39282 THE WORLD BANK INDEPENDENT EVALUATION GROUP An Independent Evaluation of the World Bank's Support of Regional Programs Case Study of the Africa Regional Trade Facilitation Project John Eriksson Director-General: Vinod Thomas Director: Ajay Chhibber Manager: Victoria Elliott Task Manager: Catherine Gwin 2006 This paper is available upon request from IEG. The World Bank Washington, D.C. ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE IN EVALUATION The Independent Evaluation Group (IEG) is an independent unit within the World Bank; it reports directly to the Bank's Board of Executive Directors. IEG assesses what works, and what does not; how a borrower plans to run and maintain a project; and the lasting contribution of the Bank to a country's overall development. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of the Bank's work, and to provide accountability in the achievement of its objectives. It also improves Bank work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. IEG Working Papers are an informal series to disseminate the findings of work in progress to encourage the exchange of ideas about development effectiveness through evaluation. The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Contact: Independent Evaluation Group Knowledge Programs and Evaluation Capacity Development (IEGKE) e-mail: eline@worldbank.org Telephone: 202-458-4497 Facsimile: 202-522-3125 http:/www.worldbank.org/ieg Contents Acronyms..........................................................................................................................iii Preface............................................................................................................................... iv Evaluation Objectives and Methodology................................................................iv Evaluation Criteria.................................................................................................iv Executive Summary......................................................................................................... vi 1. Introduction..................................................................................................................1 Challenges Facing the Sector..................................................................................1 Regional Program Summary Description................................................................2 2. Relevance: Rationale, Alignment, and Design ..........................................................4 Subsidiarity Principle ..............................................................................................4 Alignment with Country, Regional, and Bank Goals and Strategies.......................5 Regional Consensus.................................................................................................6 Design of the Regional Program .............................................................................6 Clarity and Monitorability of Objectives.................................................................8 3. Efficacy: Outcomes, Impacts, and Sustainability ...................................................11 Achievement of Objectives.....................................................................................11 Capacity Building ..................................................................................................13 Realized Distribution of Costs and Benefits ..........................................................14 Risk to Outcomes and Impacts...............................................................................14 4. Efficiency: Governance, Management, and Financing ..........................................16 Efficient Use of Resources .....................................................................................16 Governance, Management, and Legitimacy ..........................................................17 Financing...............................................................................................................19 Donor Performance ...............................................................................................19 5. Monitoring and Evaluation.......................................................................................20 6. World Bank Performance.........................................................................................21 Comparative Advantage......................................................................................21 Quality of Support and Oversight......................................................................21 i Structures and Incentives.......................................................................................22 Linkages to Other Bank Country Operations ........................................................23 Disengagement Strategy ........................................................................................24 7. Country Participation................................................................................................25 8. Conclusions.................................................................................................................26 Summary of Findings.............................................................................................26 Implications for Effective Support .........................................................................27 Annex A: Background Information on the Regional Program ...................................28 Annex B: Governance and Management Arrangements .............................................29 Annex C: Goals, Objectives, Outcomes, Outputs, and Activities................................30 Annex D: Persons Consulted...........................................................................................31 Annex E: References........................................................................................................32 Tables Table 3.1: Target Values and Premiums of ATIA Policies ...............................................13 Table 3.2: IDA Disbursements as Security on ATIA-Issued Policies.........................16 Box Box 2.1: RTFP Performance Monitoring: a Plethora of Indicators...................................10 ii Acronyms AFR Africa Region AGF African Guarantee Facility ATIA African Trade Insurance Agency CAS Country Assistance Strategy CEO Chief Executive Officer COMESA Common Market for Southern and Eastern Africa CUO Chief Underwriting Officer ECA Europe and Central Asia Region ECOWAS Economic Community of West African States ESA Eastern and Southern Africa EU European Union IDA International Development Association MIGA Multilateral Investment Guarantee Agency MTR Mid-Term Review NEPAD New Economic Partnership for African Development PAD Project Appraisal Document PHRD Policy and Human Resources Development QAG Quality Assurance Group RTFP Regional Trade Facilitation Project TAC Technical Advisory Committee iii Preface EVALUATION OBJECTIVES AND METHODOLOGY 1. This review of the Regional Trade Facilitation Project (RTFP) is one of 19 reviews undertaken as part of an independent evaluation by the Independent Evaluation Group (IEG) of the effectiveness of World Bank support for multicountry regional programs over the past 10 years (1995­2004). Twelve of the reviews, including this RTFP assessment, are desk reviews; the other seven reviews are in-depth field studies. 2. The review draws on core program documentation as well as program progress reports, related Bank country assistance strategies, and interviews with key Bank staff. In addition, the evaluation team conducted a group interview with African Trade Insurance Agency management in Nairobi, Kenya, June 30, 2005. See Annexes D and E, respectively, for a list of people consulted and references. 3. The purpose of this case study is to evaluate the relevance, effectiveness, and efficiency of the RTFP in achieving its development objective, as well as the adequacy of World Bank support. There have been some limitations in assessing the RTFP because although there a Quality of Entry Review was undertaken by the Quality Assurance Group in 2001, no commissioned or independent implementation evaluations had been undertaken at the time of this review. A formal Mid-Term Review, originally planned for 2004, then rescheduled to December 2005 and March 2006, was held in May 2006. EVALUATION CRITERIA 4. The 19 reviews use the IEG evaluation criteria of relevance, efficacy, and efficiency. In addition, they assess the Bank's performance and examine the performance of the regional program's participating countries. The key evaluative questions addressed under these criteria--designed to deal with the special characters of multicountry programs--are as follows. Relevance · Subsidiarity: To what extent is the program being addressed at the lowest level effective, and either complements, substitutes for, or competes with Bank country or global programs? · Alignment: To what extent does the program arise out of a regional consensus, formal or informal, concerning the main regional challenges in the sector and the need for collective action? To what extent is it consistent with the strategies and priorities of the region/subregion, countries, and the Bank? · Design of the regional program: To what extent is program design technically sound, and to what extent does it take account of different levels of development and interests of participating countries, foster confidence and trust among participants necessary for program implementation, and have clear, monitorable objectives? iv Efficacy · Achievement of objectives: To what extent has the program achieved, or is it likely to achieve, its stated objectives, including its intended distribution of benefits and costs among participating countries? · Capacity building: To what extent has the program contributed to building capacities at the regional and/or participating country levels? · Risk to outcomes and impact: To what extent are the outcomes and impacts of the program likely to be resilient to risk over time? To what extent have the risks to project outcomes been identified and measures to mitigate them been undertaken? · Monitoring and evaluation: Has the program incorporated adequate monitoring and evaluation processes and taken account of available findings? Efficiency · Efficient use of resources: To what extent has the program realized, or is it expected to realize, benefits by using reasonable levels of time and money? · Governance, management, and legitimacy: To what extent have the governance and management arrangements clearly defined key roles and responsibilities; fostered effective exercise of voice by program participants and coordination among donors; contributed to or impeded the implementation of the program and achievement of its objectives; and entailed adequate monitoring of program performance and evaluation of results? · Financing: To what extent have financing arrangements affected positively or negatively the strategic direction, outcomes, and sustainability of the program? World Bank's Performance · Comparative advantage and coordination: To what extent has the Bank exercised its comparative advantage in relation to other parties in the project and worked to harmonize its support with other donors? · Quality of support and oversight: To what extent has the Bank provided adequate strategic and technical support to the program, established relevant linkages between the program and other Bank country operations and an appropriate disengagement strategy for the program, and exercised sufficient oversight of its engagement? · Structures and incentives: To what extent have Bank policies, processes, and procedures contributed to, or impeded, the success of the program? Participating Countries' Performance · Commitments and/or capacities of participating countries: How have the commitments and/or capacities of participating countries contributed to or impeded the success of the program? Have one or more countries exercised a primary leadership role? · Program coordination within countries: To what extent have there been adequate linkages between the regional program's country-level activities and related national activities? v Executive Summary Background 1. Public and private sector leaders in Eastern and Southern Africa view expansion of the private sector as an essential ingredient for sustained economic growth and poverty reduction. Limited access to finance for cross-border intra- and extraregional trade is thought to be a constraint on private sector­led growth. Assessments have pointed to high levels of political risk resulting from fluctuating and arbitrary government policies, war, and civil commotion as one obstacle to the needed financing, and they have called for official action to overcome a shortage of political risk insurance in the region. Although this kind of insurance is readily available worldwide, it is practically nonexistent for trade within Africa because private insurers either will not provide coverage or they charge exorbitant premiums. In response, government members of the Common Market for Eastern and Southern Africa (COMESA)1 created the Africa Trade Insurance Agency (ATIA) in 1997, with technical support from the World Bank. Program Summary Description 2. In 2001, the World Bank launched the Regional Trade Facilitation Project (RTFP) to provide funding to ATIA and its seven founding countries--Burundi, Kenya, Malawi, Rwanda, Tanzania, Uganda, and Zambia. Specifically, RFTP provided an International Development Association (IDA) credit of $5 million to ATIA to cover its operating expenses for an initial two years, and separate IDA credits to each of the seven countries to be used by ATIA to supply short- and medium-term (up to three years) political risk insurance to domestic and foreign firms that export and import within and outside the subregion. These country credits, which varied in size according to each country's per capita income and expected demand, amounted in total to $105 million. In 2005, IDA provided another $5 million to ATIA, credits of $10.9 million to two new members (Democratic Republic of Congo, $10 million and Madagascar, $0.9 million), and a second credit of $7.5 million to Burundi. Liberia was accepted as a tenth ATIA member at the December 2005 General Assembly meeting. RTFP is expected to continue until 2011. Rationale for the Regional Program 3. RTFP's focus on impediments to private sector growth is relevant, and the project has received expressions of strong support from senior government officials of the seven founding member countries. The strongest argument for the project's regional approach is its potential for efficiency gains, which derive from having a single agency, rather than individual country ones, as the provider of trade-related insurance, especially in the context of scarce underwriting skills within the participating countries. 1COMESA played an important legitimizing and policy-making role in laying the foundation for ATIA by serving as an interlocutor among member governments and an "incubator" role by sponsoring and assisting ATIA at the early stages. It is still an active ATIA board member. vi 4. But the project substantially overestimated the demand for political risk insurance. Most regional private sector agents have emphasized that market demand for short-term trade insurance is more concerned with commercial than political risk. So the linkage between political risk insurance and needed trade financing was tenuous at best. Quality of Design and Implementation 5. While ATIA has offered solely short-term and medium-term political risk insurance to importing and exporting firms, its design anticipated that the political risk cover would be combined with short-term commercial risk insurance provided by private agents. This scheme drew from the experience of similar Bank-funded projects in Eastern Europe, which started covering only political risk and expanded to commercial risk insurance and other products. The same trajectory was envisioned for ATIA as it became financially sustainable. But the following design weaknesses, which became apparent in the initial years of implementation, prevented ATIA from progressing to that point. · Faulty product mix. The project design assumed there would be substantial demand for relatively short- and medium-term political risk insurance offered by ATIA and that ATIA would team up with an international private insurance broker that would offer commercial risk insurance. But demand for short-term political cover was lower than anticipated and the private firm that partnered with ATIA opted out of issuing commercial risk insurance to firms located in Africa. This left ATIA restricted to offering a product of limited relevance. · Limited financial structure. Until recent favorable restructuring decisions,2 ATIA was not authorized to pool IDA credit proceeds as risk capital, thereby not providing it the leverage to mobilize additional capital from the private sector as well as the ability to reinsure or to directly offer commercial risk insurance. · Reliance on promotion by the public sector. The RTFP design called for ATIA liaison offices attached to government ministries; these offices were not effective promoters of ATIA among the private sector. · Noninclusion of some major exporting countries in the subregion--for example, South Africa. Program Achievements 6. ATIA's performance against its principal objective of improving access to financing for intra- and extraregional trade has fallen short of the original projection. The issuance of political risk insurance policies by ATIA has been considerably lower than projected (with the exception of Burundi). As a result, the agency did not reach self- sufficiency in terms of premium income covering operating expenses by 2003 as originally envisaged. In fact, the total net deficit increased in 2004 to $1.3 million from 2ATIA General Assembly and Board of Directors (December 2005), and the World Bank Board of Executive Directors (December 2006). The Bank's Board also approved the consolidation of the individual IDA credits into a single regional project. vii $1.2 million in 2003. Lower than estimated demand for political risk insurance and the inability to identify a partner willing and able to carry through on a commitment to issue commercial risk insurance to African private sector entities are two main reasons for this weak performance. Another reason is the long delays and poor choices in the initial hiring of the Chief Executive Officer (CEO) and Chief Underwriting Officer (CUO). Weak marketing efforts and reliance on public sector agents in member countries are other reasons. 7. A substantial financial turnaround occurred when during the first nine months of 2005 ATIA experienced a net surplus of $408,826, in contrast to a deficit of $723,266 for the same period in 2004. This improvement was the result of three factors unrelated to ATIA's insurance sales performance: a) a sharp rise in finance income, as opposed to premium, revenues--namely, more than a 170 percent increase in interest income earned on reserves; b) a substantial write-back of staff contingency liabilities; and c) a significant reduction in some expenditure items, including not incurring salary expenses of the then vacant CEO and CUO positions. Net underwriting income actually fell over the same period by more than half, from $392,754 to $177,330.3 This disappointing operating performance occurred in spite of more aggressive marketing efforts by ATIA, greater utilization of private sector agents, performance targets for premiums, and an acceleration of inquiries. 8. Three major actions undertaken and approved by the General Assembly and Board of Directors during the last half of 2005 are critical to ATIA's ability to reverse the direction of its operating income performance and its prospects for sustainability: · The hiring of a new CEO and CUO · The decision to expand ATIA's product mix to commercial risk insurance · The agreement by countries to restructure ATIA to permit conversion of their IDA credits into pooled equity capital for the agency.4 9. In view of ATIA's performance thus far, it is not possible to predict the sustainability of the program's outcomes and impacts. While ATIA has recently generated sufficient income to meet its budget, this is mainly the result of interest earnings on its assets and other factors that cannot be counted on to continue. In the meantime, net underwriting income declined by more than half from the first three- quarters of 2004 to the same period in 2005. Long-term, ATIA's financial sustainability depends on the success of the restructuring effort and the effectiveness of new management. 3Financial data from Ernst and Young, African Trade Insurance Agency: Interim Management Letter: 31 December 2005, pp. 13­19. Net underwriting or net premium income is defined as gross premium income plus insurance commission income, less insurance commission expenses. 4African Trade Insurance Agency, Resolutions Adopted by the Special General Assembly, Nairobi, December 9, 2005, Resolution 4. The latter measure will permit mobilizing additional capital from the private sector as well as the ability to reinsure. Interview with MIGA staff member, January 26, 2006. As noted previously, the Bank's Board of Directors approved this restructuring in December 2006. viii Effectiveness of World Bank Performance 10. The Bank played an instrumental role in shaping the ATIA and getting it off the ground. But its early assistance was marked by three shortcomings. The projection of market demand for political risk cover by Bank staff was exceedingly optimistic. The Bank provided little support to ATIA during the critical early years of implementation when ATIA should have paid more attention to outreach and marketing. The Bank also provided inadequate supervision over the first CEO, whose working relationships with country counterparts and Bank Country Directors tended to be counterproductive. ATIA staff members contend that ATIA performance would have been stronger if a Bank staff member had been seconded to ATIA during the first year or two of implementation to provide more support for staff recruitment, training, marketing, and private sector orientation efforts in member countries. 11. In recent years, the Bank has worked effectively with ATIA to secure new top leadership. It has also worked intensively with ATIA and its member countries to help plan and facilitate the restructuring process. ix 1. Introduction CHALLENGES FACING THE SECTOR 1.1 Expansion of the private sector is viewed by leaders in Eastern and Southern Africa (ESA) as an essential ingredient for sustained economic growth and poverty reduction. 5 One constraint on expansion is alleged to be limited access to finance for cross-border intra- and extraregional trade. Intraregional trade has been very small, less than 10 percent of total ESA trade, which according to a recent Bank study, is "substantially lower ... compared to regional trading blocks in other parts of the world."6 Africa's share in world trade continues to be very modest, at 1.4 percent in 2001, and dominated by primary exports, which accounted for 75 percent of total African merchandise exports outside the region in 2000­2001.7 1.2 Despite the rhetorical commitment to African regional integration and economic development, private sector expansion has been very limited. Interviews and surveys have reported that banking institutions were not able to lend to most trade activities due to their own internal risk controls, and that private insurance firms were very reluctant to assume risks posed by the actions of sovereign powers. Entrepreneurs within and outside the subregion are said by project documentation and World Bank staff to perceive high levels of political and commercial risks as a result of fluctuating and arbitrary government policies, as well as damaging political events, such as war and civil commotion.8 Notwithstanding substantial political risk coverage worldwide at about US $100 billion by private insurers and by governmental and intergovernmental agencies, such as the Multilateral Investment Guarantee Agency (MIGA), political risk coverage for intra­sub-Saharan African trade is practically nonexistent because private insurers will either not provide coverage or they charge exorbitant premiums.9 1.3 In response to these challenges, government members of the subregional organization, Common Market for Eastern and Southern Africa (COMESA), developed in 1997 a proposal for an African Guarantee Facility (AGF). The Bank's supportive role 5The heads of state in the New Economic Partnership for Africa's Development (NEPAD) have identified the following priorities: regional integration, building competitiveness, promoting diversification of exports, accelerating intra-African trade, improving access to markets in developed countries, and improving Africa's share in world trade. Findings of a joint report by the United Nations Economic Commission for Africa and the World Bank, Can Africa Claim the 21st Century, estimate that seven percent economic growth per annum on average is necessary over the next 10 to 15 years to achieve "significant and sustainable poverty alleviation in Africa." Africa Regional Trade Facilitation Project, Project Initiation Document, Report 7384, August 1, 2000. 6World Bank, Salient Features of Trade Performance in Eastern and Southern Africa, Africa Region Working Paper Series No. 76, October 2004, p. 6. 7South Africa is excluded, which would otherwise distort the overall numbers. World Bank, World Development Report 2003, pp. 240­241. 8World Bank, Project Appraisal Document, March 2001. 9Interview with MIGA staff member, January 29, 2006. 1 was instrumental at this stage. A World Bank staff member, seconded to the COMESA Secretariat, identified a Bank task manager who had developed three financial guarantee projects that initially covered political risk in Albania, Bosnia, and Moldova. These projects included features that seemed relevant to conditions in the ESA region. The task manager was loaned and ultimately transferred to the Bank's Africa Region where he helped COMESA develop the AGF, which in turn, evolved into the Africa Trade Insurance Agency (ATIA). The European Union (EU) and Japan, through the Bank- administered Policy and Human Resources Development (PHRD) Fund, provided $2 million in technical assistance grants to assist project preparation during 1997­2000. COMESA in effect served as an "incubator" for ATIA, and continues to play an active role on the ATIA Board of Directors.10 Concurrently, COMESA and the Southern African Development Community (SADC) advocated complementary measures to stimulate growth and trade, including the liberalization of economies through cooperation in trade, privatization, and regulatory changes.11 REGIONAL PROGRAM SUMMARY DESCRIPTION 1.4 The Regional Trade Facilitation Project (RTFP), approved by the Bank's Board of Directors in April 2001, was designed to support the start-up and operation of ATIA. Its stated objective is to contribute to poverty alleviation through private sector­led growth in participating countries by improving access to financing for productive transactions and cross-border trade. 1.5 The project's initial financing structure entailed an International Development Association (IDA) credit to each of the seven ATIA funding member countries: Burundi, Kenya, Malawi, Rwanda, Tanzania, Uganda, and Zambia. These IDA credits, totaling US $105 million, were to provide the capital for ATIA's issuance of political risk insurance. The amount of the IDA credits to each country varied according to the countries' per capita income and expected insurance demand. In addition, a credit of US $5 million was made directly to ATIA to cover its operating and start-up expenses for an initial two years. In 2005, IDA provided an additional credit of US $7.5 million to Burundi, credits of $10 million and $0.9 million, respectively, to two new members, Democratic Republic of Congo and Madagascar, and an additional credit of US $5 million to ATIA.12 A tenth member country, Liberia, was accepted in December 2005. According to the project appraisal document (PAD), the program was designed to continue until 2011, by which 10For example, it organized a visit by members of the ATIA Board and management as well as Bank management to the headquarters of the Economic Community of West African States (ECOWAS) in Nigeria, in April 2006, to discuss membership of ECOWAS member countries in ATIA. 11In addition, a program initially called the "Cross Border Initiative" was jointly sponsored by the World Bank, International Monetary Fund, EU, and African Development Bank. 12Djibouti and Eritrea are in the process of joining ATIA. A credit by IDA to an organization rather than a government was unusual but not unprecedented. It is permissible under the IDA charter because the member governments of ATIA signed agreements with the Bank ratifying their ownership of ATIA and, in effect, certifying its creditworthiness. Interviews with Said N. Al Habsy, LEG, September 2, 2005, and Onno Ruhl, AFR, January 10, 2006. 2 time IDA assistance was expected to have been fully utilized and ATIA to have become a self-sustaining private entity.13 1.6 An alternative institutional vehicle might have been MIGA, but MIGA is restricted to longer-term guarantees (three years or more); requires equity to be insured in order to cover third-party debt, which ATIA does not; and cannot provide coverage for any commercial risks. In effect, the proposed provision of political risk insurance by ATIA is complementary to insurance offered by MIGA. 13This need not, however, imply that government members would divest themselves from ATIA. Bank and ATIA staffs argue that if this were to happen, ATIA would lose the deterrence effect of peer pressure among member states to avoid or minimize political risk losses. 3 2. Relevance: Rationale, Alignment, and Design 2.1 Summary. The focus of the RTFP on impediments to private sector growth is relevant to the development objectives of its participating countries and has received support from government and private sector representatives in the seven ATIA founding member countries. The main rationale for the project's adoption of a regional approach is its potential to achieve economies of scale in the provision of trade-related insurance in a context marked by scarce underwriting skills within participating countries. But significant weaknesses in the project's design became apparent in the initial years of implementation and required a major project restructuring. SUBSIDIARITY PRINCIPLE 2.2 The subsidiarity principle states that a program should be organized and carried out at the lowest level consistent with cost-effectiveness. The PAD gives several reasons for a regional insurance facility as opposed to independent country facilities. It also cites the paucity of private sector risk insurance in Africa. · Economies of scale. Unit operating costs should be lower for one regional agency than for a number of individual agencies, each operating in a different country. · Scarce underwriting skills. A related argument is that the use of scarce underwriting skills would be more efficient if fielded from one agency rather than several. · Improving Africa's private sector image. Project designers believed that a regional facility would send a signal that private sector­friendly policies were being pursued throughout the subregion. To the extent nonmember countries become friendlier to the private sector as a result of ATIA, this would constitute a type of "neighborhood" or "spillover" effect. · Peer pressure to maintain good policies. To avoid claims, other member countries would presumably convince a country that strays not to intervene in the transactions of its private sector. · Greater independence of ATIA from governments than in a country approach. This was believed to enhance ATIA credibility with the private sector.14 · Less risk of mismanagement in regional programs. ATIA management has argued that accountability and transparency are greater in regional than in country programs.15 14The experiences of Albania, Bosnia and Herzegovina, and Moldova tend to bear out this assertion. All three projects were heavily public-sector oriented, and subsequently shifted toward a private sector orientation. 15Meeting with ATIA management, Nairobi, June 30, 2005. 4 2.3 The validity of these reasons for a regional approach is mixed. The economies-of- scale rationale, including scarcity of skills, is the strongest among the above reasons, and it is backed by evidence from elsewhere in sub-Saharan Africa.16 ATIA management reports that coinsurers and insured banks believe that ATIA has greater autonomy in underwriting and claims settlement than parallel national institutions would have. Peer pressure from other member countries has apparently occurred in some other regional programs and this may have been a factor in ATIA experience, but evidence of this is lacking.17 The rationale that Africa's private sector image would be improved by a regional project was weakened by the initial use by ATIA of public sector liaison offices. ALIGNMENT WITH COUNTRY, REGIONAL, AND BANK GOALS AND STRATEGIES 2.4 The provision of political risk insurance18 via the RTFP is consistent with the strategies and private sector focus of the region and countries, as expressed by NEPAD, and member country development strategies.19 These strategies assume that it is the lack of financing that impedes the expansion of private sector productive activity and trade in the region. To address this presumed weakness in financial markets, COMESA initiated the RTFP as a complement to its regional integration activities, such as the introduction of a free trade area, on October 31, 2000. 2.5 The objectives of the RTFP are also consistent with World Bank goals related to private sector development, trade enhancement, and regional integration in sub-Saharan Africa.20 The Bank's Country Assistance Strategies (CASs) for each of the participating countries emphasizes the promotion of private sector­led growth, with a focus on 16PAD, pp. 12­13. The decision to create a multicountry institution resulted from a conclusion that "a country-by-country approach would not enable economies of scale in terms of operating costs as each country would have to set up its own agency." In addition, "feasibility studies for several country-based export credit agencies have shown that, with few exceptions, the necessary underwriting skills required for covering export to third countries could not be maintained in a single country scheme in Sub Saharan Africa without heavy subsidies necessary because of insufficient economies of scale." 17For some evidence from another regional program, see World Bank, IEG, World Bank Support for Regional Programs: A Desk Review of the OECS Solid Waste Management Project (2006). 18Initially, ATIA was restricted to the provision of political risk insurance to interested (i) financial institutions financing exports and imports to participating countries, and (ii) importers from and exporters to participating countries. Commercial risk on public and private obligors was excluded by IDA. Subsequently, parastatal nonpayment risk was approved, and the Annual General Assembly and Board approved in December 2005 the provision of commercial credit risk insurance and sovereign nonpayment risk. The product covers the following: inconvertibility and inability to transfer currency; inability to obtain currency of the insured in the market place; cancellation of licenses and restrictions on imports and exports; imposition or increase of import or export taxes; expropriation; government interference with entities owing insured obligations; seizure of goods; prevention of sale; or prevention of export, interference with the carriage of goods, war or civil disturbance, embargo, diversion of voyage as a consequence of any of these risks, as well as the same risks occurring while goods are in transit in a participating country. 19A recent example of high-level support for ATIA came in a letter of October 10, 2005, from President Paul Kagame of Rwanda to World Bank President Wolfowitz. 20See for example, IDA, Strategic Framework for IDA's Assistance to AFRICA (SFIA): The Emerging Partnership Mode, IDA/SecM2003-0406, June 25, 2003, p. 9. 5 diversification of exports in particular. The RTFP also supports the objectives of the Bank's Strategic Framework for Assistance to Africa to increase growth and enhance competitiveness, and it complements the priority given to regional integration in the Strategic Framework for IDA Assistance. The IDA framework underscores the importance for countries to leverage national development programs to achieve sustainable outcomes through regional approaches such as financing, capacity building, and knowledge sharing. The RTFP is to contribute to improved coordination among subregional organizations, a priority objective for NEPAD. REGIONAL CONSENSUS 2.6 Project stakeholders identified the following challenges facing the region: a) lack of financing for export and production activities due to perception of risk in Africa as a whole and in individual countries; b) political risk insurance, especially for intraregional transactions, is rarely available from commercial sources or export credit agencies, and where it is available, it is very costly; and c) cover is particularly thin or nonexistent for medium-term transactions, thereby restricting the import of essential capital goods. They said that addressing these challenges requires collective action. 2.7 Competing explanations for poor trade performance, which the project documentation acknowledges but does not address, include poor transportation infrastructure, weak markets, lack of products, and poor product quality. 2.8 The experience of the Bank in supporting political risk guarantee/insurance facilities in three countries in Eastern Europe (see footnote 13) reinforced a consensus in COMESA and seven member countries to enlist Bank help, which eventually led to the creation of the ATIA.21 2.9 COMESA assisted in the formation of ATIA, including mobilizing country members and helping formulate the international treaty that established ATIA. According to ATIA management, 50­60 percent of inquiries to ATIA now come from non-ATIA member countries.22 At its meeting in December 2005, the General Assembly of ATIA accepted Liberia as an ATIA member. DESIGN OF THE REGIONAL PROGRAM 2.10 ATIA was designed to offer short- and medium-term political risk cover and, through private insurers, short-term commercial risk insurance, to African firms importing to, and exporting from, member countries. 2.11 The design of some aspects of the RTFP was technically sound. These include: 21The Organization of African Unity (superseded by the African Union), the EU, the Government of Japan, private insurers, MIGA, and the International Finance Corporation also contributed to these discussions. 22Meeting with ATIA management, Nairobi, June 30, 2005. 6 · The governance and management structures of the ATIA, including a detailed operational manual with fiduciary safeguards · The involvement of the private (brokers, exporters, importers) and public sectors in each member country in the design and implementation of the project · An intention to include, over time, commercial risk insurance. 2.12 As previously noted, the RTFP design drew from the experience of similar projects in Eastern Europe, which started with coverage only of political risk coverage and later expanded to commercial insurance and other products. The same trajectory was envisioned by some of the designers of ATIA as it became financially sustainable. A major difference between the European and African experiences was that the latter involves a regional approach. The key rationale for this, according to some stakeholders, was a desire to promote intra-African trade when there was very little of it and exporters were reluctant to export to another African country.23 2.13 The following four weaknesses in the RFTP design became apparent during the initial years of implementation: · Faulty product mix. Project design assumed there would be robust demand for relatively short- and medium-term political risk insurance, when experience showed such demand to be longer-term, and that the stronger short-term demand tends to be for commercial risk insurance.24 IDA management opposed ATIA's issuing commercial cover, one reason being that the commercial credit information base in the region was insufficient for IDA to responsibly authorize the use of its credits to insure commercial risk. To get around this limitation, ATIA established a partnership with Atradius, a Dutch private insurance firm, in order to offer a comprehensive package that would include commercial as well as political risk insurance. But Atradius proved willing to offer commercial insurance only to enterprises in Europe doing business in Africa, not enterprises located in Africa. This left ATIA restricted to offering a product of limited demand. · Limited financial structure. Until December 2006, ATIA was not fully authorized by both its Board and the Bank's Board to capitalize and was unable to 23Bank staff interview, September 2, 2005. Trade in the Europe and Central Asia (ECA) project is mostly EU-oriented, although intercountry trade in southeastern Europe comprises a larger share of their total than does intra-African trade. At the time the ECA projects became effective, the two Balkan countries-- Albania, and Bosnia and Herzegovina--were in the midst of a region still afflicted with conflict. Moldova is separated by Romania from the southern Balkans. So a regional approach would not have made sense. Bank staff interview, January 10, 2006. 24The demand survey evidence used to justify the project grossly overestimated the demand for political risk insurance. Because such insurance virtually did not exist in the region, the survey focused on plans rather than achievements. It is a well known survey phenomenon that respondents tend to be more optimistic about plans than about past achievements. Interview with Bank staff, January 10, 2006. Other factors cited to explain lower than projected demand include overambitious financial targets and implementation schedules. Initial projections for volumes of underwritten business were apparently made on the basis of combined leverage with the private sector, while premium revenues generated occurred on the basis of ATIA lines only. 7 internally leverage its own capital. It could insure only the value of the IDA credits received by its members. In accordance with the preference of its members, it could not reinsure, or pool IDA credits provided to the different members. The Bank project team has engaged in sustained dialogue with members to shift their views in support of pooling IDA credits, which is to be a feature of the current project restructuring. · Reliance on promotion by the public sector. The RTFP design called for ATIA liaison offices attached to government ministries as key stakeholders, but these offices were not effective promoters of ATIA among the private sector. ATIA was slow to give more attention to marketing to the private sector and to change project design from a reliance on public sector liaison offices to private sector agents and brokers. A significant share of policies bound by ATIA has been for parastatals, which depending on government policies, behave more or less like commercial operations. · Noninclusion of some major exporting countries in the subregion--for example, South Africa.25 2.14 These weaknesses have contributed to a significant shortfall in RFTP's performance against its principal objective of improving access to financing for intra- and extraregional trade. The Bank and ATIA have devised corrective measures that entail a significant restructuring of the project, as discussed in the following section. On the basis of expectations of success from the restructured program, the ATIA and the Bank are designing an RTFP II that would initially cover six West African countries: Côte d'Ivoire, Ghana, Liberia, Mali, Nigeria, and Senegal. Preliminary discussions have been held with ECOWAS, which has expressed interest in playing the same "incubator" role vis-ŕ-vis ATIA as COMESA played in the ESA region. CLARITY AND MONITORABILITY OF OBJECTIVES 2.15 The principal objective of the RTFP, to contribute to poverty alleviation through private sector­led growth by improving access to financing, is so general that its relevance is questionable. Project "outcome" is to be measured by the value of the insurance policies issued by ATIA. Nowhere are the linkages between project objectives, outcomes, and outputs laid out. 2.16 The project's indicators of success also are not as relevant as they could be. Effort was made during the project's design phase to formulate monitorable indicators on the basis of project objectives.26 As shown in Box 2.1, the project established 13 indicators of 25Although South Africa is not a member, ATIA can insure transactions with South African firms, providing the other party is an ATIA member. 26The ATIA Board of Directors notes that "performance indicators have been incorporated in the Agency's Credit Agreement, as well as in every member-country's credit agreement, not only as a performance benchmark, but also as a loan covenant. These form the basis of measurement of implementation, performance, and development impact." Minutes of the 17th Meeting of the Board of Directors of the African Trade Insurance Agency. The PAD recognizes that achieving the project objective would depend 8 success. On a quarterly basis, ATIA is tracking the following key input and output indicators: volume of transactions, number of claims, number of members, and financial operations. The volume of lending by commercial banks operating in the sub-region has been difficult to monitor because not all transactions are tied to bank lending. Other indicators are monitored on a continual basis, including perception of risk (which is determined by the gearing ratio and how much risk the private sector would take on). But ATIA has not tracked the single outcome indicator that would be most relevant; namely, the increase in export value insured by ATIA policies. Box 2.1. RTFP Performance Monitoring: a Plethora of Indicators The PAD identifies the following "primary performance indicators" for the project objective: 1) volume of policies issued per country (at another point the PAD indicates that "project output would be measured by the value of insurance policies issued") 2) ATIA becomes self-financing after two years of operation 3) on average one claim or less resulting from a government performance risk per country, per year 4) ATIA enters into a leveraging partnership with private risk insurers for political risk cover.27 Secondary performance indicators include: 1) catalyzing the introduction of private comprehensive trade credit insurance 2) increase in volume of commercial bank lending for trade transactions (target of 10 percent per year) 3) decreasing trend in cost and increasing trend in maturity of trade finance 4) increased number of participating countries over the life of the project 5) broadening of ATIA's client base within a borrower's territory over the life of the project 6) diversification of participating countries' economies, reflected by a decreasing trend in the share of commodities and cash crops as a percentage of Gross Domestic Product (GDP) and exports 7) improved perception of political risks reflected by a higher leverage ratio for private risk insurers participating with ATIA, better country political risk ratings, and new private entrants providing risk coverage. At another juncture, the PAD calls for tracking both the value and number (volume) of policies issued, as well as increases in: 1) imports into the region and in exports from the region, with a target of 7 percent growth per year in the value of trade flows (US dollars) 2) volume of intraregional trade, beginning with a 7 percent growth rate, increasing to 15 percent by project completion in 10 years. 2.17 The PAD recognizes that "It will take a number of years for the project, in conjunction with other initiatives and reforms in the region, to have a measurable impact ... [at outcome and impact levels]. The indicators will be monitored throughout the project life, but it should be noted that the results in the first couple years may not be on continuation of market-oriented reforms, success of regional integration activities, and a degree of political stability in countries and in the region. 27PAD, op.cit, p. 11. A related performance indicator, the "leverage ratio" (ratio of private sector insurance coverage to that of IDA for a given transaction), is closer to an output rather than outcome measure of ATIA's performance. It measures by definition the extent to which private insurers are participating in the ATIA market. 9 representative of the impact the project will have over its ten year life."28 It is true that extra- and intraregional imports and exports are subject to a number of nonproject influences, but growth in trade is the bottom line against which ATIA should be measured.29 2.18 Two indicators were identified to trigger corrective action at the Mid-Term Review (MTR), following three years of operation: a) ATIA is not self-financing and b) the second tranche of IDA credits for individual countries is not yet fully utilized. As explained in the following section on efficacy, after four years of operation, these indicators are far from being achieved. As part of the project's Mid-Term Review (MTR), the indicators were redefined, as a reflection of the experience to date, so as to be more relevant and measurable and accompanied by a revised results framework. Instead of being couched in terms of private sector­led growth and poverty reduction, project success was to have been measured by the extent to which the project facilitates access to financing for inter-regional trade and African exports. 28PAD, op.cit. p. 4. 29Without more precise definition, the first "secondary performance indicator" shown in Box 2.1-- "catalyzing the introduction of private comprehensive trade credit insurance"--is not measurable. Country- by-country performance benchmarks for project outputs measured as total values of insurance policies issued in participating countries over the 10-year implementation period are shown in Chapter 3. 10 3. Efficacy: Outcomes, Impacts, and Sustainability 3.1 Summary. Although human and organizational capacities have been established in Nairobi, ATIA's performance has fallen considerably short of the original projection that growth in insurance volume would have resulted in ATIA operating self-sufficiently by 2003. This shortfall, discussed below, is due to the lower than projected issuance of political risk policies, reflecting an initial, substantial overestimation by Bank staff of the demand for this kind of insurance, and the failure of a European insurance broker, in partnership with ATIA, to offer commercial risk insurance in ATIA member countries or other African countries. A second reason for ATIA's performance shortfall was early difficulty in recruiting top management staff. Although ATIA expected premium income to rise in 2005, the results for the first three quarters fell below those for the same period in 2004. In fact, the expectation at the time of project approval was that premium income would cover operating expenses after the first two years.30 Therefore, financial sustainability of ATIA over the longer term is still in question. ACHIEVEMENT OF OBJECTIVES 3.2 From among the plethora of indicators described in the previous chapter, ATIA and Bank managements have focused on one output indicator--the value of insurance policies issued. But with the exception of post conflict Burundi, where burgeoning demand has required a supplemental IDA credit of $7.5 million, ATIA insurance sales have grown slowly, reaching a cumulative total of US $110 million in June 2005, less than 15 percent of the current adjusted 2011 target.31 Only in 2005, through a combination of increased interest income, mainly from IDA credits, proceeds from policy premiums, and cost-cutting measures, has there been sufficient resources to cover operating costs.32 To insure sufficient resources for operating expenses a second IDA credit of $5 million was issued to ATIA in 2005, supplementing the original credit of $5 million in 2001.33 The ATIA currently projects a break-even point in 2008. 30The PAD states that ATIA's income "would come from premium income charged to policy holders, which after the initial period should be enough to cover operating expenses." Op. cit., p. 16. 31See Table 3.1. The fact that ATIA has insured $110 million--more than the $21 million value of the IDA credits involved--reflects "coinsurance," with other insurers providing parallel coverage, as long as ATIA is the insurer of first recourse. From 2001 to 2004, ATIA and its coinsurers averaged about $28 million a year in insurance coverage. This is about 1.3 percent of the annual volume of trade (exports + imports) of the seven founding ATIA member countries combined over the same period. 32The ATIA database of past enquiries shows that while during the same period that ATIA generated about US $1.5 million in overall premium income, it had also turned down corporate and sovereign nonpayment business that would potentially have generated $129.1 million in policy sales and $34.7 million in gross premiums. Therefore, ATI could have earned substantially more in premium income had it been able to issue policies to these categories of clients. 33A previous estimate projected that ATIA would exhaust its budget by November 2005. The six-month stretch-out to April 2006 was made possible by reductions in expenses and an increase of 2 percentage points in interest earned. The ATIA Second Quarter Financial Position argues that transactions in the following quarters of 2005 are expected to earn more premium income than during the same period of 11 3.3 Table 3.1 indicates the extent to which ATIA has fallen short of original projections. The PAD in 2001 anticipated that by 2011, US $1,280 million in insurance policies would be written by ATIA. By March 2005, this target was reduced to $857 million. Actual sales by mid-2005 had reached only $110 million.34 The last four columns of Table 3.1 show that premium incomes have been falling far short of targets, and in fact declined from FY 2004 to FY 2005.35 If interest income and reserves and cost reductions are taken into account, ATIA would have just generated sufficient income by 2005 to meet its operational budget. Therefore, the sustainability of ATIA remains to be seen. Table 3.1 Target Values and Premiums of ATIA Policies (US $ million) Benchmark Outputs (target and actual ATIA Net Cash Premium Incomes values of policies to be sold by ATIA) Targets to be Actual values reached Target FY Actual FY Target FY Actual FY reached by 2011 by mid-2005 2004 2004 2005 2005 $857 $110 $2.75 $0.40 $1.62 $0.27 Sources: Benchmarks for 2011 are from ATIA, Review of the Performance Indicators under the Country Development Credit Agreements, March 2005. Actual value of policies sold is from October 2005 Supervision Aide-Memoire, Draft, Annex 1. Actual Net Cash Premium Incomes are from ATIA Management Report on Business Status, December 2005, Table 1. 3.4 Although ATIA is still not self-financing and the first tranches of IDA credits for individual countries with the exception of Burundi are not fully utilized, recent trends in ATIA operations may be precursors to better performance. After a slow start, the number of inquiries has been picking up.36 Bank management acknowledges ATIA's renewed efforts to develop a marketing strategy, which includes performance targets to increase premium revenue, and bringing new member countries into ATIA.37 ATIA has by now written policies in all but two countries (Malawi and Rwanda) of its seven founding member countries. The pipeline includes transactions in all founding member states. 2004. Nairobi: ATIA, June 2005. As of mid-2006, ATIA had not drawn on the second IDA credit, which cannot be drawn upon until the first credit is depleted. 34The project team points out that given the need for 100 percent collateral, the target would have required IDA credits of the same amount, or an eight-fold turnover of the original IDA country credits of $105 million. 35The results for the first three quarters of calendar year 2005 as compared with the same period in 2004 are even worse. Net underwriting income for the first three quarters of 2004 was $0.39 million and for the same period in 2005, it was $0.18 million. Ernst and Young, African Trade Insurance Agency: Interim Management Letter: 31 December 2005, p. 13. 36Meeting with ATIA management, Nairobi, June 30, 2005. 37Bank documentation in late 2004 reported that the RTFP had led in some countries to a) the introduction of private commercial risk insurance, b) an increase in the volume of bank lending for trade transactions, c) a decrease in cost and increase in maturity of trade finance, and d) an increase in the number of participating countries in ATIA. 12 3.5 Liberia, although it is not in the ESA region, applied and was accepted as a tenth ATIA member in December 2005.38 Bank and MIGA staff members have mixed views on the desirability of expanding the number of ATIA member countries before a solid track record has been established among existing member countries. 3.6 Other implementation efforts undertaken by ATIA in 2004 and 2005 with the potential to increase performance include: · Marketing events carried out in member countries as well as the design of new brochures and an ATIA "brand." · Greater reliance on the private sector. o Via establishment of liaison agreements with private sector representatives who are closer to the market, replacing the public sector offices appointed by member governments.39 o Appointment of a number of private sector agents/brokers to strengthen the distribution network and market ATIA's products. CAPACITY BUILDING 3.7 The RTFP has supported the creation of ATIA and its human capacity in Nairobi, but the skilled capacity needed for an effective insurance program, particularly underwriting, has been slow in coming. This has been delayed by the lags in hiring the new Chief Underwriting Officer (CUO) and new Chief Executive Officer (CEO). Three to five new underwriters were to have been in place by the summer of 2006. A major training program jointly sponsored by ATIA and the World Bank Institute, funded by a grant of US $0.5 million, was inaugurated at the beginning of calendar year 2005. The initiative includes a multiyear training program in underwriting and other aspects of the business for all ATIA staff and its country business representatives (liaison officers, brokers, and other agents); training programs for in-country investors, parastatals, exporters, and importers; and workshops and country policy sessions on trade finance for COMESA policy-makers. Training sessions have been held, including courses in July and August 2006. 3.8 The member country Technical Advisory Committees (TACs) were set up in each member country to assist in the formation of ATIA, but the TACs lapsed with ATIA's establishment. The ATIA Board of Directors and management continue to constitute a form of organizational and institutional capacity. But ATIA has not been widely known in some member countries, especially in the private sector, nor has it always had proactive representation. The identification of private sector representatives was a step in 38African Trade Insurance Agency, Special General Assembly of the African Trade Insurance Agency, December 9, 2005, Resolution 8. 39However, lead ATIA agent remains a public entity in some member countries (e.g., Rwanda and Uganda). Bank staff members assert that no private agency with sufficient capacity or credibility to take on the role of promoter and marketer exists in most countries. Bank staff interview, January 10, 2006. 13 the direction of giving ATIA a more visible and dynamic regional presence. How well the new approach works in practice remains to be seen. REALIZED DISTRIBUTION OF COSTS AND BENEFITS 3.9 Benefits foreseen at the initiation of the RTFP included: a) poverty alleviation through sustainable, private sector­led growth; b) increased employment and production resulting in poverty reduction; and c) increased exports and generation of foreign exchange. The beneficiary population would presumably consist in large measure of the working poor, including the self-employed and small business owners. Those enterprises for which policies have been issued so far tend to be large parastatal and private domestic and foreign enterprises. 3.10 Benefits across countries were intended to be proportional to the IDA credits to each member country. But as Table 3.2 shows, disbursements of IDA credits for the original seven member countries as security against ATIA policies have not correlated closely with each other. Disbursements from policies issued in Malawi, Rwanda, and Uganda have been low relative to IDA financing. But disbursements are higher relative to IDA financing in for Burundi, Kenya, Tanzania, and Zambia. Table 3.2 IDA Disbursements as Security on ATIA-Issued Policies, Seven Founding ATI Countries (US $ million)* Country IDA Financing Disbursements as of December 31, 2004 Burundi $15.00** $7.50 Kenya $25.00 $12.50 Malawi $15.00 $3.75 Rwanda $7.50 $1.88 Tanzania $15.00 $7.50 Uganda $20.00 $5.00 Zambia $15.00 $7.50 TOTAL $112.50 $45.63 *IDA financing includes supplemental credit of May 2005 of US $7.5 million to Burundi. **Includes additional IDA credit of $7.5 million in 2005. RISK TO OUTCOMES AND IMPACTS 3.11 It is not possible at this point to predict the extent to which outcomes and impacts of the program are likely to be sustainable because outcomes and impacts have yet to be realized. 3.12 ATIA is closer to self-sustaining status than it has been, but its net premium or underwriting income of $177,330 during the first nine months of 2005 was less than half 14 that for the same period in 2004, $392,754.40 More transactions are said to be in the pipeline than previously, but these have yet to be bound into insurance policies. The current restructuring process that would permit issuing commercial risk insurance as well as pooling capital, could facilitate a substantial increase in policy sales, with corresponding results for net income (as discussed in more detail in paragraphs 4.15­ 4.17). 40Ernst and Young, Ibid. 15 4. Efficiency: Governance, Management, and Financing 4.1 Summary. Despite active stakeholder participation during the design and implementation stages of the RTFP and ATIA (reflected in meetings of the ATIA Board of Directors and the General Assembly), ATIA was not effective during its initial phase. There are three reasons for this, in addition to the previously discussed design flaws: a) delays in filling top management positions; b) poor performance by the first CEO; and c) inadequate marketing of ATIA's product. These impediments combined with the limitations in ATIA's product line have put the agency two years behind its original target for reaching operational self-sufficiency. Efforts so far to attract capital contributions for the implementation phase from other donors, such as the African Development Bank (AfDB) and the EU, have been unsuccessful. But lenders, such as the AfDB, have reportedly recently reacted positively to ATIA's restructuring plans. EFFICIENT USE OF RESOURCES 4.2 The RTFP is not realizing anticipated benefits through the use of reasonable levels of time and money as was expected at the design stage. Bank management staff point out that the RTFP is an inherently risky and innovative project, and that delays are to be expected. 4.3 IDA credits to ATIA of $10 million are now double the original planned amount. IDA credits to member countries, including new members Democratic Republic of Congo and Madagascar, now stand at $123.4 million, compared with the first combined amount of $105 million. The additional $7.5 million credit to Burundi is a response to increased demand. At some point, insurance-generated income will be needed to augment reserves to cover expansion of insurance as well as ATIA operating expenses. 4.4 As shown in Table 3.1, net cash premiums generated by ATIA policy sales were US $400,000 in FY 2004 and $270,000 in FY 2005. With a total ATIA staff size of 19 these figures would imply a premium generation of $21,053 per staff member in FY 2004 and $14,211 per staff member in FY 2005. Although relevant comparisons have not been made, industry-wide averages are likely to be much higher. The average ATIA annual salary is undoubtedly significantly higher than average premium generation per staff member. 4.5 According to Bank staff, the main reasons for the weakness to date in ATIA performance, in addition to the limited demand for its product (as discussed in paragraphs 2.13­2.14) are as follows: · Startup delays. Failure to anticipate countries' inability to fulfill conditions of IDA credit effectiveness and problems in establishing insurance facilities with Lloyds of London have resulted in late disbursements of underwriting capital.41 41Minutes of the 17th meeting of the ATIA Board of Directors. 16 · Delays in hiring key senior staff or poor decisions by Bank and ATIA Board and management. The first CEO was appointed in August 2001, and the General Counsel was appointed in September. But the CEO turned out to be a poor manager and spokesman for ATIA, and on his departure in 2004, ATIA had no CEO for more than a year, until February 2006. The process of grooming a CUO from four underwriters hired in March 2002 was protracted, and a CUO was not appointed until September 2005. · Inadequate marketing. In view of the newness of the concept of political risk insurance in the region, ATIA needed a more aggressive marketing effort than management deployed in the early years. But it is unclear how much this would have mattered given the limited demand for ATIA's initially limited product. GOVERNANCE, MANAGEMENT, AND LEGITIMACY 4.6 ATIA is an international institution governed by an agreement signed by the initial seven member countries in January 2001. The governance and management arrangements for ATIA are detailed in Annex B. 4.7 The purpose of the agency is to facilitate, encourage, and develop the provision of, or the support for, insurance, including coinsurance and reinsurance, guarantees, and other financial instruments and services for purposes of trade, investments, and other productive activities in Africa to supplement those that may be offered by the private sector, or in cooperation with the private sector.42 4.8 The main governing bodies of ATIA are the General Assembly and the Board of Directors. The General Assembly meets annually and is composed of the Ministers of Finance of the ATIA member countries. The ATIA Board, which meets quarterly, is composed of 40 percent public and 60 percent private members from participating countries. COMESA is also represented on the Board. ATIA is located in Nairobi, Kenya, and currently consists of 19 staff. 4.9 The General Assembly members and Board of Directors have become increasingly vocal and have told ATIA it must perform up to expectations or close down.43 Although the Bank is not an ATIA Board member, it is invited to Board meetings and communicates frequently with directors and managers. Working relationships among Board members and with the Bank have strengthened over the years. 4.10 The governing bodies of ATIA took major steps when, at the Special General Assembly of December 9, 2005, they decided that the necessary measures should be taken to expand ATIA's product mix to commercial risk insurance and to restructure ATIA to permit conversion of IDA credits into pooled equity capital.44 The latter measure 42From the ATIA Web site, http://www.ati-aca.com/faqs.asp. 43Interview with MIGA staff member, January 26, 2006. 44ATIA, Special General Assembly, op. cit., Resolution No. 4. 17 will permit mobilizing additional capital from the private sector and the ability to reinsure. Both these actions are critical to the sustainability of ATIA.45 4.11 Senior management has been a major problem for ATIA. Whereas the first CEO had well-honed underwriting skills, he was new to Africa and his general management experience was thin, thus his performance over a three-year period was poor in key aspects and his separation was a difficult process. After a gap of more than a year, in February 2006, ATIA hired a new CEO, a former senior MIGA officer. The CUO, another key senior management position, was filled for the first time in September 2005. 4.12 ATIA management argues that having a separate Development Credit Agreement for each member country slows implementation and lowers performance. The same argument is made by Bank staff (see the next section). They also say that slow disbursement is in large part the result of the Bank procurement rule that no more than 25 percent of the total cost of a credit or loan can be disbursed at one time. They argue that such standard rules regarding loans to governments should not be applied to regional programs, in particular to programs operating through the private sector, such as ATIA.46 Another factor that reduced efficiency and delayed efficacy was the mechanism for completing transactions. It did not work well in practice and should have been revised and made simpler and more accessible to all players in the private sector market. The lack of ready access by all players and the complexities of documentation deterred potential private sector users.47 Furthermore, the current ATIA management team finds the original operational costs of ATIA to have been too high for a startup operation and that staffing and other expenses should have been lower to begin with, and only increased as revenues increased commensurately. 4.13 The RTFP derived from the interests of member countries, donors, the private sector, and other stakeholders. There is no evidence to suggest that one client has been unfairly favored over others. Active participation by stakeholders has helped to ensure that their voices have been represented. From the beginning, member countries were committed to ensure the success of ATIA. Technical advisory committees composed of both private and public sector participants prepared key aspects of project design, including the agreement to establish a regional agency, the operations manual, and the proposed arbitration and claims resolution mechanism. 4.14 There is evidence in project documents to show that countries, donor partners, and beneficiaries exercise effective voice in the various aspects of the regional program as they jointly explored different alternatives for the ATIA's legal setup. Collectively, they agreed to create a multilateral institution that was a) credible, both from a perspective of legal structure and in the sense of being free from historical or political baggage; b) an agency of professional standards, dedicated to implementing the RTFP; and c) open to 45Interview with MIGA staff member, January 26, 2006. 46Meeting with ATIA management, Nairobi, June 30, 2005. 47RTFP team member. An additional factor in deterring use of the facility may have been the role of one London brokerage firm in processing applications. While the contribution of the firm in other respects was positive, the prominent position of this broker may have deterred other brokers from utilizing the facility. 18 participation by all African countries, without creating sensitivities between the various regional and subregional organizations in the region.48 FINANCING 4.15 In the view of Bank staff members, the main finance problem for RTFP has been the Bank Board's imposed restriction on the use of IDA funds for commercial risk insurance, which has contributed to low demand for ATIA products and consequently hobbled ATIA's ability to foster private-led growth. The Africa Region's Regional Operations Committee meeting, in February 2005, recommended that explicit Board approval be sought for the use of IDA funds for commercial risk insurance. But the provision of an IDA credit to an organization such as ATIA did not become a significant issue in the design process. The legitimacy of providing a credit to a nonstate organization resides in there being national agreements in which countries agree to membership in the organization and find it creditworthy. Such agreements were signed by the seven founding member countries before the credit to ATIA became effective. 4.16 Given the participating countries' resistance to a pooling of their IDA credits, ATIA was unable to shift funds from countries experiencing low demand for political insurance to others with higher demand. After ATIA's start-up and persuasion by its management and Bank staff, member countries agreed to a pooling of credits at the December 2005 General Assembly meeting. This change will provide greater flexibility and efficiency in the view of Bank staff and ATIA, because funds can now be moved from low-demand to high-demand countries with a minimum of procedural hurdles. 49 4.17 An additional $900,000 has been raised through the contributions of $100,000 each by the nine member countries (not including Liberia) to the capital of ATIA. In addition, COMESA has made a capital contribution, as have the private sector organizations Atradius, PTA Bank, and PTA Re. DONOR PERFORMANCE 4.18 Grants totaling $2 million from the EU and Japan supported technical assistance during the design phase of the RTFP. Exploratory efforts by Bank staff to obtain support from other donors during the implementation phase have not yielded fruit. Staff members are more hopeful that recent discussions with the African Development Bank will yield positive results. ATIA management argues that for regional programs, it is better to have a multidonor approach. The validity of this argument assumes that sufficient variation among donor procedures exists to permit greater flexibility for ATIA without introducing undue transactions costs. 48PAD, p. 15. 49Meeting with ATIA management, Nairobi, June 30, 2005. 19 5. Monitoring and Evaluation 5.1 There have been no independent or self-evaluations of the RTFP, but there have been relatively frequent supervision missions, averaging twice a year. The results of the Regional Operations Committee meeting of February 2005, and the Quality Enhancement Review meeting in March, were intended to inform the Mid-Term Review meeting. 5.2 The Africa Region Quality Enhancement Review meeting of March 14, 2005, recommended that relevant outcome measures employed by MIGA also be considered in place of existing measures. The MIGA measures include: a) the gross face value of guarantees issued; and b) the total value of the investment supported. The latter measure would come closer to indicating impact on the private sector.50 50Other MIGA measures include the number of projects, number of small and medium-size enterprise projects, and South-South projects. 20 6. World Bank Performance 6.1 Summary. The Bank played an instrumental role in shaping and getting the project off the ground. It exploited its comparative advantage by effectively drawing on its expertise and experience in another region to design and prepare the RTFP. It utilized its convening power and senior leadership to move the project along at a critical juncture during project preparation. But the project team overestimated regional demand for political risk insurance and was involved in a top ATIA personnel decision that turned out to be flawed. The Bank subsequently provided thin support to ATIA during the critical, early years of implementation when ATIA should have paid more attention to outreach and marketing. The Bank also provided inadequate oversight over the first CEO, whose working relationships with country counterparts and Bank Country Directors tended to be counterproductive. Staff members contend that ATIA performance would have been stronger if a Bank staff member had been seconded to ATIA during the first year or two of implementation to provide more support for staff recruitment, training, marketing, and private sector orientation efforts in member countries. Recently, the Bank has worked actively with the ATIA Board to secure new top leadership and help plan and facilitate ATIA's restructuring. COMPARATIVE ADVANTAGE 6.2 The Bank had a comparative advantage, which it effectively exploited for the RTFP. It brought expert knowledge about trade insurance guarantee projects it had supported in Albania, Bosnia and Herzegovina, and Moldova. Its convening power helped to mobilize financial support from potential member countries but it did not mobilize capital financial support in the implementation phase from other donors (beyond the technical assistance support in the design phase from the EU and PHRD (Government of Japan ­ GOJ)). QUALITY OF SUPPORT AND OVERSIGHT 6.3 Notwithstanding dedicated project teams, the quality of the Bank's support to the RTFP in the form of strategic direction and technical performance has been modest to poor. The Bank project team's projection of market demand for political risk cover insurance was overly optimistic. The Bank provided thin support to ATIA during the critical early implementation years when more attention should have been paid to outreach and marketing. The Task Manager changed less than nine months after the project became effective. But close and effective working relationships were established between the Bank's project team and ATIA management. Working together, the Bank team and ATIA management played a critical role in facilitating restructuring and, in particular, convincing the ATIA Board of the merits of pooling capital.51 51Resolution 10 of the Resolutions Adopted by the Special General Assembly on December 9, 2005, commends the efforts of the World Bank team "to turn around the performance of the Agency and the Project." (Nairobi: 2005, pp. 6­7). 21 6.4 The Bank also should have provided closer oversight over the first CEO, whose working relationships with counterparts, including Bank Country Directors, tended to be counterproductive. 6.5 No independent evaluations have been conducted of the RTFP. Oversight mechanisms in the Africa Region of the Bank have included a February 2005 Regional Operations Committee review, chaired by the Director of Strategy and Operations in the Vice President's Office, and the March 2005 Quality Enhancement Review, which included relevant Country Directors and the Acting Executive Director of ATIA. The reports of these two meetings were relatively critical about the challenges facing ATIA and the changes required to improve performance. 6.6 Regular supervision missions have been conducted. The supervision missions conducted in October and November 2005 was particularly critical of slow implementation progress and of the basic premise of the need for political risk insurance. 6.7 The Bank's Quality Assurance Group (QAG) conducted a Quality-at-Entry Review of the RTFP in 2001. The QAG panel came to conclusions that in several key respects anticipated the conclusions of this review: · Sound institutional arrangements. The QAG panel found the institutional arrangements for creating ATIA to be sound and its handling of environmental issues to be "best practice." · Strong demand for political risk cover from foreign banks and suppliers. · Weak demand from domestic companies. The QAG panel saw domestic company demand as "the foundation of the project's purpose ... and the basis of financial projections." · The project's direct contribution to increasing exports is therefore likely to be modest, weakening its linkage with reducing poverty and increasing economic growth. · Absence of a clearly articulated strategy for addressing country specific policy and institutional reforms. The QAG panel expressed concern that absence of such a strategy, which was necessary to expand exports and private sector development, would jeopardize the broader poverty and growth objectives. · At the same time the panel recognized that reforms would need to be addressed by IDA on a country-by-country basis outside the project, and that "on balance the operation will achieve the stated development objectives and will be sustainable in the longer-term."52 STRUCTURES AND INCENTIVES 6.8 Both Bank project staff and ATIA staff members feel that other Bank staff in the Africa Region, in Washington, and in the field had a poor understanding of RTFP. Two 52World Bank, Quality Assurance Review, Quality at Entry Assessment IV, Investment Lending Questionnaire, "Regional Trade Facilitation Project." June 4, 2001. 22 possible reasons for this are two novel characteristics of the project: a) while governments stand behind the IDA credits, the project is primarily oriented and driven toward the private sector; and b) it is a multicountry project. Project staff members feel constrained by the fact that, apart from the $10 million to ATIA, the balance of the original IDA credits of $123.4 million had to be parceled out to each member country rather than being given to ATIA or pooled and managed by ATIA. In the view of project staff, the latter arrangement would have been more flexible and efficient. A credit of the total amount to ATIA would have reduced by more than 80 percent the documentation and reporting requirements for Bank staff. But as noted previously, member countries had until recently resisted any kind of pooling arrangement. Only at the December 2005 annual General Assembly meeting and Board meeting did countries agree that a pooling of capital made sense and approved exploration of this approach. 6.9 Bank and ATIA staff members point to other factors as impediments to effective project implementation. ATIA staff members point to a problem with the Bank's rule limiting any given disbursement to no more than 25 percent of total project cost. They believe this rule unnecessarily impedes the efficiency of their operation.53 Both Bank and ATIA staffs note that important Bank meetings on ATIA, such as the February 2005 Regional Operations Committee and March 2005 Quality Enhancement Review, have been poorly attended by middle and some senior management.54 Bank staff members also believe that ATIA performance would have been stronger if a Bank staff member had been seconded to ATIA during the first year or two of implementation, which would have provided more support for staff recruitment and training, and to marketing and private sector orientation to ATIA's work in member countries. LINKAGES TO OTHER BANK COUNTRY OPERATIONS 6.10 Linkages between the RTFP and Bank country operations have been weak. In general terms, there is consistency between the private sector focus of ATIA and the significant priority given to the private sector in each country CAS. Recent CASs for six of the seven ATIA member countries make at least some reference to ATIA or RTFP, but these are for the most part cryptic or indirect (e.g., a reference to COMESA in the 2004 Kenya CAS). The exception is the 2002 Rwanda CAS, which contains three references to the RTFP, one of them relatively extended, albeit descriptive.55 6.11 Project team staff members report that Bank country teams are largely ignorant of ATIA. Exceptions are a couple Country Directors who have been supportive, in contrast to one Director who was hostile to ATIA. The latter resulted at least in part from the poor communications skills of the first ATIA CEO. ATIA senior staff members observe that "World Bank [country] offices are not involved and should be." They also complain that 53Meeting with ATIA management, Nairobi, June 30, 2005. 54Only one Country Director attended one of the meetings. 55Rwanda Country Assistance Strategy, Volume 1, 2002, paragraph 84. The most recent full CAS for Burundi was issued in 1995 and therefore makes no reference to ATIA or RTFP. 23 they have little or no contact below the level of Country Director in the field and Manager or Task Manager in Washington.56 DISENGAGEMENT STRATEGY 6.12 The life of project set forth in the PAD is 10 years, from 2001 to 2011. Despite the lagging performance of ATIA, Bank and ATIA management still expect that the agency will be fully self-sufficient by 2011, and will, along with member countries, be able to service the IDA credits. The PAD also assumed that by 2011, ATIA would have become an entirely private sector entity and no country capital would continue to be invested. That assumption no longer holds. Rather, Bank staff now expect that continued active involvement of member governments in ATIA, including majority capital control (at least 51 percent), is essential into the indefinite future in order to maintain a development orientation. 6.13 Should the Bank have set an intermediate exit strategy at the outset of the project? The midpoint of project life, 2006, might have been designated as an intermediate exit point. The project staffs argue against this, noting that an experimental program such as ATIA requires a long germination period. They also point to the lags in appointments of the CEO and CUO, and to the time required for approval of the new product and capital restructuring. With the restructuring plan now approved by the countries and ATIA's Board and as well as the Bank's Board, the issue is, when is it fair to assess ATIA's performance as a restructured organization? March 31, 2009, would be an appropriate date. This would be halfway between the approval of project restructuring in December 2006 and the project completion date of June 30, 2011. 56Meeting with ATIA management, Nairobi, June 30, 2005. 24 7. Country Participation 7.1 Interest and capacity in ATIA have seemed to vary among member countries. For example, during the project preparation period, COMESA members Kenya and Uganda were early and strong supporters of the RTFP. Uganda helped organize and lead a delegation that met with the Bank's Africa Region Vice President, who in turn persuaded reluctant and skeptical midlevel managers to support the project. Kenya supported the hosting of ATIA headquarters in Nairobi. By contrast, Burundi has relatively low institutional capacity overall but strong interest from its government and Chamber of Commerce. Burundi has recently entered a period of fragile peace after a prolonged period of conflict. Political risk insurance is seen as vital, as it was in postwar Bosnia and Herzegovina. Rwanda has also been wracked by severe conflict, but has experienced a longer period of relative peace than Burundi. Rwanda probably has as much or even stronger institutional capacity than Burundi, but has apparently lacked interest in ATIA. Bank staff members attribute the much greater demand for political risk coverage in Burundi than in Rwanda to these considerations. Countries participate actively in ATIA Board of Directors meetings and General Assemblies. Each member country contributes US $100,000 to ATIA capital. 7.2 Program coordination within countries was problematic during 2001­2003, when the country ATIA liaison offices were public sector entities, such as offices in trade ministries. By and large they were poor promoters of ATIA among the private sectors and therefore did not contribute to program effectiveness. In 2004, several, but not all of these offices, were replaced by agents or brokers from the private sector who had an incentive to market ATIA services. 25 8. Conclusions SUMMARY OF FINDINGS 8.1 The RTFP became effective at a time when there was a significant decline in demand for political risk insurance worldwide, and particularly in the Africa region. Yet in extensive interviews with representatives of the private sector trade credit insurance industry, the Bank project team reported very positive reaction to the concept of countries insuring the maintenance of market-friendly policies. Moreover, as noted in Chapter 2, potential demand appeared to be considerably greater for comprehensive political and commercial risk insurance. But the firm engaged to provide commercial risk insurance decided not to cover firms located in Africa, choosing instead to do business only with those outside. Until the current restructuring, ATIA had been restricted from directly offering a comprehensive package of both covers. 57 8.2 Although ATIA human, institutional, and organizational capacities have been largely established, especially in Nairobi, its performance against its principal objective of improving access to financing for intra- and extraregional trade has fallen considerably short of the original projection that premium income growth would have made ATIA self-sufficient by 2003. There has been active participation by key stakeholders in the project during the design and implementation stages. But poor ATIA strategic and senior staffing decisions and delays have contributed to low efficiency and efficacy. As a result, ATIA has only recently reached operational self-sufficiency (two years behind its original target) and, because this is due largely to interest earnings on reserves and one- time cost reductions, its long-term self-sufficiency remains an open issue. 8.3 The Bank played an instrumental role in getting the RTFP off the ground. It took full advantage of its knowledge of similar operations elsewhere and utilized its expertise, convening power, and senior leadership to keep things moving at a critical juncture during project preparation. Yet the survey instrument employed by Bank staff members substantially overestimated regional demand for political risk insurance. The Bank has recently been actively engaged with ATIA's management and Board to design a restructuring that will enable ATIA to issue commercial risk insurance and provide it with capital on the basis of pooling member countries' IDA credits. It remains to be seen whether these major corrective measures will make the program of real value to countries and ensure its financial sustainability. 57In addition, as noted in Chapter 4, the mechanism for completing transactions worked poorly in practice and should have been made simpler and more accessible to all players in the private sector market. Another factor that has militated against ATIA achieving its targets is the departure of the first chief executive of ATIA. The project team believes the delay in appointing a successor set ATIA back at a time when its pipeline was showing signs of expanding and developing. The successor has industry credibility, but time has been lost. See footnote 2 in the executive summary for information on the restructuring decisions taken. 26 IMPLICATIONS FOR EFFECTIVE SUPPORT 8.4 Five main implications for effective support of regional programs emerge from RFTP experience in launching and supporting ATIA: · Building the capacity of a new regional institution requires continuous support from the Bank: o during the early implementation phase as well as the design phase, and o from individual country offices as well as from the regional office, o but the Bank should not get involved in decisions that in the interests of ownership should be made solely by the new institution, such as hiring. · An existing regional or subregional institution, such as COMESA, can have important legitimizing and policy-making roles in laying the foundation for a new subregional program (e.g., helping to frame and sponsor the initial international agreement establishing ATIA). o It can serve as an interlocutor between member national governments and new subregional structures. o It can also serve as an "incubator" by sponsoring and assisting the development of a new institution. o ATIA needs the capacity and mandate to ensure that sound analytical work underpins its initiatives or, in the absence of its own capacity, mobilizes external expertise to undertake needed work. · Efforts are needed up front to determine with countries the appropriate financing structures for regional projects. Where project efficiency may call for an approach that countries are not willing to accept at the outset, such as pooling of country resources, some process and tentative timetable for evolving a mutually acceptable approach should be considered at the outset. · As in regional programs involving public institutions, private sector­ oriented programs such as the RTFP require strong linkages between regional and national institutions. By design, 60 percent of the ATIA Board is from the private sector, yet until recently, ATIA's use of private sector entities to promote the program as agents and brokers has been uneven. The effectiveness with which the private sectors of ATIA member countries will become actively engaged in the program and linked to the Board of Directors remains to be seen. · A solid empirical basis is required to estimate demand for a new regional trade financing program. A survey designed on the basis of expressed need is likely to have an upward bias. 27 Annex A: Background Information on the Regional Program 1. Program (or project) number P063683 2. Program Dates August 2001­present 3. Approval Dates First credit of US $5 million to ATIA (34880): 04/03/2001 Second credit of US $5 million to ATIA (34481): 05/31/2005 4. Effectiveness Dates First credits to ATIA: 08/23/2001 and to Burundi, Kenya, Malawi, Rwanda, Tanzania, Uganda, and Zambia: 03/2002 Second credits to ATIA and Burundi: 2/15/2006 and first credits to Democratic Republic of Congo and Madagascar: 2/15/2006 5. Closing Dates Original closing date for first credits to ATIA and member countries: 06/30/2004 Revised closing date for first credits: 06/30/2006 Planned closing date for second credits: 06/30/2011 6. Sectoral or thematic areas Finance (noncompulsory pensions, insurance and contractual savings) 7. Regional or subregional Regional project­subregional (Eastern and Southern Africa): Burundi, Democratic Republic of Congo, Kenya, Madagascar, Malawi, Rwanda, Tanzania, Uganda, and Zambia. Liberia was accepted as a member in December 2005 8. Regional partnership or Regional project project Does it comprise country Separate IDA credits to nine countries but one project title. projects? If yes, do the country projects: · Conform to a template Yes · Address the same Yes problem(s) · Regularly interact with Yes--e-mails, meetings, newsletter each other (e.g., through information sharing, research, monitoring and evaluation)? 9. Rationale for the regional program · Strategic focus: Regional Integration (PAD, p. 2) regional commons, Economies of Scale (PAD, p. 13) transboundary problems, regional integration, or other cooperative actions? · Intended direct impact Export development and competitiveness, regional integration, (regional/subregional trade facilitation, and market access and/or participating countries) 28 Annex B: Governance and Management Arrangements Management Agency for Project Implementation 1. Management arrangements for project implementation Regional Coordinating Body 2. Name, location, and internet African Trade Insurance Agency address of the regional P.O. Box 10620, 00100-GPO, Nairobi, Kenya implementing Tel: +254-20-2719727/2726999. agencies/coordinating body for the Fax +254-20-2719701 project Safaricom: +254 722 205007 Kencell: +254 733 625511 E-mail: Info@africa-ECA.com 3. Current size Ten member countries 4. Current membership Public and private sectors in all countries 5. Membership criteria Agreement with terms of agreement establishing African Trade Insurance Agency 6. Membership responsibilities 7. Functions of the coordinating body Promoting, selling, and administering ATIA insurance policies 8. Meeting frequency Board meets four times a year 9. Name, location, and internet P.O. Box 10620, 00100-GPO, Nairobi, Kenya address of other implementing agencies/coordinating body for the project 29 Annex C: Goals, Objectives, Outcomes, Outputs, and Activities Goals/Mission Program Objectives Intended Outcomes Outputs Inputs/Activities To alleviate poverty To improve access to Export development Output Indicators though private sector­ financing for productive competitiveness; · Volume of policies led growth in transactions, and cross- regional integration; issued per country participating countries border trade. trade facilitation and · ATIA becomes self- by improving access market access; financing after two years to financing for international of operation transactions and financial · On average one claim cross-border trade. architecture or less resulting from a government performance risk per country, per year · ATIA enters into a leveraging partnership with private risk insurers for political risk cover Components 1) Component Creation of regional political risk insurance facilities (capital provided by IDA, an additional IDA credit would be extended to finance the operating costs of ATIA for the first two years of operation (p. 28, PAD). The facility will cover productive activity involving cross-border trade. Types of political/noncommercial coverage (see p. 29, PAD). a) Inability to convert and/or transfer currency, b) Imposition of exchange controls, c) Cancellation of licenses and restrictions on import and export, d) Expropriation, e) Seizure of goods, prevention of export, f) Interference with the carriage of goods, g) War or civil disturbance, h) Embargo. 2) Component ATIA's Operating and Capital Costs An IDA credit to ATIA will cover the costs for the startup period of ATIA's operations. Participating countries will share ATIA's operating and capital costs .... For more details see p. 35, PAD. 30 Annex D: Persons Consulted African Trade Insurance Agency, management and staff, Nairobi, Kenya, group interview, June 30, 2005. Said Al Habsy, LEG, interviewed September 2, 2005. Sherri Archondo, current Task Manager for RTFP, January 5, 2005 to present, interviewed September 1, 2005. Peter Jones, MIGA, January 29, 2006. Onno Ruhl, first Task Manager for RTFP, July 2001 to December 2001, interviewed September 8­9, 2005, and January 10, 2006. Ivan Rossignol, second Task Manager for RTFP, January 2002­December 2004, interviewed September 14, 2003. 31 Annex E: References African Trade Insurance Agency. Web site, http://www.ati-aca.com/faqs.asp. Nairobi. African Trade Insurance Agency. 2005. Special General Assembly of the African Trade Insurance Agency, Resolutions Adopted by the Special General Assembly. Resolution 4, 8. Nairobi. Ernst and Young. 2005. African Trade Insurance Agency: Interim Management Letter. Nairobi. World Bank. 2005a. Regional Trade Facilitation Project. Supervision Aide-Memoire. Washington, DC. _____. 2005b. Support for Regional Programs: A Desk Review of the OECS Solid Waste Management Project. Washington, DC. _____. 2004a. Kenya Country Assistance Strategy. Washington, DC _____. 2004b. Salient Features of Trade Performance in Eastern and Southern Africa. Africa Region Working Paper Series No. 76. p. 6. Washington, DC. _____. 2004c. Zambia Country Assistance Strategy. Washington, DC. _____. 2003a. IDA, Strategic Framework for IDA's Assistance to AFRICA (SFIA): The Emerging Partnership Mode, IDA/SecM2003-0406, p. 9. Washington, DC. _____. 2003b. Malawi Country Assistance Strategy. Washington, DC _____. 2003c. World Development Report 2003, pp. 240­241. Washington, DC. _____. 2002. Rwanda Country Assistance Strategy, Volume 1. Washington, DC. _____. 2001a. Project Appraisal Document. Washington, DC. _____. 2001b. Quality Assurance Review, Quality at Entry Assessment IV, Investment Lending Questionnaire, "Regional Trade Facilitation Project." Washington, DC. _____. 2000a. Africa Regional Trade Facilitation Project, Project Initiation Document, Report 7384, August 1. Washington, DC. _____. 2000b. Can Africa Claim the 21st Century. Washington, DC. _____. 2000c. Tanzania Country Assistance Strategy, Volume 1. Washington, DC. _____. 2000d. Uganda Country Assistance Strategy, Volume 1. Washington, DC 32 _____ .1997. Zimbabwe Country Assistance Strategy, Volume 1. Washington, DC. ______. 1995. Burundi Country Assistance Strategy, Volume 1. Washington, DC Other References: Internal Project Documents (Memoranda and Aide Memoires) wb308751 O:\CORP\REGIONAL PROGRAMS EVALUATION\Final Background Papers\RTFP Desk Review_COMPLETED.edits tracked.doc 03/07/2007 2:22:00 PM 33