Document of The World Bankdocuments.worldbank.org/curated/en/226321468082127994/pdf/ICR… · Gazelle Peninsula with government services similar to those that were destroyed by the

Document of The World Bank

Report No: ICR0000861

IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-45250)

ON A

LOAN

IN THE AMOUNT OF US$25.26 MILLION

TO THE

INDEPENDENT STATE OF PAPUA NEW GUINEA

FOR A

SECOND GAZELLE RESTORATION PROJECT

December 19, 2008 Urban Development Sector Unit Sustainable Development Department East Asia and Pacific Region

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

CURRENCY EQUIVALENTS

(Exchange rate effective September 1, 2008)

Currency unit = kina Kina 1.00 = US$0.39 US$1.00 = kina 2.56

FISCAL YEAR

January 1 – December 31

ABBREVIATIONS AND ACRONYMS

ADB Asian Development Bank

APL adaptable program loan

AusAID Australian Agency for International Development

CCI Cocoa Coconut Institute

CDP Comprehensive Development Plan

CSTB Central Supply & Tenders Board

DO development objectives

EDF European Development Fund (of the EU)

ENBPA East New Britain Provincial Administration

EU European Union

FY fiscal year

GoPNG Government of Papua New Guinea

GRA Gazelle Restoration Authority

GRA/PIU GRA Project Implementation Unit

ICR implementation completion report

IP implementation progress

ISR implementation status and results report

ITRP Immediate Term Restoration Program of GRA

KPI key performance indicator

MOAs Memoranda of agreement

MTRP Medium Term Restoration Program of GRA

NBC National Broadcasting Corporation

NEC National Executive Council

O&M operations and maintenance

PAD project appraisal document

PDO project development objective

PIP project implementation plan

PIU project implementation unit

PMR project management report

PNG Papua New Guinea

PNGSDP PNG Sustainable Development Program

PPII Provincial Performance Improvement Initiative

QER quality enhancement review

QAG Quality Assurance Group

RAMS Road Asset Management System

RAP resettlement action plan

RVO Rabaul Vulcanological Observatory

WHO World Health Organization

Vice President: James W. Adams, EAPVP

Country Director: Nigel Roberts, EACNF

Sector Manager: Keshav Varma, EASUR

Project Team Leader: Thakoor Persaud, EASTE

ICR Team Leader: Thakoor Persaud, EASTE

PAPUA NEW GUINEA Second Gazelle Restoration Project

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 ............................................... 12 Key Factors Affecting Implementation and Outcomes .............................................. 63 Assessment of Outcomes .......................................................................................... 144 Assessment of Risk to Development Outcome ......................................................... 175 Assessment of Bank and Borrower Performance ..................................................... 186 Lessons Learned ........................................................................................................ 217 Comments on Issues Raised by Borrower/Implementing Agencies/Partners ........... 22Annex 1. Project Costs and Financing .............................................................................. 23Annex 2. Outputs by Component...................................................................................... 26Annex 3. Economic and Financial Analysis ..................................................................... 29Annex 4. Bank Lending and Implementation Support/Supervision Processes ................. 30Annex 5. Beneficiary Survey Results ............................................................................... 32Annex 6. Stakeholder Workshop Report and Results ....................................................... 33Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ......................... 34Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ........................... 38Annex 9. List of Supporting Documents .......................................................................... 39 MAP

i

A. Basic Information

Country: Papua New Guinea Project Name: Second Gazelle Restoration Project

Project ID: P054238 L/C/TF Number(s): IBRD-45250

ICR Date: 12/22/2008 ICR Type: Core ICR

Lending Instrument: APL Borrower: INDEPENDENT STATE OF PNG

Original Total Commitment:

USD 25.3M Disbursed Amount: USD 25.3M

Revised Amount: USD 0.0M

Environmental Category: B

Implementing Agencies: Gazelle Restoration Authority Cofinanciers and Other External Partners: Australian Agency for International Develoopment (AusAID) European Union (EU) B. Key Dates

Process Date Process Original Date Revised / Actual

Date(s)

Concept Review: 01/20/1998 Effectiveness: 04/18/2000 04/18/2000

Appraisal: 04/05/1999 Restructuring(s):

05/12/2001 09/26/2002 06/30/2005 12/19/2006

Approval: 12/09/1999 Mid-term Review:

Closing: 06/30/2003 12/31/2007 C. Ratings Summary C.1 Performance Rating by ICR

Outcomes: Satisfactory

Risk to Development Outcome: Moderate

Bank Performance: Satisfactory

Borrower Performance: Satisfactory

C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings

Quality at Entry: Satisfactory Government: Satisfactory

Quality of Supervision: Highly Satisfactory Implementing Agency/Agencies:

Highly Satisfactory

Overall Bank Satisfactory Overall Borrower Satisfactory

ii

Performance: Performance: C.3 Quality at Entry and Implementation Performance Indicators

Implementation Performance

Indicators QAG Assessments

(if any) Rating

Potential Problem Project at any time (Yes/No):

Yes Quality at Entry (QEA):

None

Problem Project at any time (Yes/No):

Yes Quality of Supervision (QSA):

Satisfactory

DO rating before Closing/Inactive status:

Satisfactory

D. Sector and Theme Codes

Original Actual

Sector Code (as % of total Bank financing)

General public administration sector 16

General transportation sector 29 19

General water, sanitation and flood protection sector 21 30

Other social services 17 47

Power 17 4

Theme Code (as % of total Bank financing)

Access to urban services and housing

Infrastructure services for private sector development

Natural disaster management

Other social development E. Bank Staff

Positions At ICR At Approval

Vice President: James W. Adams Jean-Michel Severino

Country Director: Nigel Roberts Klaus Rohland

Sector Manager: Keshav Varma Keshav Varma

Project Team Leader: Charles M. Feinstein Stuart Whitehead

ICR Team Leader: Thakoor Persaud

ICR Primary Author: Thakoor Persaud

iii

F. Results Framework Analysis

Project Development Objectives (from Project Appraisal Document) The objectives of the project are to assist the Borrower to (i) provide to the people of the Gazelle Peninsula with government services similar to those that were destroyed by the volcanic eruption in 1994; and (ii) ensure sustainability of social and economic well-being ad minimize adverse impacts of future volcanic eruptions on lives, property and livelihood in the Gazelle Peninsula. Revised Project Development Objectives (as approved by original approving authority) No formal revisions (a) PDO Indicator(s)

Indicator Baseline Value

Original Target Values (from

approval documents)

Formally Revised Target Values

Actual Value Achieved at

Completion or Target Years

Indicator 1 : Completion of infrastructure and services and at least an 80% occupancy rate at all relocation sites

Value quantitative or Qualitative)

Four relocation sites identified but require additional planning, land acquisition and allocation, and basic infrastructure and services. About half of displaced population living at relocation sites.

Electricity, water, sanitation, and health services provided. Registration of lands allocated to displaced communities/families. At least 80% occupation of relocation sites.

Electricity, water, and health center completed. Study of sanitation options completed. Lands registered. Site occupation exceeds 80%.

Date achieved 10/01/1999 06/30/2003 12/31/2007 Comments (incl. % achievement)

Targets were fully achieved between April and December 2006, with the exception of sanitation services, for which a study was produced of recommended options for future investments.

Indicator 2 : Completion of port-related protection works in Rabaul and approval of revised zoning plan that prohibits redevelopment in the devastated areas of Rabaul


Port-related construction works not yet begun; zoning law not approved.

Port road reconstruction, erosion control, silt traps, and dredging of ports completed; zoning plan approved.

Port-related construction works completed; zoning plan approved.


Completion dates: port roads 12/01; erosion control 5/04; silt traps 04/07; dredging 9/05; zoning plan 2002.

iv

Indicator 3 : 75% completion of other infrastructure, utilities & social services under phase I, including Rabaul/Kokopo Road (except Vulcan section)


Other phase 1 infrastructure, utilities, and services not started.

75% completion. 100% of revised phase 1 program complete

Date achieved 10/01/1999 06/30/2003 12/31/2007

Comments (incl. % achievement)

Revised phase 1 program, as modified in June 2005 & December 2006 amendments, was fully achieved. Some investments were postponed, assumed by other parties, or cancelled (without compromising achievement of the PDO)-Ref. main text & Annex 1

Indicator 4 : Formulation of spatial and economic strategy framework and institutional plan for longer-term development as well as plans for costal zone management and mitigation of coastal hazards in ENB


0% complete Institutional strategy formulated

Institutional strategy formulated


Indicator 5 : Approval by ENBPA of detailed arrangements to integrate GRA into the Provincial Government


Not started

Detailed arrangements approved by ENBPA

Detailed arrangements agreed and formal approval process underway


(b) Intermediate Outcome Indicator(s)

Indicator Baseline Value

Original Target Values (from

approval documents)

Formally Revised

Target Values

Actual Value Achieved at

Completion or Target Years

Indicator 1 : Rabaul Port-related roads and key national highways in area restored and upgraded.

Value (quantitative or Qualitative)

Roads not restored and upgraded after damage caused by eruption.

Roads restored and upgraded

Roads restored and upgraded


v

G. Ratings of Project Performance in ISRs

No. Date ISR Archived

DO IP Actual

Disbursements (USD millions)

1 01/05/2000 Satisfactory Satisfactory 0.00 2 03/31/2000 Satisfactory Satisfactory 0.00 3 11/27/2000 Satisfactory Satisfactory 0.25 4 05/25/2001 Satisfactory Satisfactory 2.87 5 06/22/2001 Satisfactory Satisfactory 3.31 6 12/04/2001 Satisfactory Satisfactory 3.82 7 03/08/2002 Satisfactory Unsatisfactory 4.58 8 06/10/2002 Satisfactory Satisfactory 4.73 9 08/12/2002 Satisfactory Unsatisfactory 5.02

10 10/30/2002 Satisfactory Unsatisfactory 5.14 11 06/04/2003 Satisfactory Satisfactory 5.59 12 12/24/2003 Satisfactory Unsatisfactory 6.82 13 06/28/2004 Unsatisfactory Unsatisfactory 9.13 14 12/22/2004 Unsatisfactory Unsatisfactory 11.95 15 04/22/2005 Moderately Satisfactory Moderately Satisfactory 12.62 16 01/09/2006 Moderately Satisfactory Moderately Satisfactory 17.29 17 03/08/2006 Satisfactory Satisfactory 17.91 18 10/10/2006 Satisfactory Satisfactory 19.87 19 08/23/2007 Satisfactory Satisfactory 25.26 20 08/05/2008 Satisfactory Satisfactory 25.26

H. Restructuring (if any)

Restructuring Date(s)

Board Approved

PDO Change

ISR Ratings at Restructuring

Amount Disbursed at

Restructuring in USD millions

Reason for Restructuring & Key Changes Made

DO IP

05/12/2001 S S 2.87

09/26/2002 N S U 5.02

Reallocate loan proceeds and increase financing percentages in response to (a) redistribute project savings achieved by AusAID providing grant funding for a major road component and (b) escalating project costs and counterpart funding shortfalls.

06/30/2005 N MS MS 13.12 Reduced scope of remaining components (due to escalating costs and upgraded designs of

vi

Restructuring Date(s)

Board Approved

PDO Change

ISR Ratings at Restructuring

Amount Disbursed at

Restructuring in USD millions

Reason for Restructuring & Key Changes Made

DO IP

some subcomponents) and reallocated loan proceeds accordingly, while maintaining development objectives. Specifically: (a) upgrading of Kokopo Town solid waste disposal services replaced by a feasibility study, (b) construction of a government administration center cancelled, (c) rehabilitation of storm water drains in Rabaul Port cancelled, (d) studies on air quality and environmental health from continued emissions and research on development of various agriculture sectors cancelled. The footnotes of Annex 1 provide more details.

12/19/2006 N S S 21.13

Reallocate disbursement categories due to loan savings in some subcomponents and provision of new EU funding that freed Bank funds previously earmarked for services in relocation sites. Also reduced scope of phase 1 program by deleting Kokopo Town facility for women's activities and rehabilitation of Rabaul Port town clinic, which were inadvertently omitted from previous amendment.

vii

I. Disbursement Profile

1

1 PROJECT CONTEXT, DEVELOPMENT OBJECTIVES AND DESIGN

1.1 Context at Appraisal

1.1.1 After about 50 years of dormancy, two volcanoes in Rabaul, Mt. Tavurvur and Mt. Vulcan, erupted in September1994 for the third time in 100 years. Almost all of the physical, social, and economic infrastructure and facilities in Rabaul Town and surrounding areas were completely destroyed, severely damaged, or ended up being located within the hazardous areas after the eruption.

1.1.2 Deaths and injuries were relatively low (about five deaths and dozens injured) partly because residents and officials were aware of the threat and took appropriate actions early. However, there were significant social and economic disruptions and losses, with about 30,000 residents losing their homes and about two-thirds of them seeking permanent resettlement away from hazardous areas. Estimates compiled by the Australian Agency for International Development (AusAID) soon after the eruption placed public asset damage and losses at about US$280 million, along with similar losses for private assets such as homes, businesses, inventory, etc.

1.1.3 Both the national government (GoPNG) and the East New Britain Provincial Administration (ENBPA) acted quickly to move people from harm’s way and provide temporary shelter and services to displaced residents. They also discouraged residents from reoccupying the evacuated sites by implementing enforcement measures and offering incentives such as promising to provide safer resettlement lands and to create a new town in Kokopo. Residents were helped in demolition and storage of whatever material and belongings they could salvage from the affected areas and those who had access to their farming plots continued to use them. Many people continued to have their traditional burial ceremonies in Rabaul, an extremely important sociocultural practice which the authorities continued to allow.

1.1.4 In 1995, the GoPNG commenced its Immediate Term Restoration Program (ITRP), supported by about US$47 million from donors and lenders including AusAID, the European Union (EU), Japan, Germany, the Asian Development Bank (ADB), and the World Bank (with funds diverted from other ongoing activities). Early investments were primarily directed toward improving the immediate socioeconomic conditions of the displaced population and planning for longer term investments and development efforts. In light of the extensive damage to the town of Rabaul and the continuing existence of hazardous conditions (ashfall, toxic gases, and tremors), the national and provincial authorities decided to resettle the affected population away from the existing town. They identified prospective areas for new permanent settlements throughout the northeast Gazelle Peninsula and developed plans to build a replacement town at Kokopo, a small village about 17 miles east of Rabaul.

1.1.5 To carry out this transition, the government proposed a seven-year, Medium Term Restoration Program (MTRP) that would be implemented in two phases between 1999 and 2005. The Bank’s project appraisal document (PAD) notes that “during MTRP Phase I (1999–2002), the primary focus will be to complete the emergency restoration works, relocate displaced communities and provide services in and around the newly designated provincial capital at Kokopo.” The second Phase (2002–05) was to focus on continuing implementation of these

2

plans and consolidation of the institutional framework for planning and executing urban/regional development priorities with the integration of the Gazelle Restoration Authority (GRA) into the ENBPA.

1.1.6 The Bank’s Second Gazelle Restoration Project was designed as an Adaptable Program Loan (APL) with two phases to support implementation of the MRTP by (a) helping complete the emergency restoration works started under the ITRP and (b) facilitating the return to more normalized development planning by the provincial government.

1.2 Original Project Development Objectives (PDO) and Key Indicators

1.2.1 While the project development objectives in the PAD and the loan agreement are not identical, they are very similar. The objectives stated in the loan agreement are to “assist the Borrower to: (i) provide to the people of the Gazelle Peninsula with government services similar to those that were destroyed by the volcanic eruption of 1994; and (ii) ensure sustainability of social and economic well-being and minimize adverse impacts of future volcanic eruptions on lives, property and livelihood in the Gazelle Peninsula.”

1.2.2 The PAD sometimes frames the PDO in terms of supporting the government’s MTRP program, which is essentially the same as the project, and summarizes the project’s key performance indicators (KPIs) as follows:

Medium Term Restoration Program

Phase I (stabilization)

Phase II (consolidation)

Years 1999–2002 2002–2005

Development Objective Completion of Phase I and

institutional transition Normalization to longer-term

development process

Key Program Outputs:

Benchmarks for subsequent Adaptable Loan and Grant Financing

Completion of infrastructure and services and at least 80 percent occupancy at all relocation sites.

100% completion of port-related protection works in Rabaul and approval of revised zoning plan that prohibits redevelopment in the devastated areas of Rabaul.

75% completion of other infrastructure, utilities and social services under Phase I, including Rabaul/Kokopo Road (except for Vulcan section).

Formulation of spatial and economic strategy framework and institutional plan for longer-term development as well as plans for coastal zone management and mitigation of coastal hazards in East New Britain Province.

Approval by ENBPA of detailed arrangement to integrate GRA into the Provincial Government.

Restoration of health and education services, water supply and sanitation service to affected population.

Provision of non-project services (including passenger transport, etc.) to resettled communities.

Reduction in erosion of caldera and sedimentation in Rabaul Port.

Full occupation at relocation sties that provide sustainable livelihood for displaced families.

Consolidation of private sector economic activities in new development areas.

Establishment of Kokopo as regional government center with major increase in population.

Implementation of longer-term development planning reflecting economic, natural, and human resources in province.

Full ENBPA funding of warranted O&M costs on assets restored under ITRP and MTRP.

3

1.2.3 As explained later in this report, after achieving the triggers for APL2, the government decided against implementing the second phase. This complicated the assessment of whether the project met its end-of-project KPIs (which were also defined in qualitative rather than quantitative terms). For the purposes of the ICR, the achievements of APL1 are measured against the targets it was intended to achieve to trigger APL2. It is important to note that despite not implementing APL2, the project still managed to achieve a substantial portion of the KPIs to be achieved at the end of the second phase.

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

1.1.1 No revisions were made to the original PDO. However, as discussed later in the ICR, there were three extensions to the original loan closing date and four formal restructurings. These restructurings and extensions were required in large part because of delays and changes in promised financing by both the counterpart and assistance from other donors and lenders, as well as upgrading of some subcomponent designs, higher than expected bid prices, and inflation during the implementation period. As a result, adjustments had to be made in the work program and sequence of planned investments under APL1 as well as to the allocation of loan proceeds to ensure continued implementation progress.

1.4 Main Beneficiaries

1.4.1 The PAD indicates that the APL1 benefits of “normalized basic infrastructure provision, utility and social services” were targeted to “the population most directly affected by the volcanic eruption including, in particular, about 19,000 displaced inhabitants.”

1.5 Original Components

1.5.1 The components in the PAD and the legal agreement have slightly different names and are organized into different subparts, but include the same investments and activities. The subcomponents were defined in more general terms in the PAD but were described in the legal agreement in terms of specific investments or activities. The components and subcomponents in the PAD appear in the cost table in Annex 1 of the ICR and the ratings section of the ISRs, but the parts of the project described in the loan agreement are used to reference disbursement allocation categories and for loan reallocations, amendments, and restructuring.

1.5.2 Phase I (APL I), as defined in the loan agreement and discussed in the ICR, comprised the following:

Part A: Displaced Communities

1. Relocation of communities displaced by the volcanic eruption in 1994, through provision of relocation sites, agricultural land, internal site roads, electricity, water supply, and sanitation.

2. Provision of services for community development and economic support in the relocation sites under Part A (1) above, through agricultural extension services, improved access to schools, vocational and skills training, health and police services.

3. Resettlement and rehabilitation of Affected Persons.

4

Part B: Kokopo Town

Upgrading of Kokopo Town through: (i) restoration of the road between Rabaul Port and Kokopo Town, (ii) restoration of the Burmah Road, (iii) restoration of roads and drains within Kokopo Town, (iv) upgrading the solid waste disposal facilities, (v) provision of back-up generators for essential public service facilities, (vi) upgrading facilities at the Butuwin health center, (vii) construction of an elementary school and facility for women’s community activities, (viii) a government administration center and a commercial center, and (ix) assessment of human waste treatment options in Kokopo Town. Part C: Rabaul Port

1. Rehabilitation of port-related infrastructure and facilities at Rabaul Port through: (i) restoration and upgrading of Ah Tam and Dawapia wharves, (ii) provision of erosion control, and (iii) rehabilitation of Blanche Street and Kokopo Road and storm water drains.

2. Restoration of essential utility, infrastructure and public services at Rabaul Town/Port, through: (i) rehabilitation of mains water supply, (ii) rehabilitation of storm water drains , (iii) construction of market facilities, (iv) restoration of education facilities, (v) rehabilitation of the town clinic, (vi) provision of back-up electrical power generators for essential public service facilities, and (vii) facility for women’s community activities. Part D: Gazelle Peninsula

Establishment of a less concentrated human settlement pattern in the Gazelle Peninsula, through development of provincial townships in the North East area of the Gazelle Peninsula, including (i) restoration of the road from Baliora to Warangoi Bridge, (ii) provision of water supply at Baliora, (iii) expansion of existing high schools and health centers, (iv) construction of a facility for women's community activities, and (v) establishment and operation of the Village Development Fund. Part E: Economic Planning and Diversification

Support to develop institutional capacity in the areas of land use, water resources and wastewater management and protection of coastal zone resources, including: (i) carrying out of a study for a longer-term economic and spatial framework and implementation strategy for the North East Gazelle Peninsula, including preparation of a management plan and a hazards mitigation plan for the coastal zone area thereof; (ii) carrying out of a study of air quality and environmental health effects of ongoing ash emissions from Mt. Tavurvur; (iii) carrying out of research into cash and food crops, livestock and fisheries development, cocoa and coffee, and expansion of the Cocoa and Coconut Extension Agency; (iv) development of an agricultural diversification strategy; (v) provision of an incinerator for the Department of Agriculture and Livestock; and (vi) provision of consultants’ services for the foregoing.

Carrying out of a program of institutional strengthening for the Administration of the East New Britain Government, including planning and development, staff training and provision of consultants’ services.

5

Strengthening of GRA’s accounting, financial management and monitoring and reporting capabilities through development of a computerized accounting system, including provision of computer hardware and software, staff training and consultants’ services.

1.5.3 Support from other donors was aimed at (a) establishing sustainable livelihood for displaced communities by providing basic infrastructure, including roads, electric power and water supply as well as education and health facilities in relocation sites for displaced communities; and (b) providing health, education and public safety facilities for the rural population of the Gazelle Peninsula.

1.6 Revised Components

1.6.1 During implementation of Phase I of the MTRP, it became clear that the sequence of several activities had to be changed (e.g. agriculture and jobs in settlement areas) and a growing consensus emerged for targeting some Phase II priority investments at expanding economic activity, livelihood and social development more broadly throughout the province. 1.6.2 As discussed below, revisions were made to some components, both in the quantity of works done as well as the time within which they were done. This was due primarily to higher than anticipated cost, design changes and early problems with government counterpart and EU funding arrangements.

1.7 Other significant changes

1.7.1 After counterpart funding delays began to affect implementation, the Bank and counterpart staff got AusAID’s agreement for it to finance the Kokopo/Rabaul Road Rehabilitation Component (the largest project component). This reduced overall counterpart funding requirements by an estimated 5.2 million kina (almost US$2 million), but AusAID also required a complete engineering redesign which delayed start of construction works by about two years.

1.7.2 Delays arising from counterpart and EU funding, difficulty in finding qualified contractors, design creep, and overall inflation contributed to substantial cost increases for many components. Extensive design changes1 resulted primarily from the frequent stakeholder consultations which were maintained with the affected communities.2 The PNG Central Supply & Tenders Board (CSTB) also added to delays when GRA had to modify its system in line with new CSTB financial and procurement procedures.

1. By February 2005, the estimated cost for the original program increased to more than double appraisal estimate due largely to major design creep: Bank components (excluding Kokopo/Rabaul Road) up by 80%; EU components up by 300%.

2. While it was important to consult with stakeholders to such matters, one drawback is that they tend to demand higher than appropriate/affordable levels of service and standards.

6

2 KEY FACTORS AFFECTING IMPLEMENTATION AND OUTCOMES

2.1 Project Preparation, Design and Quality at Entry

2.1.1 The Quality Assurance Group (QAG) undertook a Quality Enhancement Review (QER) prior to loan negotiations at the Region’s request. It found the project to be well-designed and prepared and agreed with the positive assessment of GRA. The QER saw GRA as a model for decentralized development in other areas and recommended assistance for participatory local development benefiting host communities in the vicinity of relocation sites for displaced communities. Responding to these recommendations, at negotiations, a Village Development Fund component was added to the project.

2.1.2 The government’s strategy for restoration called for a shift in the focus of settlement and economic activity away from the high risk areas in and around Rabaul. Project components were selected for their expected contribution to achieving the priority MTRP government objectives. Given the range and frequency of natural hazards affecting PNG, the project strategy remains highly relevant to government’s priorities to reduce vulnerability and risk and is consistent with the Bank’s strategy.

2.1.3 Lessons of earlier projects taken into account in project design. Several key lessons noted from the earlier ITRP phase included (a) grounding project planning in understanding of the socioeconomic context and population dynamics; (b) seeking greater stakeholder consultation/participation including the business community, displaced villagers, and host communities at the relocation sites; (c) care in assessing location and characteristics of relocation sites to support sustainable livelihood; and (d) broader skills-base in the GRA/PIU to ensure robust justification for development proposals.

2.1.4 Unlike many emergency projects which stop at the reconstruction or resettlement phase, this project had the APL1 to address the short-medium term needs while the APL2 was to have a long-term focus. Even without the APL2, this added project continuity as well as relatively greater rigor in assessing investment needs, sustainability, etc.

2.1.5 Adequacy of Safeguard Policies. The key safeguard policies were related to appropriate and timely land acquisition and possible adverse social and environmental impacts. Timely location and acquisition of suitable safe and desirable resettlement sites after the eruption was a significant challenge. Keeping families from returning to their original plots and maintaining similar village and group affiliations and socio-cultural interactions as much as possible added to the challenge, as did the task of quickly providing basic services, including water and sanitation. With most of the land communally owned and owners of freehold land not willing to sell their land (or asking very high prices), a coconut plantation on the coast at Vunacambi was purchased by ENBPA for occupation mainly by fisherman from the area around Mt. Vulcan. This allowed them to continue much of their prior fishing activities and helped to reduce adaptation problems at an inland relocation site.

2.1.6 In the process of acquiring the Vunakambi Plantation, the 19 existing workers and their families had to be relocated. This was done in accordance with Bank guidelines for resettlement and a Resettlement Action Plan (RAP) was prepared and implemented to facilitate this process.

7

For those affected by the volcano, the Bank’s policy on involuntary resettlement (Operational Directive 4.30) did not apply because resettlement under the project was necessitated by a natural disaster. However, in most cases, the authorities voluntarily took several steps to ensure that preferences and community feedback were taken into account in the resettlement process. Especially challenging was the fact that some of the affected families (e.g. those in Gelagela) had previously had large rural plots and expected similar land in urban or peri-urban areas. The ENBPA worked with community representatives to locate additional sites for subsistence farming and also to give residents the option of moving to such areas if they wanted larger plots.

2.1.7 The project has an environmental assessment B category with the primary focus being upon possible adverse impacts from construction and resettlement activities. As a condition of disbursement, several components required environmental mitigation measures and safeguards that also emphasized operation and maintenance issues, e.g., (i) the Rabaul/Kokopo Road (subsequently financed and implemented by AusAID); (ii) Kokopo and Rabaul urban drainage; (iii) Rabaul Port erosion control; land-use management, water supply, sanitation and solid waste components in the relocation sites (financed by EU); and (iv) all urban solid waste and sanitation components. It is not easy to predict future volcanic activity and the impact upon the population in terms of ashfall, air quality, and related hazards so these were also noted as environmental issues to be taken into account to the degree possible.

2.1.8 The shift of settlement and economic activity away from high-risk areas was expected to concentrate development in and around Kokopo, which was just a village of about 1,000 people prior to the eruption. Planning studies were therefore done to help guide development in an environmentally-sustainable way. These include: (i) a study of the Kokopo/Tokua corridor; (ii) a Disaster Risk Assessment Study focusing on natural hazards including in the Kokopo/Tokua corridor and (iii) Economic and Spatial plans for the Gazelle area.

2.1.9 Mitigating Project Risks. Resettling the affected population away from the high risk areas was the main action taken to mitigate the risks identified. Also important was discouraging residents from returning to their land in hazardous areas and ensuring that no new settlers moved into the vacated lands. This was accomplished by providing alternative lands, educating the population on the dangers of trying to return and assisting the residents to dismantle and remove what material they could salvage from their previous land. Arrangements were made to constantly monitor seismic activities and air quality so that the population could be appropriately advised. A methodology and study design was prepared with AusAID’s assistance for a comprehensive Air Quality Study to be conducted during project implementation, but because of higher than anticipated ashfall and funding delay from AusAID, only some basic study designs were done. A World Health Organization (WHO) study is currently underway. Preliminary results indicate that while there are no immediate health risks, the long-term effects, especially on children, need to be studied.

2.1.10 To minimize the impact of settling so many residents in a new area, the authorities worked out a plan under which they maintained open consultation with representatives of the affected communities. The host communities also received special attention. They participated actively in the decision-making process and they shared in several of the benefits provided to the affected population (e.g. water and sanitation service, schools, improved roads, electricity, etc.).

8

2.2 Implementation

2.2.1 Project works were subject to such hazards and problems as counterpart funding delay, unseasonal rainstorms, equipment and material being stolen (before and after being installed) and onsite workers being attacked and injured (and in one case, killed) by robbers. Early implementation progress (IP) and achievement of development objectives (DO) were rated unsatisfactory primarily because of a slow start-up process. After the government’s initial counterpart advance to the trust account, required for loan effectiveness, its annual counterpart funding3 consistently fell well below program requirements, reflecting national budget constraints during a severe economic downturn. Delays in review and approval of parallel donor funding also had a negative impact on other downstream project activities. By the time the project was restructured in 2005 the economy was improving and so was the flow of counterpart funding.

2.2.2 The impact of initial counterpart funding problem on the EU components was more pronounced. This was because the EU’s European Development Fund (EDF) funding was available mainly for construction costs, with most of the design costs to be financed early by scarce government counterpart funding. As the designs were delayed, the EDF window closed in September 2003 with few of the EU-financed components actually implemented. Without these components in place, several other donor-lending complementary investments could not be made.

2.2.3 Even with the initial restructuring, counterpart funding continued to fall well short of program requirements. For a while, GRA was forced to stop awarding contracts. To help avoid a complete suspension of all project works, in late 2002 the Bank agreed to a government request to increase its financing percentage for civil works from 78 percent (and in one case 33 percent) to 90 percent. This was possible because AusAID agreed to finance the Kokopo/Rabaul Road as mentioned above. EU funding under the EDF7 and EDF8 financing agreements expired in September 2003, leaving many components, including those for basic services for resettled communities, unfunded. Funds from the EDF9 program for about €3.4 million (for a reduced program of activities) did not become available until July 2005 and works had to be rescheduled

2.2.4 The government submitted a new €2 million financing request under the EU’s streamlined procedures and much later, it received funding to complete works on 14 of 20 original components plus additional items in relocation sites and host communities. Unfortunately this did not include several investments important to the social and economic well-being of large catchment populations (e.g., a new rural hospital at Kerevat, a new health center at Warangoi and new markets at Kokopo and Rabaul).

2.2.5 After it became clear that these delays meant unfinished project works, the government requested another extension of the closing date, to June 30, 2007 which the Bank agreed to in

3 Government elected to use the traditional reimbursement mechanism rather than the option of Bank loan disbursements based on projected cash flows.

9

light of the reasons given. By this time, bid prices were exceeding engineers’ estimates4 on many contracts, reflecting the combined effects mentioned above. Most of the funds freed up from AusAID’s finances had been utilized and there was a projected shortfall of US$4.3 million in Bank funding. The government’s initial 2004 counterpart funding allocation plus delays in obtaining final EU approval of additional funding made it clear that project viability was in jeopardy and that a thorough review was urgently needed to redefine a core program with available resources that would maximize achievement of project objectives. Improved measures were also needed to better manage risks of further cost escalation and implementation delays.

2.2.6 Following extensive consultation with beneficiaries and other stakeholders and the Bank, several components were rephased, resized, or canceled. These included: Kokopo Civic Center (for which alternative funding was sought); studies to diversify and expand agricultural sector activities (many of which were being undertaken independently); and design studies for a possible Stage 2 of the APL. With delays and changes in EU funding for key components, the revised program for Bank financing included: the survey for titling at Sikut (Talvat) relocation site; the health center at Gelagela; and installation of water supply at all relocation sites as well as a first phase of the urgently-needed rural hospital at Kerevat, health center at Warangoi and markets at Kokopo and Rabaul, all of which had been dropped from the new EU Financing Agreement. The expectation was that if EU financing came on stream by July 2005, these components could revert back to EU for funding while the Bank could then rechannel freed funds to such items as site development works at the Kokopo Commercial Center and key sections of the Kokopo Roads and Drainage Component in partnership with the PNG Sustainable Development Program (PNGSDP)5 picking up the remainder.

2.2.7 GRA also took steps to modify the scope of work without affecting functionality on a number of components. It introduced measures to encourage increased competition in bidding for the remaining contracts as well as more active monitoring and reporting to better manage risks of further cost escalation and implementation delays for the revised program. These more proactive risk-management measures together with additional counterpart funding provided by the new government in late 2004, provided good prospects for implementation of the revised program and a sound basis for Bank approval in June 2005 of a reallocation of loan proceeds among disbursement categories, along with a two-year extension of the loan closing date to June 30, 2007. 2.2.8 With the Loan 96 percent disbursed (US$24.21 million of the original US$25.26 million) at the end of May 2007, and all remaining contracts within ten percent of completion, the Bank agreed to a third extension to the loan closing date to December 31, 2007 to ensure that resettlement and institutional targets would be achieved and restoration works would be satisfactorily completed. A long-term development strategy and plan for the area had also been completed under direct provincial initiatives.

4 In interviews with contractors about the high prices they sought, they cited GRA's closer monitoring of construction quality, rejection of work not meeting design specifications compared to opportunities from other donors and lenders in urban locations who did not have such standards. 5 The PNGSDP manages development funds generated from operations of the Ok Tedi Mine in Western Province.

10

2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization

2.3.1 Close monitoring of project implementation was maintained through frequent field visits, progress reports and phone conferences among GRA, ENBPA, and the Bank. GRA’s close site supervision and use of a contract administration manual enhanced risk-management techniques. There were frequent contractors’ meetings which helped in resolving implementation issues, ensuring good construction quality, and limiting cost escalation and implementation delays following the restructuring in 2005. While there were initial cases of design creep, consultation with stakeholders had a major positive effect on sustainable design, including those for simple community-managed water schemes under the Village Development Fund that are intended to have self-financing O&M. To facilitate greater beneficiary participation, GRA initiated a weekly call-in show, augmented by periodic workshops which were often scheduled during Bank supervision missions to provided feedback to all participants on project progress and effectiveness in addressing expressed needs. Close monitoring and evaluation of issues affecting project viability resulted in several timely and effective interventions, including restructuring of financing arrangements and project composition, to avoid cessation of works, especially during the early implementation period.

2.3.2 The active participation of beneficiaries in the construction phase continues to be seen in the utilization of the facilities provided. The schools, health centers, and port facilities are all being fully utilized and government financing is being augmented by contributions from users who have created parent-teacher associations (for the schools) and other user groups.

2.4 Safeguard and Fiduciary Compliance

2.4.1 Environmental Protection. Review of contractor compliance with environmental management plans is highlighted in GRA’s contracts administration manual and has been done in all cases by supervision engineers, GRA’s environmental specialist and Bank supervision missions. AusAID followed its own procedures for the Kokopo/Rabaul Road Component.

2.4.2 The project is successful in consolidating the shift in settlement and economic activity away from the high-risk areas in and around Rabaul. New planned development in Kokopo and other parts of the Gazelle Peninsula is considerable and orderly. Studies of solid waste and wastewater/sanitation management arrangements found the existing disposal site unsuitable due to possible groundwater contamination. Two Bank supervision missions included solid waste specialists and more suitable alternative sites were identified within the context of an overall solid waste management system including collection and waste recycling and minimization. The study of wastewater and sanitation reviewed the suitability of alternative technologies, particularly in light of the permeable ground conditions prevailing in the Gazelle Peninsula, and their financial and institutional implications. It also confirmed that septic tanks remain suitable provided septage is disposed of satisfactorily. For rural areas, other on-site technologies for treatment of human waste are also being developed locally including ventilated composting toilets.

11

2.4.3 Although a comprehensive air quality study was delayed6, the Rabaul Volcanological Observatory (RVO), WHO, and the local health authority have mitigated risks through constant monitoring and frequent advisories to the affected population, particularly in and around Rabaul Town. During periods of intense emissions, the regional hospital has been closed, with patients evacuated temporarily to safer areas.

2.4.4 All damaged buildings in Rabaul have been demolished and the devastated areas as well as the caldera walls have been seeded for vegetation to discourage redevelopment and to minimize erosion and siltation of Rabaul port facilities and the harbor where silt traps have also been constructed. Trees and grass now cover almost all of the affected vacant areas.

2.4.5 Social Issues. Before this project, the preliminary baseline survey of the target population conducted in July 1998 indicated that 53 percent of the target population was residing in the four established relocation sites that still lacked most basic services and in some cases, adequate land for sustainable livelihood. It also found that 36 percent were back in their villages of origin despite continuing emissions from Mt. Tavurvur, and the rest was either in unofficial “care centers,” temporarily residing with relatives in neighboring villages, or had resettled elsewhere at their own initiative. To address the issue of small plot sizes that were insufficient for a sustainable livelihood from subsistence farming and market gardening at some sites, the Provincial Administration purchased a plantation on the northeast coast as an alternative relocation site for fishing families. Basic services have also been provided through the project in relocation sites and by late 2007 over 80 percent of the relocation sites were permanently occupied.

2.4.6 Although it is not a requirement in the legal documents, the government was planning to conduct a social assessment (including a household survey) of the outcomes of the project. Because the government has not opted for the APL2 project, many of the issues which would have been more fully addressed under that phase will have to be done by others. These include: progress in more fully restoring or improving livelihoods and the sense of community among the resettled populations and host communities; stresses leading to problems of social control and, potentially, weakening social cohesion; concerns about security of tenure and land titling in a matrilineal society uprooted from its ancestral lands; and some tensions in the relationship between relocated villages/populations and host villages. A preliminary study design was prepared by the Provincial Administration to be undertaken by local nongovernmental organizations familiar with the Gazelle, but for funding and scheduling reasons, the assessment is being undertaken directly by social specialists in GRA and ENBPA. From mission site visits and several meetings with beneficiary groups, indications are that these groups feel that current access to health, water and sanitation, and education levels are equal to or better than before the

6 An initial study was planned to assess impacts of periodic ashfall. The ash continues to fall almost daily and there was an eruption in 2006 so a more comprehensive study is now being undertaken by the Rabaul Vulcanological Observatory (RVO) with WHO assistance.

12

eruption. Overall economic well-being is more difficult to gauge; this and more detailed impact figures are expected in early 2009 from that ongoing study.

2.4.7 Land Acquisition. The only land acquired under the project was the plantation at Vunacambi as a relocation site. Nineteen plantation workers and their families were resettled with compensation under the terms of a RAP in accordance with Operational Directive 4.30.

2.4.8 Indigenous peoples. Residents from the devastated areas in and around Rabaul who were most directly affected by the eruption and, therefore, a key target population for relocation under the project were mainly ethnic Tolai. Under the strategy for a more dispersed settlement pattern, project facilities such as schools and health facilities are also more accessible to other ethnic groups including the Bainings and other groups who are also indigenous to the area. Similarly, the Village Development Fund Component was intended to provide priority services to host communities adjacent to relocation sites but also covered villages in a wider catchment in the Gazelle Peninsula with beneficial social impacts.

2.4.9 Fiduciary Aspects. Compliance with fiduciary requirements is rated satisfactory. GRA commissioned the design and installation of a computerized project financial management system to meet the Bank’s reporting requirements under the Loan Administration Change Initiative and all financial management reports were submitted to the Bank on time. In addition, GRA maintained its existing spreadsheet-based project monitoring system that also tracked project costs and financing, procurement, and implementation progress on the overall project, including components funded by other donors. GRA’s annual project accounts were audited throughout by the Office of the Auditor General, consistently receiving unqualified audit certificates and satisfactory reports on financial management. Minor points raised in the management letters accompanying audits were addressed promptly. There are no outstanding audit reports and all critical legal covenants have been complied with.

2.4.10 Procurement. Procurement of works, goods and services was consistently carried out satisfactorily in accordance with Bank guidelines by GRA staff who had gained experience with these guidelines during the ITRP from 1995 to 1998. With no foreign interest initially in the generally small-scale works involved in a relatively remote area and a relatively small domestic construction industry in PNG, contracts typically drew only a small number of bidders. Bid prices often fluctuated markedly from engineers’ estimates. Reviews of possible causes indicated that it was possible that bidders were factoring in (a) likely extensions in the bid validity period due to the lengthy contract award approval process, and (b) GRA’s close, on-site supervision and enforcement of contract engineering performance specifications.

2.5 Post-completion Operation/Next Phase

2.5.1 Operations and Maintenance (O&M). An assessment during appraisal indicated that maintenance implications of the restoration program would be administratively and financially manageable by responsible agencies. O&M manuals have been prepared for components transferred together with some training to responsible local, provincial, and national agencies. Memoranda of agreement (MOAs) have been updated for the scope of maintenance required and procedures have been implemented to integrate transferred project assets into the maintenance planning and budgeting systems at the National Department of Works through inclusion of

13

rehabilitated national roads into the Road Asset Management System (RAMS), at the National Harbours Authority for Rabaul Port works (dredging and silt traps), and at PNG Power for power supplies to resettlement areas. However, indications to date are that on an overall basis, maintenance budgets are not yet adequate in these agencies, which reflects both low national allocations and provincial revenues for such activities and capacity constraints among provincial and lower level administrators.

2.5.2 Several initiatives are being taken at the provincial and local level to manage maintenance activities and augment low O&M budget allocations. A draft legal agreement has been prepared for a new management contract for the provincial Ah Tam/Dawapia port facility and an autonomous market authority is being established to operate and maintain the two principal markets in the province at Kokopo and Rabaul. A similar arrangement is being established by the responsible district for O&M at the Kokopo Commercial Center. The local education board and the local hospital board have been working in partnership with users and others to increase income from user charges, special fees and fundraising activities to help defray routine operation and maintenance costs. Water supplies are being operated and maintained by respective Ward Committees with revenues from user charges and technical advice as needed from the Water Board.

2.5.3 The good standard of maintenance by ENBPA of provincial roads, bridges, and other facilities prior to the eruption was widely acknowledged. New project works were estimated at appraisal to require an estimated 6 million kina annually to maintain the project assets to be transferred to ENBPA and assessed as manageable in relation to projected provincial revenues at that time. However, since the provincial sales tax was replaced by the national value-added tax in 2000, East New Britain’s revenues have declined by about 30 percent in real terms and the provincial maintenance budget in 2007 was 73 percent lower in real terms than it was before the eruption, falling from 24 percent of revenues in 1994 to 10 percent in 2007. The province therefore faces a large maintenance budget shortfall of about US$7 million per annum—about 45 percent of the province’s overall 2007 revenue budget—and much required maintenance will continue to be deferred. This is a national problem which has to be addressed at a different level from this project.

2.5.4 To strengthen maintenance planning, AusAID’s Provincial Performance Improvement Initiative (PPII) is assisting the province to update the condition assessment and inventory of infrastructure assets. ENBPA is trying to utilize alternative options to maintain its assets. For example, it has introduced a surcharge on the vehicle registration fee to help meet maintenance cost of provincial roads and bridges. However, recent indications are that this will generate additional revenue of only about 200,000 kina rather than the 2 million kina estimated and will not significantly affect the overall shortfall in provincial road maintenance funding.

2.5.5 Institutional strengthening and sustainability. GRA has established a well-earned reputation for efficient implementation of development works and introduction of risk-management techniques under the project have further enhanced its effectiveness. Similarly, ENBPA has long been noted for its effective administration and capacity to deliver services to its

14

people7 despite its relatively limited revenue base, compared to provinces that benefit from mining industry royalties. Key officials from several provinces have visited East New Britain in recent years to learn from its development planning and implementation management arrangements under the project with a view to establishing similar arrangements to accelerate development in their own provinces. The long-standing consensus between politicians and administrators in East New Britain Province on development priorities, which may not be as strong in other provinces, as well as the “seamless” relationship between GRA and the provincial administration, have been invaluable in providing the strong foundation for successful implementation of the restoration program in the Gazelle Peninsula of East New Britain.

2.5.6 The policies and procedures underpinning the successful restoration program in the Gazelle are not limited to emergency reconstruction projects and could facilitate development efforts in other provinces if consistently applied. The East New Britain Provincial Assembly has approved the integration of GRA into the provincial system, as covenanted under the project. GRA is now tasked with reviewing and managing all construction and maintenance of all ENBPA’s civil works and buildings and in light of the strong interest expressed by both the national government and several provinces and autonomous entities to emulate the GRA experience, it can also carry out “sustainable development programs” in other provinces.

2.5.7 Follow on project. Although the development priorities and opportunities now being targeted are identified in the province’s recently prepared Comprehensive Development Plan (CDP), the government has not requested a second-phase APL operation for East New Britain. It is reportedly very satisfied with the APL1 phase but it is trying to spread the impact of Bank assistance in other areas. To this end, it has recently requested Bank support for a similar operation in Manam, where a volcanic eruption has also caused severe social and economic dislocation.

3 ASSESSMENT OF OUTCOMES

3.1 Relevance of Objectives, Design and Implementation

3.1.1 Relevance of Objectives. The project provided comprehensive support for achieving the objectives of the government’s Medium-Term Restoration Program to restore the social and economic well-being of the East New Britain people that prevailed prior to the 1994 volcanic eruption and to ensure sustainability of outcomes and minimize adverse impacts of future volcanic eruptions on lives, property, and livelihood. This is consistent with the Bank’s strategy for strengthening the capacity for service provision as well as the resilience of communities to disasters. The principles underpinning project preparation and implementation management arrangements are also highly relevant for non-emergency reconstruction development projects in PNG.

7 Pre-eruption social indicators for health and education were consistently among the highest in PNG and the province’s economy attracted migrant workers from other parts of PNG.

15

3.1.2 Design and Implementation. Project design was highly relevant to and fully consistent with the government’s restoration strategy. Extensive stakeholder consultation ensured that the project design was relevant to the needs of displaced communities and all affected groups.

3.1.3 The tools designed for future area and regional development plans were particularly relevant because they provided a much clearer picture to many provincial and national policymakers on the implications of different options and they also charted a path toward achievement of the desired goals. The Village Development Fund was an important tool in assisting not only the relocated population but also the host communities and this more inclusive approach was very instrumental in getting cooperation from all major ethnic groups in the area, thus facilitating the resettlement process.

3.1.4 Finally, although GRA is a statutory authority reporting to a minister of the national government, it is based in Kokopo, the newly designated capital of East New Britain Province, where restoration activities are centered and where it has been able to partner most effectively with the province and affected population.

3.2 Achievement of Project Development Objectives

Rating: Satisfactory

3.2.1 The project achieved or exceeded the objectives for APL1 and achieved significant steps toward the objectives of APL2, even though the government chose not to implement the second phase. This decision to forego the second phase in part reflects the success of APL1 and the government’s desire to direct Bank assistance to higher priority development needs.

3.2.2 The project benefited from significant financing contributions from other partners (the EU, AusAID, and the national and provincial governments), but also experienced serious implementation delays partly due to problems with the fund flows and decisions from other participants. These delays and funding problems (including severe counterpart funding shortfalls) persisted through 2004 and caused the project to be rated unsatisfactory. However, they were largely overcome and the project turned around dramatically beginning in 2005 and ultimately achieved or exceeded its key performance indicators (KPIs) as detailed in Annex 2 on Outputs by Component. These achievements were attained despite daunting conditions and obstacles that at times made satisfactory completion of the project seem unlikely.

3.2.3 Briefly, the development objective of restoring government services similar to those destroyed by the 1994 eruption were achieved through (a) provision of services and infrastructure to those occupying the relocation sites, (b) upgrading the infrastructure and services of Kokopo Town and establishing it as the new provincial capital, (c) rehabilitating and restoring infrastructure and services of Rabaul Town/Port both for the areas considered safe for residential occupation and for port operations, which are now functioning at pre-eruption levels, and (d) providing infrastructure and other public services in various townships in the northeast part of the Gazelle Peninsula. Infrastructure and services now meet or exceed those existing prior to the eruption.

3.2.4 The development objective of ensuring the sustainability of social and economic well-being and minimizing adverse impacts of future eruptions was achieved through some of the

16

measures mentioned above related to providing services and infrastructure and also, more specifically, by successfully (a) relocating populations away from vulnerable areas, (b) avoiding resettlement in vulnerable areas, (c) implementing erosion control and silt trap measures in Rabaul Port, (d) establishing a Village Development Fund to finance community-based infrastructure and services, community development, and promote economic activities, and (e) development of a comprehensive development plan for the province as well as conducting other research on coastal zone planning and hazard mitigation.

3.3 Efficiency

3.3.1 Economic Rate of Return and cost effectiveness. This being a post-disaster resettlement project, the APL1 component was focused more upon the infrastructure, resettlement, and related needs of displaced groups. The strategy of the government was to shift the locus of settlement and economic activity away from direct exposure to future volcanic activity at Rabaul to the relatively low-risk areas of Kokopo and its surrounding hinterland. Many project benefits are therefore not easily quantified in economic rate of return analyses. Instead, for most components, e.g. health and education facilities, justification was based on a least-cost approach to satisfying demonstrable demand at warranted service standards in keeping with best practice for emergency reconstruction projects. The Kokopo-Rabaul Road was scheduled to have a cost-benefit analysis; however, it was financed by AusAID and no information on such assessment was available. The Vulcan sector was scheduled to be done at a later time to allow ash compaction and lower-cost drainage design options to be studied.

3.3.2 As indicated earlier in this report, project investments are assessed to have met their design objectives and close supervision and enforcement of design specifications by GRA ensured satisfactory construction quality. All infrastructure and facilities are already well-utilized and arrangements are being set in place to systematize and increase funding for maintenance as well as to make commercially oriented components substantially cost-recoverable, thereby improving project sustainability.

3.3.3 Financial Rate of Return. Not applicable.

3.4 Justification of Overall Outcome Rating

Rating: Satisfactory 3.4.1 The overall rating is based on the full achievement of project objectives/KPIs; safeguards and fiduciary compliance; the findings of quality assurance reviews both at entry and during supervision and acknowledgement of its effectiveness of the GRA model by various stakeholders.

3.5 Overarching Themes, Other Outcomes and Impacts

(a) Poverty Impacts, Gender Aspects, and Social Development 3.5.1 The economic and social dislocation of the eruption was considerable. About 20,000 people sought permanent resettlement in other parts of the Gazelle Peninsula including about 10,000-12,000 who were accommodated in relocation sites under the project. Relocation sites

17

were intended to maintain important socio-cultural aspects and to provide sufficient land for a sustainable livelihood from subsistence and cash crops. Although it is an ongoing process, growth in the province’s economy generally and the urban economy in and around Kokopo is providing access to increasing livelihood opportunities. Many project components particularly benefit women. The Council of Women Centers provides skills development and counseling and social services, and the community water supply schemes under the Village Development Fund have reduced the considerable burden of fetching water borne by women and children in these villages. The Village Development Fund also targeted priority needs identified by host communities and extended services in villages of ethnic Bainings people, also indigenous to the area.

(b) Institutional Change/Strengthening 3.5.2 The studies discussed above that were implemented through the project have established the tools for planning longer-term economic and social development in the Gazelle Peninsula. Amendments proposed to the Gazelle Restoration Authority Act to incorporate GRA into the provincial system and expand its mandate to assist development efforts in other provinces also confirms the widespread recognition of the effectiveness of this as a development model.

(c) Other Unintended Outcomes and Impacts (positive or negative) 3.5.3 The success of the APL1 in this province is reportedly among the reasons why the national government was reluctant to approve an APL2 in the same province since it was felt that other provinces needed similar assistance.

3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops

Not applicable. While there was no requirement for this in the legal documents, in view of the success of this model, the government is trying to document some of the more widespread project impacts on both direct and indirect beneficiaries. The results will be available at a later date and so it is not possible to put it in this report.

4 ASSESSMENT OF RISK TO DEVELOPMENT OUTCOME

Rating: Moderate 4.1.1 The shift toward Kokopo and away from Rabaul created winners and losers and a small but vocal group continues to press for redevelopment of Rabaul. ENBPA has shown its commitment to the government’s policy prohibiting redevelopment in the devastated areas through recent decisive action ordering amendments to Rabaul Town Plan to prevent such development. Recent relocation of key commercial enterprises, the main market and the post office from Rabaul is slowly weakening calls for resettlement there. Now that Kokopo has been designated the new administrative capital of the province and densification is continuing, the risk of return to Rabaul is decreasing daily. The Disaster Risk Assessment and various planning studies undertaken under the project also provide a credible regional development plan which the ENBPA is committed to implementing.

18

4.1.2 Arrangements are being made by responsible national and local agencies to integrate maintenance needs into operations and budgets. Both the national Port Authority and the national Water Board have good track records on maintenance. Maintenance of national roads has traditionally been badly under-funded, but the national Department of Works is making improvements associated with the Bank-financed Road Maintenance Project (Ln. 7119-PNG). Still, significant decline in the ENBPA revenue base following replacement of the provincial sales tax by the national value-added tax has weakened the province’s traditional capacity for adequate maintenance funding and has, therefore, imposed some risk to sustainability of project outcomes. With the current increase in national earnings from PNG’s primary mineral exports, increased revenue-sharing is more likely and this should assist in project sustainability.

5 ASSESSMENT OF BANK AND BORROWER PERFORMANCE

5.1 Bank Performance

(a) Bank Performance in Ensuring Quality at Entry Rating: Satisfactory 5.1.1 The Bank’s performance in ensuring quality at entry is rated satisfactory. The team reviewed several alternative design strategies and had extensive consultations with displaced communities and other stakeholders to ensure broad consensus on project objectives and urgent priorities, consistent with these objectives. Assessments were made of possible adverse environmental and social impacts and of project sustainability in terms of post construction O&M implications.

5.1.2 In light of the political risks in PNG at that time, the QAG was requested to conduct a Quality Enhancement Review (QER) prior to negotiations to leave open the possibility for useful changes in project design to be incorporated at negotiations The QER Panel found the project to have been well-designed and prepared.

(b) Quality of Supervision Rating: Highly satisfactory 5.1.3 The Bank’s performance during supervision is rated highly satisfactory for a number of reasons. Taking account of the status of the country as a fragile state, the Bank engaged throughout the project in close supervision and follow-up, particularly during and after project restructuring in 2005. There were at least two missions annually with frequent audio conferences with the project team in Sydney plus additional specialist missions focusing on solid waste management, social issues and other aspects. All regular supervision missions made site visits to ongoing works and often met with project stakeholders, including at periodic workshops. The Bank recognizes the special needs of fragile states and the project team demonstrated appropriate flexibility in helping resolve problems affecting project-implementation progress and/or achievement of development objectives. For example, it helped to address the adverse impact of severe government counterpart funding constraints by working with AusAID and transferring the Kokopo/Rabaul Road to AusAID in 2000 for grant financing. And in 2002, it used the resulting

19

loan savings to increase its disbursement share of civil works and reduce the impact of low counterpart funding. In 2005, when funding from another donor was being delayed, the project team was able to shuffle available project resources toward key bottleneck tasks until the other donor’s funds became available. To help manage costs, the project team worked with GRA to develop risk management strategies that proved effective in minimizing further cost escalation and implementation delays; and in making additional adjustments when other funding came back on stream in late 2005. Continuity of key Bank project team members resulted in an excellent working relationship with counterpart officials at all levels throughout the process and this facilitated constructive dialogue and quick actions throughout, including on issues affecting project viability.

5.1.4 A QAG review in September 2006 rated the Bank’s overall project supervision during fiscal 2005 and 2006 as satisfactory (the second highest rating on a six-point scale). The ratings for the main supervision assessment subcategories were: highly satisfactory for candor and quality of ISRs; moderately satisfactory for focus on development effectiveness; satisfactory for fiduciary/safeguard aspects; and satisfactory for adequacy of supervision inputs and processes. The panel’s report also noted the very high degree of commitment and effectiveness of the task team in resolving many of the key problems that were encountered during implementation. At the end of fiscal 2006, QAG had also agreed to grant the project a “golden flag” (to prevent it from being classified as potentially at-risk) in recognition of the strong advances made toward achieving project objectives despite challenging conditions (two of the project’s three risk flags were for country environment and country record) and success in overcoming earlier implementation problems that had triggered a long-term risk flag.

(c) Justification of Rating for Overall Bank Performance Rating: Satisfactory 5.1.5 The extensive analysis of design strategies and investment priorities and stakeholder consultation called for by the Bank during project preparation resulted in a broad-based and abiding consensus on project objectives and outcomes. Bank commitment to resolving problems and its flexibility in adjusting project financing arrangements to lessen the adverse impact of funding problems by the government and one other donor helped the project to meet or exceed development objectives set for it and to demonstrate an effective model for decentralized development of local infrastructure and services in East New Britain and elsewhere in PNG.

5.2 Borrower Performance

(a) Government Performance Rating: Satisfactory 5.2.1 During the early years of implementation, the project faced severe difficulties from inadequate government counterpart funding that limited implementation on the ground even after an extension to the loan period, raising concern amongst stakeholders over project viability. Even so, the government has remained strongly committed to the project and the ENBPA worked side by side with the GRA throughout, standing firm against political pressure to

20

redevelop Rabaul (which would have undercut the strategy for the restoration program and the rationale for Bank involvement); maintaining close contact with stakeholders on project implementation progress, and supervising or implementing a number of project studies directly. With increasing commodity prices, counterpart funding improved markedly, helping the restructured project to achieve its satisfactory completion and outcome. The government recognizes the value of GRA’s experience and capability and is in the final stage of formalizing its association with ENBPA to enable it to also assist similar development efforts in other provinces.

5.2.2 National, provincial and district agencies have incorporated project assets into their maintenance planning and operations and ENBPA has introduced additional revenue measures for maintenance on project assets it is responsible for. However, maintenance budgets remain well below requirements highlighting the need for national structural changes to provincial finances including government grants and transfers, in particular.

(b) Implementing Agency or Agencies Performance Rating: Highly satisfactory 5.2.3 GRA secured the confidence of the national government, the business community and the population of the Gazelle Peninsula under the ITRP initiated in the immediate aftermath of the eruption. Under the MTRP, GRA has again distinguished itself as a highly effective implementing agency, despite severe complications relating to multi-donor financing and inadequate government counterpart funding. During project preparation, GRA quickly adjusted to the need for a stronger analytical underpinning for the MTRP including in relation to environmental and social impacts and for extensive stakeholder consultation in selecting project priorities. It hired specialists in its management team and installed a comprehensive project accounting and financial management system. GRA staff quickly adopted Bank procurement guidelines to ensure cost effectiveness and transparency in its operations. Following project restructuring, GRA’s proactive oversight and enhanced risk management techniques were effective in limited further cost increases and implementation delays, ensuring satisfactory project completion and achievement of all project objectives. GRA’s vigilant, on-site supervision of the large and diverse program of engineering works has been a hallmark of project implementation and construction failures have been minimal.

(c) Justification of Rating for Overall Borrower Performance Rating: Satisfactory 5.2.4 The initial counterpart funding problem reflected a national economic downturn rather than project commitment problems. As country earnings increased, the government increased its allocation to help improve project implementation. At both the national and provincial level, sustained support throughout both the ITRP and the MTRP demonstrates the enduring commitment to achieving a good project outcome for the people of East New Britain and trust in the Bank as a key development partner.

21

6 LESSONS LEARNED

6.1 Major Projects can be successfully implemented in PNG

6.1.1 The PNG government has proved itself fully capable of creating conditions conducive to good project outcomes and achievements. Aid and donor agencies as well as government officials acknowledge that the Second Gazelle Restoration Project is a success story due to: the sound government strategy guiding the restoration effort throughout; decentralized decision-making on restoration priorities; highly effective and transparent implementation arrangements by the GRA in close partnership with ENBPA and extensive local consultation to ensure sustained and broad-based stakeholder support. Some of GRA’s systems—particularly the contract administration and quality control manuals—have been adopted by the Department of Works and the Central Supply and Tenders Board. Other countries in the Pacific region are reviewing these manuals and have expressed an interest in using them.

6.2 A time contingency in the implementation schedule would have been appropriate

6.2.1 Notwithstanding the successful implementation of the ITRP, the four-year implementation period for the first phase of the APL was inadequate. Implementation delays accumulated virtually from the outset, due largely to government counterpart funding problems. By the initial June 30, 2003 loan closing date, less than 20 percent of the Bank loan had been disbursed. Inclusion of a substantial time contingency in the implementation schedule would have been appropriate, particularly given the project complexity and that that counterpart funding constraints had been identified as a project risk.

6.3 Bank financing or channeling EU funds through Bank for works for displaced communities would have reduced delays in basic services for them

6.3.1 Involvement by several donors with different financing and disbursement procedures and information requirements imposed a considerable administrative burden on GRA’s administrative and engineering staff, in particular. In addition, reserving Bank-financing for critical components for improving the quality of life for displaced communities would have lessened the adverse impact of problems experienced with the EU financing mechanism by accelerating provision of basic services in relocation sites as well as intensifying the dialogue on social and livelihood aspects.

6.4 Comprehensive review is needed of financing arrangements for local services including government grants and transfers, in particular

6.4.1 The assessment at appraisal indicated that O&M on project assets would be administratively and financially manageable by all responsible agencies. The adverse impact on provincial revenues following replacement of the provincial sales tax by the national value-added tax apparently undercut the province’s longstanding tradition of adequate O&M, something which could not have been anticipated. The province’s present limited revenue authority and dependence on government transfers highlights the need for a thorough review of the current arrangements.

22

6.5 Landholders are key stakeholders in ensuring equitable and efficient project outcomes

6.5.1 The shift toward a more dispersed settlement pattern away from Rabaul had major implications for land availability and use in Kokopo and the Gazelle hinterland, particularly since most land is in customary ownership. The slow and limited release of land has narrowed development options, possibly increasing public sector development costs and creating winners and losers. For emergency reconstruction type projects, in particular, landowners should be seen as key stakeholders in ensuring that project development objectives are achieved in an equitable as well as efficient way as possible. The system of communally owned land adds to the challenge.

6.6 Simple risk management techniques can limit cost escalation and implementation delays significantly

6.6.1 The use by GRA of a contract administration manual and enhanced risk management measures following project restructuring in 2005 was effective in minimizing further cost escalation and implementation delays. While, implementation delays in the early years were largely out of the GRA’s control, these tools and techniques are transferable and could generate major efficiency gains on development projects generally in PNG.

7 COMMENTS ON ISSUES RAISED BY BORROWER/IMPLEMENTING AGENCIES/PARTNERS

7.1 Borrower/implementing agencies

The draft ICR was reviewed by the Borrower and GRA. A summary version of the GRA/Borrower’s Completion Report is included in Annex 7. The full text and annexes of the Borrower’s Completion Report are in the project files. No additional comments were received from the Borrower.

7.2 Cofinanciers

AusAID and EU also received draft copies of the ICR. See the Summary of Cofinanciers’ Comments in Annex 8.

7.3 Other partners and stakeholders None

23

ANNEX 1. PROJECT COSTS AND FINANCING

(a) Project Cost by Component (in USD million equivalent)

Appraisal

estimate 1,2,8 Actual costs 3,7,8

Component Notes * (USD million) (USD million) % of appraisal

1. Improving Quality of Life for Displaced Communities 9,10,11,12 2.37 6.86 289

2. Consolidation of Kokopo as Provincial Capital 13,14,15 14.90 31.22 210

3. Restoration of Rabaul as Regional Port 16,17,18 3.52 8.57 243

4. Development of Support Infrastructure and Social Services for Northeast Gazelle Peninsula

19 5.36 12.10 226

5. Economic Diversification and Longer-Term Development Planning

20,21 1.55 0.24 15

Design and Supervision 4,22 3.36 6.10 182

Total Baseline Cost 31.05 65.09 210

Physical Contingencies 3.40 0.00 0

Price Contingencies 4.75 0.00 0

Total Project Costs 39.20 65.09 166

Front-end fee IBRD 0.26 0.26 100

Total Financing Required 39.46 65.35 166

* See the detailed cost table in Section (c) of this annex for notes to this table.

(b) Financing

Source of Funds Type of

cofinancing

Appraisal estimate

(USD million)

Actual financing

USD million

% of appraisal

Borrower (including national, provincial, and district governments)

11.11 14.53 131

IBRD 25.26 25.26 100

Rabaul Volcanological Disaster Relief Committee (grant)

Parallel 0.02

Cocoa and Coconut Institute PNG (grant) Parallel 0.04

PNG Harbours Board (grant) Parallel 0.18

PNG Sustainable Development Fund (grant) Parallel 4.20

Fees, Interest, etc. 1.72

European Union (EDF 7,8,9) Parallel 2.90 7.50 259

ADB Parallel 0.15 0.00

AusAID (grant) Parallel 0.04 11.90 29,750

Total 39.46 65.35 166

24

(c) Detailed Costs by Subcomponent

c Notes

PAD estimate 1,2,4,8

(US$ million)

Actual costs for December 2007 revised program 3,5,6,7,8

(US$ million) Percentage of

appraisal

1. Improving Quality of Life for Displaced Communities

2.37 6.86 289

1.1 Land Mobilization 9 0.63 0.45 71 1.2 Infrastructure and Utility Services 10 1.56 4.45 285 1.3 Social and Public Safety Services 11 0.12 1.55 1,292 1.4 Community Dev. and Economic Support 12 0.06 0.41 683

2. Consolidation of Kokopo as Provincial Capital 14.90 31.22 210

2.1 Transportation 13 8.19 23.09 282 2.2 Utility Infra., Social Services, and Public Safety 14 1.21 3.50 289 2.3 Economic Support and Government Admin. 15 5.50 4.63 84

3. Restoring Viability of Rabaul as a Regional Port 3.52 8.57 243

3.1 Land Management 16 0.06 0.05 83 3.2 Port-related Infrastructure 17 2.73 6.92 253 3.3 Utility Infrastructure, Social, Economic, and

Public Safety Services 18 0.73 1.59 218

4. Development of Support Infrastructure and Services for the N.E. Gazelle

5.36 12.10 226

4.1 Physical Infrastructure, Social, Economic, and Public Safety Services

19 5.36 12.10 226

5. Economic Diversification and Longer-term Development Planning

1.55 0.24 15

5.1 Potential New Economic Services 20 0.76 0.18 24 5.2 Longer-term Framework Strategy 21 0.79 0.06 8

Design and Supervision Costs 4,22 3.36 6.10 182

Total Base Costs 31.05 65.09 210

Physical Contingencies 3.40 - -

Price Contingencies 4.75 - -

Front-end fee 0.26 0.26 100

Total Cost 39.46 65.35 166

Notes: 1. Figures from PAD (October 1, 1999).

2. Exchange rate: US$1 = Kina 2.7 at appraisal.

3. Average exchange rate: US$1 = Kina 3.14 during implementation.

4. Design and supervision costs of investments for APL1 and design of components originally planned for implementation in APL2 (see footnote 22 for additional details).

5. Average inflation during project implementation = 41.6%; remainder of cost increases due to substantial increase in scope of works and higher than anticipated bids.

6. Actual cost of additional investments estimated at 18.36 million kina. Investments cancelled from project (and from government development plans) estimated to cost 7.52 million kina at inflated PAD cost.

7. Current cost of investments cancelled from Bank project but which borrower plans to complete at a later date with other financing estimated at 35.27 million kina (December 2007):

Activity/investment estimated cost

Rabaul Town Clinic, Stage 1 2.60 Kerevat Rural Hospital, Stage 2 0.83 Warangoi Rural Hospital, Stage 2 1.10 Kokopo Market, Stage 2B 2.85 Kokopo Commercial Centre 4.37 Ralum Civic Centre, Stage 1 12.23 Ralum Civic Centre, Stage 2 8.49 Ralum Civic Centre, Stage 3

(ENB Public Library) 1.16 Phase 2.2 Design 1.63

Total 35.27

8. All costs exclude GRA PIU operational costs

25

9. The PAD allowed for the purchase and surveys of new resettlement estates, and further cadastral surveys at Sikut Matupit and Sikut Talwat. In the final programme less land was purchased, but the two surveys at Sikut were completed.

10. The PAD included internal roads at Sikut, electricity and water supplies to Gelagela, Clifton, Sikut, and Vunakabi, for a sanitation study, and the upgrading of Gelagela aid post to rural health centre. In the final programme, all of these projects were completed, with additional areas covered by the electricity and water supplies - Asing (water and electricity) and Warena (water only). The water supplies (8 no.) were all deep bores, except for an additional component for water tanks to the community resource centres.

11. The PAD allowed for the upgrading of an existing aid post to rural health centre. The health centre provided is considerably larger than that planned at appraisal with a maternity ward, two general wards, an operating theatre, and several clinic buildings.

12. The original project allowed for agricultural extension services to the resettlement estates, mainly in the form of the provision of cocoa seedlings to displaces farmers. The final programme included the extension services, as well as the construction of an agricultural nursery at Sikut to provide further seedlings to the resettlement estates.

13. The original project included both the upgrading of the Kokopo to Rabaul road (excluding Vulcan), and the upgrading of the Kokopo Town roads and drains. The Kokopo to Rabaul Road was moved to project component 4 (see note 19 below). The upgrading of the Kokopo Town roads and drains was completed, with increase in the length of roads and drains being constructed (7.6 km over 7.1 km drains, 5.9 km over 4.1 km roads).

14. The PAD allowed for the supply of generators to key institutions in Kokopo and elsewhere, the provision of a water supply at Butuwin health centre, a feasibility sanitation study for Kokopo, and the improvement of the solid waste dump at Raniolo. The original project also included the construction of a central community school in Kokopo, and a provincial headquarters for women's groups. The majority of the generators were supplied under component 4 (see note 19 below), the Butuwin water supply was successfully installed together with extensive civil works required by the ENBP Physical Planning Board, the sanitation feasibility study was completed, as was a study for the Raniolo solid waste dump. Unfortunately the solid waste dump at Raniolo was not feasible, and the ENB Provincial Government is seeking alternative sites. The construction of the community school was upgraded under the education reforms to a primary school. The women's groups provincial headquarters were constructed under a separate programme, and the design of a district headquarters was completed as well. Two additional components were added, being the construction of four residences for the Department of Justice and Attorney General, and the construction of an STI Clinic building at Butuwin health centre.

15. Originally the project included construction of a new market at Kokopo, the proposed Kokopo Commercial Centre, and the provincial administrative headquarters at the Ralum Civic Centre. Although the designs for all three components are complete, due to severe cost overruns the construction was not completed: 60% of the market has been constructed, 5% of the Commercial Centre (55% if the surrounding roads are included), but the Ralum Civic Centre was cancelled from the project and will be completed with other financing.

16. The original PAD included the survey of a portion of government land at section 46 in Rabaul for eventual development for public institutional use (for police station, fire station, health centre, district offices, women's facilities etc). The survey was completed successfully as well as the full civil works design of the proposed subdivision.

17. The original project included the construction of silt traps and a groyne to reduce siltation of the harbor, extensive erosion control works throughout Rabaul, the restoration and upgrading of the Provincial Government wharf, the removal of damaged structures in the town, and the upgrading of the main port road. The groyne construction was deleted as it was no longer required, but all other works were successfully completed. The silt traps are considerably larger than the pre-eruption structures, and the facilities at the ENBPG wharf have been considerably enhanced for the fishing industry.

18. The original project included the rehabilitation of the town water supply, health and education facilities for the remaining population, the construction of a new market and facilities for women's groups, and the provision of an incinerator for the national quarantine services. The provision of a mains water header tank was completed successfully, as was the restoration of a primary school at Malagunan village to cater for the town population. The design of the planned replacement health centre was completed, but the facility was not constructed due to limitation of funds. The new market and women's facilities were successfully completed at Page Park and on Malaguna road, and the incinerator was supplied and installed for NAQIA. A project has been added to this component to provide maintenance to the temporary health facilities in Rabaul.

19. The PIP and PAD included a large variety of projects to establish a more dispersed pattern of settlements in the Gazelle Peninsula. These projects included upgrading the road from Baliora to Warangoi, provision of water supplies at Baliora township and Vunamami FTC, stand-by generating facilities at National Broadcasting Corporation (NBC) transmitters and restoration of the transmitting station at Kurakakaul, completion of Utmei High School, restoration of Malabunga High School, support to the Vuvu High School / Kabaleo TTC switchover, upgrading of rural health centers at Kerevat and Warangoi, rural police facilities at Bitapaka, women's district headquarters facilities at Kerevat, a village development fund to assist host communities, and a cocoa nursery for CCI PNG at Kulaun (Londip). These projects have been successfully completed, with a 70% increase in the length of the Warangoi road upgraded. There was also a considerable increase in the scope of works at the two rural health centers, now upgraded to rural hospitals (with stage 2, at each, to be completed with other financing), and the addition of an new subcomponent at Tauran Primary School, and of five further health sector projects to be implemented in 2008. The village development project provided 20 facilities—water supply, educational facilities, electricity, etc—to a higher standard (and cost) than planned during preparation.

20. The original project envisaged feasibility studies for and implementation of agricultural development in alternative crops, beef production, and fisheries development. However, the implementation of the agricultural studies was not successful, mainly due to the failure of the province's DPI to formulate an agricultural development strategy on which to base the studies.

21. The original project included the formulation of economic and spatial development strategies and plans for the area, and an air quality environmental study. The economic and spatial development studies were completed successfully with the reports on disaster risk assessments, physical planning of the Kokopo - Tokua corridor and Kokopo Urban Development Plan, a provincial growth centre hierarchy, an economic development plan. All of these reports were presented to and accepted by the PEC at a policy level. Regrettably, the support of AusAID for the air quality study was withdrawn after completion of the first phase. WHO is currently assisting in the second phase.

22. The original project allowed for design and supervision costs for most projects - based on percentage costs of the projects themselves - and a component for the design of Phase 2 projects, based on the agreed list of projects. The designs of all projects were completed successfully, but supervision was mostly undertaken by the GRA PIU staff at no cost to the project. In addition, a sub-component was added for the provision of technical assistance by an international consultant to increase the technical capacity of the GRA PIU in areas where expertise was lacking, and for the review of designs by other consultants. Design and supervision of the Kokopo to Rabaul road was the single most costly design and supervision item (at cost to AusAID).

26

ANNEX 2. OUTPUTS BY COMPONENT

This annex provides an expanded account of project outputs, using as a point of departure each component’s link to the original triggers for APL2 (which were used as KPIs to assess implementation of APL1). In reality the outputs of APL1 went a long way toward achieving the ultimate goals and outcomes outlined for the end of APL2 (see matrix in Section 1.2), which was a key reason the government chose to target future international development financing toward other national priorities rather than implementing APL2.

Component 1: Displaced Communities

KPI 1: Completion of Infrastructure and Services and at least 80 percent occupancy at all relocation sites. Satisfactory. The electric power grid has been extended to relocation sites and water supply has been provided to 10,000 people in these sites to reduce costs of trucking in water to users. The health post at Gelagela has been upgraded to a full health center, reflecting the size of the community there and a new rural hospital and a new health center were constructed at Kerevat and Warangoi respectively that also serve communities in relocation sites. Community Resource Centers have been constructed and staffed to provide community development and economic support as well as health awareness and health education services at relocation sites and a new Council of Women’s Center that provides training programs for women has been constructed at Kerevat. Seedling nurseries have been developed in Gelagela, Sikut & Londip, access roads have been rehabilitated to facilitate production and marketing of food and cash crops in relocation sites and a study on access to micro-credit has been undertaken.

With 80 percent occupation by displaced communities at relocation sites, the KPI for this component has been fully achieved indicating more settled communities and improved livelihood.

While relocation is entirely voluntary, it has not been frictionless and there were some social issues in adjusting to new sites. These were exacerbated by the lengthy delay in provision of some basic services and facilities due to approval delays of another donor’s financing arrangements. Among such social issues, the weakening of social cohesion affecting mainly the youth in these communities seems to be part of a national trend and it will require continued attention.

Component 2: Kokopo Town

KPI 3: 75 percent Completion of Other Infrastructure, Utilities and Social Services Under Phase I, Including Rabaul-Kokopo Road (except for Vulcan Section). Satisfactory. With AusAID financing, reconstruction of the Rabaul-Kokopo Road has been completed except for the Vulcan section. Reconstruction of Kokopo town roads and drainage works that were to be done under the project have been done while some remaining (non-project) contracts are underway with funding from the PNGSDP. The project leveraged substantial additional domestic and foreign funds for such works. A new community school for 880 pupils age 8-12 was constructed. The first phase of Kokopo market has been completed. It will assist the marketing of produce from the Gazelle hinterland and along with the Kokopo Commercial Center, accelerate the

27

consolidation of Kokopo as the new town. Management contracts have been prepared through user and provincial arrangements for maintenance of these facilities. In addition, and outside the original project scope, a new sports field has been constructed for the increasing number of sporting events being held in Kokopo and alternative financing is being sought for the Civic Center to house national and provincial government agencies (this was deferred due higher than estimated cost to build to current design specifications). Overall, Kokopo has undergone a transformation from small village of about 1,100 people prior to the eruption to a large town of about 15,000 people now and the boundary has been extended to include Takobar, another village east of Kokopo.

Component 3: Rabaul Port

KPI 2: 100 percent Completion of port-related protection works in Rabaul and approval of revised zoning plan that prohibits redevelopment in the devastated areas of Rabaul. Satisfactory. The provincial Ah Tam/Dawapia port for coastal shipping has been rehabilitated and a management contract prepared for outsourcing O&M. Berthing areas at both the national and provincial wharves have been dredged and international cargo can again be shipped directly to and from Rabaul Port which is now functioning at pre-eruption levels (305,000 tons in 2007 compared to 322,000 tons in 1993), serving both international and regional shipping. Devastated areas in Rabaul have been cleared and, together with the caldera walls, have been reseeded to reduce erosion. Silt traps have been constructed around the caldera walls and at the port to reduce siltation of the harbor and port facilities. The approach road to the port has also been reconstructed. To serve the small remaining population in Rabaul, the water supply system has been rehabilitated and a market has been constructed. Rabaul Town Plan prohibits resettlement in vacated areas and this is constantly being monitored and enforced.

Component 4: Gazelle Peninsula

The project implementation plan (PIP) and PAD included several initiatives to establish a more dispersed pattern of settlements in the Gazelle Peninsula. These projects included upgrading of the road from Baliora to Warangoi, provision of water supplies at Baliora township and Vunamami Farmers Training Center, standby generating facilities at National Broadcasting Corporation (NBC) transmitters and restoration of the transmitting station at Kurakakaul, completion of Utmei High School, restoration of Malabunga High School, support to the Vuvu HS / Kabaleo Teachers College switchover, upgrading of rural health centers at Kerevat and Warangoi, rural police facilities at Bitapaka, women’s district headquarters facilities at Kerevat, a village development fund to assist host communities, and a cocoa nursery for the Cocoa Coconut Institute (CCI) at Kulaun (Londip). These projects have been successfully completed, including the upgrading of 70 percent more of the Warangoi road than originally planned. There was also a considerable increase in the scope of works at the two rural health centers, now upgraded to rural hospitals (with stage 2, at each, to be completed with other financing), and the addition of an additional subcomponent at Tauran Primary School, and five further health sector projects underway. The village development project provided 20 facilities—water supplies, educational facilities, electricity, etc.—to a higher standard (and cost) than was planned during the preparation stage.

28

Component 5: Economic Planning and Diversification

KPI 4: Formulation of a Spatial and Economic Strategy Framework and Institutional Plan for Longer-Term Development Including Coastal Zone Management and Mitigation of Coastal Hazards in East New Britain. Satisfactory. A Comprehensive Development Plan was prepared. It reflects the priorities and opportunities now being targeted as the restoration works are being completed. It draws on the Regional Development Study that identified a growth center hierarchy as a basis for future spatial development in the Gazelle Peninsula including: the Physical Development Study for the Rabaul to Tokua Corridor that set guidelines for development along the coast east of Kokopo; the Disaster Risk Assessment Study that focused on mitigating the future impact of natural hazards including in the Kokopo/Tokua corridor; the Hydrological Mapping Study for the northeast Gazelle; and the Urban and Regional Economic and Spatial Strategy Study consolidating earlier analysis, including the work done to formulate an agricultural sector strategy for the province. Taken together, these activities and documents respond fully to the objective to restore longer-term development planning tools in the province.

KPI 5: Approval of ENBPA of detailed arrangements to integrate GRA into the provincial government. Satisfactory. The provincial and national authorities have agreed to amendments to the Gazelle Restoration Authority Act to incorporate GRA into the ENB Provincial Administration system and expand its mandate to assist development efforts in other provinces. This confirms the widespread recognition of the effectiveness of this as a development model. The National Executive Council (NEC) approval process to confirm this agreement is currently underway.

29

ANNEX 3. ECONOMIC AND FINANCIAL ANALYSIS

Not applicable

30

ANNEX 4. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION PROCESSES

(a) Task Team members

Names Title Unit Responsibility/

Specialty Lending Stuart Whitehead Urban Finance Specialist EASUR Task Team Leader

Robert David Cohen Financial Management Specialist EASUR Financial Management

Richard Scheiner Procurement Specialist EASUR Procurement

Supervision/ICR

Stuart Whitehead Urban Finance Specialist EASUR Task Team Leader- to 11/04

Thakoor Persaud Lead Economist EASTE Task Team Leader Nurul Alam Sr Procurement Specialist EAPCO Procurement Lawrence Leonard Carlson Consultant EASRE Municipal Engineer

David Michael Chandler Sr. Financial Management Specialist EAPCO Financial Management

Peter Brandriss Portfolio Analyst EASOP Portfolio Analysis Colleen Mary Gollach Consultant EACNF Urban and Water Bruce M. Harris Consultant EASTE Social Scientist Bisma Husen Procurement Spec. EAPCO Procurement Cristiano Costa e Silva Nunes Procurement Spec. EAPCO Procurement Sudesh Ponnappa Senior Program Assistant EAPCO Assistant Esme Jaya Abedin Operations Analyst EACNQ Operations Melissa Sanchez Program Assistant EASTE Assistant Inneke H. Ross Sr. Program Assistant EASUR Assistant Bobbie Brown Program Assistant EASUR Assistant Sandra Walston Short Term Temporary EASUR Assistant

31

(b) Staff Time and Cost

Stage of Project Cycle

Staff Time and Cost (Bank Budget Only)

No. of staff weeks USD Thousands (including travel and consultant costs)

Lending FY98 106.25 FY99 151.98 FY00 17 59.90 FY01 0.00 FY02 1.83 FY03 0.00 FY04 0.00 FY05 0.00 FY06 0.00 FY07 0.00 FY08 0.00

Total: 17 319.96

Supervision/ICR

FY98 0.00 FY99 0.00 FY00 5 29.39 FY01 11 59.82 FY02 10 50.22 FY03 9 64.79 FY04 4 47.25 FY05 5 71.24 FY06 5 77.81 FY07 4 78.80 FY08 1 24.71

Total: 54 504.03

32

ANNEX 5. BENEFICIARY SURVEY RESULTS

Not applicable

33

ANNEX 6. STAKEHOLDER WORKSHOP REPORT AND RESULTS Not applicable

34

ANNEX 7. SUMMARY OF BORROWER'S ICR AND/OR COMMENTS ON DRAFT ICR

GRA (Borrower’s) Completion Report – Summary (Full text in project files) 1. Assessment of the project objective, design, implementation and operation experience

Overall, the project has performed well with respect to the overall goal of restoring the social and economic well-being of the population of the North East Gazelle Peninsula, and to the objectives that were set up in the PIP and the PAD. The primary constraint to completing all the components designated to achieve the objectives was one of increased costs, and the secondary constraint was the deferring or cancellation of components due to the inability of the stakeholders to agree on a scope of works or to establish the guiding criteria or strategies of their component projects. With the exception of one component - the Rabaul to Kokopo Road - the design and implementation of all projects was undertaken by the GRA PIU, with technical assistance from a variety of consultants for the physical design works. The significant increase in the overall cost of the project components, is attributable to a lesser extent to inflationary costs during the life-cycle of the project, at 33.1 percent, and to a significantly larger extent to the increase of scope of work on almost every project, accounting for a 103.4 percent increase of those projects that were implemented, and a 198.8 percent increase of those few projects that remain unfunded. During the construction of the project components, the development of the PIU’s contract administration and quality control manuals made a significant contribution to overall quality and cost control management, resulting in fair value for money being obtained on all of the works. Overall, the PIU has attempted to maintain a spirit of co-operation and teamwork with each and every contractor to reduce the risk of non-performance by the contractors, and to achieve the programmed goals and outputs with the minimum of conflict. The sound partnership and harmonious relationship that has been consolidated between the GRA and the respective provincial government agencies does much to ensure the sustainability of the various facilities constructed under the project. The vast majority of the projects were handed over to the ENB Provincial Administration, with some of these being further handed over to other agencies —PNG Ports, Department of Works, CCI PNG, PNG Water Board, and the like. The GRA has provided all end-users with an operation and maintenance manual, to assist with the management of these assets. The Provincial Administration is well aware of its obligations to manage and maintain its assets, and therefore, despite the climate of poor operation and maintenance of public assets in the country, the operation of the assets created or upgraded under the project has not been as bad as it might have been. The province has the will to operate and maintain its assets and has a plan of action for the ENB government’s operation and maintenance of the assets developed under the GRA’s programs. It is the revenue that is lacking. It is most likely that those assets that depend on funding from the National Government are by and large destined to be under-maintained. However, those that are suitable for user-pay systems have their own revenue-generating ability and are better maintained.

35

The project beneficiaries in the respective areas where amenities have been provided are accessing them and show great appreciation for the interventions. 2. Evaluation of the borrower's own performance during the evolution and implementation of the project The Borrower’s performance both during the establishment and during the implementation of the project was good. Many problems were encountered during the implementation phase. Some of the more critical of these in terms of impacts on implementation progress and / or demands on GRA management capabilities were the timely release of adequate counterpart funding by the National Government; the skills and knowledge deficiencies of key stakeholder organizations; the complexity of, and variation in the review procedures, documentation and reporting requirements of the various donors; the poor performance of consultants that were not in tune with local requirements and appropriate practices; and the practical difficulties during the construction of some infrastructure projects. During the project design period, there was an unusual level of co-operation between the relevant government agencies and the GRA. Frequent meetings were held, and a continuous dialogue - particularly with the ENB Provincial Administration - ensured that all interests were catered for as much as possible. This dialogue continued throughout the implementation period, and, combined with the high motivation of the staff at the GRA PIU, did much to keep the project running as smoothly as possible. There were many positive outcomes within the GRA’s organization as procedures and systems were developed during the life of the project. For financial and progress reporting, the GRA developed its own project management reporting (PMR) system allowing financial reporting in a variety of format—by project, by sector, by funding source, etc—with a format that is flexible enough to be consistent with the requirements of the many organizations to which the GRA is bound to report. A wide variety of progress report formats were also developed for a similar range of stakeholders. During the life of the project, the GRA entered into thirty-seven contracts various consultancy contracts for difference technical assistance requirements. With a few notable exceptions, the performance of consultants was not as satisfactory as had been anticipated, and, as the technical experience of the PIU staff increased, more and more of the consultants’ work was taken over by the PIU. Similarly, during the life of the project, the GRA entered into one hundred and one construction, and seven supply contracts, a procurement achievement in itself, and a challenge to any organization in terms of quality control and meeting donor requirements. In 2003/2004, the GRA PIU developed its own contract administration and quality control manuals in order to improve and optimize the administration and management of its construction contracts. With almost no exceptions, the PIU supervised all of the construction contracts.

36

Throughout the implementation of the project, the PIU held regular three-weekly implementers’ meetings to review the progress and discuss the problems on each and every project. These meetings were always attended by the PIU’s Community Development Officer and Environmental Officer, whose assistance in community and other consultation and in environmental management was invaluable. We are often asked why the Gazelle Restoration Authority has been successful and whether our model can be reproduced elsewhere? It is true that there are certain factors that have contributed to the GRA’s success, but we believe that there is no magic formula for reproduction of the GRA model. It is the individuals who make up the team that are responsible for the GRA’s success to date with independence, flexibility, character, and teamwork being key elements. 3. Evaluation of the performance by the Bank, any co-financiers, or other partners during the evolution and implementation of the project It would be fair to say that the co-operation and support of the Bank through its Task Managers and their teams was just about as good as it could be, during the preparation and implementation of the project. The support of AusAID on the Rabaul to Kokopo road component, and in the provision of technical experts for the initial phase of the air quality study, for the disaster risk assessment report, and for the economic and spatial development studies was highly appreciated. Further support was forthcoming from the European Commission that took over some twenty-three of the projects. Although highly appreciated, this support was often seriously hampered by the EC’s complex and impenetrable administrative procedures. Throughout the project, the co-operation and support of the East New Britain Provincial Administration was immeasurable, and, generally speaking, the GRA has received understanding, support and assistance from all sectors of the government and the community. The single most damaging event that had the widest implications on the relationship between the GRA PIU and the Bank was the investigation into corruption and fraud that was undertaken early in 2004. The GRA PIU was never given any official feed-back on this investigation. Other problems were minor by comparison, and were far outweighed by the positive relationship between the Bank’s staff and the PIU. They included the time lost in obtaining No Objection Letters, and variations in requirements. On the other hand, the positive impact of the Bank’s support cannot be overstated. From the start of the project the Bank’s Task Managers have been extremely supportive in all aspects of the project including assistance with development of the project, assistance with implementation and reporting, assistance with financing, and assistance with discussions with the national government on the GRA’s behalf, showing that there was an excellent degree of co-operation between the two organizations during the life of the project.

37

The main lesson learned from the GRA PIU’s working relationship with the various partners on the Medium Term Restoration Program has been that continuous dialogue with the partners is essential to the smooth and continuous operation of the programme. A second lesson has been that, however hard it was to assimilate the different procurement and procedural guidelines of the different donors, and, indeed, the changes to procedures during the lifespan of the project and / or individual components, the guidelines had to be followed; and the changes accepted. The GRA PIU has done its best to cope with the “moving goalposts” during the project, and gratefully acknowledges those who have worked behind the scenes to assist.

38

ANNEX 8. COMMENTS OF COFINANCIERS AND OTHER PARTNERS/STAKEHOLDERS

AusAID agrees that factors for success include the high quality tender and contract documentation and procurement process, the strong management and supervision by site staff; and the positive involvement of the local authorities and communities taking ownership of the project. The ENBPA policy of not paying land compensation in hazardous zones was also a factor for success, and this and the GRRP community involvement model could be adopted on other similar projects in PNG. Evidence indicates that the project has generated many benefits for the general population in terms of establishing a more reliable transport system and lower operating costs which also manifests itself in better and cheaper access to markets, social, education and health facilities. Some of these benefits include improved public transport, reduced travel time and vehicle operating cost, increased economic activity along the roadsides, increase in vehicle ownership, introduction of 25- seater buses due to increased demand for travel, increased attendance at schools, expansion in agricultural production (access to markets) and less traffic on secondary roads. AusAID will be conducting an Impact Assessment of the GRRP in early September 2008 to capture the socio-economic impact of the AusAID project on the local community. A number of secondary beneficial impacts were achieved, including; the accreditation of the GRA Soils Laboratory; removal of UXO which has made the Rabaul–Kokopo Road much safer, seawalls were replaced; pullover bays and shelters have been provided at bus stops. The Community Participation activities provided an effective model for future projects.

39

ANNEX 9. LIST OF SUPPORTING DOCUMENTS

1. The World Bank, Second Gazelle Restoration Project: Project Appraisal Document (Report No. 19278-PNG), October 1, 1999.

2. The World Bank, Second Gazelle Restoration Project: Loan Agreement (Loan No.4525 PNG), between The Independent State of Papua New Guinea and International Bank for Reconstruction and Development, January 19, 2000. 3. The World Bank, Second Gazelle Restoration Project: Project Agreement (Loan No.4525 PNG), between International Bank for Reconstruction and Development and Gazelle Restoration Authority, January 19, 2000. 4. The World Bank, Second Gazelle Restoration Project, Procurement Plan, June 2005 5. The World Bank, Second Gazelle Restoration Project: Amendments to Loan Agreement – September 21, 2002; June 30, 2005; and December 19, 2006 6. The World Bank, Aide-Memoires of the Second Gazelle Restoration Project, from October 1997 to May 2007. 7. ENBPA Division of Planning and Research, Resettlement Action Plan, January 1, 1999 8. Gazelle Restoration Authority PIU, Quarter Status Report 1999 - 2007 9. Gazelle Restoration Authority PIU, Medium Term Restoration Programme, Monthly Reports 2006- 2008 10. PIU, Gazelle Restoration Authority PIU, Medium Term Restoration Programme Implementation Completion Report GRA (Borrower’s) Contribution, May 30, 2008 11. Gazelle Restoration Authority – PIU Medium Term Restoration Program – Environment Management Plan, April 2007 12. East New Britain Provincial Administration, General Observation of Present and Socio-Economic Conditions of ENBP particular of Gazelle Peninsula, October 12, 1998 13. Dept. of Agriculture and Livestock, Konedobu, Land Evaluation of Kboku, Molkolkol and Aliananda Lands Open Bay, East New Britain, Report No. 574, March 1998 14. Gazelle Restoration Authority, Design and Documentation of Various Road Projects, Gazelle Peninsula East New Britain Province, Investigation Report by Over Arup & Partners Pacific Consulting Engineers, November 1996 15. Customary Land Tenure Issues in PNG Report, AusAid, July 1995

40

16. Final Report, The Study on Tokua Airport Development Project in Papua New Guinea, prepared by JICA, March 1992 17. GRA Borrower’s Completion Report, May 30, 2008