Malawi - Public expenditure review (Vol. 2 of 2) : Detailed report : statistical annex

Repowt No. 7281 -MAI Malawi Public Expenditure Review (In Two Volumes) Volume 11: Detailed Report Statistical Appendix A 7, 1990 Southem Africa Department FOR OFFICIAL USE ONLY DocuMt of ke WOrMd Baik This documert has a resricted distribution arnd may be used by recipients only in d per ce of dtheir oicial duties. Its cotents may not oherwise be dsclkxselwthout W,rd Bank audhorization.  CURRENCY AND OTHER EQUIVALLNTS Annual Averae Exchange Rate Kvacha ver US rnillar 1984i 1.413 1985: 1.719 1986: 1.861 1987t 2.209 1988: 2.561 1989: 2.822 WEIGHTS AND MEASURES 1 cubic foot (cu ft) - 0.0283 cubic meters 1 cubic meter (m3) - 35.3 cubic feet - 264 US gal 1 kilomater (km) - 1,000 meters - 0.621 miles 1 hectare (ha) - 10,000 square m4ters - 2.47 acres GLOSSARY OF ABBREVIATIONS LDD - Agricultural Development Division ADIa C Agricultural Development and Marketing Corporation cxs - Central Medical Stores DEVPOL - Statement of Development Policy 1986-1997 DLVW - Department of Lands, Valuation and Water DPMT - Department of Personnel and Management Training DSB - Department of Statutory Bodies DWSF - District Water Supply Fund EPD - Economic Planning and Development ESCOM - Electricity Supply Commission of Malawi FD - Forestry Department KCH - Kamuzu Central Hospital MDC - Malawi Development Corporation MFNR - Ministry of Forestry and Natural Resources MHC - Malawi Housing Corporation MOA - Ministry of Agriculture MOEC - Ministry of Education and Culture MOF - Ministry of Finance MOH - Ministry of Health MWS - Ministry of Works and Supplies GLOSSARY OF ABBREVIATIONS CON'T NRDP - National Rural Development Programce wcn - National Statistical Office NTC - Northern Transport Corridor PRAM - Private Hospital Association of Malawi PBC - Primary Health Care PSIP - Public Sector Investment Programme FVHO - Private Vehicle Hire Organization QEHC - Queen Elisabeth Centre Hospital RBM - Reserve Bank of Malawi SAL - Structural Adjustment Loan TEA - Traditional Housing Area - Not applicable - Not available FISCAL YEAR 41 1 '-r st  FOR Or. ILAL Uba. UNLY MALAWI PUBLIC EXPENDITURE REVIEW TABLE OF CONTENTS VOLUMF I Page No. Executive Summary ....................................................i Major Findings and Recommendations ................................... i VOLUME II I. Macroeconomic Framework for Public Expenditure Strategy ..... A. Diagnosis of Fiscal Imbalances .1 B. Medium-Term Framework .7 II. Intersectoral Public Resource Management Issues . . 13 A. Balance Between Recurrent and Capital Expenditures ..... 14 B. Planning, Budgeting, and Expenditure Control Processes.21 C. Wages and Employment ................................... 24 D. Debt ................................................... 32 E. Parastatals ............................................ 36 III. Sectoral Options ........................... ................ 41 A. Sectoral Envelopes ................ 41 B. Education ..................... 47 C. Health ..................... 55 D. Agriculture and Fisheries .............................. 68 E. Forestry ....................................... 81 F. Transport ....................................... 90 G. Water ....................................... 95 H. Buildings ....................................... 99 I. Housing ....................................... 100 J. Power ....................................... 104 Statistical Appendix ....................................... 108 This report is based on the findings of a core mission that visited Malawi in November/December 1987 and sectoral missions that visited Malawi between September 1987 and March 1988 and reviewed with the Government in October 1988 and June 1989. The missions consisted of Kathie Krumm (task manager. AF6CO), Fred King (AF6CO), Ranjit Teja (IMF), Ercis Kurtulus (Consultant), Tia Duer (AF6PH), Ola Storeng (FRM), Sudhir Shetty (AF6AG), Jane Armitage (AF6AG), Ken Clare (Consultant), Ebenezer Aikins-Afful, (AF6IN), Carolyn Gochenour (AFTIN), and Assefa Telahun (AF6IE). Mr. Steve O'Brien (chief economist) joined and advised the core (AF6CO) mission. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  INDEX OF TABLES Table PaRe No. VOLUIK I Maior Findings Table 1.: Macroeconomic Framework, 1988/89-1992/93 . . 6 Table 2.: Sectoral Programs: Comparison with Budget Shares ....... 13 VOLIME II I. Macroeconomic Framework I.A.1 Flow of Funds, 1980-88 ....................................... 2 I.A.2 Comparison of Original Estimate, Revised Approved and Actual Expenditures, 1982/83-1987/88 ..................... 6 I.B.1 Flows of Funds 1988/89-1992/93 Less Constrained Scenario .......................... 10 I.B.2 Flow of Funds, 1988/89-1992/93 Conservative Scenario .............................. 10 II. Intersectoral Public Resource Management Issues II.A.1 Development Budget Shares, Suggested Core Program as Percent of Sector Expenditures, Total Expenditures and GDP Comparison With Existing PSIP Commitments ........... 17 II.A.2 Budgetary Allocations Under Suggested Core Program .......... 18 II.C.1 Trends in Wages and Salaries Expenditures 1976/1977 - 1988/89 ........................... 25 II.C.2 Range of Increases in Nominal and Real Civil Service Wages, 1975-81, 1981187 and 1987-89 .. . . 28 II.C.3 Number and Increase in Civil Service Established Positions, 1980/81-1987/88 .................................. 29 II.C.4 Breakdown of Increases in Established Positions, 1985/86 - 1987/88 .. 29 II.C.5 Number and Budget for Non-Established Positions, 1l82/83-1986/87 .. ........... 30 II.C.6 Real Wage and Wage Share, Projections 1988/89-1992/93 ....... 31 II.D.1 Trends in Interest Service, 1980/81-1988/89 ................. 32 II.D.2 Projected Financing Shares of Fiscal Deficit by Source, 1989/90-1992/93 .................................. 34 II.D.3 Projections of Interest Service, Level, and Budget Share, 1988/89-1992/93 ...................................... 35 II.D.4 Schedule of Interest Due from Parastatal Sector ............. 35  INDEX OF TABLES Table Page No. III. Sectoral Options A. Sectoral Envelopes III.A.1 DEVPOLs Projected Sectoral Shares, 1988/89-1992/93 .......... 42 III.A.2.1 Projected Sectoral Shares of Expenditures, (Less Constrained Case) ................................... 43 III.A.2.2 Projected Sectoral Shares of Expenditures, (Conservative Case) ....................................... 44 III.A.3 Comparison of Projected Sectoral Expenditure Programs to Budget Envelope. 1988/89-1992/93 ......................... 46 B. Education III.B.1 Education Exr3nditures, Total Budget Share and Per Capita Years, 1983/84 - 1987/88 . . 50 III.B.2 MOE Recurrent Expenditures by Categories .................... 51 III.B.3 Education Investment and Recurrent Expenditure Projections, 1988/89 - 1992/93 .............................. 54 C. Health III.C.1 Health and Total Public Expenditures .. 58 III.C.2 Healths Overruns on Revenue Budget by Expenditure Category, 1983/84-1985/86 ..........63 III.C.3 Health Investment & Recurrent Expenditure Projections, 1988189 - 1992/93.. . ..... .................................. 65 III.C.4 Health Core Expenditure Options, 1988/89-1992/93 ... . ............... 69 D. Asriculture III.D.1 Agricultures Budget Shares and Expenditure Growth, 1984/85 - 1987/88 . ....... .... . . 72 III.D.2 Agricultures Revenue Budget Expenditures by Category ........ 73 III.D.3 Agriculture and Fisheries: Expenditure by Category and Program, 1987/88-1988/89 ................................ 73 III.D.4 Agriculture: Incremental Recurrent Expenditures, 1984/85-1987/88 ............................................. 75 III.D.5 Agriculture: Projected Expenditures, 1989-1992/93 .......... 76 III.D.6 Agricultures New Projects in PSIP .......................... 80 III.D.7 Fertilizer Subsidy Amount and Budget Share . .0........... S  INDEX OF TABLES Table Page No. E. Forestry III.E.1 Government Budgetary Allocaticns/Revenues: 1983/84 - 1987/88 ..................................................... 84 III.E.2 Index of Public Expenditure in Forestry: 1983/84 - 1987188 ..... 85 III.E.3 Revenue Budget by Program, 1985/86. - 1987/88 ........... 85 III.E.4 Share of Personal Emoluments in Revenue Budget .............. 86 III.E.5 Main Components of the Forestry Development Budget, 1987/88 ..... 88 III.E.6 Forecasts of Local Financing for Forestry Activities for Next 5 Years ............................................ 88 III.E.7 Forestry Expenditures Projections ........................... 89 F. Transort III.F.1 Recent Trends in Intrasectoral Allocation of Expenditures For Roads, 1983/84 - 1987/88 ................... 91 III.F.2 Development Expenditures by Sector, 1981/82-87/88 ........... 92 III.F.3 NonRoad Transport Recurrent ELpenditures,by Sub-sector, 1984/85 - 1987/88 .......................................... . 92 III.F.4 Transport Sector: Projected Expenditures, 1987/88 - 1992/93 ...... ............ .. 95 G. Water III.G.1 Public Investment in Water and Sanitation, 1982/83-1986/87 . ..................................... 97 III.G.2 Water Sector: Projected Expenditures ....................... 98 R. Building 111.1.1 Buildings Sector: Revenue Budget Expenditure and Index of Real Expenditure, 1984/85 - 1989/88 ............... 100 I. Housing 111.1.1 Malavi Housing Corp., Schedule of Current Rent Income and Fair Rent .............. ........................... 103 TII.T.2 Moutsina Sectr-r: ?ro4ectel T)e-veio"-ent Erpenditure Program .............., 10 J. Power III.J.1 Power Sector: Projected Expenditures, 1986/;--1992/93 .................. ,107  I. MACROECONOMIC FRAMEWORK FOR PUBLIC EXPENDITURE STRATEGY 1. Malawi has been subjected to a number of external shocks starting in the early 1980s that have adversely affected her prospects for growth and development. Foremost amongst these has been the war and civil strife in neighboring Mozambique which not only led to a severely debilitating increase in 'nternational transportation costs but also to a large influx of refugees. Together with sharp declines in the terms of trade in 1985 and 1986, these disturbances have imposed an enormous burden of adjustment on the Malawi economy. How have public expenditures responded to these developments and what are the implications for the medium term? As outlined below, the initial deterioration in the fiscal balance between 1985 and 1987 reflected these shocks. The expenditure pressures and need for fiscal restraint are expected to continue in the mediun-term. A. DIAGNOSIS OF FISCAL IMBALANCES. Back2round: 2. During the first four years of the decade, Malawi made considerable progress in reducing the size of the overall budget deficit (excluding grants) which declined from nearly 15 percent of GDP in 1980 to about 10 percent of GDP by 1983.1/ 3. With total revenues roughly constant over the period, the improvement was almost entirely attributable to the reduction in government expenditures which declined by 5 percentage points to 29.5 percent of GDP in 1983 (Table I.A.1), largely reflecting the completion of major construction projects in Lilongwe. At the same time, the external current account balance improved dramatically, declining by 10 percent of GDP to 11.0 percent of GDP in 1983. Thlus about half of the improvement in the current account could be traced to the decline in the budget deficit, the remainder being accounted for by a reduction in net private sector dissavings. The following year, 1984, witnessed a 1/ The fiscal year in Malawi runs from April 1 through March 31. However, to facilitate integration with balance of payments and national accounts data, this section expresses all fiscal variables on a calendar year basis. The remainder of the report expresses variables on a fiscal year basis. The methodology for the review is as follows. Budget data for a given fiscal year have three vintages -- original budget estimate (referred to as original or estimate or budget), revised budget estimate (referred to as revised), and actual expenditures (referred to as actual or not identified). In addition, the recent macroeconomic analysis is based on the Policy Framework Paper 1988/89-1990/91, including intermediate estimates of actual 1987/88 and of budget 1988189. Sectoral breakdown is available from budget data only. Intersectoral and intrasectoral analyses, often carried out before the actual 1986/87, revised 1987/88, and budget 1988/89 figures were available, may use less recent data.  -2- further decline in Government expenditures and the budget deficit, reinforcing a large increase in net private sector savings associated with exceptionally favorable external developments during the year. 4. The trend of declining Government expenditures and deficits was reversed in 1985 and again, more dramatically, in 1986. Although revenues in both these years were more than 1 percent of GDP higher thani before, expenditures rose by S percent of GDP to over 34 percent of GDP by 1986. Among the factors contributing to the surge in expenditure were substantial increases in the Government's wage bill, large extra- budgetary expenditures incurred with the build-up of a str.cegic grain reserve, and special security related expenditures. Not surprisingly, these developments--together with declining terms of trade--also led to a deterioration in the current account balance. By end 1986, gross official reservei were down to 3.7 weeks of imports as compared with 11.5 weeks two years earlier. 5- It i important to note, however, that the implications for the private sector in Malawi (defined here to include parastatals) were considerably more serious than suggested by this somewhat aggregative account. This is because in the setting of scarce foreign exchange availability, the burden of much tighter foreign excharge rationing- -starting in the second half of 1985--fell considerably harder on the private sector. Staff estimates of the share of the Central Gover.anent in total imports indicate a rising trend through the 1980s, with an especially large jump in 1985 (Appendix Table 10). Table X.A.1: FUN OF fLOW, 19001S (in S af MP) 1wo 1961 192 1ws 19"4 1965 III US1 1o Private Sector 1/ Saving 2/ 7.9 13.8 16.7 14.9 13.1 11.0 9.7 11.2 7.9 Inveatant 3/ 11.7 6.1 7.7 6.8 6.4 6.3 3.5 S.0 6.0 Stockbu; ling 2.8 2.5 6.6 9.1 -0.2 5.2 0.2 1.S 3.8 Net Pr?v'to Sevi ;e 4.3 2.9 2.2 -0.7 4.9 -0.8 6.0 4.4 -2.2 Covernment 4/ Revenues 19. 1i.5 19.1 .2 s. 7 21 2. 0.4 20.0 Ezpenitur 84.4 4.7 2.0 29.S 2S.4 80.0 34.2 2.7 26.5 of wui,c: Investment 10.5 7.0 6.9 7.2 6.7 6.5 6.5 5. 5.4 Sudget deficitX (excluding grunt) -14.9 -1S.2 -12.0 -10.8 -4.7 -4.S -12.8 -9.8 -6.5 External Accounts Current accont balance -21.2 -12.8 -10.6 -11.0 -1.0 -9.0 .4.5 -4.9 4.7 1/ Includoe Satutory diee. 2/ tncludes private tranafers. 3/ Private secter groee fixed "eptla for_tion. 4/ Calendar year basie SOLICE: iblawi Cotuntry Econoaic lomerandue. iacaI deficit. for 19U end 196S differ froa those in the CBE hbcause they arm bhad on deficit. as shown in the ablvi Conee; Reert 1909. Private *e4tor balance and the distribution of national aevinsa have beei adjusted accordinly .  -3- 6. Thus the crowding out of productive sectors of the economy- -induced by fiscal imbalances together with failure to switch Government spending away from imports towards domestically produced substitutes- -has been substantial and constituted a major drag on the economy. Indeed, an important cause of the slowdown in economic browth in 1986 was the unavailability of Imported raw materials and spare parts. Such shortages have continued to plague the manufacturing sector in 1987. The role of exogenous disturbances: 7. Having described the evolution of macroeconomic and fiscal aggregates, it is now time to relate these to the major exogenous disturbances that have buffeted the economy since 1982. Because the immediate effect of all the shocks that hit the economy was to reduce real income available to the private sector, they invariably led to a decline in private sector savings and capital formation and gave rise to a worsening of the current account balance (Table I.A.1). To the extent these shocks were expected to persist in the medium term, they warranted a reduction in the budget deficit to offset the deterioration in the current account. While this did indeed occur following the shock to the transportation sector in 1982, these gains were quickly reversed after 1984, unfortunately coinciding with another exogenous shock to the system (the worsening in the terms of trade in 1985 and 1986). This section argues, without justifying, that to a large extent, the deterioration in the fiscal balance between 1985 and 1987 was itself a reflection of the very exogenous shocks that hit the economy. (a) Transportation: 8. Prior to the disruption caused by insurgency in Mozambique in the early 19809, lalawi's two principal external transportation links were the Mozambican ports of Beira and Nacala, both accessed by rail. During 1982 and 1983, these routes came under frequent attack and became increasingly unreliable; by mid-1984 they had completely brcok.en down, forcing most of Malawi's externul trade through a circuitous route to the port of r-rban in South Africa. Depending on the commodity, this raised the price of imports and exports between 50-250 percent so that the c.i.f. margin increased from around 25 percent of c.i.f. value of imports in 1981 to 40 percent by 1984. According to a Bank staff report, the additional cost of the diversion to the economy as a whole- -directly, in the form of higher freight rates and, indirectly, in terms of loss of production and sales ari3ing from shortages and late delivery of essential inputs--was at least US$ 50 million or 4 percent of GDP in 1983, and even higher in subsequent years. Although many businesses adjusted to this shock by expanding regional trade, mainly with South Africa, the increase in transportation costs was a crippling blow to external trade. 9. While the effect of higher transportation costs was most damaging to the private sector, its ripples were also felt in the public sector through a number of direct and indirect channels that forced government expenditure upwards. First. higher transportation costs raised the Government's direct import bill. If the difference in  -4- freight rates between the Durban and Nacala routes is taken as a proxy for added costs due to the disruption (notionally, a differential of about MK 220 per metric ton in 1987/88), then the estimated 16,000 tons of Government imports would cost the Government an additional MK 3.5 million. Further, higher costs would have been reflected in higher prices for all tradable goods purchased by the Government. Assuming that the 15 percentage point increase in the c.i.f margin was fully reflected in prices, and, depending on tradable goods content of Government purchases, expenditures on 'other goods and services" (suitably adjusted to avoid double counting of Government imports) would have been between MK 12-25 million lower during fiscal 1987/88. These figures--which are quite robust to alternative methods of calculating the excess cost--imply an added cost to the Government ranging between 0.5-0.9 percent of GDP in 1987/88. 10. Secondly, the Government devoted considerable resources towards the eventual reopening of the Nacala line as well as facilitating the flow of traffic to Durban via Mozambique. While quantification of such costs is difficult. our initial estimate for such added expenditure is in the order of 0.75-1 percent of GDP in 1987/88. This figure was obtained by considering the excess of defense expenditure (including the separate budget for Nacala and extra budgetary expenditure) over and above the levels prevailing between 1982-84 and adding to that the expenditure incurred for track rehabilitation on the Nacala line. 11. Thirdly, higher transportation costs have led to larger transfers to a number of parastatals Vwose performance was adversely affected. The most obvious case is that of Malawi Railways which once relied ci traditional routes to Beira and Nacala for 60 percent of its traffic. Indeed, it has been estimated that if just the Nacala line alone were operating at normal capacity during 1987/88, Halawi Railways' revenues would be higher by MK 6.5 million, which would have obviated the need for most transf3rs and debt relief from the central Government. Again, to the extent that the Government failed to pass the higher landed cost of fertilizer to small scale farmers, transfers to the Fertilizer Revolving Fund have been higher than they would have been otherwise. 12. Thus the shock to the transportation sector has increased Government expenditures in the order of 1.5 to 2 percent of GDP during U37/88, with similar orders of magnitude applying the year before. These figures may be deemed to be distinctly on the conservative side since they exclude higher transfers to parastatals, induced by escalation in transport costs, which could not be estimatad. (b) Refugees: 13. Although the war in Mozambique continued through the 19809, the influx of refugees was light up until mid-1986, with many refugees being absorbed by families and relatives in Malawi. However, since then, the probleA has exploded into a crisis with the number of identified  -5- refugees increasing fivefold to 360,000 in the twelve months to November 1087. If these trends persist, as they are widely expected to, the number of refugees will exceed 500,000 by mid-1988. 14. Fortunately for Malawi, international donors have been forthcoming in providing timely assistance in terms of food and other essential supplies. As a result, the direct financial cost to the public sector has been small and the refugee situation-has not affected budgetary outcomes in a significant manner. Since mid-1987, the United Nations High Commissioner for Refugees (UNHCR) has been actively involved in funding relief work and will be reimbursing a number of Government departmints (e.g. the Ministries of Health and Education) and public sector enterprises (e.g. Lilongwe and Blantyre Water Boards) directly for provision of services to refugees. 15. Most of the additional burden of caring for the refugees has been incurred in an opportunity cost sense, with Mozambic4ns displacing Malawians in the consumption of public goods and services. Although the refugee population is small relative to the total, their numbers have been rising rapidly in most of the affected districts bordering Mozambique. In Nsanje district, for example, the ratio of refugees to Malawians now exceeds 90 percent. In such areas, the stress on local public services has been substantial. A few examples illustrate the point. In affected areas, district commissioners have been devoting about half their time on dealing with the refugees. The police has had to divert a number of officials to the task of maintaining law and order as well as assuming new tasks (such as escorting food distribution teams in the field). Likewise, since refugees utilize health services much more intensively than Malawians (they account for roughly one in three hospital admissions), the implied expenditure has not been negligible.2/ 16. At the time of this review, a comprehensive study of Government involvement had not been completed. The study recently completed by UNDP could provide such a base to impute a value to total expenditure thus incurred, to enable a better assessment of the impact of Government operations, and also to provide a means of eliciting more financial support for the international relief effo.:t. 17. From the above, it follows that at least half of the 4.7 percent of GDP increase in Government expenditures that occurr-d between 1984 and 1986--mainly arising from the shock to the transportation 2/ As an illustration, if it is assumed that refugees utilize health services about 50 percent more intensively on a per capita basis, then an average of 350,000 refugees in 1987/88 would cost the Government close to MK 4.5 million, a plausible figure to many health officials involved in the relief effort.  -6- sector--may be attributed to the effects of exogenous disturbances. This is based on an estimation of direct costs which presumably underestimates the impact. (ii) Other Factors 18. An important factor which increased expenditures during this period was the decision by the Government to develop a strategic grain reserve. Formal financial arrangements have been set up with ADMARC which acts as the Government's agent for this reserve. In 1985/86 and 1986187 the Government purchased maize from ADMARC totalling MK 26 million and MK 46.7 million, respectively, representing 1.2 and 1.9 percent, respectively, of GDP. The recent food deficit situation has resulted in the reduction in the reserve in 1987/88 and 1988/89, with some non-tax revenue accruing the Government for its sales of maize out of the strategic grain reserve. The required build-up of the reserve in the future is discussed further below. 19. Structural factors contributed to the remainder of the sharp expenditure increase but are by no means the dominant factor in explaining the expenditure imbalances experienced recently in Malawi. Structural problems are particularly important in the revenue budget which historically has been characterized by substantial upward revisions to the original approved budget. As shown in Table I.A.2, the increase in the revised approved budget over the original approved budget was 8.5 percent in 1983/84, increasing to over 12 percent in 1984/85. 1985/86 and 1986/87. Further overruns also have occurred in actual expenditures over approved amounts, on the order of 7 to 9 percent. In the recent period, this trend has been less proncunced as the Government shifted its focus to stricter budgatar-i control. It is noteworthy since this general tendency of overrunning both the approved and even the revised budget ceilings has been a common occurrence. 20. One of the major sectors which has experienced problems is health. As shown in Table I.A.2, the revised budget for this ministry consistently was substantially higher than the original approved budget, ranging from 18.9 percent in 1983/84 to 59.4 percent in 1985/86. The -: " I.A.2 C.p.m of fial.-i - Q..- t_ _ ,4_d l Ag sA_*, _ t 5/2/-16/ f '-I----- 4 -- -iusi._ -/ ti----_- ,-.___ _______------4oi/7---------. 0,l/up _1_ue g.Ile