Somalia - Trans-Juba Livestock Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 4438 PROJECT PERFORMANCE AUDIT REPORT SOMALIA - TRANS-JUBA LIVESTOCK PROJECT (CREDIT 462-SO) May 4, 1983 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. WEIGHTS AND MEASURES 1 hectare = 2.47 acres 1 kilogram 2.2 pounds I liter (1) 1.76 British Imperial pints 0.88 British Quarts 2.11 U.S. pints 1.05 U.S. quarts ABBREVIATIONS BTO - Back to Office Report DCA - Development Credit Agreement GOS - Government of Somalia IDA - International Development Association IFFF - Irrigated Feedlot and Fodder Farm KMF - Kismayu Meat Factory LDA - Livestock Development Authority MOLFR - Ministry of Livestock, Forest and Range NRA - National Range Agency OED - Operations Evaluation Department OPS - Operational Policy Staff PCMU - Project Construction Management Unit PCR - Project Completion Report PDTU - Project Development and Training Unit PPAM - Project Performance Audit Memorandum RMEA - Regional Mission in East Africa SAR - Staff Appraisal Report TJLP - Trans-Juba Livestock Project WDA - Water Development Agency FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT SOMALIA - TRANS-JUBA LIVESTOCK PROJECT (CREDIT 462-SO) TABLE OF CONTENTS Page No. Preface ............................................................ i Basic Data Sheet ................................................... ii Highlights ......................................................... iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. SUMMARY .................................................. 1 II. MAIN ISSUES .................................. .......... 4 A. The Irrigated Fodder Farm and Feedlot (IFFF) . ................................. .......... 4 B. The Project's Influence on Stock-Owner-s Income .. .................................. ......... 7 C. The Veterinary Clinics ............................. 9 D. The Scope and Extent of the Project .................. 9 E. Conclusion ......................................... . 10 Attachment 1: Comments from National Range Agency ................. 12 Attachment 2: Comments from Trans-Juba Livestock Development Agency 14 PROJECT COMPLETION REPORT I. Summary ................................................. 16 II. Background .............................................. 19 III. Project Identification, Preparation and Appraisal 23 IV. Implementation .......................................... 29 V. Project Impact .......................................... 41 VI. Financial Performance .................................... 44 VII. Institutional Impact .................................... 52 VIII. Bank Performance ........................................ 55 IX. Conclusions ............................................. 59 Annexes Maps This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT PERFORMANCE AUDIT REPORT SOMALIA - TRANS-JUBA LIVESTOCK PROJECT (CREDIT 462-SO) PREFACE This is a performance audit on the Somalia Trans-Juba Livestock Development Project in the southern region of the Somalia Democratic Republic, for which Credit No. 462-SO was approved on February 5, 1974, in the amount of US$10 million equivalent. The project was closed on December 31, 1981 by which time US$9.2 million equivalent was disbursed. The balance was cancelled in June 1982. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department (OED) and the Project Completion Report (PCR) dated November 29, 1982. The PCR was prepared by the Resident Mission in East Africa, Nairobi and reviewed by the Eastern Africa Regional Office. The audit memorandum is based on the Appraisal Report (261a-SO) of January 10, 1974, the President-s Report (P-1333) dated January 21, 1974 and the Develop- ment Credit Agreement (DCA) signed on February 20, 1974. Another key document perused was the report of a Review Mission dated May 10, 1977. Internal Bank memoranda, as contained in project files in Washington and RMEA, Nairobi, have been reviewed and staff associated with the project have been interviewed. An OED mission visited Somalia in August 1982. The mission toured the project area and held discussions with the Project Manager and the Agency now charged with overseeing the project - the National Range Agency (NRA). The project executing agency was disbanded before the project was formally closed and its officials were absorbed in the NRA. Discussions were also held with senior officials of the Ministry of Livestock, Forest and Range (MLFR), including H.E. the Minister and the Vice Minister. The information obtained during the mission was used to test the validity of the conclusions of the PCR. The audit finds that the PCR covers adequately the project's salient features and generally agrees with its conclusions. In addition to summarizing the objectives and achievements of the project, the PPAM, further discusses the concept and trial performance of the Irrigated Fodder Farm and Feedlot, the project's influence on stock owners- income, the veterinary clinics and, in a broader aspect, the scope and extent of the entire project. The valuable assistance provided by the Government of Somalia and, in particular, the Ministry of Livestock, Forest and Range, together with traders and stock owners with whom the mission held discussions, is grate- fully acknowledged. The draft report was sent to the Borrower on January 3, 1983 for comments. Comments received from the National Range Agency and the Trans-Juba Livestock Development Agency have been incorporated and are attached as Attachments 1 and 2 to the PPAM.  - 11 - PROJECT PERFORMANCE AUDIT REPORT SOMALIA TRANS-JUBA LIVESTOCK PRCJECI (CREDIT 462-SO) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as % of Expectations Current Estimate Appraisal Estimate Project Costs (US$ million) 11.5 10.6 92 Credit Amount (US$ million) 10.0 9.2 92 Date Board Approval 02/05/74 Date Effectiveness 06/21/74 10/29/74L Date Physical Components Completed 06/78 All incomplete Proportion Completed by above Date 100% 30%/b Closing Date 06/30/80 12/31/81 Economic Rate of Return (%) 25 negative Financial Rate of Return (%) 14 negative Financial Performance unsatisfactory Institutional Performance unsatisfactory Beneficiaries (nomad families)Lc 20,000 not known (negligible) Meat Production (tons) 3:900/d not known ',egligible) Cattle Offtake (head per annumn) 52,000 not known (negligible) CUMULATIVE DISBURSEMENTS FY74 FY75 FY76 FY77 FY78 FY79 FY80 FY81 FY82 Appraisal Estimate (US1 million) 0.2 2.0 5.0 6.5 8.0 9.5 10.0 - - Actual (US$ milLion) - 0.3 0.6 2.2 3.3 3.8 6.3 8.1 9.2 Actual as percentage of Estimate - 16 13 33 41 41 63 81 92 Date of Final Disbursement: 06/82 MISSION DATA Sent Date No. of Manweeks Specializations Performance Types of Mission 0y (Month/Year) Persons in Field Representem/e te Trendl Problems/h Identification RMEAL0 01/70 1 2.0 - - - - Preparation RMEA 11/70 2 1.0 - - - - HQ/R1EA 12/71 4 5.0 - - - - HQ/RMEA 03/72 6 6.0 - - - - RMEA 1i 06/72 3 6.0 - - - - RMEA A 08/09/72 2 4.0 - - - - Appraisal HQ 02/73 b 24.0 - - - - Pre-effectiveness RQ/RMIEA 05/74 4 8.0 - 1 - - Supervision I HQ/RMEA 02/75 3 1.5 - 1 1 - Supervision II RMEA 09/75 1 1.0 - I 1 - Supervision III RMEA 03/76 1 1.5 -- - - Review Preparation RHEA 04/76 2 4.0 - - - - Review Preparation HQ 11/76 1 1.5 - - - - Review HQ/RMEA 01/77 5 15.0 - 1 2 F, M Supervision IV RMEA 07/77 2 3.0 - 3 1 F, Supervision V RHEA 11/77 1 1.5 - 3 1 M, P, F Supervision VI RMEA 03/78 2 3.0 - 3 1 F, Supervision VII RMEA 08/78 1 2.0 - 3 2 M, P Supervision ViIlI REA 12/78 2 3.0 - 3 1 N, P, F Supervision IX RMEA 06/79 1 1.5 1 3 2 M, F, T, P Supervision X RHEA 01/80 3 6.0 ae, eg, 1 3 2- Supervision XI RMnEA 03/80 1 1.5 1 - , F, , P Supervision XII RMEA 07/80 1 1.5 1 3 2 M, F, I, P Supervision XIII RMEA 03/81 1 2.0 1 3 2 M, F, T, P Supervision XIV RMEA 11/81 2 3.0 1 3 2 , F, T, P Supervision XV/PCR RMEA 02/82 2 3.0 - - - - OTHER PROJECT DATA Borrower Somali Democratic Republic Executing Agency Livestock Development Agency!L! Follow-on Project None/ Note: Footnotes on next page. - 1.11 - /a Two postponements - to 08/21/74 and 10/29/74 (actual). /b An estimate only. No project components completed but some reached an advanced stage. /c At full development of Project (Year 6). /d Incremental production from project per annum. /e eg = Engineer; ae = Agricultural Economist; 1 = Livestock Specialist. /f 1 = Problem-free or minor problems; 2 = moderate problems; 3 = major problems. /g 1 = Improving; 2 = Stationary; 3 = Deteriorating. /h M = Managerial; P = Political; T = Technical; 0 = Other. /i As observer with USAID mission. /j Ful.l preparation mission. /k Follow up on preparation prior to final draft of preparation report. /1 Of the Ministry of Livestock, Forestry and Range (MOLFR). /m LDA was abolished in 1981; the National Range Agency (NRA) has taken over some of LDA's functions but the Bank remains involved with NRA through the Central Rangelands Development Project (Credit 906-SO). - iv - PROJECT PERFORMANCE AUDIT REPORT SOMALIA - TRANS-JUBA LIVESTOCK PROJECT (CREDIT 462-SO) HIGHLIGHTS The project aimed to facilitate cattle sales in the southern part of the country by creating markets, establishing stock routes with stock watering points, and a second holding ground for cattle prior to shipment overseas or dispatch to the Kismayu Meat Factory for processing. Unfinished (store) animals would be fattened either on two ranches of 80,000 ha each, to be established by the project, or on an irrigated feedlot and fodder farm (IFFF). Veterinary facilities would be strengthened and a disease-free zone esta- blished. The project was to have been implemented by the Livestock Development Authority (LDA) which would be assisted by design and planning consultants and would also have the services, for the duration of the project implementation period, of expatriate line managers. The project failed to achieve its objectives and must be regarded as a failure in almost all its aspects. Not only is the recalculated economic rate of return negative (25% at appraisal), but also institution-building has failed because - through no fault of project design or implementation - LDA was disbanded in 1981. The IFFF fattened about 100 animals on a trial basis, but the performance rate was about half what was expected. The farm is now virtually abandoned and the equipment unused. While stock watering points, both surface reservoirs and drilled wells have been established, the wells have not yet been equipped with pumping units. Development of the one ranch finally approved has been suspended before it can be used. The holding ground and most of the veterinary clinics have been established, but only one of the markets has been completed and this is not functioning. The only wholly positive features are the stock routes (albeit only about 50% of target) and the fact that the more-than-planned development of watering facilities along them is helping the movement of trade cattle and nomadic graziers. Other points of interest, and lessons learned, are: - there is a grave risk of designing complicated projects, requiring many specialized skills for execution, in a country where trained manpower is difficult to find (PPAM, paras. 24-26, and PCR, paras. 8.03 and 9.04); - the need to complete the stock watering points, veterinary clinics and redesign the IFFF to ensure at least part of the investments already made are not lost (PPAM, paras. 7, 15, 23, and 31); - the socio-economic aspects of nomadic livestock husbandry are poorly understood and this has a negative impact on ongoing and future livestock projects (PPAM, paras. 16-19 and 25; PCR, paras. 3.08 and 8.04).  - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM SOMALIA - TRANS-JUBA LIVESTOCK PROJECT (CREDIT 462-SO) 1/ I. SUMMARY- 1. The Trans-Juba Livestock Project, supported by IDA with a US$10 million equivalent credit (462-SO of January 21, 1974) was the Bank Group's first project in the agricultural sector in Somalia. For a total estimated cost of US$11.5 million equivalent, the project was designed to support the movement, fattening and marketing of cattle in the southern tip of the country, between the Juba river and the international borders with Kenya and Ethiopia, in an area of about 65,000 km2 or about one-tenth of the total area of the country (See Map). Except for the relatively small animal health component, the project was to have been implemented by the Livestock Development Agency (LDA) which was, at that time, responsible "to promote the marketing and export of livestock and related products, and to coordi- nate the development of the livestock industry" (SAR No. 261a-SO of January 10, 1974, Annex 2). 2. The project had the following components: (a) constructing five permanent cattle markets; (b) developing one holding ground (one other was already operating); (c) further developing, and providing water supplies, along about 1000 km of stock routes; (d) establishing two 80,000 ha grazing ranches; (e) developing an irrigated feedlot and fodder farm (IFFF) where about 25,000 head of cattle would be finished, on the produce of the farm, in five batches of 5,200 each; (f) supporting a disease control program and establish disease-free zones; and (g) providing technical support and services, including assistance with detailed project design and implementation, and the prepara- tion of further projects. 3. The objective was to increase offtake of cattle by making access to markets easier and to pass on to the pastoralists, in higher prices for their animals, the greater part of the savings in weight loss and mortality that well-designed and managed stock routes would bring about. The two 1/ Adapted from the PCR. -2- grazing ranches would enable LDA to buy younger animals that would mature on these ranches, enabling the nomads to change their herd composition to more breeding stock and consequently enable higher herd offtakes. The additional holding ground would increase LDA's capability to hold and sort cattle before passing them on to the Kismayu Meat Factory (KMF) - which was producing, principally, stewed steak for sale to Russia and also some frozen sides - the grazing ranches, live export shipment and the IFFF. The disease-free zones were expected to qualify Somalia-s frozen (i.e., unprocessed) carcasses to enter the higher-priced European markets. Manage- ment and other technical support would strengthen LDA-s ability to execute the project and plan others in other parts of the country. 4. Consultants, engaged according to Section 2.02 of the Project Agreement, had prepared their design and costings by early 1976. These indicated that project costs may be about double appraisal estimates (PCR para. 4.03). After some discussions and fact-finding missions, the Bank mounted a full review of the project in early 1977 (PCR, para. 12). The report of this Review Mission found that, in fact, project costs would only increase by about 25% and, to accommodate that, recommended some minor scaling-down of the project (see PCR, text table 4). Most importantly, however, it recommended only a relatively minor scaling-down of the IFFF (see also PCR, Annex 4). Soon after the completion of the Review Mission, GOS terminated the Consultants' contract and also lost the services of four expatriate staff, who were in line positions as a condition of Credit effectiveness [DCA Section 7.01 (e)]. 5. Achievements fell far short of expectations of both the SAR and the Review Mission report; in fact, in many fields little or nothing was accom- plished (PCR text table 4): (a) only one market was developed, but is not being used; (b) a new holding ground was duly completed, but not until the original site was changed: the original site was being heavily used by stock owners already;.1/ (c) only about 550 km of stock routes were developed and only half the dams and boreholes planned were successfully completed. However, none of the completed boreholes along the stock routes were equipped with pumps when the audit mission visited the project area in August 1982; (d) some work was done on one of the proposed ranches, but the ranch was subsequently abandoned; 1/ OPS believe that the area was deliberately settled by an additional population between the time the intention to make it into a holding ground became known and its being gazetted as a holding ground. -3- (e) the IFFF proved to be a complete failure (PPAM, paras. 8-15); (f) four or five veterinary clinics were built and are functioning, if not as well as they could. The concept of establishment of the disease-free zone had to be abandoned: it was never really fea- sible; (g) technical support has, on the whole, failed, despite two separate attempts to bring in consultants. 6. Unsurprisingly, instead of an expected 25% economic rate of return, the project's economic and financial impact on Somalia is negative. This negative impact is further exacerbated by the fact that LDA, the institution that the project was to build up, was itself disbanded in 1981, for reasons unconnected with the project. 7. However, some positive results have, or with some additional work may yet, emerge: (a) the water resource development - the ponds and, when equipped, the boreholes - will not only reduce weight losses and mortali- ties along the stock routes, but also enable nomadic cattle to keep away longer from the Juba river in the dry season, thereby making more use of upland grazing and reducing the risk of trypano- somiasis to which the cattle are exposed in the river valley; (b) the holding grounds - former and newly developed - will, if properly used, enable trade cattle to be held prior to shipment to Saudi Arabia and the Gulf. This business is now entirely in private hands and therefore some formalized arrangements are necessary before full use can be made of the holding grounds; (c) the ranch, if completed, can be similarly used by traders to improve the younger and poorer animals that they may not be prepared to buy otherwise; (d) the veterinary clinics, if supported by proper handling facilities to ease prophylactic treatments, will improve animal health and reduce mortalities; (e) just possibly - and perhaps very optimistically - the IFFF can be used to produce pelleted fodder to feed live cattle during their 7-10 day sea journey to their final market. II. MAIN ISSUES A. The Irrigated Fodder Farm and Feedlot (IFFF) 8. Documents in the project files, antedating the appraisal mission, show that during project preparation the Bank's preparation missions included the IFFF because of the insistence of the Government of Somalia (PCR para. 3.04). This reluctance, and doubts about its feasibility expressed earlier, is not apparent in the appraisal report; despite the fact that the results expected - no weight gain for five days and 1 kg per day for 70 days -- were highly optimistic, yet they were put forward without any reservations being recorded. The Project Review Mission, writing in 1977 (PPAM, para. 4), strongly supports the IFFF [in this, backing the views of LDA-s newly ap- pointed General Manager but not those of his predecessor (PCR, para. 4.11)], although by then the Planning Department of the Ministry of Finance concluded it was not a sound component (BTO, preliminary Review Mission, November 1976).I1/ A succession of experienced Bank Livestock Specialists, involved in project supervision, and the Project Consultants, also cautioned against it (PCR para. 4.10). OPS expressed its reservations on the IFFF when commenting on the Review Mission's draft report in February 1977. 9. The audit concludes that the parameters on which the techniques and economics of the IFFF were based were not supported by solid, sustained investigations. The ration suggested (d.m. = on a dry matter basis): 45% maize silage, 20% maize grain, 32% grazing and 3% oilcake (cotton and sesame) (SAR Annex 10) is expensive and unrealistic in a country where maize is a human staple and, in most years, imported. The appraisal report assumes a conversion ratio of about 11% (a daily intake of 8.8 kg dry matter and a weight gain of 1 kg per day after a settling-down period; see SAR, Annex 2). This is very optimistic for Zebu cattle, especially in an environment where natural selection has favored survival under harsh conditions rather than high feed conversion. The animals were expected to enter the feedlot weighing 300 kg and leave it weighing 370 kg, i.e. their average weight, while being fattened, would be 335 kg. Such animals, when mature, would consume 2-1/2% of their body weight per day (on a dry matter basis), i.e. about 8.4 kg. The 8.8 kg assumed by the appraisal mission, (2.63%) while optimistic, is perhaps acceptable, but was eventually proven too high (PPAM, para. 11). 10. The Review Mission reaffirmed the viability of the component (paras. 3.04-3.09, Review Mission-s final report, May 1977). However, it changed the appraisal mission-s parameters as follows: 1/ The Region wishes to draw attention to the fact that the Review Mission did suggest reducing the size of the IFFF. -5 - weight of animals: 225 kg'/ ration: (d.m.) 84% of maize silage, 13% lucerne hay, 3% urea, bone meal and salt weight increase: 750g/day feed intake at 2-1/2% of body weight: 5.6 kg conversion ratio: 13.3% 11. These expectations compare with the only actual trial that was conducted in 1981, using 100 animals (Supervision Report, April 1981). The results are as follows: weight of animals: 236 k2/ daily liveweight gain: 370 grams daily feed intake (d.m.): 5.75 kg (2.4% of body weight)3/ conversion ratio: 6.47ZY The ration fed was similar to that assumed by the Review Mission, except that it also included molasses. 12. The mission was advised that not one animal has put on more than 500 g/day (8.7% conversion). Similar trials in Pakistan have produced similar results; between 450 and 500 g/day.5/ The Pakistani trials were 1/ enter feedlot weighing 200 kg, leave weighing 250 kg. 2/ enter feedlot weighing 213 kg, leave weighing 259 kg. 3/ c.f., PPAM, para. 9. 7/ The Region feels that it would be wrong to give too much weight to this one trial and that there is no convincing evidence that the appraisal mission assumed too optimistic feed conversion. 5/ Pakistan Journal of Agriculture, 1959 and 1960. - 6 - conducted with oxen that were accustomed to being stall fed and handled, unlike the animals - most of them entire males - that were expected to be fed in the IFFF../ 13. The economic analysis in the SAR (Annex 21), in its sensitivity test, examines the case of an increase in capital and operating costs and the effect of adverse turns in purchase and sales prices, up to 10%. The ERR of their base case is 13%, the lowest, if sales prices fell 10% below assumption, would be 7%. The conversion ratios obtained nearly double the operating costs because it means feeding the animals twice as long for the same weight gain. Since a 10% increase lowers the ERR to 10% from 13%, the actual 50% increase would make it negative, even if the sale price remains as expected. 14. Another assumption of the Review Mission that the audit cannot accept (and was not found acceptable by the PCR - see para. 8.08) is on the very issue of the expected price of fattened animals. The assumption was that fattened animals will fetch a sizeably higher per kilo price than the lean ones, i.e. that the additional weight (and quality) commands a premium over the live weight of the store animal. The appraisal team assumed that the fattened animals will be sold as frozen or chilled beef and therefore assumed a live weight price for fattened animals which was over 80% higher than canning quality, and about 45% over the live export price (SAR, Annex 3, Table 6). The Review Mission, very realistically, concluded that this quality premium for frozen carcasses is unlikely to be attained in the face of Austra- lian and Argentine competition in markets open to frozen meat and assumed, also correctly, that Somalia-s future in the export cattle trade, (a-ad for all livestock) lies in selling animals live (Review Mission Report, para. .3.01). The Review Mission assumed that animals for the IFFF would be pur- chased for So Sh 2.50/kg and sold for So Sh 3.50/kg FOB Kismayu (Annex 7, Review Mission Report). According to all informants of the Audit Mission 1/ OPS states that the Review Mission recommended proceeding with the IFFF because (a) it was to have been 300 ha and not 1,200 ha, as expected at appraisal and (b) because an "acknowledged expert on feedlot pro- duction in East Africa" was used by the Mission. They (and the Region) also do not believe that one trial is necessarily conclusive. The Region also draws attention to cattle feeding trials conducted at Lanet in Kenya where cattle which were genetically identical to Somali cattle, and which came from an environment very similar to the project area in Somalia, achieved liveweight gains of close to one kilogram per day. Without going into details, (1) the audit cannot accept the Review Mission's calculations that the 300 ha that they proposed to irri- gate will produce the feed for the number of animals (13,800) they believed would be fattened in a year since it implies a production of about 17 tons of dry matter per annum per hectare; (2) the Review Mission's expert has clearly expressed a minority opinion on the economic feasibility of the IFFF, contradicting the Bank-s Livestock Specialists, including OPS itself; (3) even one trial, involving 100 animals and producing under 50% of expected results, certainly gives a strong indica- tion of likely results. - 7 - and of the Livestock Specialists in RMEA, contemporary and current, this has never been the case. Animals are sold by the exporters per head, CIF port of unloading and no quality premium is offered; therefore, they cannot pay one either.-1 Also, as the Appraisal Mission remarked (SAR, Annex 2, para. 16) and the Audit Mission observed and was advised, feeding stock in transit is not only inadequate but expensive. The animals are shipped loose in compart- ments of the hold of the ship and cannot feed properly; therefore, weight losses on the 6-10 day sea vo ae are high and would be higher still with sensitive stall-fed animals- Therefore, until marketing conditions change dramatically, the present export markets cannot pay the premium for quality assumed by the Review Mission and the operation, which gives a low ERR (12.5%) even when the Review Mission's assumptions in price and feed conver- sion are accepted (Review Mission, Annex 7, Table 4), is virtually certain to be negative, given the actual feed conversion ratio (PPAM, para. 11) and the likely prices. It is the firm view of the audit that the Review Mission, on the evidence available to it, or readily obtainable within the country, should have unequivocally recommended the abandoning of this component, and if the Government had agreed, this would have saved the cost of the IFFF (US$2.7 million) and that of its associated farm machinery (US$0.7 million), 32% of the project expenditures (PCR, text table 6) on which the likelihood of any return is, at best, remote. 15. A major remaining issue of the IFFF is how the Bank can help Somalia to salvage at least part of the investment. The audit mission, based on its observations in Kismayu harbor, discussions with officials of GOS's MLFR, livestock exporters and the remarks in the SAR regarding feeding animals in transit (PPAM, para. 14) believes that using the IFFF for producing baled or pelleted fodder, for sale to exporters in Mogadishu and Kismayu would be a great service to the country. Current prices for an estimated 200 kg of rough hay, loosely bundled, was said to be So Sh 700 delivered to Kismayu harbor, (So Sh 3.50/kg or about US$0.23/kg). On this basis, the gross income from a hectare of sorghum hay would be in excess of So Sh 25,000 (US$1,700) per crop; two or three crops a year are a possibility. The subject would certainly appear to deserve further study, even though the large sums spent on buying equipment to deliver fodder to cattle in the sheds will be lost if the IFFF becomes merely a fodder farm. B. The Project's Influence on Stock-Owners- Income 16. The SAR states that "the principal benefits of the Project would stem from two basic sources: the increased number of animals from nomads and increased weight added by the fattening operations on the ranches and feedlots" (SAR, para. 7.01) "...8,000 would represent additional offtake from the available herd ..." (SAR, para. 6.01). 1/ OPS disagrees with this view. OPS states that at the time of the review mission higher prices per kilo were paid for larger animals. At that time, animals in the 170-200 kg weight range were purchased at prices ranging from So. Sh. 2.00-2.50 per kg, while heavier animals were being sold for So. Sh. 3.50-4.00 per kg. 2/ The Region believes that cattle from a feedlot would travel at least as well as cattle coming directly from rangelands. - 8 - 17. At no point does the SAR substantiate the assumption of an increased offtake due to the activities of the project by an attempt to verify, from actual herd composition data, that there were, before the project, marketable animals that stock owners could not, or did not want to, sell. It confines itself to assuming that offtake, "estimated (audit-s emphasis) to be between 9 and 10%" (SAR Annex 1, para. 14) will increase to 10% (SAR, Annex 2). 18. While of course it is virtually impossible to conduct a meaning- ful cattle census amongst stock owners whose grazing grounds vary with the seasons, straddle international borders, and vary also from one year to another in the same season, depending where exactly the rain has fallen, it is by no means impossible to conduct sample surveys of herd compositions. This would readily reveal whether stock owners have any animals that are saleable and they cannot, or, - equally importantly, - do not want to, sell. Only such an investigation would support the claim that offtakes can be increased; no such investigations have been undertaken either during project preparation or at appraisal. 19. Information obtained by the audit mission would indicate that there is no barrier to the sale of male cattle!l amongst stock owners. The export of females of all stock is legally banned, although they may be sold for slaughter within Somalia. This of course reduces their potential value as trade cattle. Therefore, offtakes can only be increased in the long term: ensuring better survival of calves; controlling debilitating diseases; in the even longer term changing the trading pattern and creating a demand for younger, lighter animals which would give more room on the ranges for breeding females and their young offspring..2' However, in the Somali environment it is safe to assume that the range is fully used over the seasons (with the notable exception of water resources being exhausted before the grazing, PPAM, paras. 20-21) and therefore only marginal increases are likely to occur in offtake rates. The steady increase in export sales over the years does not negate the basis of this argument because, firstly, numbers are quite small compared to the total number of cattle held by Somalis (in Somalia, Kenya and Ethiopia) and, secondly, domestic demands can always shift to barren cows that may not be exported. Thus, it is unlikely that the project could have, or did, increase offtakes to any extent. 20. Without a doubt, however, the project did reduce weight losses and mortalities amongst animals using the stock routes. By supplying water, it has enabled nomadic stock fully to utilize grazing previously not com- pletely exploited. Also, in a country where the road network is not too extensive, the clearing of tracks has enabled vehicles to travel along them in the dry season, thereby generally contributing to the development of the more remote areas. 1/ Traditionally, there is more reluctance to trade camels, but the project did not extend to livestock other than cattle. 2/ OPS advises that the Review Mission has considered this factor in its calculations. - 9 - 21. However, through lack of funds and organizational difficulties, the successful boreholes along the stock routes are not equipped with pumps and engines, although these have been procured. The audit shares the concern of the project department supervising a Bank-supported range project that the installations should be completed and the wells used - after surface water sources dry up - for the benefit of the graziers as well as for the users of the stock routes. C. The Veterinary Clinics 22. Four of five veterinary clinics planned under the project were actually built. The ones seen by the audit mission were staffed and operating an on-demand consultant/curative service: treating individual animals that were brought to them. They were, originally, also equipped with crushes, dips and marshalling paddocks for prophylactic campaigns. These are not being kept up. The mission was advised that rinderpest vaccinations are carried out in a thorn enclosure where each animal has to be caught and held to be vaccinated and then marked so that they are not inoculated again. 23. It is to be regretted that the regional staff of the Ministry of Livestock, Forest and Range did not accept Bank staff views that this method of operation is not only wasteful of manpower but also slow and liable to miss animals. One of the main objectives of the clinics was to establish a centre where many animals can be given prophylactic treatments quickly and effi- ciently. The facilities to do so were constructed and their maintenance - even reconstruction from their present dilapidated state - would be a minor expense to the Government and a major benefit to stock owners. D. The Scope and Extent of the Project 24. Reviewing the planned components of the project - not necessarily the costs and logistics of the operations proposed - the audit is struck by the truly overwhelming tasks that project management were to solve. To recount them briefly: (a) arrange the flow, in five batches, of 25,000 head of cattle to the IFFF: moving 50,000 animals a year, 12,500 movements, within a short time, every three months; (b) ensuring an even flow of feed to these animals either as green fodder or silage; to make this quantity of silage, move it in and out of silos: as much as over 200 tons, or at least 70 trailer-loads a day to the feedlots and perhaps double that volume per day from fields to silos; (c) operating 160,000 ha of ranch lands, supplying them with cattle and selling or transferring them after a while. This in- volves organizing the average movement of about 550-600 animals per day, in and out of the ranches. Without previous experience, and without an existing infrastructure, building up such a volume of business is indeed a daunting task. LDA-s previous experience was confined to holding animals for a rela- tively short time, between purchase and sale; under the project, literally tens of thousands of animals would have to be kept, fed, managed, herded, all very smoothly and without major upsets, to ensure that they gain weight in the process at an acceptable rate: financial benefits to the managing organiza- tion hinge far more on weight gain than on any other factor excepting only mortalities, itself largely a function of management. - 10 - 25. The audit questions whether, even if everything had turned out exactly as planned, it would have been possible to gear up to an operation of such complexity. As noted, this was the first major project in the agricultural sector; its outcome clearly depended to a major extent on the response of a large number of people who were not closely involved in, or under the control of, project management: the nomadic stock owners. By all accounts, little was definitely known at the time of appraisal about their way of life, grazing rights, motivation, etc. Only the scantiest literature exists on these matters and not even that was examined either at project preparation or appraisal, since neither team had included a social anthro- pologist. Therefore it would have been wiser not to raise Government's expectations by devising such a complex project but to phase it, in shorter time slices and fewer components in each slice. The outcome may well have been not only a more gradual and less daunting buildup of staff, skills and equipment, but also the kind of flexibility that, for all intentions to the contrary, simply cannot be built into a single project with necessarily set objectives, and greatly hampers adjustments to newly revealed sociological facts. The lack of a solid data base at the time of appraisal (SAR, Annex 1, para. 14 and Annex 2, para. 2) would also have suggested a far more cautious approach, further underlined by the lack of local managerial expertise (SAR, para. 5.02), recognized by the preparation and appraisal missions, and not disputed by the Borrower. 26. Although a technical assistance component was built into the project - it was envisaged that, initially, four senior management positions would be filled with expatriates - this component was unsuccessful. The originally appointed senior management staff, who were provided by the Engineering Consultar- firm applinced according to Section 2.02 of the Project Agreement, were dismissed by GOS and the replacements, approved most reluc- tantly by the Bank, did not, on the whole, meet expectations (PCR, para. 4.14). Thus, LDA and GOS were very much thrown onto their own resources, in effect assisted only by Bank supervision missions, to solve their many day-to-day and planning problems. 27. It should be noted that the Ogaden war of 1978 did not affect the project directly, in that no actual fighting took place in or near the project area. However, Government has, under its Emergency Regulations, requisitioned project vehicles (PCR, paras. 4.20-4.21) which not only deprived the project of their services but also reduced their effective service time. E. Conclusion 28. While undoubtedly the project has failed to reach almost any one of its objectives (PCR, para. 9.01), it nevertheless teaches some important, positive lessons - and not only the cautionary ones expressed in the PCR (para. 9.04). 29. Probably the most significant positive contribution the project has made was the development of the stock routes, holding grounds and veterinary dispensaries (PCR, para. 9.03) to which the audit wishes to add the development of surface water resources and, when finally equipped, (which, since the pumps and engines are actually in store in Kismayu is most likely) the boreholes. - 11 - 30. As noted (PPAM, para. 6 and PCR, para. 9.02) LDA, which was to have been built up into a leading national institution in the livestock sector, partly by this project, was abolished in 1981, and therefore, in a strict narrow sense, the institution-building effects of the project were, even if not zero as suggested by the PCR (PCR, para. 9.02), certainly much impaired. But possibly the lesson has been learned that Somalia has much to gain by allowing its private-sector livestock traders to exploit the market for live animals, as they have been doing for many decades past. The Ministry of Commerce, stepping into the gap left by the abolition of LDA, has been made aware of the magnitude and importance of the problems of the Gulf and Saudi markets, and is studying them, with EEC help, with the objective of identi- fying the means to counter poten ial Australian and New Zealand competition in the market for live animals Within the country, the National Range Agency (NRA) is taking up some of LDA's former tasks and there is a general belief, shared by the projects department and the audit, that LDA's failings are being studied by them and, hopefully, will be avoided. 31. Since the Bank continues to be involved with NRA in one project (Cr. 906-SO),./ it will be in a position to offer its technical advice to follow through on the institutional issues raised by the Trans-Juba Project and assist GOS, the MLFR and NRA to formulate its policy and actions to achieve the broad objectives of the past on-going livestock projects. 1/ Information supplied to the audit mission by the Director General of the Ministry of Commerce. 2/ Central Rangelands Development Project; for further details, see also SAR, No. 2163-SO, dated April 17, 1979. - 12 - ATTACHMENT I Page 1 Jamhuuriyadda Dim. Soomaaliya 1i WAKAALADDA DAAQA QARANKA (National Range Agency) P. 0. Box 1759 - Tel: 27025 - 34007 Y aeti Telex 736 NRA Jl..l ,A Muqdisho oomaallya We: Samalrange Y) 3 o MOGADISHU 22.3.1983 Mr. Shiv S. Kapur, Acting Director General, Operation Evaluation, Worla Bank, 1818 - H. Street Washington D.C. 20433 U, S. A. Dear Mr. Kapur; Re: Project Performance Audit Report on Somalia Trans-Juba Livestock Project (Credit 462-50) I invite your kind attention to your letter January 3, 1983. The Audit Report along with Project Completion Report (PCR) has covered all aspects of the project performance, our comments are offered herewith on only those references where there has been a -mention of National Range Agency which are as follows:- REFERENCES PREFACE:- NATIONAL RANGE AGENCY:- Now charged with overseeing the Project" PROJECT COMPLETION REPORT:- IV IMPLEMENTATION 4.31 "In March 1981. ....................................... .......... Responsibility for infrastructural aspects (stockroutes, stock-water, holding grounds) was assigned to the National Range Agency (NRA) of MOLFR and the rest was assumed by the Minister, but the exact delegation of responsibility was unclear and the arrangement was slow to work..................................................f V PROJECT IMPACT 5.07 "There is the possibility that completion of the boreholes established under the Project which were not equipped at the Project's close may be achieved under another IDA financed project. Discussions with NRA of MOLFR have tentatively identified the Central Rangelands Development Project (Credit 906 - SO) as a potential vehicle. This matter is being followed up by the Bank staff responsible for the super- vision of that project. Completion of the TJLP livestock water program 13 ATTACHMENT 1 Page 2 :2: by this means would overcome the current unsatisfactory situation and unsure more effective application of the extensive investments made in the stockroute - water system under the project." COMMENTS In this connection it is be brought to the notice of all concerned that this Agency was Aharged so far to oversee the TJLP Project after its completion. The stockroutes or any of the marketing infrastructures said to have been developed were not handed over to this Agency* The Project after its completion continues to be a separate entity with the Project Director under the Ministry of Livestock, Forestry and Range. This Agency is not aware of the development which took place in the project after March 1981 and also after the completion of the Project i.e. from 31.12.1981 upto this date. This Agency took over the various old infrastructures of the ex-LDA on its abolition which are mainly the holding grounds in the northern and southern zones of the Country and Quarantine Station at the port of Berbera* With regards, T sincer, (Dr. b ah A Karani) GE R NRA CC: H.E. The Minister of Livestock, Forestry & Range Mugadishu. CC: Mr. Andrew Hayman, Resident Representative, World Bank Mogadishu. - 14 - ATTACH4ENT 2 Page il Trans-Juba Livestock p,ecj Livestock Development Agency P. 0. Box 929 Mogadishu - Somala MÅ"hruuca Koolaha Trans Juba Wa~aI4dda Horumarinta Xoolaha C4We : STOCKAGENCY" eklephorie No. 46029 18/3/1983 To, LDate ~ TJP/ 30/83Megadisho. To: The World Bauk,Vashiaton,DC.20433 V.S.A. CC. Resident Rep. World Bank, Mega__s__ Sub: Projeot Perfermaa6e aidit RepOIrt *m Trams-Juba Prøleot-Credit-462-S..... Om behalft the imister ad Viee-Ninister I have theliberty to reply and kiadly refer to yeur letter dated em Jam.3rd,1983 hih is related tø the ab*ve aubjet. V. have ae auoh te eemat eept for tw* polats,tirstly; 50 per eat of the Projeat Borevells ar. ftaotieal and the renMaer pepratiem is goig om, seoeadly; Paoilities o.g. hm&es, equipmeat, *ttioes,werkshop aad the tarm(IFF) amd steik-r@ute and Ag-Libah Keldi~g Grud are operational. Furthermore, ve approoiate aud tully spport(PPAN) polnt of views i page(1t-para. 15) amd th* Baak's fresh intervetie te salvage at least part of the tavestmet(IFFF) is a benefilial oetribution. The Prøjeot is ex the way in prodno.ag baled todder for sale to livesteok exxperts. Res.rep * is kIiMl† røquesøed for fawardiag this letterý , Qrs nd døply regrted fror the delay. Sibeerely, (Abehir Ahned Fa7 if - 15 - SOMALIA TRANS-JUBA LIVESTOCK PROJECT (CREDIT 462-SO) PROJECT COMPLETION REPORT November 29, 1982 Eastern Africa Projects Department Regional Mission in Eastern Africa - 16 - SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT I. SUMMARY 1.01 The Trans-Juba Livestock Project (TJLP) was appraised in January/February 1973 after an identification and preparation period extending from the late 1960's. The Project was approved by the Board in January, 1974, and was signed the following month and sub-- sequently declared Effective after considerable delay in October 1974. This pre-Effectiveness delay was due largely to the administrative and legal arrangements and to the other formalities required for this first Bank Group project in agriculture in Somalia. 1.02 The project aimed at providing the infrastructure and market outlets that would facilitate marketing of cattle from a 65,000 km2 project area south of the Juba river. The implementing agency for the project, the Livestock Development Agency (LDA), would purchase cattle from pastoralists for immediate canning in the Kismayu Meat Factory (KMF) or for processing after growing-out and fattening for export as frozen and chilled beef. 1.03 The principal benefits of the project were expected from: (i) improved offtake from pastoralists due to reduced weight loss and fewer deaths from better trekking and marketing facilities and improved animal health care; and (ii) improved quality of animals and higher slaughter weights from the growing-out and fattening of cattle on the ranches and feedlot. These would in turn assist the economy since they represented increased productivity and would result in higher foreign exchange earnings. The pastoralists were expected to receive higher prices which, together with increased overall herd productivity, would result in significantly higher cash incomes per family. The situation was expected to become self-perpetuating as pastoralists were induced by higher prices to breed and sell more stock. 1.04 The project included the development of stock routes and water points to penetrate the project area, markets at which pastoralists could sell animals, holding grounds for staging supplies of cattle to the KMF and to the two ranches where immature cattle were to be grown out, and an irrigated fodder farm and feedlot (IFFF) where mature cattle would be fattened to produce high-quality beef. In addition, the project would construct veterinary centres and dispensaries to strengthen the regional veterinary programme and create a disease- free zone from which exports to high quality European markets would be allowed. - 17 - 1.05 Key project indicators and actual project achievements are shown in Table 1 and in the Basic Data Sheet. On all counts, the project fell seriously short of expectations. Its economic rate of return, which at appraisal was estimated at 25%, was negative. 1.06 Of the several factors leading to the project's downfall, the failure of LDA as implementing agency was the most significant. Throughout the history of the project, the ineptness of LDA management, its adequate financial control, and the deliberate actions it took to prevent the auto- nomy of project management underlay the failure of the project to achieve its objectives; and Government did not take the necessary timely actions to control the situation. Other important factors were the size, scope and complexity of the project in relation to local managerial, staff, technical and financial resources; an over-ambitious planned rate of development; shortage of experienced Somali staff and difficulty in retaining them; dif- ficulty in attracting suitably qualified expatriates to Somalia; delays in design and planning due to late arrival and poor performance of consultants; shortages of materials and spare parts; chronic shortage of working capital and development funds; inadequate performance of the Water Development Authority (WDA) in developing livestock water; and diversion of project funds, equipment and vehicles for non-project purposes. 1.07 The project's overall impact was negligible in spite of concerted Bank efforts in the later stages to encourage implementation of the stock- route and water supply components in order to achieve some useful benefits for traditional producers. The stockroutes, holding grounds and boreholes were completed on a reduced scale but their effectiveness was largely undermined by the projects failure to equip the stockroute borehole system. The project did not contribute to institutional development, since the two agencies most involved with the project, LDA and KMF, were both dissolved due to unsatisfactory performance. Some studies were completed on range and pastoralist issues, and overseas training and a small amount of technical assistance was provided; however, the impact of these components was minimal. 1.08 The PCR has devoted a considerable proportion of its content to presentation of the history of the project, as the means to explain Bank and Government actions, the role played by the LDA and the reasons for project shortcomings. Certain lessons learnt in the implementation of TJLP--the necessity of Government commitment to a project; the need for smaller, simpler projects in an undeveloped subsector; and the need to carefully choose expatriate consultants--may have parallels in other agriculture projects presently under implementation in Somalia. It was felt that a detailed discussion here might be beneficial in avoiding a repetition of the disappointments which have attended the Bank's first intervention in the agriculture sector in Somalia. SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Key Indicators of Development of Project Components Item Unit --- Closing Date Target --- Achievement Appraisal Review Cattle Markets No 5 5 1 1/ Stock Routes - boreholes No 26 2/ 21 15 (21) 3/ - dams No 32 32 15 - developed km. 1026 1026 more than 550 km Holding Grounds %completed 100 100 100 4/ Ranch I 100 100 Suspended in 1978 Ranch II 100 100 Suspended in 1980 IFFF 100 100 Suspended in 1981 Veterinary centres No 5 5 4/ Limited assistance in constructing/equipping Kismayu Regional Veterinary Centre Veterinary dispensaries No 5 5 5 5/ 1/ Currently not used 2/Included 3 to be repaired 3/ 21 drilled, 15 with potable water, 14 not yet equipped. 4/ With the exception of equipping of 2 boreholes, 5/ Five dispensaries constructed, one not usable due to poor construction and another abandoned due to poor location; three rehabilitated by GTZ for Regional veterinary programme. - 19 - II. BACKGROUND General 2.01 In terms of production and potential fo: grow'h, tIC Somali economy is dominated by the agricultural sector. Reliable statistics on the gross domestic product (GDP) are not available, but it was estimated in 1978 to be of the order of US$ 1,000 million. Productive sectors, other than agri- culture (manufacturing, mining, construction and public utilities), contri- buted only 10 per cent of GDP and employed 7 percent of the labor force. Services contributed about 28 percent of GDP and employed 13 percent of the labor force. GDP growth was estimated at about 2.5 percent a year in real terms for the period between 1972 and 1978. However, the average growth rate was only about 1 percent a year in the productive sectors while it was 6 to 7 percent in services. Crop production, which showed a decline of 5 percent, was especially disappointing. The rapid growth in the services sectors reflected mainly a rapid increase in government employment. Over the last two or three years Somalia's external and internal financial position has deteriorated sharply. There are a number of reasons for this deterioration, e.g. the Ethiopian-Somali conflict, the rise in oil prices, Government economic policy, etc. One important factor responsible for the recent deterioration is the stagnation of production in agriculture resulting mainly from lack of producer incentives. 2.02 Of the agricultural subsectors, by far the most important is livestock production, which provides a livelihood for 60 percent of the population (total 4.2 million), about one-half of GDP and 80 percent of export earnings. The Project Area 2.03 The project area covers some 65,000 km2 and is bounded in the east by the Juba River, in the north and west by the Kenya border and in the south-east by the Indian Ocean (refer Map 1). The area includes some 80 percent of the lower Juba and some 40 percent of the Upper Juba administra- tive regions. On a geographical basis it can be divided roughly into two regions (refer Map 2): - 20 - Table 2 The Project Area 2 Area in km Average Annual Estimated Rainfall (mm) - carrying capacity 2/ (LU/sq. kn)- (a) The Uplands Region Mandera-El Wak Uplands 20,000 380 4 Upper Juba Valley 2,000 190 6 (b) The Central and Coastal Region Dugiuma Sub-Region 3,000 520 16 Marine Plain 12,000 545 13 Lower Juba-Shebeli Flood Plain 2,000 500 27 Fafadun Plain 13,000 545 11 Lac Dera Plain 11,000 500 21 Dudumali Plain 2,000 463 20 Total Project Area 65,000 2.04 The Uplands region in the north is mostly hilly, with high country bordering Kenya to the northwest. In general the area is arid with sparse vegetation with a limited grazing potential confined, because of sparse water supplies, to the period during and shortly after the rains. Up to 25 percent of the total area is bare earth. Altitudes vary from sea level to 500m. The central and coastal region can be broadly characterized as a relatively featureless plain, interrupted by river valleys and tracts of rougher relief. For the most part it is arid to semi-arid rangeland subject to recurrent droughts. It has low shrubs and infrequent trees. The type and composition of vegetative cover reflects the wide variations in soil and rainfall found throughout the project area. 3/ 2.05 Rainfall is bimodal with two dry and two wet seasons. The main wet season extends from March/April to June/July. The short rains begin in September and are followed by the main dry season. Rainfall in the two rainy seasons is extremely variable. Temperatures are high but relative humididy is low. 1/ Source: UNSF/FAO 'Somali Agriculture and Water Survey'. FAO, Rome, 1968. 2/ Source: Democratic Republic of Somalia, 'Trans-Juba Livestock Development Project: Loan Application to the International Development Association', October 1972. 3/ Further details of the project area can be found in the Appraisal Report and Loan Application Report. - 21 - 2.06 The project preparation mission estimated the livestock population of the project area to be approximately 1 million, almost all kept by traditional pastoralists. Grazing patterns involved moving into Ethiopia and Kenya in pursuit of forage and some 30,000 to 40,000 cattle were believed to be sold into Kenya as immatures each year. This informal export of animals was one of the main stimuli for the design of the TJLP. The Setting 2.07 At preparation it had been noted that "..... few of the former programs leading to the development of the livestock sector have left any lasting or tangible effect ..... In more recent times USAID established a range manage- ment project in the Afmadu area which had no impact upon changing nomadic patterns of range utilization". Thus, the TJLP was to be the first substantial integrated livestock development programme in Somalia. It was a pioneering project combining: (i) a marketing development strategy; (ii) development of infrastructure to encourage marketed offtake; (iii) initiation of a stratified beef industry involving growing-out of immatures and feedlot fattening; (iv) improved range management and utilization through the clearing of stockroutes and development of water supplies; and (v) improved disease control and veterinary care. 2.08 The LDA had overall responsibility to coordinate the development of the livestock subsector and was engaged in programs to develop the hides and skins industry, improved livestock breeding, marketing, holding grounds and the export of livestock and livestock products. The Agency was respon- sible for all port veterinary work and the issue of veterinary health certificates for export. It had also funded the vaccine requirements of the Veterinary Department in 1967, 1968 and 1969. Although the LDA was known to lack management skills, manpower, planning capability, financial support and financial control, the appraisal mission concluded that the Agency would be a suitable vehicle for implementing the TJLP, if strengthened with a special project implementation unit. External Factors 2.09 The impact of the drought in the years 1973 - 1975 had a substantial effect in developing an awareness of the need for pastoral development in Somalia. Throughout the 1970's the Government aimed to pursue the development of water supplies in rangeland areas in order to reduce concen- trations of livestock in areas of permanent water and to improve utilization of resources. 2.10 At preparation and appraisal Somalia had strong ties with the USSR, which had led to the establishment of important political and economic relationships. In particular, the USSR had provided the funds for the cons- truction of the KMF and had provided staffing for key technical posts in the factory. It was generally assumed that the USSR was to be the main long-term market for virtually all of KMC's output of canned meat. During the TJLP both the technical staff support for the KMF and the USSR market for its canned products were withdrawn thus severely undermining Government strategy for meat marketing and livestock development in the Trans-Juba area. - 22 - 2.11 The period after 1973 (Appraisal) was characterised by rapid inflation in industrialized and oil-importing countries. The effects were felt by the TJLP in the form of increases in costs of construction materials, machinery and equipment, and in project operating costs. Delays in implemen- tation prompted further increases and project financial constraints were accommodated by reductions in project components. 2.12 The Ethiopian-Somali conflict erupted in late 1977 leading to the commandeering of project vehicles and other equipment. The security situa- tion prevented freedom of movement in the western and north-western sectors of the project area, imposing a management constraint for the remainder of the project's life. The conflict marked the end of any remaining Soviet- Somali economic ties. Bank Involvement 2.13 The TJLP was the Bank Group's first project in agriculture in Somalia. Since then, the Bank has committed nearly US$ 50.0 million on livestock and crop production projects in the country as follows: Project Effective US$ million (IDA) Drought Rehabilitation 1976 8.0 North West Agricultural Development Project 1977 10.0 Central Rangelands Development Project 1980 8.0 Agricultural Extension/Farm Management Training 1980 10.5 Bay Region Agricultural Development Project 1980 12.0 In 1981 an Agriculture Sector Study (including fisheries) was also completed by the Bank. 2.14 Development of the livestock sector has long been one of the fundamental objectives of Somalia's agricultural policy, since it offers the means to increase export earnings and monetary incomes of pastoralists. The control of diseases and development of a marketing infrastructure (through the LDA) were the most prominent features of the Government's policy in the early 1970's. By the late 1970's emphasis had moved to the development of range management and utilization strategies, to maximize livestock production and prevent degradation. The Bank supported this general approach. Project Completion Report Preparation 2.15 This PCR was initiated by a consultant in February/March 1982. A two-week field visit to the project area and Mogadishu was followed by a file review at RMEA. A planned second field visit to the project area to collect data being prepared by project management and to inspect all project physical components was cancelled to reduce costs. 1/ The consultant was supported during his assignment by IDA staff familiar with the project. 2.11 Security restrictions at the time of the first field visit and the cancellation of the second field trip prevented the mission from visiting most of the project area and inspecting physical project components such as stock- routes and water points although nearly all of these had been visited earlier by Bank staff in the course of project supervision. 1/ The restrictions of movement in the project area and almost complete absence of project records and accounts resulted in the conclusion that a second field trip would be of little value. - 23 - III. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL Identification 3.01 Identification of the project commenced as early as 1967. LDA had identified specific infrastructural requirements in the southern Juba area, including holding grounds and stock markets, quarantine stations and port marshalling yards. As time progressed and the number of Bank Group missions dealing with identification of a project increased the shape of a possible project emerged. 3.02 In 1970 USAID- IDA discussions focussed on a project for the area south of the Juba River. USAID was at that time involved in the development of permanent water supplies and range management in north-eastern Kenya and feared a major influx of cattle from southern Somalia (see also para 2.06). Complementary range development was needed in that area. This coincided with Government's initiation of an agricultural development policy giving highest priority to livestock development. 3.03 From January, 1970 the identification of a project in the area south of the Juba River was pursued by the Bank with several joint missions from Washington and Nairobi. In all, four identification missions were mounted in the period up to March, 1972 (including the January 1970 "observer" mission). USAID, UNDP (Somalia),the LDA and the Wildlife and Range Development Depart- ment were heavily involved in identification. Bank Group staff and consul- tants spent an estimated 25 manweeks in Somalia during identification. 1/ 3.04 At least two important issues emerged during identification: (a) the KMF was noted by the January 1970 IDA mission as 'poor' and it was doubttul if a capacity of 30,000 head per annum of slaughter cattle could be exceeded (design capacity was 60,000); and, (b) in early 1972 the Government insisted on the inclusion- of the irrigated fodder farm. The March 1972, mission noted that ".....if it (the fodder farm) was excluded, the Government had little interest in pursuing the project further". 3.05 The KKF was generally perceived as a key piece in the increased production/slaughter/export strategy of the TJLP. The irrigation farm was central to Government's plans to promote a stratified beef industry and was to become the largest single cost item. The March 1972 mission convinced Government that the irrigation component should be small and that it should form part of a pilot ranch component, the two components together not exceeding 2,000 hectare. This was lpter construed to refer to the area of irrigation, and both the preparation and appraisal missions were to include a large-scale irrigated fodder farm and feedlot and a greatly expanded ranching component. Preparation 3.06 The project as prepared reflected Government's desire to expand the productivity of the all-important livestock subsector. Livestock exports were, and remain, the country's main foreign exchange earner. Rangeland production was low and priority was set on improving marketing systems and increasilg offtake with the basic objectives of improving export earnings and increasing pastoralists' incomes. To achieve these goals, preparation efforts emphasized improved marketing facilities and the introduction of cattle fattening opera- tions, which would enable a diversification of production into chilled and frozen beef for export. The KMF was already involved in exports of canned meat to the USSR and trial shipments of carcass beef had been made to other countries (e.g. Comoros). 1/ See summary tables preceding Table of Contents. - 24 - 3.07 The preparation effort also aimed to enhance the veterinary programmes already underway in order further to complement the improved strategy and improve offtake. This component was intended to lead to the establishment of a disease-free zone, the cattle from which could be slaughtered and exported as carcass beef to European markets where higher prices could be obtained. 3.08 It is significant that, at the time of project preparation, there was little data of substance available concerning: (i) range productivity and the existing pastoralist system; (ii) traditional livestock production and marketing; (iii) irrigated fodder production; (iv) livestock performance parameters under rangeland and improved conditions; and (v) hydrogeology Further, project planning did not include a sociological study to describe traditional grazing rights, the pattern of movement of herds, production constraints, rangeland use and the perceived needs and infrastructural requirements of the project's intended beneficiaries. Subsequently conflicts were to arise over restriction of grazing rights imposea (alDeit temporarily) by the ranches, the planned siting of a holding ground at Afmadu and the demarcation of the site for the irrigated fodder farm (the area was already partly settled and LDA was required to clear land in a new area to resettle the farmers). 3.09 The preparation team was mounted by IDA in June 1972, some 30 months after the first tentative identification. Design was the result of inputs from at least a dozen Bank, UNDP, USAID, GOS and consultant specialists. A further mission to finalize preparation visitedSomalia in late August, 1972. 3.10 The Bank's involvement in both identification and preparation was predominant because it was carrying the project to the Government as the technical personnel to carry out detailed project formulation were not available in Somalia at that time. While data and support were provided by the Somali authorities, most of the local input came from UNDP and USAID and consultants who had worked in the project area. However, the Government did eventually play a decisive role in dictating final project components, particularly the IFFF. The loan application, completed with Bank assistance in October 1972, was transmitted to the Bank by Government in November 1972. Appraisal 3.11 The project was appraised in January/February 1973 by a six-man team. Whereas the preparation mission included livestock, range and irrigation specialists, the appraisal mission included specialists to cover: - organization, management, staffing and training; - economics and marketing; - banking, credit, LDA financial structure; - irrigation and crop production; - ranch and feedlot development; and - abattoir development. - 25 - 3.12 The project as appraised contained only three significant design changes from the preparation document: the reduction from three to two production ranches, a slight reduction in the size of the irrigation farm and feedlot capacity and changes in the locations of four of the field veterinary centres. The estimated project cost at appraisal was also nearly US $ 1 million higher, largely as a result of a downward adjustment of the fixed Somali Shilling - Dollar exchange rate. Project financing arrangements were similar to those in the preparation report except that Bank/IDA funds totalled 87 per cent of project cost as against 82 per cent at preparation. 3.13 The extensive manpower resources devoted to the identification, preparation and appraisal of the project resulted from the Bank Group's desire to assist the Somali agricultural sector and the relative diffi- culty of identifying and processing suitable agricultural development projects in the country at that time. Negotiations and Effectiveness 3.14 No special difficulties or issues arose at Negotiations as most of the proposals had already been exhaustively discussed with Government by the identification, preparation, appraisal and post-appraisal missions. The project was presented to the Board on January 21, 1974 and the Development Credit Agreement (DCA) was signed on February 20, 1974. 3.15 June 21, 1974 was stipulated as the Credit Effectiveness date, following fulfillment of conditions required by the DCA. The principal Credit Effectiveness conditions arose from the findings of the appraisal mission and included: (i) finalisation of legal instruments; (ii) formalization of the Project Agreement with LDA(as the executing agency); (iii) signing of a Subsidiary Loan Agreement between Government and the LDA; (iv) transfer to LDA of an area of 1,900 km as the requirement for ranches, holding ground and IFFF; (v) appointment of management consultants and/or staff to fill six senior technical positions; and (vi) provision of LDA's audited financial statements. - 26 - 3.16 In March 1974, before the Credit became effective, Government had proposed forming a Livestock Trading Agency (LTA) and to transfer project implementation responsibility to it. The Bank advised GOS not to follow this course. The joint HQ/RMEA mission to Somalia in May 1974 convinced GOS of the complications and delays such a transfer would cause. A compromise was reached under which the responsibility for supervision of LDA was transferred from MOLFR to the Ministry of External Trade. 1/ 3.17 Inability to complete effectiveness conditions required a postponement of Credit Effectiveness to August 31, 1974. However, by the end of July all conditions had been met with the exception of appointment of consultants. LDA had chosen a company but negotiations with the firm broke down in mid-August due to disagreement on fees. Subsequently, following a further postponement of Effectiveness, LDA was successful in negotiations with another consulting company and Effectiveness was finally declared on October 29, 1974. Project Description :3.18 The principal aim of the project as appraised was to improve cattle marketing and raise livestock production (meat output) in the Trans Juba area. The strategy adopted had three major components: (i) improvement of markets and stockroutes to facilitate purchase and movement of cattle and to reduce weight losses and mortality during trekking; (ii) development of grazing ranches to allow growing-out and further conditioning of those animals not suitable for immediate slaughter and to produce feeder grade stock; and (iii) development of the IFFF to fatten cattle suitable for the KMF which would in turn sell its output as frozen or chilled beef on the world market. 3.19 The project was thus designed to increase export earnings from greater output of canned meat, and to initiate sales of frozen and chilled beef on the export market. Improved veterinary care and the greater weight gains were expected to increase returns to the economy, incomes of the nomads and profitability of the LDA and the KMF. A later sophistication of the marketing component was to be the development of a disease-free zone; the contribution to the disease control program by the project would assist both the pastoralists and LDA through access to high-value export markets for cattle produced in the disease-free zone. The project was also intended to provide an outlet for immature cattle being sold through informal cnannels to Kenya. 1/ Subsequently transferred to MOLFR in October, 1976. - 27 - 3.20 The following incremental benefits were projected for TJLP at full development in the sixth year of operation: Table 3 Project Incremental Benefits Production Parameters Amount per Annum - incremental cattle sales by LDA 52,000 head - fattened 25,000 head - canned 19,000 head - live export 8,000 head Ranch throughput (sales) 20,000 head Feedlot throughput (sales) 25,000 head Total carcass weight from fattening 4,625 tons Incremental live exports US$ 6.6 million Feedlot growth rate (kg/head) 0.93 kg/day 1/ Beneficiaries (pastoralists) 20,000 families - 1/ In project lifetime. Project Components 3.21 2The total area required for the major project components was 1,900 km and the Government made the area available to the LDA for its exclusive use at no charge for an unlimited period. 3.22 Project components are summarized below and discussed in Annex 1. The components were not designed in detail and final site selection was not completed prior to project commencement. These functions were to be executed by consultants appointed to key positions in the Project Development and Training Unit (PDTU) and Project Construction and Manage- ment Unit (PCMU) both of which were to be formed in the LDA to implement the TJLP. Project components were: (1) Marketing of Cattle: - establishment of 5 permanent cattle markets; - upgrading of over 1,000 km of existing stockroutes through clearing and development of water supplies at staging points; - development of one, 20,000 ha holding ground; and - further development of existing holding ground at Kismayu. (2) Growing and Conditioning of Cattle: - development of two, 80,000 ha grazing ranches each to provide an output of at least 10,000 cattle per annum for feedlotting/slaughter. (3) Fattening Cattle: - development of one, 1200 ha irrigated fodder farm to feedlot 25,000 cattle per annum supplied from the ranches and holding ground. - 28 - (4) Disease Control - development of 5 veterinary centres (including Regional Veterinary Centre, Kismayu); and - development of 5 veterinary dispensaries. (5) Technical Assistance and Financial Assistance with Oparating Costs - provision of staff and technical services; - development of feasibility studies and future project preparation facility; - provision of funds for operating costs of LDA for the first three years; and - arrangements for training in Somalia and overseas. Organization and Management 3.23 Overall responsibility for development of the project was vested in the LDA, which was already charged with the development of marketing infrastructure and control of all livestock exports in Somalia. The PDTU, under a Project Manager directly responsible to the General Manager, LDA, and with a total of twelve professional staff, was the entity responsible for the day to day management of all aspects of project implementation.(Annex 2). 3.24 The Appraisal Report recommended that the design consultants for the IFFF and the management consultants for the PDTU be recruited from the same organization. A PCMU was planned within the PDTU to take responsi- bility for all construction with the exception of the boreholes which were to be drilled by the Water Development Agency (WDA) and the IFFF which was to be developed by contractors. 3.25 The disease control program and feasibility studies were to be directed by the Veterinary Department and the Planning, Training and Research Department of MOLFR, respectively. Project Financing 3.26 Total project cost was estimated at So.Sh 71.8 million (US$ 11.5 million) of which So.Sh 45.6 million (US$ 7.3 million), or 64 Dercent, was expected to be foreign exchange. The IDA Credit of US$ 10 million, which was to finance the project's foreign exchange requirements plus tne equivalent of US$ 2.7 of local costs, was provided at standard IDA terms and was to finance approximately 87 percent of total project cost. 3.27 Government was to provide the equivalent of US$ 1.5 million as its contribution (including taxes and duties). In addition, the Somali Development Bank was to provide an overdraft to finance LDA marketing operations in the first four years up to a maximum of US$ 640,000 equivalent at any one time. - 29 - IV IMPLEMENTATION Project Commencement and Initial Development 4.01 The LDA appointed an international consulting firm to staff the specialist positions of Project Manager, Senior Grazing Ranch Manager, Irrigation Specialist, IFFF Manager, Master Mechanic and Project Finance Officer. The.consulting firm, using the above staff and additional short-term specialists, was also to provide for detailed design of project components. 4.02 The consultant team commenced work in January, 1975 and spent the first year on basic project development exercises, including: (a) recruitment of Somali staff; (b) detailed planning of work programmes; (c) preparation of specifications and tender documents for purchase of equipment, construction of buildings, drilling of boreholes and irrigation design on the IFFF; (d) design of a monitoring system; (e) design of an accounting system; and (f) coordination of the work of PCMU in preparing designs and tender documents. 4.03 The consultants felt that the remoteness of the project area necessitated the tendering of a single, large contract for project infrastructure. Thus, tendering did not proceed until all detailed bills of quantities were prepared. Tender documents for developments on the grazing ranches, the holding ground water supply and the IFFF were completed by August, 1975. Early in 1976, as part of the detailed planning phase, the consultants carried out an in-depth cost review of each project component from which they concluded that the project would cost in excess of 100 per cent more to develop than projected at appraisal. Cost increases in the irrigation and feedlot components were noted to be particularly high. 4.04 It was during this phase of project implementation that the first evidence of the basic weakness in organizational structure, which was later to prove such a constraint on successful project implementation, came to light. The consultants appointed to manage the PDTU were dependent on administrative and financial support from LDA to enable them to appoint complementary staff and to form the PCMU. The consultants' efforts were, however, severely constrained by limi- tations within LDA, including: (i) inconsistency, arbitrariness and delay on the part of LDA management; (ii) shortages of finance and weak financial management; - 30 - (iii) poor overall operating performance, and (iv) inadequate control over project inputs resulting in unnecessary wastage and misapplication. 4.05 Furthermore, in the local staffing of PCMU and PDTU the advice of the consultants was not sought by LDA. In their initial field survey of the project area the consultants expressed serious concern over the non-availability of qualified technicians in Somalia to staff the PCMU. LDA agreed in March 1975 that the National Construction Agency would accord the project top priority and would assign architects and draftsmen to design buildings and quantity surveyors and field staff to supervise construction. LDA itself would recruit the artisans and labour for the actual construction work. These arrangements never became fully effective, however, and shortage of qualified and experienced Somali staff developed to be a serious and chronic project constraint, 4.06 When appointed in May 1975, the Head of the PCML was instructed to undertake the work of the architect as well as to discharge his own responsibilities as engineer and he became heavily overloaded. The manage- ment consultants were continually concerned throughout 1975/76 that LDA was not making sufficient efforts to establish the PCMU as an effective unit and they warned LDA and the Bank many times during the period that serious project execution delays could be expected unless urgent remedial action was taken by LDA. Project Review 4.07 By late 1975/early 1976, about a year after the consultants were in post, it became evident that the project was not developing as anticipated. Parallel to this, KMF had begun purchasing its cattle requirements directly from traders and threatened LDA's existence since LDA expected to supply up to 85 per cent of KMF's cattle for processing. The threat was exacerbated by the narrowing spread between the purchase price of cattle and the selling price of beef (this spread was the main source of LDA revenue and a key determinant of project financial viability). The narrowing price spread was prompted by changes in international beef pricing over which the project had no control. 4.08 Other problems also emerged during 1976 which threatened the project's viability, including: (a) accelerating increases in capital and operating costs; and (b) serious discord within the consultant team and between the team and LDA management -- these differences developed to a point where the consultant team became seriously Dolarised within itself and where relations between the team and LDA became severely strained. - 31 - 4.09 In February/March 1976, IDA focussed particular attention on the question of estimated cost increases, particularly those of the irrigation and feedlot components emphasised earlier by the consultants. The Bank concluded that the cost projections needed review. There were also suspected inconsistencies in design and specifications which, if substantiated along with the cost increases, would point to the need for some restructuring of the project. 4.10 An IDA supervision mission in March 1976 updated post-appraisal cost escalations and they were further reviewed in Nairobi in July 1976 in meetings between RMEA and the management consultants. Overruns in the irrigation and IFFF components were shown to be of the order of 100 per cent and it was concluded that an irrigated fodder farm and feedlot would not be viable. The world price for beef had fallen by some 50 per cent during 1973-76, following US curtailment of imports. The consultants, supported by IDA, recommended the exclusion of the IFFF component and retention of the stockroute , water development, marketing and veterinary components for which project funds would be sufficient (Annex 3). 4.11 Government however, subsequently notified IDA in late July 1976 that it preferred to implement the project in full as designed. The basis for this decision was not clear but coincided with the appointment of a new LDA General Manager who, in contrast to the outgoing General Manager, was strongly in favour of the IFFF component. 4.12 IDA formally decided in September 1976-to mount an in-depth review and a preliminary mission 1/ visited Somalia in November of that year to prepare with Government the full Review Mission. The Mission was mounted in January/February 1977 with the prime task of reviewing the project in full, with emphasis on the IFFF component. Government fully supported the review exercise. 4.13 The Mission's main findings are presented in Annex 4. The most important, in summary, were: (i) the whole of the marketing component should be implemented as appraised with only minor change; (ii) the IFFF component was considered still financially viable and should be implemented in 190 ha of irrigated forage to provide for a reduced annual throughput of 13,800 head per annum rather than 25,000 head; (iii) the ranching component should be developed on one ranch only (Ranch II) initially,on a reduced area of 50,000-ha in the first instance -- development of the other ranch to be delayed two years; (iv) fattening cattle for live export was considered a sounder economic proposition than export of processed products; 1/ The mission coincided with the Somali Government State Planning Commission report on the project of October 1976. - 32 - (v) cost escalation overall was some 25 per cent in real terms in contrast to the consultant's earlier higher estimate; and (vi) performance of the management consultants was considered less than satisfactory. Post Review Project History 4.14 The LDA terminated the consultant's contract in February 1977, at which time a local Project Manager was appointed who lacked project development experience. Project accounting lapsed and implementation plans and timetables started to be made independently of the project development plan. Progress was slow in recruiting senior staff to replace the consultants and it was not until October, 1977 that the General Manager of IDA visited India to recruit the staff required. IDA objected to the proposed candidates for Ranch Manager and Feedlot Manager on the basis of their inexperience, but Government proceeded with the appointments. 4.15 By mid 1977 Government had agreed in principle with the findings of the Review Report and had decided to implement the project in accordance with its findings. In July 1977, LDA concluded an agreement with FAO to pro- vide an Irrigation Engineer and an Equipment Specialist for 12 months to com- plete the IFFF design and installations. At the same time, LDA also finalised a contract with WDA for the drilling and equipping of 26 boreholes on the IFFF, holding grounds, ranches and stockroutes. 4.16 In January 1978 LDA was relieved of its right to trade in cattle in the lower Juba area. 1/ In addition, the Project Manager and the Project Engineer resigned. The reasons for these latter moves were never made clear but it was believed to be due to dissatisfaction with relationships existing between LDA and the project. 4.17 By May 1978 two of the four Indian professional staff recruited earlier by the General Manager decided not to take up their appointments and only the Feedlot and Ranch Managers arrived. To add to the confusion, LDA contracted with the Somali Engineering Consultancy Agency (SECA) for design of the irrigation component, having already agreed to the appoint- ment of the FAO Irrigation Engineer who was already on site. The latter understandably became upset and the SECA contract was subsequently cancelled. 4.18 During 1978 LDA showed increasing disinterest in the project and a decreasing will to act in addressing the project situation. In May 1978 the Bank Resident Representative advised the Minister of MOLFR of the seriousness of the situation and informed him that the Bank might have to take drastic action. On technical issues LDA showed increasing disinclination to consider Bank recommendations or proposals. For example, IDA insisted on pressing ahead with the construction of the Ranch II headquarters buildings notwithstanding that they knew them to be poorly sited and had been advised to relocate them by both IDA review and supervision staff. It also insisted on constructing the whole feedlot complex instead of a small pilot facility aq advised by the Bank. 1/ Refer Annex 7 for details. - 33 - 4.19 In May 1978 IDA telexed Government to the effect that (i) a deadline should be set by which time the issues of management, technical assistance and water rights should be resolved; (ii) project performance must improve visibly; and (iii) specific performance criteria and an implementation timetable should be established. Failure of the LDA and Government to act appropriately would lead to a decision by IDA to suspend disbursements. 4.2.0 By July 1978, just prior to the PIR, it was noted that urgent action needed to be taken on the following: (i) clarification of responsibility for cattle marketing in the project area; 1/ (11) supply of working capital to be provided to the project by Ministry of Finance; (iii) discussion on a draft Water Law for the Juba River; (iv) return of vehicles and equipment confiscated during the Ethiopian/Somali conflict; (v) need for an engineer replacement in order to provide engineering supervision for the construction program; (vi) need to fill the vacant Accountant's post urgently to provide for proper project accounts; and (vii) need for WDA to appoint a Hydrogeologist to provide technical support to its drilling teams working in the project area. 4.21 At the PIR meeting on July 11, 1978 with the Ministers of Livestock and Finance and the Presidential Advisor, proposals for changes to project management were made by the Government. Of most substance was the proposal to replace LDA by MOLFR as the executing agency. Other key issues (para 4.19) were extensively discussed and the Government made a convincing case to the Bank that it could resolve all matters and put the project on course by mid-August. It also agreed to return or replace all vehicles and equipment commandeered for use in the earlier war. 4.22 Notwithstanding strong pressure to do so, Government did not relieve LDA of its project responsibility and the General Manager subsequently adopted an aggressive stance, accusing IDA of -painting a black picture*, and *forcing unilateral decisions and interfering 1/ LDA was subsequently permitted to purchase cattle in the Upper Juba area (para 4.24), but most of its marketing teams were transferred to the KMF. LDA consequently had difficulty re-establishing its teams and long delays ensued. - 34 - in purely internal matters . The telex containing this attack went unanswered although the Resident Representative discussed the substance of the allegations with the General Manager. 4.23 By early December 1978, following a full supervision, it was reported that the project situation appeared to have improved in the areas of staffing and implementation capacity. However, the ranching, holding ground, feedlot and stockroute components had shown negligible progress and the performance of the expatriate professional staff was viewed as generally unsatisfactory after some six months in post ; also, LDA continued to exercise bureaucratic control from Mogadiscio and its arbitrariness, overall inefficiency and delayed decision-making continued as serious constraints on project development. 4.24 On January 7, 1979 a law was passed fully restoring LDA's livestock marketing function in the Lower Juba area. At the PIR held the next month, it was concluded that despite remaining problems there had been an improving trend since the PIR in 1978 and IDA agreed to continue the project, the future of which had been in the balance. 4.25 The following implementation problems were noted at the 1979 PIR, some of which were readily resolvable, others were not. In discussing them Government gave strong assurances that it would do all possible to address those with solutions expeditiously: (i) price increases and cost overruns; (ii) delays in tendering and for importing materials and equipment; (iii) shortages of building components and machinery spares; (iv) dense bush cover on Ranch II and bush regrowth after physical clearing; (v) high costs of physical or chemical bush clearing; (vi) unusually wet weather in 1977 and 1979 which delayed development,and floods in 1980 which caused severe damage on the IFFF where considerable repair was needed; and (vii) shortages of working capital due to: (a) LDA's slowness in applying for reimbursements; (b) some expenditures eligible for 92 percentCreimbursement being claimed at only 70 percent; and (c) accounts and reimbursement claims being in general disarray due to lack of qualified accounting staff. 4.26 The precarious project financial situation was not unrelated to the disarray of LDA's accounts. With an advance of So.Sh 5 million from Government and reimbursement of different project cost categories at either 70 or 92 percent, the project should have had sufficient working capital in its revolving fund. However, with inadequate accounting, con- sequent weak financial control and slow application for reimbursement of eligible expenditures, chronic project funding constraints continued as a crucial bottleneck to project execution. - 35 - 4.27 Damage to the facilities of Ranch II early in 1980 by both elephants and nomads climaxed a series of ranch development constraints (bush cover, siting of buildings, lack of potable underground water). This occurred at the same time as the visiting consultant for the irriga- tion component concluded his assignment with a report indicating the viability of the IFFF to be technically questionable. Thus, by March 1980, with the project limping, IDA concluded that there was no alternative but to suspend development of the ranch and the IFFF and to limit the project to the salvage of the stockroute, marketing and veterinary components that would be of use to the traditional livestock sector. Further details of the Bank's position are in Annex 5. 4.28 The LDA requested an extension of the Closing Date from June 30, 1980 to December 31, 1981. After concluding agreements with the Government that the completion of the stockroutes, boreholes, holding grounds and veterinary dispensaries would be awarded priority, and that no further development would take place on the IFFF and ranch, the Bank agreed to the Closing Date extension. Progress during the Extension Period 4.29 To provide for further water development in the project area during the extension, the July 1977 LDA/WDA contract was extended in June 1980. IDA also pressed strongly for accounting improvements and for a system of justi- fication of project expenditures acceptable to the Bank which would provide a basis for reimbursement of Statements of Expenditure. These improvements were, however, never achieved to IDA's satisfaction and consequently withdrawal claims submitted by Government for reimbursement of Statements of Expenditure for most project expenditures incurred after June 30, 1980 were not approved because of lack of adequate justification 1/. 4.30 Contractors employed to clear stockroutes progressed well in 1980/81 but the WDA experienced several months delay in carrying out its work under the extended contract to drill seven wells in the northern part of the project area. These delays plus consequent difficulties experienced in the procurement and financing of power/pump units for the additional boreholes, prevented the proper equipping of the boreholes in the life of the project (see, however, para 5.07). 4.31 In March 1981 the Government disbanded LDA because of poor overall performance, high financial loss and weak economic impact 2/. While the Government's move was fully justified, the timing brought the project to almost a complete halt for the final nine months of the extension due to the confusion and delay in reassigning LDA's former responsibilities within MOLFR. Responsibility for infrastructural aspects (stockroutes, stock-water, holding grounds) was tentatively planned by Government to be assigned the National Range Agency (NRA) of MOLFR and responsibility for all other aspects was to be assumed by the Minister, but the exact delegation of responsibility was never clearly spelled out by Government. Consequently, the project literally limped to a close with the Project Manager being the only professional staff making any input. He with the Bank's support, concentrated on procuring the pump/power units 1/ This was a major contributing factor*to the Credit not being fully disbursed and the need for fund cancellation when the last disbursements from the Credit were made by IDA in Oct. 1982 (see also paras 6.05 and 6.14). 2/ This followed the closure of KMF by Government in December 1980 for essentially similar reasons. - 36 - to equip the boreholes. The procurement was successful but installation was not achieved. WDA adamantly refused to fulfil that part of its contract and the price required by the one private contractor in the area who was capable of doing the work was regarded as unacceptable by IDA. Given the general climate in the Bank at the time of the December 31, 1981 project closing, there was no support for extending the project to facilitate completion of the equipping of the boreholes. The Bank did, however, initiate discussions with Government on the possibility of finishing the work under another IDA-financed project. Details are in para 5.07. Project Physical Achievements 4.32 During the greater part of its life the TJLP was a problem project and from 1976 was given a -3- status rating. Delays and poor planning led to cost increases, finally greatly in excess of the 25 per cent estimated by the Review Mission, with the result that project scope (development) and activities (operating costs) were progressively reduced or suspended. LDA and project management were unable to redefine objectives and concentrate on achieving specific targets for the most important component, the stockroutes and holding grounds, with the result that most of the boreholes developed by the project were not equipped when the project closed, thus preventing a major potential economic benefit from being realised. 4.33 Table 4 presents the project components planned at appraisal, the revised plan of the Review Mission and the project status by the Closing Date. 4.34 Despite intensive supervision efforts by IDA, including technical advice on irrigation and feedlot design, evaluation of feedlot rations, chemical bush control experiments, preparation of cost budgets and a system to justify reimbursement of operating costs, no physical component of the project was ompleted th full and accounting and inventory control never improved. 4.35 The most substantial infrastructural achievements were: (i) formation of over 550 km of stockroutes; (ii) completion of 15 dams on the stockroutes; (iii) completion of cut lines and clearing of holding grounds; (iv) drilling of 15 productive boreholes (of which only one was equipped). These infrastructural achievements were the only substantial project accomplish- ments. Details of the development of project components are presented in Annex 6. Discussion on the shortcomings of the project's institution building efforts can be found in Chapter VIII. 4.36 In the DCR Mission's view the most important factors contributing to the project's failure to reach its objectives were: (i) size, scope and complexity of the project were too large in relation to local managerial, staff, technical and financial resources; (ii) planned rate of development was over-ambitious;- - 37 - (iii) inadequate performance of LDA, the implementing agency; (iv) ineffectiveness of PDTU/PCMU due to inadequate support from LDA; (v) shortage of experienced Somali staff and difficulty in retaining them (due to lack of incentives and encouragement and the competition of highly paid jobs in other areas, e.g. Middle East); (vi) difficulty of attracting suitably qualified expatriates to Somalia due mainly to LDA's reluctance to provide adequate remuneration and attractive working conditions; (vii) delays in design and planning due to late arrival and performance of PCMU staff; (viii) shortages of materials and spare parts; (ix) improper siting of major physical components including: (a) site selection of Afmadu holding ground; (b) site selection of IFFF; and (c) site selection of ranches in areas of heavy bush -- also, site selection for ranch buildings in inaccessible areas; (x) chronic shortage of working capital and development funds; (xi) inadequate performance of WDA in developing livestock water; and (xii) diversion of project funds, equipment and vehicles for non-project purposes. 4.37 These factors are discussed in more detail in Annexes 6, 7 and 8. TABLE 4 - 38- Page 1 80ALIA TRANS JUA LIVWSTFW - CRDIT 462-SO PROJECT CMPLETION REPORT PR CT PHYSICAL ACHIEVWRTS - IMPLEENTATION STATUS No. Appraisal Objectives Appraisal Design Review Mission Revisions Actual Achieved 1. MARKETS, STOCKROMUTS AND HDIING GROUNDS Improvement of marketing facilities (1) Con8truction of 3 (1 Recommended to (10 One market completed (including stockroutes, holding large cattle markets at Tabda but not grounds, livestock water and at Tabda, Kazkati maintain scope of operative; markets) to facilitate sales and Gelib; and movement of cattle and to appraisal plan but reduce weight loss and mortality during trekking. These in turn would improve productivity anu offtake. Stockroutes to connect development of markets and holding grounds; latter to serve for quarantine, temporary markets selection and regulation of flow (2) construction of 2 (2) 2) neither completed; small cattle markets prior to construction at Lugh Farrandi and Busar; Jof permanent markers; (3) development of water (3) recommended that 26 3) 21 boreholes drilled of supplies and staging boreholes be which 15 were potable; points 30 km apart constructed (13 in however, oly one on over 1000 km of 1977; 13 in 1978 ;no peimanently equipped stockroutes; develop- change in stockroute (YIF) and one ment of some 26 bore- development program temporarily equipped holes and 32 ponds! except on stockroute (Aglibah holding dams. aligment away from ground); 15 ponds/dams Juba River; successfuly constructed. (4) construction of (4) vecomended to proceed (1) holding ground completed 20,000 ha holding as soon as possible, as at Aglibah but new bore- ground at Afmadu, designed at appraisal. hole only temporarily equipped; (5) sh clearing of (5) recommended to (5) 21 bush cleared; one 12,000 he and restrict bush clearing water point developed; and development of 2 to pilot scale but to water points on proceed with water existing ground development as at Kismayu; and planned; and (6) procurement of (6) v planned at (6) pghicles, radio- telephones, vehicles and appraisal. camping equipment provided equipment for aDA to LDA. marketing teams. 2. GRAZING RANCHES To develop two, 80,000 ha Development of (i) Recommended that ranch (i) Development of Ranch II ranches for conditioning! one, 80,000 he near IFFF (Ranch II) suspended in 1980 due to fattening cattle procured ranch net IFF; be developed on lck of potable water, 'from pastoralists for and reduced scale of heavy costs of bush channeling subsequently 50,000 ha initially; removal and encroachment to !FFF and, in turn, to and and damage by elephants KM4C. A central activity and nomads. Some fire- would be to supply a breaks and roads completed; total of 20,000 cattle buildings (hich were poorly per year to I1dFF for serev) were partly constructl eb later processing by K0ne store and building materials and export as frozen or destroyed by fire and well chilled beef. casing damaged permanently by nomads resisting ranch dievelopment; and (ii) another 80,000 ha (ii) recomended that (ii) io developments undertaken unit to be developed development of Ranch I on Ranch I. near Afmadu holding be postponed two years. ground -- both ranches to be lovided with perimeter fences, firebreaks, roads, buildings, water, vehicles, machinery, etc. - 39 - SOMALIA TAL 4 TRANS JUBA LIVESTOCK PROJICT -- CREDIT 462-SO PROJECT COMPLETION REPORT PROJECT PHYSICAL ACHIEVEMENTS - IMPLEMENTATION STATUS No. Appraisal Objectives Appraisal Design Review Mission Revisions Actual Achieved 3. IRRIGATED FODDER FARM AND FARM FEEDLOT (IFFF). To produce continuous A. Development of a A. Recommended to A. 200 ha cleated of bush; supply (25,000 head per 1200 ha irrigated reduce irrigated some 90 ha semi- annum) of high quality forage farm area to 190 ha prepared of which 30 ha cattle for KMF for utilizing surface initially, building completely prepared; processing and export irrigation using possibly to 300 ha main irrigation canal as frozen and chilled water pumped from maximum (LDA, how- built to 1200 ha beef. Cattle to be Juba River through ever, decided to capacity;irrigated forage supplied to the IFFF 2 km canal;construction build irrigation program was too complex for by the ranches (20,000 of houses, warehouses, canal to 1200 ha local management; houses, head per annum) and offices, workshops and capacity); Mission con- warehouses and other Aglibah holding ground other necessary structures cluded that disposal of essential buildings (5,000 head per annum) to provide essential in- fattened cattle by export- completed; and fattened on frastructure to allow ing live to Gulf States irrigated forage IFFF to run as an was sound economic (maize, alfalfa and independent, self- proposition than export Rhodes grass fed as supporting unit, of processed products. ensilage and green feed) B. construction of B. recommended to reduce B. 40 pens, troughs, shelters fscilities (pens, facilities to cater plus accompanying facilities troughs, shelter, for throughput of completed for throughput scales, dips, etc.) 13,800 rather than of 13,800 head per annum; to cater for through- 25,000 head per and put of 25,000 head annum; and per annum; and C. provision of equip- C. recommended to dispose C. some equipment on order at ment and facilities of some surplus equip- time of eview was re- f or irrigation, ment and to purchase dundant; other equipment forage production, other essential items, for molasses feeding, etc. feed harvesting! recommended by Review was conservation, procured but not used -- storage, mixing much of it was still crated and distribution. during FCR Mission. Note:One pilot feeding trial was completed in course of pro- ject the results of which indicated that key productia coefficients were seriously overestimated at appraisal and review (Annex 6, pars 26). 4. ANIMAL HEALTH To provide additional 1. Construction and I Recommended to I. Limited support with support to existing equipping of proceed as planned construction of Regional veterinary programs Regional Veterinary but to replace Brays Centre but increased of the Veterinary Centre, Kismayu, and with Badada; assistance with drugs Department of >'ULFR establishpent of four and equipment; none of with view to de- centres at Afmadu, other 4 centres constructed; creasing disease Gelib, Giamma and the overall plan and its incidence generally Brays; implementation could have and reducing been improved if there had animal losses; besides been better contact between improved disease the project and the Regional control, project wasas t me to set up disease-free eteeinay teprojet andG zone to facilitate etwehn tus e a i no thr access of locallyins processed frozen and chilled beef to European markets. II. establishment of 5 B1. recommended to proceed BI. 5 dispensaries constructed veterinary dispensaries as at appraisal but to but workmanship was inferior at Bli Haji, Gobs, review siting to ensure one facility abandoned, one Anjail, Beles Gogani dispensaries were poorly sited; 3 and Tabda; and established at suitable rehabilitated and currently locations and operated by GTZ; and III. provision of vehicles, III. recommended to expedite III. procured as planned; some drugs, equipment and procurement, drugs misplaced; some operating expenses. equipment unsuitable. - 40 - TABLE 4 Page 3 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT PROJECT PHYSICAL ACHIEVEMENTS - IMPLEMENTATION STATUS 0. ADpraisal Objectives Appraisal Design Review Mission Revisions Actua Achieved . ZICHNICAL SERVICES To direct and (a) Establishment of Project As planned at appraisal (a) PDTU established with supervise all Development and Training appointment of management aspects of project Unit (PDTU) within LDA consultants who commenced development; to with 12 professional work early 1975. PDTU lacked provide design and staff (six expatriates); support and encouragement tendering services; from LDA. Consultants' per- to administer and formance in turn not supervise project satisfactory and their training. contract terminated by LDA early 1977. Thereafter PDTU staffed with essentially un- qualified personnel with result that standard of overall project.management and technical services was poor. PDTU was effectively disbanded when LDA was dissolved in March 1981; to undertake (b) establishment of Project (b) PCMU never fully functional; construction work Construction and Main- inadequate staffing and and supervision of tenance Unit (PCMU) funds provided by LDA; contractors involved within LDA to carry out in developing water and supervise develop- supply, IFFF, ranches ment of infrastructure; and holding ground; ) upgrading of staff (c) provision of overseas (c) 3 senior staff trained capabilities at all training for LDA and overseas; 2 returned to levels; and other Government staff; project. 5 accounting staff and in training overseas at project's close; and d) to provide for (d) undertaking of (d) feasibility study of ground- future livestock feasibility studies water potential by development in for future consultant -- no concerted Somalia. livestock projects effort to identify or in Somalia. prepare a follow-up phase or other project in the live- stock development field. 6. 'PERATING EXPENSES -3 provide bridging Provision by IDA of major As planned at appraisal Expenses reimbursed only until inance for LDA to portion of project June 1980 after which little support project operating expenses for disbursement of expenses due to jntil such time as first three years. unsupported SOE's; accounts not -t was self support- well kept after 1976; GOS _ng (projected by financing inadequate and und of year 3). constant working capital shortages. 7. ROADS pgrading of road Repair of Badada-Kismayu As planned at appraisal Some work carried out on force Road (143 km). account by PCMU. Impact minimal. - 41 - V. PROJECT IMPACT General 5.01 As a result of unsatisfactory implementation performance, the only signi- ficant project impact was in the form of the limited physical achievements described in para 4.35. Unsatisfactory LDA operating performance and aban- donment of the ranches and the IFFF resulted in the failure of the livestock purchasing and fattening components of the project and, consequently, the principal objectives of improved off-take/incremental weight gains/marketing development were not achieved. Institutional Development 5.02 The project had no beneficial role in improving institutional development in Somalia. Unsatisfactory LDA performance was not able to be rectified and this, along with high financial losses, led to its dissolution by Government in March 1981. A similar fate befell KMF in December 1980 for the same reasons. The project thus did not lead to the formation of an effective organizational/management structure suitable to support future development of the livestock subsector in Somalia, as anticipated. Studies and Training 5.03 Negligible studies were completed on bush clearing, range improvement, range forage production and carrying capacities, traditional pastoral production and constraints, extension amongst pastoralists (in terms of project objectives), or study of existing marketing systems. One Economist, an Irrigation Specialist and an Accountant were trained overseas by the project (the Accountant did not return to Somalia). The project also contributed to the overseas training of five LDA staff in accounting; they were still undergoing training at the time that the project closed. Technical assistance and a small amount of irrigation/feedlot training was provided by short term consultants. The overall studies and training effort, however, fell far short of appraisal expectations and its impact was minimal. Potential Benefits 5.04 The project contributed to the regional disease control programme by providing some infrastructure that is being used by the Regional Vete- rinary Services and which in future could benefit the economy by decreasing disease incidence and wourrality. 5.05 As noted earlier, the project created a limited but useful infrastructure of stockroutes, livestock water supplies, veterinary services, a market and a holding ground. However, the potential favourable impact on trekking and marketing of cattle by pastoralists through improvement of stockroutes and water points were not fully realised due to the non-equipping of the boreholes on the stockroutes. Of the 15 potable boreholes drilled under the project, only one was permanently equipped (at the IFFF), although one more was temporarily equipped (at the Aglibah holding ground), Map 3 provides details of the location of the above infrastructural developments. - 42 - 5.06 The stockroute water system could lead to substantial future benefits if the Somali Government completes the installation of pumps and motors on boreholes, maintains the equipment and controls range use by regulating water supplies. Benefits could include: (i) easier movement of livestock and better condition of animals at the time of marketing; (ii) lessening of range pressure around traditional water resources and better use of range in areas where water has been provided; and (iii) improved productivity of the livestock subsector generally, due to (i) and (ii) above. 5.07 There is the possibility that the completion of the boreholes estab- lished under the project which were not equipped at the project's close. may be achieved under another IDA-financed project. Discussions with NRA of MOLFR have tentatively identified the Central Rangelands Development Project (Credit 906-SO) as a potential vehicle. This matter is being followed up by the Bank staff responsible for the supervision of that project. Completion of the TJLP livestock water program by this means would overcome the current unsatisfactory situation and ensure more effec- tive application of the extensive investments made in the stockroute-water system under the Project. Rates of Return 5.08 At appraisal, separate financial rates of return were calculated for the marketing and production components. The financial rate of return for the whole project was estimated at 14 percent. Neither financial rates of return nor the economic rate of return for the project was calculated by the PCR Mission since it was concluded that the actual rate of return of each component was negative, and that therefore the financial rate of return of the whole project was also negative. 5.09 Marketing Component. Excluding costs not directly related to the marketing or production aspects of the project (eg. feasibility studies, animal disease control and training), and including a prorated share of the capital costs of project administration, the appraisal estimated a 16 percent financial rate of return for the marketing component. The sensitivity analysis revealed that an increase of only 10 percent in cattle buying prices would reduce the rate of return to as low as two percent. The appraisal addressed this risk by including the condition of Government consultation with IDA on an annual basis, to "... maintain reasonable prices and margins for all participants in the project". 5.10 During a March/April 1976 supervision, it was noted that project costs had risen sharply and that world market prices for beef had fallen by about 50 percent during 1973-76. In the project area cattle prices had risen from So. Sh 1.44 per kg at appraisal to So. Sh 1.90 per kg, a 32 percent increase (Annex 3). Furthermore, LDA had at that time been relieved of its responsibility to purchase cattle for the KMF (Annex 7). - 43 - 5.11 Details of the LDA financial performance in livestock marketing during the project could not be obtained by the PCR mission and were reported by the Development Department of NRA to be unavailable. The LDA was believed to have outstanding debts with local banks of approximately So Sh 80 million by the time of its disbandment. Most of that amount was suspected to have been lost on cattle marketing transactions. During the project IDA did not develop a regular marketing programme and conse- quently did not gain a reputation amongst pastoralists as a business-like and creditworthy purchaser, which was essential for it to compete with active and skilled traders. 5.12 It is certain that IDA lost heavily not only on its own account in the Trans Juba area but also when buying on account for the KMF, as the income Qenerated was not sufficient to cover LDA's operating costs due to the small margin agreed with KMF. The project development costs of the stockroute and markets were over So Sh 13 million (Table 6) and other significant costs were also attributable to this component. However, IDA did not generate an income sufficient to cover its operating costs and to provide a return on this investment. It was concluded, therefore, that the financial rate of return on the project's marketing component was negative. 5.13 Production Component. The appraisal report estimated the finan- cial rate of return of the production component, which included the two grazing ranches and the IFFF, to be 13 percent. The sensitivity analysis indicated a reduction to 12 percent in the rate of return for a 10 percent increase in capital costs or a reduction to 10 percent for a similar increase in operating costs. Viability was more sensitive to cattle prices, the rate of return falling to 10 percent for a 10 percent increase in the purchase price of fattening animals and to seven percent for a 10 percent fall in sale price after fattening. 5.14 The construction and equipment costs of the ranches and IFFF exceeded So Sh 30 million (see Table 6, Items 3-6). In addition there were LDA operating costs and consultants' costs also attributable to the production component. Since only a small feeding trial on the IFFF involving fewer than 100 cattle was completed, almost no revenue was generated. The two ranches generated no benefits whatsoever. The PCR mission concluded, therefore, that the rate of return on the production component was negative. 5.15 Economic Analysis. The principal intended economic impact of the project was to boost Somalia's foreign exchange earnings from value added in the livestock industry. Based on a series of adjustments to the costs and revenues of the project (detailed in Annex 25 of appraisal report), the appraisal estimated a 25 percent economic rate of return. Sensitivity analysis was restricted to a 10 percent increase in costs (which reduced the rate of return to 20 percent) and a 10 percent decrease in selling prices (which also reduced the return to 20 percent). Individual project components were also evaluated. The LDA cattle marketing component was expected to produce a 39 percent economic of return and the production component a 21 percent economic rate of return. The project did not result in generation of value added, either in livestock numbers (the marketing component) or incremental weight gains (the production component). In fact, no revenues over LDA operating costs were believed to have accrued from the project and the PCR Mission concluded that the economic rates of return of the two basic components and of the project as a whole were negative. - 44 - VI. FINANCIAL PERFORMANCE Project Cost Estimates 6.O1 Project cost estimates are based on reported Government commitments to the project and IDA reimbursements up to October 30, 1982. It is important to note that, in comparing actual project costs with those projected at appraisal, no component of the project was actually completed. Annex 4 gives project cost summaries for appraisal and review mission estimates and the costs calculated by the management consultants. Unfortunately it is not possible to categorise costs actually incurred in the same manner. 6.02 The Closing Date of the project was extended by 18 months from June 30, 1980 to December 31, 1981. An additional period of ten months through Oct. 1982 was allowed for completion of final disbursements. 5.03 Total project cost and IDA contributions were lower than estimated at appraisal, e.g.: Table 5 Total Project Cost Appraisal Actual SoSh US$ % SoSh US$ % ('000) ('000) IDA 62,300 10,000 87 57,136 9,171 86 Government: - expenditure 9,465 1,517 13 - debts 8,978 1,441 14 TOTAL 71,765 11,517 66,114 10,612 6.04 The indirect and administrative costs of the project could not be determined. The Government costs were significantly underestimated because total operating costs were believed to have been higher and accumulated debts are believed to be in excess of SoSh 1.5 million. Poor accounting unfortunately precluded accurate determination of total project costs and the breakdown given Eor project components must be treated as an approximation. No details were provided by Government for operating costs in 1981, which could have been in excess of SoSh 2 million. This estimate is not included in Tables 5 or 6. Total project cost estimates by component are given in Table 6. The data for the table were derived by the Project Implementation Unit (PIU) of the Somali Ministry of National Planning and by the PCR consultant from reimbursement applications. Categorization was completed after discussions with project staff. - 45 - Table 6 Total Project Cost by Component (as at 25/7/82) Project Component A22raisal 1/ Estimated Total Total Total Total (SoSh) (US$) (SoSh) (US$ equiv.) ('000) ('000) 2/ 1. Buildings 3/ 3/ 352,000 56,500 2. Stockroutes and markets 6,806 1,092 13,137,916 2,108,215 3. Holding grounds/ranches 4/ 7,932 1,273 7,134,000 1,145,104 4. IFFF 25,530 4,098 16,822,3CO 2,700,225 5. Machinery and heavy equipment 5,012 804 4,659,000 747,833 6. Farm equipment 5/ 5/ 5,495,000 882,022 7. Office equipment 3/ 3/ 231,000 37,079 8. Vehicles 3/ 3/ 4,404,000 706,902 9. Veterinary equipment and supplies 1,563 6/ 252 6/ 354,000 56,822 10. Consultants 5,398 866 5,321,000 854,093 11. Feasibility Studies 2,000 320 67,500 10,835 12. Training 815 130 689,000 110,594 13. LDA operating costs 2,819 7/ 452 7/ 7,447,348 1,195,401 14. Contingencies 13,890 2,230 - - TOTAL 71,765 11,517 66,114,064 10,611,625 Source: Estimated by Project Implementation Unit of Ministry of Planning and updated by consultant from disbursement files. 1/ Not including contingencies. 2/ Exchange rate of US$ 1 = So.Sh 6.23 3/ Included in other components. 4/ Appraisal projections are for ranches only. 5/ Included in IFFF component. 6/ Included as 'Disease Control' at appraisal. 7/ Included as 'Project Development and Training Unit' at appraisal. -46 - respect to the TJLP. Funds were allocated to cover the operating costs for the first three years of the project. The Government through the Ministry of Finance made available a revolving fund of So. Sh 3 million in 1975. In mid-July 1978, the MOF provided a further So. Sh 2.047 million. In addition the Government was reportedly providing 8 percent counterpart financing for costs in Categories I to VI (a) inclusive and 30 percent in Categories VI (b) to VIII inclusive. 6.08 In May 1979, plagued by shortages of operating funds, LDA requested IDA to provide a US$ 0.5 million advance. A similar approach was also made in January 1980. Both were rejected because the project should have had sufficient working capital and project management was also late in submitting reimbursement applications. 6.09 In late 1980, the seriousness of the cash flow situation resulted in non-payment of staff salaries for a period of approximately six months. The project was at a complete standstill and many staff left their posts. By April 1981 Government provided further funds (amount not known) and since that time it reportedly provided two additional contributions of So. Sh 600,000 and So. Sh 500,000 (dates unknown). 6.10 The Somali Development Bank did not provide a working capital overdraft to LDA for the project's trading activities, as planned at appraisal; the Government's contributions were expected to meet these requirements, in addition to other project operating costs. In fact, however, these funds were also used for development activities undertaken on force account and,consequently, working capital shortfalls were experienced throughout the project period. 6.11 The problem underlying the project's unsatisfactory financial situation, which seriously constrained project implementation, was the lack of proper accounts and consequent lack of financial control and cost consciousness. Neither the Government nor LDA ever made the necessary efforts to resolve the management and staffing issues at the root of the situation. 6.12 In addition to the actual costs in Tables 5 and 6, substantial debts were accumulated against other Government organisations, private companies and members of staff. Known debts were at least So. Sh 1.5 million. Accounts and Disbursements 6.13 Financial planning to determine development and working capital requirements, and accurate project accounting, were practised only in the project's initial stages. Financial recording and reporting were virtually non-existent after the departure of the management consultants. This situation contributed to delays in preparation of reimbursement appli- cations and errors in reimbursement claims. 6.14 IDA attempted on many occasions to persuade Government to improve LDA's accounting in order to instil the necessary accounting discipline and cost consciousness and to meet the accounting/auditing covenants in the DCA. Improved accounting was also essential in order to provide justification for reimbursement of Statements of Expenditure on reimbursement of which the project was heavily dependent in the - 47 - IDA Contributi.n 6.05 By Oct. 14, 1982, IDA had contributed a total of US$ 9,178,092 to the project (Table 7). The Credit was therefore underdrawn by appro- ximately US$ 822,000 and the undisbursed amount will be cancelled. An estimated So Sh 530,000 was still owing to WDA for the drilling of five productive and two saline wells under the extended LDA/WDA contract. IDA was prepared to make reimbursement but the relevant application - No. 158 - was incorrect and was returned for correction but no attempt was made by the project to resubmit it (see also paras 4.29 and 6.14). 6.06 Overspending occurred in categories I, II, III and IV. These cost overruns were attributable to price increases and purchases of equip- ment (Category IV) not anticipated at appraisal. Some of the additional equipment was also purchased as a result of project Review Mission recommendations. Overruns were met from Category IX (see para 6.20). IDA expenditures as of the date of the last disbursement were as follows: Table 7 IDA Expenditure as at 10/30/82 US$ - 2/ Category Appraisal Actual Balance- I Irrigation Works 1,550,000 1,569,748 - 19,748 II Markets, Stockroutes 1,750,000 1,877,133 - 127,133 III Vehicles 1,250,000 1,343,729 - 93,729 IV Equip. and Bldg. Matpriqls 870,000 2,570,382 -1,700,540 V Management consultants 800,000 762,714 + 37,286 VI(a) LDA staff training 180,000 101,838 + 78,162 (b) Other GOS staff training 90,000 0 + 90,000 VII Feasibility Studies 210,000 7,588 + 202,412 VIII LDA Operating Costs 1,300,000 944,960 + 355,040 IX Unallocated 2,000,000 0 (2,000,000) Totals 10,000,000 9,178,092 + 821,908 3 1/ Exchange rate : US$ 1 = So.Sh 6.23 2/ Balance in Credit (+), Overdrawn (-) 3/ Balance undisbursed; to be cancelled shortly, with cancellation back dated to October 28, 1982. Government Contribution 6.07 At appraisal, Government was to meet 13 percent of total project costs. Since project costs would include taxes and duties, the real contri- bution would be about 8 percent of total project costs. In fact, it appears that Government has covered more than 14 percent of total costs (including taxes and duties). Category VIII provided for LDA operating costs with Table 8 SOMALIA TRANS-JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT - Schedule of Disbursements (US$ '000) IDA Fiscal Cumulative Amount Disbursed Actual as % Year and Appraisal Actual Appraisal Quarter Estimate 1973/74 June 30 200 - 0 1974/75 September 30 500 - 0 December 31 1,000 - 0 March 31 1,500 - 0 June 30 2,000 323 16 1975/76 September 30 3,000 370 12 December 31 4,000 428 11 March 31 4,500 560 12 June 30 5,000 627 13 1976/77 September 30 5,200 745 14 December 31 5,500 983 18 March 31 5,800 1917 33 June 30 6,500 2164 33 1977/78 September 30 7,000 2934 42 December 31 7,200 3156 44 March 31 7,500 3276 44 June 30 8,000 3276 41 1978/79 September 30 8,200 3419 42 December 31 8,500 3557 42 March 31 9,000 3880 43 June 30 9,500 3881 41 1979/80 September 30 9,700 4659 48 December 31 9,800 4810 51 March 31 9,900 5726 58 June 30 10,000 6269 63 1980/81 September 30 7052 71 December 31 7202 72 March 31 7473 75 June 30 8097 81 1981/82 September 30 8119 81 December 31 8212 82 Final (Oct. 1982) 9178 92 - 49 - later stages for its operating funds. The accounting improvements were never introduced nor were alternative justifications for reimbursement under Statements of Expenditure designed. The result was that the Bank effectively stopped reimbursing against Statements of Expenditure for project expenditures incurred after June 30, 1980. The only exceptions were for minor amounts in cases where the expenditures were clearly justified (e.g. tuition fees for students at foreign universities). 6.15 The rate of drawdown of funds from the Credit (Table 8) was slower than expected, primarily because: (i) there were long delays in implementation of project components; (ii) poor accounting led to delays in obtaining reimbursement; and (iii) financing of the implementing agency was inadequate and delayed, leading to a lower level of claims than expected; The higher rate of drawdown after 1979 was largely attributable to payment for project components completed by contractors. Cost Escalation 6.16 Substantial cost increases occurred in the lead time between appraisal (1973) and completion of designs (1976) and before commencement of project development (mid-1977). Long delays in development contri- buted to increased costs as the post-1973/74 oil price increases took effect in raising prices of imported construction materials, vehicles, machinery and equipment. Despite cost escalations, total project costs did not exceed estimates because components were scaled down or were not completed. 6.17 Some appraisal estimates for component costs were unusually low. The reasons for this are not known but may have been due to lack of information specific to the project area or lack of understanding of condi- tions that posed risks of increased costs (e.g. logistics, access to project areas, lack of knowledge of borehole drilling conditions, shortages of materials). In addition, costs estimated in US dollars and converted to Somali shillings automatically resulted in some under-estimates due to high valuation of the local currency at the exchange rate of US$ 1 = SoSh 6.23 used throughout the life of the project. 6.18 It is difficult to compare actual costs to those projected at appraisal and review because of the incompleteness of project components at the project's close. Over-expenditure of IDA funds and estimated total costs occurred in categories I and II (Civil Works), III (Vehicles) and IV (Equipment and Building Materials). The over-expenditure of funds under Category IV of nearly 200 percent was due to: (i) price increases of equipment and materials; (ii) purchase of excessive equipment (e.g. IFFF farm and feedlot equipment); - 50 - (iii) re-purchasing of additional equipment after sale of original heavy equipment; 1/ and (iv) purchasing of certain equipment recommended by the Review Mission that was not included in appraisal estimates. Procurement 6.19 Long delays were experienced in the completion of Letter of Credit formalities, necessary for importations. No special GOS priority was awarded to development projects and it was necessary for project staff or procurement agents to follow procedures which were often excessively complicated. Consequently, the use of Procedure 3 for purchase of cons- truction materials and heavy equipment was used where possible in order to expedite procurement. 6.20 Shortages of foreign currency led the Government to request IDA to pay all foreign costs of materials and equipment purchased for the project. In September 1979, IDA formally approved an amendment to the DCA, altering the disbursement rate in Cateogry IV to "100% of foreign expenditures and 92% of local expenditures". There was also a revision of the allocations by Category at that time as detailed in Table 9 below: Table 9 Reallocation of Schedule 1 of the DCA US$ '000 Original Reallocated Allocations Amounts I. Civil Works for Irrigation and associated engineering 1,550 1,330 II. Other Civil Works 1,750 1,950 III. Vehicles 1,250 1,500 IV. Equipment and building materials 870 1,820 V. Consultants for LDA management 800 1,120 VI. Training (a) LDA Staff 180 180 (b) Government Staff 90 90 VII. Feasibility Studies for future projects 210 210 VIII. LDA operating costs 1,300 1,300 IX. Unallocated 2,000 500 10,000 10,000 1/ LDA decided to construct the IFFF on force account rather than by using contractors and hence found it necessary to replace certain key equipment items. - 51 - 6.21 LDA made two other requests for alterations of the categories of the DCA in 1979 and 1980 which were intended to allow LDA easier access to project funds. However, these were not considered necessary by IDA and were not approved. 6.22 The use of Procedure 3 and excessive delays within LDA led to the necessity for a procurement agent and in early 1980 a Nairobi-based agent was appointed. Communications and administrative delays were such that severe difficulties were faced by the procurement agent in obtaining project staff assistance in order to allow him to expedite procurement. Under the circumstances his effectiveness was limited but he made a significant contribution nevertheless. Diversion of Project Inputs 6.23 In 1978, 17 vehicles and five radio telephones, together with a quantity of camping equipment were requisitioned by the army for use in the Ethiopian-Somali war. Records surrounding this incident are vague but the Project Manager reported to the PCR mission that the project had been fully compensated. He indicated that 11 vehicles w-Thich survived the conflict were returned to the project ana tnat compensation was eventually received for the other six vehicles, the radio telephones and the camping equipment. 6.24 Apart from this major example of misapplication,there were several others of a more minor but nevertheless serious nature. Misuse of vehicles and diversion of fuel and spare parts was known to occur. Routine checking of Statements of Expenditures by supervision missions also revealed diversion of project funds for purchase of household furniture and carpets which were not used on the project. Misapplication was encouraged by lack of proper accounting and weak overall cost and inventory control. It was on these grounds, and because of the observed misuse of project funds and inputs, that IDA took a rigid stand on the question of reimbursement of Statements of Expenditures and refused to reimburse unless satisfactory justification was provided that such expenditures were in fact incurred for project inputs which were actually used on the project. - 52 - VII INSTITUTIONAL IMPACT LDA, KMF and the Project 7.01 Institutional development through improved performance and profitability of both LDA and KMF was a major indirect project objective. 7.02 At identification/preparation of the project, it was recognised that the past performance of both institutions was unsatisfactory and a special effort was made in staffing the appraisal mission to ensure that the areas of organization, management, financial structure, staffing and training were adequately treated. The appraisal mission also recoanised that LDA, in particular, lacked management skills, trained manpower, planning capability and financial discipline but concluded that it could implement the project if strengthened with a special implementation unit (the PDTU) and if key specialist posts were filled with qualified and experienced expatriates. At least six fulltime specialists were envisaged (para 4.01 and Annex 2, para 3) supported by short-term consultancies as needed (para 7.07). It was also recognised at appraisal that LDA's financial base was weak. An analysis of 1971 and 1972 accounts, for example, revealed that over 80 percent of LDA revenues was generated from non-productive and non-trading activities (e.g. port fees). Thus, it was planned at appraisal that the Somali Development Bank would provide overdraft facilities to LDA for the project's trading activities and that the project would provide the funds for LDA's project-related operating costs for the first three years of operation. Furthermore, at appraisal and negotiations Government gave assurances that the local staffing shortfalls in the marketing, production, planning, architectural, engineering and construction areas would be met by priority hiring of local staff or by employment of consultants. The weaknesses in LDA's financial control, recording and reporting were to be strengthened by similar measures. Government's obligations with respect to these and other aspects of institutional strengthening were detailed in the Credit Agreement and Subsidiary Loan Agreement. 7.03 Despite the fact that the weaknesses of LDA were recognised at each step of the project cycle prior to Effectiveness, and notwithstanding that extensive effort was made to detail them and to build into the project design the necessary measures to overcome them, LDA was unable at any stage to operate at a level of competency adequate to ensure efficient project implementation. In the course of the project LDA was headed by two General Managers -- the first was in post until July 1976 and the second until LDA was dissolved in March 1981. Unfortunately, neither brought to the post the qualifications, experience and commitment needed to develop the institution as envisaged at appraisal. The tenures of both GM's were characterised by inadequate, inconsistent and arbitrary management styles which, coupled with shortages of finance, weak financial control, financial wastage and fund misapplication, precluded development of LDA as an institution capable of serving the project's demands and of spearheading livestock subsector development in Somalia. 7.04 The General Manager in office in the later project stages enjoyed a high degree of independence from Ministerial control. Under - 53 - his stewardship, the management consultants lacked support and encourage- ment. Many legal covenants were ignored and Government took no steps, which were well within its power, to address the situation, notwithstanding constant Bank reminders. LDA's failure as an institution started at the top and it became so pervasive at all levels that Government in March 1981 finally dissolved the Agency. After years of inaction, the extreme measures ultimately taken by Government served eloquently to underscore the failure of LDA. Effectively, IDA and the project were so interdependent that, when LDA failed as an institution, the project failed with it. The way things ended, any evaluation of the project's institutional impact on LDA was irrelevant. 7.05 The nroiect also had no beneficial effect in developing KMF. For reasons outside of the project's control, the mutually supportive roles that IDA and KMF were supposed to play in each other's growth and development never materialised. The rivalry which existed between them was personally and politically inspired and it was undermining to both institutions. Responsibility for cattle trading and processing in the Trans Juba area was assigned and reassigned from one to another in the late 1970's in a manner bordering on the whimsical. LDA's attempts to expand live cattle sales to the Middle East was also an undermining influence on KMF, or at least perceived to be so by KMF management. It was all self defeating, in the extreme. Because of poor operating and financial performance, KMF was closed by Government in December 1980, three months prior to LDA's dissolution. The project thus witnessed in its lifetime the demise of the two key institutions in livestock marketing/processing in Trans Juba which it was designed and set up to promote and develop. The project's inability to expand and develop either institution was a significant failure of the project and indeed also of the Government's strategy for livestock development in Somalia as a whole. The Management Consultants' Role 7.06 The Appraisal Mission proposed that LDA's shortcomings as an institution could be rectified by several measures; among them was the appointment of a consultant firm to supply staff for key specialist posts, including project management. Thus, following Effectiveness, a first major task was to select a firm. Seven prospective bidders were short- listed by Government in mid-1974. However, negotiations with the firms were protracted (para 3.17) and it was not until October 1974 that a contract was finally awarded. 7.07 The first of the consultant staff, the Project Director, arrived in post in November 1974. In the ensuing months the other posts (para 4.01) were filled and three subcontractors to undertake topographic mapping, soil survey and IFFF irrigation design were also appointed. In the period through to mid-1976, detailed planning for the stockroutes, the holding ground and the ranches was carried out and designs, specifications and bills of quantity drawn up for construction of project buildings. These efforts and associated tendering procedures were to be the major contribution of the consultants because in the first half of 1976 the team itself split into factions and serious differences between the team and LDA management emerged over the details of borehole development and the employment of a civil engineer to oversee project construction activities. - 54 - 7.08 Matters drifted until August 1976 when there was a change of Project Directors; this was followed by resignation of the Senior Grazing Ranch Manager the next month. The departure of the two staff badly under- mined work continuity and this, combined with the inflexibility and lack of tact of the new Director, widened the gulf between the project and LDA management. Alienation of staff and low morale resulted and there were complicating problems between project management, other Government departments, and local authorities. Matters deteriorated to the point where, in Fo.briprv 1977, the consultants were terminated by LDA without replacement. 7.09 The PCR Mission's conclusion was that the consultants' performance was unsatisfactory, notwithstanding the difficulties they endured at the hands of LDA management (paras 4.04 - 4.06). In fairness, it should also be noted that the consultants worked diligently to set up a suitable management structure to serve subsector development over the long term. They also set up the project accounts and a stores control system all of which, unfortunately, fell apart after they left. 7.10 The Review Mission took particular issue with the consultants on the question of the project cost recalculations which were instrumental in prompting the review (para 4.09). The Review Mission found the consultants' cost estimates exaggerated because of -Additional items not included in the Project..........; *Faulty technical design.... *Overlavish design of various items...... etc. The Mission's overall conclusion on the consultants was summed up thus: 1/ "Government's worries and doubts over the calibre of management were undoubtedly strengthened by the mission's report of its preliminary findings; the result was that Government decided to give the Consultant the contractual ninety days notice of the termination of its services. Although the mission at first advised caution in taking such a serious step, Government's lack of confidence in the Consultant's performance on this project, and the more recent confrontations with management, made it almost impossible for the existing Consultant to continue. Despite further delays, a fresh start seemed essential to give the project a sound basis for successful implementation. The mission thus concurred with Government's action. 7.11 All in all, the story of the consultants' role in the project was an unhappy one. Their technical assignment was difficult, and the project's work climate, mainly because of the deficiencies of upper level LDA management, was not a conducive one. They were not able in the circumstances to make more than a token contribution and their impact was minimal. -Whether another firm with better qualifications and experience to tackle the stockroute, ranch and IFFF components could have Aone better, is conjectural at this stage. The PCR Mission is of the view that much more could have been achieved if the consultants had been chosen from a country with a stronger background in extensive ranchina/cattle oroduction. One thing is certain -- the project did not look up after the consultants' departure so that, no matter what their deficiencies were, no one was able to do any better in furthering the project after they left. 1/ Review Mission report page 12, para (e) - 55 - VIII BANK PERFORMANCE 8.01 The Bank's involvement in the TJLP at identification, preparation, appraisal and review was extensive. The project was the Bank's first agricultural project in Somalia and, as the country lacked the technical personnel and experience to undertake detailed project formulation, the Bank provided the assistance required. 8.02 The project as prepared and appraisal fully reflected Government priorities in the agricultural sector and its desire to expand livestock productivity. However, while the project concept was sound, and while the project area containing the country's largest livestock resource was well chosen, the project as designed was too large and complex for the existing managerial, technical, manpower and financial resources of the country. The latter constraints were recognised at preparation and appraisal and measures were detailed to rectify them but subsequent ability and willingness of Government to take the needed steps was seriously over-rated by the appraisal team. As a consequence, those responsible for managing the project were faced with insurmountable constraints and Bank/IDA performance throughout the project was inex- tricably linked with this counterproductive situation. 8.03 One can understand the Appraisal Mission's desire to move forward quickly in such an undeveloped subsector in which the potential appeared so great, particularly since the project was the Bank's first intervention in agriculture and a great deal of effort had been expended in identification/prepration and since both Government and LDA had given every indication that their strong support would be forthcoming. No hindsight, however, was needed to perceive the serious management prob- lems of attempting to develop, from a standing start, two 80,000 ha ranches, a 25,000 head per annum feedlot and a 1,200 ha irrigated fodder farm, and of attempting to coordinate these new developments successfully with LDA's existing and proposed marketing endeavours and with other pro- posed project activities. 8.04 At appraisal, there was little detailed information on production and marketing practices in the project area and virtually nothing was known of range fodder species and potential range product- ivity. Again, in Somalia, irrigated fodder production was untried and feedlot fattening was just a concept. There was such a paucity of data and local experience on which to plan development interventions that greater prudence should have been exercised. It was unfortunate that these components were not limited to research or pilot efforts initially and that the project's central development thrust was not limited to stockroute and holding ground development. If the size of these risky components had been made smaller at appraisal, there would still have remained a viable project and one which would have been much more manage- able and within the Government's resources. If that course had been followed, it is quite likely that there could have been a well established base today from which to develop a larger second phase. As it is, with a costly failure before us, the future course is uncertain. There is an important lesson to be drawn from this in the PCR Mission's view. 8.05 The performance of the project review mission in January/ February 1977 was unsatisfactory given that: (i) four years had elapsed since appraisal, during which the project had problems sufficiently serious to warrant a full-scale review; (ii) the nature of the problems was well documented by suc- cessive supervision missions, by the project itself and indeed by Government. 1/ The problems included: the size, design and complexity of the project, inadequate local financing and staffing, lack of com- mitment by upper level LDA management, serious cost overruns, LDA/KMF rivalry, significant declines in international beef prices, concern over KMF's canned meat market, serious differences between the management consultants and LDA/Government, inadequate performance of the management consultants, evidence of project design weaknesses and highly questionable viability of certain project components (e.g., ranching, IFFF); and (iii) there was clear evidence of LDA's inability to act as project executing agency and strong indications that it would not take adequate action to improve the situation--likewise, Government itself appeared unwilling to move decisively. 8.06 The Review Mission should therefore have proceeded more cautiously than it did, particularly as a pre-review visit by a-mission staff was made to Somalia in November 1976 to arrange the details of the proposed review. In proceeding as it did, the Review Mission diverged seriously from earlier supervision recommendations, particularly with respect to IFFF development. 8.07 The technical and financial judgements of the Appraisal Mission as well as those of the Review Mission were overly optimistic. The Appraisal Mission justified the financial/technical viability of the feed- lot operation on: (i) a grade premium varying from 53 to 83 percent (mean 1/ The State Planning Ministry had examined the project in September 1976 and it concluded that the IFFF was a risky venture and suggested three alternatives to the appraisal approach all of which involved significant scaling down of the component. - 57 - 57/58 percent) between the buying and selling price of cattle; (ii) a daily liveweight gain of 0.93 kg/head; and (iii) a throughput of 25,000 head per year. 8.08 These coefficients were quite unrealistic for a pioneering feedlot effort in Somalia; they would be enviable achievements for a second or third generation family in the feedlot business in a country with an advanced livestock industry with a developed infrastructure and with inputs and markets at hand. In the PCR Mission's view, a premium for grade, quality or finish as opposed to a payment for meatiness, age or size) has never been clearly demonstrated in the Somali market. Furthermore, all cattle transactions have traditionally been on an arbitrary per head basis which was so imprecise and subject to personal judgement as to undermined sophisticated analysis on which the Appraisal Mission based its case. The appraisal case was weak and it continued to be so at the time of the review. The appraisal coefficients were revised slightly downwards by the Review Mission and the throughput was cut from 25,000 to 13,800 head per annum, but the grade premium (40 percent and the projected daily lieveweight gain (0.75 kg per head) remained as the key determinants of IFFF viability. As at appraisal, the real sit,ation was misjudged by the Review Mission and this resulted in unduly optimistic conclusions concerning the financial/ technical feasibility of the IFFF. 8.09 Similar optimism attended the review of the ranching component; however, the development of one 80,000 ha unit was recommended to be post- poned by two years, while the other was to be expedited on a reduced scale of 50,000 ha initially. This approach was an improvement on that at apprasial but the challenge was still clearly too great for the project, given its past performance. Besides, development was recommended to proceed in areas which were bush covered (which was directly contrary to appraisal expectations) and which were known to be used by nomads and inhabited with game, including elephants and predators. In noting these facts, and . in observing that communications in the area were poor and that ranch development should be more gradual, the review team did not mention the risks involved. This was a serious oversight, since resistance from nomads and the complicating presence of elephants were key considerations in IDA's decision in 1980 to drop the ranching component altogether. 8.10 The PCR Mission is of the view that the Review Mission's conclusions on the future course of the project were, overall, less than satisfactory, particularly given the hindsight available to the review team and given circumstances in the project at the time. In the circumstances, there was in- sufficient justification for maintaining the size, design and complexity of the project so in line with appraisal recommendations. In the PCR Mission's view, there was at no time a real possibility that the project could develop on the scale recommended at appraisal or review. - 58 - 8.11 Finally, with respect to overall Bank input, the project consumed significant manpower resources witlh one pre-Effectiveness mission, 15 sunervision missions, two review oreparation missions, a full review together with considerable assistance from RMEA's Disbursement Section. The volume of communication was also heavy. Furthermore, the TJLP was a focus at every PIR, CIR and PPR in recent years because from 1976 the project was consistently rated 3. Supervision efforts, which were undertaken by eight different Bank staff, were satisfactory both with respect to continuity and quality. There were some lapses, but not on the important issues, which were consistently satisfactorily analysed and highlighted. The lapses were ones more of omission than commission. If supervision was to be faulted, it would be on the grounds that the missions in some cases may have been too conciliatory in their attitude, when a firmer approach may have been more warranted. This is conjectural, of course, as it has to be remembered that this was the Bank's first agricultural project in Somalia. - 59 - IX CONCLUSIONS 9.01 The story of the TJLP is not a successful one because the project did not meet any of its stated objectives and its economic rate of return was negative. There was no expansion of the managerial expertise needed to support sustained growth of the all-important livestock subsector in Somalia and, in fact, only limited training of any description took place. 9.02 In terms of its objectives, the project achieved no institutional impact because the LDA and KMF, which were the two main institutions to be developed and promoted under the project to provide a strong institutional base for longer-term subsector development efforts, were both disbanded in the project's course. No effective substitutes had been established for them at the time of preparation of the PCR Report and Government had not developed an alternative livestock marketing strategy. The future develop- ment of the livestock marketing currently represents a major challenge to Government as well as to donors. Government has approached the Bank for assistance in preparing a future marketing project. It has also indicated that it feels that more private sector involvement should be encouraged in the livestock marketing subsector. Based on the weak financial per- formance of not only LDA and KMF, but of similar public sector entities in Somalia and elsewhere, the PCR Mission feels that Government thinking in this direction should be encouraged. 9.03 The stockroutes, holding grounds and the veterinary dispensaries were the only useful infrastructural components developed by the project. They have produced no significant benefits so far but could do so quickly if Government moves expeditiously to equip the boreholes and complete the stockroute watering system. This hopefully will be accorded priority as the effort to complete the work is marginal in comparison with what has gone before and significant benefits would accrue directly and promptly to the nomadic sector. 9.04 There are lessons to be learned from the project, the most important of which are: (i) size and complexity are a dangerous mix in a first phase project in an undeveloped subsector in a difficult project environment; (ii) in the absence of strong Government commitment, there is no project; (iii) expatriate consultants need to be carefully chosen -- technical staff from some countries have quite special aptitudes which are not universally shared and, when projects require special expertise, it should be sought from those countries which can most appropriately supply it. - 60- ANNEX 1 Page ' 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Detailed Features of Project Design Introduction 1. The Trans-Juba Livestock Project was planned to provide the infra- structure and market outlets to facilitate marketing of cattle from a 65,000 km2 project area south of the Juba River in southern Somalia. The project was to integrate purchasing of cattle from traditional pastoralists with growing and fattening of immatures on ranches and in a feedlot. Mature cattle obtained from pastoralists and fattened cattle from the ranch and feedlot system would provide a regular supply of canning and carcass beef for the Kismayu Meat Factory (KMF). At the time the project was prepared, Somali authorities estimated that 30,000 to 40,000 immature cattle from the Trans Juba area were being sold through informal channels to Kenya. The project was expected to provide a more attractive market for these animals. Assistance and incentives to the traditional sector were also to be increased under the project by improving animal health services and by the development of markets and marketing infrastructure all of which were aimed at more effective range utilization, increased animal productivity and greater marketed offtake. 2. Appraisal strategy had three major components: (i) improvement of markets and stockroutes to facilitate purchase and movement of cattle and to reduce weight losses and mortality during trekking; (ii) development of grazing ranches to allow further condition- ing of those animals not suitable for immediate slaughter and to produce feeder grade stock; and (iii) development of an irrigated fodder farm and feedlot (IFFF) to fatten cattle suitable for KMF, which would in turn sell its output as frozen or chilled beef on the world market. 3. The Droject was designed both to increase the VMF'- nitnit nf canted meat as well as to initiate sale by KMF of frozen and chilled beef. The latter was expected to show a higher value added by allowing entry into higher quality markets, thereby increasing returns to the economy, incomes of the nomads, and the profitability of both the Livestock Development Agency (LDA) and the KMF. The broad design of each project component to achieve these objectives is discussed below. Detailed planning and design of the components was to be completed by consultants as the first stage of project implementation. - 61 - ANNEX 1 Page 2 Cattle markets, stockroutes and holding grounds 4. The project was to establish organized marketing outlets for livestock produced by traditional pastoralists in order to reduce liveweight losses and mortality during trekking and to increase overall offtake rate. The improved infrastructure would facilitate the orderly purchase, movement and holding of cattle by LDA and allow extension of the necessary veterinary supervision to a much larger number of cattle. The marketing component included: (i) development of three new permanent markets at Tabda, Maskati and Gelib in project years 1, 2 and 3; (ii) establishment of two smaller markets at Lugh Farrandi and Busar in project years 2 and 3; (iii) clearing of over 1,000 km of existing and new stockroutes to connect holding grounds and markets, and the establishment of staging posts at 30 km intervals with water supplies (from dams or boreholes), over project years 1, 2, 3 and 4; (iv) development of one new 20,000 ha holding ground at Afmadu with perimeter fence, internal firebreaks, roads, three boreholes, one cattle dip, three sets of yards, housing and equipment; (v) expansion and redevelopment of the existing Kismayu holding ground by clearing of bush and improvement of the access roads and water supply; and (vi) establishment of three cattle purchasing teams to operate holding grounds and markets. 5. The markets were to include sale ring, office, pens, races, crush, housing, water supply facilities, shelters for traders, weighing equipment and generators for domestic electricity and operation of water pumps. Grazing Ranches 6. The original project design was to develop two 80,000 ha ranches with a capacity to grow to maturity some 10,000 cattle. The ranches were located in the Marine Plain area (Map 2) and were to provide cattle for the feedlot and afford a market outlet for immatLre cattle from pastoralists. No breeding was planned on the ranches, only growing ("backgrounding"),mainly to prepare cattle for feedlot finishing. The main developments on each ranch included: (i) construction of boundary fencing, tracks, access roads and firebreaks to subdivide the area into approximately 30 sub-areas; - 62 - ANNEX 1 Page 3 (ii) development of boreholespstorage tanks and dams; (iii) construction of houses, offices, store and hay shed, and provision for electricity supply; and (iv) procurement of vehicles and equipment as required. 7. Ranch I was to be developed in project years 1, 2 and 3 and Ranch II in years 2, 3 and 4. Irrigated fodder farm and feedlot (IFFF) 8. The IFFF was to provide for a throughput of 25,000 cattle per year. Animals were to be trekked from the ranches (20,000 each year) and from the Afmadu holding ground (5,000 each year) after purchase from pastoralists. Cattle would be selected from pastoralists' herds by marketing teams; the majority of animals purchased would be going directly to the KMF for production of canned beef for export. The main features of the IFFF included: (i) construction of a system for flood irrigation of 1,200 ha of land for production of maize for silage, alfalfa and Rhodes grass (Chloris gayana), and construction of a main canal to deliver pumped water from the Juba River; (ii) construction of a feedlot for a throughput of 25,000 cattle per annum comprising pens, covered feed troughs, dips, and a veterinary crush; (iii) construction of housing, offices, stores, a workshop and a feed preparation area; (iv) construction of an access road and internal all-weather roadways; (v) drilling a borehole and equipping a reticulation system for domestic and livestock water supplies; (vi) provision of motor and generator for electricity supply for domestic and other use; and (vii) provision of vehicles, tractors and equipment for cultivation, irrigation, harvesting, feed mixing, feed preparation and feed delivery. 9. The IFFF was to be designed in four months and established over two years with construction by contractors employed by the LDA. The appraisal report assumed cattle weight gains in the feedlot of 70 kg in a 75-day feedinz cycle (0.93 kg Der dnv) and the achievement of full operating capacity in year 6 of the project. Irrigation was to be carried out in a 24-hour cycle for forage production. At appraisal, a tentative site for the IFFF was identified (the site of its subsequent development) but consultants were to undertake detailed soil surveys, water analyses and -63- ANNEX I Page 4 design of the irrigation system and, if necessary, select a more suitable site for the irrigation and feedlot component. Animal Health 10. The ?im of the animal health component was to expand veterinary care in the project area by providing tne pnysicai intrastructure to extend the reach and improve the effectiveness of existing preventive veterinary services. The component was to be established under the project and later transferred to the Veterinary Department of MOLFR. The main features of the component included: (i) construction of offices and laboratory for the Regional Veterinary Centre in Kismayu; (ii) construction of four district veterinary centres (with laboratories) at Afmadu, Gelib, Giamma and Brava; (iii) construction of five veterinary dispensaries at Bulo Haji, Goba, Anjail, Beles Gogani and Tabda; (iv) provision of vehicles, veterinary equipment, electricity and water supplies at each location; and (v) funds for the employment of staff and for operating costs of each facility for three years. 11. The animal health programme was also planned to create a disease-free zone in the Lower Juba area from which cattle could be supplied to KMF for the production of chilled beef, which would be permitted entry to high value European markets. Such markets were not open to beef exports from Somalia because of lack of adequate livestock disease control. Technical Services 12. The LDA was appointed as the executing agency for the project. Since it had never implemented a development project and did not have the institutional structure suitable for execution of the TJLP, the appraisal called for the formation of the Project Development and Training Unit (PDTU) within the LDA. The PDTU was to be staffed by expatriate and Somali staff and was to be responsible for implementing the project. 13. The technical services component included funds for the first three years of the project for the appointment of the management consultants to man six key positions in the PDTU and six professional Somali staff, in addition to funds for: (i) the establishment of the Project Construction and Management Unit (PCMU) within the PDTU in order to carry out construction work and supervise the work being under- taken by contractors; - 64 - ANNEX1 Page 5 (ii) the completion of designs of project components and preparation of detailed bills of quantities (the appraisal report recommended that the same firm of consultants be employed to provide the six expatriate staff and staff to prepare detailed designs); (iii) the appointment of contractors to drill and equip 26 boreholes (the appraisal report perceived this function would be carried out by the Water Development Authority (WDA)); (iv) the overseas training of four senior counterpart staff (at post-graduate and diploma levels) and of short-term study teams; (v) local training programmes for LDA and MOLF. staff; and (vi) feasibility studies for the preparation of future agricultural projects in Somalia (not necessarily related to the TJLP or the project area). Funds for LDA Operating Costs 14. The project would provide the funds for LDA's project-related operating costs for the first 3 years of the project. LDA returns on the marketing component were expected to provide sufficient funds to meet operating costs from the fourth year onwards. Project Review 15. The project was intensively reviewed in early 1977 and some changes were proposed by the Review Mission in the project as appraised. The substance of these changes is discussed in Annex 4. - 65 - ANNEX 2 Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Project Organisation and Management Introduction 1. At appraisal, the LDA was selected as the project executing agency. The appraisal mission recognised that LDA lacked management skills, manpower, planning capability and financial control but concluded that it could implement TJLP if strengthened with a special project implementation unit. The Project Development and Training Unit (PDTU) was thus created as a functional arm of the LDA responsible for implementation of the TJLP and of future livestock subsector development projects. Organisation 2. It was proposed at appraisal that the PDTU was to be headed by a Project Director directly responsible to the General Manager of LDA who was the Agency's senior executive. The PDTU was envisaged ultimately to become the Production Department of LDA at full project development. The Appraisal Mission further proposed the establishment within PDTU of a Project Construction and Maintenance Unit (PCMU) which would undertake the bulk of the project's construction work and would also provide supervision for project developments being undertaken by contractors. The organisational structure as agreed at appraisal is attached overleaf. Management 3. It was recognised at appraisal that the LDA did not have the technical staff capable of developing and managing the project. It was proposed, therefore, that expatriate specialists would be appointed to the six senior positions in the management structure, viz (i) Project Director (ii) Senior Grazing Ranch Manager; (iii) Irrigation Specialist; (iv) Irrigated Fodder Farm and Feedlot Manager; (v) Master Mechanic; and (vi) Project Finance Officer. 4. A consulting firm was to be appointed in order to supply the specialists. The specialists were to be employed on contract for three to four years with subsequent transfer of responsibilities to Somali counterparts. The overall arrangements did not work satisfactorily. Details of problems encountered are presented in paras 7.07-7.09. -66 - ANNEX 2 Page 2 SOMALIA Trans-Juba Livestock Project Livestock Development Agency ORGANIZATION AD MANAGEMEN CUART A. DURING DEVELOPMENT General anager Marketing Dept. Development Dept. Port Veterinary Service Finance & Adn. Dept. Project Development n n Training Unit P Kismayu Other Kismayu Other Kisja c Otr Regions PlannIn Pla ring aiPro- Con- & Co- duction sultancies I ordination SectioS Section Senior Irrigated Project Ha rjc niern Ranch Fodder Finance Cnt oslat Manager Farm and Ofie Mannnc Feedlot Ui Manager Grazing Ranche B.AT FULL DEVELOPMENT General Manager Marketing Dept. Development Dept. Port Veterinary Service Fnne&Am.Dp. Pouto et & rmtin Srvc isau OtherKKsa e SrieRegionsRgin s Kismayu Otheru 0h 's Coordnatio Section Secio Race Fodder Far a.nd Feedl,ot_ Sectionn - 67 - ANNEX 3 Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Background to Project Review The two documents comprising this annex have been reproduced from the files with only minor editing because of their importance in documenting first-hand the changing circumstances of the project in the period October 1974 (effectiveness) and mid-1976 by which time serious questions were being raised concerning the project's viability, in particular the key IFFF component. It was against the background described in these documents that the Project Review Mission was mounted in January/February, 1977. The Review Mission's findings are presented in Annex 4. - 68 - 'Fo,;i No. 57 W NATIONAL BANK FOR INTERNATiC '. DEVELOPMENT ANNEX 3 RECON., JCTION AND DEVELOPMENT I A S IATION Pagen 2 OFFICE MEMORANDUM TO DATE: May 21, 1976 FROM: SUBEJECT: SOMIALIA - Trans-Ouba Livestock Project - Credit 462-SO - Project Review - Full Back-to-Office Report 1, A visit was made to Somalia March 28-April 4, 1976, to determine if the project is in need of a review and to agree with the government officials concerned on timing and on the composition end duration of the mission should a review provo necessary. The first two days and last three days of my stay in Somalia wore spont in Mogadishu and the period in between in the Project Area around Kismayu. 2. The Credit was declared effective on Octobar 3, 1974 and the Project Manager arrived in Somalia in January 1975. He and his team, which joined him later in the year, have spent their time sinB then preparing for project implementation. This included recruitment of Somali staff, detailed planning of construction work and preparation of tender documents for the purchase of- equipment, construction of buildings and the drilling of boreholes. Virtually all this pre- paratory work has now been completed and the Project is on the thres- hold of implementation wnich is expected to start in July 1976 at the end of the rainy season. 3. The Project essentially aims at increasing the productivity of the traditional herds in the Trans-Juba Region through the establish- ment of cattle markets, stock routes, holding grounds, growing out . ranches, irrigated fodder farms and feedlots and veterinary clinics and laboratories. The Livestock Marketing Agency (LDA) is charged with the responsibility of developing and operating this marketing infrastructure and is expected to channel cull and fattened cattle through the already existing.Kismayu Meat Factory (KMF) for processing and export thereby generating badly needed foreign exchange. 4. A number of important developments have taken place since appraisal in January/February 1973 and these arE now threatening the viability of the Project. The most important ara the following three developments:_ ANNEX 3 -69- Page 3 - 2 - May 21, 1976 (i) increase in Project costs estimated by the consultants at about 80% and mainly associated with the irrigated fodder farm and feedlot; (ii) decrease in the world Trarket prices of beef by about 50% between early 1973 and early 1976 and uncertainly about prospects of their recovery; and iii) cancellation of LOAIs cattle purchasing contract with KMF by the Supreme Revolutionary Council because the latter company could not afford to buy cattle from the former at the ongoing price of Sogs 2.31/kg live weight and make money by selling beef on the international market. As a result LDA is now attempting to cultivate marketd for live cattle abroad in contradiction to the Project. 5. Under the Project cattle were to be bought by LDA at a price of Soh 1.44/kg live weight and sold to KMF after fattening for Scs 2.20/kg live- weight i.e. an increase in price of about 53/g live weight. The producer price of cattle has however increased in Somalia to about So9 1.90/kg live- weight since appraisal which means that LOA would have to sell these animals to KMF at a far higher price that Sob 2.20/kg live weight to realize a profit on its fattening operations, but under these circumstances KMF would be unable to market the beef abroad. The lack of a beef grading system and the absence of premium prices for quality meat in Somalia further aggravate the problem. 6. The P-oject has therefore undergone two major changes since appraisal three years ago, namely: (a) a dramatic increase in capital and operating costs; and (b) a marked decrease in the spread between the purchasing cost of cattle in Somalia and the selling price of beef abroad which has resulted in the loss of LDA's principal, if not only, buyer, Because of these changes the project is in need of an indepth review at thi stagc when most of thu pianning and design worK has been completed. Govern- ment will have to make a number of pol:cy decisions to keep the Project going such as defining LDA s role in livestock marketing, providing additional finaic) for the Project, increasing meat prices and introducing systems for grading beef before Project implementation can start. The primary objective of all these exercises would be to render the Project financially and economically viable. -70 - ANNEX 3 Page 4 - 3 - May 21, 1976 7. 'The above matters were brought to the attention of the Secretary of State for Finance and the General Manager of LDA and both agreed with the need for an carly indepth,roview of the Project. They further agreed that the review should start on or about May 17, 1976 to avoid the rains and to provide the Project Manager with sufficient time to prepare for the review mission. Also that the mission should consist of a livestock economist, a livestock specialist in ranching and feedlots, an agricultural engineer and a financial analyst. B. The next step would be to telex these agreements to Headquarters to find out if such a mission can be mounted at the time suggested. Somali authorities and the consultants should then be adviaed accordingly. 9. I estimate that the mission would require about three to four weeks in the field and a similar amount of time in report writing. File: Cr. 462-SO -. ANNEX 3 - 71 - Page 5 Attachment to BTO Report of May 21, 197 TRANS-JUBA LIVESTOCK PROJECT COST COKPARISON SUMHARY SHEET APPRAISAL COST ESTIMATE CONSULTANTS' REVISED ESTIlATED COST INCRZASE Appraisal costs COST ESTIMAT-2 ABOVE JAN. 1976 LIWi:L COMPONENT escalated by 60% Assumed Jan. 1976 Appraisal Jan. 1973 costs costs 1/ Dec. 1975 level Jan. 1976 level Jan. 1973 level - - - - - - - - - - - So. Sh. '000 rounded - - - - - - - - - - % Markets & Stockroutes 6,800 10,900 11,900 49 475 Grazing Ranches 7,900 12,700 11,300 -11 443 Irrigation 10,800 17,300 32,000 +87 *196 Irrigation Fodder Farm 6 Feedlot 14,700 23,500 25,200 47 *71 Construction & Main- tenance Unit 5.000 8,000 8.100 41 -462 Development & Train- ing Unit 2,800 4,500 3,900 -. -15 439 Consultants Fees 5,400 8.60Q. 5,400 -37 - LDA OPERATIONS, total 53,500 85,600. 97,700 414 +83 Disease Control 1,600 2,500 2,200 -11 +38 Non project related training 800 1,300 900 -31 413 Feasibility Studies 2,000 3,200 2,200 -31 +10 GOV. OPERATIONS, total 4,400 7,000 5,300 -24 +20 LDA AND GOVERNMENT OPERATION 57,900 92,600 103,000 +11 4-78 CONTINGENCIES Physical 5,800 9,300 10,3005. +11 +78 Price escalation 8,100 19,000 21,6005/ +14 4167 TOTAL PROJECT COST 71,800 120,900 134,900 +12 488 1/ Estimated cost increase of 60% caused by post price escalation according to Bank Guidelines between January 1973 and January 1976. 2/ Source: unofficially communicated by consultants March 15, 1976 - 72 - ANNEX 3 Page 6 SOMALIA TRANS-JUBA LIVESTOCK PROJECT, CREDIT 462-SO Project Review Preparation Discussions in Nairobi between Project and RMEA Staff, July 1-8, 1976. Aide Memoirs 1. The Project Director and the Senior Grazing Ranch Manager, visited Nairobi between July 1 and 8, 1976, and hand carried with them the latest cost estimates on the Project and the plans and tender documents for the various components of the Project. 2. The consultants now estimate. total Project cost at about US$ 24.4 million compared with the appraisal forecast of around US$ 11.2 million. They also show that the increase in total Project cost of US$ 13.2 million is roughly equivalent to the newly estimated cost of the irrigated fodder farm and feedlot, Their estimates are considerably higher than those made by RMEA/Local Staff two months earlier and could in part be attributed to the inclusion of provisions for depreciation of-plant and equipment not pro- vided for by appraisal and to the procurement of a much larger inventory of spare parts than was deemed necessary by appraisal. The consultants estimates of total project cost and of the costs of the irrigated, fodder farm and feed- lot are on the higb side and can be scaled down considerably possibly to about US$ 22 million and US$ 11.00 million respectively. 3. The conclusions of these Nairobi discussions were that (a) the fattening of cattle on irrigated forage crops in Somalia under existing cost/price relationships is not viable; (b) the marketing of Somali beef in the Arab world and the oil producing countries of the Persian Gulf in competition with Australian and New Zealand products will be more difficult than we previously thought; (c) the livestock marketing components of the P2oject are still viable and are now estimated by the consultants to cost about USS 11.2 million; and (d) the Credit would be sufficient to implement the livestock marketing components of the Prqject with only minor alterations. 73 Annex 3 Page 27 4. nder the above circumstances it was proposed that :- . The Irrigated Fodder Farm and Feedlot "(a) the irrigated fodder farm and feedlot should be dropped from the Project; (b) the irrigation component of the project should, if found viable, be implemented as a separate discrete project devoted mainly to cash crop production with a small pilot scheme included to investigate the technology and economics of forage crop production and cattle fattening on a limited scale; (c) the consultants should, on their return to Somalia later on this week, investigate the economics and viability of crop production under irrigation fdr consideration by the Bank and the Somali Government; and (d) the alternatives of securing funds to finance a separate irrigation project for-cash crop production should be explored. II. Livestock Marketing (a) the Credit should be used exclusively to finance the livestock marketing components of the Project; and (b) the development of livestock marketing infrastructure should commence without delay and confirmation of this will be transmitted to the Somai authorities after coRsultation with IDA in Washington. 5. RMEA recommends the adoption of the above proposals and has forwarded their views to Washington. ANNEX 4 - 74 - Pg Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Findings of the Project Review Mission Introduction 1. This annex summarises the main findings of the Review Mission of January/February, 1977 which was mounted in response to a report by the consultants predicting cost overruns of over 100 percent if the project was to be implemented as appraised. The mission TOR particularly stressed the need for a detailed review of the IFFF component,the cost increases of which were the highest of all. An IDA supervision mission in March/April 1976 had also drawn attention to high cost increases in the IFFF component and to the steep decline in the international prices of beef "by about 50% between early 1973 and early 1976" and, on the basis of these observations,sought agreement from the Somali Secretary of State for Finance and the LDA General Manager for an early in-depth review of the project, which was tentatively agreed for May 1976 (Annex 3). The review team was not actually fielded until January 1977 and in the meantime, in July 1976, the IFFF came under further scrutiny during discussions in Nairobi between the Project Director, the Senior Grazing Ranch Manager and RMEA staff,from which it was concluded that the fattening of cattle on irrigated forage crops in Somalia under existing cost/price relationships was not viable and that, under the circumstances, the irrigated fodder farm and feedlot should be dropped from the project (Annex 3). The Review 2. It was against this background that the Review Mission was mounted in early 1977. The team of five specialists was in the field from January 20 to February 10 and the mission report was circulated in May 1977, the main conclusions of which were: (a) Project Scope The mission considered that there were adequate funds to implement in the remaining project period (1977-80) the following: (i) the marketing component in its entirety comprising markets, stockroutes, water supplies and holding grounds, with minor modifications ; (ii) the two grazing ranches with some modifications in timing; (iii) the feedlot and fodder farm with the irrigated area reduced to 190 ha, sufficient for fqtt-ning 13,800 cattle per annum for live export rather than for slaughter by KMF and export as chilled or frozen beef. ANNEX 4 Page 2 (iv) the disease control componenL essentially as appraised; and (v) the technical services component as appraised. (b) Project_PlanniaE_and_ Dsign The mission expressed concern at the lengthy planning period required by the consultants - from early 1975 to late 1976- particularly for the marketing component of the project. The mission was further unfavourably impressed by the performance of consultants on the grounds that: (i) plans did not involve attempts at cost reductions; 1/ (ii) plans were technically faulty;-- (iii) the designs, plans and costings were not provided in sufficient detail; and (iv) reports by the outgoing Ranch Manager were available only in the German language. (c) Project Components MarketinZ. The mission was critical of the planning and implementation delays in developing the marketing component and recommended that it be expedited as planned at appraisal. The mission also recommended the: (i) prompt awarding of the contract for construction of boreholes; (ii) realignment of the main stockroute to join Fafadun and Afmadu rather than to run alongside the Juba River; (iii) development of one additional borehole on the Kismayu holding ground; (iv) employment of an engineer to head the PCMU to modify the consultants' plans (which were too elaborate) and prepare work plans for construction; (v) prompt award of the tender for supply of heavy equipment; (vi) development of closer links with regional authorities for pooling equipment, funds and expertise for stock- route development; (vii) expeditious award of the contract for development of the Afmadu holding ground and Ranch II; (viii) recruitment of an expatriate Senior Grazing Ranch Manager; (ix) review of the need for three markets at Fafadun, Shappi and Soia (as recommended by consultants); and (x) advisability of developing temporary trial markets for at least one year before permanent construction. 1/ For example, design of feedlot yards, proposed changes of stockroute alignment from Fafadun - Afmadu to a route along the Juba River, siting of Ranch II buildings, siting of Afmadu holding ground. ANNEX 4 - 76 - Page 3 (d) Grazin Ranches. The mission recommended that one ranch only be developed initially and that the second should not commence until 1979 in order to allow resources to be concentrated on development of the first ranch and also to provide project management with experience. The mission recommended that the first ranch (Ranch II) be implemented as planned at appraisal but with initial development of 50,000 ha rather than 80,000 ha. (e) Irrigated Fodder Farm and Feedlot (IFFF). The mission was concerned that the available information suggested that the soils on some two-thirds of the area would present technical difficulties for irrigation and cultivation and that forage yields projected at appraisal might not be achieved. 1/ The substantial conclusion of the Review Mission was that the initial engineering design for the site was poorly prepared by the sub-consultant with a number of ele&ntary design faults." An earlier suggestion of growing cash crops on part of the IFFF was discarded by the Review Mission because: (i) most of the IFFF could not be viably developed and operated; (ii) it was considered inadvisable for the LDA to become involved in crop production; and (iii) the proposed feedlot as revised by the mission was a sounder economic proposition. Commenting specifically on the feedlot, which the mission regarded as a viable undertaking, notwithstanding the serious reservations voiced by the March-April 1976 supervision mission and by the consultants during their July 1976 visit to RMEA (Annex 3), the mission concluded that the feedlot would be viable under a number of alternative throughputs and weight-gain combi- nations, which could be varied according to prices and types of animals purchased. The most viable combinations included fattening 13,800 head from 200 kg to 250 kg over 67 days or 7,800 head from 200 kg to 285 kg over 110 days on feed produced on 190 ha of irrigated land which was selected for initial development. The revised system selected by the mission had the following features: (i) production of maize for silage and alfalfa for green feeding using staggered plantings which required less machinery and land; and (ii) fattening of 13,800 head on 190 ha with potential to expand to 300 ha under irrigation. 2/ 1/ Black clay soils (impermeable, subject to waterlogging and salt accumu- lation) accounted for 34 of the 72 fields in the design layout of the IFFF. Micro5elief data indicated high land levelling costs on 21 fields (1000-4000 m of soil per ha to be moved in each field); it was also thought that such levelling would disrupt topsoil profiles with adverse effects on yields. Further, only some 300 ha of the 1,200 ha of the IFFF was considered suitable for flood irrigation. 2/ The PCR Mission felt that the basic assum;tions of the Review Mission which led them to conclude that the feed requirements of the 13,800 head could be produced from 190 ha were far too optimistic. ANNEX 4 - 77 - Page 4 (iIi) revision of machinery requirements -- past orders for some units were considered excessive and some essential machinery had not earlier been provided for; (iv) revision of plans and tender documents; (v) provision of an irrigation specialist for six months to supervise construction, plan the irrigation system and train staff; (vi) management by an expatriate feedlot specialist was recommended; and (vii) provision of a master mechanic was recommended for twelve months. (f) Disease control. The mission was concerned that the consultants had not been in communication with Regional Veterinary Office, which had consequently not taken part in the design of the veterinary component. LDA and the mission agreed that the planned disease-free zone of the Lower Juba Region would not be achieved in the life span of the TJLP. The mission recommen- ded to strengthen the Regional Veterinary Service with vehicles, equipment and vaccine supplies and to substitute Badada for Brava as the site of the fourth district veterinary centre The mission proposed that siting of the veterinary dispensaries be reviewed and that they be changed if more suitable locations were indicated. (g) Technical services. The Review Mission recommended that the project's technical services component be retained essentially as planned at appraisal. 3. The main changes in the project costs estimated by the mission compared with those estimated at appraisal in January 1974 and as revised by the consultants in July 1976 are detailed in the attached table. The mission recognized that cost increases had occurred since appraisal but not at the level estimated by the consultants, with whom the mission disagreed on several counts with respect to the costings. - 78 - ANNEX 4 Table 1 S OMAL IA Page 1 TRANS JUBA LTVEST-CK PROJECT REVIEW MISSION - JANUARY/FEBRUARY 1977 Project Cost Summary (So.Sh. '000) Appraisal Consultant Review Report Revision Mission Jan. 1974 July 1976 Feb. 1977 A. 1ARKETS AND STOCKROUTES Stockroutes and Markets 3,280.9 4,901.2 2,456.8 Boreholes (except grazing ranches and IFFF) - - 7,940.6 Afmadu Holding Ground 1,545.0 3,454.3 2,778.6 Operating Costs 3,139.6 3,900.5 1,917.1 7,964.6 12,256.0 15,093.1 B. GRAZING RANCHES Capital - 4,960.0 9,287.0 6,112.4 Operating 6,982.0 7.193.0 1,844.8 11,942.0 16,480.0 7,957.2 C. IRRIGATED FODER FARM ADI FEEDLOT Irrigation works and land clearing 10,377.0 27,257.0 3,499.0 Other capital development 11,398.0 26,507.0 7,525.0 Operating costs 8,429.0 22,983.0 4,729.3 30,204.0 76,747.0 15,953.3 D. CONSTRUCTION AND MAINTENANCE UNIT Capital 2,593.0 3,555.7 3,551.3 Operating 3,135.5 54,027.6 3846.3 5,728.5 8,583.3 7,397.6 E. DEVELOPMENT AND TRATITUG UNIT Capital 1,054.0 1,105.7 349.1 Operating 2,960.0 3,065.1 2,250.0 Training 1,185.0 1,303.5 1,101.4 5,199.0 5,474.3 3,710.5 F. DISEASE CONTROL Capital Cost 839.0 1,417.5 851.0 1,310.0 713.9 510.0 2,149.0 2,131.4 1,361.0 - 79 - ANNEX 4 Tablee1 Page 2 Appraisal Consultant Revised Report Revision Mission Jan. 1974 July 1976 Feb. 1977 G. OTHER ITEMS Management and Consultnt Surveys 5,762.9 * 1,000.0 Non-Project Related Tra ning 815.0 896.5 600.0 Project Preparation 2,000.0 2,000.0 1,000.0 Loan for Cattle Purchcse Interest Payments - - 1,000.0 8,213.0 2,000.0 3,600.0 H. CONTINGENCIES 5% Physical 2,743.6 Price 10/ 8,871.6 11,615.2 I. FUNDS ALREADY SPENT 1975-76 Nil 4,600.0 8,000.0 Total Costs 71,765.0 129,168.0 74,487.9 *Included in other categories. Source: Review Mission Report, Annex 13, Table 1. -80 - ANNEX 5 Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Closing Date Extension - Bank Position as of April, 1980. The attached Bank position paper as of April 1980 is reproduced here in full from the files in view of its importance as the background to the Bank's decision to extend the project from June 30, 1980, to December 31, 1981 (for 18 months rather than the 12 months suggested in the document). The extension was approved mainly to provide for completion of the all important stockroute program which was seen as a major potential benefit for the traditional producer. The position paper was essentially based on the findings of an in-depth supervision of the project during March 6-13, 1980, which provided updating information on the project. The mission's key findings on the IFFF and Ranch II are also reproduced here in slightly edited form. -81- ANNEX 5 Page 2 SOMALIA Trans-Juba Livestock Development Project Extension of the Project Closing Date Bank Position Paper Following April 1980, PPR. 1. On April 2, 1900, the Government of Somalia, through the General Manager, LDA, requested extension of the Closing Date of the Trans Juba Livestock Development Project from June 30, 1960, to December 31, 1961. This letter was received by RMEA with another from the GM dated April 1, 1080, requesting a real3ocation of the proceeds of Schedule I of the Crodit Agreement. Receipt of these requests was acknowledged by RMEA on April 17, 1980, but further action was delayed pending formal review of the Project by the World Bank in Washington. 2. Review of the Project was undertaken in late April 1980 as part of the annual Problem Projects Review and the following conclusions were drawn by Bank management: a) Ranch Number II In view of the serious encroachment of the ranch area by ncmais and elephants, it is considered unrealistic to proceed further ith developing the area for ranching. The damagy to the one productive borehole located on the southern boundary and the burning of the construction supplies warehouse at the head- quarters site is viewed by the Bank as evidence of active resistance by the nomadic population in the area to the prcposed developments and the fact that the food resources of the area appear already to be fully utilized by nomadic livestock and wildlife are considered to be sufficient grounds for suspending further ranch development efforts in the area. For these reasons Bank management is not able to consider an extension of the financing of this project component beyond the current June 30, 1980 Closing Date. b) irrigated Fader harm and reedlot On the basis of a current analysis by the Bank of input costs and expected benefits, using the most optimistic assumptions, this compunent is considered financially non-viable. In addition, the absence of forward planning of the procurement and marketing of livestock, the observed difficulty of coordinating the basic IFFF components (comprising crop establishment, pesticide treatment, irrigation, forage, harvesting/conservation and feeding) and the unresolved logistical problems of transporting finished animals from the project site to terminal markets in the Gulf States, are further constraints pointing to technical non-viability of the undertaking. On those grounds, Bank management is prepared to provide a limited extension of the financing of this component for one year through June 30, 1981 to enable completion of existing offices, storage buildings, houses, pumping station and roads. This would allow sufficient infrastructure for Government to test tho financial/technical viability of IFFF under practical conditions if it so wished, Annex 5 82- Page 3 nothwithstandin,-, Bink mnanemnnt's advice to the contrary. The Government uhould also be cnccuragui to 1dantify alternative uses to which IFFF facilities might Lo put in order to attempt to recoup the considerable earlier investments mad in the subproject. c) Holding Ground, Stockroutce, Borehole3 These developmants, along with development of the veterinary and markcting conLers see below) are considered by the Bank to be the most valuable of the Project as they will provide significant direct benefits to traditional livestock producers in the area. The Bank recently agreed to Government's request for an upward adjustment of the contract price for developing the Aglibah Ho3 ding Ground so that development of this component is expected to proceed expeditiously. The borehole program has been seriously delayed over recent months by a dispute between LDA and WOA concerning payment by LDA to WDA for non-productive bores but this is now believed to be resolved. WDA has agreed to drill four more wells under the existing contract (bringing the total to 19) and on March 10, 1980, wrote to LKA requesting an upward revision of the contract price for the remaining seven wells to be drilled to complete the contract. In order to ensure continuation of the development of this subprojcct, the Bank is prepared to provide a limited extension of the financing for one year through June 30, 1981, contingent on expeditious renegotiation of the contract between LDA and WDA for completion of the borehole program. d) Veterinary and Marketing Centers While the development of the veterinary centers is now largely complete, further work remains to be done on the marketing centers and this cannot be completed-before the current June 30, 1980 Closing Date. In order therefore to ensure that the veterinary and marKeting cecrs arc =plcted urder the Project, the Bank is prepared to continue financing their development for a further year through June 30, 1961. 3. The extension ,of the Closing Date of the Project to enable completion of subprojects (c) and (d) above is contingent not only on prompt renegotia- tion of the contract between LDA and WUA for borehole development as mentioned in (c) above, but also on Government providing the Bank with an undertaking that high priority will be given to ensuring that the remaining project works will be completed by June 30, 1981, after which time no further extension of the Closing Date is contemplated by the Dank. - 83 - Annex 5 Page 4 Bank management hac requested that a PMEIA mission visit Somal to review thc present clisbursncnt/commitment situation with LDA and, in the light of the foregoing conditiuns attachr-d to the extension of th,e Closing Date through June 30, 1981, to drow up a revised projection of exponditures until the proposed June 30, 1981, -Closing Date. The mission should also assist LDA in preparing an appropriate request for reallocution of the proceeds of Schedule I of the Credit Agreement based on the revised Projcctions. - 84 - Annex 5 Page 5 WORLD BANKflNTE NATIONAL FINANCE COOPORATION TO: DATE: April 2, 3C0' TOM: S -ECT: SOMALIA: Trans Juba Livestock Project - Credit 462-S0 Status of Development of Fodder Farm/Feedlot and Ranch TI I. As requested, further details are provided on the current status of development of the fodder farm/feedlot and Ranch II, both of which appe&.r at this point in time to be. non-viable. This material is intended to supplement that presented in the Iarch 26, 1980, BTC of the most recent supervision mission. 2. F0DER FARM AND FFEELOT Technical Aspects - A fundamental issue concerning the overall viability of the component relates to the question of whether the complicated organizational and managerial requirements of the technical package can be provided by the Somali staff and the Indian consultants. As presently organized, the various sections of the fodder farr/feedlot are run by managers whose efforts are unco- ordinated. Thus the irrigation section works independently of the crop production group and these in turn work independently of the feedlot section. The tractor and machinery section which provides vital services to all groups, is also independently managed. In theory, all are coordinated by the project manager but as he is stationed in Kismayu and seldom visits the site, there is no co- ordination. Repeated suggestions by Bank supervision missions to rectify this situation have gone unheeded. Along with the lack of coordination is a lack of supervision, which is equally serious. Work on the farm is of poor general standard. The maize crop seen on the last March 10-12 supervision of the forage production unit was very uneven and was heavily infested with stem borer. Seed of unknown origin had been used which had contributed to the erratic plant distribution. At the Lifaii uf Lhe mision s visit, the crop was due to be irrigated but was not because the irrigation section manager was ill and nobody else had the authority to start the pumping system. The 48-hour delay which ensued resulted in severe crop wilting and loss of plants. Loss from stem borer was also seen to be extensive but no chemicals were available for treatment and no plans were made to procure any, notwithstanding that the fodder production/feedlot consultant, had predicted the problem and recommended purchase of suitable chemicals. The indications are that farm staff do not possess either the initiative or the technical competence to produce the basic feed required by the feedlot. Execution of the technical package is beyond their qualifications and experience. The situation is aggravated by the general incompetence of the expatriate farm manager who is unable to bring to his job the essential skills and leadership required. The net result is that the vital feed production unit is functioning very inefficiently - about 10% of the level required for it to be Annex 5 -85 - Page 6 a success. Admittedly, the forogo production program i in its early stages and could he expected to improv with tim-. 3uch improverment, however, is likely to bc miniral unlcess a lae ex- ,prtriate technical input were made availablo. Because cf the difficulty and expense of recruiting quali-fied and expericiced specialists for service in Somalia, this is not considered to be a practical alternative. 3. Again, neither the supply of cattle for the feedlot (12,000-13,000 head/year projected) nor their disposal after feeding has been organized. These are major oversights. Considerable for- ward planning is required tQ procure this numbor of suitable feedlot cattle per year. Because of the variation in size and type of cattle on offer from traditional producers, 15,000-18,000 or more may have to be purchased to provide the 12,000 required for feeding. Entry of feeder cattle into the feedlot has to be coordinated with sale animals leaving it. There are complicated problems of cattle pur- chase movement, aggregation, quarantine. selection, holciing and feeding to be organized. Requests by supervision missions to LDA for plans on these matters have not been successFul. The fact is that LDA has no conception of the feedlot cattle procurement and marketing operation - no senior staff members have done any forward Planning or budgeting and there is no indication that any of them intend rectifying the situation. The technical expertise and management which is so essential to ensure the success of the com- plicated operation of ensuring a smooth flow of cattle through the system is non-existent. 4. Disposing of feedlot cattle is an even more serious problem on which no action has been taken. The overall viability of the operation is dependent on efficient disposal of cattle at the end of the feeding program. The viability of the operation was justified by the appraisal mission and the review team in terms of both quality (i.e. grade improvement) and significant weight gains. It is of concern that at this late stage LDA is unable to produce any vinrinrR tha-i rhPrP i n p4\/ nremiiim nn nuialit\/ - inrDor it is even questionable that weight and finish (as opposed to size or age) plays any significant role in determining market price. All LDA cattle transactions, both buying and selling, are on a per head basis and there are no moves to change the system. LDA justifies its present practices on the grounds that there is industrywide resistance to any suggestion of a change from a per head to a weight system. While bigger-framed animals in general conand higher prices, the per head system is so arbitrary and subject to error as to undermine the rationale in support of feedlot develop- ment - the assumption that better finished (i.e. quality) animals receive a premium is incorrect, at least in the current market. 5. A further point concerns the transport of cattle from the feedlot to Saudi Arabia and other gulf state markets. The Kismayu port lacks adequate facilities to handle live shipments. Also the type of vessel used is not suitable although steps are reportedly - P- Annex 5 Page 7 bCei3 taken to procure two specalised vessels to operate in S'omalia for livustiock transport This could. result subsequntly in the payment of a quality premium although the extent of the premium at this point is unknown. Review of loading operations by the must recent supervision mission indicated conditions to be completely unsuitable for handling quality consignments. Depending on the mix of sheop/goats, camels and cattle, and their availability for londing, a boat takes 2-4 days to load. The most disturbing aspect is the loading of cattle in which groups of 4-6 head are slung into the hold in nets. The system is wholly inappropriate for handling quality livestock. Once loaded, the minimum sailing timL to Jeddch is 8 days with 2-4 days unloading after arrival. Feedlot cattle are estimated to shrink 20-30 kg in the 12-16 day journey from the feedlot to their final destination, or roughly half of the estimated gain made on the feedlot. This loss, with mortality, bruising and damage and the imprecise per head buying and selling arrangements combine to seriously undermine the technical viability of the operation. Financial Aspects 6. The appraisal and review missions justified the financial viability of the feedlot on the grounds of: i) a grade improvement which when translated in financial terms represented a 40-58% differential between the cost of cattle for feeding and their sale price. This is not so in today's market where no premium on grade or quality can be demonstrated to exist; and ii) a liveweight increase of 0.75-0.90 kg/day which is consi- dered in hindsight to be unrealistically high, 7. Although no firm data is available on weight gains under o.mali rndition'. th. review missinn's assumption of 50 kg/head in 67 days (based on feedlot performance of cattle in Yenya managed by expert feedlot operators), is considered too optimistic. Present indications are that a gain of 50-60 kg in 100-320 days is much more realistic given the observed management deficiency in Somalia and the difficulty of fine-tuning what is essentially a very complex system. It is not doubted that some individual cattle may gain one kg/day but on a group basis cattle fed a maize silage/alfalfa regime are not likely to exceed 0.5 kg gain/head/day under the management conditions anticipated. On the assumption that the 6verage gain will be between 50-60 kg and that the buying and selling price is between 4 to 5 So shs/kg, it is projected that gross income per animal under present conditions will average some 225 So shs. A detailed income and expenditure statement has been precared to reflect current costs and prices. Assistance in preparing ne data on which the table is based was given to the mission by local staff. -87 - Annex 5 Page 8 Present indications are that the operating income of the overall operation will be negative in every year of operation. The table, if anything, understates losses in later years as some key operating costs are expected to increase at a disproportionately fast rate with time. Among these are included animal health, tractor/vehicle/machinery and fertilizer costs. Since the assumed selling price is - estimated FOB Kismayu, no allowances have been made for losses during transit from Kismayu to gulf ports. 8. The problm of ar continuous nofti7,ve operating income is compc,undcd by thn fot that component investments of some So shs 14.5 million will have boen orde 2-4 yers in advanco of any soles. The cash flow of the overall operation woul d then be totally un- realistic. Theoe is no doubt - given the present cost/price structure in Kismoyu, that the ccrponent is financially and economically non- viable. 9. The recommendation of the mission is that further feedlot development be limited to finishing the office and service bui]dings now under constructon, the pumping station and the internal roads, and that feeding be limited to a pilot scale involving four cycles of 150 nattle each to test the technical and financial/economic viability under practical conditions. If, as is prdicted, the feed- lot concept is non-viable given present conditions, alternative uses should be sought for the facilities. 10. Ranch IT Development of Ranch I was suspended on the recommendation of the 1977 Review [Mission. Development of Ranch II was recommended to continue but on a reduced scale. Development was initiated and continued until mid-1979 when LDA suspended development when two boreholes drilled near the ranch headquarters 1-re non-productive. Four bores have been drilled in total on the x h, three of which are non-productive and one highly productive. latter is situated necr the southern boundary 10Km or m, . from the ranch buildings. 11. While attention has been focussed by LDA on the problem of non-availability of water at the headquarters, detailed super- vision by the March 6-13, 1960, mission indicated problems of an even more serious nature which suggest that development should be shelved indefinitely. The problems are: i) Widespread encroachment by elephants and nomadic cattle to the extent that the feed resources of the ranch are virtually fully utilized; ii) damage to the uppercasing of the productive bore, pre- sumably by nomads - thu extent of subterranean damage could riot be ascertained but may be serious -- the effort 88- Annex 5 Page 9 required to damag the uppercasing would certainly suggest an intention to destroy the faolitys and iii) recent burning of the building supplies warehouse plus contents -t the ranch headquarters in what appears to be arson -- again presumably by noads. 12. LDA has continually maintained that the ranch site was un- occupied, but this is no longer the case. It is difficult to know the population of elephants at present but it is significant, judging from the large quantity and wide distribution of elephant dung. There was significant damage to-trees over the ranch--- also to re- inforcing steel rods lying at the headquarters site. Cattle numbers are equally difficult to estimate but five large bomas were seen during the supervision and it is reasonable to assume in the vast area of the ranch that there were more obscured from view. Natural grass- land on the ranch was well erazed and there was widespread evidence of movement of large herds. In the 80,000 ha of the ranch it is quite possible that there are 4,000-5,000 cattle being pastured. A number of this magnitude would be quite consistent with the intensity of grazing and the distribution and quantity of cattle dung seen at all points on the ranch visited during supervision. 13. The seriousness of the situation on the ranch reopens the question of ranch viability. The key problems are technical and social and they are serious problems which are not readily resolveable. Elephant damage is quite common in the general area -- early in March over Sha of bananas were destroyed by them on a farm near the feedlot. They are a particular hazard to water systems during dry weather when they have been known to destroy systems completely. The signi- ficance of their presence in numbers on Ranch II cannot be ignored. 14. More serious is the nomad question. If Ranch II forms part of the traditional grazing system of the area, there has been a major misjudgement in demarcating the area and commencing constructions on it. The acting General Manager of LDA who accompanied the mission on the ranch visit, said this was a new development - nomads had not previously camped in the area but had only passed through it. That could be true. PIU staff tends to believe it and feels that these may be new developments occasioned by general deterioration of Ethiopian/Somali relations and the refugee problem. Whatever the explanation, it is serious and the mission recommends that development of Ranch II should continue to be suspended pending a complete review of the situation. No further developments should be encouraged before LDA evaluates past efforts and elaborates a ranch development plan on which future work should be based. Future work, if any, should be limited to the area adjacent to the productive well in the south and be of a strictly pilot nature. Taking account of the nature of the present problems on Ranch II, it is highly probable that the component may not be able to be saved. In this event, losses will need to be cut by suspending development indefinitely. About So shs 3 million are estimated by LOA to have been invested - 89 - Annex 5 Page 10 in the ranch so far. It is likely that most of this will hav, to be be written off as thure are few infrastructural. developrments which can be salvaged. SOMALIA TRANS JUBA LIVESTOCK PROJECT - IFFF EST'ITATED INCOME AND OPERATING EXPENSES - ------- ---- Operati.:& Year a/ --- INCOiE Price/Head 1 2 3-20 So Shs Head So Shs Head So Shs H 0 So Shn 000 000 Cattle Seles 1,275 b/ 5.850 7,497 5,820 11,246 11,750 i. '4 CFE=ATNG XPERSES. Cattle Purchases 1.050 c/ 8,000 6,300 9.000 9,450 12,000 12,6C3 Animal Health d/ 96 144 132 Trastcr/Venicle/7achinery e/ 750 900 1,2^0 net/.abour f/ 3S0 360 !0O Purch=sSd 7esd 1/ 42 83 84 Ssed/Pasticidss h/ 60 90 120 Fertilizer h/ 90 135 ai-intenance 1/ 88 200 350 1'iscellareous J/ 35 54 .72 7,820 .11,393 15.273 1 Operating Result (223) (150) (264) Operating result as % of sales price/head 2.98 - 1.33 S/ First oceratin: year of fcndlot assumed to coincide with year 3 of construction program. / arch 1i9 price of 280-285 kg animal FOB Kismayu. Gross sales calculated on basis of 2% mortality. c/ March 1SS0 cost of 230-235 kg animal from traditional producers in Trans Juba. c/ Current LOA estimates for tags, dip, fluke drench, tryps prophylaxis, based on So She 15.00 per head. e/ Ircludes costs of all fodder farm/feedlot ru3chanical operations, cattle haulage and personnel transport. -/ Excludes tractor/machinery operators and vehicle drivers (charged under (e) above. No allowance included for expt zsecialist input. g/ Ccntin7ency feEd (hay, roughage, by-products), based on So Shs 7.0 per head. h/ Input costs qnly. 6ased on minimum levels recommended for use under local cordit4dm by ;he feedlot consultant fodder far%/Feedlot consultant. i/ Based on 2.5% cf aggregate value of onstruction investments made two years earlier. / O-ffice' Iu-r xpneI Note: Purchase ind sales prices of aninals are estimated on the basis of actual prices gathered by the mission in the field. - 91 - ANNEX 6 Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Implementation of Principal Project Components General 1. This annex describes the development history and highlights of the principal project physical components. The components comprised: (i) marketing (including stockroutes, markets, holding grounds and water supplies); (ii) two grazing ranches; (iii) the irrigated fodder farm and feedlot (IFFF); and (iv) the animal health programme. 2. The project was to be implemented in the period October 1974 to June 1980 but implementation was delayed and the project was extended for 18 months from June 30, 1980 to December 31, 1981 in order to complete the stockroute, holding ground and water supply components. Notwith- standing the extension, however, at the project's close in December 1981, none of the project components or sub-components was completed. Even seven months later, at the time of the final disbursements, all but one of the boreholes still remained unequipped despite protracted efforts by Bank staff to assist project management in completing the stockroute, holding ground and livestock water supply components. Implementation and.Status of Project Components 3. Markets, stockroutes and holding grounds Effective development of markets, stockroutes and holding grounds was central to the marketing strategy developed at appraisal and endorsed by the Review Mission. The aim was to provide facilities to reduce liveweight loss and mortality during trekking, and to increase overall productivity and offtake by development of improved markets and marketing infrastructure and expansion of veterinary services. The fact that these facilities were not completed and functioning effectively in the project's lifetime was a significant project failure because of their potential to provide important benefits to pastoralists and to the economy. As early as 1976, supervision missions emphasized the importance of the marketing component of the project. Consecutive missions concentrated on assisting project management to complete the stockroute, holding ground and livestock water components because of their crucial overall importance to project success and to the welfare of the pastoralists. 4. At appraisal it was planned that the PCMU would carry out all construction on the marketing component. The PCMU, however, was never an effective organization, being troubled chiefly by inadequate staffing. The resignation of the Head of the Unit (also the project engineer) in December 1977 effectively suspended PCMU work on the marketing components for the best part of two years, i.e. all of 1978 and most of 1979, and this lost ground was never regained. ANNEX 6 - 92 - Page 2 5. Markets In 1978 it was agreed bEtt;eei the Bank and LDA z- limit the component to the construction of one canter in Tabda which was completed in late 1979 by PCMU. However, the center never functioned. It was.,gnstructed on a black cotton soil base and, because of unsatisfactory design of the foundations, constructions settled irregularly and were in danger of collapse. Consequently the market was abandoned and fell into disrepair. 6. Stockroutes Appraisal plans envisaged forming over 1,000 km of stockroutes and development of water suFplies along the routes to provide staging points approximately 30 km apart. 1/ Water supplies were to be provided by boreholes or dams (see para 8). The stockroutes were intended to reduce livestock weight loss and mortality during trekking to markets. Limited PCMU achievements included the clearing of the Afmadu-Badada stockroute and construction of 2 earth dams in 1977, and a further six earth dams in 1979/80. 2/ The inability of the PCMU to develop the whole component as projected prompted the appointment of a private company in 1980 to clear some 525 km of stockroutes which the company did quickly and effectively. Because of earlier slippages, the costs at that late stage were such that only one-half of the appraisal targeL was achieved. The routes selected by project management were those considered of highest priority penetrating areas of high livestock population, acces& to which had previously been seriously restricted. The routes which were successfully developed under the contract under reference consisted of: (i) Khadija to Garsale; (ii) Garsale to Busar; (iii) Busar to Fafadun; (iv) Fafadun to Qorohai; (v) Qorohai to Afmadu; (vi) Aglibah to Bibi Mia to IFFF; (vii) Aglibah to Kismayu; and (viii) Aglibah to Hosweyne to Kismayu (Refer Map 3) 7. Holding Grounds The holding ground component projected at appraisal included development of a new 20,000 ha ground at Afmadu, and clearing of 12,000ha and improvement of the water supply on the existing Kismayu holding ground. The two facilities would allow LDA to control and regulate the flow of cattle to the ranches, IFFF and the KMF and for live export. The same contractors who later undertook the stockroute clearing were awarded the contract to develop the Afmadu holding ground in 1977. They were to construct firebreaks along external boundaries and internal sub-divisions, an access road, buildings and holding and sorting yards. The contractors were to commence work at the beginning of 1978 but, after investigating the site, project management susDended the work as the site was found to be an important water supply system of dams and wells used by pastoralists who were semi-settled on the area. The Regional Governor supported objections by local pastoralists to the planned holding ground development and declared the area should not be allocated to the project. In June 1979, 18 months later, a new site at Aglibah approximately 80 km south of Afmadu was selected as the location for a new holding ground. Its development was interrupted by unusually heavy rains in 1980 but was completed by April 1981. The delay in site selection led to an increase in the cost of the contract work of over 80 percent. Provision of adequate 1/ Upgrading of the Badada-Kismayu road was also planned under the program. 2/ Limited improvements were also made to the Badada-Kismayu road. ANNEX 6 Page 3 livestock water proved to be a problem at Aglibah. Two boreholes were established both of which were potable. However, only one was equipped with a temporary test pump unit which provided some water but not to the borehole's potential. Re-development work on the Kismayu holding ground was restricted to the drilling of one potable borehole by the WDA but again, as at the close of the project, it was not equipped. 8. Water Supply It was envisaged at appraisal that WDA would be contracted to complete the drilling and equipping of 26 boreholes for the project. These were to be located on stockroutes, at holding grounds, on the IFFF and at markets and veterinary centers as required. The construction of ponds/dams was planned to complement borehole development; they were to be sited in areas where borehole water was found to be unavailable or non-potable. 9. Borehole Development In July 1977, WDA signed a contract with the LDA to drill and equip 26 boreholes. In early 1978 the first borehole was completed and equipped at the IFFF. 1/ During 1978/1979 WDA established nine more potable bores. However, WDA had no hydrogeologist on its staff and insufficient equipment and vehicles in the project area. The difficult conditions in the Trans Juba area presented the WDA with technical problems with which it was unable to cope and consequently it was unable to meet the terms of the contract. The incoming General Manager of the WDA, appointed in January 1979, objected to the terms of the contract agreed by his predecessor with LDA on the grounds that: (i) the contract price did not take into account the costs of mobilizing WDA equipment, construction of access roads to drilling sites and increases in operating costs (particularly fuel); (ii) the WDA had experienced delays due to inclement weather and these delays were not provided for in financial or performance conditions of the contract; and (iii) there was a paucity of information on the hydrogeology of the project area as no test drilling had been undertaken and hence WDA was placed at a considerable disadvantage as a consequence. 10. In June 1980, LDA and WDA negotiated an extension of the July 1977 contract for developing seven more boreholes which was to include drilling, casing, test pumping of the new bores and the equipping of all boreholes developed under the earlier contract. Under the extended agreement, WDA was to provide the pumps, motors and construction materials and would assume responsibility for all associated construction work. The seven boreholes were drilled over 1980/81 after extended delays, of which five were productive, but WDA did not either supply or install the equipment 1/ At the project's close this was to be the only one of 15 potable boreholes established by WDA for the project that was permanently equipped. Another was temporarily equipped with test-pumping equipment on the Aglibah holding ground (paras 7 and 10). ANNEX 6 Page 4 as it had undertaken to do under the contract extension. LDA was disbanded in March 1981 and it was not until November that project management, now under MOLFR direction, formally terminated the contract with WDA. Hindsight indicates that WDA never had the technical capacity to meet the terms of its contract with LDA at any stage during the project. At project closure, only one of 15 boreholes with potable water was permanently equippedi also, one other was temporarily equipped. Pumps, motors and the materials for construction of pump houses, attendants' houses and water troughs were, however, procured by LDA using procedure 3. MOLFR sought IDA agreement to use private contractors to complete the equipping of the boreholes but the quote from the one contractor capable of undertaking the work was considered to be unacceptably high by IDA. In view of the potential benefits so far foregone by the inability of the project to complete the equipping of the boreholes, IDA/GoS are investigating the possibility of finishing the work under another IDA-financed project such as, for example, the Central Rangelands Project (Credit 906-SO). Completion of the equipping of the boreholes would overcome one of the major shortfalls of the oroiect and would provide important direct benefits for the project area's nomadic pastoralists and would at the same time provide substantial indirect benefits to the economy as a whole. 11. Pond Development The construction of earth ponds was planned to supplement the borehole system in areas where bores were too expensive to develop or where potable water was not available. A total of 15 ponds was developed in the project's lifetime as follows: Site Capacit (m) By PCMU Wadajir 1 x 10,000 Helina 1 x 10,000 Reebay/IFFF Stockroute 1 x 10,000 Afmadu/Fando Faan Stock- 2 x 10,000 route Reebay/Kismayu Stockroute 2 x 10,000 Hagar 1 x 10,000 By Contractor Afmadu/Qorohai Stockroute 2 x 20,000 Aglibah 1 x 10,000 Aglibah/Hoosweyne Stock- 1 X 10,000 route Aglibah/Kismayu Stockroute 2 x 20,000 Kismayu/Hoosweyne Stock- 1 x 10,000 route Total 15 12. Siting of the ponds is detailed in Map 3. Construction of the ponds proceeded in the years 1978-81 with most of the work being undertaken in the later years as the various problems related to implementation of the LDA/WDA borehole contract became apparent. Pond construction in general proceeded satisfactorily and the overall standard of workmanship was of high standard, particularly that performed by the contractor. - 95 ANNEX 6 Page 5 13. Development of Grazing Ranches The development of two 80,000 ha ranches was envisaged at appraisal as part of the overall marketing component. The rationale for the ranches was endorsed by the 1977 Review Mission which recommended, however, more gradual development and suggested that only one ranch be developed initially in a reduced area of 50,000 ha and that development of the second be deferred for two years. Most detailed planning of the ranches was completed at the time of the Project Review and the Review Mission found the plans, with some exceptions, to be generally acceptable (Review Mission Report, Annex 9, para 12). 14. From the outset of implementation in 1977/78, the development of the first ranch (Ranch II) encountered serious problems. The area, even after reduction to 50,000 ha, was unrealistically large considering that ranching was a new concept and not even a pilot ranching enterprise had been tried in Somalia previously. Again, the management consultants, who were from a central European country with no ranching tradition, lacked the qualifications and experience to develop such huge areas into ranches. Successive misjudgements on their part and that of LDA counterpart staff (who likewise lacked qualifications and experience for the task), with respect to ranch siting and the type and location of ranch infrastructural developments , resulted in errors being compounded one upon another. 15. For example, Ranch 11 was located in an area of uniformly fairly heavy bush cover despite appraisal expectations that "Two ranches would be established on natural grasslands in areas virtually free of bush-. The bush density was such that questions should have been raised by the consultants and LDA staff concerning the technical/financial feasibility of ranching in such areas. Surprisingly, the point also appeared to be missed by the Review Mission. 16. This initial serious oversight was compounded by the decision of the consultants to locate the headquarters buildings of Ranch II in the center of the ranch 1/. This was done in the belief that the headquarters should be located equidistant from all boundaries. In doing so, they overlooked the fact that the chosen site had no potable water (two bores established near the site were non-productive) and was located in an area which was inaccessible during the wet season when approaches were inundated with over a meter of water for four to five months. 17. The compounding effect of so many misjudgements could not be contained indefinitely, particularly since the decline in beef prices by some 50 per cent in 1973-76 and steeply increasing input costs throughout the 1970s, had made the financial viability of the enterprise highly questionable. Supervision of the ranching component in 1978-79 appeared to be superficial and the mounting problems did not receive sufficient attention. There were suggestions to clear bush and to implement grazing trials in order to obtain data to evaluate the viability of ranching under the conditions pertaining in the area but only token efforts were made in this direction and no hard data were collected. The development of Ranch II limped through to March 1980 when it was abandoned following intensive IDA supervision of the ranch which disclosed intrusion of a significant elephant population and the presence of appreciable 1/ Contrary to Review Mission recommendations. ANNEX 6 Page 61 numbers of nomads (and their cattle), who regarded the ranch as part of their traditional grazing areas. Damage was also observed to the upper casing of the one potable borehole on the ranch and the storage warehouse at the headquarters had been burned down. Both acts appeared deliberate and were assumed to be acts of resistance by local nomads to the development of the ranch. With hindsight, it is noteworthy that the Review Mission mentioned the presence in the ranches of both nomads and wildlife including predators and elephants in their report (Annex 9, para 11), but apparently did not assume their presence to be of significance. 18. In the development of Ranch II, considerable investments were made in construction of an access road, boundary and internal firebreaks, and in buildings (some of which was completed) and drilling four boreholes, (one of which was potable). Against these expenditures, no benefits were generated as no project cattle ever grazed the area. The shortfall in the development of the ranching component was one of the significant failures of the project. 19. IFFF Development of the 1,200 ha IFFF facility was central to appraisal strategy to increase meat production from the Trans Juba area. The IFFF was to fatten cattle for the KMF which would in turn sell its output as frozen or chilled beef on the world market. The latter was expected to show a higher value added than canned beef which was KMF's main traditional product, by allowing entry into higher quality markets thereby increasing economic returns, nomads' incomes and the profitability of both LDA and KMF. 20. The IFFF as planned was a sophisticated operation. It was by far the project's most expensive development undertaking. It involved a large upstream cattle procurement exercise to provide the 25,000 head of planned annual throughput. Some 20,000 of the cattle were to come from the two ranches and 5,000 from the Aglibah holding ground. Irrigated fodder was the main food source and it was to be fed as both silage and green feed. Successful operation of such a large and complex enterprise required considerable forward planning and close integration and management of cattle procurement quarantine, selection, movement, feeding and marketing. 21. By early 1976, some 18 months after Effectiveness,IDA, the con- sultants and GOS were raising questions concerning the project's viability. The consultants had reported large project cost increases of the order of 100 per cent above appraisal estimates and the largest increases were in the IFFF component. These observations, together with greatly reduced international prices for beef and changes in local cattle purchase arrange- ments between LDA and KMF, raised doubts about the IFFF's financial viability based on sales of chilled and frozen beef by KMF. The management consultants, who had completed most of the IFFF plans,recommended in July 1976, prior to any appreciable developments being undertaken, to drop the IFFF (see Annex 3). - 97 - ANNEX 6 Page 7 22. It was with this background that the project was reviewed in January/February, 1977. The Review Mission endorsed the concept of the IFFF but changed the strategy from producing cattle for processing by KMF to one mainly of live exports to the Middle East. The revised plan envisaged 190 ha only of irrigated crops (fed as maize silage and lucerne hay) to provide for a throughput of 13,800 cattle per year to produce average live weight gains of 50 kg (200 to 250 kg) in 67 days (0.75 kg per head per day). The financial viability of the operation was based on: (i) a grade improvement which in financial terms represented a 40 per cent differential between the purchase and selling price of cattle; and (ii) a liveweight increase of 0.75 kg per day. 23. Between late 1977 and early 1980, most of the IFFF infrastructure was established. However, the quality of all work was generally substandard. Development was hampered by lack of concern, management and organization. After the management consultants left in 1977, the various IFFF sections were run by inexperienced managers whose efforts were never coordinated. For example, the irrigation section worked independently of the crop production staff and both, in turn, worked independently of the feedlot group. The tractor and machinery group, providing services to all others, itself worked quite independently. In theory they were all coordinated by the Project Manager but he was stationed in Kismayu and visited the site infrequently. 24. Along with lack of coordination there was no overall supervision. Compounding all, it was evident from an early stage that the effective implementation of the IFFF package was beyond the qualifications and experience of local staff. Worse still, no notice was taken of repeated suggestions by Bank supervision missions nor of the recommendations of the feedlot consultant who worked on the project in late 1979 - early 1980. In fact, the consultant was so frustrated by the indifferent attitude of his counterpart staff that he declined to undertake the second three-month visit to the project provided for under his contract on the grounds that it was "a waste of project money and my time'. The project's fodder crop production difficulties were exacerbated by the general incompetence of the expatriate farm manager who was unfitted by either qualifications or experience for his post 1/. 25. In short, the irrigated fodder farm produced only enough feed for a pilot study with 94 cattle. Not more than 30 ha was ever partially irrigated (although 90 ha was prepared for irrigation) and the total dry matter produced did not exceed 100 tons. This contrasted sharply with the 4440 tons per annum required by the Review Mission plan. 26. The results of the one feeding trial undertaken in the course of the project were also in sharp contrast with both appraisal and review projections, particularly with respect to daily weight gains and feed conversion coefficients which were crucial to the attainment of optimum 1/ The appointment of the individual concerned was made by the General Manager, LDA, against the advice of IDA who considered the appointees background unsuitable for the post. - 98 - ANNEX 6 Page 8 technical efficiency of the operation. Data are summarised in the following table: Table 1 Technical!7inancial Coefficients from IFFF Pilot Feeding Trial Appraisal Review Pilot Trial ----- Projected-------- Actual Weight on leaving Feedlot (ky) 370 250 259 Weight on entering Feedlot (kg) 300 200 213 Liveweight Gain (kg) 70 50 46 Daily Liveweight Gain (g) 933 750 370 Days on Feed 75 67 124 Mean weight (kg) 350 225 230 Intake per Day (kg DM) 8.80 5.18 5.75 Total Feed DM used 660 347 713 Feed Conversion 9.43 6.94 15.5 Feed Production Cost (So.Sh./kg DM) 0.25 0.4 2.1 1/ Cost of Incremental LW (So.Sh.) 165 770 1488 Breakeven Sale Price (So.Sh./kg) 1.56 3.08 9.65 1/ Sale Price (So.Sh. per kg) 2.20 3.50 6.50 Note: Results are from trial involving 94 cattle. 1/ Does not include charges for maintenance or amortization of fixed assets. Figures, therefore, underestimate actual costs. 27. While these parameters could be expected to improve in a better managed operation, it is doubtful if the coefficients for Somali range cattle could ever approach those projected at apparaisal and review which, with hindsight, were highly optimistic. The PCR felt that all parameters used by the Appraisal Mission and, particularly, the Review Mission to evaluate technical/financial viability qf, the IFFF were similarly optimistic. For example, at appraisal, a premium ror weligt and finish (i.e. grade improvement) varying from 53 to 83 percent (mean of 57/58 per cent) was used in financial rate of returncalculations while at review a figure of 40 per cent was used. These parameters, along with high daily weight gains (0.93 kg per head at appraisal and 0.75 kg per head at review), were crucial in justifying viability. While the latter figures appear seriously to have over estimated the technical situation , the incremental financial benefits estimated from utilizing the grade premiums noted above further seriously compounded the situation. This was particularly pertinent since, insofar as the PCR Mission could ascertain (in confirmation of earlier supervision observations), there was never any evidence that the Somali market was paying a premium for grade, quality or finish, as opposed to age and size. 28. All LDA/KMF cattle buying from pastoralists was in fact on a per head basis and no moves were ever made nor anticipated (nor encouraged under the project) to change to a weight system. The best that could be said was that - 99 - ANNEX 6 Page 9 bigger-framed, meaty animals commanded higher prices in the traditional market but the per head system was so arbitrary and subject to personal judgement and error as to undermine the whole sophisticated rationale for feedlot development. 29. In summary, the feedlot failed on all counts. It was technically and managerially too difficult in the project environment and the PCR Mission4s investigations indicated that, in fact, it was never a financially viable proposition. Financial viability was only ever demonstrated in theory and, to do so , required the most optimistic (and in the PCR Mission's view quite unrealistic) financial assumptions. Significant investments of the order of So.Sh. 30 million were made in the period 1977-80. There were no accrued benefits apart from some 4400 kg of liveweight gain from the pilot trial, which was negligible in comparison with the investments made. The failure of the IFFF was the most costly and the most significant failure of the project. 30. Animal Health. At the time of project appraisal the Veterinary Department of MOLFR was already established in the Trans Juba area. Appraisal strategy was to enhance the existing health care services by: (i) developing the offices/laboratories of the Regional Veterinary Centre in Kismayu; (ii) developing four district veterinary centres (at Afmadu, Gelib, Giamma and Brava) and five veterinary dispensaries (at Bulo Haji, Goba, Anjail, Beles Gogani and Tabda); (iii) providing vehicles, equipment, services and operating costs (for three years) for each facility; and (iv) establishing a disease-free zone in Lower Juba from which cattle could be produced for providing products through KMF to high value European markets. 31. The Review Mission endorsed the component essentially as designed but considered that establishment of the disease-free zone was "neither practical nor economic to attempt to implement in the Lower Juba area within the lifespan of the project ". The mission recommended immediate strengthening of the Regional Veterinary Department including the construction of the four district centres (substituting Badada for Brava) and the five dispensaries at "suitable locations7.. 32. Implementation of the component was mixed. PCMU completed the five dispensaries in 1977-78 at Bulo Haji, Beles Gogani, Tabda, Wadajir and Badada and they were handed over to the Regional Veterinary Office in 1979 which was starting to receive German technical assistance from GTZ. The buildings were poorly sited and constructed. Those at Tabda, Bulo Haji and Wadajir were subsequently rehabilitated by GTZ but the others were not. None of the district veterinary centres was built under the project and little input was made into the Regional Veterinary Center in Kismayu. Four of the five planned vehicles were provided under the project along with a quantity of equipment, chemicals and drugs., and some operating expenses. - 100 - ANNEX 6 Page 10 33. Project staff were not sufficiently cooperative with the Regional Veterinarians in the Review Mission's view. They did not liaise adequately over building designs and procurement of equipment and vaccines and as a consequence there was significant wastage of funds. This trend unfortunately continued until the time the GTZ assumed full control of all facilities. 34. The development of the component overall fell far short of appraisal/ review expectations due to poor workmanship by PCMU and general lack of quality control at all stages, due essentially to lack of management/supervision and qualified and experienced tradesmen. It is certain that things would have been even worse still if GTZ had not provided assistance to the Region when it did. GTZ no doubt salvaged much of the infrastructural development which would have otherwise fallen into disrepair. Some impact from project investments in the component can be expected but such impact was assessed by the PCR Itission as being only minor at best. - 101 - ANNEX 7 Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Changes in Responsibility for Livestock Marketing in the Trans Juba Area during the Project Background 1. The LDA and the KMF were the principal institutions involved in the project by virtue of their roles in livestock marketing and processing in the project area. Their efficient functioning was essential to the viability of the project and the roles to be played by each institution were embodied in the clauses of the Project Agreement and Subsidiary Loan Agreement. 2. The system of LDA supplying cattle to the KMF for its slaughter requirements was to be the fundamental operating strategy of the project. Under the project, LDA would provide the KMF with canning quality cattle directly from markets and fattened animals from the feedlot for the production of carcass beef. During the project the right to purchase and trade in livestock in the Trans Juba area was transferred from LDA to the KMF and was later transferred back. These conflicting moves had destabilising and demoralising effects on LDA and the project because the financial viability of IDA (and hence the project) depended on LDA's ability to trade in livestock and to generate revenue from its position as middle man between the pastoralists and KMF. Also, the production component of the project depended on the ability of IDA to purchase immatures and other cattle suitable for growing-out on ranches and feedlot fattening. Ultimately, the issue of LDA's right to trade in livestock was overshadowed because of other failures of IDA as an institution and its inability to manage the production components of the project competently. However,the changing roles of the LDA and KMF in livestock marketing, which had the appearance of being quite arbitrarily imposed by Government, produced needless conflict and confusion within the project and had a definite undermining influence on the attainment of project objectives. Sequence of Events 3. At the outset of the project LDA was responsible for DurchasinQ the total requirements of the KMF which it did at an agreed margin. In August 1975, however, the KMF was authorized to buy its slaughter cattle require- ments directly from pastoralists or traders and it took over several of LDA's marketing teams. The KMF had successfully objected that IDA margins had become excessive due to operating inefficiency. 4. Approximately a year later, LDA was given responsibility for operating the KMF and it also assumed responsibility for procuring KMF's cattle requirements using Factory funds. 5. In September 1977 the situation was reversed, and KMF was appointed to take over the Juba Region operations of the LDA. The Ethiopian-Somali conflict of December 1977 - January 1978 resulted in the withdrawal of the - 102 - ANNEX 7 Page 2 USSR market for canned beef, on which the KMF depended for the sales of its only sign:ficant product, with the result that from early 1978 to October 1979 the KMF remained closed and the roles of both KMF and LDA in promoting cattle marketing were severely reduced. From January 1978, LDA was actually prohibited from trading in the lower Juba Region for a period of some 12 months and, between this and KMF's closure, there was no cattle trading by LDA/KMF in the project area at all. 6. During the years 1975 to 1980 inclusive, the LDA carried out live exports from the project area, purchasing from pastoralists and traders and selling to Middle East markets. The LDA is reported by MOLFR and project staff to have made heavy financial losses on its export activities but no details were available to the PCR Mission as the accounts of LDA were never properly prepared. 7. In October 1979, the KMF was reopened as an autonomous body- under the Ministry of Industry. The LDA operated separately under the MOLFR. In December 1980 the KMF was closed as a result of its poor operating and financial performance. In March 1981, the LDA was dis- banded and its responsibilities were reassigned within agencies and departments of the MOLFR. The financial losses of the LDA, its weak economic impact, its overall poor performance and its general failure to contribute to livestock subsector development were given as the main reasons for its disbandment. Thus by March 1981, the two key institutions responsible for livestock marketing/processing in the Trans Juba area, which the project was to promote and develop as centrepieces of GoS strategy for development of the area's important livestock resource, were both disbanded. With their disbandment the last shreds of hope that the project could even partially attain its livestock marketing development objectives were irretrievably dashed. 8. Livestock marketing and export data and KMF production data are shown in Tables 1, 2 and 3 attached to this annex. Data for KMF cattle purchases in Table 3 Ao not reconcile with those of sales to the KMF by LDA (Table 2) for 1972. The correct figure was not available to the PCR mission. In 1975, 1976 and 1978 it is assumed that the differences between purchases by KMF from LDA and KMF's total purchases were attributable to KMF purchases from other sources. ANNEX 7 Table 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Livestock Production and Exports: Trans-Juba Area Livestock Exports From Kismayu Port 1/ --------------Type of Animals----------------------- Year Camels Cattle Sheep/Goats 1975 3000 4400 10300 1976 5465 26155 15530 1977 NA NA NA 1978 3840 15200 18500 1979 3775 14200 10566 1980 5944 24374 9522 1981 5974 30289 4203 1/ Total by all exporters. Source: Development Department for Livestock Marketing Facilities of the National Range Agency (formerly the Development Department of the LDA). - 104 - ANNEX 7 Table 2 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT LDA Cattle Supplies to Kismayu Meat Factory Year Number of Head 1969 3681 1970 11000 1971 38295 1972 48803 1973 27823 1974 50223 1975 1/ 34867 1976 2/ 10864 1977 33491 1978 Nil 1979 3/ 2100 1980 19300 1981 Nil 1/ January to June only 2/ June to December only 3/ November and December only Source: Development Department for Livestock Marketing Facilities of the National Range Agency (formerly the Development Department of the LDA). SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Intake and Production of Kismayo Meat Factory Unit 1971 1972 1973 1974 1975 1976 1977 1978. Purchase of cattle '000 head 38 35 28 50 46 41 34 3 Average Liveweight kg/head 255 242 246 216 244 242 225 NA Meat Output tonnes 4076 5560 3127 4766 4923 4455 3217 NA Tinned Products - stewed steak '000 tins 11922 16584 5733 9934 13363 10648 6636 NA - corned beef '000 tins 360 207 44 1695 1069 508 237 NA Meat in gelatine tins - - - 748 - - 28 NA Frozen meat tonnes - - 1291 - - - - NA Sausages tonnes 2.0 0.7 0.2 0.3 0.1 - - NA C Render fat/tallow tonnes 77 153 73 94 125 77 17 NA Hides tonnes 379 557 336 531 420 366 256 NA Meat and Bone Meal tonnes 74 90 45 107 139 109 60 NA Source: IBRD "Somalia: Agricultural Sector Review " Volume II,Annex 1, Table 20 - June 1981 (D > - 106 - ANNEX 8 Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Project Design Weaknesses 1. While the concept of the TJLP conformed with Government's policy for the livestock sub-sector and while the project area was well selected (containing, as it did, the greater proportion of the country's rich live- stock resource), the project fell seriously short of its objectives and its impact was negative in economic terms. Of tqe various factors which contributed to this situation, the key one underlying all others was, in the PCR Mission's view, the basically inappropriate project design. This appeared to be the result of unrealistic perception at both appraisal and review of what might be expected to be achieved in the course of a first stage project in an undeveloped subsector and in a difficult project area, particularly in the context of the country's scarce managerial, financial, technical and manpower resources. There were a number of design weaknesses, principal among which were: (i) Size, Scope and Complexity. The PCR Mission believes that too little attention was given at appraisal and review to the serious management problems of developing two 80,000 ha ranches, a 25,000 head (per annum) feedlot and a 1,200 ha irrigation farm, and of attempting to coordinate these with LDA's existing and projected marketing activities and with other anticipated project activities. No hindsight was needed to conclude that the chances of successfully developing such a program were small. It was a complicated and risky venture and great caution was called for in the circumstances. Unfortunately, neither the complications nor the risks were sufficiently emphasised at either appraisal or review. At that time very little detailed information was available on production and marketing systems in the project area and virtually nothing was known about range fodder species and the existing pastoralist system in which massive intervention was being proposed. There was virtually no precedent in Somalia for the ranching and range interventions perceived in the project and lack of such experience should have advised greater caution. Furthermore, irrigated fodder production was untried and feedlot fattening was only a concept in Somalia. Prudence would have suggested that all of these components be limited to research and pilot trials in a first phase project, since their implementation involved such high risks. This did not happen. The components failed to develop as anticipated, the project as a whole failed and, as a consequence, there is currently no clearcut indication of which course future subsector development should follow. (ii) Organization and Management. The organizational and managemental structure developed at appraisal and subsequently endorsed at review envisaged that LDA would assume ultimate responsibility for project development, with responsibility for day-to-day management and implementation being assumed by PDTU and its subsidiary, the PCMU. The Appraisal Mission recognised that LDA ..2.. -107- ANNEX 8 Page 2 was constrained by unsatisfactory management, weak financing and a paucity of qualified and experienced staff. It also recognised that the Agency had never managed a project of the type envisaged, but the Appraisal Mission considered that, with the strengthening of key management posts through appointment of specialist staff from an international firm, these problems would be addressed. Unfortunately, the Appraisal Mission, and the Review Mission in turn, overestimated the ability of GOS and, particularly, of LDA to provide the level of support and commitment necessary to assure efficient implementation. Under the organisation and management structure established, the effectiveness of the PDTU/PCMU was entirely dependent on the support and encouragement to be provided by LDA central management in Mogadishu. This, as it turned out, was never forthcoming. The management style of the two LDA General Managers in post for the project's duration was arbitrary and inconsistent (in a word, incompetent), and neither gave to PDTU the autonomy and decision-making authority it needed in view of the project area's location hundreds of kilometres to the south and in view of the poor communications which existed between LDA headquarters and the project. Following appointment of the management consultants in late 1974/early 1975, there was a brief period of compatability, but this passed quickly and relationships steadily deteriorated over 1975 and 1976. The management consultants and LDA never developed the smooth working relationship needed to implement a large and complex project like the TJLP. There were faults on both sides but it is clear from the project files that LDA never performed at an adequate level and never assumed the strategic role expected of it. As LDA went, so went PDTU/PCMU and the project. During 1975/76, the management consultants became polarized within themselves and serious differences developed between them and LDA and between them and other Government departments and the local authorities. The situation developed to a point where, in early 1977, LDA terminated the management consultants and made no move to replace them. The Appraisal and Review Missions' expectations with respect to LDA's ability to resolve its pre-project institutional constraints were at all times overly optimistic. This was particularly so of the Review Mission which had more hindsight than the Appraisal Mission and which, in early 1977 when it was mounted, had no real basis to assume from LDA's past performance that it could be expected to improve itself to a level adequate to promote the project. The Review Mission's highly conciliatory attitude to LDA, particularly to its General Manager, was no doubt interpreted by Government and LDA at the time as endorsement for past performance (which had in fact been highly unsatisfactory and should have been subjected to detailed treatment by the Review Mission). The Review Mission's attitude effectively precluded any change in organisation and management structure and LDA continued throughout the project's life, until the Agency's dissolution in March 1981 by Government, to be a major constraining influence on the project. ANNEX 8 -n 1Page 3 (iii) Implementation Schedule. Not only was GOS/LDA implementation capacity overestimated in the context of the size, scope and complexity of the TJLP, but the implementation timeframe decided at appraisal and confirmed at review was overar,-itious. Thus, in the project design it was envisaged that the project would reach full development by year 6. Completion times for some. individual project components were even shorter -- the IFFF was to be developed in two years and the two ranches in three years. Such time schedules would be difficult to achieve even in a developed country with a highly advanced livestock industry. They were impossible in Somalia in a first phase project in an isolated area and with LDA's known pre-project limitations. Such unrealistic projections were unwarranted in the present instance -- they were also dangerous as they implied that there were quick and easy solutions to the large and intractable problem of developing a backward subsector in a difficult project environment. The serious underestimation of component/ project completion times also introduced positive biases into financial and economic rate of return calculations and resulted in misleading estimates which raised false hopes and set the stage for disappointments on the parts of both the Government and the Bank. In retrospect, the interests of all parties would have been better served had the point been emphasised at appraisal that the subsector development program being embarked upon was difficult, complex, intractable and long term (i.e. 25-30 years). Had this been done, and had the more difficult aspects been restricted to pilot efforts in the first instance, it is likely that a firm base would be established by now from which more substantial development efforts could be mounted. As it is, with a costly failure before us caused by attempting to move too far too fast, the future direction is unclear. (iv) Site Selection for Project Components. There were several examples of poor site selection in the original and subsequent project design which resulted in implementation delays and which, in some cases, were instrumental in forcing abandonment of project components. Examples were the original site selection of the Afmadu holding ground which was made at appraisal. The area was in fact inhabited by local nomads and was an important water source for them and their cattle. The abandonment of the Afmadu site in favour of Aglibah on account of the nomads' objections, which were supported bythe regional authorities, resulted in at least an 18 months delay in holding ground development and serious inconvenience to the contractor which could have been grounds for legal action against the project had the contractor so chosen. Again, the siting of the ranches in areas of dense bush and of nomad and wildlife habitation was a serious design fault attributable, in the first place, to LDA and the management consultants and, subsequently, to Review Mission endorsement. Also, the incorrect siting of Ranch II headquarters in an area without a permanent water source and which was inaccessible during the rainy season compounded these misjudgements and resulted ultimately in the decision to abandon the ranch component entirely. It ANNEX 8 - 109 - Page 4 is pertinent to note in passing that, in the PCR Missions view, even without these problems, the ranching component would have fallen far short of its objectives. While both siting and size were serious problems, the shear magnitude of the task of developing the ranches in their isolated locations was equally challenging and could not have been realistically achieved in less than a 10-15 year timeframe, so that siting, size and timing were inextricably linked in negatively influencing development of the component. Similar complicating situations also existed as constraints on the IFFF. (v) Project Justification and Risk. Perhaps the most salient design weakness of the project was in the original perception of the benefits to be generated from the project and of the justifications for the project itself. These were based on: (i) increased offtake of animals from the pastoralist system; and (ii) improvements in weight and quality from the back- grounding/fattening operations on the ranches and feedlot. The interventions proposed were large, complex, costly and untried, and hence highly risky. Surprisingly, however, the risk element was lightly treated at appraisal and review and, during processing through the Bank, the high risk nature of the project was never raised at any level so far as the PCR Mission could determine. The economic rate of return calculation at appraisal was subjected to a quite inadequate sensitivity analysis which was limited to examining the impact of only two variations viz the effect of a 10 percent increase in input costs and of a 10 percent decrease in cattle selling prices. This was quite insufficient in the PCR Mission's view -- such a risky under- taking should have been subjected to a much more detailed analysis. Furthermore, in the Appraisal Report text, risk was treated in only the most perfunctory manner as follows: 1/ -The Project is not without risks, however, particularly in the operation of the irrigated fodder farm and feedlot, which constitutes a sophisticated investment for Somalia at its present stage of development. Finally, the country has an acute shortage of experienced staff, which may present difficulties in Project implementation and management inspite of the expatriate management proposed under the Project* If this, in fact, represented the extent of the Mission's appreciation of the project's risk, then it erred seriously both in its judgement and in failing to alert Government and the Bank to the true nature of the risks associated with 1/ Appraisal Report, para 7.07. -110- ANNEX 8 Page 5 implementing the project. These errors were, in turn, compounded by the frank overestimation at both appraisal and review of the technical and financial parameters on which the viability of the IFFF (the project's most complex and costly component) was based. These have been fully treated elsewhere (para 8.07, 8.08; Annex 6, paras 26, 27), but need reiteration in the present context because they were misjudgements which should have no place in a Bank project, given the usual exemplary standard at which the Bank's quality control safeguards and the institution as a whole function. - 111 - ANNEX 9 Page 1 SOMALIA TRANS JUBA LIVESTOCK PROJECT - CREDIT 462-SO PROJECT COMPLETION REPORT Adherence to Legal Covenants 1/ Performance of the Borrower and the implementing agency with regard to major covenants of the legal agreements was disappointing. Details were as follows: A. FINANCIAL COVENANTS P.A. Section 3.01: LDA shall maintain records adequate to reflect in accordance with consistently maintained appropriate accounting practices its operations and financial position. Actual: Not effected. P.A. Section 3.02: LDA shall (i) have its accounts and financial state- ments (balance sheets, statements of income and expenses and related statements) for each fiscal year audited, in accordance with sound auditing principles consistently applied, by independent auditors acceptable to the Association; (ii) furnish to the Association as soon as available, but in any case not later than six months after the end of each such year, (A) certified copies of its financial statements for such year as so audited and (B) the report of such audit by said auditors, of such scope and in such detail as the Association shall have reasonably requested; and (iii) furnish to the Association such other information concerning the accounts and financial statements of LDA and the audit thereof as the Association shall from time to time reasonably request. Actual: Not effected. D.C.A. Section 3.05: The Borrower shall, as and when needed, cause the Somali Development Bank to provide LDA with an overdraft not exceeding two-thirds of the value of livestock in the Project held by LDA. Actual: Not effected. The Ministry of Finance provided funds to the project through the LDA, but the project suffered constant shortages of working capital. 1/ Relevant covenants are contained in the Development Credit Agreement (DCA), Subsidiary Loan Agreement (SLA) and Project Agreement (PA). - 112 - ANNEX 9 Page 2 B. IMPLEMENTATION COVENANTS P.A. Section 2.01: LDA shall carry out Parts A, B, C and E of -he Project described in Schedule 2 to the Development Credit Agreement with due diligence and efficiency and in conformity with appropriate administrative financial and engineering practices. Actual: LDA performance was unsatisfactory. P.A. Section 2.06 (b): Except as the Association may otherwise agree, LDA shall cause all goods and services financed out of the proceeds of the Credit relent to it by the Borrower to be used exclusively for the Project until its completion. Actual: There was significant misapplication of project inputs. D.C.A. Section 3.01 (a): Without any limitation or restriction upon any of its other obligations urider the Development Credit Agreement, the Borrower shall cause LDA to perform in accordance with the provisions of the Project Agreement and the Subsidiary Loan Agreement all the obligations therein set forth, shall take and cause to be taken all action, including the provision of funds, facilities, services and other resources, necessary or appropriate to enable LDA to perform such obligations, and shall not take or permit to be taken any action which would prevent or interfere with such performance. Actual: LDA performance was never satisfactory and Government did not take the necessary actions to remedy the situation. P.A. Schedule 2,2: The Project Construction and Maintenance Unit shall carry out construction and maintenance of facilities provided under Parts A, B, D and E of the Project except for the major facilities for the Afmadu Holding Ground. The Unit shall be headed by a construction officer and staffed with two field construction officers, a surveyor, draftsmen, mechanics, drivers, and construction workers. Actual: PCMU never performed satisfactorily due to staffing shortages, lack of funds and inadequate direction. D.C.A. Section 3.02: The Borrower shall not later than one year from the date of this Agreement or such other date as may be acceptable to the Borrower and the Association, take all necessary measures to enact laws for regulating the use of Juba Waters, such laws to provide inter alia that water rights for irrigation shall be allocated on the basis of prior use. Actual: Not completed. A draft Water Use Law has been prepared (see Annex 2, Appendix 4 of Agricultural Sector Review of 1981) but not yet enacted into Law. - 113 - ANNEX 9 Page 3 D.C.A. Section 3.06: The Borrower shall within three years after this Agreement becomes effective or such other date acceptable to the Borrower and the Association expand its meat factory at Kismayu to handle the cattle production under the Project and shall take all such other measures to improve the veterinary standards and sanitary practices at said factory. Actual: Not undertaken. The KMF was closed by Government in December 1980 because of overall operational inefficiency. D.C.A. Section 3.07: The Borrower shall formulate a training program for the staff of its Ministry of Livestock, Forestry and Range, Kismayu Meat Factory and LDA and obtain the Association's agreement prior to implementation of said program. Actual: Implemented at only a token level. Footnote: Some of the above covenants were also reflected in the SLA. They have not been included in order to avoid repetition.  Nm E T HOPIA Belet Uen 1N ....---ur H 1 R A N - )ur Mandera Bulo Burti :7 K ENYA Adn oido4 araeu P P E JU B A BurAcoL-. Giohar C S or rilUane Uen Fafadun - Gaad QlorohoipAig 1o ,Audegie --MOGADWISHU Sc helabmcÅ../ SOMALIA 7 9 .---TRANS JUBA LIVESTOCK Abasale PROJECT - CEDIT 462-s0 - PROJECT COMPLETION REPORT Du unBrava ~aat Project Area Bets ogari 3 - --- G rai- - --0Itraco onae LO ý W 'R iJE3 A- iabd a Re giona! enda LJiboj uto H stock R,,,ts H v e jOther Roý.ds 1acks 1< F Project Arzz g d< KENYA  NAP 2 E T H l O P lA .×3 Mandero lM A N D E R A- Lugh Ferrandi E L W AK N 0 R T H E R N oSerdale 2 0 N E oBoidoa P L A IN Ei Wako + Busor Bordera oDinsor oFofadun t F AF ADU N oC CE N T A L + Z0 NE P L A l N Dujuma-- Dif - z Maskoti MAÅ R i N E-- orisei Wcrshan - _ -_-_- + LAC DERA Afmandu - - - - E__ w Deles Gogonio / Geli __ o0 oM ido-- - - -- - i oTobdo - i -_1__ S 0 UT H EkR N Giomomo - Z' 0 Z NE - - oC E A N + P L A l NC- onto - I DESCEK UA..A SOMALIA Kismayu TRANS JUBA LIVESTOCK PROJECT - - CREDIT 462-SO -- PROJECT COMIPLETION REPORT t D U D U MAÅ L -l oBododda - Zonal Boundaries in Trans Juba Kol -2 - Kobo P L Å I N - os Karboni -c  MA P 3 E T H0PIA Khadija S 0 MALIA b(. Garsale E usar U PP E R J U B A 6BARDERA Fafadun -v- Ooroha' cl ------ andjofaan [P) Pmadu SOMALIA wk TRANS JUBA LIVESTOCE PROJECT - CREDIT 462-SO \ PROJECT COMPLETION REPORT Aglb . • Rarch \('P) /Stockroute and Water ( c Supply Development () (c)\ <p Aj(W r Hosweyne smyu oBuloHaj *:Kis rayu 0 Goba L EGE ND Ground Stock Routes completed WdHnPonds (uars) comp,eted V/a daj r p)X? Productivc borcoles Work done by contract .'ork done by force account K lL(ME TE S  二 とOEヤ0[0と91  E T H 1 0 P I A Fe,Fe, G A L G A D U D FRICA . .-.-- A F R I C A derr 3 Bi n, oi n Oddur . CENT RAL Aud,ne Bad coa . ~ ,/ ~ AtqoSH B ABER LEMLE G E D O B u A c b o UL e U e n G io h o r- B3usaur BAA BA Fofadun D,sr aad Barderai BENADIR Igo% ° 9 Audegie MOGADISHU MILES ?5AOSchelambot ____?O__________2?0____________300________ boe Golu Meca KILOMETERS - C E N T R ýA L N. J U B A O \ SOMAL_IA z FANOLE BARRAGE Moeskafier o u TRAN-JUBA LIVESTOCK PROJECT v\ Wa,shan li > b PROJECT AREA Tabdo PROPOSED PROJECT FEATURES EXISTING FEATURES Project Areo E.E.C. Holding Ground L Grazing Ronches (j5 Kismayu Holding Ground oswego Stock Routes * Market Centers J U B A (D Kismayu Livestock Markets e Boreholes, Wells or Dams Bu dod Veterinary Centers Metalled Roods G.ba Sites Grazing Ranches Other Roads, Tracks and Troils Wadajir Afmodu Holding Ground 1' Main Ports Irrigated Fodder Farm and Feedlot New International Airport Bdaddo DistYict VeteTinory Centers 9-- Regional Bundaries International Boundaries >34