Report No. 7645.BO Bolivia Country Economic Memorandumdocuments.worldbank.org/curated/en/345061468016777403/pdf/multi0... · Bs 1.00 = US$0.41 Abbreviations BAB ... 14460.0 151.5

Report No. 7645.BO

BoliviaCountry Economic Memorandum

September 15, 1989

Latin America and the Caribbean RegionCountry Operations Division ICountry Department III

FOR OFFICIAL USE ONLYU

Document of the World Bank

This document has a restricted distribution and may be used by recipientsonly in the performance of their official duties. Its contents may not otherwisebe disclosed without World Bank authorization.

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Fiscal Year

January 1 to December 31

Currency Equivalents

Currency Unit: Boliviano (Bs)Exchange Rate Effective December 31, 1988

US$1.00 = Bs 2.45Bs 1.00 = US$0.41

Abbreviations

BAB - Banco Agricola de Bolivia (Bolivian Agricul.ural Bank)BAMIN - Banco Minero (Bolivian Mining Bank)BANEST - Banco del Estado (Bclivian State Bank)COMIBOL - Corporacion Minera de Bolivia (Bolivian Mining Corporation)ENFE - Empresa Nacional de Ferrocarriles (National Railways

Corporation)ESAF - Enhanced Structural Adjustment FacilityFSE - Emergency Social FundICOR - Incremental Capital Output RatioIMF - International Monetary FundINE - Instituto Nacional de Estadistica (National Institute

of StatisticsLIBOR - London Interbank Offer RateMIGA - Multilaterai Investment Guarantee AssociationMNR - Movimiento Nacionalista Revolucionario (National

Revolutionary Movement)OLS - Ordinary Least SquaresNEP - New Economic PolicySAFCO - Systema Integrado de Administracion Financiera y ControlUNDP - United Nations Development ProgramUNICEF - United Nations International Children's Emergency FundVAT - Value Added TaxYPFB - Yacimientos Petroliferos Fiscales Bolivianos (Bolivian

Petroleum Corporation)

FOR OMCIAL USE ONLY

This report is based on a mission in December 1988, consisting ofWilliam Shaw (mission leader), Steen Jorgensen, and Mary Barton, assistedby Juan Carlos Aguilar of the Bank's Resident Mission. A preparatorymission (same participants) plus Laura Tuck visited Bolivia in August 1988.Peter Miovic accompanied the December mission and reviewed subsequentdrafts. Chapter II is based on information provided by Herbert Muller oneconomic policy, a background paper on the informal sector by OliHarrylyshin and Joseph Pelzman, and on the Bank's Financial Sector Study(16765-BO). For Chapter 3, Jorge Ospina wrote a background paper on taxpolicy, Laura Tuck contributed a paper on historical performance andprovisions of the tax reform, and Katherine Baer wrote a memorandum on taxadministration. See Annex I for a list of background papers and Bankreports recently issued on Bolivia. Mr. Jamil Mubarak assisted in theproduction of tables and in preparation of the projections. Ms. ElenaRodriguez coordinated the entire production process of the report. Thisreport was discussed with officials of the former Government in July of1989, and was subsequently reviewed with the new Government in September.Government officials have informed us that they are iLi basic agreement withthe analysis.

This document has a restricted distribution and may be used by recipients only in the performanceof their official duties. Its contents may not otherwise be disclosed without World Bank authorization,

Country Data Shoet

Area: 1106.6 POPUl0tion: Donsit : 6.1Woiu. *q.km.) (millions) e.s (For *-q ko )

Rate of growth 2.6X

Popuulot ion Chsracteristict Hea 1 thCnude bTrth rate (per 1,U1): 43 Infant mortality (per 1,166 live births): 118Crude d-eth rate (per 1,6O6): 17 Population per physician: 4800

Acces. to sate waterX of population - urban: 78

- rural: 12

Nutrition EducationCTolr;-iTntake as X of requirments: 91X imary ischool enrollment-(U) ofPer capita protein intake (g/day): 67 relevant age group: 91%

GNP per capita (USS, 1987) 1/: 670

Gross National Product 1988 Annual Rate of Growth (X. constant ricoc)us3-u T x 1980-85 1985-87 1986

GNP at Market rrices 4024 100.0 -2.5 -0.2 1Gross Domestic Investment 607 12.6 -4.2 -9.8 2.6Gross National Saving 139 3.4 -12.0 -0.9Curront Account Balance -396 -9.8 .Export of Goods, NFS 671 18.7 -4.6 6.7 12.4Import of Goods, NFS 849 12.1 -1.4 5.0 -12.9

Value Added 1988USS Mil X

Agriculture 1026 22.6Industry 1001 26.1

Mining S9 6.8Hydrocarbons 226 6.2Manuf. and Construct. 717 13.1

Services 2214 51.5Total 241 -b5

Government FinanceNon-Financial Public Sector Central Government(mi. lBs) X of GDP (mi lBs) XTof GDP

1988 1988 1987 1988 1987 1968Curront Receipts 6i 2e .8 T1218 T7&- 1iT9 171Current Expenditure 2869 26.1 1355 1386 16.4 13.3Current Surplus 78 0.8 -137 384 -2.6 3.7Capital Expenditures 842 8.2 641 863 3.0 0.8

I/ Calculated in accordance with Atlas methodology.

Country Data Sheet cont'd

Money. Credit A Prices 1984 1995 1986 1907 1988(Billion of Sol out-Randing *ndperiod)

Money Supply 4 287 811 1183 1681Bank Credit to Public Sector a -20S -638 -656 -632Bank Credit to Private Sector 2 291 732 lS1 is"

(Percentage or Index Numbers)Money as % of GOP 3.7 6.3 0.8 1.1 1.General Price Index (1966 a 1) 153 18169 68278 78265 90796Annual percentage changes in:

General Price Index 129060 11770.0 276.0 14.6 16.eBank credit Public Sector 9611 .

Bank credit to Private Sector 9606. 14460.0 151.5 64.4 44.8

Balance of Payments Merchandise Exports (Average 1986-88)

1986 1986 1987 1988 USS Mil I(MiIlions U§[5

Exports of Goods, NFS 722 674 648 671 Tin 226 30.7Imports of Goods, NFS 804 816 906 849 Zinc 32 4.4Resource Gap (deficit = -) 17 -142 - _M7 -171 Silver 99 13.5

Natural Gas 193 26.4Intorost Payments (net) -310 -233 -307 -251 Others 183 2.60Other Factor Payments (net) -63 -46 13 13 Total mI iraiNet Private Transfers 15 17 20 20Balance on Current Account -43 -405 -6i -39:

External Debt, December 31, 1988

USS MilNet Official Transfers as 82 108 117Direct Priv. Foreign Invest. 10 10 1S 30 Public Debt, in.l Guaranteed 4444Net MLT Borrowing -il8 -40 178 241 Non-Guaranteed Private Debt 200Disbursements 201 311 468 411 Total Outstanding & Disbursed 4644Amortization 319 351 281 189

Subtotal -42 62 297 388 Not Debt Serviee Ratio for 1988 1/

Other Capital (net) Public Debt inl. Guaranteed §Qand Capital n.e.i. 629 484 156 -38 Non-Guaranteed Private DebtInereaso in Reserves (-) T7 -111 71 44 Outstanding A Disbursed FQ7i

Gross Reserves (end year) 228 339 415 416

IBRD/IDA Lending (12/31/88) Million USS):Rate of Exchange (selling)

Annual AveragesJan-Apr End Period IBRD IDA

19U8 1987 1988 1989 Apr 1989 Outstanding & Disbursed 119 i47USf1.00=Cr S1.922 2i0 3 23 62518 2 Undisbursed 0 128Cr2l.00=US .620 .487 .426 .397 0.391 Outstanding inl. Undisb. 168 T7

1/ Debt Service, net of interest ear,ied on foreign exchange reserves, as a percentage of Exportsof Goods A NFS.

BOLIVIA

COUNTRY ECONOMIC MEMORANDUM

Table of Contents

PaRe No.

SUM(A1Y AND CONCLUSIONS i-vii

CWAPTER I: BACKGROUND AND RECENT ECONOMIC DEVELOPMENTS ............... 1I. Historical Background ...................................1 II. Recent Economic Developments ........................... 10

CHAPTER II: THE NEW ECONOMIC POLICY ................................ 15I. Introduction ..................................... ..... 15

II. External Sector .21III. Labor Markets .29IV. Investment Incentives .35V. Public Sector Management .38VI. The Informal Sector .................................... 43VII. Financial System .47

CHAPTER III: THE TAX REFORM ............................. 54I. Introduction and Summary ............................... 54II. Resource Mobilization Before the NEP ................... 55

III. The Bolivian Tax System ................................ 56IV. Tax Administration .................. 69V. Projections .................. 72

CHAPTER IV: ALLEVIATING THE SOCIAL COSTS OF ADJUSTMENT .77I. Introduction and Summary .77II. Social Costs of Adjustment .78

III. ESF Design and Operation .78IV. Economic Impact of the ESF. 30

CHAPTER V: MEDIUM TERM PROJECTIONS . 88I. Introduction .88

II. Moderate Growth Scenario .90III. Slow Growth: Consequences of Poorer Policies .......... 98IV. Accelerated Growth Scenario: Faster Supply Response ..102V. Conclusions ........................................... 103

ANNEX I: WORLD BANK ECONOMIC AND SECTOR WORK ON BOLIVIA ........ 105ANNEX II: ASSUMPTIONS FOR THE MEDIUM TERM PROJECTIONS ........... 106ANNEX III: OPERATION OF THE EMERGENCY SOCIAL FUND ................ 108ANNEX IV: STATISTICAL APPENDIX .................................. 111

MAP: IBRD 16591

Table of Contents (Continued)

Page No.

Tables

1.1 Indicators of Economic Performance ............................... 51.2 Output in 1988 .................................................. 112.1 New Economic Program -- Policies and Acfievements ............... 162.2 Tariff Rates Before the NEP ................................... 282.3 Wages and Productivity ................................... 342.4 Monetary Conditions ................................... 482.5 Commercial Bank Portfolios ................................... 502.6 Interest Rates and Inflation ................................... 523.1 Treasury Revenues ................................... 583.2 Fiscal Accounts of the Non-Financial Public Sector .............. 753.3 Treasury Revenues ............................................... 764.1 Emergency Social Fund Achievements Summary ...................... 814.2 Per Capita Food Consumption of ESF Workers ...................... 844.3 Educational Levels of ESF and General Population ................ 864.4 Distribution of ESF Expenditures by Poverty Area ................ 875.1 Domestic Performance ............................................ 895.2 Moderate Growth Scenario .915.3 Exports in the Moderate Growth Scenario ......................... 925.4 Moderate Growth Scenario: Balance of Paym.ents Projections ...... 955.5 Slow Growth Scenario ........................................... 1005.6 Acclerated Growth Scenario ..................................... 102

Figures

2.1 Real Exchange Rates -- Chile .................................... 242.2 Real Exchange Rates -- Peru ..................................... 252.3 Real Exchange Rates -- Argentina ................................ 262.4 Real Exchange Rates -- Brazil ................................... 274.1 Distribution Across Family Income ............................... 855.1 Debt to GDP Ratio and Current Account Deficit to GDP Ratio . 96

SUM4ARY AND CONCLUSIONS

1. The Bolivian economy suffered a crisis of staggering proportionsin the first half of the 1980s. Political instability, a crushing debtburden, a huge and inefficient public sector, extensive controls oneconomic activitv, and erratic macroeconomic management combined to producethe first recorded hyperinflation in Latin America and a severe decline inoutput. High tariffs and quantitative controls, an overvalued exchangerate, price controls and strong regulations, channeled a growing share ofproduction through the informal sector. Extreme inflation rates coupledwith controls on interest rates and other financial restrictions led to aflight from the peso and a strangling of private sector investment.Economic chaos fostered intense labor disputes, as social groups competedto at least maintain their share of a rapidly shrinking pie.

2. Immediately upon taking office in August 1985, the Paz EstenssoroGovernment undertook an orthodox stabilization program which abruptly endedthe hyperinflation. The Government increased public sector prices(particularly gasolire) to raise revenues and reduce liquidity, cut publicsector expenditures through a temporary prohibition of wage increases andpublic sector access to Central Bank credit, and increased the officialexchange rate (i.e devalued the Bolivian peso).1 Inflation fellprecipitously within a few weeks of the prcgram's announcement. Boliviahas enjoyed remarkable price stability compared to other Latin Americancountries since 1986. The consumer price index rose by only 1 in 1987and 22Z in 1988, and by about 6% (at an annual rate) in the first half of1989.

3. At the same time, the Government instituted a comprehensiveprogram to reduce the importance of the public sector and increase the roleof the price system in allocating resources. The New Economic Policyeliminated most quantitative trade restrictions and instituted an almostuniform tariff policy. Controls on interest rEtes and on denominatingcontracts in terms of foreign currencies were aLolished, along with avariety of programs for offering credit at subsioized rates. Initial stepswere made in a restructuring of the banking system to increase thesoundness of bank portfolios. Price controls were eliminated. Theexchange rate was determined in daily auctions with universal access.Restrictions on employment practices and wage negotiations were eased. TheGovernment also began a reform of public sector ente.prises, includingdrastic employment reductions in the state mining company (spurred by thecollapse of the price of tin in late 1985), a reorganization of the CentralBank, first steps towards the privatization of some state enterprises, andtighter controls on wages throughout the public sector.

1/ The Latin American Region defines the exchange rate as units ofdomestic currency per unit of foreign currency. Thereforn, an increasein the exchange rate implies a depreciation of the boliviano (prior toJanuary 1987 the Bolivian peso).

- ii -

4. The Government has been remarkably successful in putting in placeand maintaining a system of incentives to encourage greater private sectorparticipation in the economy. Rules governing trade, financial, and labormarkets have been revised to encourage competition and increase reliance onp.4Ace signals in the allocationi of goods. The New Economic Policye~ entially reversed thirty years of Bolivian development policy inencouraging private sector initiati%e and reducing the role of the state ineconomic decisions.

5. The changes in regulations had a dramatic impact on the labormarket. The easing of restrictions on dismissing workers increased labordiscipline and reduced the incidence of strikes. Wage levels were allowedto be set without Government interference (with the except:-on of minimumwage legislation), which gave firms more flexibility in their use of labor.it is difficult to precisely measure the implications of these changes foremployees. However, it is likely that the immediate impact of the newregulations was labor shedding of redundant workers and consequent declinein wages and employment in the formal sector of the economy. Over time,faster growth and the increased flexibility in the hiring of labor shouldincrease the demand for labor, and hence raise wages and employment.

6. Changes in the trade regime have resulted in a much more open andcompetitive tradeables sector. Abolition of import controls andimplementation of an almost uniform tariff policy have eliminated most ofthe distortions in relative prices imposed under the pre-1985 policies.Foreign exchange auctions generally have worked well in maintaining arealistic level of the exchange rate; the margin between the official andparallel rate has remained very small since August 1985. However, Boliviacontinues to face a huge debt burden and a large current account deficit,suffers from declining terms of trade since the collapse of tin prices in1985, and must diversify exports to reduce dependence on tin and naturalgas. This scarcity of foreign exchange points out the need to continue aflexible exchange rate policy to achieve a substantial increase in the realexchange rate over the medium term.

7. Certain key aspects of public sector administration have shownremarkable improvement. Elimination of price controls and the raising ofpublic sector prices have improved the efficiency of the price system, aswell as contributing to the public purse. The complete overhaul of taxadministration could serve as a model f-oi- tax reform in Latin America. Anewly-introduced public expenditure accounting system has greatly increasedthe amount and reliability of data on public sector expenditures, aprerequisite for efficient fiscal control. A computerized system fortracking investment projects has improved planning of the public sectorinvestment program, and recourse to procurement agents for virtually allpublic sector purchases has eased bottlenecks preventing disbursements ofexternal loans. The reorganization of the Central Bank strengthenedtraditional functions of monetary control, although the process waspainful.

- itii

8. Bolivia has achieved a significant easing of .'ts debt burden inthe past few years. An innovative buy-back and debt-equtity swap program,using funds provided by bilateral donors, has retired over US$400 millionin commercial bank debt. Given a modest increase in resources, it may bepossible to completely eliminate Bolivia's commercial bank debt within thevery near future. Bo Livia has also succeeded in rescheduling mostbilateral debt service payments through negotiations with the Paris Club,although the Paris Club has thus far refused Bolivia's request for the morelenient terms provided Sub-Saharan African countries. It can be expectedthat continued negotiations of bilateral debt service payments will benecessary, given the large size of Bolivia's official debt relative to itseconiomy and the substantial service payments that will be due for manyyea-s to come.

9. Although GDP increased in 1987 and 1988 (for the first time since1981), the recov ry was slow and consumption per capita continued to fall.A nunber of factors contributed to this. The difficulties inherent inimplementing a comprehensive change in economic policies were exacerbatedby ecternal and internal shocks. The price of tin dropped by 602 in late1985 and the price Bol.via receives from natural gas exaports has fallenalon[, with :he decline in international oil prices. Bolivia's terms oftrade declineid by 46Z from 1985 to 1988. A drought reduced agriculturalprodu:tion in 1988. Arrears in payment from Argentina fDr Bolivian gasshipmennts have created tremendous problems in managing the balance ofpaymeAts and fiscal accoi-.nts. Argentina's arrears exceedled 1TS$250 millionby mic.:-1989, depressing llolivia's int?rrational reserver and reducingfiscal revernwes below prcgrammed targets

10. Longer-term problems have also hampered adjustment and the sup?lyresponse to the new incen:.ives. While the Government has managed to reduceits cash outlays for debt service throigh buying back a portion of thecommercial bank debt and through reschemduling agreements with bilateralcreditors, debt service pald still equa1ed :.,Z of Treasury revenues in1988. Lack of trained personnel and unfamiliarity with ex-Port markets hasslowed the necessary increatse in tradeables production in response toexcharge rate depreciation. Poor transportation infrastructure has impededboth domestic commerce and internationa; trade. Most importantly, ahistory of political and ecmnomic instatility has reduced publicconfidence in 3olivian inst.tutions and the permanence of policy changes,limiting the supply of savirgs and maint.ining real interest rates atlevels well above the productivity of capital. High real interest ratesand uncertain prospects for a strong recovery have combined to constrainthe private sector investment- essential to adjust to the new set ofrelative prices imposed by the NEP.

11. Thus, despite the considerable achievements of the Paz EstenssoroGovernment, much remains to be done to sustain and improve the presentmacroeconomic policy framework. The next Govermnent will have to undertakea number of difficult measures to consolidate the gains made under the NEP.The fiscal situation must remain under tight control and essential

- iv -

institutional Improvements must be implemented, if publir confidence is tobe maintained. Also, the exchange rate will have to be managed carefullyto maintain competitiveness.

12. The restructuring of the banking system will require a number ofyears to complete, although substantial progress has been made. Therecently-completed audits of commercial banks provide a much better view ofthe quality of their assets and capital. The increase in collateralrequirements should improve the soundness of banks' portfolios, and mayreduce the concentration of loans by prohibiting past practices of lendinglarge amounts based on personal guarantees. Further progress must be madein reducing banks' operational costs, in getting rid of non-performingassets, and in revaluing bank capital to more closely reflect the presentmarket. A difficult challenge will be to impose higher capital andcollateral requirements without unnecessarily impairing financialintermediation. Over time, a sounder banking system should permit areduction in spreads and hence a fall in interest rates.

13. Reforms of state enterprises must be speeded up. While theshutting down of COMIBOL's operations in the crisis atmosphere of 1985-86was accomplished rapidly, COMIBOL's recovery as a productive enterprise hasbeen slow. Progress in the privatization program has been delayed due tofailure of the SAFCO law to pass Congress, although some of the initialstudies necessary to determine the market value of enterprises areunderway. The scarcity of qualified managers in the public sector forcedthe Government to target its efforts at the most pressing problems, whichinevitably has slowed progress in other areas.

14. Further efforts are required to strengthen public sectoradministration. While this is a long term process, priorities for theimmediate future include carrying through on planned improvements in taxenforcement, increasing efficiency and reducing evasion in the customsservice, reforming public sector wage and employment practices, and goingforward with the privatization program. The improvements in public sectoraccounting recently achieved must be made part of standard reportingrequirements for all public sector entities to ensure adequate accountingfor public funds. One of the most difficult items on the Government'sagenda is how to manage the decentralization of public s3rvices in thehealth and education sectors. While decentralization holds the promise ofincreasing the responsiveness of services to consumers, it must beaccomplished in a way that minimizes disruption of services and protectsthe poorer regions.

15. Some adjustments to the incentive structure considered by the Pazrstenssoro Government need to be carried forward. The optional net resultstax on mining would provide an appropriate incentives framework for thatsector. The guarantees for stability in the rules of the game included inthe draft investment code would improve the environment for private sectordevelopment. However, the Government will want to study whether the

- v -

guarantee of 30 years of the tax regime in the hydrocarbons and miningsectors is necessary. The provisions for joint ventures in the draftmining and hydrocarbons codes would facilitate the private investmentessential for development of Bolivia's natural resources. Some of theprovisions of these codes (particularly those concerning joint ventures)have been resisted in the interest of preserving public control overBolivia's natural resources. It must be understood that Bolivia faces adifficult future, in which obtaining the necessary external resources toensure long-term growth will be a significant challenge. Bolivia can notafford to forfeit the potential gains from private investment by neglectingthese improvements in the incentive structure.

16. One of the most intractable problems concerns how to integrateinformal sector activities into the formal economy. The informal sectorhas shown considerable dynamism over the past three years, and has absorbedlarge numbers of workers dismissed from formal sector employment. Thus,measures to increase participation in the formal sector should emphasizelowering the incentives for activities to move into the informal sector(including reducing paperwork requirements, increasing the efficiency ofpublic sector administration, and reducing the corruption incidental toobtaining official approval for business operations), rather than attemptsto suppress informal sector activity. Such reforms are a long-termprocess, and will require considerable improvements in public sectoradministration. In the short term, the Government's efforts could focus oncollecting greater revenues from the informal sector through user fees,including charges for water and electricity, site fee3 for space in publicmarkets, and other infrastructure services.

17. The next Government will enjoy a much improved fiscal situationfrom that which greeted the Paz Estenssoro Government, partially owing to aremarkable reform of tax rules and administration. Revenues from the taxreform equalled 6.11 of GDP in 1988, more than had been collected ondomestic taxes since at least 1975. Replacement of the old tax laws with alimited number of taxes has simplified both compliance and administratio.Using the commercial banking system to collect revenues and taxdeclarations has made it easier to pay taxes and improved accounting fortax payments. While the new tax system relies on indirect taxes (a valueadded tax and excise taxes) for the bulk of its revenues, it also containsincome and property taxes which could serve as important sources of revenuein the future.

18. Tax policy should emphasize stricter enforcement andadministrative improvements. The Government has extended the office set upin La Paz to monitor high-income taxpayers to Santa Cruz and Cochabamba.The next priority is implementation of the planned audit program to raiserevenues and ensure equitable enforcement of the tax rules. Work shouldbegin on improving urban property assessments and rural cadastres to assistenforcement of the presumed income tax on property. Should it be necessaryto increase revenues beyond what is now planned, the Government might

- vi -

consider reducing the deductibility of VAT payments from liabilities underthe complementary tax and/or increasing tax rates on the rural land tax,which presently are extremely low. Medium term projections of publicsector fiscal accounts show that considerable increases in tax collectionswill be essential to finance current expenditures and the domestic portionof the investment program without relying on domestic credit creation.

19. The economic crisis since 1980 has severely reduced the livingstandard of the Bolivian poor. While growth resumed in 1987, theadjustment program has imposed hardships on some groups. Given the tightfiscal constraints necessary to control inflation, options for assistingthe poor have been limited. The Government's efforts in this area havebeen channeled through the Emergency Social Fund (ESF), which has played animportant role in reducing the impact on the poor of inevitabledislocations during the adjustment process. ESF provides funds forlocally-generated projects with high labor content, and has proven highlyeffective in obtaining and disbursing foreign funds quickly, in managing alarge portfolio of projects with limited personnel, and in making asignificant contribution to development through building infrastructure.While there is some debate over ESF's contribution to total employment inthe economy, the program has increased formal sector employment and hasraised incomes among the more deprived groups in the Bolivian labor force.

20. Even assuming continued macroeconomic stability and maintenance ofthe present economic framework, Bolivia faces a difficult future. Highlevels of debt, poorly diversified exports, dependence on uncertainpayments by Argentina for a large share of export revenues, reliance onimports for most investment goods, low domestic savings, and limitedcapacity for implementation in the public sector will continue toconstrain growth for the forseeable future. We have elaborated a set ofbalance of payments projections to illustrate the magnitude of thedifficulties involved. The base case projections show the domesticpolicies and international assistance necessary to achieve an averagegrowth rate of 3.8Z, avoid a further decline in per capita consumption, andreduce dependence on external capital flows. This could be achieved byincreases in domestic savings from 91 of GDP in 1988 to 15Z by 1997;increased efficiency in the use of investment resources; strong growth inexports of non-tin minerals, soya, and other non-traditionals; developmentof alternative markets for Bolivia'a natural gas; high levels of lending byofficial donors at concessional rates until domestic savings can increasesufficiently to finance a greater share of projected investment levels; andrescheduling of Bolivia's debt service payments to bilateral creditors.

21. The projections highlight again the key elements of Bolivia'sdevelopment strategy discussed earlier. Continued marcroeconomicstability and commitment to the liberalization of markets achieved underthe NEP is essential to achieve sustained growth. Maintenance of acompetitive exchange rate is necessary to increase domestic production andrestrain consumption of tradeables. More effective public sector

- vii -

administration and increased tax revenues are required to raise the leveland improve the efficiency of public sector investment. In particular,improvements in infrastructure (especially transportation facilities andpower) are essential supports to greater industrial production and trade.Strengthened comercial bank portfolios would facilitate the efficientintermediation of the increased domestic savings necessary to achieve theprojected rates of growth. Improved health and education services areprerequisites to a better-trained and more efficient workforce, which willbe required to suppoLt the planned diversification of Bolivian production.Bolivian development will also require continued support from the donorcommunity. The remarkable progress made under the Paz EstenssoroGovernment in stabilizing the economy and restructuring the incentivessystem demonstrates that Bolivia deserves that support.

CHAPTER I: BACKGROUND AND RICENT ECONOMIC DFVELOPMENTS1

I. Historical Background

i. Before the 1952 Revolution. Until the middle of the 20th century,Bolivia's monetary economy was dominated by mining and mining-relatedactivities, with little spill-over into the rest of the economy. Ownersof mines and of large estates held great concentrations of wealth, whilethe vast majority of the population remained excluded from the modernsector. Until 1952, few Bolivians were entitled to vote and over 70Z livedin a quasi-feudal rural economy. No significant linkages developed betweenmining and the rest of the economy, as the profits from mining weregenerally invested abroad. Mining interests successfully blockedsignificant taxation of their profits, resulting in a very low level ofGovernment revenues and inadequate provision of public services.Infrastructure investments primarily focused on transport necessary to shipminerals to foreign markets. Investments essential for Bolivia's long-termdevelopment, such as transportation linking urban and rural areas, powergeneration, and basic health, education, and sanitation services, wereneglected.

2. Effective opposition to the economic and social order in thetwentieth century dates from the Bolivia's defeat by Paraguay in the ChacoWar in 1935, a disastrous conflict in which Bolivia lost a vast land areain the southeastern part of the country. The war was particularlytraumatic for large numbers of Indians from the Altiplano, who were forcedto campaign in the unfamiliar lowlands. Privation, disease, and inabilityto adjust to the lower altitude killed many more people than the actualfighting. Altogether, 5Z of the adult male population was killed orinjured in the war. Discontent with political and military leadership ledto a new generation of political leaders opposed to traditionalinstitutions, while participation in the army politicized many of the lessprivileged in the society. Labor unions were organized for the first timein Bolivia. Governments in the late 1930s expropriated the Standard OilCompany of New Jersey (partially due to charges that it had encouraged thewar with Paraguay) and nationalized the Central Bank and the Mining Bank.After a series of conservative Governments, the Movimiento NacionalistaRevolucionario (MNR) won the 1951 elections under the leadership of PazEstenssoro, the current president. The defeated Government's attempt toturn power to the military rather than accept the election results led to acountry-wide insurrection, in which armed miners and farmers defeated thearmy and police, bringing the winners of the 1951 elections to power.

3. 1952 and the Role of the State. The rew Government engineered asocial revolution. MNR came to power supported by miners and landlessagrarian workers, with the full knowledge that decades of control by thetraditional mining and landowning elites had left Bolivia backward andimpoverished. The Government's program was therefore dedicated to

1/ The historical discussion is based on the 1985 Country EconomicMemorandum (5680-BO), a background paper to that document entitled "TheEconomic Crisis", and two Updating Economic Memoranda (6455-BO and7278-BO).

-2-

destroying the traditional sources of political and economic power. A newelectoral law provided for universal adult suffrage, thus increasing theelectorate from less than 102 to 1002 of the adult population. An agrarianreform broke up the large estates and gave the land to those working it.The larger mines were nationalized under the management of the state-ownedBolivian Mining Corporation (COMIBOL). Taken together, these measures heldthe promise of a significant redistribution of income and a more equaldistribution of political power, as well as a very large increase in therole of the state in the economy.

4. Subsequent Governments steadily increased the share of the economycontrolled by the public sector. Numerous state enterprises were createdin many sectors of the economy, including manufacturing, transportation,agroindustry, and services. In the 30 years following 1952, the stateaccounted for over two-thirds of total investment. At the same time,private investment was largely conditioned by Government decisions owing tosubsidized credit programs directed at particular sectors (in the mid-1980sthe Government owned over half of the assets of the banking system) and anextremely complex and protective system of tariffs and quotas. Marketingboards and price controls extended state influence into the bulk ofdomestic trading in agricultural commodities. This process continuedthrough 1985, to a large extent independently of the ideological positionof the Government in power.

5. Impact of the State on Political and Economic Development. Inhindsight, the major achievements since the 1952 revolution have been thepolitical emancipation of the vast majority of the electorate, both interms of the extension of voting rights and the greater security andfreedom from past labor demands provided by the agrarian reform. The

redistribution of income anticipated after the revolution has been onlypartially achieved; miners and civil servants increased their share ofnational income, but little progress was made in raising the livingstandards of the rural poor.

6. Moreover, the increase in the role of the state resulted in anextremely inefficient structure of production and a continued neglect ofessential infrastructure and social services. Considerable controversyexists concerning whether the agrarian reform led to a reduction inagricultural production. While agricultural surpluses available for salein urban centers fell during the early 1950s, this may have reflectedincreased on-farm consumption of food as a result of the reform. The mostthat can be said with certainty is that little effort was made to improveroads, to provide other infrastructure, or to introduce more moderntechnology, while subsidized food imports reduced incentives for domesticproduction. Also, the confusion in land titles as a result of the chaoticnature of the reform remains a serious impediment to development in theagricultural sector.

7. The deleterious impact of state intervention was clearer in tinmining. Both Government and the miners had their own objectives fornationalized mines: the Government wished ti extract money from the minesfor investment in other sectors of the economy (through exchange rateovervaluation, export taxes, and allocating minimal investment funds formining), and the miners wished to increase their living standards. As a

-3-

result, COMIBOL was marked by overstaffing, inefficient operations, hugelosses, and lack of maintenance or new investment. Not a single largemine has been put into production by the public sector since 1952.

8. The state's emphasis on the production of private goods (i.e.goods that could be produced efficiently by the private sector) has led toa neglect of public goods. Only about 1Z of the Bolivia's area has beensurveyed. Agricultural research and extension services have done little tostimulate agricultural production. Inadequate funds are devoted to themaintenance and operation of the road network. Health and educationservices are poorly managed and inefficiently allocated. The publicsector's bias in favor of producing private goods can be traced to twofactors. First, workers typically earn more in state enterprises than inline ministries. Productive enterprises have a source of income which isindependent from the budget. This provides some insulation from CentralGovernment controls on wages, giving an economic incentive for civilservants to favor Government's involvement in the production of privategoods. Second, the pre-1952 failure of the private sector's exploitationof Bolivia's mineral and hydrocarbors reserves to be translated intoeconomic progress in the society as a whole ld Bolivia's leaders to turnto state ownership as a means of extracting rents from Bolivia's naturalresources.

9. Still, the problems inherent in Bolivia's development strategywere not immediately apparent. The disruptions of the 1952 revolutionushered in an initial period of declines in output accompanied by highinflation, but by the 1960s Bolivia began to achieve high rates of growth.GDP rose by 5.92 per year from 1962 to 1971. This growth was supported bya high rate of investment financed by foreign aid and increased inflows ofdirect investment in the petroleum sector. High growth rates continued inthe 1970s, due to substantial terms of trade gains owing to strongcommodity prices; the maturing of key investments made in the 1960s,particularly in the processing of agricultural commodities and inmetallurgy; and the discovery of sizeable petroleum and gas deposits whichmade the country an attractive client for foreign lenders. Capital inflowsincreased, largely in the form of loans to the public sector, as thenationalization of the Bolivian Gulf Oil Corporation in 1969 discouragedprivate direct investment during the 1970s. Bolivia also enjoyed lowinflation. The 1970s are generally regarded as a period of prosperity andrelative social peace.

10. The Economic Crisis. The economic crisis of the 1980s had itsroots in the prosperity of the 1970s. Bolivian development was based onlarge investments financed by external capital (lent either directly to theGovernment or to the private sector with guarantee by the Government),often borrowed at commercial terms. Much of the investments made duringthis period were poorly conceived, reflected distorted prices rather thanthe opportunity cost of the resources used, and/or resulted from politicalpressure rather than economic concerns. Examples abound of very lowcapacity utilization rates, of huge delays and cost overruns, of locatingindustrial plants in relat.vely inaccessible areas without the necessaryinfrastructure to produce and transport the product, and of having toabandon investments after hundreds of millions of dollars of expenditures.A significant portion of borrowed funds ended up as capital flight, whichtogether with unreported imports equalled 60? of the value of the debt

-4-

accumulated from 1971 to 1981. In short, during the 1970s Bolivia built upa huge debt burden without achieving the increase in productive capacitynecessary to service the debt in the future.

11. The proximate cause of the economic depression of the early 1980swas a dramatic reduction in Bolivia's access to foreign savings,exacerbated by natural disasters. A sharp fall in foreign lending (whichinitially reflected uncertainty over growing political instability) and arise in real interest rates quickly reduced Bolivia's ability to servicedebt and finance imports. Attempts to continue paying external debtservice led to a severe cut in imports. Foreign savings (the tradedeficit) fell from 4.62 of GDP in 1975-78 to a negative 3.4Z of GDP in1980-82 (i.e., to a trade surplus and outflow of domestic savings), andimports dropped by over US$200 million in 1980.2 A brief consumption boomin 1981 led to a recovery of imports (financed largely by arrears and adrawdown of reserves), but imports declined again in 1982. The fall inimports caused production bottlenecks, contributing to the decline inoutput and fueling inflation. A severe drought in the Altiplano in 1982-83and floods in the lowlands in early 1983 further reduced output. Together,the two disasters produced direct losses estimated at between US$400 to 500million, or about 72 of GDP in 1983.

12. While reduced external savings initiated the crisis, its severityand length was caused by disastrous economic management, particularly ahuge public sector deficit financed by money creation and an overvaluedexchange rate. Despite the sharp fall in external financing, politicalinstability and social conflict prevented any significant cut inexpenditures. As the economy lost access to foreign financing,expenditures were financed through increases in the supply of money,coupled with a build-up of arrears to domestic and foreign creditors. Atthe same time, maintenance of an overvalued exchange rate and declininginternational tin prices reduced export receipts on which the Governmentdepended for substantial revenues (although this was somewhat offset byrising natural gas prices). In 1982, the military Government instituted adual exchange rate regime in an attempt to stem t.he loss of reserves.Exporters were required to surrendeL 402 of their foreign exchange receiptsat the official rate of exchange. The huge discrepancy between theofficial and parallel rates (which exceeded 100Z for most of the periodbetween May 1982 and August 1985) implied a large tax on exporters, who inturn reduced to the extent possible the recording of exports, which alsolowered tax revenues. The accelerating inflation itself further increasedthe fiscal deficit, by eroding the real value of tax receipts. While it is

2/ See the 1986 Updating Economic Memorandum (6455-BO) for an explanationof the changes in savings and investment during this period.

-5-

nearly impossible to obtain meaningful estimates of the Government accountsduring the hyperinflationary period, the public sector deficit probablyequaled 301 of GDP in 198 4.

Table 1.1: Indicators of Economic Performance(Percent)

1980 1981 1982 1983 1984 1985

Investment/GDP 14.7 16.9 12.4 12.5 11.9 12.7Domestic Savings/GDP 20.2 19.5 22.7 22.0 17.5 11.6Inflation /a 47.2 32.1 123.5 275.6 1281.3 11749.6GDP Growth -0.9 0.9 -4.4 -6.5 -0.3 -0.2Budget/GDP /b -2.2 -0.8 -1.8 -18.3 -29.9 -12.7Debt/GDP /c 46.1 44.4 48.2 55.9 40.8 60.9

Memo Item

Average Margin BetweenOfficial and ParallelExchange Rate /d -- -- 259.4 262.4 290.3 140.9

/a Consumer Price Index./b Overall balance of the non-financial public sector. Data for 1983-85 are

extremely unreliable due to hyperinflation./c Includes only long-term debt of (or guaranteed by) the public sector./d Shows parallel rate as a percentage of official rate.

Sources: World Bank data

13. As inflation accelerated, it became self-reinforcing. The publicquickly perceived that extraordinary efforts were necessary to maintain thereal values of their incomes. The result was intense competition overacheiving nominal wage increases and enormous energy devoted to placingone's wealth in assets not subject to the inflation tax, in particular

3/ The hyperinflation and inadequate records make it very difficult tomeasure the size of the deficit, either in nominal or real terms, orrelative to GDP. When inflation exceeds 1X a day, the real value ofnominal expenditures depends very much on what day of the month theyare made, while our data only shows expenditures for the entire year.The problem is further complicated by the lack of accounts and thepractice of attributing expenditures made in one year to the prior yearif they relate to the prior year's budget. We can safely conclude thatthe deficit was large and increasing through August 1985, but thetiming and magnitude are impossible to determine.

-6-

foreign currency, real estate, and precious minerals. The flight from thepeso was accelerated by the forced conversion of dollar and dollar-denominated contracts among Bolivian residents to peso contracts, whichgreatly increased people's unwillingness to keep any significant portion oftheir wealth in Bolivia. The shrinking of the monetary base in turnrequired an accelerating rate of money creation to finance Governmentexpenditures. A succession of stabilization packages, some of which werequite orthodox in character and similar in intent to the 1985 program,failed owing to the Government's inability to maintain tight fiscal andmonetary policies. The dynamic of rising inflation, declining real moneybase, faster money creation, and still faster inflation continued untilhyperinflation practically eliminated the Bolivian currency as a store ofvalue or unit of account within the economy. The inflation rate reached28,0002 in annual terms in the first 9 months of 1985.

14. The hyperinflation had a number of effects on the economy.Inflation is a regressive tax which is paid by those who cannot increasetheir nominal incomes at a sufficiently rapid pace or who cannot protectthe real value of their savings, typically the least organized and poorestgroups within the society (except those totally outside the monetarysystem). The hyperinflation thus probably worsened the distribution ofincome. Extreme rates of inflation make it impossible to distinguishmovements in the general price level from changes in relative prices, thusreducing the allocative efficiency of the price system. The Government'scapturing of available savings (owing to forced savings through theinflation tax and running up of arrears) reduced the resources availablefor private investment. The hyperinflation also skewed that investmentwhich was made, away from productive activities and towards real estate andother hedges against inflation. A significant welfare cost can be seen inthe heightened social tensions and increased anxiety of every day life in ahyperinflationary environment. Most basically, economic chaos completelydiscredited public sector economic policy.

15. The decline in economic activity as a result of the crisi. isdifficult to determine. Official estimates show a fall in real GDP of over10% from 1980 to 1985, 24% in per capita terms, Per capita consumption isestimated to have dropped by a total of 162 over the period. Private fixedinvestment averaged about 32 of GDP in 1983-85, compared to 7.3Z in 1980.However, it is likely that these numbers overestimate the decline inproduction and consumption, as the huge margin between the parallel andofficial exchange rates greatly encouraged the conduct of economic activitythrough unrecorded channels. Illegal exports of coca also increased duringthe early 1980s, further expanding the share of unrecorded transactions inthe economy. Still, the trends shown in official statistics of adeteriorating economy, falling consumption, and firms living off ofinventories to maintain minimal levels of production are probably anaccurate representation of Bolivia's expe:ence.

16. The 1985 Stabilization Program. The new Government which tookpower in August of 1985 (headed by Paz Estenssoro, also president duringthe 1950s) faced an economic crisis of staggering proportions. Prices wereincreasing at about 2% per day. Government expenditures were out ofcontrol, with public agencies functioning without budgets and most

-7-

expenditures financed by credit from the Central Bank or arrears. Intenselabor disputes accentuated the chaotic atmosphere, to the point where theresponsiveness of the Central Bank to the Government was limited by laboragitation directed at individual policy decisions.

17. Similar to the experience after the 1952 revolution, the newGovernment viewed the perilous state of the economy as an indictment of thedevelopment model pursued over the past 30 years. Growing state controlover the economy with social peace preserved by the sharing of benefitsamong competing political groups was seen as leading to an inefficientstructure of production and uncontrolled public spending, followed byinflation. Thus, a recovery of the economy would require bothstabilization and a greater reliance on private sector activity guided tymarket mechanisms, as opposed to directing economic activity throughGovernment intervention and controls.

18. The Government's first task was to stabilize the economy. To doso, it was essential to reduce money creation by cutting expenditures andraising revenues, and to achieve a unified and stable exchange rate to stemthe flight from the peso. Government expenditures were limited to thelevel financeable by cash available, abruptly halting the financing ofexpenditures through Central Bank credit. The Government could notimmediately reduce the budgets of the numerous of public agencies and stateenterprises, so the drop in money creation was reflected in a large build-up of arrears to the domestic private sector. The Government temporarilyfroze public sector wages and prohibited investment expenditures. Raisingthe prices of public sector goods and services (particularly gasoline) tointernational levels increased revenues sharply. The 932 devaluation ofthe official exchange rate also increased taxes on trade. As a result ofthese measures, the public sector budget on a cash basis was approximatelybalanced in 1986.

19. Establishing a realistic exchange rate was an important tool instabilizing the economy. During the hyperinflation, the parallel exchangerate with the dollar had become a key determinant of inflationaryexpectations. The dollar exchange rate was the only quickly availableindicator of monetary conditions, so that prices tended to increase almostinstantaneously with changes in the dollar rate. The stabilization of theexchange rate in August of 1985 thus had a marked impact on priceadjustments. The availability of this nominal 'anchor' facilitated anextremely rapid disinflation. Inflation halted abruptly within two weeksof the announcement of the program; the consumer price index actuallydeclined on average in October 1985. Inflation increased again in Decemberowing to some loosening of policy and uncertainty created by the collapseof the tin market, but stricter demand management policies in january 1986restored stability.

20. In conjunction with the stabilization program, the Governmentinstituted a far reaching liberalization of markets to increase reliance onthe price system for the allocation of resources and encourage greaterprivate sector participation in the economy. Price controls and marketingboards were virtually eliminated. Public sector prices were raised usingneighboring countries' prices as guidelines. The exchange rate was setthrough an auction system with free access to all, and taxes andcommissions on foreign exchange transactions were done away with.

-8-

Quantitative trade restrictions and export licensing requirements wereeliminated, with the exception of controls on wheat and sugar, and importtariffs were reduced and simplified. The Government undertook a majorreform of the financial system: controls on interest rates and otherrestrictions on financial transactions were abolished; the Central Bank wasreorganized to reduce its role of serving as a commercial bank for publicenterprises and to strengthen traditional functions of monetary control;and a restructuring of the financial system was begun to improve bankingsupervision and increase the soundness of the banking system. Restrictionson the hiring and firing of employees were eased and private sector wageswere to be set in negotiations at the level of individual firms (althoughminimum wage restrictions were maintained). Public sector employment wasreduced as the result of the reorganization of COMIBOL, dissolution of theBolivian Development Corporation, and restructuring of the Central Bank.Efforts were undertaken to make public sector administration more efficientand the first steps towards privatization of some state enterprises weretaken. The various provisions underlying the liberalization program arediscussed in detail in Chapter 2.

21. The economy after stabilization. The first year following the endof the hyperinflation was an extremely difficult one. GDP declined by 2.8Zin 1986 and unemployment rose. The important determinants of economicactivity in 1986 wuere the stabilization/liberalization program, the fall inthe price of tin, and private sector uncertainty over the permanence of thenew policy regime.

22. Measuring the short-term impact of the stabilization andliberalization measures on growth is a difficult task. It is useful forthis analysis to distinguish between policies aimed at liberalizing theeconomy from those directed towards stabilization.4 Many of theliberalization policies may have reduced output in the short run. Duringthe hyperinflation, the bulk of production in the formal sector of theeconomy occurred in industries protected or subsidized by the state. TheNEP disrupted these activities by revamping the tariff structure, raisingthe prices of public sector goods, eliminating the opportunities for profitinvolved in the dual exchange rate system, and reducing subsidized credit.Thus, it is easy to point to examples of production lost owing to the NEP.In a sense this is not a valid comparison, as the extensive system ofcontrols had reduced efficiency, and their elimination was a prerequisitefor development over the long term. Over time, greaLer reliance on themarket should lead to a more efficient structure of production and tohigher growth rates. Still, it takes time for the private sector torecover from the loss of state protection and undertake new productiveactivities in response to the new system of incentives.

23. By contrast, the stabilization of prices may well have had apositive impact on output growth. Economic chaos during the hyperinflationhad depressed production and imposed a tremendous welfare cost on thecountry. Stability increased the effectiveness of planning and facilitated

4/ In practice the two goals were mutually dependent. For example,raising public sector prices both reduced the budget deficit andincreased the efficiency of resource allocation.

-9-

a more rapid response to developments in the market. A common fear thatdisinflation will reduce demand owing to rigidities in contracts orslowness in changing expectations to reflect the new policy regime wasprobably not very important in Bolivia. For example, it is often arguedthat an abrupt change in monetary regime (such as occurred in August 1985)will increase the real cost of wage settlements and interest rates agreedto in the expectation of continually rising prices. However, the extremelevels of inflation reached in Bolivia had largely eliminated the use ofthe peso in contracts (except with maintenance of value in terms of astable foreign currency) and had greatly shortened the maturity of mostloans. Therefore, it is unlikely that the disinflation per se had a largeimpact in reducing demand.

24. The collapse of tin prices in late 1985 accounts for a significantportion of the fall in output in 1986. The price of Bolivia's tin, whichprovided 26% of export revenues in 1985, fell by one-half, and mineralsproduction fell by 25Z. Despite the substantial rise in recorded (non-tin)export volumes encouraged by the devaluation, export revenues fell in 1986by 4X. As the bulk of tin production is in the public sector, the tinprice collapse greatly complicated the stabilization process by reducingGovernment revenues. This in turn reduced domestic demand, as revenuesfrom other sources had to be higher, or Government expenditures had to belower, than what they otherwise would have been to maintain a balancedbudget.

25. Uncertainty in the private sector over the durability of thepolicy reforms also reduced output in 1986. This was not necessarily theresult of Government policy. Although inflation increased in the first twomonths of the year, it was quickly brought under control. Overall, theGovernment maintained its commitment to monetary stability and greaterreliance on market mechanisms. However, the long history of politicalinstability and vehement protests against the stabilization program couldnot help but raise fears over the possibility of changes in policies.While a significant repatriation of assets held abroad did occur inresponse to the stabilization, the risk premium necessary to obtain funds,even at vety short maturities, was extremely high. The real interest rateon peso-denominated loans equalled 52Z at end-1986. Even dollar depositsin Bolivian banks earned 15%, or about 8 percentage points above the returnavailable in the international market (LIBOR). Reportedly, interest ratescharged by informal moneylenders were higher. While some recovery inprivate investment from the extremely low levels reached in the mid-1980sdid occur, private fixed investment remained below 32 of GDP.

26. GDP rose by 2.1% in 1987, the first increase since 1981 (itselfonly 0.9% following a fall in GDP of 0.9% in 1980). Growth in manufacturesand commerce offset a further decline in agricultural production and lackof recovery by COMIBOL. The fiscal deficit widened to 7.4Z of GDP, owingto unplanned increases in the wage bill, delays in payment by Argentina forBolivia's natural gas exports, and the costs involved in the restructuringof COMIBOL. However, consumer prices increased by only 10? (January toDecember 1987), as the increased demand generated by the higher deficit (tothe extent that the deficit was not financed bv increasing arrears) wasreflected in a fall in reserves rather than higher inflation. Export

-10-

revenues fell by 122, owing to lower prices on natural gas exports anddepressed tin production, and the current account deficit reached 12.5% ofGDP.

27. The Government adopted a Reactivation Decree in mid-1987 in anattempt to spur recovery of the economy. The Decree provided forimprovements in public sector administration, a program to settle Bolivia'soutstanding debt with the commercial banks, measures to stabilize thebanking system, increased incentives for export, and the provision ofcredit for working capital as a spur to industry. While the fund set up tochannel credit to the private sector was hampered by financialirregularities and lack of resources, the Government did move ahead onother parts of the Decree. The program for restructuring of the bankingsystem has made substantial progress, the debt buy back agreement achieveda notable success in reducing Bolivia's debt burden, and measures tostreamline disbursement procedures set in place under the Decree are nowachieving some success.

II. Recent Economic Developments

28. Fiscal and Monetary Policy. Fiscal and monetary policy tightenedin 1988. The deficit of the non-financial public sector declined to 6.5%of GDP. While revenues fell short of expectations in the second half ofthe year (largely due to a decline in VAT payments on imports), theGovernment was able to exceed its target for the overall deficit bystricter expenditure control, for example freezing the bank accounts ofstate enterprises. Savings of the non-financial public sector rose from-1.7Z in 1987 to 0.21 in 1988, facilitating a rise in public sectorinvestment, while the fall in the deficit permitted a decline in theGovernment's reliance on domestic credit from 4.92 of GDP in 1987 to 1.62in 1988. Central Bank credit given to the private sector was essentiallylimited to the level supported by disbusements of external loans.

29. GDP rose by 2.8Z in 1988, the second year of positive growthfollowing the long decline in output during the early 1980s, but no greaterthan the 2.8Z rise in population. High real interest rates, a shortage oflong-term loans (except for that provided under Central Bank credit lines),and competition from contraband imports continued to constrain domesticproduction.

30. Sectoral Composition of Output. Minerals, manufacturing, andconstruction were the most dynamic sectors in the economy (see Table 1.2).Mineral production is estimated to have increased by 33Z. COMIBOL's oucputrose somewhat from the severely depressed levels of 1986-87, although notto the extent hoped for. Private sector mining showed a strong recovery,however, particularly in the recorded production of tin and gold by smallmines. It is possible that the increase in the output of the privatesector is overstated, as a portion of the private sector productionactually represents minerals stolen from COMIBOL mines. Construction roseby 8.32, due to the many projects undertaken by the Emergency Social Fund,construction projects in the major cities, and a recovery in privatehousing. Manufacturing production also rose by an estimated 6.3Z. Thelargest increases were registered in foods, beer, tobacco, some chemicalproducts, glass and metals.

-11-

Table 1.2: Output in 1988(Percent)

Growth Rate Share of GDP /a

Agriculture -1.1 22.6Manufacturing 6.3 10.9Mining 33.0 6.8Services 1.8 37.8General Government 1.0 13.5Construction 8.3 2.8Hydrocarbons 4.8 6.2

/a Share of GDP at factor cost.

31. The agricultural sector remained depresssed, with output fallingby l.lZ. A severe drought reduced harvests and beef production in theeastern section of the country, where much of the commercial agriculturalproduction is located. Agricultural output has declined for the past threeyears, indicating that problems in the sector extend beyond this year'sevents. The difficulties since 1985 can be traced to high interest rates,a shortage of long-term capital (except for loans provided through CentralBank credit lines) increased competition from imports, and higher transportcosts owing to the rationalizaticn of gasoline prices. In addition, thesector suffers from severe structural deficiencies owing to the confusedland tenure situation, inadequate transport infrastructure, andunavailability of technical assistance services.

32. Fixed investment increased by 21.8Z in real terms in 1988. Publicsector investment expenditures rose to US$348 million from US$268 millionin 1987. Some of the bottlenecks which had constrained disbursements ofexternal funds for public sector investment were eliminated. In particularimproved management of the investment program helped increase investmentexpenditures in the latter half of the year. Private fixed investment alsoincreased in real terms, but still investment remained less than 122 ofGDP. High real interest rates and low investment reflect the very lowdomestic savings rate in the economy. While consumption rose less rapidlythan GDP, indicating a much-needed increase in savings, domestic savingsequaled only 9.3Z of GDP in 1988. Given that access to external savings isconstrained by the supply from official lenders, a continued recovery indomestic savings is essential to finance the investment necessary forincreased growth rates.

33. Inflation. Despite a restrictive monetary stance and more successin controlling the budget deficit than in 1987, inflation increased to 22Zin 1988, compared to llZ the year before. Prices rose very rapidly in thesecond and third quarters. A number of reasons have been suggested for therise in inflation. The 40Z jump in the price of gasoline increased pricesthroughout the economy, and the subsequent indexation of gasoline and otherpublic sector prices exacerbated the inflationary impact of nominal

-12-

devaluations. The depreciation of the boliviano against the parallelexchange rates of most neighboring countries, a reduction in the supply ofimports owing to the closing of the Peruvian border, and higher prices foragricultural imports owing to the drought in the United States contributedto price increases.

34. By mid-year, inflation had reached unacceptable levels, creating aserious dilemna for the Government. The policy of tying the prices ofpublic sector goods to changes in the exchange rate (to avoid adetrioration in government revenues) coupled with successive devaluations(to achieve, over time, a real depreciation of the exchange rate) during aperiod of rising inflation had increased instability. In response, theGovernment fixed the USS/boliviano exchange rate for a period of fourmonths, beginning in August. Consequently, inflation fell during the lastquarter. In conjunction with a tightening of monetary policy, the fixingof the exchange rate was successful in slowing inflation. However,controlling inflation by fixing the nominal exchange rate raises the dangerof a deterioration in competitiveness. Subsequently, the Governmentresumed the policy of making discrete devaluations of the nominal exchangerate.

35. Trade and Balance of Payments. Bolivia achieved somestrengthening of its external accounts in 1988. Most notably, a decline ofUS$52 million in recorded imports (CIF) offset falling export receipts, andreduced Bolivia's trade deficit from US$234 million in 1987 to only US$159million in 1988. A number of reasons have been offered for this unexpectedfall in iLaports. Some observers argued that the high levels of imports in1986 and 1987 reflected hoarding caused by concern that the open tradingsystem would not last. Consumers and firms therefore bought an unusuallylarge volume of foreign goods while they were available. After threeyears, however, stocks of imported goods had increased sufficiently, sothat demand fell. Alternatively, the fall in recorded imports may reflectgreater under reporting of imports and a switch from legal to contrabandpurchases. The private enterprises monitoring import price declarations in1988 did not check the prices of goods with a declared value of underUS$1000, which increased the number of declarations of import purchasesbelow US$1000. Also, there appears to have been a large, unrecordedprivate capital outflow in 1988, which may reflect undervaluation ofimports or greater purchases of contraband. This is a particularlyimportant question to resolve in evaluating the Bolivian balance ofpayments position. Bolivia will find it easier to maintain an adequatereserves level if the fall in recorded imports reflects a permanentreduction in import demand, rather than simply a switch between recordedand unrecorded imports.

36. Disturbingly, export revenues rose only slightly in 1988. Exportsby private mining companies increased strongly owing to higher productionand a sharp rise in the international prices of tin, lead, and zinc.However, a fall in international oil prices reduced the price received byBolivia for its natural gas exports to Argentina, resulting in a decline ofUS$38 million in export revenues from hydrocarbons. Nontraditional exportsdeclined by US$7 million, largely due to the effect of the drought anddeclines in the international prices of a number of Bolivian exports, butperhaps also hampered by low levels of private investment over the past fewyears. Despite little increase in export revenues, the fall in imports

-13-

reduced the the current account deficit by US$135 million, or from 12.5Z ofGDP in 1987 to 92 in 1988. This dramatic improvement is well in excess ofthat anticipated under Bolivia's program with the IMF (see below).

37. Despite the fall in the current account deficit, Boliviaexperienced a decline in international reserves of US$44 million. Thefailure of the current account improvement to lead to a strengthening ofthe reserves position can be traced to four factors. Net lending fromofficial sources (including rescheduling) fell by US$50 million. Argentinafailed to settle their 1987 arrears on payments for Bolivian natural gasshipments as promised, and even incurred a rise in arrears. The accountsshow a large net outflow of private capital, compared to a substantialinflow in 1987 (discussed above). Finally, Bolivia paid US$88 million tosettle its arrears with two foreign oil companies.

38. Bolivia concluded an arrangement with the Fund under the EnhancedStructural Adjustment Facility (ESAF) in July 1988 for an amount of SDR 136million to be disbursed in three years. The major objectives of the ESAFprogram are to increase output by 3 to 4 percent a year and to reduce therate of inflation to below 10 percent during the period. To achieve theseobjectives, the authorities are to maintain monetary and fiscal policiesand structural reforms within guidelines agreed to by the Government andthe Fund. Exchange rate policy will be compatible with a strengthening ofBolivia's competitiveness. Bolivia met all performance criteria as ofSeptember 1988 (the first six months of the program) and reached agreementon policies and measures for the remainder of the first year of the program(through March 1989) in the context of a tentative economic program for thewhole of 1989.

39. External Debt. Bolivia achieved a significant easing of itsexternal debt burden in 1988. An agreement with its commercial bankcreditors provided for Bolivia to repurchase US$253 million of its debt at112 of the face value, using funds donated by foreign governments. Thedebt-equity swap program and a separate donor contribution retired anadditional US$97 million. This agreement left US$332 million of commercialbank debt. As of April 1989, commercial bank debt had fallen further, toabout US$220 million. Bolivia is now attempting to retire the remainder ofthis debt through negotiations with the commercial banks.

40. The Paris Club agreed to a rescheduling of all interest andprincipal due participating bilateral creditors (which excluded Argentinaand Brazil, Bolivia's largest official bilateral creditors) on loans madeprior to 1986, including amounts due on debt previously rescheduled.Interest rates are to be negotiated separately with individual creditors.Bolivia had hoped to be given terms similar to those offered the poorerSub-Saharan countries under the initiative proposed at the Toronto Summit,but the creditors did not accept this proposal. However, proposals forrelatively low interest rates on the rescheduled debt service are now underdiscussion, which coupled with the cash flow relief obtained by Boliviawould only be slightly less favorable terms than offered the Sub-Saharancountries.

41. Despite these considerable successes, Bolivia's debt burdenremains very high relative to its output and exports. By end-1988,Bolivia's long-term public sector debt (including guarantees) of US$4.3

-14-

billion equalled 972 of GDP and was more than five times the value ofexports of goods and services. The debt service owed reached 51? of exportreceipts, although Bolivia only paid 23Z in 1988. Interest paid made up102 of Treasury revenues, and 2.92 of GDP. As disc(ussed in Chapter 4,Bolivia will have to continue to depend on highly _oncessional borrowing,and will not be able to afford even the harder terms offered by some of themultilateral creditors. The country has little prospect for being in aposition to comfortably service commercial bank debt for considerable timeto come.

42. Coca Trade. The export of coca products for use in cocainecontinues to be a serious social and economic problem for Bolivia. Cocaexports provide a significant source of foreign exchange and employment.However, Bolivia's role as a major supplier of coca has greatly complicatedexternal relations, and the power and wealth of the drug traffickers is apotential threat to Bolivian social and political institutions. Mostseriously, greater involvement in the processing of coca has encourageddrug abuse within the country, particularly the very dangerous practice ofsmoking coca paste. Thus, the export of coca is a serious threat to thefabric of Bolivian society, one which the Government is committed toattacking. However, elimination of the coca trade is complicated by thelarge number of people economically dependent on coca and the need topreserve traditional uses of coca, which remain important to large segmentsof the Bolivian population.

43. Legislation was passed by the Congress in 1988 in support of aprogram to reduce the production of coca leaf for processing into cocaine.The strategy underlying the program is to restrict the area in which cocacan legally be grown so as to eventually limit production to that used fordomestic consumption of coca, and to provide financial incentives tofarmers for destroying coca fields and turning to some other form ofagricultural production. By mid-1989, approximately 3400 hectares of cocafields had been eliminated. Further progress in reducing coca productionwill depend greatly on the Government's political commitment, support fromexternal donors, and reduction in the demand for coca exports.

-15-

CHAPTER II: THE NEW ECONOMIC POLICY

I. Introduction

44. Immediately upon taking office, the Paz Estenssoro Governmentundertook an extensive liberalization of the market for goods and servicesin Bolivia, reversing the trend over the past 30 years of increasing stateintervention in the economy. The New Economic Policy (NEP) constituted aconsistent and comprehensive program designed to increase reliance ofproductive activities on the price system and on private sector initiative,reduce the influence of the state on production, and increase theefficiency of public sector administration.

45. The policy changes inaugarated by Supreme Decree 21060 touchednearly every aspect of the economy. Restrictions on firms' employment andwage decisions were eased. The public sector compensation system wasrationalized and the wage bill reduced. The tariff system was simplifiedand most quantitative restrictions abolished. Controls on interest ratesand foreign exchange transactions were eliminated, and subsequently arestructuring of the banking system begun. The exchange rate5 was unifiedand an auction system established to set the rate in the future. TheGovernment raised public sector prices to levels reflecting the opportunitycost of public sector goods, and eliminated most price controls andmarketing boards. Work was begun on legislation to improve incentives forprivate investment, particularly in the hydrocarbon and minerals sectors.A reform of public sector institutions was initiated, including areorganization of major public sector enterprises, a program to privatizesome enterprises, a revision of procedures to control public sectorinvestment, the introduction of standardized accounting practices in thepublic sector, and a reform of the tax system. Table 2.1 summarizes themajor policy initiatives under the NEP.

46. The policy initiatives of the Paz Estenssoro Government haveprovided a set of incentives conducive to greater efficiency and a recoveryof the Bolivian private sector. The policy framework should serve as amodel to many developing countries now attempting to manage the difficulttransition to a more liberal economic environment. However, the economyhas responded slowly to the new set of policies. Output increased in 1987and in 1988, but growth remains slow, underemployment has risen, andcapacity utilization is low. Recovery under the NEP has been hampered byunavoidable dislocations in economic activity from such a radical policychange, the grave structural problems afflicting the Bolivian economy, thehistory of economic chaos and public sector mismanagement which the PazEstenssoro Government inherited, severe declines in the terms of trade, andarrears in payment by Argentina for Bolivian gas exports.

5/ The Latin American Region defines the exchange rate as units ofdomestic currency per unit of foreign currency. Therefore, an increasein the exchange rate implies a depreciation of the domestic currency(in this case the boliviano).

Table 2.1: New Economic Program -- Policies and Achievements

Status Before NEP NEP Measures Objoctives Results

External Sector:

Highly overvalued official Devalued official rate and set Unify exchange rate and increase Auction system in place; veryoxchange reate, large auction system to deternine competitiveness of Bolivian small differential betweendifferential between official exchange rate, production. official and parallel rate.and parallel rate. Exchange rate appears compe-

titive although complaintsvoiced over high lavel ofcompeting imports.

Trade controls and complex, Eliminated most quantitative Unify effective protection rates, Largely successful; plans todifferentisted tariff restrictions, set uniform increase competitiveness of reduce tariff rate to 10% from 17Xstructure. tariff rate. Bolivian goods. currently delayed to avoid fall in

tariff revenues. Import controlson wheat and sugar still in place.

Export tax rebate of 1OX for Increase and diversify exports. Payaents delayed for over a yearnontraditional exports. owing to lack of resources, but

now being made.

Capital controls and All restriction on foreign Encourage repatriation of capital, Successful. Some repatriation ofrestrictions on holding assets currency transactions allow exchange rate to reflect capital in 1988 but little indenominated in foreign eliminated, mrket conditions, permit 1987-88. Rise in foreign-currency. diversification of assets. currency denominated assets in

economy.

Labor Market

Wage levels sot in industry- Wages to be sot through Reduce influence of national labor Successful. Reduced frequency ofwide negotiations. High collective bargaining at lnvel unions. Permit greater labor disputes.frequency of strikes and other of firm. flexibility in setting wages.labor actions.

Restrictions on employer's Restrictions eased, but not Increase flexibility in firm's use Successful. Som reports ofability to fire workers. eliminated, of labor, increase incentives for abuses (i.e., firing labor union

work efforts, leaders). Impact on productivity

difficult to measure.

Complex system of in-kind Consolidated non-wage payments Increase transparency of Successful; r-introduction of

payments, bonuses and access into basic wage. compensation syst-e and reduce some bonuses reported.

to subsidized food stores administrative costs.

large share of public sectorcompensation.

Ovorstaffing and v-ry low roal Reduced employment through Incre-se effectiveness of public Control over real wages largely

wage levels in public sector, reorganization and increasing sector administration while effective. Inefficient

menagers' incentive to controlling *xpendituros. administration and uncompetitive

economize on labor. wages remain long term problems.

Investment Incentives

Reduced private investment Developed inv-estment code to Encourage greater private Investment code written, but not

owing to uncertainty over past provide assurances against inves*ment. approved by Congress.

nationalizations and complex uncompensated nationaliza-

administrativ, procedures for tions, guarantee of stable

approval of foreign long-term tax regime, andinvestment, simplified administration.

Lack of investmnt in mining Developed mining and Encourage greater private Codes written, but not approvedand hydrocarbons owing to hydrocarbons codes to investment, by Congress.

state control of most facilitate joint ventures and

attractive properties, provide alternative tax

inappropriate tax regim. system.

Sottled dispute with foreign Encourage exploration and Successful.

,oil companies over development of Bolivia's oil

compensation. reserves.

Public Sector Administration

Large share of state Reorganized COMIBOL. Reduce drain on public finances Successful. Large fall in

involvement in production and increase efficiency of employment. Recovery of COMIBOL

through state enterprises, minerals szctor. slow.m*ny of which roquiredsubsidios to m*intain Curtailed access of state Lower inflation. Successful.

operations. enterprises to Central Bank

credit.

Studies of state enterprises Improve *fficiency of state Studies now underway.

with a view to future enterprises and eventually sell

privatization. some to the private s-ctor.

Lack of accounting for public Program to standardize public Improve information on public Substantial improvement in budget

sector expenditures hampered sector expenditures reports sector expenditures to facilitate information.

fiscal control. (SAFCO). more efficient expenditurescontrol.

Poor administration of tax Reorganization of tax Higher Government revenues and Revenues increased and

system, corruption and lack of administration, use of private more equitable tax system. substantialimprovement in

enforcement of tax code banks to collect taxes, new administration achieved. Mu'h more

reduced Government revenues, auditing procedures, and work to be done, particularly in

stepped up *nforcement. audits and enforcement.

Tax reform with Value Added Increase Government revenues and In place. Revenues generated by

°° Tax as main source of reduce complexity of tax system. new tax system reached 81 of GDP

revenues, in 1988.

Price controls, marketing Eliminated almost all price Increase reliance on price system. Successful. Retail prices of sugar

boards and low public sector controls and marketing boards, and wheat still set by Government.

prices reduced efficiency of

the market and required large

subsidies to maintain. Increased price of gasoline Raise Government revenues; and Successful.

products to international reduce liquidity to stabilizelevels, economy.

Increased railroad tariffs, Increase Government revenues and Largely successful.

electricity rates and promote efficiency.

telephone charges.

Financial Policies

Controls on interest rates and Eliminated ceilings on lending Increase domestic f inancial assets In place. Real interest rates have

subsidized credit programs and deposit rates. Reduced and improve efficiency of credit risen to extremely high levels,

during hyperinflation reduced subsidized credit schemes and allocation, indicating continued uncertainty

monetary base and impaired chanelled development loans over permanence of policy regime.

efficiency of credit through private banks.

allocation.

Reorganized Central Bank. Limit direct lending to private Ulti;mtly successful, although

sector and strengthen traditional reorganization accompanied by

monetary control functions of disruption of Central Bank

Central Bank. operations.

Hyperinflatlon greatly reduced Introduced boliviano equal to Simplify accounting and increase In place.

value of peso. one million pesos. confidence in currency.

Lack of supervision, financial Program to strengthen Increas- soundness of banking Substantial progress made, although

controls, and hyperinflation commercial banks, including system. restructuring will take years to

impired portfolios of banking establishing Superintendent, complete.

system. requiring increased capital

asset ratios, collateralrequirements, and shedding of

nonperforming assets.

Reorganization of public Reduce role of public sector in Progress is slow. Proposals for

sector banks, including lending and improve efficiency of reorganization of BAB and BANEST

agriculture bank (BAB), mining public sector banks, still being formulated.

bank (BAMIN), housing bank, Reorganization of BAMIN underway.

and state bank (BANEST). Banco de Vivi-nda closed.

Lack of bond mrket meant Issuance of certificates of Attract foreign savings and Market in operation, but sm ll.

Govornmnt deficits financed deposit by Central Bank. oventually provide additional

through Centrol Bank credit. monetary policy tool.

-20-

47. Initially, the liberalization program was unable to halt thedecline in output (GDP fell by 2.92 in 1986), and thereafter recovery wasslow. The shift from a productive structure determined by state subsidies,directed credit, and a highly distorted trade regime to one determined bymarket forces disrupted existing productive activities. Therationalization of effective protection rates and increases in publicsector prices greatly changed the profitability of many firms. Reductionof directed credit schemes and freeing of interest rates hurt enterpriseswhich had formerly enjoyed access to credit at subsidized rates. Theunification of the exchange rate reduced profits in various activitiesdevoted to obtaining access to foreign exchange at the official rate andselling foreign exchange or goods at the parallel rate. Therationalization of COMIBOL directly reduced output in the public sector.

48. For each of these losses to output, over time one would expect acorresponding, and hopefully greater, contribution to production. The newtrade regime provides opportunities for profit in areas of the economyformerly closed to domestic production. The reduction in directed creditschemes should encourage a channelling of savings to more efficientactivities. The depreciation of the boliviano increased incentives fortradeables production. Reduced output by state enterprises lowered thedrain on Government revenues, facilitating increased public sectorinvestment. Still, the output losses mentioned above are immediate, whilethe benefits take time to emerge as the economy adjusts to the new set ofrelative prices. Experience in other countries which have undergonestructural adjustment programs suggests that this process can take a numberof years. While in the end the new policy regime should produce a moreefficient sLructure of production, the transition has involved relativelyslow rates of growth.

49. Lnw confidence in Bolivian institutions, owing to economic chaosduring the 1980s, has slowed adjustment through reducing the willingness ofthe private sector to commit the resources necessary for a more rapidrecovery. Hence, real interest rates remain much higher than the likelyproductivity of capital in the economy, there is a shortage of long-termloans, and private sector investment remains low. It will probably take anumber of years to regain private sector confidence and achieve theincreases in investment necessary to revamp the structuire of production.This process is further constrained by the long history of stateintervention, which accustomed Bolivian entrepreneurs to rely on guaranteedmarkets from public sector purchasers, and to compete through increasingtheir access to state subsidies rather than through cutting costs. Whiledevelopments since 1985, particularly the dynamism shown by the informalsector, indicate that the entrepreneurial spirit is far from dead inBolivia, delayed adjustment in the formal sector may be partially due toslowness in changing attitudes.

50. External and internal shocks have impaired growth. The collapseof the tin price in 1985 greatly reduced exports and Government revenues.The decline in the price of oil and a build-up of arrears by Argentina onBolvian gas shipments have also reduced foreign exchange receipts. Asevere drought in 1988 depressed agricultural output.

51. Some difficulties in implementation of the reform also havehampered adjustment. Delays in passage of important legislation designedto improve investment incentives may have slowed inflows of foreign

-21-

investment. Little progress has been made on privatization of stateenterprises. Severe problems continue to beset public sector administra-tion. Broadly speaking, however, the Government's record in implementingthe program is very impressive.

52. Despite these considerable obstacles, the economy does show signsof responding to the changes in relative prices. For example, a recoveryin private sector mining is taking advantage of reduced state controls andthe more positive climate for investment. The experience of othercountries has shown that it can take many years following imposition of anadjustment package for the economy to achieve the expected high rates ofgrowth. Thus, slow recovery in the Bolivian economy does not indicate thatthe reforms are not working or that radical departures in policy arenecessary. On the contrary, continued commitment to the policy regime isan essent al prerequisite to sustained development.

Il. External Sector

53. Introduction and Summary. The Paz Estenssoro Government undertooka radical reform of external sector policies in unifying the exchange rate,eliminating foreign exchange controls and prohibitions against denominatingdomestic transactions in foreign currency, dismantling the system ofquantitative restrictions, and imposing a low and uniform tariff rate.These policies achieved a much more realistic official exchange rate andgreatly reduced the wedge between domestic and international prices whichexisted prior to August 1985. While the Government has been remarkablysuccessful in improving incentives for tradeables production, Bolivia mustcontinue to pursue a flexible exchange rate policy to increasecompetitiveness. Further, adjustment to the new set of relative priceswill take time, arid will require increased investment, which presently ishampered by the low supply of savings. Increased savings, a reduction ofinterest rates, and greater availability of long-termi loans are thereforenecessary to take full advantage of the new system of incentives.

54. ExchanRe Rate Policy. In the early 1980s, hyperinflation andattempts to avoid a depreciation of the boliviano resulted in a highlyovervalued exchange rate, accompanied by foreign exchange controls. Theexchange rate on the parallel market exceeded the official rate by anincreasing margin; by August 1985 the parallel rate was 15 times theofficial rate. To the extent possible, transactions used the parallelmarket, resulting in a sharp decline in officially recorded exports andaccess of the formal sector to foreign exchange. The official exchangerate was largely a means for individuals and firms with access to foreignexchange at the overvalued rate to reap huge profits.

55. The Paz Estenssoro Government reversed past policies of attemptingto support an overvalued exchange rate with controls. An auction systemwas instituted for determining the rate, with free access to all.Residents were permitted to hold foreign exchange and enter into contractsfor domestic transactions denominated in foreign currency. All taxes and

-22-

fees on foreign exchange transactions were abolished. The immediate resultwas a 932 depreciation of the exchange rate, followed by a substantialrepatriation of capital.

56. The auction system, or bolsin, is not designed to determine theexchange rate without Government intervention, but rather to enable theGovernment to adjust the rate periodically in response to the market. TheCentral Bank fixes the minimum bid price in the daily auction and theamount of foreign exchange supplied for auction. All bids are acceptedwhich exceed the minimum, until the total amount of exchange offered by theCentral Bank is exhausted. The Central Bank has generally been able toaccommodate the demand for foreign exchange. The rate on the bolsin hasremained close to the parallel market rate since September 1985.

57. Exchange rate policy since August 1985 has carefully balanced thegoals of maintaining price stability and competitiveness, a particularlydifficult task in Bolivia where inflationary expectations are verysensitive to changes in the exchange rate. The depreciation of theboliviano in late 1985 was followed by a fixing of the nominal rate in lateJanuary for the remainder of 1986, resulting in a strong appreciation inreal terms over the year. In 1987, the Government maintained a policy ofsmall, discrete adjustments to the nominal exchange rate. This policyachieved a real depreciation of the boliviano of 6.5Z against the USdollar, or 12.5Z on a trade-weighted basis, while holding inflation to 107.The Government continued the policy of the "crawling peg' adjustment of theexchange rate in the first half of 1988, but as a result of a sharp rise ininflation decided to freeze the US$ exchange rate at Bs 2.41 in July of1988, and maintained roughly that level until resuming small depreciationsin November. From December 1987 to December 1988. the bolivianoappreciated in real terms against the dollar by 6Z, and against a trade-weighted basket of currencies (which includes all of Bolivia's majortrading partners) by 3Z. However, the boliviano depreciated against thedollar (in both real and nominal terms) in the first half of 1989.

58. Exchange rate management was greatly complicated by thecorisiderable variability and conflicting trends in real exchange rates withneighboring countries.6 The real exchange rate with Chile (see Figure 2.1)closely follows the pattern of the trade-weighted effective rate in showinga real appreciation of the boliviano throughout 1988.7 However, the data

6/ For this analysis we use the parallel exchange rates, which provideinformation on changes in the relative prices of contraband imports, animportant source of competition for Bolivian production.

7/ The trend line shown in Figure 2.1 is the ordinary least squares (OLS)trend line, which indicates a steady fall in the exchange rate. TheOLS result is confirmed by time series analysis showing a downwarddrift in the series of about .3 index points per month.

-23-

on real exchange rates with other neighboring countries show substantialvariation. Bolivia's real exchange rate with Peru (see Figure 2.2)increases about 40Z in real tevms through 1988. The depreciation of theboliviano against the Peruvian currency probably reflects some slowness inmarket adjustment to the very high rates of inflation in Peru. Theboliviano exchange rate with Argentina (see Figure 2.3) shows the samepattern, rising strongly in 1988. The real exchange rate with Brazil hasno clear pattern (the trend line given in Figure 2.4 is not significantonce allowance is made for cyclical effects), but the index showsconsiderable instability, the level varying between 100 and 200 during theyear.

59. The high degree of variability in real exchange rates with someneighboring countries has contributed to instability within the Bolivianeconomy. Changes in the real exchange rate may help explain the rapidshifts in the supply of contraband imports observed in 1988, which have hada role in the large month-to-month variation in the Bolivian inflationrate. The increased uncertainty facing Bolivian producers owing to rapidand unpredictable changes in supply of goods from neighboring countries mayhave reduced production.

60. Highly-variable real exchange rates can not be used as reliableindicators of the appropriateness of the level of the real exchange rate.Rather, the Government must rely on longer-term measures to determineexchange rate policy. Bolivia faces a huge external debt burden, dependson a large inflow of foreign capital, needs to diversify exports to reducedependence on tin and natural gas, and continues to face low terms of tradeafter the collapse of tin prices in 1985. All of these problems point tothe need to continue a flexible exchange rate policy to achieve an increasein competitiveness. The Government will have to carefully consider itsmanagement of the nominal exchange rate, financial policies (particularlythe possibility of increased public sector savings) and measures to holdwages in check and raise productivity, in light of the need to providesufficient incentives for tradeables production.

61. Trade Restrictions. Before the liberalization program undertakenby the Paz Estenssoro Government, a complex and restrictive system ofcontrols governed commercial transactions with the rest of the world.Export receipts had to be surrendered to the Central Bank at the overvaluedofficial exchange rate. A highly-differentiated tariff system, importprohibitions, licensing requirements for numerous import and export goods(particularly agricultural products), and access to credit and foreignexchange at preferential rates for certain activities resulted in verydifferent rates of nominal and effective protection for different goods.

62. This system provided substantial protection for favored goods,although the complexity and variety of trade restrictions makes itdifficult to precisely measure the degree of protection afforded differentindustries. The average nominal tariff rate in July, 1985 (weighted by thestructure of imports in 1982) was 12.1Z (see Table 2.2). Durable consumergoods had the highest rates (38.1Z), and capital goods for agriculture thelowest (6.2?). The average effective protection rate is reported as 44.4z.Consumer goods produced in Bolivia had very high effective protection rates(wood products 145.82, textiles 169.1?, and meat products 80.4?), but evenmachinery had an effective protection rate of 57.4Z. Further, these

Real Exchange Rates - Chile(increase implies depreciatilon)

106-- _ _

104

102 Z~

100

911 -

96 >-'~

94 -

-N

w :D2 -

90

88-

80 q----r

January April July October .January April July October

01 Actual + Trend Line

Real Exchange Rates - Peru(increase implies depreciation)

120- -__

110

o 1~I00 -

-14

90

80

c'4 ~ ~ ~ 0

January April July October January April July October

D Actual + Trend Line

Real Exchange Rates - Argentina(increase implies depreciation)

130 -

125 -

o 120-

115

K~~~ of; -E;f~Z 95 -

b-4

90

85- , ,

January April July October January April July October

0 Actual + Trend Line

Real Exchange Rates - Brazil(increase implies depreciation)

210 -

200 -

190

-~ 180

1 70 /160/

150z~~~~~~

N~~40I

130z

1290

100 '' r-T

January April July October January April July October

01 Actu-al + Trend Line

-28-

numbers understate the degree of protection afforded Bolivian firms priorto 1985. The existence of quantitative restrictions means that the weightsused to calculate average protection rates are not representative of theactual demand for imports. Tariff rates also do not reflect subsidizedcredit programs and the system of allocating foreign exchange at subsidizedrates. Nevertheless, the high level and large dispersion in effectiveprotection rates provide some indication of the great difference betweenBolivian and international prices.

Table 2.2: Tariff Rates Before the NEP(Percent)

Year

National Tariff (July 1985) /a 12.1of which:

Non-Durable Consumer Goods 20.2Durable Consumber Goods 38.1Capital Goods for Agriculture 6.2Capital Goods for Industry 8.2

Effective Protection Rates (1982) /b 44.4of which:

Textiles and Leather 169.1Wood and Wood Products 145.8Meat 80.4Machinery and Equipment 57.4

/a Weighted by the structure of imports in 1982./b Based on an input-output matrix from 1978.

Source: Tariff and effective protection rates in the tableand in the text were pruvided by Juan Antonio Morales.

63. This system had enormous implications for development. Thestifling of international competition encouraged inefficiency and impededdiversification of Bolivia's exports. The extensive system of tradecontrols and subsidies meant that nearly all economic decisions wereconditioned by Government policy. Enormous energy and resources were spentin obtaining preferential treatment from the state rather than in cuttingcosts.

64. The NEP undertook a radical revision of trade policy to eliminatequantitative restrictions and establish a uniform tariff rate. All importprohibitions and trade licensing requirements were abolished, with theexception of controls on sugar and wheat, and restrictions on importinggoods harmful to public health or safety. Initially, a tariff of 10 plus

-29-

1OZ of the rate in effect in August 1985 was imposed. In 1986, this wasmodified to a uniform 20Z rate on most goods, although exemptions totariffs established under the Investment and Hydrocarbons laws continued tobe recognized, and donations were imported duty free. Finally, in March1988 a new decree introduced a 102 tariff for capital goods and a schedulefor reducing the tariffs on other goods from 20Z to 102 by January 1990.The Government froze the non-capital goods tariff at 172 in late 1988 owingto concern over lower than expected fiscal revenues.

65. The Government has thus accomplished an impressive liberalizationof the trade regime. Remaining differences in tariff rates are minor,except for the 7 percentage point differential between consumer and capitalgoods, which is due to be phased out if the fiscal situation permits.Exemptions to the payment of tariffs granted under previous laws are beingphased out. The Government is now studying tariff exemptions granteddonations not resold in the domestic market; there is some suspicion thatconsumer goods are being imported as donations to evade the tariff. Alicensing system exists for imports of goods which, while having legitimatecommercial uses, also are necessary to the processing of narcotics. Someprivate sector managers complained that obtaining approval for such importscan take considerable time (apparently approval must be received from anumber of ministries), and that bribes have been necessary to speedprocessing. Still, with these few exceptions, the Government hf.s providedfor the free importation of goods with a low and practically uniform rateof protection.

66. The tariff reform was a wrenching experience for many enterprises,but was essential to ensure an efficient structure of production. The newsystem greatly reduced the protection that consumer goods had enjoyed underthe old regime, resulting in a sharp rise in consumer goods imports, bothlegal and illegal. Over time, firms should take advantage of the newprofit opportunities opened by the tariff reform; increased production insome sectors of industry is already apparent. However, successfuladjustment will require increased investment, which is now constrained bythe low level of savings. Thus, the profitability of production using theexisting capital stock has fallen, while the cost of acquiring capital toproduce goods more consistent with the new structure of incentives isprohibitive. While the NEP has effected a tremendous improvement in theincentives governing the production of tradeables, an increased supply ofsavings and fall in interest rates is essential before the full benefits ofthe system can be attained.

III. Labor Markets

67. Introduction and Summary. A key element of the Paz EstenssoroGovernment's liberalization program was reduced state intervention in the.Labor market. Supreme Decree 21060 eliminated a number of the laborregulations imposed in past years, essentially returning to the provisionsof the Ley General de Trabajo of the 1940s. This change allowed wagelevels to be set without Government interference (with the exception ofminimum wage legislation) and eased restrictions on firing employees.These changes in labor regulations provided a structure of incentivesconducive to harder work and allowed firms more flexibility in their use oflabor. Reliable data on the impact of this policy are difficult to obtain,

-30-

although interviews with business and labor representatives suggest thatthe new labor regulations:8 a) increased labor discipline and reduced theincidence of strikes; b) may have had a small, positive impact onproductivity; and c) may have led to the dismissal of union leaders andincreased evasion of social security payments. The Government also movedto control public sector wage expenditures through reducing staff andconsolidating the extremely complex system of bonuses and in-kind paymentsinto the basic wage. This reform achieved a more transparent system ofwages and reduced the costs of administering the public sector compensationsystem.

68. Labor Policies Before 1985. Before inaugaration of the NEP,Government regulations severely restricted firms' wage and employmentdecisions. Wages were determined on the basis of large scale negotiationsaffecting many levels of the economy, limiting the ability of employers touse salary adjustments as incentives. Restrictions on firing weakenedlabor discipline. Essentially, most workers could expect little reward forgood performance, and a low probability of punishment for poor performance.

69. Mismanagement of the important Government role in wage andemployment decisions exacerbated social tensions. The early 1980s weremarked by considerable labor strife, reflecting the intense competitionamong social groups for real wage adjustments which culminated in thehyperinflation of 1984-85. A confusing wage policy increased theseconflicts by heightening uncertainty. Most of the failed stabilizationprograms prior to August 1985 were unclear regarding the extent and timingof required adjustments in the real wage. For example, while the November1982 progran included a lagged indexation of nominal wages, in practice thepromised automatic adjustments were not made. Rather, increases in nominalwages were granted in response to pressure from strikes, resulting in widefluctuations in real wages and a sharpening of labor disputes.

70. Serious deficiencies also existed in public sector wage andemployment policy. Efforts to maintain public sector employment in theface of the economic crisis resulted in a severe fall in real wages,declining efficiency, and serious morale problems. Many public sectoremployees took on additional jobs during this period to maintain theirstandard of living, reducing the amount of energy and time spent on theirpublic duties. Low wages and overstaffing encouraged corruption. Anextremely complex system of compensation increased administrative costs andmade it impossible to determine whether relative salary levels wereequitable or rational. Up to 1982, the minimum wage had remained unchangedin nominal terms since the 1950s. Wage increases were granted largely

8/ Much of the analysis of the impact of labor regulations is based oninterviews, owing to the lack of hard data on many of the moreinteresting issues involved. Every effort was made to talk torepresentatives of different social and political groups, and onlythose opinions supported by a number of sources are included. Still,the reader should note that many of the judgements in this sectionreflect the (possibly biased) opinions of interviewees, and are notsupported by data.

-31-

through bonuses, access to goods at subsidized prices, and in-kindpayments, rather than changes in the structure of basic wages. Thesebenefits, which made up 702 of total compensation in 1982, varied greatlyby sector, firm, and employee. Efforts were made to rationalize thestructure of wages beginning in October 1982, including a consolidation ofsome (but not all) bonuses. However, other forms of compensation,particularly the costly system of subsidized company stores managed byCOMIBOL, continued to exist up to 1985.

71. Labor Policy Under the NEP. The Paz Estenssoro Government adopteda new policy aimed at raising efficiency in the public and private sectorsof the economy. Supreme Decree 21060 mandated three important changes inlabor market legislation. The Decree consolidated a variety of forms ofcompensation into the basic wage. This involved eliminating most bonusesand cutting off access to subsidized goods formerly enjoyed in many stateenterprises and ministries, accompanied by an increase in the basic wage.The Decree eased restrictions on employment decisions, making it easier foremployers to fire workers. Finally, wages were allowed to be determinedthrough collective bargaining at the level of individual firms, and allGovernment restrictions on private sector wage levels were abolished, withthe exception of the minimum wage.

72. Impact on the Public Sector. The consolidation of bonuses wasintended to reduce administrative costs and the public sector deficit, makethe wage scale more transparent, and lower the potential for abuse of therules. It is not clear whether the new wage scales fully compensatedworkers for the loss of access to subsidized goods, as the value of suchservices is hard to measure. In any event, their elimination should haveeased the administrative burden on the state. Certainly the consolidationdid not by itself establish a more equitable compensation system. However,the increased transparency of the compensation system was a prerequisite tothe rationalizaLion of salaries and civil service procedures now beingstudied by the Government. Disturbingly, it appears that bonuses have beenreintroduced in some public sector entities, contrary to the provisions ofSupreme Decree 21060. The Government will need to eliminate these bonusesif the civil service reform is to proceed smoothly.

73. The Government also attacked the problems of overstaffing and lowwages in the public sector. Voluntary incentives for retirement wereintroduced. Government agencies were given boliviano budgets, so thatincreases in wages could be achieved only through reductions in employment.While the public sector reduced its labor force by 33,000 persons between1985 and 1987, the bulk of these cuts were in COMIBOL and YPFB. Employmentin the central ministries may have increased somewhat since 1985, as somepublic sector entities have increased their wage bill, contrary toGovernment policy. Further, little has been done to eliminate existinginequities in salary levels and personnel practices. For example, the gapbetween salaries in the ministries and the state enterprises remains large.

74. Real wages in the public sector, particularly in the centralministries, remain very low, making it difficult to compete with theprivate sector for Bolivia's limited supply of well-trained managers andtechnicians. In key areas of public administration, the Government andinternational donors have paid salary supplements to maintain qualifiedpersonnel. This system has also proven unsatisfactory because of the lack

-32-

of standardized criteria for salary levels and resulting high turnover, andresentment by staff not eligible for supplements. The Government iE nowstudying means to rationalize the public sector salary structure, wiLhassistance by the World Bank and the UNDP.

75. Impact on Labor Relations. The rules reducing protection againstfiring and providing for wage negotiation at the firm level contributed toa fall in hours lost through strikes. Many enterprises cited a reductionin the frequency of labor disputes since inaugaration of the NEP. Theregulatory changes no doubt weakened the bargaining position of the largelabor unions and helped to insulate private sector wage decisions frompolitical pressure. Of course, the observed decline in strike activity cannot be attributed solely to changes in labor legislation. The contributionof intense competition over wages to the hyperinflation reduced popularsupport for strikes, weakening the solidarity among various groupsessential to the success of strikes during the early 1980s. Also, the fallin inflation lowered uncertainty and reduced the need for frequentstruggles to maintain real wage levels.

76. Labor market representatives claim that the new regulations haveeroded traditional barriers against unfair or undesirable labor practiceson the part of management. The most serious charge is that labor unionofficials have been dismissed to impede union activities. Observers alsonoted an increase in hiring on short term contracts. Under presentregulations, workers hired for less than three months are not eligible fortermination benefits, and management does not have to make social securitycontributions on their behalf. To avoid these costs, many unskilledlaborers are hired for 89 days and then dismissed. The large pool ofunemployed apparently makes this a feasible strategy for firms who do nothave to make even minimal investments in training. However, firms thatmust train employees in specific tasks would find such practicesinefficient. The private sector participation in the social securitysystem is relatively small, so it is not clear how widespread this practiceis. Interviewees also claimed that management has used the easier rules toget rid of older employees, who have more seniority and hence higher wages.

77. Some of these practices may conflict with values of importance tothe society. If collective bargaining by labor unions continues to be thestandard form of wage negotiation in the formal sector, then safeguards arenecessary to protect union organizers against retaliation by management.The implications for equity of firms reducing social security tax paymentsare more complex. Requiring employer contributions to social securityreduces the demand for labor, so efforts to avoid paying such taxes mayincrease employment. However, as long as the provision of social securitybenefits (largely health and insurance services) is an agreed social goaland the Government provides social security benefits tied to employment, itis logical to tax firms to finance part of the costs. Social securitybenefits increase the attractiveness of working in the formal sector, andthus constitute a subsidy to the firm's wage payments. It is thereforeappropriate to tax the firm to provide this service.

78. Impact on Productivity of Labor. The contribution of the changesin labor regulations to raising productivity is unclear. Improvements inthe productivity of labor from the new regulations can be separated intothree categories: a) increased individual efforts owing to a more efficient

-33-

structure of incentives; b) one-time gains through firing redundant workersprotected by the old regulations; and c) gains in efficiency throughsubstituting capital (or other inputs) for labor.9 Interviews withmanagers strongly support the view that improvements in incentives haveincreased individual labor effort. Employees who fear losing their jobs ina period of depressed economic activity tend to work harder. It is notpossible to judge whether significant lay-offs of redundant workersoccurred, except in the public sector, where a substantial fall in theworkforce was achieved through the reorganization of COMIBOL. However,given intense competitive pressures on firms owing to the recession andrise in imports, it would be surprising if economies were not made. It isunlikely that substantial productivity gains were achieved through adoptinga more rational capital/labor ratio, which would have required increasedinvestment. Private sector investment has declined in real terms since1986, owing to low demand and high real interest rates, so there has beenlittle scope for such improvements. Thus, while anecdotal evidencesupports the view that the labor regulations changes have increasedproductivity, their impact has probably been attenuated by the economiccrisis.

79. Available data do not show a rise in labor productivity since1985. Employment in the formal sector of the economy increased marginallyfrom 1985 to 1987 and GDP declined, indicating a fall in productivity (seeTable 2.3). However, information on both GDP and employment in Bolivia isnot very reliable, particularly as a portion of production and employmentin the informal sector of the economy is not captured in officialstatistics. Further, we lack the data required to distinguish the impactof other factors on GDP from changes in the productivity of labor. It isthus difficult to draw a firm conclusion concerning the impact ofregulatory changes on the basis of the data. In addition, the labor marketregulations directly affected only the formal sector of the labor force,defined as nonagricultural workers in enterprises reporting earnings to theMinistry of Labor, theoretically all enterprises with five or moreemployees. As the most recent estimates indicate that 60 to 65Z of thecivlian labor force work in the informal sector, the importance of theregulatory changes for overall economic activity is accordingly smaller.

9/ This discussion is limited to labor productivity, as measured by outputper employee. We lack the data necessary to measure the productivityof capital or total factor productivity.

-34-

Table 2.3: Wages and Productivity(1983 - 100)

1983 1984 1985 1986 1987

Real Wages /aAverage Wage 100 96 129 91 110Unemployment Rate (Z) 12 15 18 20 21

ProductivityEmployment /b 100 98 97 98 100Output 100 99 97 95 97Productivity /c 100 101 100 97 97

/a Deflated by the Consumer Price Index./b Excludes temporary employment created by the Emergency

Social Fund./c Defined as the output index divided by the employment

index.

Sources: IMF and Herbert Muller

80. Impact on Wages and Employment. The theoretical impact of theregulatory changes on the demand for labor, and hence on wages andemployment, is ambiguous. Firms which under the old rules had been unableto reduce their workforce despite declining demand probably took theopportunity to lay off redundant workers. On the other hand, to the extentthat firms had avoided hiring emplcyees because of the difficulty and costsof termination, permitting a freer labor market should have increased thedemand for labor. Also, it can be expected that the weaker bargainingposition of the labor union movement as as result of the regulatory changeswould slow growth in real wages.

81. While it is difficult to isolate the impact of regulatory changesfrom other events, the formal sector of the Bolivian economy has suffered arise in unemployment and a fall in real wages since 1985, consistent with adecline in the demand for labor. Unemployment in the formal sector hasincreased, by some measurements to over 20Z of the labor force.1 0 Thedata on real wages in the formal sector show a decline since 1985. Whilereal wages fell substantially in the first year under the NEP (the averagesalary as defined by Muller dropped by 29? in real terms from 1985 to

10/ However, this estimate (from Ministry of Labor Statistics) is based onvery imprecise measurements of the labor force and unemployment. SeeKlinov, Ruth, Public Sector Employment and Wages in Bolivia.

-35-

1986), they have since recovered, althouSh by end 1987 real wages remainedbelow their 1985 level. Another sourcel sees a similar pattern of realwages, with the average real wage falling by 142 in 1986, and then risingby 92 in 1987. Still, falling demand for labor is more easily explained bychanges in economic activity than by changes in policy. In particular, thesharp drop in real wages and slow recovery thereafter mirrors the patternof changes in output. It is thus likely that changes in economic activitywere more significant determinants of wage levels than the regulatorychanges.

82. The numbers given above provide a distorted view of the Bolivianeconomy, as they refer only to the formal sector. As savings are low andthere is no unemployment assistance, it is unlikely that the recorded risein unemployment actually occurred (although hours worked may well havedeclined). Rather, it reflects a shift of individuals from the formal tothe informal sector of the economy.12 The switch from formal to informalsector employment also implies that real wages declined by more thanindicated in the official statistics, which include only those workers inthe urban, formal sector. The rise in informal sector employment waslargely in labor-intensive commercial activities where wages are relatirelylow. For example, many of the thousands of miners lismissed during thereorganization of COMIBOL probably ended up carrying contraband importsacross the border or as micro-vendors in La Paz. Stagnant wages in theformal sector thus imply a decline in wages on average in the economy.

IV. Investment Incentives

83. Introduction and Summary. The framework of incentives adoptedunder the NEP was designed to encourage greater private sectorparticipation in the economy. Free convertibility of foreign exchange,elimination of price controls, reduced Government intervention in laborcontracts, and commitment to price stability greatly improved the climatefor private investment. All of these issues are discussed elsewhere inthis report. Here we would like to focus on policies specifically aimed atencouraging domestic and foreign investment. While the Government hasdrafted a series of new laws which would greatly improve investmentincentives, these laws were not approved in the last session of Congress.It is imperative that the new Government move promptly to consider thislegislation, to encourage the domestic and foreign investment essential forgrowth.

84. Investment Code. The investment code drafted by the Jovernmentwould provide a number of important benefits. A comprehl.usive statement ofthe regulations affecting foreign investors, particu.arly administrative

11/ An article by Rolando Morales in Coyuntura Economica Andina, June 1988.

12/ Data from the Urban Household Survey show a decline in labor forceparticipation rather than a rise in unemployment. See the paper byKlinov cited in preceeding footnote.

-36-

requirements, will reduce the costs involved in gathering this information,thus increasing the supply of investment. A formal statement of Governmentpolicy also may raise the credibility of the Government's commitment to thepresent policy framework. The law would reduce the delays imposed byadministrative requirements by permitting foreign investment without priorauthorization, except for investments in education, commiunications media,arms, food distribution, mining and hydrocarbons (the latter two are to begoverned by separate investment codes). However, foreign investors wouldbe required to register with the Government within 30 days of beginningoperations. Responsibility for overseeing this process is given to theCentro de Promocion de Inversiones, which should speed the administrativeprocess and avoid inconsistent treatment by different Government agencies.Foreign investors (except in mining and hydrocarbons, which are discussedbelow) are guaranteed an unchanged tax regime for a period of 10 years.Guarantees of equal '.reatment with domestic investors, equal access todomestic credit (except subsidized investment credits), and full payment inthe case of expropriation are also included.

85. Discussions with private sector representatives revealed divergentviews over the importance of the investment code and its probable impact inencouraging increased investment. Some managers felt that passage of thecode was essential to provide necessary guarantees of stability and fairtreatme.at to foreigni investors. Others, however, believed that theinsurance available under MIGA and individual countries' investmentinsurance agencies (for example, the US Overseas Private InvestmentCorporation) provided a much more credible guarantee against expropriation.There was considerable skepticism over the value of long term guarantees ofthe present economic policies and tax regime. Given recent politicalinstability in Bolivia, it is impossible to ensure that subsequentGovernments will not change the law. While the degree of credibility ofthe guarantees in the proposed investment law may be open to question,failure to pass the legislation at this point would raise questionsconcerning the degree of political support for the present economicframework and would probably increase potential investors' skepticism overthe stability of the present economic regime. Thus, passage of the code isessential to create the proper climate for increased foreign investment.

86. Mining Code. Investment in the mining sector in Bolivia has beenlow since the nationalization of the major tin mines and increasedGovernment involvement in the sector beginning in 1952. Legal constraintson private development and the reserving of some of the more promisingareas for public use have limited the role of the private sector, while thepublic sector has generally neglected investment in mining in favor ofother sectors. Given its great importance for Bolivia's development, theestablishment of a framework of incentives which would facilitate recoveryof the mining sector is essential.

87. The Government has completed work on a new mining code which,while not proposing a radical change to existing rules, would providevaluable improvements over presenit legislation. A more effective systemfor registering claims would be established, and a specialized office setup to administer claims. COMIBOL's ability to enter into partnership andjoint venture contracts with the private sector would be increased,facilitating the exploitation of COMIBOL's mining potential with minimalinputs of public financial resources and under majority participation and

-37-

management by the private sector. The system of taxation for mining wouldbe revised to allow investors to choose between a tax on profits and thepresent system of regalias (a tax on the presumed profit from mining ascalculated by the Government). The profits tax would enable foreigninvestors to deduct their payments of Bolivian taxes from their taxliability to their home country, which under the tax rules in most of theindustrial countries is not possible under the regalia system (see Chapter3 for a discussion of the new tax provisions). Foreign investors would bepermitted to engage in mining activities within 50 miles of the border,despite the present Constitutional prohibition, by entering into servicecontracts with the Government.

88. Foreign investment is essential to develop the potential of themining sector in Bolivia. Bolivian firms lack the capacit-; to make thelarge-scale investments 3ften required, and a relatively high level oftechnology, only available from ,oreign firms, is needed to exploit variouspotential sources of minerals. Unfortunately, the mining code was notapproved in the last session of Congress. While investors have recentlyexpressed considerable interest in the Bolivian mining sector, we do notexpect a significant rise in foreign irvestment in mining without a reformof the tax system and the provisions for joint ventures. A priority forthe new Government in August should be to review the code and present it tothe Congress.

89. Hydrocarbons. The investment climate in hydrocarbons sufferedgreatly under the policies of previous Governments. The nationalization ofvarious oil companies discouraged private sector activities. Under thehydrocarbons law promulgated in 1972, YPFB (the state oil company) wasgiven primary responsibility for the development of the country'shydrocarbons reserves. YPFB could enter into service contracts withforeign oil companies, but could not provide these companies a share of therisks and profits from exploration, limiting the extent of foreigninvestment in the sector. The consequence of low levels of investment hasbeen a steady decline in proven reserves. Without development of new oilreserves, it is estimated that Bolivia will become a net oil importer bythe early 1990s.

90. The Government has made some progress in improving incentives forinvestment in hydrocarbons. Most importantly, a longstanding dispute wassettled with Occidental Petroleum and Tesoro, under which Bolivian paymentswill be reinvested in petroleum exploration in Bolivia. This agreementshould also encourage exploration by other oil companies. As only about20Z of the area of the country has been investigated for oil deposits, itis likely that further discoveries will increase Bolivia's proven reserves.The Government has also been working on a new hydrocarbons code which wouldprovide for joint ventures by privatc sector oil companies with YPFB todevelop the areas reserved to YPFB under existing law. However, theGovernment did not introduce this legislation to Congress, so that privateinvestors continue to be barred from exploiting some of the areas withgreatest promise for finding oil reserves. The next administration shouldgive high priority to considering the prosposed changes in the hydrocarbonscode.

-38-

91. Agriculture. The Paz Estenssoro Government prepared acomprehensive law reforming policies and the institutional framework of theagricultural sector, although the law was not submitted to Congress. Thedraft includes a land reform program to consolidate small farms into moreviable economic units, establishment of a court to settle disputes overland titles, and a new institution to promote rural development. The lawalso provides for other measures to promote growth in agriculture, anddefines the roles of .he various public sector entities involved in thesector.

92. While many of the proposals in the draft have merit, the law needsto be implemented through specific actions. The Government needs tocarefully study many of the issues raised in the agricultural law, andpresent more concrete proposals for policy and institutional changes. TheBank and the Government are now considering a series of studies of theagriculture sector which could be financed under IDA's Economic ManagementStrengthening Operation. These studies would review public sectorexpenditures in agriculture, analyze financial institutions active in thesector, propose reforms of the system of land tenure and tax incentives,investigate how to improve coordination of donor activities, designspecific means to implement the Government's sectoral strategy, andrecommend administrative improvements in the Ministry of Agriculture.

V. Public Sector ManaRement

93. Introduction and Summary. Political instability, economic chaos,and diminishing financial resources severely eroded the effectiveness ofthe public sector during the 1980s. The quality of services provided bythe public sector deteriorated. Public sector expenditures were poorlycontrolled, as decision makers lacked basic information on actual spendingby the multiplicity of public sector entities. Huge losses suffered bymany state enterprises constituted a severe drain on Government revenues.

94. The Paz Estenssoro Government moved immediately to address themost pressing deficiencies in the public sector. The Government also begana longer-term. -rogram designed to improve administration of public servicesand increase the efficiency of state enterprises. This program includesimproved cost-control procedures, efforts to improve tax administration, astudy of public sector wage levels and civil service procedures to improvethe equity and efficiency of the compensation system, a review of stateenterprise operations to increase their efficiency or sell them to theprivate sector, and a series of studies to design an effectivedecentralization of public services. Despite the major gains in publicsector administration under the Government of Paz Estenssoro, the newGovernment will face a number of urgent challenges, most importantlymanaging the decentralization process.

95. State Enterprises. The state greatly increased its share in theproduction of goods and services after the 1952 revolution. This was mostnotable in the exploitation of minerals and hydrocarbons (where by theearly '980s the state controlled, respectively, 65Z and 802 of production),but also included a variety of smaller productive activities. While theGovernment's share of manufacturing activities is more difficult toestimate, it probably exceeded 50Z. The state owned glass, textile, and

-39-

cement factories, and produced or marketed dairy products, sugar, rice,maize by-products, edible oils, and tea. Many of these enterprisesrequired substantial transfers from the Government budget to cover losses.

96. Regaining control of public finances through controlling stateenterprise deficits was a key element of the stabilization program. TheGovernment dissolved the Bolivian Development Corporation and transferredthe state enterprises formerly under its control to the regionaldevelopment corporations, in the hope of increasing efficiency by placingadministration of the enterprises closer to their operations. To reducetheir ability to run deficits, access of state enterprises to Central Bankcredit was curtailed. A reorganization of COMIBOL was undertaken to reduceemployment, streamline administration through eliminating the system ofsubsidized food stores, and close uneconomic mines. The collapse of tinprices in late 1985 exacerbated COMIBOL's problems, requiring deep cuts inproduction and employment (COMIBOL's labor force eventually fell from30,000 to 7,000 employees). The Central Bank was reorganized to strengthenits traditional monetary and regulatory functions, and reduce its role as asource of credit for the private sector. The Government also began areorganization of YPFB, the state hydrocarbons vompany. While changes inthe administrative structure of YPFB have been implemented, little progresshas been made in divesting elements of YPFB's operations w.,ich could bemore efficiently carried out by the private sector. In particular, theGovernment should move forward with plans to allow moze private sectorparticipation in the distribution and marketing of natLral gas, as well asin exploration and development of new gas fields.

97. The Government has initiated a program designed to improve theoperation of state enterprises and sell some of them to the private sector.A series of studies of the state enterprises are underway to locatedeficiencies, determine what kinds of assistance would improve operations,and estimate the fair market value of the enterprises. A report on thestatus of a selected group of enterprises is scheduled to be submitted tothe Government by mid-1989.

98. The Government has proceeded cautiously in the privatizationprogram, and prospects for the program are uncertain. The next step afterreview of the report will depend greatly on the nature of and receptiongiven its recommendations. Cooperation by the regional developmentcorporations, which have direct responsibility for the enterprises, wouldbe voluntary. The Government has introduced legislation (as part of theSAFCO law) which would ease the rules for the transfer of corporationsthrough public auctions, which now requires Government approval on anindividual basis. Unfortunately, the SAFCO law was not enacted in thisCongressional session, which means that no progress can be made onprivatization until the next session in August.

99. The next Government will face difficult decisions concerningprivatization. The selling of state enterprises is a sensitive issue inBolivia; the charge that companies were sold for insufficient compensationis difficult to disprove, given the complexity of determining fair marketvalue. Also, the market within Bolivia for these companies is probablyrather thin, given the great concentration of wealth. Thus, the newGovernment will need to ensure a sufficient political consensus behind theprivatization measures, and should take steps to interest foreign investors

-40-

to increase the competitiveness of any auctions. While the difficulties inthis process are clear, the potential benefits are well worth the effort.Privatization of many of the state enterprises w3uld both reduce the drainon Government revenues in supporting deficit operations and provide theopportunity for their more efficient operation.

100. Administration. The Government has taken a number of measures toimprove administration. Under the SAFCO program, standardized accountingpractices and methods of cost control are being introduced throughout thepublic sector to increese the information available to managers on publicsector expenditures. This program includes the introduction of modernprocedures for accounting, auditing, budgeting, and management of publicsector credit; the establishment of systems in most public sector entitiesto implement these procedures; and a large training program in how to usethe new systems. These reforms should help enable the Government to reduceexpenditures in a mcre efficient manner than the ad-hoc measures (freezingof accounts, across the board budget cuts, build-up of arrears, and generalprohibition against depending on Central Bank credit) used since 1985, andgenerally improve the control of public sector funds. To date, effortshave concentrated on obtaining accurate cash flow data from 26 major publicsector ertities, many of which lack acceptable accounting systems. TheGovernment also has formulated new guidelines for auditing, under whichaudits will have tzo be performed according to internationally acceptedstandards and the opportunities for corruption during the audit processwill be greatly reduced. While an excellent beginning has been made inimproving the control of public expenditures, passage of the SAFCO law isessential to ensure that the new procedures become permanent.

101. Significant improvements in tax administration also have beenaccomplished, including a restructuring of the Internal Revenue Service,adopting of new enforcement procedures, and a system for tax collectionsand compilation of data through the private banking system. Work is nowbeginning to address the considerable deficiencies of the customs service,ex-.end the new tax administration system to the regions, and introduce animproved auditing system (see Chapter 3 for a discussion of taxadministration). Increasing the efficiency of the public sector should beviewed as a long-term process; the progress made since 1985 is only abeginning. Just consolidating the improvements made and implementingcurrent plans will require a considerable commitment of resources by thenext Government.

102. Beyond the areas already discussed, the most urgent problem inpublic sector administration is to increase the efficiency of Governmentagencies providing basic health and education services. The Paz EstenssoroGovernment has formulated plans for a decentralization of these services tothe regional development corporations and municipalities; it is hoped thatdecentralization will improve services through increasing localaccountability for their quality. The new Government will quickly have tomake decisions on whether and how to progress with the decentralization.It is essential that the public entities given the responsibility forprovision of services have the administrative capacity and resourcesnecessary to do so, and that the transition proceed smoothly.

-41-

103. More generally, the question of a broader decentralization of thepublic sector must be carefully studied to establish legal and proceduralnorms for the process, define an equitable and efficient allocation offinancial resources, determine the forms of technical assistance necessaryfor the corporations and municipalities, and clearly delineate the divisionof responsibilities among different levels of government. Work on thesehighly complex and extremely important issues is now begining, with supportfrom a technical assistance loan by IDA.

104. Price Policy. Prior to 1985, the state subsidized variouseconomic activities through maintaining unrealistically low levels ofpublic sector prices. Slow adjustment of official prices during thehyperinflation led to extremely low prices for public goods, resulting in aconsiderable drain on the budget and a misallocation of resources.Gasoline prices remained at levels well below international prices or thoseprevailing in neighboring countries, in part to facilitate commerce in theface of the difficult geography and poor road system. This policy resultedin substantial illegal exports of gasoline to take advantage of the pricedifferential. The price of electricity was also low, reducing capitalinvestment in the power sector. State enterprises generally priced theirgoods well below their opportunity cost, and covered their losses throughsubsidies from the budget. In transportation, trucking prices were kept athigh levels by a legally-sanctioned cartel of private firms. Lack ofcompetition from trucking allowed the state railroads to charge highfreight rates, the revenues from which were used to subsidize the very lowpassenger fares.

105. The Paz Estenssoro Government raised public sector prices tolevels prevailing in the international market or in neighboring countries.This change reduced subsidies from the Central Government budget to thestate enterprises and helped to control the demand for public sector goodsand services. Thus, the new policies contributed to the stabilization ofthe economy and increased the efficiency of the use of resources.

106. The price of gasoline and other oil derivatives was increased tolevels approximating their cost at the border, as calculated beforesubsidies by neighboring countries. The price of gasoline increased bymore than three times (in real terms) between August and September 1985.This proved to be the essential support for Government revenues in late1985 and in 1986 and also helped to control inflation by soaking upliquidity in the economy. Some sectors of the economy experienceddifficulties in adjusting to the jump in transport costs. Transportationof some industrial and agricultural products became unprofitable,particularly those with low value per volume. There are reports thatsignificant portions of the 1986-87 fruit harvests could not be shipped tothe cities, which would help explain the very low prices for fruit observedin rural areas.

107. The Government has had mixed success in maintaining gasolineprices at international levels. The price of gasoline was fixed at theSeptember 1985 price until February 1988, when it was raised by about 40Z,in part to discourage smuggling of Bolivian gasoline in some neighboringcountries. Small, monthly adjustments were made until July, when this

-42-

policy was discontinued due to concern over rising inflation. By July thegasoline price was well above the border price. Periodic instability inthe supply and demand for gasoline in some neighboring countries isinevitable, owing to distorted pricing policies and wide variability ininflation in those countries. The Government can not hope to compensatefor this instability, and should instead continue to focus on maintainingthe gasoline price close to international levels. If increased revenuesare required, a retail tax on gasoline might be considered. Oneimprovement in allocative efficiency would be to end the requirement thatgasoline prices be equal throughout the country, to better reflectdifferences in transportation costs.

108. The Government increased rail freight rates substantially in 1985.However, in response to increasing pressures from users, freight rates werelowered in 1988, in part compensated by a 50? hike in passenger fares. Thefare structure adopted for freight consisted of continued high rates forimports, a 30? discount for internal traffic, and a 60? discount forexports. The discounted fares corresponded fairly closely to marginalcost. However, the rigid discrimination between export and import farescreated periodic imbalances in demand between outgoing and incoming freighton the Arica-La Paz line, resulting in substantial excess capacity in thetrip from Arica to La Paz. An IDA-financed consultant recently recommendeda new fare structure for freight which will provide for more flexibility insetting fares and allow more efficient operations. This proposal is nowbeing implemented. Proposals for a reform of passenger rates are alsobeing considered.

109. Supreme Decree 21060 also had provisions for the prices of variouspublic services. The maximum rate on electrical power for industrial usewas fixed at US$.45 a kilowatt/hour, while the rate for electricity used byhomes, as well as urban transportation fares, were to be set by localgovernments after negotiations with concerned parties. Tariffs ontelephones were set at levels close to those prevailing internationally.

110. The role of the state in determining the prices of goods andservices produced by the private sector was reduced. Supreme Decree 21060eliminated the extensive system of price controls and marketing boards,which covered most agricultural commodities sold in the country. Prices onall goods were allowed to be set by market forces, with the exception ofsugar and wheat products. The retail prices of wheat flour and bread areset by the Government. Sugar prices are set by agreement with producers,and are substantially higher than the world market price. With these twoexceptions, however, the NEP greatly increased the role of the market inthe allocation of resources through returning pricing decisions to theprivate sector.

1ll. The benefit to agricultural producers appears to have been modest,as agricultural production has declined steadily since 1985. Other policychanges impaired growth in the sector, for example the reduction ofsubsidized lending and higher gasoline prices. It is also possible thatprior to 1985 most production was not actually sold at the state price, butrather through informal markets, so that removing the price controls hadlittle impact on output.

-43-

VI. The Informal Sector13

112. Introduction and Summary. The economic crisis since 1980 has

increased the importance of informal sector activity in the Bolivianeconomy. Informal economic activity reflects the desire to evade taxes,

avoid costly and time-consuming registration requirements, and minimize

payment of bribes. The presence of a large informal sector imposessignificant costs on the economy (compared to the same level of economic

activity carried out through formal channels), owing to lower tax revenues,

the smaller and less efficient scale of operations, and lack of legal

sanctions for business agreements. A more efficient structure of

production and commerce could be achieved by reducing the incentives for

informal sector operation, for example by easing registration requirements

and reducing corruption. However, even under a more efficientadminist-ative structure, the evasion of taxes would still provide a

significant incentive for informal sector activity. Thus, we can expect

that a large informal sector will continue to exist for some time to come.

Since stricter enforcement of the law would probably be counterproductive,Government efforts should focus on requiring informal sector enterprises to

pay for public services, which would both raise needed revenues and reduce

the competitive advantage that informal sector firms enjoy relative to

formal ones.

113. Definition. Before analyzing the importance and implications of

informal economic activity, it is helpful to define what we are talking

about. The informal sector is viewed in different ways by differentobservers. We will adopt the convention that the informal sector includes

all economic activity done without formal approval or registration from

Government authority, where such is normally required. This definition

excludes activities which are illegal in and of themselves, such asinvolvement in narcotics, prostitution, gambling, and theft. Included asinformal sector activities would be importing contraband, unlicensedproduction, unlicensed retailing, and various forms of tax evasion.

114. Size. Although reliable data are difficult to obtain, it is clear

that the informal sector makes up a large portion of the Bolivian economy.Some observers have claimed that about 60-652 of the civilian labor forceare involved in informal sector activities. This figure includes the vast

majority of Bolivian farmers, most of whom pay no taxes and many of whom

lack titles to their land. Informal economic activity is also high in

urban areas. It is claimed that 74Z of labor employed in manufacturing and

56? of construction workers in La Paz are informal, and that 86? of

commerce, 43Z of transportation, and 25? of services are providedinformally. These figures are admittedly unreliable. Besides the inherentdifficulty in measuring activities designed to evade Government scrutiny,

estimation of the impact of informal sector activity is impeded by the fact

that many formal sector establishments engage in some form of informaleconomic activity. For example, registered manufacturers may retail a

portion of their goods to unregistered micro-vendors, or may produce aportion of their goods in an unregistered establishment.

13/ This chapter is based on a background paper by Oli Havrylyshyn andJoseph Pelzman entitled Informal Activities in Bolivia. References for

the data cited can be found in that paper.

-44-

115. Incentives for Informal Sector Activity. The motivations forinformal sector activity can be divided between tax evasion, avoidance oftime-consuming bureaucratic requirements, and avoidance of the costs ofcorruption. Contraband imports avoid payment of import tariffs, the valueadded tax, customs fees, and the transactions tax, which presently total322 for consumer goods (see Chapter 3 for a discussion of contrabandtrade). The avoidance of import fees on intermediate inputs and capitalgoods in turn encourages informal provision of the final good. Forexample, the availability of contraband automobiles (viewed as an input)has resulted in a huge supply of informal taxi cabs (final output) in LaPaz and Santa Cruz. The importance of regulations in stimulating informaleconomic activity is harder to gauge. Formal bureaucratic requirements donot appear to be excessive, although requirements for dual registry ofbusinesses at the local and national level may unnecessarily raise firms'costs, and a number of managers (in the formal sector) complained about themultiplicity of laws governing their work. More than the formalrequirements, managers cited corruption in administering the law. Privatesector representatives noted repeated demands for bribes by Governmentinspectors and tax collectors. One person claimed that his small retailbusiness was visited an average of four times a day by various officials tocollect bribes. The actual costs involved are impossible to measure, andinformal enterprises are probably also subject to extortion by Governmentofficials. Still, frequent reports of corruption and over-regulationsuggest that they are important impediments to participation in the formalsector of the economy.

116. Incentives for informal sector activity increased during the early1980s. The highly-overvalued official exchange rate prior to August 1985provided a huge incentive for undertaking some foreign exchangetransactions outside of official channels. The sharp fall in Bolivia'srecorded export receipts after 1980 in part reflects a switch in productionto informal channels in order to earn the parallel market rate. Evenregistered firms which enjoyed access to foreign exchange at the officialrate to purchase imports could channel at least part of their outputthrough informal exporters at the parallel rate. The growing importance ofthe coca trade also encouraged informal sector activity. The system ofexchange controls made it difficult to introduce foreign exchange earnedthrough selling coca into the legal economy. The increased supply of cocadollars thus increased the demand for goods traded through the informalsector. Also, the hyperinflation greatly increased incentives fortransacting business in dollars rather than pesos, to take advantage of thegreater stability of the dollar. As this became difficult to do throughlegal channels following the dedollarization decree (which required theforced conversion of foreign exchange holdings into pesos at the officialrate), incentives for evading official recognition increased. Finally, thefall in real wages in the public sector may have increased corruption amongpublic employees, thus making it more attractive for businesses to operatewithout official registiation.

117. Some of the reforms under the NEP reduced incentives for informalsector activity. For example, the unification of the exchange rateeliminated the need to export outside of official channels to earn areasonable return in pesos. However, reduced incentives for informalsector operations may not have resulted in a substantial fall in informalsector activity. In this view, there is an initial cost to switching to

-45-

the informal sector in terms of learning how to operate without detectionby the Government, establishing contacts with other informal sectorsuppliers, and building a network of clients. Once these costs are paid,-versons will continue to operate informally despite some decline in theprofitability of informal versus formal sector business, particularly aspenalties for informal sector activity are hard to enforce. Further,significant incentives for informal sector participation continue to exist,particularly avoidance of taxes.

118. Other changes in the economy since 1985 may have raised theincentive for informal sector activity. The fall in public sectoremployment increased the supply of labor to the private sector. This mayhave reduced real wages sufficiently to make profitable highly labor-intensive activities typical of the informal sector, such as carryingcontraband goods on foot across the border and selling goods in the street.The increased demand for consumer goods since 1985 increased the importanceof commerce to the economy, a business traditionally done by the informalsector. Further, greater reliance on a value added tax and more aggressiveefforts to collect taxes by the Paz Estenssoro Government may have drivenmore businesses into the informal sector.

119. Impact. The cost to the economy of informal sector activity maybe high. Lack of formal sanction by the state imposes a number ofconstraints on efficient economic activity. In order to avoid detection,informal enterprises may restrict themselves to a suboptimal scale ofoperations. Access to credit and insurance is more difficult for informalsector enterprises, which reduces the size and increases the risk ofinformal activities. Reportedly, money lenders to the informal sectorcharge astronomical rates of interest, much higher than the very highlending rates charged by banks. Since informal sector enterprises pay lesstaxes than formal ones, Government revenues are reduced below what theywould be if all businesses were registered. While some taxes are collectedfrom the informal sector, for example site fees charged by municipalities,the major sources of Government tax revenue (value added tax, socialsecurity taxes, profits tax) are avoided by the informal sector. Finally,to the extent that Government regulations are effective in reducingnegative externalities or achieving desirable social goals, theiravoidance imposes costs on the economy. For example, contraband medicalsupplies of very poor quality have reportedly circulated widely in thecountry. We also heard claims that milk imported from Europe byneighboring countries, but taken off the market owing to suspectedcontamination from the Chernobyl accident, was showing up in Bolivianshops.

120. The economy also derives some important benefits from informaleconomic activity. To the extent that informal businesses avoidunnecessary or excessively costly regulatory procedures, they represent anet gain to the economy compared to full compliance with the law. Theavoidance of corruption, which is a tax on business which does not raiseGovernment revenues, also raises efficiency. Finally, the informal sectorhas been the most dynamic element of the Bolivian economy over the pastthree years and has absorbed large numbers of unemployed workers, whileemployment declined in the formal sector.

-46-

121. Policy. Government policy should be geared to integrating theinformal sector into the mainstream economy while not unduly restrictingeconomic activity. Such a strategy would require reducing the complexityof Government regulations on productive economic activity to somereasonable minimum consistent with acceptable health and safety standards.Simplified licensing procedures for new ventures in small and medium-sizedindustry would encourage greater formal sector participation. Reducingcorruption would be an effective means of reducing the incentive forinformal sector activity, but salaries (along with Government resources topay them) are so low that it is difficult to envision any dramatic gains inthe near future.

122. The Government should probably avoid more vigorous enforcement ofregulations as a means of increasing the cost of nonregistration. Morestringent enforcement would drive producers into smaller, less mechanised,and often less efficient, operations. Government employees would takeadvantage of stricter enforcement to collect more bribes, thus increasingthe cost of doing business while accomplishing little in terms of raisingrevenue for the state. Thus, treating contraband trade as a criminaloffense (as suggested by many formal sector managers) would probably becounterproductive. The criminalization of informal sector activity wouldbe difficult to enforce and would create great social tensions, given thelarge portion of the workforce employed in the informal sector. However,legal reforms may be useful to speed procedures when control of contrabandactivity is desirable.

123. More vigorous efforts to collect user fees for infrastructureservices may be a useful means of increasing Government revenues from theinformal sector. Such fees would include site rentals for commercialactivities, charges for electricity or water, and maintenance of publicmarkets. Evasion of user fees is difficult (compared to other forms oftaxation) because they can be linked to provision of specific services. Itis also likely that fees are more willingly paid, since the benefits aremore immediate and specific to the payer than other kinds of publicservices. Increased tax collections from the informal sector would bothraise Government revenues and reduce the competitive advantage enjoyed byinformal sector businesses over legal businesses. To facilitate taxing theinformal sector, it makes sense for local governments to provideinfrastructure for a centralized market, as is being done in some cities.Admittedly, these measures would not greatly increase Government revenuesor reduce the competitive advantage of informal versus formal sector firms.Given the tremendous problems affecting public sector administration, itcan be expected that a large portion of economic activity will take placein the informal sector for years to come.

-47-

VII. Financial SIstem14

124. Introduction and Summary. The Bolivian financial system hasstarted to recover from the economic chaos of the early 1980s.Hyperinflation severely reduced financial intermediation in Bolivia,limiting the supply of capital to firms and the efficiency of itsdistribution. Financial assets increased as a result of stabilization, butnot sufficiently to reduce real interest rates to reasonable levels. Banksavoided a severe decline in profits during the hyperinflation, owing toopportunities for profitable arbitrage based on access to credit andforeign exchange at subsidized rates. However, their portfoliosdeteriorated. Continued high real interest rates reflects a lack ofconfidence in the continuation of present policies, low savings due todepressed incomes and the impaired portfolios of some banks. Maintenanceof the present policy regime and a restructuring of the banking system toincrease soundness and the effectiveness of supervision should over timereduce interest rates, but this process may take years to complete.

125. Bolivian Financial System. Financial intermediation in Bolivia iscarried out by commercial banks, private-sector development banks (of whichonly one has a significant role in the market), Government owned banks,non-bank financial intermediaries (including savings and loans associationsand credit unions), and a large network of informal moneylenders. Thecommercial bank market is highly concentrated, with 6 banks (the five majorprivate commercial banks and the Banco del Estado) holding 611 of totaldeposits and 63Z of total capital. The degree of competition within thesystem fell during the economic crisis, as most international banks leftBolivia. In addition, five domestic banks have been liquidated. Someobservers have charged that the major banks fix both deposit and interestrates, and that competition is largely carried out through advertising inorder to increase market share. Commercials for the major commercial banksare frequently seen on television in La Paz. On the other hand, thevariety of financial intermediaries and the presence of foreign banks(albeit largely inactive ones) implies a substantial potential for acompetitive market should the opportunities for profitable investmentsincrease.

126. Chaotic economic conditions prior to August 1985 greatly reducedrecorded financial assets within the economy (see Table 2.4). Interestrate controls in conjunction with hyperinflation resulted in stronglynegative ex post deposit and lending rates, reducing the attractiveness ofdomestic assets. Total liquidity (M2) in the economy fell by 69Z in realterms from 1980-85. The dedollarization decree in 1982 converted allforeign currency contracts into pesos at the official, highly-overvaluedexchange rate. People with dollar accounts in Bolivian banks suffered hugelosses, exacerbating a decline in confidence in the banking system andleading to a sharp increase in the share of foreign currencies in domesticassets, (although that change was only captured in official statisticsbeginning in 1985, since holding dollar assets was illegal during much ofthe 1982-85 period).

14/ This section relies heavily on the Bank's Banking Sector Study (BO-6765)

-48-

Table 2.4: Monetary Conditions(Bolivianos of January 1980, End of Period)

1978-81 1983 1984 1985 1986 1987 1988 /a(average)

Money (Ml) 12.3 6.8 5.6 4.1 4.5 5.6 5.0Currency 8.3 4.8 4.9 3.6 3.6 4.4 3.6Demand Deposits 4.0 2.0 0.7 0.5 0.9 1.2 1.3

Quasi-Money 8.5 3.5 1.2 1.9 5.7 8.0 8.9Bolivianos 5.8 3.5 1.2 1.1 2.1 1.7 1.8Foreign Currency 2.8 0.0 0.0 0.8 3.6 6.3 7.1

Total Liquidity(M2) /b 20.8 10.3 6.8 6.0 10.2 13.6 13.8

/a October./b Money plus quasi-money.

Source: Central Bank

127. The decline in the real value of the officially-recorded monetarybase was accompanied by an increase in the importance of informal sectormoney lenders. Typically, these lenders tend to be less efficient thanofficially sanctioned banks, because they operate largely on the basis ofpersonal relationships without the protection of the law. This constrainsthe size of their operations and increases their costs, hence increasingthe interest rate they charge on loans. While the presence of a largeinformal financial system no doubt increased the supply of credit over whatcould have been provided through official, highly-regulated channels,greater reliance on money lenders raised the cost of financialintermediation.

128. Paradoxically, the hyperinflation may not have greatly damagedbanks' profitability. While the shrinking of the money supply (in realterms) cut their volume of business, banks were well situated to takeadvantage of the poasibilities for profitable arbitrage inherent in thesystem of financial controls. For example, banks (or their customers)could profit from subsidized credit available from the Government and theopportunity to convert pesos into foreign currency at the official exchangerate. Banks also developed strategies for increasing profits in a highinflation environment. Many branches were opened to encourage deposits, onwhich banks paid highly negative rates. To hedge against inflation, banksinvested a large portion of their assets in real estate, which enjoyed aboom until the stabilization program reduced the attractiveness ofinflation hedges. These measures helped to maintain banks's profits, butcould not avoid the deterioration in the value of their financial assetsowing to the steady decline in the economy.

-49-

129. Financial Policies under the NEP. The new Government institutedfundamental reforms in financial regulations a few weeks after takingoffice. Supreme Decree 21060 abolished all controls on interest rates andauthorized financial transactions in foreign currency and in local currencywith maintenance of value. Reserve requirements in foreign currency werelifted, and local currency reserve requirements were changed to 102 onsavings deposits and 502 on sight deposits. Domestic residents werepermitted to borrow and lend abroad. A number of directed credit schemesat subsidized interest rates were eliminated.

130. The Central Bank has begun to issue debt instruments, largelycertificates of deposit paying close to market rates of interest. It ishoped that these certificates will circulate widely in the economy, so thateventually they can be used as an additional instrument of monetarycontrol. So far, the market is not that large. As of December 1988, US$73million was outstanding, mostly held by the commercial banks and publicsector enterprises.

131. Until recently, banks could satisfy reserve requirements throughpurchasing Central Bank certificates of deposit, which provided someremuneration of reserves. In order to reduce the Central Bank's losses,Supreme Decree 22139 (issued in February of 1989) forbids banks from usingcertificates of deposit in lieu of reserves. As a result, banks can nolonger earn interest on their reserves, which will reduce profits. TheGovernment believes that the banking system is sufficiently strong toabsorb this change without difficulty, and that existing high spreadsprovide banks with sufficient profits; higher priority should be placed inreducing the Central Bank's losses. The Centr.l Bank should closelymonitor the profitability of the banking system to ensure that profitsremain within an acceptable range.

132. The Government has begun work on a reorganization of the publicsector banks to increase their efficiency, redefine their role, and reducethe public sector's participation in direct lending. The Government closedthe Banco de Vivienda, and has sold off some assets of the Banco Minero.Proposals for a restructuring of the Banco de Agricola and the Banco delEstado have now been formulated. The new Government should move forward onimplementing the reorganization of public sector banks, both to eliminateunneeded services and to increase the efficiency of those credit operationswhich are retained.

133. Impact on Finnacial Assets. Financial liberalization and the endof the hyperinflation brought an immediate increase in recorded financialassets. Bank deposits increased by 1802 in real terms from 1985 to 1986,owing to a repatriation of funds from abroad to take advantage of highinterest rates and some switching from informal to formal financialintermediaries. Disturbingly, the share of financial assets denominated inforeign currency has actually increased since 1985, reflecting continuedfears over prospects for the exchange rate and inflation. Foreign currencydeposits equalled 552 of total bank liabilities to the private sector asof June 1988, as compared to 29Z in June 198b. The availability offoreign deposits has certainly increased the su-ply of capital to theBolivian economy by providing depositors an asset less subject to the riskof devaluation. However, the large share of dollars in the financialsystem reduces the Government's ability to extract seignorage, severely

-50-

limiting the flexibility of fiscal policy. The availability of foreigncurrency deposits means that the Government must finance its deficits byforeign credit (credit from the domestic private sector is very limited) orrisk a rapid increase in inflation. Foreign currency deposits,particularly as the bulk are very short term, may be more sensitive tochanges in perceptions of Government policy than peso deposits, and thusmore likely to leave the country with a decline in confidence. The growingshare of bank deposits denominated in foreign currency thus increases therisk of banks suffering abrupt changes in their deposit level, raising thepotential for instability.

134. Impact on Conmercial Banks. Adjustment to disinflation proveddifficult for many banks. With low rates of inflation and liberalizeddeposit rates, the large network of bank branches was no longer profitable.Employment in the formal banking sector has declined sharply.Stabilization also ended the real estate boom, reducing the market value ofbanks' capital. The sharp change in relative prices which accompaniedtrade liberalization and increases in public sector prices lowered profitsin many enterprises with large debts to the major banks. Uncertainty overeconomic prospects also greatly increased the share of short termliabilities in banks' deposits, which increases the risk and limits theamount of long-term lending the banks can do. Banks thus experienced ageneral deterioration in the quality of their portfolios and in profits.

135. Some improvement in bank portfolios can be seen in 1988 (Table2.5). Loan overdues fell from 8.O0 of assets in 1987 to 6.3Z in 1988, andfrom 67.3Z of capital to 61.OZ. Observers generally believe that thebanking system is in better shape than in the years immediately followingstabilization. However, these figures may understate the weakness ofbanks' portfolios. The level of overdue loans excludes nonperforming loanswhich are refinanced continually by the banks. As banks are not permittedto accrue interest on loans even one day overdue, there is a strongincentive to refinance loans. Further, the book value of banks' capital isprobably overstated, as the sharp fall in the market value of banks' realestate holdings has not been reflected in balance sheets.

Table 2.5: Commercial Bank Portfolios(Percent, End of Period)

1986 1987 1988

Overdue Loans/Assets 7.5 8.0 6.3Reserves/Overdue Loans 70.4 68.7 78.7Capital/Assets 8.9 11.9 10.3Overdue Loans/Capital 83.4 67.3 61.0

Note: Refers to private sector commercial banks only. Excludes banksclosed in 1987.

Source: Superintendent of Banks

-51-

136. Many Bolivian banks also suffer from administrative deficiencies.A scarcity of middle management has impeded the decentralization ofoperations and the adoption of more modern systems of control. Informationsystems in some banks are inadequate, making it difficult to controlembezzlement. Some observers blame an overly traditional approach tobanking for the lack of diversification into different credit services orgreater reliance on commissions, although the demand for such services (forexample consumer loans) in Bolivia may not be great.

137. The Bolivian Government, with the assistance of the World Bank, isnow undertaking a program to strengthen the commercial banks. As acondition for participation in the on-lending of IDA's Financial SectorAdjustment Credit, banks were required to submit to an audit of theircperations. The Government has begun measures to strengthen bankingsupervision. Regulations now require banks to increase their collateral onoutstanding loans and increase their capital asset ratios to moreacceptable levels. A large proportion of banks' assets have been lent withno guarantees beyond personal signatures, in many cases to people withfinancial interest in the bank. It is hoped that formal collateralrequirements will strengthen banks' portfolios and facilitate a decline inthe concentration of bank loans. Banks will submit a plan for the sheddingof nonperforming assets within two years. Failure to get rid of theseassets within two years will lead to their purchase by the Central Bank,using half cash and half certificates of deposit. In conjunction with thisprogram, the Central Bank will provide loans to assist in the refinancingof banks' assets into longer maturities at lower interest rates than nowprevailing in the market.

138. Impact on Interest Rates. Real interest rates rose to extremelyhigh levels immediately following the stabilization and liberalizationprogram begun in August 1985. Bank lending rates in pesos dropped sharplyfrom 462 (a month) in August 1985 to 142 in September, as the depreciationreduced the required premium for holding peso-denominated assets. Sincethen, nominal interest rates have declined steadily. However, inflationhas fallen much more rapidly, and real interest rates (comparing nominalinterest rates with inflation over the same period) have remained above152, severely constraining domestic production (according to many Bolivianbusinessmen). Real interest rates declined somewhat during 1988, asnominal interest rates failed to rise along with the increase in inflation.Still, the interest rate on boliviano loans remained about 12 percentagepoints higher than the increase in the consumer price index during 1988(see Table 2.6).

139. High real interest rates since August 1985 reflect a shift in themeans used to ration the available supply of capital, not a rise in thescarcity value of capital. Prior to August 1985, capital was allocated tofavored borrowers at negative real rates of interest through variousdirected credit schemes by the Government, and on the basis of personalrelationships by the commercial banks. The liberalization of financialmarkets under the NEP allowed capital to be priced at its opportunity cost,which despite some increase in supply remains very high.

-52-

Table 2.6: Interest Rates and Inflation(Percent)

1983 1984 1985 1986 1987 1988 /a

Interest Rate /bDollar Deposits /c -- 5.0 11.2 14.9 17.8 16.1Peso Deposits 45.0 140.0 110.0 33.4 32.7 26.6Dollar Loans /c -- -- 16.8 22.0 28.5 24.2Peso Loans 69.0 157.0 232.1 64.8 43.2 36.4

Consumer Price Index /d(percentage change) 328.5 2177.2 8170.5 66.0 10.7 21.5

Real Interest Rate /ePeso Deposits -66.2 -89.5 -97.5 -19.6 19.9 4.2Peso Loans -60.6 -88.7 -96.0 -0.7 29.4 12.3

/a Through November./b End of period./c Dollar lending and deposits were prohibited by the 1982 dedollarization

decree./d January to December (except 1988)./e Comparing the interest rate and the rate of change in the consumer price

index for the equivalent period.

Source: World Bank, IMF, and Instituto Nacional de Estadistica.

140. The reasons for high real interest rates on boliviano loans can beseparated into the determinants of the interest rate on deposits and thespread between deposits and loans. The interest rate on dollar deposits atBolivian banks averaged 16Z in October 1988, or 7 percentage points higherthan the interest rates on dollar deposits in the international market.Bolivian banks must pay a higher rate to attract dollars owing touncertainty over Government policy. Despite the Government's demonstratedcommitment to a liberal5zed economic environment and considerable successin reducing inflation rates, it appears that public confidence in themaintenance of the present policy regime remains low. Given pastexperience, depositors evidently can not ignore the potential for a returnto capital contzols or another dedollarization decree. It is difficult tosay how long it will take for suificient confidence in the system to permita significant fall in dollar rates.

141. Deposit rates also have been driven up by the poor portfolios ofsome commercial banks, particularly a high share of nonperforming assets.These banks have had to choose between refusing to capitalize interest due(and perhaps endangering the- solvency of the bank) and paying very highdeposit rates to attract the deposits necessary tu do so. To maintain

-53-

liquidity, a few of the weakest banks began to take deposits at interestrates higher than their lending rate, which led to bankruptcy and theclosing of a few banks. In turn, competition for deposits from the weakerbanks has driven up the whole structure of deposit rates. Strictersupervision of the banking system and greater disclosure requirements toincrease the amount of information available to denositors should increasethe soundness of the system and permit the safer banks to charge lowerinterest rates.

142. Commercial banks' spreads between deposits and loans declinedduring 1988, but remained in excess of 101 in October. These spreadsreflect reserve requiroments, operating costs, and the large share ofnonperforming assets. Present regulations require banks to hold 20% ofdemand deposits and 10% of time deposits as reserves. While paying someinterest on these reserves would contribute to a fall in interest rates ina competitive market, there is some question concerning the degree ofcompetition in the Bolivian banking system. In any event, lack ofresources severely constrains the Government's ability to do so at thistime. It is also difficult to lower reserve requirements while remainingwithin the stringent monetarv program necessary to control inflation. Highoperating costs reflect the short maturity of financial contracts owing tothe uncertain economic environment, which requires substantialadministrative costs to manage; and the high percentage of nonproductiveassets, particularly excessive investment in fixed assets with low yieldsand overstated book values, left over from the hyperinflationary period.In 1987, operating costs of the private sector commercial banks equalled12% of the average loan portfolio, a very high ratio compared to othercountries. While banks have taken steps to reduce their costs, principallythrough reduction of personnel, further measures are necessary to sell offunneeded physical infrastructure and generally increase administrativeefficiency. Still, the maturity structure and low volume of loans can notbe improved until there is greater confidence in the financial system.

143. One can conclude from this analysis that a fall in real interestrates will require increased confidence in maintenance of a liberalfinancial environment and a restructuring of the banking system to improvethe soundness of portfolios, increase competition, and reduce operatingcosts. Confidence should increase over time as people observe that theessential framework of economic policy remains stable despite changes inpolitical regimes. The restructuring of the banking system willnecessarily take a period of years, and care must be taken to maintainstability during this difficult process. Thus, while it is likely thatreal interest rates will continue to decline, it will probably take aperiod of years to reach levels consistent with the marginal productivityof capital.

-54-

CHAPTE III3 Th TAX REFORM15

I. Introduction and Su=ary

144. The Paz Estenssoro Goverrnment responded to the catastrophicallylow level of state revei;ues in 1985 by instituting a complete overhaul ofthe tax system. The tax reform simplified administration by reducing thenumber of taxes and relying on the bulk of revenues from consumption ratherthan income taxes. The Value Added Tax (VAT) is the main source ofrevenues. While taxes on income, profits, and property are also included,at present they generate only limited income for the state owing to low taxrates and difficulties in enforcement.

145. The Government's achievements in raising revenues and improvingadministration are impressive. The tax reform contributed 36Z of Treasuiryrevenues and 6.1Z of GDP in 1988, despite the fact that the taxes wereintroduced only in May 1987 and in the face of considerable difficulties incollecting taxes in Bolivia owing to limited administrative resources andthe large size of the informal sector. Decentralization of taxcollections through the commercial banks and computerization of the systemhave increased control of tax collections and reduced the opportunities forcorruption. A promising beginning has been made in targetted enforcementefforts for the larger taxpayers.

146. For the near future, tax policy should focus on consolidating thepresent system through stricter enforcement, auditing, and broadening thetax program to the regions, rather than fundamental changes to the taxrules. The planned implementatior of the audit program, and improvementsin customs administration would provide the greatest revenue return.Random checking of declarations under the simplified tax system might helpprevent cheating, although this would have a lower priority foradministrative resources. The presumed profits tax could be improved bychanging the base from net worth to total assets (to eliminate the presentbias against equity financing) and by raising the yield-equivalent taxrate. Adoption of the proposed net results tax for mining is necessary toencourage foreign investment.

147. Over the longer term, the tax system provides a number ofpotential avenues for efficiently raising revenues. Reduced deductions ofVAT payments under the complementary tax would increase revenues and helpshift the burden of taxation from consumption to income. While thepresumed property taxes contributed only limited revenues in 1988, improvedurban and rural land cadastres could raise their yield. Such improvements,however, should be given lesser priority than those mentioned above.

15/ This chapter is based on a background paper, The Bolivian Tax System,by Jorge Ospina (consultant). Other material used include a backgroundpaper by Laura Tuck entitled Resource Mobilization in Bolivia Under theNew Economic Program and information from Katherine Baer onadministrative aspects of the system.

-55-

II. Resource Mobilization before the NIP

148. The Bolivian public sector suffered a catastrophic fall inrevenues during the early 1980s, to approximately 52 of GDP by 1984. 16 Thecollapse of revenues was due to an erosion of the tax base and poorcollection of taxes due.

149. Declining international prices for Bolivia's exports, anovervalued official exchange rate, the collapse in domestic production, andhyperinflation severely reduced the tax base from 1980-85. Governmentrevenues heavily depended on exports by state encerprises, and the worldprices for Bolivia's export commodities fell sharply during this period,with the exception of natural gas prices. The dollar price of tin fell by272 between 1980 and 1984, while the prices of sugar and coffee fell by 332and 48X respectively. The overvaluation of the exchange rate furtherdiscouraged exports and drove production into illegal channels. The fallin imports over the period also reduced tariff revenues. Most importantly,the hyperinflation severely eroded the real value of tax receipts. As itwas difficult to successfully index tax receipts to the rate of inflation,late payments became essentially worthless after only minor delays. Theseproblems exacerbated longstanding difficulties in tax collection inBolivia, owing to exclusion of large portions of the economy from the taxrolls. For example, few taxes are collected from the informal sector.Estimates of informal sector activity range as high as two-thirds of thelabor force and 502 of GDP (see Chapter 2). Taxation of the agriculturalsector is minimal, despite the existence of a number of large-scale,commercial operations, especially in the Eastern Lowlands.

150. Weak administration and the complicated tax system impairedcompliance and enforcement of the tax rules, thus reducing the collectionrate on taxes due. Tax administration deteriorated during the early 1980s,as the fall in real public sector wages lowered morale and increased theneed for alternative sources of income (through taking extra jobs orcorruption). The tax base included a number of variables that weredifficult to calculate and/or to verify, for example personal income,corporate profits, or land values. A large share of personal income inBolivia is not derived from wage earnings that can be easily documented.Various accounting techniques reduced recorded profits, and the Governmentfrequently lacked the resources and expertise to measure profits. Non-existent or poor land cadastres plus the dearth of actual land transactionsprecluded the validation of declared land values. Further, falsedeclaration of tax liabilities and outright failure to pay taxes werewidespread.

16/ It is impossible to accurately measure revenues cr the public sectordeficit during the hyperinflation. The real value of income dependedgreatly on what day of which month it was received, while our data onlyshow annual totals. Further, weakness in accounting practices andexpenditure control make the nominal data particularly difficult tointerpret. Various sources show different estimates of the deficit in1983-85. The data provided are intended to give an idea of themagnitude of the problem; all we can say with certainty is that thedeficit was large and increasing.

-56-

151. The Paz Estenssoro Government immediately increased revenues bystopping the hyperinflation, raising public sector prices (particularly ongasoline) and increasing the official exchange rate. Government revenues(including sales by state enterprises) rose to 19Z of GDP in 1986.Combined with severe controls on expenditures and a runup of arrears,higher revenues achieved an approximately balanced budget (in cash terms)in that year. While this was a remarkable performance, the methods used toreduce expenditures quickly (for example, the freezing of state enterpriseaccounts and across the board expenditure cuts) and the increase in arrearswere not efficient instruments of fiscal control over the longer term.Further, while the fiscal accounts were in considerably better shape in1986 than they were prior to 1985, total Government revenues remainedheavily dependent on sal,s of hydrocarbons, with internal taxescontributing only 9Z of Treasury revenue.

III. The Bolivian Tax System

152. Introduction. In 1986, the Government reformed the tax system toincrease revenues while minimizing the impact of the tax structure onproductive activities, and to improve administration and enforcement of thetax code. The Government eliminated many cf the numerous taxes which hadcomplicated the previous system, and did away with the traditional incometax. A simplified tax system was introduced, with the largest source ofrevenue to come from indirect taxes, particularly the Value Added Tax.The tax rules were carefully designed to simplify administration byreducing the number of taxes, by defining a tax base which can be measuredwith the available data, and by providing incentives for self-enforcement.

153. The new tax system went into effect in May 1987, and has beenresponsible for an impressive rise in Government revenues. Internal taxesequaled 6.1Z of GDP in 1988 (see Table 3.1), far better than the levelsachieved over the past ten years. This performance is particularlyimpressive, taking into account that an entirely new system of taxadministration and control is still being implemented. Still, the revenuesgenerated by some of the new taxes remain well below their theoreticallevels. It is likely that the revenue potential of the tax system willincrease with further efforts to improve tax administration and control.

154. Increased internal revenues have reduced the Government'sdependence on taxes and transfers from YPFB (the state oil and gas company)which fell from 62Z of Treasury revenues in 1986 to 50Z in 1988 (the latterfigure excluds YPFB's payments on the VAT and transactions tax, or 5X ofTreasury revenues). Still, revenues remain highly vulnerable tofluctuations in international prices and market conditions in thehydrocarbons sector. Further efforts are necessary to increase anddiversify the source of revenues.

155. We will first discuss the major taxes in the new system, evaluatetheir potential for increasing revenues, analyze their impact on economicactivity and degree of progressivity, and suggest some modifications toincrease revenues and facilitate administration. After covering each ofthe taxes separately, we will turn to administrative issues of the systemas a whole, and then to projections of tax revenues.

-57-

156. Value Added Tax. The VAT is the primary source of revenues underthe new system, and made up 57? of Treasury revenues from the tax reform in1988. A 10? rate applies to all transactions, including foodstuffs andmedicines, but excluding real estate, export activities, interest payments,most capital market transactions, and the selling or transfer of goods dueto capital contributions or to a firm's internal restructuring processes.In the c&se of imports, the 102 rate is applied to CIF values, afterincluding all customs charges. The tax credit method of computation isused, under which the tax rate is applied to the total sales (output, ofthe firm and the tax paid on goods purchased (inputs) is then deducted.Firms can subtract purchases of capital goods in computing the VAT base.Thus, total value added (the base for the tax) equals total retail sales offinal consumer goods.

157. Exports are excluded from the VAT. Exporters receive a fiscalcredit for VAT payments made on inputs, which can not exceed 10X of thevalue of the export. The difference between fiscal credits and VATpayments owed by the exporter, if positive, is returned via negotiabledocuments (Notas de Credito Negociables), which can be used for payingother taxes.

158. VAT revenues in 1988 were 3.4% of GDP (see Table 3.1), whichsuggests that ample room remains for including business transactionspresently outside of the control of tax authorities. How to do so is a keyquestion for the tax authorities. Strict enforcement of compliance isessential to avoid a deterioration in tax discipline, particularly giventhe unfamiliarity of many Bolivians with paying taxes. However, whilestricter enforcement is certainly necessary to increase revenues, there isthe danger that it will drive more firms into the informal sector and thusreduce the tax base (see Chapter II for a discussion of the impact ofenforcement on the informal sector). Therefore, the appropriateenforcement measures must be carefully chosen. Accordingly, the Governmenthas focused VAT enforcement on medium and large retail outlets, which haveless potential for leaving the formal sector.

159. A uniform tax rate of 10% was adopted for revenue andadministrative reasons. Simple procedures and easy control devices areessential to quickly implement a new tax system, particularly in Boliviawhere administrative resources are severely limited. Differentials in taxrates would complicate tax management and compliance, as switching of goodsbetween different categories to pay a lower rate would be difficult tocontrol.

160. The heavy dependence on the VAT and other consumption taxes, alongwith the uniform VAT rate, has raised complaints that the system as a wholeis regressive. As the poor generally save less than the rich, taxingconsumption may mean disproportionately taxing the poor. While it is truethat the introduction of different tax rates for different goods couldintroduce a measure of progressivity, it would also greatly complicateadministration. Moreover, the VAT together with the 1OX complementary taxare neutral with respect to income, excise taxes have been imposed onluxury goods, and the lowest income groups (those with incomes of less than130 bolivianos per month) are exempt from some taxes and evade othersthrough working in the informal sector. Thus, the system as a whole isslightly progressive, perhaps more so than tax systems in some countries

Table 3.1 : Tr.aury Revenue

1967 1968 1989 1987 1988 1989 1987 1988 1989

--------Million Bolivian …------- --Percent of Treasury Revenues -- ---------Percnt of CDP---------

Tax Reform Revenue 348.0 636.2 868.0 28.6% 35.86 43.1% 4.60 6.1% 6.91

Value Added Tax 190.8 348.3 466.1 15.7% 19.8% 23.4% 2.2% 3.4% 3.7%Complementary Tax 24.3 48.4 66.3 2.0% 2.7% 3.3% e.3% 0.6% O.5%Simplified Tax 2.9 4.6 7.6 0.2% 0.3% 0.4% 9.OX 9.OX O.1XTransactions Tax 23.3 58.5 71.2 1.9% 3.3% 3.6X 0.3X 9.6X 9.6XSpecific Taxes 23.8 45.9 77.7 2.0% 2.6% 3.9% 0.3% 0.4% 6.6%Prosumed Tax on Profits 10.7 28.4 49.2 0.9% 1.6% 2.6% O.1X 0.3% 0.4%Presumed Tax on Property 31.7 51.6 61.4 2.6% 2.9% 3.1X 0.4% O.5X 0.6XAmnesty Tax 40.5 1.5 2.7 3.3% 9.1% O.1X O.SX 0.0% 6.OXTransport Tax 0.0 2.9 7.5 O.OX 0.2% 0.4X O.OX 6.61 O.1X

°° Other 0.0 40.1 48.4 O.OX 2.3% 2.4X 6.OX 0.4X 0.4X

Direct Taxe* YFPB 574.8 883.8 844.8 47.2% 60.2X 42.5X 6.6% 8.6X 6.7X

Other State Enterprises 42.1 19.2 0.0 3.6% l.lX 0.O% 0.6X 0.2% 0.0%

Custom Duties 194.7 209.0 228.1 I6.0X 11.9X 11.61 2.2% 2.O% 1.7X

Mining Royalities 6.4 7.5 18.7 0.4% 0.4% 0.9% O.1% O.1X 0.1%

Other 53.0 10.3 48.5 4.4X 6.6% 2.4% o.ex O.1X 0.4%

Total Treasury Revenue 1218.0 1760.0 1988.9 16W.O% 166.0% 19.0% 13.9% 17.6% 1S.9%

-59-

with high marginal taxes on income accompanied by numerous deductions.Modifications are possible which would increase the progressivity of thesystem as a whole if that is desired. We will suggest a few ideas indiscussions of individual taxes.

161. Complementary Tax to the VAT. The complementary tax applies a 102rate to the incomes of individuals (personas naturales) from the followingsources: wages, rents, interest payments, royalties, patents, trade marks,technical assistance, commissions, and operations derived from changes inequity participation. Social security receipts and persons with incomes ofless than 130 bolivianos are exempt. However, persons can deduct fromtheir payments 1OZ of all consumption expenditures, if registered VATinvoices are presented as a record of such expenditures. In other words,VAT payments are deductible from payments under the complementary tax. Ifconsumption exceeds income, and VAT payments are 1OZ of consumption, theindividual earns a tax credit. 17

162. The major purpose of the complementary tax is not to collectadditional revenues (although to the extent that consumption is less thanincome it may have that affect), but to provide incentives for self-enforcement of the VAT. To reduce their liability under the complementarytax, i.dividuals have an incentive to demand documentation of their VATpayments. To do so, they will ask for receipts from retailers for theirpurchases. This formal recording of transactions will encourage theretailer to pay the VAT, which will in turn lead the retailer to requestreceipts from suppliers. This system has been very useful in increasinguse of tax receipts, which has served as an effective means of monitoringcompliance by retail stores.

163. It has also been argued that the relationship between the VAT andcomplementary taxes will prompt informal sector retailers to pay VAT,because their clients will demand receipts. This is possible, but may notalways be the case. Consumers should be more or less indifferent between:1) paying the VAT tax on a given transaction and deducting it entirely fromthe complementary tax; and 2) avoiding the VAT at the time of purchase butpaying the full amount of the complementary tax. In fact, consumers may bebiased towards avoiding their VAT payments for several reasons. First, thecomplementary tax is assessed quarterly, so the second option given aboveallows for a delay of up to three months in paying the tax. Second, theVAT is assessed in conjunction with the one percent transactions tax(discussed below), which is not deductible from the complementary tax.Avoiding the VAT and paying the complementary tax thus permits the consumerto avoid the transactions tax. Finally, it is not clear that theGovernment can effectively cross-check between VAT receipts provided bytaxpayers and VAT payments from retailers. To keep track of the wholechain of VAT payments from the manufacturing level to the retail saleslevel, and to compare VAT payments by the retailer with VAT deductions byindividuals (even on the basis of samples) is a difficult task. Thus, theretailer may have the option of charging the customer for the VAT, issuing

17/ For a critical review of the VAT and complementary tax system, seePablo Ramos Sanchez, Temas de la Economia Boliviana III, EditorialPuerta de Sol, 1988, especially pp. 185-203.

-60-

a receipt, but in the end not paying the tax to the Government. Insummary, while the complementary tax may have encouraged greater VATcompliance, this effect should not be overemphasized.

164. Conversely, it appears that the revenue potential of the taxsystem is reduced by deductibility of VAT payments from the complementarytax. A black market in VAT receipts has arisen, in which individualsliable for payments under the complementary tax obtain documentation ofhaving made more VAT payments than they actually did. The receipts areobtained from taxpayers with incomes below 130 bolivianos a month, who areexempt from the complementary tax and therefore have no use for thereceipts. It is also likely that receipts are falsified. While the sizeof this market is unknown, it is an open question whether the benefits toenforcement from increasing the demand for VAT receipts are not offset bythis loss in revenue.

165. The above considerations raise the question of whether greaterrevenues could not be collected through the complementary tax with littlecost to other objectives. If it is true that the self-enforcing aspects ofthe VAT/complementary tax system are not great, and that substantialrevenue is lost through false VAT declarations, then the Government mightconsider reducing the level of VAT deductions from the complementary tax.Presently, the complementary tax is the only tool available for taxingincome and capital sources of earnings, but only 0.4Z of GDP was collectedin 1988 with this tax, well below international standards for incometaxation on individuals. The complementary tax is an attractive candidatefor collecting revenue because much of it is deducted directly fromsalaries, making it more difficult to evade payment. If it is desired tointroduce more progressivity in the system, different tax rates dependingon income levels could be introduced. For example, establishing threeincome categories would increase progressivity without greatly complicatingadministration or excessively raising marginal tax rates. In short, thecomplementary tax offers a significant potential for raising revenue andincreasing equity. Of course, decisions of this nature will depend on thelevel of revenues which are necessary, what alternatives are available forincreasing revenues, administrative resources and the values of thesociety.

166. It should be noted that the proposals discussed for increasingtaxation of income through the complementary tax are not the most efficientmeans of raising income taxes. In theory, it would be preferable tointroduce a tax on all sources of income, not connected to the VAT.However, raising taxes on income through the complementary tax is moreconsistent with the present structure of the tax system (and thus would beeasier to administer than introducing a new income tax) and retains some ofthe self-enforcement aspects of the VAT/complementary tax system. Giventhe present tax system, the complementary tax appears to be the mostconvenient instrument if higher taxes on income are necessary.

167. Transactions Tax. The transactions tax is a charge of one percenton sales of goods and services. Unlike the VAT, it is a cascading tax;payments under the transactions tax. on intermediate inputs cannot bededucted from the VAT or transactions tax on the final good. Exports (butnot inputs used to produce them), interest payments, domestic help, andpublic duties are all exempted. Supposedly the transactions tax was

-61-

established as a substitute for nine different taxes that existed in theprevious system.18 While it could easily be replaced by a small increasein the VAT rate, the Government did not want to change the tax rate of 10X.Its revenue effects are not insignificant. In 1988 the transactions taxaccounted for 3.3Z of Treasury revenues, or 0.6Z of GDP.

168. The transactions tax is a very attractive source of revenue in theBolivian context. It is easy to administer, as the base (gross earnings)is simple to calculate. Still, it is noL regarded very favorably in theeconomics literature, owing to its cascading effects and the fact that theflat rate does not distinguish among the costs to the community ofdifferent businesses.19 A more theoretically efficient tax would be aseries of license fees depending on the type of business. The problem withthis system is that it creates insurmountable administrative difficultiesand is more open to political pressure to favor different industries than aflat rate tax is. Despite its unpopularity, the transactions tax is auseful source of revenue, and should not be eliminated unless replaced withan alternative source.

169. Taxes on Specific Consumption. The tax system includes a seriesof excise taxes on specific consumption goods. Originally, a 302 rate wasimposed on alcoholic beverages, perfumes and cosmetics, and a 50% rate ontobacco products and imported jewelry. For imports, the base is the CIFcost before customs duties and the VAT. More recently (Decreto Sunremo21991 of 1988) the Government extended specific consumption taxes toelectrical goods, pottery and chinaware, and automobiles. A the same time,the rate on imported jewelry was reduced from 50% to 10Z. The recently

F approved budget also includes a 20Z tax on gaseous beverages, a rise in thetax rate on beer to 45X, and a 20Z tax on certain types of electricityconsumption. Specific consumption taxes provided 2.6Z of Treasury revenuesin 1988.

170. There is some question as ti the effectiveness of excise taxes inthe Bolivian context. Generally, one of the main objectives of excisetaxation is to place the burden of the tax on consumers through taxinggoods for which the demand is relatively inelastic. This is difficult todo in Bolivia owing to the close competition between informal and formalsector suppliers. Many of the items subject to the consumption tax can beimported illegally. Thus, intensive use of excise taxes may significantlyreduce the tax base via a decrease in official sales of taxable articles.In addition, the introduction of a series of excise taxes with differentrates has complicated the tax system. On the other hand, excise taxes are

18/ According to Juan Cariaga, the Finance Minister at the promulgation ofthe tax reform. See Foro Economico, "El Proyecto de ReformaTributaria", April 1986, p. 15.

19/ See discussion of a similar tax in Colombia in IntergovernmentalFinance in Colombia, Final Report of the Mission on IntergovernmentalFinance, Harvard Law School International Tax Program, 1984 pp.275-280. Taverns, for example, are more costly to a community (interms of demands on police and higher incidences of drunk driving) thanbarbershops and should therefore bear higher taxes.

-62-

a convenient means of quickly raising revenues. The decision of theGovernment to impose a tax on beverages is understandable from thisperspective. Excise taxes also introduce some progressivity to the systemby taxing goods largely consumed by high-income taxpayers. In the longerterm, it may be useful to explore other means of increasing revenues (seethe discussion under the complementary tax) or improving the progressivityof the system, perhaps through a luxury tax with a well-defined base.However, such changes are of lesser priority than efforts to improveadministration and increase the yield from existing taxes.

171. Presumed Profits Tax. The presumed profits tax is a 2.5Z levy onnet worth, or the difference between assets and liabilities. This tax isused as a proxy for a profits tax. All enterprises, public and private,must pay this tax, with the exception of non-profit organizations,religious groups, savings cooperatives, mutual societies, and companiesengaged in mining, hydrocarbons and electricity, which are subject tospecial taxes. Taxes paid on real estate, automobiles, motor boats, andairplanes can be deducted to avoid double taxation (they are subject to thetax on presumed income of property owners described below). The dollarvalue of fixed assets is maintained through indexation to the dollar.

172. Collections under the tax are relatively low. In 1988, thepresumed profits tax contributed only 0.3Z of GDP to the Treasury. Each ofthe 25,456 companies registered in the general tax system paid an averageof abount US$600. One reason for the low level of revenues is thatpayments under the tax on personal income of property owners (see below)are deducted from the tax liability, to avoid double taxation. Even so,actual levels of corporate taxation are below potential levels.

173. A number of criticisms have been made of the presumed profits tax.It does not discriminate between those companies which earn income andthose that do not. More broadly, the tax is assessed on the original bookvalue of assets adjusted for depreciation, not on their economic value.The tax discriminates in favor of those activities which operate with lowfixed capital. As a result, the services sector is encouraged at theexpense of industry. Thus the presumed profits tax does not conform to anycriteria of horizontal equity. The tax also encourages debt as compared toequity financing, as net assets appear lower when a corporation's growth isbased on credit financing rather than profit reinvestment or contributionsby shareholders. Finally, foreign companies can not deduct the presumedprofits tax from their home tax liability.

174. The motivation for introducing the presumed profits tQx appears tohave been the difficulties involved in collecting taxes on corporateprofits under the former tax system. It is not evident that correctlyvaluing assets and liabilities is easier than measuring profits; in fact,the administration of the tax is very difficult without informationconcerning the corporations, income during the year.

175. Technical assistance provided by the IMF has suggested twoimprovements to the presumed profits tax: a) changing the tax base from networth to total assets so as to eliminate the present bias against equityfinancing; and b) :-aising the yield-equivalent tax rate to generate morerevenue. These changes should provide a better incentive structure and

-63-

raise corporate taxation to more reasonable levels. However, if over timethis new system continues to produce low levels of revenue, then theGovernment might consider reintroducing a tax on income, perhaps retainingthe tax on presumed profits as a minimum tax.

176. Tax on Presumed Income of Property Owners. Like the presumedprofits tax, this tax is designed to capture income through measuringassets. The assets chosen for this purpose can be divided into threecategories: urban real estate; vehicles (automobiles, motorboats andairplanes); and rural real estate (scheduled to be implemented in 1989).The tax rate on urban real estate is 0.35% plus an additional amount thatvaries (from 0.15% to 2.652) according to the value of the property. Inthe absence of reliable property assessments, the tax is based on sworndeclarations of property values. Taxes on vehicles are based on the ex-customs value, depreciated at 201 a year until a residual value of 16.8Xremains. The tax rates range from 1.5% to 5%, depending on the value ofthe vehicle. The rural land tax depends on land use (cultivable land,cattle raising, or pasture), on location (in accordance with the agrarianreform classification), and on the size of the estate. The annual ratefluctuates between US$.01 and UF$.94 per hectare. These rates areextremely low; the Government expects to collect only US$17 million in 1989f,rom the rural land tax. Land Jeemed uncultivable under the AgrarianReform Act of 1953 is exempt.

177. It is safe to say that the full potential of the presumed incometaxes has not yet been realized. Difficulties ir administration in thelocal governments responsible tor collections and the lack of land surveyshave greatly 'imited revenues, which amounted to only 0.5% of GDP in 1988.A somewhat larger amount probably went to the municipalities. TheGovernment is already in the process of implementing a joint auditingeffort with the municipalities, mainly directed at the exchange ofinformation and provision of technical assistance.

178. Although the very low rates of the rural land tax mean that itshould have little impact on resource allocation, it is useful to discussthe implications of the structure of the tax, in case rates are raised inthe future. Taxation of land according to its productivity may stimulateincreased efficiency in land use. This is a particularly relevant toBolivia, wherp many of the wealthy hold large, unproductive estates. Onthe other hand, the system provides a disincentive for land improvements(compared to taxation of profits) because no reduction in tax is allowedfor the extra costs associated with investment and the tax rate increasesonce the land is improved. This is an unavoidable shortcoming as long asaccurate cadastral surveys are not available. Performing such surveyswould facilitate an important improvement in the rural taxation system, butthe tremendous costs _.nvolved make this a project for the longer term.Finally, linking the tax rate to the size of the holding could encouragefragmentation of parcels, although presently this is a useful shortcut forfocusing taxation on the largest estates.

1,9. Despite the very low rates, the introduction of the rural land taxhas been met with considerable political opposition, probably because ruralareas have traditionally escaped taxation from the Central Government andlandowners see even a relatively light tax as a dangerous precedent.Indeed, the major importance of the rural land tax is as a first step in

-64-

taxing rural activities, particularly commercial agriculture in thelowlands. This taxation is desirable both to raise revenues and to improvethe equity of the tax system.

180. Property taxation in general has the potential for becoming asignificant source of revenues and improving the progressivity of the taxsystem. Improved cadastral surveys and more effective administration ofthe presumed tax on property could make a substantial contribution toBolivian development. However, these changes will take time and resourcesto accomplish.

181. Simplified Tax System. This is an alternative tax system designedto ease compliance with the tax laws for small businesses which lack therecord keeping necessary to calculate their taxes under the regular taxrules. A business can enroll in the simplified system if capital does notexceed 9600 bolivianos, total sales do not exceed 48,000 bolivianos, and noitem is sold for more than 100 bolivianos. The businesses in thesimplified tax system pay no other taxes. Each business declares itsannual earnings when enrolling in the system, is placed in one of threecategories according to gross earnings, and pays a fixed amount dependingon the category.

182. The amount paid under each category is calculated using acomplicated formula to approximate what tax liabilities would be under thenormal system for businesses in each category. The formula assumes thatcapital is equal to 20Z of earnings, that profits equal 20Z of earnings,and that 95Z of income is consumed.

183. The Government collects only minimal amounts of revenues throughthe simplified tax system, equivalent to only 0.3Z of Treasury revenues in1988. Tax payments by category ranged from 400 bolivianos per business inthe highest to 13 bolivianos in the lowest. The principal reason for thesystem is to simplify the tax process for the small taxpayer and toincorporate the largest possible number of taxpayers in the NationalRegistry of Taxpayers. The latter benefit should not be discounted, asmaintaining a list of even potential taxpayers is an important tool infacilitating future extensions of the tax system.

184. The simplified tax system may be subject to abuse through falsedeclarations of earnings. There is talk about medium size businesses whichdo not meet the requirements enrolling in the system. However, many taxesare vulnerable to evasion through under-declaration of income. The dangerin the simplified tax system is that failure to check earnings declarationswill encourage more medium sized businesses to participate, thus reducingtheir tax liability and the likelihood of their being audited, compared toregistering under the regular system. It might be useful to do some samplechecking of businesses in the system and impose severe penalties forcheating. This would probably not directly generate much greater revenues,but may serve as a deterrent to ussing the system for evading taxes.

-65-

185. Import Tariffs.20 Taxes on imports make up a significant shareof Government revenues, but remain well below their pctential. Customsduties alone equalled 11.4Z of Treasury revenues in 1988, and the VAT onimports contributed another 4Z. Tariff receipts were 132 of the CIF vaiueof legal imports in 1988. While the theoretical tax rate on imports isdifficult to determine (owing to changes in rates during the year and thevarious exemptions still in force under the old investment law), theweighted average of the 102 rate on capital goods and approximately 17?rate on consumer goods is close to 13!, indicating that the customs servicecollects a reasonable level of revenues compared to the official value ofimports. However, imports are reportedly subject to considerableunderevaluation, and the value of imports which escape all legal controlsis reliably believed to be very large. Thus, considerable revenues couldbe raised by achieving more effective controls on imports.

186. Such control is particularly difficult to achieve in Bolivia owingto enormous administrative problems. Long and sparsely inhabited bordersand limited resources make it difficult to effectively control illegalimports. Corruption and inefficiency are reportedly very high in thecustoms service, facilitating underdeclaration of imports and othermeans of tax evasion. The Governmen.'s preference for a uniform tariffrate in part reflects the difficulties involved in administering acomplicated tariff structure; the opportunities for misclassifying goods tolower the rate are just too great.

187. The problem of controlling contraband imports is particularlydifficult to resolve. Anecdotal evidence suggests that importing goodswith.out notifying customs is extremely common. For example, the train fromBrazil to Santa Cruz has been seen to slow down just short of the city (andcustoms control), and most of its freight unloaded for shipment to thecontraband market. Newspaper advertisements have been seen in Santa Cruzoffering to sell import ceritificates or automobiles without papers.Private sector representatives in the formal sector claim that it isimpossible to compete with contraband importers who evade the tariffs, VAT,transactions tax, and fees.

188. The Government adopted an interesting approach to reducingcontrqband in encouraging privately-financed comites to assist in customsenforcement. This experiment has had mixed results. The comites arefinanced by producers associations, which results in their checking mostlyfor goods which compete with the associations which pay their salary.Thus, at some control points one might see a number of comitesrepresentatives, each looking for particular forms of contraband goods.This illustrates the inefficiencies which can be introduced by expectingthe private sector to furnish public goods. It also appears that lack ofcooperation between the customs service and the comites has hampered thework of both agencies, although private sector representatives in SantaCruz felt the comites had been effective in reducing contraband. Inprinciple, reliance on the private sector to enforce the law may raiseserious questions concerning due process and equity, although we have noinformation on comite activities in this regard.

20/ This section considers only the revenue and administrative aspects ofcustoms policy. The impact on incentives is discussed in Chapter 2.

-66-

18). The Government has made some progress in improving customsddministratior. by hiring private verification agencies to check on theprices and quantitites in import declarations. Although there has beensome criticism of the verification agencies, especially since it isdifficult for the customs service to supervise their activities, the newimport monitoring system seems to be working better than the old one. Oneissue which has arisen is that imports of less than US$1000 are not subjectto verification of their prices, on the grounds that it is not worth theadministrative expense. However, there appears to have been a substantialrise in the number of declarations below US$1000, raising suspicions thatthe policy has increased incentives for undervaluation. It might make moresense to verify the prices of imports with relatively small values on arandom basis, with stiff penalties if obvious cheating is discovered.Depending on the costs of verification and the frequency of cheating, thisapproach could be extended to imports of much greater dollar value.

190. While fundamental improvement of the customs service will take along time, there are some steps that the Government can take quickly whichwould raise revenues. Presently, import tariffs, the VAT, and specificconsumption taxes are all paid separately, and there is some evidence thatall three taxes are not being collected on legal imports. The Governmentintends to introduce a single form for payment of these taxes in 1989,which should aid enfoLcenient. Another serious problem is that imports tendto drop out of the system between entry at the border and registration atthe customs administration centers in the major cities, where the taxes areactually paid. The coordination between border points and customs centerscould certainly be improved. The Governmnent's plans to cross-check importdocuments and customs declarations should help reduce this problem. Itmight also be useful to require prepayment of customs duties and taxes as acondition of clearing goods at the border.

191. Hydrocarbons Revenues. Despite the considerable success inraising revenues through the tax reform, the Government remains greatlydependent on revenues from YPFB. In 1988, taxes and transfers from thehydrocarbons sector equaled 50Z of Treasury's current revenue. TheBolivian authorities have been able to extract a large proportion of YPFB'soperating surpluses through a combination of direct taxation and a transfersystem based on negotiations between YPFB and the Government. At the sametime, a realistic exchange rate policy and periodic increases in the priceof gasoline have enhanced the profitability of YPFB's commercialoperations.

192. The system for calculating YPFB's transfers to the rest of thepublic sector generally has worked well despite its complexity, althoughoccasional problems have emerged. According to SAFCO, weakness in YPFB'saccounting systems have complicated the transfer of revenues.2 1 In 1988YPFB's payments to the regional develcpment corporations were greatlydelayed; the same complaint is sometimes heard in relation to its payments

21/ Sistema Integrado de Administracion y Control Gubernamental, "Programa.de Emergencia", Informe No. 3, pp. 5-6.

-67-

to the Treasury. On the other hand, YPFB's accounts were frozen in late1988 (along with the accounts of other state enterprises), which no doubtcomplicated their compliance with the transfer system.

393. Despite the overall success of hydrocarbons taxation (especiallyin comparison with the experience in many other countries) one couldenvision some improvement. For example, it would be possible to unify alltaxes in an ad-valorem tax on internal and external sales. The tax rateshould be high enough so as to reduce to a minimum the proportion of YPFB'soperating surplus which is negotiated as an additional transfer to theCentral Government, although some negotiations would still be necessary totake into account changes in cash flow requirements and YPFB's investmentprogram. The goal would be to increase the transparency of the tax systemand to provide YPFB's managers more ircentive to increase efficiency. Atleast, some reform along these lines is what general considerations wouldsuggest. However, more careful study would be necessary. Given that thepresent system has been successful in capturing revenues, there is probablyno dire urgency in changing it.

194. Mining Taxes and Royalties. After the collapse of world tinprices in 1985, mining taxes almost disappeared as a source of Governmentrevenue. In 1988, taxes on the mining sector represented only 0.4Z ofTreasury revenues. Since 1985, the only tax on mining activity is aroyalty on presumed net income, calculated as the difference between worldprices and presumed costs. For tin, wolfram, antimony, silver and bismuth,the royalty is 532 of net income, while for lead and zinc the rate is 20Z.

195. Two kinds of presumed costs are set by the Government. Presumedsale unit costs (costos de realizacion) are established as a proportion ofworld prices, depending on the quality of the mineral ore extracted.Presumed operating costs are set in US dollars for two categories ofproducers defined according to monthly production levels. Supposedly, costestimates are periodically updated by the Comite Permanente deActualizacion de Regalias Mineras, but in practice this committee hasseldom met and changes in unit costs have been rare. It is likely thatsome revision of presumed costs would be useful. No royalties have beencollected on tin over the past few years because presumed costs have beenhigher than world prices.

196. This system has been criticized on a number of grounds. As taxesare not based on actual costs, firms can not deduct the expenses ofexploration and development from their tax liability. This reduces theincentive to invest relative to a tax on profits. Since the unit costestimates made by the Government are based on the average costs for allgrades, extraction of low grade (and low profit) ore may becomeunprofitable after paying taxes, despite the fact that extraction of thisore would be profitable under an income or profits tax. The present systemdoes not meet the tax credit requirements in the countries of origin ofinternational mining companies, thus discouraging much-needed foreigninvestment in the sector. Moreover, taxrtion on the basis of presumedcosts is arbitrary, difficult to adjust to take account of technologicalchange, and subject to unpredictable revision by the Government in power.This is a serious drawback in the mining sector, which requires stabilityin tax rules over a considerable period of time. The principal benefit ofthe system is that companies do not need to keep much in the way of records

-68-

to calculate their tax liability. This is an important consideration inBolivia, where many mining enterprises are simply one or two personoperations.

197. The Government has been working on an alternative tax system forthe mining sector. This system would be optional, and is designed toappeal to multinational firms. A tax of 102 would be levied on 'netresult.' (sales less administration and production costs, lessreinvestment) until the initial capital investment is recovered, and 302thereafter.22 The idea is to provide an incentive for reinvestment ofearnings (reinvestment lengthens the recovery period, thus reducing thecurrent tax rate) and to permit the deduction of taxes paid the BolivianGovernment against home countty taxes.

198. The primary drawback would be the system's complexity. Itsadministration would require a very reliable auditing system and theestablishment of all sorts of limits on allowed expenditures.Multinational firms have considerable expertise in hiding profits fromlocal governments through various accounting techniques. It is thereforeuncertain how much revenue would actually be collected under the newsystem. Given that the proposed investment code for the mining sector-wcull guarantee ch'e present tax regime for 30 years, it is extremelylu,portant that Bolivia not lock itself into a system under which it isdifficult to successfully collect taxes. For this reason, the Governmentis considering instituting a minimum tax in conjunction with the netresults tax, to ensure that companies do not succeed in claiming anunreasonable level of deductions.

199. Coparticipation. The coparticipation system guarantees thatvarious regional entities automatically receive a share of tax revenues.Internal revenue taxes 23 are distributed in the following manner: 75Z toTreasury, 102 to municipalities, 102 to the regional developmentcorporations, and 5Z to universities. In additioti, a portion of thepresumed tax on income from property is retained directly by themunicipalities. The regional distribution of the coparticipation funds ismade in accordance with the locdtion where the tax collection is made.

200. This system has introduced some rigidities into the allocation ofstate revenues. For example, the 52 allocation to the universities is notaffected by the costs of education or the effectiveness of universityprograms. Also, the practice of distributing revenues ir the region inwhich they are collected means that regions with hydroca-bons resources or

22/ The investment recovery period is over when the capital account of theproject reaches zero. The capital account is defined as in-estment inthe current year plus engineering and project adm'ninistrationexpenditures plus interest charges attributed to investment of theprevious year minus net results in the current year after tax payments(or plus net losses).

23/ VAT, the complementary tax, transactions tax, specific consumptiontaxes, the simplified tax, the presumed profits tax and that part ofthe presumed income tax on property that is collected by the Treasury.

-69-

relatively higher-income populations (for example Santa Cruz) enjoy arelatively greater level of income. The poorer regions, on the other hand,receive little from the coparticipation program.

201. One's view of this system depends on the importance of regionalversus national ties. Regional loyalties are very strong in Bolivia, sothat perpetuation of inequities among regions is accepted by large parts ofthe population. While the Government has made an effort (through theRegional Development Fund) to assist the poorer regions, so far theresources available to the Fund are not great. The question of whether andhow to address regional disparities in income and public services willbecome even more pressing as the decentralization process continues.Considerable resources will have to be devoted to fashioning anadministrative structure which reflects the Bolivian view of theappropriate balance between regional and national interests.

IV. Tax Administration

202. Introduction. The Government has established a variety of newprocedures to implement the tax reform. A new administrative structure,use of the commercial banking network to collect taxes, various publicitycampaigns, and targatted enforcement efforts have played a key role inachieving the increase in internal revenues since 1986. While the newsystem is a dramatic improvement over tax admir.istration in the early1980s, much remains to be done. Further administrative improvements shouldinclude extension of the office for monitoring high-income taxpayers toCochabamba and Santa Cruz and implementation of the new audit system.

203. Tax Administration Prior to the Reform. Monitoring andenforcement of the tax rules were highly inadequate under the old taxsystem. Uniform standards for tax documentation did not exist, so that themany agencies involved in collections could not adequately coordinate theiractivities. There was essentially no mechanism for checking actualpayments against declared obligations. The processing of accounting datafor tax collections suffered from long delays. While the commercial bankswere partially used for tax collection, the administrative arrangementswere such that the system could not function properly. Governmentregulations prevented expanding the number of commercial banks involved intax collection, even though this resulted in extremely poor coverage ofsome geographic areas. Contracts between Internal Revenue and commercialbanks did not include any sanctions against retaining funds beyond thetargetted amount agreed with Internal Revenue. Monitoring of banksubmission of revenues to the Treasury was poor, resulting in considerabledelays in the receipt of funds. Documentation submitted by commercialbanks was not uniform, so that Internal Revenue could not verify or cross-check tax information. Direct participation by Government employees inaudits and tax collection encouraged widespread corruption and reducedGovernment revenues. Administrative deficiencies were compounded by thecomplicated income tax system, which made taxpayer control extremelydifficult.

204. The new tax system was designed to address many of these problemsby computerizing and dccentralizing tax collections. The system is basedon greater use of the commercial banks, reliance on computers to accountfor tax receipts and monitor taxpayers, improved auditing procedures,

- 70-

targetted enforcement efforts for large taxpayers and retail stores, arevamped taxpayer registration system, and a reorganization of the agenciesreponsible for managing the system. We will cover each of these areasseparately, although to fully understand the system it is important to keepin mind their interrelations.

205. Organizational Aspects. As a first step towards implementing thetax reform and employing a more effective group of administrators, theGovernment established a separate ministry to administer the tax reformprogram. The Tax Ministry had responsibility for monitoring andimplementing the new taxes, and for overseeing the work of the variousagencies responsible for taxes. This structure was successful inchannelling resources for tax administrat>. and in emphasizing theoverriding importance of revenue generation ror the success of theGovernment's economic program. The Tax Ministry was reintegrated into theFinance Ministry in 1988, as a separate subsecretariat reponsible foradministering the tax program.

206. The Government has moved to improve the basic structure andfunctions of Internal Revenue so that it can eventually take over theresponsibility for implementing the tax reform. In June, 1987, theGovernment issued Supreme Decree 21628, which provides for thereorganization of Internal Revenue to improve its ability to administer thetax regime. New departments were established to reflect the most importantaspects of the system, including collection, auditing, data processing, andtechnical and legal affairs. Regional offices have also been reorganizedalong the same lines. Despite the organizational improvements, much workremains to be done in training Internal Revenue personnel in new taxadministration techniques and generally improving the quality of InternalRevenue staff.

207. Commercial Banking Network. The system for tax collection throughthe banking system was designed to simplify administration and make iteasier for individuals to pay taxes. Tax collection is administered byeleven commercial banks and their branches, which together represent 210offices thrcughout the country. Taxpayers submit their tax statements andpayments directly to commercial bank branches. The branches transfer tothe regional development corporations, municipalities, and universities theamounts to which they are entitled under the tax coparticipation scheme(see previous section for a description), and prepare a daily summary ofcollections. These summaries, along with tax statements and receipts, aresent to the bank's main office in La Paz, which transcribes the informationonto computerized tapes. The main office then transfers the funds due tothe Treasury, and transmits the information collected from the branches tothe centralized automatic data processing center in Internal Revenue.

208. This system has a number of important advantages. The computersenable Internal Revenue to automatically compute collections for the entirebanking system on a daily basis, and to cross-check banks' computerizedstatements against actual TLeasury deposits. Internal Revenue then submitsto each bank's headquarters a statement of all transferral orders issued bythe bank w;hich did not result in an actual deposit to the Treasury.Control is also facilitated by the centralization of all tax documents,information, and payments in La Paz through each bank's central office.

-71-

The removal of Government employees from direct collection of taxes andimproved accounting for tax payments has reduced the opportunities forcorruption in the system, thus increasing equity and revenues.

209. On the whole, the collection of taxes by the commercial banks hasfunctioned smoothly with only minimal costs. Recent analysis of the systemhas shown that information transferred by the banks to Internal Revenue islate only about 102 of the time, a creditable performance given thedifficulties which plagued the former system. Commercial banks haveretained about 0.8Z of tax revenues to cover their administrative costs.It is difficult to believe that the public sector could perform the sameservice for less.

210. Taxpayer Registration. A key element in increasing taxcollections is obtaining information on potential taxpayers. TheGoverrnment carried out a national registration campaign in early 1987 forindividual and corporate taxpayers, with registration effected through thecommercial banking system. Two hundred thousand taxpayers were registeredas a result of the campaign, double the number under the previous taxregime. Registration records are organized according to the type ofeconomic activity, geographical location, and income category, providingInternal Revenue with a much broader information base on taxpayers than hadexisted previously. Information provided in the taxpayer register also hasbeen useful for auditing.

211. Publicity Campaigns. The Government has carried out severalpublicity campaigns to familiarize taxpayers with the new taxes and taxadministration system, and more generally to encourage compliance with thetax laws. Since early 1987, the publicity department of the taxdirectorate has used radio, television, and newspapers to explain how topay taxes, to appeal to taxpayers' sense of patriotism and civicresponsibility, and to warn against attempts to evade taxes. Initialpublicity efforts are considered to have been effective. The increase intax collections was higher after each campaign than at any other timeduring implementation of the reform.

212. Enforcement. To obtain the largest revenues from scarceadministrative resources, the Government has focused enforcement efforts onthe 'largest' taxpayers (those with significant taxable resources). Anoffice was established in La Paz exclusively devoted to monitoring largetaxpayers, with its own data base on large taxpayers in La Paz, along withsoftware for quickly checking taxpayer accounts. This office also reviewstax declarations of large taxpayers for accuracy, and notifies taxpayerswho fail to file or file faulty declarations. A special audit program hasbeen set up for large taxpayers, which is being used as the pilot programfor a nationwide auditing system. This program is designed to perform asimple, quick auidit of the taxpayers' major financial accounts, and comparethis information with data kept by the office on large taxpayers.

213. Another successful technique used for tax enforcement has been themonitoring of retail stores in La Paz cu-- VAT payments. Investigatorsdetermine which stores are rot issuing reciepts for purchases with theamount of VAT paid included. These stores ar- closed, and a large bannerattached stating the reasons for closure and warning against removal of thebanner until the business has complied with its legal tax obligations.

-72-

Since April, 1987, Internal Revenue has carried out three such campaigns inLa Paz, although there have been no closures in other cities. VATcollections from small businesses have increased significantly after eachcampaign.

214. While the Government has made an excellent beginning inestablishing effective procedures for enforcing the new tax laws,improvements are necessary in a number of areas. Progress made in taxcollection through the La Paz office of large taxpayers needs to befollowed up by establishing similar offices in Santa Cruz and Cochabamba,cities which together have as many large taxpayers as La Paz. Little hasbeen accomplished in tax auditing beyond the program for large taxpayerscited above. The Government has postponed implementation of a morecomprehensive auditing program until a more automated system is in place,owing to corruption among the auditors. A more aggressive andcomprehensive enforcement program is essential to collect taxes that wouldotherwise not be paid, to provide a credible threat of penalties totaxpayers who are tempted to cheat, and to convince the public that the taxsystem is equitably administered.

V. Projections

215. Introduction. Despite the considerable increase in Governmentrevenues achieved over the past few years, further improvements will berequired to ensure that the public sector has sufficient funds to meetcurrent expenditures and finance an increasing portion of investment fromdomestic resources rather than foreign funds. In the short term, moreaggressive enforcement of the tax laws and continued expenditure constraintshould achieve a rise in savings and permit non-inflationary finance ofnecessary investment expenditures. However, the future is clouded by theexpected decline in revenues from natural gas exports to Argentina in 1992(see Chapter 5), which now contribute 24Z of Treasury revenues. Continuedeffo:ts to diversify the sources of Government revenues is essential toachieve acceptable revenue levels in the mid-1990s.

216. Revenues in the Short Term. In 1988, Treasury revenues areestimated to have increased to 17Z of GDP, three percentage points higherthan in 1987. This rise was due to increased revenues from the tax reform,which outweighed the slight decline in import duties (relative to GDP)owing to the fall in registered imports. The increase in the value addedtax (the main component of the tax reform) by 1.2Z of GDP is impressive, asVAT revenues also were affected by the decline in collections on imports.Broadly speaking, the 1988 data show a strong improvement in domestic taxcollections, partially offset by reduced taxes on imports.

217. Savings of the nonfinancial public sector rose to 0.7Z of GDP in1988 compared to -1.7Z in 1987. The available data indicate that thisimprovement came from better savings performance in the State Enterprises,although our information on their accounts are not sufficiently reliable todraw firm conclusionrs Tnrreased savings permitted a decline in theoverall deficit of the nonfinancial public sector, despite a rise ininvestment in relation to GDP.

-73-

218. We anticipate some increase in Central Government revenues in1989. The decline in taxes and transfers from YPFB to 6.7Z of GDP(compared to 8.52 in 1988) should be offset by rising collections under theVAT (0.3Z of GDP) and increased collections from most other internal taxes.Net savings of the nonfinancial public sector should rise, and the overalldeficit should fall to 5.72 of GDP. Increased foreign financing of thepublic sector deficit in 1989 (including an assumed receipt of delayed gaspayments from Argentina) would imply that net domestic credit from theCentral Bank to the nonfinancial public sector is negative in 1989. For1988-89, Bolivia will have achieved a substantial rise in public sectorsavings and received additional external savings, permitting an increase ininvestment to be financed without relying on domestic credit creation bythe Central Bank.

219. Medium Term Scenarios. A crucial issue for Bolivia's future iswhether the public sector will continue to raise sufficient revenue to meetcurrent expenditures and finance the investment necessary to achieveacceptable rates of growth. We have constructed two possible scenarios forfuture revenue efforts, to illustrate the challenge facing the Bolivian taxauthorities in meeting this goal. The first (called the "no improvements"scenario) shows the implications for savings and the overall balance of thenonfinancial public sector if the rates of collections (actual collectionsdivided by the tax base) on existing taxes remain the same as the levelsanticipated for 1990. Thus, revenues from each tax would increase at thesame rate as the base for that tax. This scenario implies no change inlegal tax rates after 1990, no introduction of additional taxes, and noincrease in collection rates as a result of administrative improvements.This is by no means a likely scenario; recent progress in raisingcollection rates and the Government's plans for improving administrationand increasing resources devoted to enforcement make it probable thatcollection rates will rise substantially. Rather, our purpose is to devisea baseline to quantify the increase in revenues which will be necessaryfrom these improvements.

220. Table 3.2 shows the implications of this scenario for the fiscalaccounts of the nonfinancial public sector, expressed as a share of GDP.Revenues can be expected to increase significantly over the next threeyears, owing to better prospects for the price of Bolivia's natural gasexports. This result assumes that the present pricing formula remains inplace through the expiration of the contract with Argentina in 1992, andthat Argentina pays all arrears and maintains payments under the contract.This is by no means certain. but it seems to be a reasonable workingassumption. It will also be essential for the Government to continue toincrease the price of domestic gasoline in line with inflation (or keep theprice the same in dollar terms) to avoid an erosion in YPFB's revenues fromdomestic sales. Government expenditures remain at about the same ratio toGDP. The decline in the net savings of other public sector entitiesreflects the increasing share of YPFB (which pays a relatively greaterpercentage of revenues to the Central Government) in total revenues, ratherthan any deterioration in performance. Altogether, savings of thenonfinancial public sector rise to 2.8Z of GDP by 1991, permitting a sharpincrease in investment expenditures. Investment rises in the early 1990salso because of planned expenditures on the Brazil power plant project,financed by foreign loans (see Chapter 5).

-74-

221. The halving of Bolivia's exports of natural gas to Argentina in1992 anticipated in our projections (see Chapter 5) would have seriousrepercussions for Government finances. The dee.line in export revenues fromgas results in a fall of US$137 million in YPFB's transfers to theTreasury, or 1.32 of GDP (see the figure under 1992 for direct taxes onYPFB in Table 3.2). The accompanying fall in imports would also reducecustoms receipts. Altogether, savings of the nonfinancial public sectorwould decline to -0.42 of GDP in 1992. compared to a projected 2.8Z in1991. The overall deficit would rise to 9.92 of GDP, unless the Governmentundertook an even sharper cutback in investment than assumed in theprojections.

222. The projections after 1992 imply a continuing deterioration in theGovernment's ability to finance investment under the no improvementsscenario (see first column under 1993-97 in Table 3.2). Revenues wouldaverage the same percentage of GDP as in 1992, and expenditures woulddecline as interest payments fall relative to output, resulting in highersavings of the nonfinancial public sector. However, external financecaptured by the public sector is expected to decline after 1992. The levelof external finance in the early 1990s is artificially high owing to loansprovided under the Brazil power plant project, and the 1992 level isfurther increased by external assistance to help Bolivia adjust to thesharp fall in export revenues from gas sales to Argentina. It cannot beexpected that this emergency assistance would continue for the rest of theforecast period. The decline in external finance implies a rise ininternal finance (credit from the Central Bank, averaging 1.9% of GDP from1993-97) if the projected investment level is sustained. This is clearlynot consistent with acceptable rates of inflation. Either rising inflationwould undermine the macroeconomic program, or the Government would beforced to cut investment. In either case, GDP growth would fall.

223. To finance public sector investment without engendering increasedinflation, it will be necessary to make greater revenue efforts thanassumed here. Table 3.2 shows our projections (consistent with the basecase presented in Chapter 5) for increased tax collections thaL would besufficient to finance projected investment expenditures. The specificimprovements assumed include: a) a rise in VAT collections on domestictransactions (excluding YPFB) from 2Z of GDP in 1989 to 3.52 in 1993-97(note thac the figures given for VAT collections in Tables 3.1 and 3.3include YPFB payments); b) an increase in VAT collections on imports from42 of the CIF value in 1989 to 6% in 1993-97; and c) a rise in specificconsumption taxes from 0.7Z of consumption in 1989 to 1.52 in 1993-97. Ofcourse, it is easy to imagine other combinations of increased taxcollections which would provide the same level of revenues, and theseassumptions hardly exhaust the opportunities for increasing taxes. Ourpurpose is simply to quantify one means of achieving the necessary revenueeffort.

224. Utnder the base case assumption of increased collections, Treasuryrevenues would rise to 17.8% of GDP in 1993-97 (compared to 16% in the noimprovements scenario), and the savings or the nonfinancial public sectorwould increase to 3.4Z of GDP (versus 1.5% in the n) improvementsscenario). Higher savings would reduce the overall balance, thus limitingthe state's demand for finance to that available from external sources.

Table 3.2:Fiscal Accounts of the Non-Financial

Public Sector (as a percent of CDP) 1993-97 1993-9'

No Improv ement Base Came

1987 1988 1989 1990 1991 1992 in Tax Effort

Treasury Currrent Revenues 13.9X 17.6O 15.9X 17.2X 17.0% 16.0X 16.0X 17.8X

Tax Reform 4.0X 6.1X 8.9X 7.1X 7.0X 7.2X 7.2X 8.51

of which:Value Added Tax 2.2X 3.4% 3.7X 4.1X 4.1X 4.11 4.1X 5.0X

Transactions Tax 0.31 0.6f 0.6X 0.5% 0.6% 6.51 6.51 6.5%

Specific Tax 0.3% 0.4% 6.81 0.6% 0.6% 0.7X 0.7% 1.21

Direct Tax"e on YPFB 6.8% 8.51 6.71 7.65 7.4% 6.21 6.21 6.81

Other Stat. Enterprises 0.51 0.2% 6.61 0.3% 0.31 0.21 0.21 0.3%

Trade Taxes */ 2.31 2.1% 1.9X 2.21 2.1X 2.21 2.21 2.1%

Other e.81 6.1X 0.41 0.21 0.2X 0.21 0.21 0.21

Treasury Current Expenditures 16.41 13.31 17.1X 16.91 16.61 16.21 16.21 16.21

1 Net Savings of Other Public -0.91 -2.9X 2.6X 2.65 2.31 -0.3X 1.71 1.7%

un SSctor Entitiesr- of which:

SEE Net Transfers to Treasury 6.71 8.51 8.71 7.71 7.61 6.31 7.0% 7.06

Net Savings of the Non-Financial -1.71 0.71 0.81 2.8% 2.8X -0.4% 1.FX 3.4X

Public Sector

Capital Incom 0.65 6.91 1.0X 1.31 1.0X 1.0X 1.01 1.0X

C.pital Expenditure 6.2X 8.3X 7.6X 11.2% 10.7% 10.6X 8.8X 8.6x

Overall Balance -7.3X -6.71 -5.71 -7.1X -6.9% -9.9X -6.0X -4.21

External Finance 2.4X 4.0% 7.0X 8.8X 6.9X 9.4X 4.1X 4.1X

Internal Finance 4.9X 2.7X -1.3% 0.3X 1.0% 6.6X 1.91 6.11

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - --_ _-

- - --_-

- - - - - - --_-

- -- - - -

-76-

225. The structure of Government revenues would look different by 1997(see Table 3.3). In 1988, transfers and taxes from state enterprises madeup 51.3Z of Treasury revenues, and revenues from the new tax system 35.8?.By 1997, state enterprises would provide 40.72 of a substantially higherlevel of revenues, compared to 47.52 from domestic taxes. The share ofrevenues from the foreign sector (excluding state enterprises) would alsofall. Customs duties would decline to 10.1? of Treasury receipts. Boliviawould have achieved a much more diversified tax base and established a taxregime more responsive to changes in domestic economic activity than tochanges in the international economic environment, goals that repeatedlyhave eluded administretions in Bolivian history.

Table 3.3: Treasury Revenues(percent of total)

1988 1997

Tax Reform 35.8 47.5of which: VAT 19.8 23.4State Enterprises 51.3 40.7Trade Taxes 11.9 10.1Other 1.0 1.7

226. The projected level of revenues would be sufficient only tomaintain current expenditures (excluding interest payments) at presentlevels relative to GDP and finance the investment necessary to achieve the3.8Z rate of GDP growth assumed in the base case. Any increase in theeconomy's resources devoted to improving Government services would requireadditional revenues. In particular, to the extent that the Governmentwishes to raise GDP growth beyond the base case, improve the inadequatelevel of health and education services, or even to perform necessarymaintenance regarded as a current expenditure, additional taxes or highertax rates would be required.

227. The Government has a number of alternatives which would raiserevenues well above the levels projected in the base case. For example, asuccessful program to integrate larger portions of the informal sector intothe legal economy could generate substantial revenues. While theGovernment plans to implement the rural land tax this year, the rates ofthe tax are very low, well below initial proposals made in tne early yearsof the Paz Estensorro Government. Given the large amount of land held onunproductive estates, an increase in the land tax rates (if successfullyenforced) would yield substantial revenues while having littledistortionary impact on the economy. Lower deductions for VAT payments ora change in the tax rate on the complementary tax also could increaserevenues. Thus it is safe tc conclude that the Bolivian tax s,stem hasconsiderable revenue potential. Assuming that future Governments continuethe impressive administrative and poliry performance of the Paz Estenssororegime, public sector revenues should be sufficient to finance the publicsector's contribution to development.

- 77 -

CHAPTER IV: ALLEVIATING THE SOCIAL COSTS OF ADJUSTMENT2 4

I. Introduction and Summary

228. The economic crisis in the first half of the 1980s imposed severehardships on the poor in Bolivia. Given the tight fiscal constraintsnecessary to control inflation, options for assisting the poor are limited.The Government's efforts in this area have been channeled through theEmergency Social Fund (ESF).

229. ESF was intended to soften any harmful impact of the adjustmentprogram on the Bolivian poor and increase the acceptance among thepopulation of the dramatic changes in economic policy under the NEP.Traditional food assistance and subsidy programs were felt to beinefficient and contrary to the principles underlying the NEP. Therefore,ESF was designed as a financial institution to provide funds for locally-generated projects with high labor content. To respond as quickly aspossible to the emergency, the Government exempted ESF from procurementregulations and restrictions on wage levels, in order to attract a highly-qualified staff.

230. Experience with the initial phase of ESF operations suggested somemodification in design, including setting up of an outreach program toimprove project submissions, tightening of supervision to improveimplementation, and a shift in emphasis from low-cost, high employmentprojects towards social assistance and more difficult infrastructureprojects with higher long-term benefits. These changes, along with ahighly-efficient management information system, have greatly increased theeffectiveness of ESF operations. ESF disburses about US$2 million eachweek and controls around 1000 projects with a staff of less than 80professionals.

231. ESF has had a large impact on the economy. The program hasattracted and effectively employed substantial funds from internationaldonors, thus easing the adjustment process. The concentration of projectsin non-traded activities (largely construction) may have slowed the desiredtransfer of resources to the tradeables sector, but in the longer run theinfrastructure improvements should help increase tradeables production.ESF has increased formal sector employment (by end-1988 27,000 persons wereworking in ESF projects), and has increased the incomes of ESF workers.ESF has generally been successful in reaching the poorer elements of thepopulation, as evidenced by the fact that ESF households have lower foodco.nsumption per capita, and (to a lesser extent) nave lower incomes thanthe population at large. Less success has been achieved in targettinginfrastructure improvements to low-income communities.

232. ESF may have had some role in increasing acceptance of the NEP bythe population. ESF clearly enjoys wide political support, and ESF staffhave been successful in working with groups traditionally npposed to

24/ This Chapter is based on preliminary work for a forthcoming study onthe ESF.

- 78 -

Government policies, including some of the more politically active NGOs andlabor unions. Opinion surveys reveal generally favorable views of ESFamong recipients.

II. Social Costs of Adiustment

233. Introduction. From the outset of the NEP, the Bolivian Governmentattached a high priority to measures to alleviate any further deteriorationin social conditions. This was due to fears over the effect of economicdislocation on the already very low living standards of the Bolivian poor,as well as concern that increased hardship would strengthen opposition tothe Government's policies.

234. Social Conditions Before 1985. Social conditions in Bolivia wereamong the worst in the hemisphere before the implementation of the NEP.Available data showed very low levels of life expectancy and high mortalityrates. Life expectancy for males and feniales in rural areas was estimatedto be 47 and 51 years, respectively, the low numbers mainly due to highinfant mortality rates. A 1983 UNICEF survey found that more than 20X ofall children born in the Altiplano died before the age of 1 and more than30Z died before the age of 2. The infant mortality rate for Bolivia as awhole was 123 per 1000 live births on average during 1980-85, according toofficial estimates. Most deaths among children are due to diarrheal-parasitic and acute respiratory diseases; protein-energy malnutrition isassociated with more than a third of the deaths. A national nutritionalsurvey found that 411 of all children under the age of 5 were affected byprotein-energy malnutrition.

235. Impact of the NEP. Data on the determinants of wplfare haveworsened substantially si.nce 1985. Per capita GDP has fallen by 6Z, andper capita consumption by 4Z. Real salaries fell in most sectors of theregistered economy, and a shift in employment to the low-wage informalsector has further reduced average income (see Chapter 2). Governmentspending on social programs was cut substantially as a part of thestabilization measures in 1985 and has still not recovered fullv, while thepopulation has increased. However, it is difficult to estimate the impactof this decline on the poor, as we lack data on the distributional effectof traditional public programs. In fact, there is evidence that socialconditions improved, at least for some groups. For example, it appearsthat nutritional status in rural areas showed some improvement over thepast few years. More data would be required to determine the overallimpact on social conditions of the adjustment program.

III. ESF Design and Operation

236. Introduction and Summary. ESF was structured as a financialinstitution to fund labor-intensive infrastructure and social assistanceprojects proposed by local agencies. ESF was organized outside the normalbureaucratic structure, with exemptions from wage limits and procurementrequirements to increase efficiency. After the initial uhase ofopcratiorio, changes in prQje.. pzuinmotion and supervision were adopted.More detailed information oni the strucLure of the ESF and opcrationalimprovements can be found in Annex III.

- 79 -

237. Initial Proposals. Various options were considered foralleviating the social costs of adjustment before setting up the ESF. Theearly discussions in 1985 revolved around more traditional programs ofwidespread distribution of food or medicines, or direct subsidies forselected items. A number of difficulties were perceived with thisapproach. First, the administrative costs in directly providing, orsubsidizing the purchase of, commodities to low income groups would havebeen substantial, with no return in terms of development. Second,providing free food (particularly if taken from international donations)would have discouraged the production of food, increased the country'sdependence on food imports, and reduced the efficiency of resourceallocaticn. Third, permitting recipients more discretion in the use ofassistance (providing money instead of food, for example) was viewed asmore consistent with the reduced role of the state inherent in the NEPreforms. Finally, the Government wished to avoid imposing a uniform methodof assistance, but rather to respond to the demands of local groups whichwould be in a better position to judge the most immediate needs.

238. ESF Projects and Structure. Following these principles, the ESFwas structured as a financial institution to provide funds for small-scale,labor-intensive projects, mostly in infrastructure. The projects weremeant to be demand-driven, meaning that the ESF would finance projectsproposed by local groups. The focus on labor-intensive, infrastructureprojects was adopted to increase employment among the poor while at thesame time providing services to the community which would aid development.

239. ESF was designed as a temporary institution, scheduled todisappear in December 1989. It was hoped that in three years time growthwould have picked up enough to solve the employment problem, and that theneed for austerity measures would be less, allowing for increases in social

i spending as part of the regular budget. Another likely reason for theplanned three-year life of ESF was to let ESF's termination coincide withthat of the Government of Paz Estenssoro. Subsequently, the life of ESFwas extended through 1990 to allow an orderly transfer of the experienceand the projects to other institutions.

240. To maximize the speed and efficiency of operation, ESF was set upessentially outside of the normal bureaucratic structure. The Governmentexempted ESF from several regulations ordinarily imposed on public sectorinstitutions. In particular, ESF was permitted to avoid the complicatedand slow procurement procedures followed by other public sector entities,and was allowed to pay higher salaries than those received by employees ofthe Central Government. This helped to attract the most qualifiedpersonnel. Organi:ationally, ESF was placed outside the line ministriesand depended directly on the Ptesidency of the Republic. The Presidenttook a personal interest in the institution, giving it the necessarypolitical weight.

241. Size and Distribution. ESF's total program was planned as US$180m.illon over the lifc of the institution. Funds are committed by ESF onlywhen it has received commitments for financing. By early 1989. fundingappears to be falling short on targ-t. Therefore. ESF .may be forced tosluw down the current pace of its activities to ensure that commitments forapproved projects do not exceed available funds.

- 80 -

242. ESF funds projects in four basic categories: economicinfrastructure, social infrastructure, social assistance, and productionsupport. Economic infrastructure (covering 37.81 of committed funds)encompasses infrastructure closely related to productive activities,including road maintenance and upgrading, urban improvement, irrigation,flood control, and reforestation. Social infrastructure (47.6? ofcommitted funds) covers infrastructrue for health and education, water andsanitation, basic housing (mostly self-construction), and some culturalprojects (for example, repairs of historic ouildings). Social assistance(8.82 of committed funds) covers recurrent costs in education and training,vaccinations, school breakfasts, and production of school materials.Production support (5.7X of committed funds) is mainly credit providedthrough NGOs to productive units which are outside of the formal financialsystem, such as microenterprises producing for the informal sector andsmall cooperatives in mining and agriculture. Table 4.1 shows ESF'sresults in physical and financial terms as of December 31, 1988.

243. ESF Operations. Experience with the initial phase of ESFoperations suggested a number of modifications to the initial design,particularly in the areas of project promotion, supervision, and the kindsof projects approved. An outreach program was instituted to improve thequality of projects submitted, and supervision was tightened to avoid casesof poor implementation. There has been a shift in ESF projects from low-cost, high employment projects to a greater emphasis on infrastructureimprovements that will aid long-term development, and to social assistanceprojects. ESF has achieved a remarkable success in administering a largeprogram with relatively few staff, aided by a highly-efficient managementinformation system. Annex III provides a description of these operationalimprovements, as well as more detailed material on the structure of theESF.

IV. Economic Impact of the ESF

244. Introduction and Summary. The success of the ESF in obtainingforeign assistance has eased the adjustment process. While in the shortterm, the concentration of ESF projects in non-traded goods may have donelittle to aid the necessary transfer of resources to the traded goodssector, ESF's contribution to infrastructure will aid long-term developmentof tradeables. Available data indicata that ESF has increased formal sectoremployment, and has raised t';e wages of participants beyond what they couldhave expected to earn in the job market. ESF has had some success intargetting benefits to the poor. ESF households generally consume lessfood per person than the population at large. Although the per capitaincome of ESF households is not concentrated in the lowest deciles of percapita inconme in the population, the expected income of ESF households ifthe ESF worker had another job wsuld place almost half of ESF households inthe lowest three deci1.es. However, ESF has had less success in targetinginfrastructure improvements to the poorest areas of the country, owing tothe lack of institutions in poor, rural areas able to prepare projects.

_45. Macroeconomic impact. The ErF has eased adjustrmient throughcapturing more than TTS$100 million in foreign currency by the end of 1988,or about one-fourth of the current account deficit in 1498, Most of thisfinancing has been in the form of grants. There is substantial reason tobelieve that foreign aid inflows would have been considerably lower without

Table 4. 1:

Emergency Social FundAchievemkents Summary

ASOF DECEMBER 31, 19'8

AMIOUNT _NUMBER OF _ AMOUNT _WORKtS IN EXECUllON OR APPROVEC _ BENEFICIARIES _ PEOPt.E WORKING

CONMITED PROJECTS DISBURSED direct berlciei fe PROJ. EN EXECUTION)

CAL - i... indirect ben6hcarios

NIURWflN 3.04 49 1.25 3.968.959 br"kfas nd M news 39.990 (d)

HALTH CAF 2.4 2 1.22 1.603.000 vcctbons 2.64.836 (,)

PAeWGANDEJCATNAL SLPRP 4.01 71 197 64.000 l xt: 48.000 sdo dosks 771.204 (d)

5rKI1T ENkW ORT 0.8 11 0 29 12 proects (SUS. 517,38 1)

TOTAL 10.3 16 4.7 3 3.455.030 (d.1)

CCIA.L SIFRASTRUCTURE

CMWNGWATERANOSEWAGI 18 0 23 8 S 996 298 mete ISo (d) 4.40

HOIGM 13.2 a 3 6 p S."34 houswa 28.025 Id) 2.33

{ILMTH CO61RXU16 4.93 114 2 1 354 constucbone 1.16,4.928 (0) 19 91a

SCH aA N6TRIUC:IW 12.7 334 5. 66 215 coubuc6ons 590 (di 2.315

00 CUtJnO RDECIS 1 0 28 D 65 (d) 131

> 6.1 149 3 7 49proects (US 1)467,505) 79.763 (-)

TOTAL 56.1 94 27.81 1.273.406 (d.I) 2.71

rCOMO WFRASTRUTURU 1 -

7N0NEWNT 11.30 163 7.26 838.340in2olsbo P fret p| |g 454.750 0. 3.621

RW t1 MANIEPWAM l i 41| 11| 119.41 17 10.1 6.681 Km. 752.230 5.) 5,396

6lIGATK N 3.68 56 1.84 27,775 Has. 65.947 (d) 861

10PAROGSYSl IS 3 20 27 0.94 212.773 rmetrs ol walsI 137.420 ) 492

FIO DCE IASI1RJCT|OiE 0.|13 32 piojecb (5US 93Z292) 36.015 e) 1.053

|E NTN DFORES'TIO 4.61 5 3 4 102.343 rye olgavione 128.161 C) 864

1.46 43 1 1 31 prtmects 235.575 C') 25

ITOTAL 445 5I 24.6 1 _ _|_| 1.610,098 (d.I _ 12_ _3_ _

PsOOUCTION SUPPORT L I -11 Ir 60 1-1-

lRnATMRUND61 2.231 91 1.S81 | 7718 arml u s 600 (d1)

PffDUCN( lEIS 25 pA 3r50 22| 1.11r | pojects (SUS 3.5001.869) 60 (1C) 493

CMOs 1.10 1 28 1 0.34 1 1 11 projocts ($US. 1.097,756) 21,156 (1) 1 7TOTAL I_3_ 18.31 L * 91__ 3_.131 _|_I_| 21,816 f1) 61 6

|10B TOTAL L [ 7.L ZE3 [7 [.3530 [ 773]

- 82 -

ESF. Several donors have begun or resumed programs in Bolivia because ESFwas able ensure speedy and efficient implementation. Furthermore, ESFfunds have been disbursed to Bolivia a lot faster than under traditionaldevelopment projects. The average lag between date of commitment anddisbursement of all funds is less than one year in ESF, shorter than evenfast-disbursing balance of payments support. Rapid disbursements have beenfacilitated by the simp]e nature of the projects and the mass-productionprinciples adopted for ESF operations.

246. The impact of the projects themselves on adjustment is difficultto judge. On the one hand, the composition of ESF projects may have slowedadjustment somewhat. Most of ESF's projects are in construction and thusbenefit the non-traded goods sector, while a key goal of the adjustmentprogram is to increase incentives for tradeables production. Thus,resources which might otherwise be used to increase exports or produceimport substitutes are instead being used to produce non-tradeable goods.On the other hand, to the extent that ESF projects are highly laborintensive and succeed in adding jobs (rather than simply competing foralready-employed workers), the emphasis on non-traded goods does not affectadjustment. ESF also has made a significant contribution to Bolivia'sinfrastructure, which should aid tradeables production over the long term.The many projects in road ms_iatenance will help circulation in the countrysubstantially; so far more than 6000 km of roads have been improved. Lowerroad costs will assist agricultural production, which faces a severeconstraint in high costs of transport. Similarly, projects in irrigationand flood control have reduced the costs of agricultural production.

247. An independent Bolivian study estimated that the effect of theUS$37 million committed during 1987 would be an increase in GDP of aboutlZ. However, the study is based on an old input-output matrix arid needsupdating. ESF has clearly had some effect in the depressed constructionindustry and through multiplier effects on production in general. ESFinvestments, which will be about US$100 million in 1989, will add 30Z tothe public sector investment program. If it is assumed, ratherconservatively, that there is only a negligible demand multiplier effect ofthe investment, an extra US$100 million of demand (investment) would givean increRse of about 2Z in GDP.25

248. Impact on Employment. ESF appears to have had a substantiJ. rolein increasing formal sector employment. By the end of 1988 as many As27,000 people were working on ESF financed projects.2 6 The NationalInstitute of Statistics (INE) estimates that for each person directlyemployed through ESF, an additional 1.1 persons drE employed indirectlythrough the production of materials for ESF projects. This rather highcoefficient is due to the low import content of ESF projects.

25/ Assuming that the demand effect Is not lost to inflation and that thereis no crowding out of onher invesLm.cnt.

26/ There is some evidence that this estimate is about 102 too high, as itis calculated from dividing the total ESF wage bill by the standardwage. Some studies found that many ESF projects pay more than thestandard wage, which would result in an over estimate of employment.

- 83 -

249. It is very difficult to measure the increase in employmentattributable to ESF. We lack the data to trace the effects of theincreased demand in the construction industry through the different sectorsand on to labor supply. One clue to determining the impact on employmentof the ESF concerns changes in wages in the construction sector, where mostESF projects are located. The increase in ESF employment in 1988 isestimated to have been about 17,000 jobs, somewhat greater than totalemployment in the construction sector before the ESF. Despite thistremendous increase in demand for labor in construction, real wages inconstruction (and in other sectors of the economy) rose only slightly in1988. Lack of pressure on wages indicates a high level of unemploymentprior to inauguration of the ESF, implying some increase in employment as aresult of the ESF.

250. This rise in employment in the formal sector of the economy mayhave come from workers in the informal sector, rather than from people whootherwise would have had no job at all. True unemployment is very low inthe Bolivian economy, owing to the lack of unemployment insurance orwelfare services (i.e., people must make a living somehow). Even so,increasing formal sector employment is of considerable value to theeconomy. Greater participation in the informal sector eLodes the tax baseand may resu-lt in dn inetficient scale of production (see Chapter 2). Moreimportantly in this context, workers who turned to the informal sectorafter losing jobs during the adjustment program generally received very lowwages. For example, miners fired from COMIBOL may have tried to make aliving through selling goods on the street in La Paz. Viewed this way, theESF may not have increased employment so much as increased wages andproducti-vity of already 'employed' workers.

251. Thuqs a better proxy for the ef.-?ct or ESF workers is to estimatetheir increased income in the ESF project compared to their expected incomein the economy as a whole. For most ESF workers (90Z in one study) theincome earned in the project is the principal source of income for theirhousehold (802 of the workers had more than one dependent). Based onhousehold survey information a wage equation was estimated using gender,age, location and schooling as independent variables. The estimatedcoefficients were then used to determine what ESF workers would have beenable to make on average in the labor market as a whole (both infornal andformal sectors), if they could have found a job. This exercise shows thaton average ESF workers earned 43Z more than they would otherwise have made,if they had been employed. Future work will refine this result.

252. Ttp relative improvement in income is higher in poorer cities.The negative effect on earnings of being in a depressed city is less forESF workers than for the population at large, indicating that the relativeimprovement in income is higher for ESF workers in depressed areas. Forexample, the difference between what ESF workers are paid and what theywould earn in other occupations is twice as high in Potosi as the samecomparison for the nation as a whole.

253. Income Distribution of ESF Workers. ESF workers are generallypoorer and spend less on food per capita than the average Bolivian. Theonly reliable information we have about ESF workers is from a national

- 84 -

household survey which only covered the capitals of the departments. Thus,the results presented here are only valid for urban projects, whichrepresent about 55Z of all projects. Table 4.2 shows the percentage of ESFworkers in each decile of population, ranked by household per capita foodconsumption27 . Of the total ESF workforce, 43.1X of ESF households fallwithin the lowest three deciles and 8.8? fall within the top three deciles.It is interesting to note that there may have been a fall in the percentageof ESF workers in both the highest and lowest deciles since 1987. Asimilar study in that year found 52.32 in the lowest three deciles but 162in the top three deciles.

Table 4.2: Per Capita Food Consumption of ESF Workers(In percent of each decile of the population)

Decile Percent of ESF Workers(percent of population) In Each Decile

9-1OZ 11.6210-202 15.4220-302 16.1230-40? 12.3240-50Z 9.3250-60% 11.6?60-70Z 4.8?70-80Z 1.2280-90 4.8Z90-10OZ 2.8?

254. ESF appears to be less progressive when measured against ofhousehold income rather than per capita consumption (see the black bars inFigure 4.1) In fact, the lowest income decile is under represented (8.7?),while 37.5? are in the lowest three deciles, and 21.1? in the top threedeciles. When looking at these numbers it should be remembered that thisfamily income includes the income earned in the ESF project. If incomeexcluding earnings from the ESF is predicted for each worker using the wagefunction for the population at large, and this is added to other householdmembers's income (which is negligible in relative terms in the lowerdeciles) the pattern changes substantially (see the shaded bars in Figure4.1). This exercise shows that in the absence of the ESF (estimating theESF worker's income on the basis of the wage function explained above),46.0? of ESF families would have been in the lowest three deciles and 14.2Zin the highest three deciles. As all projects do require some skilledworkers (e.g. bricklayers, nurses, architects, and trainers), it is notsurprising that there is some "leakage" of ESF employment into higher-income deciles.

27/ This was chosen because it is closer to a "permanent incombe measurethan household income.

FIGURE 4. 1:

distribution across family incomedeciles with and without ESF

percent in EPH sample25 - -

20

15

10

5

01 2 3 4 5 6 7 8 9 10

decile of per capita income

M with ESF m without ESF

Note: dociles calculated from EPHi sample

- 86 -

255. Households of ESF workers are generally less educated than thepopulation at large. Table 4.3 shows the distribution of adult householdmembers among educational categories for ESF households and the householdsurvey sample at large. For instance, 59.82 of household members of ESF-workers have only basic or no education whereas the same percentage is37.61 for the population at large. At the other end of the scale 3.22 ofESF household members and 20.52 of the population at large have highereducation. A statistical test strongly rejected the hypothesis that theprobability of being in any one educational category is the same in the twosamples.

Table 4.3: Educational Levels of ESF and General Population(percent)

Educational Level ESF Households Population

none 14.7 8.8basic 45.1 28.8intermediate 19.8 16.0secondary 16.2 25.8technical 0.5 3.8normal 0.5 4.1university 2.3 12.3others 0.9 0.3

256. GeoRraphical distribution of ESF projects. ESF has invested morein depressed departmentsZ8 than in the more affluent ones, although thisresult breaks down if the distribution of ESF projects at the provinciallevel is analyzed. As no independent survey information exists as to whothe beneficiaries of the ESF projects are, inferences on the distributiveeffects of the infrastructure built has to be based on geographicalinformation.

257. The programming of ESF resources geographically has attempted tofollow the poverty contours of the country. A methodology was establishedto set funding targets by department. The distribution of projects wasbased on GNP per capita, per capita transfers from the Central Government,unemployment, infant mortality, urban social infrastructure, condition ofrural life, and percentage of school-age children enrolled. As of August1988, no department had a greater than 0.3Z deviation from the targetinvestment share.

258. However, the success of these targetting efforts breaks down atthe provincial level, as shown by Table 4.4, which summarizes thegeographical distribution of ESF commitments by province. Each province isgiven a designation between 1 and 5 based on the poverty index, with 5

28/ Bolivia is divided into nine departments and one hundred provinces. ESFhas projects in eighty of these provinces.

- 87 -

being the areas of most critical poverty. Provinces have been aggregatedinto areas corresponding to poverty level (rather than departmentallocation), as have ESF expenditures. Table 4.4 shows that ESF projects arelargely concentrated in the higher-income provinces. Of course, evenprovince level data often mask large disparities in the standard of living,particularly within urban areas. For example, Poverty Area I includes onlydepartmental capitals, and in these cities ESF projects are generallylocated in the poorest urban areas. It is therefore possible that ESFprojects are reaching pockets of poverty within generally better-offprovinces.

Table 4.4: Distribution of ESF Expenditures by Poverty Area

ESF CommitmentsPoverty Area 1985 Population ESF Commitments per Capita

(millions) (US$ millions) (US$)

I 2.3 56.1 24.0II 0.7 20.3 27.8III 1.3 17.3 13.5IV 1.4 16.3 11.7V 0.7 6.2 9.5

259. The poor performance of targetting resources on the poorerprovinces is a direct result of the demand-driven nature of the ESF.Typically, the poorest communities are also those with the weakestinstitutions and lowest levels of communication. These two factors arecrucial in determining a community's ability to prepare an acceptableproject. Moreover, the lack of Government and NGO presence in many of themost needy areas of Bolivia has meant that institutional intermediaries arenot available for organizing projects in the poorest areas. In the absenceof any institution to work with, even outreach efforts by ESF staff couldnot succeed in starting projects.

-88-

CHAPTER V: MEDIUM TERM PROJECTIONS2 9

I. Introduction

260. Bolivia faces an extraordinarily difficult set of developmentchallenges. The public sector's external debt equalled 972 of GDP andmore than five times the value of exports in 1988, while debt service owedmade up 52Z of export receipts. Export revenues are dominated by mineralsand natural gas; the former are highly vulnerable to changes ininternational economic conditions, while payment for the latter fromArgentina is becoming increasingly uncertain. Domestic savings equalledonly 4Z of GDP in 1987, so that even the relatively low level of investment(1OZ of GDP) was largely financed from external sources. The low supply ofsavings results in very high rates of interest, severely constrainingprivate sector investment. A shortage of skilled personnel coupled withhigh levels of corruption and heavy political influence in administrativedecisions in the past have produced an inefficient bureaucracy and severeshortcomings in the management of public services. Most seriously, lowlevels of education and lack of access to basic health services seriouslyimpair the efficiency of the Bolivian labor force.

261. Given these difficulties, enormous efforts will be required toincrease the rate of growth of GDP while reducing dependence on externalcapital inflows. The projections in the moderate groeth scenario show thedomestic policies and international assittance necessary to achieve anaverage growth ratp cf 3.8Z, avoid a further decline in per capitaconsun.pLion, and reduce the size of the current account deficit relative tooutput. This will require increases in domestic savings from 9Z of GDP in1988 to 15X by 1997; increased efficiency in the use of investmentresources; strong growth in exports of non-tin minerals, soya, and othernontraditionals; development of alternative markets for Bolivia's naturalgas after the expected reduction in gas sales to Argentina in 1992; highlevels of net lending by official donors at concessional rates; andrescheduling of Bolivia's debt service payments to bilateral creditors.Continued macroeconomic stabilitv, commitment to the liberalization ofmarkets achieved under the NEP, an appropriate exchange rate policy, andfurther improvements to the incentives structure to encourage privateinvestment are the essential prerequisites to achieving these goals.

29/ The projections outlined here are very sensitive to estimates for therecent past. It is extremely difficult to obtain reliable economicdata in Bolivia, which increases the likelihood of error in theprojections beyond the normal level of uncertainty for exercises ofthis type. Still, our information is sufficiently reliable to drawbroad conclusions on the likely evolution of Bolivia's economy.

-89-

Table 5.1: Domestic Performance(average for period)

History Moderate Slow Accelerated(1980-87) (1989-97) (1989-97) (1989-97)

GDP Growth (S) -2.1 3.8 2.2 6.2Fixed Investment/GDP (2) /a 7.3 14.9 10.3 16.7Domestic Savings/GDP (2) 9.2 13.5 10.3 14.0ICOR -4.4 4.0 5.0 3.0Import Elasticity 1.0 1.2 1.2 1.2Consumption Per Capita (2) -1.1 0.3 -0.8 2.5

/a The high investment ratio in the projections reflects investmentfor Brazil power project.

262. The slow growth scenario illustrates the impl'cations for theBolivian economy if the domestic policy improvements assumed in themoderate growth scenario are not forthcoming. In this scenario, failure topass legislation necessary to establish an appropriate framework forinvestment (including the investment code, hydrocarbons code, and miningcode), lack of commitment to the privatization process, and lack ofimprovements in public sector administration result in lower domesticsavings (owing to reduced confidence), reduced capital inflows, andimpaired efficiency. Under these conditions, domestic savings would riseto only 122 of GDP by 1997 (3 percentage points below the moderate growthscenario), export growth would average 52 from 1989-97 compared to 6Z inthe moderate growth scenario, and GDP would rise by only 22 a year. Percapita consumption would decline by almost 12 a year from 1989-97, raisingserious doubts as to the political sustainability of the economic program.

263. "he accelerated growth scenario shows the policies and supplyresponse necessary to achieve a significant increase in per capitaconsumpti(n over the next decade. In this scenario, it is assumed that allof the poLicy improvements in the moderate growth scenario are undertaken,and, in addition, that: (a) the private sector responds more rapidly tothe liberalization program, facilitating a more rapid rise in investmentthan in the moderate growth scenario; (b) the Government achieves a morerapid than expected improvement in public sector administration, raisingefficiency and increasing confidence, and hence savings; and (c) Argentinacontinues to buy Bolivian gas at the same level (in vulume terms) through1997, compared to the 502 cut in 1992 assumed in the moderate growthscenario. With these assumptions, GDP growth would average about 6.02 ayear, supported by a more rapid increase in domestic savings and a fall inthe ICOR, compared to the moderate growth scenario. Per capita consumptionwould increase by 2.5Z a year, implying a significant improvement in livingstandards by the end of this century.

-90-

264. The Government has recently completed an overview of Bolivia'sdevelopment strategy through the year 2000, which presents recommendedpolicies and investment priorities for each of the major economic sectors.This exercise was extremely useful in setting out a comprehensive view ofthe measures necessary to achieve sustained development. One element ofthe strategy document is a set of medium-term macroeconomic projections,which are somewhat more optimistic than those in the moderate growthscenario in this chapter. The Bank and the Government have held detaileddiscussions concerning our differing views of Bolivia's prospects. Themain differences revolve around the level of private capital inflows whichBolivia will receive during the 1990s, and the efficiency of investment.The accelerated growth scenario presented in this document is closer to thep.ojections presented in the Government's strategy.

.I. Moderate Growth Scenario

265. Introduction. This scenario shows what would be required for theBolivian economy to increase GDP growth rates to above 3.52 a year, avoid afurther decline in per capita consumption, and reduce its current accountdeficit and debt burden relative to GDP. The domestic performancenecessary to achieve these goals includes a strong rise in domesticsavings; an ICOR of about 4; a rapid expansion of non-tin minerals andnontraditional exports; and the sale of electrical power to Brazil, asenvisioned under the recent agreements between the two countries.Increased inflows of external savings play an important role in financingnecessary investment until the projected rise in domestic savings can beachieved. Improvements in domestic policies are required to raiseefficiency and encourage the private sector confidence essential to achievethe projected growth rates.

266. Output. Output increased by 2.8Z in 1988, and is anticipated torise by 3.52 in 1989. Under the moderate growth scenario, GDP growthexceeds 4Z in 1991, but then falls to below 22 in 1992 as the economyadjusts to the 50X reduction in natural gas exports to Argentina (seebelow). Thereafter, growth recovers to about 4? per year, resulting in anaverage annual growth of 3.8? from 1989-97. Improvements in efficiencyplus higher savings and investment should permit more rapid growth after1997, perhaps on the order of 4.5-52 per year. The projected GDP growthrate will require a substantial rise in investment. While the ratio ofinvestment to GDP is inflated in 1990-93 (averaging about 18z) owing to thepower plant project, investment excluding the project is projected toincrease from 10? of GDP in 1988 to 162 by 1 9 97 .3U

267. The projected investment levels will require a sharp increase indomestic savings. As the public gains more confidence in the maintenanceof price stability and the NEP incentives structure, and as the benefits of

30/ Note that the investment ratio given in Table 5.2 includes the changein stocks. The ICOR calculated from fixed investment averages 4 overthe forecast period.

-91-

more appropriate economic policies are seen through higher GDP growth,savings should increase. High marginal savings rates result in an increasein the ratio of domestic savings to GDP from 9Z in 1988 to over 15Z by 1997(see Table 5.2). By 1997, nearly all investment would be financed bydomestic savings, compared to one-fourth of investment projected to befinanced from external sources in 1989. The rise in savings is achieved inthe short term through a fall in per capita consumption. However,consumption per capita increases by almost 1Z a year in the latter half ofthe forecast period, as GDP growth rises and the marginal savings ratedeclines somewhat. On average over the forecast period, per capitaconsumption increases marginally in real terms.

268. Exports. Export volumes are projected to rise by 6.52 a year from1989-97. Minerals exports increase strongly in the short term owing to arecovery in COMIBOL (the state mining company) and increased private sectorinvestment. Evidence of a rebound in the mining sector has already beenseen in 1988, and we have received a number of reports of increasedinterest in Bolivian mining oa the part cf foreign investors. Exports ofsoya are also projected to rise rapidly over the next eight years. Japanappears ready to provide a secure market for Bolivian soya, and preliminarydata indicate the availability of large areas of land suitable for growingsoya. Other non-traditional exports are assumed to increase by 10% a year.It is hoped that the new incentives structure will facilitate somediversification of Bolivian exports into textiles, wood and wood products,and simple industrial products. Even sa, by 1997 non-traditional exportswould only account for 30% of export revenues, compared to 402 for mineralsand 30% for energy.

Table 6.2: Moderate Growth Scenario(perc-nt)

1987 1988 19Pa 1990 199' 1m 1997

Constant Prices

GDP 2.4X 2.8X 3.5% 4.4X 4.2X 2.1X 3.91Consumption per Capita 1.61 -6.56 -0.8X -0.7S 1.60% 1.6 0.4XInvestm.nt/GDP JI 16.63 11.63 13.2X 18.06 17.83 17.73 16.13Domestic Savings/GOP 4.06 9.3X 10.63 12.86 12.9X 11.63 15.33External Savings/GDP 6.9X 2.33 2.3 6.543 4.9X 6.7X 0.83

Current Prices

Current Account/GDP / -12.65 -9.06 -9.73 -12.93 -12.1% -12.7% -6.63Public Debt/GDP 166.06 96.83 90.83 96.3X 96.73 99,3X 84.33

/u Incr6ase in investment and current account to GDP ratios in 1990-92 refloct expenditures underproject to sell eloctricity to Brazil. Investment data includes change in stocks.

-92-

Table 5.3 Exports in the Moderate Growth Scenario(percentage change per annum in constant prices)

US$ Level1970-80 1980-85 1985-88 in 1988 1989 1989-97

Total Exports 1.0 -8.1 8.2 599.0 7.0 6.5Mi.Lerals -1.4 -10.0 1.7 276.5 14.7 6.2Energy 4.3 -0.3 1.4 217.9 -1.5 2.6Soya -- 6.3 52.5 20.3 23.7 17.2Other 16.2 -19.3 27.4 84.3 1.9 11.3

Terms of Trade 4.1 1.9 -19.0 -- -0.7 0.5

269. Prospects for natura'l gas exports are extremely uncertain, as theydepend in the near term on Argentina maintaining its payments under thepresent contract, and after 1992 on reaching a new agreement between thetwo countries. Argentina's arrears on payment for Bolivia's natural gasexports reached US$124 million by the end of 1988 (a US$4 million increasefrom the end-1987 level), and have increased alarmingly during 1985.Bolivia is in a partirularly difficult bargaining position, as it presentlyhas no means of selling the gas in other markets. Nevertheless, we assumethat an easing of the economic crisis in Arger;tina will allow Argentina tomaintain payments for current gas shipments, as well as clear up pastarrears in 1989, as envisioned under the IMF program. This assumption iscrucial to the balance of payments forecasts. Without continued payment byArgentina under the terms of the present agreement, it will be impossiblefor Bolivia to achieve the improvement in the current account deficit orthe growth rates envisioned under the moderate growth scenario.

270. Likewise, it is difficult to anticipate whether a new agreementwill be reached after expiration of the present contract in 1992. Finaldecision on the contract will depend on developments in the internationaloil market, economic conditions in Argentina, and political relationsbetween the two countries. In order to satisfy domestic demand withoutrelying on Bolivian gas shipments, Argentina will have to raise domesticgas prices to encourage conservation and new investment by the privatesector. Even if the Argentine Government decides to do this, it willprobably take a number of years before gas production increasessubstantially. Therefore, we assume that a new agreement is reached in1992 to continue natural gas exports to Argentina through 1997. However,given the present difficulties with obtaining Argentine payments and thepotential for increased domestic gas production in Argentina, it appearsless likely that the new contract would have the same terms as the presentone. Some conmbination of lower price or lower volumes of sales might beexpected. To reflect the greater uncertainty over the terms of any newcontract between the countries, we arbitrarily reduce the 'olume of gasexports from Bolivia by half from 1992 onwards. It must be noted that thisassumption has an enormous impact on the forecast results, and isparticularly uncertain.

-93-

271. Brazil Gas Project. Given the prospect of losing a significantshare of export revenues from gas sales to Argentina, the BolivianGovernment has been exploring the possibility of exporting products madefrom natural gas to Brazil. A recent agreement between the two countriesprovides for the construction of a petrochemical complex to produce ureaand polyethylene, the installation of a gas-fired thermoelectric plant inBolivia to produce 450 megawatts of electric power, and building of a gaspipeline from the Santa Cruz area to Puerto Suarez (near the border withBrazil) where the urea and power plant will be located. The total cost ofthe investments would be US$1.2 billion.

272. A number of questions concerning the the feasibility of theproposed projects must be answered before work can proceed. Whileagreement in principle has been reached by the two countries, many of thedetails have yet to be ironed out. Flurther, it is not clear whether therewould be a sufficient market for the products, the size of thepetrochemical projects may not be large enough to produce at a costcompetitive in today's market, and the infrastructure requirements of theprojects would be considerable.

273. In addition to technical considerations, the projects haveimportant implications for the economy. The required investments would beenormous in Bolivian terms, more than three times the size of the publicsector investment program in 1988. Even if construction and managementwere largely contracted to private firms, the project (particularly theurea and polyethylene plants) would constitute a significant drain onscarce managerial talent in the Bolivian Government. There is somequestion whether such a highly capital-intensive project with fewemployment or other linkages with the Bolivian economy would be worth sucha price. Most seriously, unless considerable commitments could be obtainedfrom private investors, it appears that Bolivia would shoulder the bulk ofthe risk by guaranteeing the necessary finance, most of which would beborrowed at close to commercial terms. Should the projects fail (eitherbecause of unexpected technical difficulties, unfavorable changes in marketconditions, or political or economic instability in either of the twocountries), Bolivia would still have to repay this debt, a large portion ofwhich would be owed to preferred creditors and hence not be subject torescheduling. The risks involved for Bolivia are thus substantial.

274. A number of possibilities exist for reducing the risks to theBQlivian public sector, which the Government is presently exploring. Giventhe great technological and financial requirements of the urea andpolyethylene plants, it would be sensible to find p--ivate investors willingto shoulder tie bulk of the risk. The public sector could still share inthe profits, both through normal taxes and through selling the natural gasused as an input to the plants. In addition, it may be worthwhile toinvestigate having Brazil provide some of the investment resourcesnecessary for the power plant. This would reduce the risk to Bolivia andhelp ensure the success of the power plant by providing Brazil more of anincentive to continue its participation.

275. In the moderate growth scenario, we assume that the Bolivianpublic sectcr builds the power plant (and related natural gas pipeline) tosell electrical power to Brazil. However, owing to the risks involved, the

-94-

urea and polyethylene plants are excluded from the projections. Even thepower plant by itself would have a significant impact on the Bolivianeconomy, raising investment by US$.50 million a year from 1990-93, andincreasing export revenues by over US$100 million (in 1988 dollars) from1993 on.

276. Imports. The official import data show a decline of US$53million in merchandise imports in 1988 compared to 1987, despite the risein GDP growth. This may reflect a reduction in stocks; in this view, therapid increase in imports in 1986-88 was partially due to speculativepurchases in anticipation of a reversal in the policy of low tariffs andelimination of quantitative controls undertaken in the NEP. With increasedconfidence in the maintenance of a liberal trading environment, thespeculative demand for import goods fell. Alternatively, the decline inimports may reflect a switch from formal sector to contraband imports inresponse to the closer scrutiny of prices introduced by the procurementagents. This creates a dilemma for the medium-term projections, as usingthe official base year may provide too optimistic a picture of balance ofpayments prospects. On the other hand, is is preferable to rely onofficial statistics where possible. We have therefore adopted thefollowing convention. We will use the official estimates for 1988, butadjust the import levels upward in 1989 to partially compensate for theunusual fall in import demand recorded in 1988. Even so, the share ofimports in GDP in 1989 remains well below the level reached in 1987. Thisadjustment explains the rise in the current account deficit projected for1989.

277. Over the medium term, we asssume that Bolivia achievesconsiderable success in reducing its reliance on imported goods. Importvolumes are projected to rise by 4.6Z a year from 1989-97, above the rateof growth of GDP. The overall import elasticity of 1.3 masks differentimport behavior between capital and consumer goods. Given the heavy importcontent cf investment, it is likely that capital goods will increase at thesame rate as investment, or about 7Z a year from 1989-97. To achieve somenforeign exchange savings despite the rapid increase in capital goodsimports, we assume that consumer goods increase less rapidly than GDP.Slow growth in consumer goods imports is due to the rise in domesticsavings and some switching between domestic and foreign consumer goods asadjustment to the new set of incentives makes domestically-produced goodsmore competitive. Achievement of the implied foreign exchange savingsgreatly depends on maintenance of an appropriate exchange rate policy.

278. External Lending. Lending from official donors plays a crucialrole in the Bolivian adjustment program. Over the long run, Bolivia shouldexpect to finance the bulk of investment through domestic savings.However, savings are presently very low, and it will take time for thesavings rate to recover. Therefore, high levels of foreign assistance willbe necessary over the next few years to finance investment until domesticsavings can increase. External savings (the trade deficit) is projected at2.6Z of GDP in 1989. Of course, capital inflows (net of amortization) will

-95-

be even higher, as a portion of external funds must be used to coverinterest. The current account deficit is thus expected to be almost 102 ofGDP in 1989.31

Table 5.4:Modrate Growth Scenrio:

Balanc. of Paywmnts Pro *cUons(In Millions of USE)

1987 18e8 1969 1996 1991 1992 1997

Current Account Balance -531.3 -396.3 -463.6 -661.6 -678.8 -761.7 -515.4

Net Official Transfors 166.e 117.0 120.0 123.6 126.0 129.6 126.0

Direct Foreign Investment 16.0 36.6 85.0 6 5.0 62.5 71.1 164.6

Net LT Lending 161.1 114.6 Z86.3 343.0 841.7 369.7 211.3Gross Disbursements 215.6 264.0 441.4 689.0 596.6 646.0 419.6Amortization 280.6 169.4 156.1 246.6 263.9 280.3 208.3of which: paid 64.3 88.8 39.6 112.6 122.9 124.3 169.3

not paid o.0 86.6 68.6 133.4 131.0 156.6 39.6

Other Flows (Net) /a -69.7 -116.4 -81.3 -39.3 -40.0 -29.0 0.1

Change in Reserves 78.2 44.0 -26.0 -26.0 -6.2 i6.8 -27.7

Exceptional Finance fromBilaterals 240.7 207.1 124.6 204.8 193.5 204.2 62.1

Required Rollover ofAmortization 0.0 80.6 68.6 133.4 131.0 156.0 39.0

Required InterestCapitalization 240.7 128.6 68.0 71.4 82.8 48.2 13.1

I/ Tncludes short-term, private non-guaranteed, errors ano omissions, and unpaid gas receiptsfrnm Argentina.

279. The ratio of the current account deficit to GDP averages about12.5X in 1990-92 (see Table 5.2), reflecting a variety of influences on theBolivian economy. First, the deficit rises owing to the large capitalgoods imports necessary to construct the natural gas pipeline and powerplant envisioned in the recent agreements with Brazil. Second, investmentin the economy as a whole is projected to increase as a result of greaterconfidence, further raising the need for savings before domestic savingscan rise sufficiently. Finally, some increase in external capital isprojected in 1992 to compensate for the anticipated cut in Bolivia'snatural gas exports to Argentina (befoze sales of power to Brazi. come onstream).

280. The rise in the current account deficit is reversed following1992. Export revenues rise and imports decline with completion of thepower project, and the continued rise in domestic savings permits a greatershare of investment to be financed internally. Together, theseimprovements reduce the current account deficit from 12.7Z of GDP in 1992

31/ An additional reason for the difference between the ratio of externalsavings to GDP and the current account deficit to GDP is that theformer is calculated at cor.stant prices, while the latter is at currentr.rices.

- 96 -

Figure 5. 1:

DEBT TO GDP RATiO

120 -

100=PR 80 -CEN 60 -TAE 40 -

20

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997

CURRENT ACCOUNJT DEEOMT TO GDP BAE,

PERCENTAGES14-

12 -

10

8

6

4-

2

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997

-97-

to 5.62 by 1997. The fall in capital inflows (relative to GDP) will alsopermit a decline in debt and interest owed in comparison to GDP. Chart 5.1provides a graphical illustration of the ratios of the stock of debt andthe current account deficit to GDP anticipated in the projections.

281. Table 5.4 shows how the projected deficit is financed. Higherexternal lending in the early 1990s comes from a sharp rise in assistancefrom official sources. US$150 million a year (from 1990-93) of thisincrease represents loans to finance the power plant, US$200 million a yearis provided through the rescheduling of debt service payments through theParis Club, and the remainder reflects project disbursements. Totalexternal capital falls after 1993. While project disbursements ondevelopment loans remain stable, lending for the power plant ends anibilateral debt service payments eligible for rescheduling fall. However,the decline in the availability of foreign capital does not reduce growth,as domestic savings will have risen sufficiently to finance projectedinvestment levels.

282. The moderate growth scenario provides a relatively somber view ofprospects for the next decade, as living standards would not improve overcurrent, depressed levels. However, this scenario does envision continuingper capita growth, and would facilitate a more rapid increase in growthfollowing 1997. By that year, the economy would have achieved the rise insavings and investment necessary for sustained development, while reducingdependence on external borrowing. The contrast between the Bolivianeconomy by 1997 and today would be great. The current account deficitwould have fallen below 6Z of GDP, compared to 9X today. While the debtstock is projected to decline onl.y marginally in relation to output, debtwould equal four times the value of export revenues in 1997 compared tomore than five times today. A more diversified export base (both in termsof products through higher nontraditional exports, and markets through theBrazil power plant agreement) would have reduced Bolivia's vulnerability toexternal shocks. Most importantly, high levels of investment would havebuilt the capital stock necessary to achieve even higher rates of growth inthe next century.

283. Policies. Commitment to the basic macroeconomic framework put inplace by the Paz Estenssoro Government, and adoption of the further policyimprovements discussed earlier in this report, are essential to achievingthe projections of the moderate growth scenario. Economic stabili:y and anappropriate exchange rate level geared to maintaining the competitivenessof domestic production are prerequisites for the increased production andimproved efficiency envisioned in the moderate growth scenario. An opentrading system with low and uniform tariff rates, no quantitativerestrictions, and continued unrestricted access to foreign exchange willencourage higher export growth in nontraditionals and ensure an efficientallocation of resources among productive activities.

284. Future Governments will need to carry through initiatives begununder the Paz Estenssoro Government to achieve the rise in investmentenvisioned in the moderate growth scenario. Vigorous implementation of thefinancial sector reforms discussed in Chapter II will increase theefficiency of financial intermediation, facilitate a decline in interestrates necessary to raise investment, and help encourage the public to keeptheir savings in Bolivia. Modifications to present laws, particularly the

-98-

draft legislation prepared by the past Government but not passed byCongress, will be required to improve incentives for private sectorinvestment. The investment code would provide stable rules of the game forinvestors and facilitate foreign investment through streamliningregistration procedures. The mining and hydrocarbons codes would providefor joint ventures of Bolivian state enterprises with private investors,essential to obtain foreign capital to develop Bolivia's minerals andhydrozarbons resources. The improved land surveys and tax systemenvisioned under the mining code would also spur development in thatsector.

285. Future Governments will face tremendous challenges in achievingthe improvements in public sector administration envisioned in the moderategrowth scenario. Higher tax revenues will be required to finance thepublic sector investment program without recourse to credit from theCentral Bank, and better tax administration will be essential to raiserevenues. The extension of the system for controlling higher-incometaxpayers to Cochabamba and Santa Cruz, the expanded audit program, andplanned improvements in the efficiency of the customs service must becontinued by the next Government. Progress will have to be made inimplementing a more rational salary structure, reorganizing key publicsector entities, and improving the operations of the regional developmentcorporations. Implementation of the privatization program would reducethe drain on the budget from poorly-managed state enterprises and improvethe efficiency of Bolivian production. Work begun on improving the controlof investment expenditures, removing bottlenecks to successful completionof projects, and reallocating investment resources to ensure consistencywith a coherent set of sectoral strategies must be continued to achieve anefficient public sector investment program. Perhaps the most challengingquestion facing the next Government will be how to design adecentralization strategy which results in more efficient provision ofpublic goods, improves equity, and avoids a disruption of health andeducation services.

286. The policy framework inaugarated by the Paz Estenssoro Government,plus the improvements discussed above, should provide an efficientstructure of incentives and inspire the confidence essential forachievement of the moderate growth scenario. Given this policy framework,official lenders should he willing to finance a considerable portiol. of theprojected investment levels for the next few years, until domestic savingscan increase sufficiently and Bolivia's reliance on external financedeclines. Achievement of the growth rates envisioned under the moderategrowth scenario will require considerable good luck, difficult to acceptrestrictions on consumption growth, and enormous efforts by all levels ofBolivian society and by official donors. However, the achievement ofsustainable growth is well worth the risks and efforts involved. Anillustration of alternative outcomes for the Bolivian economy provided bythe slow growth scenario might put matters in better perspective.

III. Slow Growth: Consequences of Poorer Policies

287. The slow growth scenario illustrates the consequences for growthand consumption of failure to continue and improve on the economic policiesof the Paz Estenssoro Government. We assume that the following policy

-99-

reforms suggested in this document are not implemented: a) the investment,mining and hydrocarbons codes are not passed into law; b) improvements intax administration and enforcement are not adopted, particularly the newaudit system and a reform of the customs service; c) certain improvementsin public sector administration begun under the Paz Estenssoro Governmentare not maintained, including the system for controlling public sectorexpenditures (SAFCO) and better monitoring of public sector investmentexpenditures; d) no progress is made in privatizing state enterprises; ande) the restructuring of the financial system is not aggressively pursued,so that banks continue to carry highly overvalued assets on their books,the expected reorganization of the public sector banks does not occur, andcollateral requirements are not enforced. However, we continue to assume(as in the moderate growth scenario) that inflation is kept under controland that the basic policy reforms of the NEP are not reversed: interestrates and capital flows remain uncontrolled, the liberal trade system iskept, and rules limiting the rnle of the state in wage setting andemployment practices remain in force. Also, the Brazil power plant goesforward as in the moderate growth scenario.

288. Failure to carry through on the policy reforms assumed in themoderate growth scenario would have very somber implications for theBolivian economy. Lack of passage of the investment, mining, andhydrocarbons codes would reduce inflows of foreign direct investment (andreduce access to necessary foreign technology), thus irmpeding developmentof Bolivia's substantial minerals and hydrocarbons resources. Poorer taxadministration would lower fiscal revenues and reduce the funds availableto provide public services and finance public sector investment. Theconsequent deterioration in infrastructure and continued low levels ofsocial services would have serious implications for productivity in theBolivian economy. Failure to implement the proposed privatization of somestate enterprises would mean losing the benefits of higher efficiencyexpected in private sector production and a continued drain on the budget.Delays in the restructuring of the financial system would keep interestrates above levels assumed in the base case. More generally, lack ofimprovements in the incentives structure would impair confidence and makeit less likely that domestic savings or investment would increase to thelevels projected in the moderate growth scenario.

-100-

Table 5.5: Slow Growth Seonario

(percent)

1901 1919 1909 1991 1992 1t90

Constant Price.

GDP 2.3X 5.51 1X.1 2.5X 1.3A 2.21Consumption per Capita -5.51 -J.9X 1.9X O.9. 6t.5 -1.31Investment/GDP /I 11.6x 13.21 12.51 12.7X 12.71 12.5XDomestic Savings/CDP 90.X 10.6X 9.4X 9.7x 7.91 12.1XExternal Savings/GDP 2.3X 2.6X 8.1X 2.81 4.6X 9.3x

Current Prices

Current Account/GDF /a -90.1 -9.7X -10.41 -10.ex -10.SX -6.1XPublic Debt/GDP 96.61 0s.81 93.7X 94.7X 95.9X 83.5X

/a High investment and current account to GOP ratio* reflect expenditures underproject to sell gas to Brazil. Investment data includes change in stocks.

289. It is of course difficult to precisely quantify the impact ofparticular policy choices on growth. To provide an idea of the sensitivityof the economy to policy choices, the slow growth scenario includesrelatively small reductions in savings, investment, and efficiency comparedto the moderate growth scenario. We represent the reduced efficiency ofinvestment by a rise in the ICOR from 4 in the moderate growth scenario to5 in the slow growth scenario. It is assumed that domestic savings fallsby 3 percentage points of GDP (on average from 1990-97) compared to thelevels achieved in the moderate growth scenario. Thus, savings rises from9? of GDP in 1988 to 12Z in 1997 in the slow growth scenario, compared to15Z of GDP in the moderate growth scenario. Lending from official donorsalso falls owing to reduced commitment on the part of Bolivia to promisedreforms and lack of necessary counterpart funds to obtain loans. Lower netlending would be reflected in some combination of reduced willingness toreschedule old debt, reduced commitments, or slower disbursements fromexisting commitments owing to failure to comply with loan conditions. Forthe purposes of illustration, we assume that net disbursements fromofficial sources are 202 below the level in the moderate growth scenario.Private direct foreign investment also falls by 202 compared to themoderate growth scenario, owing to failure to pass improvements to theincentive structure recommended above. Lower domestic and external savingsmean that investment reaches only 131 of GDP by 1997, compared to 16Z inthe moderate growth scenario.

290. Reduced irnvestment in turn implies a decline in the rate of exportgrowth compared to the moderate growth scenario. The impact of reducedinvestment on exports is difficult to forecast. Production for export isrelatively more investment-intensive than for domestic use, given Bolivia'sheavy reliance on the capital-intensive hydrocarbons and mining sectors forexport revenues. On the other hand, it is likely that the Government would

-101-

give first priority in allocating its resources to production for export.Given the uncertainty involved, we have adopted the assumption that theshare of exports in GDP (in constant prices) is unchanged from the moderategrowth scenario.

291. It is important to emphasize that these levels of domesticperformance and foreign assistance are not extreme, compared to theexperience of other countries. Despite reduced savings rates relative tothe moderate growth scenario, savings rates must still rise substantiallyin the slow growth scenario. In a survey of major World Bank borrowers,the average ICOR for the 1973-80 period was found to be 4.7, compared toour assumption of S in the slow growth scenario. Bolivia continues toattract sufficient foreign capital to run a trade deficit (while many otherLatin American countries must reduce imports sufficiently to achievesubstantial trade surpluses), despite reduced lending compared to themoderate growth scenario. The assumptions for savings, investment, andefficiency have such drastic consequences for growth in Bolivia because ofthe severity of present economic problems, not because the assumptions areso much worse than average performance in developing countries. Putanother way, the results of the moderate growth scenario were onlyachievable by policy performance far superio- to the average.

292. Results. Lower levels of investment and reduced efficiency lowerGDP growth to 2Z per year over the forecast period. This implies a furtherdecline in consumption per capita of about 1Z a year. The fall inconsumption would be even greater, except that savings rates fall from thelevels of the moderate growth scenario. Even so, by 1997 living standardswould be 30? below their 1980 level. Although external borrowing fallsrelative to the moderate growth scenario, slower exports and output growthresult in no improvement in debt indicators compared to the moderate growthscenario. Debt equals 84Z of GDP in the slow growth scenario in 1997 andinterest payments take up 15Z of exports, about the same as in the moderategrowth scenario. Thus, despite reduced access to external savings, Boliviawould remain with a debt burden approximately the same as that projected inthe moderate growth scenario.

293. These results have a strong downside risk because of thedifficulties in managing a deteriorating economy. Recall that the basicNEP policy framework and relatively low inflation rates are assumed tocontinue even under the slow growth scenario. However, there is somequestion as to whether the present macroeconomic framework would survivecontinued declines in per capita consumption. The view of many observersof the Bolivian economy is that it would be impossible to maintain the NEPreforms and avoid destabilizing increases in public expenditures in theface of a continued fall in living standards. Failure to achieve therecommended policy reforms would result not only in somewhat lower growthrates, but in a return to something resembling the economic chaosexperienced in 1980-85. This extremely sobering possibility illustratesthe crucial importance of implementation of the policy changes discussed inthis report and embodied in the moderate growth scenario.

-102-

IV. Accelerated Growth Scenario: Faster SuDDpl Response

294. Given the lack of improvement in living standards in the moderategrowth scenario, it is important to explore what would be necessary toincrease GDP growth. Barring an unexpected terms of trade gain, higherrates of GDP growth could be achieved through: a) a more rapid supplyresponse by the private sector to the adjustment program; b) more rapidthan expected improvements in public sector administration; and c)continued sales of gas to Argentina at the same level (in volume terms)through 1997, compared to the 502 cut in gas export in 1992, which isLssumed in the moderate growth scenario. The former would facilitate amore rapid increase in savings (from both external and domestic sources)than projected in the moderate growth scenario, permitting a more rapiddecline in interest rates and a rise in investment. Increased efficiencyin public sector administration would result in faster and more productiveimprovements in infrastructure, and reduced barriers to private sectordevelopment. These improvements would raise the efficiency of investment,enabling a rise in GDP growth rates.

295. The accelerated growth scenario illustrates the assumptionsnecessary to achieve a growth rate of 6Z per year over the forecast period.Domestic savings rises by three percentage points of GDP above the level inthe moderate growth scenario, and external savings contributes anadditional one percent of GDP, due to increased inflows of direct foreigninvestment plus a rise in official lending (because more efficientadministration permits faster disbursements). Higher investment permits arise in export growth to 8Z a year, compared to 6.52 in the moderate grwothscenario, led by rapid growth in non-tin minerals and agriculturalproducts. In addition to increased resources, better administrationfacilitates a decline in the ICOR, from 4 in the moderate growth scenarioto 3. Higher GDP growth implies an increase in per capita consumption,which averages about 2.5Z over the forecast period.

Table 6.6: Accolerated Growth Scenario

(percent)

1988 1989 1990 1991 1992 1997

Constant Price,

GDP 2.81 3.65 6.8X 8.4 8.O1 6.61Consumption per Capita -5.51 -0.8% 1.1% 3.3X 3.0X 2.6%Inve-tment/GDP /I 11.16 13.2% 20.0% 19.6% 19.51 17.7XDomestic Savings/GDP 9.31 106.6 13.6X 13.31 11.71 16.91External Savings/GDP 2.31 2.6x 7.1X 6.41 80.6 2.6X

Current Prices

Curront Account/GDP /a -9.0X -9.71 -14.61 -13.16 -13.6X -.0.6Public Debt/CDP 96.61 96.81 511 96.2X 96.7X 79.6x

/a High investment and current account to GDP ratios reflect expenditures under

-103-

296. While this scenario represents a more optimistic view of theBolivian economy than in the other two scenarios, even here the improvementin living standards would be modest. Despite fairly rapid growth inconsumption, by the end of the forecast period per capita consumption wouldremain slightly below the level achieved in 1980. However, more rapidgrowth would provide the resources necessary to improve significantlyhealth and educational services. Such improvements could make a dramticcontribution to achieving a more equitable society, as well as raising theskills and health of the Bolivian labor force, a prerequisite to sustaininghigh rates of growth over the long term.

297. It is important to note that much of the improvements assumed inthe accelerated growth scenario are outside of the control of theGovernment. The key element of this scenario is the rapid increase inconfidence in the stability of the adjustment program, so that privateagents respond quickly to the more liberal macroeconomic frameworkinaugarated under the NEP. Experience with adjustment efforts in otherdeveloping countries indicates that it can take considerable time forpolicy reforms to have a marked impact on savings and investment, and thuson growth. However, it is certainly possible that the adjustment processcould be smoother in Bolivia, thus achieving economic performance similarto that presented in the accelerated growth scenario.

V. Conclusions

298. These scenarios highlight the extreme challenges facing Boliviaand the policies necessary to achieve acceptable rates of growth. Despitethe markedly improved economic policies undertaken since 1985, the severestructural problems faced by the Bolivian economy will continue toconstrain growth through the end of this century. Even the minimumachievement of avoiding further declines in per capita consumption willrequire continued policy improvements, strong support from official donors,a favorable international environment, and considerable luck. Failure topress on with policy reforms could have disastrous consequences, furtherreducing living standards and perhaps leading to a reoccurrence of theeconomic chaos experienced prior to 1985.

299. The scenarios provide some insight into other questions of greatimportance to the Bolivian economy. Barring a major, positive terms oftrade shock or a more rapid than expected supply response, Bolivia probablywill not be in a position to service loans at commercial terms within theforseeable future. Despite relatively optimistic export projections,highly concessional terms on lending, and restrained import growth, theexternal capital required to achieve the GDP growth rate in the moderategrowth scenario results in a stock of external debt equal to over 802 ofGDP by 1997. All donors should attempt to provide assistance at highlyconcessional rates; we would advise the Bolivian Government to reduce tothe extent feasible reliance on loans with interest rates above 3-4Z.

300. The projections show the need for bilateral creditors to rolloveramortization and capitalize interest on loans incurred before 1986.However, the projections indicate that even with most new loans beingprovided on concessional terms, the debt to GDP ratio will remain very

-104-

high. Therefore, initiatives for debt relief on bilateral loans should beexplored. This would improve Bolivia's capability to service the newbilateral loans implicit in the projections. Also, debt relief wouldimprove the overall investment climate by reducing uncertainty about thefuture and the need for future taxation of investment income. Thus, debtrelief would facilitate the recovery of investment necessary todevelopment.

301. Finally, the fact that achievement of acceptable rates of growthwhile improving debt service indicators is within reach of the Bolivianeconomy underscores the remarkable progress made in economic policy since1985. The Bolivian economy in July of 1985 faced hyperinflation, seeminglyunending declines in GDP, chaos in the workplace, and depressed livingstandards. Despite the short term costs of some policies, the NEP hasbegun a recovery in production and set in place the prerequisites forsustained growth. However, the improved policy performance achieved since1985 is not sufficient to ensure a viable future. Continued policy reformand strong support from official donors are the essential elements of astrategy which would achieve acceptable rates of growth, reduce Bolivia'sreliance on external capital, and build the base for faster growth in thefuture.

- 105 -.

Annex I

WORLD BANK ECONOMIC AND SECTOR WORK ON BOLIVIA

Memorandum on Financing Requirements and Public Investment (BO 86-4),December 1986

Updating Economic Memorandum (6455-BO), December 1986

Transport Sector Strategy Paper (6882-B0), July 1987

Population Health and Nutrition Sector Memorandum (6965-BO), February 1988

Regional Development Strategy for the Eastern Lowlands (7158-BO),April 1988

Updating Economic Memorandum (7278-BO), June 1988

A Review of the Public Investment Program and Financing Requirements,1987-90 (7248-BO), June 19588

Banking Sector Study (6765-BO), November 1988

Resource Mobilization in Bolivia Under the New Economic Program,November 1988

Export Corridors (7298-BO), February 1S89

- 106 -

Annex I1Page 1 of S

ASSUKPTIONS FOR THE MHDIU( TERM PRO'JECTIONS

1. Introductioi. The projections are based on the following sources:a) data for 1987 and 1988 are taken from World Bank data, officialstatistics and estimates developed by the IMP; b) projections for 1989-90are designed to match the IMF program, although minor differences with theIMF forecasts exist owing to differences in definitions and data sources;ane c) projections for international commodity prices atd interest ratesare taken from the Bank's International Economics Departmnent.

2. National Accounts. For 1988-90, projections of CDP and investmentare taken from the .UMF program. In the longer term, GDP is assumed toincrease at a rate sufficient to avoid a decline in per capita consumption,on ave.age over the forecast period. However, GDP growth is reduced in1992 to reflect the halving of gas exports to Argentina on expiration ofthe present contract. Fixed investment is determined by assuming anaverage ICOR of four. The change in stocks is assumed to equal the averageincrease with respect to GDP (about 2Z) during 1970-78, the most recentperiod of relative economic stability in Bolivia.

3. Trade. Projections of export volumes are our best judgment ofprospects, based on maintenance of a competitive exchange rate and theLiacroeconomic framework underlying the New Economic Policy. Recovery ofCOMIBOL in the short term, and greater private sector participation in thelonger term, result in a rapid increase in minerals exports other than tin.The assumied stagnation in tin exports, in which Bolivia is a relativelylarge supplier, is based on projections of market growth done by the Bank'sCommodities Division. Argentina is assumed to make all payments due underits natural gas contract with Bolivia and to clear up all arrears.However, the expiration of the contract in 1992 greatly increases theuncertainty of projections beyond that date. We have assumed that a newcontract provides for a 50 cut in Bolivia's natural gas exports toArgentina. The rapid expansion of soya exports is based on estimates ofthe land available for increased soya production. Given Japanese interestin buying the Bolivian soya crop, we have assumed that Bolivia will be ableto sell as much soya as it can produce. Other nontraditional exports areassumed to increase at 10 a year.

4. Short-term projections for import demand are extremely uncertain,owing to the sharp fall in import volumes in 1988 indicated by officialstatistics. Some reason exists to suspect that underevaluation of importsincreased in 1988, although the decline may also be due to the presence oflarge stocks of import goods after the rapid increase in imports in 1986-87(See Chapter 4). We have assumed that imports recover in volume terms in1989 (as anticipated in the IMF program), but the share of imports in GDPremains below the level reached in 1987. Thus, the import elasticitiesshown in Table A2.1 reveal a strong decline in imports relative to GDP in1988, and subsequentlv a sharp recovery in 1989. For the medium term,increased savings results in a relatively low growth rate of consumer goods

- 107 -

Annex IIPage 2 of S

imports relative to GDP. However, there is probably less scope foreconomizing on imports of intermediate inputs and capital goods, so theyincrease at roughly the same rate as investment during the forecast period.

5. All price projections are taken from the Bank's InternationalEconomics Department. The price of Bolivia's natural gas exports isexpected to rise at the same rate as energy prices in the internationalmarket, as specified in the September 1987 agreement with Argentina. It isassumed that the prices of non-factor services, all import goods, andnontraditionals (other than soya) will increase at the same rate asinternational prices for manufactured goods.

6. Balance of Payments and Debt. Short-term projections for grantsand loans from official sources are taken from donor commitments at theConsultative Group meetings in November 1988. This results in a sharp risein disbursements (including grants) from official sources, to US$560million per year in 1989-91. We do not believe that this level issustainable in real terms over the next 10 years. More likely, this highlevel of lending represents a redirection of donor assistance to Boliviaowing to the extreme nature of the economic crisis. Still, we do notexpect that donors will sharply reduce their assistance either. To reflectthe likelihood of some moderation in official assistance, developmentassistance from official sources average about the same from 1992-97 as in1989-91, implying some fall in real terms. In addition, US$150 million ayear from 1990-93 is provided at close to commercial terms for the projectto sell electrical power to Brazil. We assume that net flows of short termand private nonguaranteed debt are zero throughout the forecast period.

7. The assumptions made for the rescheduling of public debt arespelled out in Chapter 4. Bolivia reschedules all debt service payments toParis Club creditors owed on debt contracted prior to 1986. However,payments on loans contracted in 1988 or later are made as scheduled.Bolivia is assumed to retire the remainder of its commercial bank debt(owed or guaranteed by the public sector) at the same terms as receivedunder the buy-back agreement reached in 1988.

Table A2.1TRADE VOLlUES AND PRICES

1987 1988 1989 1990 1991 1992 1992-97Exports USS Millions ------------percentage change per annum at constant prices----------------__________________

Tin 68.8 8.9X 20.5% 6.2% 6.6X -1.31 -0.8XOther Met. A Min. 40.0 60.7% 8.eX 9.6% 5.5% 7.7% 13.3X

11-4 4 Hydrocarbons 256.0 3.6X -1.5% 13.7% 0.9X -42.7% 13.4X0 Soya 19.2 -14.4X 23.7X 16.0X 14.7X 20.90 17.8%X j Other Nontraditionals 87.1 -7.0X 1.9X 12.3X 18.7X 10.9% 18.9%

Mon-Factor Services 129.3 -7.2X 10.8X -0.8X 6.4X 6.o% 4.0%

Total Exports 625.3 7.6% 7.0X 11.1X 5.8X -11.9X 9.8%(less realization costs)

Import Elsticities____________…__ _-- _lasticiti-s…

Consumr Goods (wrt COP) 126.2 -0.3 -1.6 0.3 0.5 0.8 1.0Raw Mat A Int Goods (wrt COP) 320.6 -9.4 -1.0 1.0 1.0 1.0 1.0Capital Goode (art Invost.) 306.9 -0.1 1.6 1.0 1.9 1.0 1.0Mon-Factor Services w(rt CDP) 153.3 -3.2 0.4 -3.6 0.8 0.8 0.8

Total Imports (Growth) 906.9 -13.2% 8.6x 24.0% 3.2% 1.5X 2.0X

Trae Pricoe…______________ ----------------percentage change per annum an constant prics …----------------

o Tin - 3.0X 18.6X -2.1X 6.e% 5.3X 6.1XOther Met. A Hin. - -1.6X -12.2X 4.3X 3.9% 2.0% 3.09Hydrocarbons - -17.9X -19.1X 8.8X 6.4X 11.6X 6.6XSoya - 23.1X 4.4% -9.3X 3.7% 4.9% 2.4%Other Nontraditionsls - 4.0% 6.2X 3.1X 4.2X 3.7X 3.8%Non-Factor Services - 8.3% 3.7X 3.7X 3.7% 3.7X 3.8X

Total Exports - -3.0X 3.9X -1.4X 4.9% 9.9X 4.31

Import Prices - 8.3X 3.7% 3.7X 3.7X 3.7% 3.8X

Table A2.2

:s un P9t.IVIA - NATIONAL INCOME ACCOUNTS 1987-97

o 1987 1968 1989 1990 1991 1992 1992-97

§T <US1 Millionse---------------prcontage change per annum at constant prices-----------------

k CDGP 4266.7 2.81 3.6S 4.41 4.2X 1.8% 4.0X

Conasumption 4089.6 -2.7X 2.0X 2.1X 3.8% 3.9% 3.0%

Public 417.1 5.91 6.01 2.1X 2.7X 3.0X 3.9X

Private 3872.4 -3.81 1.61 2.2% 4.0% 4.0% 2.91

Gross Investmant 1/ 428.6 18.3X 18.3X 42.0X 3.2% 1.61 2.0X

Public Fixed Inv. 280.0 30.21 -7.8x 56.11 0.0% 6.08 -1.3X

Private Fixed Inv. 182.4 16.9% 13.3X 0.0X 10.0X 5.0X 7.81

Change in Stocks -33.8 77.51 -152.0% 197.41 4.3% 1.71 4.11

Export WIES 664.6 4.7X 7.6X 8.9X 6.7% -9.0X 8.8%

Imports ONFS 906.9 -13.2% 8.8X 24.0X 3.2X 1.5% 2.08

Level

GDP Deflator 1.0 15.21 16.5% 7.6% 9.7% 16.4% 9.4%

World Inflation 1.0 8.3X 3.7X 3.7X 3.71 3.7X 3.8X

LIBOR (X) 7.2X 8.0X 8.5X 8.5S 7.7X 7.7X 7.71

Exchange fate (Be/ 3) 2.1 14.41 11.51 4.6% 4.6X 4.9X 4.4X

Table A2.3BOLIVIA - BALANCE OF PAYMENTS 1987-1997(in millions of USS)

^4 u1987 1988 1989 1990 1991 1992 199 1993 1994 1996 1996 1997

0

x un Resource Balance -258.2 -177.9 -203.0 -420.1 -427.4 -s03.6 -283.3 -431.8 -273.7 -288.1 -239.8 -203.40 0 Exports GNFS 647.7 671.1 745.7 807.0 893.1 888.3 1338.7 1071.3 1183.9 1307.9 1469.7 1850.8

< 00 I ports CNFS 906.9 849.0 948.7 1227.1 1320.5 1389.9 1620.0 1502.9 1457.6 1576.0 1709.3 1854.3

Factor Service Receipts 26.6 18.0 15.0 17.0 18.0 19.0 22.0 20.0 21... 22.0 23.0 24.0

Factor Service Payments 319.6 256.4 298.5 284.5 297.3 307.2 353.3 323.4 346.8 357.7 380.8 377.9LT Interest Due Public Sector 303.3 203.4 229.5 215.9 227.8 235.2 269.4 248.4 288.1 274.1 272.1 284.3of which: paid 62.6 78.9 171.5 144.5 165.1 187.1 285.2 212.9 239.2 251.8 258.5 271.2

Other Interest 1/ 4.4 48.0 54.0 48.6 44.5 41.9 38.9 40.1 38.6 38.6 38.6 38.6Other Factor Service Psyments 12.0 5.0 15.0 20.0 25.0 30.0 45.0 36.0 40.0 45.0 60.0 55.0

Not Private Transfers 20.0 20.0 23.0 26.0 28.0 30.0 39.2 32.0 38.0 42.0 42.0 42.0

Current Account Balance -531.3 -396.3 -463.5 -681.6 -678.7 -701.7 -575.4 -703.1 -581.4 -581.8 -535.4 -515.4

Net Official Transfer. 106.0 117.0 120.0 123.0 128.0 129.0 132.6 133.0 140.0 135.0 130.0 125.0

Direct Foreign Investment 16.0 30.0 35.0 65.0 62.5 71.1 118.7 84.4 100.2 118.9 135.6 154.5

I Net LT Lending to Public Sector -65.1 114.6 285.3 343.0 341.7 369.7 222.5 364.4 159.4 159.8 217.8 211.3o Gross Disbursements 215.6 284.0 441.4 589.0 s95.6 640.0 474.3 817.6 427.6 441.1 465.8 419.8Amortization 280.6 169.4 156.1 248.0 253.9 280.3 251.8 2S3.2 268.2 281.3 247.9 208.3

of which: paid 64.3 88.8 89.8 112.6 122.9 124.3 149.0 122.2 138.2 150.3 168.9 189.3not paid 216.2 80.6 66.5 133.4 131.0 156.0 102.8 131.0 132.0 131.0 81.0 39.0

Other Flows (Net) 1/ 36.3 -110.0 -84.4 -39.0 -40.0 -29.0 -4.4 -22.0 0.0 0.0 0.0 0.0

Unpaid Gas Receipts -96.0 -a6. 123.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Change in Reserves 78.2 44.0 -20.0 -25.0 -E.2 26.8 -20.0 -23.3 0.9 -5.2 -44.7 -27.7

Exceptional Finance 456.9 207.1 4.5 204.8 193.8 204.2 125.9 188.5 180.9 153.3 96.7 52.1Unpaid Amortization 216.2 80.6 68.5 133.4 131.0 156.0 102.8 131.0 132.0 131.0 81.0 39.0Unpaid Interest 240.7 126.6 68.0 71.4 62.8 48.2 23.1 36.5 28.9 22.3 15.7 13.1Paymnts to Oil Companies 2/ -120.0

Unidentified Finance 0.0 -0.4 0.1 -0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1

1/ Includes short term, private non-guaranteed, and IMF.2/ Payments in foreign currency made to two oil companies under an agreemnt reache$ in 1988.

- 1.11 -

Annex IIIPage 1 of 3

ODeration of the Emergenc, Social Fund

1. The Emergency Social Fund (ESF) has an excellent record in boththe speed of disbursing funds and the quality of the projects funded. Thisannex provides some details on ESF management which was not relavent to theevaluation of the economic impact of ESF (Chapter V).

2. Proiect Cycle. The project cycle in ESF essentially consists ofproject promotion, initial evaluation, appraisal, and final approval. (seeFigure t. 3.1 for a graphical presentation, and Figure A 3.2 for 3norgariizational chart). ESF evaluates its desired allocation of projects interms of the regions served, the type of project, and the type of financingavailable. According to this master plan, ESF promotes the submission ofprojects in target areas. Once the project proposals are received from thesponsors, who can be local, regional or central governmental institutionsand non-governmental organizations (NGOs), the projects are checked forcompleteness in design, are reviewed against the programing targets ofESF, and are passed on to the appraisal department. During appraisal avisit is made to the project site, cost estimates are checked against alist of pre-set unit costs, the beneficiaries are checked for consistencywith ESF's target population, and potential benefits are estimated. If theproject meets all criteria for acceptance, it is presented to the board ofdirectors for approval and is passed on to the legal unit for signing ofthe contract.

3. Each contract is a four-party agreement, ESF signs as financier ofthe project, the sponsoring agency as sponsor, a supervising agency(suggested by the sponsor and approved by ESF) as supervisorl , and acontractor as the implementing agency. Most contractors are small privatecontracting firms, often only one or two-person operations, or in the caseof assistance projects an NGO. During implementation, the supervisor hasthe technical responsibility for the quality of works. ESF evaluatesimplementation and approves further disbursements based on progress.

4. Project Promotion. It was quickly realized that outreachactivities would be essential to generate projects. Initially, theexpectation had been that many unfunded projects existed in private andpublic institutions, and that projects would be brought forward forappraisal once ESF activities were publicized. However, it became clearthat the ideas and plans of local institutions were a long way fromprojects that could be appraised. Many proposed projects were poorlydesigned or not efficient in their use of funds. Some institutions hadideas for projects but lacked the technical expertise necessary to presenta project which would meet the scrutiny of the ESF (or that ofinternational donors). Thus, most of ESF's initial projects were sponsored

1/ If the sponsor has the necessary technical capacity the sponsor canalso be supervisor.

- 112 -

Annex IIIPage 2 of 3

by Government agencies in urban areas. The statd had the necessaryexperience in preparing project proposals, and Government organizationswere better-organized and more efficiently run in urban areas than in thecountryside. The nature of these early projects led to the notion that ESFhad an urban bias and that it was primarily designed to support Governmentorganizations.

5. In response to these problems, the ESF instituted an outreachuniit, referred to as the project identification department, which has beenvery successful in promoting .aew projects. ESF staff have tiavelled to allparts of the country to help co=nunities and other potential sponsorsdesign projects. By late December 1988, it was estimated that eight out often projects presented to ESF was appraisable. cmopared to a ratio of cnein ten in 1987. While initially a shortage of fundable projects existed,presently ESF has a large bank of suitable projects for which there is nofinancing, totalling more than US$200 million dollars. The outreach efforthas also helped spread knowledge of ESF to remote areas of the country, andthe technical assistance provided local entities in preparing projects hasbeen a tremendously useful institutional development device.

6. Supervision. After more projects entered the implementationphase, serious problems in supervision emerged. The institutionalcapacities of many supervising agencies were poor. In several projects thesupervising agency never showed up at the site, increasing theresponsibility of the contractor. Howevei, Lhe contractors were mostlysmall businesses and often rather inexperienced, and substantial sub-contracting took place, generally endangering good implementation. Thisproblem may have been exacerbated by little concern among contractors forreputation effect, due to the short term nature of ESF and procurementrules (set by IDA, among others) which limit the total amount eachcontractor can implement.

7. Ev.dence of poor supervision caused serious concerns for projectquality. ESF hired an experienced engineer to do an independent technicalaudit of a sample of projects and the supervision department hired severaltechnical specialists to help the regional supervisors with specialproblems. ESF also found it necessary to pay a fee to some supervisingagencies to enable them to free up staff to supervise the projects as wellas to pay for travel expenses to the often remote project sites. Thesupervision responsibilities are now an integral part of the contract.With the extra specialists the quality of work is under better control.

8. Over the life of ESF there has been a marked shift in the focus ofthe institution from employment generation towards longer term development.There has been a slow move to different types of projects and slightlydifferent appraisal criteria. Early in the life of ESF, the institutionwas primarily concerned with quickly disbursing to projects with a highlabor content. This contributed to the urban bias, as a lot of urbanimprovements (such as street paving) were undertaken because of theirrelatively high employment content and simple technical nature. Later

- 113 -

Annex IIIPage 3 of 3

these projects were phased out, and economic infrastructure projects likeirrigation, drainage improvements to protect roads, and flood controlreceived a higher priority.

9. In the first six months of active life of ESF very few socialassistance projects were undertaken. This was probably due to severalfactors besides the focus on employment generation: (i) worries about thecomplicated nature of these projects; (ii) the skill profile of ESF staffemphasizing engineering and economics; and (iii) the lack of suitablesponsors, as the NGOs were hesitant to participate. To increase theemphasis on social assistance, staff with more exDerience in che socialareas were hired, and special. social assistance units were created in eachof the operational departments. These changes have increased the emphasison social assistance projects, and have also helped to convince the NGOsthat ESF indeed was serious about social assistance. The shift in emphasisis also apparent in a shift in appraisal criteria. Concerns likesustainability, cost recovery and long term social benefits have becomemore important, as compared to the intitial focus on quick implementationand low cost per employee-month generated.

10. Efficiency of Operations. ESF has become a true mass producingunit with limited inputs. Each week projects worth about US$2 million areapproved and a similar amount is disbursed. There are around 1000 projectsunder execution at any point in time financed by at least 10 differentinstitutions. All this is administered by a staff of less than 80professionals; including support staff, total staff is around 130. Ahighly-efficient management information system has facilitatedadministration with only limited personnel. ESF is particularly proficientin producing statistics, and can immediately print out exactly whatprojects are being financed by any specific donor, the budget for theseprojects, level of implementation, and expected benefits. This ability hasboth helped to control operations and to raise funds through convincingdonors that their money is being spent effectively.

- 114 - Annex IVPage 1 of 23

TABLE 1.1IREAL EFFECTIVE EXCHANGE RATE

(19=0 a 100)

1 9giG

October 86.339November 84.676

December b'.917

1967

January 83.438February 83.236March 82.169April 80.487May 79.474June 78.961July 79.783August 79.438,ept.Wber 77.401October 7.8306November 73.946December 71.896

1968

January 71.086February 72.024March 70.804April 72.753May 73.646June 74.225July 77.092August 60.084September 78.709October 77.762November 76.075December 76.262

Source: NMF

-115 - Annex IVPage 2 of 23

TABLE 1.2:

CONSUMER PRICF INDICES

(oercent ctange)

---------------- Base 1S80------------------Portiod Bolivia Argentina Bruzil Chile Peru

'.986

January 32.96 3.02 16.71 2.69 5.16

February 7.9t 1.70 12.66 0.10 4.22March 0.07 4.63 7.76 1.48 5.26

April 3.S9 4.73 '.11 1.40 4,05

May 0.97 4.03 (.8V 0.72 3.36

June 4.26 4.66 (.61 1.33 3.56

July 1.78 6.76 0.58 1.00 4.69

August 0.64 8.78 0.88 0.64 3.97

September 2.28 7.23 0.94 1.62 3.57

October 0.59 8.05 ;..01 1.52 3.96

November 0.11 6.31 2.12 1.41 3.56

December 0.65 4.72 7.46 1.51 4.68

1987

January 2.46 7.68 14.30 1.91 6.57

February 1.23 6.60 14.46 1.81 6.59

March 0.70 8.20 13.31 1.64 6.34

April 1.59 3.37 21.63 2.37 6.59

May 0.34 4.17 25.10 1.62 6.90

June 0.22 8.00 27.16 0.68 4.68

July 0.06 10.16 3.60 1.70 7.33

August 0.99 13.70 6.00 1.43 7.36

Soptember 0.68 11.70 5.70 1.88 7.00

October 2.09 19.10 9.00 1.62 6.80

November 0.28 10.10 12.90 1.62 7.20

December 0.80 3.40 29.64 1.69 8.51

1988

January 0.46 9.10 21.32 0.56 12.77

February 1.96 10.40 17 89 0.37 11.80

March 0.84 14.70 18.88 1.87 22.60

April 4.74 22.06 19.28 0.80 17.90

May 1.41 11.09 17.78 0.50 8.50

June 2.06 18.00 19.53 0.60 8.80

July 3.99 28.82 26.14 0.08 30.96

August 2.40 25.02 20.66 0.82 21.70

Septembe,r 0.37 11.70 24.01 0.88 114 .iO

October 2.08 9.00 27.25 1.50 40.60

Novor,ber 0.19 5.70 26.92 1.90 24.40

December 1.33 6.80 78.79 1.90 41.90

- 116 - Annex IVPage 3 of 23

TABLE 1.3:

PARALLEL EXCHANGE RATES(Bolivian* per local curr-ncy)

23-U u UU=== UuUU= UUUU=== - U S-U ww- .suwU :wss *

ARGENTINA (AUSTRAL) BRAZIL (CRUZADO) CHILE (PES() PERU (INTI)AVG 8JY AVG SELL AVG BUY AVG SELL AVG UY AVG SELL AVG WUY AVG SELL

1987

January 0.9900 1.1300 0.0680 0.0730 0.0084 0.0093 0.0620 0.0940

F-bruary 1.0800 1.1400 0.0630 0.0720 0.0083 0.0098 0.0880 0.0930

March 1.0800 1.1C00 0.0630 0.0730 0.0083 0 0094 0.0o40 0.0560

April 0.9900 1.1200 0.0600 0.0690 0.0094 0.0093 0.0810 0.090

May 0.9300 1.0600 0 0690 0.0660 0.0062 0.0092 0.0660 0.0790

Ju,ne 0.8600 0.9900 0.0510 0.0620 0.0061 0.0092 0.0590 0.0700

July 0.6300 0.9400 0.0860 0.0460 0.0090 0.0089 0.0610 0.0640

August 0.7000 0.6000 0.0280 0.0400 0.0078 0.0088 0.0440 0.0660

September 0.5700 0.6600 0.0280 0.0380 0.0079 0.0087 0.0430 0.0610

October 0.4900 0.56800 0.0260 0.0370 0.0079 0.0068 0.0390 0.0460

November 0.4600 0.6400 0.0270 0.0360 0.0079 0.0067 0.0320 0.0390

December 0.4600 0.5300 0.0260 0.0320 0.0078 0.0086 0.0260 0.0360

1988

January 0.3700 0.4600 0.0210 0.0270 0.0079 0.0088 0.0200 0.027CFebruary 0.3000 0.3800 0.0180 0.0240 0.0076 0.0085 0.0200 0.0280

March 0.3000 0.3600 0.0170 0.0210 0.0074 0.0082 0.0180 0.0240

April 0.2900 0.3300 0.0120 0.0150 0.0076 0.0086 0.0160 0.0210

May 0.26e0 0.3000 0.0100 0.0110 0.0076 0.0085 0.0120 0.0140

June 0.2278 0.2647 0.0067 0.0106 0.0076 0.0084 0.0126 0.0137July 0.2018 0.2261 0.0079 0.0094 0.0073 0.0082 0.0116 0.0130

August 0.1742 0.2062 0.0070 0.0084 0.0074 0.0083 0.0108 0.0122.

September 0.1677 0.1798 0.0047 0.0066 0.0072 0.0081 0.0074 0.0091

October 0.1629 0.1683 0.0041 0.0050 0.0070 0.0079 0.0040 0.0053

November a 0.1652 0.1662 0.0033 0.0033 0.0083 0.0084 0.0037 0.0037

December a 0.1492 0.1492 0.0028 0.0028 0.0086 0.0086 0.0014 0.0014

a Average of buy and sel rates

- 117 - Annex IVPage 4 of 23

TAiBLE 1. 4:

OFFICIAL EXCHANGE RATE

(Bolivianos per locol currency)

USA ARGENTINA BRAZIL CHILE PERU(1US) (Austral) (Cruzado) (Pos*) (Inti)

1987

January 1.9300 1.4676 0.0222 0.0096 0.1837

February 1.9200 1.3820 0.1047 0.0096 0.1372

March 2.0000 1.3046 0.0961 0.0096 0.1342

April 2.0100 1.3094 0.0907 0.0096 0.1364

May 2.0000 1.303o 0.0661 0.0096 0.1357

June 2.0700 1.3003 0.0650 0.0096 0.1336

July 2.0600 1.7470 0.0464 0.0094 0.1 312

August 2.0900 0.9345 0.0443 0.0094 0.1310

September 2.1100 0.8281 0.0418 0.0094 0.1331

October 2.1300 0.6103 0.0396 0.0094 0.1344

November 2.1700 0.8218 0.0364 0.0094 0.1086

December 2.2100 0.6332 0.0335 0.0096 0.1105

1988

January 2.2200 0.5678 0.0280 0.0091 0.0673February 2.2500 0.5683 0.0273 0.0092 0.0o62March 2.2900 0.4646 0.0213 0.0094 0.0094

Apr il 2.3200 0.3973 0.0189 0.0096 0.0703

May 2.3600 0.3615 0.0190 0.0096 0.0712

June 2.3900 0.3037 0.0139 0.0097 0.0724

July 2.4100 0.2736 0.0123 0.0096 0.0730

August 2.3900 0.2006 0.0099 0.0097 0.0724

September 2.4000 0.2013 0.0006 0.0097 0.0727

Octobor 2.4000 0.1964 0.0062 0.0097 0.0096

November 2.4214 0.1884 0.0040 0.0098 0.0048

December 2.4479 0.1831 0.0032 0.0099 0.0049

| TA,E 2.1: 0LIVTA - GRSS 0OISTIC PROOUCT SY SECTOR AT CURRE9T PRICES. 1978-86

1-4 '01r IlIions ol p.soa Bi. ianoa)

10X n

1Q8 1;970 1980 1981 1982 1983 1984 1985 1986 1981 1988(.)

G-oan Oo-eatic Prod.ct at M.P 1; 75,172 90.210 122,946 154.896 402,164 1.387.401 18,590,710 2.150,497.804 8,249,530.167 9.238.458.037 11.2900500,179

Indir-ct Taes on 1.po,ts 2,157 2.343 3.292 4,635 3.694 5),40 55.800 18,049.812 102.031-293 118.477.471 123.413,183

Cro-a Dostic P'gdut t P.O. 1/ 73,005 87.867 119.654 150,211 398,470 1,382.151 18.534,910 2.132,447,992 8,147,498,874 9,119,980,566 11,167,086,956

Ag,i'cltur 14,017 16.431 22.615 27.939 78,897 299,927 4.170,355 592.820.542 2,012,432,222 2.207.035,297 2.702.435.053

moinir~ 11.097 12.477 19,384 19,377 76,108 178.297 1,594,002 198,317.663 562,177,422 611.038,698 748.194.829

Hydroca.bona 3,212 3.339 6.701 10,064 44.629 96,751 1,037.955 130,079,328 431.817.440 483.358,970 591.855,611

Min.,aIa 7,885 9.138 12.683 9,313 31,479 8',547 556.047 68,238,336 130,359.982 127.679,728 156.339.218

9,418 11,950 17,948 23,884 41,839 211,469 2,947.051 294,277.823 1.051,027.355 i.240.317.357 1.518.723.831

Elct6icity CGa and Wat-a 584 615 838 1,352 2.789 6.911 55.605 8.529.792 57,032.492 63.839.864 78,169.609

Conatr.ction 3,504 3,603 4.547 5,708 13,548 37,318 741,396 106,622.400 252,572,465 300.959,359 368,513,871

_ Ce_re- 8,761 10,808 13.162 19.527 70.131 211,469 2.891,446 385.973.087 1.857.629.743 1,988,155,7F3 2,434,424.065

Trno.o,t Stomag and CoMmunicat;on 5.475 7.557 7,299 8,562 21,916 127.158 1,631.072 196,185.215 684,389.905 784,318,329 960,369,482

Finncial So-o.as 10,732 12.653 17.230 21,030 47,0)19 154,801 1 853.491 204,715.007 977.699,865 1,103,517,648 1,351.217,527

Baniing 4,161 5.360 7,179 7,811 19,S25 87.075 1.149.164 155.668.703 554.029,923 629,278,659 770.529.003

0Ona-hiap of Duallina 6.570 7,293 10.051 13,219 27.494 67.725 704.327 49.046.304 423.669.941 474,238.989 580.688.524

COe_naI SoCal and Paraonal S ricesa 1,679 2.460 2.991 4,356 5,977 19.350 18.535 6,397.344 16.294.998 18,239,961 22,334.174

Can-ral Co-.'nant 7,373 8,875 12,923 17.575 38,253 128,540 2.557,818 179.125.631 643.652,411 766.078,368 938.035.308

Do.atic esia... 365 439 718 901 1.992 6,911 74,140 8.509.792 32,589,995 36.479,922 44,668.348

(a) E-at;a._.

1/ ft.p a.ket rc.sa

pp. Pod.uca' Prue. In Bohiis.n ntional aoc'ont th.us .ncod all ind-,'ct t....a .. cpt -.. o,t.

Sou,c-: National Institto of Statistics and staff cat;atea

> ""ABLE 2.2: BOLIVIA - ROMS DOMESTIC PRODUCT BY SECTOR OF ORIGIN AT CONSTANT 1980 PRICES, 1978-19881-4 4-4

0x

c to

to X1978 1979 1980 1981 1962 1963 1984 1986 1986 1987 1988(e)

Cross Doestic Product at a.p. 1/ 124,490 124,656 122,960 124,083 118,674 110,943 115,811 116,446 107,211 109,479 112,627

Ind r.ct Taxes on Importe 3,689 3,238 3,292 3,763 1,069 427 332 927 1,326 1,454 1,230

Gross Dometic Product at p.p. 1/ 120,961 121,418 119,658 120,330 117,684 110,516 110,279 109,618 106,885 108,075 111,297

Agriculture 21,495 22,262 22,653 22,364 23,906 19,981 24,562 26,789 26,634 26,483 26,204

Mining 21,146 19,478 19,407 20,139 19,626 18,614 16,336 14,284 12,662 12,266 14,448

Hydrocarbons 6,976 6,423 6,728 7,072 7,476 6,838 6,869 6,736 6,488 6,664 6,881

Minerals 14,164 13,066 12,679 13,067 12,560 11,776 9,466 7,649 6,694 6,691 7,667

Manufacturing 18,881 18,678 17,974 16,681 14,631 13,863 11,926 16,816 11,038 11,423 12,142

Eloctricity Cos and Water 714 734 806 907 930 938 938 946 987 926 978

Construction 6,681 6,479 4,621 4,068 3,698 3,639 3,666 3,171 2,918 2,896 3,134

Co m rce 13,283 13,622 13,261 14,418 13,464 11,796 11,652 12,110 12,895 13,634 13,421

_ Transport Storge and Comunication 6,688 7,577 7,321 8,174 7,799 7,069 7,266 7,337 7,657 7,971 8,410

Financial Services 17,448 17,466 17,248 16,829 16,308 16,984 15,464 16,102 14,892 16,063 16,164

Banking 7,541 7,496 7,189 6,704 6,126 6,829 6,068 4,716 4,406 4,673 4,643

Ownership of Owellings 9,907 9,969 10,069 10,125 10,182 15,336 16,386 16,386 10,492 10,490 10,611

Comunal Social and Personal Services 6,0e2 6,038 4,881 4,867 4,710 4,710 4,239 3,942 3,904 3,922 3,949

Ceneral Governmnt 11,596 11,283 12,940 13,193 13,749 14,836 16,149 16,643 14,646 15,171 16,027

Domestic services 838 652 e68 687 702 709 710 714 719 726 730

(P) Preliminary figures.

(-) Estiatos.

1/ .p. market prices.p.p. producers price. In Bolivian national accounts those include all indirect taxes except

on imports.

Source: National Institute of Statistics and Staff Estimates.

TABLE 2.3: BOLIVIA - GROSS DOMESTIC PRODUCT BY SECTOR OF ORIGIN AT CONSTANT 1980 PRICES, 1978-1987

(Implicit d flators)

0X 1978 1979 1980 1981 1982 1983 1984 1985 1988 1987 1988(e)C~ F-

<Co@. Gross Domestic Product at mp. 1/ 1.604 0.724 1. 000 1.248 3.389 12.606 188.073 19471.210 78941s.677 84386.683 100,336

Indirect Taxes on Imports 0.604 0.724 1.000 1.248 3.389 12.506 168.073 19471.210 76948.677 84385.683 100,336

Cross Domestic Product at p.p. 1/ 0.604 0.724 1.0W0 1.248 3.389 12.506 168.073 19471.210 76948.877 84386.863 100,336

Agriculture 0.652 0.738 1.000 1.250 3.301 16.011 169.858 22129.262 78813.826 88608.143 107,222

Mining 0.525 0.641 1.000 0.962 3.898 9.579 97.682 13883.902 46807.314 49860.369 51,786

Hydrocarbons 0.460 0.620 1.000 1.423 S.970 14.149 161.107 19313.931 68762.127 73637.869 86,013

Minerals 0.557 0.700 1.000 0.713 2.812 6.926 68.742 9039.387 23303.636 22435.377 20,581

Manufacturing 0.499 0.643 1.000 1.440 2.879 15.254 247.132 27210.165 95219.003 108580.702 126,080

Electricity Gas and Water 0.818 0.838 1.000 1.491 2.999 7.368 59.407 9026.235 s7783.680 68941.538 79,928

Construction 0.526 0.566 1.000 1.407 3.6e4 10.26F 208.660 33624.219 86656.705 103968.328 117,586

2 Commerce 0.660 0.793 1.000 1.354 5.209 17.927 248.150 31872.261 144068.142 146900.826 181,389

Transport Storage and Comunication 0.821 0.997 1.000 1.047 2.810 18.014 226.349 26739.160 90563.703 98396.478 114,194

Finarcial Services 0.615 0.724 1.000 1.260 2.883 9.897 119.936 13565.490 65662.690 73260.151 89,166

isanking 0.552 0.716 1.000 1.166 3.187 15.469 226.749 33008.631 125915.892 137607.404 165,956

Ownership of Dwellings 0.663 0.732 1.000 1.306 2.700 6.563 67.816 4722.348 40380.284 45208.674 S5,246

Comm unal Social and Personal Services 0.336 0.488 1. 000 0.897 1.26C 4.108 4.372 1822.868 4173.924 4660.679 5,656

General Government 0.664 0.787 1.000 1.332 2.782 8.664 168.844 11460.849 43947.317 50496.234 62,423

Domestic services 0.572 0.674 1.000 1.312 2.838 9.747 104.422 11946.487 45326.837 50247.827 61,190

(e) Estimates

1/ m p. market prices

p.p. producers prices. In Bolivian national accounts these include all indirect taxes except on imports.

Source: National Institute of Statistics and staff estimates

TABLE 2.4: BOLIVIA - GROSS DOMESTIC PRODUCT BY SECTOR OF ORIGIN AT CONSTANT 1980 PRICES, 1978-1988(Growth rates)

1979 1980 1981 1982 1963 1964 1966 1966 1987 1988(e)

Cross Domettic Product at e.p. 1/ 6.1 -1.4 0.9 -4.4 -6.6 -0.3 -0.2 -2.9 2.1 2.8

Indirect Tax" on Imports -9.0 1.7 14.0 -71.0 -66.8 -22.2 179.2 43.8 6.9 -12.4Gross Dometic Product at p.p. 1/ 6.4 -1.4 6.6 -2.3 -6.6 -0.2 -0.7 -3.3 2.1 3.8

Agriculture 3.6 1.4 -0.9 6.9 -16.4 22.9 9.1 -4.7 -0.2 -1.1Mining -7.9 -0.4 3.8 -3.0 -4.7 -12.2 -12.6 -16.6 1.6 17.9

Hydrocarbons -7.9 4.7 6.1 6.7 -8.5 6.5 -2.0 -4.6 1.6 4.8

Minerals -7.8 -2.9 3.1 -7.8 -2.3 -19.6 -20.3 -26.9 1.7 33.6

Manufacturing -1.6 -3.3 -7.8 -12.4 -4.6 -14.0 -9.3 2.1 3.5 6.3Electricity Gas and Water 2.8 9.8 12.6 2.6 0.9 -0.2 1.0 4.4 -6.2 5.6Construction -3.0 -30.2 -10.2 -8.9 -1.6 -2.3 -10.8 -8.0 -0.8 6.3

t" Coamerce 2.6 -2.7 6.7 -6.6 -12.4 -1.2 3.9 6.5 6.0 -4.8

Transport Storage and Communication 13.7 -3.4 11.7 -4.6 -9.5 2.1 1.8 3.0 6.6 6.6Financial Services 0.1 -1.2 -2.4 -3.1 -2.1 -3.2 -2.3 -1.4 1.1 6.6

Banking -6.6 -4.1 -6.7 -8.6 -8.1 -10.0 -8.9 -6.7 3.9 1.6

Ownership of Dwollings O.6 6.9 0.7 6.8 1.6 0.5 0.0 1.0 0.0 0.2

Coi_unal Social and Personal Services 0.7 -3.1 -0.6 -3.0 0.0 -10.0 -7.0 -1.0 6.5 0.7

General Government 1.7 14.7 2.0 4.2 7.9 2.1 3.3 -6.4 3.6 -0.9Domestic services 2.2 2.5 2.6 2.2 1.f 6.1 6.6 6.7 1.0 6.6

(-) Estimates

1/ m.p. mrket pricesp.p. producers prices. In Bolivian national accounts these include all indirect taxes except on imports.

Source: National Institute of Statistics and Staff Estimates.

TABLE 2.6: BOLIVIA - GROSS DOMESTIC PRODUCT BY SECTOR OF ORICIN AT CONSTANT 1980 PRICES, 1978-1988(Percentage distribution)

cn> " 1978 1979 1980 1981 1982 1983 1984 1985 1988 1987 1988(-)

0x o

ross Domestic Product at mp. I/ 103.06 102.7X 102.U! 103.1X 199.9! 199.4! 100.3! 100.8! 101.3! 101.3! 101.1X

< Indirect Taxes on Imports 3.0X 2.7X 2.8! 3.1! 9.9s 0.4X 0.3X o.8x 1.3X 1.3X 1.1X

G Cross Domestic Product at p.p. 1/ 199.O! 199.9X 199.OX 100.09 19.9X 199.9X 1O.9! 199.9X 1O.OX 199.9! 190.OX

Agriculture 17.8X 18.3! 18.9X 18.8! 20.3! 18.1! "!.3% 24.5! 24.1! 23.6! 22.8!

mining 17.6X 18.6X 18.2X 18.7X 18.8x 18.8X 14.8X 13.06 11.4! 11.3X 13.6X

Hydrocarbons 5.8! 5.3X 6.8! 6.9X 6.4! 8.2X 6.2X 6.1X 8.1X 8.1! 8.2!

Minerals 11.7X 19.8x 19.8x 10.9X 10.2X 10.7! 8.8! 8.9! 6.3X 5.3! 6.8x

Manufacturing 15.6B 16.3! 16.0% 13.8! 12.4! 12.56 10.8! 9.9s 10.4! 10.6X 10.9!

Electricity Gas and Water 0.6X 0.6X 0.7! 6.8! 0.8! 6.8! 0.8! 0.9X 0.9s 9.9! 0.9!

Construction 6.6% 5.3X 3.8X 3.4X 3.1! 3.3! 3.2X 2.9! 2.8! 2.7! 2.8x

f Commerce 11.9! 11.2X 11.1X 12.0X 11.5X 10.7X 10.8X 11.1X 12.2X 12.65 12.1!

Transport Storago and Communication 6.6X 8.2X 8.1X 8.81 8.8x 6.4! 8.51 6.7X 7.1X 7.4! 7.6!

Financial Services 14.4! 14.4X 14.4! 14.0! 13.9X 14.4! 14.0X 13.8! 14.1! 13.9X 13.8X

Banking 6.2! 8.2! 8.09 5.8! 5.2! 6.1! 4.8! 4.3X 4.2X 4.2X 4.2X

Ownership of Dwellings 8.2X 8.2X 8.4X 8.4! 8.7X 9.4! 9.4! 9.5! 9.9! 9.7x 9.4!

Com_unal Social and Personal Services 4.1X 4.1X 4.1X 4.0! 4.0X 4.3! 3.8x 3.6! 3.7! 3.6X 3.5!

Coneral Government 9.2! 9.3x 10.8X 11.0X 11.7X 13.4! 13.7X 14.3X 13.8X 14.09 13.5!

Domest;c services 0.5! 0.65 0.8! 0.6! 0.6X 6.6X o.6x 0.7! 0.7X 0.7! 0.7x

(-) Estimates

1/ u.p. arket pricesp.p. producers prices. In Bolivian national accounts these include all indirect taxes except on imports.

Source: National Institute of Statistics and Staff Estimates.

TAKE 2.6: OLIVIA - CROSS DOMESTIC PRODUCT BY EXPENDITURE CATEGORY.

(In millions of Pesos Solivisnoo)

1966 1961 1962 19#3 1984 1986 1966 1987 1988(e)

x 0

C CDP at M.P. 122,946 154,096 462,164 1,387,491 18,698,716 2,160,497,904 6,967,175,624 8,311,386,980 10,0651,964,714

*socCa° Imports of Goods and WIS 24,796 33,416 99,797 297,S20 1,866,238 466,216,680 1,793,401,400 2,322,861,440 2,826,662,491

Exports of Goods and WFS 31,521 37,473 141,508 430,246 2,912,386 444,836,161 1,687,236,446 1,830,461,088 2,716,102,063

R-eource Cap 6,726 4,068 41,711 132,726 1,062,148 (23,381,629) (126,161,966) (492,390,352) (111,460,428,

Available Resources 116,220 160,638 360,463 1,264,786 17,638,662 2,173,879,333 7,113,337,479 8,863,776,332 10,163,415,142

Consumption 96,162 124,690 310,677 1,081,747 15,330,018 1,901,060,625 6,723,645,466 8,274,461,632 9,600,691,704

Public 15,904 19,814 42,4t4 113,326 2,396,944 301,536,636 677,384,961 1,173,400,60M 1,334,06M,000

Private 82,268 106,076 268,223 968,422 12,934,074 1,599,613,989 6,046,260,616 7,101,051,532 8,268,091,704

Investment 16,068 26,146 49,776 173,618 2,208,544 272,826,866 389,692,013 629,324,800 562,823,436

Fixed Investmnt 17,514 21,568 52,103 162,784 2,263,841 233,333,100 426,65,638 492,266,692 576,473,038

Public Sector 6,633 14,666 29,699 117,199 1,562,050 174,999,831 222,902,439 260,895,517 326,153,173

Private Sector 6,961 6,902 22,404 45,585 701,881 58,338,846 206,739,213 231,377,746 246,301,673

Changes in Stocks 644 4,680 (2,327) 10,234 (66,297) 39,496,70a (38,966,623) 37,069,108 (12,649,608'

Creoss Domestic Savings 24,784 30,206 91,487 306,744 3,260,692 249,447,279 263,636,068 36,934,448 461,373,010

Net Factor Income (7,260) (9,683) (29,788) (98,463) (1,288,60M) (181,476,800) (696,276,W00) (666,804,112) (693,845,009,

current Transfers 1,487 1,2el 3,976 35,749 438,271 46,482,360 182,106,606 206,070,886 470,006,000

Cross National Savings 19,621 21,684 66,674 243,036 2,410,963 114,453,629 (160,639,334) (413,798,779) 327,526,010

---- --- ---- --- ---- --- ---- --- ---- --- ---- -_---- ---------- ----- - - - -- - - - ------ ----- --------- --- _--- ___________________________________.

p/ Preliminary ; el Estiested. 23 Aug 1989

Source: National Institute of Statistics and Staff Estimates.

TAKE 2.7: lOLIVIA - GOSS DOiESTIC PRODUCT BY EXPENDITURE CATEGORY(In ml ii.ons of 196 Bolivian Posoo)

('4

0--4 0 1960 1961 1982 1983 1984 1966 1986 1987 1988(e)X__cl _

M GP at M.P. 122,960 124,668 118,674 11,943 116,611 110,445 107,211 109,479 112,627., Term of Trade Effect 6 (23) (1,973) 3,284 731 (2,967) (6,924) (8,302) (7,991)

CDY 122,966 124,66 116,761 114,227 111,342 167,478 100,287 101,177 164,636

Imports of Coods and WFS 24,796 28,990 20,839 21,486 17,828 23,122 22,616 26,472 22,080Exports of Goods and IFS 31,621 32,634 31,622 27,786 27,180 24,934 27,943 28,376 29,200Exports TOT adjusted 31,621 32,611 29,649 31,070 27,911 21,967 21,019 20,073 21,209Resourco Gap 6,726 3,644 16,683 6,301 9,362 1,812 6,333 2,903 7,120

2 Availablo Resources 116,224 120,639 107,991 104,642 101,269 108,633 :01,878 1015,76 106,407

Consumption 96,166 99,777 96,829 92,389 90,326 94,088 93,196 94,736 93,442Public 16,908 17,236 16,734 18,106 18,523 19,162 11,927 17,686 18,620Private 62,268 82,641 79,096 74,283 71,803 74,936 76,268 77,160 74,822

Investment 18,068 26,762 12,162 12,273 16,933 14,646 8,683 11,840 11,966Fixed Investment 17,614 17,086 12,149 12,227 11,534 10,476 10,166 10,648 12,906

Public Sector 8,633 11,618 6,926 8,803 7,968 7,866 6,281 6,642 7,348Privato Sector 8,961 6,467 5,224 3,424 3,676 2,619 4,874 6,064 5,649

Changes in Stocks 544 3,677 13 46 (6m1) 4,070 (1,472) 1,194 (1,600)

Cross Domestic Savings 24,784 24,306 22,845 18,674 20,286 16,367 14,016 14,743 19,086Net Factor Inco (7,260) (8,628) (8,943) (7,870) (7,129) (8,476) (7,400) (7,089) (6,164)Current Transfers 1,487 1,024 1,193 2,867 2,426 2,171 2,266 2,224 4,483

Cross National Savings 19,021 17,302 16,096 13,661 16,582 10,062 8,876 9,878 17,384

p/ Preliminary ; */ Estimted. 23 Aug 1989

Source: National Institute of Statistics and Staff Estimates.

TABLE 2.6: BOLIVIA - CROSS DOMESTIC PRODUCT BY EXPENDITURE CATEGORY

(Crowth Rates)

(N> 44 1981 1962 1963 1984 196 1986 1987 1988(o)

_ 0

o _N

a w GDP at M.P. 6.9X -4.41 -6.6x -0.3X -0.2X -2.91 2.11 2.8X

*c T-rm of Trade Effect .. 8338.SX -266.41 -77.7X -5o5.6e 133.4X 19.9X -3.8X

IC4. CDY 0.91 -6.9X -2.1% -2.51 -3.51 -6.7X 6.9X 3.3X

Imports of Goods and NFS 16.91 -28.11 3.11 -17.06 29.7X -2.21 12.71 -13.31

Exports of Goods and NFS 3.2% -3.11 -11.91 -2.2X -8.3X 12.1X 1.51 2.9X

Exports TOT adjusted 3.1X -9.1X 5.11 -10.21 -21.3X -4.3X -4.56 5.71

Resource Cap -47.3X 201.4X -41.01 48.41 -80.6X 194.3X -45.6X 14S.31

Available Resources 3.7% -10.41 -3.11 -3.21 7.31 -6.21 4.61 -1.1X

Consumption 1.6X -4.0X -3.61 -2.2X 4.21 -0.91 1.71 -1.41

Public 8.31 -2.9X 8.2X 2.31 3.41 -6.41 -1.9X 5.9X

Private 0.31 -4.21 -6.11 -3.31 4.41 0.41 2.51 -3.0X

Investment 1l.61 -41.41 0.91 -10.91 33.0X -40.3X 36.41 1.1l

Fixed Investment -2.41 -28.91 6.61 -6.7X -9.21 -3.11 4.81 21.81

Public Sector 36.21 -40.41 27.11 -9.81 -1.31 -32.81 6.9X 30.2X

Private Sector -39.11 -4.4X -34.51 4.41 -25.8X 86.11 2.71 10.91

Changes in Stocks 575.9X -99.61 263.81 -1406.65 -777.21 -136.21 -181.11 -183.81

Cross Domestic Savings -1.91 -8.61 -18.71 9.2X -19.41 -14.31 6.21 29.65

Not Factor Income 10.71 11.41 -12.06 -9.41 18.91 -12.71 -4.21 -13.91

Current Transfors -31.11 16.65 139.41 -16.11 -10.6. 4.11 -1.6X 98.06

Gross National Savings -9.61 -12.81 -10.21 14.91 -36.51 -11.71 11.31 76.0X

p/ Preliminary ; */ Estimated. 23 Aug 1989

Source: National Institute of Statistics and Staff Estimates.

> t&. TABLE 2.9: BOLIVIA - COSS DOMESTIC PRODUCT BY EXPENDITURE CATEGORY" 0 (As percent of CDP in current Bolivianos)X rn

f198 1961 1982 1983 1984 1986 1986 1987 1988(e)

CDP at M.P. 1N.9% 166.91 166.0% 166.6 160.61 166.61 166.1% 166.61 196.6%

Imports of Goods end NfS 20.21 21.6S 24.8X 21.4X 16.61 21.8% 26.7? 27.9% 26.1XExports of Goods and NFS 26.61 24.2X 36.2% 31.0% 16.71 20.7x 23.9% 22.0% 27.0%Resource Gap 5.51 2.61 10.41 9.6% 6.71 -1.1l -1.8% -6.9% -1.1X

Available Resource- 94.6X 97.4X 89.6X 90.41 94.3X 191.1 161.8% 106.9% 101.11

Consumption 79.61 80.6% 77.31 78.0% 8' 5% 88.4% 96.2% 99.6 96.6XPublic 12.9X 12.7% 10.61 8.2% lZ.YX 14.0X 9.7X 14.1% 13.3XPrivate 66.9% 67.81 54.7% 69.81 69.61 74.41 86.51 86.41 82.21

Investment 14.71 16.9% 12.41 12.6X 11.9% 12.7X 6.61 6.41 6.6CFixed Investment 14.2X 13.9X 13.0 % 11.7X 12.2% 10.9% 6.11 6.91 6.71Public Sector 6.91 9.65 7.41 8.41 8.4X 8.11 3.2% 3.11 3.2%Private Sector 7.3% 4.6% 5.6% 3.31 3.8% 2.71 2.9% 2.8% 2.65

Change. in Stock* 0.4X 3.0X -6.61 6.7% -0.3% 1.81 -0.6X 0.41 -O.11

Gross Domestic Saving. 20.2% 19.6% 22.7% 22.0% 17.65 11.61 3.81 0.4% 4.51Net Factor Income -5.9X -6.4X -7.4% -7.11 -6.9% -8.4% -8.51 -7.9% -6.91current Transfer. 1.21X .8X 1.91 2.61 2.4X 2.2% 2.6X 2.51 4.71

Gross National Savings 16.5S 13.91 16.31 17.5X 13.0% 6.31 -2.21 -6.61 3.31

p/ Preliminary ; */ Eati"ated. 23 Aug 1989

Source: National Institute of Statistics and Staff Estimates.

TABLE 3 BOMLIVIA - aA,.A)WE OF PAYMENTS. 1970-1987(M. I o- o1 LISS)

1970 1971 1972 1q73 1974 1975 1976 1977 1978 1979 1980 1901 1982 1983 1984 1985 1986 1987 1988

T,ede Selenc. ~~~~ ~~~~~316 16.1 24 9 30 3 1903 -1306 -255 154 -1399 -1345 2638 -63 0 2502 1660 2329 -694 -1175 -2341 1508

6e.o..te~~~~~~~F06~~~ 191 0 16G44 197.9 2605 5584 4440 9882 6343 6288 7598 942 912.4 8277 7551 7245 623.6 5565 5184 5413

C" la" to CIF~~~ 159 2 169 6 173.0 230 2 366 1 574 6 593 7 618 9 s68o 894 3 76 4 975 4 577 5 589 1 491 6 693 0 674 0 752 5 700 1

allianceof WS -6.2 -5 0 -1168 -12' -87 -6 6 -9 5 -318 -392 -263 -35.6 -470 -269 0 7 -180 -126 -250 -240 -169

> Racfte. to 14.6 16 4 21.6 26 4 49 4 60 8 68 6 70 8 65s0 98 0 61 4 87 2 76 3 97 1 87 5 98 0 117 0 129 3 130 0

0 Paymente 2068 21 4 33 4 388 56S1 87 4 78 1 1026 1242 1243 117.0 1342 1032 9684 1055 1106 142.0 1533 1489

X *:r on Ience -Of -Goods-and -WS 25.6 11.6 13 1 17 9 1363 -137 2 -35 0 -16 4 -179 1 -160 8 226 2 -110 0 223 3 168 7 214 9 -82 0 -142 5 -258 1 -177 7

C (LIFact, e.-ces(Nt) -25 3 -17 2 -21 6 -229 -405 -369 -400 -752 -1174 -1947 -2950 -3953 -4591 -411.0 -4312 -3623 -2795 -2931 -2364

<06mv Neceipte 2.1 4 2 0 2 68a 3.9 8 7 16 8 7 0 5 1 15s6 22 1 22 0 14 4 46 8 36 0 16.0 12 4 26 5 18s0

10.. Intereat 2.1 4 2 0.2 68 3. 9 67I 1689 70 5 1 1.42 14 1 14 5 67 36 9 265 15 0 12 4 265 180

09t,., 0.0 0.0 0.0 0 ~0.0 00 00 00 0 0 82 80 75 77 7 9 75 0 0 00 00 00

Payment. 27 4 21 4 21 6 29 7 44.4 4.3 6 54 8 82.2 122 5 212 3 317 1 417 3 473 5 457 8 467 2 378 3 291 9 319 6 25644

lnte,.et 8 3 126 154s 22 6 281 33 4 46 3 692 1083 1655 2655 348A 412.0 3419 3613 325.5 2456 3076 2514

InWop.t ft.d 6.~3 1268 15.4 22 6 261 33 4 463 692 1063 1653 2810 3312 290.5 3003 2921 1515 127 669 1249

I.6e'..t DeIf'.-d A n A,,,. 0~0 0~0 0 0 0 0 00 0.0 0 0 00 0 0 00 A45 17.2 64 6 25 6 69.2 1740 1'229 2407 1265

Int4ee"tCaptalized Sho,t T*. 0.0 0.0 0 0 0 0 00 0 0 0 0 00 0 0 0.0 0 0 0.0 369 16 0 00 0.0 00 00 00

Inte,p.t Caetelised La" To,* 0.0 0.0 0 0 0 0 00 0 0 0 0 00 0 0 0.2 0 0 0 0 0.0 0 0 0.0 0 0 00 00 00

0th,.,. I/ 19.1 6 6 6 4 7 1 - 3 10 2 10 5 13 0 14 2 46 8 51 6 66.9 61 5 115 9 105 9 52 8 46 3 12.0 5 0

Not P,,v.st T,enwfo,a 1.5 2.1 4.6 4 9 3 0 2 6 4 2 5 0 7 8 12 0 12 7 13.3 16 7 40.2 21 8 14 5 17 0 20.0 20 0

Bele.c. on C,.-,e., Ac~.,t 1.6 -3 3 -3 7 -0 1 146 1 -171 5 -70.8 -86 6 -288 7 -345 5 -54 1 -492.0 -219 1 -204 1 -194 5 -429.8 -405 0 -531l2 -396 1

Net Off.cial T,enefere 2 4 5.0 a 6 10. 10 7 7 a 9 8 14 0 30 0 39 2 46 8 26 3 26 8 56 0 66 7 65 5 82 0 106 0 117 0

Oreoct Po,e.qn Inveeteent -75 S 1 9 -10 6 4.6 258 35 2 19 5 22 0 23 0 37 5 4.3 9 75 6 31.0 6 9 7 0 10 0 10 0 I5s0 30 0

r-. Not C0ffciel N & LT 109.1 74.2 1056 21 8 74 0 107 6 223 4 338 3 251 0 178.9 280.4 205 1 -164 4 53 0 -174 5 -118 1 -40 4 175 7 241 1

CsI 0.eb.weeeento 129 3 95 6 1475 57 3 1295 1724 2949 4.391 3671 3232 4661 454 3 2109 507.5 1598 2006 310.5 4562 4105

- o,tizat, one 20.2 21 4 41 7 35 5 5.55 64 8 71 5 100a6 116 1 14.43 185 7 249 2 375 3 454.5 334 3 318 7 350 9 2805 169 4

LEceptionel Fvnencsne 0.0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 188 1 2942 352 4 378 8 261 6 358 0 359 6 240 7 126 5

0th*' LoneTom Capital 0 0 0 0 0 0 0.0 00 0 0 0 0 07 51 518i -291 5 3 -147 -1 5 -12

Receipte 0.0 0.0 0.0 0 0 0 0 0.0 0 0 0 7 5 1 51a8 3.8 10 4 0 0 1 1 0 0

peymente 0.0 0.0 0 0 0 0 00 0.0 0 0 00 0 0 00 32.9 5 1 147 2 6 12

NatSAo.t Te. Capital -0.4 7 3 -0.4 -1 9 -20 -2. 0 7 374 337 -335 -452 -889 -1475 -1139 1805 24 5 990 00 00

Receipt. 1.3 10 8 3.3 0 0 0 0 0 0 1 8 38 7 97 1 54 9 19 3 14 0 125 4 279 7 568 5 24 5 99 0 0 0 0 0

Poymento 1 7 3 5 3 7 1 9 2 8 2 4 1 1 I 3 63 4 Be 4 64 5 102 9 272 9 393 6 388 0 0 0 0.0 0 0 0 0

Net LIse of Dv Rseoe,Coe -2.0 1.1 1 4 13 2 -45 0 5 -23 5 -11 16 9 0 0 62 5 -6 1 10 6 3 1 -25 5 -12 4 92 6 0 0 -44 0

E.ree end Oe,esoons -32 0 -95.2 -81lI -50.0 -144 1 -27 5 -121 5 -293 5 -177 A -76 4 -397 5 -178 3 49 7 100 2 27 4 179 7 -84 3 12 6 -112 5

Uno.4 A,dA"ntine Coo Receipts 0 0 0 0 0 0 0 0 00 0 0 0 0 00 0 0 00 0 0 0 0 00 0 0 00 -205 -31 9q60 -60

Qene.n..R.se-ase -3 4 9.0 -20 0 1 9 -1052 50 1 -376 -312 1064 148C -958 1586 732 -2885 -1475 56 9 -1104 782 44 0

Nlee Itoop

Ipo,te(F011) 135.2 144.2 142.8 194 0 303 9 50'i 7 517 1 537 9 659 1 873 6 5764 A 38 5 519 3 485 5 409 6 567 2 563.9 631.6 595 4

Coose R."e,.. Level 2/ 60.4 .. 61 7 .. 76 5 88 6 206 175 5 211 3 287 4 2..9 8 259 7_128 5.. 136.4 182 0.. 176_7 17112 228 1_3385.5 .415 0_415 8

22A.9g89

1/Ic,depeymente to pet.,ole,e comoa,es2/ Th. eceerne inconesetehcy boteeen the level of Crose Ree*-* -nd Ch-a.. -e ..... (N.t) n. d~.e ti.. r.d.ct,r of re. .,.tvof th. C.vt-Iu 8-kh tlrgh theo,..so

of eh.,ot t..,. debt of ftE 62 -11 n.n -1961 end P 369 9 ..Il - -v 1983 ,,t. eed... tere d.bt

So,rCe Cent.l Benk of slvs

7A8LE 3 2 BOLIVIA - MEI9CHANOISE. EXPORTS (CIF), 1970 I1988(W 1 on LI~.rn SS)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988

14,.rlI 204.9 173.3 174.1 225.9 387.3 314 2 393 5 491.4 515.0 591 7 638 5 556 1 419 3 347.4 364 0 261 7 206 6 206 9 276 5

in40 Ti.. Mial. 23.9 24.6 32.3 55.6 51 4 74 3 13.38 2201.0 228.0 238.8 265 9 237 3 175.5 190 8 120 5 55 2 12 6 44 3Anti mony I/ 30.9 9.0 9.1 17 4 29.1 17 1 31 4 18.1 16 a 29.6 11.8 14 1 6 7 6.3 1 9 13.7 12 2 3 2 2 010L Alloys12.4 6.8 3.5 3 4 2 7 3 0 2 3 1.5 1,3

00 Ti. Concentrated 102.0 82.0 as 9 98 7 .74.5 120.0 142.1 192 9 172 7 167.6 139.3 77.2 41.0 32 4 57.0 52 7 48.8 56 2 33.2Zinc ~~~~~~14.3 15.3 15.4 26.0 37.7 40.3 39.1 44 7 31.3 42.7 36.7 40.4 38 4 33.4 37.3 29.5 28.0 32 7 60.2

Cov*., 1235 83 868 13 4 1s.0 7.3 6.5 4 1 4.0 3.3 3.5 4.4 3 1 3.0 189 1 7 08a 00 0.1Anti ony .- 14 6 20.2 11 I 10.0 21 0 13.7 12 2 19 6 15.4sit.*r 10.5 8.3 7.6 12.6 26.8 28.5 24.3 30 8 3.3.8 58.3 118.3 71.7 37.1 58.3 21 4 10.2 27 3 33.4 47 0Lead 7.8 6.0 5.7 8.3 11.5 7.7 8.4 12.4 10.7 18 0 14.5 11.5 6.5 4.0 1.0 0.5 5 0 4.2 5.98simmi,th.

0 2Cod . . . . 5.5 7.2 375 61 0Oth*,sI/2 9.3 6.9 37 6.1 15 0 196 32,6 9 5 5 4 91I 1.1 0.9 0.8 1 1 10 2 0.4 0. 1 0 1 0

Flyd'"Ocbons: 13.2 23.9 41.6 67.0 193.1 153 9 167.9 134 8 122.3 149 7 245.1 346.5 398.4 420 1 388.9 374 5 333.7 256 0 217 9C.ommi.. 3/ .. 22.6 3.3 4 5 4.9 0 1 0.1 6.7 3.7Naow,ol 0mg 9.9 18.1 29.2 42,5 54 9 66 8 70.5 105 0 220 9 336.7 381 6 376.2 375 7 372.6 328.7 268 5 214.2PrOop... 0 6 15 0.7 1.6 3.1 5.9 3.7 2 3 0.9 0.2Ill.ts.0 ~ ~ ~ ~ ~ ~ ~ . .0 1 3.4 6 4 4.0 2 4 0.9 0 2

Crud* e Ptrol.. 4/ 13.2 23.9 31.7 46.9 163.9 111,4 112 6 67.4 42.3 4.4.0 . 29.3 8 4 . 4 6 0.8 0 0

Agriciiiltiiral products: 7.6 13.1 16.4 36.8 613 54 4 60.1 74 3 609 97.7 123.8 56.9 52 6 44 1 26 1 31 2 82.0 93.4 69 0C,11tto. 0.7 3.8 7.6 9.7 22.0 16.1 12.0 17.7 14.8 10.6 0 900 Soge' 1.0 1.0 0.4 12.4 21.9 17.4 42.8 22.9 14.2 30.8 51.2 5.7 6.1 12~3 6 6 1.8 5.2 86 4.1

IN ths Y. . . . 72 4.9 5,2 3.2 068 0.8 1 4 8.1 8.1 19 3Colff. 356 3-5 . 5 4.3 7.0 13.1 18.7 16.7 19 7 20.8 15.8 15 5 12-9 6 6 13.8 13.0 11.5 14.9Caste.... . . . 2.8 2.9 2.5 2.2 1.6 2 3 1.4 3.1 6.6 5.40 Map ~~~~ ~~~~ ~~~~ ~~~~~~~~~1.9 4.7 3.2 4.2 2.7 08a 0.5 3.2 1.8 2.0ftat m.sd Cattle 0.5 2.0 18s 3.1 0.2 0.8 2.1 3.0 2.6 3.0 1.3 0.9 0 6 1.2 1.5 1.2 12 8 6.7 2.0Timber 1.9 2.6 3.8 7.7 12.9 11.1 10.0 12.0 12.6 21.7 31.1 16.0 11.6 7.6 6.0 5.8 21.3 30.9 21.0saw. .. . 6.1 3.6 7.4 4.8 1.5 5.3 17.3 19.2 20 3

OtMe rmrductm 3.1 5.7 6.3 6.6 6.8 86s 8~5 18.6 25 7 18.1 32.2 41 4 3,51 10.6 4.6 3 0 21.2 1129 15.616.tela.chesic frodacta . 4,8 86 60 4.9 . . 5.8 00A,tesionisa . 2.5 4 2 4 6 1.1 0.6 0.1 0.3 4.8 1.0 1.7063... 5/ 3.1 5.7 6.3 V6 6 6.6 a 6 8' is18. 25.7 20.8 19.2 28 7 29 1 10.0 4 5 2.7 10.6 11 9 13.9

Total W.cada aorts (CIF) 226.8 216.0 240.4 336.3 650.5 531.1 649,16 719.3 723.9 857.2 1039.6 996 9 90-5,6 622.0 783.6 670.5 643.6 569.2 599 0Freight and Incu"ranco 6/ 37.S 29.6 42.3 77,6 94.1 87.1 81,4 85.0 95 1 97.4 97.4 86.5 77,9 669 59 1 46 9 67 1 50.8 57 7Total Pl6m,chandia so porte (FOB) 1,1.0 166.4 197.9 260.5 5584 44.4.0 588 2 634 3 628 8 759.6 942.2 912.4 827 7 755 1 724 5 623 6 556.5 518 4 541.3

No.-Facto, Se.-ncos 7/ 14.8 16.4 21.6 26.4 49.4 60.6 686 70.8 85.0 96.0 81.4 67.2 76.3 97.1 87.5 98 0 117.0 129.3 130.0Total Eaporta and NFS (FMI) 206.6 202.8 219.5 266.9 606.8 504.6 636 6 705. 713.8 857.8 1023.6 99 6 904 0 652 2 812 0 721.6 673.5 647.7 671 3

P/ Prealiminary ;e/ Estimates. 7A..g89I/ Inclued Antimony concentrates 1970-1979-2/ Incluille alloys 1970-1979. No *nidenica of gold .. port. befo-. 1965.3/ Includes di.eel oil4/ In-lude Gasoline 1910-1979.5/ Inclimdas otbea. .ricultural p,oducts; includoe Goa..Cata9.ans and L..tn.r 1970-1978. Tlr .. r. n..o an y .. p.rt.e before 1980.6/ Differ.ence betasen CIP andi PFM price of *nport..7/ Nomn-factoar eso-ica inclmde shwipment, other transport, other goda. er ..... and rant ma it app..,. ini th. Cantral

Bank of fbi i.ivse balanc.a of paysants. Froa 1979 to 1966 a h.r.l.dosn -n othr goods. se-rn na ..d .. nt a..modo to sam.n thes latter. Frain 166 on th. mooirkal.eta ohm. the a .. ni... citI.o.t fo.torrait.

So.rcesx: Centr'al Bank ef Iloli.is and staff astimatea.

TABLE 3.3 BOLIVIA - MER1CHANDISE EXPORTS (C .). 1970-1988(N10, f .o.to.at 1960 153)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1988 1967 1988

014

> 4 Minermls: 738.8 7M.0 727.9 728.7 747.7 703.8 811.5 812.4 773.4 711.0 636.5 688.9 593.2 490.0 503.2 377.0 483.6 390 6 481 3

1-4 0 Ibtel.Tin Metallic .. 107.7 103. 8 103.6 111.6 118.6 186 3 211.4 267.9 237.4 238.8 314.6 313.4 227.4 262 4 178 9 137 4 29.3 99.3

X %0 Antie.nyIf 24.2 24.8 27.6 31.1 27.4 24.9 36 5 29.0 22.0 34.6 11.8 14 9 7 7 9-5 1.9 10.4 10 9 2 5 1 4A., Ioya .. . . . .. 12.4 9.0 5.5 5.4 4.9 5. -

4C cC Concentrate.CO Tin Concentratod 444.7 375.0 360.2 34. 4 350.1 276.2 3161 304 7 239.8 169 2 139.3 92 5 54.3 42 1 78.2 74.3 144.3 151 9 87 1

0~Zinc 36.9 36.1 31.4 39.4 38.9 39.3 39.0 486 41.0 42.8 38 7 35 5 35.4 32.5 29 2 27 2 28 0 30 9 42 2

Copper 16.6 16.5 17.6 17.3 16.7 12.7 9 a 6.7 6.4 3 7 3.5 5 4 4.5 4.1 2 6 2.6 1 2 TunsIe 3O .3 36.4 36.0 36.1 36.1 35.2 45 1 39 0 39.4 36.5 47.4 42.8 45 1 35.6 34 1 22.7 20 9 13.8 11 6

Antimony ~ ~ ~ ~ ~ ~ ~ ~ ~ . . 14.6 22 1 148a 17.3 19.8 13 8 14.5 20 2 15 2Sii.*, 128 0 115.1 98.2 111.3 120.2 137 0 116 19 140.3 133 3 121 8 118.3 137 2 101 4 106.2 53 8 34.9 106 2 101.8 154.2Leed 2323 21.3 17.2 18.2 17.6 16.0 16.8 18.0 15.0 14 2 14.5 14~2 10.3 8.6 2.2 1.2 11 4 8 7 8.9

0ld1 . 6 5 7 4 32.1 53 00th.r. /2 32.6 29.7 15 7 20.4 29.1 42.1 64 0 14 6 6.8 10 6 1.1 0 8 0.6 1.4 14.0 0.6 1 3 1.4 8 2

tVdroca,orn.t: 161.5 293.0 488-6 581.1 541 5 456.8 449 8 334 1 273.2 235.2 245 1 246.2 265 0 284 3 253.1 241 6 240 3 145 3 151.1Gasoline - .. . 22.6 3 1 5.7 5 4 0.2 0 1Natur.al 0. - 109.2 170.1 167 5 168.7 170 1 177.4 171 2 186.9 220.9 237 5 248.6 240.7 239.2 239.7 239 7 143 7 150.2

Propane - ~~~~~~1.1 2 6 0.9 1.6 2 8 5 3 3.2 2.1 1 0 0 3Sassn. -. 0.1 2.8 5.4 3.3 2 2 1 0 0 3Crude Petroleums SI 161.5 293.0 379.3 411.0 374.1 267 1 279 7 155 5 99 2 47 3 -. .31.7 9.5 11.2 1 6 0 0

Agricultural P,oducts: 211.6 39.0 53.3 91.5 102 9 119.1 138 2 105.9 99 1 154 4 123.8 51 4 67 6 72.9 44 9 49 4 75 0 119.7 127 2

CIN Cotton 3.2 68. 15.6 23.4 33.1 37.1 13.9 14.2 21.8 13.7 0.9cl llues. 8.6 3.4 1.4 332. 23.2 26.4 72 5 58.1 37 6 62.4 51.2 9.7 16.5 24.7 9.9 3 1 8.6 12 8 5 9

- Lmmsh.., 17.13.6 4.9 4.9 66a 217 25 3.7 13 9 16 9 38 7Coffee 14.1 137 16.1 1568 12.6 19 6 19.5 17. 20 6 26 5 20.8 19 4 28.2 25.3 12.6 25.5 19 7 29 5 36.7Coo""*. . . . 3-0 2.9 1 7 2 2 4 2 5 0 4.3 7 1 13.6 10.8Gamea2. 4.0 4.7 2.6 4 3 6 9 5 9 3.9 5 7 2.9 3.1Meet end Cattle 0.4 1.5 1.7 2.1 1. 23 0 9 0 7 08a 2-5 1.3 0~9 0.7 006 1.0 0.6 7 0 3.2 0.9Timber 7.3 11.5 16.4 17.1 31 6 34.7 31.3 15 7 18.3 26.7 31.1 12.2 8 9 8 4 8.0 6.1 17 7 17 7 11.5

say ~ ~ ~ ~ ~ ~ ~ .. . 6.1 2.6 78a 9 1 3.5 8.3 21 3 22 9 1986

Other Products 3.5 28.4 15.6 14.2 15 . 11.1 5.1 9.6 11.1 14.3 32.2 21 3 20 2 12 6 5.4 8.1 16 4 14 5 10 5HMtalmockenic Prod.clio -. .-. . 3.3 6.8 7 7 3.8 0.0A.'tes.nios. . . . 2.1 4.2 3.4 1.3 0.9 0 4 4!f 2~3 0.4 0.7Oth.., 4/ 3.5 26.4 15.8 14.2 15.6 11.1 5 1 9.6 11.1 9.0 19.2 10.3 15-2 11 7 5 0 3.6 14 1 14.1 9 6

Total Merchandise Eapaot. (CIF) 932.2 1122.5 1265.5 1415.6 1407.8 1209.6 1404.6 1261.9 1156.8 1114.8 1039.6 1007.6 946.1 8599 9606.7 676.3 815.3 670 1 770 1Freight and Xncur.nce Ji/ 106.3 60.7 106.3 167.7 168.5 136.7 128.0 121 6 116.1 106.8 97.4 66.1 78.6 69.3 62.3 49.1 76 8 40 5 42.4Total Narchandiso Exinrts (FOB) 618.9 1061.6 1179.3 1247.9 1241.1 11512. 1276.6 1140.3 1036.7 1006.0 942.2 921.7 867.5 790.6 744 3 627.2 738.5 629.6 727.7"me-Factor Seruicee 6/ 41.8 44.7 54.0 54.9 67.4 9.8. 107.9 101.3 105.6 107.5 61.4 868. 77.0 100.6 92.3 102.6 1032 103.0 95.61.5.l E.porte sad WS5 (FOS) 668.6 1066.5 1223.3 1304.6 1326.5 1247.9 1364.5 1241.6 1144.3 1115.4 1023.6 1006.5 944.5 691.2 636 6 729.8 641 7 732.7 823 3

of Preliminary ; * Eatitest..If Include An5ie.ny cencentbrate 1970-1679.2/ Includes *lloy. 3070-1979. MD e,idee of gold e.port.a before 1965.3/ Inc lwde Gasol ine 1970-1979.4/ ln;l..des other. agricultural products; include. Ooin,Cesten. and Loath., 1970-1976. Tb.r. .a.r 8 ..yep.,t. W.fe 1960.8; iff.e'nCe between CIP end PM8 p'ice of *.po'ts.Of Non-factor eer,iceo include shipment. oth., trenepert, other gwoodo..Cr.iee a.. rent as it appears in the Centrel

111na of SOli.im's balanc, of payments. From 1979 to 1965 a br.akdoen 0. otler good.. aevic.a end rent a.... d. to rem,, the letter. Prom 1966 on the ewoo.&eh..t. who. the ma.rnice. aithout fact., rec-ipt..

Sources: Central Sank of l.komi,i end staff e.tie.tes.

TABLE 3.4: BOLIVIA - MECHANDISE EXPORTS (CIF). 1970-1988(Prrce indices 1980e100)

e141970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988

1-40

Minerals.

Tin~ Metalic 22.2 23.7 29.8 49.8 43.3 45.0 63.3 75.0 88.6 100.0 84.5 75.7 77.2 72.7 68.1 40.2 43.0 44.6

1 61 Aati on 127.6 37.0 33.0 56.0 106.4 66.7 66.1 62.4 75.4 85.0 100.0 94.9 86.0 66.4 97.7 131.6 112.0 128.0 141.0

00 Alloys 100.0 75.8 63.8 62.5 54.2 52.1Ca

Concetitrata..Tin Concentrated 22.9 21.9 23.4 26.7 49.8 43.1 46.0 63.3 72.0 88.6 100.0 83.4 75.5 77.1 72.9 70.9 33.6 37.0 36.1

Zinc 38.8 42.4 49.0 66.1 96.9 102.7 100.3 91.9 76.4 99.7 100.0 113.9 106.3 102.8 127.8 108.3 100.1 105.7 142.6

Copper 67.1 50.4 49.3 77.4 95.8 57.5 66.0 61.2 62.9 89.8 100.0 81.3 68.8 72.9 64.6 64.6 64.6

Tungsten 52.9 37.4 27.1 29.2 58.4 63.4 77.1 115.5 100.4 96.3 100.0 100.8 14.9 56.1 56.6 45.4 32.1 36.3 44.0

Anti mony 100.0 91.4 75.3 58.1 106.1 98.9 64.2 97.0 101.2

Silver 8.4 7.2 7.9 11.3 22.3 20.8 20.4 22.0 25.4 47.9 100.0 52.3 36.5 54.9 39.9 29.2 25.7 32.8 30.5

Land 3W.4 28.1 33.2 45.7 65.4 48.1 50.1 68.8 71.4 126.9 100. 80.5 63.4 46.3 46.3 43.9 4.3.9 62.5 65.9

Si soth 107.2

cold 85.1 97.0 117.0 115.1

Others I/ 28.5 23.2 23.5 29,9 51.6 46.6 50.9 65.1 61.6 65.6 100.0 105.9 129.9 80.9 72.9 65.2 61.5 71.4 12.2

*%drocarbons: 100.0 123.1 118.1 119.9 111.0 93.4 78.1

Gasoline 100.0 107.5 79.0 90.9 51.9 51.2

Natural Gas 9.1 10.6 17.4 25.2 32.3 37.6 45.9 56.2 100.0 141.7 153.5 157.1 157.1 155.5 137.1 172.9 142.6

Propane 56.0 53.5 77.9 100.0 110.9 111.0 116.1 107.3 94.8 80.6

Sutano 100.0 123.1 118.1 119.9 111.0 93.4 78.1

Crud. Petroleum 8.2 8.2 8.4 11.9 43.8 38.8 40.3 43.3 42.6 93.1 92.5 88.5 40.7 51.0 0.0

C) Agricultural Products: 26.6 33.6 34.5 42.4 59.6 45.7 56.0 70.2 61.4 63.3 100.0 106.8 78.1 60.4 58.2 63.3 109.3 78.0 70.0

-4 Cotton 22.0 43.0 48.6 41.4 66.5 48.8 86.2 124.3 68.0 77.1 100.0

Sugar 27.4 29.1 28.4 37.4 94.3 65.9 59.0 39.-' 37.8 49.3 100.0 58.3 43.5 49.8 67.2 56.5 60.4 67.0 69.7

1 Leather 53.1 100.0 104.7 46.5 30.7 30.4 38.0 43.7 48.0 49.9

Coffee 24.9 25.5 26.5 37.3 34.1 35.7 67.6 109.3 80.9 69.1 100.0 81.6 59.2 51.1 52.3 54.0 66.1 39.0 40.6

Caetan. 92.2 100.0 153.2 102.3 42.0 46.8 33.2 43.9 48.0 50.0

Goes 47.4 100.0 124.8 96.6 39.0 12.7 13.0 56.0 62.0 64.5

Meet and Cattle 131.8 131.9 105.9 149.4 8.3 62.5 230.3 448.1 315.3 122.4 100.0 105.1 87.8 151.5 150.0 18.0 182.3 209.4 217.8

Timber 2S.9 24.3 23.1 45.0 40.8 32.0 32.0 76.5 68.9 81.3 100.0 147.0 130.6 92.9 75.0 71.5 120.1 175.0 182.0

Say 100.0 139.3 96.0 49.9 43.5 64.4 81.4 84.0 103.4

Other ProductsMoetl mechanic Product. 147.2 100.0 104.4 129.6 0.0

Art.san ies 121.3 100.0 141.3 63.9 71.0 38.3 6.6 210.1 229.5 238.7

Other 21 68.8 20.1 40.0 46.4 56.5 77.7 185.8 196.6 230.9 119.8 100.0 279.0 191.7 85.0 88.6 75.5 74.7 84.2 141.5

Total Merchbandise Exports (CIF) 24.5 19.2 18.7 23.9 46.2 41.2 45.2 57.0 82.6 76.9 100.0 99.1 95.7 95.6 97.1 99.1 78.9 84.9 77.8

Freight andi Incurance 34.9 36.7 40.0 46.4 56.5 62.8 63.6 69.9 80.5 91.2 100.0 100.5 99.1 96.5 94.8 95.5 113.4 125.5 138.0

Total Merchandise Export. (FOB) 23.2 17.9 16.8 20.9 44.8 38.6 44.5 55.6 60.5 75.4 100.0 99.0 95.4 95.5 97.3 99.4 75.4 82.3 74.4

Mon-Factor Services 34.9 36.7 40.0 46.4 58.5 62.8 63.6 69.9 80.5 91.2 100.0 100.5 99.1 96.5 94.8 95.5 113.4 125.5 136.0

Total Exports and WfS (FMS) 23.7 16.7 17.8 22.0 46.6 40.5 46.0 56.8 62.4 76.9 100.0 99.1 95.7 95.6 97.1 98.9 60.0 88.4 81.5

p/ Preliminaery ; o Eistiestes.I/ Inds. of market price. in U.5S terms for mntel. and phosphate rock e.ported by do,oloping countries used from 1963 on.

2/ Index of market price, in US I terse for egricultural raw material. used frme 1983 on.Sources: 1970-7`9 Data: Central Bank for mineral* end for the rest aission estiastas based on noluass and prices aleo given by the Central Bank.

19604611 Data: Central Sank for Bolivia, unpublished eorkaheeta dated April 25,1986.1996647 date: Central Sank worka1heeta dated Dec.23, 1986 and July 27, 1987.

TABLE 3.5: BOLIVIA - MERCHAMISE EXPORTS (CIF). 1970-1988(Real CrGoth Rates - based on 1980 pricea)

1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988

l4inerals: 3,2 -4.5 0.1 2.6 -5.9 15,3 0.1 -4.8 -8.1 -10.2 7.9 -13.9 -17.4 2.7 -25.1 28.3 -19.2 23.2> 4-4 "StaelH 0 Tin Metallic -3.6 4.6 2.8 6.2 39.4 27,9 26.8 -3.9 -7.2 31.7 -0.4 -27.5 15.4 -32 6 -22.4 -78.7 239.1

x 00 Antimony 0.6 13.3 12.7 -12.0 -9.0 45.5 -20.4 -24.1 58.2 -66.0 25.8 -47.9 22.2 -79.5 435.0 4.6 -77.1 -43.3O _. Alloys 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 -27.6 -38.8 -1.2 -10.1 19.0 -100.0 0.0 0.0

C 4 Concentrates-cc t ° Tin Concentrated -15.7 1.4 -9.4 1.6 -20.5 13.6 -3.6 -21.3 -21.1 -26.4 -33.6 -41.3 -22.5 85.8 -4.9 94.2 5.3 -42.6

4 Zinc -2.1 -13.0 25.3 -1.1 0.9 -0.7 24.7 -15.8 4.5 -14.3 -3.2 -0.1 -8.4 -10.2 -6.7 2.8 10.6 36 4Coppor -11.5 8.3 -3.0 -3.6 -23.9 -22.5 -31.9 -5.1 -42.2 -3.8 53.2 -15.1 -10.2 -30.4 -7.0 -54.5 -100.0 0.0

Tungsten 9.2 4.5 0.2 -5.2 -2.6 28.3 -13.5 0.8 -7.4 29.9 -10.0 5.8 -21.0 -4.3 -33.5 -7.9 -34.0 -15.8

Antimony 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0-0 0.0 0.0 51.2 -33.2 17.0 14.9 -30.3 4.6 39.8 -24.7

Siler -7.9 -16.4 15.6 8.0 13.9 -13.2 18.0 -5.0 -8.6 -2.8 15.9 -26.0 4.7 -49.4 -35.0 204.0 -4.1 51.5

Lead -8.5 -19.6 5.8 -3.3 _0a9 4.8 7.4 -16 9 -5.3 1.9 -1.5 -27.6 -16.2 -74.9 -43.7 833.0 -41.1 33.2

Gold 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 14.5 332.9 65.3

Others -8.9 -47.0 29.5 42.7 44.6 52.2 -77.2 -39.9 21.2 -89.7 -22.7 -27.5 120.8 929.0 -95.6 112.0 7.7 485.7

!drocearbons: 81.4 66.7 1869 -6.8 -15.8 -1.3 -25.7 -18.2 -13.9 4.3 0.4 7.7 7.3 -11.0 -4 5 -0.6 -39.5 4.0

Os-el in- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -86.3 84.3 -5.5 -96.5 -27.3 -100.0 0.0 0.0

Natura Ca 55.7 -1.5 0.7 0.9 4.3 -3.5 9.2 18.2 7.5 4.7 -3.2 -0.6 0.2 0.0 -40.1 4.5

Propane 0.0 0.0 0.0 0.0 0.0 0.0 0.0 161.9 -67.9 81.4 71.0 89.3 -39.1 -33.9 -54.1 -69.7 -100.0 0.0

Butane 0.0 0.0 0.0 0.0 0.0) 0.0 0.0 0.0 0.0 0.0 4861.1 96.0 -39.1 -33.9 -54.1 -69.7 -100.0 0.0

Crud- Petroleum 81.4 29.5 8.4 -9.0 -23.2 -2.6 -44.4 -36.2 -52.4 -100.0 0.0 0.0 0.0 -70.2 -100.0 0.0 -86.0 -100.0

Agricultural Products: 36.3 36.7 71.7 12.4 15.8 16.u -23.4 -6.3 55.8 -19.8 -58.5 31.5 7.8 -38.5 10.0 52.0 59.5 6.3

Cotton 177.2 77.1 49.7 41.3 12.2 -62.5 2.3 52.9 -36.9 -93.2 -100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0Sugar -5.9 -59.0 2254.4 -29.9 13.8 '74.5 -19 8 -35.3 66.1 -18.0 -81.0 90.2 33.2 -59.9 -68.4 176.2 48.7 -54.2

Leather 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -64.2 1.4 38.8 -61.2 -5.4 48.2 274.2 21.2 129.1

cn Coffee -2.5 32.3 -12.7 -20.4 55.7 -0.4 -12.4 20.6 38.1 -27.1 -6.7 35.2 -3.6 -50.3 103.0 -22.7 49.6 24.6

-4 Castena 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -5.8 -42.0 32.2 90.7 18.5 -12.5 64.1 93.3 -21.5Co_m 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 17.2 -44.7 65.5 61.6 -14.4 -34.2 46.7 -49.4 6.8

MI at and Cattle 299.5 12.1 22.1 15.5 -46.6 -28.7 -26.6 23.2 197.3 -47.8 -31.8 -23.3 23.2 24.2 -39.1 1022.5 -54.3 -71.3

Timer 57.0 42.7 4.1 84.5 3.8 -9.9 -49.8 16.6 45.8 16.5 -60.6 -27.3 -6.0 -4.6 2.1 117.6 -0.4 -34.7

Say -58.1 203.3 17.5 -61.5 135.8 156.9 7.5 -14.1

Other Products 712.0 -44.6 -9.7 9.5 -28 ^ -53 7 86.3 16.5 28.8 124.7 -33.9 -4.9 -37.8 -57.1 50.0 102.5 -11.6 -27.6

Mataleechanic Products 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 170.2 -13.2 -50.4 -100.0 0.0 0.0 0.0 0.0 0.0

Artouonies 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 103.9 -19.7 -62.3 -30.4 -54.8 1025.0 -48.9 -80.9 61.9

Other 714.3 -44.6 -.. 7 9.5 -28.9 -53.7 86.3 16.5 -19.0 112.9 -46.5 47.7 -22.8 -57.1 -27.8 289.2 -0.1 -30.5

Total Merch,ndlee Eport- (CIF) 20.4 14.5 10.1 -0.6 -8.4 8.9 -10.2 -8.3 -3.6 -6.7 -3.1 -6.1 -9.1 -6.2 -16.2 20.6 -17.8 14.9

Freight and Incuronce -25.5 31.7 57.8 -0.7 -16.7 -7.7 -5.0 -2.8 -9.6 -8.8 -11.6 -8.7 -11.8 -10.1 -21.2 56.4 -47.3 4.9

Total Warcbandime Exporte (FOB) 26.4 13.2 5.8 -0.5 -7.3 10.9 -10.7 -8.9 -3.0 -6.5 -2.2 -5.9 -8.9 -5.8 -15.7 17.7 -14.7 15.6

Nan-Factor Services 6.8 20.8 5.4 53.7 10.7 11.4 -6.1 4.2 1.8 -24.2 6.6 -11.3 30.7 -8.3 11.2 0.5 -0.1 -7.2

Total Exports end NF5 (FOB) 25.5 13.5 5.8 1.8 -6.1 10.9 -10.3 -7.8 -2.5 -8.2 -1.5 -6.3 -5.6 -6.1 -12.8 15.3 -13.0 12.4

…-- - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - -- - -- - - - -- -- - - - - - - -- - - - - -- - - - - - - - - - - -- -- - - - - - - - - - - - - - - - - - - - - - -- - - -- - - -- - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - -- - - - - --

p/ Preliminary; */ Esti_mtes. 23

Aug89

Source: Table 3.3.

TABLE 3.6: BOLIVIA - MERO4M1ISE IIPRTS (CIF), 1970-88 1/> ju4l'- O (Mill !ions of a ... et US56)

x Cl'OJ -C:

ff4 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988

Consumer Goode 32.3 34.2 41.8 45.9 76.3 105.3 102 6 122.1 158.6 184.9 172.2 248.4 100.0 66.9 95.5 133.8 133.7 126.2 134.6

Durable 22.5 24.7 29.9 30.6 49.6 $1.2 59.5 73.1 96.1 73.2 65.6 125.2 40.8 19.6 53.3 46.1 61.9 56.2 65 1

Non Durable 9.6 9.5 11.9 15.3 26.7 44 1 43.1 49.0 62.5 111.7 106.6 123.3 59.2 47.3 42.3 87.7 71.8 70.0 69.7

Rle Materials end Intermediate Goods 59.9 53.4 60.3 78.4 135.4 199.5 186.0 187.3 215.4 287.2 222.8 311.3 221.9 244.2 172.4 252.5 201.7 320.4 243.3

Fuel 1.1 1.2 1.1 1 2 2.4 9.6 9.3 7.7 7.2 9.7 1.8 14.8 9.4 4.2 1.5 2.4 2.7 3.7 3.0

For Agriculture 2.4 2.0 4.3 6.6 6.1 9.0 7 5 7.9 9.4 9.9 11.4 15.7 7.7 11.5 17.6 18.4 15.8 25.0 16W1

For Industry 56.4 50.2 54.9 70.6 126.9 180.9 169 1 171.7 198.8 267.6 209.6 280.8 204.9 228.5 153.4 231.7 163.2 291.7 224.2

4 Capital Coods 66.2 71.1 69.9 104.3 152.7 268.8 303.8 303.9 390.8 415.6 279.8 408.5 243.9 271.2 218 2 280.1 315.2 305.9 322.0

Construction 868 15.4 1:.1 12.2 22.9 42.9 60.6 45.4 41.7 38.1 32.6 52.4 37.0 44.3 31.9 26.3 33.6 38.0 29 8

For Agriculture 3.2 3.0 3.9 7.2 9.7 17.3 13.2 16.5 21.F 16.6 13.8 21.7 5.4 7.5 13.6 27.4 40.1 24.0 26.2

For Induetry 26.4 31.3 30.7 50.6 56.4 98.2 126.3 156.5 216.5 214.1 150.9 207.7 144.7 159.6 104.1 137.1 151.7 155.1 174.9

Transport and Equipment 27.8 21.4 24.2 34.3 63.7 1105. 101.7 85.6 111.0 146.8 82.5 126.6 56.8 59.9 68.6 89.3 89.8 88.8 91.0

Other Goods 2/ 0.8 10.9 1.0 1.6 1.7 1.0 1.4 5.5 3.9 6.6 3.6 7.2 11.6 6.8 5.4 26.6 23.4 0.0 0.0

Total Merchandi-s Imports (CIF) 159.2 169.6 173 0 230.2 366.1 574 6 593.7 618.9 768.7 894.3 678.4 .75.4 577.5 589.1 491.6 693.0 674.0 752.5 700.1

Freight and Insurance 24.0 25.4 30.2 36.2 62.2 71.9 76.6 81 0 109.6 220.7 102.0 138.9 58.2 103.6 62.0 125.8 110.1 120.9 104.7

Total hMachandies Imports (FOG) 135.2 144.2 142.8 194.0 303.9 502.7 517.1 537.9 659.1 673.6 576.4 636.5 519.3 485.5 409.6 567.2 563.9 631.6 595.4

Non-Factor Services 20.8 21.4 33.4 38.8 56.1 67.4 78.1 102.6 124.2 124.3 117.0 134.2 103.2 96.4 105.5 110.6 142.0 153.3 148.9

Total Imports end NFS (CIF) 160.0 191.0 206.4 269.0 422.9 642.0 671 8 721.5 892.9 1018.6 795.4 1109.6 680 7 685.5 597.1 803.6 816.0 905.8 849.0

1/ In Bolivie the three biggest custom offices are not on the frontier, thus thers exists the lAugQ89

so called CIF-Fronter end CIF Aduans values. This is one equivalent to the usual understan

2/ Includes sheat end other sovernment imports.

e/ Estimates exist only for select 1986-1987 import categories.

p/ Freliminary.

Source: Central 8 nk of Bolivia, National Institute of Statiatics and staff setimetes.

TA8LE 3.7: BOLIVIA - I14PORTS IN CONSTAIT 1980 PRICES

-40 (i l lion- of con-tant 1980 US3)

x O

4900

0.. 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1901 1982 1983 1984 1985 1986 1987 1988

Consuer Goods 77.1 80.4 69.6 60.8 105.4 140.1 138.5 151.9 196.7 221 4 172.2 283.7 139.1 93.9 121.1 206.5 164.5 142.7 138.4

Ourobl 64.6 67.2 74.7 66.0 87.9 97.5 93.A 104.5 119.4 80.2 65.6 124.4 41.8 20.3 56.3 48.9 55.5 45.9 49.3

Non Wurable 12.7 13.2 14.9 14.8 17.5 42.6 45.1 47.4 77.3 141.2 106.6 159.3 97.3 73.6 64.0 157.6 109.0 96.8 89.1

Raw Materials and Int reediate Coods 193.7 162.8 165.4 180.2 241.8 329.4 302.0 275.2 275.7 320.1 222.8 308.0 226.4 253.3 182.1 268.1 181.2 262.1 184.3

Fuel 25.3 20.7 17.5 13.8 6.5 26.8 24.4 18.3 17.1 15.9 1.8 13.2 8.6 4.4 1.6 2.7 2.5 3.2 2.4

Other Interediate Goods 168.4 142.1 147.9 166.4 235.3 302.6 277.5 256.9 258.6 304.2 221.1 294.8 217.8 248.9 180.5 265.5 178.6 259.0 181.9

Ca.itnl Coods 189.6 193.5 174.7 224.9 270.3 428.2 477.3 434.7 485.3 455.6 279.8 408.1 250.0 281.3 230 4 297.3 282.9 250 1 243.8

c.,Cfl Other Cood- 2/ 2.3 29.7 2.5 3.4 3.0 1.6 2.2 7 9 4.8 7.2 3.6 7.2 11.9 7.0 5.7 28.2 21 0 0.0 0.0

-4

Total Narchandiee Imerte (CIF) 462.8 466.4 432.2 489.3 620.5 899.2 919.9 869.8 962.5 1004.3 678.4 1005.0 627.4 635.4 539.3 800.2 649.8 655.0 566.5

Freight and Insurance 68.8 69.1 75.5 78.0 110.1 114.5 120.4 115.9 136.1 241.9 102.0 138.1 59.6 107.4 8686 97.0 114.0 108.2 86.5

Total Merchandie Imwrts (FOB) 394.0 397.2 356.7 411.3 510.4 784.7 799.6 753 9 826.4 762.4 576 4 866.9 567.7 528.0 452.7 703.2 535.6 546.7 480.0

Non-Factor Services 59.6 58.2 83.5 83.6 99.3 107.4 122 7 146.8 154.2 136 3 117.0 133.4 105.7 100.0 111 4 117.4 140.8 125.3 112.7

Total Imports and IFS (CIF) 522.4 524.6 515.7 573.0 719.8 100.66 1042.7 1016.6 1116.8 1140.6 795.4 1138.4 733.1 735.4 650.7 917.6 790 4 780.3 679.2

1/ Estimated using PJV only as surrogate for weighted VN &rI Food ind;ci*s. 1Aug89

2/ Eoti_eted using MIN only in place of weighted JV and Fuei indicies.

*/ Estiate

p/ Preliminary.

Source: Central Bank of BSlivia. National Institute of Stat.stics and staff -atiates.

(f) TABLE 3.8: HOI_IVIA - MERCANDISE IMPORTS, 1970-1988

Im.p-t Price Indices (1980 - 100)

l-4 0

C4

<00 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988c

Coenuer COed 41.9 42.5 4t.7 56.8 72.4 75.2 74.1 80.4 80.7 63.5 100.0 87.6 71.9 71.3 78.9 64.8 81.3 88.4 97.4

Durable 34.9 38.7 40.0 46.4 56.5 62.8 63.6 69.9 80.5 91.2 100.0 100.6 97,6 96.4 94.7 94.2 111.4 122.3 132.1

Non Durable 77.4 72.2 80.0 103.5 152.2 103.5 95.7 103.5 80.9 79.1 100.0 77.4 60.9 64.3 65.2 55.7 65.9 72.4 78.2

Raw Materials and Intermediate Cood. 30.9 32.8 36.4 43.5 56.0 60.6 61.6 68.1 78.1 89.7 100.0 101.1 99.0 96.4 94.7 94.2 111.3 122.2 132.0

Fuel 4.3 5.8 6.3 8.7 36.7 35.7 38.2 42.0 42.0 60.9 100.0 112.1 108.7 95.2 93.2 89.9 106.4 116.8 126.1

Other Intermediste OeDO 34.9 36.7 40.0 .6.4 5t.5 62.8 63.6 69.9 80.5 91.2 100.0 100.6 97.6 96.4 94.7 94.2 111.4 122.3 132.1

Cap itl Coods 34.9 36.7 40.0 46.4 56.5 82.8 63.6 69.9 80.5 91.2 100.0 100.6 97.6 96.4 94.7 94.2 111.4 122.3 132.1

Other CoDe 34 .9 3b 7 40.0 46.4 56.5 62.8 63.6 69.9 80.5 91.2 100.0 100.6 97.6 96.4 94.7 94.2 111.4 122.3 132.1

Total Mnrchandia taports (CIF) 34.4 36.4 40.0 47.0 59.0 63.9 64.5 71.2 79.9 89.0 100.0 97.1 92.0 92.7 91.2 86.6 103.7 114.9 123.6

Froight and Insurance 34.9 36.7 40 0 46.4 56.5 62.8 63.6 69.9 60.5 91.2 100.0 100.6 97.6 96.4 94.7 94.2 111.4 122.3 132.5

Total Merchandise Imports (FOB) 34,3 36.3 40.0 47.2 59.5 64.1 64.7 71 3 79.8 88 4 100.0 96.5 91.5 92.0 90.5 80.7 105.3 115.5 124.0

Non-Factor Ser.ice- 34.9 36.7 40.0 46.4 56.5 62.8 63.6 69.9 80.5 91.2 100.0 1OC.6 97.6 96.4 94.7 94.2 111.4 122.3 132.5

Total Imports and NFS (CIF) 34.5 36.4 40.0 46.9 58.7 63.8 64.4 71.0 80.0 89.3 100.0 97.5 92.8 93.2 91.8 87.6 103.2 116.1 125.0

p/ Preliminary; */ Estimtes. 1Aug89

Source: ISM, Coemodity Prie- Oate.

TABLE 3.9: BOLIVIA - MEOIANDISE IMPORTS, 1970-1988

cn (Crr-th Rat. Percsntsges, constant 1980 prices)

> qi1-.40

>C'4

1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988

Cae

104

Consumer Goods 4.2 11.4 -9.9 30.5 32.9 -1.1 9 7 29.4 12.6 -22.2 64.8 -51.0 -32.5 28.9 70.6 -20.3 -13.3 -3.0

Ourable 4.3 11.1 -11.7 33.2 10.9 -4.2 11.9 14. 2 -32.8 -18.3 89.7 -66.4 -51.4 177.0 -13.1 13.5 -17.3 7.3

Non Durable 3.9 13.0 -0.6 18.5 143.1 5.8 5.2 63.1 82.6 -24.5 49.4 -38.9 -24.4 -11.9 143.2 -30.9 -11.2 -7.9

Raw enterials and Interediate Goods -16.0 1.6 8.9 34.2 36.2 -8.3 -8.9 0.2 16.1 -30.4 38.2 -26.5 11.8 -28.1 47.2 -32.4 44.7 -29.7

Fuel -18.2 -15.4 -21.2 -52.7 311.1 -9.0 -25.0 -6.5 -7.0 -88.9 645.9 -34.5 -49.Q -63.8 67.8 -4.9 25.4 -25.3

Other Interadiste Coods -15.7 4.1 12.5 41.4 28.6 -8.3 -7.4 0O. 17.6 -27.3 33.3 -26.1 14.3 -27.5 47.0 -32.7 45.0 -29.7

Capital Coods 2.0 -9.7 28.7 20.2 58.4 11.5 -8.9 11.8 -6.1 -38.6 45.2 -38.5 12.5 -18.1 29.0 -4.8 -11.6 -2.5

Cl-4

I Total Merchandise Iports (CIF) 0.8 -7.3 13.2 26.8 44-9 2.3 -5.5 10.7 4.3 -32.5 48.1 -37.6 1.3 -15.1 48.4 -18.8 0.8 -13.5

Freight and Insurance 0.6 9.2 3.4 41.0 4.1 5.1 -3.7 17.5 77.7 -57.8 35.4 -56.8 80.1 -19.4 12.0 17.5 -5.1 -20.1

Total Merchandise Importa (FO8) 0.8 -10.2 15.3 24.1 53.7 1.9 -5.7 9.8 -7.7 -24.4 50.4 -34.5 -7.0 -14.3 55.3 -23.8 2.1 -12.2

Mon-Factor Services -2.2 43.3 0.2 18.7 8.1 14.3 19.8 5.1 -11.7 -14.1 14.0 -20.7 -5.5 11 4 5.4 19.9 -11.0 -10.1

Total Imports end NFS (CIF) 0.4 -1.7 11.1 25.6 39.9 3.6 -2.5 9.9 2.1 -30.3 43.1 -35.6 0.3 -11.5 41.0 -13.9 -1.3 -13.0

p/ Prelisinary; e/ Estiates. 23Aug89

Source: Table 3.8.

04 0

x r

TABLE 4.1:

ai WBOLIVIA - CCNISWER PRICE INDEX

(1986 = I")

CPI 1976 1979 1969 1981 1982 1983 1984 1986 1986 1987 1988

Januery 376.6 431.9 633.2 866.9 989.2 3,848.4 17,959 9 629,971.8 41,050,237.0 62,493,355.2 56,446,698.6

F.bru.ry 378.9 449.1 636.5 883.6 1,148.1 4,243.8 22,091.9 1,781,384.8 44,314,302.9 53,139,102.0 67,644,001.0

March 374.1 436.1 646.1 981.0 1,241.7 4,744.3 26,781.9 2,225,674.6 44,343,478.3 63,508,884.8 68,027,00.0

April 374.9 434.3 651.7 881.9 1,398.3 5,147.3 43,614.1 2,487,922.3 46,937,004.4 64,361,163.6 80,775,069.8

May 372.1 446.8 871.4 901.4 1,464.5 5,617.0 64,121.1 3,375,313.9 46,381,626.9 54,646,991.5 61,634,686.6

June 386.9 463.7 766.5 902.6 1,523.0 5,796.8 86,736.1 6,623,841.8 48,357,008.1 b4,426,923.6 62,902,999.0

July 419.9 467.4 718.9 916.6 1,851.7 6,388.3 70,184.4 10,017,602.2 49,219,576.e 54,397,849.0 85,414,806.0

August 4e6.1 476.6 742.9 970.6 2,182.8 8,036.1 80,709.2 16,876,225.7 49,533,372.9 54,938,108.2 86,983,068.0

Septembr 409.9 482.3 726.3 986.7 2,687.8 9,353.7 110,835.7 28,098,528.4 60,661,830.1 55,255,270.9 86,736,869.8

October 425.6 490.3 736.2 972.4 2,984.4 16,432.8 176,371.1 25,611,915.8 50,960,568.0 58,409,000.0 88,121,0.090

Nov_mber 425.7 511.7 743.6 971.1 3,650.1 13,018.2 232,025.7 26,431,706.3 56,906,634.1 6e,249,880.8 87,992,006.8

December 427.7 622.1 771.9 964.8 3,825.6 18,392.4 373,293.5 30,873,311.6 51,238,413.0 56,702,000.0 88,896,000.0

Sourco: National Institute of Statistics 22 Aug 1989

MAP SECTION

61~~~~~~~~~~~~~~~~~~6

BO L I VI A,/- , < Mo,. Roods Rel,ef (jetersj Cl.not.c Zonxes

. r t / S-cond.ry Roods 6000 AND A8OJ,E

< - t@s>V ERAt Ra,l,..d, 400U 5000o .~~~o (3, 00 .5OO.,'

-A N D; t DeporhmenI Boundories . ?000 _ lEMPLRAEL10(2000 S.B TOPICA&Inme-rotIonol Bo-ndor,es 0 000 ITROPICAt

* ~ ~ ~ 4' K.,i ~~~~~ B RAZIL

( LA PAZ IB E N E

BOLIVI A-14

T|j .-

-- >\AR TI OCf4N I TRINIDAD I,L A4.4 \ 55'

05,ARGA