Mexico - Rio Colorado Irrigation Project

Document of The World Bank FOR OFFICIAL USE ONLY FILE COPY Report No. 2591 PROJECT PERFORMANCE AUDIT REPORT MEXICO RIO COLORADO IRRIGATION PROJECT (Loan 527-ME) June 29, 1979 Operations Evaluation Department 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.  FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT MEXICO RIO COLORADO IRRIGATION PROJECT (Loan 527-ME) Table of Contents Page No. Preface i Basic Data Sheet 11 Disbursement Table 111 Highlights iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. Project Summary Project Description 1 Targets and Goals 3 Sector Significance 4 Project Achievements 4 II. Main Issues Inadequacy of Bank Loan Amount 5 Procurement Procedures 7 Rate of Return 7 III. Conclusions 9 Appendix: Evaluation of Results 11 Tables: 4.2.1/2/3; 4.2.3; 4.3.4; 4.4.1/2/3/4/5 17 PROJECT COMPLETION REPORT I. Background 26 II. Formulation 32 III. Implementation 40 IV. Agricultural Impact 68 V. Rates of Return 79 VI. Institutional Performance and Development 81 VII. Bank Performance 88 VIII. Conclusions and Recommendations 90 Annex I CTables 1 - 2) 94 Annex II (Tables 1 - 2) 98 Tablee 1 - 25 101 Charts 1 - 3 126 Map 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 MEXICO RIO COLORADO IRRIGATION PROJECT (Loan 527-ME) PREFACE This report presents the results of the audit of the Rio Colorado Irrigation Project in Mexico, the fourth in a series of seven Mexican irrigation projects assisted by Bank loans during the period 1961-1975. Loan 527-ME, for US$25 million was signed in January 1968 and closed in June 1977, fully disbursed. The audit is based on interviews with Bank staff and on a reading of the appraisal report, relevant legal documents, supervision reports, project files, the project completion report (PCR) prepared by the Bank's Latin American and Caribbean Regional Office in August 1978 (copy attached), and a report entitled "Evaluation of Results of the Rehabilitation of Irrigation District No. 14, Rio Colorado, B.C.N. and Son.", prepared by the Sub-directorate of Evaluation of the Mexican Secretariat of Agriculture and Hydraulic Resources (SARH) in April 1978 (translation of concluding chapter attached as appendix). An OED mission visited Mexico in December 1978. The mission held discussions with the staff and consultants of SARH; Nacional Financiera S.A. (NAFINSA); Banco Nacional de Credito Rural (BNCR), and project par- ticipants. The information obtained during that mission was used to test the validity and completeness of the conclusions of the PCR and the Borrower's evaluation report on which it is based. The audit finds the PCR unusually detailed, comprehensive and accurate with respect to the project's achievements and shortcomings. Some minor differences in emphasis and interpretation are needed to reflect the Borrower's views and the findings of the audit. These differences are described in the Project Performance Audit Memorandum included in this report. The draft report was sent to the Borrower for comments; no comments have been received. The valuable assistance provided by the staff and consultants of SARH, NAFINSA, and BNCR and by participating farmers is gratefully acknowledged.  - 1i - PROJECT PERFORMANCE AUDIT REPORT BASIC DATA SHEET MEXICO: RIO COLORADO IRRIGATION PROJECT (Loan 527-ME) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$million) 95.0 262.0 /1 Overrun () - 176 Loan Amount (US$million) - 25.0 Disbursed as of December 31, 1977 25.0 25.0 Date Physical Components Completion 12/74 12/85 Proportion Completed (%) 86 Proportion of Time Overrun M%) (not yet completed) Economic Rate of Return (%) 11 15 /2 OTHER PROJECT DATA Original Actual Item Plan Revisions Current Estimate First Mention in File - 09/66 Government's Application - - 02/67 Negotiations 11/67 - 12/67 Board Approval - - 01/25/68 Loan Agreement Date - - 01/26/68 Effectiveness Date 08/01/68 09/01/68 02/20/69 Last Disbursement 01/--/77 Closing Date 06/30/75 06/30/76-12/31/76 06/--/77 Borrower Government of Mexico Executing Agency Secretariat of Hydraulic Resources (SRH) Fiscal Year of Borrower January 1 - December 31 MISSION DATA Month/ No. of No. of Date of Item Year Weeks Persons Staffweeks Report Appraisal 02-03/67 6 3 n.a. 12/21/67 Supervision I 08/68 2 1.0 n.a. 08/21/68 Supervision II 08/69 1 1.0 n.a. 10/13/69 Supervision III 03/70 2 1.0 n.a. 04/30/70 Supervision IV 03/71 3 1.0 n.a. 05/21/71 Supervision V 07/72 2 1 0 n.a. 09/15/72 Supervision VI 02/73 2 1.0 9.0. 03/20/73 Supervision VII 02/74 2 1.6 3.0. 03/14/74 Supervision VIII 02/75 3 1.6 9.3. 04/07/75 Supervision IX 05-06/76 /3 3 1.0 9.1 08/19/76 Supervision X 02-03/77 7- 1 0.2 3.4 03/31/77 Supervision XI 12/77 /4 1 0.2 0.4 01/05/78 Completion 04/78 3 2.0 32.7 01/05/78 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) Pesos (Ps.) Year: Appraisal Year Average Exchange Rate: US$1 = 12.500 Ps. Intervening Years Average US$1 = 12.825 Ps. Completion Year Average US$1 = 22.573 Ps. /1 Total cost was reassessed to be successively 98/129/161/220/250/262 million dollars. /2 Should the revised total project not be fully invested by 1985, the IRR would then be reduced to about 14%. /3 This mission also supervised Loan 970-ME. /4 This mission also supervised Loans 969-ME and 1111-ME.  MEXICO RIO COLORADO IRRIGATION PROJECT (Loan 527-ME) Disbursement Table US$millions cumulative Appraisal Actual Actual as % Year ending Estimate Disbursement of Estimated 12/31/1968 4.4 .0 0 12/31/1969 8.2 0.08 1 12/31/1970 12.6 2.14 16 12/31/1971 15.9 6.91 43 12/31/1972 19.2 11.32 59 12/31/1973 22.3 17.40 78 12/31/1974 25.0 19.50 78 12/31/1975 - 23.23 - 12/31/1976 24.84 12/31/1977 25.00  - iv - PROJECT PERFORMANCE AUDIT REPORT MEXICO RIO COLORADO IRRIGATION PROJECT (Loan 527-ME) HIGHLIGHTS The Bank loan for the Rio Colorado Irrigation Project in Mexico was signed in January 1968. The project was intended to support the rehabilitation of the irrigation and drainage works and related facilities over an area of about 250,000 ha in the Rio Colorado irrigation district. The planted area was to be increased by 40,000 ha, the diver- sification of crops was to be introduced and crop yields increased by improved water application. The project was the fourth of a series of eight Mexican irrig- ation projects assisted by Bank loans since 1961. In common with most of these projects, the Rio Colorado Project has experienced substantial cost overruns and delays in implementation; nonetheless, the project is regarded by the Mexican authorities as one of the most successful in their irrigation rehabilitation program. The project has facilitated the rapid integration, commercialization and industrialization of the ejidos and other rural institutions in the area, creating important social benefits for the project participants. The reestimated economic rate of return is around 15% compared to 11% at appraisal. Other points of special interest are: - because of the scarcity of Bank loan funds, the loan amount was too small, contributing to delays in project implementation (PPAM paras. 13-17; PCR paras. 2.08 and 3.49); - Bank procurement requirements conflicted with the Govern- ment's existing policies and procedures, creating delays and excessive work loads at both ends (PPAM paras. 18, 19; PCR paras. 3.34-3.39).  PROJECT PERFORMANCE AUDIT MEMORANDUM MEXICO RIO COLORADO IRRIGATION PROJECT (Loan 527-ME) I. PROJECT SUMMARY 1/ Project Description 1. Schedule 1 to the Guarantee Agreement describes the project as follows: (a) The project consists of the rehabilitation, improvement and operation of the irrigation and drainage works and related facilities serving a gross area of about 250,000 2/ ha in the Rio Colorado Irrigation District. (b) The main works included in the project are: (i) reconstruction and lining with concrete of about 2,330 kilometers of canals, and construction of necessary canal structures and gauging stations; (ii) improvement of about 1,180 kilometers of existing drains and construction of about 335 kilometers of new drains, together with necessary drain structures; (iii) rehabilitation of about 485 existing deep wells, drilling and equipping of about 105 new deep wells, and drilling of about 220 observation wells; (iv) construction or improvement of about 555 kilometers of roads; (v) land leveling on up to 178,000 ha; and (vi) construction of telephone lines, office and repair shop buildings, and houses for ditch-riders. (c) The project includes the purchase, installation and utiliz- ation of equipment and vehicles for repair and maintenance operations. 1/ Adapted from PCR. 2/ The cropped area would increase from 176,000 ha to 203,000 ha, with about 12,700 ha double cropped, bringing the total annual cropped area to 215,700 ha. - 2 - (d) The project also includes the establishment and operation in the Rio Colorado Irrigation District of a continuing program of hydrogeological and cropping pattern studies, investigation of on-farm drainage needs and soil-testing in order to determine the adequacy of the underground water supplies, identify the cropping patterns which would make the most productive and advantageous use of available water supply, and detect salinity or alkalinity problems developing during the operation of the project. 2. Other project programs or actions, such as agricultural research and extension services, seed inspections and pest control, irrigation tech- nical assistance to farmers, and agricultural credit were to be undertaken by the Secretariat of Hydraulic Resources (SRH) and the Secretariat of Agriculture and Livestock (SAG). At appraisal, it was recommended that assurances that such programs or actions that would be undertaken should be obtained from the Mexican Government at negotiations. Commitments for carrying them out were reached through the side letters. These programs at the time of their inception were perhaps difficult to quantify and, therefore, were only lightly described and no implementation schedule or cost estimates were worked out for them at appraisal except for the amount available for credit. 3. The project cost estimates, including only civil works, relocation of farmers and equipment, was US$95 million. The investment period would be seven years, from January 1968 to June 1974. In addition, farmers were expected to invest about US$12 million for on-farm drainage over a period of about 15 years. The cost increase that some of the programs described before would impose on the SRH budget was apparently included in the O&M cost, while the cost of extension services to be provided by SAG was not considered at all. 1/ 4. Contracts for civil works were to be awarded on the basis of international competitive bidding in accordance with the Bank's guidelines. International competitive bidding did not have to be called, however, for works to be undertaken by the Government's own forces or by local contractors with an aggregated amount equivalent to 20% of all construction cost to be financed by the Loan. 2/ 1/ It was also omitted by the appraisal mission when estimating the rate of return in the investment. 2/ Side letter 1/26/78. -3- Targets and Goals 5. The basic objective of the project was to maintain the District's agricultural productivity through prevention of water losses and increasing soil salinity. It proposed the diversification of crops, intruducing barley and oats or forage crops that would permit full utilization of winter water which would otherwise not be utilized beneficially, and the gradual reduction of the cotton area from 127,500 ha to about 100,000 ha to eliminate areas that produced low yields, which was inefficient use of water. Forage was in short supply in Mexicali, and livestock farmers have to import part of their feed from the United States. The expansion of forage production would meet current demands and, at the same time, enable expansion of livestock in the area. The livestock industry was expected to benefit also from the increased production of grain sorghum, especially in the conditioning of beef cattle for export and in the growing dairy industry. The diversification program was expected to be achieved through water control which SRH exercised through the District's Advisory Committee and medium- and long-term credit availability. In addition, increases in crop yields would result from improved water application and from relocation of farmers to better suited areas. The project intended as well to avoid adding unemployment of both manpower and existing capital facilities in the region. 6. The planted area was to be increased by 40,000 ha (from 176,000 ha to 215,000 ha), the production tonnage by about 570,000 tons (from 540,000 tons to 1,111,000 tons) and the gross value by Mex$180 million 1/ (from Mex$952 million to Mex$1,132 million). However, the projected alternative uses for the land (para. 5) would, on the whole, cause a reduction of the average productivity for the extended area, from Mex$5,409/ha to Mex$5,248/ha.2/ Net farm income after rehabilitation was to be increased, on average, by about Mex$9,000 ha. 3/ 7. Intermediate appraisal targets were 91% 4/ efficiency on conveyance and distribution of water and 80% on farm water applications. 1/ Current prices of 1967. 2/ Appraisal Report, Annex 8.: Productivity Area (ha) Tons ('000) GPV (Mex$'000) Mex$/ha Mex$/ton Present 176,000 540 951,607 5,409 1,763 Future 215,700 1,111 1,132,040 5,248 1,019 3/ The Appraisal Report does not provide figures to support this statement. 4/ 87% for the gravity water and 96% for well water. These targets were too optimistic, however, and will not be reached, even with completion of the proposed works and with excellent handling of water within the farms. If such goals were to be achieved, drastic changes in operation of the systems and schedules for irrigation water delivery would be necessary.,_.together with the introduction of mechanical methods of irrigation (sprinklers, dripping), which had not been proposed at appraisal (PCR para. 3.16 ). 8. Control of the soil salinity and of the water table were two other intermediate but basic targets. Although these targets could not be quantified, it was assumed that they would be reached during the project implementation period. Soil salinity control, however, depends largely on the availability of adequate amounts of water with low salt content, on the efficiency of drainage and on the efficiency of the leaching technique. This is a long process and may take many years. It should be mentioned that the quality improvement of the Colorado River water (PCR para. 1.14) was not foreseen at appraisal. The water table control, is also a trial and error procedure that could take many years to achieve, perhaps exceeding the implementation period. Sector Significance 9. Although the Rio Colorado is a large irigation project by any standard, it covers only about 4% of the irrigated land in Mexico and about 1.5% of the total cultivated land. At the regional level -- Northwest region -- the Mexicali and San Luis Rio Colorado valleys cons- titute about 11% of the cultivated area and produce 7% of the grains, 5% of the oilseeds and 16% of the cotton. Project Achievements 10. In brief, the PCR mission estimated conservatively that the ex-post internal rate of return (IRR) is 14.6% (at appraisal, it was estimated to be 11%), although there is still some degree of uncertainty since works are not fully completed (72% as of December 1977). Project goals for both cropped area and production are nearly met (as of December 1978, the cropped area was 98.5% of that expected at full development while production was 79%), and the soil-salinity hazard has been adequately overcome. To fully meet project objectives, it is essential that remaining works be completed and that current studies and programs to control salinity and water pumpage be continued. The project cost rose from US$95 million at appraisal to US$262 million, or an increase of 176%. As of December 31, 1977, about 76% of the revised total project cost had been invested; it is expected that the rest will be invested by 1985. Should this investment not take place by then, the IRR would be reduced to about 14.0% rather than 14.6%. - 5 - 11. With respect to Bank performance, despite the adequate ex- post internal rate of return, the mission judged that if certain delays could have been avoided, due mainly to inadequate feasibility studies and insufficient budgetary allocations, the IRR would have been substantially higher. At the time of appraisal, these deficiencies were not adequately recognized by the Bank. A situation of over-confidence on the part of the Bank had developed over the years due in large part to the fact that SRH had had a record as a strong organization with capabilities generally far better than in most countries. 12. The delays suffered by the Rio Colorado project were similar to those appearing concurrently in other irrigation projects. Since early 1975, the Bank and SRH (as well as SARH, the ministry which in early 1977 was formed through amalgamation of SRH and SAG) have con- ducted an active dialogue on planning and implementation of irrigation projects in order to minimize delays resulting from inadequate feasibility studies and institutional weaknesses. The fact that the Division in the Bank dealing with Mexican irrigation projects deals as well with other agricultural projects -- so as to have a more integrated approach to the entire agricultural sector in Mexico -- has helped in conducting these dialogues and in attempting to overcome the inadequacies and weaknesses mentioned. II. MAIN ISSUES Inadequacy of Bank Loan Amount 13. The PCR concludes (para. 8.03) that the project's rate of return would have been much higher if delays in implementation could have been avoided. The PCR attributes these delays mainly to the inadequacy of the feasibility study and implementation schedule, and insufficient budgetary allocations by the Borrower. The audit agrees with this finding and with the emphasis in the Borrower's evaluation report on the financial (vs technical) constraints. 1/ One important factor which contributed to the financial stringency surrounding the project's execution, and therefore to its delay, however, was the inadequate amount of Bank financial assistance, which amounted in the end to only about 9% of actual project cost (PCR para. 3.49). 1/ See Appendix, Sec. 4.2 "Comments on Project Execution". - 6 - 14. At the time the project was appraised in 1967, Bank funds were scarce, and loan amounts for large, active borrowing countries like Mexico, were subject to a form of rationing at the country level which was translated into a pressure on the projects staff to scrutinize with extra care the Borrower's estimates of project foreign exchange cost, in addition to a number of management decisions simply to postpone certain loan commitments, particularly for countries with relatively stronger economies. Accordingly, the head of NAFINSA was consulted by the Bank in early 1967 as to whether the Mexican Government wanted to proceed with the appraisal of the project in spite of the risk that a smaller, or even no loan might be possible. The appraisal mission was instructed to keep in mind a loan amount of US$35 million as a maximum, but to discuss no figure with the Mexican Government, nor to make any promises as to when the loan might be made. 15. After their return to headquarters, the appraisal mission estimated the foreign exchange cost at US$43 million (40.6% of the US$106 million total cost), as compared with the original Mexican request of US$33.1 million. Most of the increase resulted from higher contingencies (20% of total cost vs. 4.5%), and the inclusion of interest during construction. Six months later, the mission had whittled the foreign exchange component down to only US$25 million (out of a revised total cost estimate of US$99 million). The Loan Committee agreed to proceed on that basis. 16. The Mexican delegation came to negotiate the loan armed with their own revised cost analysis which indicated a foreign exchange component of US$39.2 million rather than the Bank's estimate of US$25 million (PCR para. 2.08). After a detailed revision of SRH's' analysis, the Bank agreed to increase the estimate for six types of work by US$3.6 million. The Mexicans, on the other hand, agreed to finance all cement, concrete additives, reinforcing steel and locally-procured equipment with their own funds to "facilitate procurement of these items". While this obviously helped to reduce the amount of the loan, it should not have affected the estimated foreign exchange cost of the project. Nonetheless, the projects staff incorrectly reduced the estimate in the appraisal report to reflect the switch in source of finance for these items. The net effect of these changes was a foreign exchange cost estimate of US$26.2 million. Since the Loan Committee had only authorized a loan of US$25 million, SRH agreed to reduce contingencies by US$1.2 million, from 20% to 15%, to bring the estimated foreign exchange cost down to the predetermined loan amount. - 7 - 17. In retrospect, it seems unfortunate that the Bank exerted so much pressure during appraisal and negotiations to reduce the amount of the loan, albeit for what may have seemed to be valid reasons at the time, with the result that the burden imposed on the Government's budget by the project was much greater than anticipated, contributing to the delay in implementation. Procurement Procedures 18. In this project, as in the other Mexican irrigation projects, the Bank's procurement guidelines were superimposed on existing Mexican Government procurement policies and procedures. This resulted in the preparation and submission to the Bank for approval of voluminous, detailed international tender documents covering 31 separate civil works contracts, all of which were awarded in any case to local contractors. In spite of the fact that, in Mexico, Bank ICB procedures have never resulted in a civil works contract being awarded to a foreign bidder, they have been followed faithfully in all of the seven Mexican irrigation projects, and the delays and administrative burden they created added to, rather than substituting for, the delays and burden already imposed by the Government's own cumber- some procedures, which require civil works to be broken up into US$2-3 million lots to enable a large number of small local contractors to participate (PCR para. 3.36). 19. SARH and the Bank's projects staff are currently discussing ways of streamlining procurement procedures under Bank-assisted irrig- ation projects in an effort to reduce paperwork and delays to the minimum needed to ensure least-cost results and free competition among bidders, both local and foreign. This dialogue is opportune, but should have been initiated much earlier. Rate of Return 20. There is a substantial difference between the 20.5% rate of return calculated in the Borrower's evaluation report (Appendix, section 4.4, "Indicators for Project Evaluation") and the 14.6% in the Bank's PCR (para. 5.02). This difference is due to the following factors: (a) The authors of the PCR chose different base year to determine the net value of production without the project (1972 vs 1971). As a result, their assumed project benefit stream begins at a significantly lower level, and one year later, than that assumed by the Borrower; - 8 - (b) The PCR assumes a more gradual buildup of benefits in the future than does the Borrower; on the other hand, the Borrower's evaluation does not include livestock benefits; (c) The PCR assumes further Government investment in the project up to 1985 at an average annual rate of US$7.7 million, whereas the Borrower's evaluation assumes no further investment after 1977; 1/ (d) The PCR assumes on-farm investment in improved drainage and irrigation over a 20-year period, 1985-2004; the Borrower's report assumes none; and (e) The PCR's estimate of increased cost of operation, maintenance and extension due to the project is 65% higher than that of the Borrower, again due to a difference in base years used to define the without project case. 21. The audit believes the PCR's assumptions are generally more realistic than those reflected in the Borrower's evaluation report, particularly with respect to the time curve of future benefits and the further investment needed to realize them. The selection of base period to determine the without-project production value has a somewhat more rational basis, too, than that used in the Borrower's report, which serves to maximize the project's assumed benefits in the crucial years. The Borrower's estimate of without-project recurrent cost, however, is based on the 1968-1971 average and appears more realistic than the corresponding figure in the PCR, which is based on one year. The project's actual rate of return can, of course, only be measured at a later date; in the meantime, the audit supports the PCR's estimate of roughly 15%, which is satisfactory and in line with other successful irrigation rehabilitation projects in Mexico. 2/ 1/ Expenditure and budget data given to the audit mission by SARH, however, indicate Government investment in land levelling, drilling and interconnection of wells and resettlement of farmers continuing beyond 1977 and into the future. 2/ See for example OED's recent Impact Evaluation of the Third Irrigation Project (Loan 450-ME), Report No. 4348, dated June 18, 1979. - 9 - 22. The differences in approach to the rate of return calculation between the PCR and SARH's evaluation reflect a divergence, albeit minor, in the way the Bank and SARH regard the status of the project: While the PCR stresses the importance of making further, water-saving investments to increase the project's overall efficiency and eliminate overpumping, SARH tends to regard the project more in terms of the area benefitted, which is already very close to the appraisal target, and thus to think of the project as already being completed for practical purposes. 23. Neither calculation adequately reflects the project's important indirect benefits, which are immediately evident during a visit to Mexicali, an important center of economic growth which is virtually dependent on the project for its rising prosperity. The project has also facilitated the rapid integration, commercialization and industrialization of the ejidos and other rural institutions in the area, creating important social benefits for the project participants. III. CONCLUSIONS 24. The Rio Colorado Irrigation Project, in spite of the delays in its implementation, is regarded by the Mexican authorities as one of the most successful projects in their irrigation rehabilitation program. They argue that the rapidly deteriorating situation in the district prior to appraisal necessitated hasty action in the preparation, appraisal and negotiation of the loan, "in the full knowledge that, in the event, the proposed actions could be improved". 1/ The audit agrees with this view and finds that, in spite of the inadequacies of the feasibility study, acknowledged by both the Bank and the Borrower, the overall design of the project was appropriate and its timing opportune. 25. The Bank's appraisal, as was the fashion at the time, did not adequately reflect the many uncertainties surrounding the project's costs, benefits and rate of implementation. In particular, it did not foresee the favorable impact of the outcome of negotiations between Mexico and the USA regarding salinity of the Rio Colorado waters or the unfavorable impact of underestimation of costs and delays of construction and overestimation of projected future irrigation efficiency and SRH's technical and administrative capacity. These latter errors were common to some of the other appraisals of Mexican irrigation projects as well. 2/ The more favorable economic return now expected for the project, 1/ Appendix, section 4.1, "Comments on the Feasibility Study ". 2/ For example, see the PPAR on the Mexico Third Irrigation Project (Loan 450-ME), Report No. 1067, dated March 18, 1976. - 10 - however, could probably not have been entirely foreseen at the time of appraisal either, since it depends partly on more recent information about the recharge rate of the underground aquifer. 26. In conclusion, the project was quite successful. The Bank's intervention may have contributed to that success by inducing the Mexican authorities to accord somewhat more priority to the project in the provision of budget resources and technical assistance than they might have done if there had been no external agency involved, but it is impossible to be certain about this point. Otherwise, the success of the project is due to the efforts of SRH and the Mexican Government. - 11 - APPENDIX Page 1 of 6 pages EVALUATION REPORT PREPARED BY SARH, APRIL 1978 Chapter IV EVALUATION OF RESULTS 4.1 Comments on the feasibility study Prevailing conditions sharply curtailed the time available for carrying out the feasibility study and, very probably, largely determined its approach. According to knowledgeable sources, the study was completed hastily because of the pressing need for a technical response that would alleviate the problems faced by farming in the Mexicali Valley at that time, in the full knowledge that in the event the proposed actions could be improved. Given the limitations of the soil and the amount of water available, it may be thought that too large an area was considered for the project. However, the existing social climate would have certainly made it difficult to reduce the proposal further. Moreover, a move in that direction could have been prejudicial to the diplomatic negotiations underway regarding the salinity of the Rio Colorado. In addition, the project is soundly structured and the proposed solutions are realistic in view of the situation, which was approximately identified. Even though a series of changes and adjustments -- necessary in any project -- were made later, they did not as a rule represent sweeping changes, as will be seen below. This suggests that the feasibility study was reasonably supported by basic studies, with the possible exception of soil studies. In point of fact, the definitive location of the tracts in the District was somewhat significantly changed in relation to the feasibility study. According to the technical personnel who were following project execution, the change had to be made because of the impossibility of obtaining high quality land acceptable for beneficiary relocation. However, there is also the possibility that, between the soil studies and project execution, soil characteristics changed because of deterioration of the water used to irrigate them. - 12 - Page 2 of 6 paw-Q 4.2 Comments on project execution The construction program reflected numerous changes with respect to the feasibility study, owing to the greater detail of designs plus additional needs and specific situations detected during its execution. Table 4.2.1. shows the proposals made in the feasibility study, some changes considered, and the actual results. Among the principal categories the major changes fall under rehabilitation and construction of canals, whose final length was only 85% of that originally planned. Although the number of canal structures did not change too much in absolute terms, it should be noted that in terms of canal-kilometers they increased 30% over the feasibility study. The beneficiaries exerted a good deal of pressure, often successfully, to obtain more facilities, as the figures show. Moreover, the drainage network involved is the same as that originally considered. One of the items that were expected to change significantly during construction is land levelling. The feasibility study proposed the levelling of 178,638 ha; in the first change proposed, the figure became 138,150 ha, and in the second 160,000. In the event, by 1977, 191,815 ha had been levelled, including some 20,000 regraded ha, which is not too far from the original figure. Local opinion has it that this work was affected by the beneficiaries' demands based on the specific benefits they derived from them. Finally, the significant lengthening of the construction period, which was originally to last from 1968 to 1972, and which actually took nine years, was reportedly due to financial problems more than to technical causes, which led to low budgetary allocations being earmarked for the project. Capital expenditure was estimated by the Evaluation Officer, on the basis of information available at the time, at Mex$2,050 million in 1966 prices. In other words, the actual budget was 94% higher than the original estimate. It should be noted that the category of indeminifications which was included under expenditures and thus not considered at the feasibility stage, affects only a very small portion of the cost overrun. In 1977 prices, expenditures totaled Mex$6,000 million. - 13 - Page 3 of 6 pages By categories, the estimates, which are subject to a greater margin of error because the investment programs were not broken down into as great detail as they should have been, indicate that the major headings where the greatest discrepancies were noted are canal structures (227% cost overrun), drainage network (167% overrun) and land levelling (where 109% above the amount shown in the feasibility study was spent). Tables 4.2.2. and 4.2.3. show some of the results of the estimates. With regard to support programs for farming and stockraising, efforts were generally fruitful and had a significant impact on the development of those activities. In particular, regarding the contributions of technical assistance and agricultural research, the relative level of development attained for some time by local farmers may suggest that gains in agriculture were not necessarily contingent on such services. However, in view of the water supply problem that the District faced in the 1970s and the resulting deteri- oration of soils, as shown by the data for that period, the most urgent need was to adapt farming techniques to the new situation, which, without technical assistance, may have been extremely difficult to achieve in a relatively short time. Moreover, the impact of beneficiary resettlement and of the District's operation and conservation programs is important from the stand- point of the efficiency they represent for water management, thereby facil- itating the working of larger areas in the District. In addition, those programs made it possible for farming to be done on better soil, which has certainly had a major impact on yields. 4.3 Some effects of the project With regard to the creation of jobs, thus far farming and stock- raising in the Valley have experienced a decrease of approximately 6% in the number of day workers demanded between 1970 and 1977. Since,this estimate does not take into account changes in the labor intensity of field work, it is likely that the decrease was due to diversification of activities. In the long run, it is hoped that the increases in the area will offset that decrease and it is estimated that demand for manpower will be about 10% higher than the 1970 figure under the same agricultural technology hypotheses.. With regard to the labor force engaged in farming and stockraising in the area in 1970, none of the estimated demands -- up to 1970, 1977 and at full development -- would be sufficient to provide full employment in the region, except in Septemberwhen cotton is harvested. Table 4.3.4. gives the relevant figures. - 14 - Page 4 of 6 pages In addition to jobs in farming, stockraising has also required rural labor. Although the numbers involved are not large, the upward trend is due in good measure to the project. With regard to the effect of the project on farmer income, given the higher yields and the introduction of high-yield crops, as well as the increase in areas harvested, overall profits in 1977 prices shot up from some Mex$730 million to almost Mex$2,860 million between 1971 (which is taken as representative of the situation without the project) and 1977. Over the long run, also in 1977 prices, the benefits are expected to reach Mex$2,650 million annually. In view of the slight variation in the number of beneficiaries, average profits have risen similarly. Lastly, in the absence of more precise data, according to local opinion project benefits are rather evenly distributed, given the relative homogeneity of plot size. With regard to the revenue that execution of the project has brought to the Government in the form of taxes, as previously mentioned, percentages were virtually unchanged, but a rough estimate based on prices and tax legislation in effect in 1977 shows a notable increase, from Mex$40 million in 1971 to Mex$59 million in 1977. 4.4 Indicators for project evaluation The economic evaluation of this project is based on the application of cost/benefit analysis, measuring the contribution made thus far by the project to national value added and that expected in the future. The calculations are expressed in 1977 prices. As pointed out in Chapter I, without the project, the long- term forecasts predicted a stationary situation from 1972 onward, at figures for the 1970/71 season. For the structure of production from 1968, when the project began, to that date, historical data were used since prior to 1971 the project had virtually had no impact on activities. In addition, as previously noted, the system of production expected after full development was in line with the figures given in Table 3.4.1 of the preceding chapter. For activities at full development, data for District production were used up to 1977. From 1978 onward, an activities model similar to that given as the long-term target (although involving smaller areas and yields than those targeted) was used. -15- Page 5 of 6 pages The crop yields for 1978 were then meshed with those that had been obtained in recent. years in the District. Producer prices included in the analysis were, without exceptioin, those prevailing in the area in 1977. Table 4.4.1 shows expected yields for 1978, product prices and production costs for each activity. Those costs were assumed to be constant throughout the evaluation period, both with and without the project, having been calculated on the basis of the figures in Table 3.3.2, less taxes, labor and machinery depreciation. In point of fact, taxes are considered to be mere transfers and labor to be abundant, as indicated above. Machinery depreciation was deducted because it was felt that the same costs would be incurred with and without the project. No other types of benefits were included -- such as those that could be derived from stockraising -- because it was not known to what extent they were dependent on the project, and because there was a lack of adequate information on the operations themselves. Typically, without the project, costs included maintenance, operating costs and extension. For the period from 1968 to 1971 the figures were compared with actual expenditures in 1977 prices. After that time the average for each period was used. Table 4.4.2 gives the figures. The costs attributable to the project are made up of capital expenditure, (from which indemnifications were deducted) operating, maintenance and agricultural extension costs. apital expenditures were deducted from Table 4.2.2. Operating, maintenance and agricultural extension costs include actual expenditures between 1968 and 1977, shown in 1977 prices in Tables 4.4.3 and 4.4.4; beginning in 1968 the same costs as for 1977 were used. The indicators of the cost/benefit analysis were then calcu- lated for three different cases, internally known as optimistic, average and pessimistic, depending on certain parameters used in the analysis. The factors that were varied were: (i) Total harvestable areas in 1978; it was assumed that 205,000, 200,000 and 195,000 ha respectively would be harvested. (ii) The rapidity with which the value of production increased over the previous season; in a period of 8, 11 or 13 years 80% of the increase expected over the long run was reached, depending on the hypothesis used. (iii) The social rate of discount was assigned values of 10%, 12% and 14% for the three cases. The indicators of the project evaluation are shown in Table 4.4.5. Their magnitude reveals the project's high economic rate of return. -16- Page 6 of 6 pages 4.5 Final recommendations The following is a summary of some considerations which, on the basis of the analyses made, are recommended to ensure the success of the project: (i) With regard to surface and underground water quality, the water should continue to be monitored so that, should a salt content considered dangerous to agricultural use be detected, timely and necessary action can be taken. In particular, in order to protect groundwater reserves, extraction should be restricted so that potentially permanent damage can be avoided. (ii) With regard to the participation of beneficiaries in the financing of the upkeep and operating costs of the system, a campaign to promote the notion that such expenses will be fully covered by the farmers should be carried out, so that the services can be rendered expeditiously and properly. In addition, farmer cooperation should be sought in the amortization of works. (iii) Although technical advisory programs for the beneficiaries appear to be proceeding at a suitable pace, special attention should be given to persuading the beneficiaries to maintain the original levelling of their land. (iv) It would clearly be desirable to terminate the resettlement of beneficiaries as quickly as possible, in order to avoid losses caused by continuing to provide water to parcels in the southern part of the old District. (v) It would be productive to promote the introduction of more efficient plot irrigation systems, such as pressurized irrigation, which would make it possible to use the water resources in the region more intensively. EVALUATION OF THE REHABILITATION OF IRRIGATION DISTRICT 14 RIO COLORADO, B.C.N. & SON CHANGES IN CONSTRUCTION Rehabilitation lst Change 2nd Change Actual ITEM Unit Project Amount A Qunttyfuantity QuantityAmut/ Quantity Canal rehabilitation and construction km. 2,334 1,520 1,520 1,992 Canal structures No. 4,685 4,000 4,337 5,221 Rehabilitation and construction of drains km 1,515 1,350 1,360 1,515 Drain structures No. 1,702 1,200 1,245 1,226 Land levelling ha 178,638 138,150 160,000 191,815 /3 Well drilling No. 328 /2 89 89 189 Well rehabilitation No. 533 246 246 297 Road construction km 555 555 555 1,383 /1 Construction and repair of houses and buildings houses 83 70 83 55 Repair and expansion of telephone system km 467 120 120 120 Installation of measurement stations stations 76 92 92 92 /1 Includes repair of existing roads. /2 107 wells for irrigation and the rest for level measurement. /3 Included some 20,000 levelled ha. /4 Includes the works carried out in the Mesa Arenosa de San Luis Rio Colorado. EVALUATION OF THE RAHABILITATION OF IRRIGATION DIETRICT 14, RIO COLORADO, B.C.N. 6 SON. INVESTHENT PRORAT (EN PESOS) TEM 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 TOTAL IN CONSTANT 1966 PRICES Contracted Work 37 830 399 97 590 627 163 772 310 254 406 989 104 479 540 96 694 022 139 570 210 103 309 100 n.d n.d levelIing 618 377 2 548 227 9 750 000 26 082 401 35 387 594 28 229 422 33'683 214 36 660 343 19 173 000 192 132 578 Engineering, 10 475 100 26 022 232 44 648 475 54 170 862 54 342 045 50 261 050 37 878 096 51 867 544 26 725 448 n.d n.d administration, and District vehiclea Subtotal: 10 475 100 64 479 008 144 787 329 227 693 172 334 831 435 190 128 184 162 801 540 225 121 018 166 694 891 n.d n.d Payment for 00 land, indemnifica- 3 013 055 8023 902 1 349 801 1 957 488 6 021 978 4 765 217 2 939 914 1 712 500 n.d. n.d. t ion (resett lemuent) ToTAL: 10 475 100 67 492 063 152 811 231 229 042 973 336 788 923 196 150 162 167 566 757 228 060 932 168 407 391 496 800 001 2 053 595 533 IN CONSTANT 1977 PRICES Contracted Work 140 142 220 361 446 091 606 563 666 942 455 320 386 961 706 358 126 033 516 926 881 382 626 388 n.d n.d Levelling 2 483 444 10 233 846 39 156 627 104 748 603 142 118 863 113 371 179 135 273 973 147 230 321 77 000 000 771 616 856 Engineering, 38 796 667 96 378 641 165 364 742 200 632 836 201 266 854 186 152 062 140 289 256 192 102 801* 98 983 163 n.d n.d adminiistra- tion, equip- a nent and Dis- trict vehicles , Subtotal: 38 796 667 239 004 305 537 044 679 846 353 129 1 248 470 777 715 232 631 611 786 460 844 303 655 628 839 872 n.d n.d Paytent for land 11 002 050 29 302 470 4 933 770 7 141 260 22 000 000 17 400 000 10 725 000 6 250 000 n.d. n.d indemnification ( react t leient) TOrAL: 38 796 667 250 006 355 566 347 149 851 286 899 1 255 612 037 737 232 631 629 186 468 855 028 655 635 089 872 184 000 000 6 002 586 733 * Includes expenditure in Mesa Arenosa de San Luis Rio Colorado. Source: General Resident's office for Rehabilitation of Irrigation District 14. EVALUATION OF THE REHABILITATION OF IRRIGATION DISTRICT 14 RIO COLORADO, B.C.N. & SON COMPARISON OF PLANNED AND ACTUAL INVESTMENT PER SELECTED CATEGORIES FEASIBILITY STUDY ACTUAL Estimate ITEM (millions of pesos) millions of in millions 1966 prices current pesos of 1966 pesos DISTRIBUTION NETWORK Canals 465.93 859.082 645.65 Canal structures 45.97 200.171 150.44 DRAINAGE NETWORK Drains 44.58 158.525 110.19 Drain structures 42.61 99.899 75.08 LEVELLING 143.59 398.621 299.59 OTHER WORKS Well rehabilitation 65.72 172.245 129.45 Construction of level stations 0.38 4.150 3.12 Road improvement and construction 39.92 44.312 33.30 Installation and repair of telephone lines 2.50 1.950 1.47 Construction and repair of houses and buildings 20.69 5.300 3.98 TOTAL: 871.89 1,944.255 1,461.42 Includes expenditures in the Mesa Arenosa de San Luis Rio Colorado. Source: General Resident's Office for Rehabilitation of Irrigation District 14. EVALUATION OF THE RERABILITATION OF IRRIGATION DISTRICT 14, RIO COLORADO, B.C.N. & SON. SUPPLY AND DEHAND FOR MANPOWER IN AGRICULTURAL ACTIVITIES (NO.OF MANDAYS/HO) hinthly MAnth of JAN FEB KAR APR NAY JUN JUL AUG SEPT OCT NOV DEC AV Peak Demand 1959-60 Required 6 230 6 640 6 710 30 350 56 935 57 216 57 298 57 625 115 416 8 312 6 248 6 508 34 624 115 416 Existing 47 623 47 623 47 623 47 623 47 623 47 623 47 623 47 623 47 623 47 623 47 623 47 623 47 623 47 623 Surplus 41 393 40 983 40 913 1 17 273 39 311 41 375 41 115 21 864 Shortfall 9 312 9 593 9 675 10 002 67 793 67 793 1969-70 Required 12 722 14 434 28 141 23 863 19 937 20 994 21 489 22 630 50 664 15 707 13 678 13 696 21 496 50 664 C) Existing 32 820 32 820 32 820 32 820 32 820 32 820 32 820 32 820 32 820 32 820 32 820 32 820 32 820 32 820 1 Surplus 20 098 18 386 4 679 8 957 12 883 11 826 11 331 10 190 17 113 19 142 19 124 12 811 Shortfall 17 844 1 487 17 844 1976-77 Required 15 388 17 636 26 109 21 916 13 685 18 302 18 058 18 817 41 910 13 575 15 026 16 249 20 223 41 910 Long-term 15 177 19 886 31 496 28 631 22 025 22 052 22 293 24 461 47 513 18 428 18 329 17 128 23 952 47 513 requirements Sourcet General Population Census, General Statistics Bureau, secretary of Industry-commerce, and estimates by the Evaluation Office, S.A.R.H. (Secretariat of Agriculture and Water Resources). EVALUATION OF THE REHABILITATION OF IRRIGATION DISTRICT 14 RIO COLORADO, B.C.N. & SON YIELDS, AVERAGE FARM PRICES AND PRODUCTION COSTS WITH THE PROJECT YIELDS AVERAGE FARM Production Cost 1978 Potential Prices (Pesos/ha) (Ton/ha) (Ton/ha) (Pesos/Ton) 1/ Garlic 7.0 9.0 10,000 3,578 Sesame 1.5 1.5 7,540 3,151 Alfalfa 12.6 15.0 1,450 3,378 Cotton 3.5 4.0 7,925 4,180 Dats 3.5 5.0 1,830 2,588 Sunflower 2,3 3.0 4,525 2,588 Barley 3.7 4.2 2.062 2,728 Asparagus 3.5 4.0 9,375 2.605 Miscellaneous 1.0 1.0 23,340 j'o0 Maize 2.5 3.5 2,900 2,441 Grass 50.0 50.0 156 2,341 Sorghum 3.0 4.5 2,030 2,474 Forage sorghum 50.0 70.0 150 3,148 Wheat 4.3 5.0 2,170 2,704 Vines 14.6 14.6 6,720 4,675 1/ Does not include labor, taxes or depreciation of farm machinery. EVALUATION OF THE REHABILITATION OF IRRIGATION DISTRICT 14 RIO COLORADO, B.C.N. & SON REGULAR COSTS WITHOUT PROJECT (thousands of pesos) YEAR MAINTENANCE OPERATING COSTS AGRICULTURAL EXTENSION 1968 30 305 93 183 11 928 1969 36 785 95 477 10 197 1970 27 975 86.253 8 353 1971 23 160 108 176 9 739 1972 29 556 95 773 10 054 onward in 1977 prices. EVALUATION OF THE REHABILITATION OF IRRIGATION DISTRICT 14, RIO COLORADO, B.C.N. 6 SON. EXPENDITURES MADE BY THE IRRIGATION DISTRICT FOR MAINTENANCE AND OPERATION IN 1977 PESOS Average ITEm 1968 1969 1970 1971 1968-71 1972 1973 1974 1975 1976 1977 Studies/Projects 2 753 429 2 396 170 1 933 537 808 318 1 972 863 4 356 452 4 009 683 3 978 547 3 770 143 2 851 548 2 277 000 Maintenance 23 096 716 31 227 278 25 692 865 21 723 613 25 435 118 37 489 880 31 700 970 43 111 209 39 250 239 42 428 342 54 805 000 lorchasee and 4 455 091 3 161 757 348 993 627 591 2 148 358 855 069 2 949 662 1 693 178 1 553 138 - - Equ i pment Total Maintenance 30 305 236 36 785 205 27 975 395 23 159 522 29 556 339 42 701 401 38 660 315 48 782 934 44 573 520 45 279 890 57 082 000 Operation 74 240 564 76 506 339 69 678 731 91 634 805 78 015 110 76 919 161 69 773 114 76 524 558 75 595 359 53 551 513 57 121 000 mnagement and 18 215 483 17 887 311 14 908 798 14 099 011 16 277 651 15 555 183 16 788 754 20 786 837 20 798 530 15 264 538 17 347 000 Administration Miscellaneous 727 057 1 083 686 1 665 759 2 442 650 1 479 788 1 712 995 2 204 311 5 335 933 L 355 086 740 773 1 737 000 Operations Total Operations 93 183 104 95 477 336 86 253 288 108 176 466 95 772 549 94 187 339 88 766 179 99 627 228 97 748 975 69 556 824 76 205 000 EVALUATION OF THE REHABILITATION OF IRRIGATION DISTRICT 14, RIO COLORADO, B.C.N. & SON. EXPENDITURES FOR TECHNICAL ASSISTANCE IN 1977 PESOS Average ITEM 1968 1969 1970 1971 1968-71 1972 1973 1974 1975 1976 1977 Irrigation and 10 307 408 8 755 693 7 452 005 8 838 369 8 838 369 14 197 199 11 514 922 9 466 604 12 435 537 12 789 768 20 388 000 Draina3e Engineering Agric-Extension 1 620 916 1 440 815 900 509 900 509 1 215 687 2 521 425 2 701 527 8 284 682 10 806 107 14 408 142 14 408 142 Services Total Agric- 11 928 324 10 196 508 8 352 514 9 738 878 10 054 056 16 718 624 14 216 449 17 751 286 23 241 644 27 197 911 34 796 142 Extension Source: Main Office, Irrig. District 14, Rio Colorado, B.C.N. & Son. EVALUATION OF THE REHABILITATION OF IRRIGATION DISTRICT 14 RIO COLORADO, B.C.N. & SON EX-POST ECONOMIC EVALUATION INDICATORS LEVEL C/B R.P. P.V.N.B. I.R.R Optimistic 2.32 14 5,620 20.8 Average 1.88 15 3,202 20.5 Pessimistic 1.55 16 1,835 20.1 P C/B Cost/Benefit ratio; R.P. Recovery period, in years; PVNB Present value of net benefit, in millions of pesos; I.R.R. Internal rate of return, in %.  - 26 - PROJECT COMPLETION REPORT MEXICO RIO COLORADO IRRIGATION PROJECT I. BACKGROUND 1/ General 1.01 In 1967, agriculture in Mexico contributed about one-sixth of the total gross domestic product and employed nearly half of the population. Agricultural output was growing at a rate of 4% per year, coping, therefore, with the annual population growth of 3.6%, although consumption per capita was increasing. Agricultural commodities made up 65% of all exports, with cotton, cereals and livestock products representing 35%. Mexico's agricultural policies were aimed at production of adequate food and fiber products for national consumption and at increasing agricultural exports. 2/ 1.02 More than 4 million ha of farmland were under irrigation, 2.5 mil- lion ha of which were operated by the Government through the Secretariate of Hydraulic Resources (SRH) and 1.5 million ha by private irrigation systems. About 30% of the Mexican crop production at that time came from irrigated land. 3/ 1.03 The total cultivable area of the country is estimated to be 30 mil- lion ha, of which about 40% was cultivated in 1967. 4/ It was felt that a continued increase in agricultural outputs could be obtained not only by increasing the area of cultivation but also by using the already cultivated land more intensively. The Government devoted, at that time, more than 90% of its agricultural investments to build and improve irrigation facilities while giving priority to rehabilitation of existing irrigation facilities over the construction of new schemes because of greater economic benefits. In the early 1960s, the Mexican Government prepared a plan to improve the irriga- tion system in several districts, among them the Rio Colorado. As part of this plan, studies were initiated with great urgency in 1965 to prepare a rehabilitation project for the Rio Colorado which would assure adequate use of the available water in 176,000 ha already irrigated and permit the expansion of the irrigated area by about 27,000 ha. On October 27, 1966, the feasibility report prepared by the Secretaria de Recursos Hidraulicos (SARH) was sub- mitted to the Bank. 1/ For more detail on the agricultural sector, see "Mexico - Background Paper on Agriculture," December 1, 1977, Division LCPA4, IBRD. 2/ In 1976, Mexico's Agricultural Sector's contribution to gross domestic product (GDP) amounted to about 9% and constituted 23% of total exports and 4% of total imports. The sector is the main source of income for Mexico's 24 million rural population that makes up about 40% of the country's labor force. 3/ The total irrigated land in Mexico is now 5 million ha (of which 3.9 million are Government operated and 1.1 million are private), estimated to contribute about 60% of the gross production value of crops. 4/ In 1977 between 15 and 20 million ha, or about 60% of the cultivable land, was under cultivation. - 27 - 1.04 The Mexican Government requested on February 8, 1967, a Bank loan to help finance the rehabilitation and consolidation of the Rio Colorado Irrigation District comprising 203,000 ha. Three previous loans had been made by the Bank for the rehabilitation of irrigation projects 1/ and they were progressing satisfactorily. In February/March 1967, a Bank mission 2/ visited Mexico to appraise the Rio Colorado project. The project was found economically justified (internal rate of return, 11%) and suitable for a Bank loan, and a US$25 million loan for a term of 20 years, including 7 years grace at an interest of 6-1/4%, was signed on January 25, 1968. The Rio Colorado Irrigation Project was the fourth and, thus far, the largest irri- gation rehabilitation project financed by the Bank in Mexico. 1.05 During 1968, preparation of engineering designs were undertaken by SRH but no construction contracts were awarded. The loan did not become effective until February 20, 1969 (see Chapter III); construction started about the same item, is now (December 1977) about 72% completed 3/ and is expected to be fully completed in 1985. By the end of 1976, as loan funds were being exhausted and project completion was reaching about 85% of appraisal estimates, the closing date was set for June 30, 1977. The Secretariat of Agriculture and Hydraulic Resources (SARH), following the Bank's sugges- tions, 4/ started preparing an ex-post evaluation report (Informe de eval- uacion de Resultados) on the project. A draft of this report (considered by SARH to be almost complete) was available to the Project Completion Report mission of April 1978. The Rio Colorado Area 1.06 The project area includes the Rio Colorado Irrigation District within the Mexicali Valley. This valley is part of the Colorado River delta which includes as well the Imperial Valley of California. The District ex- tends south from the United States-Mexico border into the States of Baja California and Sonora. The gross area of this District is about 330,000 ha, of which 176,000 ha were irrigated in 1967. Water from the Colorado River, the amount of which is established under International Treaty, was used at that time for the irrigation of 113,000 ha, while 63,000 ha were supplied with water from deep wells. The climate of the District is arid, with an average annual rainfall of only 36 mm. Temperatures during the summer months average about 320C (900F), while temperatures in winter are about 130C (550 1/ Loan 275-ME in January 1961 for US$15 million; Loan 336-ME in April 1963 for US$12.5 million; and Loan 450-ME in May 1966 for US$19 million. 2/ The Bank mission was composed of Messrs. Greenshields, Bartsch, Kordik, Zarandin, Garnier (FAO), and Knights (Consultant). 3/ If compared with revised estimates of work. If compared with appraisal estimates, completion as of December 1977 was 86%. 4/ March and December 1977 Supervision Missions. - 28 - Freezing conditions during nights may be expected between the months of November and February. Hailstorms during winter months have been reported only twice in the past 34 years. 1.07 The District is a relatively flat plain within the Colorado River delta. The plain is divided by a 10 to 20 m elevated ridge so that drainage from the northwestern part flows toward the Salton Sea in the United States and from the remainder of the area toward the Gulf of California. Both parts of the area have topography favorable for irrigation and effective water con- trol. The soils of the District are medium to fine-grain alluvium with no coarse sands or gravel deposits. Most of the soils are classified 1/ as I and II, which are suitable for sustained production under irrigation. In general, they are free draining except for about 33,000 ha of heavy clays for which an intensive drainage network is necessary. The organic matter content is low in all series; therefore, nitrogen fertilizer applications are necessary for satisfactory production. 1.08 At the time of appraisal (1967), salt had accumulated in the soils due to faulty irrigation practices under the desert-type climate; limited application of irrigation water had, moreover, precluded leaching and resulted in increasing salinity over about 35% of the area. 2/ It was estimated that 20% of the area was free of salts, that about 35% would require moderate leaching and that 22% would require heavy leaching; the remaining 23% would not be cultivated including 13% with high salinity 3/ and three areas of highly alkaline soils. Further, the constant rise of salt content 4/ in the Colorado River water since 1962 was causing a still more severe salinity condition for the Colorado River District. The proposed leaching for saline soils at appraisal would not have been possible had the salt content in the river water not been drastically lowered since 1972 due to the measures taken by the USA in compliance with an international treaty 5/ as explained further (para 1.14). 1.09 The availability of water resources for the District has suffered a drastic change over the last 70 years recorded. At the beginning of the century, there was plenty of water in the river and supposedly also in the underground aquifer. Annual floods occurred and water covered extensive 1/ According to the United States Bureau of Reclamation's system of land classification. 2/ Appraisal Report, paragraph 3.09. 3/ More than 2% salts. 4/ In 1962, the drainage sy3tem of the U.S. Welton-Mohawk Irrigation Dis- trict began discharging highly saline water into the Colorado River upstream of the international border. 5/ "Minute 242" signed between the USA and Mexico sets 1,050 ppm as the maximum salt content of the Mexican share of the Colorado River water. - 29 - areas and caused great damage. One of the larger floods broke through the Alano River, a branch of the Colorado River and replenished the USA's Salton Sea, which was a dry closed basin. As water utilization and flood regulation of the river water took place in the USA, less water reached Mexico and less water went to recharge the aquifer. During the 1930s, the Hoover and Parker dams were built in the United States, which greatly reduced the threat of floods but also greatly reduced the water volume that the Colorado River brought into Mexico. Thus, Mexico and the USA initiated negotiations for the control of the water volume that was to be received in the Mexican terri- tories. 1/ These negotiations ended with the signing of the 1944 trSaty by which Mexico's share of the Colorado River water was set at 1,850 Mm /year. At that time, the total water resources for Mexico was estimated to be 2,800 Mm /year, as shown below. However, as the water table rec9rds showed a pro- gressive depletion during the previous years when 1,150 Mm /year had been pumped, and because infiltration from canals was to be eliminated, the tota water availability for the area was estimated, at appraisal, to be 2,550 Mm / year as follows: Total Availability Total Resources (at appraisal) -------------Mm 3/year------------- Gravity Water From the Colorado River 1,850 1,850 Underground Water /a From US through the common aquifer 300 300 Deep percolation from canals 150 Deep percolation from fields 500 950 400 Total 2,800 2,550 /a No percolation from the river channel or from flooded areas occurs, as had been the case many years ago. 1.10 The Colorado River water has now been almost completely tapped at the international border and the river flow at that point consists almost en- tirely of return flow. In order to comply with the international treaty, as to both quantity and quality of water, the USA is practicing constraint in further use of fresh water, is channeling highly saline water to the sea and, in the near futurs, it will construct a large desalinization plant with a capa- city of 340,000 m per day. 1/ International water treaty between the USA and Mexico, February 3, 1944. - 30 - 1.11 The latest reviews of the hydrologic studies show that although the seepage from canals will 3be eliminated and deep percolation on the fields may diminish, pumping 900 Mm /year from the aquifer will not seriously affect the aquifer and that no saline intrusion will occur for more than 30 years. Water availabilit for the Rio Colorado District will then, in the long run, be about 2,750 Mm /year from which, water supply fSr municipal and industrial uses will be subtracted, leaving a net of 2,370 Mm /year for irrigation. 1.12 Irrigation in the area began in 1908, through the efforts of mainly private US firms. In 1937, the Government, as part of its agricultural reform policy, decided to colonize and develop the valleys of Mexicali and San Luis Rio Colorado. The former National Irrigation Commission took over the Irriga- tion District and started to improve the irrigation facilities, principally through the construction of canals, drainage works and the Morelos Diversion Dam. The cultivated area, which was 4,000 ha in 1908, grew to about 60,000 ha by 1937 (Table 1). From 1938 onwards, expansion of the cropped area continued and by the time of appraisal, 176,000 ha were being cultivated. The main crops were cotton and wheat, occupying 72% and 26% of the area, respectively. As surface-water resources diminished in quantity or had to be disposed of because of high salinity, the use of underground water increased considerably. In 1937, about 91,000 ha were irrigated by gravity and about 100 ha by wells, whereas in 1967, 113,000 ha were irrigated by gravity and 63,000 ha by wells. 1.13 Early in the 1960s, the salinity level of the Colorado River in- creased substantially as a consequence of drain water that was being dis- charged into the river by US farmers and the irrigation works that were built several decades before on a cost-sharing basis, partly by the Federal Govern- ment, partly by private enterprises and partly by the users themselves, were developing great deficiencies. Little attention was given to the construction of facilities that would make full utilization of the potential water re- sources, and excessive water losses resulted from unlined canals and ineffi- cient water distribution. Maintenance was poor since the water tariffs were insufficient to cover costs. Most of the canals suffered loss of capacity and the drainage system, although large, was incomplete and deficient. Nearly all of the structures made of wood were in poor condition, insufficient in number and obsolete. Inadequate land leveling also contributed to neffi- cient water use. Under these circumstances, about 50% of the 1,850 Mm received under the 1944 international treaty was being lost in the canal systems at a time when the salt content of -the water had increased from about 950 to 1,500 parts per million, with a maximum of up to 3,000 parts per million. The farmers had no alternative but to continue using this water, producing serious increases in the salinity of their soils but with no means to leach them. It was necessary then to reduce water allotment per user, first from 20 to 18 ha and later from 16 to 14 ha. Deep well drilling was accelerated in order to increase the irrigation water supply. Every year the volume of water pumped exceeded the aquifer safe yield. 1.14 Although the area (because of soils and climatic conditions) is well suited to irrigation farming and has considerable potential for expand- ing agricultural output, even with limited water supplies, the water quality, - 31 - the condition of the irrigation works and the salinity of soils had reached, by that time, disastrous levels. Even so, agricultural production levels of some crops were among the highest in Mexico and farmers had large enough farms to achieve acceptable levels of income and to use modern mechanized methods. The authorities, as well as the farmers, were aware of these poor conditions in the project area and of the risk of further deterioration and they were in general agreement that remedial action had to be taken. The farmers, through their organized groups, expressed strong support for the proposed rehabilitation program. The Government initiated discussions with the US to find a solution to the problem of the high salt content in the Colorado River water and prepared the studies for the rehabilitation of the District. 1.15 In 1965, after five years of negotiations, Mexico and the US signed a provisional agreement 1/ whereby they concurred that the problem should be observed and studied for the following five years. The US constructed at its own expense a lined canal to carry the Welton-Mohawk drainage water to Morelos Dam where it could be discharged either above or below the Dam, as required, to maintain acceptable salinity of water for use by the Rio Colorado District. Experience on the operation of the Morelos Dam, in conjunction with the drainage bypass channel since its completion in November 1965, appeared to indicate, and the appraisal mission took it for granted, that an acceptable salinity level could be 3attained and that practically all of the 1944 Treaty allocation of 1,850 Mm /year would be available for use in Mexico. However, this was not the case since salinity increased constantly and the water share decreased because large amounts of highly saline water had to be disposed of. After the Mexican President visited the US in 1972, "Minute 241" was signed between the two countries. This document established further corrective measures to improve the salinity of the water, reducing it to 1,050 parts per million on average annually. In 1973, "Minute 242" was signed, giving a final solution to the problem. Mexico now receives the agreed volume of water with a maximum annual average salt content of 1,050 parts per million, while the high salt content water is disposed of by the Welton-Mohawk canal which was further extended through Mexican territory so as to discharge into the Santa Clara estuary. The construction, operation and maintenance costs of this canal were assumed by the US Government. 1.16 At appraisal, about 7,000 ejidatarios occupying 115,000 ha and about 5,000 private owners occupying 82,000 ha were settled in the area, mainly along the banks of the Colorado River and the Alamo River, a branch of the Colorado. The occupied area was not, however, compact with properties scattered over 328,000 ha; this situation has been corrected as part of the rehabilitation project. At present, within the project, gross commanded area of 275,000 ha, about 7,000 ejidatarios have 123,000 ha, and 6,000 private farmers have 84,000 ha. 1.17 At appraisal, the population of the area was about 300,000 of which about 220,000 were in Mexicali City. The actual and projected area populations are as follows: 1/ Protocol of March 22, 1965. - 32 - Total Population 1930 1950 1970 1980 Mexicali 30,000 124,000 396,000 574,000 San Luis - - 80,000 120,000 Total - - 476,000 694,000 Mexicali: Rural - 59,000 102,000 150,000 Urban - 65,000 294,000 424,000 San Luis: Rural - - 25,000 35,000 Urban - 7,000 55,000 85,000 Source: Statistics of the Mexican Economy, NAFINSA, Mexico 1977, SARH and Estatistical Agenda, Bajo California, 1977. II. FORMULATION Origin of the Project 2.01 The fundamental reason for undertaking the project was the rapid deterioration of agriculture in the project area due to the loss of land to salinity and the ever smaller water availability due to high losses through infiltration and faulty operation of irrigation systems The problem was even greater since total iater availability of 1,850 Mm /year from the Colorado River and 950 Mm /year from the aquifer were not enough to serve the 203,000 ha with water rights, excluding all the remaining arable land. In 1960, 177,000 ha were irrigated with river water and 21,000 ha with well water. By 1967, however, 113,000 ha were irrigated with river water and 63,000 ha with well water. While river water with high salinity was disposed of in the sea, the aquifer was -being over-pumped. Further, the properties were scattered over all the area and there were independent diversions of water to irrigate areas of different sizes. 2.02 Therefore, rehabilitation of the Irrigation District #14, Rio Colorado was urgently needed, and a program had to be worked out to achieve maximum use of the water and land at hand. The decision was made to build a whole new network of conveyance and distribution canals completely lined in concrete, including the necessary laterals to irrigate a maximum tract of 60 ha. Conveyance losses were expected to be reduced from 50% to a maximum of 15%. Land irrigated with pumped water was to be fixed with regard to well capability, salinity and the topography of the land. Well water serving an area of over 50 ha was to be conveyed through concrete-lined canals to the delivery gate of each 50-ha subdivision. Consideration was also given to completing the drainage system required for allowing the proper disposal of salts from cultivated lands. A large part of the existing drains were to - 33 - be rehabilitated and new ones built to facilitate and expedite the disposal of leaching water. Because of the lining of the irrigation system, a reduc- tion of the aquifer recharg was expected. After a study of hydrologic con- ditions, a volume of 700 Mm /year was established for the more rational usage of ground water. Using hydrogeological studies ai a basis, plans were also laid to make use of an estimated volume of 150 Mm /year of the underground stream which empties into the Salton Sea. In order to use this water for irrigation purposes and at the same time keep the water table low and relieve the rising artesan pressure in the northwest area of the District, deep wells were to be drilled. In the area of 3mesa de San Luis, it was planned to drill 60 deep wells to increase by 180 Mm /year the water volume available. The existing wells and equipment were to be rehabilitated and wells beyond repair because of slides or bad water quality were to be replaced. A considerable number of small-diameter watertable observation wells were to be constructed. A road network, ditch rider houses, a telephone network, a repair workshop for machinery and pumping equipment and a building for the District offices were also to be constructed. Further, and one of the most important aspects for correcting the District's problems was that the area was to be one body, leaving out all tracts of land which, because of their condition, did not qualify for good crop production or make their reclamation worthwhile. Thus, the Rehabilitation Project included plans to relocate irrigated areas within the District's newly fixed limits. Unirrigated settlers' land in excess of that with water rights would be expropriated, and the owners would be reim- bursed for the value of the land at tax assessed prices. Users whose land with water rights would fall outside the District's limits, were to be relo- cated in the expropriated area. Likewise, in accordance with the Department of Agrarian Matters and Colonization, 1/ the land owned under the provisions of Agrarian Law would be relocated, reducing the parcel to an area entitled to water rights. To assure that farm irrigation efficiencies would be high, it was decided that land would be levelled and special quarters, including labs and equipment, would be provided for the Engineering and Drainage Depart- ment of the Irrigation District. 2.03 The rehabilitation of the Rio Colorado District, had priority in the national program of rehabilitating and expanding existing irrigation districts (para 1.03), since it would contribute, not only to the agricul- tural economy of the District in particular and of Mexico in general, but also to the more effective utilization of the country's water resources. Preparation, Evaluation, Negotiations and Approval 2.04 SRH, which replaced the National Irrigation Commission in 1947, was responsible for the project. It took care of project preparation, exe- cution, operation, maintenance and yearly agricultural planning. Between 1965 and 1966, a comprehensive semi-detailed feasibility report for the rehabilitation of the District within the lines described above (para 2.02) 1/ Now Secretaria de Reforma Agraria (SRA). - 34 - and including technical assistance to farmers 1/ was prepared by the General Direction of Irrigation Districts (Direccion General de Distritos de Riego) of SRH. Project implementation was to take five years (1967-71) at a cost of Mex$1,054 million (US$84.2 million). 2/ 2.05 Two relevant aspects regarding the work volume estimates and implementation time and the decreasing benefits should the project not have been undertaken are worth mentioning. The study was based on the preliminary designs of 56,000 ha, or 28% of the project area. Quantity estimates and costs were extrapolated for the entire project area. In the case of drains and some of the larger canals, specific preliminary designs and quantity estimates were prepared to obtain greater accuracy than that available from some area extrapolation. Over time, however, this procedure was misleading since work quantities grew (para 3.03) and implementation time almost doubled, as explained further in this report. The feasibility study considered that, without the project, conditions in the area would further deteriorate because of over-pumping and increased soil salinity. Should the over-pumping not stop, larger areas than those irrigated with the over-pumped water would have to be abandoned in the long run. 3/ Should the irrigation and drainage works not be accomplished, salinity would have increased in the 127,645 ha already affected. In 30 years time, class II soils would become class III and yields would fall by 30%. In 50 years time, class II soils would become class IV and yields would drop by 60%. Cotton and wheat could not be planted after 50 years and alfalfa after 30 years. Benefits, therefore, were imputed to the stoppage of over-pumping and the prevention of increasing salinity, neither of which would be accomplished without the project. 2.06 The feasibility study was sent to the Bank in October 1966 and the Bank appraisal mission visited Mexico in February/March 1967. The Appraisal Report is a clear document, but it provides little detailed information to explain or support some of the claims and conclusions reached in it, such as those on on-farm works, farm budgets, crop yields and cost projections, and the additional quantified benefits from replenishment of lost water and pre- vention of salinity. It also appears that, even though SRH technical capa- bilities were known to be very good, the appraisal mission overlooked the communication and managerial shortfalls. It did not study sufficiently the institutional aspects and the communication problems within and between agencies and their responsibilities in carrying out the project. The economic rate of return on the investment estimated by the appraisal mission was 10% (Chapter V). 1/ Technical assistance to farmers was to be provided within the context of the Technical Assistance Plan (Servicio de Asistencia Tecnica Agropecuaria) for the State of Bajo California. 2/ The Feasibility Report estimated the B/C ratio for the project as 2.398. 3/ It was estimated that, in 1964, benefits of Mex$2 .114 million were produced3in the 11,627 ha irrigated wiSh 206.4 Mm (actual pumping, 1,186 Mm against safe pumping, 980 Mm ) of over-pumped water. - 35 - 2.07 The project was considered economically justified and suitable for a Bank loan of US$25 million for a term of 20 years, including a grace period a Bank loan of US$25 million for a term of 20 years, including a grace period of seven years (the 20-year term was used on the three previous rehabilitation loans and appeared appropriate for this type of project). It was recommended that, during negotiations, assurances be obtained that: (a) the Ministry of Agriculture and SRH would continue and strengthen technical service to farmers (Appraisal Report, para 5.10); (b) soil studies would be continued to the satisfaction of the Bank and SRH would provide farmers with technical and such other assistance as may be required to enable them to install on-farm drainage when it becomes necessary (Appraisal Report, para 5.11); (c) adequate financial resources would be provided for medium- and long-term credit (Ap- praisal Report, para 5.13); and (d) the Government would take all necessary legal action to establish the rehabilitation charges and water rates to be paid by the farmers (Appraisal Report, para 6.06). 2.08 The main issues discussed within the Bank and with the Mexicans were water tariffs and recuperation of the investment expenditures from farmers. During negotiations, discussions centered on fixing a minimum target for medium- and long-term credit, the project foreign component, and allocation of the Bank loan proceeds. Also at negotiations, the Mexican representatives stated that their analysis of project costs had shown a foreign currency component of US$39.2 million rather than the Bank's estimate of US$25.0 million. Based on the detailed revision of the Mexican's analysis, the Bank representatives agreed to increase the foreign currency component estimate on six specific types of work by US$3.6 million. However, the Mexi- cans agreed that cement, concrete additives, reinforcing steel and equipment did not involve foreign exchange expenditures and were to be financed solely with local currency. This facilitated the procurement process on these items for the Mexicans. The net effect of these adjustments was a foreign exchange requirement of US$26.2 million. Since a loan in the amount of US$25 million had received preliminary approval by the Loan Committee, the negotiators agreed to reduce the contingency allowance by US$1.2 million, giving a total foreign exchange requirement of US$25 million. The reduction of the contin- gency allowance and of US$2.8 million of local costs, resulted in a reduction of the total project cost estimate from US$99 million to US$95 million. The total cost reduction helped to bring the internal rate of return up to 11% from 10%. 2.09 A loan of US$25 million to assist in financing the foreign exchange cost of the Rio Colorado Irrigation Project was approved by the Board on January 25, 1968 and the loan documents were signed one day later. The borrower was the Nacional Financiera, S.A., a financial agency of the Mexican Government and the only authorized channel for foreign loans guaranteed by the Government. The provisions of the Loan and Guarantee Agreements were similar to those contained in loan documents for the previous Mexican irri- gation projects and for other projects of this kind in other countries. As in previous Mexican irrigation loans, the Nacional Financiera, S.A. acted only to channel the loan proceeds to the Government, which, through SRH, was to carry out and execute the project. Hence, provisions relating to the project which are normally in the Loan Agreement were, in this instance, in the Guarantee Agreement. - 36 - 2.10 In addition to the standard legal requirements, the Loan Agreement (Art. VIII, Section 7.01), required as a c6ndition of effectiveness, that the Guarantor take the initial required legislative or other action to ensure that water charges levied for operation and maintenance of the project would be set and maintained at levels necessary to cover those costs and that the Government make provisions for the recovery from the users of as much as practicable of investment expenditures. Further, the Guarantee Agreement (Section 3.09) states that the Government shall make available or cause to be made available medium- and long-term credit on reasonable terms to farmers in the project areas to enable them to carry out the neces- sary on-farm works and make full use of the project facilities, and (Section 3.10) that the Government shall provide or cause to be provided agricultural extension and other technical services needed in the project area. 2.11 It was agreed 1/ that the disbursement percentage of the loan funds would be 35% of the civil works (excluding cement, concrete additives and steel to be furnished by the Guarantor) whether performed under contract or by the Government's own force. The schedule of disbursements was estimated 2/ as follows: 1968 $4.4 million 1969 $3.8 million 1970 $4.4 million 1971 $3.3 million 1972 $3.3 million 1973 $3.1 million 1974 $2.7 million 2.12 Two side letters 3/ provided details and clarified the covenants contained in the Loan documents. The first letter referred to (a) continua- tion of the hydrogeological studies and the establishment and enforcement of regulations to prevent depletion of the underground water supply; (b) the undertaking of continuous soil testing in 150 representative areas of the District to detect the development of the soil salinity, the encouragement of farmers to make appropriate chemical applications if required, and provi- sion of assistance by SRH to the farmers in installing on-farm drainage when necessary; (c) the undertaking of studies of agricultural patterns and encour- agement of the farmers to adopt the most productive and advantageous cropping pattern; (d) the provision of agricultural research and extension services to the farmers, an increase in activities of the Instituto Nacional de Investigaciones and the provision of at least 14 extension officers by 1970; (e) the improvement of seed inspection and certification services and plant pest and disease measures, and the enlargement of the irrigation technical assistance to farmers; and (f) the actions to ensure that not less than Mex$5 million would be available for agricultural credit from 1969 onwards, increas- ing to Mex$15 million by 1974. The second letter referred to the allocation 1/ Side letter of January 26, 1978. 2/ Appraisal Report, paragraph 4.16. 3/ The two letters are dated January 26, 1968. - 37 - of the loan proceeds and to the procurement procedures, as explained in other parts of this report. The Bank sent a letter to SRH 1/ enclosing an outline of the progress reports that should be submitted to the Bank. 2/ 2.13 The initial effectiveness date was August 1, 1968; however, this date had to be extended three times, and finally, the loan became effective on February 20, 1969. While the Government recognized the need to take action on water tariffs, it felt that the political risk was high so decided not to act before carrying out an educational campaign leading to an agreement with all or most of the people concerned. In previous Mexican irrigation loans, the fulfillment of the condition of effectiveness that dealt with legislative or other actions to introduce appropriate water charges had also caused delays. The document setting up legal requirements was finally received as requested by the Bank, but actual action on tariff level did not take place until October 1977, and the recuperation of the investment has not yet started. In light of these experiences, it would have been desirable to require that at least the tariff level had been raised before Board presentation, while the collection of the investment expenditures could have been tied to the work schedule, as discussed later. Project Description 2.14 Schedule 1 to the Guarantee Agreement describes the project as follows: (a) The project consists of the rehabilitation, improvement and operation of the irrigation and drainage works and related facilities serving a gross area of about 250,000 3/ ha in in the Rio Colorado Irrigation District. (b) The main works included in the project are: (i) reconstruction and lining with concrete of about 2,330 kilometers of canals, and construction of necessary canal structures and gauging stations; (ii) improvement of about 1,180 kilometers of existing drains and construction of about 335 kilometers of new drains, together with necessary drain structures; (iii) rehabilitation of about 485 existing deep wells, drilling and equipping of about 105 new deep wells, and drilling of about 220 observation wells; 1/ April 29, 1968. 2/ Guarantee Agreement, Article III, Section 3.04(b). 3/ The cropped area would increase from 176,000 ha to 203,000 ha, with about 12,700 ha double cropped, bringing the total annual cropped area to 215,700 ha. - 38 - (iv) construction or improvement of about 555 kilometers of roads; (v) land leveling on up to 178,000 ha; and (vi) construction of telephone lines, office and repair shop buildings, and houses for ditch-riders. (c) The project includes the purchase, installation and utiliza- tion of equipment and vehicles for repair and maintenance operations. (d) The project also includes the establishment and operation in the Rio Colorado Irrigation District of a continuing program of hydrogeological and cropping pattern studies, investigation of on-farm drainage needs and soil testing in order to determine the adequacy of the underground water supplies, identify the cropping patterns which would make the most productive and advantageous use of available water supply, and detect salinity or alkalinity problems developing during the operation of the project. 2.15 Other project programs or actions, such as agricultural research and extension services, seed inspections and pest control, irrigation tech- nical assistance to farmers, and agricultural credit were to be undertaken by SRH and SAG. At appraisal, it was recommended that assurances that such programs or actions that would be undertaken should be obtained from the Mexican Government at negotiations. Commitments for carrying them out were reached through the side letters. These programs at the time of their inception were perhaps difficult to quantify and, therefore, were only lightly described and no implementation schedule or cost estimates were worked out for them at appraisal except for the amount available for credit. 2.16 The project cost estimates, including only civil works, relocation of farmers and equipment, was US$95 million. The investment period would be seven years, from January 1968 to June 1974. In addition, farmers were expected to invest about US$12 million for on-farm drainage over a period of about 15 years. The cost increase that some of the programs described before would impose on the SRH budget was apparently included in the O&M cost, while the cost of extension services to be provided by SAG was not considered at all. 1/ 2.17 Contracts for civil works were to be awarded on the basis of inter- national competitive bidding in accordance with the Bank's guidelines. Inter- national competitive bidding did not have to be called, however, for works to he undertaken by the Government's own forces or by local contractors with an aggregated amount equivalent to 20% of all construction cost to be financed by the Loan. 2/ 1/ They were also omitted when estimating the rate of return in the invest- ment. 2/ Side letter 1/26/78. - 39 - Targets and Goals 2.18 The basic objective of the project was to maintain the District's agricultural productivity through prevention of water losses and increasing soil salinity. It proposed the diversification of crops, introducing barley and oats or forage crops that would permit full utilization of winter water which would otherwise not be utilized beneficially, and the gradual reduction of the cotton area from 127,500 ha to about 100,000 ha to eliminate areas that produced low yields, which was inefficient use of water. Forage was in short supply in Mexicali, and livestock farmers have to import part of their feed from the United States. The expansion of forage production would meet current demands and, at the same time, enable expansion of livestock in the area. The livestock industry was expected to benefit also from the increased production of grain sorghum, especially in the conditioning of beef cattle for export and in the growing dairy industry. The diversification program was expected to be achieved through water control which SRH exercises through the District's Advisory Committee and medium- and long-term credit availability. In addition, increases in crop yields would result from improved water appli- cation and from relocation of farmers to better suited areas. The project intended as well to avoid adding unemployment of both manpower and existing capital facilities in the region. 2.19 The planted area was to be increased by 40,000 ha (from 176,000 ha to 215,000 ha), the production tonnage by about 570,000 tons (from 540,000 tons to 1,111,000 tons) and the gross value by Mex$180 million 1/ (from Mex$952 million to Mex$1,132 million). However, the projected alternative uses for the land (para 2.18) would, on the whole, cause a reduction of the average productivity for the extented area, from Mex$5,409/ha to Mex$5,248/ha. 2/ Net farm income after rehabilitation was to be increased, on average, by about Mex$9,000 ha. 3/ 2.20 Intermediate appraisal targets were 91% 4/ efficiency on conveyance and distribution of water and 80% on farm water applications. These targets were too optimistic, however, and will not be reached, even with completion of the proposed works and with excellent handling of water within the farms. If such goals were to be achieved, drastic changes in operation of the systems and schedules for irrigation water delivery would be necessary, together with the introduction of mechanical methods of irrigation (sprinklers, dripping), which had not been proposed at appraisal (para 3.16). 1/ Current prices of 1967. 2/ Appraisal Report, Annex 8. Productivity Area (ha) Tons ('000) GPV (Mex$ '000) Mex$/ha Mex $/ton Present 176,000 540 951,607 5,409 1,763 Future 215,700 1,111 1,132,040 5,248 1,019 3/ The Appraisal Report does not provide figures to support this statement. 4/ 87% for the gravity water and 96% for well water. - 40 - 2.21 Control of the soil salinity and of the water table were two other intermediate but basic targets. Although these targets could not be quantified, it was assumed that they would be reached during the project implementation period. Soil salinity control, however, depends largely on the availability of adequate amounts of water with low salt content, in the efficiency of drain- age and on the efficiency of the leaching technique. This is a,long process and may take many years. It should be mentioned that the quality improvement of the Colorado River water (para 1.14) was not foreseen at appraisal. The water table control, is also a trial and error procedure that could take many years to achieve, perhaps exceeding the implementation period. Sector Significance 2.22 Although the Rio Colorado is a large irrigation project by any standard, it covers only about 4% of the irrigated land in Mexico and about 1.5% of the total cultivated land. At the regional level -- Northwest region - the Mexicali and San Luis Rio Colorado valleys constitute about 11% of the cultivated area and produce 7% of the grains, 5% of the oilseeds and 16% of the cotton. III. IMPLEMENTATION Effectiveness and Start-up 3.01 The Loan Agreement (Article VII) specified as additional conditions to effectiveness that: (a) the Borrower would make arrangements with the Guarantor, which would be satisfactory to the Bank, providing for the trans- fer of the proceeds of the loan from the Borrower to the Guarantor and (b) the Guarantor would take such legislative or other action to levy water charges for operation and maintenance and for the recovery of the investment, as much as may be practicable, together with reasonable interest thereon. The date of effectiveness was set for August 1, 1968. However, due to the difficulties in making the final arrangements with the water users regarding water charges, the date of effectiveness was extended first to September 1, 1968 and then to February 20, 1969. By this time, the Government had complied with all the conditions for effectiveness. A Presidential Decree was issued (Dec. 1968) stating that farmers should pay 50% of the rehabilitation invest- ment, or about Mex$4,800/ha, over a period of 25 years with 6% interest, which represented an annual installment of Mex$192/ha. 1/ No mention was made of O&M charges. Construction, which should have been started at the beginning of 1968, was also delayed a year. During 1968, construction activity was limited to planning and preparatory works such as collection of field data, design and preparation of specifications for construction. The two first contracts (for the central feeder canal) were awarded on March 27, 1969. The technical assistance to farmers 2/ also suffered a setback. Of five extension agents operating in the area at appraisal in 1967, only three remained until 1970, 1/ See paragraph 6.18 2/ Appraisal Report, paragraph 5.09. - 41 - even though the commitment of the Guarantor was to provide at all times adequate agricultural extension and other technical services as required in the project area and to have by the beginning of 1970, at least 14 extension offices within the project area. The main cause of delay in starting proj- ect implementation appears to have been the lack of a clear construction program and engineering design. Even if the legal actions concerning water charges could have been fulfilled at any time before the final date for effectiveness (February 20, 1969), construction could have started no earlier than 1969. 3.02 The appraisal mission (February 13, 1967) attributed to Irrigation District #14 staff engineering design capabilities they did not have and estimated that final engineering design could be done in sufficient time to allow for the commencement of the works in 1968. It was a common practice of SRH to transfer experienced design engineers to the field office of a given project only after the Government had allocated money for construction, which, in this case, happened in 1968. The year's delay in starting construction and the low speed of staffing and implementing the technical assistance pro- gram for the farmers caused further delays in implementing the project but had little impact on project performance. Revisions 3.03 Some important adjustments in physical infrastructure, construction schedules and financial aspects were made after effectiveness. At the time of appraisal, there were no complete designs for the whole project but only semi-detailed layouts for the secondary systems of 56,000 ha (28% of the project area) and a few more advanced designs for some of the main canals. These preliminary designs and quantity estimates were then extrapolated for the entire project area; therefore, many changes of design in the remaining area (147,000 ha) of both major and minor works and changes in the construc- tion schedule had to be expected (para 2.05). Changes in design were reflected in the final amount of work and represent an increase of roughly 40% over the overall quantities estimated at appraisal. The design changes are also re- flected, although only partially, in the revised cost estimates made in the early stages of the project as shown below. Major changes in design that occurred during project implementation are summarized in paragraph 3.27. - 42 - Millions of US$ Appraisal Revised Cost Cost Estimates Estimate 1971 1972 a/ Change 1. Rehabilitation, construction and lining of canals 29,500 33,288 49,920 + 2. Canal structures 3,200 3,283 5,640 + 3. Rehabilitation of existing drains 2,300 2,086 7,440 + 4. Structures of existing drains 1,800 2,040 6,240 + 5. Construction of new drains 1,200 1,098 2,000 + 6. Structures for new drains 900 1,003 3,280 + 7. Well rehabilitation 2,500 2,212 2,304 + 8. Well construction 2,800 2,484 2,800 0 9. Roads 3,200 2,851 3,280 + 10. Telephone system) 200 179 800 + 11. Gauging stations) 27 40 + 12. Buildings 1,700 1,477 840 - 13. Land leveling 11,500 10,260 9,600 - 14. Relocation of farmers 800 1,360 1,760 + 15. Cement and Steel 10,500 16,328 11,152 - 16. Local purchase of equipment 1,400 1,336 1,360 - 17. Engineering and supervision 9,000 6,600 13,200 + 18. On-farm assistance - - 2,704 + 19. Offices in Mexico City - 720 840 + 20. Contingencies 12,500 9,144 4,000 - Total Project Cost 95,000 97,776 129,200 + a/ SARH Letter #5850 of November 17, 1972. 3.04 From 1974 onwards, project cost continued to rise, mostly due to inflation although some changes in design were still being introduced. These changes, however, did not alter the project as a whole since the same area, within the same perimeter as considered at appraisal, was to be served by the irrigation works. The construction schedule, the percentage of loan disburse- ment and the closing date of the loan had to be revised because of the cost increase of the project. The construction schedule was extended twice allow- ing two years of extra time over the original completion date of June 1974. The second extension was made on the assumption that, by December 1976, coin- ciding with the exhaustion of the loan funds, all works would be completed, but by that date very little progress had been made. As of December 31, 1977, only 86% of the works considered at appraisal, or 72% of the revised amount of work, had been completed. There is no firm schedule for the completion of the remaining 28% of the works. The project cost estimate rose from US$95 million at appraisal to US$262 million as of December 31, 1977, as shown below. The loan disbursement percentage for civil works was reduced from 35% to 16% to keep the Bank involved until the works were completed. Although - 43 - the Bank's participation in the following years dropped to about 10%, the disbursement percentage of the loan funds was kept at 16% because such funds were close to exhaustion. Cost Bank Share % Disbursement & Year US$ Millions Over Total Cost Over Civil Works 1968 95 26 35 1971 98 26 35 1972 129 19 35 1974 160 16 16 1976 250 10 16 1977 262 10 Loan Funds Exhausted 3.05 By December 31, 1977, US$199 million had been expended, of which US$25 million were loan funds and US$174 million local funds. The overall participation of the Bank up to that date had decreased from 26% at appraisal to 12.5% as shown below. Since the amount of the loan remained unchanged, the Government provided an additional US$104 million and it will have to pro- vide approximately another US$63 million to complete the works. No additional financing or cancellation of the loan has occurred. US$ Millions Revised Estimate As of December 31, 1977 Bank Participation Local Loan Total Completed Percentage Original Cost 70 25 95 95 26 Revised Cost 237 25 262 199 13 Extra Government Contribution 167 - 167 104 - Physical Progress 3.06 Because of the initial delay in starting the works (para 3.01) and the low funding of the project, 1/ the works were only 17% completed by the end of 1971 against the 63% estimated at appraisal. By this time also, the revised cost estimate had increased by US$3 million (from US$95 million at appraisal to US$98 million). Modifications in design (including changes in canal and drainage alignments), specifications and construction, as well as interference of on-going irrigation with construction, have caused further delays. The modifications were found to be necessary to adjust the works to physical features of the area (topography, soils) and because some original designs and specifications did not appear to yield the desired results. Due to increases in quantity of work, changes in specifications and, consequently, in unit costs as well as to underestimation of costs at appraisal, the esti- mated project cost rose again to US$129 million. Since 1972, Government 1/ As of December 1971, investments totalled US$43.7 million instead of the US$58.7 million considered at appraisal. - 44 - funding has exceeded the amounts of investment assumed at appraisal. 1/ By December 1974, when all works should have been finished, only 49% had been. 2/ By this time, the project cost estimate had risen to US$160.1 million, or 69% more than the appraisal estimate. The effort of the Government to catch up with the construction schedule was hampered by the sudden inflation of 1974 and following years. Therefore, although Government investment continued to be high in 1975 and 1976, only 72% of the works proposed had been completed by December 1977. The remaining works will be carried out from 1978 onwards at a pace set by future Government funding. The following table summarizes implementation progress: Physical Progress as Percentage of Completion 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 Appraisal 18 33 50 63 76 88 100 - - - Actual: A a/ 0 6 26 40 50 69 81 83 n.a. 86 B b/ 0 3 8 17 31 40 49 58 68 72 Total Project Cost Estimates (US$M) 95 - - 98 129 - 160 - 250 c/ 262 d/ e/ a/ Compared with appraisal estimate. Progress reported by the Supervision Missions and Quarterly Reports. b/ Compared with revised estimate. c/ At US$1 = Mex. Pesos 12.5 (April 1976). d/ Cost to complete all project works at US$1 = Mex. Pesos 12.5 from 1968 to June 1976 and US$1 = Mex. Pesos 22 thereafter. e/ As of December 31, 1977, only US$199 million equivalent had been expended. 3.07 The physical progress overall percentage .of 86% (compared with appraisal) is somewhat misleading since: (a) it compares actual progress with appraisal estimated quantities and not with the revised work quantities; (b) for many project components, actual work quantities considerably exceeded those assumed at appraisal and further works in these components are envi- saged; and (c) some other components fell short compared with appraisal and no further investment is intended. Judging from which works should still be carried out to fulfill project needs (such as new drains, new wells at San Luis, interconnection canals, structures and leveling), the physical progress 1/ As of December 1974, investments totalled US$124.3 million instead of the US$95 million considered at appraisal. 2/ Compared with the revised estimate. -45 - of the project appears to be 72%. With the amount of work done so far, how- ever, the project targets of 207,000 ha of physically irrigated land has almost been reached, 1/ overall water conveyance efficiencies have risen from 44% to 76%, harmful salinity has receded in nearly 60,000 ha, and the water tables have dropped to safe levels in about 14,000 ha. 3.08 Project Area. Physical area increased from 156,000 ha irrigated in 1963 to 200,000 in 1976, as shown in Table 2. The Appraisal Report indi- cates that 176,000 ha were planted in 1968 and the third supervision report (April 30, 1970) mentioned an average area of 180,200 ha for the period 1964-68. These figures include double cropping which was possible because of over-pumping. The project goal area at appraisal was 203,000 ha physical plus 12,700 ha of second crop. In the agricultural season, 1977/78, 202,600 ha were planted. Present estimates are that 207,000 ha physical could be irrigated when all project works are completed, and that from 15,000 to 20,000 ha could undergo double or permanent cropping when final conditions regarding water allocation for other uses and increases in irrigation effi- ciencies are reached. 3.09 Water Sources. The project area has three sources of water: (a) the Colorado River, which, a cording to the treaty between Mexico and the USA, should provide 1 850 Mm /year; (b) the local aquifer where the potential is about 500 Mm /year (previously estimated as 700 Mm3 /year); and (c) the aquifer of San Luis where the actual estimated potential is about 400 Mm3/year. These gross quantities, however, do not give a firm basis for estimating future water availability since some assumptions about groundwater recharge have to be revised, and, because the use of both surface and underground water depends on physical and legal constraints which have not yet been resolved. 3.10 The average quantity3 of water received from the Colorado R ver from 1973 to 1975 was 1,802 Mm /year and from 1975 to 1977, 1,831 Mm year. According to SARH, the difference between 1,831 Mm3/year and 1,850 Mm /year, as stated in the international agreement, is due to the fact that, occasion- ally, the monthly water demands do not coincide with the monthly minimum and maximum volumes of water to be delivered by the USA. 2/ FSr the purpose of the analysis, we will consider the amount of water as 1,850 Mm /year, as agreed between the two countries, on the assumption that the Rio Colorado Irrigation District will improve its operations 3/ and will be able to use all the water 1/ In the winter agricultural season, 1977/78, 203,000 ha will be planted including 20,000 ha of perennial crops. 2/ The water is delivered to Mexico in accordance with monthly requests which must be within the minimum and maximum water quantities stipulated by the Treaty. Operations under the International Treaty are supervised by the International Boundary and Water Commission, United States and Mexico. 3/ Improving operation would mean also enforcing planting time, irrigation application schedules and pumping synchronization. - 46 - it is ntitled to. The latest reviewed potential of underground water is 900 Mm /year (paras 1.09 and 1.11), from which about 300 Mm /year flos from the US Colorado River Basin across the international boundary; 300 Mm /year comes from percolation of irrigation water from the lisht soils located in the northwestern part of the project; and about 300 Mm /year can be safely mine-pumped from the San Luis area at the east side of the project area. For the local aquifer, Inderlying the project, the underground flow has been estimated at 2003Mm /year and the deep percolatisn from canals and irrigated fields as 500 Mm /year, giving a total of 700 Mm /year. However, as most of canals are now lin5d and as irrigation application efficiencies tend to im- prove, only 300 Mm /year 1/ as infiltration frsm the fields has been considered as giving a total exploitable 1mount of 500 Mm /year. Total extraction in this area, however, averaged 900 qm /year in the last five years, which represents an over-extraction of 400 Mm /year and causing a general drawdown of 30 to 50 cm/year. In the area of SaS Luis, according to the hydrogeologic study men- tioned before, up to 400 Mm /year can be pumped safely. No severe effects in the drawdown of the aquifer will be present for more than 30 years. Pres- ently, only 120 Mm /year are being pumped; therefore, the total availability of the water can be estimated as follows: Availability Estimates Previous Latest Present Use Difference --------------------(Mm 3/year) Colorado River 1,850 1,850, 1,802 - 48 Project Aquifer 700 500 924 +424 San Luis 200 400 125 -275 Total 2,750 2,750 2,851 +101 3.11 The mission feels that further observations and studies of the underground reservoir are needed and that the legal actions (as mentioned in the Loan document) to control excessive pumping should be taken to avoid possible irreversible damage from the intrusion of sea water. The Mexican Government continued the hydrogeologic studies 2/ to determine the adequacy of the underground water supply (para 3.33) and did not take all the legal actions regarding the control of pumping as stated in Section 3.12 of the Guarantee Agreement. On December 15, 1975, a presidential decree was issued forbidding the drilling of new wells in the project area. SARH officials told the mission that a program for reducing pumping from wells would be enforced as the interconnection of canals (of wells with a gravity system) progress and conveyance efficiencies rise. At present, private owners of wells get an annual pumping permit from SARH according with the area to be planted, regardless of the quality of water. All wells have water meters 1/ The SARH fSasibility report (1967) considers that deep percolation is only 98 Mm /year for the pump irrigation area. 2/ As agreed in the side letter of 1/26/78. - 47 - and SARH personnel check the area irrigated and the hours of pumping to com- pare with the actual pumped volumes with the volume specified in the annual permit. However, this control is weak. Also, when a private owner of a well wants to rehabilitate it, he has to get authorization from the Govern- ment. The regulations of the Irrigation District covering underground water include private wells. 3.12 Available Water and Distribution. Because of ov r-pumping, total water availability between 1973 and 1977 averaged 2,851 Mm /yeir as shown in Table 3. In the long range, as gumping decreases to 900 Mm /year, total water availability will be 2,750 Mm /year. As this amount (Rio Colorado and and underground aquifer) of water is the only source of water in the region, municipal supply for the area and for Tijuana City will have to be diverted from this source. Further, some allowances have to be made because other small uses and unavoidable operat onal losses at the head structure gates occur so that only about 2 370 Mm /year would actually be available for irri- gation instead of 2,511 Mm /year, as considered at appraisal, as follows: Distribution of Water Mm 3/year Present Medium-Term Long Range Average Year Year 1985 Year 2000 1. Total Resource 2,851 2,750 2,750 2. Other Uses Waste and other uses 80 80 80 Mexicali water supply 100 50 70 Tijuana water supply 7 130 230 Total 187 250 380 3. Irrigation 2,663 2,500 2,370 3.13 Well Irrigation. At appraisal, 60,000 ha were irrigated with under- ground water and the intention was to reduce this area to about 50,000 ha. SRH records, nevertheless, show that 75,175 ha were irrigated with underground water in 1971 and that a substantial reduction through the years brought this figure down to 69,200 ha in 1977, thus transferring 5,975 ha to gravity irri- gation; however, 19,000 ha more than expected at this time still remain under pu ping. Reduction of gumping (in the project area) from the present 1,050 Mm /year to only 500 Mm /year would force a further reducSion in the pump irrigated area towards the proposed limits. About 500 Mm will be 9umped within the project area to irrigate adjacent lands and about 400 Mm will be pumped in San Luis fields, out of the project area to be delivered through the canal network to supplement irrigation in both the wells and gravity - 48 - areas. 1/ Salinity of underground water is very high and progressively in- creasing from south to north forcing the abandonment of some wells (para 3.17). If irrigation efficienSies improve to desired levels, about 50,000 ha could be irrigated with 500 Mm pumped from t e aquifer. At present, water gross allocation from wells of about 12,777 m /ha is high considering that commanded areas are not saline, and, therefore, no leaching is necessary. This shows that farm and conveyance efficiencies are low. However, it is expected that when the well interconnection canals are finished and SRH take over the private wells, gross allocations could be reduced to about 10,000 m /ha. 3.14 Irrigation Water Use. The agricultural pattern proposed at appraisal which inclded 99,000 ha of cotton and 8,700 ha of alfalfa, required an average of 6,990 m /ha for plant use. The §ross allocation, including leaching re- quirements and losses, was 11,650 m /ha for 215,700 ha. 2/ During the last five years, the average amount of water used in irrigation has been 2,664 Mm / year, from which 1,770 3Mm /year was gravity water from the Colorado River and San Luis, and 1,894 Mm /year from the wells, as shown in Table 4. The areas, however, grew from 168,248 ha in 1973 to 199,002 ha in 1977, still keeping annually about 20,000 ha of pas ure. Gross wate5 distribution varied greatly each year (from about 12,000 Im /ha to 15,000 Im /ha in the area irrigated by wells and from about 13,000 m /ha to 16,000 Mm /ha in the area irrigated by gravity). However, this unstable water distribution showed a tendency to stabilize through the years. This fact is consistent with the increase of conveyance ef iciencies from 52% in 1973 to 76% in 1977,. In the long run, only 2,370 Mm /year will be availabl for irrigation, which 9eans a grosh allocatisn reduction, from 10,710 Mm /ha to between 9,643 Mm /ha and 9,159 Mm /ha, depending on whether permanent crops occupy 10,000 ha, 15,000 ha or 20,000 ha, and, if the wells are fully connected to the gravity system. 3.15 For the purp"ose of our analysis, three stages have een analyzed, as shown in Table 5: (a) actual as of 1977/78, with 2,664 Mm /year 3water availability; (b) a transitory situation during 19§5, with 2,500 Mm /ha water availability; and (c) the year 2000, with 2,370 Mm /ha water availability. It is considered that both conveyance and application efficiencies will rise, respectively, from 76% to 80% and from 70% to 80%. The latter case assumes that sprinkler systems will be used in much of the area. A resume of the analysis is shown below: 1/ The installed capacity of the 76 deep wells in San Luis is 300 m 3/year; however, according t*o the International Treaty, only 197,350 Mm /year can be pumped in this area. The new proposed wells would be drilled south of the area. 2/ Appraisal Report, Annex 5. - 49 - Available Water allocation Area (ha '000) Water (m 3/ha) Physical Cropped Mm /year Gross a/ Net b/ Present Situation, 1978 200 220 2,660 10,710 6,872 Transitory Situation, 1985 207 227 2,500 10,164 6,789 Future Situation, 2000 207 217 2,370 9,643 6,989 222 9,159 6,598 227 9,159 6,566 a/ At farmgate. .b/ Water available to plants (excluding leaching). 3.16 Irrigation Efficiencies. Irrigation efficiencies have increased substantially since the beginning of the project. Conveyance efficiencies rose from 44% in 1968 to 51% in 1973 and to 76% in 1977. Although the 5anal lining and the structures are almost complete (para 3.26), about 530 Mm /year. of water is still being lost. As leakage from the canals is nil (perhaps 5%), all losses are operational and it will be hard to lower them unless the present system of having water in all the canals simultaneously and distri- buting the water on demand is abandoned. If water is delivered to the different areas of the project on a fixed schedule and farmers are given an irrigation quota, the operational losses could be reduced. Although it would be difficult to convince the farmers to regionalize crops and to set fixed planting times, some progress in this respect could be expected. Therefore, for the purpose of the analysis, it has been assumed that conveyance effi- ciencies will rise to 78% by 1985 and to 80% by the year 2000. There are no firm data on farm efficiencies although there seems to be wide variance between pump irrigated and gravity irrigated areas. Estimated field efficiencies run from 50% to 75%. Comparing consumptive uses with water deliveries as farmgates, including the water used in leaching salts, the average field application efficiency appears to be 70%. If leaching water is excluded, field effi- ciencies are only 60%. As the saline nature of the soil will always require additional water with each irrigation, we will assume that plant water requirements are somewhat higher than consumptive use and, therefore, that the efficiency of application is 70%. Since the water for irrigation will be scarce, field efficiencies should increase. It has been estimated that with appropriate drainage to wash the salts with limited amounts of water and using more advanced irrigation techniques, the application efficiencies could increase to 80% by the year 2000. Regarding the amount of wat r saved, the project goals have been partially reached since more than 600 Mm /year of water have been saved through the lining of canals. Operational and field application sfficiencies, however, can still be improved in order to save about 250 Mm /year additional. The estimated overall irrigation efficiencies are shown in Table 6. - 50 - 3.17 Water Salinity. Salinity in water diverted from Rio Colorado has diminished according with the "Minute 242" signed between the USA and Mexico in 1974, from an average of 1,500 ppm to a range of 1,020 ppm to 1,051 ppm. Water from San Luis wells has only about 900 ppm of salts and the average salinity of the old wells within the project area is of 1,400 ppm 1/. This clearly indicates the need of shutting down pumping in the most saline wells and using water from the San Luis wells, as recommended in the hydrogeologic study. It is estimated that average salinity of the water for the whole pro- ject will be, in the long run, about 1,060 ppm in Table 7. It is also expected that salinity in the aquifer may drop in the future due to the diversion to the Gulf of California of the drainage water brought by the Wellton-Mohawk canal. A benefit of the project not attributable to the investment is the dropping of salinity content of the Colorado River water. Before 1960, river water with 900 ppm of salt 2/ was mixed with drainage water of the Wellton- Mohawk irrigation, which had more than 9,000 ppm of salts. The mixed water very often contained more than 3,000 ppm of salt 3/. The progressive loss of agricultural area that took place before the rehabilitation project should be attributed, therefore, not only to the lack of drainage and to the faulty irrigation practices used in the area, but also to the high content of salts in the Rio Colorado River water. It has to be borne in mind that, without the project, but with the enforcement of the "Minute 242," the salinization of the soils could have been controlled and that from 1974 onwards only very small areas, if any, were going to be abandoned because of salinity. 3.18 Soil Salinity. By nature, both the soils and water in the area are highly saline, with the exception of the deep sandy soils located in the north- east. The project intended to improve the soils by means of good water man- agement and by providing adequate drainage on about 14,000 ha. The Appraisal Report did not treat the salinity problem in depth even though it attributed to the project the benefits of stopping the spread of salinity in the project area. However, it did recommend a continuous program of soil' testing so that any salinity or alkali problems (para 3.33) could be dealt with effectively. It was further agreed (side letter of January 26, 1968) that soil samples would be taken annually at three depths at 150 representative areas. Data on the subject in supervision reports are scarce and sometimes contradictory. On the other hand, SARH has taken, as agreed, thousands of soil samples and has compiled abundant information but failed to identify the real nature of the soil salinity problem (para 3.22). As further explained, it has been found that salt accumulates in the project area at the rate of at least 2.5 million tons per year but the nature of the salts has not been determined. At the moment it appears as though the accumulation of salt in the soil profile is not harmful to the plants. An agreement has been signed between SARH and the National School of Agriculture of Chapingo 4/ to study the problem for a period of three years. Results can therefore be expected by 1982. 1/ Bureau of Reclamation water salinity classification. Water salinity of the deep wells varied between C 3- S and C - S3* 2/ Bureau of Reclamation water salinity classification C3-S1 3/ Bureau of Reclamation water salinity classification C -S2' 4/ Post-graduate college. Irrigation and Drainage Engineering Department. - 51 - 3.19 The first soil salinity survey was made in 1964-65 and showed that about 100,000 ha were free of salt and that the rest of the project area was either composed of saline or saline alkali soils. At appraisal, it was estimated that about 50,000 ha were free of salt; that in about 90,000 ha, salinity had increased because of inadequate leaching; and that about 60,000 ha needed heavy leaching. The supervision report of August 1968 reported that salinity had increased in over 130,000 ha, had decreased in 60,000 ha, and that were no changes in salinity in about 30,000 ha. Therefore the soil's condition between 1964 and 1968 had steadily deteriorated. A second soil salinity survey was carried out in 1977 and, although analyses are still being made, it appears from the preliminary results that about 110,000 ha are free of salt and that in the remaining area 97,000 ha have either saline or saline alkali soils. The supervision report of September 1976 states that 147,000 ha are salt-free or have moderate salinity and that areas of moderate salinity have dropped from 100,000 ha to 60,000 ha; therefore, the deterioration trend has been changed and from 1968 onwards, the salinity in the area as a whole has decreased. This result can be attributed, not only to the project works that made more water available (canal living) and allowed for better irriga- tion practices and leaching (land leveling and drainage) but also to the fact that salinity of irrigation water has dropped considerably because of the international treaty with the US. 3.20 The following table compares the results of the first and second soil salinity studies with the assumptions at appraisal. It can be seen that some improvement has been achieved but further improvements should still be made to assure agricultural success. Soil Survey Appraisal Soil Survey 1964/65 1967/68 1977/78 b/ Depth cm Area Depth cm 0-20 20-50 % a/ 0-20 20-50 ------------------ -'000 ha ------------------ Free of Salts 105 83 20 51 114 98 Saline 93 83 35 91 83 82 Highly Saline 5 37 22 61 0 27 Project Area 203 203 203 207 207 Outside Project Area c/ 69 69 23 61 25 25 Surveyed Area 272 272 100 264 232 232 a/ Described as soils free of salts, soil requiring moderate leaching and soils requiring heavy leaching. b/ Provisional results. c/ Saline and saline-alkali soils. - 52 - 3.21 Besides the described surveys, a salt balance trial has been done every year since 1970. It consists of determining the amount of soluble solids that remain in the fields or are removed from them. Total salt in- put brought by irrigation waters and total salt removal by drainage are calculated and compared. Sampling is more intense in the salty critical areas of the three parts of the project: zone of wells (light soils), inter- mediate zone (loamy soils) and saline areas (heavy soils). Presently, 800 samples are taken in selected locations and the annual and cumulative amounts of salt that presumably remained in the fields are the following: Year 1970 1971 1972 1973 1974 1975 1976 1977 Million of ton/year 2.8 2.8 2.4 2.5 2.7 2.6 2.7 2.6 Cumulative 2.8 5.6 8.0 10.5 12.7 15.3 18.0 20.6 3.22 To check these results, soil samples at six different depths are taken in 800 selected places. The results indicate that, in a four-year period, not less than 8 million tons of salt have accumulated in the area as follows: 1 million tons in the light soils, 1.5 million tons in the medium textured soils, and 5.5 million tons in the heavy and saline soils. The accumulation of salts in the soil profile takes place in the first 50 cm in the heavy soils, between 50 and 200 cm in the medium soils, and below 200 cm in the light soils. The described accumulation of salts in the pro- file is somewhat inconsistent with the rejent soil salinity survey, which shows a constant decreasing of salts in alt the depths of the soil profile. Further, as previously explained, these salts at the moment do not harm the plants but further accumulation may. It is believed that the Colorado River waters that have a high content of calcium (Ca) and magnesium (Mg) salts help to precipitate the harmful salts in soils. No carbonates or bicarbonates are present in the water. 3.23 Several trials of leaching have been made in fields that were aban- doned because of the salts and in areas where farmers were to be settled. In all cases, the results were positive in a sense that salts diminished in the higher stratas of the soil profile and accumulated at different depths, depending on the texture of the soil. The tested crops showed a considerable yield increase. For the purpose of this analysis, it will be assumed that soil salinity will not decrease nor increase in the coming years and that any future increase of the actual yields is due to other technical inputs. 3.24 Drainage. At appraisal, 137,000 ha irrigated by gravity and located mostly in heavy soils were to be provided with a drainage network for the disposal of leaching water and excess irrigation water and, in particular places, for lowering the water table. At present, 125,000 ha, or 91% of the appraisal estimate, have been provided with drainage networks. High water table affected areas located in the northwestern section have decreased from about 22,000 ha in 1968'to 8,000 ha in 1977. However, as the cause of the problem in this area seems to be the leakage from the unlined All-American Canal in US, the problem may not disappear completely in that area. There are still 164 km of drains to be dug in the western area, where a high water table and heavy soils are present. Construction has started recently and it is expected to be completed by 1979. On-farm drainage works have not yet started (para 3.28). The accomplishments as of today are as follows: - 53 - Drainage Improvement Project Goals Accomplishments Appraisal Revised 1972 1973 1974 1975 1977 Gravity Area ha 137,000 137,000 n.a. n.a. 110,000 120,000 125,000 Length of Drains km 1,515 1,691 444 779 1,170 1,350 1,527 % - 100 26 46 69 80 90 Water Table Depletion ha 18,000 22,000 n.a. n.a. 9,000 12,000 14,000 3.25 A problem which could have affected the project in the long run was the fact that the lower course of the Colorado River had been silted and that a sand bar of about 4 m in height has closed its mouth to the sea. Drainage water therefore would not able to reach the sea and the formation of a lake in the southern part of the project might have had adverse affcts. There- fore, in 1974, a channel 41 km long and 13 m wide to carry 6 m /sec was built from the point where the river bed reached an elevation of 4 m over sea level to a large depression located to the west of the project close to the Mexicali- Tijuana highway. The area of this depression is about 40 km in length and 15 kt in width, and the bottom is flat and about 4 m below sea level. At presen%, a lake has been formed with a water surface area of more than 100 km . The volume of water which evaporates from the lake will greatly exceed the drainage water.input and an equilibrium will be reached. 3.26 Project Works. Table 8 shows work quantities as estimated at appraisal, as revised in 1975/76 and as completed by December 31, 1977. The work quantities to be completed from 1978 onwards are also shown. The explanation of the difference in work quantities is in the footnotes of the table but the following are worth mentioning. (a) Canal lining. In addition to building the 307 kM of canals for connecting existing wells within the project area or installing a gravity system to replace the wells in the south portion of the area, about 120 km of canals should be built to connect the new wells to be dug in San Luis area to replace wells in the south part of the project. (b) New Wells. At present, out of a total of 725 wells dug since 1950, 693 are in operation and 32 have been shut down. By mid-term it is assumed that the same number of wells will be operating but about 60 will have been abandoned and replaced by new wells. About 70 private wells will also be closed and farmers will be given water by gravity. SARH will then dig 70 new wells in the Mesa de San Luis, as follows: - 54 - Number of Wells Present Future Project Wells Project Area 390 390 San Luis 67 137 Abandoned (32) (60) Private Wells 236 176 693 693 (c) Roads. The Appraisal Report proposed the improvement and construction of 555 km of roads, mainly on the berms of the main canals (main roads already existed). However, it was also necessary to widen and improve the roads along secondary canals to assure access and for protection of the canals. Some roads in the sandy soils areas where the wells are located were also built. The roads in the project totalled 1,835 km. An extra 120 km of roads will have to be built to connect the new wells in San Luis. (d) Land Leveling. There is great demand by the farmers for more leveling, although there are no significant advantages to be gained. However, since farmers do not have to pay for it, they demand it from the Government. This situation is reflected in the increase of about 10,000 ha in land leveling over the appraisal estimate. 3.27 Changes in Design. Many changes in design were introduced to project works, as summarized below. All of them were necessary in order to get the desired results and because new project needs arose (paras 3.03 and 3.06). (a) Major Changes (i) Change of location of main canals and laterals. The first half of the main canal (Alamos canal) has been abandoned and the new alignment is a by-pass route to the Central canal. The former main canal's laterals have been cut off and instead interlinked with the Central canal, which became the main supply of the system. A main canal (the Delta No. 2 lateral) that served the central part of the area has been abandoned and the supply of its commanded area is derived from the network of wells at San Luis. These changes brought about considerable adjustments in the distribution and drainage systems. The change of location of the Central canal was seen as necessary as the topographic survey advanced. The new location is in the highest part of the area, along the dividing ridge (para 1.07). An appraciable amount of earth works (embankments) have been saved and operation is easier. - 55 - (ii) Change of drainage flow direction on about 20% of the area. Main drain collectors (Norte and Ferrocarril) which previously drained into the California Gulf are discharging into Rio Nuevo and Salton Sea in the United States. All changes in the drainage network originated also from the topographic surveys and studies after appraisal. It would seem that the projection (from 56,000 ha to 203,000 ha) made for the irrigation and drainage works overlooked the particular topographic characteristics of some areas. (iii) Widening and surfacing of about 1,000 km of canal berms in order to accommodate gravelled roads for general traffic. It was cheaper than building special embankments for the roads (not foreseen at appraisal). It also protected the berms of the canals and permitted faster communication for general traffic and operation of the canal system. (iv) Enlarging and compacting the project irrigated area from 203,000 ha to 207,000 ha. About 4,000 ha of land within the project perimeter, not considered suitable for irrigation at appraisal, had been reclaimed and, about 400 ejidatarios had been settled there. The western portion of the project with about 4,000 ha which was going to be abandoned in order to transfer its water rights to the municipal supply of Tijuana are now being included again in the project because water savings on the project, due to lining, permits the water transference to Tijuana without sacrificing area. (v) Changing the project boundaries. In order to relocate farmers, within the project area, who had their land (10,300 ha) and their water rights in the south and out of the project boundaries and to accommodate farmers (7,000 ha) who were within the project area but had very bad soils or inconvenient locations, the project bounda- ries were changed. (vi) Inclusion of an outfall drain to avoid using the lower reach of the Colorado River that had silted (para 3.25). (vii) Addition of 70 wells in Mesa Arenosa de San Luis to enable the full use of the aquifer by extracting 200 m /year additional to the present pumping volume (para 3.15). This aspect however is yet under study and no decision of the exact location of the wells nor of the time for constructing them had*been reached. The works required are estimated as follows: - 56 - New deep wells 70 Lined canals 120 km Electric line 130 km Roads and ancillary works (b) Minor Changes (i) Due to the major changes discussed above, other changes were necessary such as: - Increased length of sub-laterals and number of canal structures - Increased number and change of characteristics of drain structures. (ii) Change of the telephone system by a radio system. (iii) Reduction of the number of ditch-rider houses. 3.28 Farm Investments. According to the appraisal, about 33,000 ha needed intensive drainage and about 18,000 ha needed on-farm drainage (tile). At present those drainage works have not been initiated and it seems that if credit is made available,to farmers for these works, not only the 51,000 ha seriousy affected, but about 50,000 ha more will be provided also with on- farm drainage works of both types, surface and tile drains. Because of the heavy investment required, it has been assumed for this analysis that on- farm drainage will be installed over a 20-year period, between 1985 and the year 2005. 3.29 As a result of a number of factors the on-farm investments in the project area have concentrated basically on the acquisition of farm machinery, transportation equipment, expansion of the area and alfalfa, some fencing, acquisition of breeding stock and the establishment of some perennial crops. Occasionally, some small land leveling works have been carried out by farmers. Farm machinery use has spread in the area, totally replacing drafting animals. In 1968, there were 9,896 operations completely mechanized, 871 partially mechanized and 470 not mechanized, and the equipment owned by the farmers in the area included about 6,800 tractors with 7,000 trucks. In 1976 (last year with statistical information), all farms had become totally or partially mechanized and were using about 4,600 tractors, larger ones, on average; about 470 combine harvesters; and 6,800 trucks (see comment on farm machinery ownership, para 4.25). - 57 - Farm Machinery 1969 1972 1976 Number Area Number Area Number Area Mechanized Farms 9,896 161,022 9,028 114,405 9,547 152,831 Partially Mechanized Farms 871 12,152 3,526 56,876 2,808 50,989 Farms without Mechanization 470 3,542 17 280 0 0 11,237 176,716 12,571 171,561 12,355 203,820 Tractors a/ 6,885 5,662 4,608 Harvesters b/ 1,435 1,233 1,965 Various 4,540 2,932 2,420 Trucks c/ 7,017 5,999 6,828 a/ All sizes of crawler and wheel tractors. The tendency is to have bigger tractors. b/ Of all types and sizes, from simple cutters to combines. c/ Includes jeeps, pickups and trucks. 3.30 The area under alfalfa (perennial crop) increased from 7,500 ha in 1967 to 19,600 in 1978 and constitutes the basis for the dairy industry. Fences have been placed to handle both dairy herds and fattening steers. About 3,000 ha have been planted to asparagus and vineyards. Some very in- teresting investments have been realized under the ownership and management of farmer organizations. Barns for milking cows and feed-lot corrals for steer fattening have been established and are successfully functioning as a service for different breeders or owners. 3.31 For the future and in order to achieve the full potential of the project, investent on improvement of irrigation and drainage will be encour- aged by the District authorities and the extension agents. Higher water use efficiency at farm level and on-farm drainage improvement are expected to play a relevant role in further raising the yields and increasing the output value for the whole project area. Processing and Marketing 3.32 Although agro-industry has developed to some extent in the region and the valley of Mexicali, it is not possible to relate its growth only to the project because there are other factors that have influenced the economy and activities in the region and because the information on these matters is not available. However, there are some data that reflect the situation to a point. The three most important dairy plants have increased their milk processing capacity by 77%, from 21.3 million liters in 1970 to 37.8 million liters in 1977. This increase reflects the growth in production (para 4.08) and is also related to the growth in production of feed grains and the increase in production of feed for livestock. Tomatoes grown in the area in increasing quantities have permitted the development of the juice and paste industries, - 58 - juice and paste industries, and part of the asparagus produced in the valley are processed and canned in Mexicali. An important and increasing percentage of the cattle fattened in the project area flows toward the local market, where slaughter houses have had to expand and modernize their facilities. The marketing channels for wheat are the Flour Company of Mexicali (Co. Harinera de Mexicali), which absorbs important quantities of wheat, and CONASUPO (Compania Nacional de Subsistencias Populares), which purchases all the sur- pluses and distributed them among the rest of the existing and new milling industries. CONASUPO also buys some oilseeds and barley, the rest being absorbed by oil extraction plants and the brewing industry. Most of the considerably increasing vegetable production flows into the domestic market, .fresh or canned, and some is exported to the U.S. Summarizing, the processing industries and the marketing channels and systems have expanded accordingly, with the expansion of the overall production in the region, although there is no specific and detailed data to quantify the different items. Special Studies 3.33 Both the Appraisal Report and the Loan Documents provide for the undertaking by SARH of some special studies aimed at solving the critical prob- lems of the area. It was recommended and agreed upon in the side letter of January 26, 1968 to continue or start observations of ground water 1/ and soil salinity and drainage 2/ and to initiate a study of the cropping pattern 3/. These are discussed below: (a) Hydrogeologic Studies. The Federal Government has continued hydrogeological studies to provide the basis for efficient integration of surface and groundwater supplies, to ensure effective control of pumping of the underground water in the project area and to provide data for the issuance of suitable regulations. However, only limited effort has been devoted to this study. Between 1971 and 1972, Mex$30 million were assigned to this study 4/. Perforation of observation wells cost Mex$27 million. Out of the 221 observation wells considered at appraisal, only 51 were drilled, of which four are producing wells and 47 are observation wells. Well depths vary from 150 m to 800 m. In addition, another 60 production wells are being used as observation wells, making a total of 110. Underground water levels are read twice a year in each of the 110 wells. The water table in the project area is falling from 30 to 50 cm/year and 10 cm/year 1/ Appraisal Report para 4.06 and Guarantee Agreement, Section 3.12. 2/ Appraisal Report para 5.11. 3/ Appraisal Report para 7.03. 4/ Estudio Geohidrologico del Valle de Mexicali B.C. y Mesa Arenosa de San Luis SON. SRH 1972. - 59 - in the San Luis area. Some preliminary hydrologic balances have been worked out that shSw that the underground water availability is about 900 Mm /year. Observations should continue, but, more frequently and in more places than have been studied so far, and a hydrodynamic model should be tried. Also more accurate estimates of the resources than the ones available will be needed (see para 3.11) in the near future. (b) Salinity and Drainage Studies. Soil salinity and water table studies are being carried out (paras 3.18 to 3.23). A total of 124 observation wells for water table and drainage studies were dug. They are of 4 to 6 m in depth and located at 2 km intervals or closer. Readings are being taken annually or semi-annually, depending on the location of the well and of the level of fluctuations detected in the pre- vious year. Maximum levels occur in June (summer) and minimum levels in January (winter). Recharges of up to 50 cm have been observed due to other causes than deep infiltration from irrigation. Isobatas and Isohipsos curves are drawn every one or two years. It is istimated that drains 2.5 m deep, with densities of 1 km/km would lower the water table to a safe depth. At present there is a large amount of information recorded on soil salinity, water table depths and flow, but SARH has not tyet found practical solutions to the problems. Regarding soil salinity, determination of salt types in the soil profile and methods of controlling salts have to be found. Regarding the shallow water tables, mainly in the heavy soils area, design parameters for de- signing the drainage have to be worked out. (c) Cropping Pattern Studies. These studies have not been under- taken. Yearly agricultural plans are drawn, taking into consideration: (a) Government production goals, (b) market and price opportunities, (c) farmers' wishes, and, (d) water constraints. Through the accumulation of experience some guidelines are now used as a basis for recommendations to farmers on crops to be planted. Procurement and Construction 3.34 The Guarantee Agreement states that "the Guarantor shall cause the public works included in the Project to be constructed, rehabilitated or im- proved by contractors satisfactory to the Guarantor and the Bank employed under contracts providing therefore," and the side letter of January 26, 1978 spells out the procedures to be followed for contracts to be awarded under international competitive bidding and for those to be let under local bidding procedures not exceeding 20% of all construction work to be financed by the - 60 - Bank 1/. Additionally, SARH follows the general procurement procedures used by the Government in all public works, which are considered by the Bank as "acceptable." Construction contracts include all costs of the inputs (materials, equipment, hand labor, administration and profit) with the exception of cement and steel, which are furnished by the Government at the work site. 3.35 Participation of foreign construction companies is precluded as in all other Mexican projects, mainly because of the small size of the contracts, and because of the cumbersome procedures that have to be followed to partici- pate in the biddings. All contractors for civil works in the project were Mexican firms, either from Mexico City or from the Mexicali area. The farmers participated to a large extent in the land leveling. SRH awarded contracts to the farmers on the basis of unit prices for the whole area, and the farmers used their own equipment, under the supervision of the SRH engineers. 3.36 It is a general practice of SARH to let contracts of US$2 million to US$3 million value for major divisible works such as canals and roads. Out of 31 contracts let under ICB procedures, only 26 exceeded the value of Mex$20 million 2/. The reason given is that the Government has to provide work for as many contractors as it can and to favor local (regional) con- struction enterprises, which, in general, are small. However, the main reason appears to be the way and rate in which final designs and procurement documents are prepared and the need to invest immediately the partial allocation of money that projects receive through the year from the Government. Both reasons, even though valid, interfere with timely execution and may effect the quality and cost of works. Contracts generally had to be extended because of changes in specifications or revision of quantities and, recently, because of cost increases. 3.37 Between 1969 and 1977, 46 major contracts and 170 minor contracts for civil works and about 800 contracts for land leveling were awarded for the Rio Colorado Project. Only 31 major contracts were awarded under ICB procedures and the 185 remaining civil works contracts were awarded, either through local bidding or through negotiations. Land leveling contracts were let to individual farmers and groups of farmers through direct negotiations in small contracts, ranging from Mex$100,000 to Mex$500,000. The distribution of contracts and their total value are summarized in the following table: 1/ Appraisal Report, para 5.07. 2/ Appraisal Report, para 5.07, states that: "In order to make practicable the participation of international contractors, schedules and specifica- tions would be prepared for civil works contracts having a minimum value of 20 million pesos insofar as possible." - 61 - Local ICB Procedures Total Executed Civil Works Major Contracts 31 15 46 46 Minor Contracts - 170 170 170 31 185 216 216 Land Leveling - a/ a/ a/ Value of Contracts (Mex$ Million) Civil Works 1,210.25 704.92 1,915.17 1,719.00 Land Leveling 401.58 391.30 2,316.75 2,110.30 a/ SARH does not have the figure but estimates that 800 contracts were let. 3.38 From the total value of the contracts of Mex$2,110 million, Mex$1,511.10 million were partially financed by the Bank. The remaining Mex$599.2 million were financed solely by the Government. From Mex$1,511.10 million worth of work financed by the Bank, Mex$l,210 million, or 80% were let under ICB procedures, and Mex$300 million, or 20%, were let under local procedures 1/. Due to cost increases of project works and because the dis- bursement percentage of the loan proceeds was lowered from 35% to 16%, the Bank's participation in financing the works has been about 21% of the 72% of the costs of all project works as of December 1977, as shown in Table 9. 3.39 Although Mexico followed ICB procedures, it was difficult for the Bank to keep track of the progress of contracts because (a) related documents (invitations to bid, bidding, awarding and such) generally arrived at the Bank out of date; (b) contracts were small, awarded in a disorderly way, and scattered over the area; and (c) disbursements were requested at partial progress of contracts. The large number of contracts did not allow the project offices to review them in detail and such procedures overburden Bank staff. This is a problem that causes concern in the Bank since it occurs in other irrigation projects as well. It could, to a certain extent, have been avoided if the Bank had insisted on having better prepared projects and firm procurement schedules. The experience of having leveled nearly 170,000 ha through contracts with farmers or groups of farmers has still to be analyzed. On the one hand, it appears to have been beneficial for the area since about 1/ Side letter of Jandary 26, 1968: "International competitive bidding procedures will not have to be followed in the case of items of con- struction work which may be carried out better by the Government's own forces or by local contractors, the total cost of such construc- tion not to exceed 20% of all construction work to be financed out of the proceeds of the Loan." - 62 - Mex$400 million worth of works were given to local farmers, but, on the other hand, it seems that works were not always being carried out properly and that farmers purchased too much equipment that will have very limited use in the future. Construction was generally made through capital-intensive procedures. Many small contractors have used hand labor extensively in lining small reaches of canals and in building canal and drainage structures. The quality of the work in general appears to be good. Cost and Disbursement 3.40 Table 10 and Chart 1 compares all the revised cost estimates (worked out by the supervision mission with information furnished by SARH) with those of the appraisal. Project cost has steadily increased during project imple- mentation, from US$95 million at appraisal to US$262 million in 1978, as follows. Project Cost Estimates US$ M Percentage Appraisal 1968 95.0 100 1971 98.0 103 1972/74 129.0 136 1975 160.0 168 1976 250.0 263 1978 262.0 276 3.41 As explained before, changes in design and in work quantities should have been expected, since, at appraisal, engineering was not completely pre- pared. In 1972 as engineering designs advanced, larger work quantities appeared and SARH 1/ advised the Bank that the cost of the project had risen from Mex$1,222.2 million (US$97.7 million) to Mex$1,615 million (US$129.2 3 million) due to various modificitions. Additional excavation of 10,790,000 m for the canals and 12,300,000 m for the drainage and about 1,300 more struc- tures were needed. SARH asked the Bank to lower the loan disbursement per- centages from 35% to 16% (as of January 31, 1973). As work quantities and unit costs continued increasing between 1973 and 1975, some of the project components were decreased in order to keep the cost of the project at US$129.2 million. However, this exercise was fruitless because all the works were required to reach -the project's objectives. By 1975 most of the engineer- ing designs were completed and the project cost estimate rose to US$160.5 million. From then onwards, project cost increases have been due mostly to in- flation. 3.42 Together with the underestimation of work quantities, the unit prices at appraisal seem to also have been underestimated. Between 10% to 20% of the civil works cost increases are attributable to deviation of water courses and to provisional protectin'g works because the area was under cultivation. If appraisal unit prices had been applied to the lately revised work quantities, an increase of about US$40 million, or 40% of the appraisal cost, would have resulted. Cost increases due to inflation rose more than two times over those 1/ Letter #5850 of November 7, 1972. - 63 - of appraisal and may account for about 80% of the project cost increases. Therefore, unit prices may have been underestimated by about 10% as follows: Project Cost (US$ M) Appraisal 95 Increase of work quantities (95 x 1.4) 133 Underestimation of cost (133 x 1.1) 146 Inflation (146 x 1.8) 262 The mission was not able to analyze in detail the cost increases because of the incomplete and unreliable information furnished by SARH. 3.43 Total Unit Costs. As of December 31, 1977, US$199.2 million had been invested to complete about 86% of the works considered at appraisal, or 72% of the works included in the latest estimate of work quantities. For the purpose of this analysis, we assume that the Mexican Government will invest the additional US$62.8 million required to complete all project works by 1985. The breakdown of investments made up to 1977 and for 1978 are shown in Table 11. The total cost of the project would be US$262 million at 1977-78 construction prices but this estimate is tentative since no adequate work quantity estimates nor a construction schedule are available for works to be carried out from 1978 onwards. The estimate includes all the required works for adequate use of the water resources and for appropriate operation. 3.44 Construction unit prices, including materials and equipment, rose more than 100% during the 10 years that elapsed since appraisal to'December 1977 as shown in the following table. However, its impact on the project cost appears to be only 80%. Unit Costs (Mex$) 1968 1973 1974 1975 1979 Excavation M3 2.58 3.47 4.16 6.15 -- Structures unit 8,176 9,299 14,099 17,420 -- Drilling well 175,682 275,285 -- 433,333 -- Pumps and Motors unit 129,635 152,222 -- 191,592 -- Cement ton 390 560 -- 626 893 Steel ton 1,995 -- 2,346 4,761 5,420 3.45 The unit cost per ha has risen 105%, from US$468/ha at appraisal to US$962/ha as of Decembe' 1977, and will rise to US$1,266/ha, or a 170% in- crease, when all project works are completed. This final cost, however, is within the cost range currently accepted for this kind of work. 3.46 As explained before, after the inflation effect, the cost increase is due mostly to underestimation of work quantities and changes in design. - 64 - Both arose because the project was not completely prepared when appraised. This situation has occurred in three other irrigation projects financed by the Bank in Mexico and doubts arise as to the procedures used by SARH (formerly SRH) in preparing engineering projects. It would seem that pres- sures are placed on the technical staff which tend to prevent them from thoroughly studying projects. Further, institutional constraints do not allow for timely revisions of planning and designs. It would also appear that the Bank had been overconfident of SRH's capabilities and did not thoroughly analyze the project nor the institutional constraints that affected project designs and supervision. 3.47 Expenditure Categories. At appraisal, US$92.8 million (98%) was the estimated cost of civil works, materials, engineering and administration categories, and US$2.2 million (2%) was the cost of equipment and relocation of farmers. In the revised cost estimate, US$259 million (99%) accounts for the first three categories and US$3 million (1%) for equipment only, since the relocation of farmers category had been assimilated by the civil works category. While the cost of civil works, materials, engineering and super- vision rose 180%, the cost of equipment rose only 110%. The following table shows the cost increases of the different categories and the amount invested as of December 31, 1977. Invested as of Total Cost to Complete Categories Appraisal December 31, 1977 the ProJect US$ M US$ M a! US$ M a/ Civil Works 73.3 156.2 210 205 280 Materials 10.5 14.4 140 20 190 Engineering and Supervision 9.0 26.0 290 34 380 92.8 196.6 210 259 280 Relocation of farmers 0.8 - - - - Equipment 1.4 2.6 190 3 210 95.0 199.2 230 262 280 a! As percentage of appraisal. 3.48 The materials category rose only 90% over appraisal. It seems that this category was accurately estimated while other civilworks were not. One of the reasons appears to be that the change in canal lining material of reinforced concrete by asphalt concrete and clay prevented further pur- chases of cement and steel. However, a higher increase of materials' cost could have been expectec because the number of structures increased con- siderably. The mission was unable to clarify this matter. Engineering and supervision category cost rose 280%. At appraisal it represented 9% of the project cost but when the works were completed, it was 13%. This in- crease, although high, has not exceeded the percentage usually applied to - 65 - projects in which engineering designs are preliminary at the time cost estimates are made which is the case in the Rio Colorado project. On the contrary, given the fact that designs for about 70% of the area had to be worked out from very preliminary layouts and that supervision of construction may have been very demanding because of the numerous contracts, the engineer- ing and supervision cost could have easily exceeded 15% of the project cost. Obviously, appraisal estimates were too low. Physical contingencies were estimated at appraisal as 20% of the project cost and it might have been appropriate if adequate preliminary designs had been available for the whole area. 3.49 Financial Resources. Out of the US$199.2 million so far invested in the project, US$174.2 million has been furnished by the Government and US$25 million by the Bank loan. Future investments of US$62.8 million to complete the project will be provided solely by the Government. Therefore, while the Government's share in the financing grew from 74% at appraisal to 87% of December 1977 and will reach 91% when all works are completed, the Bank share dropped from 26% at appraisal to 13% in 1977 and will drop to 9% of the pro- ject total cost, as shown in the following table. Bank financing was in- adequate due to underestimations of project cost at appraisal. Taking the final work quantities and appraisal unit prices, the Bank financing would have been about US$35 million. However, without the loan, the project would have not been carried out, or if it was, the pace of implementation might have been slower. Mexican Bank Loan Government 527-ME Total ---------- (US$ million) --------- At appraisal 70 25 95 As percentage (74) (26) (100) As of Dec. 31, 1977 174.2 25 199.2 As percentage (87) (13) (100) To complete the Project 62.8 - 62.8 Total 237.0 25 262.0 As percentage (91) (9) (100) 3.50 A crucial factor in the achievement of project objectives has been the improvement of the Rio Colorado water quality due to the investment made by the US Government in accordance with the international treaty and amend- ments signed between the US and Mexico. In 1974 the US extended the Wellton- Mohawk drainage canal through the Mexican territory at the cost of US$70 million. In addition, the US will build a large desalinization plant at a cost of US$200 million to comply with the water quality requirement in the - 66 - treaty without reducing water availability to the US uses 1/. The US will also take care of the operation and maintenance of both works. 3.51 A disbursement schedule was prepared in the appraisal report 2/ pro- jecting the expenditures over seven years, from 1968 to 1974. The project was to be closed by June 30, 1975. Disbursement of loan proceeds started two years late and funds were totally withdrawn three years later than the year stated in the Loan Agreement, as shown in Table 12. The actual pace of disbursement, however, was very close to the one agreed upon. The disburse- ment percentage was lowered, because of cost increases, from 35% to 16% (January 1973), when only 45% of the loan funds had been disbursed to permit disbursement of the loan over the entire execution period of the project. The loan was closed on June 30, 1977. The Mexican Government made an extraordi- nary effort in funding the project between 1973 and 1976, trying to catch up with the increasing costs. Loan Disbursements (Cumulative US$ M) 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 Appraisal 4.4 8.2 12.6 15.9 19.2 22.3 25.0 - - - Actual - - 2.1 6.9 11.3 17.4 19.5 23.2 24.8 25.0 Disbursement % - - 35.0 35.0 35.0 16.0 16.0 16.0 16.0' 16.0 Schedules and Delays 3.52 At appraisal it was proposed that construction works start in 1968 and be completed by mid-1974. Due to the initial delays in starting the works (para 3.01 and 3.02) and to the limited funding that could not offset the effects of inflation (that caused increases in construction costs), construc- tion progress was slower than expected at appraisal. Furthermore, design changes and increases in work quantities caused additional delays. Table 13 and Chart 2 compared both the investment and the construction schedule at appraisal with the actual investments and construction progress. The break- down of the actual investments between 1968 and 1978 is given in Table 14. 1/ .Date of Work Cost Completion (US$ million) Wellton-Mohawk canal USA territory 16 miles 24 1973 Mexican territory 37 miles 56 1974 Desalinization plant 190 to be built in 1980 Total 270 2/ Appraisal Report, para 4.16. - 67 - 3.53 As of December 1971, project investments were US$43.7 million in- stead of the US$58.7 million anticipated at appraisal and physical progress was 17%, compared with 63% at appraisal. In this period, delays and low funding were the main causes of the little progress made. Between 1972 and December of 1974, the date on which work should have been completed according to appraisal, project investments were US$124.8 million as against US$95 mil- lion at appraisal. Nevertheless, physical progress was only 49% instead of 100%. Again, in this period, inflation and increase of work quantities were the main constraints. From 1975 to December 1977, project investments were US$199.2 million, or 110% higher than the appraisal total cost estimate, but only 72% of the works were completed. (If compared with appraisal work quantities, progress would have been 86%). As in the previous periods, higher work quantities and inflation were the causes of the limited progress between 1975 and 1977. In 1977, investments dropped substantially, from over US$20 million/year, an investment that was maintained since 1970, to about US$11 million. The reason was that, with the exhaustion of loan funds that were used by SARH as a leverage to obtain Government funding, and, having completed most of the major works, SARH felt that the remaining works could be under- taken progressively over the following years with smaller funding provided solely by the Government. For 1978, an allocation of US$8.6 million has been made and works will be 76% completed by December 1978. 3.54 With the works so far completed, the project can operate satis- factoril as long as availability of water remains at its present levels of 2,660 Mm /year (para 3.15). As municipal demand increases (Mexicali and Tijuana cities) and pumping from the project aquifer diminishes, all the remaining works (para 3.26) should be carried out to assure adequate supply of water, higher conveyance and application efficiencies, and better water quality. Although no assurances from the Government have been obtained that the remaining works will be completed by 1985, for the purpose of this analysis, we have assumed that the remaining US$54.2 million necessary to complete the aforementioned works wiT1 be invested from 1979 to 1985, when total water 3 supply would be 2,500 Mm /year. From 1985 to year 2000, when only 2,370 Mm / year will be available, operation efficiencies of the main systems will have to be improved through better overall water management and application effi- ciencies will have to be increased by the use of more advanced irrigation methods and by the provision of on-farm drainage where required. If the Govern- ment does not continue financing, then irrigation of the whole project area could be impossible. 3.55 The implementation schedule at appraisal was optimistic. Even if the work quantities had not increased and no inflationary effect had been present, project works would not have been completed by 1974 because of the size of the works and institutional constraints. Neither would overall water use efficiency been improved since the many factors affecting it (such as cropping patterns, date of planting, zoning, farm water deliveries and farm water applications) are, and will continue to be for some time, difficult to control. It is worth mentioning that in no project in Mexico has there ever been a thorough program produced or at least a clear bar chart or a CPM (critical path method) analysis, least of all a PERT, to show the sequences and timing of actions and their interrelationships. - 68 - Exchange Rate Adjustments 3.56 Loan 527-ME was disbursed in 14 different currencies. As of December 31, 1977 the debt outstanding - after amortization and before currency adjustments - was US$21.6 million. Because of variations in the exchange rate of the different currencies, the debt outstanding as of that date increased in US dollar terms by US$5.4 million, bringing the total debt outstanding - after amortization - to US$27.0 million. In terms of local currency, the original loan was the equivalent of Mex$312.5 million. The Mexican Peso vnderwent a sharp devaluation in 1976, but it fell from US$0.080 before the devaluation to about US$0.045 by December 31, 1977. Thus, as of this last date, the oustand- ing debt, after other currency exchange adjustment and amortization, totalled Mex$594.0 million. Special Conditions 3.57 None of the special conditions included in the Loan Documents were fully complied with. The degree of compliance of the special conditions by SARH is summarized in Table 15. IV. AGRICULTURAL IMPACT Initial Project Goals and Constraints 4.01 The project was carried out as part of a national program to promote better utilization of available water in publicly operated irrigation districts. The principal aims were to improve the District's land use and agricultural productivity through prevention of salinity and water losses, combined with an improvement of farming practices and crop diversification and to expand the area under irrigation from 176,000 ha to 203,000 ha. The introduction of some seasonal crops would permit double cropping in 12,700 ha, bringing the irrigated crop area per year to 215,700 ha. The livestock industry would benefit from the production of feed grains and improved pastures, especially in the fattening of beef cattle and in the growing dairy industry. The in- troduction of oil-bearing crops was in line with Government policy. 4.02 The main constraints for the increase of production and productivity in the project area were the shortage of water caused by the deteriorating water distribution system, the salinization of soils and the lack of adequate technical assistance services for farmers. The winter water allotment to Mexico, under the existing international water agreement, forced the expan- sion of winter crops (wheat and forrages) to utilize the water (which other- wise would not have been utilized beneficially). Further, the very high summer temperatures forced a change of planting time for certain crops to avoid the harmful effect's of the heat on the plants and the seasonal increase of salinity (due to high evaporation and shortage of water for leaching). - 69 - The Beneficiaries 4.03 According to the appraisal report, in 1967 there were 11,785 farmers established in the project area, of which 6,948 were ejidatarios and 4,837 were colonos or private farmers, occupying a total of 196,780 ha. At that time also, several hundred farmers who had been granted water rights were scattered around and outside the District borders, thus making it difficult to control the use of their water rights. Excessive water losses occurred as a result. As a solution a reallocation program was undertaken by which these farmers were gradually reallocated within the District, occupying idle land and bringing it into production. The reallocation program is expected to be completed during 1978. Including the farmers involved in the re- allocation program, the total number of farmers established in the District will be 12,355, of which 6,727 are ejidatarios and 5,628 are private farmers, covering 207,000 ha, and all of them have benefitted or will benefit from the project through the improvement of the irrigation facilities and tech- nical assistance programs. The change in land tenure and area is as follows: Year Ejidatarios Colonos Total Number ha Number ha Number ha 1967 6,948 114,824 4,837 81,956 11,785 196,780 a/ 1977 6,727 123,008 5,628 84,012 12,355 207,000 Change -222 +8,184 +791 +2,056 +570 +10,220 a/ Water rights 203,000 ha. 4.04 No selection of farmers was necessary since the project objective was to provide improved irrigation and drainage facilities for the farmers who possessed land and already recognized water rigths. Some of these farmers, however, have sold their land to third parties, generally landless farmers. Technological Changes 4.05 Among technological objectives to be realized through the project were (a) high efficiencies in water management, and (b) stoppage or amelio- ration of soil salinization. To achieve the first target, extensive canal lining and numerous control structures were provided and about 90% of the area was leveled. However, the desired water conveyance and application efficiencies assumed at appraisal have not yet been reached and if they are to be some measures and procedures not foreseen at appraisal have to be adopted (such as better coordination of water supply by gravity and wells, the imposition of irrigation schedules, the organization of irrigation by zones, and the use of m9re advanced irrigation technologies such as sprinkler systems). Conveyance efficiencies have increased substantially due to the lining of canals but operational water losses remain high because of the on- demand water distribution. It will take some time and much effort to raise water management efficiencies to the desired level and therefore actions should be taken right away. Farmers are already used to cultivating leveled - 70 - land and some of them are now re-leveling. About 20,000 ha would be re- leveled in 1978. Water application efficiency appears to be 70%. However, to be able to irrigate all the project area in the long run, farm efficien- cies have to increase to at least 80% (para 3.16). Already some farmers are using sprinkler systems, and drip irrigation is being tested in the vineyards. The District will also soon install two pivotal systems on an experimental basis. 4.06 Successful experience has been gained in soil salinity as far as leaching soils and making land productive (paras 3.18 to 3.23). Extensive areas of soil that were highly saline are now under cultivation. However, the accumulation of salts in the soil profile is a matter for concern. Soil sampling and many physiochemical tests have been made but investigation has gone nowhere since the salt problem has not been fully identified. Work is half complete and an interpretation of data and trials for corrective measures are lacking. Steps are being taken to determine the nature of these salts and their harmful effects on plants but until accumulated data is interpreted and corrective actions have proved to be efficient, no firm assumptions about soil salinity evolution can be made. It should be borne in mind that, had the salinity of the Colorado River water not diminished, the soil salinity problem would have not 3receded as fast as it did. With the present water allocations of 9,100 m /ha (at farm gate) in the gravity irrigated area, it is possible to raise an annual3crop and spare some water for leaching. The excess of water, about 1,000 m /ha over the average farm irrigation needs, permits the lowering of the salt concentration near the plant and the washing away of,a portion of the salts, regardless of the irrigation methods and the imperfect drainage. The apparent irrigation efficiency is about 70% if the water for leaching is included and 60% if it is not. However, as the water allocation at farm gate will decrease in the future, irrigation methods have to be improved to save water, and adequate on-farm drainage has to be provided to readily carry the excess water with salts to the drainage system. 4.07 The appraisal objectives were too ambitious both on conveyance and application efficiencies and on the speed and extent of the saline soils reclamation. In the first case, because of the difficulty of subjecting the whole area to irrigation by zones and schedules to obtain 80% conveyance effi- ciencies,. pipes (close conduits) would have to have been used in lieu of canals, and, in the second case, because of the quality of water of the Colorado River ,prior to the signing of "Minute 242" (not envisaged at appraisal), soils reclamation should have appeared more difficult. 4.08 On the agronomic side, although it cannot be exclusively attributed to the project, a gradual technological change has taken place in the project area over the last 10 years. The evolution of the region's overall activities on the one hand, and the permanent although slow growth of technical support to farmers on the other, have had an impact at farm level diversifying the cropping pattern, expanding the cropping area, increasing productivity and upgrading the farmers' attitude and performance. Over the years, agricultural research activities have undertaken new programs and the extension services have been enlarged and improved and this has caused the introduction of new - 71 - crops and farming practices. At the time of appraisal cotton, wheat and barley accounted for 170,000 ha of the 176,000 ha cropped. At present, there are large areas planted with oil-seeds, alfalfa, rye grass for cattle fattening, vineyards, asparagus, and various vegetables for fresh marketing. The use of farm machinery is generalized and so is the use of improved seeds, fertilizer and chemicals for pest controls. Special emphasis is given to irrigation methods, to soil management under irrigation conditions and to proper planting dates. In general most farmers have become more aware of the relevant tech- nical problems in farming and more careful in handling them. This general technological change is reflected in the yields presently obtained in the project area, which, in some cases, have already reached the levels projected at appraisal for the full development stage (1985). Taking into account the on-going arrangements to cover all farmers with technical assistance services (para 6.08), it is expected that the technological change experienced in the area will make even more progress over the next years. Agricultural Production 4.09 Expansion of Cultivated Area and Cropping Intensity. In the year of project appraisal (1967), 178,000 ha were actually cropped and irrigated, which represented 88% of the 203,000 ha of the then estimated total irri- gated project area. Over the following six years, the irrigation system further deteriorated and the area irrigated was reduced to 168,000 ha in 1973. From this year, as the project works advanced and were put into use, the area under irrigation increased rapidly, reaching the present level of almost 203,000 ha, or 98% of the now estimated maximum of 207,000 ha irri- gable area (para 3.08). Originally, it was estimated that the maximum area under irrigation (set by the water rights granted to farmers) would be 203,000 ha, of which about 12,700 would eventually be double cropped. At present, both these estimates have been revised. Due to changes in the District's border line and in water availability, it will be possible to irrigate a total of up to 207,000 ha, but no double cropping is presently permitted or will be permitted in the future because of the limited avail- ability of irrigation water. This provision from District authorities is based on the present and future water availability, on the need to keep the presently existing 20,000 ha of perennial crops properly supplied with water and on the need to conserve the water underground reserves. The expansion of the cultivated areas from 1969 to date and the future expansion up to 1985 are shown in Table 16. 4.10 Cropping intensity at full development was estimated at appraisal as 1.06 over 203,000 ha physical, not considering the 8,700 ha of perennials as double cropping. The revised cropping intensity at full development is 1.1 over 207,000 ha physical but considering the 20,000 ha of perennials as double cropping (Table 17). As water resources will be less than estimated at appraisal (para 3.12)%, a valid comparison of the impact of the project should be based on the relation of water availability to the total areas cropped at full development and considering perennial crops as double cropping, as shown below. Such comparisons reveal that the revised cropping intensity is 9% higher than that assumed at appraisal because of a better use of the water resources than was originally assumed (para 3.15 and Table 5). - 72 - Unit Appraisal Revised Change % Water Available Mm 3/year 2,511 2,370 Physical Area Ha 203,000 207,000 +2 Cropped Area a! Ha 220,400 227,000 +3 Cropping Intensity 1 1.09 1.1 +1 Cropped Area Equivalent b/ Ha 220,400 240,500 +9 a/ Including perennial crops as double cropping. b/ Considering perennials as double cropped. 4.11 Evolution of Cropping Pattern. Presently, about 60% of the area is planted with winter crops and 40% with summer crops in order to make the best use of the Colorado River water which, according to the treaty is delivered in fixed quantities each month 1/. Under this constraint, sub- stantial changes have taken place in the area's cropping pattern since the project was appraised. In 1967, 120,000 ha, or more than 70% of the total area cropped, was -planted with cotton, because farmers were encouraged by very high prices and readily available credit from firms operating the ginning plants. Then, as the international price for cotton declined, the area planted fell to 40,000 ha in 1973, while grains and oil-seeds occupied a steadily growing area 2/. Simultaneously with these changes, the area under alfalfa grew from 7,500 ha in 1967 to almost 20,000 in 1978. In 1973, following successful experiments in the region and attracted by the good prospects for the cattle fattening business (especially for exporting steers to the US), farmers started to introduce rye grass as a winter grazing re- source. From 6,000 ha in 1973, the area occupied by this pasture increased to 11,000 ha at present. In 1978 soybeans will be introduced for the first time in the project and it is expected that about 1,900 ha will be planted. Corn has been cultivated for a long time but summer heat affects its yields. While the planting time is occasionally advanced to avoid the warmest months, new varieties, presumably resistant to the heat, are being tested. Areas of vegetables have increased and vineyards are being planted, although on a small scale. The two latter crops have a high potential because they are suitable to the climatic conditions of the area. However, the relatively free changes farmers make on their own cropping patterns that take into account, to a certain extent, the national needs but especially the profitability of different crops, will have to be subjected to tighter control by the District authorities. Given the area of the District, the availability of water, the water rights granted to farmers and the water required by the crops, the District authorities will have to grant planting permissions in a balanced 1/ The treaty set minimum and maximum amounts of water to be delivered to Mexico each month. 2/ The reduction of the area of cotton was envisaged at appraisal as a measure to make more economical use of water. However, the reduction of the area was due mainly to market constraints than to the enforce- ment of SARH policies. - 73 - way in order to (a) assure the water supply for the area under perennial crops, including alfalfa which basically supports the dairy industry; (b) allocate the highest possible water amount to the most profitable crops, including cotton; and (c) distribute the balance of water available for crops requiring less irrigation water. Table 17 details actual and projected cropping patterns, and Table 18 shows the expected changes at full development. 4.12 Incremental Physical Output and Yields. Although a number of factors have contributed to the increment of farm output in the project area (para 4.18), the basic one is the increase of the area cropped and yields made possible by the regularization of availability of irrigation water and by soil leaching resulting from the project works. Table 19 compares yields projected in the Appraisal Report with yields actually achieved at the present stage of development. It shows that yields projected at appraisal for the full development stage have already been obtained for some of the most important crops such as cotton, wheat and safflower and that goals for others have almost been reached. In general, yields are expected to further improve in the future under the excellent natural conditions in the project area and as a result of a substantial improvement in technical assistance services and a higher general efficiency in the use of water. 4.13 Physical production has already substantially increased as shown in Table 20 and it is firmly expected that it will continue to increase to reach the revised productive targets. It can be seen in the table that winter crop (mainly wheat and barley) production has risen from about 140,000 tons in 1967 to 385,000 tons at present and that, by 1985, it will reach nearly 600,000 tons. Summer crop physical production will reach about 300,000 tons by 1985, or almost as much as in 1967. This is due to the reduction of the cotton areas which occupied about 68% of the area in 1967> and that will occupy only 24% by 1985. The all-time low production of summer crops in 1972 was due to the drop of cotton prices (para 4.11) and the acute salinity conditions of the soil (Annex 1). Production of peren- nial crops, where alfalfa is the most important, has steadily increased, from about 46,000 tons in 1967 to 244,000 tons at present, and will reach about 330,000 tons by 1985. 4.14 Although at appraisal, it was considered important to seek to support livestock future development through long-term credit, beef production was not considered as significant and it was not included in production eval- uation and projections. In 1972 and 1973, as a result of previous successful research and experiments, rye grass pastures were introduced in the cropping .programs of the District, strongly influenced by the attractive prices the neighboring US market was offering for both feeder and slaughter cattle. In 1974, almost 16,000 ha were planted with rye grass, which, with a grazing period of 130 to 160 days, produced a net liveweight gain of 660 kg/ha. In later years, both prices and area under rye grass declined somewhat and this industry has stabilized at a level of about 11,000 ha, grazed by 65,000 to 75,000 feeder steers that come from the range lands (agostaderos) of the neighboring states of Sonora and Chihuahua. Following the development of the rye grass-based fattening business, the feedlot fattening industry also - 74 - developed for the final conditioning of steers. The prospects for this industry lie basically in improvement of pasture/herd management and of herd quality. 4.15 The development of the livestock industry in the project area shows some interesting aspects and innovations since appraisal. Production of milk has increased, from 35 million liters in 1970 (first year with statistical information) to 62 million in 1977, supported mainly by alfalfa, complemented by some other forage crops. Milk production by 1985 will drop from its present level to 59,000 -tons, or about 10%. Due to future limitations in -irrigation water supply, any expansion of milk production can only be expected from better yields for alfalfa and higher production of milk per cow. 4.16 Variation of Prices. Prices of the crops cultivated in the project area deteriorated rapidly in the first three years of the project and remained at low levels until 1973, when they started to grow rapidly upto 1977, as shown in Table 21. With the exception of 1969 and 1973, unit average produc- tion costs followed approximately the same tendency of prices. However, the rebound in prices was followed by costs with a year lag and at a slower pace, as shown in Chart.3, thus improving the price-cost coefficient for the farmers. Regardless of the price increase trend, market prices for grains (barley, maize and sorghum) and for oil-seeds (safflower and soybeans) remained below the CONASUPO's support prices 1/. Prices of forages, vegetables and grapes are farm-gate prices. In the other cases, the price includes transportation to the plants or collecting centers. 4.17 Production Value. The increment in net value of production is quite significant at the present stage of development and subtantially for the projected period. In fact, net output value (at 1977 prices) increased from Mex$1,275.8 million in 1968 to Mex$1,706.2 million in 1977 and is expected to reach Mex$3,080 million in 1985, as shown in Table 22. Product- ivity is expected to increase from 1967 to 1985 by 47%, as shown in Table 23, which presents, as key indicators, the actual gross and net value of pro- duction for the project area-and per hectare for 1967 (year of appraisal) and for 1973 and 1977 as well as for the Appraisal Report projection and recently revised projection, both for 1985. 4.18 - The increase of physical production and the increment of the net value of production until the present stage can be attributed to: (a) the expansion of the cropped,area (from 168,000 ha in 1973 to 202,600 ha in 1978); (b) the improvement of yields, resulting from the regularization of irrigation water availability derived from the project; (c) the availability of agricul- tural credit for production inputs and livestock; (d) the supporting services for farmers; and (e) the favorable price/cost ratio. Future increases are expected to result from the combination of: (a) the incorporation of about 4,000 ha (to reach 207,000 ha physical); (b) obtaining the maximum avail- ability of water by completing project works; (c).the further improvement of water conveyance and application efficiencies; and (d) the effect of the expanded technical assistance programs. 1/ For the IRR, economic prices were used for wheat, beef and cotton and local market prices for the other commodities, both at constant 1977 prices in Mexican Pesos (Table 19). - 75 - Employment 4.19 The last two population census were taken in 1960 and 1970. Between these two dates, the total population in the valleys of Mexicali and San Luis Rio Colorado increased from 323,467 to 459,928, while the active population in the primary sector went down from 53,368 to 39,595. This trend seems to have continued in the following years. Estimates of rural population for the period 1970-76 for Mexicali shows an average decrease of 7.4% per annum. This appears to have been caused mainly by the decrease in the cultivated area before the 1970s and the cotton crisis in the early 1970s. There is no information on the present active population and employment for the project area, but the Mexicans presume that the employment generated by the now larger areas being cultivated must have stopped the past trend and that the demand for labor will grow by about 10% as the project reaches its full potential. 4.20 The total labor requirements for the project area were calculated for the years 1967 and 1977 based on the actual cropped area and for 1985 based on the revised cropping projection. For such calculations, the labor requirements per ha attributed to each crop were those provided by the District authorities and they compare reasonably well with those obtained from other Mexican regions. The reduction of labor requirements in the area originates from the dramatic reduction of the area planted to cotton, which was 121,000 ha in 1967, 62,000 ha in 1977 and is expected to be only 50,000 in 1985. Wheat and barley, which require about 10 man-days per ha per year, have mostly replaced the area previously planted to cotton, which requires over 45 man-days per ha per year. On such basis, the estimated labor require- ments are as shown below: Gross Average Year Area Cropped Total Labor Requirement Man-Day/ha 1967 178,181 5,400,000 to 7,300,000 36 c/ 1977 199,002 4,000,000 to 5,500,000 a/ 24 c/ 1985 207,000 4,100,000 to 5,780,000 b/ 24 a/ The "Evaluacion de Resultados" report of SARH considers 7,300,000 man-days. b/ The "Evaluacion de Resultados" report of SARH considers 8,600,000 man-days. c/ Assuming full mechanization. 4.21 Assuming that 30-ha farms or larger will not use family labor for agricultural work and that 70% of the required hand labor on farms under 30 ha would be supplied by the farmers' families, the employment opportunities would increase by about 55,000 man-days/year between 1977 and 1985 as follows: - 76 - Year Farm's Area Total Area Hired Hand Labor (ha) (ha) (man-day/ha) (total man-days) 1977 <30 8,214 24 197,136 >30 190,788 7 1,335,516 199,002 8 1,532,652 1985 <30 8,214 24 197,136 >30 198,726 7 1,391,082 207,000 8 1,588,218 Equity Issues 4.22 The land tenure pattern did not undergo significant changes due to the project. In 1967, before the project, 11,400 of the total 11,800 farmers owned 20 ha or less, with 8,500 of them concentrated in the 10-ha to 20-ha bracket. In 1977, the proportion is about the same with only a slight increase in the total number of farmers, due to the reallocation program. Another slight change is the average area per farm, which increased moderately in the case of the ejidatarios and decreased for the private farmers, as shown below: Ejidatarios 1967 1977 Total Average Area Total Average Area No. Area Per Farm No. Area Per Farm (ha.) (ha) (ha) (ha) 6,948 114,824 16.5 6,727 123,008. 18.3 Private Farmers (Colonos) 1967 1977 Total Average Area Total Average Area No. Area Per Farm No. Area Per Farm T(al (ha) Ta) (ha) 4,837 81,956 16.9 5,628 84,012 15.0 All Farmers 1967 1977 Total Average Area Total Average Area No. Area Per Farm No. Area Per Farm (ha) (ha) (ha) (ha) 11,785 196,780 16.70 12,355 207,020 16.70 - 77 - 4.23 Benefits from increased production have actually benefited the farmers in general, although in different proportions. As described at appraisal, many private landowners had at that time higher incomes than the ejidatarios because they operated bigger farms on average, had more financial resources, and used more modern techniques. Nine-tenths of the farms with over 20 ha were privately owned and more than 70% of the farms of 5 ha or less were operated by ejidatarios. With the project, although large farmers have higher total incomes because of the size of their holdings, incomes per ha are now the same all over the area as a result of project works, credit, extension services and other help made available to the farmer through the project. Since there are no statistical data, a subjective observation is that farmers with less than 10 ha, who occupy 21,000 ha, or 10% of the area, have benefited the most since, on their own, they would not have reached their present stage of development; farmers in the bracket of 10 ha to 20 ha, who occupy 166,000 ha, or 80%, have also benefited but they could have accomplished development if the project had not been undertaken; and, farmers with more than 20 ha, occupy- ing 20,000 ha or 10%, have benefited the least. 4.24 Most of the crops grown in the project area are subject to Govern- ment-set support prices and farmers deliver their products directly to CONASUPO's storage facilities (para 3.32). Milk is sold to dairy plants at a Government-set official price, and cotton is grown under contract between the producers and the ginning companies. Therefore, benefits from increased production are not captured by middlemen or other agents. On the other hand, the farmers have organized themselves into cooperatives or marketing groups (such as in the case of livestock) through which they get an extra income. 4.25 Ownership of farm machinery, however, does not appear to be evenly spread. From SARH statistics on the subject it, is clear that many farmers, especially ejidatarios and very small colonos, do not have enough tractors and implements of their own and thus rent machinery services. In 1967, the fully mechanized farms covered 161,000 ha and the partially mechanized farms covered 12,000 ha. There were 3,000 ha of farms without machinery, whereas, by 1976, fully mechanized farms covered only 153,000 ha and the partially mechanized farms covered 51,000 ha (see para 3.29). In the matter of equip- ment ownership, there are two things to consider: (a) that many farmers prefer to use custom rental machinery instead of having their own when the size of their individual plots cannot justify such equipment; and (b) that it appears that equipment is very heterogeneous and, in some cases, not very well taken care of, which forces the renting of equipment. However, no shortage of machinery is felt in the area. Secondary Goals and Unforseen Effects 4.26 There are some interesting aspects concerning the farmers' perform- ance that should be pointed out, although it is not possible to determine whether or not they can be considered as effects derived from the project. In general, the farmers in the District are fairly good according to Mexican standards. This is reflected in the use of technical inputs, their - 78 - willingness to accept technical guidance from extension agents, their ini- tiative and imagination in creating different kinds of organizations of their own (for maketing, input supply and others), and their ability to obtain and use credit. These are all signs of a certain know-how and entrepreneurial capability which appears to be a solid base for a further developed and stable rural community. A good example of this enlightened attitude is the creation of "Association of Commercial Agencies" among farmers (Asociacion de Sociedades). In several sectors of the District, farmers have organized a number of commercial companies or agencies for inputs supply, tractor rentals, crop airspraying and the like, for the benefit of any farmer (ejidatario or private) who wants to join as a member-and use their services. These agen- cies, in turn, are merged into an association headed by a board formed by representatives of the base agencies, which coordinate all the activities and, businesses on behalf of and for the benefit of all members. Stage of Present Development 4.27 In order to evaluate the impact of the project it is necessary to take into consideration that the project has not been fully completed. Loan funds were exhausted by January 1977 but project works will take about eigth more years to reach completion, i.e., by 1985 1/. Meanwhile, agricultural development will continue to follow at the same pace since project works would be immediately put into operation and used by the farmers. It is assumed that the full stage of agricultural development and production will be reached also in eight years from now. Agriculture has evolved very favor- ably as has been explained in the preceding parts of this report. Yields have surpassed the ones assumed at appraisal while the original target area of 203,000 ha will be reached this year (1977/78) 2/. The impact of the present stage of development can be judged, to a certain extent, by the internal rate of return (IRR) of the investments made up to date, assuming that the remaining investments will not be made and that the water supply will decrease to its final limits (para 3.12) 3/. Calculations give an IRR of about 14%, which compares favorably with the 11% estimated at appraisal and with the 14.6% of the ex-post calculations for full development (para 5.02). 1/ By December 31, 1977 project works were 86% completed if compared .with appraisal and 72% completed if compared with the revised estimates. 2/ Table 22, Gross Production Value as of 1977 is 75% of the revised target at full development and incremental Net Production Value, 32% of its revised target. 3/ Present ivailability, 2,660 Mm /year; final availability (year 2000), 2,370 Mm /year. - 79 - V. RATES OF RETURN 5.01 The appraisal report estimated an internal rate of returo-_of 11% and it was felt in the Bank that it was on the low side. 1/ However, it appeared to be compensated by the great certainty that the return would be realized, since the project involved merely rehabilitation of an existing set of irrigation and drainage works serving an already developed area. The benefits as shown in Annex 1 included (a) the incremental net value of production between 176,000 ha, that was being cultivated at the time of appraisal and 215,000 ha that was expected to be cultivated with the project at full development, and (b) the net value of production attributable to the replenishment of water lost (in the aquifer) and to the prevention of soil salinity. It was esti- mated that, without the project, there would have been an annual reduction in productivity in the area because of increased soil salinity in about 128,000 ha and a permanent reduction of 11,627 ha because of the need to stop over- pumping the wells to preserve the aquifer in the long run. The prevention of these reductions in output was an important component (50%) of benefits. However, there are only two footnotes in the Appraisal Report which partially explain the way in which these last two benefits were calculated. The pro- posed reduction of the areas devoted to cotton, which was not to be replaced by high value crops but rather with low value crops such as forage and grains, affected the rate of return. On the cost side, the analysis included the estimated project cost plus on-farm drainage, works and soil treatment on 18,000 ha and an increase in operation and maintenance expenditures. 2/ The costs of hydrogeologic studies and soil sampling and of the irrigation extension services seem to have been included in O&M costs, but costs for the enlargement of agricultural extension services were omitted. 5.02 The recalculated economic rate of return, is about 14.6%. Table 24 shows the stream of benefits, which include the incremental net value of production over a stable cultivated area without the project. The project benefits started to be felt in the 1972/73 campaign (Annex 1) and it is estimated to reach full development by 1985, as orignally projected at appraisal. Without the project, it is assumed that the area cultivated could have remained at 172,000 ha, that is, the area cultivated in the 1971/72 campaign, because a decrease in productivity due to the increase in soil salinity was not likely to happen since the irrigation water salt content has dropped to acceptable limits (para 3.17). At present, the physical area that will be irrigated at full development is 207,000 ha while at appraisal, the water rights were for 203,000 ha. Of the 207,000 ha, it is projected that 20,000 ha will be cultivated with perennial crops, while in the appraisal it was expected that about 12,000 ha would have been under double cropping. Double cropping is not presently permitted in the project. As 1/ The bank staff was'also somewhat puzzled by the fact that the calculated economic return was a little lower than it was for the Lagunera Project, notwithstanding the fact the Mexican and Bank experts seem to regard the new project as more promising than Lagunera. 2/ Appraisal Report, Chapter VII and Annexes 8 and 9. - 80 - mentioned above, it is assumed that the cultivated area could have remained stable without the project since there is no definite evidence to support the assumption that it would have decreased because of soil salinity, especially after the international agreement was reached on water quality. Over-pumping is expected to be gradually reduced over an adjustment period. Water manage- ment technical assistance to farmers and on-farm investments are expected to improve the efficiency of the system to compensate for the reduction of water availability in the long run (stoppage of over-pumping and increase of water supply for Tijuana and Mexicali cities). .5.03 On the cost side, it is assumed that the remaining project works will be gradually carried out and fully completed by 1985. 1/ It is proposed to invest in non-farm drainage on 100,000 ha and water conservation works 2/ on 140,000 ha both to be done over a 20-year period, starting in 1985. The maintenance program projected includes well replacements. All the above investments and maintenance expenditures will secure the productive 6apacity of the area and guarantee the incremental water supply for Mexicali and Tijuana city needs. 5.04 A sensitivity analysis shows that if a drop of 30% in the incre- mental net benefits is experienced, the economic rate of return will be reduced to about 11%, which is at present the opportunity cost of capital in Mexico. 3/ 5.05 The more favorable economic return expected now from the project is a combination of several factors: a larger area to be cultivated - facilitated in part by a higher safe well pumping volume - and favorable crop output productivity and improvement in price-cost of production coefficients. While the underground water maximum safe pumping was estimated at appraisal to be 700 Mm3, the mission's review of the studies has revealed that the safe level of pumping is 900 Mm3. In the recalculated rate of return, no account has been taken of the direct benefits of securing the additional water supply for the cities of Mexicali and Tijuana. It must be noted here, that the project's success had been greatly increased by the 1972 and 1974 international agreements that guarantee a minimum water quality to the project. Without these agreements, the infrastructure that was being built may not have solved the problems of the area and the economic rate of return might have been substantially lower. 1/ Including 70 new wells in San Luis area with the necessary connection canals, rods and power line. 2/ Pipe conduction, sprinklers, pivotal or dripping systems. 3/ Memo of J. A. Holson, February 8, 1977. - 81 - VI. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT Institutional Design 6.01 Since the project's inception until December 1976, SRH was in charge of the physical execution of the project while the Secretaria de Agricultura y Ganaderia (SAG) was responsible for agricultural aspects. This separation of responsibilities worked against coordination in the two basic aspects of the project - physical and agricultural implementation. The Bank supervision missions repeatedly expressed concern for the poor performance of the agri- cultural extension services, and only in the last years of the project was a very substantial improvement observed. This duality adversely affected also the collection of data and reporting on the progress of the project. 6.02 The new Government that took office on December 2, 1976, promul- gated a new Organic.Law of the Federal Public Administration later in the same month. This law established the basis of the Central Government administration and the semi-autonomous entities. It merged the former SRH and SAG in a new Secretariat of Agriculture and Hydraulic Resources (SARH) with the functions and responsibilities of the former two. The by laws of SARH were issued in August 1977. SARH is assisted by four sub-secretaries - Agriculture and Operations, Livestock, Hydraulic Infrastructure, Planning, Forestry and Fauna - one administrative department and one coordinating office for integration and development of the entities and programs in the sector. SARH at the state level is headed by a General Representative, who in the case of Baja California is a Doctor in Agricultural Sciences. He is assisted by eight department chiefs: Planning, Hydraulic Programs, Agricultural Program, Livestock Program, Forestry and Fauna Program, Financing and Administration, Irrigation Districts and Rainfed Districts. The state delegation of SARH, not including the irri- gation and rainfed districts have a total staff of 1,136, of which 251 are university graduate professionals and 398 intermediate level professionals. The Rio Colorado Irrigation District employs 768 people, of which 50 are university graduates and 97 are intermediate level professionals. 6.03 It is early yet to judge the performance of SARH, but it is expected that coordination and communciation within the agricultural sector will improve. The newly created coordination office would play an important role to this end. Further, the office of the Sub-Director General for project appraisal within SARH has been strengthened. This office is preparing a post-appraisal report on Rio Colorado (para 1.05). The mission discussed its preliminary findings with it and reviewed the draft. The Sub-Director General could be very useful in future for project monitoring performance, since he intended to keep records on all aspects of the project to facilitate post- appraisal. 6.04 The project did not contemplate institutional building requirements because it was assumed SRH had sufficient experience and trained personnel to successfully implement it. While this perception was basically correct, there was ample room for strengthening SRH internal control and communication. - 82 - The mission was faced with a number of contradictory data between different offices in Mexicali SARH delegation and between this and SARH offices in Mexico City, on important matters such as cost and investments, amount of work performed and water use. Further, through a review of project files and from experience in other ongoing projects financed by the Bank in Mexico, it appears that SRH was not and that SARH is not yet able to plan and supervise projects satisfactorily. In all the projects, and especially in the Rio Colorado, it has been difficult for the Bank to obtain, from the engineering department, reliable figures on work quantities, cost estimates and works scheduling. The inconsistencies of work quantities given in the reports (quarterly and supervision) indicate that no planning had taken place and that unified criteria at decision level did not exist. 6.05 At project level, the rehabilitation works were the responsibility of the Rehabilitation Section headed by an engineer under the former SRH manager and since 1977 under the SARH representative. Its performance is discussed in Chapters III and IX. Regarding operation and maintenance, the situation is more promising, although not satisfactory. The Rio Colorado District is organized into six units, each of which is under the direction of an engineer. These units are parcelled into 85 sections, each with a ditch- rider in charge. Efficient communication by road, telephone and radio networks exist, but water measurements and water deliveries and records still need further improvement. Water charges collection is made either through the banks that furnish agricultural credit to farmers or directly at District offices. Although all farmers pay, the correct amount is' apparently not charged and this is reflected in the low level of collections (paras 6.15, 6.16 and Annex 2). Budgetary practices are very confused and the mission was unable to determine what the costs of the different district's activities were (operation, PLAMEPA, technical assistance, maintenance and so on). Supporting Services 6.06 At the time of appraisal, the agricultural research and extension services under SAG (Secretariat of Agriculture and Livestock) were minimal. Some modest research programs were carried out in one experiment station near Mexical. and extension services had only five agents for the whole district. However, the lending institutions and the cotton ginning companies provided important technical support to farmers whose operations they financed. The Appraisal Report contained no quantification on how the technical services would be expanded through the project, but it stated that SAG had drawn up a comprehensive plan for such services for the Rio Colorado District, based on the cooperative effort of several agencies, and that the work program would be coordinated by the State Secretary of Agriculture (State of Baja California). During negotiations, assurances were obtained that the Government would con- tinue to provide and strengthen these technical services planned by SAG and SRH. The side letter of January 26, 1968 stated that the Government would take all necessary action to provide agricultural research and extension services to the farmers of the District including experimental farm investi- gations, small farm technical guidance, livestock promotion and cooperative extension services. For this purpose, the activities of the Instituto Nacional de Investigaciones Agricolas (INIA) were to be increased and at least 14 extension officers were to be provided by the beginning of 1970. - 83 - In addition, the services of seed inspection and certification and of plant pest and disease control measures were to be strengthened. SRH was also going to enlarge its technical assistance to farmers in relation to irrigation and drainage practices, improving its laboratories and strengthening its staff and its activity in demonstration plots. The extension and research services in the project area were to be reviewed from time to time and modified if necessary by mutual agreement between the Guarantor and the Bank. 6.07 In reality, it took a number of years to create a technical support system properly geared to-farmers' needs. In 1968 an important addition was made by the establishment of PLAMEPA (Plan de Mejoramiento Parcelario), which concentrated specifically on helping farmers to improve their irrigation practices in order to make more efficient use of water, control salinity and improve yields. PLAMEPA grew steadily over the years. While in 1968 it had five engineers, two topographic brigades, an a budget of Mex$ 2.2 million, it now has 20 engineers, six topographic brigades, 25 extension agents and soil experts, soil laboratories and support personnel, financed by a budget of Mex$ 40 million. By 1971, the agricultural extension services were organized to operate in close coordination with PLAMEPA, and in 1977 the two services were merged into one called "Supporting Brigades" (Brigadas de Apoyo). This service, with the personnel indicated above, operates as a team capable of covering all the technical needs of the farmers it can serve. 6.08 However, the needs for technical support continued to exceed the capacity of existing services. Recognizing this fact, a new effort is presently being implemented. In November 1977, an agreement was signed in Mexicali by which four institutions have joined efforts to put in operation a program aimed at covering the entire project area with an adequate technical assistance service. The four institutions are: SARH, through its General Directorate of Production and Agriculture Extension; the Government of the State of Baja Califorina; the National Bank for Rural Credit (Banco Rural del Noroeste) and the farmers organizations, which include five different groups. Through this agreement, all four institutions will contribute funds, so that in two or three years all the District farmers will receive intensive and integral technical assistance from 100 agents, each one responsible for not more than 2,000 ha (if, on average, the farms have about 20 ha, each agent will deal with 100 farmers). This program, once fully implemented, will fulfill the target set at appraisal for the supporting services scheme and it appears that, due to the experience gained over the years, the present set up better meets the actual needs in the District. Farm Credit 6.09 At the time of appraisal, it was thought that an essential element in the development of the project was going to be the availability of sufficient medium- and long-term credit. It was expected that the expansion of livestock production, the replacement of machinery and the anticipated need for on-farm drainage works was going to increase substantially the demand for this kind of credit. The Mexicans recognized-the importance of this element for the success of the project, but were reluctant to fix minimum targets on the loan documents. Finally, during negotiations, they agreed to establish such minimum targets in a side letter. By the side letter, the Guarantor agreed to - 84 - take all necessary action to ensure the availability, in the project area, of medium- and long-term credit resources, expected to amount to not less than -_ $ 5,000,000 annually beginning in 1969 ai,a incieasing up to approximately Mex$ 15,000,000 annually, by 1974, these figures to be reviewed from time to time and modified, if necessary. These limits were not revised. 6.10 There is no institution responsible for centralizing the information on credit operations in the project area. The information on credit volumes submitted to the Bank by SRH in the semi-annual reports was collected, usually by mail. Such data were global and incomplete, thus preventing good monitoring. Often, not all the banks or ginning companies cooperated in sending the infor- mation. Nevertheless, credit availability appears to have been adequate to the requirements of the project area. Although the mission could not obtain information on medium- and long-term credit from all financial institutions in the area, and for all the period, it seems that the minimum targets were, overall met. The Mexicali office of the Guarantee Fund for Agriculture, Livestock and Poultry Development (FIRA) reports the following rediscount figures, which amount to about 85% of the original credit to the final borrower. (Mex$ million) - 1971 1972 1973 1974 1975 1976 1977 Rediscount 6.4 3.0 12.0 12.1 9.4 16.2 52.8 Average Loan Volume 7.5 3.5 14.1 14.2 11.1 19.1 62.1 6.11 The big increase in lending in 1977 was due, not only to the produc- tion cost increases caused by inflation, but also to the devaluation of the Mexican peso which made the farmers prefer loans in pesos rather than in US dollars (which were obtained in US commercial banks); therefore, the FIRA rediscounts grew fast that year. 6.12 In addition to the medium- and long-term loans with FIRA redis- counts, the private banks lend on medium- and Long-terms with their own resources. Up until 1975, the public credit institutions in the agricultural sector were Banco Ejidal, Banco Agricola and Banco Nacional Agropecuario. In August of that year, they were united to create the Banco Nacional de Credito Rural S.A. The principal sources of credit to farmers when the project started were the cotton ginning companies; however their role has been sub- stantially reduced. In the early 1970's, as the cotton area was reduced and the cotton crop went through a crisis, the companies were severely affected, and several ejidos and farmers' associations built ginning plants with the financial and technical support of the official credit institutions. As a result of these events, the number of ginning companies since 1968 has been reduced from 14 to five and those with oil mills, fom five to two. The ejidos and farmers' associations have been able to compete successfully with the ginning companies, partl!y because they are subject to lower tax brackets and the fact that, the labor force is to a large extent integrated by farmers' family members and has non-unionized farm workers whose wages are not high. In the cotton crisis years, the ginning companies were faced with serious problems to recover their credit from farmers, but the Government finally stepped in to avert financial disaster, guaranteeing the farmers' debt and - 85 - re-scheduling it for payment in five years. The ginning companies then discounted this debt with the private banks. Most of the financial resources used by the ginning companies were raised in the US; however, since the devaluation of the Mexican peso, they have been using the internal market. Financial Performance and Returns 6.13 Perhaps the most serious problem that the District faced before the project was the nearly complete lack of conservation of the irrigation system and drains. The main factor behind this situation was the insufficient funds generated from the users to cover the costs that proper maintenance of the district would have required. This was an endemic problem to most of the irrigation districts throughout the country. Therefore, it was felt that, to preserve the works and keep them operating efficiently once the project was completed, the water tariff levels should be high enough to cover operation and maintenance cost instead of having to rely on fiscal transfers. In December of 1968, a Presidential Decree was issued for the collection of the investment costs and it reaffirmed the need to adjust water tariffs periodically. It went further to establish the suspension of water service to those users that did not meet payments. It failed, however, to fix the O&M charges at an appropriate level at that time. 6.14 In 1967/68, water charges were Mex$ 1.75/1/sec-day for gravity water and Mex$ 3.20/1/sec-day for pumped water from SRH wells. Municipal supply cost Mex$ 3.5/1/sec-day and farms having their own wells paid Mex$ 30/year/ha. These tariffs remained in effect for a long time because of political con- siderations. It was assumed, however, that water charges would increase during project implementation and that, because of the fully integrated system for water distribution, the areas served by gravity systems and by SRH wells could be charged Mex$ 2.06/1/sec-day and the areas served by private wells Mex$ 30 ha when the project was completed. It was projected at appraisal that maintenance and operating costs would increase gradually throughout project construction from Mex$ 38 million up to Mex$ 48 million. However, unforeseen inflationary trends that developed in the 1970s substantially increased the fund requirements in current terms. 6.15 The information on operation and maintenance costs reveals dis- crepancies as reported by different offices of SARH (para 6.05), but it is clear that the funds generated by water tariffs are dropping. In 1967, water tariffs generated about 80% of the funds required for operation and maintenance, while in 1977 they contributed only about 35% (Table 25). As work on interconnection of the gravity and wells distribution system (includ- ing the 76 wells of San Luis) advanced, the water from wells was delivered and charged to the farmers as if it were only gravity water. In August 3, 1977, the "Comite Directivo" of the District approved an increase in the price of gravity water from Mex$ "1.75/1/sec-day to Mex$ 3.10/1/sec-day, and the Govern- ment approved it on October 20, 1977, stating that it should be applied to the current agricultural season (winter, 1977/78). - 86 - Water Tariffsa/ Before After 1967/68 October 1977 October 1977 Gravity 1.75 1.75 3.10 Federal Wells 3.20 1.75 b/ 3.10 Private Wells 30.00 30.00/ha 53.15/ha Mexicali City 3.50 3.50 6.20 a/ Pesos/i/sec-day, except for Private Wells. b/ Mainly delivered through the irrigation canals network. 6.16 Despite this increase, it is projected that the tariff collection will cover only about 37% of the costs in 1978. The authorities had reached an agreement with water users to increase the tariff level annually until the District becomes self-sufficient, and they expect that this target could be met by 1981 or 1982. Using 1977 figures, the mission estimates that full rate tariff could be immediately applied (Annex 2), since the incremental O&M charge would represent only between 2% and 16% of the net incremental value of production. 6.17 The Government informed the mission that it has initiated a campaign throughout the country to make the larmers aware of the consequences for the projects and the Government treasury of maintaining unrealistically low water tariffs. In 1977, tariff increases similar to the one in Rio Colorado were made in other irrigation districts such as Culiacan (64% increase), Alto Rio Lerma (67% increase), Rio Mayo (103% increase), Rio Yaqui (100% increase), and Lagunera (20% increase). According to the Mexican authorities, this is the beginning of a new effort to make the irrigation districts self-sufficient, a policy that they intend to pursue firmly. 6.18 The December 1968 Presidential decree established the repayment conditions of the project investment expenditures by the users. The farmers were supposed to pay Mex$ 192/year/ha over 25 years for a total of Mex$ 4,800/ha, as proposed by the farmers and in line with their income payment capacity. These payments were going to cover 50% of the investment cost, plus 6% annual interest rate. The Bank felt that the proposed water rates were not fully satisfactory, but accepted them because they represented the maximum that could be obtained by the Government after difficult nego- tiations. The Bank also felt that some additional collection should be made from those farmers who were going to have their land leveled, but the appraisal mission opposed this on the grounds that the agreed charges were the maximum that the farmers will be able to afford. The farmers have not yet started to pay, but the project costs increased from Mex$ 1,187.5 million at appraisal to Mex$ 2,691.4 million as of December 1977. Since the Decreee fixes the payment quotas in absolute terms, it means that the users will end up paying a small fraction of the project costs. The authorities feel that the collection of the investment repayments will probably start after the water tariffs have reached the self-sufficiency level. - 87 - 6.19 The failure of the Government to act on the tariff and on the investment repayment was caused by fear of protests from the users and the political repercussions it may have brought. In the case of Rio Colorado, there were no indications of any serious opposition during project construc- tion nor are there any presently to tariff adjustments. It is advisable in future projects that action on tariff level be made a conditions of Board presentation, not necessarily to raise it to the level that would allow projects to be self-sufficient, but in those cases, where tariff rates have lagged too much, such action could result in a first adjustment in an agreed schedule to reach self-sufficiency. A full adjustment in many projects would probably mean increases by much more than 100%, which could likely encounter strong opposition from farmers, even if they have ample income capacity to absorb the increase. In the case of the investment repayment expenditure by the beneficiaries, if an agreement is negotiated between the Government and farmers or a Government decree is issued establishing payment conditions, these documents should specify conditions on rates and proportions of the costs to be repaid in relative terms, not absolute terms. The users could also be given a more direct and active supervisory role such as the right to annually audit the project accounts through independent auditors, if necessary financed by project resources. A work schedule by units could be developed as was proposed in Rio Colorado under which works would be advanced in a given area. If benefitting farmers start to repay the investment costs and water charges, works will go forward in the next unit. The Government and the Bank should be prepared to stop further development of the project if the users already benefiting refuse to pay, unless it is demonstrated that their income capacity at that moment suggests postponment. This action will help to bring pressure on the users already benefited from the potential beneficiaries. Staff and Training Issues 6.20 In connection with the expansion and improvement of the supporting services described in paragraphs 6.06 to 6.08, it is important to point out that the District authorities and SARH high officials in the State of Baja California have taken additional measures for staff training which appear wise and practical. Recognizing the importance of efficiency in the use of water and of adequate management of soils and drainage under irrigation conditions, the said authorities have made provisions for all the management personnel, the field engineers and the extension agents to attend courses at the "Benito Juarez Irrigation and Drainage Training Center" at El Carrizo in the State of Sinaloa. The District Director has stated that such training has proved valuable to all the staff responsible for handling the problems related to soils, irrigation, drainage and salinity control. Further, ditch-riders who cannot meet minimum requirements for their posts are being replaced with more qualified people and all are being trained to become agricultural extension agents so that they can help to provide technical assistance to farmers. Accounting and Reporting 6.21 The Appraisal Report did not recommend that a separate account be kept within SRH for project works and no specific recommendations were made on project auditing since, apparently, the internal auditing provided by the Government - 88 - was satisfactory. Thus, the completion mission had difficulty in obtaining firm figures on expenditures made for the project and the amounts of work done. It also had difficulty in separating expenditures that belonged to the project from those related to normal non-project activities of the former SRH delegation. On the basis of discussions with responsible sub-directors for accounting at SARH headquaters in Mexico City and a quick review of their procedures, it would appear that little effort would be required, in large projects such as Rio Colorado, to establish separate project accounts. It was not possible to obtain any internal auditing reports prepared by the former SRH on Rio Colorado or to learn the number of visits that SRH auditing personnel made to the project or the results of such visits. The SARH Director General for auditing appears to have the technical staff (accountants, economists, engineers) and the organizational capacity to potentially carry out the kind of auditing needed in projects such as Rio Colorado. In future projects, advantage should be taken of this office by specifically establishing its participation in the loan documents. Auditing requirements should be spelled out in the documents and the necessary budgetary allocations to cover the task to be performed should be secured. 6.22 The quarterly, semestral and annual reporting by SRH was regular, but content was unsatisfactory. Information on the construction, physical progress, cqst estimates, expenditures and financing was more complete than that on other aspects, but no information was ever received on operation and maintenance. Information on crop production, medium- and long-term credit, technical services, soil and farm drainate studies and on-farm development by farmers, was also deficient, and the guarantor never provided data on progress or difficulties with rehabilitation charges and water rates, or con- tractual arrangements with farmers related to the above charges. Apparently at appraisal and during negotiations, although the reporting needs and specific tables were discussed (the outline of Periodic Progress Reports were sent to SRH on April 29, 1968), insufficient attention was given to the capacity of SRH to fully supply the material needed. The Bank, during the contruction period, made some attempts to correct the problem and met with some success. It will be desirable that for future projects, an assessment be made of the capacity of the project unit or institution responsible for the project to supply the information and/or to work out an arrangement under which it can be secured from other agencies. The Bank should attempt, as early as possible, to have the reporting deficiencies corrected, or, if the agreed system proves impracticable, to have it modified. VII. BANK PERFORMANCE 7.01 Although the project has an adequate ex-post internal rate of return (IRR) (see Chapter VIII) it is the mission's judgment that if certain delays could have been avoided, due mainly to an inadequate feasibility study and insufficient budgetary allocations, the IRR would have been substantially higher. The inadequacies in the feasibility study were moreover not dealt with by the appraisal report as regards the following points: technical - 89 - aspects (para 1/), estimates of farmers' income missing (para 2.19), cost of technical assistance omitted (para 2.16), under-estimation of quantities (para 3.03) and implementation inadequate (paras 3.02 and 3.55). It would appear that despite these inadequacies it was correct to have proceeded with the appraisal of the project although in hindsight it appears that a better procedure would have been for the Bank to have been involved earlier--during the preparation stage--thereby minimizing the inadequacies in the feasibility report used by the appraisal mission. Such a procedure is in fact currently being followed in several instances for projects that the Bank has on its future lending program (Los Naranjos/Los Macines Drainage, Apatzingan Irriga- tion, Aguamilpa Irrigation Drainage and Bajo Bravo Rehabilitation II). 7.02 Another reason for the delays was institutional weakness within SRH (absence of adequate project authority or its equivalent). The appraisal mission might have insisted on better institutional arrangements or alter- natively a similar result could have been achieved through subsequent supervision missions. 7.03 Both of the foregoing deficiencies were no doubt due in part to the fact that SRH was (and as SARH continues to be) a strong organization with capabilites far better than in most countries. A situation of over- confidence on the part of the Bank therefore developed over the years. With the delays that occurred not only on the Rio Colorado project but on several other projects as well the Bank since 1975 (when coincidentally reorganization took place whereby one division dealt with all agricultural projects in Mexico) has come to understand better the deficiencies mentioned and to work out with SRH (later SARH) how to correct them--a process still underway. 7.04 During negotiations, later ratified by the loan documents and two side letters, the Government committed itself to carry out "programs and studies" dealing with technical assistance to farmers and improvement of the implementation schedule, and various technical aspects of the project. The loan documents and side letters did not however quantify these programs and studies (paras 2.14 and 2.15) with respect to cost and time. Although the loan documents stipulated that these programs would be reviewed periodically by the Bank and the Guarantor, such reviews did not take place during supervision missions because of the absence of supporting data. 7.05 Eleven supervision missions visited the project between 1968 and 1977, averaging one mission per year. The Bank took no action following the first three missions. Following the fourth to the ninth missions, letters were sent to SRH giving comments and requesting specific information about costs, design changes and such, but in no case did SRH answer the letters or submit the requested information. It appears that supervision missions spent too short a time in the field; moreover, SRH would have been unresponsive to pressure brought by the,Bank since SRH was not organized for corrective actions and suffered from lack of an implementation plan (para 6.04). 1/ Paras 2.06, 2.15, 2.20, 2.21, 3.01, 3.06, 3.25, 3.26, 3.27, 3.41, 3.46 and 3.48. - 90 - 7.06 The Bank reviewed 31 Quarterly Progress Reports prepared by the project staff. Although the reports were sent regularly (para 6.22) they were generally incomplete and Bank efforts to improve them were only partially successful. 7.07 The Bank also reviewed the tender documents of 31 major contracts let under ICB procedures and their contract extensions; also a number of contracts awarded under local procedures. The large number of contracts and the disorderly way in which they were awarded caused an excessive burden on project staff (para 3.39). 7.08 Because of the project's approaching completion date (para 1.05), the Bank, following the ninth supervision mission (June 1976), advised SRH about the preparation of the Project Completion Report and asked it to compile the basic information needed. The tenth mission (March 1977) and eleventh mission (December 1977) were devoted exclusively to analyse the collected information and to prepare guidelines for SRH to prepare its version of the Completion Report. The date of the Project Completion Report mission was postponed three times, from March 1977 to April 1978, when finally the mission took place. The delays were due to the extension of the closing date to allow full loan funds disbursement and to the delay in collecting the desired information for the PCR. 7.09 A factor contributing to success of the project--not foreseen at the time of appraisal--was improvement of the quality of the Colorado river water (1,100 ppm rather than 1,500 ppm assumed, at appraisal) resulting from negotiations between Mexico and the USA. In summary, it appears that in retrospect the final project outcome, although quite favorable, could have been still better if the Bank had been involved at an earlier stage of project preparation and if the Bank had conducted a more active dialogue with respect to correction of some of the institutional weaknesses within SRH; on both these points the Bank has in the past few years begun to correct its approach as regards other irrigation projects in Mexico with which it is involved. Still to be worked out is how best to reduce the large number of civil works contracts (para 8.06). VIII. CONCLUSIONS AND RECOMMENDATIONS Internal Rate of Return 8.01 The ex-post internal rate of return as estimated on a conservative basis (with respect to cropping pattern) by the mission is 14.6% which is considerably higher than the 11% estimated at appraisal. The IRR of 14.6% is believed to be highly probable even though there is a degree of uncertainty since works are not fully completed (presently 72% 1/), since the nature of the salt accumulation in the soil profile has not yet been determined, and since on-farm works have not been implemented. On the other hand, there are 1/ 86% if compared with appraisal estimate. - 91 - clear indications: (a) that SARH will continue with the on-going project programs 1/; (b) that SARH will carry out the remaining project works com- pleting them by 1985; (c) that whatever soil salinity problems arise, they will be controlled with the available techniques; and (d) that eventually the on-farm works 2/ will be accomplished. The sensitivity analysis (Annex 1) shows that, within a reasonable range of negative circumstances, the IRR could drop as low as 11%, which however is the economic cost of capital in Mexico. 3/ 8.02 The IRR as above determined does not take into account benefits from increased water supply to the cities of Tijuana and Mexicali amounting to 300 million m3 annually--water made available through improved water- management efficiency resulting from the project. As foreseen at appraisal, the project provides additional indirect, intangible and/or social benefits such as increased employment (para 4.19), crop diversification and a more socially stable farming population. 8.03 The IRR of the project would of course be substantilly higher if the delays encountered had been avoided. The loan was signed January 1968 but construction did not start until a year later and other delays resulted from an inadequate implementation schedule and insufficient funding. How such delays could be avoided in the future is discussed below. Recommendations for SARH 8.04 Extra funding: It is expected that SARH through the regular budgetary funding of the Colorado District will complete the project works, continue with studies and technical-assistance programs, and improve operation and maintenance. Nevertheless, if benefits of the project are to be accrued in the proposed time span, SARH should contemplate extra funding in order to intensify these works and programs, giving special attention to: (a) completing all the remaining project works by 1985 4/; (b) encouraging farmers to continue trials on more effective irrigation methods and eventually assisting them in installing the required irrigation and drainage works on their farms; 5/ and (c) intensifying the hydrogeologic observations and testing a hydrodynamic model of the aquifer to determine with more accuracy the underground water availability. 1/ Hydrological, drainage and salinity studies and the technical assistance program. 2/ Mechanical irrigation systems and surface and internal drainage. 3/ As calculated by the Bank of Mexico. 4/ Irrigation of 207,000 ha, full control of soil salinity and water table, reduction of pumping. 5/ Water management assistance at farm level is efficiently provided by PLAMEPA at present but it needs to be intensified. - 92 - 8.05 Water charges: Water tariffs were raised substantially recently (paras 3.01 and 6.15) and the farmers have agreed that water tariffs will progressively increase until they cover 0&M costs, which is expected to happen by 1981 or 1982. 1/ The farmers have also accepted the Presidential Decree of December 1968 (paras 3.01 and 6.18), which set an annual payment of Mex$ 192/ha for 25 years to pay for the irrigation project investments and now have the right to examine and review the Irrigation District's accounts. SARH should nevertheless carry out a socio-economic study as stipulated by the water law to determine the farmers' paying capacity. Through such a study it should be possible to persuade the farmers that they are able to pay, not only the full O&M costs, but also a major part of the investment. Ultimately, the District should be made administratively and financially autonomous; see dis- cussion of this problem in the Background Paper on Agriculture, o2.cit., para 4.29. 8.06 For future projects the annual payment for recuperating the investment should not be fixed in advance, since final project costs are always different from preliminary estimates, to a large extent because of inflation. Farmers should also be granted the right to review construction progress and project costs. The Rio Colorado farmers and project staff said that farmers have always been willing to pay for the works and that collection of appropriate rehabilitation charges would not have been a problem. 8.07 Project Authority: Experience with the Rio Colorado Project as well as two other current projects partially financed by the Bank (Panuco, Loan 969-ME and Bajo Rio Bravo, Loan 1111-ME) have pointed to a prevailing institutional weakness needing correction. This weakness concerns divided responsibility for the project as between SARH central offices and the local authorities, namely the Irrigation District. As another example, the chief of rehabilitation works of the Rio Colorado District gave only partial information to the SRH's central Evaluation Department. Feasibility studies and construction designs are normally prepared at the central offices without adequate participation and feedback from the local authorities. In the case of the Panuco Project, the officials in charge were vigorous enough to over- come this basic weakness, whereas in the case of the Rio Bravo Project, after numerous discussions (in which staff of the Bank was involved) a local official was designated as principal coordinator, in effect setting up a "Project Authority" although this was not formalized. It would appear that for future projects serious consideration should be given to formalizing the Project Authority concept. 8.08 Accounting and control: For no projects in Mexico have individual accounts been kept (para 6.21). In SARH's central office, a record of the money specifically assigned to the project is kept but the changes throughout the year, e.g., decreasing or enlarging the amounts for specific purposes, 1/ In no project in Mexico are water rates revised periodically. Given the high rate of inflation in recent years it is hardly surprising that the overall district's operating deficits are growing rapidly. However, SARH has announced that the districts as a whole will be financially self-sufficient on a current account basis by the year 1982. - 93 - are not recorded. Also, data from the project administration when available do not match with those of the central offices. Recommendations for the Bank 8.09 The Bank should be involved with preparation of irrigation projects in Mexico at an earlier stage than was the practice several years ago but which is the practice now; the quality of the feasibiity study available at the time of appraisal will thereby be improved. The feasibility study available at the time of appraisal should in any case be adequate to justify key elements of the project including costs and benefits and to present a well prepared implementation schedule; the appraisal could then proceed with minimum risk and implementation could proceed with minimum delay. To further speed up implementation, measures recently proposed in the CPS memo of July 14, 1978 "State of Project Preparation Necessary for Loan Approval" could be adopted, including consideration of retroactive financing of final design of key elements of the project especially those to be undertaken during the first two years of the implementation of the project. 8.10 Key institutional matters such as the designation of a special project authority might in future cases be made a condition of Board presen- tation. 8.11 With respect to cost accounts, their establishment on an adequate basis should be made a condition of effectiveness. 8.12 With respect to progress reporting, the Bank should at appraisal specify who will be responsible for report preparation. The first super- vision mission should devote part of its time to reviewing and assuring that appropriate measures are taken for preparing the progress reports. 8.13 Since SAR civil works contracts are excessively fragmented, the Bank should make an effort to solve this problem on a uniform basis by means of a joint SARH-Bank task force dealing with the entire portfolio of SARH/Bank projects. - 94 - ANNEX 1 Page 1 COMPLEMENTARY INFORMATION FOR THE PROJECT EVALUATION A. Costs and Benefits Streams at Appraisal 1. As shown in Table 1, benefits at appraisal included the incremental net value of production, the replenishment of groundwater and the prevention of salinity. These two last benefits accounted for more than 50% of all benefits. B. Determination of the Base Year for the Ex-Post IRR Evaluation 2. To evaluate the benefits attributable to the project a starting point had to be set since on the one hand, the appraisal's presumption that benefits would decrease without the project can now be seen in perspective and, on the other hand, unforeseen events that helped the area have ocurred. Effectively, the cultivated areas were decreasing as soil salinity progressed dropping from 176,000 ha in 1967 to 168,000 ha in 1973, and over-pumping grew also to compensate for areas that could not be irrigated by gravity because of the large salt content in the Colorado River water. Between 1961 and 1962, gravity irrigated areas shrunk from 113,000 ha to 98,000 ha while pumped irrigated areas grew from 63,000 ha to 75,000 ha in 1971, giving a total drop of 4,000 ha in the period. However, in 1972 when the second lowest area of 172,000 ha 1/ were cultivated and only 42% of project works had been completed, the tonnage and gross and net value of agricultural production rose substan- tially in respect to the previous years and kept increasing over the following years as can be seen in Table 22. As the area shrank rather than grew and the half completed works could not have an impact on the project, the production increase can only be attributable to increase of yields caused by the improve- ment of the Colorado River water quality whose salinity dropped from 1,200 ppm average to 1,050 ppm, 2/ according to the "Minute 241" 2/ signed in 1972 and whose covenants were immediately applied. 3. Over and above those benefits, between 1972 and 1973 construction speeded up and reached 57% of completion, and fresh water from San Luis was made available to the project. The gravity irrigated area rose by 2,000 ha to 100,000 ha and the pump irrigated area fell by 4,000 ha to 70,000 ha. Although the cultivated area reached its lowest figure of 168,000 ha in 1973 (2,000 ha less than the area served by irrigation), tonnage and gross and net values of production increased again in that year. It is obvious to assume that project works started to give the desired results. Yield increases in 1/ Average salinity in 1971 has been 1,200 ppm but increasing through the years to 1,500 ppm with a maximum of 3,000 ppm. 2/ "Minute 241" of July 1972 between Mexico and US established that water delivered to Mexico would have the same quality as the water at the imperial dam, i.e., 1,050 ppm as maximum. - 95 - ANNEX 1 Page 2 certain crops were noticeable. Wheat yields, which, up to 1971, had been under 3.5 tons/ha gross rose to over 4.5 tons/ha from 1972 onwards; cotton yields were below 2 tons/ha up to 1970 and over 3 tons/ha from 1971 to date, and barley yields, which were 2.5 tons/ha on average until 1971, have been 3.5 tons/ha on average since 1972. Therefore, for the purpose of this analysis the year of 1972 is considered the base year and it is assumed that without the project, production would have stabilized at the 1972 level. C. Circumstances that May Affect the IRR Sensitivity Analysis 4. The limited water supply for the area puts the project in a critical position for obtaining the expected results in agricultural production. All factors of the project such as conveyance and application efficiencies and pumping volume must reach the assumed targets. Should efficiencies continue to be low or over-pumping in the project area be halted without installing new wells in San Luis, the project goals may not be reached. Under the assumption that all project works will eventually be completed but that opera- tional and farm efficiencies would remain at the present levels, agricultural production would drop about 10% in the long run and by 1985 would represent about 15% of the net incremental value of the production. 5. Any drop in physical production due to other factors is not fore- seen since there are no climatic hazards in the area and pests and plant and animal diseases are being taken care of. However, a drop in prices could happen in relation to exports or substitution of imported commodities such as cotton, wheat and meat. If prices dropped 5% for all commodities, the incre- mental net value of production would drop 15% by 1985. Therefore, for the purpose of the sensitivity analysis, it is assumed that the combination of these two factors, low water use efficiencies and drop in prices, would repre- sent 30% of the incremental net benefits. The economic rate of return so calculated is 11%. MEX I CO Loan 527-ME Rio Colorado Irrigation Project Estimated Costs and benefits at Appraisal (Basic Data for Calculating the Internal Rate of Return) Costs Benefits Difference Increase in Increase in Replenishment Prevention between Project On-farma/ Operation & Net Value of of Lo of c/ Benefits Year investment Investment- Maintenance Total Farm Output Water Salinity- Total and Costs -------- -----------------------------in thousands of Pesos------------- 1968 20/,500 207,500 25,114 25,114 -182,386 L969 172,500 1,720 174,2?0 25,114 25,114 -149,106 1970 201,250 3,190 204,440 25,114. 25,114 -179,326 1971 152,500 4,880 157,380 21,653 25,114 2,080 48,847 -108,533 1972 152,500 6,160 158,660 38,418 25,114 4,161 67,693 -90,967 1973 145,000 7,430 152,430 56,923 25,114 6,241 88,278 -64,152 1974 156,250 10,000 8,620 174,870 72,514 25,114 8,322 105,950 -68,920 1975 10,000 9,770 19,770 87,511 25,114 10,402 123,027 103,257 1976 10,000 9,770 19,770 88,200 25,114 12,483 126,497 106,727 1977 10,000 9,770 19,770 90,290 25,114 14,563 129,967 10,197 1978 10,000 9,770 19,770 91,679 25,114 16,644 133,437 113,667 1979 10,000 9,770 19,770 93,068 25,114 19,903 138,085 118,315 1980 10,000 9,770 19,770 94,458 25,114 23,162 142,734 122,964 1981 10,000 9,770 19,770 95,847 25,114 26,422 147,383 127,613 1982 10,000 9,770 19,770 97,236 25,114 29,681 152,031 132,261 1983 10,000 9,770 19,770 98,625 25,114 32,940 156,679 136,909 0' 1984 10,000 9,770 19,770 100,015 25,114 36,200 161,329 141,559 1985 10,000 9,770 19,770 101,404 25,114 39,459 165,977 146,207 1988 10,000 9,770 19,770 101,404 25,114 49,237 175,755 155,985 1998 9,770 9,770 101,404 25,114 54,291 180,809 171,039 2008 9,770 9,770 101,404 25,114 78,376 204,894 15,124 2017 9,770 9,770 101,404 25,114 78,376 204,894 195,124 a/ This would cover the installation of tile drains in an area comprising about 18,000 ha, and the application of neutralizing chemicals to reduce soil salinity. b/ Due to previous overpumping of groundwater, about 11,625 ha of presently cropped cotton land would have to be taken out of cultivation without the Project. The annual figure in this column represents the estimated net value of cotton grown on this land. Continued cultivation of the area will be made possible by the Project by reducing water losses. L/ The very arid climate and the heavy clay soil in the western section of the project area causes a substantial portion of irrigation waters to evaporate from the fields, leaving behind a residue of salts. The installation of drains and increased availability of water will provide greater opportunities for leaching and, hence, prevent the accumulation of salts in the soil. The figures in this column are the estiated annual values of the net output on lands that would otherwise he retired from cultivation if salinity is not checked. d/ '1This columni will add up to approximately zero when discounted at 11 percent. MEXICO Loan 527-ME Rio Colorado Irrigation Project Relevant Figures of Project Impact Works Water Available Physical Area erVed Agricultural Output c/ Year Comil et ion QuqnLtLy Salt Gravity Wells- Total Area Volume GVP NVP NVP- Observations n. Mn /year ppm -----------ha------------ --- ha Ton'000 -----Mex$'000------ 1967/68 0 2,342 1,210 1.1.3 63 176 178d/ 449 3,214 1,276 - Time of Appraisal. - - - - - - - - - - Period of Deterioration. 1971 33 2,283 1 99 75 174 174 658 2,35 577 - 7 42 2,550 1,120 98 74 172 172 746 2,843 1,116 - "Minute 241" International Treaty. 10/3 57 2,419 1,100 Itu/ 70 170 168 823 2,924 1,306 191 San Luis Wells and Project Works. 014 67 1,070 117 72 189 189 781 4,062 1,851 736 S- - - - - - - - - - Period of Recuperation. 1"71 72 2,746 933 132 68 200 199 953 3,740 1,706 591 I/l8 - - 1,050 136 67 203 203 963 3,897 1,797 692 1)85 100 2,500- 1,050 157 50 207 207 1,281 5,184 3,080 1,869 Note: Underlined figures are the minimum for all items but for salinity, water available and pump area which are the maximum. .i/ n Colorado River water only. Yearly averages. T/ Only in the project area. Does not include San Luis wells. c/ Incremental. df Incldes,, double cropping. / Fr-i 1'73 onwards the water from San Luis wells has been mixed with the Colorado River water and di,tributed thILrough lhe irrigation canal network. i/ ,roni tory Situation. In the long run only 2,370 Mm /year will be available. M-l - 98 - ANNEX 2 Page 1 COMPLEMENTARY INFORMATION FOR THE PROJECT EVALUATION Operation and Maintenance, Costs and Charges Using 1977 figures, th mission estimated the unit cost of irriga- tion water (per ha and per 100 m of water) delivered at farmgate and the revenue generated by the current tariff for 1977, 1978, 1985 and year 2000. It was assumed that gravity and well water would be delivered through the canal network and that only one tariff would be applied. Table 1 shows that under such assumptions with the present tariff from 1978 to 1985 only 58% of the O&M costs would be recovered and that, by the year 2000, about 52% would be recovered. It is worth noting that the difference of 58%, as cal- culated by the mission, with 37%, as estimated by the District for 1978, is due to undercharging for water. Should the tariff be raised to the sug- gested level of Mex$5.41/l/sec-day, the O&M cost would be fully covered. The suggested increment of water tariffs would represent (1977/78) only 16% of the net incremental value of production for the low value crops as shown in Table 2. It appears, therefore, that SARH could speed up the water tariff adjustments without any problem. MEXICO Loan 527-ME Rio Colorado Irrigation Project Theoretical Unit Water Cost and Cost Recovery for 0AM (tentative) Available Water- Water Cost- Revenue Suggested Gross Farm Gate- Cros Farm Gate Tari Tariff Equivalent Total Deficit Tarif s Year Mm31y Mm3/7 3/haZ' Mex$/1000 m3 Mex /1000 m, Mex$/ha ex$- Mex$/1000 ms- Mex$/ha Mex}M MexsM Mox$e zent 1977 2,660 2,142 10,710 49 61 591 1.75 20 195 43 87 (5.4) 1978 2,660 2,142 10,71.0 49 61 591 3.10 36 350 77- 53 (5.4) isi ory 1985 2,500 2,104 10,164 52 62 573 3.10 36 330 75 55 5.4 ire 2000 2,370 1,896 9,159 55 69 573 3.10 36 300 68 68 6.0 Net available for irrigation excluding third party's rights. Assuming the average efficiencies of 0.81, 0.84 and 0.8 for 1977, 1985 and 2000. Physical area 1c00,000ha 1977-78, 207,000 Ia 1985-2000) considering double cropping and permanent crops. Assuming an annual 0&K budget of Mej$130 million per year. Mex$/1/sec-day equivalent to 86.4 m is the same amount as estimated by SARII. However SARH used several tariff categories and included third party's rights. NEXICO Loan 527-Mr Rio Colorado Irrigation Project Inckdence of the Suggested Tariff Increase lncrement l Water Charges Crop Area NPV/ha With Current Tariff- With Suggested Tar f Mex$/ha Mex$/ha 7- Mex$/ha .- Low Value Crops Barley 27,151 1,365 11 29 Sorghum 2,441 2,560 6 15 Wheat 63,460 2,962 5 13 Total and Weighted Average 93,052 2,485 155 6 395 16 Medium Value Crops Alfalfa 18,182 3,525 310* 9 790 22 Safflower 5,536 4,659 155 3 395 8 I Rye Grass 10,762 4,798 155 3 395 8 Sesame 390 6,285 155 2 395 6 C) O Total and Weighted Average 34,870 4,128 236 6 601 15 High Value Crops Asparagus 1,422 14,400 310* 2 790 5 Cotton 62,091 18,151 155 1 395 2 Grapes 150 29,000 310* 1 790 3 Other (Vegetables) 1,183 32,500 310* 1 790 2 Total and Weighted Average 64,846 18,070 161 1 412 2 * Perennial and vegetables hiavv been charged twice since water demand is nearly double that of seasonal crops. a/ Current tariff in 1977 Mex$195/ha Current tariff in 1978 Mex$350/ha Increment Mex$155/ha b/ Incremental water charges Mex$/ha 100 NPV Mex$/ha c/ Suggested tariff Mex$590/ha Current tariff in 1977 Mcx$195/ha increment Mex$395/ha  - 101 - Table 1 MEXICO Loan 527-ME Rio Colorado Irrigation Project IRRIGATED AREA IN THE MEXICALI VALLEY FROM 1903 THRO;uGOH 1965 IRRIGATION WITH IRRIGATION WITH TOTAL PERIOD COLORADO RIVER WATER PUMPED FROM IRRIGATED WATER DEEP WELLS AR EA Hectares Hectares Heclares 1908-1910 4,169.0 4,169.0 1911-1915 12,214.4 12,214.4 1916-1920 48,762.0 48,762.0 1921-1925 69,554.0 69,554.0 1926-1930 67,367.0 138 67,505.0 1931-1935 54,052.0 138 54,190.0 1936-1940 90,731.0 138 90,S69.0 1941-1945 113,052.0 138 113,190.0 1946-1950 144,329.0 1,053 145,382.0 1951-1955 192,612.0 2,556 195,168.0 1956-1960 177,142.0 21,391 198,533.0 1960-1961 129,220.0 56,870 186,090.0 1961-1962 115,865.0 63,236 179,101.0 1962-1963 113,992.0 62,689 176,681.0 1963-1964 118,746.0 66,465 185,211.0 1964-1965 112,282.0 66,755 179,037.0 MEXICO Loan 527-ME Rio Colorado Irrigation Project Project Area (thousand ha) -IrriRated Physical Area Gross Double Water (Mexicali Valley) Commanded Gravity SRH Wells Private Wells Crop Total Rights (not added) Before the Project 328.0 275.0 1963/64 (5.1) 156.4 1964/65 156.3 1965/66 159.1 1966/67 1967/68!2 113.0 63.0 (incl.) 176.0 Before the Project 0 Proposedb/ 137.0 66.0 (incl.) (12.7) 203.0 1968/69 1969/70 168.2 1970/71 98.5 49.5 25.7 173.6 1971/72 98.0 49.2 24.4 171.6 1972/73 100.0 46.1 24.1 170.2 1973/74 117.4 47.1 24.9 189.3 1974/75 119.8 46.1 24.4 190.2 1975/76 129.8 46.0 23.2 200.0 After the Project Proposed (medium term) 147.0* 40.0 20.0 (20.0) 207.0 207.0 Revised Estimate 157.0** 50.3*** - (15.0) 207.0 207.0 * Includes areas of combined gravity-well irrigation since about 90,000 ha could be irrigated with the amount of water pumped (900 Mm3/year). ** With transferred (400 Mm3/year)pump water from San Luis to gravity system. *** Pumping only 500 Mm3/year. a/ Area mentioned in the Appraisal Report. b/ Assumed at appraisal. c/ Possible only with 80% conveyance and 88% farm efficiencies. __ ~r3 MEXICO Loan 527-ME Rio Colorado Irrigation project Total Water Available and Distribution (millions of cm/year) Appraisal Current Actual Average soures and li,es Assumptions Assumptions 1967/68 1968/69 1969/70 1970/71 1971/72 1972/73 1973/74 1974/75 1975/76 1976/77 5 years 3 years C,dl., River 1,850 1,850 1,01 1,782 1,762 1,788 1,780 1,679 1,839 1,754 1,820 1,919 1,802 1,831 1'nqping lt" 700 900, 669 711 632 643 958 992 1,204 1,038 1,002 1,010 1,049 1,017 S.a, Luis (incl.) 400g - - - - - 93 151 132 121 129 125 127 Iroject Are, (incl.) 500- 669 711 632 643 958 899 1l053 906 881 881 924 889 1,l1r1 2,550 2,750 2,470 2,493 2,394 2,431 2,738 2,671 3,043 2,792 2,822 2,929 2,851 2,848 W.-e and .ther -ses 39 80 7 4 27 15 47 129 25 33 129 86 80 83 1inicipal supply - 170- 121 115 130 133 134 123 105 96 91 85 100 9L C lr.,nsferen,ce ,, ri_n_- 230 - - - - - - - - 12 12 7 12 I rIg a 1i 2,511 2,370 2,342 2,374 2,237 '2,283 2,557 2,419 2,913 2,652 2,590 2,746 2,664 2,663 .. A-suming that the fel potential of the aquifer is used (hydrogeologic study SARIl) the actual pumping Installed capacity is 300 x 106m3/year but -Ily about 125 . 10 m /geir is being pumped. 6 3 A/ Ac,,,ly a.,ut 900 10 m /year are being pumped. Ptunping of 200 x 106m3/year should be tF1sAerred to San luis (hydrogeologic stgd SARII). Deep percolation in the area (from canals and farm irrigation) previously estimated .is 500 x 10 3m /year is now estimated to be 300 x 10 m /year. W.I Was hta been minital since US is delivering the full quantity of Mexican Water (1,850 Hm ) with salinity of only 1,050 p.p.m. Water with high salinity is being by-passed by the Weliton-Mohawk canal. d/ E*Llimated for the 3year 2000, l Equivksalen t,, 4 m /sec to be delivered by year 2000. One n, /sec of water will be delivered during 1979 when it expected that the aqueduct will be complted. MEXICO Loan 527-ME Rio Colorado Irrigation Project Total Irrigation Witer Allocation (Mm-) Current Actual Appraisal Assumptions 1972/73 1973/74 1974/75 1975/76 1976/77 Average Available for IrrigationAl 2,511 2,370 2,419 2,913 2,652 2,590 2,746 2,664 Source: Wells 700 900 Colorado river 1,811 1,470 Gross Allocation 1. Pump Irrigation Volume 700 500 830 1,023 890 866 863 894b/ Asea (ha/000) 60 50 70 70 70 70 70 70 1 m /ha (gross) 11,670 10,000 11,857 14,614 12,714 12,371 12,329 12,777 2. Transference from Wells to Gravity 400 162 181 148 136 147 1559/ 3. Gravity Irrigation 1,811 1,470 1,589 1,890 1,762 1,724 1,883 1,770d/ A5ea (ha/000) 155e/ 177f/ 98 119 120 130 129 119 m /ha (gross) 11,680 10,747 16,214 15,882 14,683 13,262 14,597 14,928 a/ See table 3. b/ All pumped water volume less transferences to the canal systems. c/ Includes all pumped water from San Luis plus water pumped in the project area. d/ Includes the Colorado river water plus water transferred from wells (less third party users. See table3 ). e/ Includes double cropping on 12,000 ha. f/ Includes double cropping on 15,000 ha. M MEXICO Loan 527-ME Rio Colorado irrigation Project Actual and Projected Use of Water for Irrigation Number of Area Available Farm-Cate water volume Crops per l a '000 6Wgter Conduction al Farm Water Applications Leaciin§, To Slants Year sr Efficiecy 10 m /year ni/ha Efficiency m /ha m3/ha- m /ha 1resent Situation (1977-1978) iloll. l 70 860 0.9 774 11,057 0.7 7,740 - 7,740 Gravity 1 110 1,800 0.76 1,368 9,120 0.7 6,384 833 5,551 2 20 (nc.) 0.76 18,240 0.7 12,768 1,666 11,102 Wh,le Project Average - 200 2,660 0.81 29142 10,710 0.7 7,497 625 6,872 iransuiry Sltuation (1985) ls I 60 700 1.0 700 11,667 0.7 8,167 - 8,167 Cravity 127 1,800 0.78 944 7,433 0.7 5,203 679 4,524 2 20 - 0.78 460 23,000 0.8 18,400 2,40u 16,000 Wla.e Pr-j,ct Ave-age 207 2,500 0.84 2,104 10,164 0.72 7,337 548 6,789 Ful ure i ou i in (2000> A. WlI, 50 500 1.0 - 500 10,000 0.8 8,000 - 8,000 (,.viiy I 147 1,870 0.8 1,266 8,612 0.8 6,80 889 5,991 2 10 - 0.8 230 23,000 0.8 18,400 2,400 16,000 WhlIe P rjeL Aveiage 207 2,370 0.84 1,996 9,643 0.8 7.714 725 6,989 is. Wll and Giavity i 187 2,370 0.8 1,436 7,679 0.8 6,143 587 5,556 2 20 - - 460 23,000 0.8 18,400 2,400 12,000 Whol- Pr,. c Av1rage 207 2,370 0.8 1,896 9,159 0.8 7,328 762 6,566 C. Wlil.,an Gravity 1 192 2,370 0.8 1,551 8,078 0.8 6,463 60) 5,863 2 15 - - 345 23,000 0.8 18,400 2,40D 16,000 Whe - lrect Average 207 2,370 0.8 1,896 9,159 0.8 1,328 730 6,598 zil r cii,tel as 15% of the plants conlsump>tive ,se t o eddi tonaly aeppliled, ei Ltier at planting time or in each, pication, ppr,ii:,a) Reprlonsiderel 177. , th,e g-o, te,r allotment for the project. MEXICO Loan 527-ME Rio Colorado Irrigation Project Irrigation Efficiencies Present 1985 2000 Irrigated Area Conduction Application Overall Conduction Application Overall Conduction ApplicationlDyerall Pumps 0.9a/ 0.7 0.63 1.0h/ 0.7 0.7 1.0 0.8 0.8 Gravity 0.76 0.7 0.53 0.78 0.7 0.55 0.8Q/ 0.8 0.64 Pumps and Gravity 0.81 0.7 0.57 0.84 0.7 0.59 0.84P/ 0.8 0.67 a/ Lined canals about 50%. o b/ All canals lined. c/ Irrigation on schedule by zones. d/ Sprinkler in one third of the area. e/ Full interconnection of wells and gravity system. M - 107 - MEXICO Table 7 Loan 527-ME Rio Colorado Irrigation Project Water Quality (Water: Mm3/year) (Average Salinity: ppm) Colorado San Luis Project Year Average River Pumping Wells Three Sources 1967/68 1,215 - - - 1968/69 1,210 - - - 1969/70 1,210 - - - 1970/71 1,200 - - - 1971/72 1,120 - n.a. - 1972/73 1,100 - - - 1973/74 1,070 n.a. - - 1974/75 1,040 n.a. - - 1975/76 1,078 900 - - 1976/77 993 890 - - Last Five Years Average - Salinity 1,056 900 1,400 1,160 - Water Volume Mm3/year 1,802 125 924 2,851 Future (estimate) - Salinity 1,020-1,051a/ 900 1,300 1,064 - Water volume Mm3/year 1,850 400b/ 500 2,750 a/ Range of salinity estimated on the basis of "Minute 242" signed between Mexico and the USA: 900 + 121 - 30. b/ Includes 200 Mm3/year pumped within five miles from the international border and 200 Mm3/year that could be pumped upto 15 km from the border. MIX I Co Loan 527-i,; Rio Colorado Irrj_&jtion Project ProjecL Works Completed Percentage of Loan Current as of CompleLion compared Io lie cormpleted Unit Appraisal Agreement Estimate 12/31/77 to Appraisal from 1978 onward 1. Cana; Lining km 2,334 2,330 2, 504b/ 2,07721 89 42 c/ 2. Sit i , uri s # 4,685 - 5,685- 5,221 il 1 464- 3. D rai ns Rehabi Ii tatIon kin 1,181 1,180 1,191 1,191 l1 - 11. St ruictunre I (incl . in 6) - (incl. in 6) (incl. in 6) - '. Nk Drain, km 336 335 500dl 336' 100 L64 6. St Iurs # 1,702 - 1,347 - 1,226P 72 121 I Well Reliabi I itation # 482 185 604 297 62 30 7- H. Nw Well : Pr-oct ion # 107 105 192 122 114 709/ t1Observat i on 4 221 220 51 51 23 10 Ro-ts km 555 555 1,505 1,385 250 ti. lelephoie km 467 - 120- 120 - (2. Galo,uing Stations # - - 113 92 - 21 13. Ditch Rider louses # 81 - 55n/ 55 68 14. [ui ldi d s og # 2 - 5- - 0 5 H IS. L;uol L-el ing ha 178,000 178,000 185,000 175,286 98 9, 714/ :boit 100 km of canals for cuniecCting wells have not yet been built. They are pending decision on final well set-up. It/ Proposed bv Sil (letter #5850 of 11/7/72) to improve operation of the canal system. c/ :lainlY farm turn-nuts in order to facilitate water delivery and water measurement. d/ About 160 km of new drains in the western part of the area (Progreso) have not been built. Construction of tlie out-fall drain for this area started in 1977. el mainlv in the southern part of the project, including 40 km of the main collector to Laguna Salada. f/ About 600 individual farm drainage outlets have been eliminated, or replaced by outlets serving two or moie farms, and about 250 bridges have been added. Stritires; in: New Drains 429 Existing Drains 797 1 ,226 h! e I r0ihbi lit at ion is a continuous process. Wells have to be reblii It every 15 years on average and pumps cIangel I vIrV 10 years. t/ Ini-] rdes boring f 3M7 wells and replacing of 205 pumps and 235 electrical motors. j/ tirclude 67 wells of lesa Arenosa de San tuis,already installed and 70 new wells to be constructed om tje adjacent area. k/ In nablition 60 prodction wells have been selected to monitor underground water levels and salinity. T/ Ntldiiional 830 km of canal service roads have been widened ani improved to be used for feneral traffic, and 120 kim of roads for connecting tlie wells in c n/ Radio sstim!i have been adopted in liei of telephone lines. San Luis will be consiructed. n / In, luies hou,es for gate operators which are equivalent to two ditch rider houses. l/ ih repair shop will be built in 1978. Other four offices in units III , V and VI of the district will be built later. p/ ine lini' ru btout 6,00 ha in the western part of thie project area where drainage has not yet been installed and riout 4,000 h- srattered all ovrer the project. / WlIs to be dog in Mesa de San Luis to extract 200 million cubic meters per year, replacing wells that have to he shi down in the so1tern part of the project area- MEXICO Loan 527-ME Rio Colorado Irrigation Project Project Works Financing Loan dis- Exchange Financing Total Cost of bursements Rate Loan proceeds Disbursement Cost of Wprks Civil Works Period US$M Mex$/US$ Mex$M Percentage Mex$M Mex$M c/ Jan. 69 - Dec. 72 11.32 12.5 141.50 35 404.29 757.20 Jan. 73 - June 76 12.90 12.5 161.25 16 1,007.81 963.50 July 76 - Dec. 77al 0.78 22.0 15.84 16 99.00 389.60 I0 25.00 318.59 21b/ 1,511.10 2,110.30 Percentages 15.00 72.00 100.00 a/ Loan funds were totally disbursed by January 1977. b/ Average percentage of disbursement. c/ As of December 31, 1977. m. MEXICO Loan 527-ME Rio Colorado Irrigation Project Revised Cost Estimates (US$ Millions) Revis[sLns Expended as Civil Works Appraisal 1971 1972 1974 1975 1976 1977- of 12/31/77 1. Canal Lining 29.5 33.3 49.9 52.2 58.7 89.1 96.0 72.4 2. Structures 3.2 3.3 5.6 9.5 10.7 19.2 23.0 17.6 3. Drain Rehailitation 2.3 2.1 7.4 8.1 10.1 (5.0) 8.5 8.5 4. Structures 1.8 2.0 6.2 5.4 6.2 (6.0) - 5.1 5. New Drains 1.2 1.1 2.0 2.4 3.6 14.3 8.0 4.4 6. Structures 0.9 1.0 3.3 2.1 2.5 10.3 4.0 3.0 7. Well Rehabilitation 2.5 2.2 2o3 1,6 2.4 3.4 5.0 2.1 8. New Wells 2.8 2.5 2.8 2.5 3.3 6.4 10.4 7.4 9. Roads 3.2 2.8 3.3 2.4 2.7 6.5 5.1 5.1 10. Telephone System 0.2 0.2 0.8 0.3 0.3 0.2 0.4 0,4 I-. 11. Gauging Stations - - - 0.1 0.1 0.6 - - 12. Buildings 1.,7 1.5 0.8 0.6 0,,6 0.6, 0.4 04 13. Land Leveling 11.5 10.3 9.6 9.6 19.2 21.6 37.0 29,0 14. Contingencies 12.5 9.1 4.0 - - 20.6 7,2 0.7 Suibtotal 73,3 71.5 98.2 97.0 120.4 190.8 205.0 156.2 tLther Works 1. Reallocation of Farmers 0.8 1.4 1.8 h/ b/ 5,7 - - 16. Materials 10.5 16.3 11.1 12.0 16.8 17.9 200 14.4 17. Engineering and Supervision 90 6.6 13.2 18.0 21.2 33.5 34.0 26.0 18.. Equipment (local) 1.4 1.3 1.4 1.3 1.3 1.6 3.0 2.7 19. offices in Mexico - 0.7 0.8 0. 8 5 0.5 - 20. On-Farm Assistance - - 2.7 b/ b/ . - - Total 95.0 97.8 129.2 129.2 160.5 250.0 262.0 199.2 Percentage 100.0 103.0 136.0 1360 169.0 263.0 276.0 a/ Estimated by the mission h/ The items concerning relocation of farmers and on-.farm assistance are related to social costs and are indefinite. Item dropped from estimate. (Supervision Report, Rio Colorado, March 20, 1973). MEXICO Loan 527-ML Rio Colorado Irrigation Project Project Cost (millions of Pesos) To be At Appraisal Actual as of D , 31. 1977 Completion Expended Local Foreign TLocal ocal Loan Total Percentage in 1978 Canal Lining 228.8 140..0 368.7 979 89 Related Structures 37.5 2.5 40.0 238 Ill Subtotal 266.3 142.5 408.7 997 220 1,2j Rehabilitation of Drains 12.5 16.2 28.7 115 101 Related Structures 21.2 1.3 22.5 69 72 Construction of Drains 6.3 8.7 -15.0 60 100 Related Structures 100 1.3 11.3 41 72 Subtotal 50.0 27.5 77.5 245 40 285 Rehabilitation of Wells 27.5 3.8 31.3 28 62 Well Construction 30.0 5.0 35.0 100 114 Subtotal 57.5 8.8 66.3 118 10 128 Roads 27.5 12.5 40.0 69 250 Telephone and Gauging Stations 2.5 - 2.5 6 Bui ldings 12.5 8.7 21.2 5 68 Land Leveling 71.2 72.5 143.8 391 3 95 Other Works - - - 9 Subtotal 113.7 93.7 207.5 430.2 50.1 480,3 Relocation of Farmers 10.0 - 10.'0 (Incl.) - (Inc.) SubLoLal of civil works excluding 497.5 272.5 770.0 1,790.2 320.1 2,110A 86 148' c,ment and steel Cement and Steel, Local Purchase 131.3 - 131.3 193.9 - 193.9 *8 Total Civil Works 628.8 272.5 901.3 1,984.1 320.1b/ 2,304.2 Engineering and Supervision at 12 57. 112.5 - 112.5 351.2 - 351.2 *21 Equipment., Local Purchase 17.5 - 17.5 36.0 - 36.0 20.0 lotal, Construction, Engineering G Equipment 758.8 272.5 1 2,371.3 320.1 2,691.4 Con,tingency 157. 116.2 40.0 - 156.2 - - - Overall Total, including Contingency 875.0 312.5 3201 197.1 A, Percent Ige 74 26 100 200 27 227 86 1/ For estimIting the distribution of loan funds among catego-les the timing and amount of major conLtracts have been used. b/ For v-,timating the equivalent of the USS25 million loan in Mexican Pesos, the rate of exchange of Iet- 12 5/dollar from 1968 to June 1976 and of Pesos 22/dollar from July 1976 onwards have been apldi ed to the actual lace of the loan disbursemenLts c/ Includes: L-nd Leveling Mex$ 33 million ILnte'rcnnection of wel l.s 35 million srip.ion 1and drainage system:; 80 million FotLl Mex$ 148 million - 112 - Table 12 MEXICO Loan 527-ME Rio Colorado Irrigation Project Loan Funds Disbursement As of December 1977 (US$ million equivalent) Second Year Disbursements Accumulated Semester by Semester Disbursements 1969 June - December 0.08 0.08 1970 June 0.58 0.66 December 1.48 2.14 1971 June 1.87 4.01 December 2.90 6.91 1972 June 2.44 9.35 December 1.97 11.32 1973 June 3.14 14.46 December 2.94 17.40 1974 June 0.90 18.30 Decefber 1.20 19.50 1975 - June 1.47 20.97 December 2.26 23.23 1976 June 0.99 24.22 December 0.62 24.84 1977 June- 0.16 25.00 a/ Original closing date June 30. 1975. b/ Final disbursement - January 17, 1977. MEXICO Loan 527-ME Rio Colorado Irrigation Project Investments and Physical Progress 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979-1985 Investments (US$ Million) Appraisal 16.6 13.8 16.1 12.2 12.2 11.6 12.5 Cumulative 16.6 30.4 46.5 58.7 70.9 87.5 95.0 Actual 0.8 5.9 14.0 23.0 37.0 21.9 22.2 34.0 29.7 10.7 8.6 54.2 Cumulative 0.8 6.7 20.7 43.7 80.7 102.6 124.8 158.8 188.5 199.2 207.8 262.0 Physical Progress (Percentage of Completion) Appraisal 18 15 17 13 13 12 12 Cumulative 18 33 50 63 76 88 100 Actual 0 3 5 9 14 9 9 9 10 4 4 24 Cumulative 0 3 8 17 31 40 49 58 68 72 76 100 I- MEXICO Loan 527-ME Rio Colorado Irrigation Project Investments Breakdown (Mex. Peso Millions) Cumulative up to 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1977 1978 Civil Works - 43.3 118.9 208.8 331.6 148.7 163.5 278.0 311.2 115.0 1,719.0 115.0 Land Levelling - .8 3.3 13.0 37.5 55.0 45.4 73.3 88.0 75.0 391.3 33.0 Subtotal - 44.1 122.2 221.8 369.1 203.7 208.9 351.3 399.2 190.0 2,110.3 148.0 Cement and Steel - 6.4 25.2 30.7 24.8 22.8 28.6 32.9 19.0 3.5 193.9 8.0 Engineering and Supervision 9.8 22.7 28.2 35.2 58.0 46.6 36.8 33.52/ 46.83/ 33.6 351.2 21.0 0 & M Equipment - - - - 5.6 1.1 2.8 7.2 10.8 8.5 36.0 20.0 Total 9.8 73.2 175.6 287.7 457.5 274.2 277.1 424.9 475.8 235.6 2,691.4 197.1 M l- 4- Loan 527-ME Rio Colorado Irrigation Project Check List of Compliance with the Covenants Document Subject Compliance Loan Agreement Section 5.02(c) Regarding the maintenance of Was not complied with. No project accounts records on operations and finan- were ever sent to the Bank nor did Bank missions cial conditions and furnishing had access to the records. See paras 6.21-6.22 them to the Bank Guarantee Agreement Section 3.01 Regarding the funding of the Partially complied with. Funding was very project low at the begining of the project and although increased afterwards it could not offset the inflation. See paras. 3.06 and 3.52-3.55 Section 3.01(a) Regarding the submission -to the Partially complied with. Engineering drawings Bank of plans, specifications, and specifications were submitted to the Bank work schedules and modifications as part of the tender documents for contracts but v in instances late. With the exception of the first design changes (SHR Letter #3677) the Bank was not advised of changes on schedules and of project modifications which were numerous. See paras. 3.04, 3. 39 and 3.52-3.55. Section 3.03 (See Loan Agreement Section 5.02(c) above) and 3.04(b) Section 3.11(a)(b) Regarding water charges to cover Partially complied with. Some regulations have been and (c) O&M costs and investment recovery issued. Water charges do not cover O&M costs. Investment recovery is not yet being collected. See paras 3.01 and 6.13-6.19 Section 3.12 Regarding control of pumping of Partially complied with. Some regulations underground water and coordination have been issued but no full control can be of the use of surface and under- attained since project works are not finished. ground water See paras 3.11 and 3.17 Schedule 1 (4) Regarding the programs for hidro- Partially complied with. The first two have not geologic and cropping pattern - been taken care of in the proper way. The studies, and drainage and soil second two have to be intensify. See paras studies 3.11, 3.18-3.23 and 3.33 MEXICO Loan 527-ME Rio Colorado Irrigation Project Expansion of Cultivated Area and Cropping Intensity Area Cropped Cropping Intensity Considering only Seasonal Perennial Physical Area Considering Area Planted Area Cultivated Crops Crops Total Cropped to Perennials as Double-Cropped ha ha ha % % Actual 1967 170,230 7,951 178,181 88 92 1968 168,108 9,073 177,181 84 92 1969 161,142 12,840 173,982 86 92 1970 157,180 15,455 172,635 85 92 1971 157,951 15,695 173,646 86 93 1972 153,053 18,508 171,561 85 94 1973 147,637 20,611 168,248 83 93 1974 171,995 17,285 189,280 93 102 1975 169,477 20,732 190,209 94 103 1976 179,844 20,040 199,884 97 106 1977 179,248 19,754 199,002 96 105 1978 181,000 21,600 202,600 98 108 Projected 1979 185,500 20,000 205,500 99 109 1980 185,500 20,000 205,500 99 109 H 1981 185,000 20,000 205,000 99 109 1982 185,000 20,000 205,000 99 109 1983 185,000 20,000 205,000 99 109 1984 187,000 20,000 207,000 100 110 1985 and thereafter 187,000 20,000 207,000 100 110 a/ Until 1975 the total district area with water rights was 203,000 ha. Starting in 1976 it was increased to 207,000 ha. Consequently, cropping intensity refers to those areas for the indicated years. MrXICO Loan 527-MI7 Rio Colorado Irrigation Project Composition aud Evolution of Cropping Patterns (physical ha) Wheat Sorghum Other Crops: Barley Maize Sudan (mostly Rye Grass Alfalfa Asparagus Cotton Oats Oilsceds Sorghum (f. silage) vegetables) (annual) (perennial) Vineyards Total AC I U.1 120,777 43,451 4,119 571 - 1,312 - 7,547 404 178,181 88,709 64,970 8,692 2,402 - 3,335 - 8,664 409 177,181 1' 76,651 50,114 29,033 1,340 - 4,004 - 12,516 324 173,982 1170 58,791 79,472 8,393 7,037 - 3,487 - 14,752 703 172,635 1)71 57,550 65,549 21,765 9,775 758 2,534 - 14,284 1,411 173,646 1172 49,759 72,312 16,682 8,624 1,436 4,240 - 16,665 1,843 171,561 1113 40,895 69,573 10,046 10,654 5,893 4,660 5,916 18,869 1,742 168,248 1974 91,982 44,081 9,670 4,173 2,798 3,528 15,763 15,453 1,832 189,280 1975 43,591 61,985 18,410 19,755 6,936 4,701 11,099 18,775 1,957 190,209 171, 41,672 93,191 13,418 11,412 2,151. 5,007 12,993 18,356 1,684 199,884 1(77 62,091 90,938 5,926 2,930 2,441 4,160 10,762 18,182 1,572 199,002 1L78 62,000 81,000 16,100 3,400 1,600 6,000 10,900 19,600 2,000 202,600 Projcc ted 1171 64,500 85,800 11,400 5,000 1,500 7,300 10,000 18,000 2,000 205,500 11180 62,000 91,000 1.0,500 5,000 500 7,500 9,000 1.7,900 2,100 205,500 1181 59,000 95,000 10,000 5,000 500 7,500 8,000 17,800 2,200 205,000 1112 57,000 98,000 8,500 5,800 500 8,200 7,000 17,700 2,300 205,0(00 1083 54,000 102,000 7,000 6,500 - 8,500 7,000 17,600 2,400 205,0 100 '84 5(),000 106,000 7,000 8,500 - 8,500 7,000 17,400 2,600 207,010 1'?A .indl thwrcif(er 50,000 106,000 7,000 8,500 - 8,500 7,000 17,100 2,900 207,010 MEXICO Loan 527-ME Rio Colorado Irrigation Project Relevant Comparisons of Cropping Patterns and Area Planted (Thousands of ha) SARH Appraisal Revised Change Year Base Year Full Develop- Actual Base Year Full Develop- at Full 1967 ment 1985 1977/78 1972/73 ment 1985 Development Cotton 121 128 99 62 50 50 -49 Wheat/Barley 43 43 50 81 72 106 +56 Oil Seeds 4 - 15 16 17 7 -8 Maize/Sorghum 1 - 10 3 9 9 -1 Sorghum/Sudan - - - 2 1 - - Other Crops 1 1 10 6 4 9 -1 1 Other Forages - - 23 - - - -23 Rye Grass - - - 11 - 7 +7 ( Alfalfa 8 5 9 40* 34* 34* +25 Asparagus/Vineyard - - - 4* 4* 6* +6 Cropped Area 178 176 216 224 191 227 +11 Physical Area n.a. n.a. 203 203 n.a. 207 +4 Cropping Intensity - - 1.06 1.10 - 1.10 - * Counted as double cropping. M co - 119 - Table 19 MEXICO Loan 527-ME Rio Colorado Irrigation Project Comparison of Yields of the Main Crops (tons/ha) 1967 1973 1977 Projection 1985 Appraisal At Crops Estimate Actual Actual Actual Appraisal Revised Cotton 2.9 2.1 3.5 3.2 3.5 4.0 Wheat 2.9 3.0 4.0 4.3 3.8 5.0 Barley 2.9 2.1 3.6 3.5 3.8 4.5 Alfalfa 9.0 6.0 12.0 12.0 15.0 18.0 Rye grassb/ n.i../ n.i. 0.525 0.505 n.i. 0.700 Safflower n.i. 1.5 2.0 2.25 2.25 2.5 Sorghum n.i. 4.0 3.0 2.5 4.0 4.5 Soybeans n.i. n.i. n.i. n.i. 2.5 2.2 Maize n.i. n.i. 3.0 2.5 n.i. 3.5 Other cropsc n.i. 5.0 8.0 10.0 n.i. 15.0 - - - - - - - - - Index (1967 = 100.0)- - - - - - - - - All cropsd/ - 100.0 159.2 175.8 251.9 305.1 a/ Not included. b/ Cattle fattening, net liveweight gain. c/ Mostly vegetables. d/ Weighted average yield for all crops and milk in the project area. - 120 - Table 20 MEXICO Loan 527-ME Rio Colorado Irrigation Project Physical Agricultural Out-put (thousands of tons) 1967 1972 1978 1985 Final Development Appraisal Revised Base Year Present Approx. Revised Winter Crops Wheat 125 121 172 267 190 390 Safflower 6 31 25 34 8 Barley (ncl.) 7 51 61 (incl.) 126 Oats nil 1 - 345 - Garlic nil 2 1 6 Rye Grass - - 6 5 Other 3 13 24 57 Subtotal 125 140 270. 385 569 595 Summer Crops Cotton 370 259 149 198 347 200 Sesame - 1 4 Sorghum (Grain) 3 21 4 18 Sorghum (Forage) - - 22 40 - Maize - 5 5 16 Beans - - - Sudan - 5 - Other 3 19 34 60 Soybeans - - 3 25 9 Subtotal 370 265 200 270 412 305 Perennial Alfalfa 45 45 210 235 131 308 Asparagus 1 6 5 - 6 Vineyards - - 4 - 14 Subtotal 45 46 216 244 131 328 Milk - 47 67 59 MEXICO Loan 527-ME Rio Colorado Irrigation Project Domestic Market Price for the Main Agricultural Productsa/ (Valley of Mexicali, Baja California) (Mex$/ton) Products 1967 1970 1972 1974 1976 1977 CONASUPO CONASUPO Cotton 2,580 2,725 2,592 - 3,859 10,476 10,500 - Wheat 913 913 913 1,100-/ 1,300 1,732 2,048 2,050 Barley 950 800 700 n.a. 1,150 1,654 1,980 2,030 Alfalfa 425 420 450 - 900 925 925 - Rye Grass (cattle lwt.) - - - - 13,570 12,050 17,600 - Safflower 1,500 2,000 1,500 1,500 4,000 3,326 4,526 3,900 Sorghum 650 720 720 725 1,400 1,760 1,760 2,030 Soybeans - - - - - - -c/ 5,500 Maize - 1,150 1,150 n.a. 1,500 2,100 2,100 2,900 Vegetable Crops 1,725 1,725 1,725 - 1,875 4,250 5,000 - Milk - 1,700 1,980 - 2,900 3,760 3,980 - Note: Underlined figures are the minimum prices in the period. a! In current pesos. b/ The Mexicali support prices for wheat were higher than those for the rest of the country, because of the high transport cost that would otherwise be involved in bringing wheat to the District from other producing areas of Mexico. c/ Sorghum will be planted for the first time in 1978. No record of local prices exist. - 122 - Table 22 MEXICO Loan 527-ME Rio Colorado Irrigation Project a/ Summary Qf Actual and Projected Value of Production Year and Type Gross Value Gross Cost Net Value of Production Area Cultivated of Production of Production of Production ha --------------MEX$ million (1977)------------------ 1968 Crops 168,517 3,214.2 1,938.4 1,275.8 Livestock 8,664 N.A. N.A. N.A. Total 177,181 3,214.2 1,938.4 1,275.8 1971 Crops 159,362 2,207.8 1,650.2 557.6 Livestock 149.0 129.9 19.1 Total 173,646 2,356.8 1,780.1 576.7 1973 Crops 143,463 2,693.4 1,440.2 1,253.2 Livestock 230.6 177.6 53.0 Total 168,248 2,924.0 1,617.8 1,306.2 1975 Crops 157,335 2,639.3 1,538.6 1,100.7 Livestock 342.9 229.6 113.3 Total 190,209 2,982.2 177 1,214.0 1977 Crops 170,058 3,405.4 1,782.6 1,622.8 Livestock 343.1 259.7 83.4 Total 199,002 3,748.5 2,042.3 1,706.2 1979 Crops 177,500 3,912.5 1,879.9 2,0326 Livestock 375.7 254.3 121.4 Total 205,500 4,288.2 2,134.2 2,15'.0 1981 Cr-ps 179,200 4,188.2 1,835.5 2,352.7 Livestock 2580 371.0 243.7 127.3 Total 205,000 4,559.2 2,079.2 2,480.0 1983 Crops 180,400 4,448.5 1,803.0 2,645.5 Livestock 24.600 361.5 237.2 124 3 Total 205,000 4,810.0 2,040.2 2,769.8 1985 Crops 182,900 4,718.7 1,774.1 2,944.6 Livestock 366.7 231.3 135.4 Total 207,000 5,085.4 2,005.4 3,080.0 a/ All values computed at constant 1977 prices in Mexican pesos. Wheat, beef and cotton prices were adjustcd tc reflect World Bank's price projections i. constant U.S. dollars (October 21, 1977). MEXICO Loan 527-ME Rio Colorado Irrigation Project Comparison of Key Indicators 1967 1973 1977 1985 Appraisal Appraisal Report Revised Unit Estimate Actual Actual Actual Projection Projection Indicators Gross Value of Production Mex$M 4,187.6 3,098.3 2,924.8 3,748.5 4,834.8 5,085.4 Net Value of Production " 1,926.2 868.8 1,313.2 1,706.2 2,649.6 3,080.0 As Percentage* % (222) 100 151 196 (305) 354 Gross Value of Production Mex$/ha 23,793 17,388 17,383 18,833 22,414.4 24,567 Net Value of Production/ha " 10,944 4,876 7,805 8,573 12,283.7 14,879 Physical Area ha'000 176 178 168 199 203 207 Productivity GPV/Area Mex$/ha 23,800 17,404 17,410 18,839 23,818 24,565 As Percentage* % (143) 100 104 113 (143) 147 a/ 1977 Mexican Pesos. * Appraisal figures are in brackets. j MEXICO Loan 527-ME Rio Colorado Irrigation Project Estimated Costs and enefits (1977 Mex$) BENEFITS COSTS Increase in Difference Project On-farm Increase Net Value of between Bene- Year Investment Investment in O&M Cost Total Farm Output fits & Costs 1968 32.7 - - 32.7 - -32.7 1969 236.9 - - 236.9 - -236.9 1970 533.9 - - 533.9 - -533.9 1971 859.2 - - 859.2 - -859.2 1972 1,268.7 - 36.0 1,304.7 - -1,304.7 1973 704.2 - 14.3 718.5 191.2 -527.3 1974 572.0 - 36.1 608.1 735.9 127.8 1975 712.2 - 30.7 742.9 99.0 -643.9 1976 657.8 - 10.0 667.8 230.3 -437.5 1977 227.1 - 26.0 253.1 591.2 338.1 1978 148.3 - 54.0 202.3 692.0 489.7 1979 196.0 - 54.0 250.0 994.7 744.7 1980 153.0 - 54.0 207.0 1,127.7 920.7 1981 153.0 - 54.0 207.0 1,285.6 1,078.6 1982 153.0 - 54.0 207.0 1,446.4 1,239.4 1983 153.0 - 54.0 207.0 1,575.4 1,368.4 1984 125.0 - 54.0 179.0 1,695.1 1,516.1 1985 125.0 218.5b/ 54.0 397.5 1,868.7 1,471.2 1986-2004 - 218.5 54.0 272.5 1,868.7 1,596.2 2005-2017 - - 54.0 54.0 1,868.7 1,814.7 a/ The historical series of costs (1968-77) were inflated by price indices provided by SARH for contracted civil works, land leveling and operation. For crops output and its costs, prices of individual crops and cost per ha for 1977 were used for historical series and projections. b/ This would cover installation of drains comprising of about 100,000 ha and irrigation improvement of a 140,000 ha, both in a period of 20 years. Estimated cost for drain works is Mex$11,500 per ha and Mex$23,000 per ha for irrigation improvements. MEXICO Loan 527-ME Rio Colorado Irrigation Project a! Operation and Maintenance Costs and Financing- (millions of Pesos) Projection 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 A. Expenditures 40.2 40.6 41.5 45.2 58.9 50.0 82.4 84.1 92.2 151.7 205.8 Maintenance 18.0 20.2 17.5 13.7 28.0 29.4 42.6 44.5 53.8 88.0 128.7 Operation 22.2 20.4 24.0 31.5 30.9 30.6 39.8 39.6 38.4 63.7 77.1 B. Financing 40.0 40.4 37.8 45.8 58.2 57.3 82.0 84.6 95.4 166.0 205.8 Water Tariffs 31.7 32.6 28.7 33.1 33.2 30.2 46.5 39.7 40.0 52.7 77.0 (% of A) (78.9) (80.3) (69.1) (73.2) (56.3) (50.3) (56.4) (47.2) (43.4) (34.7) (37.4) Fiscal Transfer 8.3 7.8 9.1 12.7 25.0 27.1 36.5 44.9 55.4 113.3 128.8 C. Balance (B-A) -0.2 -0.2 -3.7 0.6 -0.7 7.3 -0.4 0.5 3.2 14.3 - a/ Does not include purchase of heavy equipment for maintenance. I-n MEXICO LOAN 527 - ME N.e ,. RIO COLORADO IRRIGATION PROJECT UsA I milion 130 -- - - - - 260 PROJECT INVESTMENTS AND LOAN DISBURSEMENTS 17f - 240 1110 - - - ----- - ------ --- - ----_ - - - - - - ----~ -~~ ---.- ---- - - - -- --- --- 22 Projec~ Expenditure Right Scat 100200 g0 --- 0 go 8160 120 (00 Project Exenditre Appri~a Righl Scale 40- -0 3o Loan Dbursemnt 0 AppraI Apiraisal L ..t Sr .... ..... ..''...- '' Loönisbursement Left Scale lo- .20 19118 196 1910 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 15¯4 1985 YEAR Wrk Bnk - 19357 Som,. la r des - 127 - MEXICO LOAN 527 - ME CHART-2 RIO COLORADO IRRIGATION PROJECT % CONSTRUCTION SCHEDULE 210 200- WEIGHTED AVERAGE UNIT PRICE - - - WEIGHTED AVERAGE UNIT COST OF PRODUCTION 190- 1967 = 100.0 180- 170- 160- 150- I 140- 130 I / 120 - 110 - 100 / 90 - \ I 60 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 JUNE 1978 World Bank - 19065 M'~X!CO LOAN 527 - ME R!O COLORADO !RR!GATION PROJECT CHAnT-3 CONSTRUCTION SCHEDULE 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 WORK ITEM FiRIST SECOND THIRD FOURTH FIFTH SIXTH SEVENTH EIGHTH N1NTH TENT i FIEVENTIV v9m 207 Km. CANAL REHABILITATION ' . AND CONSTRUCTION y(101) DRAIN REHABILITATION " "~i ' P z (100Y 1179 Km DR AIN CONSTRUCTMON - .c t '07 W,,is WELL REIIABILITATION WE LL DR I LLING - 7 S250 r 1 12i K ROAD CONSTRUCTION (25 .2 t..-,..I V)o TELEPHONE AND GAUGING --- ..i..,r0 e ae STATIONS × BUILDINGS..=, ---. , ----- -- LAND LEVE L ING -. ~ 7s v Starting Date Appraisal & Actual 1B11E11 ltf Planned ( ) Percent of Completion Word Bank -3975 2033 묘 4PC RI 꺼