Tanzania - Port Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 13763 PROJECT COMPLETION REPORT UNITED REPUBLIC OF TANZANIA PORT REHABILITATION PROJECT (CREDIT 1536-TA) DECEMBER 9, 1994 Energy and Infrastructure Operations Division Eastern Africa Department African Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENT Currency Unit = Tanzanian Shilling (at Appraisal Year 1984) Tsh. 1.0 = US$0.06 US$ 1.0 Tsh. 17.0 (At Completion Year, 300 Annual Average) Tsh. 1.0 = US$0.003 US$ 1.0 = Tsh. 340 ABBREVIATIONS AIDAB Australian International Development Assistance Bureau CIDA Canadian International Development Agency DANIDA Danish International Development Agency FINNIDA Finnish International Development Agency FLTs Fork Lift Trucks ICB International Competitive Bidding IDA International Development Association MCW Ministry of Communications and Works NORAD Norwegian Agency For Development Cooperation ODA Overseas Development Administration PCR Project Completion Report PCU Project Coordination Unit RMG Railway Mounter Gantry Crane RTG Rubber Tired Gantry Crane SAR Staff Appraisal Report SIDA Swedish International Development Authority SSG Ship-to-Shore Gantry Crane TA Technical Assistance TA & T Technical Assistance & Training TANZAM Tanzania-Zambia Highway TAZARA Tanzania-Zambia Railway TEU Twenty Foot Equivalent Unit (Container Size) THA Tanzania Harbour Authority TRC Tanzania Railways Corporation MEASUREMENT 1 meter (m) = 3.2808 feet (ft) 1 kilometer (km) = 0.6214 miles (mi) 1 metric ton (tonnes) = 1.023 short tons FISCAL YEAR Government: July 1 - June 30 TANESCO: January I - December 31  THE WORLD BANK FOR OFFICIAL USE ONLY Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation December 9, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Tanzania Port Rehabilitation Project (Credit 1536-0 TA and Grant 1741-0 TA) Attached is the Project Completion Report on Tanzania - Port Rehabilitation (Credit 1536- O TA and Grant 1741-0 TA) with Part II contributed by the Borrower. It contains detailed observations and gives an adequate account of project preparation, implemention and results. The US$27 million Credit was IDA's contribution to a US$91 million rehabilitation project for the port of Dar-es-Salaam. Nine bilateral donors -- Australia, Norway, the Netherlands, Denmark, Finland, Italy, the United Kingdom, Canada and Sweden -- would provide the balance. Approved in December 1984, the Credit was fully disbursed. It was closed on December 31, 1992, two and a half years behind schedule. The project aimed to increase port capacity and to strengthen the managerial and operational capability of the Tanzanian Harbours Authority (THA). Components of the original project design included conversion of two berths to fully containerized operations; purchase of container and general cargo handling equipment; modernization of grain handling facilities; rehabilitation of the lighterage wharf and oil jetty; and provision of technical assistance, training and consultancy services. During implementation, the project was expanded to include paving of eight berths, increased technical assistance, and additional equipment. As a result, total project cost increased from US$91 million to US$189.4 million, with the bilateral donors contributing the necessary funds. The project did modernize port facilities but traffic volume increases did not match expectations and the grain handling facility is grossly underutilized; the reestimated economic rate of return is 23% vs 33% at appraisal. Technical assistance was unsuccessful at all levels: it failed to strengthen the managerial and operational capability of the Port Authority even though institution- building has been the principal objective of Bank assistance to the Tanzanian port sector ever since the 1955 Railways and Harbours Administration Development Project (Loan 110-EA). The latest failure should be ascribed to the same shortcomings which plagued all previous efforts: lack of basic skills and motivation among THA staff; lack of suitable counterparts to work alongside expatriate experts; and diversion of technical assistance staff into dealing with operational matters at the expense of providing training to THA staff. The project outcome is rated as marginally satisfactory, its sustainability as unlikely, and its institutional impact as negligible. An audit is planned. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not othervise be disclosed without World Bank authorization.   FOR OFFICIAL USE ONLY TANZANIA PROJECT COMPLETION REPORT PORT REHABILITATION PROJECT (CREDIT 1536-TA) Table of Contents Preface ......................................................... Evaluation Summary .................................................. I. PROJECT REVIEW FROM THE PERSPECTIVE OF THE BANK ................ 1 Project Identity .................................................... 1 Background ....................................................... 1 Project Objectives and Description ........................................ 2 Project Design and Organization .......................................... 2 Project Implementation ................................................ 3 Project Results ..................................................... 5 Project Sustainability ................................................. 8 Bank Performance ................................................... 8 Borrower Performance ................................................ 8 Project Relationships ................................................. 9 Consulting Services . ................................................. 9 Project Documentation and Data . .......................................... 9 II. PROJECT REVIEW FROM THE PERSPECTIVE OF THE BORROWER (Summary Report) 10 III. STATISTICAL INFORMATION ..................................... 11 Table 1: Related Bank Loans and/or Credits ................................. 11 Table 2: Project Timetable ............................................ 12 Table 3: Disbursements .............................................. 12 Table 4: project Costs and Financing ...................................... 13 Table 5: Project Results .............................................. 14 Table 6: Status of Covenants ........................................... 16 Table 7: Use of Bank Resources ......................................... 17 Annex I PROJECT REVIEW FROM THE PERSPECTIVE OF THE BORROWER (Detailed Report) This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.   i TANZANIA PORT REHABILITATION PROJECT (CREDIT 1536-TA) PREFACE This is the Project Completion Report (PCR) for the Port Rehabilitation Project, for which Credit 1536-TA in the amount of US$ 27 million was approved on December 15, 1984 and signed on February 5, 1985. The Credit was closed on December 31, 1992, about 2 1/2 years behind schedule. It was fully disbursed and the last disbursement was on November 15, 1993. The PCR was jointly prepared by the Energy and Infrastructure Division of the Eastern Africa Department in the Africa Regional Office (Preface, Evaluation Summary, Parts I and HII), and the borrower (Part II). Part II comprises a summary report which is included in the main text and a detailed report prepared by the borrowers' consultant which is included as an Annex to this report. Preparation of this PCR was initiated during the Bank's final supervision mission of the project in March 1993 and is based, inter alia, on the Staff Appraisal Report; the Credit and Project Agreements; supervision reports; correspondence between the Bank and the borrower; and internal Bank memoranda.   ii EVALUATION SUMMARY Background and Project Design 1. The port of Dar-es-Salaam occupies a key role in the Tanzanian transport network, being situated at the confluence of the two rail networks (the Tanzania-Zambia Railway, and the Tanzasia RailWays Corporation) and the major road arteries. It is an important source of foreign exchange for Tanzania as a result of the transit traffic that passes through the port, and it acts an important gateway to the world for the land-locked countries of Uganda, Malawi, Rwanda, Burundi and Zambia, as well as for eastern Zaire. 2. Although prior to the project productivity in the port was at reasonable levels, operations were highly labour intensive and geared towards general cargo handling. There were fears that the port would not be able to cope with the world-wide shift towards containerization of freight, and that it would be unable to provide the levels of service necessary to attract transit traffic. In addition, a review of the previous IDA project (Loan 865-EA) suggested that the managerial and operational capability of the Tanzanian Harbours Authority (THA), responsible for operation of the port, was weak. 3. The project was designed to remedy these problems by providing funds for (i) the conversion of two cargo berths to fully containerized operation; (ii) the purchase of container handling equipment and the replacement of deteriorated general cargo handling equipment; (iii) modernization of grain handling facilities; (iv) rehabilitation of the lighterage wharf and oil jetty; and (v) provision of technical assistance (TA), training and consultancy services. The total estimated project cost was US$91 million, of which IDA would lend US$27 million. Nine bilateral donors provided the remaining funds (paras. 4.02 and 4.03 in Part I). Implementation Experience 4. The credit became effective in July 1985, about two months after the scheduled date, and the contractor was in place the next month. The bulk of the civil works were completed broadly on time. Some delays occurred in the conversion of berths to containerized operation due to adverse soil conditions. In addition, delays in the construction of the Ubungu depot caused some problems for port operation, indicating that co-ordination of construction activities could have been improved. After handing over the container yard to THA, soil settlement led to cracking of the pavement in localized areas. This, however, did not greatly disrupt port operations (para. 5.03 in Part I). 5. Implementation of the TA was slow and was hampered by THA's reliance on the TA personnel to perform on-line management tasks necessary to keep the port in operation and on a failure to provide adequate counterpart staff. The TA programme expanded greatly over that envisaged under the SAR. THA did not recruit a TA coordinator until after the mid-term review of the project, and evaluation of the TA was not undertaken until May 1991 (para. 5.10 in Part I). 6. Delay in the procurement of an item of container equipment required an extension to the credit to December 31 1992. The credit had already been extended from June 30 1990 to June 30 1991, principally to allow TA to be funded. One component of the project, rehabilitation of the lighterage wharf, was not undertaken and the funds were transferred to the construction of the containerized facility. (para. 5.08 in Part I).  iii Project Results 7. The overall scope of the project was substantially expanded during the implementation phase through inclusion of additional donor funds to support: (i) paving of berths 1 to 8 (about US$ 43.8 million); (ii) increased TA and-consultancy services (about US$ 14.3 million); and more equipment (about US$ 7.3 million) (Part III, Table). The additional funding amounted to US$65.4 million. But, even on the components included at the appraisal stage, there was a substantial cost over-run of about US$33 million, of which about US$16 million represented over-runs on civil works costs (conversion to containerization and the grain handling facility) and about US$ 17 million resulted from increased equipment quantities and prices. Of the increase in civil works costs, about US$10.5 million was caused partly by the appreciation of the Yen and partly by unforeseen adverse soil conditions on site. 8. The project succeeded in its aim of modernizing facilities at Dar-es-Salaam. Ship waiting time and berth time were both drastically reduced. Productivity (boxes handled per gang hour) in container handling was initially disappointing as construction activities limited the amount of available stacking space, but improved by 25 %; this was partly due to the introduction of an incentive payment scheme for workers in these berths. Traffic volume at the port did not increase as fast as had been planned, partly due to lower economic growth in the region and a suspected shift in Zambian traffic to South Africa. Container dwell time was initially very high at the port but fell after THA made considerable efforts to resolve the problem (para. 6.03 (i) in Part I). 9. The TA aspect of the project was largely unsuccessful at all levels. The TA program expanded greatly from that envisaged in the SAR but was not coordinated or evaluated until after the mid-term review of the project. There has been a decline in the availability of cargo handling equipment attributed to the failure of the THA technicians to absorb the training provided by FINNIDA for three years; this may partly reflect the poor educational backgrounds and low morale of these technicians; aspects which were not fully taken into account when the TA component was designed. The TA personnel were often unable to provide training to their managerial counterparts because they had to devote a large amount of time to port operations and because the counterparts were often not made available (paras. 5.10 and 6.03 (iii) in Part I) and in some cases the donor-supplied TA personnel did not possess the required experience and skills. Without improving its commercial orientation and strengthening accountability for performance for both management and staff, it is unlikely that further provision of technical assistance would be of benefit to THA. 10. The grain handling facility has been grossly underutilized. Even during the famine relief efforts of 1992 much of the grain was bagged at berthside using portable bagging machines, by-passing the grain terminal (paras. 5.06 and 6.03 (iv) in Part I). Project Sustainability 11. Given the limited impact TA in terms of terms of improving THA's institutional capabilities, the extent to which the project benefits can be sustained will depend on THA's ability to improve its organization and management. THA has already made progress in a number of areas e.g. reducing container dwell time, and the introduction of incentive schemes for container handling. THA has also embarked on a new corporate strategy comprising measures to: (i) decentralize its operating functions away from headquarters to the port managers; (ii) restructure its organization including divestiture of some of its important businesses such as cargo handling. Improving the organization and management of port activities is important not only to sustain the benefits of this project, but to ensure that the port  iv retains its commnercial attractiveness in relation to other regional ports given the given political developments in the region (paras. 7.01 and 7.02 in Part I). Lessons Learned 12. The size and complexity of the project was not commensurate with THA's capacity to efficigntly implement and ensure the sustainability of project benefits. The effectiveness of the TA program designed to overcome THA's institutional weaknesses was handicapped by; (i) the lack of basic skills and motivation aniong THA staff; (ii) the lack of suitable counterparts to work with THA staff; (iii) the diversion of TA personnel into dealing with operational issues at the expense of providing training for THA staff. The problem of the size and complexity of the project was compounded by the large number of donors involved and the need for THA to deal with their different procurement rules and procedures. To their credit, THA management undertook a number of positive steps during this project (such as the reduction in container dwell time and the introduction of incentive payments in the container yard). However, until the fundamental issues regarding THA's commercial orientation and accountability for performance are addressed, it is questionable if additional provision of further TA will be of benefit. As indicated above, THA has started to seriously address these issues, with IDA assistance, under the ongoing Port Modernization Project. THA is also in the process of signing a performance contract with its parent ministry, the Ministry of Works, Communications and Transport. 13. TA needs to be coordinated and monitored on a continual basis throughout the period in which it is received. THA did not undertake this until prompted to do so by the mid-tern review, and evaluation was not begun until May 1991. IDA should have played a greater role in encouraging THA to do this earlier, given the latter's known organizational weakness. Provision of long-term TA to THA should be avoided unless the organization is restructured and made commercially-oriented and procedures for improving accountability for performance are introduced. 14. THA should improve its procurement processes to ensure that there is a continual monitoring of procurement by higher management and that approvals for defined smaller purchases are decentralized to appropriate levels; this can help to avoid delays which require an extension of credit. 15. Investment needs should be carefully assessed at the time of appraisal. The grain terminal is currently redundant; an evaluation of the possible volatility of grain trade should have been undertaken to assess the impact that this might have had on the investment requirements.   1 TANZANIA PORT REHABILITATION PROJECT (LOAN 1536-TA) PROJECT COMPLETION REPORT PART I: PROJECT REVIEW FROM IDA'S PERSPECTIVE 1. Project Identity Project Name - Port Rehabilitation Project Credit Number - 1536-TA RVP Unit - Africa Region Country - Tanzania Sector - Transport Subsector - Ports 2. Background 2.01 Tanzania has a large territory (945,000 square kilometers) with a widely dispersed population (about 26 million) and a low population density. The Tanzanian transport sector has both significant national and regional importance. The Tanzania-Zambia Railway (TAZARA) and the Tanzania-Zambia Highway (TANZAM) provide Zambia's main outlet to the sea. The central line of the Tanzania Railways Corporation (TRC) provides Rwanda, Burundi and Zaire with their most important access route and Uganda with an important alternative route to the sea. The road network in Tanzania is important both for domestic traffic and for that from neighboring landlocked countries. The port of Dar-es-Salaam is situated at the confluence of the 2 railways systems and the major road arteries, and is by far the largest of the 3 main sea ports operated by the Tanzanian Harbours Authority (THA). 2.02 At the time of project conception, Tanzania was suffering from severe shortages of foreign exchange, reflecting poor export performance. Rehabilitation of the transport infrastructure was seen as important in enhancing exports and in gaining much-needed foreign exchange from transit traffic. Bank objectives for the transport sector focussed on the modernization of existing facilities, reducing the costs and improving the capacity of the rail system and strengthening the planning and institutional capabilities of the key transport sector entities in Tanzania. Major investments in road and rail infrastructure were planned. 2.03 Due to its pivotal role, modernization of the port of Dar-es-Salaam was identified as an important pre-requisite for the success of these wider objectives. Although port productivity was at a reasonable level, operations were geared towards general cargo handling and were highly labour intensive. It was feared that the port would not be able to cope with an increased volume of containerized traffic, which was becoming the dominant mode of shipping. Also, there were concerns that the port would not be able to provide the level of service that would continue to attract transit traffic.  2 2.04 This project followed on from Loan 865-EA implemented from, 1972-1979 (the port of Dar-es- Salaam being administered at that time by the East African Harbours Corporation), which provided financial assistance for the modernization of the port of Dar-es-Salaam. This project successfully added to the physical capacity of the port; however, managerial and operational capability remained at a low level. 3. Project Objectives and Description 3.01 The Port Rehabilitation Project was designed to: (i) modernize the physical infrastructure of the port of Dar-es-Salaam, in particular to allow it to cater efficiently for rapid containerization of general cargo traffic; and (ii) improve the capability of THA staff through a manpower training program aided by technical assistance (TA). 3.02 The project consisted of the following components: (i) conversion of the existing cargo berths 10 and 11 to a solely container handling facility (the major component of the project); (ii) the provision of special container handling equipment and replacement of deteriorated general cargo equipment; (iii) modernization of grain handling facilities and equipment; (iv) rehabilitation of the lighterage facilities and the petroleum jetty; and (v) training, technical assistance and consultancy services. 3.03 IDA funds for the project were on-lent to THA under a Subsidiary Loan Agreement from the Government to THA. Cofinancing was obtained from NORAD, the Netherlands, DANIDA, FINNIDA, ODA, ITALY, CIDA, SIDA and AIDAB. All funds from bilaterals were provided to THA on a grant basis. IDA was to provide credits of US$27 million compared to a total project cost of US$91 million. 4. Project Design and Organization 4.01 The project design fitted with the perceived requirements of THA, both in terms of upgrading infrastructure and providing training/technical assistance. The executing agency for the Project was THA, under the supervision of the Ministry of Communication and Works (MCW). A Project Coordination Unit (PCU) was established, headed by the Director of Engineering of THA. The consulting firm, Bertlin and Partners, carried out the feasibility study for the project and were selected as the supervision consultant for the civil works financed by IDA. 4.02 The contributions of the bilaterals were as follows: Netherlands: grain terminal; DANIDA: ship- to-shore gantry cranes (SSG) and one rail-mounted gantry crane (RMG); FINNIDA: 5 rubber tire gantry cranes (RTG), fork lift trucks and trailer units, standby generator, and container workshop equipment; NORAD: rehabilitation of the oil jetty, training at Bandari College; and Italy, ODA, CIDA and AIDAB financed some minor components of the project. In addition, SIDA (with some funds from NORAD) financed work on Berths 1 to 8. 4.03 The PCU, consisting of three expatriates and two local counterparts, coordinated the various components of the project by meeting regularly with the bilateral donors in Dar es Salaam. Although financed by different donors (chiefly SIDA, FINNIDA and Netherlands), the team functioned well during the project implementation period.  3 5. Project Implementation 5.01 Overall. The date of credit effectiveness was postponed from May 1985 to July 1985 because of a delay in the finalization of an agreement with NORAD. The project commenced without this agreement, which was not signed until October 1987.7 The Closing Date of the project was extended from Junqr30, 1990 to June 30, 1991, and finally to December 31, 1992 to enable THA to finalize the procurement of the sixth RTG. There was an overall time over-run of 61 %, but this was primarily related to the procurement of the sixth RTG and the TA. Civil works were largely completed on time. 5.02 At the time of project appraisal, the most significant risks to the project were estimated to be the possibility that traffic would not increase as quickly as forecast, physical risks associated with the civil works and that the institutional benefits would not materialize as rapidly as was anticipated. As para. 5.03 describes, some of the physical risks did materialize, although they did not greatly influence the operations of the port. Although, at appraisal stage, it was felt that THA was relatively adaptable to reap significant institutional benefits from the provision of the TA, in the end THA's performance in this area was far below expectations (para.6.03 (iii) in Part 1). 5.03 Container Terminal. The container terminal and depot were completed on time, but at cost over- run of US$10.5 million (or 29% of the budget estimate, para. 6.02). After handing over the container yard to THA, uneven settlement of the pavement blocks and block breakages occurred locally in the container yard. Investigations revealed that the high level of ground water in the affected areas was the cause. Remedial work to lower the ground water level was incorporated into the second phase project, the ongoing Port Modernization II. Soil investigation prior to design failed to predict the underground water seepage and its results.There were also delays in the completion of the Ubungo depot which had an adverse impact on storage capacity at the port; Ubungo should have provided storage for the period after the transit sheds at Berths 10 and 11 were demolished. Roof leakage delayed the use of the Kurasini depot. The leakage was fixed by THA's own work force, although the repairs took some time to organize. 5.04 Equipment. Equipment for container handling was procured broadly as scheduled with the exception of the sixth RTG. Delays in procurement of this item and on the part of the supplier required an extension of the credit to December 31 1992. The new container workshop was completed and trained mechanics started container operation in June 1988 close to schedule. However, a decline in the availability of equipment suggests that the training was not fully successful (see para. 6.4 below). Rehabilitation of the existing Fork Lift Trucks (FLTs) started in May 1992. SIDA, CIDA and ODA assisted THA to rehabilitate portal cranes, mobil cranes, FLTs and parts of the central and satellite workshops. 5.05 Rehabilitation of Berth 1-8 Pavement. The rehabilitation of the pavement behind Berths 1-8 was not included in the SAR. However, there was an urgent need to undertake this work, and SIDA (principally) and NORAD agreed to supply the required funds. The pavement design in an area behind the berth 8 transit shed was upgraded to an extra heavy duty pavement to give THA additional container stacking capacity to preclude serious congestion in container operations. Construction was completed within three and a half years without delays. Cargo handling has become more efficient and handling equipment depletion substantially reduced due to the new paving. 5.06 Grain Terminal. A grain terminal was constructed, financed by the Netherlands, near THA's headquarters. Underground and roof belt conveyers, bucket elevators and other equipment for handling bulk grain were installed in the terminal.  4 5.07 Kurasini Oil Jetty. The two phases of rehabilitation of Kurasini oil jetty were completed behind schedule, partly due to NORAD's condition that the ownership of the pipelines and manifold being transferred from Tipper oil refinery to THA, but mainly because of increased scope of work and heavy delays caused by work stoppages due to vessels utilizing the jetty for discharging of petroleum products. The rehabilitated and upgraded jetty, however, is not designed to accommodate the larger vessels currently calling at Dar es Salaam, and would need to be further strengthened to allow for 40,000 D4WT vessels. 5.08 Rehabilitation of lighterage Wharf.Rehabilitation of the lighterage wharf was not undertaken, due to the failure of the chosen contractor to obtain a letter of credit. The funds were transferred to the construction of the container terminals. 5.09 Training and Training Equipment. The procurement of the training equipment for the Bandari College was not carried out within this project. Restrictive conditions on the initial bid (which required all bidders to bid for all lots) meant that only one potential supplier replied to the tender invitation. The bid was re-tendered under the Port Modernization Project and was finally awarded in November 1992. The training in Bandari College financed by NORAD was seriously hampered by the fact that the equipment was not procured. Engineering training did take place at the port when FINNIDA brought in equipment into the container workshop. However, during this period there was considerable uncertainty over Bandari College's role in meeting THA/s future training requirements. 5.10 TA Program. The SAR contained a detailed TA program specifying THA's institutional weaknesses and the TA required for their resolution. Once the project was underway, however, this TA schedule was abandoned as the TA program expanded into a diverse program that was not initially well coordinated and not evaluated adequately. THA recruited the TA coordinator in 1990 after consensus was reached with IDA and donors during the mid-term review on the need for more systematic coordination and evaluation of TA performance and effectiveness. With hindsight it could be suggested that IDA should have encouraged THA to take such actions earlier in the project. THA started to monitor performance of the TA personnel in May 1991 but evaluation was found to have lapsed by a Bank mission in July 1992. The TA team encountered difficulties in transferring knowledge to THA staff. For the higher level staff, there was a tendency for TA personnel to concentrate on operations necessary to run the port as line managers rather than TA personnel. THA then became reliant on these personnel, and in some cases their contracts were extended to ensure continued operation of the port. At the technician level, low educational backgrounds and poor morale hindered the acquisition of new technical skills; these constraints were probably not fully taken into account in the TA design. FINNIDA attempted for three years, unsuccessfully, to train technicians in the workshop, and attributed much of their lack of success to the poor educational background of the THA staff. 5.11 Lessons From Implementation Experience. The main lessons learned during implementation were: (i) effective implementation of TA requires that the recipient is committed to the concept of TA and to providing adequate counterpart staff; this did not happen under this project; in addition, TA needs to be coordinated and monitored and evaluated (both donors and THA needed to place greater emphasis on this); although the latter commenced in May 1991, this evaluation lapsed during 1992; the issue of TA is also discussed in para 6.03 (i);  5 (ii) THA needs to improve its procurement processes to ensure that there is a continual monitoring of the procurement; this can help prevent delays which may require an extension of credit; and (iii) construction activities need to be coordinated to minimize their disruptive impact upon port operations; this did-not always happen in this project. 6. Project Results 6.01 Project Costs. The project succeeded in its aim of modernizing port facilities at Dar-es-Salaam. There was a cost over-run of about US$33 million on the project items that were included in the SAR, representing about 36.2% of total project costs estimated at appraisal stage. Of this, about US$16 million represented over-runs in civil works costs (associated with berths 10 and 11 and the grain terminal) and about US$17 million was a result of increased equipment prices and quantities. The overall scope of the project was substantially expanded during the implementation phase through inclusion of additional donor funds to support: (i) paving of berths 1 to 8 (about US$ 43.8 million); (ii) increased TA and consultancy services (about US$ 14.3 million); and (more equipment about US$ 7.3 million). The additional funding amounted to US$65.4 million. Table 4 in Part III provides a comparison of the cost and project financing estimates at appraisal stage to the actual results including the costs added during the implementation phase. 6.02 Reasons for Cost Overruns. Apart from the increased scope of works, the main reasons for the cost overruns were: (i) about US$ 10.5 million or 29% more than appraisal estimate for the container terminal due to unforeseen adverse soil conditions at the container yard (para. 5. ) and the appreciation of the Yen; (ii) increased quantities and prices of container terminal equipment compared to appraisal estimates; and (iii) higher construction costs of grain terminal (about US$5.7 million, or 39% more than appraisal estimate. The higher costs were covered by supplemental financing from SIDA for the container terminal and by funds originally allocated for Bandari College equipment and training (para.6.09 in Part 1). 6.03 Achievement of Performance Objectives and Targets. Many of the performance targets set in the SAR have been met, although the increase in traffic volume at the port has been lower than anticipated (in 1992 total throughput of non-petroleum traffic was 2,308,000 tonnes against a forecast of 2,778,000 tonnes). As a consequence, berth utilization is also lower than forecast. Slow economic growth in the region and diversion of Zambian cargo to South Africa are the likely causes of this (although there is no official confirmation of the latter). Ship waiting time in 1993 was reduced to 40% of the 1981 level reported in the SAR, average berth time fell by a similar amount and productivity for break-bulk cargo increased fourfold over 1981 levels. Container handling productivity was initially disappointing partly because stacking space was occupied by the civil works contractor. The introduction of an incentive scheme covering 30% of the staff in the container terminal may have contributed to a promising 25% increase in productivity. The share of container traffic increased from 19,234 TEU in 1982 to about 100,000 TEU in 1993 (see table below for traffic growth between 1988 and 1992) or average of 16% p.a.  6 Comparison of SAR Traffic Estimates to Actual! (000 ton) 1988 1992 SAR Estimate Container 345 596 Dry Cargo 296 282 Other 803 851 Total 1,444 1,729 Actual Container 690 820 Dry Cargo 1,370 1,600 Other 2,050 2,320 Total 4,110 4 740 1/ Traffic growth estimates by mode were included in the SAR for only two years. Overall performance in terms of achievement of objectives and targets was as follows: (i) Container Dwell Timne. Container dwell time increased from 35 days in 1987 to 53 days in 1988 and remained at a high level even with the newly completed container yard, due to the: (i) lack of reliable communication system among the shipper, consignee, forwarding agent and the shipping company; (ii)lack of container handling equipment; (iii) lack of foreign currency; (iv) erroneous import declarations; and (v) long bureaucratic procedures in the Custom clearance process. THA made considerable effort to reduce the container dwell time by meeting with relevant parties to find solutions. This worked well and by September 1992, the average dwell time was reduced to about 13 days. THA continues to monitor this. (ii) Equipment Availability. Container equipment availability was kept at 95 % at the initial stages of the project. However, it started deteriorating toward the end of the project period to around 70% - 80% for the different types of crane. Equipment availability in general cargo is about 50% for FLT of smaller lift capacity in 1991 and 1992. This decline is a cause for concern to both THA and IDA. (iii) TA Program. The TA program has largely been unsuccessful. Evidence for this can be seen in the decline in the availability of equipment (see par 6.4 above) and in the reliance on the TA personnel to play key roles in port operations. It is argued below that this is largely due to THA's lack of commitment to genuine TA, although co-ordination and evaluation of TA could have been encouraged more by the donors and IDA. (iv) Grain Imports. Grain import volume has been far smaller than the forecast at appraisal, almost nil, except for emergency food import in 1992. Even in handling the emergency  7 food at that time most grain was bagged using portable bagging machines placed at berth aprons then transported inland directly from aprons by trucks and railway, leaving the grain terminal grossly under utilized. (v) Economic Rate of Return. A post-implementation economic rate of return was re- f estimated as 23 % compa-red to 33 % projected at appraisal. The ERR has been calculated as the discount rate that equalizes the present value of net benefits from the project to initial investment costs. The net benefits from this project are the reduction in ship waiting time and the avoided costs of rerouting cargo through neighboring countries and through land routes less O&M costs. The revised ERR is about 30% lower than that estimated at appraisal because of higher investment and lower traffic levels. The ERR is not as low as the higher investment costs, and lower traffic levels might suggest because of the higher growth of container traffic than anticipated at appraisal. (vi) Financial Performance. In FY1992/93 THA achieved a net profit after tax of Tsh. 2.5 billion representing a net profit margin of 8.9%. Although this is clearly a low profit margin, it represented a significant improvement over FY1991/92 when a net after tax loss of Tsh. 2.2 billion was recorded. In recent years THA profitability has declined due to the following several factors: (i) non-adjustment of tariffs on a regular basis (except in FY1992/93 when a substantial increase was implemented); (ii) continual depreciation of the Tsh. from Tsh. 17.0 = US$1 at appraisal to Tsh. 500 = US$1 in 1994; (iii) failure to improve operating efficiency and reduce costs; etc. As a result THA was not able to consistently achieve the rate of return on assets of at least 7 % as agreed under the Credit. THA did, however, comply with the other agreed financial covenants, i.e., (i) reduced and maintained its arrears from Government to less than 90 days during most years; (ii) consulted with IDA on new capital investments exceeding US$2 million; and (iii) often submitted its audited financial statements within 6 months of the end of a financial year. As of end of FY1992/93, accounts receivable represented less than 30 days worth of sales on average, although a few items have been outstanding for more than 60 days. Adequate provision has been made in the accounts for the long outstanding debts. Under the Port Modernization Project, a new financial improvement strategy was agreed comprising: (i) repayment to the Government of the fixed assets component of existing donor grants plus the past and future World Bank on-lent funds to THA and all future grants; (ii) debt to equity ratio of 60:40; (iii) adoption of a dividend policy; (iv) institution of well-funded and administered productivity incentive scheme; (v) investment of cash revenues in higher yielding assets; and (vi) use of tariff structure as a device for both altering the behavior of port users and for cost recovery. These measures are being implemented with varying degrees of success with the exception of (iii) and (v) (under consideration). In addition, THA has invited proposals for a study to recommend a new corporate strategy and a plan for commercialization and/or divestiture of its activities. Implementation of the study's recommendation is expected to lead to improved operational efficiency and profitability. 6.04 Lessons from Project Results. The main lessons that can be learnt from an assessment of the results are that: (i) The size and complexity of the project was not commensurate with THA's capacity to efficiently implement it and ensure the sustainability of project benefits. The effectiveness of the TA program designed to overcome THA's institutional weaknesses was handicapped by; (a) the lack of basic skills and accountability for performance among  8 THA staff; (b) the lack of suitable counterparts to work with THA staff; (iii) the diversion of TA personnel into dealing with operational issues at the expense of providing training for THA staff. Until the morale and skills of THA are strengthened, it is not clear that additional THA will provide any further benefit; (ii) investment needs should-becarefully assessed at the time of Appraisal; the grain terninal is currently redundant and a careful assessment should have been made of the volatility in grain traffic and the impact that this might have had on the choice of options for modernizing grain handling facilities; and (iii) incentive based payment schemes can have a positive impact on productivity (eg container handling) and consideration should be given to their application throughout the organization; however they need careful preparation before implementation; and 7. Project Sustainability 7.01 It is not clear that the institutional capability exists to deliver a sustainable improvement in results. The recent decline in equipment availability indicates that THA needs to improve its maintenance procedures if the benefits of the project are to be permanent. The reductions in waiting and service time (ship dwell time) are encouraging, but progress on this, and other aspects of port operations needs to be continued. In addition, THA are in the process of signing a performance contract with the MCW, their parent ministry, and are considering substantial organizational changes such as restructuring or privatization. It is likely that major changes will be needed to improve THA's weak organizational capability. These need to be targeted at improving the basic skills and quality of THA's workforce and the introduction of accountability in management and greater discipline at the workplace. Without reforms to introduce this, it is likely that the long term objectives of the project will not be realized. 7.02 Concerns over the future prospects of the port must be raised by the political and economic developments in Southern Africa which have made competitors to Dar-es-Salaam more acceptable and attractive. 8. Bank Performance 8.01 The Appraisal report made a solid foundation for the project implementation and additional project components were included by donors based on consultations among donors. The annual donors meetings were held jointly by the donors and THA every year. The Bank carried out supervision missions regularly, about two to three times a year throughout the project period. As a result, the Bank was able to monitor and take some corrective actions during the implementation of the project. The Bank played a considerable role in pushing THA to reduce container dwell time. However, it could have played a greater role in the earlier stages of the project, in particular in encouraging THA to coordinate and evaluate the TA. With hindsight, the Bank could have gained a better understanding of the institutional weaknesses in THA at the project design stage and could therefore have supported a smaller project with better monitoring and control of technical assistance. 8.02 A mid-term review mission which included consultants financed by IDA and other donors was carried out in June 1988. Based on mid-term review and subsequent supervision missions, the preparation of the Port Modernization II was initiated.  9 9. Borrower Performance 9.01 There were delays in the compliance with the covenant of the Development Credit Agreement regarding submission of the audited financial statements. This was due to (i) shortage of qualified staff; (ii) lack of a clearly defined accounting and procedural manual; and a (iii) lack of systenmatic intwnal accounting audits. 9.02 THA's management capability was not strong enough to adequately implement this project. There were delays in procurement that caused extensions to the credit; THA relied substantially on TA personnel for running port operations and were not fully committed to the concept and practices of the TA. THA did make progress in some areas, and should be congratulated on their success in reducing container dwell time. 10. Project Relationship 10.01 The working relationship between the Bank and the Borrower was satisfactory during the development of the port rehabilitation project and the modernization of the Port of Dar-es-Salaam. There was constant dialogue between IDA and the Borrower which facilitated the timely actions needed during the revisions in the scope of the Project. The annual donor meetings chaired by the Borrower proved to be beneficial to all parties in terms of generating additional resources to cover the cost overruns, eg SIDA grant, which made project implementation easier. 11. Consulting Services 11.01 The PCU and the Engineering Department functioned well during this project and a major part of the success is attributable to their coordination with the donors and with the consultant teams. The team in the PCU worked well together even though they were funded by different donors. However, given the problems referred to in Section 9, conditions were not ideal to enable smooth transfer of technology during the project period. Many expatriates, as a matter of necessity, performed as trouble shooters or temporary line managers, rather than trainers, to their Tanzanian counterparts. 12. Project Documentation and Data 12.01 The Development Credit Agreement and Project Agreement for the project incorporated the Borrower's priorities. Amendments were made to the Development Credit Agreement (May 29, 1990, June 28, 1991 and June 30, 1992) to allow for the extension of the Closing Date of the project. 12.02 Supervision memos and correspondence were largely well documented and proved to be useful in the preparation of this PCR.  10 TANZANIA Port Rehabilitation Project CREDIT 1536-0 TA - GRANT 1741-0 TA PROJECT COMPLETION REPORT PART II: Borrower's Assessment of Project Implementation (Summary Report)l/ 1.01 Reference is made to two Credits, programmed to act on one integrated rehabilitation project for the port of Dar es Salaam (Credits 1536-0 TA and 1741-0 TA refer). 1.02 The project formally consists of 7 parts, but these are actually concentrated in 3 major elements: civil & building infra-structure, equipment and Technical Assistance & Training (TA & T). There is however a fourth major element, but it has never figured as such and has never been targeted for concentrated input. This element is the element that should make the other three investments worthwhile, and that element is the functioning of management. 1.03 Taken one by one, infrastructure has performed satisfactorily. So has the equipment element. The longer-term performance of TA & T is uncertain, but management performance is unsatisfactory by modern standards. Since Dar es Salaam Port is facing tough times ahead, the last qualification of having to measure management performance by a modern yardstick has become a necessity for survival in the international port scene. The outcome measuring management however, is negative. THA will have great difficulty in meeting stiffer competition from neighboring ports, while at the same time having to accept a considerable downturn in traffic. 1.04 A major contributing factor to the less than modem performance of THA's management is due to structural omission of this element in the project design stage. In relative terms, heavy (over) investment has occurred in infrastructure, equipment and TA & T in comparison with upgrading management capability. The Project did not allow for providing THA management with a modern port institution and integrated port management methods 1.05 Since the Borrower, the Tanzania Harbours Authority (THA), is fortunate enough to have available right now a further Credit from the Bank, it should most seriously be considered to undertake corrective action rectify the shortcomings of the TA. I/Detailed report prepared for the borrower by its consultant is shown as an annex to the PCR.  11 TANZANIA PORT REHABILITATION PROJECT PART III STATISTICAL INFORMATION Table 1. Related Bank Loans and/or Credits Loan/Credit Title Purpose Approval Year Status Railways & Harbours Administration To assist the East African Railways & Harbours 1955 Completed in 1958 Development Project Loan 11O-EA Administration (EAR&H) in financing its four-year (1954-57) Development Program. Second East African Railways & Harbours To assist EAR&H in financing its three-year Capital 1965 Completed in 1971 Administration Development Project Renewals and Expansion Program Loan 428-EA Harbours Corporation Project To help increase the capacity and efficiency of the ports 1969 Completed in 1977 Loan 638-EA of Dar-es-Salaam and Mombasa Third Harbours Project To assist the East Africa Harbours Corporation (EAHC) 1972 Completed in 1979 Loan 865-EA in financing its five-year Development Program (1972-76) Credit S024-TA To help assist the Tanzania Harbour Authority (THA) in 1979 Completed in 1983 Dar es Salaam Port Engineering the first phase of the proposed expansion of the Port Dar-es-Salaam.  12 Table 2. Project Time Table Item Planned Date Revised Date Actual Date Identification 07/78 Preparation 05/81 Appraisal 1 09/78 11/81 11/812/ Appraisal 2 07/82 Negotiation 07/82 01/83 01/83 Board Approval 09/82 03/83 12/84 Signature 02/85 Effectiveness 01/83 07/85 Closing 06/90 12/92 Completion 12/89 04/93 Table 3. Disbursement (US$ million) Year Appraisal Estimate Actual Annual Cumulative Annual Cumulative 1.3 1.3 1985 1.4 7.0 8.3 1986 8.6 10.0 7.8 16.1 1987 8.8 18.8 13.8 29.9 1988 6.9 25.7 1.5 31.4 1989 1.0 26.7 0.3 31.7 1990 0.3 27.0 0.3 32.0 1991 0.0 1.7 33.7 1992 0.0 0.5 34.2 1993 0.0 2/ No agreement was reached with the Government on the scope of the project during appraisal.  13 Table 4. Project Costs and Financing A. Project Costs (US$ million) Item Appraisal Estimate Actual Local Foreign Total Advance Civil Works 3.0 1.7 4.7 6.6 Main Project Civil Works 11.9 24.4 36.3 111.53/ Main Project Equipment 0.6 17.6 18.2 50.24/ Technical Assistance/Training 0.9 3.1 4.0 15.15/ Consultative Services 0.7 2.5 3.2 6.06/ Sub-total 17.1 49.3 66.4 189.4 Contingencies: Physical 1.9 2.1 4.0 - Price 13.0 7.6 20.6 - TOTAL PROJECT COSTS 32.0 59.0 91.0 189.4 3/ Of this, US$ 43.8 million was spent on projects not originally included in the SAR. 4/ Of this, US$7.3 million was spent on equipment outside the scope of the SAR. 5/ Of this, US$11.9 million was spent on the provision of TA outside the scope of the original SAR. 6/ Of this, US$2.4 million represents expenditure on consultancy services outside the scope of the original SAR.  14 B. Project Financing (US$ million) Sources Appraisal Estimate Actual Local Foreign Total IDA = 27.0 27.0 34.2 Cofinanciers: NORAD 2.8 2.8 18.6 NETHERLANDS 10.3 10.3 21.0 DANIDA 7.9 7.9 16.7 FINNIDA 6.8 6.8 25.8 ITALY 3.8 3.8 4.8 ODA 0.4 0.4 2.8 CIDA 1.3 SIDA 34.7 AIDAB 3.5 Government 32.0 32.0 26.0 r - -~~Total 32.0 59.0 91.0 189.4 Table 5. Project Results A. Direct Benefits Indicators Appraisal Estimate Actual 1. With Container Terminal only Ship Waiting Time (days) 1420 624 Berth Utilization (percentage) 90 83 2. With Container Terminal and Grain Facility Ship Waiting time (days) 402 227 Berth Utilization (percentage) 80 56  15 B. Economic Impact Indicators Appraisal Estimate Re-estimated Rate of Return | 33% 23% Underlying Assumptions 1. Traffic 13% lower 12% 2. Costs 10% higher 30% 3. Traffic 13% and costs 10% lower 11% 4. Container traffic 20% lower than 11% forecast, overall traffic 13% lower 5. Port efficiency reduced by 33% 23% C. Studies Title Purpose Status Impact Equipment Rehabilitation To determine the Completed Study will form the Study requirement for spare basis for the parts for existing THA rehabilitation of spare equipment parts component of the _____ ____ ____ ____ _ I_I project.  Table 6. Status of Covenants Covenants Status Development Credit Agreement Section 3.02 Annual review of tariffs and approve the increase. OK Section 3.03 Consult with IDA on any capital investment more than US$2.0 OK million. Project Agreement Section 2.06(b)(i) Records and monitors the progress of the project, identifies services to be OK financed out of the Credit Section 2.06(b)(iii) Report to IDA at regular intervals Section 2.08(b) Informs IDA of any condition which interferes with the accomplishment of OK the purpose of credit. Section 2.09 The result of the study on rehabilitation submitted to IDA OK Section 3.02 THA shall maintain its machinery, equipment and other properties. OK Section 3.04 Commence a comprehensive manpower training program for its OK personnel Section 4.02(b) Submit the independently audited financial statements not later than nine Complied but with some delays. months from the end of the fiscal year Section 4.03 THA shall not incur any debt unless its net revenues are not less OK than 1.5 times the estimated maximum debt service requirement for any succeeding fiscal year. Section 4.04 Re-evaluation of fixed assets every three years starting 1987 done in 1992 Section 4.05 Revision of tariff based on the actual cost. revised in 1992  17 Table 7. Use of Bank Resources A. Staff Inputs STAGE OF PROJECT CYCLE STAFF WEEKS Through Appraisal 55.5 Appraisal through Board Approval 146.2 Board Approval through Effectiveness 12.7 Supervision 79.2 PCR 3.5  18 B. MISSIONS Stage of Month/Year No. of Specialization Performance Types of Project Cycle Persons Represented ' Rating Status 2 Problems I Preparation 07/78 1 FA Appraisal 1 11/81 6 PE,EC,PS,FA,RA Appraisal 2 07/82 Supervision 05/85 1 EC 1 2 Supervision 01/86 2 EC,PS 1 1 Supervision 02/86 2 EC,PS 1 2 Supervision 08/86 1 PS 2 Supervision 08/87 1 PS 2 Supervision 08/88 2 PE,EC 2 2 Supervision 08/89 2 PE,FA 2 3 Supervision 02/90 2 PE,EC 2 2 Supervision 06/90 2 PE,EC 2 2 Supervision 10/90 2 PE,FA 2 2 Supervision 11/90 1 FA 2 1 Supervision 01/91 1 PE 2 2 Supervision 04/91 2 PE,FA 2 2 Supervision 07/92 3 PE, FA, PROC 2 3 Supervision 11/92 2 PE,FA 2 2 Key to Specialization: PE=Port Engineer; PS=Port Specialst; FA=Financial Analyst; EC=Economist; OA =Operations Analyst; PROG=Procurement Specialist Key to Problems: 1 =problem free or minor problems, 2=moderate problems, 3 =major problems  19 Annex I TANZANIA Port Rehabilitation Project CREDIT 1536-0 TA GRANT 1741-0 TA PROJECT COMPLETION REPORT PART II: Borrower's Assessment of Project Implementation (Detailed Report) INTRODUCTORY REMARKS This Part was drawn up without prior knowledge of the contents of Parts I and III of this report. Of the three recent Bank Credits involving the Tanzania Harbours Authority (THA) for respectively Engineering, Rehabilitation and Modernization, this PCR covers the second Credit. An associated Grant (no. 1741-0 TA) is also included in this project assessment. The previous Credit (no. S-24-TA for USD 2.5 million during 1980-1983) was rounded off with a PCR in 1992, which stated that the objectives of the project were achieved within the estimated costs. The Borrower is in agreement with those PCR findings, indicating that both the Bank and the Borrower were satisfied with the starting position of the Credit being dealt with in this PCR. The third Credit (no. 2095-0 TA) is currently effective and forms a natural extension of the second one. INTEGRATED PERFORMANCE The project was targeted to rehabilitate and modernize the port of Dar es Salaam, in which it succeeded as a whole. Of the project objectives almost all were fully met. Quoting the 1984 Staff Appraisal Report the aim of the project was to realize a "Transformation .... to a modern, medium sized, efficient container port" with the "thrust of the project, .... emphasis firmly placed on rehabilitation and on increased efficiency in the use of existing facilities". All but the very last statement were executed more than satisfactorily. Of the five main risks identified at appraisal, only two became really significant. A third risk, the possible but undesired decrease in grain traffic, did materialize temporarily during years of substantial domestic production. The two risks that really did feature were firstly and mainly the long- term failure of traffic to increase as planned, and secondly the too slow growth in institutional benefits. In the assessment following below no attention is paid to the failure of traffic to grow as expected, as the main causes for that occurrence were the political developments in Southern Africa and the stagnation in regional economic development. The emphasis is therefore placed on the issue of institutional development performance. The project formally consisted of 7 parts, but these were actually concentrated in 3 major elements: Infrastructure (civil & building works, including consultancies), Equipment and Technical Assistance & Training (TA & T). If the well-known '3M' formula (Manpower, Methods and Means) is considered, which is a requirement for any production entity to perform satisfactorily, then the above elements constituted the Means (Infrastructure and Equipment) and the Manpower (TA & T), but not the Methods. It is the latter aspect which forges the others into making a success out of a project such as the one considered here. Represented by the System of Management, the Methods-aspect is the one that should make the other investments become sustainable and worthwhile. However, this aspect was never recognized as such at appraisal and was therefore never really targeted for a concentrated input. Though in general it can be concluded that the IDA-project identified at appraisal, as well as the overall integrated  20 Annex I project, both have performed satisfactorily, it is the Borrower's view that the crucial Methods-aspect was overlooked at the project's formative stage. It is noted here that inputs channelled along the route of TA & T may at first sight appear to be inputs feeding into the abovementioned 'System of Management' element. This is not sb, however, because experience in this project has shown that TA & T only produces personalized, ad-hoc and often uncoordinated outputs with primarily very localized results. What has been found missing is the existence of an Integrated Institutional Development Plan, to function as the reference point for developing an efficient and integrated System of Management. The TA & T inputs did indeed produce an improvement of the Manpower aspect, but did not produce a general "house style" or a new system of management which today can be recognized throughout THA. In view of the very competitive environment in which Dar es Salaam Port must now operate, there is increasingly a need to examine THA's institutional structure and its functioning in order to let it successfully meet the challenge of a modernized and commercialized port. It is recommended that THA engages professional external expertise, using the currently effective Credit, to execute a fundamental review of THA's management challenges and to produce an Integrated Institutional Development Plan. The primary objective of this exercise must be to put THA on a new and modern footing, doing away drastically with inefficient modes of operation in order to be able to survive as a major East African port. Summarizing the project scope, one needs to distinguish between three constituent parts. First, there was the scope covered by the Bank's Credit and the unplanned Swedish Grant; and second the expanded scope covered in the Staff Appraisal Report (SAR), which dealt with several additional co- financed subprojects. The third part however, was for further additional subprojects not covered by the SAR. Briefly examining the first two parts in retrospect, the actual costs exceeded the appraisal estimate by 41% and the project duration experienced a 61% time overrun. Placing the total project in the right perspective though, the size of the abovementioned additional projects is seen to have been such that the Bank's involvement, costwise, was limited to only 20% of the final actual project cost (at 1992 exchange rates). One cannot therefore assess the Bank's contribution in isolation. Particularly the financial consequences for THA are a point in case when considering that the original SAR project size was USD 91.0 M, whereas the final overall cost actually turned out to be more than twice as high (USD 189.4 M). The consequences for THA's forex repayment obligations to the Treasury are even more severe: the Principal Sum to be repaid is almost 3 times higher than envisaged at the outset (USD 163.4 M instead of USD 59.0 M). Further scale effects are present which were not covered by the lesser scope of the SAR project at the outset; for instance in the case of TA & T the USD 15.5 M actual cost compares with a figure of USD 4.0 M estimated at the time of appraisal. Returning to the project elements, a further distinction can be made between the 'hardware' type (Infrastructure and Equipment) and the 'software' type (TA & T and Management). Costwise the final overall proportions for these were respectively 92% and 8%, but with 0% input for a System of Management. Taking the performance of project elements one by one, Infrastructure (USD 124.1 M, or 66% of actual total cost) has performed satisfactorily. So has the Equipment element at USD 50.2 M (26% of total). The longer-term performance of TA & T (USD 15.8 M or 8%) is uncertain, but Management performnance was and still is unsatisfactory by modern standards. Since Dar es Salaam Port is having to accept a considerable downturn in traffic, THA will face great difficulty in meeting stiffer competition from neighboring ports if it continues to operate with the current institutional structure and methodology. The imbalance in project software inputs is clearly outlined above and should be redressed in the near future to support the high hardware inputs already implemented. Maior lessons learned  21 Annex I (i) For an integrated project of this size and complexity (approximately USD 320 M in two main phases, some 70-odd contractual relationships, and 12 financiers during a project span of 10 years) it is necessary to include in the project setup a fundamental review of the Borrower's institutional structure and management methods. The review should be carried out with the aim to strengthen as much as possible the Borrower's capacity ,A do its business successfully in the new environment created by the project. The review should be carried out by competent external advisors with experience in the field of third world port management. (ii) Initial overoptimism regarding the mobilization of local \inputs and a much increased project scope caused a virtual trebling of forex funding requirements, which has resulted in an excessive financial burden now resting on the Borrower. A revision is required of the financial conditions under which the Borrower is expected to operate successfully. BANK PERFORMANCE The Staff Appraisal Report was generally well executed and the risks properly anticipated. As already stated above, the important 'Methods' element was found to be considerably under-emphasized. This proved at the end of the project to have turned into a serious drawback for THA in moving ahead with strength and efficiency in an increasingly difficult business environment. Early attempts by the Bank to get the institutional issue off the ground unfortunately floundered after a couple of years. The Bank also pointed early on to the increased needs for project coordination, TA & T, management information systems and a port development study, but not specifically to an Integrated Institutional Development Plan. In retrospect it is a pity that the Bank did not pursue the follow-up of these issues more intently, even in the face of the Borrower's lack of active support, so that more benefits could have been reaped today. The financial scale effects were not foreseen of a more than twice as large project than anticipated at appraisal (USD 189.4 M instead of USD 91.0 M). The actual share of forex in the overall project costs proved to be a figure of 86% instead of 65 %, indicating overoptimism at appraisal for the capacity of the local economy to deliver really significant inputs. The forex increase is even greater than the overall project cost increase: final forex demand had grown to 280% of what was estimated at appraisal. Indications existed at the time that the project scope estimated in the SAR would not be sufficient. As it turned out the repayment of the almost trebled amount of forex, at a high interest rate, has severely depressed THA's financial performance. Added to this, the 11 % interest rate adopted for the Subsidiary Loan Agreement is far too high by current standards and is now seriously threatening the financial security of THA. The requirement for project coordination was properly recognized, but vastly underestimated. This is reflected by a current staffing level of 7 coordinators and a total input of USD 3.4 M, instead of the single coordinator planned originally for an input of approximately USD 0.4 M. The local situation and the complexity of the project has in this case proved to be much worse than envisaged at appraisal. The problematic shortage and mis-timing of coordinators was aggravated by the bilateral funding formula adopted for the staff. The tasks carried out by project coordinators expanded far beyond pure coordination within the Directorate to which the Project Coordination Unit (PCU) was attached. During the project's lifetime it became apparent that the Bank's head office procurement staff were becoming more instrumental in the decisionmaking process at that end. Though their professional expertise was certainly a help, it is the Borrower's distinct impression that the Bank is not aware enough  22 Annex I of the drawbacks this move has created. Principally, experience showed that Bank procurement staff are quite insensitive to the management requirements in the field. Bureaucratic (rigid and inflexible) application of procedural rules has at times resulted in considerable delays and even outright management problems in the field. Examples are for instance the considerable delay (approx. 6 months) in the early procurement stage of the second phase of the Container Terminal (covered by the next' Credit)r the procurement of Bandari College training equipment (a failed ICB round), the review of management information systems (so far not executed yet) and the manning of the project coordination unit (threatening discontinuity in services). All the abovementioned delays have caused and are causing time misalignments in the field. The benefits of strict adherence to procedural rules are out of proportion with the time lost in implementation in the field. In the SAR mention was made of the Entrance Channel not being an immediate priority at the time. In relation to the other rehabilitation needs that was a correct conclusion - at that time. However, at the end of the project's life cycle this situation had changed, and with the main rehabilitation having been implemented the need for modemizing the channel became more obvious and even pressing. The Borrower carried out a re-evaluation exercise which confirmed the economic attractiveness of the subproject, and subsequently arranged a financier and a dredging contractor to carry out the work. At a time when shipping lines were increasingly feeling the drawbacks of the narrow, twisting and shallow entrance channel the Bank effectively vetoed the project without presenting really significant arguments. It is felt that this decision has damaged the international image of Dar es Salaam port just when it needed a boost, but also that higher dredging prices will now unnecessarily reduce the project's economic returns when it finally is executed. With 12 financiers, some 70 contractual relationships and vetting of contract documents being carried out by THA, by the Tanzanian government authorities and by the financier's authorities, the amount of procedural paperwork became immnense. A large amount of this work could have been avoided if at the start a firm decision had been made to opt for one single standardized approach for all procurement and for all financiers. This would then have required some involved negotiations between the many parties, but those would definitely have paid off in the end. More timely implementation of subprojects would then have been possible, and more time available for coordination difficult matters such as TA & T. Maior Lessons Learned The following are the major lessons of experience derived from implementation of the project: (i) future large, integrated projects should be screened at an early stage for their financial sensitivity to possible additional subprojects not included in the basic project scope established at appraisal. This screening should be to determine effects on the repayment regime after the project is completed; (ii) future Credits be loaned direct to capable Implementing Agencies such as THA, and not via the Ministry of Finance; (iii) adopt a Special Account, enabling a more direct management of the credit funds. This approach has already been followed up for THA in the currently effective credit (no. 2095-0 TA). several percent above LIBOR) to alleviate the financial burden in the future seems more appropriate than the current fixed rate.  23 Annex I (v) For local economies experiencing the first stages of revival and liberalization, it is unrealistic to expect that even moderately high shares of local project costs can be executed successfully. Without local industry being able to contribute substantially to project inputs, the local share should initially remain in the 15 - 20% range. (vi) in aid of sustainability, more attention should be paid to having middle, senior and top management trained in modern management techniques - quite apart from the training in their own specializations. This training should also be institutionalized locally, enabling repetition of training and making it a formal management tool in itself. (vii) in future projects of this size a serious and detailed assessment has to be made at appraisal of the Borrower's real 'shop-floor' capability to coordinate the project. Only after this exercise can the TA demand be specified realistically. Project coordination TA for large and complex projects should never be delivered by bilateral financiers. A single consultant should be engaged for all coordination, preferably with financing from the leading donor. (viii) in order to avoid excessive growth of unscheduled PCU tasks periodical reviews with the Borrower's management should be carried out and appropriate measures taken. It is healthier to institute several smaller coordination units in appropriate directorates, than to have one large unit in one single directorate trying to attempt to manage everything at the same time. For top management a coordination unit at that level should seriously be considered, to connect with the other units at directorate level. All these units should have a clear task to transfer knowledge to local counterparts. (ix) adopt a single procurement procedure for all projects. The Bank did in fact do some work on scrutinising procurement documents for other financiers. Spend time at the start to get this one clearly sorted out, for which the Bank's formats and procedures could be taken as a model. Force all new financiers to adhere to this system. BORROWERS's PERFORMANCE During Design Stage Project preparation under Credit S-24-TA ensured a proper starting position for this Credit. Some of the statistical data provided at appraisal later proved rather unreliable, meaning that certain assumptions may have become invalidated. There is no evidence of this having happened though, and statistical data collection and processing have improved since with the aid of TA inputs. For the Grain Terminal a fundamental design change was introduced, away from a steel silo structure to a concrete one. This no doubt increased the facility's technical quality, but at the same time made it impossible to operate the Grain Terminal on a sound commercial basis due to the much higher costs. The commercial soundness was already very marginal in the steel case, so this change only made matters worse. The facility therefore could only function effectively as a national storage facility for emergency food relief use. In practice no recognition was given to this aspect, thus burdening THA financially with a commercially non-viable product. During Implementation As stated above, the prime objectives of the project were achieved. However, as a consequence of much higher forex demands than envisaged at appraisal (details given elsewhere) the profitability of THA dropped considerably towards the end of the project. This now presents a major risk for THA and for the Tanzanian government, which relies heavily on the financial performance of THA to provide it with forex revenue.  24 Annex I The overall time overrun of the project was 61 %, primarily caused by belated disbursements for Equipment (6th Rubber-Tyred Gantry crane) and TA & T. The latter still continues on into the next Credit and the former was a late decision after a reallocation move. This concerned the switch from procuring backup generators (which FINNIDA then funded) to procuring a 6th RTG instead. All hardware costs have been exceeded well beyond reasonable margins (arbitrarily chosen at i 20 %), except those for the Container Terminal Ship-to-Shore Gantry cranes and the Rail-Mounted Gantry crane. The largest absolute and relative increase concerns the very much higher input of Other Container Terminal Equipment, i.e. for 5 nos. Rubber-Tyred Gantry cranes and for transfer equipment for in-port and off-port use. An increased scope in Navigation and Communication Equipment also resulted in high cost increases. These investments were well justified priority choices and were partly offset against lags in expenditures for General Cargo Equipment and Spare Parts. The next-largest increase in costs was incurred for the urgent Rehabilitation of Kurasini Oil Jetty. This was the result of an increased scope of work, and heavy delays (208 % time overrun) caused by work stoppages due to berthed vessels utilizing the much-used jetty berth. In absolute terms the largest Civil Works cost increase arose from the Container Terninal and Ubungo Inland Container Depot subproject, though percentage-wise the excess was moderate. Prime causes for the cost increases were threefold: firstly, the forex content in the contract was 74% as against an anticipated 52 % at appraisal. Secondly, the inordinate and unforeseen shift in exchange rates between the sharply rising value of Yen (a Japanese contractor carried out the works) and all other currencies by a factor of approximately 4.5. Thirdly, the 46% time overrun resulting from delayed handovers of work areas. The forex shortfall in the IDA Credit was gratefully compensated by a Swedish Grant worth 15 % of final project costs. The quality of the delivered hardware elements was generally good, save for two specific cases and a number of minor defects. One specific lack in performance was the delivery of a number of cranes which contained too many prototype components instead of plain, tested-out, production-line components (the first set of RTG's for the Container Terminal). The other lack in performance was the localized failure of Container Terminal pavement, caused partly by heavy unscheduled overloading and partly by an inadequate design which did not allow for proper drainage of groundwater in the substrata. Only one subproject was cancelled: the Lighter Wharf Rehabilitation. This was a straight result of the selected contractor failing to provide an acceptable performance bond. The project was subsequently abandoned as the forex was needed for the Container Terminal project instead. One instance of failed ICB bidding occurred with training equipment for Bandari College. After a revision of the tender instructions (a relaxation of tight conditions) the procurement was successfully executed in a next round. The SAR assessment for TA & T indicated anticipated disbursements for both TA and Training, but only the TA part was actually programmed explicitly at the time. Evaluating final outcomes, it is seen that no disbursements were actually made for Training under this Credit (but this was covered in the next one) and TA disbursements took quite a different course from that expected at appraisal. Out of the 252 TA manmonths planned, orly one TA assignment (24 marnonths) was more or less carried out as originally planned. Several reasons can be quoted for this deviation, the most significant one being the takeover by other donors of whole 'blocks' of TA. Of the six original blocks five were transferred to other donors, while the 6th one is still due (to be executed under the next Credit). As it turned out, the Bank's TA funds were in practice utilized for filling shortfalls arising in TA programs funded by  25 Annex I other donors. This kind of 'reverse' takeovers amounted to 65 % (131 mr/m) of the Bank's final TA input. The rest was for the as-planned TA, which was extended by another 46 rn/m. The most significant "mis-judgement" at appraisal becomes evident if one realizes that eventually over 3000 manmonths of TA has been -delivered for this integrated project. An indication of botllthe Bank's and the Borrower's inability to cope properly with TA is the subsequent mis-judgement concerning factual TA requirements, which as late as 1989 (5 years after the project took off) erroneously stood at approx. 1000 manmonths while then in actual fact being 100% higher. A further case in point is the so far non-mobilized Head Office TA, even though preparations for this package started as far back as 1986. In those days the PCU lacked the manpower to cope with the task of managing TA, but today basic questions have risen about the form in which this TA must be implemented. Such a decision problem is symptomatic for the way TA is managed in general. In order to place the TA question in the right context one should distinguish between different types of TA: (i) PURE TA & T, operating with local counterparts and aimed at providing long-term sustainability; (ii) ADVISORY TA, assisting local staff with their decisionmaking, but without a dedicated training obligation; (iii) LINE TA, fulfilling line management duties to compensate for shortages in the local establishment's staff; also without a dedicated training obligation. Of the abovementioned types of TA, only the first kind is structurally aimed at sustainability. This type formed the vast majority of TA assignments, generally implemented at the lower levels in the Borrower's hierarchy. The other two types of TA appeared at higher levels within the Borrower's organization and did of course include some element of training and knowledge transfer, but generally not in a pre-planned way. Those types were therefore not monitored for any performance in that sense and as a consequence the sustainabilty at the more senior levels showed considerable gaps in performance. A frequent omission at the start of many TA assignments for the integrated project was to miss out on specifying very clearly the outputs expected from each of these forms of TA. Without clear targets outputs could not be monitored and evaluated properly, thus leading inherently to underperformance. A further contribution to underperformance followed from the non-delivery, delayed startup and often low quality of local counterpart staff. Added to this came the unavoidable phenomenon of local staff, once trained, leaving the Borrower's establishment for more lucrative jobs in the private sector. Management of TA & T remained weak on all sides. Generally less priority was given to TA & T than to hardware (engineering) matters. This is a direct consequence of placing the PCU in the Engineering Directorate and assigning the task of monitoring TA & T to it, instead of to a sister unit in the Manpower Directorate (this situation is now rectified). The evaluation methodology remained ineffective until very late in the project's life, and even then it is not assured that monitoring and control is really effective. Institutionalized training of local staff was limited to the lower echelons. No real training at the higher levels occurred under this Credit, but was taken care of in the next one. However, that programme is a temporary one and a permanent management training feature still needs to be incorporated into the Bandari College.  26 Annex I A consistent lack in effective management performance by the Borrower has been the slow to very slow decisionmaking process within its organizations, i.e., both within central government and THA. The Bank repeatedly addressed this subject, but real reforms were not instituted or felt. Apart from many examples of slow procurement, a further case was the vital exercise of asset revaluation, which took more than 3 years to get organized and executed. The Subsidiary Loan Agreement took quite'a number of- years before it was signed, well beyond the terms of the Credit Agreement. A positive example of management's performance was the very laborious process of reducing container dwell time in the face of many obstructions outside THA. In this field a review of THA's institutional structure and methodology is still very much needed to, among other matters, greatly improve the speed of decisionmaking. The Borrower's purchasing and supply system increasingly lagged behind the requirements of a modern port, thus inhibiting efficient performance. As with other systems in the establishment, this system needs to be fundamentally reviewed and overhauled to give real customer service to port staff. Ship Waiting Times and Average Berths Times have improved very considerably, resulting in average waiting times less than one tidal cycle per vessel. When considering that night-time navigation is still not possible this performance is very satisfactory. The cargo handling rates also improved in a major sense, though container handling was expected to be at higher rates. The SSG-RTG transfer cycle at the Container Terminal lags behind in this respect, in combination with the RTG-offport delivery cycle. It should be noted that no lighterage operations have been carried out during the project's lifetime, as the quayside cargo handling improved sufficiently to avoid this. For both imports and exports of non-petroleum traffic the appraisal estimates proved higher than in fact realized. Both types of estimates were generally too high by approx. 10-20%, indicating a systematic error of fairly minor importance. Two major deviations occurred, for imports in 1990 and 1991. In both cases the deteriorated economic situation in Zambia and ZBR caused traffic drops to occur of some 40%. In fact the trend continued in 1992, but was compensated that year by the high inflows of relief foodstuffs to offset the drought. A more detailed analysis shows that transit traffic to and from Malawi grew significantly and that Zambian copper export lagged consistently behind expectations. Ugandan traffic in fact reduced as Kenya managed to claim more of that transit traffic. Domestic Tanzanian food imports were reduced very considerably due to increased home production, as well as imports for oils & greases and metal sheets & pipes for the same reason. On the export side a shift occurred away from Tanzanian agricultural products to more non-traditional (new) commodities listed under "other exports", and less cotton products were exported because more were processed in the local textile industries. In relative terms the mode of handling cargo in the form of containers remained in 1992 as anticipated at appraisal (40%). In the same year a shift towards an increased dry bulk share occurred as a result of the regional drought situation. General cargo/break bulk consequently lagged on expectations, but in 1988 there was no dry bulk at all, meaning the general cargo handling mode met the appraised figure. Major lessons learned (i) Implement a review of the THA institution, with the aim of making the whole organization managerially much more effective than it is now.  27 Annex I (ii) Future Loans to be made directly with the Implementing Agency such as THA and not through governrnent institutions, because this will very considerably improve management performance of the Borrower. (iii) No equipment prototypes to be tested out at the expense of Clients in develcaing countries. Donors should beware of encouraging suppliers in their countries to do so, passively or actively. (iv) In order to obtain maximum sustainabilty, much better and stronger identification and monitoring of TA & T is needed. This problem is, however, one of the most persistent and difficult problems in the field of development assistance. Really workable and successful solutions have so far proven to be singularly elusive, even on a world-wide basis. One recommendation that can be made from the experience in this project is to insist on the TA expert to step back completely out of line duties during the last 25 % of his/her assignment, leaving all the daily work to be done by the local counterpart. Since the TA expert will still be around, the local counterpart can thus be assisted while having being pushed into "the deep end of the swimming-pool". (v) Avoid repetition of a major training omission: not having organized a permanent training facility for middle, senior and top management in port management disciplines. (vi) Instill much more of a "customer service" attitude throughout the Borrower's organization, which now is not present and an inhibition to fast and efficient working of management. For example, overhaul the purchasing & supply system to become really user-orientated instead of head-office dictated. (vii) Since sustainability is a prime target in development assistance, it is proposed to selectively adopt a different kind of approach towards the use of consultants. Since the Bank generally allows a preference for local firms, the hybrid solution suggested below may be of interest in future. Since the preference given is only in quantitative terms (a higher bidding cost being accepted) and not in qualitative terms, local consultants always face imminent defeat when bidding for new contracts. If however, the Bank (possibly other donors may follow suit) accepts local consultants to be temporarily infused with foreign TA during the duration of the project, then these firms stand a much greater chance to compete successfully. In this way a much greater degree of (national) sustainability is created. Also, this arrangement is much cheaper than full-scale foreign consultants 'imported' for the occasion at much greater expense. (viii) Not being a major issue, but still worth mentioning, is the use of an "Expenditures" overview in the reporting to the Bank. This reporting requirement was faithfully met, but what may seriously be questioned is whether the overview was of any practical use at all. Producing it each time required quite an effort, but the data were meaningless as the currency employed was TSh instead of USD. This meant that the constantly sliding value of the local currency (over a 8 year period) resulted in 1985 TSh's and 1992 TSh's being mixed together in the same table. This in fact debased the expenditures figures and made any meaningful comparisons impossible. If overviews such as these are to be really meaningful, the record should from the start be maintained in USD or SDR.  28 Annex I RELATIONSHIP BANK-BORROWER Considering the 12-year life span of this project, the Bank-Borrower relationship has always been very good. A common understanding always existed of the problems encountered and also with regard to establishing a joint approach to solving-them. The later idea for involving a locally based Bank advisor was welcomed and appropriate. The setup worked in principle, but proved less effective in this case than hoped for by both sides. Causes: the concept arrived late in the project's life, i.e. long after the Borrower had established his own mode of project management, but also the advisor was less readily available due to heavy involvement with another (much larger) major project. During the long lifetime of this project only one significant and unfortunate shadow was cast over the good working relationship. The Borrower still fails to comprehend and regrets the Bank's actions to postpone the implementation of the Entrance Channel Improvement: a project with an attractive financial and operational return, for which a donor and supplier was successfully lined up by the Borrower. PERFORMANCE OF CO-FINANCIERS Denmark (DANIDA) financed the two Ship-to-Shore Gantry cranes and the Rail-Mounted Gantry crane for the Container Terminal. This subproject was successfully implemented, though certain delays occurred due to trouble with the supplier. Finland (FINNIDA) executed a large TA & T project (still ongoing) and supplied Rubber-Tyred Gantry cranes for the Container Terminal as well as terminal and highway tractors & trailers. At a later stage FINNIDA also financed a much-needed backup generator power station for the same Container Terminal, thus ensuring continuity in the port's container handling capability. All projects were successful, with the usual reservations placed on sustainability of TA & T. The concerted Finnish effort regarding the latter subject has continuously been a guiding example for others during the whole duration of the integrated project. The Netherlands funded the provision of the bulk Grain Terminal, which proved not to be immediately necessary in an operational sense after it was completed. This was simply due to the fact that Tanzania was successful in producing enough domestic grains at the time to prevent importation, but not enough to generate exports. Since then the Grain Terminal has been successfully used to store relief grains for alleviating the severe regional droughts. Norway (NORAD) was instrumental in successfully maintaining the operative status of the strategically essential Kurasini Oil Jetty, which today is still the most heavily occupied berth of Dar es Salaam port. NORAD also put a great deal of effort into the technical and institutional upgrading of Bandari College, the port's training institution. Due to the very large gap which originally existed in the institution's capacity at the start of the programnne, the final results are now still uncertain despite the considerable effort to close the gap. This TA & T program suffered more than the usual share of implementation problems, which has had the positive result of highlighting where attention should be paid in future. In addition to the above, NORAD also co-financed the extensive rehabilitation of the general cargo Berths 1-8, and also financed the creation of a new Dhow Wharf and the rehabilitation of offshore structures for navigation purposes.  29 Annex I Italy financed a considerable quantity of port terminal, navigational and communication equipment, which was all put to good use upon receipt. The Bank funded the new Container Terminal and the conversion of two Inland Container Depots, as well as equipment for the Container Terminal- (6th RTG), Bandari College and spare parts fof the SSG's. Further funding involved various TA & T and consultancies for civil works as well as institutional affairs. Sweden (SIDA) provided considerable funding for the successful rehabilitation of Berths 1-8, copper handling equipment, and TA & T for general cargo operations and for the PCU. In addition to this SIDA provided funds to fill the large shortfall created by the spectacular rise in the Yen exchange rate, which was linked to the Container Terminal project. The United Kingdom (ODA) continued with a longstanding programme of TA to the marine staff in the port, to which it added the first TA input for the PCU called for under this Credit and general cargo equipment. At a later stage ODA assisted with the successful rehabilitation of many mobile cranes in the general cargo section. Canada (CIDA) also continued successfully with the funding of a longstanding TA and spare parts supply programme for the general cargo section's central workshop. Australia (AIDAB) provided for a considerable supply of general cargo equipment, which was successfully employed. Sweden, Finland, The Netherlands, the United Kingdom and the Bank all contributed towards maintaining the Project Coordination Unit, which was an essential tool for implementing the whole integrated project.