ANNUAL REPORT ON PORTFOLIO PERFORMANCE Fiscal 1994 Annual Report on Portfolio Per formance Fiscal 1994 The World Bank Washington, D.C. @ 1995 The International Bank for Reconstruction and Development/ rHE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing September 1995 This report is a study by the World Bank's staff, and the judgments made herein do not necessarily reflect the views of the Board of Executive Directors or of the government they represent. ISBN 0-8213-3446-8 Annual Report on Portfolio Performance Fiscal 1994 The World Bank Annual Report on Portfolio Performance (ARPP) for fiscal 1994 is the third in a series of annual reports prepared in accordance with the recommendations of the Portfolio Management Task Force report, "Effective Implementation: Key to Development Impact," issued November 3, 1992.' The main objective of the fiscal 1994 ARPP is to assess the development effectiveness of the Bank's portfolio. To this end, the report provides an assessment of portfolio performance, utilizing an enhanced rating system that emphasizes the achievement of development objectives. The report also reviews progress in portfolio management and provides special analyses on specific topics requested by the Board of Executive Directors, such as project restructuring and resettlement. Portfolio Size and Composition Portfolio Size The Bank's portfolio of operations under implementation in fiscal 1994 comprised 1,762 operations in 134 country portfolios (table 1). Between 1992 and 1994 the overall portfolio has fluctuated within 0.5 percent of that level. The nominal commitment value remained virtually constant at $141.8 billion, but the real value declined by 1.9 percent, to $136.6 billion. At the end of fiscal 1994, roughly half the commitments had not yet been disbursed. The commitment value represented 40 percent of the total cost of all operations in the portfolio ($355.1 billion). Countries in nonaccrual status at the end of the fiscal year accounted for 26 operations and $0.5 billion in commitments. 1. The World Bank consists of the International Bank for Reconstruction and Development (IBRD) and the International Development Association (IDA). The term operatlons portfolio refers to Bank operations under implementation and is used interchangeably with portfolio in this report. The terms Regions and Regional refer to the Bank's Regional Offices. Dollar amounts are current U.S. dollars. 2 TABLE I BANK PORTFOLIO IN NOMINAL AND REAL TERMS, FY92-FY94 Lending Commitnent Value Average Loan Size Number of Projects FY92 FY93 FY94 FY92 FY93 FY94 FY92 FY93 FY94 Nominal Terms Adjustment /a 135 126 117 22,088.5 22,376.3 17,298.4 163.6 177.6 147.8 Investment 1,620 1,644 1,645 113,808.6 119,169.4 124,513.7 70.3 72.5 75.7 Total 1,755 1,770 1,762 135,897.1 141,545.7 141,812.1 77.4 80.0 80.5 Real Terms /b Adjustment /a 22,088.5 22,002.3 16,661.9 163.6 174.6 142.4 Investment 113,808.6 117,177.4 119,932.3 70.3 71.3 72.9 Total 135,897.1 139,179.6 136,594.2 77.4 78.6 77.5 a. Adjustment includes both structural and sectoral adjustment and debt reduction loans. b. Deflated to FY92 prices using the following commitment deflators: FY92, 100- FY93, 101.70; FY94, 103.82. Portfolio Composition In terms of regional distribution of the portfolio (table 2), Africa still has the largest number of projects (34 percent), but East Asia and Latin America have the largest value of commitments (24 and 22 percent, respectively). The share of South Asia continued to decline in number and commitments (to 14 percent of projects and 19 percent of commitments). Europe and Central Asia continued its rapid expansion, but from a low base, while the share of the Middle East and North Africa declined slightly. The portfolio is concentrated in a small number of countries. The five largest country portfolios in terms of commitment value in fiscal 1994 (India, China, Mexico, Indonesia, and Brazil) accounted for 22 percent of all projects and for 42 percent of total commitments. Large country portfolios, defined as portfolios over $1.0 billion or with twenty or more projects, accounted for 62 percent of the number of projects and 83 percent of total commitment value in fiscal 1994, 33 countries were in this category. TABLE 2 BANK PORTFOLIO DISTRIBUTION BY REGION, FY92-FY94 FY92 FY93 FY94 Commitmert Commiment commionent M jgo Value % No Value % N % Value oL & Rexio P-i. Total WSW To-a fr4 Totl~ f T-1a Pro Total WSW) Total Africa 606 34.5 22,304.1 16.4 604 34.1 21,970.7 15.5 590 33.5 22,235.6 15.7 East Asia 300 17.1 28,390.3 20.9 298 16.8 30,624.2 21.6 300 17.0 33,538.1 23.6 South Asia 266 15.2 30,335.8 22.3 255 14.4 28,517.9 20.1 242 13.7 26,570.1 18.7 Europe & Central Asia 106 6.0 13,271.6 9.8 122 6.9 15,611.2 11.0 150 8.5 16,919.8 11.9 Middle East & North Africa 169 9.6 10,124.9 7.5 168 9.5 11,298.2 8.0 154 8.7 10,722.6 7.6 Latin America 308 17.5 31,470.4 23.2 323 18.2 33,523.4 23.7 1 326 18.5 31,825.8 22.4 Bankwide Total: 1,755 100.0 135,897.1 100.0 1,770 100.0 141,545.6 100.0 11,762 100.0 141,812.0 100.0 3 TABLE 3 PORTFOLIO DISTRIBUTION BY SECTOR, FY92-FY94 No. ofPrje. commiment Value (7SM FY92 FY93 FY94 FY92 FY93 FY94 No. Total No. Total No. Total Amount Total Amount Total Amount Total Agnculture 474 270 442 25.0 431 245 28.504 1 21 0 27,4527 194 27.855.0 196 Education 170 97 184 104 187 106 9,3907 69 10,655 5 7 5 11,8646 84 Energy 52 30 53 30 59 33 5.2192 38 5.6112 40 6,219 44 Environment I ) 1 4 02 12 0 7 18 0 () 84 3 0 1 832 1 0 6 Financial 115 b 6 106 6.1 98 5 6 14,059 6 1013 12.231 0) 8 6 10.232 1 7 2 Industrial/lIDF 77 44 65 3 7 55 3 1 7,038 9 5 2 6,339 I 4 5 5,731 8 410 Mining & Other Extractive 16 0)9 16 0)9 IS 09 1.071 2 08 1,213 4 09 1.020 1 u 7 Multi Sector 90 5 1 93 5 3 91 52 8.9993 66 11,128 3 79 8.665 5 6 1 Pop Health & Nutrition 111 63 126 7 1 135 77 5,1194 3 8 6,441 6 46 7.1109 S 0 Power 150 8 5 149 84 133 75 20.3524 15 0 19,936 6 14 1 18.776 1 13 2 Public Sector Management 65 3.7 82 46 90 5.1 3.0648 2 3 3,533 0 2 5 3,388 6 24 Social I 0 1 I 0 1 6 0 3 28 0 010 28 0 0 0 178.6 I I Telecommunications 31 1 8 33 1 9 35 20 2.310 2 1 7 2.508 5 I 8 2.8149 2 0 Transport 200 11 4 207 11 7 204 II 6 16,373 1 12 0 18,405 3 130 20.1748 142 Urban 110 6 3 113 64 114 6 5 7.432 8 5 5 8,2674 5 8 8.636 8 6 1 Water Supply & Sanitation 92 5.2 96 5.4 97 5 5 6,915.5 5 1 7,7097 54 8,3184 5 9 Bankwide Total: 1,755 100.0 1,770 100.0 1,762 100.0 135,897.1 100.0 141,545.6 100.0 141,812.0 100.0 a The Social sector category comprises social safety net and social investment fund operations b The Environnient category includes only free-standing environmental operations There is a much larger number of operatons with sub%tantal environmental objectives, components. and dimensions Only minor shifts have taken place in the composition of the portfolio by sector (table 3). In terms of numbers of projects, agriculture remains the largest sector (24.5 percent of projects), followed by transport (11.6 percent), education (10.6 percent), population, health, and nutrition, or PHN (7.7 percent), and power (7.5 percent). The number of agriculture and power projects continued to decline, while transport projects remained at about the same level and education and PHN projects continued to increase. In terms of commitment level, agriculture is again the largest sector (19.6 percent of commitments), followed by transport (14.2 percent), and power (13.2 percent). For all other sectors, the share of total commitments is below 10 percent. The commitment level of agriculture remained stable, while transport continued to increase rapidly and power to decline. Commitment levels for education and PHN also continued to increase, and their combined total is now about the same as that for power. Portfolio Performance Rating Methodology For this review of the portfolio as of the end of fiscal 1994, the Bank's rating methodology was substantially modified in order to shift the focus of the rating process to a more realistic and transparent assessment of the development effectiveness of Bank operations. As noted below, this resulted in some downgrading of project ratings and made trend analysis more difficult. More projects would be expected to receive unsatisfactory ratings than under the previous system, but in practice it is not possible to disentangle this effect from that of other factors affecting performance ratings. 4 The traditional methodology was strongly oriented toward physical and financial inputs, although a rating on the likelihood of achieving development objectives (DO) was assigned, it was subsumed under the rating for overall implementation status (OS). The revision included the following features: * A dual rating system with separate summary ratings for development objectives and implementation progress (IP). * Replacement of the OS rating by the new IP rating based exclusively on progress in input delivery and component implementation relative to the original staff appraisal report implementation plan or a formally revised plan. * An explanation for the DO ratings giving the specific reasons for judgments regarding expected development results. Portfolio Performance Measures The basic measure of performance used in this report is the percentage of problem projects-the proportion of projects in the portfolio rated unsatisfactory or highly unsatisfactory. Reflecting the dual rating scale described above, the percentage of problem projects was calculated with respect to development objectives and implementation progress. Because the emphasis of this report is on the development effectiveness of the Bank's portfolio, the report uses the percentage of problem projects with respect to development objectives as the key measure of portfolio performance. The percentage of problem projects with respect to implementation progress is used as a secondary measure. Bankwide and Regional Portfolios Performance with Respect to Development Objectives. The percentage of problem projects, in terms of development objectives, in the total Bankwide portfolio increased from 13.9 percent in fiscal 1993 to 15.2 percent in fiscal 1994 (see table 4).' The regional exceptions to this Bankwide trend are South Asia and, to a lesser extent, Africa. The sharpest decline in performance occurred in the Middle East and North Africa region, which has the highest percentage of problem projects (followed by Europe and Central Asia and by Africa). East Asia has by far the lowest percentage of problem projects. If countries that are in nonaccrual status or are suffering from civil strife are excluded from the portfolio, the performance picture is substantially different. As shown in table 4, the DO-based proportion of problem projects when these problematic country portfolios are netted out is only I 1.3 percent (slightly lower than in 1993 when a similar adjustment is made). Both the 2 Based on unweighted averages of the performance of individual projects. See the discussion below of project performance weighted by loan size. (that is, in terms of commitment at risk). 5 TABLE 4 PORTFOLIO PERFORMANCE BY REGION, FY92-FY94 PT92 FY93 FY94 No of % Problem Prol. .t % Problem Pro. too % Problem Pro, Proj OS D0 Pro, 0S DO Pa IP 1X0 Africa 606 24.6 21.2 604 24.3 19.7 590 20.2 18.7 East Asia 300 3.3 20 298 4.4 2.8 300 8.0 4.7 South Asia 266 18.0 9.1 255 16.1 13.5 242 13 6 10.8 Europe & Central Asia 106 15.1 12.0 122 20.5 20.7 150 26.0 22.2 Middle East & North Africa 169 17.8 9.9 168 17.3 15.4 154 27.9 27.0 Latin America 308 18.2 9.9 323 13.6 10.3 326 199 13.3 Bankwide Total 1,755 17.6 12.5 1,770 16.9 13.9 1,762 18.3 15.2 Excluded Portfolios 100 51.0 43.6 137 56.9 56.5 Bankwide Total (net) 1,670 14.9 11.9 1,625 15.1 11.3 a (ountries that were in nonaccrual status or suffered from civil conflicts in FY93 (Angola, Chad, Congo. Liberia. Rwanda. Somalia, Sudan. Togo and Zaire. Yugoslavia, I laiti) b iountries that were in nonaccrual status or suffered fron civil conflict in FY14 (Angola, Burundi. ljihouti, Rwanda. Somalia, Sudan. and Zaire. Algeria and Yemen. Yugoslavia. Fornier Yugoslav Republic of Macedonia, Rep t bosnia and i ferzegovina, I laiti number of projects in these portfolios and the proportion of problem projects increased significantly between 1993 and 1994. The effects of these portfolios on regional performance are evident in several regions, particularly in Africa and the Middle East and North Africa. When those country portfolios are netted out, the percentage of problem projects is only 12.5 (rather than 18.7) for Africa and 13.8 (rather than 27.0) for the Middle East and North Africa. As table 5 shows, when project performance is weighted by loan size (that is, is analyzed in terms of commitment at risk), the DO-based performance of the fiscal 1994 portfolio is much better (only 9.3 percent of commitments at risk, compared with 15.2 percent problem projects for unweighted project performance) and remains fairly constant (9.6 percent of commitments at risk in 1993 and 9.3 percent in 1994). If the portfolios of countries that are in nonaccrual status or are suffering from civil strife are excluded, overall performance is even better: only 8.3 percent of commitments are at risk, compared with 9.1 percent in 1993 when a similar adjustment is made. The effects of these portfolios on regional performance were again quite significant in Africa and in the Middle East and North Africa. When these portfolios are netted out, the percentage of problem projects is only 13.1 (instead of 15.9) for Africa and 13.4 (instead of 16.6) for the Middle East and North Africa. Consistent with the more pro-active measures taken to address portfolio problems. the proportion of problem projects that have been in that status for three consecutive years or more decreased from 26 to 20 percent between 1993 and 1994. Moreover, of 134 country portfolios in fiscal 1994, only 21 (including the 13 countries in nonaccrual status or experiencing civil conflict) had a very low likelihood of achieving project development objectives, while 72 had a high likelihood of achieving development objectives. As a group, the 33 large country portfolios performed better than the Bankwide average, with 13.6 percent problem projects, the 5 largest country portfolios had 6.5 percent problem projects. 6 TABLE 5 PORTFOLIO PERFORMANCE BY REGION WEIGHTED BY LOAN SIZE, FY92-FY94 FY92 FY93 a/ FY94 f Problem Pro. % Problem Proi. % Problem Pro, Commitment OS ) 'ommilment Ois DO Commitment IP DO Value (USSM - Value (US5A) Value (US$M) Africa 22,304.1 23.0 17.5 21,970.7 23.4 16.6 22,235.6 18.8 15.9 East Asia 28,390.3 2.8 1.0 30,624.2 4.0 0.9 33,538.1 8.3 1.5 South Asia 30,335.8 16.6 7.5 28,517.9 14.7 10.7 26,570.1 10.3 7.2 Europe & Central Asia 13,271.6 19.3 13.2 15,611.2 26.6 19.4 16,919,8 29.1 14.7 Middle East & North Afica 10.124.9 15.5 7.6 11,298.2 13.2 11.9 10.722.6 25.7 16.6 Latin America 31,470.4 16.1 6.9 33,523.4 10.8 6.5 31,825.8 21.7 9.1 Bankwide Total 135,897.1 14.8 8.2 141,545.7 14.0 9.6 141,812.0 17.1 9.3 Excluded Portfolios 2,457.4 42.6 34.3 4,890.4 49.2 35.3 Bankwide Total (net) 139,088.3 13.5 9.1 136,921.6 16.0 8.3 a. Countries that were in nonaccrual status or suffered from civil conflicts in FY93 (Angola, Chad, Congo, Liberia. Rwanda. Somalia. Sudan. Togo and Zaire: Yugoslavia; Haiti) b. Countries that were in nonacemral status or suffered from civil conflict in FY94 (Angola, Burundi, Djibouti, Rwanda. Somalia, Sudan, and Zaire: Algeria and YemenYugoslavia, Former Yugoslav Republic of Macedonia. Rep. of Bosnia and Herzegovina Haiti). Performance with Respect to Implementation Progress. Like the DO ratings, the IP-based performance of the portfolio was also downgraded, except in Africa and South Asia. Tougher rating standards and, possibly, the switch from OS to IP as a criterion may have played some role in the apparent deterioration (particularly since the OS rating included DO ratings that were unduly optimistic). As was shown in table 4, IP-based performance remains somewhat worse than DO- based performance. Two reasons may be that implementation problems are more immediately apparent than failure to achieve a more distant development impact and that such problems may be considered transitory and not necessarily impeding the achievement of development objectives. Factors Underlying Portfolio Performance. The major substantive factors underlying portfolio performance at the country level fall into three categories: political stability, macroeconomic framework and policies, and institutional strength and implementation capacity. Political instability or civil strife has severely damaged portfolio performance in several countries, especially in Africa, but also in the Middle East and North Africa and in Europe and Central Asia. In many other countries the vicissitudes of electoral cycles (leading to diversion of attention, personnel turnover after elections, and other problems) have negatively affected project performance. Macroeconomic conditions set the context, for good or bad, for the implementation of projects. On balance, these conditions were more favorable in 1994, although budgetary constraints continued to lead to severe local funding problems in many countries. Although institutional weaknesses often affect implementation capacity across the board in particular countries, there are frequently significant differences in the institutional capacity of sectoral ministries and agencies within a country. Apart from systematic factors that tend to affect the performance of all projects in a country or sector, there are many factors that only affect particular projects. A review of the 233 problem projects indicates that some of the principal factors explaining probable failure to achieve development objectives are largely under the control of the implementing agency. The most important of these is deficient project management, followed by ineffective use of technical assistance. In low-income countries a shortage of skilled managers and staff seems to be the key constraint, in other countries inadequate project management is often the result of political changes that lead to high rates of personnel turnover. In many cases the recruitment and retention of good 7 management and staff is hampered by inadequate salaries and related civil service deficiencies. Other factors amenable to control by the recipient government also affected a significant proportion of the problem projects-in particular, the lack of adequate counterpart funds. The latter is closely interrelated with inadequate government commitment, which was cited in some projects and which often leads to inadequate sector policies, such as failure of power and water tariffs to reflect costs and insufficient autonomy for public utilities. Factors generally not subject to government control were cited for at least one-fourth of the problem projects, they were largely concentrated in a few countries facing wars or civil disturbances. The review of problem projects pointed to some factors affecting implementation that are under Bank control. These included excessive complexity of project design, unduly optimistic expectations regarding institutional capacity and the effectiveness of technical assistance, and, in some cases, overestimated demand projections based on overoptimistic assumptions. The review also noted that performance is sometimes affected by continuing differences between the Bank and governments. Difficulties associated with government commitment sometimes reflect the role of the Bank (as a development institution) in urging governments to undertake new initiatives, make basic policy and institutional changes, or address neglected priorities, such as operations and maintenance. These measures are often particularly difficult, and government commitment is therefore inherently uncertain. Some project failures appear to be the result of the clash between the Bank as an agent of change and the political and other difficulties governments face in making fundamental changes. Average portfolio performance is also affected by the numbers of new projects that enter the portfolio and of older projects that leave the portfolio during the year. New projects rarely encounter major problems during the first year or two and therefore improve the average performance rating of the portfolio, while projects that are completed or terminated and therefore leave the portfolio include a relatively higher proportion of problem projects. The net effect of entries and exits in 1994 was an improvement in the ratings of the overall portfolio compared with 1993. Portfolio Management During fiscal 1994 the Bank intensified its efforts to promote a strong results orientation. The resources devoted to supervision continued to increase, from an average 16.3 staff weeks per project in 1993 to 17.6 staff weeks in 1994, and all regions now assign high priority to implementation and "results on the ground." "Next Steps," a program of actions intended to reorient the Bank's business practices so as to improve the development impact of Bank-financed activities, is in its second year of implementation. Initial assessments of several of these actions are reported below. Rating Methodology As mentioned above, improvements in the Bank's rating methodology, by shifting the focus of attention toward the achievement of development objectives, have reinforced the emerging "culture of implementation" and enabled the Bank to identify, report, and address more effectively the real constraints on development impact. A further step will be the introduction of an indicator-based rating methodology to provide a more objective basis for assessing project progress in achieving development objectives. Project Restructuring Projects may need to be restructured if the objectives are no longer of high priority or if the project components cannot be implemented as originally designed. A review of project restructuring found that the Bank approved the restructuring of 47 projects in 1993 (representing about 2.7 percent of the Bank's portfolio) and 55 projects in 1994 (about 3.1 percent of the portfolio). Restructuring was particularly important in some countries and in some sectors, such as agriculture. Although improvements in performance take time, the ratings of nearly three-fourths of the problem projects restructured in 1993-94 had improved by the end of fiscal 1994, thereby reducing the number of problem projects significantly and indicating that restructuring is a useful instrument for improving the quality of the Bank's portfolio. Quality at Entry and Supervision The Bank's Regional Offices have taken a number of measures to enhance the quality of new projects before they enter the portfolio. These include requiring up-front actions that are critical to project success; avoiding excessive complexity, ensuring readiness for implementation, strengthening borrower ownership and beneficiary participation; and pilot-testing new approaches. These initiatives are part of a continuing effort to improve quality at entry; it is, however, recognized that considerable room for improvement remains. The Regions have also continued to strengthen portfolio management practices through such measures as involving borrowers more closely in portfolio management; tightening the management of extensions of project or loan effectiveness and closing dates; using midterm review and supervision missions to identify needed midcourse corrections; taking pro-active measures, such as project restructuring, to increase the chances that projects will achieve their development objectives; using country portfolio performance reviews to address generic issues affecting performance; relating economic and sector work to critical implementation issues; using thematic supervision and intensifying the use of Resident Missions and local consultants to increase the cost-effectiveness of supervision; and addressing systemic problems in borrowers' procurement, accounting and auditing, and legal systems. Resettlement It is the Bank's policy that if projects involve involuntary resettlement, the incomes and living standards of the displaced people should be restored, enabling them to participate in the benefits of development. A review of all projects involving resettlement found that compliance with the requirements for resettlement planning was nearly total. The aggregate number of people displaced by ongoing operations in the portfolio remained stable in 1994 (with new entrants and exits balancing); the sectoral composition of resettlement issues is changing, with electric power declining in importance and transport and urban projects becoming more important. Project performance in terms of implementation of resettlement-related actions was rated as good by responsible staff, although resettlement specialists rated it somewhat lower in some cases. A high level of supervision is clearly needed to ensure sustained attention to this issue. 9 Future Steps In fiscal 1995 the World Bank plans to complete the actions contained in the "Next Steps" action program. By December 1994 almost all these actions had in fact been completed or were in the process of being implemented. While this represents substantial progress, the key issue for the longer term is the overall impact of these actions on the development effectiveness of Bank-assisted operations. As is common in the development field, this impact can begin to be measured only after a few years. In the meantime, maintaining the focus on implementation will remain a Bank priority. THE WORLD BANK A partner in strengthening economies and expanding markets to improve the quality of life for people everywhere, especially the poorest Headquarters European Office Tokyo Office 1818 H Street, N.W. 66, avenue d'I6na Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1, Marunouchi 3-chome Chiyoda-ku, Tokyo 100, Japan Telephone: (202) 477-1234 Telephone: (1) 40.69.30.00 Facsimile: (202) 477-6391 Facsimile: (1) 40.69.30.66 Telephone: (3) 3214-5001 Telex: MCI 64145 WORLDBANK Telex: 640651 Facsimile: (3) 3214-3657 MCI 248423 WORLDBANK Telex: 26838 Cable Address: INTBAFRAD WASHINGTONDC IN-34446 9 78S0821 334461- ISBN 0-8213-3446-8