Implementation Completion and Results Report

Document of
The World Bank

Report No:ICR00002995

IMPLEMENTATION COMPLETION AND RESULTS REPORT

(TF-90206,TF-90207)

ON A

GPOBA GRANT

IN THE AMOUNT OF US$ 5.18 MILLION

TO THE

PLURINATIONAL STATE OF BOLIVIA

FOR A

Decentralized Electricity for Universal AccessPROJECT

December24, 2013

Sustainable Development Department

Bolivia, Ecuador, Peru and Chile Country Management Unit

Latin America and the Caribbean Region

CURRENCY EQUIVALENTS

(Exchange Rate Effective October 5, 2013)

Currency Unit = Bolivianos (Bs.)

1.00 Bs. = US$ 0.14

US$ 1.00 = 6.96 Bs.

FISCAL YEAR

January 1—December 31

ABBREVIATIONS AND ACRONYMS

APL / Adaptable Program Loan / MTR / Mid term review
CAS / Country Assistance Strategy / MSC / Mid-term Service Contract
CO2 / Carbon dioxide / MSE / Micro and Small Enterprises
EA / Environmental Assessment / NPV / Net Present Value
GOB / Government of Bolivia / OBA / Output-Based Aid
GPOBA / Global Partnership on Output-Based Aid / O&M / Operation and Maintenance
ICT / Information and Communications Technology / OTM / Technical Office for Monitoring (OficinaTécnica de Monitoreo)
IDA / International development Agency / PAD / Project Appraisal Report
IDTR / Decentralized Infrastructure for Rural Transformation Project / PCU / Project Coordinating Unit
IERR / Internal Economic Rate of Return / PDO / Project Development Objectives
IP / Implementation Performance / PPP / Public-Private Partnership
IPDF / Indigenous Peoples Development Framework / PV / Photovoltaic
ISR / Implementation Status and Results Report / SHS / Solar Home System
KPI / Key Performance Indicators / TA / Technical Assistance
kWh / Kilo-Watt-hour / VMEEA / Vice Ministry for Electricity and Alternative Energy
M&E / Monitoring and Evaluation / Wp / Watt-peak
WTP / Willingness to Pay
Vice President: / Hasan Tuluy
Country Director: / Susan Goldmark
Sector Manager: / Malcolm Cosgrove-Davies
Project Team Leader: / Lucia Spinelli
ICR Team Leader / Enrique Crousillat
BOLIVIA
Decentralized Electricity for Universal Access
CONTENTS
Data Sheet
A. Basic Information
B. Key Dates
C. Ratings Summary
D. Sector and Theme Codes
E. Bank Staff
F. Results Framework Analysis
G. Ratings of Project Performance in ISRs
H. Restructuring
I. Disbursement Graph
1. Project Context, Development Objectives and Design
2. Key Factors Affecting Implementation and Outcomes
3. Assessment of Outcomes
4. Assessment of Risk to Development Outcome
5. Assessment of Bank and Borrower Performance
6. Lessons Learned
7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners
Annex 1. Project Costs and Financing
Annex 2. Outputs by Component
Annex 3. Economic and Financial Analysis
Annex 4. Bank Lending and Implementation Support/Supervision Processes
Annex 5. Beneficiary Survey Results
Annex 6. Stakeholder Workshop Report and Results
Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR
Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders
Annex 9. List of Supporting Documents

I N S E R T

D A T A S H E E T

H E R E

AFTER APPROVAL BY COUNTRY DIRECTOR

AN UPDATED DATA SHEET SHOULD BE INSERTED

MANUALLY IN HARD COPY

BEFORE SENDING A FINAL ICR TO THE PRINT SHOP.

NOTE: The Data Sheet is generated by the system

using the information entered in the Operations Portal

each time you use “Send Draft”, “Print” or “Submit Final” functions.

1. Project Context, Development Objectives and Design

1.The Decentralized Electricity for Universal Access Project, comprising two GPOBA Grants(TF090206 and TF090207) for a total ofUS$ 5.175 million was approved on July 2, 2007 and became effective on September 26, 2007. The original closing date —June 30, 2010— was extended three times throughout the implementation of the project. Final closing date was June 30, 2013.

1.1Context at Appraisal

2.With an average nominal income of $960 (2004) and a poverty rate in rural areas (where about 40 percent of the population lived) of 82 percent, Bolivia was at the time of appraisal the most economically challenged country in South America. Only 30% of Bolivia’s rural population had access to electricity, the second lowest in Latin America, and the majority of schools and health centers in rural areas were not electrified. The provision of infrastructure services to rural Bolivia was, therefore, an urgent necessity but also comparatively costly.Responding to this challenge, in 2003 the World Bank approved a US$20 million credit as the first phase of a 10-year Adaptable Program Loan (APL) package for Decentralized Infrastructure for Rural Transformation (IDTR), a major component of which focused on rural electrification. This project developed a pioneering model for off-grid rural electrification through Public-Private Partnerships (PPPs) – “Medium-Term Service Contracts” (MSCs) for the provision of electricity with solar systems for dispersed rural population whereby the service provider would be responsible for the operation and maintenance of the installations during the initial years. The GPOBA Decentralized Electricity for universal AccessProject assessed in this report built upon the experience of the IDTR project, while also aiming to enhance the output-based aid (OBA) approach, and the scheme’s sustainability.

3.Government policy. In 2006, the Vice Ministry for Electricity and Alternative Energy (VMEEA) launched a new national strategy for Universal Access (“Electricity to Live with Dignity”). The strategy recognized the magnitude of the universal access challenge and the need to mobilize both public and private sector financing and expertise. The Universal Access Fund has several funding windows for: (i) grid extension and densification, (ii) off-grid, and (iii) productive uses. At that moment, the VMEEA was interested in exploring Medium-Term Service Contracts as the leading mechanism for the off-grid window. The GPOBA project was proposed with view to help the Government mainstream the service contract approach into its new Universal Access Strategy. New service contracts would be implemented in 10 new areas throughout the country, with larger replication to be financed in the medium term by the Universal Access Fund, once operational. To this end, the Vice Ministry carried out a comprehensive analysis of the contracts and proposed modifications to: (i) adapt the model to new regions with different socio-economic conditions; (ii) increase implementation speed to contribute to faster scale-up; (iii) enhance focus on market development; (iv) improve accountability for outputs; and (iv) improve poverty targeting.

1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved)

4.The Project Development Objectives, as approved in 2007, were to increase affordable access to electricity in remote rural areas of Bolivia. To this end the project was expected to:

1. Extend sustainable electricity access to at least 7,000 households, micro-enterprises, schools and/or clinics in remote and dispersed rural areas using solar home systems at a subsidy of $650 per unit, and develop basic improved lighting solutions at a subsidy of $150 per unit for an additional estimated 2,000 very poor households (therefore directly benefiting an estimated 9,000 households, or 45,000 people).
2. Consolidate and mainstream a mechanism for electrification of dispersed households in Bolivia using output- based “Service Contracts” with private sector for the provision of electricity with decentralized, renewable energy technologies under the framework of the Government’s universal access policy.

5.Key outputs of the project were:

1. Installation of at least 7,000 solar home systems (SHS) for dispersed rural households, schools, clinics and micro and small enterprises (MSEs);
2. Distribution of an estimated 2,000 “Pico PV” systems for lighting and basic ICT (Information and Communications Technology) services for the poorest households (below 20Wp);
3. Provision of sustainable service by the SHS providers over four years (including maintenance visits, training, and market development activities);
4. Consolidation of output-based service contracts between government and private sector service providers as a key mechanism for the Government’s electrification program for dispersed households under the new Universal Access Policy.

1.3 Revised PDO(as approved by original approving authority) and Key Indicators, and reasons/justification

6.No changes were made to the PDO or any related performance indicators.

1.4 Main Beneficiaries

7.The beneficiaries of the project identified at the design stage were an estimated 45,000 low-income people (mostly indigenous) living in the rural dispersed areas of Bolivia. The project was aimed to address the needs of new areas not covered by the former IDTR project. In addition, it was foreseen that the poorest households who would not be able to afford the subsidized solar home systems fees would always benefit from the electrification of rural schools and health posts as well as from the multiplier effects of electricity in making possible rural business.

8.It was also expected that the project would benefit the private sector in enabling the creation of sustainable business operations through medium-term service contracts and the expansion of the market for renewable energy equipment. Also, central and departmental governments were expected to benefit through the achievement of their rural electrification targets.

9.While on a relatively limited scale, the project was expected also to provide global benefits through the reduction of greenhouse gas emissions.

1.5 Original Components (as approved)

10.The project’s components originally approved in 2007 were the following:

11.Component 1: New SHS service contracts (US$4.6m). The project was designedto fund output-based subsidies to private providers for market development activities, the sale, installation and after-sale service for solar home systems of different sizes for rural households, micro-enterprises, schools and clinics, for domestic, social and productive uses. At least 7,000 systems would be installed in the rural areas of five departments of Bolivia. The project was expected to bid out 10 new service contracts. Each area was to be awarded to the qualified bidder promising to service the largest number of users at given total subsidy per area. The contractors were expected to (i) market and install the systems, (ii) provide satisfactory service over four years, (iii) carry out at least one visit a year to each user, and (iv) provide specified market development services, and (v) perform basic monitoring and evaluation tasks. Although the subsidies would be channeled through the service providers, thesewere aimed to benefit the users through a reduction in the price of the panels and overall installation. As in IDTR, the actual user payments would be verified regularly via external audits and internal audits on random sample basis, with penalties for providers’ if overcharged.

12.GPOBA subsidy would cover 61% of project costs of a typical 50Wp system for households, and 69% of a typical school system. Larger systems for household or productive uses are not subsidized but would be also offered by the service providers. It was considered at appraisal that the funding for a typical 50Wp SHS would come from costumers and GPOBA subsidy. GPOBA will cover US$650 and costumers would pay: (i) upfront fee of about US$50; (ii) repayment of remaining system costs either in cash or through micro- credit of about US$335; and (iii) replacement of battery of about US$65 if battery fails before the project end. In addition, the user would be responsible for replacement of batteries and spare parts for the rest of the system’s operation, estimated at 20 years.User contribution was estimated based on a comprehensive demand survey, assessing willingness to pay and the IDTR experience. However, it was expected that these subsidies could be further reduced through the competitive bidding process.

13.Component 2: Pico-PV (US$0.3m): This sub-component consisted of a pilot project to facilitate the market development of pico-PV systems by providing incentives for participating companies proven to meet well specified project quality criteria. This new generation of PV systems below 20Wp offer basic lighting and communication services such as cell phone and battery charging, and radios. Such systems could be distributed via a broad range of existing sales points, without installations. These pico-PV systems are considered the most promising modern lighting application for the very poor households worldwide and the project aimed to introduce this solution to Bolivia. A target of 2,000 systems was considered at appraisal. However, it was contemplated that the final number of Pico PV would depend on market response and absorption capacity. Pico-PV subsidy was estimated to average $150 per unit, covering market development costs as well as about half of an average pico-PV system price, but given the supply and demand uncertainties, the subsidy amount would be adjusted as needed during implementation.

14.The introduction of the pico-PV pilot aims at developing affordable lighting solutions for the lowest income strata, which cannot afford the subsidized SHS. The pilot built on successful experience of similar systems in Asia, but this was their introduction in Bolivia and Latin America. It was recognized, however, that pico-PV offers much more limited energy service compared to SHS and is therefore not a substitute for a SHS. This compound had a limited budget relative to the SHS because: (i) there was limited experience with supply of these systems worldwide and none in Bolivia; (ii) the response of the market to these new products was not known; (iii) it takes time to build markets for a totally new product; and (iv) the systems provide service at a level below the minimum required under the Government’s Universal Access Program of 20 Watts, therefore the role of such systems still needs to be defined.

15.Component 3: Transaction Support Technical Assistance (TA) (US$ 0.275 million)

(3.1) External audits and Output Monitoring and Verification (M&E): Component to finance external technical and financial audits of the project (US$0.1mn over three years) and of the providers by an independent auditor, as well as the monitoring and verification of outputs over a three- year period (US$0.3mn). The monitoring and audits are relatively costly due to high logistics costs of remote areas with dispersed population.

Pico PV

(3.2) Recipient-executed Technical Assistance for Transaction Support, Coordination and Supervision: While the majority of the technical assistance was covered by a specific Bank-executed TA component previously awarded by GPOBA in 2006, a further US$100,000 was earmarked for activities directly related to the coordination and supervision, which are executed by the recipient. These funds needed to finance the staff required for the project implementation in the event that the IDTR project (and its PCU) would close earlier than the proposed GPOBA activity.

16.The provision of TA was consistent with emerging best practices for off-grid electrification, which indicate that successful projects require substantial accompanying TA resources for both project transactions and for supervision. Experience showed that the investment in TA and capacity building can be offset by a reduction in subsidy requirements. For example, successful bidding transactions in the IDTR project resulted in subsidy savings equivalent to US$2.5 million, compared to about US$600k in preparation and transaction TA.

1.6 Revised Components

17.While the nature of the project components remained unchanged, savings stemming from a successful bidding for SHS made possible an increase in the total number of SHS and Pico-PV systems delivered, as well as an increase in the allocation of funds for the TA component. In practice, this involved the incorporation of 1,650 Pico-PV systems into the bidding process (and budget) of Component 1, together with 126 solar systems for public buildings (schools) that were not contemplated initially.A final reallocation of loan proceeds was formally agreed in March 2013 as part of the third project re-structuring, as shown in the following table.

Funds allocation (in US$)
Original allocation / Amendment of March 2013
TF090207 / TF090206 / TF090207 / TF090206
Component 1 / 1,380,000 / 3,220,000 / 1,380,000 / 3,220,000
Component 2 / 90,000 / 210,000 / 90,000 / 103,805
Component 3 / 82,500 / 192,500 / 82,500 / 298,695
Total / 1,552,500 / 3,622,500 / 1,552,500 / 3,622,500

1.7 Other significant changes

18.The original closing date of the project was June 30, 2010. In October 2007, legal documents were amended toadapt them to the new organization of the Government of Bolivia (i.e. to move the PCU, including its staff, from the Ministry of Public Works, Services and Housing to the Ministry of Hydrocarbons and Energy). Further amendments extended the project’s closing date in three occasions till June 30, 2011, March 31, 2013 and June 30, 2013 respectively. Also, the design of the project was amended twice in order to reduce the service contract period for SHS (starting immediately after installations) from four years to three years and, then, two years.

2. Key Factors Affecting Implementation and Outcomes

2.1 Project Preparation, Design and Quality at Entry

19.The project was designed following the Output-Base Aid (OBA) approach of its predecessor IDTR, approved in 2003. It incorporated general lessons on service contract design as well as lessons specific to the IDTR transaction that was under implementation. These included the importance of market information in ensuring the sustainability of SHS systems, awareness of the high risks associated to long term exposure of providers, the need for transparent tender processes, the potential of SMEs in ensuring long-term sustainability and the importance of setting up an adequate incentives system for market development as well as establishing an accurate monitoring system.

20.The project was designed to maximize the benefit of available resources by adopting an OBA through which: (i) payments to private or public operators were made against functioning services; (ii) the bidding process aimed selecting a responsive bidder that minimized the use of the subsidy per unit, thus maximizing the number of installations. By that time, this approach was being tested in the IDTR project and had been applied in other Bank projects elsewhere.

21.Although at the time of project preparation the Bank had accumulated considerable experience in the implementation of rural electrification involving private-sector participation, the project assessment recognized that a project in Bolivia in general, and an intervention in rural areas in particular, could confront a set of exogenous obstacles. Economic and/or social instability were rated as high risk at appraisal, as well as political interference that could affect project implementation and the withdrawal of support to PPPs in infrastructure. These risks (that could not be mitigated by the project itself) were correctly diagnosed at appraisal and, to greater or lesser degrees, affected project implementation. Given this situation, attracting private-sector participation as service providers was indentified as a main challenge. In hindsight, it can be argued that, given the perception of high risk associated to external factors, the original project design that required a four-year commitment of SHS service providers after installations was overly optimistic.