Mexico - Highway Sector Project

Document of The World Bank L- FOR OFFICIAL USE ONLY Report No. 2281b-ME STAFF APPRAISAL REPORT HIGHWAY SECTOR PROJECT MEXICO March 7, 1979 Projects Department Latin America and the Caribbean Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Currency Equivalents Currency Unit = Peso (Mex$) US$1 = Mex$ 22.5 after 1976; Mex$ 12.5 previous to and including 1976 Mex$ 1 = US$0.044 Mex$ 1 million - US$44,444 Fiscal Year January 1 - December 31 Weights and Measures Metric: British/US Equivalent 1 meter (i) = 3.28 feet (ft) 1 kilometer (km) = 0.62 mile (mi) 1 kilogram (kg) = 2.20 pounds (lb) 1 metric ton (m ton) = 2,205 pounds Abbreviations ALALC Latin America Free Trade Association ASA Airport Authority BANOBRAS National Bank for Public Works and Services DGAF Director General of Auto Transport ERR Economic Rate of Return IDB Inter-American Development Bank PEMEX Mexican Petroleum Monopoly PFVF Public Freight Vehicle Fleet PPVF Public Passenger Vehicle Fleet SAROP Ministry of Human Settlements and Public Works SCT Ministry of Communications and Transport SPP Ministry of Programing and Budgeting TOFC Trailer on Flatcar (Piggyback Traffic)  FOR OFFICIAL UJSE ONLY STAFF APPRAISAL REPORT HIGHWAY SECTOR PROJECT MEXICO TABLE OF CONTENTS Page No. I. INTRODUCTION ........................................... 1 II. THE TRANSPORT SECTOR ................................... 2 A. General .......................... 2 B. Highways ........................ 3 C. Railways ........................ 3 D. Ports .................................... 3 E. PEMEX .................................... 4 F. Aviation ........................ 4 G. Coordination and Planning ..... ............ 5 H. Past Bank Participation and Experience ............ 6 I. Current Issues and Prospects .... ........... 7 III. THE ROAD TRANSPORT INDUSTRY ............................ 10 A. General ........................................... 10 B. Organization and Structure of the Trucking Industry 11 C. Organization and Structure of the Bus Industry 12 D. Regulation of the Trucking Industry .... ........... 12 E. Regulation of the Bus Industry .................... 1:3 F. Current Issues and Prospects ...................... 14 IV. THE HIGHWAY SUBSECTOR .................................. 15 A. The Network ...... ..............1................... B. Highway Administration ..... ............. 16 C. Highway Planning and Finance ......... ..18........... E D. Highway Engineering and Construction ............. . 20 E. Maintenance ....... ............ .................... 21 F. Rural Roads ................... .................... 22 G. Past Bank Participation and Experience ............ 22 H. Current Prospects .............. .. ................. 24 This report is based on the findings of an appraisal mission which visited Mexico in July 1978. The mission comprised Messrs. R. Burns (Transport Economist, Project Officer), A. van Dijck (Senior Highway Engineer), J.L. Forcina (Loan Officer), C. de Castro (Road Transport Specialist) and R. Anson (Regional Development Specialist, Consultant). The report has been edited by Miss V. Foster. This document has a restricted distribution and may be used by recipients only in the performance1 of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  TABLE OF CONTENTS (Continued) - ii - Page No. V. FRAMEWORK FOR THE PROPOSED SECTOR LOAN 1979-1982 ....... 25 A. General....25 A. Gnea .............................................. 2 B. The Prospect for the Highway Subsector, 1979-1982 26 C. The SAHOP Highway Program, 1979-1982 .28 D. Institutional Framework .33 E. Sector Objectives .34 F. Achievement of Sector Objectives .36 VI. IMPLEMENTATION .37 A. Basis for Bank Pa,rticipation .37 B. Project Preparation, Evaluation and Selection 39 C. Procurement .40 D. Assessment of Risks .44 VII. AGREEMENTS REACHED AND RECOMMENDATION .45 TABLES 1. Distribution of Public Investment by Sector and within the Transport Sector, 1971-1976 .47 2. Railway Operation, Productivity and Financial Indicators, 1971-1977 ..................... I ...... 48 3. Port Traffic Volumes Major Mexican Ports, 1971-1977 .... 49 4. PEMEX Transport Statistics, 1977 ....................... 50 5. Air Traffic Statistics, 1972-1977 ...................... 51 6. The Vehicle Fleet, 19S50-1977 ........................... 52 7. Composition of Public: Vehicle Fleet, 1978 .... .......... 53 8. Comparative Automobile Ownership Rates, 1960-1975 ...... 54 9. Fuel Consumption and Major User Tax Revenues, 1965-1977 55 10. Design Standards ....................................... 56 11. Estimate of SAHOP Highway Budget, 1979-1982 .... ........ 57 12. Estimated Schedule of Disbursements ..... ............... 58 ANNEXES 1. Economic Analysis of Representative Projects .... ....... 59 2. Highways and Regional Development in Mexico .... ........ 73 3. Tentative List of Projects for Commencement in First Year 77 4. Assessment of Financial Intermediary BANOBRAS .... ...... 79 5. Related Documents and Data Available in the Project File ............ . . .............. 85 CHARTS I. Organization Chart cf SAHOP II. Organization Chart of General Directorate for Federal Highways MAP IBRD 13977 - Mexico Highway Sector Project - Main Highway Network  I. INTRODUCTION 1.01 Over a period of 18 years, since 1960, the Bank has made seven loans for highways in Mexico. During this long association, the Ministry of Human Settlements and Public Works (SAHOP) has achieved considerable progress in all aspects of highway planning and development and has provided Mexico with a nearly complete highway system. SAHOP has established a strong main- tenance organization, has pursued construction programs in close collaboration with state and municipal agencies and has built up a fairly good system for project planning and preparation. Mexican highway programs have generally been economically well founded and responsive to the needs of the various regions; they have been flexible einough to meet changing circumstances and to intitiate new programs, particular:Ly in the area of rural roads. 1.02 The last of the Bank highway loans to Mexico was made in FY74 (US$90 million) for the Seventh Highway Project, which provided for upgrading or construction of 16 road sections totaling about 1,800 km of federal high- ways. While SAHOP performed well in matters concerning procurement and exe- cution of civil works, the project encountered considerable initial delays because of the combined effect of [Large inflationary cost increases and bud- getary compressions. The initial project scope and implementation schedule implied an overall level of highway expenditures that was well above Govern- ment intentions and capability. In order to pursue the project as or:iginally planned, SAHOP would have had to give it preferential budgetary treatmaent at the expense of other highway expenditures of equal or greater priority. The scope of the project was eventually reduced from 16 to 12 sections totaling about 1,200 km, and the implementation schedule was extended by two and a half years, up to 1981. The problems encountered with the Seventh Highway Project illustrated the limitations of conventional project lending in Mexico and pointed out the need for improved coordination between SAHOP and the Government agencies in charge of overall programing and budgeting. They also showed the need to consider carefully the future level of highway expenditure at the time of project preparation and appraisal. 1.03 A review of the transport sector conducted in the framework of the economic mission at the end of 1976 concluded that the realization of Mexico's development objectives over the next decade would depend, inter alia, upon the continued availability of reliable and efficient road transport, requiring local improvements and pavement overlays, and upon provisions of upgradled services to rural communities and to regions not yet fully linked with the national network. It was also clear that, over the next five years, SAHOP would have to operate under more stringent budgetary constraints than in the early 1970s, which meant that the improved budgetary coordination called for above would be essential. The sector review served as a basis to iidentify highways as a suitable field for further lending and to recommend the sector approach. While the conventional approach focused Bank attention primarily on the relatively narrow subjects of economic feasibility and engineering of specific subprojects, the sector approach would allow the Bank to integrate its contribution much better into the SAHOP planning process and to support improvements in budgetary coordination, program composition, project selection criteria, and support of overall development objectives.  1.04 After the change of Administration in December 1976, the Government, through SAHOP, explored the possibility of further highway lending. The objectives and implications of the proposed sector approach were reviewed with the three ministries concerned with the sector, i.e., SAHOP, the Ministry of Finance and the recently reorganized Ministry of Programing and Budgeting (SPP). The preparation of the proposed sector loan proceeded over the mid-1977 to mid-1978 period. In July 1978, the project was appraised by a team consisting of R. Burns (Transport Economist, Project Officer) and A. van Dijck (Senior Highway Engineer), assisted by J.L. Forcina (Loan Officer), C. de Castro (Road Transport Specialist) and R. Anson (Regional Development Specialist, Consultant). 1.05 Because of the nature of a sector loan, this report is organized and presented slightly differently from that for a project loan. The second and third chapters are intended to provide a comprehensive view of the transport sector in Mexico and the road transport industry. The fourth chapter on the highway subsector gives more attention than usual to the nature of the highway administration and planning machinery since these are the bases for successfully carrying out a sector loan. The fifth chapter presents and evaluates the frame- work of the proposed sector loan in the context of the Government's highway program and budgeting process. Chapter VI deals with the modus operandi of the proposed sector loan concerning subprojects evaluation and selection, tendering, disbursing and monitoring. The final chapter lists matters which were discussed and agreed during negotiations. II. THE TRANSPORT SECTOR A. General 2.01 Transport has not been a "problem" sector in Mexico in the sense that it has not been a serious constraint to economic development, partly because, over the years, it has had consistent support in terms of public investment, receiving slightly less than 20% of the budget allocation through 1974 (Table 1). For 1975 and 1976, the percentage dropped to 14% and 12% respec- tively, indicating a possible change in its future status in the overall public sector plan. Other than public investment, however, a major reason for the general adequacy of transport has been the existence of a privately financed bus and truck industry that has operated under few Government restrictions. Thus, on the supply side, the public sector has until recently provided ade- quate support for transport investment in all modes, while the private sector has provided an adequate fleet of road vehicles and levels of service. Avail- able evidence in the form of GDP components, road vehicle traffic counts, and actual rail and air traffic figures indicated that demand for transport was growing in excess of 8% annually during the period 1966-1976, in which GDP was growing slightly in excess of 6% annually. 2.02 Within the transport sector, the two major modes are road and rail. Pipelines and coastal shipping are important, but specialized, while aviation, the most dynamic mode in terms of growth, is relatively small in terms of output and is highly specialized. Ports and the merchant marine are naturally  related to road and rail traffic and have been of increasing importance in recent years. The only attempt at estimating the relative importance of the road, rail, and air modes in terms of output was made in 1972, when it was estimated that road transport accounted for 69% of intercity freight ton-km and rail accounted for 31%. For passengers, it was estimated that road trans- port accounted for 90% of intercity pass-km, with 6% for rail and 4% for aviation. All available evidence indicates that the road and aviation share of intercity traffic has been increasing relative to rail. 2.03 In terms of public investment over the 1971-1976 period, highiways received 57% of total transport investment followed by railways with 26%, ports with 10% and aviation with 7% (Table 1). If consideration is taken of private sector investment in the form of road vehicles, total investments were roughly proportional to traffic shares. B. Highways 2.04 The highway subsector is dealt with in considerable detail in Chapters III, IV, and V. Here it is sufficient to state that, after 30 years of road construction, the network is extensive (approximately 62,000 km of paved roads), requiring a large maintenance budget (26% of the 1978 road budget). The vehicle fleet is large (over 1.0 million trucks, 58,000 buses, and 3.0 million auto- mobiles), with traffic counts exhibiting high growth rates, typically in excess of 10% annually. The administration of the road network is complicated (federal, state and local jurisdictions) and is dominated by SAHOP, which has, over the years, developed considerable proficiency at planning and administration. C. Railways 2.05 The Mexican railway network has about 19,000 route-km of standard gauge track, compared to the approximately 39,000 km of paved federal highways (there are another 23,000 km of paved state and local roads), and carried approximately 31% of the intercity ton-km in Mexico in 1972, the rest moving by road. The railways are well run and have experienced an average annual growth rate in traffic approximating that of GDP, i.e., in excess of 6% per year (Table 2). The presence of large amounts of bulk, long-haul traff:ic such as foodgrain, fertilizer, and mineraLs has ensured an important role for rail freight traffic. In spite of favorable traffic conditions, good management, and improving operational efficiency, the railway runs at a large deficit, requiring an annual Government subsidy of approximately US$220 million (Table 2). D. Ports 2.06 Mexican port traffic (Table 3), excluding petroleum, is surpri'singly small for a large country with a 1976 GDP of US$54.2 billion and combined imports and exports of US$10.2 billion, chiefly because much of Mexico's foreign trade moves overland to and f'rom the United States by road and rail. Thus, Veracruz, Mexico's largest port, moved only 819,000 tons of general cargo in 1977, while Tampico, the second largest port, moved 626,000 tons of general cargo, accounting for the bulk of the approximately 2 million tons  - 4 - of general cargo that moved through the 11 principal ports of the country. Most of this traffic moved through the Gulf coast ports of Veracruz, Tampico, and Coatzacoalcos, to and from the U.S. east coast and Europe. 2.07 Mexican port operations have generally been less than satisfactory in the past, partly because of theiLr relative lack of importance to the economy and partly because they were controlled by a department of the Navy until 1976. In addition, strong unions have dominated decision-making and operations in the most important port of Veracruz. The Government is paying much more attention to the port system now irn connection with a regional development plan which calls for the shift of population and economic activity away from the central plateau to the well watered and lightly populated coastal areas. The development of industrial port growth poles is part of this strategy. E. PEMEX 2.08 PEMEX, the state oil monopoly, is one of the largest industrial establishments in Latin America. 'It is a fully integrated operation from ex- ploration through refining to retail sales of refined products. In 1977, it had about 14,000 km of pipelines for major movements of gas, oil, products and petro-chemicals, with about 7,000 kcm under design or construction (Table 4). It has a modern fleet of 30 tankers for coastal movements of crude oil and re- fined products, and it owns or rents over 3,600 railroad tank cars and approxi- mately 2,800 tank trucks. With production in the southern oil fields increasing rapidly, the transport capacity of PEMEX is also being expanded, with US$1.6 billion earmarked for transportation purposes in the 1977-1982 PEMEX investment program (excludes new gas pipeline to U.S.). 2.09 The implication for the public freight transport system can be con- siderable since PEMEX has increasingly chosen highway tank trucks rather than rail tank cars for products distribution in Mexico because of the superior service characteristics of truck delivery. Since petroleum products were 11.6% of rail ton-km in 1976, such decisions can have a significant impact on rail traffic and earnings. For highways, tank truck axle loads are considered a primary cause of road damage by the Ministry of Communications and Transport (SCT), and efforts are now being made to ensure that new tank trucks have an appropriate number of axles to reduce the magnitude of unit loads. For ports, the size of PEMEX tankers and the frequency of their calls can affect the dredging requirements as well as the extent and location of shore facilities. Although PEMEX constitutes a specialized and independent operation, its transport decisions clearly need to be considered in any sector discussion. F. Aviation 2.10 Air passenger transportation has been growing at an average rate of 15% for the last five years (Table 5). In 1977, there were 50 airports capa- ble of handling medium or larger size aircraft, of which 24 could receive jets. Approximately eight million commercial passengers and 71,000 metric tons of cargo were carried in 1977. The Mexico City airport accounts for about 41%  -5- of commercial passenger traffic and 55% of air cargo. Two trunk airlines of approximately the same size serve the domestic and the international markets. Aeromexico is the Government-owned service, Mexicana is 10% Government--owned and 90% private sector. Aeromexico is the chosen instrument of the Government for ensuring services to small or remote communities that would not be justi- fied on financial grounds. Losses on these services have been partly cross- subsidized by profits on high density routes and partly passed on to thbe Government through the annual deficit. In recent years, the annual deficit has been greatly reduced because of a conscious effort on the part of the Government not to subsidize air services. G. Coordination and Planning 2.11 The organization that comes closest to fulfilling the function of a Ministry of Transport in Mexico is SCT. This ministry, however, does not have responsibility for the planning, construction and maintenance of roads or airports or the PEMEX transport investments. Neither is it responsible for planning the rolling stock, locomotive maintenance, and rehabilitation require- ments of the railroads. In terms of investment planning, then, SCT has responsibility only for ports and for new rail line construction. The only place where all transport investments (excluding PEMEX) are scrutinized together is in SPP, where budget examiners who specialize in transport investments negotiate the annual budget cuts among the modes. By directing each agency to prepare a multi-year perspective plan for investments, as well as the annual budget request, SPP has attempted to incorporate a longer view into the budget exercise and to make it more nearly a transport investment planning exercise. 2.12 While SCT has limited authority for investment planning, it does, in principle, have much more authority in the area of transport policy and coordi- nation, including such matters as tariff setting, road user charges, regulation, intermodal transport, licensing, and control of overloading. In fact, this organization devotes most of its resources to dealing with, and attempting to regulate, the well organized bus and truck industry which is described in the following chapter. 2.13 The only truly intermodal activity into which SCT has chosen to enter is that of tariff setting. The pricing of transport services has become in- creasingly important in a period of strong inflation. For the private sector bus and truck operations, this has not presented a problem since failure to cover costs quickly leads to a cessation of services, and the industry has been able to negotiate tariff ceilings adequate for its needs. Moreover, the policing of official trucking tariffs is a near impossible task. It is public sector tariffs for aviation, railways and ports that are a perennial issue. The tariff increase proposal for an individual mode must first be presented to SCT, which then places the case, with the SCT recommendation, before an inter-ministerial committee for prices and tariffs. This committee, composed of representatives from the Ministries of Finance, Commerce, Programing and Budgeting and Communications and Transport, examines the case in the light of national concerns and priorities. Each ministry, to some extent, represents a distinct point of view and has its own constituency. The public sector deficit is the chief concern of the MirListry of Finance, but it has only  -- 6 - one vote. The usual outcome of such deliberation is a tariff increase that temporarily reduces the size of the annual deficit or keeps it from accel- erating. Prospects for the railway, for which this issue is fundamental, are reviewed in paragraph 2.17. H. Past Bank Participation and Experience 2.14 The Bank's major lending activities in transport have been in high- ways and railways. Seven highway loans have been made, the first beginning in 1960 and the seventh in 1974 (Chapter IV, Section G). These loans have been made on a regular basis every two or three years. In railways, three loans have been made: the first (Loan 103-ME) in 1954, followed after 18 years by the second 1/ (Loan 825-ME) in 1972, and the third in 1976. The first and only port loan (Loan 820-ME) 2/ was made in 1972. The sole airport loan was made in 1974. Of the total of 12 transport loans made in Mexico since 1954, five have been made since 1972, covering all four modes of transport. Thus, the Bank's lending operations in the sector have broadened in recent years. 2.15 A number of institutional and policy objectives have been sought over the years in the course of making these loans. For the sector as a whole, the goal was to have transport administration and planning along functional lines, with most policy and investment decisions reached in SCT. Only partial moves have been made toward this goal, most notably with the transfer to SCT of port planning and operations and the responsibility for implementation of new rail construction. Attempts have been made by SCT to take a national view of trans- portation (Chapter IV, Section C), but the broad consensus required to make this a continuing effort in Mexico has been missing so far, and, in the absence of serious problems in the sector, no feeling of urgency has been generated that might lead to this consensus. For roads, the emphasis has been on better planning and project analysis, together with an effective system for control- ling truck overloading and a rational system of road user charges. Advances in planning and project analysis (Annex 1) have been successful enough to allow the consideration of a sector loan for the eighth highway project. Some improvement has been made in the area of road user charges (para 2.18), while work is only beginning on the system for controlling the overloading of trucks (para 2.21). For rail, the emphasis has been on controlling the deficit and improving operational efficiency. Substantial gains have been made in operational efficiency, but little has been achieved in terms of controlling the deficit, for reasons set forth in paragraphs 2.13 and 2.17, although progress has been made in understanding the nature and origin of the deficit throughi studies of passenger operations and low density branch lines. The far-reaching revision made in 1977 in the railway investment program would require a reformulation of the scope of the ongoing project. A new investment program and financial targets are c:urrently being reviewed. For ports, the Bank attempted with the first loan to promote more efficient operating methods in the existing ports and encouraged a commercial approach to tariffs and services. It also recommended that investment planning for the ports be cen- tralized in a national port authority. Substantial operational improvements 1/ PCR issued in January 1979. 2/ PCR issued in November 1978.  - 7 - have been achieved in some of the ports, but little has been accomplished in rationalizing port tariff policies. Rather than a national port authority, the Government has provided for a new subsecretariat in SCT to monitor port operations and control investment planning, a clear improvement over the old arrangements. In aviation, a basic requirement under the first loan wELs to install a commercial accounting system in the airport authority (ASA) for better knowledge of costs and charges. Such a system has been installed recently, and preparation for a second aviation loan is now under way. I. Current Issues and Prospects 2.16 Some of the more important issues in the Mexican transport sector today are: - The rail deficit - Road user charges and diesel fuel pricing - Road and rail competition - Intermodal movement of cargo - Truck axle loads - Food grain imports - Support for regional development objectives. Some of the issues have been mentioned above in connection with previous Bank lending in the transport sector; others are fairly recent developments. Each is described briefly below with an assessment of future prospects. 2.17 The rail deficit is chiefly due to: - Uneconomic passenger services with costs of operation that cannot be recovered throughL tariffs. Even tariffs covering variable costs would be well above those for alternative bus services. - Rapidly rising rail costs in a period of inflation accompanied by slow and inadequate tariff increases. - Strong rail labor unions that make labor cost reductions extremely difficult to achieve. The Government and the railway management are well aware of the causes of the rail deficit and what needs to be done to eliminate it (Table 2). The most important steps would be abandonment of most rail passenger services and frequent freight tariff increases anticipating cost increases. The Government appears to be willing to forego the public revenues needed for covering the rail deficit as well as those of other public enterprises in other sectors. In the context of the Third Railway Project, discussions are under way w:ith the Mexican authorities with a view toward achieving a reduction in the rail subsidy (US$221.0 million in 1977).  - 8 - 2.18 Road user charges and diesel fuel prices are interrelated issues in that fuel taxes are the major public revenue earners from road users, generat- ing over US$400 million of revenue per year (Table 9). In 1975, the Bank financed a road user charge study in SCT; the major findings were that: - PEMEX underprices all ref-ined products since crude oil inputs are valued at less than half world market prices. - Automobile users and private gasoline driven trucks paid gasoline taxes in 1975 sufficient to cover the total annual maintenance cost of highways plus most of the annual construction costs. - Diesel-driven trucks and gasoline-driven public service trucks (para 3.02) paid virtually nothing for the use of the highway system. The matter of underpriced refined products for all users goes beyond the transport sector and is essentially a macroeconomic issue. The policy of the Mexican Government to cross-subsidize long haul intercity trucking with reve- nues from gasoline users is a long-standing one, based originally on the desire to promote commerce and to integrate the country with low cost transport. The wisdom of continuing this policy is open to question, and the issue has been discussed a number of times with the Government from the point of view of public finance, equity, and economic efficiency. A partial move to remedy the situation was made recently when gasoline-driven public service vehicles were also made subject to the large gasoline tax. The Government, however, has decided to continue the policy for diesel-driven trucks with full know- ledge of the arguments against doing so. The proposed sector loan would afford the Bank an opportunity to review the matter with SCT and SPP through periodic consultation (para 5.35). 2.19 Because Mexico has well developed road and rail networks, the ques- tion of road-rail competition is raised frequently. The two issues mentioned in paragraphs 2.17 and 2.18 are also related to modal competition since sub- sidies tend to favor one mode over the other by depressing tariffs below true costs. In the case of the rail subsidy and the long haul trucking subsidy, the two tend to offset each other in the intermodal competition sense. One clear tendency, however, is to encourage the purchase of more transport serv- ices, both road and rail, than would otherwise be the case. In assessing intermodal distortions, it is necessary to consider that the existing highway network is parallel to all the major rail links, and this has been the case for many years. Shippers have been offered alternative road and rail services for a long time, and most of them have adjusted their operation to a modal choice based on a combination of service characteristics and tariffs. The Government, through ownership of the foodgrain distributor (CONASUPO), the fertilizer industry (FERTIMEX), part of the iron and steel industry (SICARTSA) and the petroleum industry (PEMEX), has a great deal of control over the modal choice for the inputs and outputs of these industries. With the notable excep- tion of PEMEX, the great bulk of the traffic associated with these industries moves by rail. Foodgrains, coal, coke, iron ore, fuel oil, fertilizer, and  - 9 - cement form the rail freight traffic base. Of these, only cement is in the private sector, and it moves by rail as a matter of choice. These bulk com- modities move long distances in large quantities, and their movement by rail is appropriate. Less-than-carload rail freight has almost disappeared -- lost to trucking long ago -- an entirely satisfactory and economic development. The modal split for freight in Mexico is not inappropriate in that rail is spe- cializing in long haul, large volume movements and has experienced sustained growth in freight traffic in these areas. The main concern of the railways is not competition from trucking, but, rather, the problems indicated in para- graph 2.17 preceding. 2.20 The intermodal movement of cargo is an issue that has usually been raised in the context of port developments. Container technology has been late arriving in Mexico, partly because of the relatively small amount of gieneral cargo moving through Mexican ocean ports (para 2.06) and the difficulty of dealing with the stevedores' union iin the country's largest port, Veracruz. Handling of containers through the ports and inland will become an increasingly important intermodal transport issue. With the ports under SCT, the subject is now likely to receive the attention it deserves. Piggyback traffic (TOFC) to and from the U.S. via Mexican railways is already well established whereas potential TOFC traffic between Central America and North America via the rail- ways has hardly been explored. Serious consideration has, however, been given to a container land bridge across the narrow Isthmus of Tehuantepec frcm Salina Cruz on the Pacific to Coatzacoalcos on the Gulf of Mexico. Most of th,e infra- structure exists, and a viable alternative to the Panama Canal for high value container traffic is a distinct possibility. 2.21 Conventional attempts at controlling truck axle loads with manned weigh scales have not been successful because of trucking industry opposition. Trucking industry representatives insist that they oppose the system only be- cause it was abused by the weigh scale operators. The current plan to ideal with this potential problem (the extent of overloading is not known) is to determine the extent of violations and subsequent road damage by using dlynamic weigh scales that do not require stopping the trucks; typically, a smal:L per- centage of trucks are responsible for the bulk of the damage. With the in- formation obtained, SCT can, through its newly established contacts with the trucking industry, present the problem and seek a mutually satisfactory way for dealing with it. There seems to be a consensus developing that weight controls would benefit the industry as a whole. SAHOP is planning a survey of axle weights on selected primary roads (para 5.15) which would provide a basis to assess the situation. The periodic consultations (para 5.35) planned over the period of disbursement of the proposed sector loan would allow the Bank to continue to press for progress in this area. 2.22 Food grain imports into Mexico are large (roughly one million tons annually through Tampico and Veracruz alone in the last three years) and likely to increase. The Bank-financed bulk grain facility at Veracruz was operating at capacity the day it opened, and much grain is still moving by slow and costly methods. With rising oil revenues and personal incomes, food imports could easily increase more rapidly since agricultural production will  - 10 - almost certainly grow more slowly t:han grain and grain-related consumption. This is an unpleasant development for Mexican planners to contemplate and is the chief reason why more bulk grain facilities are not being constructed in 4-he ports. The need for this cannot, however, be ignored for long as an obvious and costly bottleneck develops. 2.23 The Bank is attempting, with the proposed loan, to aid SAHOP analysts in developing an approach to assessing the regional development impact of specific road projects (paras 5.28 and 5.34 and Annex 2). The Mexican authorities are also examining the possibility of developing selected port areas into industrial zones capable of generating employment opportunities away from the Mexico City area. This would probably involve the expansion of port capacity and the provision of container-handling facilities. 2.24 In summary, there are areas where the Mexican authorities can improve their transport plan and policy formulation. Failure to do so to date is not based on lack of knowledge. It is largely due to the size and complexity of the country and economy together with a political system that seeks consensus among numerous interested parties. Progress is made, but it is taken in small steps and not before all are in agreement that the step should be taken. Such an approach has led to a transport system that has supported a remarkable record of sustained economic growth. The Bank has made a useful impact on the sector in the past and could continue to do so through financing of projects in all modes and continued dialogue on policies affecting the sector. III. THE ROAD TRANSPORT INDUSTRY A. General 3.01 The most recent data indicate that slightly over one million trucks and 58,000 buses are currently operating in Mexico (Table 6). The trucking fleet has been increasing at about 10% annually in recent years, while the bus fleet has been increasing on an average of 7% per year. The industry that operates these vehicles is large and growing, presumably profitable, and ex- tremely complicated. The bus industry, and especially the trucking industry, have been allowed to expand and operate in a free market environment even though Government regulations to control them were on the books as early as 1932. After almost 40 years of development under a free market regime, the Government decided, in 1971, to establish contact with the road transport industry and to begin enforcing regulations. The Federal Directorate General of Auto Transport (DGAF) was formed in SCT to study this largely unknown industry and to make recommendations on how it could be made to work better. Since 1971, studies have been completed, and attempts to control and influence a portion of the industry have been made. In March 1977, the Government endorsed a five-year road transport program prepared by DGAF, the basic objectives of which are to: - Provide better, more reliable road transport service. - Improve the distribution of services. - Improve the working conditions of drivers.  - 11 - In order to discuss the objectives and the means for achieving them with what was regarded as an excessively fragmented industry, it was considered necessary to develop manageable groups of industry representatives to confer with the Government. The very act of creating these groups is likely to influence the organization and structure of the industry. For this reason, the following discussion on the organization and structure of the industry treats what existed in the recent past as well as what is likely to happen in the future. B. Organization and Structure of the Trucking Industry 3.02 As far as the Federal Government is concerned, the important classi- fication for trucking in Mexico is whether or not the vehicle is registered as "for hire", is over 3 tons capacity, and operates on federal highways. In 1977, it was estimated that approximaLtely 82,000 trucks (8% of the total fleet) fell into this category, which is referred to as the public Freight Vehicle Fleet (PFVF). The PFVF is much more important than its numbers indicate in that it contains the largest trucks engaged in long-haul intercity freight operations and can be characterized as the "modern" element of the industry. The 975,000 vehicles that do not belong to the PFVF are either less than 3 tons, do not operate on Federal highways, or are not for hire. Very little is known statistically about this fleet, but the PFVF has been subject to considerable scrutiny by DGAF in its attempt to regularize the operations of this portion of the industry. 3.03 Table 7 gives the latest information available on the PFVF in terms of vehicle types and firm size. Of the 82,000 automotive units, 61% are rigid body two-axle trucks, 16% are rigid body three-axle trucks, and 23% are truck units for semi-trailers. Approximately half the engine units are single owner- operator units, while the other half are owned and operated by 1,670 separate companies with an average fleet size of 25 units per company. The smaller two-axle trucks predominate in the single owner-operator enterprise while almost all of the heavy trucks (three-axle and tractor trailers) are owned and operated by companies. Prior to 1971, when DGAF began examining the PFVF, the single owner-operator units were 91% of the industry as opposed to 49% in 1977. This increasing concentration of the Larger vehicles in large firms has been a direct result of DGAF efforts since 1971. 3.04 The mechanism for achieving the consolidation of the many single vehicle operations has since become an important structural element of the public trucking industry. The basic approach of DGAF was as follows: - Issue no new federal operating rights to single-vehicle operations. - Encourage association of the single-owner operations into larger entities. - Set up state committees made up of representatives of trucking companies or cooperatives to, deal with state-related problems of the truckers. - Set up route committees to deal with route-related problems.  - 12 - The first two initiatives of DGAF led to the grouping of single-vehicle oper- ations into companies as described in paragraph 3.03. The second two initia- tives had the effect of further grouping the companies into regional cartels with which DGAF could communicate through the route committees and the state committees. In this sense, DGAF has succeeded in its goal of understanding and partially controlling an important portion of the trucking industry. The potential implications of such developments are discussed in Section F. C. Organization and Structure of the Bus Industry 3.05 The bus industry differs substantially from trucking in that the Public Passenger Vehicle Fleet (PPVF) is much smaller (58,000 buses vs 1,054,928 trucks in 1977) and that the proportion of buses in public service is much higher (30% vs 8%). Also, the percentage of the PPVF falling into the single owner-operator category is much lower than in public trucking (10% vs 49%). A smaller vehicle fleet operating on schedules on well defined routes has been easier to regulate by DGAF, and the case for doing so is stronger than in trucking. The organization and structure of the PPVF reflect this reality. DGAF has also formed the equivalent state and route committees for buses. D. Regulation of the Trucking Industry 3.06 For public service trucking, the major areas of regulation are: 1/ - commodity restrictions - geographical operating rights - tariffs DGAF is attempting to manipulate these three major control variables to achieve the objectives set out in paragraph 3.01. 3.07 The commodity restriction is implemented by issuing three different kinds of licenses to PFVF: - general freight - agricultural commodities only - special services Special services include mainly tank trucks for liquids and dump trucks for bulk minerals and construction activities; 16% of the PFVF was operating under this license in September 1977 (Table 7). Agricultural haulers moved farm products, including live animals, and accounted for 42% of the PFVF. General freight is made up of those items not in the other two categories, principally 1/ Axle load controls are applicable to all trucks and are dealt with in paragraph 2.21.  - 13 - manufactured goods, and accounted for the remaining 42% of the PFVF fleet. The smaller trucks predominate in hauling agricultural commodities, while 73% of the tractor--trailer units are hauling general cargo. 3.08 Geographical operating rights are implemented in connection with the commodity restriction. Thus, a general freight license allows the truckers to operate in a specific corridor or route; there are 40 routes in the present system, and agreements of pooling of routes by haulage companies provide con- ditions of greater flexibility than the "corridor" restriction implies. Load- ing and unloading are permitted in zones at the end of each route. An agricul- tural license specifies the geographical constraint differently in that the trucker can operate freely on all federal roads within a radius of 225 km from his base town, or a specific route may be specified, as in the case of general cargo. Special transport licenses allow certain goods (such as bulk peltroleum products) to travel in specialized vehicles on all federal roads. All 'License holders can carry goods between terminals, railway stations and depots f-or loading and unloading. 3.09 All PFVF vehicles are directed by law to use the official tariffs published by DGAF. A tariff revision was completed in 1975 after years of official tariff stability. The tarifff structure is complicated by the oper- ational classification of roads (federal, state, and municipal); federal tariffs apply only to transport on federal roads, while the states have tariffs for state roads; a load originating on a state road and traveling through various states, for instance, may show a cumulative tariff, provided it is as high as the corresponding federal tariff applicable to the portions of federal roads where it will travel. In general, high value commodities such as manufactured goods pay high rates, and low value commodities pay low rates, an attempt to follow the so-called "value of service pricing" structure employed by the railways. As in the case of the railways and ports, these rates bear little relation to costs and are viewed in some cases as a means to influence the cost structure of certain industries and sectors favored by the Government. Thus, a number of exceptions are provided for: the Federal Government is to receive a 50% discount while coal, coke, petroleum products, food grains, sulphur, cement, salt and various others receive an 8% discount. Imported goods, on the other hand, are subject to a 15% surcharge except for foodstuffs and inputs to agricultural production. Although complicated, the official tariff structure is not sufficiently flexible to take into account the size of shipment, empty backhauls, loadability of cargos, and other such items, and there is little doubt that considerable deviation from published rates does take place with the active cooperation of the transporter and the client. This raises issues of enforcement which are discussed in Section F following. E. Regulation of the Bus Industry 3.10 The equivalent to commodity classification in the bus industry is the type of service: first class, second class, or mixed. Otherwise, the  - 14 - geographic and tariff restrictions are similar for public passenger services operating on federal roads. PPVF services are authorized on the same 40 routes defined for general cargo operation, and passenger tariffs are regu- larly published by DGAF in the same manner as for freight transport. However, bus passenger transport is a homogenous enough product to have a well behaved and predictable cost structure which allows the tariffs to reflect costs. Official tariffs are therefore usually adhered to. F. Current Issues and Prospects 3.11 In its desire to establish control over an important segment of the trucking industry, DGAF has created a number of regional trucking cartels for general cargo, imposed commodity restrictions on other truckers, and published official trucking tariffs that are clearly not cost-based. All of these actions are based on the philosophy that trucking on federal highways is a public service and that federal regulatory power is needed to ensure that the public is well served. The question now is to what specific ends will DGAF use its authority and what, in practice, will be the impact of doing so. 3.12 DGAF lacks well defined objectives. Improving the efficiency and distribution of trucking services is a supportable general objective, but, in the absence of data and analyses to indicate the presence or lack of an accept- able level of efficiency or distribution of public trucking services, it is not clear that recent DGAF actions have been in the public interest. The basic implied DGAF assumption that large trucking firms are more efficient and less costly than single vehicle operations has never been established. Most economic studies for other countries have indicated that truck and bus operations do not exhibit so-called "scale economies," i.e., the utilization rates and unit costs of small firms are about the same as those for large firms and, there- fore, encouragement of larger operations will have to be justified on other than economic grounds. The case can be made that most of the DGAF action to date has been anticompetitive and likely to lead to less trucking service at higher prices than previously. Not allowing agricultural truckers to haul fertilizer on an otherwise empty 'back haul is not likely to lead to efficient utilization of trucking capacity, nor is value-of-service pricing an efficient allocative mechanism. Perhaps the most troublesome initiative, however, is the existence of the self-policing and self-regulating state and route com- mittees that have been formed to make implementation of DGAF regulations possible. The large truckers clearly see these committees as a means of reducing competition on major routes. By limiting the number of operators and not allowing new entrants, rrstability" can be brought to the industry. 3.13 The DGAF desire to commiunicate with a stable, well organized portion of the trucking industry does, however, have its merits. These can be listed as follows. - Allows large enough firms to develop to obtain commercial financing at reasonable terms.  - 15 - Makes industry operation of jointly owned terminals easier to organize. Facilitates the development of large scale intermodal movement such as containers and piggyback. Provides a manageable forum for discussion of issues such as axle weight controls, road user charges, and working conditions for drivers. To date, most DGAF efforts have gone toward the organization of the sltate and route committees, together with developing a data base for the federal for-hire trucking service. The industry, as, should be expected, has attempted with some success to turn these initiatives to its own advantage. The danger of such developments should, however, not be overstated, Own account as opposed to for-hire trucking is always an alternative for many shippers shoulcl services decline and rates be raised too high. Then, too, the route committees do not serve as perfect monopolies, and competition within the group for traffic does exist to some degree. Finally, the pressure for entry by smaller non-federal operators will always be there and will increase as profits rise. Excessive profits in the "organized" federal for-hire service will also provide a strong incentive for evasion of the DGAF rules by the more cost-conscious users of transport. On balance, the benefits of organizing the modern portion of the industry may well be in excess of the actual costs. Much depends on the DGAF staff, the tasks that they undertake, and the use to which they put the committees. 3.14 SCT efforts to organize and regulate the industry have not substan- tially altered the fact that: (a) trucking in Mexico, as in most developing countries, is a fragmented and keenly competitive industry even in the presence of regulation; and (b) there is still little understanding of the impact of regulation on the industry. The proposed sector loan would provide the Bank a framework within which to review these matters periodically (para 5.29). IV. THE HIGHWAY SUBSECTOR A. The Network 4.01 The Mexican highway network has evolved rapidly since the introduc- tion of the road transport technology in the 1930s, as shown in the following tabulation:  - 16 - Evolution of Mexican Highway Network 1950-1977 (km) Year Paved Gravel Earth Total 1950 13,595 6,836 2,024 22,455 1960 26,979 11,203 3,860 42,042 1965 34,431 18,373 6,693 59,497 1970 41,947 21,079 6,539 69,565 1975 60,643 77,723 46,393 184,759 1977 62,000 88,310 48,300 198,610 In 1950, there were only 22,500 km of road in Mexico, of which 13,600 or 60% were paved. In 1977, the network had grown to almost 200,000 km, of which 62,000 or 31% were paved; between 1950 and 1975, the length of paved roads grew at an annual rate of 6%. Automobile ownership has been relatively high in Mexico, 33% above the average for Latin America and only 44% below current ownership levels in Europe (Table 8). Recently, the growth rate in paved roads dropped to about 1% per annum reflecting, in part, the near maturity of the main highway network in terms of coverage and the increasing emphasis given to low standard roads in lightly populated areas. The vigorous and successful labor-intensive rural roads program led to the construction of 73,000 km of rural roads during the 1971-1976 presidential period alone. These developments, together with the steady increase in the number of vehicles and in the volume of traffic, have resulted in a large maintenance load and extensive requirements for rehabilitation and strengthening of existing roads and for localized expansion of capacity to relieve emerging bottlenecks. B. Highway Administration 4.02 Administration of the road network has been developed to reflect the distribution of responsibilities of the Government at the federal, state and municipal (local) levels. (a) Federal Highways - SAHOP is directly responsible for the planning, construction, and maintenance of all federal highways; financing is 100% from federal funds. SAHOP has 42 directorates grouped under three undersecretariats; its organization is satisfactory, and it has clearly established lines of authority (Chart I). SAHOP is staffed with experienced engineers and well qualified heads of departments. (b) State Highways - PLanning, construction and maintenance of state roads are the responsibility of the highway administration in each state. Coordination with the federal authorities is carried out by the State Highway ]Board (Junta de Caminos), of which the Governor is chairman and in which SAHOP is represented through its Centro SAIIOP, the fully staffed and equipped dependency of SAHOP in each state. The annual state highway program is submitted to SAHOP for  - 17 - review and approval; financing for construction as well as main- tenance is on a bi-partite basis with one-half federal and one-half state funds. (c) Local Roads (Caminos Vecinales) - Planning and construction of these feeder roads are carried out by SAHOP through its Centro SAFiOP after agreement with the State Highway Board and the municipalities in- volved. Construction is financed on a tri-partite basis: one-third federal, one-third state and one-third local (either municipalities or private enterprise) funds. On completion, such roads are trans- ferred to the corresponding State Highway Board for maintenance with Board funds. (d) Rural Roads - Since 1971, a very large rural roads program hLas been carried out, aimed at joining small towns and villages with the nearest main road and providing temporary employment to the poorer level of the population by using labor-intensive methods. These roads are planned, constructed, and maintained by SAHOP and were financed, until recently, entirely with federal funds. In 1978, funds for construction andl most maintenance were passed directly to the states and do not appear in the SAHOP budget. (e) Toll Facilities - While SAHOP plans and constructs toll roads and bridges, they are operated and maintained by a semi-autonomous public agency "Caminos y Puentes Federales de Ingresos y Servicios Anexos". Funds for financing the toll facilities are provided partly from tolls levied on the users of existing facilities and partly from foreign and domestic borrowing. 4.03 By 1977, the above institutions had provided the following highway network: 1977 Mexican Highway Network by Administrative Unit (km) Paved Gravel Earth Total Federal (federal) 39,470 2,720 1,117 43,:307 State (bi-partite) 17,435 16,788 3,630 37,853 Toll (federal) 1,062 - - 1,062 Local (tri-partite) 4,038 6,338 3,596 13,972 Rural (federal and state) -- 62,461 39,955 102,416 Total 62,005 88,307 48,298 198,610 More than 90% of the federal system and all the toll roads are paved, forming the backbone of the system; 46% of the state system and only 29% of the 'Local system are paved, while none of the rural roads are paved. Coverage is generally adequate, although unevenly distributed. About 25% of the network is concentrated on only 10% of the country's area, with another 50% of the network located on about 40% of the area; the remaining 25% of the network  - 18 - serves about half of the country. However, a major portion of the country is underpopulated, and, in fact, the density of kilometers of road per thousand inhabitants is higher in the underpopulated areas than in the heavily populated ones. 4.04 While it would appear that the division of responsibility set forth above could lead to confusion, the system works smoothly because of the over- riding authority of SAHOP for design standards and specifications and the financial participation of the Federal Government at all levels. The key to successful federal participation at all levels is the institution of the Centros SAHOP. These organizations serve as nearly self-contained federal highway offices in each state, looking after federal interests at each of the three levels of Government and providing the technical expertise that would otherwise be lacking in some areas. Centros SAHOP thus serve both decentral- ization of authority and unification of standards and quality levels. A program of building appropriate offices, warehouses and workshops is being carried out in each State Capital, and the proposed loan may be used for extension of these centers. 4.05 The SAHOP presence does not come without a price, as reflected in the 1978 budget (para 5.10). The management, design and supervision costs of SAHOP (virtually all engineering design is done in-house) account for almost 37% of the total SAHOP road budget. This is a somewhat distorted picture in that 1978 was a very low year for the SAHOP construction budget, and permanent staff expenditures cannot be cut back. The widespread federal presence, represented by well qualified and well paid engineering staff, has been a key factor in the continuing and largely successful development of the Mexican highway system. The cost of this management function will be kept under review during the bi-annual consultations (para 5.31). C. Highway Planning and Finance 4.06 Two recent efforts have been made at national highway planning in Mexico. In March 1976, SCT completed the "Plan Nacional de Transporte", which was an attempt to prepare a comprehensive transportation plan dealing largely with roads and railways. It makes projections of investments and the network for 1982 and 1988, with a general view of the year 1994. It was completed at the end of the Echeverria administration, and, although it was never officially published, it serves as a major unofficial source document for planners in all modes. SAHOP published its own perspective highway plan in November 1976 entitled "Esquema Director para la Infraestructura Carretera." The analysis makes use of city population projections, GDP growth rates, gravity models, and highway capacity assumptions to predict the required highway network in 1995 for two alternative national economic growth rates. Thus, in the long run, SAHOP has a fair idea of the task ahead. 4.07 For more immediate use, consultants retained by SAHOP have recently prepared the first phase of a pavement overlay study in which data relevant to the modernization of the federal highway system have been gathered. A second phase of this study, in which an implementation program for modernization works will be prepared, is scheduled to start early in 1979 (para 5.15). On a wider  19 - scale, SAHOP has prepared several drafts of a six-year program for the current 1977-1982 presidential administration, indicating what is desirable at the federal, state and local levels in terms of investments and maintenance. This material is presented in project det:ail at the federal level, major expense categories at the state level, and allocation by state at the local level. 4.08 The SAHOP planning exercise is based on sound inputs; inventcry and condition surveys are being kept up to date, and a satisfactory system of traffic counts is being enforced. Design standards and cost estimates used are generally realistic. The economic evaluation methods are generally adequate, although some fine tuning to define optimum improvement levels under conditions of scarcity of funds is recognized to be necessary. 4.09 The six-year program and the annual plan are updated and defined in an iterative process of analysis and negotiations involving the states, SAHOP and SPP. This process, in which SAHOP plays the key role, has proved generally responsive to the needs of the various regions and flexible enough to respond to changing circumstances (general reduction of highway expenditures) and to initiate new programs (rural roads and modernization). The continuing planning process outlined above provides adequate guidance to identify requirements and effect shifts in emphasis; it does not result, however, in firm commitment of resources. The annual budget prepared by SAHOP and SPP, with the latter asserting overall control, is the instrument to effect resource allocation and firm up SAHOP's annual program. Ceilings defined in the budget are strictly enforced, and SAHOP's actual expenditure for a given year corresponds closely to its budget allocations. Since no firm commitments are made beyond the next budget year, the Government retains flexibility to respond to shifts in fiscal and development policies. The system rests on the strength of the coordination between SAHOP pluri-annual planning and SPP annual budgeting. This has been a problem in the recent past. One of the important objectives of the proposed sector loan is to assist in bringing necessary improvements to this area (para 5.26). 4.10 The following figures indicate past levels and patterns of road investments and maintenance expenditures for the previous two presidential administrations as well as current estimates of what is projected for the present administration. National Road Construction and Maintenance Expenditures Actual 1965-1976 and Projected 1977-1982 (Billions 1977 Mex$) 1/ 1965-1970 Actual 1971-1976 Actual 1977-1982 Projected Type of Road Amount % of Total Amount % of Total Amount % of Total Federal 23.7 66 27.9 48 30.3 59 State 7.1 20 11.4 19 10.9 20 Local 4.8 13 8.8 15 8.2 15 Rural 0.4 1 10.7 18 3.4 6 Total 36.0 100 58.8 100 52.6 100 1/ Mex$ I Billion = US$44.4 million. For the purpose of comparison with the table in paragraph 5.10, Mex$ 1.00 1978 is approximately equal to Mex$ 0.85 1977.  - 20 - 4,11 During the Diaz Ordaz administration, 1965-1970, emphasis was on the federal highways, which took 66% of the budget; rural roads during that period received only 1% of expenditures. Much of the high level of federal investment was due to a major toll road construction program. During the Echeverria period, 1971-1976, federal road expenditures dropped from 66% to 48% of the total, largely because of a cessation of toll road construction. Rural roads increased greatly from 1% to 18% of total expenditures as a major push was made to produce rural access and rural employment through labor-intensive con- struction methods (para 4.18). For the Lopez Portillo administration, 1977- 1982, all indications are that toltal expenditures will drop below those achieved in the 1971-1976 period in real terms and that emphasis will shift back to con- struction, and especially maintenance and modernization, of federal roads, while the rural roads program, which received so much emphasis during the peak period 1972-1974, would continue at a more balanced pace of 6% of total expend- itures (para 5.08). 4.12 Road user charges in Mexico (para 2.18) are a combination of (a) charges levied by the Federal Government on the manufacture of vehicles and tires, (b) license taxes levied by the State Governments, (c) taxes on fuel and lubricants collected by the national oil company, and (d) revenues from toll roads and bridges. With the exception of toll revenues, none of the charges are earmarked for roads. Federal and state taxes go to the federal budget while PEMEX retains part of the taxes on fuel and lubricants for the financing of its investments, with the remainder going to the general budget. The user charges paid by gasoline-driven vehicles are substantial, generating revenues of over US$400 million per year, 90% of the annual cost of construct- ing and maintaining the entire federal, state and local network (Table 9). While a large portion of this revenue is generated by automobiles in Mexico City, a substantial amount is generated on intercity roads. D. Highway Engineering and Construction 4.13 The General Directorate of Federal Highways in SAHOP (Chart II) has a large, well organized staff that engineers most of the highways and bridges in the country employing modern dlesign techniques; the standards (Table 10) to which these are engineered are adequate. Since funds have been scarce in recent years, SAHOP has responded by selecting lower than strictly justified design standards for new construction and/or by postponing well justified upgrading of existing roads to higher standards. There are competent local consulting firms engaged in highway and bridge design work in Mexico, but SAHOP employs them only when its own staff is fully occupied, and then mainly for bridge design. 4.14 One of the results of 30 years of a sustained level of road con- struction in M1exico has been the development of a large number of experienced and well equipped road contractors and a well defined relationship between contractors and SAHOP. SAHOP has an office in which contracting firms must register, and 2,000 firms have done so; the firms are also obliged to update the register information annually. There is a wide range of capacity among them, with the emphasis on medium-sized firms. Two factors are at play here: (a) SAHOP's practice of contracting by phase corresponding to annual budget  - 21 - tranche; and (b) SAHOP's deliberate policy of distributing its funds among a large number of projects and thus effecting a regional balance. The size of the country, as well as SAHOP's policy of keeping a steady volume of work in all regions, has led to a strong regionalization of contractors. 4.15 Highway construction is carried out under unit price contraLcts awarded on the basis of competitive bidding. Construction contracts are satisfactory and contain an adequate price escalation clause. Under the Seventh Highway Loan, the Bank agreed that invitations to bid would be limited to eight prequalified local firms and eight prequalified foreign firms. SAHOP does not provide advance payments in its contracts; however, BANOBRAS (Banco Nacional de Obras y Servicios Publicos) provides short term loans to con- tractors for work performed but for which payments have not yet been made. Supervision of construction is carried out satisfactorily by SAHOP staff. Competition among contractors is intense, and construction prices are com- petitive, two factors which have discouraged foreign firms from bidding for highway contracts in Mexico, although they have been invited to do so under Bank- and IDB-financed projects. E. Maintenance 4.16 Maintenance expenditures in 1978 account for approximately 26% of the SAHOP road budget (para 5.10). This relatively high proportion reflects two facts: first, the network is aging, and, second, road maintenance require- ments in Mexico are well recognizecl and maintenance quality generally is good. SAHOP, through its Directorate General of Maintenance, is responsible for the maintenance of the federal highway system and part of the rural road system. Routine maintenance is carried out by the Government's own forces; only major betterment works are carried out by contract. The Directorate, which is well organized and staffed with about 20,000 workers, operates in each state through divisions attached to the Centros SAHOP. The Directorate alsoa employs an advanced system of efficiency indicators and road quality indices in monitoring the effectiveness of its operations. Maintenance of state and local roads and some rural roads is the responsibility of the State Highway Boards. 4.17 The maintenance, operation, and allocation of SAHOP equipment are controlled by the Directorate General of Equipment and Transportation, which operates in close cooperation with the Directorate General of Maintenance. The Equipment and Transportation Directorate is well organized and has a strong planning and programing department; systematic records are kept of each item of equipment and each vehicle, recording its use and maintenance data and forecasting the timing of its replacement. The Directorate is experimenting with a system of hire charges in three pilot zones, but SAHOP considers that the experience gathered so far is not sufficient to decide whether to expand hire charge funding nationwide or to continue the present system of direct budget allocations for purchases of equipment and vehicles. The replacement value of the maintenance equipment is about US$120 million with a fleet consisting of approximately 7,500 vehicles and 2,500 equipment items. With an average economic lif'e of about eight years, this implies an annual replacement of about US$15 million, not counting additional requirements due to the increasing workload and the backlog of renewal. The low priority given to highways, reflected in the relatively small budget allocations in the last three years, has led to little replacement of maintenance equipment since committed construction is much more difficult to cut back and equipment pur- chases are easily delayed. As a consequence, the average age of the equipment  - 22 - fleet has increased from a normal level of five-six years to about nine years, and that of the vehicle fleet from three-four years to about six years. What was thought to be a temporary shortage of funds has lasted for several years, and an effort must soon be made to replenish SAHOP's equipment fleet to --oid a sharp decline in maintenance standards. The proposed loan includes funds for the purchase of maintenance equipment and vehicles (paras 5.12 and 5.32). F. Rural Roads 4.18 As indicated previously (paras 4.01, 4.02 and 4.11), the administra- tion of President Echeverria (1971-1976) made a major commitment toward providing basic road access to the villages and rural communities through a largely labor-intensive rural roads program (para 5.07). This was made possible by the presence of four mLajor factors; - an extensive primary and secondary road network; - the extensive SAHOP organization at the state and local levels; - adequate budgetary support; and - political support at the highest level. The extensive existing road network meant that most villages were reasonably close to an all-weather primary or secondary road, and the construction of numerous 5- or 10-km access roads was all that was required, rather than con- struction of long and costly penetration roads. The SAHOP organization again was the key to successful planning and implementation, providing decentralized techinical and administrative support when required. Federal. budget allocations were increased sharply to pay for the works, while the full support of the President and the Minister of SAHOP was required to resist the critics of the program, mainly the engineering and contracting professions, which decried the low standards employed, and the road construction industry, which would have preferred to see the budget suppc,rt primary and secondary road construction. By most measures, the program was a success. 1/ The program has now been fully institutionalized in SAHOP with t:he creation of a Directorate General of Rural Roads and a regular budget allocation, and it will continue at a reduced but steady pace (para 5.08). The Bank has contributed to the construction and improvement of rural roads through various loan operations, especially irri- gation and PIDER rural development projects. Since early 1974, about US$40 million in Bank financing has been included in eight loan packages, which would support an estimated 6,000 km of feeder roads to be constructed or improved. G. Past Bank Participation and Experience 4.19 The total Bank participation of about US$267 million for the devel- opment of the Mexican highway network has had a great impact on the improve- ment of the primary highway system; out of a total length of about 44,000 km of federal and toll highways, the Bank has assisted, or is assisting, in the improvement or construction of about 14,300 km, approximately one-third of the total network. Although the quaLlity of works executed is satisfactory, cost increases, construction delays, and insufficient budget allocations have generally resulted in delayed completion of the projects. 1/ SAHOP, with support from UNDP and participation of Bank staff, organized, in Oaxaca, in October 1978, an international conference on labor-intensive methods for the construction of rural roads.  - 23 - 4.20 Key information on past loans is as follows: Loan No. Date Amount Works Present Status (US$ million) 268-ME Oct. 1960 25.0 3,150 km of federal roads Completed 1968 317-ME June 1962 30.5 385 km of toll roads, 5 toll bridges and 1 ferry boat Completed 1967 354-ME Sept. 1963 40.0 6,0010 km of federal highways and maintenance equipment Completed 1972 401-ME Feb. 1965 32.0 350 kcm of toll roads and 3 toll bridges Completed 1970 528-ME Jan. 1968 27.5 2,200 km of federal highways Comleted 1973 695-ME June 1970 21.8 1,040) km of federal highways Completed 1977 968-ME March 1974 90.0 1,216 km of federal highways, Completion initially 1,875 km expected in 1981 4.21 Loan 695-ME, signed in June 1970, was completed only in Septenmber 1977, or about three years behind schedule. The main reason for the delay was the lag in budgetary appropriations after substantial cost increases had occurred as the result of strong inflation and of increased quantities caused by design revisions. In November 1974, the Bank decreased the disbursement percentage from 33% to 18% since price increases were not yet matched with the devaluation of the Mexican peso. Because benefits were inflating at least as fast as costs, and traffic generally exceeded projections, the projects re- mained economically feasible. The Project Completion Report (PCR) of May 31, 1978 1/ for this loan concludes, in fact, that, with Bank funds tied to a few large projects, failure to allocate domestic funds for these projects for rea- sons of budget limitations or changes in priorities leads to a slow disbursing project, and that designing a loan to finance large numbers of small segments and allowing for changing priorities over time would be desirable if the Bank wishes to disburse larger loans over shorter periods and increase the relevance of its support in the subsector. 4.22 Loan 968-ME was signed in M.arch 1974, and substantial delays were en- countered from the outset because of a shortage of budgetary allocations in the face of sharply higher bid prices caused by rapid inflation. Macroeconomic control measures were taken to reduce the impact of the public sector deficit on inflation; the road budget was cut in real (though not nominal) terms, and the allocation for all projects was reduced across the board, implying that execution would be stretched over several additional years. After the Mexican 1/ The PPAR is scheduled to be issued early in 1979.  - 24 - peso was devalued late in 1976, SAHOP re-evaluated its programs and proposed a curtailment of the project from 16 to 12 road sections (including one section reduced by half). The estimated cost of the reduced project is about the same as that of the original project; Bank participation has been maintained at 40%. In September 1978, works had been completed for 45% of the project, and completion is expected early in 1981. 4.23 The project format of previous loans has, in times of rapid inflation and fiscal restraints, led to slow disbursement and eventual curtailment of project scope. Budget allocations in Mexico for one year reflect the budgetary imperatives of that year. Scarcity is spread around so that projects are slowed down but seldom abandoned. Traditional Bank lending for a specific group of subprojects, to be completed over a four- to five-year period, has proven ill-adapted to the system because it does not provide sufficient flexibility in the selection of subprojects and in the pace of their execution. H. Current Prospects 4.24 A subject that deserves mention is that concerning the distribution of the benefits of highway construction and maintenance. Benefits from federal highway expenditures accrue to the economy at large in the form of savings in, or avoided increases of, vehicle operating costs. Stepped-up maintenance and timely pavement strengthening also result in savings in future reconstruction requirements which could eventually have to be borne by the Government. The direct impact of federal road expenditures will be felt first by users of pri- vate automobiles and by the road transport industry. The relatively affluent auto users, with approximately 3 million vehicles making up between 40 and 60% of traffic volumes on intercity roads, are major beneficiaries from federal road programs. However, the above proportions, compared, for example, with an average of 80 to 90% in the US and European countries, indicate that commercial intercity traffic in Mexico still plays a determinant role in defining infra- structure requirements. This is reflected specifically in the treatment given to time savings in establishing the economic ranking of projects (Annex 1). Savings on the operations of not-for-hire freight vehicles (para 3.02) would directly affect production costs. Savings by public trucks and buses would, in the long run, be passed on to the rest of the economy although, in the case of public trucks (para 3.02), this may, as indicated in paragraph 3.13, be delayed until the competitive pressure generated by increased profit margins is sufficient to force new entries. Small truck operators who generally operate with a smaller margin would benefit directly, as would bus riders who are generally not affluent. 4.25 The overall level oiL road user charges (para 4.12) indicates that major beneficiaries pay for the cost of road investment. In that sense, road investments are not regressive. An additional consideration is that the truck- ing industry and the road construction industry are major employers of semi- skilled and unskilled labor, and that the poor benefit indirectly through employment in industry made possible by an adequate road network. One area where a specific statement seems possible is the regional distribution of highway expenditures and resulting benefits. Road development in Mexico has  -- 25 - played a major role in facilitating the integration of remote or undeveloped regions in the economy. This will continue on account of the structure of the highway planning process which provides for participation of regional and municipal authorities (para 4.02); and a deliberate policy of spatial devel- opment and decentralization (para 5.28 and Annex 2). The information neces- sary to approach the subject of benefit distribution in terms other than the general considerations made above is not available. These considerations suggest, nevertheless, that benefits are broadly distributed with a limited income redistribution effect. 4.26 The prospects for the highway subsector depend very much on the time frame considered. Adjusting to the lower short term priority given for transport in general and roads in particular has been painful for SAHO]?. It is no longer sufficient for a project to be economically justified in order to be included in the road program. The need now is to select the highest yielding projects in order to maximize the return on investments. This fre- quently means the early completion of high yielding elements of a larger project such as reconstruction or dcoubling the carriageway of only a small percentage of an overall road link. SAHOP's attitude as the dominant factor in highway planning and engineering is changing, and the proposed loan is designed to support continued adjustments. 4.27 There are still large areas on the map of Mexico which do not indi- cate the existence of primary or secondary roads. However, most of these areas are mountainous and lightly populated and have little potential for development. The era of providing access to promising territory in Mexico with new penetration roads is drawing to a close. The highest yielding road investments are now maintenance and reconstruction of existing roads (Annex 3). This should set priorities for the next five years. However, longer term projections of traffic densities based on past trends (Table 8) and the anticipated expansion of the vehicle fleet point to the need for considerable expansion of capacity on the primary network in the second half of the next decade. V. FRAMEWORK FOR THE PROPOSED SECTOR LOAN 1979-1982 A. General 5.01 The implications of the preceding chapters concerning the possible role of the Bank in Mexico's highway sector can be summarized as follows; (a) Highway infrastructure in Mexico is in the hands of mature and well developed institutions. The resources allocation process ensures the participation of states and municipali- ties and provides both continuity and flexibility. Improved coordination is needed between pluri-annual sectoral planning and central programing and budgeting. (b) The realization of Mexico's development objectives over the next decade would depend, inter alia, on the continued avail- ability of reliable and efficient road transport. This would  - 26 - require the implementation of balanced highway programs, with special attention given to maintenance and improvement of existing roads. (c) Over the next five years, SAHOP would have to meet the country highway requirements under more stringent budgetary constraints than in the early 1970s. The relative scarcity of resources would call for careful planning and for selective design rely- ing on limited improvements to existing facilities. (d) Past highway loans under the project format have lacked flexibility and suffered from slow disbursements. The sector approach proposed for the present loan would overcome these limitations and would allow the Bank to better integrate its contribution into SAHOP's planning and budgeting process and to broaden the sector objectives of the proposed loan. 5.02 The proposed highway sector loan would finance the foreign exchange component of acceptable capital expenditures included in the annual SAHOP budget and consultant services. Specific subprojects would have to meet agreed requirements concerning economic justification and engineering standards and would have to be implemented in accordance with agreed tendering and procurement procedures. The basic elements of the proposed sector loan are: (a) Definition and agreement on scope and structure of an economically justified and financially feasible highway program covering the period 1979-1982; (b) Definition and agreement on appropriate sector objectives and development of related covenants; and (c) Definition and agreement on specifications and procedures for loan implementation, in particular, project selection, tendering, disbursement and monitoring. The first two elements are reviewed hereafter while the third point, imple- mentation, is developed in Chapter VI. B. The Prospect for the Highway Subsector, 1979-1982 5.03 The projections set forth in paragraph 4.10 represent initial estimates of feasible levels of expenditures. These projections reflect all public highway expenditures, including the federal, state and local components. The program as administered directly by SAHOP, which includes some state and local roads (paras 5.10 and 5.16), is of most interest to define the framework for the proposed sector loan. This program accounts for approximately 59% of national road investment and maintenance expenditures and includes most of the large projects in the country. Even though 41% of national maintenance and construction expenditures takes place at the state and local levels, it should be borne in mind that Mexico has 32 states, and 35 cities with populations in excess of 100,000, which necessarily leads to a large number of small projects in the state and local categories, most of which  - 27 - are too small to be considered for financing under a SAHOP-administered sector loan. For the purpose of the proposed sector loan, the SAHOP budget is there- fore the key document. Before proceeding with an examination of the ',AHOP pro- gram, it is necessary to assess the outlook of overall Government expenditure in roads over the period of the proposed loan. (i) Financial Prospects 5.04 Total expenditures for roads in Mexico for the two previous presi- dential administrations (1965-1971 and 1971-1976) grew in real terms at an average annual rate of about 8.5% (para 4.10). Average annual expenditure for roads in the period 1971-1976 was about Mex$ 9.8 billion per year (1977 prices). Given the rapid increases; in GDP and road traffic during the period (paras 2.01 and 2.04), the 8.5% growth rate in road expenditures was reason- able and desirable. For the 1977-1982 period, it is expected that public expenditure in roads will be about 13% below the 1971-1976 levels, averaging Mex$ 8.8 billion (1977 prices) per year. This will take place in a period when the economy, after two years of very low growth, is expected to expand at a rate in excess of 6% per annum. 5.05 The levels of expenditure projected for roads are less than neces- sary to maintain the current level of services for the demand generated by the growth of the economy. This will manifest itself chiefly in terms of congestion. The Government is aware of this and has consciously made the choice to channel public resources into agriculture and petroleum production. Not only are highways affected, but also the transport sector in general (Chapter II and Table 1). An increased level of allocation would not be possible given the present national tax effort and current pricing policies for public services. A period of austerity is clearly in view for road planners, requiring more careful consideration of how the scarce funds are used. The adequacy of the program can therefore be assessed only in relation to the composition of the program and the criteria for selection of subprojecits. If the policies outlined below on these matters are followed, it should be possi- ble to accommodate the growing demand through 1982 within the envelope of the proposed program without excessive congestion and reduction of serviceability. However, beyond that time, the level of public support for road infrastructure will have to be substantially increased. (ii) Structure of the Program 5.06 An analysis of the structure of the program outlined in paragraph 4.10 shows that the shares going to state and local roads have been almost constant for the last 14 years (approximately 34%) and are expected to be so for the balance of the 1979-1982 period. This reflects, in part, the absorptive capa- city of the lower administrative units and the political realities in Mexico concerning the relative power of federal, state and local Governments. 5.07 Over the period under review, major shifts have taken place or are expected in the relative shares of federal roads (primary network) vis-a-vis  - 28 - rural roads. In the 1965-1970 period, only 1% of expenditures was for low standard rural access roads while the federal roads commanded 66% of expendi- tures. For the period 1971-1976 as a whole, rural roads received 18% of the public funds while federal roads dropped from 66% to 48%. This dramatic shift in the structure of SAHOP's highway program was made possible by a large increase in the level of expenditures in 1971-1976, which allowed real expenditures for federal roads to increase 4.2 billion pesos even though their share was decreasing. This moderated the opposition from the capital-intensive road construction industry since it still had an adequate volume of activity. A liberal allocation of funds for all types of roads plus the active Presidential and Ministerial support made the strong performance in rural roads possible. 5.08 By 1976, three factors developed that made it impossible -- and probably undesirable -- to sustain such a pace in rural roads: - budgetary constraints were instituted to control the high inflation; - the rural road projects with the highest priority had been completed; and - the maintenance burden was becoming significant. The program for 1977-1982 reflects these developments, and the rural roads construction program has been tuned down in absolute terms as well as in shares. It is anticipated that approximately half a billion pesos per year will be spent on rural roads in 1977-1982, representing 6% of total road expenditures (compared with 18% in 1971-1976, and only 1% in 1965-1970). More planning effort is also going into the organization of maintenance operations (currently a SAHOP responsibility with state funds) rather than accelerated construction activities. The 1977-1982 program is by no means a return to the neglect of the 1965-1970 period, and it represents a serious commitment on the part of the Mexican Government to continue supporting basic rural access in a time of austerity when the argument for more expenditure for primary roads is very strong. In terms of traffic volumes handled, the contribution of rural roads is almost insignificant compared to that of main roads. The overall program strikes an adequate balance between rural roads and federal roads. C. The SAHOPllighway Prqgram, 1979-1982 5.09 As indicated in paragraph 5.03, the SAHOP highway program has been taken as the framework for the proposed sector loan rather than a formally approved highway plan. The built-in flexibility of the budgeting process in Mexico would not allow rigid programing for one sector. The Government feels instead that a rolling four-year program providing careful estimates of SAHOP's budget to be updated annually is a better way to improve the budget exercise and to assess the scope of future highway programs and policies and, therefore, to define the framework for the proposed sector loan. Preparation of a four-year estimate of SAHOP's highway budget for the period 1979-1982 has been the focal point of loan preparation. The depth and significance of these estimates lie in the fact that they are recognized as realistic, given the public finance situation in Mexico and the current priorities among sectors. These projections may actually be of more value than usual highway plans, which tend to be a reflection of perceived needs rather than a statement of  - 29 - fiscal capability. In that sense, the four-year estimates provide an adequate framework to assess the absorptive capacity of the sector and project the future rate of disbursement of the proposed sector loan. The initial four-year estimates presented below were confirmed during negotiations and will be updated annually in consultation with the Bank (paras 5.26 and 5.31). 5.10 The table below summarizes the highway expenditures budgeted by SAHOP for 1978 and presents the initial estimates of SAHOP's budget for the period 1979-1982. A more detailed breakdown is given in Table 11. Estimates of SAkHOP Highway Budgets 1979-1982 (Millions of Current Mex$) 1978 Y e a r Total Budgeted 1979 1980 1981 1982 1979-1982 Management, design and supervision 3,300 3,500 4,000 4,600 5,300 17,400 Construction Federal Roads 2,360 2,900 3,600 4,700 5,600 16,800 Purchases Maintenance Equipment 120 400 500 600 700 2,200 Other Maintenance Expenditures 2,180 2,500 2,900 3,400 3,900 12,700 Urban Bypasses 60 100 200 300 400 1,000 SAHOP Centers 100 150 150 100 100 500 State Roads SAHOP Responsibility 540 600 650 700 750 2,700 Rural Roads SAHOP Responsibility 330 400 450 500 550 1,900 Total 8,990 10,550 12,450 14,900 17,300 55,200 (Millions Current US$ Equivalent)/ (400) (469) (553) (662) (769) (2,453) /1 Based on the average market exchange rate of approximately Mex$ 22.5 US$1.0. Percent Distribution of SAHOP Highway Budgets 1979-1982 1978 Y e a r Total Budgeted 1979 1980 1981 1982 1979-1982 Management, design and supervision 37 33 32 31 31 32 Construction Federal Roads 26 27 29 32 32 30 Purchases Maintenance Equipment 1 4 4 4 4 3 Other Maintenance Expenditures 24 24 23 23 23 23 Urban Bypasses 1 1 2 3 2 2 SAHOP Centers 1 1 1 1 1 1 State Roads SAHOP Responsibility 6 6 5 5 4 5 Rural Roads SAHOP Responsibility 4 4 4 3 3 3 Total 100 100 100 100 100 100  - 30 - (i) Overview 5.11 Federal road construction and maintenance expenditures between 1965 and 1976 grew in real terms at an average rate of 2.7% per annum while total road expenditures were growing at 8.5%. From 1977 to 1982, the growth rate for expenditures for federal roads is expected to be approximately 1.7% per year in real terms. The approximate 42,000-km paved federal network carries the bulk of the national traffic, which has been growing continuously in excess of 8% per year. In 1975, it was estimated that approximately 10,000 km of the federal road system had vehicle flows in excess of 3,000 vehicles per day. Further, the age structure of the federal network is as follows: Age Number of km More than 30 years 6,000 20-30 years 8,000 Less than 20 years 28,000 Given the age structure and traffic volumes on the network, it is important to realize that the bulk of future federal road expenditures is determined by maintenance and capacity expansion along existing routes rather than by construction of new routes. The SAHOP program reflects these financial and physical realities and indicates a shift toward maintenance and reconstruction. The change in emphasis would, of course, be gradual as ongoing commitments for new construction, notably under existing Bank and IDB projects, dominate the two first years of the program, 1979 and 1980. (ii) Maintenance 5.12 Maintenance expenditures have grown steadily in absolute terms and in share of SAHOP's program, since the late 1960s. As mentioned in paragraph 4.17, equipment purchases over the last few years were below renewal requirements; the 1978 allocation was only Mex$ 120 million (US$5 million) compared with a rough estimate of average renewal requirements of Mex$ 340 million (US$15 million) annually. Unless the backlog of renewal is made up, there is a real danger that SAHOP's equipment fleet would not have sufficient capacity to support an adequate maintenance program. Projected level of maintenance expenditures (including equipment purchases) would grow from Mex$ 2.3 billion in 1979 to Mex$ 4.4 billion in 1982 and would remain at 27% of SAHOP's budget. The proposed level of maintenance should be adequate. The above budget estimates allow for replenishment of SAHOP's equipment fleet. Equipment renewal studies are completed and up-to-date and would be the basis for an expanded purchase program rising from the 1978 level of Mex$ 120 million to Mex$ 400 million in 1979, and to Mex$ 500, 600 and 700 million in 1980, 1981 and 1982 respectively. Confirmation of SAHOP's equipment renewal program through the period 1979-1982 was obtained during negotiations (para 5.32). (iii) Construction and Improvements 5.13 Construction of federal roads and urban bypasses accounts for 27% of SAHOP's expenditures in 1978, expected to rise to 34% in 1982. This is the category where the careful planning and the shift in project selection  - 31 - criteria called for in paragraph 4.26 would take place. Completion of ongoing new route construction accounts for an important share of this category in the early years. The project to be initiated during the loan period would include mostly: - modernization of existing routes, which implies both structural strengthening (by overlay), and rehabilitation and partial reconstruction; - limited expansion of capacity at congested sections through construction of extra lanes; and - urban bypasses to alleviate urban congestion and separate through traffic from local traffic. 5.14 The emphasis on incremental improvement implies adaptation in project planning and design methods, calling for more attention to identify local inadequacies (stretches of deteriorated pavement, and congestion points). Considerable advances have already been made during preparation and appraisal of the proposed loan in adjusting to the necessary shift in type of projects. The modernization study now under way (para 5.15) has already provided aL basis to plan a specific work program. The proposed sector loan would allow the Bank to support specific subprojects in this category and to advise SAHCIP on adaptation of design methods (para 5.31). (iv) Studies 5.15 The SAHOP program also contains studies, in particular: the second phase of the modernization study (para 5.33); research into the design of pavement management systems, including the effect of heavy axles and the methods for monitoring pavement condition and design overlays. SAHOP also participates in the activities of the major non-profit "think tank" in the country, Fundacion Javier Barros Sierra. This group is organized to explore long term perspectives for the development of the country (horizon 2000); as such, SAHOP intends to undertake, through this group, studies to investigate the future requirements for road transport infrastructure under various scenarios concerning motorization policies, economic growth rates, and spatial distribution of economic activities. The emergence of Mexico as a major oil exporter makes such an exercise particularly relevant provided that it is grounded on facts and takes into account the effects of constraints already apparent today. In view of its implication for planning and policies, the significance of the study component o:E SAHOP's program goes far beyond ilts modest share of the program resources. The proposed sector loan would provide the Bank an opportunity to be associalted with specific studies and to point to specific needs in this area. Should the Bank be requested to finance studies, the terms of reference and qualifications of the consultants would have to be acceptable to the Bank. (v) State and Rural Roads; Centros SAHOP 5.16 The SAHOP road budget also includes state roads, for which SAHOP has contracting responsibility. These are roads partially financed by IDB. Any Bank-financed state roads would also appear in the SAHOP budget for administra- tive purposes. Expenditures for this type of road are expected to remain at  - 32 - 5% of the SAHOP budget for the period 1979-1982. The rural road item in the SAHOP budget is purely for maintenance of rural roads in areas where this cannot be turned over to local Government for lack of funds. At 3% of the SAHOP highway budget, it is a significant outcome of the successful rural roads construction effort of previous years. Finally, SAHOP's budget also provides for extension of facilities in the Centros SAHOP. (vi) Administration 5.17 Currently, the administrative costs of running SAHOP account for 37% of the total SAHOP road budget. This relatively high level reflects the excess design and supervision capacity of a ministry that has expanded regu- larly even though the actual physical workload in recent years has stabilized. It also reflects the generally adequate level of salaries for engineers, which is responsible for the high retention rate of qualified and experienced people. This allocation has the attribute of a fixed cost and would require little expansion to accommodate large budget increases. It is expected that the share of this item in the total budget would drop from 37% in 1978 to 32% in 1982. The periodic consultations scheduled under the sector loan would provide opportunities to scrutinize this item (para 5.31). (vii) Financing of SAHOP's Ilighway Programs 5.18 The SAHOP budget is f:Lnanced by the Bank, IDB, and the federal budget. The figures presented below indicate how the 1979-1982 budget projec- tions in paragraph 5.10 would be financed and the relative-importance of the proposed Bank loan of US$120 million during this period. Financing of SAHOP Highway Budget 1979-1982 (Millions of Current Mex$) 1979 1980 1981 1982 Total Federal Roads IDB-financed 400 800 1,000 1,200 3,400 State Roads IDB-financed 400 400 500 550 1,850 Federal Roads Bank Seventh Project 968-ME 1,100 1,100 600 - 2,800 Proposed Sector Loan Projects 1,000 1,400 2,050 2,350 6,800 Total Foreign Financed Projects 2,900 3,700 4,150 4,100 14,850 Percent of Total SAHOP Budget 27 30 28 24 27 Foreign Contribution as Percent of SAHOP Budget 13 14 13 10 12 Of the estimated Mex$ 55 billion SAHOP program for the 1979-1982 period, about 8.0 billion (14%) is associated with previous Bank and IDB loans. (This assumes a negotiated US$60 million IDB loan in effect.) The proposed loan would add about Mex$ 6.8 billion to the Bank involvement (12%) so that, in total in this period, foreign financing agencies would cover about 27% of SAHOP's program. Since the Bank and IDB finance only the estimated  - 33 - foreign exchange component, their contribution would cover about 12% of SAHOP's budget. External financing obtained by the Government in relation to highway expenditures are includedl in the SAHOP budget allocations and do not come in addition to such allocaitions. (viii) Assessment and prospects 5.19 SAHOP's highway program for the period 1979-1982, as reflected by the four-year budget estimates, would, on the whole, be adequate to carry the road system through what would clearly be a holding period. The balance of the program and its structure are reasonable and prudent, given the austerity imposed on the sector. The program would be followed by a phase of renewed expansion made possible by increases; in oil revenues. Such expansion, starting in the mid-1980s, would be necessary, to meet mounting requirements for capacity increases and to accommodate changing consumption patterns related to the rapid rate of motorization (Table 8). D. Institutional Framework 5.20 The implementation of SAHOP's road program and the financial aspects of the proposed sector loan would involve the following four departments and institutions: SAHOP, SPP, the Ministry of Finance and BANOBRAS. 5.21 As executing agency, SAHOP would prepare and submit subprojects; it would prequalify contractors, tender contracts, supervise their implementation and report on completion of subprojects. SAHOP would also carry out studies and surveys to develop the basis for the periodic assessments of the program and the updating of the rolling four-year estimates of SAHOP budgets. As indicated in Chapter IV, SAHOP has the capacity to carry out these functions. 5.22 SPP is the most important ministry for achieving the sector objec- tives associated with the proposed loan. SPP is the single authority in charge of reconciling the investment plans of ministries such as SAHOP with overall budget constraints and national priorities. The need for strengthening coor- dination between sectoral programs and overall fiscal and development policies has been recognized by the present administration. SPP is generally well organized and has adequate authority and staff. In the specific area of transport, SPP does not have the technically experienced personnel to compete with SAHOP expertise, but this questiLon is being addressed. SPP staff who played a key role in the preparation of the proposed sector loan are fully qualified to assume their functions Linder the proposed sector loan. The role of SPP would be to provide SAHOP with the necessary guidance to review and adjust its pluri-annual budget projections in line with inter-sectoral allocations and specific development policies affecting transport demand and the resulting requirements for road infrastructure. 5.23 The Ministry of Finance is interested in the proposed sector loan as part of its plan to finance the public budget. It also plays the important role of indicating the likely constraints on public finance that SPP musIt use in preparing sector budgets. It is a well developed ministry with a full com- plement of experienced professionals and can provide the information needled by SPP on a timely basis.  - 34 - 5.24 The financial intermediary and borrower would be BANOBRAS, which functions as the principal Mexican Government entity for contracting domestic and foreign source credit for public works construction in Mexico. Credit obtained by BANOBRAS is relent primarily to the Federal Government for public works, low-cost housing and the improvement of public transportation facilities. Loans are also made to states and municipalities. The Federal Government has been the major stockholder of BANOBRAS since its creation in 1933 and currently owns, directly or indirectly, approximately 98% of the outstanding capital stock. Under the Organic Law, the Federal Government controls the policies of BANOBRAS, and, in practice, substantially directs BANOBRAS' borrowing and lending activities. The chief function of this insti- tution would be to request disbursements and to repay the loan out of the Federal Government's budget, the bulk of the record-keeping and administration remaining in SAHOP (para 6.14). The institution has a well established financial basis and adequate staff and sufficient experience in similar activities to carry out its assigned responsibilities in relation to the proposed sector loan (Annex 4 provides a detailed assessment of BANOBRAS). E. Sector Objectives 5.25 While the Mexican highway sector is generally well organized and planned, there are several areas where improvements can be made. The objec- tives sought through the proposed loan are: - to induce improved coordination between SPP and SAHOP in producing a well balanced, financially feasible road investment program - to assist SAHOP in shifting emphasis from new road con- struction to reconstruction, rehabilitation and maintenance - to support research aimed at assessing the regional impact of specific highway projects and - to promote rational policies affecting vehicles using the highway system. Paragraphs 5.26 through 5.29 following expand on the nature of these objectives, while Section F outlines the specific agreements obtained from the Government to achieve these sector objectives. (i) Coordination 5.26 SPP has attempted (para 2.11) to incorporate a longer view in the annual budget exercise by directing each agency to prepare a multi-year pers- pective plan. As in most such exercises, the agency plan tends to understate the true costs of its proposed undertakings and to overstate the minimum physical facilities required. In order to be of any real use in the budget exercise, a realistic financial constraint must be combined with accurate cost estimates. The proposed sector loan seeks to support achieving this through the preparation of a rolling four-year program of SAHOP highway budgets pre- pared and updated annually in cooperation with SAHOP, SPP, Finance, and the Bank. The budget estimates in paragraph 5.10 were prepared in this manner  - 3 5 -- and will serve as the major point of reference in assessing the progress and evolution of the Mexican highway program as well as being the source of Bank- financed subprojects. These estimates were confirmed during negotiations. (ii) Balance of SAHOP's Progran 5.27 The emphasis of SAHOP's program has to shift from construction of new roads to reconstruction, rehabi:Litation and maintenance of what is now a nearly mature highway network. This involves more attention to relatively small incremental improvements, requiring fine-tuned design, including, in particular, pavement overlay programs and limited upgrading and construction of bypasses to remedy local inadequaLcies and bottlenecks. The proposed sector loan would provide opportunities for progress in the following ways: - through the annual consultations on SAHOP's highway program; - through the review and analysis of specific subprojects proposed for Bank financing; - through direct support and supervision of specific studies necessary to assess requirements and adapt design methods, in particular, the preparation of an overlay program; and - through support of systematic planning and funding on an annual basis for the replacement of maintenance equipment and vehicleas. (iii) Highway Planning and Regionial Development 5.28 As mentioned in paragraph 2.23 and Annex 2, decentralization of population and economic activity is a key development objective in Mexico. The National Urban Development Plan recently approved by the President makes explicit identification of geographical regions that can best support future development. The proposed sector loan would support further research ini- tiated jointly by SAHOP and the Bank during loan preparation on the concept of regional development indices as a way of gauging the effectiveness of specific highway investments in supporting regional development policies. SAHOP would, under the loan, pursue an agreed program of research on the matter (see Annex 2 for a description of work completed to date and paragraph 5.34 for proposed loan conditions). (iv) Transport Policies 5.29 The proposed sector loan would promote rational policies that affect the use of the highway system. SCT is chiefly responsible for policies concern- ing vehicle regulations, road user charges and intermodal transport. A good study on road user charges has been completed under the Second Railway Project. During loan preparation, the road transport industry (Chapter III) and the prospects of intermodal transport were examined in considerable depth, laying the ground for future dialogue with SCT on policy matters affecting transport services. The loan would provide periodic joint reviews (para 5.35) between the Bank on one hand and SCT and SPP on the other hand on the following mattersi  - 36 - axle weight control road user charges - road transport regulation and - intermodal transport F. Achievement of Sector Objectives 5.30 In order to pursue the sector objectives set forth in Section E above, agreements will be sought to: - specify the nature and timing of the annual discussion and updating of the SAIIOP highway program; - specify the steps to be taken to ensure adequate maintenance of the road system; - ensure execution of a pavement overlay study leading to a comprehensive national overlay plan; - ensure the completion of an agreed program of research on the regional development impact of specific road projects; and - provide regular reviews with SCT and SPP of progress in areas concerning road transport policies. The following paragraphs indicate the essential elements to be included in the proposed agreements. 5.31 A four-year projection of tentative SAHOP highway budgets will be prepared by SPP in conjunction with SAHOP and will serve as the focus for annual consultation with Bank staff in September of each year. The projections will be prepared in sufficient detail to indicate the level and balance of projected SAHOP spending in both constant and current currency units. Bank staff will then be given the opportunity to comment on the program in relation to the previous year's estimates and in the overall balance and composition of the program. The relationship of ongoing and proposed candidate projects for financing under the loan will also be discussed. The Bank will then sub- mit its comments and estimate cif what constitutes the four-year SAHOP program that is to serve as the point of reference for progress in the sector. The Mexican Government will indicate its acceptance or objections to this estimate and the accompanying comments. The estimate resulting from this process will be referred to as the "tentative SAHOP highway program". It will be adjusted after the budget is published in January at a second meeting with Bank staff and will serve as the budget framework for the sector loan until the program is revised and updated the following September. This procedure was confirmed during negotiations, and the initial four-year projection was set forth in signed minutes of negotiations. 5.32 The Government agrees to purchase equipment and vehicles for highway maintenance in accordance with technical studies satisfactory to the Bank prepared by the maintenance and equipment departments in SAHOP and will  - 37 - otherwise provide sufficient funds in the SAHOP budget to ensure adequate and timely maintenance for the federal highway system. The minimum estimates of budget support necessary for this item were confirmed during negotiations. 5.33 SAHOP will carry out a second phase of the national road moderniza- tion study, the first phase of which has been completed recently. The second phase will conclude with a five-year program of modernization of the exist- ing road network and will include, as basic technical data, frequency distri- butions of axle weights on major road sections. A draft report should be made available to the Bank before December 31, 1980. This was confirmed during negotiations. 5.34 SAHOP will continue the joint Bank/SAHOP research effort in quan- tifying the ranking of specific highway projects in supporting regional development objectives. The composite index concept developed by the initial Bank/SAHOP joint effort will be developed further, and experiments with dif- ferent combinations of variables and weights will be performed on an agreed sample of subprojects submitted undier the sector loan for Bank financing. The Bank will be given the opportunity to comment on the draft final report. Copies of the final report will be made available to the Bank on or before December 31, 1980. This was confirmed during negotiations. 5.35 The Ministry of Programing and Budgeting will, at the time of the annual January budget review, cause the Ministry of Transport and Communica- tions to provide information and/or reports on development in the four areas mentioned in paragraph 5.29. Bank staff will prepare comments on the review and will submit these comments to the concerned ministries for their irLforma- tion. This was confirmed during negotiations. VI. IMPLEMENTATION 6.01 This chapter describes the modus operandi of the proposed sector loan to serve as a guide for future supervision efforts and as the basis for the loan agreement. A. Basis for Bank Participation 6.02 The tentative SAHOP budgets set forth in paragraph 5.10, the finan- cial plan for the budget in paragraph 5.18 and the more detailed budget presentation in Table 11 serve as the basis for assessing the Bank's role in SAHOP's highway program over the period 1979-1982. The analysis of the financing of SAHOP's highway program, 1979-1982, in paragraph 5.18 indicates that the proposed sector loan would cover an increasing share of SAHOP's federal road construction (including strengthening and reconstruction) program as the ongoing IDB loans and Bank Loan 968-ME reach completion. The specific areas which could be supported by the! proposed sector loan are indicated below. Ongoing subprograms as well as work scheduled for tendering in 1979 were carefully scrutinized to ascertain that they would provide a basis to commit at least 40% of the loan resources by June 30, 1980 and to disburse an expected US$35 million, i.e., 29%, by end 1980.  - 38 - Estimate of Potential SAHOP Budget Items to be Financed by Sector Loan 1979-1982 (million current Mex$) Y e a r Total Item 1979 1980 1981 1982 1979-1982 Construction Federal Roads 650 900 1,400 1,650 4,600 Purchases Maintenance Equipment 1/ 250 250 200 100 800 Other Maintenance Expenditures 2/ - 100 200 300 600 Urban Bypasses 50 100 150 200 500 SAHOP Centers 50 50 - - 100 State Roads SAHOP Responsibility - - 100 100 200 Rural Roads SAROP Responsibility 3/ - - - - - Total 1,000 1,400 2,050 2,350 6,800 (Millions current US$) (44) (62) (91) (104) (302) 1/ Financing mainly of maintenance equipment and a small amount for selected consulting services. 2/ Financing only of "isolated reconstruction" element of this budget item. 3/ Rural road financing under the project is not envisaged. The Bank would continue to help finance rural road construction through other operations such as Bank-supported rural development projects (notably the PIDER program) and regional development projects. 6.03 The total cost of the component of SAHOP's 1979-1982 program that could be partially financed under the proposed sector loan totals about Mex$ 6.8 billion, out of a total of Mex$ 55.2 billion, i.e., approximately 12%. This could accommodate a loan as large as US$150 million. The amount of the proposed loan has been set prudently at US$120 million, which would ensure full commitment over the four-year period 1979-1982. Should the Government decide to step up the level of commitment, the preparation of a possible followup operation could be advanced. The volume of external financing has not influenced the total size of SAHOP's budget. External contributions are offset by equal reduction in net allocation from Government funds (para 5.18). 6.04 In September 1978, SAHOP prepared, and submitted to the Bank for review, 22 subprojects for which funds have been requested under the 1979 budget. A preliminary assessment indicates that the proposed subprojects would be eligible for financing under the proposed sector loan. The port- folio of construction subprojects presented by SAHOP for 1979 is briefly summarized in Annex 3; it includes six new constructions, six urban bypasses, and 10 modernization projects. The proposed portfolio is representative of the type of subprojects that would be supported under the loan. Total cost of the various subprojects is about Mex$ 6.4 billion, with a 1979 budget requirement of about Mex$ 900 million. Further projects are expected to be submitted for 1979, particularly in the categories of equipment purchases and studies.  _ 3s ( B. Project Preparation, Evaluation and Selection 6.05 The two important concepts to be considered here are the individual subprojects and the contracts associated with each subproject. In general, the completion of a subproject involves more than one contract. The sub- project is the unit considered for purposes of project preparation, evaluation, and selection, while contracts associated with the subproject are the basis for disbursements. Thus, not all eligible subprojects would be completed by 1983, the end of the disbursement. period, and SAHOP would pursue their implementation under its program for subsequent years in accordance wit:h the schedule defined for each subproject. All contracts accepted for disbursement under the loan would have an initial completion date before December 31, 1982. The commitment of funds would be related to accepted contracts pertaining to eligible subprojects. Commitments would be monitored closely to ensure ade- quate progression in disbursement. It is expected that SAHOP would submit subprojects well in excess of what would be strictly necessary to meet the commitment and disbursement estimates outlined in paragraphs 6.17 and 6.18. Since most road construction contracts in Mexico cover only one year, the proposed system would provide ample flexibility to adjust the progression of commitment and disbursements. The process for declaring a subproject eligible is outlined in paragraphs 6.06 through 6.09 while that for acceptance of specific contracts is explained in p.aragraph 6.13. 6.06 Subprojects are defined and prepared in SAHOP with some input from SPP, and, unless the Bank disagrees, the SAHOP definition stands. OverLy large subprojects with large fluctuations :Ln traffic volumes and engineering standards over their length would be avoided, and the Bank could request that a large subproject be broken down and analyzed as smaller, more homogenous subprojects. This is not expected to be a problem.. The third example in Annex I is such a case. While the 201-km Queretero-San Luis Potosi road could easily be jiusti- fied as a single project of doubling two lanes to four lanes, the Mexicans have chosen the most congested 46 km as a subproject and submitted it for eligibility. Such an approach is also consistent with the resource picture and priorities outlined in paragraphs 5.13 and 5.14. Each subproject submitted under the loan would have to be prepared with the depth and level of detail normally required for feasibility studies. Each subproject would have to be technically sound and economically justified, with an economic rate of return of at least 12%. SAHOP has the capacity and the basic information to carry out economic analyses acceptable to the Bank and has demonstrated this capacity on several occasions in connection with Bank appraisals and project completion reports. Equipment purchases would be supported by a renewal program including an analysis of capacity requirement and of availability of existing equipment. Studies would be evaluated on the basis of the potential merit of the speacific investment or policy improvements which they are expected to bring about. Subprojects concerning workshops and Centros SAHOP (which would be of re:Latively small size) would be supported by an analysis of functional requirements and justification of location and general layout. The above was confirmed at the time of negotiations.  - 40 - 6.07 SAHOP would prepare, for each subproject, a basic data sheet presenting the key technical, design, cost, economic and tender information as well as further preparation planned. The attachments to Annex 1 illustrate the form of the information for actual subprojects. Bank staff would study these data and the supporting documentation mentioned above, ask additional information as required (especially on new construction projects), and determine subproject eligibility. Rejected subprojects may be re-entered for future consideration when lack of information is corrected, if the project scope is amended, or if changed conditions improve supporting analysis. 6.08 Declaring a subproject eligible would mean agreement of the Bank to participate in the financing of the contracts related to the implementa- tion of this subproject, provided budget funds are allocated and the tender process has been executed according to agreed procedures and in accordance with the agreed design standards. Subprojects resulting in average annual total expenditures (not counting the year of smallest expenditure) of less than Mex$ 10 million (US$450,000) would not be considered under the proposed sector loan; a similar limit of Mex$ 5 million would be applied to maintenance equipment purchases. This restriction would not apply to studies. 6.09 Once a project has been declared eligible, SAHOP would complete its preparation involving, for example, detailed engineering for road construction, drawing up specifications for equipment or detailed terms of reference for a study. If the data supplied in the original basic data sheet would change significantly during the design stage of the subproject, SAHOP would send a revised evaluation sheet to the Bank for its confirmation prior to inviting tenders. This procedure was agreed upon during negotiations with the terms set forth in a draft supplementary letter. C. Procurement 6.10 Civil works and equipment eligible for financing under the proposed sector loan would be procured through international competitive bidding in accordance with Bank guidelines for procurement. Suppliers of equipment pro- duced within Mexico would enjoy, as usual, a margin of preference of 15% on CIF prices or the prevailing customs duty, whichever is lower. (i) Prequalification 6.11 Civil works contractors would be subject to prior prequalification. For Mexican contractors, the system of prequalification now in use under the Seventh Highway Project (para 4.15) would be continued. For foreign contractors, invitations to prequalify would be sent annually to embassies and published in Development Forum; the notices would contain information on the program as a whole. Prequalified contractors should, however, take the initiative to inform themselves in Mexico on future tenders; all, up to a total of eight, foreign contractors who wish to be invited for any subproject should be included in the short list of tenderers. If there are more than eight prequalified foreign firms, they may be invited on a rotational basis, following current practice. Pending the new  - 41 - prequalification for foreign contractors, foreign contractors would be invited to bid according to the procedures agreed for Loan 968-ME. As to Mexican contractors, SAHOP would continue to invite at least eight tenderers following current practices. SAHOP would keep the Bank informed on the list of pre- qualified foreign contractors and on those refused prequalification, iLf any, and the reasons thereof, and on the interest shown by contractors to be invited for specific tenders. (ii) Tender Process 6.12 SAHOP would (without the Bank being involved at this stage): - prepare tender documents - update cost estimates - prepare short list of eight Mexican contractors and of all interested foreign contractors (for civil works only) - issue tender invitations - open tenders - prepare tender evaluations - award contracts - sign contracts and - issue notices to proceed. (iii) Bank Acceptance of Contracts 6.13 SAHOP would send to the Bank, after signature of the contract, the following documents: - an updated basic data sheet - the tender evaluation report indicating the award decision and its justification and - the contract. The Bank would review this information and, if satisfied, indicate no objection, thereby declaring the specific contracts accepted, or request further information. The Bank would have the right to reject a contract and request new tendering in case agreed tender procedures were not followed. If the Bank should reasonably determine, after consultation with the Borrower, that the execution of an eligible subproject (in the sense of paragraph 6.08) is proceeding on the basis of a contract inconsistent with the procurement  - 42 - procedures agreed under the loan, the Bank may cancel from the loan the amount corresponding to its scheduled participation in such contract. In case a higher-than-expected tender price would make the project economically not feasible, SAHOP and the Bank would review the subproject justification and explore possible reduction in design standards. The above was confirmed during negotiations. (iv) Maintenance of Records 6.14 The proposed loan would provide financing for a great number of contracts. As indicated in paragraph 5.24, a relatively small part of the work involved in administering disbursements would be delegated to BANOBRAS. The bulk of the administration would fall on SAHOP, which would keep records of all tenders and contractual documents and official decisions pertaining to each eligible subproject. SAHOP would prepare periodic statements of the list of approved contract(s) for each subproject, showing total expenditures to date and monthly payments. The Bank would reserve the right to inspect and review the documentation maintained by SAHOP and BANOBRAS and to be sent specific portions of the detailed records. This arrangement was confirmed during negotiations. (v) Disbursements 6.15 Disbursements for various types of subprojects would be made against normal documentation on the following basis: - 47% of total expenditures for civil works; - 100% of foreign expenditures for imported equipment or 100% of the ex-factory cost net of taxes of locally produced equipment; and - 47% of total expenditures for consulting services. The foreign exchange component of the civil works has been estimated at 45% on the basis of an analysis prepared by SAHOP. The analysis has been checked by the Bank and is acceptable. The contribution of 45% would be achieved by disbursing 47% of all payments to contractors except the repayment of the 5% retention withheld from contractors during the guarantee period following the completion of works. The analysis assumes that the works would be carried out by firms established in Mexico and excludes the cost of fuel for which Mexico is a net exporter. The foreign exchange component of the previous highway loan (968-ME) was 40%; however, in 1974, the Mex$ was overvalued; it was subsequently devalued to about half its value against the US$, resulting in higher foreign exchange components than for Loan 968-ME. Although no foreign contractor is expected to be the lowest bidder for civil works, foreign suppliers are likely to be successful for heavy equipment, such as bulldozers, rollers and motorgraders. It is expected that the studies to be financed under the loan would n,ot deal with usual feasibility and engineering matters but with problems of planning and methodology (para 5.15). As such,  - 43 - they would require the participation of foreign experts (estimated US$7000 equivalent on average per man-month). While it is the Government's policy to encourage the use of local consultants, specialized foreign consultants are engaged routinely. Since the foreign exchange costs of such studies would vary depending on whether local consultancy firms (with or without association or subcontracting with foreign firms), or whether foreign firms (with or without association or subcontracting with domestic firms) would be engaged, there is likely to be a wide range in the foreign exchange component of such studies, estimated to range between 20 and 80%. The proposed disbursement percentage of 47% (same as for civil works) would simplify loan administration and would, on the average, broadly reflect the foreign exchange component of such studies. 6.16 Unless substantial changes are made in the scope of the subproject (for which Bank approval would be necessary), the Bank would disburse 47% of expenditures for civil works contracts, including cost overruns of 10% of the contract value for one-year contracts, and, respectively, 25% and 40% for two and three-year contracts in order to take into account small change ord'ers and escalation payments. Documentation justifying such increases would not be required in these cases. 6.17 Tentative estimates of the progression of disbursements are given below and in Table 12. Disbursement Schedule for Sector Loan (Million US$) Total 1979 1980 1981 1982 1983 1979-1983 Construction 5 22 27 34 88 Equipment and Consulting Services 6 12 10 4 32 Gross Disbursement 11 34 37 38 120 Less Slippage 1/ -5 -5 -7 -7 24 - Net Disbursement 6 29 30 31 24 120 Accumulated Disbursement 6 35 65 96 120 - 11 Slippage would be due mainly to construction delays reflecting past experience and to normal time laig in payments and disbursements. 6.18 The achievement of the above progression in disbursement would imply the commitment of about 15% of loan funds by June 30, 1979, 40% by June 30, 1980 and 75% by June 30, 1981. The progress of commitments, together with the rate of disbursements, would be reviewed bi-annually during the September and January consultations (paras 5.31 and 6.19). The above schedule of disbursements and commitments was reviewed and agreed during negotiations. (vii) Monitoring 6.19 The annual consultations on SAHOP's highway program and the budget would provide the major opportunity for reviewing the progress of the loan. The basic data sheet for each eligible subproject would be periodically up- dated by SAHOP to show progress under various contracts. The basic data sheets  - 44 - would be the repository of all important information concerning the implementa- tion of the loan. After completion of the subproject, the Bank would receive the "acta de recepcion de las obras", the Mexican document which summarizes the technical, financial and legal aspects of a completed work; also, a final updated basic data sheet would be submitted, which would serve as a basis for the completion report. 6.20 Monitoring and supervision of the proposed sector loan would focus on the four following areas: (a) The evolution of SAHOP's highway program to be reported on annually by SAHOP in advance of the September consultation (para 5.31). (b) The progress made in pursuing sectoral objectives of the loan, to be reviewed jointly by SPP, SAHOP and SCT on the basis of ad hoc documentation. (c) The implementation of eligible subprojects. In preparation for the January consultation, SAHOP would prepare: (i) a complete file of updated basic data sheets for all eligible subprojects; and (ii) an assessment of overall progress and common problems. (d) The progress in commitments and disbursements of loan funds. SAHOP would prepare: (i) monthly statements of expenditures; (ii) a quarterly summary of expenditures and physical progress on all eligible subprojects; and (iii) a bi-annual review and updating of a commitment schedule in which existing and projected contract commitments are set forth in order to achieve target disbursements as indicated in paragraph 6.17. The above monitoring arrangements were discussed and confirmed at the time of negotiations. D. Assessment of Risks 6.21 SAHOP's budget estimates have been carefully reviewed and are con- servative; they may, in fact, understate the real intention of the Government. It is highly improbable, given the economic outlook of Mexico, that actual expenditure would be below these estimates. The amount of the loan has been set prudently below the potential disbursement basis. There is no serious risk that SAHOP's program would not provide sufficient basis to commit and disburse the loan as scheduled. The wide range of subprojects and the flexi- bility afforded in the commitment of loan funds would prevent a situation  - 45 - wherein the loan would be tied to a relatively small number of large sub- projects stretching over an extended completion period (para 6.05). 6.22 The modus operandi of the loan rests on the capacity of SAHOP and, to a lesser extent, of BANOBRAS. Past BANOBRAS performance has been satis- factory, and SAHOP has an established record of competence and stability. Past and ongoing experience under Bank highway projects suggests that the risks affecting project implementation are below normal. Past experience on the economic performance of Bank-financed highway projects, as well as the current orientation of SAHOP's program toward so-called incremental projects which usually yield high returns, suggests that there is no undue risk affecting the economic soundness of subprojects. 6.23 One area of uncertainty concerns the objectives of the loan in the area of transport policy under the jurisdiction of SCT. The possibility to effect improvement in the areas mentioned in paragraph 5.29 through scheduled consultations would depend on the opening of a full dialogue with SCT to review the implications of its policies concerning, in particular, the regula- tion of the trucking industry. The relative modesty of the objectives set in this area reflects the tone of the Bank's lending relations with Mexico and an appreciation of the greater effectiveness of structured consultations as opposed to formal undertakings on policy matters. VII. AGREEMENTS REACHED AND RECOMMENDATION 7.01 During negotiations, agreement was reached with the Governme;nt on the following: (a) the initial four-year SAHOP highway budget estimates (paras 5.09 and 5.26); (b) SAHOP's equipment renewal program through the period 1979-1982 (para 5.12); (c) the nature and timing of the bi-annual discussions and updating of the SAHOP highway budget estimates (para 5.31); (d) minimum estimates of budget support to ensure adequate maintenance of the road system and adequate replacement of maintenance vehicles and equipment (para 5.32); (e) execution of a national road modernization study (para 5.33); (f) completion of an agreed program of research on the regional development impact of specific road projects (para 5.34);  - 46 - (g) provision for annual review of progress on selected issues in the area of road transport policies (para 5.35); (h) project evaluation and selection procedures (paras 6.05 through 6.09); (i) prequalification, tendering, and award of contracts (paras 6.11 through 6.13); (j) maintenance of records (para 6.14); (k) disbursement procedures (paras 6.15 through 6.18); and (1) monitoring procedures, in particular the preparation of an initial commitment schedule with provision for bi-annual updating (paras 6.19 and 6.20). 7.02 Subject to the above, the project provides a suitable basis for a Bank loan to BANOBRAS of US$120.0 million equivalent; the terms would be 17 years, including a four-year grace period. March 7, 1979  - 47 - TABLE 1 MEXICO HIGHWAY SECTOR PROJECT A. Distribution of Public Investment by Sector in Mexico, 1971-1976 (Percent) Sector 1971 1972 1973 1974 1975 1976 Industry 4:L.7 34.1 31.3 34.0 39.5 42.6 Welfare 211.6 24.1 28.6 25.2 20.4 19.2 Agriculture and Rural Development 14.5 14.7 13.6 16.0 17.2 16.7 Tourism 0.3 0.4 0.4 0.6 1.0 1.1 Administration and Defense 1.3 3.4 1.7 1.6 2.0 3.4 Communications 1.6 1.8 6.0 6.0 5.4 5.5 Transport 19.0 21.5 18.5 16.6 14.3 11.5 B. Distribution of Public Investment Within the Transport Sector in Mexico (Percent) Mode 1971 1972 1973 1974 1975 1976 Average 1971-1976 Highways 64 68 64 50 48 50 57 Railways 22 15 22 29 34 34 26 Ports 9 7 11 10 10 10 10 Aviation 5 9 3 11 7 6 7 Source: Table 5.18,Special Study of ithe Mexican Economy: Major Policy Issues and Prospects, IBRD working document Dec. 2, 1977. November 1978  - 48- TABLE 2 MEXICO HIGHWAY SECTOR PROJECT Ferrocarriles Nacionales de Mexico A, Freight and Passenger Traffic,1965-1978 Year Net Ton-Kilometers Passenger-Kilometers (Billions) (Billions) 1965 14.6 3.0 1970 18.1 3.4 1975 27.0 2.6 1976 26.8 2.5 1977 283.7 3.1 1978 1/ 30.0 N.A. B. Productivity Measures Rail Freight Operations, 1965-1977 Average Speed Net Ton-Km Average Load of Net Ton-Km Wagon of Train per Loaded Wagon per Available Turnaround Year (Km/Hour) Train/Km (Tons) Car-Day (Days) 1965 23 674 39 1,414 13 1970 22 833 42 1,401 16 1975 23 1,051 47 1,577 16 1976 26 1,006 48 2,122 12 1977 25 1,161 47 2,186 12 C. Financial Indicators, 1965-1977 Average Cost Average Income Average Cost Net Operating 2/ Average Income per Ton/Km per Pass-Km per Pass-Km Deficit per Ton/Km Cargo Cargo Passengers Passengers Millions Year (Mex$) (Mex$) (Mex$) (Mex$) (US$ equivalent) 1965 9.9 11.2 4.4 14.7 48.6 1970 9.8 13.1 4.6 17.3 96.6 1975 14.0 19.5 6.6 34.8 188.6 1976 14.2 26.8 7.3 56.3 221.9 1977 19.9 31.2 8.3 46.2 221.5 1/ 1978 traffic estimate based on four months actual data. 2/ Exchange rates used 1965-1975 is 12.5 Mex$ per US$. For 1976-1977 the rate was 22.5 Mex$ per US$. Source: Ferrocarriles Nacionales de Mexico July 1978  MEXICO HIGHWAY SECTOR PROJECT Port Traffic Volumes Major Mexican Ports 1971-1977 (1,000 Tons) Veracruz Tampico Manzanillo Mazatlan Guaymu Coatzacoalcos I/ Year GC DB L GC DB L GC DB L GC DB L GC DB L GC DB L 1971 766 801 361 335 934 2,162 173 187 125 77 - 126 84 130 322 112 1,151 1,112 1972 780 1,178 398 416 1,074 2,647 132 603 96 143 - 30 112 84 270 111 1,471 900 1973 901 1,298 465 427 1,335 3,641 152 434 161 218 - 92 111 16 450 143 1,850 1,364 1974 1,061 1,223 437 613 1,557 1,855 295 534 169 123 91 80 81 290 446 104 2,721 2,623 1975 932 933 355 790 1,290 1,338 240 678 241 115 371 64 80 295 651 83 2,150 6,927 1976 867 560 357 789 1,545 1,130 366 364 152 96 176 25 37 171 642 92 1,449 7,123 1977 819 593 352 626 2,002 770 176 165 82 100 67 18 159 187 154 90 1,537 9,130 11 Increase in liquid cargoes 1975 on, reflects increases in Mexican petroleum exports, Note: GC * General Cargo, DB = Dry Bulk, L = Liquid Source: SCT July 1978  - 50 - TABLE 4 MEXICO HIGHWAY SECTOR PROJECT PEMEX Transport Statistics 1977 Pipeline Network: Under Design Planned Existing (Km) or Construction (Km) Network (Km) Crude Oil 4142 1579 5721 Refined Products 4674 2428 7102 Gas 5637 2642 8279 Total 14453 6649 21102 Vehicle Fleet: Number Number Number Total Capacity Rented Owned Ships 30 654,136 DWT - 30 Rail Tank Cars 3635 194,472 m3 1834 1801 Tank Trucks 2873 85,095 m3 2100 773 Other Trucks 229 N.A. - 229 Traffic Flows: Millions Ton/Km % Total Crude Oil Pipelines 26130 38 Refined Products Pipeline 6105 9 Gas Pipelines 6112 9 Other Pipelines 430 1 Coastal Shipping 22301 33 Rail Tank Car 1200 2 Tank Trucks 5658 8 Other Trucks 315 _ Total 68251 100 Source: PEMEX July 1978  - 51 - TABLE 5 MEXICO HIGHWAY SECTOR PROJECT Air Traffic Statistics 1972-1977 _/ Passengers Freight Year Million P'assenger-km Million Ton-km 1972 4,259 54 1973 4,948 62 1974 5,893 68 1975 6,657 76 1976 7,564 85 1977 8,470 93 1/ Figures above are the sum of the two major Mexican airlines, Mexicana and Aeromexico, and include both dormestic and international traffic. Source: ICAO Statistics September 1978  - 52 TABLE 6 MEXICO HIGHWAY SECTOR PROJECT The Vehicle Fleet 1950-1977 Year Private Cars Truiks Buses Total 1950 173,080 111,252 18,466 302,798 1960 483,101 293,423 26,126 802,650 1965 771,118 :388,684 30,702 1,190,504 1970 1,233,824 524,985 33,059 1,791,868 1971 1,338,404 .554,497 34,480 1,927,381 1972, 1,520,144 592,772 35,723 2,148,639 1973 1,766,504 645,323 37,043 2,448,870 1974 2,053,241 728,965 41,053 2,823,259 1975 2,400,930 887,912 50,762 3,339,604 1976 1/ 2,689,042 967,824 54,315 3,711,181 1977 3,011,727 1,054,928 58,117 4,124,772 1/ Figures for 1976 and 1977 are mission estimates Source: Asociacion Mexicana de la Industria Automotriz, 1976 September 1978  - 53 - TABLE 7 MEXICO HIGHWAY SECTOR PROJECT Composition of Public Vehicle Fleet (By type of ownership and type of vehicle) Companies 1/ Single- Type of Vehicle (Units) Vehicle owners Total (Units) A. The Public Freight Vehicle Fleet as of Early 1978 2/ Rigid trucks: 2 axles 15,908 33,784 49,692 Rigid trucks: 3 axles 10,575 4,812 13,387 Semi-trailers (truck): 2 axles 2,521 263 2,784 Semi-trailers (truck): 3 axles 12,600 1,528 14,128 Sub-total units with engines 41,684 40,387 82,071 Semi-trailers (trailer): 1 rear axle 422 31 453 Semi-trailers (trailer): 2 rear axles 12,900 1,112 14,012 Trailer: 2 rear axles 184 16 200 Sub-total detachable trailers 13,506 1,159 14,665 Other 5 9 14 Total Units 55,195 41,555 96,750 B. The Public Passenger Vehicle Fleet as of Early 1978 Bus 15,074 334 15,408 Other (under 10 pass.carrying capacity) 298 1,425 1,723 Total 15,372 1,759 17,131 1/ DGAF estimates there were 1,670 companies operating at the time. 2/ Percentages of PFVF by type of license are as follows: General license 42 Agricultural license 42 Special license 16 Source: DGAF September 1978  - 54 - TABLE 8 MEXICO HIGHWAY SECTOR PROJECT Comparative Automobile Ownership Rates 1960-1975 (Passenger cars per 1,000 population) Country or Region 1960 1970 Most Recent Estimate 1/ Mexico 13.4 24.5 40.1 Europe 16.8 49.9 90.0 Latin America and Caribbean 11.7 22.1 29.9 N. Africa and Middle East 7.5 11.5 15.4 Africa South of Sahara 2.1 3.4 3.9 South Asia 0.9 2.2 2.1 East Asia and Pacific 2.7 7.0 9.3 I/ "Most recent estimate" in most cases is around 1975. For comparison, 1975 figures were computed for Mexico. Source: Mexico - Mission estimates, July 1978 Rest of World - "World Economic and Social Indicators", IBRD, October 1978 November 1978  - 55 - TABLE 9 MEXICO HIGHWAY SECTOR PROJECT Fuel Consumption and Major User Tax Revenues 1965-1977 Diesel 2/ 4/ Diesel 1/ Engine Gasoline 3/ Gasoline Consumption Tax Consumption Tax Total (Million Revenues (Million Revenues Revenues Year liters) Million Mex$ liters) Million Mex$ Million Mex$ 1965 2,028 25.2 5,415 - 1970 3,433 69.3 8,131 - 1971 3,582 72.5 8,711 - 1972 4,067 79.9 9,402 - 1973 4,606 85.1 10,438 - 1974 4,811 86.0 10,638 449.0 535.0 1975 5,271 96.3 11,171 4,989.0 5,085.3 1976 5,779 109.9 11,730 5,950.4 6,060.3 1977 N/A 119.9 N/A 9,498.4 9,618.3 1/ In 1976, approximately 66% of diesel fuel consumption for road transport was by trucks, 19% by buses, 12% by rail and 4% by coastal shipping. 2/ Fixed annual tax depending on size of vehicle. 3/ In 1976, approximately 70% of gasoline consumption was by automobiles, 24% by trucks, and 6% by buses. 4/ Special gasoline sales tax introduced in late 1974, extended to trucks 1977. Source: Fuel Consumption - Asociacio'n Mexicana de la Industria Automotriz, 1976 Tax Revenues - Ministry of Finance, July 1978 November 1978  MEXICO HIGHWAY SECTOR PROJECT Design Standards Geometric Flat and Hilly Medium Heavy Plan Design Unit Rolling Country Mountainous Mountainous Special Design speed km/h 110 90 70 60 A " 90-110 80 60 50 B 80 70 50 30 C 70 60 40 25 D 60 50 30 20 Special 2 Width Surfacing m 7 7.3 7.3 7.3 Special 4 " 2X7.5 2X7.5 2X7.5 2X7.5 AP 6.7-7.3 6.1-6.7 6.1 6.1