33 14 337 164 8 173 Champasack 7,612 28 193 345 68 14 49 113 36 5,326 993 35 412 Attapeu 1,438 29 69 92 40 7 14 46 41 713 181 11 195 Saysomboun Sp. Zone 782 9 45 57 13 2 9 13 17 435 68 - 114 Total 70,016 742 297 658 223 2,545 4,442 1,286 389 616 1,430 603 42,595 9,260 460 4,470 Note: Total Ministerial and provincial civil servants 70,016 Total number of civil servants belonging to mass organizations at the central level 518 Total number of civil servants as of Januaryl, 1996 70,534. Source:Civil Service Administration. - 185- CHAPTER 16: SOCIAL SECURITY A. REVIEW OF EXISTING SOCIAL PROTECTION PROVISIONSI 1. The pattern of social protection provision in Lao PDR illustrates the legacy of a centrally planned economy dominated by the public sector. This is gradually changing under the NEM as a response to both privatization and new opportunities for private sector development. In every country the social protection system is, to some extent, a reflection of the economic system and the stage of development. For countries such as Lao PDR, which are in the process of economic transition, it is essential that the social protection system is restructured to support that process. A system which closely links benefit entitlement to public sector employment inhibits labor mobility and private sector development. However, the formal sector is still relatively small in Lao PDR and the majority of the population is dependent on subsistence agriculture, has little cash income, and has very limited access to basic health services. Social protection thus needs to be developed at the basic lev-el as part of poverty alleviation programs, while remaining affordable within the public resource cGnstraints. 2. Pervading these general observations is the recognition that human resource development is still at a low level in Lao PDR, with skilled and experienced personnel very scarce. This will have inevitable consequences for the development of the administrative systems required for social protection delivery. In this context, this chapter focuses on the existing social protection provisions for (i) the public sector and touches briefly on (ii) the private sector and (iii) general social protection provisions, and identifies their weaknesses. Public Sector Provisions 3. The relevant current provisions relate to civil servants in both the central and provincial administration, and employees of mass organizations of the people.2 The public sector schemes provide comprehensive protection against a broad range of contingencies (retirement, death, invalidity, employment injury, maternity, and sickness) and there is also provision for the payment of children's allowances and for severance payments on termination of employment. Some civil servants also qualify for housing and assistance with the education of their children. The provisions are also applied retrospectively to war veterans who were disabled in the war which ended in 1975 and who are civil servants. The annual totals of payments by types of benefits and their distribution are provided in Table 16.1. This chapter largely draws on an ILO report (1995), "Lao PDR- Development of Social Security," prepared by Clive Baley (ILO/UNDP/Lao R.7), together with an IMF report (1993), "Integrating Social Safety Nets into an Adjustment Program," and background work for the World Bank's SAC III. 2 The armed forces and the police have their own scheme, which is financed from a specific budget allocation, and they are not included within the scope of this chapter. - 186- TABLE 16.1: ESTIMATE OF PAYMENTS IN 1994-95 AND DISTRIBUTION BY TYPE OF BENEFIT (MILLION KIP) Old Age Children Disability Artificial Health Matemity Death Total Limbs Care 3,252 332 1,003 7 183 56 509 5,343 60.9% 6.2% 18.8% 0.1% 3.4% 1.0% 9.5% 100.0% 4. The introduction of a contributory system and the establishment, albeit nominally, of the Social Security Fund represent a change in the character of the social protection provisions available to public servants, but this has not as yet been reflected in many of the individual provisions, and there is some confusion as to the position of these provisions both in law and in practice. There are also a number of anomalies and administrative weaknesses in the system which has been established for the financing and payment of benefits, as listed below: * There is no specific matching "employer's" contribution (from the Government). The Fund and the employees' contribution rate were established without any actuarial analysis of current or projected liabilities. There is a deficit which is increasing annually and which must be met directly from the Govermnent budget. * The benefits covered by the Fund are not clearly defined. Initially, the Fund was intended solely to finance the payment of pensions, but under Regulation 2282, based on Decree 178/PM, public servants were obliged to pay for medical services and then claim reimbursement from the Fund. Pensions to disabled war veterans are also paid from the Fund. On the other hand, lump-sum payments paid to those who do not qualify for a pension are met from the Government budget. * There is confusion over the coverage of those persons who were formerly employed in SOEs, and it seems that in law, but not wholly in practice, such persons lost their pension rights on the change in status of their employer. 5. Old-age Benefits (Retirement Pensions). There are various qualifying conditions which include special provisions for those who are war veterans or who have been employed in a dangerous occupation for at least five (5) years. On the other hand, those who were employed before 1975 under the formner regime are penalized in that three (3) years' service only counts as one year for pension purposes. The following weaknesses are noted: * The qualifying period is long and there is no benefit for those who do not satisfy it: this tends to discourage labor mobility and to inhibit public sector voluntary retrenchment. * Given the fact that female life expectancy is higher than male, there would seem no logical justif ication for a lower pension age for women. * In the context of the introduction of the contributory concept, it would be advisable to finance separately the special provisions for war veterans. - 187- 6. Furthermore, the benefit formula is very generous with replacement rates of between 75% and 90% of the last basic salary (up to 100% in the case of revolutionaries) plus child allowances. The effect is compounded by the additional provision of a lump-sum payment equivalent to 15% of the last basic monthly salary for each year of service (which is not paid from the Fund). It should be noted, however, that many civil servants receive, in addition to their basic salary, a responsibility allowance graduated according to the level of their post but with a current maximum of 10,000 kip per month. 7. It may be argued that the high level of benefits is a reflection of low wages in the public sector. But it is implicit in the goals of the NEM that the public sector is progressively restructured on a leaner and more cost-effective basis, with a revised salary structure and conditions of service. In this context, it would be appropriate to ensure that available resources are prioritized to ensure adequate remuneration for productive and efficient service which could, it is hoped, gradually provide a realistic income from which the individual public servant could make some provisions for retirement. 8. The high replacement rates are further enhanced by full indexing to earnings by reference to the grade and scale applicable at the date of retirement. This has the effect of maintaining the high burden of pensions expenditure, whether it is met by the Social Security Fund or directly by the Government, and inhibits progress in improving the public service salary structure because of the implicit additional cost. Under SAC III, the Government has declared its intention to de-link pension adjustments from salary increases. 9. Projections made in the actuarial analysis are based on a benchmark of an annual influx of 350 new awards of old-age pensions with an average pension of 32,000 kip. On this basis it is considered that the total payments for old-age pensions would remain stable in the next few years and would then decline. But if there were a reduction in the number of contributors because of the public sector retrenchment, or if the number of new pension awards exceeded 350, then this would have implications for financing. It is predicted that it will not be possible to hold to this benchmark under the present circumstances, and that expenditure levels for old-age pensions will not decline and could increase by up to 15%. 10. Invalidity Benefits. The situation that arises when a public servant must retire early from service because of permanent illness or non-work-related injury is rather confusing. This contingency was formerly included under Regulation No. 54 of 1986, included within the provisions for "convalescence," however, under Decree 178/PM and Regulation 2282, the convalescence provisions were discontinued and the "invalidity" benefit is now payable at 70% of basic salary to those who have suffered a loss of earning capacity of at least 81% and who have at least five (5) years of service. 11. Survivors' Benefits. Child allowances continue to be paid following the death of a public servant or pensioner, and, in addition, the child receives an orphan's allowance of 10% of the deceased parent's basic salary until age 18. However, there is no provision for payment of survivors' benefits to a spouse (widow or widower). Provision is limited to the payment of funeral expenses and of a lump sum of up to 15 times the last basic monthly salary. Although this is a significant amount, it does not provide an adequate level of protection for those who were dependent on the deceased and is inconsistent with the protection provided to widows of the disabled. - 188- 12. Employment Injury (Including War Veterans). Employment injury provisions cover civilian employees in respect of current risks and in respect of disability during the wars that ended in 1975. The vast majority of claims are made by war veterans, and the Department of War Veterans of the Ministry of Labor and Social Welfare (MLSW) administers the disability provisions covering four groups: (i) war veterans; (ii) civilian victims of employment-related injury; (iii) civilian victims of non-work-related injury; and (iv) invalidity (resulting from serious sickness). 13. There is no provision that restricts the payment of benefits to disabled persons when they reach retirement age and become entitled to pensions. This results in combined replacement rates well in excess of 100%. This is another aspect of the scheme that should be reappraised in the context of social security principles. The overall objective is to provide the beneficiary with a pension which reflects membership in the scheme and which is based on earnings when the beneficiary was working. It is inconsistent with this approach to use social security resources to finance benefit entitlement under two provisions at the same time. The usual approach would be to restrict entitlement to the higher benefit. Periods of entitlement to employment injury or sickness benefit would be regarded as periods of service for retirement pension purposes. 14. There is a need to reconsider the basis of the disablement provisions and, in particular, their application to war veterans. There are special considerations applicable to war pensioners which may justify their special treatment, but it is inadvisable to merge these with provisions to meet loss of earnings in a social security scheme financed through the contributions of current workers. 15. Medical Care. Formerly, public servants and pensioners and their spouses and children under age were entitled to free medical treatment. By virtue of Regulation 2282, however, the patient is required to pay the appropriate charges and then claim reimbursement from the Social Security Fund through his or her employing Ministry. Where services are provided by a Ministry of Health facility, there is no charge for the accommodation, medical equipment, food, etc. The public servant may attend a private clinic for treatment if public health facilities are not available, but there are limits to the extent of the reimbursement. 16. The intent of this legislation is to provide additional resources for the health sector, and this is consistent with the policy being developed by the Ministry of Health to introduce cost sharing. The arrangement, however, does not seem to be widely known either among health facilities or public servants and is not fully operational. Furthermore, steps have not been taken to set aside a specific part of the 6% social security contribution for health care purposes. 17. Sickness Benefits. Under Decree 178/PM, a public servant who is unable to work because of sickness is entitled to full salary for 30 days and thereafter to 80% of salary for up to another 12 months. In reality, however, salaries are paid in full and the amount is not claimed from the Social Security Fund. 18. Maternity Benefits. A pregnant public servant is entitled to three months of maternity leave with full pay but, in addition, also receives a lump-sum payment of 60% of her basic salary on the birth of her child. Maternity leave can be extended by paid sick leave. Under Regulation 2282, the cost of medical treatment, but not the allowance, is met by reimbursement from the Social Security Fund. 19. Termination of Employment. Severance payments for public servants have been introduced in order to support a voluntary retrenchment program which resulted in a reduction of the civil - 189- service from nearly 95,000 in 1989 to some 73,000 in 1993/94. Further reductions were targeted to reach 55,000 by the end of 1995, but the pace of the reduction has slowed, partly because of the need to avoid a loss of skilled people through a more selective and controlled approach and partly because the compensation package offered has not proved sufficiently attractive (see Chapter 15 for more detail). The allocation for retrenchment payments in the 1995/6 budget has been reduced to 2.6 billion kip from 3.8 billion in 1994/95. The formula applied for determining the payment is: Last monthly basic salary x years of service + one month's salary + one year's children allowances. 20. According to an IMF study, the average severance pavment in the first six months of 1994/95 was 486,000 kip (about 115% of average gross salary). The IMF study reflected on the effectiveness of the retrenchment policy and on the suggestion that take-up had been low as the payments were not sufficiently attractive. But in the subsequent analysis it was concluded that it would be inadvisable for the payments to be increased. It should be noted, however, that termination of employment would result in a loss of entitlement to what are seen as important pension provisions. It is usual in most pension schemes for there to be a provision for compensation by a lump sum for those who are unable to qualify for a pension, and a formula on this basis was introduced, taking effect from November 30, 1993, under Decree 178/PM. This formula is almost identical to, but not additional to, the one already in place as a severance payment. Thus, there does not seem to be any additional provision to encourage further public sector retrenchment. 21. Employees of SOEs. There is some uncertainty as to whether employees of SOEs continue to be covered under the public sector pension scheme. It seems from the provisions of Decree 178 and Regulation 2282 that they are excluded and this is the view of the Ministry of Finance (MoF) and Department of Public Administration. However, it does not seem that they were specifically compensated for what was a loss of pension rights. Lao Beer is jointly owned by a Thai consortium, which has a 51% shareholding, and the Government of Lao PDR, which has 49%. Salaries and conditions of service are determined by an 11-person Committee which reflects the ownership division. Representatives of the Ministry of Industry and the MoF sit on the Committee. The 304 workers are all in pensionable employment under the public sector scheme, and several workers have retired since 1993 and receive their pensions through the Ministry of Industry. The company does not deduct the 6% contribution from workers but pays this monthly from company funds direct to the MoF. 22. Nevertheless, where the enterprise is still operating as a joint venture with partial state ownership, it seems that, at least in some cases, employees continue to be regarded as insurable in the pension scheme. For the workers of the 30 strategic enterprises that remain within the public sector (including major employers such as the public utilities -- the Lao Electric Company; the water company [Nam Papa]; post and telecommunications [EPTL] -- and the pharmaceutical and state printing companies), neither Decree 178 nor Regulation 2282 seems to apply. 3 See IMF (1993), "Lao PDR: Integrating Social Safety Nets into an Adjustment Program." -190- Nam Papa has 571 employees throughout the country; water supply in Savannakhet Province is the responsibility of a separate water company. Nam Papa is continuing to deduct 6% from each employee as a contribution to the pension fund. These payments are not made over to the MoF or the Social Security Fund but are placed in a special account. If a worker leaves before pension age he is paid a lump sum from this account representing one month's salary for each year of service. Those who have reached pension age have claimed and received their pensions through the responsible Ministry. Other benefits to Nam Papa workers are paid from company funds (80% of health care costs are reimbursed andfull basic pay is paid during maternity or sick leave). 23. Financial Administration. The present arrangements for financial administration are cumbersome and it is difficult for the Department of Social Security and the Department of Budget to either project expenditure or determine it subsequently. Financial arrangements relating to income from contributions for persons employed in the provinces and for pensions paid to persons resident in the provinces are decentralized. The result is a complex web of accounting and information flows among the central Treasury, the provincial treasuries, the finance offices of the provincial departments, and the provincial departments of social security. 24. The Department of Pensions in the MLSW is responsible for determining the rate of pension and for issuing an award notice and a pension identity card to the responsible department. However, the Department has only five staff members and has not yet been able to establish records of pension awards. Pensions for provincial staff are awarded in the provinces, and there are as yet no central records. There is no basis for coordination with the Department of Social Security, which has a staff of only eight but which is responsible for the payment of pensions and the management of the Fund. The Department of Public Administration in the Prime Minister's Office is establishing a computerized data base for serving civil servants, but there are as yet no prospects of a data linkage with the MLSW. *25. There is no data base for pensioners either centrally or at the provincial level. Entitlement and payment depends on the three-monthly submission of schedules to the Department of Social Security (centrally or at the provincial level) by the responsible employing departments: this is effectively a claim for payment from the Fund. The information provided is detailed, but is deficient with regard to the data required for actuarial purposes (i.e., there are no data for date of birth, or for age, sex, or length of service). 26. There are major differences in the financing of pensions from the financing of other social security benefits, such as health care. Expenditure on pensions will increase annually as more persons become entitled to pensions at higher rates and with longer life expectancies. As noted above, the financial plan will need to be formulated to fit these circumstances. Health care is a short- term benefit where the pattern of expenditure can be assessed on an annual basis and financed accordingly. In these circumstances, it is inadvisable to mix the financing of both types of benefits within one fund, and separate accounts supported by distinct contributions should be established. Private Sector Provisions 27. Current Status. From the situation in 1986, when almost all economic activity was controlled by the state, there has been significant development of the private sector under the NEM policy. This reflects the privatization of many SOEs and the establishment of new enterprises with - 191 - both foreign and local investment. In the absence of either an analysis of the 1995 census or nationwide surveys of establishments and the labor force, it is difficult to quantify the size of the private sector. Most SOEs have been privatized in ways that have involved a change of status for about 3,000 workers. Many factories have been established, particularly in the manufacturing and services sectors.4 But these represent only a partial picture, since trade, services, and finance establishments are excluded, and it is estimated that there are currently between 60,000 and 70,000 people employed in the private sector. 28. In those industries which were privatized or which are operating on a joint venture basis, it seems that those workers who were retained following the change of status generally lost their pension rights, since their membership in the public servants' pension scheme ended. The new companies generally provide higher levels of remuneration but do not provide pensionable service. Private sector workers do not enjoy the same level of social protection as the public sector, and there are indications that in some cases they are working under adverse circumstances with inadequate conditions of service. The following paragraphs review the situation in more detail. 29. The Labor Act of 1994. The Labor Act applies to all workers and employers who carry on activities in any economic sector in Lao PDR but it does not apply to civil servants employed in state administration and technical services, national defense, and public order.5 Section 62 of the Labor Act places the responsibility for the implementation of the Act on the labor administration -- the Department of Labor -- and provides for the issuing of regulations regarding implementation, direction, inspection, etc. The Act cannot yet be implemented (or enforced), as it requires an implementation decree (which has been drafted but is awaiting enactment) and supporting regulations. 30. The Labor Act has a number of references to "social security" which are intended to place obligations on employers to provide social protection for their workers. While most of these provisions relating to social security require additional legislation before they can become operational, Section 48 obliges each labor unit to establish a "social security fund" or to pay contributions to a compensation fund. 31. Thus, the Labor Act provisions represent a mixture of reliance on an obligation imposed directly on employers and references to a social security fund. But these provisions are not being generally applied. It is understood from discussions with the Chamber of Commerce, the Lao Federation of Trade Unions (FTU), individual employers, and the MLSW that many employers do not know about, do not understand, or merely ignore the provisions of the Labor Act. Those employers that provide social security benefits to their workers at or above the level of the Labor Act are in the minority. 32. A few employers have purchased employment injury compensation insurance from the only insurance company currently operating in Lao PDR (Assurance Generale du Laos), which is associated with a French insurance company and which operates in the country with a 20% Government shareholding. Policies are offered with three options for the payment of medical 4 A survey of manufacturing establishments in the public and private sectors conducted by the Ministry of Industry in 1994 identified 5,900 industrial establishments employing 41,000 workers with 9,700 working for employers with less than 10 employees. 5 Decree No. 24/PR promulgated Act No. 002/NA of March 14, 1994 and replaced Act No. 10/90/ASP of November 29, 1990, which had been found to have a series of operational weaknesses. - 192- expenses and lump-sum payments for disablement. The insurance offered provides some protection, although it falls far short of that envisaged in international standards, or even of that provided to public servants. The premium rates, which vary according to four categories of risk, are also too high to be generally applicable throughout the private sector. 33. It is understood that there are no private pension plans or occupational pension schemes (except as an extension of foreign schemes to cover expatriate workers in Lao PDR). Some foreign employers who have established businesses have asked about pension provision, but the perception of workers varies. It is said that many public sector workers are attracted by the possibility of employment in the private sector but are inhibited by the loss of pension rights. Many of the new private sector workers are young, and their social protection priority is health cover (sickness, maternity and medical care). According to the Lao FTU several employers with union support have established their own "social security funds" which are essentially welfare schemes that provide assistance in situations of need such as death or serious illness. 34. The overriding impression of social security provision in the private sector is one of employers operating in an uncontrolled environment, with protection essentially dependent on their awareness and their benevolence. The Lao FTU has apparently established trade union activities in nine garment factories and has about 12,000 private sector members, but the majority of these are in former SOEs. General Provisions 35. General Social Protection Provisions and Health Care. The vast majority of the people of Lao PDR are not employed in either the public or private sector and are thus outside the scope of whatever social security The largest hotel in Vientiane has 207 employees of whom 75% are male provisions exist. In this and 90% under age 30. The hotel is privately owned and is shortly to be for social protection is taken over by an international hotel group. Staff turnover is high: only h sealth care. Health 16% have more than two years' experience. The hotel follows the services are the provisions of the Labor Law and provides full pay for three (3) months for responsibility of the maternity and up to 30 days sick pay per year. It has an arrangement with Ministry of Public Health a private clinic for medical treatment for its staff' the workers receive and provincial health free treatment and the hotel is billed. The hotel has purchased a workers' authorities with a network compensation insurance policy from AGL. There has been one claim -- of hospitals and clinics at for a death due to a motor vehicle accident. Medical care and workers' the district level. At the compensation insurance account for about 8% of the payroll. It is central level, there is a understood that the new owners will provide membership in a pension medical science plan for its management, most of whom are foreign. university and hospitals which have the responsibility of procuring and producing drugs. 36. As is discussed in Chapter 4 of these Background Papers, the private health sector is beginning to be established and there are now about 60 private clinics, about half of which are in Vientiane. There is as yet no regulatory system to monitor the private sector or to control costs. Those who can afford to do so seek medical treatment in Thailand. Some private sector employers meet the health care costs of their workers. The AGL offers premium health insurance cover, aimed largely at expatriate workers. Government workers are now being required to partially contribute, albeit indirectly, through the Social Security Fund, to the cost of their health care, although the arrangement is not yet fully operational. - 193 - 37. Social Welfare. There are no administrative mechanisms in place to provide assistance to those who lack the means to support themselves. The Department of Social Welfare does have some resources to alleviate hardship among some vulnerable groups, but this cannot be said to constitute the basis for any comprehensive social support system, and the limited resources make it difficult to apply systematic targeting. The main components of the social welfare program cover: (i) the aged and disabled, (ii) orphans, (iii) victims of natural disasters, and (iv) refugees. B. OPTIONS FOR REFORM AND DEVELOPMENT 38. The Government of Lao PDR has achieved considerable success in its implementation of the NEM and has demonstrated its commitment to economic growth as a means of providing improved living standards and welfare for the people. The process of transition from a centrally planned economy to one based on market forces implies a greater role for private enterprise and a gradual withdrawal of state ownership and control. This shift in policy has inevitable implications in the area of social policy, as in this area there is also a need to consider the role of public expenditure in providing social protection. 39. Roles and Responsibilities for Social Protection. In any country the arrangements for the provision and financing of social protection reflect a necessary division of responsibility among the state, employers, individuals, and families. The role assumed by the state will invariably influence the overall pattern. In Lao PDR it is reasonable to expect that the orientation toward the market economy and the development of the private sector should be complemented by measures to foster individual initiative and responsibility. In this context it can be seen that, in the future, the state should not be the sole source of social protection but that, instead, employers will be obliged to look after their workers, individuals will be expected to plan for their future, and traditional support systems within families will be maintained. 40. This will be a gradual process but the following steps have already been taken in this direction: * The introduction of a social security contribution for public servants * The introduction of cost sharing for health care * The obligations imposed on employers by the Labor Law. 41. As discussed in Chapter 15, in spite of the fact that, at present, levels of remuneration among civil servants remain low, based on a minimum wage of 26,000 kip, efforts are being made to transform the public service into a leaner, more efficient unit with increased emphasis on productivity rather than on protection. 42. These general considerations provide the background for the need to shape a social protection system that reflects the direction in which the country wishes to go as regards its social policy but which also has regard to the legacy of the past. 43. Lao PDR has a severe shortage of trained and experienced administrators and of specialist skills in such areas as accountancy and computerization. Thus, first, it is essential that the pace and scope of development in that direction takes into account the capacity of the system to administer it. Second, opportunities should be sought to devise mechanisms for some public functions to be - 194- undertaken by autonomous agencies (or even by the private sector) in order to relieve the burden on the public sector and enable it to concentrate on core activities. 44. The Need for a Social Protection Strategy. Even though it is envisaged that the responsibility for providing and financing social protection should be shared among the state, employers, and workers and their families, this activity should take place within an overall policy framework. The Government should develop a strategy for the future development of social protection which should encompass both social security and social welfare/social assistance elements and which should also ensure policy coordination among ministries. Health care is one of the most important of the social security contingencies and is an issue which merits priority in Lao PDR. The relationship between social security and health care provision and financing should be developed. 45. In formulating a strategy for the development of social protection, the Government will be faced with a number of choices. One of these has already been referred to, namely, the division of responsibility for providing and financing social protection. To what extent should the state directly provide social protection, to what extent should it use statutory systems based on collective responsibility (social insurance), to what extent should it place the obligation on employers, and to what extent should it leave the matter to individuals? The choices are restricted, at the present time, by limited fiscal and human resources, but decisions made now may establish systems and expectations that will prove difficult to change in the future. This theme will underlie many of the policy issues discussed in the following paragraphs. 46. The state has an obligation to ensure adequate living standards for all its citizens, but it also has more specific social protection obligations to its own workers. As regards the latter, current provisions reflect the former role of the state as a source of protection for workers in almost every respect. This notion carries with it obligations which are increasing and which are inconsistent with the economic development policy. There is thus a strong case for scaling down the public sector social protection system and for seeing that its financing is controlled and shared with those protected. But this cannot be achieved rapidly because many public servants have become dependent on this system and have acquired rights. Furthermore, given the generally low level of salaries in the public sector, the present scope for individual saving is very limited. Change must be gradual. 47. At the other end of the spectrum there is a virtual absence of protection for the emerging private sector workers. The Government will not wish to discourage investment and private sector development by the imposition of heavy burdens, but some protection needs to be provided to prevent workers from being exploited and disadvantaged. There is thus a clear need to develop some form of social protection for the private sector. Reconciling Public and Private Sector Provisions 48. Can these two objectives coincide? Is it possible to develop a framework for one social protection system that embraces both the public sector and the private sector? Here, again, there are difficult choices to be made. 49. In principle, and in the long term, it would be desirable to establish a public social security system where there is uniformity of treatment for both private and public sector workers. It would also be in the interest of labor mobility. But not only are the provisions for public servants too - 195 - generous (as regards their replacement rates) to be extended to the private sector, but, in addition, they reflect to a great extent conditions of service rather than social security concepts. If uniformity with appropriate social security provision is to be achieved, it must be in stages through the restructuring of the public sector scheme so that it progressively takes on the characteristics of a social security scheme. If it is considered that social protection for public servants, or certain sections of the public service such as the armed forces, merits special treatment in some respect, this would best be achieved through separate provision supplementary to the social security principles which would determine the core of the scheme. A Policy Framework 50. Basic Social Protection. Responsibility for the provision of social protection within a policy framework should reflect the source of financing. The state should retain responsibility for the provision of basic social welfare services, within the limit of available resources, and the provision of primary health care and preventive services. This basic layer of social protection would thus be financed from taxes. Ideally, this provision should also include social assistance payments to those without other means of support and determined through a means test based on uniform criteria. But in a country such as Lao PDR, where the resources are not yet available to meet such needs, this has to be seen as a long-term objective. 51. Protection against Loss of Income: Social Insurance. For those who are employed and thus dependent on a regular income, there is a need for protection against the risk of temporary or permanent interruption of earnings because of incapacity, unemployment, retirement, or, from the viewpoint of family dependents, death. Such persons will also often seek a higher level of health care than that available through the public health service. 52. To an extent, it is possible to provide this protection through obligations imposed on employers, either voluntarily, through collective agreements, or, compulsorily, by public statute such as the Labor Law. However, in most countries experience has shown that employer-based protection cannot be wholly relied upon, particularly in respect of long-term protection, and that the heavier the burden or the smaller the enterprise, the less likely it is that, in reality, the protection will be provided. Reliance on this type of arrangement can set worker against employer and can lead to disputes or discrimination (for example, against women or older workers). The liability on the employer may be underwritten by insurance, and in many countries this is a statutory requirement, particularly in the case of employment-related injury. But the universal nature of the risk may be unattractive to commercial insurance and could lead to the imposition of high premiums or selectivity. 53. Most countries have therefore developed some form of system based on social insurance principles supported by legislation involving the payment of contributions by both employers and workers (and sometimes by the government) to a_fund. This fund is typically managed either directly by a government department or by an autonomous organization, often under the supervision of a board constituted on a tripartite basis to ensure independence and to safeguard the interests of contributors and beneficiaries. 54. Benefits are paid on a "defined benefit" basis in accordance with a prescribed formnula to those who are insured with the fund. However, the arrangements can be devised to reflect policy choices on such aspects as the scale of the benefit and the extent of redistribution either between generations or between the lower paid and the higher paid. The benefit formula selected will - 196- determine the level of contribution required to support it. Both will influence the extent to which this level of social protection meets the overall needs of those covered. 55. Protection Based on Individual Responsibility or from the Employer. If the level of protection provided by the social insurance scheme is deliberately modest (for example, benefits could be provided on a flat-rate basis), there will be increased scope and demand for supplementary provision. If, on the other hand, replacement rates (as a percentage of earnings) are high, the reverse will occur. The "supplementary" social protection can in fact be so significant as to not be supplementary at all but the dominant component. In these circumstances, the defined benefit would provide a basic level of protection with principal reliance placed either on occupational (employer- based) provision or on arrangements based on individual responsibility (such as savings schemes and personal pension plans). 56. A division of responsibility that recognizes distinct roles through three tiers of protection -- from the state (providing basic services and perhaps income from taxes), from collective solidarity (providing modest defined benefits derived from contributions), and from individual or employer- based schemes (providing occupational provision or private arrangements) -- provides a framework within which individual schemes can be designed. 57. The determination of the respective lines of demarcation between these three tiers is, however, crucial. Virtually all social protection systems reflect a desire to share responsibility for financing and providing benefits, although there is considerable variation in where the line is drawn between collective and individual responsibility. The stage of development will be material factor. The role played by the state in the first tier will be limited by the state's resources: many of the policy initiatives that are desirable will be seen as long term. Similarly, the extent to which the social protection can rely on occupational provisions or on individual arrangements will inevitably be limited for some years to come in a country such as Lao PDR, where there are no major private sector employers, where the capital market has not yet been developed, and where there is limited insurance or private pension provision. Nevertheless, it is within the scope of a social protection policy framework to define the limits of responsibility and to then foster the establishment of an environment which will, when circumstances allow, conform with these objectives. 58. In the context of these observations, the following paragraphs consider the options for reform and development of public and private sector social security provision, and also identify the priorities that require attention. 59. Pensions and Defined Benefit Social Insurance Schemes. Provision of social protection in respect of the consequences of termination of employment because of old age, invalidity, or death is the major component of most social security systems. It has also proved the most difficult to administer, particularly as regards financing. Most countries have chosen to adopt the defined benefit system, so that benefit entitlement is effectively a promise derived from the prescribed formula. As a country develops, its people tend to live longer and earn more. Most social security schemes provide pensions related to earnings in order to cushion the effect of retirement; for the pension to be meaningful, arrangements must be made to ensure that its value is maintained in real terms. Therefore, in most countries the cost of pensions is increasing annually, as more pensions are awarded and as pensioners live longer and receive higher pensions. 60. Financing of Pensions. There are basically two ways to finance a pension scheme. Under a fully funded financial system, funds are accumulated in advance so that current resources match - 197 - future liabilities. Contribution rates need not increase unless the expected return from investment of the accumulated funds falls or benefit expenditure increases beyond the extent anticipated. But this system requires a high initial contribution rate which may be politically or economically unacceptable, and many developing countries either lack the opportunities to invest the funds with a real rate of return, or lack the expertise to do so, or else lack the data to make long-term projections reliable. The fully funded system is generally adopted by occupational pension schemes to ensure long-term protection of those insured. 61. Most public schemes instead adopt a financial system which is either based on the pay-as- you-go (PAYG) system or where there is some partialfunding. Under a PAYG system, this year's contributions are used to pay this year's pensions. But since pension expenditure increases over time, this means that the contribution rate will increase and that effectively, as the scheme matures, the contribution burden on employers and workers will increase. Furthermore, it also means that under the simple PAYG system there will be a very limited accumulation of reserves and thus limited opportunity to secure additional income to the fund from investment. 62. In order to offset these disadvantages, and where there is scope for a real rate of return on investment, a partial funding system may be developed with a higher contribution rate. This permits a partial accumulation of funds and thus an investment income. When the return from investments is insufficient to meet the pension expenditure, the financial system is reviewed and the contribution rate is increased. This financial system is flexible, and thus the contribution rate and the period of equilibrium (between reviews) can be adapted to suit the circumstances in the country. Its success does, however, depend on (i) the understanding that the initial contribution rate will have to be increased at some point, and (ii) the effective management of the accumulated fund to secure the anticipated return. Where this is not the case -- either through mismanagement or as a consequence of inflation -- the financial structure of the pension system is threatened and some action may be necessary to adjust either the benefit formula or the contribution rate. 63. In Asia, social insurance pension schemes on a defined benefit basis have been established in the Philippines (separate schemes exist for public and private sector workers), in Vietnam, in Malaysia (for invalidity benefits only), in Pakistan, and in Iran. They are under consideration at present in China, India, Mongolia, and Sri Lanka. Most countries in the region provide pensions on a defined benefit basis to their public servants, although this is generally not done through a social insurance scheme (Vietnam and the Philippines are exceptions). Expenditure in respect of public service pensions in most countries in the region is still met directly from the government budget, but many countries have recognized, as in Lao PDR, the need to cover this increasing liability through the establishment of a special fund. 64. Provident Funds. In many developing countries, social protection on termination of employment has been provided by the provident fund system, which is essentially a form of compulsory savings derived from contributions by both workers and their employers. The benefit paid as a lump sum on final termination of employment is based on the accumulated savings in the member's account, and there is no insurance element which reflects the pooling of risks. Since there is no defined benefit, the amount paid out is based entirely on the contributions paid in respect of the individual member plus the return on the investment. If the investment performance is poor, as it has been in many countries, because of inappropriate decisions or inflation, the member bears the risk. In some countries the value of members' accounts stand at considerably less in real terms than the value of the contributions at the time of payment. Furthermore, the benefit paid has no bearing on the needs or circumstances of the individual at the time of retirement, there is no protection in the - 198- event of early termination because of invalidity or death, and the benefit is paid as a lump-sum instead of a pension. For these reasons, many countries are reviewing their provident funds and contemplating their conversion into a defined benefit pension scheme. This has already been done in Ghana and Nigeria and is under consideration in Zambia, Tanzania, Uganda, Sri Lanka, and India. Provident funds remain well established, however, in several countries (for example, Singapore, Malaysia, Indonesia, and the Pacific Island states). 65. Mandatory Savings Schemes. An important variant of the provident fund system has been developed in some countries (notably, Chile and Argentina), either as an alternative or a supplement to a social insurance pension scheme. This is commonly known as a mandatory savings-annuity scheme and differs from the provident fund in the following respects: * Contributions are paid to any one of a number of approved pension funds at the choice of the member * The choice can be varied * The approved pension funds must comply with regulations intended to protect the member * The member is paid an indexed annuity (not a lump sum) on retirement based on the accumulated balance in his or her account at that time. 66. In a defined contribution scheme of this kind, the benefit is not related to the social need of the individual member, who also bears the risk, since there is no pooling of risks which will guarantee a defined benefit as there is under a social insurance scheme. It is common, however, for such schemes to be underpinned in several ways. First, a portion of the contribution serves as an insurance premium to cover risks associated with early termination of membership because of invalidity or death. Second, the govemment may guarantee that a minimum pension level comes from public funds. 67. Other Supplementary Provisions. Supplementary social protection is provided in many countries under private pension arrangements, savings schemes, or occupational pension schemes, or through collective agreements, and the public system may be set at a level which encourages the development of such schemes. In some countries (Switzerland and Australia are examples), it is mandatory for employed persons to be members of an occupational pension scheme, and in the United Kingdom it is possible either for an employer or an individual who has an approved occupational or pension plan to contract out of part of the public scheme, apart from basic coverage. It should be emphasized, however, that where there is reliance on individual or employer-based arrangements, either as the main pillar of the pension scheme or as a supplementary pillar, it is usual for these provisions to be supported by a regulatory framework to protect the interests of workers, and by a public body to administer this framework. Pension Options for Lao PDR 68. The Public Sector. In the context of the above, what are the pensions options for Lao PDR at the present stage of its development? As regards the public sector scheme, there is a need to reduce the average cost of pensions which is currently higher than average wages. This could be - 199- achieved by various modifications which could either be introduced piecemeal or as part of a general restructuring of the benefit formula. In this context, the following reforms should be considered: => Provide for benefits to be based on the consumer price index =* Increase insurable earnings to include all cash allowances such as responsibility allowances > Reduce replacement rates in real terms > Separate expenditure in respect of pensions from expenditure regarding war disabilities, health care, and other benefits = Revise the qualifying conditions to provide for a lower minimum qualifying period and an annual accrual rate => Provide for pensions to be paid as of right and not subject to permission => Ensure that employees of privatized SOEs continue in pensionable employment = Improve protection for surviving spouses in the event of death in service or after pension. 69. Adjusting Pensions. It is important to maintain the purchasing power of pensions through a mechanism for regular review and adjustment. The present system with a direct linkage to the earnings of the pre-retirement post is too generous. 70. Extending Insurable Earnings. Many civil servants receive regular payments which are not included within insurable earnings for pension or contribution liability (for example, responsibility allowances of between 2,000 and 8,000 kip are paid to about 19,000 civil servants). It is estimated that these allowances represent on average 12% of basic salaries. Inclusion of these payments within the definition of insurable earnings would result in an increase in contribution income and would also provide a more realistic replacement rate. But in the long term, this reform would result in an increase in pension costs of 9.8%. 71. Replacement Rates. A reduction in replacement rates should also be seen as a step towards orienting the scheme away from an occupational pension scheme for public servants and towards a social security scheme which could be applied, in the future, to civil servants. It also presumes that priority should be given to improving salaries rather than pensions. This objective can only be achieved gradually because although pension rates are high the salaries on which they are based are low. It is considered that replacement rates should be progressively reduced, through transitional provisions. An accrual rate of 1.5 per cent for each 12 months of pensionable employment, with a minimum qualifying period of 15 years would be a reasonable rate for a social security pension scheme: a replacement rate of 52.5% after 35 years. 72. Although this might seem low in relation to present earnings levels, the effect of the change would be limited for some years because of the need for transitional provisions to protect acquired rights under the present arrangements. This would also allow time for the restructuring of the public sector to be completed and for improvements in salary levels to be implemented. It would be feasible to apply a scheme operating at this level (probably with the imposition of a ceiling on insurable earnings) to private sector workers at the appropriate time. It is estimated that the net long-term - 200 - impact of the introduction of a revised pension formula with a minimum pension provision would be to reduce pension expenditure by 28%. 73. Equality of Pensionable Age. There is also a case for the progressive raising of retirement age for women to achieve equality with men. This could be effected over a 10 year period to avoid it inhibiting efforts to reduce overall staffing levels. While such a change would be justifiable on equity grounds it would also result in an overall reduction of pension expenditure of approximately 7% when fully implemented. Consideration should also be given to phasing out special qualifying conditions, i.e. for those in dangerous employment. 74. Financing Options. Various options exist as regards the financing of the public sector pension scheme: (i) Option 1. Retain the present system whereby a deficit in the Social Security Fund is met by the Government on a PAYG basis. (ii) Option 2. Determine the contribution rate that will be sufficient to meet the pension expenditure on a PAYG basis and identify an appropriate division between the Government (and SOEs) as the employer and the employees. (iii) Option 3. Determine the extent of actuarial liability to the Fund and the contribution rate required to partially fund future pension expenditure. 75. The actuarial liabilities of a pension system are the present value of pensions in force and of the acquired rights of the participants. They represent the amount which would have to be paid at the date of calculation, assuming that assets were invested at the rate used in the calculation and that there were no further contribution receipts. They illustrate the pension debt arising from a pension scheme which is being financed (as in Lao PDR) according to the PAYG financial system. It is estimated that actuarial liabilities in respect of old-age pensions alone are 60 billion kip -- twice the current payroll. If it were thought desirable to adopt a funded financial system to replace the PAYG system, this would require a payroll contribution of about 13% for the next 40 years in addition to the contribution rate required to meet current pension costs. 76. On this basis, funding the scheme as in Option 3 would seem impractical. But the financial basis for public service pensions needs to be regularized. The establishment of a Fund implies the formulation of a financial plan which seeks to match income and expenditure over a period. The level of the contribution should determine the length of that period, and if there is no provision for advance funding it is inevitable that the cost of the scheme (and thus the contribution rate) will increase in the future. Furthermore, it would be preferable for the protection of social security provisions if there were to be a specific employers' contribution paid by Government. Option 2 is therefore preferred. 77. Financial Responsibility. The benefits for war veterans should probably not be met from the Social Security Fund. Entitlement relates to events that took place before the Fund was established, and contribution liability should not be imposed on current civil servants (even though some of these are beneficiaries). Compensation for war veterans should be made, but from another source financed directly by Government. 78. A similar concern relates to other liabilities imposed on the Fund by Regulation 2282. These concern medical expenses and maternity pay, which should be distinguished from the financing for pensions. A separate fund, or at least an account, should be established and financed or, preferably, these matters should also be taken as a direct charge on Government finance. The Social Security - 201 - Fund should be used to finance pensions unless special arrangements are made to apportion part of the financing for other purposes and it is accounted accordingly. 79. Survivors' Benefits. As pointed out above, the public sector pension scheme does not provide for pensions to be paid to surviving dependents (widow or widower). Although separate lump sums are paid for the funeral and for the dependents, the only pensions are paid in respect of children. It is considered that pensions should be made to dependent spouses -- those who are not employed, who are over age 45 or under age 45, and who have the care of young children. The cost of this can be partly offset by combining the two lump sums (and reducing the aggregate amount) currently paid and removing entitlement where an orphan's allowance is paid. The combination of the modification to the lump-sum benefits and the introduction of pensions for spouses would generate a 70% increase in the cost of death benefits. But this would appear only gradually, depending on the transitional provisions. Furthermore, since death benefits are a small proportion of total costs, the impact of the reform would be limited and would represent only 4% of overall long- term pension cost. 80. Invalidity Benefits. In order to achieve consistency with the proposed old-age pension provisions, it will be necessary to revise the provisions for invalidity pensions to provide for an accrual rate which reflects length of service. The cost impact of these changes would be very small, although it should be noted that the proposed new benefit formula would represent a significant cost containment measure. 81. The impact of all the changes discussed above would depend greatly on the transitional rules adopted. Savings would arise only gradually and would also be subject to other considerations such as wage policy. It is emphasized that social security reform cannot redress the problems related to either low public sector pay or overstaffing. But the concept of lower replacement rates and greater individual responsibility for social protection through savings is closely linked to higher real incomes. Replacement rates are high now because salaries are low: this linkage will continue. 82. The Private Sector. As has been noted, social protection for the private sector labor force is limited. The sector is growing rapidly and many new businesses are being established. Many of those employed are young and there is a high proportion of females among them, particularly in the important textile sector. For many of these workers, pension provision is not seen as a priority. There is no experience, either in the public or the private sector, in Lao PDR in the administration of a social security scheme. There is only one commercial insurance company and it is difficult to envisage that it has either the capacity or the desire to expand beyond its present role. It offers life insurance cover but aims principally at the expatriate market. Among the vast majority of the labor force, earnings levels are not sufficiently high to provide much scope for saving, and capital markets are undeveloped. 83. In this context, a very cautious approach should be taken to the provision of pension arrangements for private sector workers. This should await the stabilization of the sector, the development of public administration, and an increased awareness of social security concepts. The Government will wish to give careful consideration to the various options which have been discussed above. It is considered that the public sector scheme should be restructured to provide a system which could be adopted by the private sector, but this will take time. Meanwhile, social security provision for other benefits could be introducedfor private sector workers which would have a more immediate impact than pensions while being simpler and cheaper to administer. 84. Employment Injury. The incidence of employment injury within the public sector is said to be very low but it is not possible to distinguish such cases, on the one hand, from sickness and medical expenses, and, on the other hand, from war disabilities. For civil servants, it is proposed - 202 - that the cost ofprotection within the present rules should be met directly by the Government and not from the Social Security Fund. Employing departments and the War Veterans Departrnent should, however, be required to identify the number and types of claims in respect of employment injury or occupational disease. 85. An employment injury scheme based on social insurance principles should, however, be established to protect private sector workers and those employed by state enterprises. This would supplement the provision in the Labor Law which envisages a role for a social security scheme. Such a scheme would be simple to administer, would have an immediate impact as there are no contribution conditions, and would help to establish social security principles in the private sector. Initially, the scheme could be limited to employers with at least 10 employees. 86. The scheme should cover accidents arising out of and in the course of insurable employment and occupational diseases but could also, optionally, be extended to include accidents occurring between an insured person's home or the place where he takes his meals and the place where he works. Normally, such a scheme would provide protection against the cost of medical treatment relating to the injury or disease, loss of earnings arising from temporary incapacity, permanent total or partial disablement, and death. It is proposed, however, that employers should retain direct responsibility for the payment of wages during the first 30 days of incapacity in order to ease the administrative burden on the new scheme. The cost of medical treatment could, however, be met from the scheme but it would be necessary to make arrangements with suitable health facilities, which could be accredited to the scheme following satisfaction of prescribed conditions. 87. The scheme could be financed by employers through a contribution based on a percentage of the payroll. There should be no contribution deducted from workers for the financing of a benefit which is seen as the direct responsibility of employers. The contribution should be paid to an employment injury fund to be administered by the social security scheme and from which benefits should be paid direct to beneficiaries. A decision would have to be taken as to the basis for the contribution liability. There are essentially three choices: (i) A standard percentage of the payroll applicable to all employers (ii) A differential rate by industrial group which takes account of the perceived different degrees of risk among industrial sectors (iii) A merit rated system which is based on (ii) but with modification in the light of accident experience. 88. Option (iii) is related to the objective of encouraging accident prevention measures by employers. This should, in any event, be an objective of the scheme, and some resources should be devoted to this and liaison should be established with officials responsible for occupational safety and health. This option does, however, require a sophisticated system for monitoring accidents and a close linkage with accident prevention and may be too ambitious for a newly established scheme in a developing country, where occupational health and safety standards are undeveloped. Option (ii) would be consistent with the approach taken by the insurance company AGL, which has four different premiums based on industrial risk perceptions. Option (i) would be the simplest to administer and if commuting accidents are included the application of industrial risk rating would be inappropriate. 89. Maternity Benefits. Given the high maternal mortality and infant mortality rates in Lao PDR, it is important to ensure that women have access to adequate medical care during their pregnancy and their confinement. It has been noted that some private sector employers are not - 203 - applying the Labor Law provisions and that there is also some evidence of discrimination against young married women. In these circumstances, it would be reasonable to include within the scope of the private sector social security scheme some provision for the medical treatment of pregnant women and for the payment of an allowance as compensation for presumed loss of earnings for a three-month period before and after the confinement. As in the case of employment injury, treatment would have to be provided at an accredited health facility. 90. Sickness Benefits. In the public sector, the same principles as for maternity and employment injury are applied. Health care reimbursement from the Social Security Fund reached 36 million kip in the first quarter of 1994/95. It is considered that this expenditure should not be charged to the Social Security Fund and should be met directly by the Government. The establishment of a health insurance scheme should be considered in the future but its feasibility is dependent on general improvements in the quality of health care facilities and a more general acceptance that such facilities can be supported by cost-sharing systems. 91. In the private sector, it would be consistent with the other proposals suggested above to provide some modest protection in respect of the consequences of sickness as regards both loss of earnings arising from temporary incapacity and medical treatment. But it would be inadvisable to initially provide general protection for all periods of incapacity and, in any event, the Labor Law places this responsibility on employers for at least the first 30 days of sickness in any year. It is thus recommended that the social security scheme should only assume responsibility for periods of incapacity beyond 30 days and should provide a sickness benefit of perhaps 60% of average earnings and medical treatment from an accredited provider. 92. Death Grants. The proposed social security scheme for the private sector could also include provision for the payment of a death grant intended to cover the cost of financial expenses for a deceased insured person. The amount could be 6 x the average monthly earnings of the deceased. 93. Administration. Under the present arrangements, responsibility for social security at both the policy and operational levels lies with the MLSW. The Ministry was established in July 1993 by Decree 04/PM. The Decree also covers the internal structure of the Ministry and provides for the establishment of six departments and for a network of administration in each province. Social security cuts across the division of responsibility between several of the departments, thus: * The Department of Social Security is responsible for managing the "pensions fund" and the "social security fund" * The Department of Pensions is responsible for "studying, guiding, and inspecting provisions and policies for retired staff' and for coordinating with the Prime Minister's Office on anticipated retirements pensions * The War Veterans Department is responsible for policies and provisions for war veterans * The Department of Labor is responsible for labor and for monitoring performance at the work place on various aspects, including social security * The Department of Social Welfare is responsible for policies relating to the poor, the handicapped, and other vulnerable groups. - 204 - 94. The justification for the departmental distinction between Social Security and Pensions is not sufficiently strong. Both departments are small and there is a need to establish a social security unit responsible for all aspects of policy and operations. There is a clear linkage between the award of pensions and their payment, among the records necessary to identify those in personable employment, those receiving pensions, and the management of the fund. It is recommended that consideration should be given to merging these two departments. 95. Both departments are new and their staffs are still developing a basic knowledge of social security concepts and administration. Priority needs to be given to improving their capacity and to the efficient organization of their administrative functions. This is an essential foundation for the assumption of the new responsibilities which will gradually emerge as the plans for the development of social security progress through the policymaking process toward implementation. In particular, steps should be taken to establish a data base for all pensioners (initially in Vientiane and then covering the whole country) and to collect the basic information necessary to monitor the financing of the pension scheme. 96. The new social security obligations envisaged above will have additional administrative obligations. These include the establishment of two separate funds: = The Social Security Pensions Fund, which involves the following broad functions: * Establishing a national data base for all those public servants who are insurable under the social security pension fund * Establishing a data base for all pensioners of that fund; reviewing and revising, as appropriate, the arrangements for the payment of contributions to the fund and for the award and payment of pensions = The Social Security Benefit Fund (with two branches -- the employment injury branch and the short-term benefits branch), which involves the following broad functions: * Registering all employers liable to be covered under the proposed social security scheme for the private sector * Establishing systems for the processing, payment, and accounting of benefits * Managing the fund with regard to the collection of contributions due and the payment of benefits. 97. The basic objective should be for the social security system to be self-financed from its own income from contributions, and from interest from the investment of reserves not immediately required for expenditure on benefits or administration. In this respect, it will be important to separate those elements in the current social security program that are outside the scope of the self- financing mechanism. Thus while the system can be used for the payment of pensions to war veterans, the financing of those pensions should be specifically provided for by additional funding from the Government. 98. Responsibility for the administration of the proposed social security scheme requires very careful consideration. There are a number of factors which need to be taken into account: * The funding of the scheme would be based on contributions paid by employers and workers. - 205 - * The Government would have a role as employer, and thus as contributor, and also as the final guarantor for the scheme and for the promises of benefit entitlement that it effectively makes. * The scheme is intended to be self-financed but its success will depend heavily on the quality of its administration: it will need high caliber staff with specialist skills in such aspects as accounting, investment, and computerization. 99. It is accordingly recommended that the Government should establish an autonomous statutory body, perhaps called the Lao PDR Social Security Organization (SSO), responsible to the Minister of Labor and Social Welfare on policy matters but with its own Board of Directors representing the Government and the social partners. As a transitional measure, the SSO could assume responsibility only for the management of the Social Security Benefit Fund -- the proposed scheme for private sector employees. On this basis the public sector scheme could remain directly under Government administration until the time appears right to extend pension provisions to private sector workers. C. SUMMARY OF RECOMMENDATIONS Social Protection Policy Framework 100. It is considered that Lao PDR should develop a strategic framework for the future development of social protection. This should be consistent with the economic policies of the NEM and should reflect a new division of responsibility for the financing and provision of social protection. This should recognize that the state cannot be expected to assume responsibility for protecting the people against all social contingencies and hardships. Private sector employers and individuals should both play an important role, and emphasis should continue to be placed on the strength of the traditional support provided within families and communities. 101. While the current stage of development and the low level of incomes do not permit a major shift in this division of responsibility, policymaking for new social protection initiatives, such as the reform of the public sector scheme and the design of social protection arrangements for the private sector, should take account of this. Thus self-financed schemes should be developed, individual initiatives encouraged, and the responsibility of the state for directly providing social protection focused on basic social protection provision. 102. These principles are also applicable to the social protection of public servants. Current salary levels are low and provide some justification for high pension replacement rates. But there needs to be a gradual shift in priorities toward an improved salary structure which reflects higher levels of efficiency and skill and away from the network of protective conditions of service which inhibit labor mobility and individual initiative. 103. Social security schemes should also be designed to avoid disparities in treatment between public and private sector workers. The objective should be to gradually develop a common core of social provisions applicable to all formal sector workers, but at a level which envisages supplementary provision either from employers or individuals or both. 104. Policymakers in Lao PDR should take time to study the policy options. The introduction of a social security scheme, particularly one which is applicable to private sector workers, brings with it a new and wide range of administrative responsibilities. At present, human resource capacity is low and both caution and training will be required in large measure. It would be advisable to commence with a modest scheme excluding pensions and to adopt an administrative system which makes the - 206 - most of available skills and experience, for example, through the establishment of an independent agency. Outside the formal sector, living conditions are low in a country still dominated by subsistence agriculture and undeveloped infrastructure. Here there is no scope for social security in the traditional sense and priorities should be focused on satisfying basic needs, particularly as regards health, water, and sanitation. Improving income levels must be part of integrated rural development and poverty alleviation programs. The following summarized recommendations should be considered for both the public and private sectors. Recommendations for the Public Sector > Restructure the public sector pension scheme in accordance with the following objectives: * Establishment of a self-financed pension fund based on contributions from employer(s) and workers * Extension to the private sector in future * Separation of pensions financing from other benefits (war veterans, health care) o Revision of the pension formula with a link to the CPI and lower replacement rates over a transitional period * Revision of the system for determining insurable earnings * Improved provision for survivors and equal retirement age for women * Coverage for workers of SOEs. => Administration of the social security pensions fund. The Department of Social Security should administer the scheme pending the establishment of an autonomous agency. A tripartite advisory board should be established on an interim basis with representatives of the LFTU, the Department of Public Administration, the MOF, SOEs, and the Chamber of Commerce. The functions of the Department of Social Security should be integrated with those of the Department of Pensions. Recommendations for the Private Sector = Create a modest scheme for a social security benefit fund establishing the concept and principles of social security in the emerging private sector with the following objectives: * Initial focus as a supplement to the Labor Law * Provision for employment injury, sickness, maternity, and death (funeral grant) * Pensions provision to be included in a later phase. - 207 - 105. Administration of the social security beneft fund This fund should be established as a trust fund for those covered, under the supervision of a newly established and independent Social Security Organization (SSO) with a Board of Directors to advise the Minister of Labor and Social Welfare on policy issues, and to control the fund and safeguard the interests of contributors and beneficiaries. The Board should include representatives of the LFTU, the Chamber of Commerce and the Government (MoF, MLSW). The Board would appoint a Director to be responsible for the administration of the scheme. The staff of the SSO should comprise partly civil servants on loan and partly those recruited directly. Administrative Reforms 106. Strengthen the administration of social security. Steps should be taken to increase the capacity of the MLSW to monitor the financing and administration of the Social Security Fund. This applies to the Departments of Social Security, Pensions, and War Veterans. Each department is responsible for part of the administrative process, but there is a need for their respective functions to be more closely integrated and for those functions to be developed so that the Ministry is able to achieve the following: * Identify its pensioners by category of pension * Determine expenditure on pensions * Ensure that all contributions payable are received * Carry out a regular analysis of the financial status of the scheme. It is recommended that priority should be given to the following: v Establishing a data base for all pensioners, starting in Vientiane e Revising procedures for the collection of information on award, financing, and payment of pensions * Introducing computerization for the management of the data base * Reviewing and integrating the functions of the Departments of Social Security and Pensions. - 209 - STATISTICAL APPENDIX TABLE 1: NATIONAL BUDGET REVENUE, 1988-1995/96 1988 1989 1990 1991 1992 1/ 1992/93 1993/94 1994/95 1995/962/ Compounded Buoyancy Growth Rates w.r.t. GDP (Million kip) Total Revenue 28,531 35,556 60,960 74,672 90,456 113,256 135,818 165,502 217,362 28.9 1.0 A. Tax Revenue 21,474 27,421 37,644 54,355 63,513 85,928 106,730 134,699 175,958 30.1 1.1 1.IncomeandProfitTaxes 6,288 6,196 7,956 14,597 17,604 13,325 18,631 28,022 34,236 23.6 0.8 Profit Tax 6,288 6,196 7,956 13,274 14,615 8,552 11,476 17,345 20,723 16.1 0.6 Income Tax neg. neg. neg. 1,323 2,989 4,773 7,155 10,677 13,513 59.2 2.1 2.AgricultureLandTax 1,213 1,358 1,520 2,041 1,604 1,235 2,187 1,630 2,116 7.2 0.3 3.DomesticGoods&ServicesTax 7,352 10,918 7,145 11,867 15,051 18,663 24,982 33,017 49,454 26.9 1.0 TumoverTax 7,352 10,916 7,145 9,994 11,717 14,265 19,910 27,557 33,946 21.1 0.8 Excise Duties nil nil nil 1,873 3,334 4,398 5,072 5,460 15,508 52.6 1.9 4. Intemational Trade Taxes 6,463 8,588 18,036 20,728 24,848 25,068 33,780 38,668 46,656 28.0 1.0 Import Duties 4,846 7,127 5,364 9,976 14,104 21,980 25,700 33,767 40,632 30.4 1.1 Export Duties 1,617 1,461 12,672 10,752 10,744 3,088 8,080 4,901 6,024 17.9 0.6 5. Natural Resources Taxes na na na 4,017 2,794 23,550 21,757 27,371 35,285 54.4 2.0 Timber Royalties na na na na na na 21,000 26,727 34,584 28.3 1.0 TaxesonOtherResources na na na 4,017 2,794 23,550 757 644 701 6.OtherRevenue 158 361 2,987 1,105 1,612 4,087 5,393 5,991 8,211 63.9 2.3 BusinessLicenses 30 54 1,160 136 120 115 128 184 138 21.0 0.8 Registration Fees na na na na na 855 2344 na 1,497 Other 128 307 1,827 969 1,492 3,117 2,921 5,163 6,576 63.6 2.3 B. Non-tax Revenue 7,057 8,135 23,316 20,317 26,943 27,328 29,088 30,803 41,404 24.8 0.9 I. State-owned Enterprises 2,667 2,746 10,889 8,511 9,287 9,590 7,666 8,003 9,070 16.5 0.6 DepreciationandDividends 2,667 2,746 4,589 4,755 4,138 4,695 3,024 2,980 5,231 8.8 0.3 Leasing Fees na na 6,300 3,756 5,149 4.895 4,642 5,023 3,839 -7.9 -0.3 2.On-lendingofForeignFunds na na na 5,821 5,958 7,004 11,999 10,618 15,483 21.6 0.8 InterestandCapitalReturn na na na 5,821 5,958 7,004 11,199 10,618 15,483 21.6 0.8 3. State Assets 4,390 5,389 12,427 5,985 11,698 10,734 9,423 12,182 16,851 18.3 0.7 Concessions na na na 1,071 345 553 375 397 396 -18.0 -0.6 Overflight Rights na na na 4,618 9,746 8,360 7,236 7,898 9,790 16.2 0.6 Other 4,390 5,389 12,427 296 1,607 1,821 1,812 3,887 6,665 5.4 0.2 Memorandum Items (Billion kip) Sale of State Assets na na na 4 1 15 7 2 15 27.6 1.0 NominalGDP(FY) 229 427 613 722 848 942 1,069 1,323 1,630 27.8 1 Total GovemmentExpenditures 74 104 123 151 192 203 253 352 362 21.9 0.8 Current 27 37 67 82 101 109 131 163 165 25.4 0.9 CapitalandOn-lending 47 66 56 69 91 94 122 189 197 19.6 0.7 Grants .. .. .. .. .. 31 67 73 58 I/ In 1992, the fiscal year changed from January-December to October-September. 2/ Data for 1995/96 are preliminary. Abbreviations: na - not available; neg. - negligible; nil - nil. Sources: Govemment authorities and World Bank staff estimates. -210 - TABLE 2: COMPARATIVE REGIONAL KEY ECONOMIC INDICATORS GDP GDP Per cap. Savings Exports Inflation Per cap. Foreign Pop. (PPP) growth GNP % of GDP 12 mos. CPI GDP debt (m.) China $3,172b 9.8% $435 36% $135.0b. 9.3% $2,660 $100.0b. 1,21 Japan $2,668b 0.6% $38,750 34% $443.0b. -0.4% $21,350 0 125. India $1,180b 5.3% $310 24% $29.9b. 9% $1,280 $85.2b. 928. Indonesia $709b. 7.4% $780 38% $45.4b. 10.5% $3,690 $90.0b. 196. South Korea $468b. 9.9% $8,550 35% $125.0b. 5.1% $10,534 $18.0b. 45.2 Thailand $408b. 8.5% $2,315 37% $54.2b. 7.4% $6,870 $62.1b. 60.7 Taiwan $279b. 6% $11,236 27% $112.0b. 3.7% $13,235 0$ 21.5 Philippines $185b. 4.5% $1,010 15% $15.8b. 11.8% $2,800 $37.3b. 68.1 Malaysia $171b. 9.3% $3,530 34% $68.7b. 3.4% $8,763 $23.3b. 20.2 Bangladesh $159b. 4.5% $220 7% $3.5b. 6.4% $1,350 $14.8b. 122. Hong Kong $140b. 4.2% $21,558 30% $172.0b. 5.9% $23,080 0$ 6.3 Vietnam $93b. 8.5% $220 7% $4.3b. 9% $1,263 $24.7b. 75.0 Singapore $66b. 9.1% $22,520 48% $118.0b. 0.9% $21,493 0$ 3.1 Sri Lanka $53b. 5.7% $635 16% $3.6b. 11.2% $3,030 $6.4b. 18.2 Myanmar $31b. 6.4% $890 12% $0.8b. 21.8% $676 $5.3b. 47.0 Cambodia $12b. 4.9% $215 8% $0.3b. 18% $1,266 $1.6b. 10.1 Lao PDR $9.7b. 8% $325 4% $0.3b. 6.7% $2,071 $1.2b. 4.8 ; Net creditor. - 211 - TABLE 3: PRELIMINARY BUDGET REVENUE PROJECTIONS PREPARED BY THE GOVERNMENT OF LAO PDR (MILLION KIP) ~~~~~~~~~~~~~~~~~~~~*.. / . * * s.s ............... ... 1995/96 1/ 1996/97 1997/98 1998/99 1999/2000 Taxes 175,958 210,818 214,700 244,200 277,800 Profit tax 20,723 29,330 31,000 35,700 41,700 Income tax 13,513 16,600 15,000 17,300 20,200 LandTax 2,116 3,150 6,000 7,000 8,000 Tumover tax 33,946 40,300 42,000 50,000 57,000 Import tax 40,632 47,480 55,000 66,000 75,000 Export tax 6,024 5,610 6,500 6,800 7,000 Timber royalties 34,584 33,000 29,700 31,200 32,800 Other 24,420 35,348 29,500 30,200 36,100 Nontaxes 41,404 47,160 75,300 100,800 129,200 Leasing 3,839 4,910 2,100 2,200 2,300 Interest and Amortization 15,480 17,112 6,600 6,700 6,800 Others 22,085 25,128 66,600 91,900 120,100 Total revenue 217,362 257,978 290,000 345,000 407,000 (Percent of PER's GDP) Revenue 13.3 13.6 13.3 14.1 14.9 Tax 10.8 11.1 9.8 10.0 10.2 Nontax 2.5 2.5 3.4 4.1 4.7 Expenditure 22.2 24.4 24.3 24.6 24.2 Current 10.1 10.8 10.4 10.6 10.2 Capital 12.1 13.6 13.9 14.0 13.9 Overall Balance (Exl. Grants) -8.8 -10.8 -11.1 -10.5 -9.3 Memorandum Item Nominal GDP (Billion kip) 1,630 1,901 2,187 2,445 2,734 1/ 1995/96 numbers are actual estimates while the remaining years' numbers are original government projection. 2/ Revenue does not include asset sales of 14,525 and 18,712 in 1995/96 and 1996/97 respectively. Source: Lao government authorities (CPC and MoF). - 212 - TABLE 4: LAO PDR - PROGRAMMED BUDGET FOR PIP, ONGOING AND NEW PROGRAMS 1995/96-1999-2000 (BILLION KIP) Total '95/96 1995/96 1996/97 1997/98 1998/99 1999/00 '999/960 Agriculture & Forestry 23.4 28.9 34.1 38.5 44.0 168.9 Research & extension 3.6 3.5 4.0 2.5 3.6 17.2 Ongoing & committed 3.6 2.8 2.7 - - 9.1 Research & extension projects 3.6 2.8 2.7 - - 9.1 New programs - 0.7 1.3 2.5 3.6 8.1 Research extension & crop promotion - 0.4 0.7 1.1 1.4 3.5 Programs Rural credit & cash cropping programs - 0.4 0.7 1.4 2.2 4.6 Irrigation 7.2 9.5 10.2 10.6 10.8 48.3 Ongoing&committed 2.2 1.7 1.1 0.7 0.7 6.4 Medium scale system 1.4 1.0 - - - 2.4 Small scale system 0.7 0.7 1.1 0.7 0.7 4.0 New programs 5.0 7.8 9.1 9.9 10.1 41.9 Support for small scale system - 0.7 1.0 1.4 1.4 4.6 Rehabilitation of existing system 1.8 2.8 2.7 2.8 2.9 13.0 Small and medium scale system 3.2 4.2 5.4 5.6 5.8 24.3 Forestry 8.3 9.9 9.4 11.3 11.5 50.4 Ongoing & committed 3.2 4.9 4.7 5.6 5.8 24.3 Re-afforestation 1.1 1.4 1.3 1.4 1.4 6.7 Conservation & Management 2.2 3.5 3.4 4.2 4.3 17.6 New programs 5.0 4.9 4.7 5.6 5.8 26.1 Re-afforestation 5.0 4.9 4.7 5.6 5.8 26.1 Livestock - 0.7 1.3 1.4 1.4 4.9 Ongoing & committed - - - - - New programs - 0.7 1.3 1.4 1.4 4.9 Disease Control programs - 0.4 0.7 0.7 0.7 2.5 Breed - 0.4 0.7 0.7 0.7 2.5 Other Agriculture 4.3 5.3 9.1 12.7 16.6 48.0 Ongoing & committed 4.3 5.3 9.1 12.7 16.6 48.0 AreaDevelopments 3.6 4.2 8.1 11.6 15.5 43.1 Narcotics reduction related development 0.7 1.1 1.0 1.1 1.1 4.9 Industry - excluding Electricity 9.3 13.2 16.3 18.5 20.0 77.3 Ongoing & committed 4.7 4.9 5.8 - - 15.4 Credit for private industrial sectors firms 4.7 4.9 5.8 - - 15.4 New programs 4.7 8.2 10.6 18.5 20.0 61.9 DGM institutional strengthening 1.6 1.6 1.9 2.5 2.5 10.1 Institutions for trade & skills transfer 1.6 1.6 1.9 3.7 5.0 13.8 Credit for private industrial sector firms 1.6 4.9 6.7 12.3 12.5 38.0 Electricity 1/ 21.0 23.1 26.9 32.1 37A 140.4 Ongoing& committed 18.7 13.2 5.8 - - 37.6 Medium & large generation projects 9.3 8.2 5.8 - - 23.3 Distribution network extension 4.7 - - 4.7 Small scale isolated generation systems 4.7 4.9 - - - 9.6 New programs 2.3 9.9 21.1 32.1 37.4 102.8 Distribution network extension 2.3 8.2 11.5 17.3 20.0 59.3 Small scale isolated generation systems - 1.6 9.6 14.8 17.5 43.5 (continued) - 213 - 1995/96 1996197 1997/98 1998/99 1999/00 Total '95/96 '9912000 Transportation 63.4 75.5 83.0 83.6 82.1 387.6 Roads 58.3 66.7 72.0 71.5 70.6 339.1 Ongoing & committed 41.6 36.4 23.3 20.1 11.3 132.7 Rt. #13 N Vientiane-Luang Prabang-Pakmong 11.6 7.1 - - - 18.7 Rt. #13 S Vientiane-Cambodia, pavements 4.8 10.1 9.5 8.5 6.0 39.0 Rt. #13 S Vientiane-Cambodia, bridges 2.9 3.0 4.3 3.9 5.3 19.4 Other national road links-pavement 14.5 10.1 9.5 7.7 - 41.9 Other national roads-bridges 7.7 6.1 - - - 13.8 New programs 16.7 30.3 48.7 51.4 59.3 206.4 Second Mekong Bridge-S. Laos (Completion 2002) - - - - 6.8 6.8 Other national links - C. & S. Thailand, Vietnam 1.9 4.0 7.6 7.7 3.8 25.1 Regional links North to Thailand, Vietnam - 3.0 7.6 9.3 11.3 31.2 Rt. #1 2.9 5.1 7.6 6.2 4.5 26.3 Other national roads - - 1.9 2.3 3.8 8.0 Feederroads(province) 2.9 4.0 5.7 7.0 9.1 28.7 Feeder roads (rural development program) 5.8 10.1 13.3 13.9 15.1 58.3 Road maintenance (national) 1.5 2.0 2.9 3.1 3.0 12.4 Roadmaintenance(provincial) 0.7 1.0 1.1 1.2 1.1 5.1 Bridge maintenance 1.0 1.0 1.0 0.8 0.8 4.5 River Transport 1.3 1.7 1.5 1.2 0.9 6.6 Ongoing & committed 0.5 0.6 - - - 1.1 Hydrographic surveys - - - - Port reconstruction 0.5 0.6 - - - 1.1 New programs 0.8 1.1 1.5 1.2 0.9 5.6 Surveying and installation Nav. aids 0.2 0.2 0.2 0.2 0.2 0.9 River channel improvements 0.3 0.3 0.3 0.2 0.2 1.3 River port rehabilitation/upgrading - - 0.3 0.2 0.2 0.7 Bank stabilization 0.3 0.6 0.8 0.6 0.3 2.6 Aviation 3.9 7.1 9.5 10.8 10.6 41.9 Ongoing & committed 3.9 7.1 7.6 8.5 6.8 33.9 Navigation system improvements 1.0 1.0 - - - 2.0 Airport rehabilitation 1.9 3.0 4.8 4.6 2.3 16.6 Masterplan implementation 1.0 3.0 2.9 3.9 4.5 15.3 New programs - - 1.9 2.3 3.8 8.0 Navigation system - - 1.9 2.3 - 4.2 Airport rehabilitation - - - - 2.3 2.3 Terminal rehabilitation - - - - 1.5 1.5 Railways-No public commitments - - - - - - Telecommunications 8.4 11.6 12.4 13.9 14.7 61.1 Ongoing & committed 5.5 2.5 1.9 2.7 1.9 14.5 Trunk links 1.9 - 1.0 1.2 0.8 4.8 Exchange systems 0.7 0.5 1.0 1.5 1.1 4.8 Intemational links (private investment) - - - - - Rural telecommunications 2.9 2.0 - - - 4.9 New programs 2.9 9.1 10.5 11.2 12.8 46.5 Regional intemational links - 2.0 2.9 3.9 3.8 12.5 Domestic trunk link extensions 1.0 2.0 1.9 1.5 1.5 7.9 Exchange systems in remaining provinces 1.0 2.0 1.9 1.5 1.5 7.9 Subscriber network installation 1.0 1.5 1.9 1.9 2.3 8.6 Rural telecommunication - 1.5 1.9 2.3 3.8 9.5 Other Infrastructure (MCTPC) 11.6 12.1 12.8 13.1 15.1 64.8 Ongoing&committed 11.6 9.1 5.7 3.1 - 29.5 Water supply systems 9.7 8.1 5.7 3.1 - 26.6 Public building construction 1.9 1.0 - - - 2.9 (coeduti%d) -214 - 1995/96 1996/97 1997/98 1998/99 1999/00 Total '95/96 '99/2000 New programs - 3.0 7.1 10.1 15.1 35.3 Water supply systems - 2.0 4.8 6.2 9.1 22.0 Public building construction - 1.0 2.4 2.3 2.3 8.0 Drainage systems - - - 1.5 3.8 5,3 Culture 3.8 4.9 6.0 7.3 8.5 30.4 Ongoing & commifted 3.8 4.9 6.0 7.3 8.5 30.4 Education 19.4 23.6 28.7 34.3 39.4 145.5 Ongoing & committed 18.1 13.4 12.2 7.3 6.0 56.9 Primary & secondary education 8.8 7.1 7.4 5.8 6.0 35.1 Technical & vocational training 4.1 3.1 2.4 1.5 - 11.0 National polytechnic 3.1 - - - - 3.1 Dong Dok tertiary institution 1.8 3.1 2.4 - - 7.2 Non-fonmal education 0.3 0.2 - - - 0.5 New programs 1.3 10.2 16.5 27.0 33.5 88.6 Primary & secondary education - 6.1 9.4 14.6 17.9 48.0 Technical & vocational training - - 1.8 4.4 7.4 13.6 Non-formal education - 1.0 1.8 2.9 3.0 8.7 Rural primary & vocational programs 1.3 3.1 3.5 5.1 5.2 18.3 Health 15.4 19.1 23.3 28.6 34.5 120.9 Ongoing & committed 8.0 1.9 0.9 1.0 1.0 12.8 Health & nutrition program 3.4 - - - - 3.4 Basic health services Savannakhet, Sekong 0.7 0.6 0.7 0.7 0.8 3.5 Other projects 3.9 1.3 0.2 0.2 0.3 5.9 New programs 7.4 17.2 22.4 27.7 33.5 108.1 Malaria control (IDA) 3.4 5.4 7.8 8.4 7.7 32.8 Vaccination - 1.1 1.7 1.2 1.3 5.2 Drinking water programs - 1.1 1.7 2.4 3.9 9.0 Sanitation - 1.1 1.7 2.4 3.9 9.0 Nutrition 0.6 1.1 1.1 1.2 1.3 5.3 Birth spacing & matemal & child health 0.6 1.1 1.1 1.2 1.3 5.3 Hospital&curative services - 1.1 1.1 3.6 6.4 12.2 New rural upland programs 2.9 5.4 6.2 7.2 7.7 29.3 Housing Offices 6.9 8.1 9.4 10.8 12.0 47.1 Social Welfare 8.6 17.2 20.5 23.5 26.8 96.5 Rural Development 5.6 21.8 30.8 38.7 46.9 143.8 Total 196.8 259.0 304.0 343.0 381.2 1,484.0 1/ Excluding govemment equity in large schemes. Sources: Department of Public Investment Plan, CPC. IMAGING Report No.: 16094 LA Type: SR