Bolivia - Ingavi Rural, Omasuyos-Los Andes Rural, Ulla Ulla Development Projects

Document of The World Bank FOR OFFICIAL USE ONLY Repon No. 7574 PROJECT PERFORMANCE AULLT REPORT BOLIVIA INGAVI RURAL DEVELOPMENT PROJECT (LOAN 1211-BO) ULLA ULLA DEVELOPMENT PROJECT (LOANF 1510-BO/CREDIT 762-BO) OMASUYOS-LOS ANDES RURAL LEVELOPMENT PROJECT (CREDIT 933-BO) DECEKBER 30, 1988 Operations Evaluation Department hs doment bas a resticted di_buton and may be used by recipients only in the perfonnanee of t '"* 'XE,,i. .........  WEIGHTS AND MEASURES Metric System 1 ton (t) - 1,000 kg - 2,205 pounds 1 hectare (ha) - 10,000 square meters - 2.47 acres 1 kilometer (km) - 0.62 miles LIST OF ABBREVIATIONS AND ACRONYMS ASI - Agricultural and Social Infrastructure Division of INFOL BAB - Agricultural Bank of Bolivia BE - State Bank BEP - Bolivian Electric Power COMBOFLA - Bolivian Committee for Promotion of Wool EEC - European Economic Community ERR - Economic Rate of Return FAOICP - Food & Agricultural Organization/World Bank Cooperative Program FOMO - National Manpower Development Service FOTRAMA - Alpaca Wool Processing and Marketihg Company GDP - Gross Domestic Product GOB - Government of Bolivia ICB - International Competitive Bidding IDA - International Development Association IDRA - Institute for Rural Development of the Altiplano IFAD - International Fund for Agriculture Development IFC - International Finance Corporation INBOPIA - National Institute for the Handicraft and Small Industry INFOL - National Institute for the Development of Wool LAC - Latin America and the Caribbean Region, World Bank LCB - Local Competitive Bidding LOALE - EMENA & LAC Disbursements Division, World Bank MACA - Ministry of Peasant Affairs and Agriculture M&E - Monitoring and Evaluation NGO - Non-Government Organization OAS - Organization of American States OED - Operations Evaluation Department, World Bank PCR - Project Completion Report PIL - Milk Pasteurizing Industry PPAM - Project Performance Audit Memorandum PPAR - Project Performance Audit Report PU - Project Unit RD - Rural Development SAR - Staff Appraisal Report SNC - National Road Services UNDP - United Nations Development Programme UPOLA - Project Unit of Omasuyos-Los Andes  FOR OMF AL US ONLY rHE WORLD BANK Washington. D.C. 20433 U.S.A. Ofke of ODictai.G"Wal Opetatmm EAIvuI4M December 30, 1988 IMMORANDUN TO TEE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report on BOLIVIA INGAVI Rural Development Project (Ln. 1211-BO) ULLA ULLA Development Project (Ln. 1510-BO/Cr. 762-BO) OMASUYOS-LOS ANDES Rural Development Proiect (Cr. 933-BO) Attached, for tvnformatlon, is a copy of a report entitled 'Project Performance Audit Report on Bolivia Ingavi Rural Development Project (Ln. 1211-BO), Ulla Ulla Development Project (Ln. 1510-BO/Cr. 762-BO), Omasuyos- Los Andes Rural Development Project (Cr. 933-BO)" prepared by the Operations Evaluation Department. Attacbment This document ha a resuicted distribulon and may be used by reipients only In the perfonmnce of their official dutdase Its contents may not otherwise be disclosed without World fak authoriztion.  FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT BOLIVIA INGAVI RURAL DEVELOPMENT PROJECT (LOAN 1211-BO) ULLA ULLA DEVELOPMENT PROJECT (LOAN 1510-BO/CREDIT 762-BO) OMASUYOS-LOS ANDES RURAL DEVELOPMENT PROJECT (CREDIT 933-BO) TABLE OF CONTENTS Page No. Preface .... ..................................................... i Basic Data Sheets ................................. .. .. . iii Evaluation Suinnary ....... ....................................... xii PROJECT PE&rORMANCE AUDIT MEMORANDUM I* BACKGROUND ... ........... 1 A. Agricultural, Economic and Social Setting .... ...... 1 B. Objectives and Design of the Projects .... .......... 2 Ingavi Rural Development Project ................. 3 Ulla Ulla Development Project ................. ... 3 Omasuyos-Los Andes Rural Development Project ..... 4 C. Board Concerns ............................................... 4 D. Related Projects .......... . . .. . . . .. . . . . .. . . ..*. . . . . 4 _I_ IMPLEMENTATION AND OUTCOMES ........ . . . . . . . . . . . ........ . 6 Ao Ingavi Project *.-*-*.. .................+ --v*--8 ! ~~~~~~~Management ................... ... 000-0-000.......... 8 Achievements by Component ........... 8 Beneficiariees....................-............. 9 Disbursements and C o sts ............... 9 Project Outcome ............. * .. .................. . 10 B. Ulla Ulla Project o..o ..... 10 Achievements by Component ........................ 11 Disbursements and Costs....... .. 14 Project Outcome. .o.. ... ......... .1.4 ...... .. 14 C. Omasuyos-Los Andes ................... . 15 [ Maagemnt .................................................. i 5 Achievements by Componentoooooo ... 00-0-00 ...... 15 Beneficiaries ...... .................... .... . 16 Disbursements and Costs ........................... 17 Project Outcome.... .. ...... ... 17 III. CONCLUSION AND MAIN ISSUES ............................. 18 A. Overview ....* .................................. 18 B. Improper Understanding of the Socio-economic Framework of the Project Areas ... ... .......... 18 C. Project Design and Complexity .. *oooo ............ 19 D. Incorrect Implementation of the Credit Scheme ...... 19 Tlhis document has a restricted distribution and may be used by recipients only in the performance of their offcial duties. Its contents may not otherwise be disclosed without World Bank authorization.  TABLE OF CONTENTS (cont'd) Page No. ATTACHMENTS: I. Comments from Cofinancier (EEC) ......................... 23 II. Comments from the Borrower ............ .......... ....... 25 PROJECT COMPLETION REPORTS INGAVI RURAL DEVELOPMENT PROJECT I. Introduction ............... 29 II. Project Formulation .................................... 30 III. Project Implementation ................................. 32 IV. Agricultural and Economic Impact ....... ................ 38 V. Institutional Performance .... ........ ......... 41 VI. Bank Performance . ......... O-* .... ....... ........... .. 41 VII. Conclusions and Recommendations . ........................ 43 Annexes .*see*** ....................................... 45 ULLA ULLA DEVELOPMENT PROJECT OVERVIEW .... o....... . .......................................... 55 I. Introduction .. ......................................... 67 II. Project Formulation ....... ............................. 69 III. Project Implementation ....... *........................ 72 IV. Institutional Performance .............................. 79 V. Agricultural and Economic Impact ... ................. 81 VII. Bank Performance . ...... . ..0.0. . ... .. ... ............. . 82 VII. Conclusions and Recommendations . **..... ***............ 84 Tables . * ......................... . .87 Annex. ........$.*..I ........................... 89 OMASUYOSILOS ANDES RURAL DEVELOPMENT PROJECT I. Introduction .... ..... .............................. 93 II. Project Formulation .* .. ............................ ... 94 III. Project Implementation ................................. 96 IV. Institutional Perforwance .............................. 100 V. Agricultural and Economic Impact ....................... 100 VI. Bank Performance ....................................... 103 VII. Conclusions and Recommendations ........................ 105 Tables ... . .............................................. 109 Annex .......................................................... 117 MAPS: IBRD 11737: INGAVI Rural Development Project IBRD 12907: ULLA USLLA Development Project IBRD 14079: OMASUYOS-LOS ANDES Rural Development Project  PROJECT PERFORMANCE AUDIT REPORT BOLIVIA INGAVI RURAL DEVELOPMENT PROJECT (LOAN 1211-BO) ULLA ULLA DEVELOPMENT PROJECT (LOAN 1510-BO/CREDIT 762-BO) OMASUYOS-LOS ANDES RURAL DEVELOPMENT PROJECT (CREDIT 933-BO) PREFACE This is a performance audit of three projects designed to develop the rural sector in five provinces of Bolivia's Highlands region, the Altiplano, through on-farm credit, extension, and rural infrastructure. The Ingavi project loan was approved in February 1976 in the sum of US$9.5 million and closed in October 1984 with US$3.1 million (33%) being cancelled. The Ulla Ulla project loan and credit in the sum of US$9.0 million each were approved in January 1978 and closed in December 1986 with US$5 million (29%) being cancelled. The Commission of the European Communities (EEC) provided parallel financing of the equivalent e' S$2.0 million. The Omasuyos-Los Andes project credit was approved In . 1979 in the sum of US$3 million and closed in December 1986 with US$0.9 million (30Z) being cancelled. The three projects have been covered in a single audit as they were specifically poverty-oriented, their implementations ran in parallel, they were located in contiguous geographical areas, and have been affected by the same economic and social frameworks. The audit report consists of an audit memorandum (PPAM) prepared by the Operations Evaluation Department (OED), and three Project Completion Reports (PCRs). The PCR for the Ingavi project was prepared by the Bank's Latin America and Caribbean Regional Office, and the PCRs for the Ulla Ulla and Omasuyos-Los Andes projects by the World Bank/FAO Cooperative Program. The audit memorandum is based upon a review of Appraisal Reports, the President's Reports, the L-oanlCredit and Project Agreements and PCRs. Correspondence with the Borrower and internal Bank documents have been consulted and Bank staff associated with the projects have been interviewed. t OED mission visited Bolivia in April 1988 about 18 months after the loans were closed. Discussions were held with officials of the Ministry of Peasant Affairs and Agriculture, the two institutions implementing the projects (Institute for Rural Development of the Altiplano and Natioul Institute for the Development of Wool), the Central Bank, the Agricultural Bank of Bolivia, and several bilateral and multilateral donors. A field trip to the project areas provided additional information to validate the conclusions of the PCRs. The audit finds the PCRa generally accurate with respect to the projects principal achievements and difficulties. The points discussed by the audit have been selected in the light of their importance for the design and implementation of future rural investments aimed at assisting  - ii - the Government in the development of Bolivian agriculture in general ard Bolivia's Altiplano in particular. The valuabln assistance prcrided by the Government, its offiekals and other donors and individuals met during the preparation of this audit memorandum is gratefully acknowledged. Following customary OED procedures, copies of the draft audit were sent in October 1988 to the Government and to the Commission of the Euro- pean Communities (EEC), cofinancier of the Ulla Ul'a project. Comments from EEC and from the Borrower have been taken into account in prepariag the final report and are also included as Attachments * and II.  - III - PROJECT PERFORMANCE AUDIT REPORT BOLIVIA INOAVI RURAL DEVELOPMENT PPRJECT (LOAN 1211-80) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as X of Expectation Current Estimate Appraisal Estimate Project Cost (USf million) 12.9 8.4 8s Loan Amount (US8 million) 9.5 8.4 67 Amousht Cancelled - 8.1 - Date Board Approval 02/17/78 02/17/76 - Date of Signing 08/09/78 Date Effectiveness 65/67/78 10/07/76 %2 Closing Date 08/80/82 08/81/64 6s Economic Rate of Return (N) 1s Les than 10 Number of Direct Beneficiaries /a 4-10,000 4,800 /b Number of Credit Beneficiaries 4,600 2,060 c Be CUMULATIVE DISBURSEMENTS FY77 FY76 FY70 FY96 FYs1 FY82 FYsB FY84 FY8s Appraisal Estimate (USS million) 0.5 2.1 5.0 7.2 8.6 9.6 - - - Actual (USS million) 0.4 1.6 8.0 8.6 4.6 5.6 5 8.1 e 6 .4 Actual as X of Estimat so 48 6f 60 52 58 64 08 68 Date of Final Disburement November 9, 1984 Principal Repald (US$ million) 1.64 as of May 81, 1988 STAFF INPUTS (staff weeks) FY74 FY76 FY76 FY77 FY7s FY79 FYs8 FY61 FY62 FY8s FY84 FY8s FY88 TOTAL Identification/ Preparation 55.4 77.1 18.2 - - - - - - - - - - 146.7 Appraisal - 49.4 47.6 - - - - - - - - - - 97.0 Negotiations - - 16.7 - - - - - - - - - - 16.7 Supervision - - 4.2 18.2 27.8 18.0 19.5 16.8 28.4 7.6 11.8 18.1 0.9 165.4 Other - 0.2 0.8 0.2 - - - - - - - 0.1 - 1.2 Subtotal 55.4 126.7 82.5 18.8 27.6 18.0 19.5 18.8 28.4 7.6 11.8 18.2 0.9 416.1  - Iv - BASIC DATA SHEET INAVI PAGE 2 MISSION DATA Date No. of Staff,Days Specialization Performance Types of (oayr) Persons In field represented id Rtinj /a Trend If Problems La Identification 10/78 8 60 a,b,h Proparation 10/74 4 64 arc,s,h Appraisal Lu 05/75 7 210 *,e,c,d,o,g Post Appraisal 01/76 1 6 o Supervision 1 /h 06/76 1 14 o 2 1 U Supervision 2 Li 11/76 1 8 e 2 1 T Supervision 8 08/77 8 16 d,e,e 2 2 T Supervision 4 10/77 8 19 a,d,f 2 2 T,P SupervisIon 5 LI 08/78 1 7 S n.a Supervision 6 04/78 a 26 a,e,- 2 1 T,M Supervision 7 08/78 5 28 e,d,i 2 1 T,M Supervision 8 02/79 1 7 a F,T,M Supervision 9 10/79 4 44 alb,d,e F,T,M Supervision 10 Lk 04/80 2 16 a,b 2 1 F,T,M,P Supervision 11 Li 06/60 1 8 i n.a Supervision 12 / 12/80 1 14 a 2 1 F,T Supervision 13 Lr 07/81 2 21 * ,c 2 1 F,T,M Supervision 14 Lk 02/82 2 18 *,c 2 2 U,T,F,P Supervision 15 fr 08/62 1 20 1 n.e Supervision 16 10/62 1 1 a 2 2 M,T,F,P Supervision 17 08/83 1 11 a 3 a P,M,F,T Supervision 16 Lo 08/83 2 28 a,o n.e Supervision 19L/o/ 04/64 1 a a n.e Supervilion 20 lo 06/64 1 14 a n.a Completion 11/84 2 11 *,c n.e  v BASIC DATA SHEET INGAVI PAGE S OTHER PROJECT DATA Borrower Republic of Bolivia Executing Agency Ministry of Peasant Affairs and Agriculture - through the Project Unit and Institute for Rural Development of the Altiplano (IDRA) Fiscal Year January 1 - December 81 Name of Currency (abbreviation) Bolivlan Peso (lb) Currency Exchange Rate Appraisal Year Average 1976 US81.96 0 20.00 Intervening Years Average 1978-81 US81.00 = 21.65 1982 US81.00 = 196.98 1988 US81.00 a 690.00 Completion Year Average 1984 US31.9 O 8,671.00 Follow-on Prolect Name Ulla Ulla Development Project Loan/Credit Number Loan 1519-89/Credit 762-BO Loan/Credit Amount 9.0/9.0 Date Board Approval 01/12/18 La farmer families. L& Familles organized Into groups. /c Annual average. d s=Xagricultural it/b=agricultural economist/c=financial aaalyst/d=irrigation engineer e=economist/ftlivestock specialist/g=heolth specialist/h=credit specialist. Le improblem free or minor problem/2=moderate problems/3=major problems. /j 1=improving/2=stationary/S deteriorating. Lg FPtinanclol/M=managerial/T=technical/P=political/O=other !h With supervision of Credit 261-80 and 681-8O. /I With supervision of Ulla Ulla Rural Development Project. Li Health component supervision. /k Together with supervision of Omasuyos-Loe Andes Project. 1I Supervision of road components. Together with supervision of roads component of the Ulla Ulla RD Project and preparation of the Santa Cruz Settlement Project. Lm Limited supervis'on to determ!ne the status of the auditing and financial reporting systems for the Ingavi and Ulla Ulla RD projects. n Supervision of the Water Supply and Sanitation components of the Ingavi, Omasuyos-Los Andes and Ulla Ulla RD projects. /o Together with supervision of Credit 988-80 and Loan 1510-B0. Le Mission had to rvturn to headquarters due to a general strike In Bolivia.  - vi - PROJECT PERFORMANCE AUDIT REPORT BOLIVIA ULLA ULLA DEVELOPMENT PROJECT (LOAN 1610-BOICREDIT 762-80) BASIC DATA SHEET KEY PROJECT DATA Appralisl Actual or Actual as X of Expectation Current Estimate ApDraleal Estimate Project Cost (USS million) 24 18 Ls n Credit/Loan Amount (US8 mil lIon) 9/9 9/4 100/44 Cofinencing: EEC (US$ million) 2.0 n.e. n.o. Date Board Approval 01/112/S 01/12/78 _ Dote of Signing 04/08/78 Date Effeti veses 07/6/78 07/81/79 Closing Date 06/8983 12/81/8 - Economic Rate of Return (3) 25 Negativo - Number of Diroct Beneficlarloo 8,400 1,780 La 51 Number of Credit Beneficiaries 2,606 n.n. - CUMULATIVE DISBURSEMENTS FY78 FY79 FY86 FY81 FY82 FY8B FY84 FY86 FY86 FY87 Appraisal Estlmat* (USS million) 1.0 6.2 11.5 14.8 18.0 - - - - - Actual (USS million) 6.2 1.6 2.6 6.a 9.a 10.2 11.0 12.0 12.9 18.0 Actu.l as X of Estimate 26X 268 22X 87X 52X 67X 61X 67X 72X 72X Date of Final Disbursement October 27, 1987 Principal Repaid (US$ million) 2.6 as of May 81, 1988 STAFF INPUTS (Staff weeks) FY78 FY77 FY78 FY79 FY86 FY81 FY82 FY88 FY84 FY86 FY86 FY87 TOTAL Identification/Preparation 11.5 24.2 6.1 - - - - - - - - - 86.8 Appraisal - 166.1 82.8 - - - - - - - - - 138.8 Negotiations - - 11.1 - - - - - - - - - 11.1 Supervision - - 7.2 17.5 19.9 18.8 24.2 23.4 21.8 14.9 6.6 28.0 182.8 Other - 6.1 1.2 - - 0.2 - O.1 - - - - 1.6 Subtotal 11.5 180.8 62.8 17.5 19.9 19.0 24.2 28.6 21.8 14.9 6.8 28.0 W89.6  - vii - BASIC DATA SiEET ULLA ULLA PACE 2 MISSION DATA Date No. of Staff/Days Specialization Performance Types of (fmo/yr) Person In fiold repreonted Lb Rating Ic Trend /j Problems Is Identification a a Preparation 06/76 (OAS Team) n.n n.a Appraisal 08/77 11 41 Supervision 1 08/76 2 7 *,b Supervision 2 08/76 a 10 a,b,c 1 1 U,F Supervision a 02/79 1 6 a 2 1 F Supervision 4 06/79 a 20 a,d 2 1 F Supervision 6 11/79 1 6 a 2 1 F Supervision 6 64/80 2 21 a,d 2 1 F,P,M Supervistor 7 12/o0 2 14 a,b 2 1 N Supervitoen 8 bVe81 8 29 ald,e 8 2 U,F,P Supervislon 9 66/a1 2 14 *,b 8 2 M,F,P Supervision 10 10/81 2 16 b a 8 M,F,P Supervioton I1 02/82 1 18 a 8 2 M,F,P Supervision 12 08/62 1 10 a a 8 O,M,F,P Supervision 18 10/82 a 27 *,f,f a 8 P,O,F Supervision 14 04/88 1 2 b 8 1 P,O,F Supervision 15 06/64 1 B a 2 8 F,P Supervision 16 12/64 1 5 a 8 1 P,O Supervision 17 04/86 1 10 a a i P Supervision 16 09/85 1 10 a 2 Lf 8 a Supervision 19 95/66 1 26 a 8 /j 8 ag Supervision 20 09/86 2 40 a 8 Lf a La Supervision 21 11/86 4 6 a,b 8 f a La Coletion 66/67 2 15 * b  - vill - BASIC DATA SHEET ULLA ULLA PAGE 3 OTHER PROJECT DATA Borrower Republic of Bolivia Executing Agency Ministry of Peasant Affairs and Agriculture - through Institute for the Development of Wool (INFOL) Fiscal Year January 1 - December 81 Name of Currency (abbreviation) Bolivian Peso (lb) Current Exchange Rate Lt Appraisal Year Aveege 1979 US81.00 = 20.00 Intervening Years Average 1979-81 USS1.00 24.31 1982 US31.00 a 198.00 1988 US81.00 = 600.00 1934 USS1.00 a 8,671.00 1986 US81.00 - 1,692,900.00 Completion Year Average 1988 US81.00 = 1,920,000.00 Follow-on ProJect Name Omasuyos-Los Andes Rural Development Credit Number Credit 938-B0 Credit Amount USS8.0 million Date Board Approval 07/27/79 Is Provisional eatimate. No exact figures available. ,b *=Agricultural it/b=Agricultural Eceonomilt/caEngineer/d=Soclologiet/e=Anthropologi.t/f=Financial Analyst. Le 1aproblam free/2oderate probl.m/8major problem. /d isimproving/2ntationary/8=deterlorating. Le Fftinancial/M=managerial/T=technical/P=political/O=other. Lf Dt *elopment Impact (new supervision report format). jg Overall status (now supervision report format). h Soureez Central Bank of Bolivia.  - xl1 - BASIC DATA SHEET OMASUYOS-LOS ANDES PAGE a Is Families organized Into group. lb Annual average. /c a Agrlculturallst/b=Agricultural Economist/c=Financial Analyst/dzForestry Specialist/e=Economist/ 'tConsultant. id lproblem tre or minor probless/2=moderate problems/8=major problems. 1., lui.proving/2=tationary/Sudetrlorating It Ftinanci.l/U=managsrial/T=technical/P=political/O=othor. Lg Mission *lso supervised Ingxvi Rural Development Project. Lh Sou ce: Central Bank of Bolivia.  BASIC DATA SHEET OMASUYOS-LOS ANDES PAGE 2 MISSION DATA Date No. of Stoft/Daye Specializatlon Performance Types of Cmo/Yr) Persons in field representd c R Ratina /d Trend / Problems /t Identification 02/78 Preparation 08/78 Appraisal 10/78 7 n.e. o,b,c,d,e Supervieson 1 11/79 7 6 * 1 1 - Supervision 2 05/80 1 8 * 2 1 F Supervision 8 12/80 1 11 a Supervision 4 08/81 1 11 a 2 1 M,P,F Supervision 5 10/81 1 8 a 2 1 F,P,M Supervision 8 La 02/82 1 8 s Supervision 6 02/82 1 2 d 2 1 F,P Supervision 6 02/82 1 1 c Supervision 7 08/82 1 '2 f Supervision 7 08/82 1 7 e Supervision 7 08/82 1 1 a Supervision 9 10/82 1 1 * 2 2 F Supervision 9 08/881 1 17 * B 2 F,P,T Supervision 10 11/88 1 6 a 8 1 P,F,T Supervision 11 04/84 1 8 a a 1 P,F,T Supervision 12 11/84 1 C a 8 1 P,F Supervision 18 08/85 1 8 a 8 Supervision 14 09/85 1 9 a B Supervision 15 0S/88 1 24 a 4 Supervision 16 09/86 2 16 a 4 Completion 08/07 2 1S *,c OTHER PROJECT DATA Borrower Republic of Bolivia Executing Agency Ministry of Peasant Affairs and Agriculture, through the Project Unit and Institute for Rural Development of the Altiplano (IDRA) Fiscal Year January 1 - December 81 Name of Currency (Abbreviation) Bolivian Peso (lb) Current Exchange Rate jh Appraisal Year Average 1979 USB1.00 = 24.61 Intervening Years Average 1979-81 US81.00 = 24.51 1981 US$1.00 = 24.51 1982 US21.66 3 196.00 1988 US$1.00 = 600.00 1984 US$1.60 = 6,671.00 1985 US$1.00 a 1,692,000.00 Completion.Year Average 1988 USB1.W0 s 1,S920,m0.00 Follou-on Project Nono  - xi - BASIC DATA SHEET OMASUYOS-LOS ANDES PACE 8 /a Faila I organized Into groups. /b Annual average. le m=Agricultural Iat/b=AgrlculturaI Economist/c=Flnenclal Analyst/d=Forestry Special st/.=Economist/ *=Consultant. /d laprobln tree or minor problems/2=modorate probloms/3=aJor problem. lo lu=sprov 1ng/2=stntlonary/3=detsrlorating /f Faftnnenal/umanagertal/T=techn'-osl/P=poIiticsl/Oothor. Lg Mission also supervised Ingavi Rural Devolopment Project. Lh Source: Central Bank of Bolivia.  - xii - PROJECT PERFORMANCE AUDIT REPORT BOLIVIA INGAVI RURAL DEVELOPMENT PROJECT (LOAN 1211-BO) ULLA ULLA DEVELOPNENT PROJECT (LOAN 1510-BO/CREDIT 762-BO) OMASUYOS-LOS ANDES RURAL DEVELOPMENT PROJECT (CREDIT 933-BO) EVALUATION SUMMARY Introduction The three projects covered by this audit report comprise the entire IDA/Bank assistance to Bolivian agriculture in general, and rural development in particular, between 1976 and 1987. In the mid-1970s, the Government's strategy for agriculture and rural development emphasized the reduction of the country's dependence on imported foodstuffs and the alleviation of acute poverty and underemployment in Bolivia's Highlands -- the Altiplano. For the latter objective, public agricultural investment was directed towards integrated rural development projects to provide basic social services to increase the productivity of subsistence agriculture. The Ingavi project was the first example in Bolivia of such an integrated approach and was followed, relatively quickly, by the Ulla Ulla and OmAsuyos-Los Andes projects as development models for the region of the Altiplano and the adjoining intermediate valleys. Earlier IDA assistance to Bolivia included two simple agricultural credit projects for commercial beef development in the tropical lowlands and another two credit operations (1971-1980) in support of small-scale farmers in the Altiplano. 1/ The four IDA operations had a positive impact on cattle management and beef production but did not achieve their institutional and sectoral objectives. Obiectives The Ingavi, Ulla Ulla and Omasuyos-Los Andes projects aimed at increasing crop or livestock production and beneficiary incomes by forming farmer groups to facilitate delivery of services (technical assistance and credit), and providing social infrastructure on a province-wide basis in Bolivia's Altiplano. These multi-objective projects focussed on income generation and distribution, community participation and food production and emphasized replicability. The Ineavi Rural Development Proiect scught to achieve its objectives in the Ingavi province, benefitting 10,000 families by establishing a new and autonomous unit to administer the project and provide improved production services (organization of farmer groups, extension, animal health); providing agricultural credit through the Agricultural Bank of Bolivia; and constructing 20 deep wells to irrigate 1,000 ha, 4,000 shallow wells for drinking water, rural roads and other 1/ PPARs of Bolivia, First and Second Livestock Development Proiects, (OED Report No. 537, dated October 11, 1974); and Third Livestock and First Agricultural Credit Proiects, (OED Report No. 3996, dated June 23, 1982).  - xiii - social infrastructure. The project cost was estimated at US$12.9 million and the Bank loan of US$9.5 million was to be disbursed over five and half years. The Ulla Ulla Development Prolect was designed as a phased development of the alpaca/llama livestock sector in the Ulla Ulla region to the Northeast of Lake Titicaca and of related fiber processing capacities in the La Paz area, benefitting 6,400 families. The project included 16 components for livestock development (experimental sta4ion, alpaca/llama commercial farms, credit, extension services); a wool collection network and a wool processing plant; rural roads; social infrastructure; and women's participation in health and nutrition training. The project cost was estimated at US$24 million and IDA/Bank financed US$18 million to be disbursed over five years The Commission of the European Communities provided parallel financing in the amount of US$2 million equivalent. The Omasuvos-Los Andes Rural Development Project aimed at reaching 6,000 small-scale families in two provinces to increase their agricultural, livestock and forestry production. The beneficiaries were to be organized in farmer groups which were to receive integrated agricultural packages, including credit and technical assistance; other instruments comprised roads, and drinking water and waste disposal facilities. The project cost was estimated at US$9.3 million and the IDA credit of US$3 million was to be disbursed over four and half years. The International Fund for Agricultural Development provided parallel financing in the amount of US$4 million equivalent. Implementation Experience The implementation of all three projects experienced delays by lack of final design of several components, lack of counterpart funds, shortages and frequent changes of staff, and technical difficulties. Post- appraisal developments such as political instability and the economic and social crisis beginning in the early 1980s caused increased turnover at ministerial and project management levels which made it extremely difficult to carry on with development as was envisaged at the time of appraisal. The agricultural credit components in each of the three projects functioned during the early phases of the projects but were managed by the projects' autonomous units, contrary to appraisal expectations. In the Ingavi project the deep well irrigation component was not carried out for lack of background studies which would have revealed unsuitable soil and climatic conditions. The rural infrastructure was generally carried out, including an access road. Of the loan of US$9.5 million, US$3.1 million was cancelled. In the Ulla Ulla project the road component was completed as was much of the rural infrastructure. The improvement of alpaca and llama herds did not take place, and the main intended investment activity of the project -- an integrated scheme of wool purchases and processing by a specialized yarn plant -- was left uncompleted. A conservation program for the endangered vicuina was carried out successfully. Of the credit/loan of US$18 million, US$5 mil3ion was cancelled.  - xiv - In the Omasuyos-Los Andes project, the technical services operated during the earlier years, and farmer training took place. Both male and female extension agents (the latter very important in the socio-cultural context of the Altiplano), delivered extension programs in homemaking, family health and horticulture. An access road was completed but the health infrastructure reached only 4% of targets and the forestry component was a failure. Of the IDA credit of US$3 million, US$0.9 million was cancelled. Results and Sustainability Government activities (irrigation, forestry, research, and extension) under all three projects fell short of their envisaged scope in virtually every area, and the actual achievements have still to be proven sustainable at present input levels. The credit and technical assistance components cannot be maintained by the existing institutional framework. However, public social investments (rural roads, health and education facilities), while also falling short of their phydical targets to some extent, were relatively successful and appear to offer better prospects for sustainability. As for private investments (on-farm inputs and investments supported by agricultural credit) the achievements generally fell short of their targets, but were successful, if only for a few years. In the absence of follow-on support activities, their longer-term outlook is in doubt. Findings and Lessons Having been prepared and implemented under almost the same macroeconomic conditions, sharing the conceptual basis and being directed at nearly identical populations and environments, it should not be surprising that the implementation experiences of the three projects and their present situation have much in common: - the project design was too complex, particularly in view of Bolivia's scarcity of administrative, managerial and technical personnel, the geographic isolation of the project areas, and the novelty of some of the projects' components (PPAM, para. 18); - the projects were designed to be managed by ad-hoc units rather than by line agencies. These units ceased to function at the end of the projects and are not sustainable (PPAM, para. 19).2/ The following lessons can be drawnt - a proper understanding of the project areas, their socio- economic structure and their traditions and values would have 2/ The EEC notes that its recent experience suggests that "ad-hoc' units are more operational than line agencies for implementing rural development projects, although there is a greater risk that they will not be sustained.  - xv - prevented or at least mitigated, some serious project shortcomings (PPAM, paras. 55-57); the outcomes of the three projects confirm the conclusions made in a recent review of rural development investments, particularly in regard of the negative effects of project complexity (PPAM, paras. 58-60); and lack of adjustment for inflation in one credit scheme has completely undermined its sustainability (PPAM, paras. 61-64) and given rise to substantial unintended subsidies and losses to Government (PPAM, paras. 65-66).  PROJECT PERFORMANCE AUDIT MEMORANDUM BOLIVIA INGAVI RURAL DEVELOPMENT PROJECT (LOAN 1211-BO) ULLA ULLA DEVELOPMENT PROJECT (LOAN 1510-BO/CREDIT 762-BO) OMASUYOS-LOS ANDES RURAL DEVELOPMENT PROJECT (CREDIT 933-BO) I. PROJECTS' BACKGROUND A. Agricultural. Economic and Social Setting 1. Bolivia consists of three major agro-climatic regions extending, at different altitudes, in a North-South direction over the entire country. By far the largest is the zone of the Eastern Lowlands, a vast, sparsely populated and little developed expanse of tropical forest which occupies 60% of the country's land area. With high temperatures and rainfall throughout the year, the Lowlands can produce two annual crops. Commercial farming concentrates on sugarcane and cotton and subsistence farming, on cassava and bananas, with rice being grown for both purposes and cattle raising complementing agriculture. 2. The Valleys amount to about 15% of Bolivia's land area; they are a transition zone between the Eastern Lowlands and the Altiplano/Mountains area. Altitudes range from 800 to 3,500 m and soil and climatic conditions vary greatly. Agriculture is limited to about one tenth of this zone. Besides fooderop production (grains, vegetables), pork and poultry are raised and market-oriented dairy farming is practiced. 3. The Altiplano/Mountains region occupies roughly one quarter of Bolivia's land area, in the Center and West of the country. About half of this area lies at elevations of more than 4,500m and does not permit either agriculture or animal husbandry. The other half, the Altiplano, an inter- mountain plain stretching from Lake Titicaca to the Argentinian border, is characterized by cold and dry winters and low rainfall, a climate which limits severely the agriculturale options: limited foodcrop production (tubers and cereals) alternate with low-productivity livestock ra.4ing (cattle, pork, poultry, and llama and alpaca at higher elevations), mmstly for on-farm consumption. 4. Bolivia's population is very unevenly distributed; there is a striking difference between the sparsely populated and largely undeveloped sub-tropical lowlands of the East (where most commercial non-food crop production and 25% of the food for national consumption is produced); and the highly populated Altiplano-Valleys area of the West and Center where small-scale traditional agriculture supplies a large part of the food consumed in the urban centers of the Altiplano, including La Paz. Less than 20% of the Altiplano area has potential for crop production; most of this area is in the Department of La Paz, where the three project areas are located. The average holding on the Altiplano varies from 2 ha near Lake Titicaca to 4-5 ha in the high valleys, but the planted areas are much smaller due to the need to leave land in fallow. The entire Altiplano region is highly dependent on livestock production, with a population of 6 million sheep, 1.6 million llamas and alpacas, and 1 million cattle.  5. There are intricate demographic links between these three zones. Bolivia's main ethnic and linguistic groups (Quechuas, Aymaras, Guarayos and others), all with distinct sccio-economic and cultural traits, have for thousands of years, occupied areas that extend across the country in an East-West direction. This permitted them to avail themselves of the production diversity of three different climatic and agricultural systems. The movements of people and goods along an East-West axis maximized employment and production by farming or livestock-keeping thoughout the year in dry and cold highlands, in humid valleys, and in warm tropical areas.l/ B. Obiectives and Design of the Project 6. Gevernment's previous and present strategy in the agricultural sector supported by the Bank, is directed at reducing the country's dependence on imported foodstuffs and alleviating acute poverty and underemployment in the traditional agricultural sector -- the Altiplano. For the latter objective, public agricultural investment was directed towards integrated rural development projects to provide basic social services, infrastructure, marketing and credit, and extension services. The Ingavi project was the first example in Bolivia of such an integrated approach and was followed, relatively quickly, by the Ulla Ulla and Omasuyos-Los Andes projects, as development models for the region of the Altiplano and intermediate valleys. Within that strategy, Government recognized the difficulties in bringing about an orderly development of the Altiplano, because of its limited resources and harsh climatic conditions, and where the population has traditionally been little receptive to change and even, to some extent, suspicious and unapproachable. The Bank also was aware of these difficulties and the loans were approved by the Board recognizing these risks. 7. The three projects covered by this PPAR provide the total of the Bank's assistance to Bolivian agriculture in general, and rural development in the Altiplano in particular, between 1976 and 1987. The Ingavi Rural Development Project was Bolivia's first integrated rural development project and followed similar undertakings elsewhere in Latin America.2/ All three projects aimed at increasing crop or livestock production and beneficiary incomes by forming farmer groups to facilitate delivery of services (technical assistance and credit), and providing social infrastructure on a province-wide basis in Bolivia's Altiplano, where 60% of the country's population is located, with severe climatic and resource- base constraints. These multi-objective projects focussed on income generation and distribution, community participation, food production and replicability. 1/ Romero, Hugo, Movimientos sociales y planeamiento andino en Bolivia, Ediciones Populares Camarlinghi, La Paz, 1980. 2| PCR of Colombia - Integrated Rural Development Prolect (OED Report No. 6807), dated May 29, 1987, and PPAR of Mexico - Integrated Rural Development - PIDER I (OED Report No. 4617), dated June 30, 1983.  -3- Inpavi Rural Devel2oment Proiect 8. The project was a follow-up of two previous activities in the Altiplano: (a) the IDA-financed Livestock Credit Project (Credit 261-BO) which earmarked part of the livestock credit for the Altiplano;3/ and (b) the United Nations Development Program (UNDP) Groundwater Survey of the Northern and Central Altiplano, which indicated the availability of groundwater for irrigation. 9. The five-year project sought to achieve its objectives in the Ingavi Province by: establishing a new and autonomous Project Unit (PU) to administer the project and provide improved production services (extension, investment and short-term credit, farmer organization, animal health); and constructing 20 deep wells to irrigate about 1,000 ha, 4,000 shallow wells for drinking water, pit privies for about 3,000 families, six health centers, and 85 kms of rural roads. The project was to provide the opportunity for Government to gain experience and devise a model for development in the Altiplano. The autonomous PU was to be attached to the Ministry of Peasant Affairs and Agriculture (MACA) and its Departments of Engineering and Services would implement components or coordinate their execution by six other Government departments. The PU had a ten member Board of Directors and its initial staffing was to be accomplished by hirings from other institutions. The appraisal mission considered the strengthening of an existing institution but decided a new unit would be a more effective way to coordinate the project (para. 49). The basic productive benefit assumptions were that irrigation, higher levels of production inputs, and improved agricultural practices would increase yields and the volume and value of production. The project cost was estimated to be US$ 12.9 million, and the Bank loan of US$ 9.5 million was to finance the foreign exchange equivalent plus US$ 4.4 million of local costs. Ulla Ulla Development Prolect 10. The original project concept aimed at a phased development of the alpaca/llama livestock sector in the Ulla Ulls region and of related fiber processing capacities in the La Paz area. Eventually, however, the five- year project included 16 components for: livestock development (including an experimental station and commercial farm, credit, and extension services); a wool collection network and a wool processing plant; rural roads; social infrastructure; and women's participation in health and nutrition training. The newly created National Institute for the Development of Wool (INFOL) was to be responsible for project implementation and was to enter into operational agreements with other Government departments for the implementation of the rural and social infrastructure elements. The livestock component would benefit 2,500 families, the wool handicraft industry 3,400, and the social infrastructure 3,000. The project's costs were estimated at US$ 24 million; the IDA credit (US$ 9 million) and the Bank loan (US$ 9 million) would finance 100% of the foreign exchange component of US$ 9 million and 60% of local costs. 3! PPAR of Bolivia - .hird Livestock and First Asricultural Credit (OED Report No. 3996), dated June 23, 1982.  - 4 - The Commission of the European Communities and the Bolivian Government were to share the financing of the remaining US$ 6 million. Omasuyos-Los Andes Proiect 11. The Omasuy,s-Los Andes project was identified in 1978 on the basis of the initial experieace of the Ingavi project which had become effective in July 1976. The four and a half year project aimed at reaching the small-scale farm sector in the provinces of Omasuyos and Los Andes in the northern Altiplano, to increase agricultural, livestock and forestry production of 6,000 families. The beneficiaries were to be organized in farmer groups through which integrated packages were to be provided, including credit and technical assistance, roads, and drinking water and waste disposal facilities. The Omasuyos-Los Andes project was a replication of the (then still ongoing) Ingavi project. The project components were: establishment and partially financing of a project unit under a new agency created as a condition for the Credit, the Institute for Rural Development of the Altiplano (IDRA); investment credit for 2,500 and production credit for 6,000 farmers; a forestry program (establishment of a nursery and long-term credit to reforest 600 ha); improvement of 50 km of feeder roads; 4,000 pumps for shallow wells to supply drinking water; and 3,000 privies for sanitatior.. The productive benefits were to derive from a more intensive cultivation of foodcrops, both for sale and on-farm consumption. The project costs were estimated at US$ 9.3 million. The IDA credit of US$ 3 million was to finance administrative and technical support (US$ 2.3 million) and rural infrastructure (US$ 0.7 million). The Internatior l Fund for Agricultural Development (IFAD) cofinanced the project through a loan of US$ 4 million to support the agricultural credit and forestry components, and Government was to finance the remaining US$ 2.3 million. C. Board Concerns 12. The Ingavi project was presented to the Board in February 1976 and was explicitly characterized as a high-risk operation. While the Board endorsed this approach in principle, concerns were raised about the feasibility of forming farmer production groups or associations in heterogeneous sub-areas and populations. However, the explanation that these groups would not be formal cooperatives or have f.ll legal responsibility wa deemed satisfactory. The Ulla Ulla project was not explicitly seen as a high risk operation during Board presentation in January 1978, but concerns about the possibility of quickly establishing a multi-disciplinary organizational structure cutting across institutional lines, and requiring well paid staff were evident. The Board did not consider the Omasuyos-Los Andes project (July 1979) as a high risk operation because of the Bank's previous experiences in the Altiplano, but it did link the project's chances of success to the ability of retaining well paid staff. D. Related Projects 13. The first two IDA operations in Bolivia's agricultural sector were two simple agricultural credit projects for commercial beef development in the tropical lowlands which were implemented between 1967 and 1974. The projects were administered by a specially created Livestock Project  -5- Division in the Agricultural Bank of Bolivia (BAB). Most of the project's objectives were achieved one year earlier than expected.4/ The PPAR concluded that the two projects had a positive impact on cattle management and beef production, and that the Livestock Division performed satisfactorily in appraising and supervising ranchers' investment programs, even showing a modest profit in its operations. The PPAR also concluded that BAB's financial situation steadily deteriorated during the projects' implementation largely due to measures taken by the Government: imposition of unprofitable lending programs, reduction in the interest rate for on- lending and devaluation of the Bolivian currency (most of BAB's borrowings were in foreign currency). 14. The third and fourth IDA operations (Credits 261 and 561-BO) were also agricultural credit projects (1971-1980) through BAB, with the following aspects distinguishing them from their predecessors: the extension of the beneficiary groups to include small-scale farmers in the Altiplano; an . tempt to charge positive interest rates by indexation of the loan principal to large-scale beef farmers; and several studies aimed at reorganizing BAB and rationalizing beef pricing and marketing.5/ The achievements of the third and fourth operations were: BAB's agreement to accept chattel mortgages for small-scale farmers instead of land titles, the reinforcement of BAB's Livestock Division, and the establishment of an evaluation and monitoring system. The projects were also successful in diversifying lending operations from the lowlands with its large farms to small-scale farmers in tha Altiplano and Valleys regions. The projects failed, however, in their institutional and sectoral objectives, which were and remained beyond their control. Because the indexation formula hqd not been agreed upon during negotiations, lengthy discussions took place between IDA, Government and ranchers, and it was not until June 1975 when agreement was reached by all three parties concerned. The agreement was not fully respected either by Government (which in 1978 unilaterally lowered interest rates) or by farmers who, after two devaluations, responded by increased defaults on repayments. 15. The three rural development projects under review were identified and appraised during a period of favorable economic conditiovs. Between 1970 and 1978, real GDP grew at an annual average of 5.4%, bolstered by buoyant export commodity prices, relative political stability, easy access to external financing and high gross investments. By contrast, their implementation fell into a period of prolonged crisis: net foreign lending ended abruptly in 1979/80 as Bolivia's foreign debt was deemed to have become excessive, and it oecame apparent that past lending was not having a significant development impact but was instead stimulating serious capital flight. Since 1980 the Bolivian economy has steadily declined, reaching an all time low in 1985 when inflation exceeded 20,000% in annual terms. Between 1978 and 1985 agricultural GDP declined 7% but its share of total GDP increased to 20%, reflecting the sharp drop in mining and manufacturing. The economic policies pursued by Government between 1978 41 PPAR, Bolivia - First and Second Livestock Development Projects (OED Report No. 537), dated October 11, 1974. 5/ PPAR, Bolivia - Third Livestock and First Agricultural Credit Projects (OED Report No. 3996), dated June 23, 1982.  - 6 - and 1985 worsened the general economic decline. Interventionist pricing policies, maintenance of an overvalued exchange rate, and excessive government expenditures (with the accompanying drain on domestic credit, expansion of the public debt, and runaway inflation) created a macro- economic environment hostile to growth in all sectors, particularly agriculture. Since August 1985 the Government has been implementing a new economic policy which so far has led to remarkable progress in controlling inflation and in reestablishing order in the economy. II. IMPLEMENTATION AND OUTCOMES 16. The performance and -'utcomes of all three rural development projects need to be considered in the context of the following factors: (a) Bolivia's Altiplano is populated by different indigenous peoples, with traditions and farming techniques reaching back thousands of years; (b) all projects were appraised before the economic and social crisis in the early 1980s which affected Bolivia deeply, but even under normal circumstances these projects appeared optimistic in most of their underlying assumptions; and (C) the administrations of the three projects were unable to resist local pressures injurious to the projects' interests, started but did not always complete new undertakings, and could not incorporate the experiences of their partner projects. The three projects had several common experiences or suffered from the same factors. Hyperinflation occurred between 1979 and 1985, the period during which most of the projects' implementation took place. Several components in each of the projects were poorly identified, designed or appraised. All projects utilized enclave management units with complex coordination arrangements with other institutions. The management units have ceased to exist. The agriculture credit components met with some success in their productive objectives but failed in their institutional and sustainability objectives. The projects were affected by low staff morale and continuous changes in local staff. The common and particular outcomes of the three projects are discussed in the following sections. 17. Having been prepared and implemented under almost the same macto- economic conditions, sharing the conceptual basis and being directed at nearly identical populations and environments, it should not be surprising that the implementation expe_ience of the three projects and their present situation have much in common. 18. First, there can be little doubt that the projects' design was too complex, particularly in view of Bolivia's scarcity of administrative, managerial and technical personnel outside the capital, as well as the geographic isolation of the projecL areas and the novelty of some of the projects' components. 19. Second, and in response to the before-mentioned shortages of specialized staff, it was decided to entrust project implementation to ad-  - 7 - hoc management units rather than to Government line agencies. While such an arrangement may look appealing because of the (apparent) avoidance of inter-departmental rivalries, in many cases it creates more problems than it helps resolve: the necessary coordination mechanisms are usually quite complicated, the lack of a firm place of these "enclave' units in the Government structure sometimes makes them dependent on the goodwill of the participating line agencies, the problem of personnel shortages is exacerbated rather than mitigated by the setting-up of additional agencies, and because of their location outside the regular structure of line agencies, their continued existence beyond the implementation period is not automatically assured but requires continuous provision of adequate funds to sustain follow-up activities. 20. Third, such project implementation units are subject to increased staff turnover, partly in response to the temporary nature of these posts, partly because at the top levels staffing is particularly affected by political changes. A volatile political situation tends to have destabilizing influence on PIU staffing and hence, on project implementation. 21. Fourth, several project components were poorly identified, designed or appraised, leaving the implementing agencies with an unrealistic framework for implementation. This problem is linked to the first issue (of design complexity) but adds to it through overly ambitious physical targets and a compressed time schedule. 22. This set of basic weaknesses would have presented a formidable challenge to the implementing agencies even under favorable economic and political conditions. In the event, the external circumstances were extremely adverses a prolonged period of hyperinflation in the early 19808 combined with extraordinary political and social instability which left their marks on all three projects. 23. It is interesting to r te that these three projects, while taking place in a rather unique physical environment, present insights almost identical to the conclusions derived from a survey of issues in about 200 rural development projects financed by the World Bank.6/ There the root causes of a project's difficulties were weaknesses in design, which could take the form of excessive scope and complexity (particularly in view of the frequent leck of previous relevant experience); awkward management structures and implementation arrangements; and unrealistic implementation plans. Other shortcomings, such as an absence of arrangements to assure an adequate flow of funds to the projects; inadequate or untested technologies; and an unsatisfactory incorporation of technical assistance inputs, have also made themselves felt repeatedly in the three projects under review. The following sections will examine their experiences individually. 6/ Rural Development. World Bank Experience, 1965-86, World Bank/Operations Evaluation Department, April 1988, in particular Chapter 4, pages 21-36.  A. Ingavi Project Manasement 24. The five-year project (1976-1981) was closed after eight and half years in March 1984. The project unit became operational in December 1976, and was established under MACA. The project unit had to deal with seven other Government agencies and was burdened by a Board of Directors of ten members with conflicting interests, who rarely met. The project unit was absorbed in March 1979 into the newly created Institute for Rural Development for the Altiplano (IDRA), which also coordinated the follow-on project of Omasuyos-Los Andes. During the project's life, the project unit and IDRA were plagued by frequent changes of Directors and consequently, of almost all key staff, while ignoring Loan Agreement covenants which required Bank approval of appointments of directors. This called for continuous training of new staff and antagonized the farmers' associations which had to cope with an ever changing project management. Since 1986, however, the same management team has been in place, allowing the consolidation of some of the project's gains. At project's initiation in 1976/77 there was a staff of 120. Because of low salaries and diminishing morale, at project's closing only 30 remained, and at audit there were only four agents to serve 4,000 families in the province at Ingavi and 8,000 in Aroma (a province added to the project in December 1981 to receive technical assistance and credit). IDRA estimates that for a continued and proper coverage of the beneficiaries the required ratio of technicians to families is 1:60, and that consequently a total staff of 67 would be needed for Ingavi alone.7/ Achievements by Component 25. Irrigation. The deep well irrigation component, accounting for 16% of project costs, failed (PCR, paras. 3.06-3.07). Out of a target of 20 wells to irrigate 1,000 ha, only five test wells were installed and only two maintained from 1983 to 1985. All were eventually abandoned, because only a few micro-regions have the proper soil and climatic conditions to obtain an adequate production response to irrigation, and because of the difficulty to organize farmers' participation in irrigation arrangements. A design flaw in the irrigation component arises from the fact that when water is desired during the dry season, temperatures in the Altiplano are often below freezing point and water cannot be provided. 26. Extension Services. Agricultural extension was inadequate to meet the full requirements of the Altiplano traditional rural population (PCR, paras. 3.08-3.09). There were simply not enough staff knowledgeable in the local language and customs to serve the large and widely scattered beneficiary population. But where and when the extension system operated properly, it succeeded in organizing farmers' organizations to receive collective credit and in giving instruction on simple farm practices, 7/ While a ratio of 1:60 is probably too high, the actual situation (four agents for 12,000 families) is obviously inadequate. A supervision report for the Omasuyos project (dated January 27, 1984) suggested a ratio of 1:250 as suitable for the Altiplano; this would translate into a total requirement of 48 agents for the two provinces.  - 9 - mainly the use of improved seeds, fertilizers and land preparation. Between 1977 and 1987, 350 courses were given to a total of 9,000 farmers, and as a complementary activity, project veterinarians treated about 350,000 animals. 27. Infrastructure. The rural infrastructure components were generally carried out as foreseen (a'CR, para. 3.02). The road component exceed appraisal expectations by 142 and the planned six health centers were constructed in locations which can be reached by project and non- project beneficiaries. The centers were staffed and are functioning seven days a week. Only 30% of appraisal targets for waste disposal were met; family water supply benefitted from the use of a simple but durable handpump designed and patented by the project (over 1,200 pumps were delivered to beneficiaries). 28. Credit. Several basic objectives of the credit component were achieved, despite inadequate institutional and administrative arrangements which restricted the scope of the programs farmers in the project area were, for the first time, exposed to institutional credit, organized into farmer groups to apply for credit, and experienced higher yields as a result of introducing agrochemicals. Unfortunately, the failure to improve the management of the credit component resulted in excessive staff use, almost fully subsidized credit, and erosion of the value of credit repayment. (These latter issues are discussed further in paras. 61-67). Almost all credit was production credit (short-term) and was granted in kind (mainly improved seeds and fertilizer) by the project unit. About US$ 3.0 million were spent on the purchase of imported agricultural inputs, including improved dairy breeds, of which the Bank financed about US$ 1.7 million equivalent (56%) and the Government of Bolivia the remainder. About US$ 2.5 million were granted as credit in kind during the 8-1/2 years of project execution, through 321 subloans to associations, benefitting about 8,000 individual farmers. It appears that a large portion of the difference of US$ 0.5 million between the totals for import purchases and credit was lost during 1982-85 when inputs were delivered to farmers at prices well below acquisition cost, in response to heavy pressure from farmers who claimed that market prices were lower than those being charged by the Project Unit. Beneficiaries 29. Eight thousand families, or twice the appraisal estimate, benefitted from the technical services in two provinces (Ingavi and Aroma). The increased number of beneficiaries was at the expense of decreasing the amount of the average short-term subloan per beneficiary (US$166 at completion as compared with US$246 at appraisal). Five thousand families (45X of the appraisal estimate) benefitted from the drinking and sanitation components; most of them also received credit. Disbursements and Costs 30. Because of chronic lack of counterpart funds and the Bank's obligation to reimburse only against incurred expenditures, the project suffered delays and was scaled down, resulting in loan disbursements of only US$ 6.4 million, US$ 3.1 million being cancelled. Actual total costs were estimated at completion to be US$ 8.4 million as compared with USS  - 10 - 12.9 million at appraisal. Actual expenditures for expatriate and local technical services amounted to US$ 4.8 million (or 57X of actual project costs), while at appraisal it was estimated that such expenditures would account for only 25% of project costs. Proiect Outcome 31. A large part of the production benefits were estimated to derive from the 1,000 ha to be irrigated, a component which was not carried out. However, rainfed areas which received credit do show some impact; with inorganic fertilizers, improved land preparation, and improved seeds, average yields for potatoes (the main staple) have doubled or tripled (depending on weather conditions) from 3.5t/ha prior to the project, and oats and barley yields (the main forage crops) have doubled to 3 t1ha. Because of the mixed cropping practices followed by all farmers, there are no reliable records of other production components. A visit to the project area, however, confirmed the importance of alfalfa and oats for livestock production and the increasing role of dairy cattle. About 960 improved dairy cattle were introduced and distributed to 200 families. The demand for dairy cattle in the project area is still very large but constrained by two factors: scarcity of improved breeds, which often must be brought from Peru, and lack of credit. Farmers who have received credit for dairy cattle and tractors often have repaid their loans within two years, partly because the investments were productive, partly because loans could be repayed with devalued currency. At audit (May 1988), the economic situation appeared stationary, with no evidence of continued development. This may be due in large part to the cessation of technical assistance and credit in 1988. The issue of sustainability is further discussed in para. 42. The ERR at appraisal was estimated at 15%. Because of the failure of the irrigation component, the uncertainty regarding sustainable yields on rainfed crops and the continued lack of an adequate support structure, the project cannot be considered successful. B. Ulla Ulla Proiect 32. Management. Project implementation lasted almost nine years compared to an appraisal estimate of five years. As in the case of the other two projects, implementation was hampered by economic and political instability, hyper-inflation which made the provision of adequate and timely counterpart funding illusory, and frequent changes in project management which in turn had repercussions on senior project staff. The fact that implementation was entrusted to an agency which was a new institution in the process of being established, with project responsibilities well beyond the narrow technical scope of this institution (i.e., the development of a national wool industry) contributed to implementation problems and delays, as did the extreme isolation of the project area (the most remote by far of the three projects). As in the case of the Ingavi Project, however, there has been a degree of institutional stabilization in recent years, and the new management is looking at the role and possibilities of the institution, and the project's legacy and future, with a commendable sense of realism.  - 11 - Achievements by Component 33. Infrastructure. Among the infrastructure components realized under this project, the most important ones in terms of their potential impact on the project area are new access roads (whicn was also the only physical component with a target overachieved -- 56 vs 54 km estimated at appraisal.8/ During the field visit, the mission found circumstantial evidence that the roads have not only facilitated movement of goods into and out of the project area (cutting the travel time to La Paz by about five hours) but are also enabling the population in the area to avail themselves more easily of temporary employment opportunities outside the area. However, the physical improvement will need to be complemented by a corresponding marketing structure for the area's main commodity, wool, in order to reap fully the intended benefits of this investment. 34. Next to the roads, the other physical investments pale in significance, although in the longer run they will also contribute to the process of social transformation and of integration of the beneficiary population in the national mainstream. The school facilities (17 primary classrooms and one intermediate school) appear to be well used, and the implementation shortfall was comparatively minor (none at the intermediate level, about one third for the primary classrooms). One supervision report claimed that the classrooms were not always established in the places of greatest demand but rather in response to the political weight of specific local groups, but such observations are difficult to verify or reject on the basis of spot observations, and project personnel felt that by and large the school facilities represented sound investments. It should not be overlooked, though, that while education will facilitate the out- movement of people from an overpopulated area (i.e. relative to its resource base), it may also, over the long run, deprive this region of its wuost energetic and best qualified people.9/ The provision of new health facilities (two health posts and five "micro-posts") corresponded to 100% and 171 of targets respectively, whereas the sewage disposal program (construction of pit privies) reached only 81 of its planned size. The shortfalls in meeting the targets of the health and waste disposal components were a result of the lack of knowledge of the social structure of Ulla Ulla. The health component was opposed by the traditional healers. The local customs and traditions were difficult to overcome when trying to introduce new sanitation practices. Two out of five wool collection posts were constructed in the project area and one central warehouse near La Paz. 35. stool Processina. This component, expected to be the main intended investment activity under the project, was the one with the most disappointing results to date: the establishment of a yarn spinning plant near La Paz which was to provide the outlet for the increased production of alpaca and llama fiber in the project area and reverse the flow of (illegal) export of fiber to Peru which deprived Bolivia of additional income, public revenue and employment opportunities in the downstream 8/ The figure of 141 km given in the Bank overview preceding the PCR (para. 4) appears to include 85 km of minor upgrading works, in addition to 56 km of new construction. 9/ The EEC notes with satisfaction this positive element of the project.  - 12 - production process. The plant was to have a processing capacity of 595 t of finished fiberlyear. Various supervision reports (whose basic findings appear not to be contcsted by the Government personnel with whom the audit mission consulted) suggest the following set of shortcomings and potential problems:10/ (i) The equipment (total purchase price of US$ 5.1 million) had been bought through international competitive bidding at prices well above the level prevailing at the time of procurement (1982).11/ While well protected in its original crates and in good physical condition (a statement which also applies to the situation at the time of the audit mission in May 1988), it was of course subject to technical obsolescense; (ii) Purchases were apparently based on badly balanced equipment lists resulting in equipment bottlenecks at some stages of the production process and in excess capacities at others, because actual purchases deviated from the appraisal report specifications. Bringing the entire production facility to an output level of 595 t/year would require additional equipment purchases on the order of US$ 1.1 million (in 1986 prices); (iii) The Government has prepared a factory site at the outskirts of La Paz at the cost of US$ 0.2 million equivalent. Developing the site for factory readiness would require an additional investment of US$ 1.5 million equivalent; (iv) Even under optimistic output/cost/price assumptions, the economic rate of return was not expected to exceed 5%. This assumed a satisfactory resolution of fiber collection, transportation and marketing problems, and was based on an implementation schedule which has already been invalidated oy the passage of time; and (v) Given these conditions, the Bank recommended the selling of the equipment to the private sector rather than completing the investment. This conclusion was unacceptable to the Government then and is even more so now, given the reduction of the equipment's resale value in the intervening years. More serious for the Government, however, are the negative social and political implications of such a decision which it feels are of paramount importance. 36. Having to make the choice between a course of action devoid of economic and financial attraction and a decision with some potentially damaging political and social consequences is probably the most difficult legacy of this project. The present management of the National Institute for the Development of Wool (INFOL) is favoring a gradualist approach, proceeding in stages with the transformation of the raw material (better grading, cleaning, etc.) before embarking on factory operations proper, to 10/ Supervision Report, Annex 2, December 24, 1986. ll/ Due in part to the inclusion of erection charges (20%) which were never consummated, but also to a premium for incremental risks.  - 13 - make sure the necessary collection and transportation arrangements are firmly in place, and also to put greater emphasis on llama (rather than alpaca) fibers, recognizing the comparative advantage/disadvantage Bolivia has vis-a-vis her main competitor (Peru). While this approach is to be commended for its caution and flexibility, it should be realized that fiber processing as originally envisaged is unlikely to become a financially viable proposition unless the funds for the necessary additional investments can be obtained under highly concessionary terms. 37. For project activities which did not result in physical infrastructure, an outcomes assessment (or even an enumeration of project results) is much more difficult. The most important achievement (with one exception to be dealt with in the next paragraph) seems to have been improved livestock practices, manifest in better fiber shearing techniques (with scissors instead of knives, sharpened pieces of sheet metal or glass shards), parasite control (construction of dips) and, in some cases, supplemental feeding and attempts at pasture improvement. 38. If one abandons a narrow financial view, the success of the vicuna conservation program has to be rated high. At the verge of extinction in the Ulla Ulla area at the beginning of the project, the stock has since more than tripled to an estimated 2,800 heads. While at the present time there is no viable technique of exploiting this potentially very valuable resource 12! short of destroying it (i.e. of killing the animal), the non- monetary value of conserving this unique animal should not be dismissed out of hand. 39. Credit. Credit operations under this project were minimal compared with appraisal targets. Only 26 subloans for livestock development and 18 subloans for handicraft development totalling US$ 0.1 million were granted from 1981 to project closing in 1986. It appears that the main reasons for this low demand for credit were: (a) the failure of the Agricultural Bank of Bolivia in its contractual obligations to open a branch office in the project area 13/ and the resulting need for credit applicants to make a ten hours' journey to La Paz; (b) linked to the first shortcoming, the absence of a dynamic credit promotion campaign in an area where most producers did not have any knowledge of institutional credit; and (c) the opposition of middlemen--the traditional leaders and credit suppliers. 40. Since subloans granted during 1981-84 were medium- to long-term, with grace periods of up to three years, and at highly negative interest rates, all producers repaid their subloans before or soon after the grace period had expired, because the hyperinflation prevailing during the 1980-85 period had reduced drastically their real-terms debts. It was only 12/ Because of its superb quality, (illegally obtained) vicuiia fleece commands prices several times higher than the top alpaca quality. 13/ BAB argued that the small credit volume would not have justified the rather high fixed expenses (two permanent staff plus operational costs) of maintaining a field office. While this seems a valid argument, it opens the question of why this point was not considered during the project design stage.  _ 14 - in 1986 that interest rates were raised to be positive in real terms. Outstanding amounts of subloans for livestock developments granted during 1986 are equivalent to USS 0.1 million, and those of subloans granted by the State Bank for the handicraft development are equivalent to only US$ 11,000. Discussions with former project staff suggest that subloans were granted less on their economic and technical merits than as a result of local political pressure, and that not seldom the proceeds were used for purposes different from those envisaged in the original investment plans prepared by the project unit, which suggests lack of supervision. 41. Beneficiaries. Project benefits cannot be quantified as there is at present no reliable information on herd sizes, fiber production, or crop yields.14/ Nor are there satisfactory estimates of the numbers of project beneficiaries; they vary greatly according to the specific activity/ component under review, ranging from a handful in the case of the credit scheme to a substantial portion for the social infrastructure (where most effects will have a long gestation period), to almost the entire population if one looks at the benefits of the road or of improved husbandry practices. Disbursements and Costs 42. The PCR provides a global estimate of total project cost of US$ 18 million, or 77% of appraisal estimates. Disbursements totalled US$ 13.5 million of a total loan/credit amount of US$ 18 million. Equipment, materials, and vehicles accounted for US$ 6.3 million (47%) of disbursements, project administration/operation absorbed more than US$ 4.7 million (35%), and civil works nearly US$ 1.2 million (9%). Prolect Outcome 43. It is probably fair to say that the project has led to a change in perceptions among the project population (although the mission is aware of the possibility of response biases in the peasants with whom it had an occasion to talk), which could be reinforced and amplified by subsequent education, extension and training activities (inadequate as they may be at the present time), which together with the basic physical infrastructure provided under the project and future additions to it, could lead to a gradual betterment of living conditions and income levels in this distressed area. Whether the same result could have been achieved at a lower cost, or whether more could have been done under better macroeconomic circumstances, remains an unanswered question. The project has to be considered unsuccessful because its main productive components were not implemented.15/ 14/ It should be stressed that much of the project area lies at too high an altitude to permit agriculture, and is even beyond the range of the llama, so that the alpaca is the basis of the peasants' livelihood. 15/ The EEC stresses thp.t the project's failure to demonstrate a way to implement such a project and to institutionally support the national administration is as serious as its failure to achieve its productin- objectives.  - 15 - C. Omasuvos-Los Andes Management 44. The four-year project (1979-1983) was closed after seven years in December 1986. The project unit was quickly established in Achacachi (province of Omasuyos) and by April 1980 a full complement of staff was in place. The location of the project unit quickly became a problem which affected project implementation. The province of Omasuyos has traditionally been an area of unrest and a source of social and political pressures (The appraisal report indicated that the Project Unit should be located in either Achacachi or Huarina, both in Omasuyos province--para 4.02). Had the project unit been located in the neighboring province of Los Andes which had tio such tensions, project implementation would have been less affected by local pressures. As with the Ingav4 project (para. 24), frequent staff changes in IDRA, and of directors and key personnel of the Project Unit prevented any long-term planning throughout project life. The