Policy, Research, and External Affairs WORKINQ PAPERS Country Operations Southern Africa Department Afri-.a Regional Office World Bank August 1990 WPS 497 The Measurement of Budgetary Operations in Highly Distorted Economies The Case of Angola Carlos Elbirt A proper measurement of Angola's 1989 budget should reflect the tremendous difference between official and parallel market prices at which transactions are undertaken. If all transactions are "valued" at parallel market prices, the budget deficit fcr 1989 would drop from 22 percent to 12 percent. a l Thc Policy, Research, and External Affairs Complex disunbutes PRE Working Pape todisseminatethefindings of work in progress and to encourage the exchange of ideas among Bank staff and all othes interested in development issues. These papers carry the names of the authors, mfilct only their views, and should be used and cited accordingly. The findings, inteipretations, and conclusions are the authors' own. They should not be attributed to the World Bank, its Board of Directors, its management, or any of its member countries. Policy, Research, and External Affainr| Country Operations WPS 497 This paper- a product of tne Country Operations Division, Southern Africa Department, Africa Regional Office - is part of a larger effort in the World Bank to study a highly distorted economy with a view to identifying appropriate development policies. Copies are available free from the World Bank, 1818 H Street NW, Washington DC20433. Please coniactTerry Gean, room J11-250,extension 34247(16 pages). In a highly distorted economy such as Angola's, Should Angola's exchange rate and prices be budget accounts can be misleading - because liberalized, the budget for 1989 would tend to prices in the parallel market, including the look like this adjusted budget - because the exchange rate, represent up to 100 times official conversion prices used to adjust the budget prices. resemble matket prices. So the impact of policy adjustment would be as follows: Parallel prices are the real opportunity costs for consumers and guide them in their decisions. * The deficit would decline from 22 percent So budget accounts may not truly show the of GDP toward 12 percent, depending on the resources they are supposed to measure. extent of price and exchange rate adjustments. Nominally the budget deficit would increase, but Angola's government collects taxes and pays that would be a mere accounting change. expenses in two currencies: strong kwanzas (with attached buying rights, such as access to * Any additional reduction of the budget hard currencies or goods) and weak kwanzas deficit would require active budget policies. In (which must be used in the parallel market). Angola, the wage bill and extrabudgetary expenditures account for more than half of In adjusting Angola's 1989 budget, Elbirt government spending, so those items would have assumed that all tax revenues or transfers from to be cut. Personnel costs account for more than oil companies were in strong kwanzas, all other 16 percent of GDP: cutting them by 10 percent taxes and revenues in weak kwanzas, and would reduce the deficit by more than 10 per- expenditures varied. He found the composition cent. of revenues in the adjusted budget to be totally different from the nonadjusted one. Oil revenues * Wage remonetization should be neutral in represent 98 percent of revenues, not 48 percent, terms of the nominal deficit of the consolidated as in the original budget. So dependence on oil public sector (the central government, revenues is underestimated. parastatals, and financial institutions). The extra betief:ts parastatals would enjoy from price The composition of spending also changes, liberalization would compensate for the central but not as radically. Wages remain the largest government's extra spending on wages, if price item. Extrabudgetary items rank second, rather liberalization and wage remonetization are than third, among expenditures - at 24 percent, strictly synchronized. not 12 percent. The PRE Working Paper Series disseminates the findings of work under way in the Bank's Policy, Research, and Extemal AffairsComplex. An objective of theseries is to getthese fundings outquickly, evc a if presentations are Icss than fully polished. The findings, interpretations, and conclusions in these papers do not necessarily represent official Bank policy. Produced by the PRE Dissemination Center THE HEASUREMENT OF BUDGETARY OPERATIONS IN HIGHLY DISTORTED ECONOMIES: THE CASE OF ANGOLA by Carlos Elbirt A. Introduction In highly distorted economies budgetary accounts could be very misleading. In Angola for example, prices in the parallel market, including the exchange rate, represent up to one hundred times official prices. Access to official prices depend on special rights ("cartaosw) enjoyed only by formal sector workers. Some goods obtained at official prices are traded in tne parallel market. Therefore, budgetary accounts do not necessarily provide a good picture of the real resources they are supposed to measure. Parallel prices are the real opportunity costs for consumers and guide them in their decisions. The Government collects its taxes and pays its expenLses for all practical purposes in two currencies: Kwanzas that have attached buying rights, such as access to hard currencies or goods, and Kwanzas which have to be used in the parallel market. A Ministry may have the budget to hire some personnel but will not find anybody at official wages unless some buying rights are offered as part of the wages (at parallel prices, wages are ridiculously low). In a general sense, a more accurate measurement of the budgetary accounts should help policy makers at least in three aspects. First, to 4 2 know with more precision the relative size of the Government. Second, to know more accurately the relative size of the Government deficit. Third, to provide a more accurate picture of the composition of both revenues and expenditures. An appropriate measurement of the budgetary account could be crucial in determining the type of policies required by the country. B. A "Conversion" Problem Proper measurement of the budget in Angola requires "homogenization". We can assume that, for all practical purposes, there are two "currencies": Strong Kwanzas (S) which are the ones attached to special buying rights at official prices and Weak Kwanzas (W), with access only to the parallel market. Sn* W n - Pp/Po Where n is the conversion rate between strong and weak kwanzas. It could be the ratio between parallel (Pp) and official prices (Po). Strictly speaking, for measurement purposes, it should be ratio between the "shadow" exchange rate (or prices) and the official one. Each budgetary item, revenue or expenditure, can be converted into weak or market Kwanzas using the following equation: *I - a * I + (1-a) * I * n 3 where *1 - the adjusted budgetary amount for a particular item I - is the original, non adjusted budgetary amount a - is the proportion of weak Kwanzas in the budgetary item (1-a) - is the proportion of strong Kwanzas n - is the conversion factor The GDP can also be expressed in adjusted Kwanzas (*GDP) thus allowing us to show the budget as percentage of *GDP. Since, in the case of Angola, tentative GDP estimates have been done in US$, its adjustment into weak or market Kz is quite simple for a given conversion factor. In this exercise we will convert the budget for 1989, into weak or parallel market price Kwanzas. Extra-budgetary expenditures are also included in the conversion. It is rather difficult to estimate n in view of the absence of comprehensive price information for Angola. However, for indicative purposes we will use a variable n in this exercise (from, say, 10 to 66; which are consistent with parallel exchange rates of 300 to 2000)1/. A/ n - (Pp/Po) where Pp - pirallel prices Po - official prices (- 30 - official exchange rate) If Pp- 300 : n- (300/30)m 10 If Pp- 2000 n- (2000/30)= 66.66 4 In order to estimate a, the following criteria were followed: (a) all taxes on oil or transfers from oil companies were considered, in a simplifying assumption, to be 100 percent strong Kwanzas. The reason for this assumption is the fact that those revenues have, as counterparts, dollars in the Banco Nacional de Angola. (b) all other taxes and revenues were considered weak Kwarzas. We assume that tax payers ot'ler than oil companies, pay their taxes after bartering and wbridging* the official and parallel markets. Since we are looking at the real resources behind each budgetary item, it appears that their payments to the Government are all in weak Kwanzas. (c) the composition of each expenditure item in weak and strong Kwanzas was based on the import-component attributable to them (See Table 2.3). Wages have about 35 percent in Strong Kz, according to estimates carried out in another work2/. Extrabudgetary expenditures, debt service and war-related payments (mostly imported war material), are considered to include a 100 percent import-component. They are, therefore, made up all of strong kwanzas. 2/ Angola: Selected Issues on Adjustment, in preparation. 5 TABLE 2.3. ANGOLA: FOREIGN EXCHANGE COMPONENT OF CENTRAL GOVERNMENT EXPENDITURES (2 IN 1980 BUDGET) Service commissions 60.0 Personnel training 65.0 Medical supplies 100.0 Medical services 100.0 Other services 10.0 Food 80.0 External cooperation 60.0 Scientific research 100.0 Representation allowances 100.0 Visitors and conferences 10.0 Other current expenses 30.0 Capital expenditures 60.C Source: Ministry of Finance, Mission estimates. C. The results of the "conversions' The results of adjusting the budget for 1989 are shown in Tables 2.4 and 2.5. Table 2.6 shows the overall results of using a parallel exchange rate of 2,000 (n-66) and 300 (n-10). We will focus on the results using a parallel exchange rate of 2,000. The choice is based upon the fact that 6 this exchange rate is close te the parallel market rate3/. Several conclusions were reached frou analyzing the adjusted budgett (a) the composition of revenues in the adjusted budget is totally different from the non adjusted one. Oil revenues represent 98 percent of the total revenues as opposed to 48 percent in the original, non adjusted budget. The actual dependence on oil revenues is, therefore, underestimated in the non-adjusted budget. (b) the composition of expenditures also changes, but not as radically as revenues. Wages continue to be the largest item. Extrabudgetary expenditures are more important than originally shown: they represent 24 percent of the total expenditures instead of 12 as originally thought and rank second instead of third among all the expenditures. Table 2.7 shows the structure of expenditures and the "rank' of each item before and after the adjustment. 3/ Strictly, we should use the shadow exchange rate. However, it is not possible, at this stage, to estimate the shadow exchange rate in Angola due to the lack of information. Am advantage of the parallel market rate is the fact that it measures the opportunity costs for consumers prevailing at the present moment in the economy. In any event, as it is referred to bellow, the conclusions will not be altered when different rates covering a wide range are used. 7 TABLE 2.4. ANGOLA - 1989 GOVERNMENT REVENUES AT PARALLFL PRICES (Billions of Kz.) z z Budget 1989 "Strong' "Soft" In soft Kz Z Budget GDP 'otal Revenue 81.1 47.0 53.0 2,576.4 17.4 Tax Revenue 61.0 62.0 38.0 2,556.3 17.3 Taxes on Oil Prod. 38.0 100.0 0.0 2,533.3 17.1 Petroleum Output 11.0 100.0 0.0 733.3 5.0 Profits of Comp. 14.0 100.0 0.0 933.3 6.3 Other 13.0 100.0 0.0 866.7 5.9 Taxes on Int. Trade 6.8 0.0 100.0 6.8 0.0 Customs Duties 5.0 0.0 100.0 5.0 0.0 Customs Fees 1.6 0.0 100.0 1.6 0.0 Tom.age Tax 0.0 0.0 100.0 0.0 0.0 Lighthouse Tax 0.0 0.0 100.0 0.0 0.0 Export Duties 0.2 0.0 100.0 0.2 0.0 Taxes on Incomes 9.2 0.0 100.0 9.2 0.1 Industrial Tax 5.5 0.0 100.0 5.5 0.0 Labor Income 3.3 0.0 100.0 3.3 0.0 Capital Income 0.4 0.0 100.0 0.4 0.0 Taxes on Goods 7.0 0.0 100.0 7.0 0.0 Stamp Duties 3.5 0.0 100.0 3.5 0.0 Petroleum prod. 1.0 0.0 100.0 1.0 0.0 Beer 1.6 0.0 100.0 1.6 0.0 Other 0.9 0.0 100.0 0.9 0.0 Non Tax Revenue 8.3 0.0 100.0 8.3 0.1 Tans. from NFPEs 4.8 0.0 100.0 4.8 0.0 Property Income 2.5 0.0 100.0 2.5 0.0 Tans. from Mixed Enterprises 0.5 0.0 100.0 0.5 0.0 Other Revenue 11.8 0.0 100.0 11.8 0.1 Memo Items: Market Exchange Rate 2,000.0 Official Exchange Rate 30.0 Ratio of Market to Official ER 66.7 GDP in US$ Billion 7.4 GDP in Kz 14,800.0 Source: Ministry of Finance and Mission Estimates. TABLE 2.5. ANGOLA - 1989 GOVERNMENT EXPENDITURES AT PARALLEL PRICES (Billion of Kz.) 1989 Z Z Budget Budget "Strong" "Soft" In soft Kz z Total Expenditure Incl. Extrabudg. Exp. 128.6 50.7 49W 4,408.6 29.8 Total Expenditure Inc. Extrabudg. Exp. 109.2 43.7 56.3 3,241.0 21.9 Current Expenditure 89.1 40.0 60.0 2,429.0 16.4 Personnel 53.5 35.0 65.0 1,283.1 8.7 Salaries 43.5 35.0 65.0 1,283.1 8.7 Other Compensation 4.5 35.0 65.0 1,043.3 7.0 Travel 4.2 35.0 65.0 107.9 0.7 Training 1.3 35.0 65.0 31.2 0.2 Goods 8.4 62.5 37.5 353.2 2.4 Fuel 2.9 100.0 0.0 193.3 1.3 Repairs 2.7 30.0 70.0 55.9 0.4 Medical Supplies 0.8 100.0 0.0 16.6 0.1 Durable Goods 0.75 50.0 50.0 23.7 0.2 Other 1.3 30.0 70.0 26.9 0.2 Services 6.8 65.4 34.6 299.0 2.0 Food 4.4 80.0 20.0 235.5 1.6 Linen, Clothing 0.8 30.0 70.0 16.6 0.1 Medical Services 0.3 100.0 0.0 21.0 0.1 Other Services 1.3 30.0 70.0 26.9 0.2 Other Current Exp. 5.0 37.5 62.5 129.0 0.9 Foreign Coop. 1.2 60.0 40.0 48.5 0.3 Scientific Research 0.1 100.0 0.0 6.7 0.0 Representation 0.0 100.Ar 0.0 2.5 0.0 Conferences, etc. 0.4 10.-r 90.0 3.0 0.0 Other 3.3 30.0 70.0 68.3 0.5 ":ansfers 15.4 34.5 65.5 364.7 2.5 Pension Benefits 0.2 0.0 100.0 0.2 0.0 Losses of UEES 0.0 Price Subsidies 0.0 Other Subsidies 15.2 35.0 65.0 364.5 2.5 Capital Expenditure 20.1 60.0 40.0 812.0 5.5 External Debt Interest 3.4 43.7 56.3 10Q.9 0.7 Other Extrabudg. Exp. 16.0 100.0 0.0 1,066.7 7.2 Deficit including Extrabudgetary Exp. (47.5) 20.4 (1,635.8) (12.4) Deficit Excluding Extrabudgetary Exp. (28.1) 18.4 (474.1) (4.5) Source: Ministry of Finance and Mission Estimates. 9 (c) the budget deficit incrAases tremendously in nominal terms. This reflects, basically, a change in the "numeraire'. However, it decreases in relative terms from 22 percent of tne GDP to 12 percent of the *GDP41. (d) adjusted Government expenditures account for 30 perce::t of the *GDP, not over 50 percent ae originally estimated. It should be pointed out, however, that since GDP estimates include production for auto- consumption or non-marketed GDP, 30 percent continues to indicate that a sizeable part of the economy is controlled by the Central Government. TABLE 2.6. ADJUSTED GOVER1MENT DEFICIT AND EXCHANGE RATE Deficit Z of Exchange Rate Trillions of K GDP 2000 1.8 12.4 300 0.3 13.2 4/ The use of different conversion factors, which underline the use of different parallel exchange rates, will not change the budget deficit relative to *GDP substantively. For example it was found that the budget deficit varies from 12 to 13 percent of *GDP, depending on whether the exchange rate is 300 or 2,000. Since even "low values' for parallel or market exchange rates are considerable higher than the official rate, transactions in weak Kz become a small proportion of the total (See Annex B of this Chapter). 10 D. The budtet deficit. exchange rate, and prices The budgetary situation for 1989 is more accurately represented by the adjusted budget, particularly from the resource mobilization view point. It could be said that, should exchange rate and prices be totally liberalized, the budget for 1989 would tend to look, in the absence of other budgetary changes, like the adjusted budget shown in Tables 2.4 and 2.55/. The reason for this is the fact that conversion prices used in adjusting the budget are considered to resemble, by and large, market conditions. however, we do acknowledge the fact that the conversion factors used in the exercise are highly ter ative. Based on the converted estimates, the impact of policy adjustment on the budget could be the followings (a) the deficit would decline from 22 percent of the GDP in the direction of about 12 percent, according to the extent of exchange rate and price adjustments. In nominal terms, there is little doubt that the budget deficit would increase thus reflecting the change in the nnumeraire"6/. However, these would be mere accountina changes; I/ Non oil revenues would be probably underestimated in such a case because they could increase as the base increases. 6/ It is possible that the larger nominal deficit also results from the fact that the Government is a net foreign exchange user (See Annex C). Additional analysis will be needed for a conclusive view on the matter. For the public sector as a whole, including the financial sector, it is generally agreed that the impact of exchange rate movements on the nominal deficit depends on whetLr or not the public sector is a net user of foreign exchange resources (See Lizondo, J. S. Multiple Exchange Rates and Black Market Exchange Rates. A Non-technical Survey of Theoretical Results. University of Tucuman, Argentina, mimeo, 1990) 11 (b) any additional reduction in the budget deficit would necessitate active budgetary policies. In the case of Angola, since the wage bill and extrabudgetary expenditures add up to more than 50 percent of total expenditures, active policies to reduce the budget deficit will most likely require some reduction in those items. Personnel expenditures account for 16 percent of the *GDP. Cutting them by 10 percent would reduce the deficit by more than 10 percent. (c) an overall wage remonetization should, under certain conditions, be neutral vis-a-vis the public sector deficit7/. Such neutrality refers to the nominal deficit of the consolidated public sector (Central Government, parasta:als, and financial institutions). This is a theoretical assumption based on the fact that extra benefits enjoyed by parastatals from price liberalization would compensate for the additional expenditures in wages of the Central Government, if price liberalization and wage remonetization are strictly synchronized. The fiscal neutrality does not refer to each individual component of the public sector, including the Central Government. Moreover, given the special characteristics of the Central Government, the increasing autonomy enjoyed by parastatals in Angola, and the eventual privatization of most of them, it is very appropriate to asses the budgetary estuation separately from the consolidated public sector, as intended in this paper. 71 Angola: Selected Issues on Adjustment, in preparation. 12 TABLE 2.7. 1989 - STRUCTURE OF "ADJUSTED" AND "NON-ADJUSTED" BUDGET Non-Adjusted Adjusted Expend Expend Kz Bill X Rank Billions x Rank Revenues 81.1 L00 2576.4 100 Oil 38.8 48 2533.3 98 Non oil 42.3 52 43.1 2 Expenditures 128.6 100 4408.6 100 Personnel 53.5 42 1 1283.1 29 1 Goods & Serv. 20.2 16 2 781.2 18 3 Transfer 15.4 12 3 364.7 8 4 Capital 20.1 16 2 812.0 18 3 Extra Int. 3.4 3 4 100.9 2 5 Extrabud 16.0 12 3 1066.7 24 2 -47.5 -1832.2 Sources: Economic Mission 13 Annex I An "adiusted" Budget and the exchange rate 2.46 The budget deficit can be adjusted taking into account the existence of Strong Kwanzas (with some purchasing rights at official prices) and weak Kwanzas (without any purchasing rights), and can then be expressed in terms of GDP. If the operatic.xs in weak Kwanzas are a small proportion of the total, the budget deficit relative to GDP is likely to change very little in response to char.ges in the exchange rate used for conversions. D =J-E (1) D - deficit J - Government revenues E - Government expenditures D* J*-E* (2) D* (Jl + N*J2)-(E1 + N*E2) (3) where * means "adjusted" J1 - Gov. revenues in Weak Kwanzas 14 J2 - Gov. revenues in Strong Kwanzas N - Conversion factor (the number of weak Kwanzas that make a Strong one) N = ( Pp/Po) Pp - Prices parallel market Po - Prices official market D* = (Jl - El) - fN (J2 - E2)1 GDP* GDP us * ER GDPus - GDP in USS ER - Exchange rate (ER approx= Pp) D* - Jl - El - (J2 - E2) GDP* GDPus*Pp GDPus* Po If (Jl-El) is close to zero, then (D*/GDP*) will not be affected by changes in ER. In these estimates, the assumption that the GDP is constant in US$ is a strong one. It can be defended in the case of Angola pezhaps because of the high proportion of tradeables in the GDP (oil, diamond, etc.). 15 Annex II The Government budget and the external accounts ANGOLA: Dollar Component of the Government Budgot. 1989 (Percent and USX) 1989 Official Dollar Component of Budgot Budgot X USi Kw Total Revenue 81.1 46.9 1.206 Tax Revenuo 61.0 62.8 1.2667 Taxes on Oil Prod. 86.o 100.0 1.2667 Petroleum Output 11.0 100.0 0.3067 Profits of Comp. 14.0 100.0 0.4867 OtherV/ 13.0 100.0 0.4333 Taxes on Int. Trade 6.6 0.0 0.0000 Customs Duties 5.0 0.0 0.0000 Custom Foes 1.6 0.0 0.0000 Tonnage Tax 0.0 0.0 0.000 Lighthouse Tax 0.0 0.0 0.0000 Export Duties 0.2 0.0 6.000w Taxes on Incomes 9.2 0.0 0.6000 Industrial Tax 6.5 0.0 0.0000 Labor Incom 838 0.0 0.006 Capital Incom 0.4 0.0 6.0600 Taxes on Goods 7.0 0.0 0.0000 Stamp Dutios 8.6 0.0 6.0000 Petroleum Prod. 1.0 0.0 0.0000 Beer 1.6 0.0 0.000 Other 0.9 0.0 0.900 Non Tax Revenue 8.8 0.0 0.0060 Trans. from NFPEs 4.8 0.6 0.0609 Property Incom 2.6 0.0 0.6060 Trans. from Mixed Enterpriseo 0.5 0.0 O.00O Other Revenue 11.8 0.0 0.0000 Source: Ministry of Finonce and Mission Estimates 8/ Excludes profits tax of enterprises on special regimes 16 ANGOLA: Dollar Componont of the Govornment Budget, 1989 1989 Official Dollar Component of Budget Budgot X US3 Kw Billion Total Expenditure inc. extra-budgotary exp. 128.6 41.6 1.7837 Total Expendituro excl. *xtrabudgetary exp. 109.2 31.2 1.1371 Current Expenditure 89.1 29.2 0.8691 Porsonnel 53.6 36.6 0.6627 Salaries 43.5 36.0 0.6076 Other Compensation 4.6 36.0 0.0526 Travel 4.2 60.0 0.0840 Training 1.3 20.0 0.0087 Goods 8.4 17.6 0.0493 Fuel 2.9 0.0 0.0000 Repairs 2.7 10.0 0.0090 Medical Supplios 0.8 100.0 0.0267 Durable Goods 0.7 40.0 0.0093 Other 1.3 10.0 0.0043 Services 6.8 68.1 0.1317 Food 4.4 80.0 0.1173 Linen, Clothing 0.8 0.0 0.0000 Medical Servieoo 0.3 100.0 0.0100 Other Services 1.3 10.0 0.0048 Other Curr. Expend. 6.0 21.1 0.0354 Foreign Coop. 1.2 60.0 0.0240 Scient. Research 0.1 100.0 0.0033 Roprosentation 0.0 100.0 0.0012 Conference* etc. 0.4 10.0 0.0013 Oth-r 3.3 6.0 0.0055 Transfers 16.4 0.0 0.0000 Pension Bonefits 0.2 0.0 0.0000 Losses of UEEu 0.0 Price Subsidie 0.0 Other Subsidies 16.2 0.0 0.0000 Capital Expenditure 20.1 40.0 0.2680 External Debt Interest 3.4 100.0 0.1133 Other Extrabudg. Exp. 16.0 100.0 0.6333 Deficit Including Extrabudgetary Exp. -47.5 -0.6171 Deficit Excluding Extrabudgetary Exp. -28.1 0.1296 Source: Ministry of Finance and Mission Estimates PRE Working Paper Series Contact v AuLtho for paper WPS483 An Evaluation of the Main Elements Refik Erzan August 1990 G. llogon in the Leading Proposals to Phase FI -la Holmes 33732 Out the Multi-Fibre Arrangement WPS484 Stock Markets, Growth, and Policy Ross Levine August 1990 R. Levine 39175 WPS485 Do Labor Market Distortions Cause Ram6n Lopez August 1990 R. Luz Overvaluation and Rigidity of the Luis Riveros 34303 Real Exchange Rate? WPS486 A RMSM-X Model for Turkey Luc Everaert August 1990 S. Aggarwal Fernando Garcia-Pinto 39176 Jaume Ventura WPS487 Industrial Organization Implications Timothy Condon August 1990 S. Fallon of QR Trade Regimes: Evidence Jaime de Melo 38009 and Welfare Costs WPS488 Prepaid Financing of Primary Health Per Eklund August 1990 K. Brown Care in Guinea-Bissau: An Knut Stavem 35073 Assessment of 18 Village Health Posts WPS489 Health Insurance in Zaire Donald S. Shepard August 1990 K. Brown Taryn Vian 35073 Eckhard F. Kleinau WPS490 The Coordinated Reform of Tariffs Pradeep Mitra August 1990 A. Bhalla and Domestic Indirect Taxes 37699 WPS491 How Well Do India's Social Service Nirmala Murthy August 1990 E. Madrona Programs Serve the Poor? Indira H.rway 37483 P. R. Panchmukhi J. K. Satia WPS492 Automotive Air Pollution: Issues and Asif Faiz August 1990 P. Cook Options for Developing Countries Kumares Sinha 33462 Michael Walsh Amiy Varma WPS493 Tax Reform in Malawi Zmarak Shalizi August 1990 A. BhalIa Wayne Thirsk 37699 WPS494 Alleviating Transitory Food Crisis Victor Lavy August 1990 A. Murphy in Africa: International Altruism 33750 and Trade WPS495 The Changing Role of the State: Arturo Israel August 1990 Z. Kranzer Institutional Dimensions 37494 PRE Working Paper Series Contact lb AULb2l for paper WPS496 Issues in Evaluating Tax and Robert Conrad August 1990 A. BhalIa Payment Arrangements for Publicly Zmarak Shalizi 37699 Owned Minerals Janet Syme WPS497 The Measurement of Budgetary Carlos Elbirt August 1990 T. Gean Operations in Highly Distorted 34247 Economies: The Case of Angola WPS498 The Build, Operate, and Transfer Mark Augenblick August 1990 D. Schein ('BOT') Approach to Infrastructure B. Scott Custer, Jr. 70291 Projects in Developing Countries