Tanzania - Morogoro Textile Project

Document of The Wod Bank FOR OFFICIAL USE ONLY ReportNo. 8696 PROJECT PERFORMANCE AUDIT REPORT TANZANIA MOROGORO TEXTILE PROJECT (LOAN 1607/CREDIT 833-TA) MAY 31, 1990 Operations Evaluation Department This document has a restricted distribution and nmay be used by recipients only la the performance of their offivial duties. Its contents may not otherwise be disclosed witheat World Bank authorization. ABBREVIATIONS AND ACRONYMS DFI - Development Finance Institution EEC - European Economic Community ERP - Economic Recovery Program LM - Linear Meter MAF - Management Agency Firm MWATEX - Mwanza Textiles Limited NBC - National Bank of Commerce NDC - National Development Corporation PAF - Project Advisory Firm PCR - Project Completion Report PEF - Project Engineering Firm PIU - Project Implementation Unit Polytex/Company - Morogoro Polyester Textiles Limited PPAR - Project Performance Audit Report PR - President's Report RTC - Regional Trading Company TDFL - Tanganyika Development Finance Company, Ltd. TEXCO - National Textile Corporation TIB - Tanzania Investment Bank CURRENCY EQUIVALENTS Currency Unit Tanzanian Shilling (TSh) - 100 cents FY78: 1US$ m TSh 7.7 FY79: 1US$ TSh 8.2 FY80: 1US$ TSh 8.2 FY81: 1US$ = TSh 8.3 FY82: 1US$ = TSh 9.3 FY83: 1US$ = TSh 11.1 FY84: 1US$ - TSh 15.3 FY85: 1US$ TSh 17.5 FY86: 1US$ - TSh 32.7 FY87: 1US$ TSh 64.3 FY88: 1US$ = TSh 99.3 FY89: 1US$ - TSh 143.4 Rate at appraisal = TSh 8.3 WEIGHTS AND MEASURES (All weights and measures are in metric units) 1 metric ton (t) 1,000 kilograms (kg) 1 metric ton (t). 2,204.6 pounds 1 kilometer (Km) = 0.62 miles FISCAL YEAR Government: July 1 - June 30 TEXCO and Polytex: Janiary 1 - December 31 FOR OFFICIAL USE ONLY Tn WORLD SANK Wasington. DlCi 20433 1.S A ONKr A DuectWGePAl OpeaEnmns Evuatnia May 31, 1990 =AM TO TiB ECUTIVE DIRECTORS AND THE PRESIDENT SUBJECTs Project Performance Audit Report on Tanzania Moroxoro Textile Prolect (Loan 1607 /Credit 833-TA) Attac ed, for info pation, is a copy of a report entitled "Project Performance Audit Report on Tanzania - Morogoro Textile Project (Loan 1607/ Credit 833-TA)" prepared by the Operations Evaluation Department. Attachment This document Ma restricted distribution and maay be used by recipients only in the performance of their official duties. Its contents naay not Otherwise be disclosed without World Bank authoriation. - NIR OFéAL USE ONLY RJECT URN UME= (LOAN 1607/CREDIT 833-TA) TAILE OF CONTENTS PREACE ........................... BASIC DATA SET ......................... .. ......... . iii EVAIJATION SUAY *................................................. V The Industrial Setting . . ............ The ýextilel6 Sector ............................. 4 Objectiv*s ............ . .. ............ .......... 4 II. PROJECT DESIGN AND SCOPE ............ .................. 5 Project Configuration 4........... .............,.. 5 Technology ........ ... ..... ....... ........ ........ -:: : : 9 Location and EnvironmUnt4 Aspcts ................... 9 Changen In Scope ........................ ............... 9 II.KMPEMETTON EEERENE ........................... 10 Project anagement ......................... 10 Policy Interventions . ...... ................ . 11 Role of the an ........................................ 11 iv. 12 WF a y e . . . ................................... 12 Operatging Performance ........ .. ......... ..... 12 Finandtal and Economic Performance l Sustainabi11t ..... 12 Institutional Development ................................ 12 Sectoral Impact ................................. ... .. 13 V. OVERALL ASSESSENT AND O1TANDING ISSUs ................. 13 Overall Assessment ............... ........ 13 Outstanding PoUcy Issues Requiring Initlatives by Tanzanian Authorities ...... ................ 14 VI. LESSONS OF EEPERIENCE ....................... ............. 15 This document has a restricted distribution and my be ubed bP recipients only im the eformance of thei official dutis. Is o ts may mot odewis dibied without Wr a dm TABIJ! OF CONTENTS (catit PROJECT CONPIETION REPORT . ITRODUCTION . ...................... ............. 21 The Economy .................................. .........21 The Ipdugtrial Sector ....... '.................. 22 The Textila Subsector ...... ............................ 22 II. ROJECT BACKGROUD................................. ...... 23 Origin and Evolution of the Project'...................... 23 Project Description and Objectives ...........\......... 25 III. PROJECT IATION AND fANACEMEN .................... 25 \ V. OPERATING PrFORNANCE .............................. .32 V. INACIAL PEREHRMNCE ..................................... 34 Financial Resultp ............ ....................... 34 VI. ECONOMIG PEM E ...................................... 38 Economic Rate of Return .............................. 3é Foreign Exchange ffects ..................... ...... .... 38 VII. INSTITUTIONAL PERFORMANCE ................................. 38 VIII. CONCLUSIONS ............................................... 39 Lessons'to be LearnÀd ....................................39 ANNEEES 1. tchedule of Project Implementation ........................ 41 II. Schedule of -Disbursements ................................. 42 III. Polytx: Projected and Aòtual Incowe Statements, iFY82-90 44 IV. PolyteÉ; Projected and Actual alance Sheets, F782-90 46 V. Financial Rate og Return .....................;............. 48 VI. Economlc Rate- ofe*in -.- ~ -- .-.49 - ATTACEMMT Co=ments Received from tilytex ........ 51 \. . . . . . . . . j ~ I PUROJCT A UDITRT K0R000R0 TEXTILI PROJECT (LOAN 1007/CREDIT 833-TA) This Is a Project Performance Audit Report (PPAR) on the Norogoro Textile Project, supported by Loan .1607 ini\the amount of US$25 milion and Credit 833 in the amount of US$20 million. Additional funding for the US$110 million project was provided through suppliers' credits from India, Japan, the Un4ted States, and the Tansaia Investment Bank (TIB).\ The loantcredit were approved in June 1978, became\ effective in May 1979, asd were closed in June 1987, two years after the original date. The PPAR vas prepared by the Operations Evaluation Department (OED), while the Project Comple*ion Report (PCR) by the Africa Regional Office -of the Bank, with significant technical input from the National Textile Corporation (TBUCO), Horogoro Polyester Textiles Limited (iolytex), and, the Technical Department of the Burope, Middle East and North Africa Region.- The PPAR i based on the attached PCR, Staff Appraisal and the President's Report, sector and economic reporta, OED's study "World Bank/Tanzania Relations, 1961-1987," (chapter on Tanzania's Industrialization Effort 1961-87), the loan documentsi eummary of the Boord discussions, study of the project files and discusions with Bak\staf. The OED mission discussed the effectiveness of the Bank's assitance with TEXCO and Polytex, Government offi4ale, and the banking and business community.. Their kind cooperation and ialuable assistance in the preparation of this report is gratefully acknowledged. The \PCR ably discusses the project experience with regard to the orig n, evolution, implementation, management, operations, d financial, economic and institutional performance, and\ draws some of the lessons learned. The PPAR focuses on the framework 'of the country"s industrial strategy and policies at appraisal, the project design and scope, the quality of the Bank's appraisal work ad effectiveness of supervision effort, the outcome of the project and the \act of the' policy aviroment o4 project performance, the extent of the Baik's p licf and a toral interven ions, ostatadding policy iseses i'apacting on the project'a viability which require Governmet\ initiatives, and the project's susiainability. The PPAR then draws additional lessons from- the project experience, and makes recommends- tions for potentially more efative courses of action ts foster the evelop- ment of sustainable industrial projects. Copies of the drait PPAR were sent to the .Government and the enter prises invoved for their review and comments. 961 tex's comments are repro- Ueed as asttacbment to the PPAR. (LOAN 1607/CREDIT 833-TA) nssc DATA sHlET- wA/CEDITPOITIONON (AmIounts In US$ Nillion) An of nat. 31. 1990 Orx a Di d Ca lld Rai Iman 107-TA 25.0 25.0 0.0 10.9 14.1 Credit 833-TA .20.0 20.0 0.0 0.3 19.7 CUMUZATIVE ESTIATE ADAAL DIURSEII D D ~ppa~8sal atlmt (U8$ M) 2.3 14.3 S*,6 44.3 44..5 -44.7 45.0 43.0 45.0 45.0 45.0 ,tm(i $ (M m 0.0 0.0 -0.0 14.4 .14.4 22.3 25.9 35.6 38.4 44.5 45.0 s .of Appsa4 (3> 0.0 0.0 0.0 32.4 32.3 49.9 57.4 79. 85.3 99.0 100.0 of . 2Ia& 0U$sigtag 1ebuunay 16, 1988 Start ot\ rreparation n.a. 0615/77 Appraisal. n.a. 05/20/78 Board Approval n.a.. 06/29/78 Loan/CredIt Signing n.a. 07/28/78 Ian/Credit Effectiveness . 11/28/78 05/07/79 Ioan/Credit Closing . 06/30/85 05/30/87 Completion Date: Plant 12/31/81 12/31/85 Nanagement 06/30/84 06/30/88 Coipletion Time (o:bs) 44 85 Fixed Coat (US$ N) 80.2 JR 107.4 ta Cost Overrun (%> -- 34 Total Financing (US$ N) 106.9 110,6 Financial Rate gf Return () 18.9 -11.3 Economie Rate gt Return (t) 19.2 -18.1 / Excludes vorking capital and technical assistance to TEKCO. Preappataal 21.s 34.6 -, - - -ma - - 5. 34.9 -34.9 Neoitos- 10.4 - - - - - - - - - 10.4 ra 21.5ý 80.0 32.2 12.7 1s.8 1s.2\ 1o.8 .7. 2s.s 1s.3 8.8 9.s 20.s \ ------ -£ gtafIf DaATA - No. of ooftaf Dtaof Identification/\ Pre-appraisal '406/77 3.6 5 18.0 08/02--77 Ippraisal 10/7,7 1.6 2 3.2 11/30/77 Supervision 1 08/78 1.4 2 2.8s 10/05/78 Supervision I, 05/79 2.8 2 . 5.6 06/15/79 Supervisoxi I1I 1281 1.4 • 2 2.8 12/10/81 Supervision IV 02/82 1.4 1 1. \ 04/09 Supervision V 1/82 2.4 2 4.8 12/06/8 Supervision VI 06/83 1.4 1 1.4 07/07/83 Supervision VII 02/84 1.4 1 1.4 03/27/84 Supervision VIII 03/84 1.6 2 3.2 08/21/84 Supervision IX. 11/84 1.8 2 2. 12/106/84 Supervision I. 07/85 1.4 1. 1.4 07/31/85 Supervision XI 05/86 1.4 2 2.8 06/16/86 Supervision XII 01/87 1.4 2 2.8 02/26/87 Completion 09/87 1.4 1 1.4 04/10/89 OTH1ER MRJET AT Borx er: 'Govéiinmnt of Tanzania\ Executing Agency: Korogorø Polytex/TECO Flo-ót) Projfacts: Non\ PROECT PERfOANCE AUDIT RPORT MOROGORO TEXTIL PROJECT (LOAN 1607/CREDIT 833-TA) EVALUATION SUtafARY\ Introduction L* 'A centrally directed strategy of 'industrial development and policy framework "a adopted in earnest by Tanzania's political leadership during Vthp late 1960s to advance the objectives of self-reliance, -economic growth danl structural transformation. Industry wae seen as a powerful engine of growth that \could modernize the economy through more capital-intensive, higher-product4mity processes, promoted within a protected environment. The Government' firdly 'believed that a vigorously pusued industrialization strategy with emphaois of resourc4-based, producer goods indust ies that cater to basic needs, 'dubbed as the basic industry strategy, coupled with an extensive state \nvolvemet in the econigmy, would accelerate growth and, achieve a more equitable sodio-econbmI development. The Bank endorsed this approach to industrial dpvelopment frm its inception, and even became its staunch supporter throuihout the 1970s (PPAR, paras. 1-8). Objectives it. Witiin the ontext of the basic industry strategy, the project under review ipvolved the establishiient of an integrated mill with spinning, weaving ahd finishing -acilities, capable of poducing 21.5 million square meters of fabric annually, under 'a new -state-owned company -- Morogoro Polyester Textiles Limited .(Polytex). This imprt shbstituting project wai originally designed to meet the domestic demand for blended fabrics, but it' was modified subsequently to produce all cotkon yarn and fabrics (PPAR, para. 17). In addiion, the project waW to extend financial assistance to the National Textile .rporati n (TXCO), the state-owned holding company, to carry out, and eventually implement, a study that would recommend measures to improve the productivity and dapacity utilization of the existing textile mills in the TEXCO group which faced serious technical, management, manpower, "yd financial problems (PPAR, para. 9). Taplementation Experience II. In view of the c itical sortage o foreign exchange, reinforced by the increasig cost of Iported raw materiales,plant facilities were modified at -a nominal cost (US$0.4 million) - to make possible the production of \all cotton yarns and fabrics. This suggeste that the deep-seated problems of the economy in th late 1970s had t been fathomed at appTaisal, a situation which. wae exacerbted Ty U 4eMc1 of ettlillIy strnaiH to -at recurr nt foreign exchange requirements once projects were completed. A second Important modification of' the project scope after appraisal ,as the propos; to install at a mod"st t cremeagal cost (US$1.7 million) looms capable of producing wider fabrics to satisfy the requirements of the local garment manufacturers. In the circumestances, hoth changes were consequential, and the Bank (and TEXCO) deserve credit for the decision taken. Finally, the Bank agreed, rightly so, to finance the foreign exchange and part of the local cost of constructing 60 housing units for local staff because, contrary to earlier expectations, they had no access tp financing (PPAR, para. 17). iv. The set up for the management of the project was appropriate, the selected engineering and advisory firms were experienced, and their perfor,- mance on the whole was satisfactory* Although the Implementation schedule was realistic, an array of events, such as political intervention In the appointment of the civil works contractor, difficulties in the evaluation and award of contracts, problems with project mazagement, disputes between the project 'pgineering firm and TEXCO over contract negotiations with suppliers and contr ctora, high turnover of phe staff of the project Implementation unit, belated projgect commission dud to \late delivery of raw mateialU and inadequate infrastructure, resulted in substantial delays In project so- mentation -- 41 months. It is noteworthy, however, that timely completion would not \have necessarily ensured immediate start-up of opetations due to delayh in completion of the requisite infrastructure. Cost overrune amounted to 341 due -to higher costs of civil works, machinery and equipment, and engi- neering services, stemming from changes In project scopa, delays in implemen- tation, inflation, and underestimation of costs (PPAR, para. 18). Dutcom v. Polytex's operating performance to date has been very disappointing. largely for reasons beyond\ the management's control. Capacity utilization was only 17% in 1986, 27% in 1987 and 10% in 1988, despite the fact that the plant produces all cotton fabrics and relies exclusively on domestic sources for its. rr material requirements. Polytex's oor performance is, attribut- able, to week demand for Polytex' s line of PV= ts, high costs of* production and inability to cnip -to with (legally and illegally) Imported textiles, lack of foreign exd*ange to import essential chemicals and spares, and staffing problems (e.g., high turnover, absenteeism, lack of motivation, limited problem-solving capability, inadequate remuteration). Under Tansania's dire economic circumstances, the effectiveness of Polytex's management team under- standably is highly conAtrained (PPAR, Oara. 29). vi. The financial performance of Polytex has been extremely poor. Due to very low capacity utilization, price controls affecting some 401 of out- put, and pressure from Imports, accumulated losses exceed TS4 5 billion, and Polytex is not in a position to service its debt which keeps rising bydause of the continued devaluations. Recalculated financiaX and economic return. araegative compared t an estaat 191st appraisla - -The st ability of te\'project is problematic, as it is heavily conditioned on building Lp production at'. respectable levels, resolving the issue of the denominated in foreign currencies debt (PPAR, para. 28),. carying out a financial restrqc- \turing, 'continued emplo" of Postly expatriate management to fill 1ey positions, acceess, t\workkn\capital (PPAR, para. 30), and government ac ione to resolve important indust*ial, trade, and financial policy isues (PPAR, para. 22) - a tall order. vii. Polytex's institutional development (e.g., produatiop management systems, marketing, human resource development) have yet to be fully develb oped. It is disconcerting that a mill capable of producing superior quality\ fabrics turns out run-of-the-mill itoms, a melange quite differen from those originally envisaged. Also, while a comprheive training pr4gram was put in place early o and Implemented, Polytex has been unable to retain most of the original trainees, particularly those who were'tralned abroad. This has frustrated expectations to develop indigenous staff capable of replacing expatriate management and, as a result, Polytex's reliance on expatriates to fill key technical and managerial positione. persista (PPAR, para. 23). viii. The study of the productivity and capacity utilization of TEXCO's mills was completed on time (December 1978), and Its recommendations were partially implemented. Vowever, since the early 1980s mill performance deteriorated rapidly, intet al1a, due to shortages of water, electricity, fuel, and foreign exchange for procurement of dyes, chemicals, spare parts, acce4qories,- etc., which were beyond management control. As a result, t e impact of the actions taken cannot be assessed due to the disruptive effects - of these factors 4hich severely influenced individua mill performance. Nonetheless, having to c,oncentra ,e its efforts on pressIng issues ateming from the implementation of the integrated mill, the Bank did not monitor closely the implementation and effectiveness of the technical assistance program, and only half of the earparked funds were ultimately disbursed. Generally, the Bank has not been able to influence the pace of growth, reorientation and institutional development of the textile parastatals, albeit in part due to distortive economic policies and lack of receptivity Wy, the autho ities and TE%CO (PPAR, paras. 19, 24). Overall Assessment ix. Two pivotal factors underly the design of the project under rev ew and, by extension the development of the textile enbsector as a whole. First, demand projediqn for woven textiles turned out to be ovqrly optimis- tic, as consumption of\ textile fabrics has groin at a substantially lower rai than projected at appraisal. Secondo binding operational constrains permesting the - Lubsector, though evident and .-cknowledged, did not elicit sober thinking both in the Government and the Bank. Thus, the Bank, as early as 1974, had impressed on the Government the' need for undertaking a 'market and marketing study to establish the 4ong-tern potential for cotton fabrice, including export possibilities, before embarking on another project. Also, because of resource constraints, shortage of trained technical staff, low productivity, and high production costs of existing mills which presented serious obstacles to launching a major export drive, it was suggested ihat \'the Government would be well advised to develop the textile industry at a more realistic pace. The warning was not heeded. Ip the ensuing years, and without con#ulting with\ the Bank, in addition to the Bank-sponsored MorOro textile mill : iee oThe cotfon texti rofts ve uadertaken bythfei with no apparent regard to the excess capacity being built up and the Impact of the crippling constraints on operating efficiency. At .the appialsal of the project under review, the Bank fel that a loan evenant to ensure con- sultation on sectoral development was sqpractical, because of the political sensitity shrin 'y the Government on the issue (PPAR, para. 10, 11). . a#k appraisale (and the Govepament) 4A not consider seriously the alternative 'of rehabilitatiS and expaAding the-Jour older m4le in TEXCO? group before proposing the\iatablisment of a new plant. - The ikore so, etnce by the mid-1970s it was quite. evident that the economic situation was deteriorating, that the difficulties o the textile sector were compounding, and the Bank was so concerned with developments in the textile sector, low capacity utilization and productivity in particular, that a dated covenant was included in the Moragoro project obligating TEXCO to carry out a study of the productivity and capacity utilization of all its operating mills, and see to the implementation of its findings. A textile sector rehabilitation pro- gram loan addressing the sector's deep-seated problems could have saved eub- stantially on capital Investment and foreign exchange, made the existing mills more competitive, and avoided the creation of excess capacity. It would seem that the entertained strong belief in the need for constructing an integrated textile plant to meet the assumed large pent-4p lool demand for blended fabrics, albeit heavily based a imported raw materials and at a time when the country was facing acute balance of payments problems; and the Bank's understandable disinclination to launch an involved rehabilitation operation, preference for a tidier and less staff-intensive green field activity, and greater convenience in financing studies, and relying on the sponsor agengy for implementation, were major determining factors of the course of action taken (PPAR, para. 12). xi., Given the insignificant scale economies in textile mills beyond the relatively low minimum economic size and the severe constraints in human skills and infrastructure, passing consideration was given during appraisal to building a smaller plant with provision for phased expansion. Rstab- lishing a plant close to the amnimum economic size would have been a sounder decision in the Tanzanian citcumstances as it would have avoided straining scarce resources, would have made the management of the plant more wieldy,' and would still attain satisfactory economic and financial retuins. Apparently, confidence in market projections and conviction that the Additional capacity was needed outweighed all other consideratio4o and options, including potential operating difficulties (PPAR, para. 13). xii. Since tho country was already facing acute foreign exchange problems, Bank missions as early as 1971 had stressed the need for export promotion, *anzania was not subjected to quota restrictions or Import duties when exporting to the BBC )market, and the objective was to de elop a viable resource-based industry, it is striking that the Banks, in sponsoring a capital-intensive and heavily raw material import-dependent project, did not insist on its partial export orientation (yarn, fabric or both) to take advantage of the abundance of good (coaliiy cotton, modern technology, ex- patriate management team, and lower wages, while ensuring higher productivity by supplemental training of a selected group of workers drawn from existing mills under TEXCO's control, introduction of worker incentives, and\ strict enforcement of labor requirements to avoid overstaffing. An export )subsidy could have compensated fot the unrealistic foreign exchange rate and other price distortiona. Thd 4ifficulties of such a task cannot be underratedt but th notion that it ws risky to establish export-oriented spinning and weaving mills because of low value added, relatively high freight cost, vola- tile export markets, and the need for long-termvcontracts was simplistic, and appears to have been a major deterrent in following iirough (PPAR, para. 14).' xiii. The heavy import dependence, which was detected diring implementa- tion and resu4ed in the modt4cation of the plant to make poss*ble the pro- duction of all cotton yarn gd fabrics; thp subsequent decision to install looma capable of producing widpr fabrics to "eet the requirements of garment manufwturers the \overly opttinistic demand projections and the expedient disregard of alternatives, suggest that the market ,analysis was thin, and that the quality of appraisal was not up to standard ** in contrast to the Bank' creditable supervision effort -and timely interventions. The decision to go ahead with a follow-on textile operation before the outcome of the first could' be fully assessed and the lessons be discerned and assimilated was premature, probably prompted by mioperceived externalities and reinforced by institutional pressures to lend. The genuine need to initiate the project (i.e. whether the project was justified because of its benefits and not just because it supported the Government's industrial strategy) was not estab- lished 'convincingly (PPAR, paras. 10-14, 17, 21, 26). xiv. The project's performance to date has ben poor, albeit largely for reasons beyond the management?s control. The project is financially unsound and its sustainability remains uncertain. The choice of technology per as was appropriate; however, given the mill's capability to produce fabrics of superior quality, the use of sophisticated equipment and expensive expatriate management talent for the production of low quality products defies economic rationale. Furthermore, efforts to improve the productivity and capacity utilization of the public sector's textile wills were frustrated. Thus, the project did not contribute tangibly to the sector's performan:e and regional development, while the anticipated benefits from the transfer of modern tech- nology and management techniques have yet to materialize. In the circum- stances, the project cannot be considered as successful, albeit to a signifi- cant extent due to deep-seated systemic failurei (PPARt para. 27). xv. Important Issues (e.g., pegging outstanding debt denominated in foreign currency at ;some reasonable exchange rate, capital restructuring, access to institutional credit) affecting- the project's performance and viability (and of many other industrial enterprises as well) demand bold initiatives by the Tanzanian authorities, as discussed in PPAR, paras. 28-31. Lesabns of Experience xvi. The project experience offers instructive lessons and suggestions that may provide a better defined framework for shaping the banks -posture and approach to country macro- and micro-economic issues and lending opera- tions. Specifically, the experience affirms that projects are unlikely to succeed in a hostile policy environment, and this raises the iesue whether or not the Bank should be leanding to sectors if conditions are uninviting, it is unable to exercise any influence, and project sustainability is likely to be impaired; in an environment of policy-induced distortions. monitoring the Borrower's entire investment program becomes essential, -if there is to be tome assurance that aid funds which command an opportunity cost are put into productive use an essential task 'during projebt design is to consider thoughtfully alltrnative project configurations to reflect more 'appositely the country's resource endoaent and binding constraints; costly changes in project scope after loan effectiveness suggest weak appraisal work and can be avoid4d- through a more diligent appraisal effort and meaningful risk anage'Ls; it is important that unequivocal, rules be established rega*ding the relationship between parastatals and government authorities,, and such under- standinge should be adhered tol the serviceability of the 'holding company concept in state-owned, enterprises is questionable; finally, provision -f training should be accompanied by "asures ensuring that, after completion of their train1*5, the trainees stay with the sponsbring the program enter- prise and donot seek more remunerat , opportunities elsewhere. The lessons pf experience are detailed In PPAR, para. 32, and PCR, paras. 8.03-8.06. S'PERFQMANCE AUDti REPORT MDROGORO TRXTILE PROJpCT (LOAN 16071CREDIT 833-TA) I. BACKGROUND The Industgial SettLn& 1. A centrally directed strategy of industrial development and policy framework was adopted in earnest by Tanzania's political leaderihip during the late 1960s to advance the objectives of self-reliance, economic growth and structural. transformation. Industry was been as a powerful engine of growth that could modernize the economy through more capital-int4nsive, higher-productivity processes, promoted* Vthin a protected environment. The Government firmly believed that a rigorously pursued industrialization strategy with emphasis on resource-based, producer goods industries that cater to basic needs, dubbed as the basic industry strategy, coupled with an extensive state involvement in the economy, would accelerat4 growth and achieve a more equitable socio-economic development. The Bank endorsed this approach to Industrial development from its inception, and even became its staunch supporter throughout the 1970s. 2. The implementation of\the adopt*d Aindustrialization strategy implied inter Alias creatioi of effective organizational structures and planning - processes to ensure appropriate inter-sectoral (e.g., industry, agriculture, infrastructure) and Intra-sectoral (e.g., rational project selection) alloca- tion of resourceit the substitution of an administrative apparatus for the market mechanism to effect resource allocation decisions and imilement policy measures; development, of public management capability to ensure efficient project implementation and operatios; significant levels of investment in relatively large, capital-intensive and import-intensive industries; a con- stant and dependable stream of foreign exchange to finance imp rted capital and decurrent Import requirements; a relatively large. dos*atic mrket to reap scale economies; and an adequate supply of techically trained manpow!r, including managers, to operate industrial undertakings -- a tall order. Considering Tanzania's level of deveapmt and the demanding requirements of a basic industry strateg, it is arguable whether the .strategy could have been effectively Implemented it the projected pace and time frame. 3. Major setbacks set in early on and the achievement of these goali in the immediate future is far from assured. After two dcades of pisperceived and ,poorly implemented inward-looking industrialization, and kdespite the injection of an inordinate amount of external financial and te bnical assis- tance, the industrial structure that has evolved to date *s. not7 ignificantly different from that which existed in the early 1960*, real per' capit income and wages in ihe late 198?s are lover than in the mid-1960s, ,the balance of paymente situation remaind critical, and heavy d4pendence on foreign inputs, financial resurce It technology and expertie persists -- in defiance of the- ( -2- stated goals. A confluence of factors, encompasoing an excessive and poorly managed planai6g system, Inadequate policies and Ineffective institutional arrangements., extremely low productivity of the work force and the invest- ments undertaken and, to some degree, external shocks, has led to the development of an inefficient industrial structure which is incapable of generating the hoped for sustainable growth and transformation of the economy. 4. Investment in manufacturing was strong during 1967-79, growng at over. 15% annually in real terms, induced inter alia by substantial inowe of official aid .t concessionary terms. It tapered off i the early 1980s, and declined dramatically (by one half) by the mid-1980s, mirroring the dire economic difficulties afflicting the sectpr and the economy as a whole. Real value added in manufacturing increase 6.8% per annum in 1967-79, but declined precipitously through the 1980s -- by about 5; per annum. During the same period, the share of manufacturing in GDP rose from 9.5% to 12%, but fell to 7.3% in 1986. Capacity utilization averaged around 50% in the 1970s but dropped to 25%, in the 1980s, due to over+design, heavy dependence on imported inputs, ma*agement and technical problems, poor aisintenance, and insufficient infrastructure. Despite industrial investments amounting to some US$3 billion in real terms during the past two deades, labor produic- tivity.rdmains below the 1966 level. Real output per employee in manufac- tuting has been ded,lining constantly since the late 1960s, and by 1979 was 701 of the 1966 level; by 1986, it had fallen further to 561. Real earnings per employee increased by 11% between 1967 and 1973, but declined sharply 2hereafter; by 1979 they were more than a quarter lower than in 1966, aid by the aid-1980s only 301 of the 1966 level. Finally, exports of manufactures have been erratic, averaging 13% of total exports during 1967-79 and 7Z during 1980-86. They grew (from a small base) 8% annually in nominal terms Muring 1967-79 but declined by 7% a year throughout the 1980. In real terms, manufacturing exports increased mariinally in 1967-72 and declined constantly thereafter through the late 1980s. 5. The string of intractable problems afflicting industrial perfor- mace, and whose severity assumed unwieldy proportions during the 1980s, ine des oversized or sub-optimal plant scalel completed but inoperative plant capacity due to lack ofUinfrastructure shortage of technical and mana- gerial skills and continue. reliance on high-priced exp#triatej acute shortages of imported inputs due to the lack of foreign exchange; lo factor productivity due to overmanning, rigid labor lar, poorly maintained equip- ment, absence of incentives to reward perfo rman e, and poor plant utilisa- tion; inadequate economic infrastructure (shortage of power and inadequate transport system in particular) reflecting the poor inter-sectoral allocation of ithestment and recurrent resources; under-capitalizd and heavily indebt d, if not virtually bankrupt, parastatals due to poor management and performance, rainforced by rising indebtedness due to successive \drastic devaluations. The poor condition of the *rastatals and their inablity to I/ Bet4een 1966 and 1989, the shilling has depreciated dramaticallys in in66 TSh 7 US$1; in 1989 TSh 145 - US$1. service their debt has in turn undermined the viability of the financial intermediaries (TIB and TDFL), which disploy an extremely weak portfolio and are strivin1 to remain afloat.2 6. Inadequate allocation of resources and deficient Incenti4e strue- tures for the development of the agricultural sector, excessive administra- tive controls over economic activities, and the continued growth in the size of the public sector without due regard to the limited administrative capacity available distorted the pattern of development' process and stifled progress. Ineffective organiational structures anO planning processes in- capable of ensuring proper coordination and inter- and intra-sectoral alloa- tio# of resources; ambitious investment programs, largely supported by the largess of the Bank and sympathetic donors; poor project screening procedures and haphazard project selection; underestimation of the human, institutional and organizational capacity of the country; poor organizational structure and performance of parastatals; and a distorted policy framework resulted in the development of an industrial sector w).ch is over-extended in relation to the size of the mark(et and th- country's technological and skill (managerial, technical and labor) capabilities. The policies adopted ended .up favoring industries linked to imported capital, skills and materials, rather than the use of domestic resources, thereby creating mounting pressures on th6 balance of payments at a time when, largely because of the anti-export bias, foreign exchange earnings were dwindling. Industrial investments made in the 1970s lacked selectivity, were grossly unproductive, and did not succeed in changing ppreeptibly the industrial structure, as no effective mechanism was developed to ensure that they interlocked and made the best use of available resources. The flaws and underlying problems of the strategy became more forcefully evident in the late 1970s and persist since then, suggesting that exteal dependence in all its dimensio s will continue well into the 1990s. 7. In 1986, the Government prepared an Economic Recove?y Program (ERP), taking the first init:L steps to rectify the distortions rreated- by earlier policies in an attempt to reverse the protracted deterioration of the economy. The areas addressed include exchange rate policy, the trade regime and foreign exchange allocations, performance of parastatals, the agricul- tural marketing system, pricing policies, indstrial restructuring, transport sector efficiency and public expenditures. 'The Goverament's objectives in the industrial se§tor aim at -improving capacity utilis tion rehabilitat*on of major industries, completion of on-going projects, and at ensuring that resources are directed toward the more productive and efficient firms in the g/ For details on the concept of the basip industrial strategy, policy framework, and the atr cture and performan"e of the manufacturing' sector see:. ORD, Report No. $329, World Bank/Tanzaia Relations, 1961-1987, January 16, 1990, Vol. II, "Tanzania's Industrialization ffort 1961-87," pp. 45-142. The present condition of the manufacturing actor has been ably diagnosed and analsed in a recent World Bank sector study entitleda Tanzania: An Agenda f Industrial Reco4ery, Report No. 357, June 30, 1987, 3 volumes. See also OED, PPAR No 7744, Tanzania TIB (Loans 1172, 1491 and 1750) and TDM. (Loan 1745), May 4, 1989. K -4- sector. The ERP "visagee a recovery period of ive to\ seven years.3 Furthermore, in the framework of an active macroeconomic dialogue, the Bank and the Government have ben working jointly to develop specific action pro- grams to rationalize the industrial sector, to be supported by sector lending. The proposed industrial restructuring scheme envisages three phases. Phase I will involve a diagnostic study of selected subsectors, including textiles, and the Identification of pressing issues both at the enterprise and subsector levels. Durixg Phase II, rehabilitation and re- structuring programs will be developed,\ including timetables for Implemen- tation., Phase III will lead to the implementation of agreed action programe.4 The Jextiles Sector 8. ~ The textile sector (including factory-made garments) to the largest in manufacturing in terms of employment (24%) and the second largest in terms of production (212) and valued added (19%). There are 87 establishments \engaged in textile production, but only 16 are relatively large. Six state- owned integrated nille account for 72% of the total spindles and, 772 of the looms installed. The sector continues to e4ffer from severe problems, the most common being frequent powei outages, inadequate water supply, shortages and delays in delivery of local raw materials and Imported intermediate inputs, shortage o# managerial and technical staff, premature replacement of qualified expatriates in key operational positions by nationals V th limited professional experience, inadequate train*ng of operating staff, limited capacity for production planning, acute finaicial diffidulties, overstaffing, and weak domestic market.5 As a result, capacity utilization remains low -- around 302. The public sector textile mills are under the control of the National Textile Corporation (TEXCO), a holding company which is responsible for overall management and production, as, well as 'for sector planning and investment. Obiectives 9. In, the context of the basic industry strategy, the public sector assumed a leading role in th development of the textile sector. The projeqt under review was aimed at import substitution and, originilly, was designed to meeti the growing and unsatisfied demand (pent-up demand plus projected future growth based particularly on rising farm incomes) for blended fabrics, "which feature greater durability compared to pure cotton fabrics and easier care properties. Tho project involved the establiabment of an integrated textile mill with spinning, weaving and finishing facilities, capable of producing about 21.5 million square metars of blended fabrics and 650 tons of _I The government's progra6 of policy 4d institutional reforms for 1987-90 are- outlined in Tazanias P21icy \Framework Paper, 1987-90, No. SecM87-1077, October 2, 1987, pa e. 6-24. 4/ For detai1s see PR No. P-4944, Industrial Rehabilitation and Trade K Ad1ustme t Program,\, November 22, 1988, paras. 93, 100-103. Recent economic developments and the impact of the RP are discussed in Iw., parae. 12-19. / For details see World Bank, Tansanias An AZenda for Induetal Recovery, Vol. II. Annex I. paras.\1-13. -5- surplus polyester rayon yarn annually for sale in the domestic market (sub- sequently redqqietad for the production of all cotton yarn and fabrics). The project enviiqged the installation of 42,000 ring spindles (44,688 were actually installed), 664 auomatic shuttle looms (594 actually installed but of larger width) of modern c4nventional design, and finishing facilities with modern devices to ensure product quality, under a new subsidiary, the *Morogoro Polyester Textiles Limited (Polytex). In addition, the Government aA0 TEXCO agreed to appoint a Technical Management firm to assist TWCO in carrying out a study by December 31, 1978, that would recommend measires to 4*prove the productivity and capacity utilization of the existing textile mills in the TEXCO group which faced serious technical, i,anagement, manpower, and financial\problems.' The study would be financed under the IDA Technical Aesistance Project (Cr. 601). On completion of the study, TEXCO would imple- ment\its recommendations with the assistance of additional technical person- nel at an estimated cost of US$2.2 million, of which US$2 million in foreign exchange would be provided by the proposed ptoject (SAR, pars. 2.08-2.13, 3.04-3.14, 5.0k-5.05; Project Agreement, Section 2.03). II. PROJECT DESIGN AND'\SCOPE Ptoect Confiuration 10. Two pivotal factors underly the design of the project under review and, by extension, the development \of the textile sector as a whole. First, demand projections for woven textiles were overly optimistic. Consumption of textile fabrics has grown at a substantially lower rate th*n projected at appraisal. Thus, in 1989 projected demand would have been 265 'aillion square meters (SAR, para. 4.11) compared to an estimated actual of 140 million -- the estimate for 1977 being 147 millio .6 Second,, binding operational constraints permeating the lector, though evident and cknovledged, did not elicit sober thinking both i* the Government and the Bank. Thus, at the time of the appraisal of the Hwanza textile projpct (November 1974), the Bank had impressed on the Government the need for undertaking a market and marketing study to establish the long-term potential for cotton fabris, including ,export possibilities, before embarking on another project.7 Also, because of resource constraints, shortag.; of trained technical staff, low pkoductivit\ and high production costs of existing mills which presented serious obstacles to launching a major export drive, it *as suggested that the Government would' be well advised to develop the textile industry at a more realistic pace. 6/ Interestingly, the consumption 'of textile fabrics had also been overestimated at the time of appraisal (1974) of the earlier financed Hwanza Textile Project (OED, PCR No. 5187, June 29, 1984, para. 4.65). Z/ Rapid expansion of the cotton-base,. textile dustry was a Jcey element of the basic industry strategy, aiming' inter alia at a gradual' restructuring \of the economy by 'iinking the production pattern more closely to domestic resource av*ilability and local demand. World Bank, Report No. 647, Tanzaniat ' Industrial and Mining Survey, March 31, 1975, Vol. III, Annex IV, para. 3. The waring was not heeded.8 IXu the ensuing years, and without consulting with the Bank, in addition to the Bank-sponsored Norogoro textile mill, three other cotton textile projects were undertaken by TCO with no apparent regard to the excess capacity being built up and the impact of the crippling constraints on operatthg efficiency. At the appraisal of the project under review, the Bank felt that a loan covenant to ensure consultation on sectoral development was impractical, because of the political sensitivity shown by,, the Government for a similar covenant on sector expansion during negotiations of the Kwansa textile project (Decision Hemorandum, January 3, 1978, para. 6). Instead, at the Bank's request, TEXCO would "provide information on major developments in the Borrower's textile sector" (Project Agreement, Section 5.02). The Bank did not follow up on this iportant issue. 11. At present, annual demand for textiles is estimated at 140 million meter ,9 while the existing rated production capacity at 246 million meters and actual production at 75 million meters. This suggests a capacity utili- zation rate of 30% for the sector. It is noteworthy, that 752 of the produc- tion capacity is less than to years old and cannot be viewed as outdated. With total annual production in the past few years limited to only,1,5 million meters, the unsatisfied demand of some 65 million meters was largely met through legal and Illegal imports which, aside from the low level of produc- tion, have been further encouraged by the high domestic prices due to rising costs of production (e.go, high price ok cotton, high cost of Imported inputs due to devaluations, low operating capacity and productivity levels, rising cost of external debt servicing, and heavy sales taxes). 12. Aside from the fact that the Bank's demand projections were optimis- tic, Bank appraisals (and the Government) did not consider seriously the alternative of rehabilitating (replacement of machines and equipment, balancing, and rationlization of processid* facilittes) and expanding the four older mills in TBXCO's group (Friendship Textile Kills -- AIKI, TanganyLka Dyeing and Weaving -- SUiURA , and Kilimanjaro Textiles -- TILTEX) -- 2 plants in different locat.ons) before proposing the establish- ment of a new plant. The or so, gince by the mid-1970s it was quite evident that the economic situation was deteriorating and that the difficul- ties of the textile sector (PPAR, para. 8) were compounding.10 It is interesting (and odd) that, while URAYI and SUNGURATE had just completed (1977) a partial rehabilitation, balancing and expansion program, ATEX had a\ similar program underway, and KILTEX was contemplating a modernizationlex- pansion program, the Bank was so concerned with developments in the textile sector, low capacify utilization and productivity in particular, that a dated covenant was included in the lagal agreement of the Norogoro project obli- gaping TERCO to carry out a e,udy of the productivity and cajacity utiliza- ti6n of all its operating mile, and see to the Implementatide of its findings (PPAR, para. 91. SAR, pares. 3.09-3.14, 3.25; Project Agraeent, A1 OED, PCR, Tazania - Nansa Textile Project, pares. 2.02, 8.08. 9/ "ansumption of textile fabrics has growr' at a substantially lower rate than projected in Bank appraisals, even illowing for the impact of the deteriorati6n of economic conditic6 on demand. 10/ World Bank, 1975, Vol. III, Annex IV, pp. 1-21. -7- Section 2.03). A textile sector rha46.11tation program loan addressing the sector's deep-.seated problems is likely to have saved substantially on capital investment and foreign exchange, made the existing mills more com- petitive,, and avoided the creation of excess capacity. It would seem that (a) the e tertained strong belief in the need for constructing an integrated textile plant to meet the assumed large pent-up local demand for blended fabrics, albeit heavily based on imported raw materials (650 polyester/35% cotton for shirtings, 65% polyester/35% rayon for suitings), and at a time when the country was facing acute bAlance of payments prolems and the Government 'was requesting every industry to export part of its productionf and (b) the Bank's understandable disinclination to launch an involved reha- bilitation operatikn,l prefetence for a tidier and less staff-intensive green field activity, and greater convenience in financing studies and relying on the sponsor agency for implementation, were major determining factors of the course of action takn. 13. Given that scale economies in textile mills beyond the relatively low mini u economic size are marginal and the severe constraints in human \skills and infrastructure, passing consideration was given during appraisal to building a smaller pant with provision \for phased expapsion. The appraisal (SAR, Annev 5-1, p. 4) indicated that a smaller than the proposed siz of plant would reduce the economic and financial rates of return for the project due to scale., economies. Yet, the minimum economic size plant being around 5,000 spindles and 250 looms (roughly one-third of the installed capacity), establishing a plant close to the strkimum economic size would have been a sounder decisida in the Tanzanian circumstances as it would have avoided str4ining scarce resources, would have made the management of the piapt more wieldy, and would still attain satisfactory economic \d financial returns. Apparently, confidence in market projections, and coiiction that the additional capacity was needed outweighed all other considerations and options, including potential operating difficulties. 14. Furthermore, since the country was already kacing acute foreign exchange problems, Bank missions as early as 1971 had stressed the need \for 11/ The design of rehabilitation projects poses some complex technical probtems, requiring inger .alia assessment of the physical condition of plant and equipment, including infrastructure; estimation oi the remaining ueful life of existing facilites; judgment on the degree of economic and technological oboolescene; functional. iniegration of existing and new facilities; and, more importantly, a judgmint on the economic merit of modernization/rehabilitation/balancing exercis6. Furthermore, the restructuring design exeFcise necessitates the formulation and in Vandem implementation of a financial restructuring g1n, as well as of a corporate stratear 2lan outlining specific actions that would ensure the firm's future viability." export promotion,12 and the objective was to develop'a viable resource-based industry, .it is puzling that the Bank, in sponsoring a capital-intensive and heavily raw material import-dependent project (PPAR, para. 12), did not insist' on its partial export orientation (yarn, fabric or both) to take advantage of the abundance of good quality cotton,13 modern technology, expatriate management t , and lower wagoswhile ensuring higher produd- tivity by qupplemental training of a seleted group of workers drawn from existing mills under TEXCO's control, introduction of worker incentives, and st ict enforcement of labor requirements to avoid overstaffing. An export subTidy could have compensated for the unrealistic foreign exchange rate and other price distortions. Interestingly, altho6gh Bank, staff acknowledged that there was no inherent reason why Tanzania should not be able to become an efficient producer in the future, id making international cost comparisons to estalish \competitiveness, they were comparing costs based on existing highly inefficiently operated plants dramatically affected by weak manage- *pent, poor production planning, extremely low labor productivity, and inflated cost structures (e.g., eight times more, staff than comparable western plants, shortage of well-trained managers and workers, lack of atten- tion to preventive and \pperational maintenance, shortage of spare parts resulting in excessive equipment breakdowns, inflated prices paid for equip- ment and recurring itiported inputs, high overhead costs),14 and not on projected cost structures based on availability of trained managers and oQerators. a modern and efficiently run lant. and a mill oerating at high levels of' capagity utlization as envisaged by the project. Indicatively,\ the break-even point at full production was estimated at 40% of effective \capacity (SAR, para. 7.10), suggesting a very efficient plant. The rather simplistic notion (with reference to the Tabora and Mbeya mills) that it was risky to establish export-oriented spinning and weaving.mills because of low value added, relatively high freight cost, vol tile er_Art markets, and the need for long-term contracts apparently was a major deterrent in following through (Issues Paper, August 2, 1977, para. 15). But there is no compelling reason to believe that the Mo ogoro project, assuming strong commitment and IV/ "If Tanzania does not succeed in encouifging and diversifying exports, the country's development will probably be hampered by a scarcity of, foreign exchange." World Bank, Reporp "iNo. AE-26, The Economic DeveloMet Prospects of Tanzania, 1972, Vol. I, para. 8; also _jd., "paras. 120, 121. Sde also, World Bank/Tanzania Relations. 1961-1987t Vol. II, "Tanzania' Industr fal iz-tion Effort," para. 4.22, and references to the economic reports issued in 1971, 1975 and 1977 there cited. ' J2/ Technically, Tanzania's medium-long staple cotton is ideally suited' foy blnding with polyester fibers which gives a ania an advantage in the manufacturing the blended fabrics compared to many. other primary cotton producers. World Bank, 1975, Vol. III, Annex IX, para. .6. J/ World Bank, 1975, V10. III, Annex IV, paras. 20-30. \91 diligent supervision, could not have developed into a partially export- oriented mill, the more so since tanzania was not subjected to quota re6tric- tions o/ import duties when exporting to the EEC market.15 Technologa 15. T e following technology altern tives were considered and debated at length at \appraisal: (a) pre-conventional'. technology, based on narrow mechanical (semi-aftomatic) looms and direct spinning; (b) conventional tech- alo.gX, based on broad automatic looms and ring spinning; qud (c) advanced tehnoJly, based on such devices as automatic opening of bl6v rooms, .chute- fed high speed carding machines, open spinning, unifils and electronically operated shuttleless looms. With respect to sinnin and weaving, which \accounted for about 85% of the total project cost, conventional technoloy was found\,o be the most economical in Tanzonia's conditions. Since the main 1-im of the use of advanced technology is to save on labor, such technoogy was not considered for spinning and weaving operations. The conventi al technology was recoAmoended as it showed definite advantages, over pre-conven- tional technology with respect to: loom and spinning efficiency; capital, operating and training costs; lack of a sufficient pool of skilled weavers; decreasing availability of equipment of pre-convention*l technology; production of wider, finer, Aigh-density fabrics of uniform quality; and raw material consumption per unit\ of output. For the finishing facility, which represented 15% of the total project cost, advanced technology was considered most economical. The rationale being not to - save on labor, but to achieve better efficiency in the processing of fabrics by reducing the\consumption of water, fuel, dyestuff and chemicals, at an estimated cost savings amounting to 20%. There are virtually no savings on labor regardless of whethqr advanced technology or older technology is used (SAR, Annex 5-1, pp. 1-3)\ The choice 'of technology VAr se appeaii sensible; however, in view of the mill\s capability to produce superior quality fabrics for domestic and export markets, the use of sQphisticated equipment fok the production of low quality\ products defies. economc rationale. Location and AnromentalAspects 16. Moragoro, a new industrial town being developed as a growth re'nter, was selected over Tanga at a slightly higher project cost as the place to establish the mill\ becausp of its. central location with respect to major markets, good rail and road transport connections, and access to labor. The project was designed to meet international standards for pollution control and exemplary working conditions (PCR, para. 3.13). 2Chanes in ScoMe \ 17. In view of the dontinual shortage of foreign exchange, reinforeel by the increasing cost of iuiported raw materials as result of the successive devaluations, plant facilities were modified in 1983 t a nominal cost Early on in project preparation, \the Bi had felt that there were good propects for exports especially & neighboring countVies, and, recommended that the \Company make serious efforts to exp?rt 3-5 mil ion square meters a year -- about 20% of production (Issues Ps er, para. 19). -10- (US$0.7 million) to make possible the production 'f all cotton yarns and fabrics. This development was contrary to statements during appraisal assuring that "availability of foreign exchange for the import of raw materials and spares is not expected to be a constraint since the Government gives priority in foreign exchange allocation to the import of raw materials and spares, and has agreed to make satisfactory arrangements for the timely inmportation of such" (SAR,, para. 7.14). This risk assessment suggests that the deep-seated problems \f the economy in the mid-1970s had not been fathomed, a situation which was exacerbated by the absence of centrally planned sourcing to yIeet recurrent foreign exchange requirements once projects were completed.16 A second important modification of the project scope after appraisal was the proposal in 1978 to install at a modest incre- mental cost (US$1.7 million) looms capable of producing wider fabrics to satisfy the requirements of the local garment manufacturers. In the circum- stances, both changes were consequential, and the Bank (and TEXCO) deserve credit for the correct decision taken which added flexibility to the mill's design. Finally, the Bank agreed, rightly so, to finance the foreign exchange and part of the local cost of constructing 60 housing units for local staff because, contrary to earlier expectations, tLey had no access to financing (see also PCR, para. 3.03).17 \ \\ III. IMPLEOENATION IMPERIENGE Project Managemnt 18. The set up for the management of the project (hMCO's Project Imple- mentation Unit -P- IU, the Project 'Advisory Firm -- PAF, 'the Project Engineering Firm -- PEF, and Management Advisory Firm -- MAF) was appr?- priat&, the firms selected were qualified and experienced, and their perfor- mance on the whole was satisfactory. Nevertheless, the project experienced difficulties dring implementation, and was completed with a delay of 41 months (85 months compared to 44 originally envisaged).18% The implementation schedule was realistic, but a number of events, such as political interven- tion in the appointment of th civil works contractor (PCR, para. 3.12), difficulties in the \evaluation and award of contracts, problems with project ]j/ World Bank/Tanzania Relations. 1961-87, Vol. II, "Tanzania's Industrialization Effort 1961-87," p4ras. 3.07-3.11, 7.06, 7.07, 7.10, 9.02, 9.03. 12/ Due to the increase in the cost of equipment and foreign exchange realignments as a result of delays in implementation OPAR, para. 8), by .1981 it becamb evident that there was a funding gap of some US$20 million. A modification ot_ the project scope was avoided thanks to TEXCO's ability to se ure external financing. \ However, procurement of the yarn dyeing equipment (US$1.5 million> was postponed (se0 lalso footnote 18). It should be noted that the yarn dyeing equipment (footnote 174\ was not ntalled until June 1988. Purchase of the equipment was postponed bcause of lack ,of funds a result of cost overruns, and was ultimately financed by TEXq0. management, disputes between PEF and TERXCO over contract neotiations with suppliers\ and contractors, high turnover, of PIU key staff, elated project commission du to late delivery of raw materials and inadequate infrastruc- ture, resulted in substantial delays in project implementation (for details see PCR, \paras. 3.01-3.08). Procurement of major equipment and services was made through international competitive bidding, in accordance with Bank guidelines. Nonetheless, the process of evaluating and awarding contracts was not always smooth and caused implementation delays (PCR, paras. 3.11. 3.12). It is noteworthy, that timely completion would not necessarily have ensured immediate start-up of operations due to delays in completion of the requisite infrastructure. Cost overruns amounted to 34% due to higher costs of civil works, machinery and equipment, and engineering services emanating - fror changes of the project scope, delays in implementation, inflation, and underestimation of costs. HRever, the additional financing required was only 4%, following the use of funds earmarked for contingencies and working capital, lower interest rate during construction, and the reduction of the technica asAistance component (PCR, pare. 3.14; PPAR, para.' 24) Policy Interventi&s 19. The project did not introduce macro- or micro*economic condition- Aliz 'on poJicy issues. affecting textile parastatals, pricing in particular. a Prices were (and are) based on the cost structure of the least efficient firm in the sector plus a fixed profit margin, a (formula which offered no incen- tive to a mill to improve productivity and reduce costs, particularly as it was insulated from 4omestic and international competition. Aside from the fact that the Bank deterspined, contrary to the assertions of all economic and sector reports, that "the present pricing system in textile is adequate" (Memo of April 6, 1978, para. 5), policy matters were thought of as best be taken up in the context of the' macroeconomic dialogue. The Bank merely extracted -a non-covenanted promise that the Government would ensure that the Company earns sufficient reverups to cover all its costs, srvice its debt, and ensure a reasonable return\on, its invested capital. But the dialogue during the 1970s was ineffective -- in contrast to the more constr4ctive macroeconomic dialogue in recent years, where the Bank and the Government have ,been working jointly to develop action programs to rationalize the iindustrial sector. Nevertheless, inimical macroeconomic conditions and policies have had an overwhelming impact on project performance. Role of the Bank 20. The Bank was actively involved, in' the conception, design, prepara- tion,\and implementation of the project. The Bank's technical experts sug- gested important design modifications during the implementation stage (PPAR, para. 17), interceded and helped resolve differences among the parties involved in project management, and followed up, though, not always succe- fully, lon the progress of infrastructural facilities implemented by various government agencies. -12* Operatina Performance 21. Polytex comenced operations in January 1986 (appraisal estimate January 1982), and Its operating performance to date has been very disap- pointing largely for reasons beyond the management's control (PCR, paras. 4.01-4.08). Capacity utilization was only 17% in 1986. 27% in 1987 and 10% in 1988, despite the fact that the plant produces all cotton fabrics and relies exclusively on domestic sources for its raw material requirements. Polytex's pog performance is attributable to weak demand for Polytex's line of products, high costs of production and inability to compete with legally and illegally imported textiles, lack of foreign exchange to import essential chemicals and spares, and staffing problems (e.g., high turnover, - absen- teeism, lack of motivation, limited problem-solving capability, J;nadequate remuneration). Under Tanzania's dire economic circumstances, the effective- ness of Polytex's management team understandably is highly constrained. Financial and Economic Performance. Suntainabilit 22. The financial performance of Polytex has been disaal (PCR, paras. 5.01-5.08, 6.01, 6.02). Due to poor production performance, price controls affecting some 40% of output, and pressure from legal and illegal imports, accumulated losss exceed TSh 5 billion, and Polytex is not in a position to servine its debt which keeps rising because of the coittinued devalbations. Recalculated financial and economic returns are negative, compared to an estimated 190 at appraisal. The *ustainability of the project is problematic, as it is heavily cpditioned on building up production at respectable levels, resolving the issue of the denominat;ed in foreign curren- cies debt (PPAR, para. 28), carrying out a financial restructuring, continued employment of costly expatriate management to fill key positions, access to wprking capital (PPAR, para. 30), and government actions to resolve important idustrial, trade, and financial policy issues. Institutional Developeant 23. Polytex's production management systems (e.g., process and cost controls, conservation/recycling of energy, quality control, etc.) are not yet ip to par, while TEXCO has provided limitd guidance on Improving produc- tion planning and management. The sub-optimal and low value added product- mix also attests to weaknesses in marketing. In the absence of a feedback and intelligence system, this modern mill, which is in a position to produce superior quality fabrics for domestic and potentially for export markets, produces run-of-the-mill items for local consumption. Efforts to upgrade and produce fabrics of uniform quality have been frustrated, in part because of inability to contr'1 fabric quality with reApect to required degree of white- ness, uniformity of dye and color fastness. On the other hand, efforts to adjust production to local demand requirements have not been made in earnest. 12/ Market demand developments and cpacity build-up sie themid-1970s at the kectoral level, indirectly \affecting Polytexa performance, are discussed in PPAR, pares. 10 and 11. -13.- The mill makes its 6\ procurement and marketing arrangements, tbou ; at always without external interference. Finally, concerted effor, -Z training, ,staff advancement and succession planning leave much to be dkssed. Although a comprehensive training program was put in place early on and implemented, Polytex has been unable to retain most of the original trainees, particularly those who were trained abroad (PCR, paras. 3.09, 7.02). This has frustrated expectations to develop indigenous staff capable of replacing expatriate management and, as a result, Polytex's reliance on expatriates to fill key technical and managerial positions persists. Furtherqore, the envisaged at appraisal transfer to Polytex of middle level managers,' foremen and skilled workers under TEXCO's control did not materialize, largely due to the drain in TEXCO's porsonnel and additions to-existing capacities,, thereby adding further to Polytix's operational difficulties (PCR, para. 3.10). Sectoral laat 24. The studT of the productivity and capacity utilisation\ of the operating mills under TEXCO was completed on time, and its recommendations were partially implemented. However, since the early 1980s mill performance deteriorated rapidly due, inter alia, to widespread shortaies of water, elec- tricity, fuel, and foreign exchange for procurement of dyes, chemicals, spare parts, accessories, etc., which were beyond management control. As a result, the impact of the actions taken cannot be assessed because of the disruptive effects of these factors which severely inluenced individual mill perfor- mance. Nonetheless, having to concentrate its efforts on pressing issues steming from the implementation, of a green field project, the Bank did not 'ionitor closely the preparation of the study and the implementation and. effectivenesq of the technical assistance profram. Much later, in 1984, because of dtfficulties in assessing the level of additional assistance \that was required, and to obtain a better understanding of the structure of\ the sector and the severity of -its problems, the Bank suggested to the Government that a full review of the sector be carried out. After ,a long delay, a review was carried out in late 1985 which outlined in broad terms the scope of a rehabilitation program.20 Having disbursed US$1.2 million for the improvement of the T XCO's mills cost nd fpancial control system under the technical assistance \component, the remainder amounting to US$1 mil1on was reallocated. In general, the Bank has \not been able to influence the pace of growth, reorientation and institutional deyplopment of the textile para- statals, 41beit in part due to distortive economic policies and lac. of receptivity by the authorities and TEXCO. \ V. OVERALL AUSESSMENT AND OUTSTANDING ISSUES gntall Aaeasswnt 25. The Bank's lending atrategy to industry in the 1970s was based on three dnsiderations: Tanzania was one of the least developed countries in the world; the country's leadership had shown a high degree of coumitment tA Z97 For details see World Bank, Tanania: . -An Aenda for IndustM Recovery. Vol. II, paras. 1-13. -14 growth and development; and Tanzania faced an enormous resource gap. The Bank's direct. lending for textile manufacturing ostensibly supported the development of a resource-based, import-substituting industry, within the precepts of the basic industry strategy and the goal of self-reliance. The project allegedly would rely on an important domestic raw material exported unprocessed, geterate value added, create job opportunities, and save foreign exchange. Yet,' in its original design the project was heavily dependent on imports of synthetic fibers -- and at A time when shorage of foreign exchange had become a critical issue. Also, in designing the project, the Bank did not consider at the requisite depth alternatives more suitable for the country's conditions -- rehabilitation of existing facilities, export orientation, smaller size with provision for phased expansion. Furthermorej the link between industrial sector work and lending has been somewhat tenuous, particularly concerning policy issues and their implications for the operating efficiency and financial viability of the project. 26. The heavy import dependence, which was detected during implementa- tion and resultod in the modification of the plant to make possible the production of all cotton yarn and fabrics; the susequent decision to install looms capable of producing wider fabrics to meet the requirements of garment manufacturers; the overly optimistic demand projections; and the expedient disregard of alternatives, suggest that the market analysis was thin, and that the quality of appraisal work was not up to stande rd -- in contrast to the Bank's supervision effort and timely interventions which were creditable. The decision to go ahead with a follow-on textile operation before the out- come of the first (Mwanza Textiles Project) could be fully assessed and the lessons be discerned and assimilated was premature, probably prompted by misperceived externalities (e.g., accumulated sector knowledge, possibilty of expeditious appraisal, and less costly implementation due to past experi- ence) and reinforced by institutional pressures to lend. The genuine need to initiate the project (i.e. \whether the project was justified because of its benefits and, not just because it supported the Go*ernment's industrial strategy) was not established convincingly. 27. The project's performance to date haq been poor, albeit largely for reasons beyond the management's control (see' also PCR, para. 3.02). Re- estimated financial and economic rates of return aie negative. The poject is fiancially unsound and its sustainability remaina\uncertain. The choice of technology 2r s appears appropriate; but given ths plant's capability to produce fabrics of superior quality, the use of sophisticated equipment and expensive expatriate management talent for the production of low quality products, defies economic rationale. Furthermore, efforts to improve the productivity and capacity utilization of the public sector's textile mills were frustrated. Thus, the project did not contribute' tangibly to the improvement of the sectors' performance and regional development, while the anticipa"Ad benefits from the transfer of modern technology and mana ement techniques (SAR, p4ras. 8.05-8.09) have yet to \materialize. In the circum- stances, the project cannot be considered as successful, albeit to a signifi- cant extent due to deep-vsated systemic inadequacies. butstanding Policy Issues Reairin \Initiatives by Tanzanian Authorities 28. One important issue thAt concerns Polytex and, by extension, many other industrial enterprises, is that debt denominated in foreign currencies is expected to be repaid in highly depreciated (and depreciating) Tanzanian shillings. This has imposed an unbearable financial burden on the enter- prises concerned and renders extremely difficult the development of a finan- cial restructuring plan. In addressing the present impasse, the Tanzanian authorities would be well advised to consider seriously pegging the liability of firms on ortstanding foreign currency denominated loans at some "reason- able" exchange rate, to be negotiated between the Treasury, kthe financial intermediaries and the p oject sponsor, taking into account the surrounditkg circumstances, with the difference to be forgiven and absotbed by the Treasury. To insist that enterprises repay debt at 17 times the original amount (In 1980, TSh 8.2 - US$1; in 1989, TSh 143.4 - US$1), a dramatic depreciation of the local currency which to a large extent was the direct consequence of the Government's own policy decisions and management of the economy and over which enterprises had no control, appears impractical, unfair, and counterproductive, the more so since the bulk of the debt is owed by parastatals. 29. Polytex's financial position is patently unsound. The company therefore needs extensive capital restructuring, possibly through conversion of Tre4sury and TIB loans and accrued interest into equity. 30. Access to institutional credit for working capital is another critical issue faced by Polytex and ,many other firms, emanating from the credit restrictions and the bias in credit allocation favoring enterprises catering to basic needs, producing revenue for the treasury, etc. Frozen credit ceilings for a long period under strong inflationary pressures, coupled with the additional financial, requirements, to service the rising indebtedness in foreign currencies, has created sev4re liquidity problems. This has reduced the capacity of many firms to procure needed raw materials and other inputs, even if they had access to foreign exchange, and has con- tributed to continued underutilization of capacity and low supply re6ponse. 31. Finally, a related issue which has created consternation among industrial undertikings is the treatment for tax purposes of thel "loss" arising from the repayment of loans denominated in foreign currenkies in depreciated shilli-qgs. In a Technical Circular issued on July 18, 1987, the Ministry of Finance, Economic Affairs and Planning, ruled that such an exchange loss is of 'a "capital" nature and should be capitalized, "to be taken into account when the asset is disjosed of,7" and cannot be viewed as an expense in determining the firm's taxable income. The matter deseLves con- sideration by the r sponsible authorities. VI. LESSONS OF EXPERIENCE 32. The project ex-arience offers instructive lessons and suggestions that may provide a better defined framework for shaping the Bank's posture and approach to country macro- and micro-economic issues and lending opera- tions (see also PCR,paras. 8.03-8.06). (i) The Tapaanian experience affirms that a project is unlikely to succeed in a hostile policy environment and an unconducive economic_mil . This raises the question whether the Bank should be lending to sectors (or at all) if conditions are unin- viting, it is unabl to exercise any influence, and project implementation and sustainability are likely to be impaired. Lending under such circumstances may not be consonant with prudent banking policies and responsible development assistance. (t1) An essential task during project design is to consider thouaht- fully alternative project configurations to refleqt More appositely the couUtry's resource endowment and bindin constraints. (III) State intervention and policy-induced distortions (e.g., ill- designed industrial, fiscal, monetary and foreign exchange policies) tend to undercut the contribution of individual projects to industria4 development and economic growth. Monitoring of the Borrover's entire investment grogram with respect to, inter,a1la. aggregate size, composition, and availability of foreign exchange and local funds to meet imple- mentation and operational requirements, becomes essential. if there is to be some assurance that aid funds are put into Broductive Mse. (iv) Closer litik with issue-oriented, economic and sector work and greater emphasis on detail during the preparatory phase, thorough market analysis and circumspect demand projections in particular, is likely to ensure more innovative projeit designs, deeper analysis of the broader sectoral policies and issues impacting QR performance, more accurate assessment of project risks, and better quality of appraisals. (v) Significant changes in Rroiect scoe after loan effecti4eness, impacting Lnter alia on the project's capital cost, financing plan, and implementation timetable, suggest weak appraisal work and can be avoided through a more diligent appraisal effort. ( ) At the implementation phase, timely aailability of specialized knowledge .nd experienced staff can help avert or correct promptly design problems. In this regard, clearly defined obli- gations of the engineering and advisory groups, and establish- ment of a good ranart between sponsor ,and contractors are crucial for problem solving and smooth project xecution. (vii) Project implementation can suffer from the lack of a determined effort to assess realistically early on the imolementation caRacity of the Borrgwer, as reflected in the Bokrower's inability or disinclination during implementation to carry out undertakings, e.g., provision of infrastructure. Since covenants committing the GovernmeAt and its agencies to provide infiastructure in a timely fashion do not necessarily guarantee compliance, there is need to make sure early on that work on such facilities has been initiated and that progress is being monitoreo. In the same vin, and, yet another diiension of risk analysis, the. Governmenti abilii; to honor assurances (e.g., a0 ess to foreign exchange) needs to be assessed in the light of economic developments and other surrounding circumstances. -17- (viii) Wrovision of training for managers, operators and craftsmen is not enough. Steps would have to be taken to ensure that. after comoletion of their training, the trainees sta on and do not seek more remunerative opportunities elsewhere to the detriment of the sponsoring the program enterprise. (ix) It is 1moortant_. that._uneaLcal rules be etablished remgarding the relationshi between narastatals and the government authori- ties concerned, and such understandings should be adhered to. Parastatals should, have clear objectives, and be assured of a degree of autonomy sufficient to enable them to make management decisions free from political pressures. At the same time, a well-designed systas of gg.,gat accountability, as opposed to IM. A= controls, should be instituted based on appropriate and monitorable indicators for the industry to assess their perfor- mance. Introduction'of standard cost accounting and establish- sent of profit centers at key stages of the production process are important management tools for cost and efficiency control and for taking remedial action. If such elemental conditions cannot be assured, the implication is that state ownership is not a' workable option and privatization may be the only viable alternative. (x) The experience in Tanzania, as well as elsewhere, casts o on .the .serviceability of the holding .Comany concest in advising, coordinating and directing the entities under its control. In general, holding parastatals tie up large numbers of scarce indigenous managers, accountants and technical specialists, spread thinly senior management, and create greater need for expatriates, while their overhead costs become a burden on the operating subsidiaries and raise production costs. Indeed, it is not clear whether the cost of sustaining holdtng parastatals is justified by the accruing benefits, and whether their subsidiaries would not perform equally well or even better without them. Furthermore, the central control exercised by the holding company does not allow competition among subsidiaries and, as a result, it provides no incentive to them to improve efficiency, The concept and usefulness of holding companies needs to be reconsidered. 19 'PROJECT COMPLETION REPÖRT TANZANIA MOROGORO TERTILE PROJECT (LOAN 1607/CMEDIT 833-TA) February 10, 1989 Industry and Energy Operat,ons Division Southern Africa Depaytment Africa Region -21- PROJECT COMLETION REPORT MOROGORO TEXTILES PROJECT (LAN. 1607/CREDIT 833-TA) 1. INTRODUCTION The Econom 1.01 After independence in 1961, the economy of Tanzania underwent several years of rapid economic growth. In 1967, Tanaania's leadership embarked on an era of socialism. The now priorities of the country, as enunciated in the Arusha Declaration, emphasized self-reliance, basic education and broad-based rural development, and led, in the late 1960s and early 1970s, to a rapid growth of the public sector. In the late 1970s an unfortunate combination of external factors (worsening terms of trade, the war with Uganda, and the break-up of the East African Commnity) weakened Tanzania's economy, and attempts at policy reform made in the early 1970s failed to address the key problems. 1.02 The 1984/85 budget provided the first indication of a new pragmatism in the Government's economic management; the exchange rate was devalued by one third, parastatal subsidies were cut, and an import liberalization program was initiated. In mid-1986, the Government produced a medium-term 'Economic Recovery Program' (ERP) aimed at achieving a positive growth rate in real per capita income, reducing the rate of inflation, and restoring a sustainable balance of payments position. The cornerstone of the ERP was the adjustment of the Tanzanian currency. 1.03 After more than two years of implementation, the Government remains committed to the EP. In 1,986, for the first time since 1980, the rate of growth of real GDP, at 3.4 j,ercent, exceeded the population growth rate. The 1987 real GDP growth rate accelerated to an estimated 4.5 percent. Much of the growth in economic activity is outside the official sector and is not captured in the GDP statistics. 1.04 Over the next decade the Tanzanian economy is expected to grow at an average real rate of 4.6 percent per annum. Reduced distortions (resulting from the reforms in the exchange rate, trade and pricing regimes), alleviation of serious infrastructural bottlenecks, and institutional improvements are expected to lead to significant improvements in the efficiency of resource allocation. The process of resource reallocation, economic restructuring,and rehabilitation is also expected to result in a reduction in import dependence over time, and in smaller import requirements to sustain the projected GDP growth rate, particularly over the next three years. - 22- The Industrial Sector\ 1.05 Until the mid- 1960s, Tanzania had a small industrial sec or dominated by private firms that had largely developed since independence. The largest subsector was cotton ginning. Hanutacturing employed about 25,000 people and contributed very little to GDP. In 1974, the Government adopted the Basic Industrial Strategy (BIS), emphasizing import- substitution and the production of capital and iAtermediate goods and centering around the parastatal sector. Implementati6n of the BIS resulted in. large capital and impodt-intensive investments in industry. After US$2 billion of investments In the mid- 1980s, Tanzania's industry was estimated to be producing only about 3 percent of the country's GDP (measured at world prices), while using imported inputs equivalent to almost one-third of total imports. ,Capacity utilization averaged .5 percent and one-third of industrial activities were estimated to be producing with n6gative value-added at vorld pices. 1.06 A Bank review of the induoty sector was undertaken in October 1985, and 1l toths publicition of 1anani4. An Aenda I ndustrial Recovery" (Repott No. 63574A) ia Jude 1987.: Atong ldustrial subsectors, only a few were found to be relatively efficient. Among the inefficient subsectors, however, there were several considered to be potentially viable activities, particularly textiles, tanneries, metal products and machinery. The report indicates that: parastatals are significantly over-representd in the highly inefficient activities; on average, import and capital-intensive activities are significantly less efficiept than labor- itensive activities that utilize local resources; and smaller Irms, employing ever than 100 employees, are more efficient than larger firms. 1.07. The major ongoing reforms have already resulted in some reallocation of resources. The industrial sector is currently Operating at the same leel of average capacity utilization (25 percent) as'-i 1984, but is using fewer resources. A decrease in production and.utilization of inputs is already apparent among many inefficient industrial activities while some efficient fiz*s have managed to increase capacity utilization. Nowever, a major restructuringV of inefficient industries and a significant expansioA, of efficient activities will only take place as the Government continues to implement the macroeconomic and trade reform program and initiates industrial restructuring measures in key subsectors and firms. In addition to the policy distortions that have plagued industry, there are significant structural, operational and managerial inefficiencies in the sector' at present, including managerial and skilled labor shortages, infrastructural bottlenecks, and shortages in raw materials and other inputs. The Textile Subsector 1.08 The textile subsector inKTanzania is the largest i total employment (24 percent of m nufacturing) and the second largist by gross value of production (21\ percent) and value-added (9 percent) after the food, beverage, and tobacco subsector. The public sector plays a leading role in the development of the textile industry. After the Arusha Declaration, the ajor textile companies were incorporated into and operated as subsidiaries f te N#tional Development Corporation (NRC). In January 1974, all four large integrated textile companies, as well As the manufacturing of garments K -23- and jute bags. were transferred from NDC to a new parastatal, the National Textile Corporation (TEXCO). T9XCO was organized as a holding company controlling the majority of shares and exercising complete responsibility for overall management and production of its subsidiary companies, and planning and expansion of the sector. 1.09 Annual demand for textiles in Tanzania is about 140 million linear meters (UM). In order to reach self-suffiviency and to enable exports of textiles, productio4 capacity was expanded from about 110 million LM in 1976 to about 153 million LM by the end of 1986. Part of this expansion was accomplished under the Bank-sponsored Mwanza Textile Project (Ln.ig28-TA), under which the capacity of the Mwatex Textile Mill was doubled. About 80 percent of productive capacity is in the parastatal sector. The basic industries (spinning, weaving, fLaishing) are largely unde- parastatals, while -the down-stream industries (knitting, garmenting) are under the private sector and depend on the parastatals for raw materials. ,Nearly 75 percent of 'production facilities have been created in the last decade. These facilities were largely for the new companies, while much needed rehabilitation in some of the older enterprises, with demonstrated good track records, was neglected. However, despite this comparative technological well-being and, until recently, a protected market situation, production levels at 24-30 percent of installed capacity have been poor and have, in fact, deteriorated since the completion of the Mwanza Project, when the textile industry was operating at about 40 percent of installed capacity. Poor production has been due 'to: (i) acute shortages of imported spare parts and inputs; (ii) lack of power; (iii) water shortages; (iv) faulty plant locations and project design/implementation; " and' (v)' transportation bottlenecks. As a result of these factors, nearly US$500 million worth of newly installed productive equipment is inoperative. The low production levels has encouraged large-scale importation (legal and illegal). 1.10 the consumption of synthetics as a percentage of textile consumption is increasing in developing countries, including Tanzania, because of their favorable characteristics - durability, easy care, crease-recovery iad improved stability - compared. to all-cotton fabrics. During the 1974-80 period, the global per capita consumption for synthetic i0hers increased from 1.3 to 1.8 kg while the increase for all natural fibers-was limited to 0.1 kg. In line with these worldiide changes, textile demand in Tanzania has also been shifting towards man-made fibers. Although per capita consumption doubled from 0.3 to 0.6 kg during the 1974-80 period, much demand was unmet due to import restrictions. The Morogoro Project was primarily intended to address the unsatisfie demand for blended fabrics. II. PROJECT BACKGROUND Origin ind Evolution 'of the Project 2.01 The Morogoro 'Textile Project was proposed to the Bank in June 1985 to help Tanzania meet the growing but largely unsatisfied domestic demand ior, blended fabrics. As such, it was part of the Government's strategy to reduce impot dependence in essential consumer items. The project was * 24- identified by the consultants who carried out a 1976-textile market study financed under a previous Bank operation, the Mwanza Textile Project (Loan 1128-TA). Following identification, a feasibility study for the project was completed in May 1977. The project was pre-appraised in JunelJuly 1977. The appraisal mission of October 1977 was followed by discussions with TEXCO in Washington in November 1977, and a follow-up mission in February 1978 to firm up the project scope, estimated capital cost, financing plan and implementation arrangements,. and also measures to improve productivity and capacity utilization of the operating :extile mills within the TEXCO group. 2.02 At appraisal, three major issues were identlfied concerning the project and the sector: (i) The choice of technology, between pre- conventional, conventional, and advanced technology, was extensively discussed at the decision memorandum stage. It was agreed to usemodern but conventional technology for the spinning and weaving operations, and the latest processing technology for finishing to ensure the economic utilization of water, fuel, dyes and chemicals. It was expected at appraisal that this technology would provide the necessary flexibility to meet market competition in Tanzania. (ii) Because of Bank 69ncerns regarding low capacity utilization at virtually all Tanzanian textile factories, a dated coven)ant was included in the legal agreement to ensure that TEXCO would carry out a study of the capacity utilization and productivity of its operating textile mills and that TEXCO would 'take the necessary measures to improve utilization and productivity. (Iii) Finally, possible changes in the project scope were discussed because the required financing looked as if it would not materialize; adequate financing, however, became available. 2.03 On June 29, 1978, the Bank/IDA's Executive Directors approved a loan of US$25 million to the Government of Tanzania (the Government) for a period of Z0 years, including five years gr ce, and a credit of US$20 million on standard IDA terms. The Government agreed to channel the IDA funds to- the Morogoro\Textile Company (also known as Polytex) through TEXCO, partly as loan (US$8.5 million) and partly as equity (US$3.5 million). The remaining US$2 million of the IDA credit was made available to\'TEXCO \to finance the tethnical -as istance program. All Bank/IDA funds onlent to the, company .carried an a=ual interest rate of 7.5Z with repayment over 15 years including 4-1/2 years' grace. The Project, Loan and Credit Agreements were signed on July 28, 1978, and the subsidiary Loan Agre ment between the\ Government and Polytex was siined on January 20, 1979. The Loan and CrAdit became effective on May 7, 1979, after the p qject advisory firm and the -25- p&oject engiVeering firm were appointed, and both the suppliers' credit and Tanzania Investment Bank loan that were required to complete the foreign cost tinancing of the project became effective. Project Description and Objectives 2.04 The objective of the project was to help the country meet the growing but largely unsatisfied demand for blended fabrics. The project \consisted of the establishment of the first fully integrated plant in Tanzania capable of processing polyester-blended fabrics and yarns, and technical assistance to TEXCO. The plant, known as Polytex, has spinning, weavifg and finishing facilities capable of an annual production of about 21.5 million square metersof blended fabrics of medium-to-fine quality for shirtings and suitings. and about 650 tons of surplus yarn. The project size was decided after taking into account projected dema4, .economies of scale and availability of financing. Morogoro, 120 miles 'southeast of Dar es Salaam, was selected as the location for the project as it is well connected by ;oad and rail transportatidn and centrally located with respect to major markets. The site includes space for possible future expansion. 2,05. The plant includes 44.688 ring spindles, about 594 shuttle looas and finishing facilities for the production of polyester/cotton and polyester/rayon fabrics; plant equipment was modified during implementation to enable production of all-cotton fabrics. Basic raw materials required for the project are cotton, polyester and rayon fibers. Cotton is available locally and is well-suited for blending with polyester. Polyester and rayon fiber are imported. Sufficient quantities for comissioning "and operation for about one year were to have been financ&4 under the project, although actual financing ot, working capital was much l\wer than had been budgeted at appraisal. Infrastructure and uiilities, \Ucluding electric substations, air-conditioning, steam generators, raw water treatment, effluent treatment, mechanical and electrical workshops, were,provided. Water supply was secured with the completion of the Bank-financed Hindu Dam Project in 1985. Electric power was also secured through the installation of a new transmission line from the national grid, which was partially financed under this project. 111. PROJECT IMPLEMENTATION AND MANAGEMENT 3.01 Implementation Schedule. The project was completed in seven years, versus an expected four years at appraisal. Project implementation was originally scheduled to commence in the beginning of 1979 and the test runs were to have been completed in October 1981. Actual implementation took until October\19851 and the test run cotinued from October 1985 through the end of December 1985 (Annex I). Thus, project completion was delayed by 41 months, due mainly to delays in: (i) appointing the main civil. building contra tor due to political reasons (13 months); (ii) awarding the contract to the lectrical contractor because of the need to re-tender the package 9n the basis of suppliers cri;it (14 months); (iii) commissioning of the mills dud\to late delivery of raw materials (6 months); and (1y) various reasons, including late payment of letters of credits, and lack of adequate power and water. In view of these delays, a two-year extqnsion of the original closing -26- date, from June 30, 1985 to June 30, 1987, was appro*ed in order to finance the services of the management advisory firm that was to provide technical and management assistance during start-up and the initial operating period. 3.02 Achierkment of Project Objectives. e primary objective of the project whs to help Tanzania,meet the growing but largely unsatisfied demand for blended fabrics through the establishment of an integrated plant capable of processing polyester-blended fabrics and yarns. .Due to continual shortages of necessary inputs and their increasing costs in terms of the depreciating Tanzania shilling, complicated by long delays in plant commissioning, the plant, Polytex, has been able to achieve only minimal production of the intended product. In response to these shortages, modifications were made to plant facilities to enable the production of all cotton fabrics and yarns4 which have become Polytex dominant products. Thereftre, it cannot be said that the project's original objective of meeting domedtic demand for blended fabrics has been met, nor does it seem likely to be met in the foreseeable, future. 3.03 Chgmes in PpJect Scope. Several changes in project scope became necessary during implementationt (i) Introduction of all-cotton production. The plant design was modified in 1983,at a cost of $0.7 million to allow for the processing of all cotton fabrics. This modification was considered necessary in view of difficulties in importing the polyester and \rayos fibers needed for production, and their increasing cost. Although the introduction of all-cotton processing was done very early in the project, it was felt in 984 that the preferred course for the mill would be to continue to produce blended fabrics and yarns, because using all cotton would still require the importation of chemicals and spares, and Polytex would be competing, in a depressed market, against other, established mills., This situation led to the development of the "cotton swap- scheme (para.4.04), (ii) Increase in housina units. It was envisaged at appraisal that housing units for expatriate staff and top management would be financed under the project; other staff were expected to build their own houses vi h assistance from the Tanzania Housing Bank and the National Housing Board. The Government subsequently agreed to provide sites and servips for workers' housing in Horogoro in view of the unexpectedly strong demand for housing in that area, bdt financing did not materialize. Because of the Bank's concern that inadequate housing facilities might prompt middle managers and the Tanzanian employees who yere trained abroad under the project to leave Rorogoro, the \Bank/IDA 'agreed to finance ibe foreign cost as well as part of the local cost of con tructing 60 housing units. (iii) Acquisition of additional equipment. When after appraisal it became apparent that garment manufacturers preferred to use wider fabrics, looms capable of producing the wider cloth were installed at a cost of US$1.7 million. (iv) Phasing of Project. It was learned early in project implementation that many Items that had been expected to be locally available, had to be imported at an unexpected foreign expense of about US$18 million. Two approaches to address these additional expenses were considered: either reducing the scale of the project, or developing it in phases in accordance with available finances. Phasing was considered a better option since procurement arrangements were very advanced. However, TEXCO was successful in bridging the foreign exchange gap through a credit facility from India; in addition, project cost estimates revised in March 1984 indicated that it would be possible to procure some remaining equipment at lower than expected costs due largely to favorable changes in currency exchange rates. The modif*cation in the project scope wascancelle4 and the project was completed as originally designed. (v) Modification of Technical Assistance Component. The financing of technical assistance to TEXCO did not take place as planned. On the basis of recommendations of a 1978 productivity study that was partially implemented with the assistance of UNIDO, TEXCO',had carried out a number of measures that led to improved performance of its group companies. Subsequently, however, widespread shortages of water, electricity, fuel, and foreign exchange began to hove a serious offect on their performance. Because it became virtually Impossible to assess the effedtiveness of the TA measure , in early 1984, the Bank recommended a full review and assessment of the entire subsector before mill- specific improvements were undertaken. However, to improve TE]CO's financial management. of its' companies, tAe Bank agreed to finance the services of specialists in finacial, planning and control system.under this component. 3.04 ProJect Mana&Vment, and use and performance of consultants. ,At ap#a1val, it was expedted that TEXCQ's Project. Implementation Unit (P1U) would be responsible for the implemeniation of the project. The PIU was to be assisted by a project advisory firm (PAF)\And a project engineering firm (PEP); the signing of contracts with the PAP and the PEF was a condition of effectiveness. The PAF was to i* responsible for supervising the project, includingt (a) preparation of a project execution plan; (b) selection and monitoring of consultants and contractors; (c) preparation of periodic progress reports; (d) reviewing and approval of billing submitted by consultants and gontractors; and (e) supervision of procurement. TEXCO'S corporate staff was to provide the PAP 4ith the necessary engineering, leial and accounting services.\ The PEF was to be responsible for project design, engineering, *,procuremen , construction and installation, training of personnel, pr6ject start-bp, and comissioqing. At least six months before the mechanical' completion of the project, \TEXCO and Polytex were to have appointed a management advisory firm (MAF) to perform key technical and i 181 - 28 - management functions for a least the first three years of initial plant operation. By Board presentation, TEXCO had already selected a PAF and a PEF. 3.0 Project implementation was difficult for much of the period under review. Although difficulties with procurement were the primary reason for slow implementation, problems with project management were also a significant factor. The PIU's relationship with TEXCO, particularly on procurement, was not always smooth. The PIU experienced frequent changes in key positions; between 1980 and 1982, two project managers left TEXCO. In 1982, TEXCO finally\took charge of all procurement matters, and limited the role of the PTV t6 construction supervision. TEXCO actively supervised the project through its Di;ector of Development until 1984. Mhen the Director of Development left TEkCO in that year, the project fe.l1 behind schedule. S.06 The PEP had excellent technical qualifications as the engineering arm of an internationally knon textile, company with a successful track record in joint-venture textile operations in developing countries, mainly in, South Asia and Latin America. #SF performance was satisfactory in almost all aspects of their work program. TEXCO, however, indicated to Bank missions that the PEP did not adequately safeguard TEXCO's interests during contract negotiations with suppliers and contractors. Due to their limited international experience, the PEF tended to evaluate bids in light of their local experience, which drew criticism from TZXCO and the Bank. Since the payment terms of its contract with TEXCO were related to completion of specifications, the PEF was under pressure to speed up implementation. The PEP tended to be more agreeable to contingencies, specifically against shortages, that suppliers (in particular contractors) built into their offers bepause of the country conditions. TEXCO would not agree to provisions in the contract concerning unforeseeable shortages, which it claimed always resulted in the %*justified use of these provisions. As a result, TEXCO often found itself engaged in lengthy negotiations to reach acceptable agreements, based on its own experience on other projects. On the other hand, the PEF poitsted out that the 'benefits ga4ned through these lengthy negotiations were substantially outweighed by the additional costs to the project due tp these long delays. 3.07 The appointment of a PAF that had broad internationals experience, particularly in Tanzania, and that was supported by good back-up\operational resources, was advantageous. to the project and the, PIU. The PAF was especial11 helpful, because of their extensive knowledge of loca,l conditions, ind "ealUig with Government authorities and in complementing the work of the P# in areas where wide international experience was required. In contrast with that of'the PEF, TEXCO's c?ntract with the PAP was time-based, and the PAP's knowledge of local conditibns was undoubtedly instrumental in requiring regular payments. When implementation was idled for ab9ut 18 months due to a dispute over the main civil works contract, the PAP continued to receive payments in spite \of a very modest contribution. When the PAF's contract came up for renegotiation, it was deci ed to discontinue their services as a major part of design and procurement wdrk was completed; however, agreement was reached that the PAP would continue supervising overseas traiOing until c apl tion. The remaining major activities that should have been cariied out - 29 - by the P.AF, namely local training and commission atditing, were later included in the.contract for the HAF. 3.08 The management'. advisory firm "(MAF) assisted Polytex in mill commissioning and operation, and oversawtraining. Although this was the first time the HAF had been employed on a Bank-sponsored project, MAF has had extensive involvement in developing countries, including in Africa, and its performance in all areas has been satisfactory. Giving the MAF responsibility for training ensured accountability for the quality of training, and continuity of operations as different shifts were trained. Responsibility for commission.auditing by a party not involved in design and implementation but responsible for running the mill, was effective in ensuring take-over fts suppliere and contractors. MAP'e original two-year contract expired in Juni 1907 but was extan4ed for a year as plant management was not fouAd to be sufficiently traiae'to manage the plant. 3.09 at$nn. Tht training togram *ncluded overseas study and on-the. Job trainli as well as local training in the plant training center. Long- term study and trainingIprograms in the UK and the US for about 50 Tanzanians were completed in 1984, in time for them to be Involved in machine erection and plant commissioning. Twelve technicians and loom fi#ers were also trained abroad. The cost of foreign training was abou* US$1. million and was generally successful. Most of those trained abroad took an active part in training local staff unde'r the overall supervision of the MAF. It was expected at appraiial that the company would be able to retain most of t#e trainees, at least for 4 number of years, And that they wpuld be an important link between the expatriate management and local staff, and rould graduaAly \take over the key positions of the company as expatriate services were phaed out. However,there were some problems in getting all the Tanzanians triined aproad .to return to vork at Polytex\and a number of those trained abroad abandoned Polytex to ork for other mill in Tanzania or abroad, andlor fot unrelated-private entirprises. A well-designed and fully equipped taining center was established to train the large number of operators required. The target was to train 1,018 operators and 146 mechanics, loom fixers, electricians, and other technicians. As of 1987., 1,085 had been trained and the quality of training was considered good. 3.10 It was envisaged at appraisal that some middle management, foremen and skilled workers voild likely be drawn from existing t4xtile.mills with supplemental traininj to be provided under the project. This did not mateialize due to the substantial new capacities added to the texco Group since appraisal, -the drain on Texco's,'personnel to p*ivate business, and Texco's desire to safeguard against the slipshod practices\ tha\ could be t ansferred to the new company. This contributed to the sustantiallik highqr cost of trairi$ng incurred under the project.\, 3.11 Procurement. Procurement.ok major equipment and services was made through international competitive bidding in accordance with Bank guidelines. Items costing\ less than US$100,000 were purchased through internatio al. shopping. \ Spiing and weaving \equipment, and electrical equipment were financed through suppliers' creVdit facilities from Japan ana India, respectively. All tender documents were prepared by the PEF with the assistance of the PAP, and issued to prequalified tenders, package-by- AV -30- package. Tender evaluations were done by the PEP with the assistance of the PAP. 3.12 Procurement relations in all areas .- goods, civil works and consultants -- were at times difficult. Even, after contracts had been negotiated and signed, there were disagreements over the precise responsibilities of the suppliers, delivery terms, etc. Over half of the delays in implementation are attributable to problems in procurement. Of ,particular interest were the problems surrounding the selection of the main civil works component. Tenders for this component, expected to cost about US$19 million (US$8.5 million to be financed by the Bank) were floated in February 1980. After evaluation of the bids, the PAP and PEP recommended award to a non-Tanzanian company. TEXCO, on the other hand, wanted the contract awarded to a local contractor, which had not been pre-qualified, on the basis that the price quoted by the expatriate constractor was unrealistically low for a foreign contractor and that award to the local company would result in fewer payments overseas. After much discussion, TEXCO and the foreign contractor agreed to sign a contract in September 1980. However, Government instructed TEXCO not to sign the contract because it believed the contractor to have South African connections, and that its work on another Bank-financed project had been unsatisfactory (which was not verified by Bank missions). After protracted discussions between the Bank, the relevarit governments, and the project entities, a contract was finally signed. ,This contkact was extended in 1984 to enable the contractor to build additional staff housing for the project. Negotiations were protracted, and in fact broke down over the issue of the applicable, exchange rate because TEXCO believed that foreign exch"ge could be saved if a new international tender were carried out. Again, wile the differences were finally resolved, project implementation was delayed.- 3.13 Environmental Aspects. As there are no Tanzanian sta!dards for pollution control, the project was designed to meet intern4tional standards for gaseous and liquid emissions. The new steam boilers were equipped with automatic regulating devices to ensure maximum combustion so as to control the discharge of carbon monoxide in the fuel gases. The blow room and cards were equipped with floor waste exhaust filters and air filters installed in the spinning and weaving sheds to reduce the level of dust in the working arlas. Adequate settling tanks and effluent treatment equipment were initalled to treat effluent from the converting operations, and chemical discharge levels meet international standards. Proper humidification and vengilation systems have been installed In the factory. 3.14 Cost, Disbursements and Financing. A comparison of the actual costs with the appraisal estimates is presented on the next page. Although the total actual costs of the project were within 3.5 percent of estimates at aptaisal, the large variances in two categories deserve special notice. Engineering and other services required twice as much financing as had been expected. Overruns in other categories appear to have been compensated for by reducing by 86 percent the appraisal allocatipn for working capital. Proiect Costs - Appraisal Estimates vs. Actual Costs Appraisal Estimates Actual Costs Local Foreign Total Local Foreign Total Local Foreign Total Local Foreign Total Variance ---TSh Millii--- ---US$Killions--- ----TSh millions-- ---US$millions--- (1) I. Morooro Textile Hill Plant & Hachinery 10.0 240.7 250.7 1.2 29.0 30.2 104.1 696.3 800.4 9.4 39.8 49.2 62.9; Civil Works 110.4 44.0 154.4 1363 5.3 18.6 285.4 248.6 534.0- 25.8 14.2 40.0 115.1% Engineering & _Other Services 11.5 --27.5 39.0 1.4 3.3 4.7 2.7 271.8 274.5 0.3 15.5 15.8 236.21 Study & Training 3.3 12.4 15.7 0.4 1.5 1.9 0.0 26.4 26.4 0.0 1.5 1.5 -21.1% Other 39.9 45.7 85.6 4.8 5.5 10.3 0.0 0.0 0.0 0.0 0.0 0.0 . 0.0?Z Contingencies 44.0 65.5 109.5 5.3 7.9 13.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0? Installed Cost 219.1 435.8 654.9 26.4 52,3 78.9 392.2 1243.1 1635.3 35.5 71.0 106.5 35.02 Working Capital 73.0 47.3 120.3 .8.8 5.7 14.5 35.0 0.0 35.0 2.0 0.0 2.0 -86.22 Interest During Construction 30.7 -63.1 93.8 3.7 7.6 11.3 0.0 15.8 15.8 0.0 0.9 0.9 -:92.02 Financing Required 322.8 546.2 869.0 38.9 65.8 104.7 427.2 1258.9 1686.1 37.5 71.9 109.4 4.5? II. Tech. Asst. to TEXCO . 1.7 16.2 18.3 2.0 2.0 2.2 - 20.7 20.7 0.0 1.2 1.2 -45.5Z Total Financing Required 324.5 562.8 887.3 39.1 67.8 106.9 427.2 1279.6 1706.8 37.5 73.1 110.6 3.52 ' - 32- 3.15 Financing for the project was provided from the following sources; Project Financng - Appraisal Estimates vs. Actual Financing (US$illions) Appraisal Estimates Actual Source Debt Equity Total Debt Equity Total IBRD 25.0 -- 25.0, 25.0 -- 25.0 IDA 8.5 11.5 20.0 8.5 11.5 20.0 Bilateral Credit 20.0 -- 20.0 20.5 -- 20.5 TIB (TEXCO) 10.0 31.9 41.9 10.0 35.1 45.1 Total 63.5 43.4 106.9 64.0 46.6 110.6 Disbursements on the IBRD and IDA funds were completed in FY79, about 2-1/2 years behind schedule. The schedule of disbursements is presented in Annex IV. OPERATING PERF0MNANCE 4.01 Commissioningstart-up and Capacity and Production. At appraisal, the project was expected to be commissioned in January 1982. Production was expected to. reach 50 percent of capacity in 1982, 75 percent in 1983, 90 percent in 1984 and 100 percent from 1985 on. At full production, Polytex was expected to produce about 21.5 million sq.m of blended fabrics and 650 tons of poly/rayon yarn for sale to the Arusba mill of Kiltex. 4.0 Actual start-up of Polytel took place in January 1986, with a one- shif operation in' spinning, weaving and processing. Ini4ial capacity utilization (85 percent in spinning and 40 percent in weaving) vas. above planned production targets. In July 1986, the mill started operating three shifts in spinning and weaving and two shifts in process4ng but shortly thereafteF production was slowed down due to scarcity of inputs. During 11987, production continued to be slowed by lack of raw materials, dyes/chemicals and sizing materials, but production quality reached an acceptable level and encouraged TEXCO to export a small amount of cotton and grey shelpting that !yer. Although it had been envisioned at appraisal that capacity utilization i 1986 and 1987 would have reached 100 percent, actual capacity utilization was 17 and 28 percent respectively. The table be;ow gives actual production levels for 1986 through the irst five months bf 1988, and compares these levels to the levels planned at the beginning of each year: - 33 - Planned vs. Actual Production (1986-1988) --198- --1987-- -Jan.-ey, 1988- Panned Actual Planned A4tuaI Planned ActuaI Spinning - Blended 1,048,000 981,081 2,988,000 980,049 650,000 12.9is Cotton 0 89,830 927,000 423657 1,031,000 304,038 Total 1,048,000 1,050,667 8,915,000 1,408,705 1,681,000 817,036 Weaving - Blended 8,750,000 8,184,780 18,392,000 5,195,721 2,64,000 929,701 Cotton 0 0 0 6,98 I 056.876 Totol 8,758,000 3,104,7608 18,892,000 5,201,414 0,117,000 1,88,577 Ao. B Of 0tOl capacity 19.8Z 1817S 70.85 27.45 132.25 9.95 4.03 The operating performance of Polytex has been disappointing' compared to appraisal expectations. The most significant contributing factor to the low capacity utilization of Polytex has been the almost total dependence of the company on imported inputs (polytester, rayon, 4yes, chemicals and sizing materials) and imported spare plrts. Because of a lack of foreign exchange, Polytex has been unable to maintain sufficient levels 9f the necessary imported inputs. Although the bank sent several letters and telexes to government and project management expressing concern at Polytex' poor performance, no action was taken. 4.04 Due to the lack of foreign exchange, it v4s decided that all cotton fabrics and yarns would be produced for which the primary raw material, cotton, \was locally availpble, and for which demand, both domestic and international, was assured. As discussed earlier (para. 3.03), mill equipment was modified to enable Pqlytex to produce all cotton yarn and wovens. In addition, because cotton, although expensive, was at least available to Polytex, a cotton swap arrangement was introiuced. In February 1984, the Bank/IDA supervision mislion proposed to the Government the exchanging of part of the cotton to be allocated to Polytex for an equal amount of imported polyester and rayon, thus enabling some production of blended fabrics. The mission also proposed the, use of the price differential, between cotton and polyester or rayon, for importation of dyes and chemicals. While meeting the market and Government demands for blended fabrics, the implementation of such a proposal would also result in the saving of cotton consumption in the long-term due to the higher durability and less processing waste (about 7-8%) when using polyester instead of cotton. Unfortunately, although the proposal was favorably received by concerned officials, the arrangement propose# was allowed to operate for only one year. 4.05 In retrospect, making the project heavily dependent on imported inputs was not' totalXy appropriate, particularly given that it was not anticipated at appraisal that any -outputs would be! exporthd. Although the magnitqde of the foreign exchange shortages in. Tanzania could not have been predicted at -,34- the time of appraisal, some questions should have been raised about the wisdom of designing an import-dependent projec with (at the time)\ no prospects for export. Although the changes in project design made during Implementation have decreased Polytex' dependency on imports, the project remains import dependent. 4.06 Nopower Development. At full capacity, employment will reach 1,929, compared to an estimated ;,525 made at the time of appraisal. Average daily employment toward the 44d of 1986 Vas 1,270, versus an estimated total of 1,960 required to run thi plant with three shifts a day. At the end of 1987, 1,734 were employed, including general management, sales, service personnel as well as factory staff for a three-shift operation. 4.07 Market Development and Performance. The Tanzanian market for textiles, blended and otherwise, has not developed as expected at appraisal. Consumption of textiles is in general a functioi of GDP, and consumption of textiles in Tanzania has been much lover, by about half, than expected at project appraisal. Despite this slower growth and an increase in production capacity in Tadzania's textile companies, there persists a gap of nearly 65 million meters between demand in Tanzania and supply. To counter this, the Government in 1986 allowed imports. of textiles and second-hand garments on a restricted basis. 4.08 Polytex, like all other TEXCO companies, was obliged tq market its products through the Regional Trading Corporation (RTC), a parastatal under the Miiqstry of Trade and Industry. However, in view of marketing problems faced by RTCs, the Government has allowed TEXCO mills to market their products independently. Polytex has mareted directly to the armed forces, Government departments, hospitals, local fterchants and garment manufacturers and European importers. In 1986, about 13 percent of production was sold to the armed forces, government departments and hospitalsi garment industries bought 10 percent and textile merchants 77 percent. In '1987, Polytex exported 400,000 kgs of yarn and 1.0 million meters of cloth. A more dynamic marketing set-up for Polytex p oducts is needed. Development of export marketing would bb particularly important to balance required imported inputs. V. PINANCIAL PEFORMACE Financial Results 5.01 At he end of 1987, Polytex had been operational for only two years, versus six years as expected at the time of appraisal. During those two years, 1986 and 1967, Polytex incurred losses of TSh 802.801 million and TSh 4202.645 million respectivelyl the losses on operations before administrative, selling and financial expenses, were TSh 383.627 millipa and TSh 262.237 million respectively. Annexes\III and IV pr sent the income and balance statements of Polytex, as estimated at appraisal and as actually took placel projections for 1988 to 1990 were prepared by Polytex management in mid-1987. Highlights from those financial statements are presented in the table on the next page. It should be noted that the devaluation of thi Polytez: Selected Uad4cators of Vinancial Position *-*1988--- -1990--- ---*1982---- ****1984** * -*.1986**** ****1987-*-- Projections Projections SAR Actual SAR Actual SAW Actual SAR Actual SAR Current SAR Current o . ... .. moo., e --- ------ -- --- Current ratio 1.5 0.0 1.7 0.0 1.5 0.1 1.500 -0.148 1.5 0.3 1.8 - 0.2 Debts equity 1.2 1.2 0.7 2.5 0.3 -10.4 6.2 -3.1 0.1 -28.0 0.0 -1.2 Net sales 11 247.5 0.0 510.0 0.0 606.3 63.7 606.330 266.423 606.3 1831.1 606.3 1944.7 Net profit 11 4.6 0.0 87.0 0.0 122.4 -802.8 124.930 -4202.645 127.9 -1296.4 132.9 -1174.1 Return on sale(Z) 1.9% - 12.9% Z - 20.2% -1260.12 20.6? -1577.4? . 121 -70.8 21.91 -60.41 Return on equity (3) 1.32 - 10.8? - 17.92 - 15.83 - - 14.23 - 11.8? - Return on total assets CM) 7.11 - 23.81 - 27.0? -13.7? 25.7? -5.82 24.4? -13.3? 21.92 -14.32 Debt service coverage 1.3 - 2.8 - "-3.8 - 4.000 - 4.5 - 7.4 - 11 Notes Figures shown are in Tsh alitions. -'36- Tanzanian shillihg during the life of this project had a significant impact on the financial conditign of Polytex. The official exchange rate during project appraisal was TS# 8.3 to the US dollar; the shilling had not depreciated much by the expected date of start-up (in FY82, TSh 9.3:US$1), but by the actual start-up, the exchange rate had plummeted to TSh40SUS$l. In 1987, the foreign exchange loss alone for Polytex was P3,255 million. Excluding financial charges, the financial loss in 1987 was TSh 410 million. Given the magnitude of the losses, the calculation of certain financial ratios are irrelevant. 5.02 Debt servic4 coverage has not been calculated, since Polytex is making interest payments only on the Government loan, and is not making, nor does it intend to make, any repayments of principal in the foreseeable future. 'All interest due on loans is shown as "accrued interestO in the balance sheet and there is no separation of current long-term debt liability. From 1985 to 1986, 'interest unpaid and accrued increased from TSh 819.027 to TSh 910.918 mi4ion; it leapt to TSh 2,232.57 million iij 1987. Polytex does not foresee *an' improvement in this situation, and expects that by 1990, unpaid and accrued interest would amount to TSh 3,041.225 million, using an exchange rate of TSH90sUS$1. 5.03 Financial Rate of Return. The financial rate of return has been calculated for the project as shown in Annex V. Calculation of the FRR is based on actual investment costs, actual sales and production costs for 1986 and 1987, and Polytex' projections for 1988 on. The financial streams have been adjusted to 1977 dollars after incorporating changes in the exchange rate. A S0 percent salvage value for the plant has been assumed at the end of 20 years. With this scenario, a financial rate of return of -11.3 percent is achieved, in contrast to a FRR of 8.9 percent estimated at pppraisal. If all investment costs are considered sunk costs and written off,\the FRi for the 1986-1996 period is about 38 percent before debt servicing; 'wver, if estimated debt servicing is included in the cosi\streams, the FRR again becomes negative (-17.1 percent). 5.04 It should be kept in mind that even the above scenario is optimistic. Polytex's projections indicate a rapid increase in production, seemingly to near 1002 capacity. This woukd assume that many of the external constraints on production will be alleviated. However, in view of the long- standing presence of these constraints and the probability that\they will not and cannot be alleviated until the economy-widq industrial problems have been addressed, it is highly unlikely that Polytex will be able to meet its short- termiproduction projections. 5.05 Lpancial Prospects. The difference between Polytex' expected financial performance and its actual performance is the result primarily of two factors. First, demand for Polytex products has been weak, beaause Polytex prices have not been competit ve with iported -textile prices due mainly to the large increases in the costs of production. In addition, imported raw materials, which accounted for 75 percent pf inputs, became prohibitively expensive due to the devaluations and virtually unavailable. 5.06 The latest c mpany projections, which have been. used in t4is report, were prepared before the actual 1987 results were known and cover 31 - only the three years 1988 through 1990. ,In hindsight, these projection were very optimistic: 1987 sales were forecast at TSh 970.49 million,and net loss at TSh293.076 million, whereas attual sales were TSh266.423 million leading to'a net loss of TSh,202.645 million. Polytex management has made several attempts to address its current problems. It will continue to emphasize Production of altlcotton fabrics and yarns, which have a larger share of the domestic market and are more profitable than blended fabrics (in view of lower sales taxes). There is also potential for some export of all-cotton grey cloth. The comp#ny hopes that the cotton swap scheme will be re- introduced.' In addition, it has made the fpllowing recommendations to the Government concerning its fV4ancial structure, which at present is severely under-capitalized; Government should allow Polytex to restate all foreign currency loans, including interest, (except the Treasury loan) at historical rates. The differences arising from this r4stating would be borne by Government in the way of increased equity in the company. Government should convert the Treasury loan and accrued interest payments into equity. A r iratorium on loans and interest repayments gould be declared for tt :ee years, 1988-90, with a nominal interest of one percent during is periol., The balance would be considered for resceduling in 991; loans'would have maturities of 10 years and the same interest rates as in the originak agreements. Government would exempt Polytex for at least three years, 1988-90, from payment of import duty and sales tax on polyester, rayon and dyes and chemicals (if this proposal were accepted, the repayment perio after the proposed moratorium would be reduced t6 five years). 5.07 While the above proposals would helpalleviate some of the problems facing Polytex, they would not solve the underlying problems. However, it should be.stressed that the major problems affecting Polytex are the result of existing problems throughout the textile sector. To assist the Government and TEXCO in addressing these problems, the Bank is financing a restructuring study under the Bank's Industrial Rehabilitation and Trade Adjustment Credit, which was approved by the Bank's Board in December 1988. Phase I of the study would be a diagnostic review of the textile subsector, tncluding identification of isses at the firm-specific as well as subsectoral level. During Phase I, rehabilitation \and restructuring programs would be developed, with a specific implementation timetable. Phase III would be the actual implenientation of the action program, which is expected to take place over the next few years. iK -.38 - Vi. EONOMIC PERFORMANCE Economic rate of return 6.01 The revised economic rate of return (ERR) for the project i -18.1Z, compared to an appraisal estimate of 19.22. The main reasons for the lower return ares lower than forecasted capacity utilization due to shortages of imported inputs (raw materials and spare parts), and delays in project implementation. The calculations of ERR are shown in Annex VI. Financial cost and benefit streams were adjusted using economic shadow prices. 6otton prices were based on Bank commodity prices. Whenever applicable, transfer payments to the, Government (customs duties and sales taxes) were excluded. Foreign Exchange Effects 6.02 It was anticipated at appraisal that the foreign exchange effects of the project would be very favorable, and that by 1985, four years after commissioning of the project, accumulated net foreign exchange savings from import substitution would exceed the net foreign investment costs in real 1977 prices and in fact, generate not foreign exchange. savings. These savings have not yet materialized, 'and in view of the low capacity utilization of the plant, due largely to scarcity of imported inputs, are not likely'to. Tanzania continues to be a large importer of blended fabrics. Vi. INSTITU TONAL PERPORMANCE 7.01 During the implementation period of the project, wh4ch began in 1978 and continued to the end of 1985, Polytex was guided by two expatriate management teaWps. Input from local authorities, primarily through the project implemeftation unit of TIXCO, was minimal. When implementation vas completed in 1985, the management advisory fim took over and ran the firm until its contract ran out in June 1988. 7.02 Although the number of local personnel tFained and qualified to run the' plant has been increasing, TEXCO and Polytex expect that expatriate assistanci in key management and technical roles, through the MAF, wll be needed at least through 1989. Because Polytex has never been operated for any notable length of time by a local team, it is virtually impossible to isolate\and assess at this time the institutional performance of Polytex and TEXCO in the context of this project. It is also difficult to assess their performance in view of the many factors affecting the implementation of the, project and the operations of the mill that have been outside the control of management, and the constraints imposed on them by a lack of qualified personne\l and frpquent changes in key staff.\ The performance of local tanagepent, specifically of TEXCO could have been significantly improved in procurement. However, current Polytex and TEXCO personnel are very involved in and concerned by the current problems of the company. 39' VIII. CNCLUSIONS 8.01 Polytex, the first integrated textile mill in Tanzania capable of producing blended fabrics and yarns, was set up under this project to help meet the growing domestic demand for blended fabrics that vae proving to be a drain on the country's foreign exchange resources. The plant has proven capacity to produce quality blended fabrics suitable for domestic, and eventually export, markets. However, because of problems largely beyond management's control, Polytex has virtually no prospects of operating profitably until these problems are addressed. 8.02 Fe,* of the major problems facing Polytex are firm specific and solvable by the company. The major problems, which are subsectoral (textile) and sectoral (industry) arre being addressed under the Government's Economic Recovery Program, and more specifically under the Bank's Industrial Rehabilitation and Trade Adjustment Credit. Lessons to be Learned 8.03 Better judgment could have been shown by both the Bank and the Government in designing the project. More questions should have been raised about the wisdom of designing a project to be heavily import-dependent with (at.the time) no prospecs for export, and sources of the foreign exchange needed for the initial years of implementation should have bee" better identified. As it turned out, the needed foreign exchange was not forthcoming from either the Government or the project. 8.04 It should be noted that when problems arose, the Bank and project management showed flexibility in addressing them. The conversion of the plant to all cotton was achieved rapidly and at a reasonable cost. The development of the cotton,swap scheme was resourceful, although due to lack of Government support, the swap has had only minimal impact on\the project. 8.05 The training component for this project, which was on a considerably larger scale than under the previous usans, Textile Project, was also well- thought out and implemented; industrial operations and product quality attained a high lovel not found in other TEXCO units within a short time. Use of expatriate 'staff has by and large been successful, although it is clear that Polytex still cannot be operated without expatriate assistance. One clear lesson to be learned from the project advisory firm is that contracts with consultants should be performance, and not time, based. 8.06 Because of the difficult country conditions, more than normal supervision was required, particularly concening procurement 7atters.. It, is therefore recommended that the resident mission resume the\practice of' conducting regular monthly meetings with project entities -and other goverpment agfncies involved with the project. This practice was effective during the early years of impementation in resolving od tstanding issues among local agencies and alerting project officers of current or potential problems that need immediate attention. PROJECT COMPLETON REPORT MORQGORO TEXTILE'PROJECT (Ln. 1607, Cr. 833) Schedule of .Pro ect Implementation Site Proration Vorks Auzxilia.ry Bildin .- --- Main Civil and Buildiåg Vorks Structural Ste4P5 Vorka -Eectrical 1orka Air-Conditioning Vorks Utilit §aui nt - -f-flu-enft - f - - - etent -- - ~- Erection of Pro- duction Machinery- A8reed Implementation schedule vith PEF. E egual Implemetation. 42 - Pae if 2 NWO3ECT tNPLET1n PORT E~gero loxtile Project Wr. 833-1A, L,n. 1607-TA) Schedule of Disuremnats Cuniatative 0issurseoents Flsad year Appraisal Actual to and quarter Estieatus Actqal Appraisal (D1 32.30 0.00 0.0% Narch 3 5.30 0.00 0.08 June 3, 8.30 0.00 0.0% FMN SptmØpr3 10.30 0.92 8 91 8agesba 31, 14.30 1.34 9.42 Uarch 31 1e.30 1.42 7.91 June 30 24.60 1.67 6.09 'Fat 8eptum0e 30 30.40 2.25 7.42 0ecesber 31 3É.60 2.44 .22 rth31 35.0 2.77 7.81 ie 3 39.30 3.23 L2& FMS 39 43.30 5.1 £3 64 DG~ 31 44.30 J4 14.\ Uarch 31 44.40 9.S4 21.7% Juni30 44ý40 9.97 22.5U FM \. K s$emer $0 44.50 11.05 24.8% Datunbar 31 44.50 1.84 31.1l Narch 31 44.0 - 13.95 35.05 Jun 30 \ 44.40 19.73 44.22 -43 - Page 2 of 2 CueuIative Otsburseaents . ...... ... f a.. FiMal vear ppraitatActuel to dquater Ettleates åcioål Appraillat hsptenbor 30 44.70 20.93 46.81 Deceaebr 31 44.0 22.33 50.0% arch 31 4480 23.43 52.3% is 30 44.90 23.19 53.31 FM'5 Sept~aber 30 44.90 24.53 54.6% Øcbar 31 45.00 25.83 57.4% Øarch 31 45.00 2713 60.31 Jaes 30 45.00 29.86 6.4 septe r 30' 45.00 31.36 6.7% aec~er 31 45.00 35.61 79.11 Nwcb,t 45.00 36.76 e1.7n i= 30 45.40 \ 37.57 83.5% lepteer 3 45.00 39.34 øL4, 2% geubhr 31 45.00 38.40 85.31 larch 31 45.00 39.90 98.72 \= 33 45.00 39.90 ia.7 FM8 Sptber 30 45.00 41.03 91.21 Rucetr 31 45.00 42.53 94.5 uarch~ 31 45.00 45.00 " 100.01 \ \ \- ,44·、ANNUI華1 ×實乏于付于括鬥r－一 Pa卯lofZ l \ 望！細發糁日藝曆藝藝I讓露響發藝韋霆餐發各屆藝19藝賽遺啟騷鐵發日l合 。絕！悲言二二奮二二︰l露二：:：二麾：:：言：!::：兀盡乏黑亡江I: ．響j！蘆ו鳥。。•，。！發”。。發乏丰’”。。！。”囊甲’&,&，。‘× 蒼＿&_____________________;_ 〔｝-I乏i豐蠶整豐擊！I雲豐豐曇豐豐：：豐豐尊擊邑―】，11豐｝纏 :！芝×婦秀‘為遜造織！基露寫寫洛騷藝言騙邇•藝登釁I唱騷奪絕若馳擊寥渥I鰓‘ 。曆I方忽江江：二：︰l:：忽言乏忽忽：::：言：:I：忽：︰三：:::｝方 遲盔l逐藝編•審•寫”I參叉。”‘。。徐安且‘寫。”I•’蔆華“&,’萬！丰 奮一×＿_______＿、＿___- 才蘆―華鳥審彎離名騷彎I醫響奉審賽賽狀華離華鳥翁藝―娥彎離賽權且 萋易―&&,&～。’‘秀”。’。。‘騙，，。”&&！纏騙“、！‘× 言！購鳥騫禺矯輿賽賽―彎賽離離攤彎攤彎攤邊群戀寫l賽賽彎鳥嗚藝翁纔禺！視 ．藝j―賽擊視寫尋粈寫。！身’。’&，。•提日’，。！擊’賽甲名‘,&,l鳥 ＼雙I萬 賽！變！g酒I星騷彎粅甚離巷彎I整離濺賽賽華華攤寫攤賽攤彎騷！攤賽彎賽！攤 、、豐！留！，基！’、‘&”一”&，、”&&，。。：：。，。。！，。。·，。 方！爹I，蔆晝’釁、糅基轎糅尋藝規I露華藝涌靈譽騙編言藝言騷蠶號I言、戀藝馳認擊藝籐l造 蠱―震―｝邊！&&‘認‘撾”!&&&&&．滷織，&&&-‘騰‘每×膩鉉幼‘!& 蒙！谷―呂―奪廈―寫鳥寫鳥離鳥離露l離彎趣攤離賽彎彎離寫藝攤攤禺l攤離寫賽、I§ 爹i!I邊11倡！騷禺號，•”。•！秀禺。。。o。視零，o•o•馱秀翁露。’｝。 戶’不屆纏：蘿C言’、 必：,;___--------- 至I雙×I！唱！日藝酒遺二徑屆a！嗯露屆日群露雙蘿江盔著騷翁寫I尋醒細寫’騷馳騷遜I擊 露！馴I露l•“。•。“!.&‘。‘&‘藝妒“&I藝纏專紹纏頗‘I很 亂〕―韶、攀螂鰓攀綱劉11&!,\ e:&‘• 二渥―賽‘、彎華彎攤鳥離馴發朋離騫變寫究登震離攤賽藝―攤離寫攤I戀 擊劇粤為號寫男～。，隱號”。。。。””•。•I鰓婦編甲I。 i引一‘、 江廈―騫騷離騫賽賽攤！買朋離藝寫藝發攤安各鳥各騷I遺、曇藝翁I名’、 季引寫、”闢‘&I藝、”。。“&，神。•｝翁＼號森‘}& 暴引”×× 入、！I！豐飄蠍誹’攣鉀I&i莖”―擊、 邊廈I鳥、藝•華賽為雜基l攤雜離遲霆藝§發擊尋藝號各緣I寫離震織言藝藝藝！鳥×． ’藝侈I寫、，響卜，。‘&I甚跚“&‘滷緣“&&！織著‘&‘戶‘戶｝& 邊《L一：二一－一一＼、，〕 一。。。。甲恤。，＿、瓏。＿、 、，，、亂I》弱蹺排呈拼！h劉畫―矓t ，×、。：觔，二者．蒲記三；劊記望蒼徑莖蒲I唱荒當群、：低勰劉唱―震豐罐 ,”汰權〕惡縫峰馨弱馳藝曄暹―h蒲呂驕叢望廈鑪I打居計龍；―奮！鰓權遺、：。 、．”露！j州”綴“’縱m州謝排藝‘挪‘引I矓聯鉀＼ 、’&&&’、N\”蠶•尸‘。 ＼·入k \｀、‘＼、， proj~ P~J«md Pr*JG~ pfoj~ "0~ P-åo~ APPMODI OWM04 AMMIGD, Cormnt Affroicot currcot sele in@ 0.116 0.58 0.429 on~ O.M O-OM 8.M 0~ 11~ 8.240 A4~ 0509 O.OM l.M &.NO 1~ X.CM »."3 18.444 18.4m 41.10l 0.748 0.80 0.1315 2R.«r 0.-m -7.-m -7m -0. zo» 77cm ~101 0~ 3~ 4 cp lc~. J 0»~ 4 M.4119 48.4m 48.~ 40.m 94.m OWM5 29.= M.485 ffi.IIM OL919 19.= BU.44%) la~ au.WQ ~ Ck~ *.CM G.1W G.M #~ 0.M l.us It~ 4.8w 4.= 4-m Gm~ c~colcm O.OM 0.0111111 CM O.GICk 0~ O.OMl 97.~ 4511.« 46.9» 45.0W O.M OMPI) 9~ T~$ JM.«» 40~ 49.~ 40.= M.M SNAW mm SM.~ O.M 811.41*9 18.= mo.~ la~ 818.2W orkum 0f olm O.CM 0~ *ÄM O.M 0.~ *.ou Mwo q~ 0~ 0.0m 0.~ tis.M =Am 2MAM ^$72 90.6W ^wo mäsý -1m~ mm -me.en g~ -^on *.M O.M O.OMI *.m *.en *.M 7~ O.CM Il-UD M.~ in.m mmlo ~ .84 20,~ -1MÅN 2».M -IM.ön 20.1W -1174.091 4.409 81.445 O-M Ut.«§> M.4M M.2173 51-M e&.«5 U9'4» M'4M M.M IM.00 1~ divid~ 9~ O.OMI 19.«» 18.8» 101.10: Ig-M AfflUtm f» ftwimd .^44 ».*5 M. 1% im. m A~§ M«ffm *M f~ ~Ud ~ to. ftlyt« 41d mg m^ILIr4 «fkål IM. four rum ~ad 0~ 10. ftqj«*~ förl~ om f~ 0~1c01 R~ . fw IIIwim wo f~ polv~ ~ W ofim. fil,Roi &&&.&’、．、· ’一心6一越個廈X IV 于’、文·Page 1 ofZ _I邊―111豐―11擊豐豐彎！l暮l雙彎藝I奪魚透彎―l&- ·l〕！,:I,-:11-11：華豐，,,,,,-I;,,”、”叢！造 -I】〔】11奮―l、I華叢擊擊―11；叢擊藝I蠶藝華擊―l、― 、’｝〕I〕〕矓矓―〕！蘿〕〕！蘿〔〕〕：！。 ；。。！劊”,,,-,,，磨”,!,｝、，,,&,,,,-&&,&,’馴； .,l〕｝〕｛-;l;-i;!;；蠶―!}l！蠶；蠶；}!!；蠶！;-！、 霎I森I蔆1．。一。二。－...-.-－一、－-----－一！ ”弄！l!i廈―’莽婪，-:.,l透豐豐三―墓；豐奮｝認震離驕曄曄―I；擊曇奮、靨豐―薯 鬍懿！獸量蘿畫，-;!!&&-!；乏釁忽纏擊擊豐賽，-;;,&’酌｝ 韋！}!！廈―曆曆豐藝―曆曆曆l蠶雲藝―11豐擊邊豐會豐豐豐豐―雲！I豐雲濫邊擊！I !!―豐奪】擊！莖11擊擊擊擊―l:I忿透黑縴擊擊擊彎―I；擊豐擊擊l-I ，、｛〕；】；};;；】〕認｝l&;;；〕；調；;｝〕｛〕、I！、 ·、！！〕；!！〕！!!!!!!l!111〕、〕11!！〕！ ，廈―磨擊彎擊―縴擊·豐當雲！擊―韋透豐豐豐豐豐豐豐奮豐―覃豐盡豐邊會豐！發 ．。一：；〕。！}’〕：―〕愚｝’〕’〕＇,―】雜l&{-，媒鄴！〕’〕 --- - ------ - --- --- ...... ...... ...... ~~6 ~ ~~ 4~1 1 Amnifflot chwe~ ~1490 c~ ~1600 c~ .. .. . . ......... .. lamt =12. Ta om* 41 0~ 12.010 22.478 0 am 22.41C o~ 72.9m C.Olm O.CM 72.995 01= 75 m 0.«»' IC.CM 0.00 0.00 ma. 0.0x: o~ 0~ 2-om 0.0m G:CO6- o.om Cl~ - OMID O.1W 01~ dohok 22.478 o~ 96.40 72.m 0~ 72.m 12.m &.GO W«» X:.M 78 : m 0.0w OMX) v~I L-t 4*4 omie iis~ mmé «w.= miu co-an to-m c~ m~ mm em= m.m om~ In.mo C~Ä-T ~ 0.01: O-M 0-015 85-00 li-CM N.~ 13.70 0.~ 25,10 N.~ 59.04 2=1.6W 89.00 84.2m 2».0» -w;-- ;z- - - - - -- ;;:'; 4*4 M.;; m ;;m " ;;;gw 5;;7 off.7m sa.mö M.M M-m o~ GD.CU off~ 80.7m ~ D '30.7m Cm.Wo 2147.7m MO.M ~.7m 40.moo d~ 82.0101 IM.911 45~ =.41215 0.4110 MM D.410 Mtig 0.«» MM 0.001: kol-l% O-M 0.00 =.70 OMID CM 4~ 010» W~ *-M M.M *~ M.= omlik m~ ~~ 441.m ~ .4d7 M-40 M-90 M.20D ~ am opsciaft amt OAM 6.00 O-all: 0,00 0.0m 0.0119 0~ 0-10 0~ 4.0»ý 0~ 0~ 9.60 0~ 110.= 0.00 to.M 0~ m~ ~.170 M1.11414 w~ lus-cici up.m sm.ug ^om Iffl.m 1t45-m 7~.0n i00.7w 4~ We.= am.Wo 1/ =41m4 fiffimlof aha~ . . 0~1 wed ~ch, 90 10 j~w~ 6~ nu d~ 9~ m4 amid 1" 70~ off wo# oftw ~ at M~GVoi or~cn. <w, ~~. @om 10 10 10 ~0 #ma iämjo. 10 owemb by f;ma ~ m fiffiml m~. 11-99.610 1- lo ~~ #ma &MåffiB. 10 " 14~ . s/ 8~ by ~mm. ~. 9~19 &4 1 ~~ OMJO. 0~ 1 WO~ 4~10 9~ 10 ~ro fna t~. ms i by omgammy ~. n ~ ro f. 9~ . im~ 0~ MW k~lå lad. 00 (D rn 48- Nk -40- PROJECT CMLET10# REPDRI TARIANiA- RoRoSORO TEXTILE PRMECT tt. B33-TA, Ln. 1407-TA Financial Rate of Return Capital Operatng kmnef its Salvage bet linefit Year Cost Streatn Cost Streae Streas falue Streaos 1978 0.000 0.000 0.000 0.000 1979 0.000 0.000 0.000 0.000 190 3.195 0.000 0.000 -3.195 1981 3.mS 0.000 0.000' -3.1% 1982 9.50 0.000 0.000 -9.585 193 17.942 0.000 0.000 -17.942 198 20.891 0.000 0.000 -20.891 19m .602 0.000 0.000 -8.602 1986 2086 4.738 1.761 -27.064 1987 6.882 4.036 4.143 -6.774 1988 10.077 18.940 19.2T5 -9.742 1989 6.14 19.347 20.41 -5.021 19% 0.000 18.01 20.471 1.620 1991 0.000 18.851 .471 1.620 1992 0.000 18.851 0.471 1.620 1993 0.000 18.851 20.411 1.620 1994 0.000 18.851 20.471 1.620 9 0.000 18.051 20.471 1.620 0.000 18.851 20å7» 22.120 23.740 la cmnstaut 1987 dollrs. 20Z sulvage valus ANEX VI -49 PROJECT COPLETION REPORT TANZANIA N000R0 TEXTILE PROJECT ,Cr. 83 TA, Ln. 107-T. Econosic Rate of Return Capital Operating Benefits Salvage Net Benefit Year Cost Streaes Cost Streans Streas Value Streans 19860 2.469 0.000 0.000 -2.469 1981 2.470 0.000 0.000 -2.470 1982 7.437 0.000 0.000 -7.437 1983 14.009 0.600 0.000 -14.009 1964 16.665 0.000 0.000 -16.665 1985 6.921 0.000 0.000 -6.921 1996 20.564 3.067 0.796 -22.835 1987 6.167 2.313 2.422 46.058 1989 9.238 15.778 13.564 -11.452 1989 5.632 16.215 14.405 -7.442 1990 0.000 16.382 14.405 -1.977 1991 0.000 16.382 14.405 -1.977 1992 0.000 16.382 14. 05 -1.977 1993 0.000 16.382 14.405 4.977 1994 0.000 16.382 14.405 -1.977 1995 0.000 16.382 14.405 -1.977 1996 0.000 16.32 14.405 20.500 18.523 In constant 1997 dollars. 201 salvage vale Capacity f aon 1988 an assned at near 100 percent. -51 ATTAC~MENT Page1 of 8 RO RO PQLYESTER TEXTILES \LTD P.oBOX 269 Telephone 4340/7 MOROGORO T~la x5083 -MORTEX >OENT4 ?tECEIVED FROM POLYTE our Ref. I£M/GW5/9/7/90 Daom: 4th April, ý0 Mr. Rene Ribi, Aoting Division chief, Policy Based Zendn Indastry >ablie Utilities and Urban Seotors, The World lank, 1818H Street N.W. 1JOS.AO Dear Mr. Ribi, Re: MROGæRO PM|STR JROJMT (IOAN 1607/CREDIT 833-TA) I wish to acknowledge the receipt of your letter of Pebruary 28th, 1990 in which the Draft Prpject Per- formanoe Adit Report was attaohed, and y asked for coemante. 1closed hørewith please find y coments on the report. You should please note that it writlng this rep~it I have the benefit of a hindsight and slap the ~rience of actua1,y "rking at IMYTEK fot sligtL re than one year nov. Por this reason, there are some areas where I have made observations different from those in the report. Like you, I have distributed y comments to all' who have received the draft report from yon. I am also looklmg forward to receiving the final report. Tous faithfally, MROO P0LESR TEtIL ITED. Ao Wawm~a GEEi!AZs ~$AER copy to: Prof. S. nbilnyi, t ecretary, of r cinane, D.0.AB 9111, DAR E S ALAAL ,/ 52 ATTACHENT Page 7 of8 Jopy to: 'lr. J. Sepe#a Permanent Secretary, Ministry of Industries & Trade, Po0. Box 9503, Dar es Salanm. "s "s Mr. a. Hi. rncya, Maning Director, llational Textile3 Corporation, 2.0. Box 9531,. DAR 33 STAM kW/gem -53 ATTACIMENT Page 3 of 8 MOROGORO POLYESTER TETILES LIMITED COMMETS ON THE WORLD BANK DRAPT PROJECT PERFOM ANCE AUDIT REPORT 1.0 INTRODUCTION The comments of the draft Project Performance Audit Report shall cover only those areas which POLYTME is at variance with the presentation in the report. There has been an attempt to make the comments on the points in the order appearing in the report. 2.0 CHOICE O TECHNOLOGY: It is apparent that after careful consideration, the Bank settled for "modern but conventional technology" which given the prevailing conditions in Tanzania, was the correct thing to do, However, the choice of techno- logy does not mean choice of machinery within this given technology parameter was correct. There are two areas where the choice of machinery could have been better, these pre: (i) Carding section where significantly the supplier never bothered even to erect the equipment supplied because the company/TE=CO failed to pay the insta- Iments due. As a result the equipment was ereced by the civil works contractor. The cards themselves are old cards which have been reconditioned. It is not surprising, therefore, to see that during the early stages of operation, there were excessive breakages of parts in the carding section, as the older parts broke down. At least 9 out of the 41 cards never worked, (ii) Loomshed where a number of looms have never been operated, whether they were commissioned oaf not that is not the main point at issue. T4e signifi- cance of this is that the operable capacity is different from the installed (bought?) capacity. (see 4.2). *9/2 - 54 - ATTACHMENT Page 4 of 8 Besides the reduction in the operable capacity, the loom speeds are lowT Given the high cost of energy in Tanzania, and considering the fact that over 50% of the electricity in the factory is used for "climate control" rather than motive force for direct production., it would have been economical to ablect high speed looms, Which in any case fre available within the tech- nology acceptable to the Bank, 3o0 PROJZUT MANAGEMENT: 3.1 The overall set,up of the, management may have been appropriate, given the fact that the basis for the selection was previous performance elsewhere, While not faulting the selection of the Project 2-ngineers, one can surmise that their presence on the team influenced the selection of machinery and equip- ment, most of which are of Japanese origin. TAe Banks coments on the Project Advsory PirM (PAP) are apprp- priate. However, there is a difference ot opinion on the Management Advisory Firm (MAP). There is no doubt that in 4urope, and perhaps elsewhere, the WAF are a reputable and high performance company, but it would appear, judging from low calibre of most of the manpower which were deputed to Morogoro as well as the frequency which they were otated from the project, MAP gave this project very low priority* To top that, MA' s lack of familiarity with the Japanese machinery and equipment means that during the early stages when production started parts breakages were high and fabria/yarn quality poor* 32 The o erseas training program was successful, but 1he ret ltion rLte of the trainees has been ,ow, at less than 20% of the original traineeso Contrary to the repoA, those who trained abroad,never took part in tia g the local staff. In any case the training of oe../3 ATTACHMENT Page 5 of 8 local staff was more of quantitativ4 statistics rathe than the actual trai;ning; otherwise iAP whuld not, during the third year of the contract, have to resort to excessive reliance on expatriate staff for low supervisory work, The\ asumption that "middle management would come froA the existing textiledmills was not realistic because there was no, and still there isn'tiany institute for training in textiles. .ven the existing mills at the time of apprais l mission had skilled aanpower deficiencies. 3.3 The dati on cost, disbursement and P-inancing reveal that provision of working capital was \nfortunately given very low priority. The company up to now has o live with the negatve consequences of inadequa4e Working Capital at the start of operations. The low capacrIty utilisation can also be attributed to, in addition to the other reasons given in the report., \inadequate working capital. The problem of lack of \ raw materials iz'iessentially the conseq ence of inadequate working capital. 4.0 OPERATG PERPOWRANCE: The rep rt correctly points out that the "operating p rfoace has been disappointing compared to the appraisal expectations". Apart froi the reasons given in the report, other reasons ghich contributed to low capacity utilisation include the following: 4.1 LLOK OF W7RKIMTG CAPITAL: The appraisal mission provi io for working capital was Tahs. 73mi/= and USD 5.7mi, where as the acturl provision was Tahe 35mi/=. Hence, apart from being devastated by the fUll ffects of devaluation, the compahy lacked wherewithal to purchase te necessry\ .../4 ATTACHMENT 56 -Page 6 of 8 inputs from the time when production started. The company had to adopt unorthodox finance practices whose consequences continue to affect the company till now, and shall do so in future unless some drastic measures are taken. 4.2 OPERABLR CAPACITY: The actual opepable capacity was significantly lower' than the installed capacity especially In the critical sections, i.e. the carding section and the loomahed. 9 cards out of the 41 supplied and 86 out of 516 single shuttle looms and 26 out of 52 4x1 'color looms have never worked, either by not being commissioned or lack of parts during drection. The loom speeds have been reduced by 7% on the average. There are also 1pe*heral" capacities which have never worked, (combers, roving, ring frame,. cne winding electronic clearers etc.) but the effects shall not be evident until capacity in card and loomshed is rectified, 4.3 MANAGEMET: Given the prevalence of the Japanese machinery and equipient, a Japanese firm should have been Oelected as AP. A choice of 4 European firm, as was the case, resulted in excessive breakage in machinery lod equipment (and thez\efore high parts consumption). Contrary to the claims made in PPAR, production never reached "acceptable quali* levels". It is true there wer4 some exports made, but the e resulted in quality claims from the foreign buyers which the company has to settle., MAP, who buy grey cloth for their other compa- nies, never dared to procure grey from'Morogoro. The yarn exports, an the other hand had fewer export claims. \Qn the local scene, the company held excessir (relative to production) stocks of grey cloth, some 'of which had to be sold to a local textil*s producer at near donatiomal prices, The reason for these grey cloth pile up was not 57 Page 7 of 8 veaik demand, as put in the report, but poor product oua.i.ty. Admittedly, the local production quality to a degree may have been les of NAF performance and *oie of as a consequence of the choice of machinery, but 4AP who commissioned the ill should have mentioned the equip- ment deficiences in their commissioning report. The substantive difference between plan and actual performanqe in the two tnd half years is attributed to th fact that the AF contact called for production at near capacity levels by 1988, hence the production klan was propared to meet the reqiirements/pro sions of the management contract rather than the realistic assessment of the prevailing conditions. 5.0 PINANCIAL PERFORMANGE: The financial performance has been bad and is a result of a-combination of factors including low captity utilisa- tion, devaluation of the 'anzania shilling, non repayments of the loans/interests \due and ofcourse the "excessive investment cost" which could have been reduced without materially the key project parameters, ege the excessive civil works one sees around POT. The company has already made some recommendations to the goverament which have not been fully taken up. Howeve, in concurrence with lPAR, these recommendations, "though alleviative", wold not solve the underlying problems. The key is in the.impro4\Wnt of capacity utilisation ,,through the scrapping of all the unoperable capacity in cards/weaving and gabstituting the same by high speed machinery capable of producing export quality products, as welU poviding the necessary working capitals To .some ATTACHMENT -58- Page 8 of 8 oxtent the World Bank diagnos-ic review team, which han been workin: Tanzania haa touchod on thoue aubject'3, týhoughi thoro in utill some variancc in opinion on their concluastiong. MOROGORO POLYETE TTILES TILrTED 31st IJarch, 1990