ANNUAL REPORT ON PORTFOLIO PERFORMANCE FISCAL YEAR 2002 DECEMBER 12,2002 (VOLUME I - MAIN REPORT) ACRONYMSANDABBREVIATIONS AAA Analytical and Advisory Activities LIL Learning and Innovation Loan ACS Activity Completion Summary M&E Monitoring and Evaluation AFR Africa Region MDG Millennium Development Goals AIDS Acquired Immune Deficiency Syndrome MIC Middle Income Countries ARPP Annual Report on Portfolio Performance MNA Middle East and North Africa Region ARDE Annual Rpt. on Development Effectiveness MP Montreal Protocol APL Adaptable Program Loan MTEP Medium-Term Expenditure Program BW Business Warehouse OED Operations Evaluation Department CAE Country Assistance Evaluation OPCS Operations Policy and Country Services CAS Country Assistance Strategy OPR Operational Performance and Results CDF Comprehensive Development Framework PAD Project Appraisal Document CEM Country Economic Memorandum PER Public Expenditure Review CFAA Country Fin. Accountability Assessment PIP Portfolio Improvement Program CODE Committee on Development Effectiveness PRSC Poverty Reduction Support Credit CPAR Country Procurement Assessment Review PREM Pov. Reduction and Econ. Mgmt. Network CPIA Country Policy and Institutional Assessment PSD Private Sector Development CPPR Country Portfolio Performance Review PSI Private Sector Devt & Infrast Network CRM Corporate Resource Management Group PSAL Programmatic Structural Adjustment Loan DEC Development Economics PSR Project Status Report DO Development Objective PTI Poverty Targeted Interventions EAP East Asia and Pacific Region QEA Quality at Entry Assessment ECA Europe and Central Asia Region QAG Quality Assurance Group EFT Education Fast Track QER Quality Enhancement Program ERL Emergency Recovery Loan QSA Quality of Supervision Assessment ESSD Env. and Socially Sustainable Dev. Network RVP Regional Vice President ESW Economic and Sector Work SAD Sectoral Adjustment Loan FIL Financial Intermediary Loan SAL Structural Adjustment Loan FSE Finance Network SAR South Asia Region GEF Global Environment Facility SB Sector Board HD Human Development SIL Specific Investment Loan HNP Health Nutrition and Population SSP Sector Strategy Paper IAD Internal Auditing Department SWAp Sector Wide Approach ICR Implementation Completion Report TAL Technical Assistance Loan IP Implementation Progress TF Trust Fund IRIS Integrated Records and Information System TTL Task Team Leader JSA Joint Staff Assessment WBI World Bank Institute LCR Latin America and the Caribbean Region WDR World Development Report LICUS Low Income Countries Under Stress wss Water Supply & Sanitation ANNUAL REPORT ON PORTFOLIO PERFORMANCE FISCAL YEAR 2002 CONTENTS PAGE No. EXECUTIVE SUMMARY ............................................................................................................. 1 REPORT I. Introduction.. ............................................................................................................................ 1 II. Portfolio Size and Composition ............................................................................................... 2 III. Portfolio Performance .............................................................................................................. 13 IV. Economic and Sector Work ..................................................................................................... 23 V. Millennium Goals and Results.. ............................................................................................... 28 VI. Recommendations.. .................................................................................................................. 35 TEXT TABLES Table 2.1 FY02 Portfolio by Lending Instrument.. .......................................................................... 6 Table 2.2 Portfolio by Sector of Focus and Theme ......................................................................... 10 Table 3.1 Trends in Portfolio Indicators.. ........................................................................................ 13 Table 3.2 Riskiness by Region and Network.. ................................................................................. 17 Table 3.3 Riskiness by Sector and Theme.. ..................................................................................... 18 Table 3.4 Quality Indicators ............................................................................................................ 18 FIGURES Figure 2.1 Trends in Portfolio Size and Resource Transfer .......................................................... 2 Figure 2.2 Portfolio Distribution by Region, FY97 and FY02.. ...................................................... 8 Figure 2.3 Portfolio Distribution by Network, FY97 and FY02 ..................................................... 11 Figure 3.1 Trends in Portfolio Status.. ............................................................................................. 14 BOXES Box 2.1 Portfolio Dynamics.. ........................................................................................................ 3 Box 2.2 Was there a September 11 Portfolio Effect?. ................................................................... 5 Box 2.3 What is Programmatic Lending? ..................................................................................... 7 Box 2.4 Sectoral and Thematic Codes.. ......................................................................................... 9 Box 2.5 New Approaches to Lending ........................................................................................... 12 Box 3.1 Fine-Tuning Portfolio Management Indicators ............................................................... 16 Box 3.2 Riskiness in Crisis Adjustment Lending.. ........................................................................ 17 Box 3.3 Status of FYOI ARPP Recommendations.. ...................................................................... 21 Box 3.4 FY03 Portfolio Improvement Program ........................................................... 2 1 Box 3.5 Elements of Best Practice ................................................................................. 22 Box 4.1 A Reformed Approach to ESW-AAA Assessment.. ........................................................ 25 Box 5.1 Strengthening Data for Policy-Making.. .......................................................................... 29 Box 5.2 MDGs - A Pilot Country Assessment.. ........................................................................... 30 Box 5.3 New Style CPPR: EAP and AFR.. ................................................................................... 32 Box 5.4 MDGs and the CAS ............................................................................................. 33 ANNEXES Annex 1: Guarantees.. ....................................................................................................................... 37 Annex 2: The Portfolio - An Overview Table.. ................................................................................ 39 Annex 3: Basic Portfolio Definitions.. .............................................................................................. 40 Annex 4: Millennium Development Goals .......................................................................... 42 Annex 5: FY03 PIP Countries and PIP Projects ............................................................ 43 Volume II : STATISTICAL APPENDIX EXECUTIVE SUMMARY 1. The Annual Review of Portfolio Performance (ARPP) provides the Board and Senior Management with a strategic overview of the effectiveness of the Bank's portfolio in delivering results to its clients. It assessesthe portfolio's present status, examines issues and opportunities, and recommends measures to maintain portfolio quality, including enhancements to its measurement. As a special theme, this ARPP explores approachesto making the Millennium Development Goals (MDGs) operational in the context of the Bank's corporate commitment to reinforce attention to Results. 2. Portfolio Size and Composition. The portfolio of active Bank projects ($102 billion), has decreasedby 19 percent since FY96. A slight gain for IDA is more than offset by a 25 percent reduction in the IBRD component. This decreasewas entirely in investment lending, in particular for the energy sector. Although investment loans still dominate the portfolio, their continuing decline has focused attention on the need to modernize the procedures and business practices that guide the delivery of key lending products. Portfolio composition shows a modest trend towards more diversity, with some shrinkage in the existing largest regions and networks (notably EAP and PSI). The new thematic coding shows that Finance & PSD and Environment & Natural Resources are the two largest themes, as they were in FY97. But there has been some growth in the share for the Social and Governance themes. Adjustment lending has grown, mainly driven by financial crises. It accounted for 40 percent of new lending in last five years. Much of this is fast-disbursing and hence not counted in the end-year portfolio. The old divide between investment and adjustment lending is starting to fade; programmatic instruments (still under 10 percent of the portfolio) - and reformed approachesto investment - can both contribute to a portfolio which is more strategic and country-driven. 3. Portfolio Performance. After five years of steady improvements, several portfolio risk indicators for FY02 returned to levels comparable to FYOO. A similar pattern is shown by key quality indicators (projects-at-entry, supervision). The detailed analysis from these latter exercises points to a vulnerability arising from the limited experience of many new frontline staff and gaps in the support available from a stretched sector management. This is a cautionary signal, not a cause for alarm. After a period of sustained improvement, this result is not unexpected, and needs to be seen against the pre-FY96 context when about 30 percent of the portfolio was at-risk and operational quality was less than satisfactory for one in four projects. Two complementary messagesemerge: first, the need to ensure the recent quality gains do not slip away; second,to deepenand broaden the internalization of a quality culture, shifting the focus to a higher level, that is to Results and the overall effectiveness of country programs. In the portfolio context specifically, thesetrends triggered a review of the current methodology for measuring riskiness, as well as the appropriate balance between calculated risk-taking in program design and implementation and judicious exercise of the Bank's fiduciary responsibilities. 4. The ARPP presents a fine-tuning of the approach to measuring riskiness. This responds in part to a conclusion that these key indicators of quality may have modestly overestimated portfolio improvements in the last four to five years. It will also help reinforce Annual Report on Portfolio Performance. FY02 ii Management's efforts to encourage more candid reporting, on which there appearsto have been some progress. The ratings for Realism and Net Disconnect indicate that portfolio management can improve further. The modified system will strengthen the technical and predictive quality of leading indicators as an instrument for portfolio oversight, whilst bringing them into closer alignment with OED outcome data. In policy terms the overall conclusion is that portfolio riskiness continues to be at realistic levels, in the range 15-20 percent. This corresponds to an 80-85 percent Sati;fuctory rating in OED outcome terms, which leaves appropriate space for innovation and calculated risk-taking in the increasingly complex environment bearing on the Bank's portfolio. 5. Economic and Sector Work. ESW volume grew again in FY02, driven by increases in the number of policy notes and informal products. There is an increased coverage of the key Core Diagnostics products, but this program may need fine-tuning to better respond to country priorities. ESW is now moving towards parity with lending in terms of levels of resources and management attention within the Bank. Following the efforts of the past few years, ESW quality has now largely caught up with that of projects. There has been increased use of participatory approaches; this mode of implementation shows significant quality gains. A new approach to assessingESW-AAA will be built around the quality of its integration into the CAS process. 6. Millennium Goals and Results. The Bank's focus on Poverty Reduction has now been enhancedby the commitment to support the Millennium Development Goals (MDGs). The new development paradigm, as recently endorsed by the Development Committee, is a results-based approach, built around MDG+ goals. The year saw a surge of activity across the Bank exploring the implications of operationalizing the MDGs. This will also involve engaging client-countries (including support to them in improving data systems), updating of country programs and strengthened partnerships. The next stage, now underway, is mainstreaming in Operations. Over a maybe 4-5 years transition, these new perspectives should be built into most individual CASs and PRSPs, including localization of the targets. A new generation of results-basedCASs will serve as a key vehicle for institutionalizing this shift in perspectives. These will also provide for a new focus in developing Bank operations and structuring the portfolio. Quality assurance and evaluation will also need to embrace these same, more strategic, results-oriented approaches. Inter alia, it is proposed that the ARPP itself be broadened to cover Results. All this will be a demanding agenda that will need adequateresourcing as well as a rebalancing of incentives to staff. Annual Report on Portfolio Performance. FY02 111 KEYRECOMMENDATIONS o Upcluting Business Processesand Instrunzellts to be more client-responsive, providing for a more integrated, country-driven approach to the portfolio and an enhanced approachto both adjustment and investment lending. o Improving the balance between Risk alzd Innovution, inter alia by reconfirming that portfolio riskiness levels in the range 15-20 percent are broadly appropriate in terms of providing incentives for innovation and calculated risk-taking. q Strengthening staff und management cupucities, notably by enhancing the skills and work environment in ways that will permit staff and managers to function more effectively in frontline operations. o P&zing the Measurement Frontier by extending assessmentsto a more strategic and aggregate level, while ensuring adequatecoverage at task level. Country and CAS will increasingly become the primary unit of account. o Deepening the Agendu by ensuring that the MDGs, are well-integrated within the framework of the new rest&s-based approach, recognizing the importance of country ownership. q Reporting on Results should be more systemically incorporated into the portfolio, including expanding M & E at a more aggregatelevel. As part of corporate reporting, future ARPPs should cover new res&s-based approachesand indicators. Annual Report on Portfolio Performance, FY02 I I. INTRODUCTION OBJECTIVES AND APPROACH 1.1 The Annual Report on Portfolio Performance (ARPP) provides the Board and Senior Management with a strategic perspective on the Bank's portfolio-the primary vehicle for delivering results to our clients. The main objectives, as defined in the Approach Paper reviewed by CODE, are as follows: l Assess the current status of the Bank's portfolio, and review management efforts and results in relation to the FYOl ARPP recommendations and the Bank's strategic priorities. l Assess likely trends and challenges to the portfolio over the medium-term. l Examine two special topics: the management of risk in the portfolio, and the progress the Bank is making in measuring the poverty orientation in the portfolio and in incorporating the Millennium Development Goals into Bank operations. l Identify priority actions needed to broaden/deepen the gains in portfolio quality during the past few years. 1.2 The FY02 ARPP draws primarily from materials that are prepared as part of regular portfolio monitoring functions carried out by the Regions and Networks, supplemented by project/portfolio data in the Bank's management information systems. Consistent with past ARPPs, the report uses a five-year time frame (FY97-02) to examine medium-term trends in the portfolio. In preparing the ARPP, extensive consultations were held with managers and staff from around the Bank. An array of self-evaluative processes developed in the Bank over the past few years has also helped make the ARPP process more analytical and forward- looking. STRUCTURE AND COVERAGE 1.3 The report is organized into five subsequentChapters. Chapter II reviews the recent trends in size and composition of the lending portfolio. Chapter III reviews overall portfolio performance as well as revised approachesto riskiness measurement. Chapter IV reviews the Economic and Sector Work (ESW) portfolio. Chapter V examines the progress in incorporating the Millennium Development Goals into Bank operations, the poverty- orientation of the portfolio and approachesto incorporate a results-basedapproach into Bank operations. Recommendations are summarized in Chapter VI. A second volume contains a detailed set of supporting statistical material. Annual Report on Portfolio Performance, FYO2 2 II. PORTFOLIO SIZE AND COMPOSITION The size of the Bunk project portfolio has decreused sign$icuntly since FY97, with the reduction centred on the IBRD component. Portfolio structure has been broadly stable; although, on a commitment basis, there is a trend towards a portfolio that is more diversified by region and sector. Regionully, EAP still has the largest share of commitments but also the largest proportionate decline; AFR still accounts for the largest number of projects in the portfolio. By Network, there has beena marked sh$?away from PSI (still the largest), whilst PREM shows the largest increase by number. The new product coding shows FinanciaVPSD and Environment/Natural Resourcesas the two largest themes. By lending instrument there has been a relative sh$ towards adjustment, but investment loans still dominate the portfolio. Programmatic approaches muy start to blur the dividing line. Portfolio size will be sign$xntly influenced by how well the Bank makes thut shift, us well us the matching trunslation into new guidelines, processesand delivery capabilities. PORTFOLIO SIZE 2.1 The Bank's lending portfolio (IBRD/IDA plus TFs) for FY02' consisted of 1,542 operations representing commitments of $102 billion.2 The latter figure represents a 4 percent drop in nominal terms from the previous year. However, the decline in the portfolio has beenabout 19 percent' from a peak of $126 billion in FY96. Over the last five years,the dominant factor has been a drop of just over 25 percent in IBRD net commitments, with the IDA component increasing modestly (Annex 2 provides an overview of the portfolio). In resource transfer terms, both new approvals and disbursements show a major peak in FY98- 99, with approvals declining sharply in FYOO then recovering in FYOl and 02. Disbursements show a smoother trend following the peak due to crisis adjustment lending (Fig. 2.1). FIGURE 2.1: TRENDS IN PORTFOLIO SIZE AND RESOUKCE TRANSFER (FY95 = 100) I 160: 140- 120 j \ too! 1995 1996 1997 1998 1999 2000 2001 2002 Total Approvals Disbursements Actiw Portfolio 1I I Portfolio as used in this report refers to the end-fiscal year situation 2 The portfolio also includes 16 Guarantees, totaling $1.6 billion still in force under the Guarantee Program. Their performance is reviewed separately in Annex 1. 3 Twenty-one percent in real terms. Annual Report on Portfolio Performance, FY02 3 Annual Report on Portfolio Performance, FY02 4 2.2 The pattern of entries and exits defines the portfolio. Roughly one-in-six projects are new to the portfolio each year. New entrants by numbers peakedin the crisis years of 1998/9 and are now in FY02 about 20 percent lower (254 projects, including 25 trust fund approvals). Exits, mainly projects approved in the mid 90s have hovered around 280 for the last three years. It should be noted that the formal portfolio definition doesnot capture many important fast disbursing loans: principally single-tranche adjustment operations which typically open and close within the same fiscal year. There were sixteen such operations in FY02 totaling $3.1 billion (Box 2.1). 2.3 The most striking change is the decline in the IBRD portfolio. Net commitments have fallen by a quarter since FY97, roughly equally from reduced numbers and smaller individual commitments. These reductions were all in investment lending. By Network, the reduction is overwhelmingly from PSI. At $65.5 billion in FY02, IBRD commitments are at their lowest level in more than a decade (but still almost twice those for IDA). The volume of new IBRD lending in FY02 rose to $11.5 billion from $10.5 billion in FYOl, substantially influenced by a few large adjustment loans, particularly for Turkey. However, these recent levels of new approvals are just over half the peak in the crisis FY98-99 period. For the third consecutive year, IBRD exits exceedednew approvals. 2.4 IDA net commitments in contrast have beenessentially stable; for FY02 the portfolio showed a small increase to $35 billion. New IDA lending rose for a third year to $8.1 billion for FY02. Contributing to this increase was $2.4 billion in adjustment lending. The FY02 increase was substantially influenced by growth in AFR, including a second $500 million Multi-country HIV/AIDS Program (Box 2.5). 2.5 The portfolio of trust funds and other operations, continues to grow with new approvals of $0.4 billion in FY02. There were 115 operations in this portion of the portfolio, with net commitments reaching $2.0 billion in FY02. The Montreal Protocol portfolio now stands at $800 million, now more comparable to the GEF portfolio in commitment terms. The GEE; portfolio had $1.01 billion in commitments, split 65:35 between free-standing projects and blended or partially blended projects. Special Financing programs, which are funded out of the Bank's net income, are now a significant component of the portfolio in terms of number of operations. 2.6 Looking back at the recent trend in portfolio size and the crisis-linked volatility in new approvals, it is clear that there continues to be a high level of uncertainty in making lending projections. For FY03-05 planning purposes, corporate and regional median projections assume an essentially flat profile to FY05. In a larger context, it might indeed be useful for OPCS/CRM to explore how to improve corporate level forecasting. This would be useful for Bank wide planning, budgeting and staffing. As Fig. 2.1 illustrates, the recent historical trend has been a slow reduction to the present $100 billion level with the decline having been in the IBRD portfolio. However, a new crisis, a recovery in demand by IBRD clients in investment borrowing and/or a sustained growth of programmatic lending could push it upwards. Annual Report on Portfolio Performance, FY02 5 Box 2.2: WAS THERE A SEPTEMBER 11 PORTFOLIO EFFECT? September11affectedtheportfolio in theshortterm,but overall, theBank copedwell. Therewas a modestslowdown in new lendingin the 2-3 monthsimmediately following. This reflectedthe generaldisruptionof work-patterns,mostnotablydelaysin travel critical to key stages- appraisal, negotiations,etc. - in project developmentas well as supervision. After that slowdown, as reflected in a smaller than usual sharein new approvalsin those months, there was a slow recovery. By end-yearthelevel of new approvalswasactuallyrunning abovethetrendrateof the previous threeyears. The decline in some quality indicatorsmay reflect some residualeffects from delaysin supervision.The genera1weaknessin theglobal economythat followed September 1lth is still affectingthe Bank portfolio asnew circumstances,suchaslower exports,declinesin commodity prices,weakenedtravel andtourism, andthe like, affect the detailedstructureof new lending, as well as leading to some project restructuring. The growth in emergency-linked adjustmentlending certainly has part of its roots in the post-Sept II" generalizedeconomic weakness.The major programto supporttheAfghanistanrecoverywas a very specificcollateral outcome. INSTRUMENTS 2.7 The Bank's lending portfolio is traditionally divided into either investment or adjustment, with the former very much dominant. Looking forward, the dividing line is becoming less clear, as programmatic approachesstart to emerge. The shift to-date has been relatively slow. Year-to-year fluctuations are much higher for new adjustment lending, but there seems to have been a slow secular increase, with its shareof the portfolio rising about four percent points since FY97. This increase was concentrated in IBRD borrowers. However, in portfolio terms, investment operations still accounts for 964 percent of projects and 89 percent of commitments in FY02. Straddling these two groupings, programmatic lending (Box 2.3) is well under 10percent of the portfolio by commitments. 2.8 ,Specifc investment loa~zs(SILs) are the most widely used instrument, representing about 72 percent of total commitments. The APL, now just over five percent of the portfolio and 18 percent of FY02 new investment lending, is steadily gaining in popularity. It has largely displaced use of the less flexible Sector Investment and Maintenance (SIM) loan. The LZL, the APL's twin as a new instrument, has had a more difficult start. Its use has fallen by almost half in the last two years as its early promise of low-cost, speedy delivery, design flexibility and experimentation often failed to materialize. Two other investment instruments have significant profiles: in a world of endemic crises, use of the Emergency Recovery Loan has grown steadily, whilst the commitment value of Technicul Assistance Loans (TALs) has declined about 15 percent since FY97, while remaining a significant vehicle for skills and knowledge transfers. 2.9 Adjustment lending has accountedfor over 40 percent of total IDA/IBRD new lending volume over the last five years. A major driver here was its active but inevitably volatile role5 as a responseto financial and other crises. However, becauseadjustment operations are 4 This portfolio basis of comparison exaggerates the importance of investment loans due to their longer lifespan. A comparison in resource transfer terms is better made using cumulative net disbursements. 5 A secondary role is that of a resource transfer mechanism to support long-term policy or institutional reform. This is the programmatic mode of adjustment lending (e.g., PRSC). Annual Report on Portfolio Performance, FY02 6 relatively fast-disbursing, in formal active portfolio terms they comprise just four percent of operations and 11 percent of commitments at end-FYO2. Formally including seven different types of operations, this area is dominated by Structural (SAL) and Sectoral (SAD) adjustment loans. The important new PRSC instrument is a modest one percent of FYOl-02 lending. The sharp increase in adjustment lending in FY02 was dominated by Turkey, with substantial operations also in Brazil, Argentina and Pakistan. Investment No. $B Adjustment No. $B SIL Specific Investment Loan 1,017 73.6 SAL Structural Adjustment Lending 36 5.8 APL Adaptable Program Loan 128 5.5 SAD Sector Adjustment 20 3.9 TAL Technical Assistance Loan 122 2.4 PSL Programatic Structural Adj Loan 3 1.8 LIL Learning and Innovation Loan 96 0.4 PRC Poverty Reduction Support Credit 2 0.3 SIM Sectoral Investment and Maintenance 52 4.0 EC0 Expanded Cofinancing Operation P 0.0 ERL Emergency Recovery Loan 45 3.3 FIL Financial Intermediary Loan a j.J 1,481 90.6 61 11.7 TOTAL INVESTMENT + ADJUSTMENT 1,542 102.3 2.10 The optimal balance between investment and adjustment lending for any country is heavily influenced by macro-performance characteristics, such as fiscal and balance of payments situations, absorptive capacity and governance, as consolidated in the CAS process. However, the supply side is also a relevant factor in practice: the approach of many staff is still influenced by their often greater experience with investment lending which historically has defined many Bank practices. Borrowers are also sometimes not well-versed on the choices of instruments available. Another complicating factor is the high staff turnover in recent years. Over time the choices are seen to have become more complex as new instruments, responding to client needs,have been addedto existing ones. Table 2.1 is a partial indication of that diversity. Both the Bank and its clients are now seeking to gain more value from this array of lending instruments. This is a particularly pressing challenge when at the same time the programmatic approach, often framed within a CDF/PRSP, is demanding greater cohesion across all lending, indeed also non-lending, activities and a multi-sectoral team approach. New instruments such as the PRSC and APL, as well as Sector-Wide Approaches (SWAP) are custom-designed to fit this new style, but may still need further adaptation to meet some important client needs, for example for budgetary support. 2.11 Work is now underway on proposals for Management on revised Bank operational policies for adjustment lending as well as a complementary re-examination of the effectiveness of present Bank investment lending. These two reform exercises are running in parallel, but the ultimate solutions will be integrated and complementary. The goal should be a system where the choice of instrument is driven by country needs and development effectiveness, not by preferences internal to the Bank. On the investment lending side, the main challenge is how best to rationalize the accumulation of several decadesof procedures, processesand business practices, which can sometimes weaken the Bank's competitiveness. An early improvement may come through revisions to the rules on eligible expenditures. Procedural simplifications are another avenue being examined. As was spelt out in the Cost of Doirzg Business report, Bank clients increasingly practice a competitive approach in Annual Report on Portfolio Performance. FY02 7 selecting their lending partners. They want transparency, a set of rules and practices in project design and implementation that is more consistent amongst donors and less burdensome. Some of the solutions under consideration include: simplifying project-specific fiduciary and safeguard requirements while helping clients strengthen their national standards/capabilities; strengthening task manager skills as well as providing them scope for more discretionary judgments; and harmonizing donor policies and procedures. 2.12 In the adjustment domain, there are other distinct concerns'. An important tension in the past had been around the perceived array of performance conditions - in terms of their number, focus, complexity and the demandsthey can place on borrowers. The challengelooking forward is to build upon emerging good practices,including environmental, social and fiduciary considerations,supportedby substantive,results-basedM & E. As notedin ChapterV this needs a close integration of ESW-AAA activities and lending within a unified portfolio plan. As the Bank moves forward with Poverty Reduction and now the MDGs as key policy drivers, it will be critical to addressthese issues as part of the ongoing operational policy update, in a manner that is also consistent with country ownership. REGIONS AND COUNTRIES 2.13 The FY02 portfolio includes operations in close to 130 countries, with a heavy concentration in just five countries (China, India, Brazil, Turkey, Mexico, in order) which account for 45 percent of net commitments. In comparison, the 50 smallest borrowers account for only 6 percent. The concentration is less intense by numbers of projects; here, the top ten representroughly 30 percent of the portfolio. ' From Adj~lstrner~t Lem@g to Development Policy Support Lerzdirlg.OPCS. Annual Report on Portfolio Performance, FY02 8 2.14 Portfolio data inevitably lags, as well as smoothes, the trend in new lending. There was an overall decline in commitment terms concentrated in EAP, ECA and SAR; within these regions there were particularly large declines in several key countries: China, Indonesia, Russia, Hungary, Ukraine. Possible factors that account for this include: transition to IBRD status, portfolio management concerns, and country exposure. AFR has the largest number of projects in the portfolio among the six regions, accounting for 24 percent of the active portfolio in FY02; its shareby value is significantly lower at 15 percent (Figure 2.2). After several years of contraction, the AFR portfolio has begun to stabilize and the outlook for future growth - as reflected in more active IDA lending - is promising. The AFR share of IDA lending is now just short of 50 per cent. The number of projects in the ECA portfolio, after a period of growth centered on its IDA countries, is now stabilizing; new lending grew in FY02 but mainly due to specific factors, especially adjustment lending to Turkey. EAP continues to have the largest shareof commitments in the active portfolio, but both number andcommitment valueshave fallen in the last year or so, in part due to substantial restructuring/ streamlining of key country portfolios (Indonesia, Philippines) and lending exposure limits to China. The MNA portfolio, which is moving towards a knowledge- sharing orientation, remains by a large margin the smallest in commitment terms; it continued to decline albeit slowly in FY02. FIGURE 2.2: PORTFOLIO DISTRIBUTION BY REGION, FY97 AND FY02 By commitments (%) AFR WP ECA LCR MNA SAR 2.15 Portfolio composition has consciously shifted over the last five years away from countries with weak policy and institutional environments7; in particular the share of the portfolio in low CPIA countries declined from 15 percent in FY97 to 7 percent in FY02 on a commitment basis. The share for strong performers grew more than fourfold to 18 percent. The LICUS Report has reminded us that caution on the project portfolio front in those countries may need to be balanced by an expansion in activities to support institutional development and capacity-building. Not unexpectedly the portfolio for such LICUS countries is very modest. 7 As measured by the Country Policy and Institutional Assessment (CPIA). Low CPIA countries are those with a rating of 3 or less on a scale of 1to 6. Annual Report on Portfolio Performance, FY02 9 NETWORKS, SECTORSANDTHEMES 2.16 For the first time this section reports on the new outcome-orientated data from the recoding of the Bank's portfolio completed in July, 2002. The traditional database- still the most reliable for historic comparisons - is a coding based on the structure of Bank administrative units, Networks and Sector Boards. The new coding structure has been retrofitted to the existing portfolio, although not (yet) fully integrated into the SAPIBW system. Box 2.4 provides some details on this major new database. 2.17 Because the coding framework is so new, and because the data was retrofitted, conclusions need to be treated with caution. The portfolio composition by rectory of focus for period FY97 and FY02 shows there are no radical changes on a commitment basis but rather a steady shift over the five years out of Energy and into Transportation (the largest individual sector), Law and Public Administration, and Health and Social Services. The sharesof other sectors remained fairly stable. The shareof Health and Social Services grew fastest over the period (45 percent), and Energy declined the most (37 percent). Using this new data to isolate cross-sectoral projects, defined as those reporting activities in at least three sectors each absorbing at least 20 percent of the total project budget, these remain a modest 13 percent by value of all projects. Annual Report on Portfolio Performance. FY02 10 TABLE~.~: P~RTF~LIOBYSECTOROFFOCUS AND THEME (Commitments, % Shares) Sector FY97 FY02 Theme FY97 FY02 Agriculture 10 10 Econ. Mgmt. 2 I Law & Public Admin. 13 15 Public Sector/ Gov. 7 8 ICT 2 2 Law 2 2 Education 9 9 Finance & PSD 25 19 Finance 5 5 Trade 3 3 Health & Social 9 13 Social Protection 5 6 Industry & Trade 6 5 Social Development 5 8 Energy 19 12 Human Dev. 10 11 Transportation 17 19 Urban 12 13 Water & Sanitation 10 10 Rural 13 14 Environ & Nat. Res. 16 15 2.18 Reviewed against the eleven themes, portfolio composition shows a similar slow but steady shift to a stronger social/governance orientation over the FY97-FY02 period. Finance and PSD dropped the most (25 to 19 percent), with the largest declines within that theme occurring in Private Sector Infrastructure Services. The largest increase was in the Social Development theme as its share rose from 5 to 8 percent, with the bulk of the gains concentrated in one sub-theme, Civil Engagement, Participation and Community Development. Other themes showed only marginal shifts. 2.19 As noted, whether the pace of change in the past is reviewed by economic sector or theme, the result - a measured pace - does not change. To see what the implications are for moving towards the Bank's strategic goals, projected commitments were reviewed under the FY03-07 lending pipeline. These results should be treatedas indicative; they representrather incomplete planning intentions which can be subject to large changes, particularly in the outer years. By economic sector, there is continuing growth in the share of Law and Public Administration, and now in Finance; there are indicators of substantial declines in Energy and Mining, Transport, and Industry and Trade. By theme, most shifts are modest; the only substantial movements indicated are for Economic Management and Public Sector Governance. 2.20 Using traditional definitions, by Network, the Private Sector and Infrastructure (PSI) portfolio of 543 projects ($44 billion) remains the largest with a 43 percent portfolio share, but it has shrunk substantially over the past five years, confirming the trend shown in the new coding by econonzicsector of impact. No single Network grew to fill the gap but the largest increaseis in PREM. This portfolio has nearly doubled since FY97, but still representsonly eight percent of the portfolio.* 8 Including the fast-disbursing adjustment operations that is not recorded as `portfolio' would raise the PREM share, but only by about 2 percentage points. Annual Report on Portfolio Performance, FY02 11 I By corm-&rents (%) ESSD FSE HDN FREM PSI 2.21 Within the PSI Network, the largest decline (more than 50 percent) was in the Energy and Mining sector. Almost all the portfolio reduction was in IBRD countries. Explanatory factors are the greater availability of private funds to this sector, the perceived burden of Bank safeguards and fiduciary requirements, especially for larger IBRD borrowers with market access, and a de-emphasis by the Bank on lending to these sectors. The Human Development Network (HDN) portfolio is now the second largest. Its size has been roughly constant since FY97, with a small decline for Education and an increase for Social Protection. New lending in FY02, despite some positive pointers for the future, was still well below peak levels of FY98 and FY99. In FY02 there was an upturn for Education, HNP was flat and Social Protection declined sharply. Looking to the future the MDGs and new global health issue funds should create new incentives for HD lending. The newly launched Education Fast Track Initiative is another important innovation. 2.22 Many specific innovations are occurring in lending approaches. Two important examples, both supportive of a more programmatic approach to key developmental challenges, are outlined in Box 2.5. Annual Report on Portfolio Performance, FY02 2.23 Concluding comments: The portfolio is perhaps at a turning point in structural terms. The old divisions of investment and adjustment are becoming less meaningful in the context of greater integration in programming, more country-driven agendasand the growing focus on aggregateResults. The middle zone of a more programmatic approach is likely to expand in future. Portfolio size will be significantly influenced by how well the Bank makes that adjustment, as well as the matching translation into new guidelines, processes and delivery capabilities. Annual Report on Portfolio Performance, FY02 13 III. PORTFOLIO PERFORMANCE Following ,five years qf improvements, several por