Mexico - Special study of the economy - major policy issues and prospects (Vol. 1 of 2) : Main report

Report No. 2307-ME FILE CZ'py Special Study of the Mexican Economy: Major Policy Issues and Prospects Volume I Main Report May 30,1979 ..; ,A,. f Country Programs Department I Latin America and the Caribbean Regional Office *i. w [t 5iI FOR OFFICIAL USE ONLY Document of the World Bank 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 Unit - Peso (Mex$) Since September 1, 1976 the Mexican peso has been floating and has fluctuated around Mex$ 22.60 to the US dollar since mid-1977. On May 4, 1979, the peso traded at 22.83 per US dollar. US$ 1.00 Mex$ 22.83 (May 4, 1979) Mex$ 1.00 US$ 0.0438 ( " " " ) ABBREVIATIONS AND ACRONYMS BANRURAL Banco Nacional de Credito Rural CFE Comision Federal de Electricidad CIMMYT Centro Internacional de Mejoramiento del Maiz y el Trigo CONACOSA Comite Nacional de Coordinacion del Sector Agricola CONASUPO Compania Nacional de Subsistencias Populares CPNH Comision del Plan Nacional Ridraulico FERTIMEX Fertilizantes Mexicanos, S.A. IMF International Monetary Fund INFONAVIT Instituto del Fondo Nacional de la Vivienda para los Trabajadores NAFINSA Nacional Financiera, S.A. PEMEX Petroleos Mexicanos PRONDAAT Programa Nacional de Desarrollo Agricola en Areas de Temporal SAHOP Secretaria de Asentamientos Humanos y Obras Publicas SPP Secretaria de Programacion y Presupuesto SRH Secretaria de Recursos Hidraulicos  FOR OFFICIAL USE ONLY This report was originally prepared by Messrs. Juan Villarzu and Joel Bergsman on the basis of the findings of two missions that visited Mexico in April/May 1977 and February 1978. It has been reshaped and adjusted to new information by Messrs. Alexander Nowicki and Jorge Garcia-Mujica, following extensive discussions held in Mexico in January 1979. The following persons participated in the 1977 mission: - Miguel Martin Fernandez Public Finance - Darrel Fallen-Bailey Energy - David Hughart Energy - Catherine Pierce Demographic Trends and Prospects - Marto Ballesteros ) - Renato Rossi ) - Carlos Moret ) Mexican Agricultural Sector - H.L. Manning (Consultant) ) - Gunther Schramm ) - Frank Veneroso (Consultant) Mexican Financial Markets - Christine Wallich Mexican Tax System Important contributions to the Main Report were also made by Marisa Fernandez-Palacios (Economist), Richard Sheehan and Rene Ribi (Power Engineers), and Ruth McCrea and Isabelle Wieviorka (Research Assistants). 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.   TABLE OF CONTENTS Page No. SUMMARY AND CONCLUSIONS CHAPTER I. THE CRISIS AN]D STABILIZATION PERIOD ..............1 The Origins of the Crisis ..... ................. I' Legacy of tlae Crisis ........................... 7 CHAPTER II. MAJOR DEVELOPMENT ISSUES ......................... 12 Poverty and Productivity ..... .................. 12 Income Distribution ............................ 18 Population, Employment and Basic Needs ......... 21 Agricultural Potential ......................... 28 Regional Imbalances ............................ 37 CHAPTER III. INVESTMENT FINANCING .42 Public Sector Finance .45 Tax Revenues .45 Government Expenditures .50 Public Enterprises .53 Private Savings and the Capital Market 58 External Savings .63 CHAPTER IV. GROWTH PROSPECTS ................................. 65 Development Strategy: Past and Future ......... 65 Open or Protected Economy ..... ................. 67 Development Bottlenecks ........................ 69 Some Non-Alternatives .......................... 70 Growth and Investment .......................... 72 Inflation and the Rate of Exchange .... ......... 77 Some Alternatives .............................. 79 ANNEX I POLICY OPTIONS AND DEVELOPMENT OUTLOOK ......... 83 ANNEX II TECHNICAL NOTE ON DATA AND ADJUSTMENTS - INCOME DISTRIBUTION 1963-68-75 .101 ANNEX III TECHNICAL N'OTE ON THE ENERGY SECTOR: POWER, ANALYSIS OF MARGINAL COST .110 ANNEX IV VOLUME II - STATISTICAL APPENDIX  -2- TABLE OF CONTENTS (Continued) TEXT TABLES CHAPTER I Table 1: Short-Run Economic Indicators, 1965-1976 Table 2: Production of Manufactured Goods, 1972-1976 Table 3: Short-Run Economic Indicators, 1976-1978 CHAPTER II Table 4: Population and Labor Force, 1950-1969 Table 5: Patterns of Land Tenure and Production, 1968 Table 6: Theil Indices of Inequality for Urban and Rural Areas Table 7: Population Projections, 1970-2000 Table 8: School-Age Population, 1970-2000, According to Projection 2A Table 9: Labor Force Projections Table 10: Agricultural Production, Share in GDP, and Foreign Trade Table 11: Agricultural Production, 1945-1975 CHAPTER III Table 12: Fixed Investment and Its Financing, 1965-1976 Table 13: Investment and Its Financing Table 14: General Government Consumption, 1971-1976 Table 15: Federal Government Current Transfers Table 16: Government Subsidies for Industrial and Agricultural Production, 1977 Table 17: Private Sector Savings Table 18: Ratios of Total Assets of Financial Institutions to GDP CHAPTER IV Table 19: GDP and Investment Requirements, 1965-1982 Table 20: Public Sector Investment, 1977-1982  -3- TABLE OF CONTENTS (Continued) ANNEX TABLES ANNEX I Table 1: High Oil Exports Scenario (A) Investment and Its Financing Table 2: High Oil Exports Scenario (B) Balance of Payments Projections, 1978-1982 Table 3: Public Sector Savings Scenario (A) Table 4: Gas Balance Table 5: Domestic Consumption of Fuel and Gas Table 6: High Oil Exports Projections Scenario (A) Exports of Petroleum Derivates, 1978-82 Table 7: High Oil Exports Scenario (A) Required Oil Production and Aggregate Balances Table 8: Comparison Between Two Policy Scenarios A & B ANNEX II Table 1: Basic Data on Size Distribution of Family Income: 1963, 1968 and 1975 Table 2: Estimates of Disposable Personal Income Table 3: Summary Indicators of Income Inequality (First Adjustment Methodology) Table 4: Income Distribution by Deciles: 1963, 1968 and 1975 (First Adjustment Methodology) Table 5: Summary Indicators of Income Inequality (Second Adjustment Methodology) Table 6: Income Distribution by Deciles: 1963, 1968 and 1975 (Second Adjustment Methodology) ANNEX III Table 1: CFE Investments: Average Incremental Cost   Page 1 of 2 pages COUNTRY DATA - MEXICO AREA 2 POPULATION DENSITY (Persons) - 1978 1972.5 (thousand km ) 66.9 million (mid-1978) * 34 per km2 Rate of Growth: 3.5 (from 1970 to 1978) 70 per km2 of arable land POPULATION CHARACTERISTICS (1973-1977) HEALTH (1974 preliminary figures) Crude Birth Rate (per 1,000) 42.0 Population per physician 1282 Crude Death Rate (per 1,000) 8.6 Population per hospital bed 667 Infant mortality (per 1,000 live births) 3.0 INCOME DISTRIBUTION (1977) DISTRIBUTION OF LAND OWNERSHIP (1970) % of national income, highest quintile 54.4 % owned by top 10% of owners 37.1% lowest quintile 2.9 % owned by bottom 10% of owners 0.3% ACCESS TO SAFE WATER (between 1973 and 1977) ACCESS TO ELECTRICITY (1970) % of population - urban 70.0 % of dwellings - urban 80.7 - rural 49.0 - rural 28.0 NUTRITION (between 1973 and 1977) EDUCATION (between 1973 and 1977) Calorie intake as % of requirements 117.0 Adult literacy rate % 76.0 Per capita protein intake (grs.per day) 66.9 Primary school enrollment % 112.0 GN2P PER CAPITA IN 1977: US $ 1110 GROSS NATIONAL PRODUCT IN 1977 ANNUAL RATE OF GROWTH (%, constant prices) US $ Mln. % 1965-1970 1970-1975 1975-1977 GNP at Market Prices 72869.6 100.0 7.0 5.7 2.2 Gross Domestic Fixed Investment 14756.5 20.2 9.4 8.0 -6.6 Gross National Savings 12973.9 17.8 6.2 5.6 1.1 Current Account Balance 1550.3 -1.6 - - - Exports of Goods, NFS 7867.2 10.8 3.9 2.9 2.6 Imports of Goods, NFS 7807.6 10.7 6.9 8.8 -15.5 OUTPUT, LABOR FORCE AND PRODUCTIVITY IN 1970 (at 1972 prices) Value Added Labor Force V.A. per Worker US $ Bln. % Million % US $ % Agriculture 4.1 11.2 5.1 39.2 804 28.3 Industry and Mining 12.4 33.6 3.0 23.1 4133 145.6 Services 20.4 55.2 4.9 37.7 4163 146.7 Total 36.9 100.0 13.0 100.0 2838 100.0 GOVERNMENT FINANCE General Government Federal Government Pesos Bln. % of GDP Pesos Bln. % of GDP 1977 1977 1972 1977 1977 1975-1977 Current Receipts 287.51 17.15 12.57 216.94 12.9 12.6 Current Expenditures 248.49 14.83 10.08 202.61 12.1 12.2 Current Surplus 39.02 2.33 2.49 14.33 0.9 0.5 Capital Expenditures 86.28 5.15 6.39 70.74 4.2 5.4 * More recent estimates show that the total population growth rate is estimated at about 2.9 percent per year in 1979.  Page 2 of 2 pages COUNTRY DATA - MEXICO MONEY, CREDIT and PRICES 1973 1974 1975 1976 1977 1978 (Billion pesos) Money and Quasi Money 217.6 260.9 330.9 348.7 435.1 575.0 Bank Credit to Public Sector 166.7 202.7 255.1 325.6 427.0 n.a. Bank Credit to Private Sector 25.8 29.5 35.8 66.6 90.1 n.a. (Percentages or Index Numbers) Money and Quasi Money as % of GDP 35.1 32.1 33.5 28.4 26.0 27.3 General Price Index (1972=100) 112.4 138.4 161.5 198.0 269.1 317.5 Annual percentage changes in: General Price Index 12.4 23.1 16.7 22.6 32.7 18.0 Bank Credit to Public Sector 8.2 21.6 25.9 27.6 31.1 n.a. Bank Credit to Private Sector 32.3 14.3 21.4 86.0 35.3 n.a. BAULANCE OF PAYMENTS MERCHANDISE EXPORTS (AVERAGE 1976-1978) 1976 1977 1978 (Millions US $) US $ Mln. % Exports of Goods, NFS 6770.3 7867.2 9990.9 Coffee 393.1 8.7 Imports of Goods, NFS 8241.9 7807.6 10354.4 Cotton 239.9 5.3 Resource Gap (deficit = -) -1471.6 59.6 -363.5 Other Agricultural Goods 748.7 16.5 Minerals and Metals 281.2 6.2 Interest Payments (net)** -1572.6 -1817.8 -2314.3 Wcrkers' Remittances 193.6 211.3 232.1 Petroleum 1130.9 24.9 Other Factor Payments (net) -346.1 -171.8 -215.5 Manufactured Products 1658.8 36.6 Net transfers 153.0 168.5 198.6 Non-classified 80.2 1.8 Balance on Current Account -3044.3 -1550.3 -2462.5 Total 4532.8 100.0 Direct Foreign Investment 219.3 326.3 268.1 Net MLT Borrowing (Public) Disbursement 5417.9 6232.3 8343.3 EXTERNAL DEBT DECEMBER 31 1977 Amortization -1153.4 -2295.0 -4264.3 ETRA D D Subtotal 4264.5 3937.3 4079.0 US $ Mln. Other Capital (net) 823.9 -1750.3 -1743.6 Public Debt, ndcl. guaranteed 19208 Other Items n.e.i. -2596.6 -458.7 81.6 Non-guaranteed Private Debt n.a. Increase in Reserves 333.1 -504.2 -222.5 Total Outstanding & Disbursed DEBT SERVICE RATIO for 1977 Public Debt, incl. guaranteed 46.9 Non-guaranteed Private Debt - Total Outstanding & Disbursed - RATE OF EXCHANGE IBRD/IDA LENDING (Dec. 1977) (Million US $) Averages - 1977 IBRD IDA lJS $ 1.00 = 22.58 Pesos Outstanding & Disbursed 1374 _ Pesos 1.00 = US $ 0.044289 Undisbursed 754 - Averages - 1978 Outstanding incl. Undisbursed 2128 - US $ 1.00 = Mex$22.7 Pesos 1.00 = US$0.04405 *Does not include .ilver. *'ncludes interest on private and short-term debt. Note: Data contained in the present table are the latest available (as of May 1979).  ________b - ' - -''>q- <RD X m4354&-#C3 V ~ ~ ~ ~ ~ ~ ~ ~~ ~ ~~~~~~~~ % L I a- qup So' -/ 'A~' -"\> _,- '-s ;' J!14-',40i t '>__ ~ ~ ~ ~ ~ ~ 4 " __ ' / '_ __.R ._.__ t vr^ATWA,r ,, 7 ( , r ,~~~ <, r <. Ec-!/ ~ >Gu F O F MEFXICO ( M E X I T I O-------- / * - ' 11~ ~ ~ ~~~~~C -A '4I osi 7rWr O .siu E 1inCJ S ;= - _ 0 rUrDE G . I N [z5 i -_ 1 < P BE 4C5   SUMMARY AND CONCLUSIONS 1. The economic situation of Mexico has evolved quickly since the Bank mission visited the country and collected information for this report. For a country which is successfully emerging from a recent serious economic crisis, whose foreign exchange availability is increasing and whose new economic administration focuses on design and implementation of far-reaching changes in organization of the economy, the quickness of economic change and its depth are hardly surprising. 2. While this report could not incorporate all most recent policy changes or intentions, the basic strategic choices and dilemmas outlined in it remain fully valid. Previous versions of the report have been discussed at length with Mexican public administration on several occasions, and the policy options that the report presents could be accepted as real choices, because they are being actually considered. I. The Recent Past - A Disorder in Economic Variables 3. On taking office in December 1976 the present administration inherited a heavy burden of economic disequilibria. While Mexico had been developing quickly and with a remarkable price and balance-of-payments stability through most of the three decades of 1940-70, the country began to experience in the early 70's increasing public sector and balance-of-payments deficits, as well as price inflation, depreciation of the peso, capital flight and marked slow-down in real rate of growth of the GDP. By 1976 inflation reached 21 percent, the current deficit on the balance-of-payments increased to 4.3 percent of GDP, the public sector deficit was equal to 7.4 percent of GDP and GDP was only 2.1 percent over the preceding year. These magnitudes were alarming and most unusual for the Mexican economy. 4. One of the major causes of Mexico's recent cycle of inflation and ensuing economic stagnation was that the Government was spending in the early- to-mid-1970's more than it was willing or able to raise in revenues. The total current and investmerLt expenditures of the public sector grew from an average of 12.6 percent of the GDP in the mid-to-late 60's to 19.6 percent in 1972 and to 28 percent of the GDP in 1975 and in 1976. At the same time the Federal Government gross tax revenues, although growing rapidly, were not keeping pace with these expenditures. They increased from 10 percent of the GDP in 1970 to an estimated 13 percent in 1976. Total current revenues of the public sector were higher than the tax revenues, amounting to 11.3 percent of GDP in 1965-70, 15.2 percent in 1972 and 20.3 percent in 1976. Beside a still inadequate tax effort, other underlying weakness of Mexico's public finance were--and still are--the subsidies, implicit in prices charged by public sector enterprises and agencies for goods and services they produce. These subsidies for few main items only are estimated to have been equivalent to about 6 percent of the GDP and were roughly equal to the entire public sector deficit in 1977. About 60 percent of the subsidies are accounted for by sales of petroleum products at less than their opportunity cost.  - ii - 5. Strong corrective actions of the Central Bank and the Ministry of Finance, taken within the framework of a broad national consensus, helped to attenuate these disequilibria. On September 1, 1976 the authorities abandoned the fixed exchange rate of 12.5 pesos per dollar that had remained unchanged since 1954 and let the peso float. The exchange rate market went briefly into a disarray with the exchange rate reaching 28 pesos, but descending after- wards to 22.6 pesos per dollar. Reacting to massive capital flights and financial disintermediation the Central Bank had to support the banking system with a special line of credit. For 1977-79 an Extended Fund Facility accom- panied by a comprehensive three-year stabilization program were negotiated with the IMF. 6. Strict control of budgetary expenditures has been introduced and the average increase in public sector expenditures during 1977-79 has been only about 4 percent per annum in real terms. Moreover, while the annual overruns of actual over budgeted expenditures during the preceding period (1972-1976) had been, on the average, 18.5 percent, in the past two years (1977 and 1978) this overrun had been reduced on the average to only 5.4 percent. The Government not only managed to keep public sector expenditures under control but succeeded in increasing somewhat its tax ratio, to slightly over 14 percent of GDP in 1978. 7. The measures instituted by the new administration have yielded positive results. Inflation was brought down to 18 percent in 1978 from more than 30 percent in 1977, the deficit on the current account of the balance-of- payments declined to US$2.5 billion in 1978, or 2.7 percent of the GDP, and the GDP growth was 6.6 percent over the preceding year. II. The Present - Inheritance of Social Parameters 8. The initial equilibria, recently disturbed and now on the way to return to a balanced state, involved mainly the set of short-term economic variables. While they were balanced among themselves, the economy kept devel- oping quickly, but this hardly affected the basic parameters of the Mexican social situation. In reality the economic variables were brought into a dis- equilibrium during the early-to-middle 70's, mainly because a steep increase in public spending was decided in order to improve the social parameters. Paradoxically, this seems to have affected the social parameters negatively, because related policies were ill-designed and led to perverse, even though short run, outcomes. Public expenditures led to capital-intensive technologies, inflation turned against earners of low incomes, and the recession affected employment. 9. Now, as the process of restoring equilibria among short-term variables is well advanced, an increasing amount of thought is devoted to the crucial questions--how to attack effectively the basic parameters of the Mexican social situation--its poverty, its population growth, its regional imbalances. Most of those, but in particular the living standards of the poor, were unavoidably affected by the decline in the rate of the growth of the economy from an average of 6.2 percent during 1968-74 to 4.1 percent (1975), 2.1 percent (1976), and 2.8 percent (1977). The poor, the unemployed and the underemployed were also affected by the combined effect of increasingly cheaper imports, negative real interest rates, and high additional charges  - iii - superimposed on the wage bill--which conjointly tilted the choice of production techniques toward increased capital intensity and have reduced employment oppor- tunities. Finally, the ensuing inflation took its toll on small entrepreneurs, on earners of fixed incomes and on poor farmers and peasants. 10. Population is clearly one of the main parameters of the Mexican scene. By 1972 Mexico's population was growing at an annual rate of about 3.5 percent--faster than that of any large country in the world. Births outstripped deaths by some 1.9 million in that year, an amount which topped the combined 1972 natural increase of Canada and the United States by 370 thousand people. Until 1972 Mexico had believed that, in demography, more is better. However, that year also marked the beginning of a radical change in the population policy. A Government-sponsored National Family Planning Program was formalized--through a new General Population Law, followed by a creation of the National Population Council. The World Fertility Survey found that by late 1976 nearly 42 percent of Mexico's women of child-bearing age and exposed to the risk of pregnancy were currently using contraception devices. In cities of 20,000 or more inhabitants, as much as 61 percent of such women were practicing contraception. By mid-1978 unofficial estimates put the population growth rate at 2.9 percent--within sight of the 2.5 percent growth rate set by the new Government administration for 1982, the end of its term of office. 11. The goals of the family planning program are ambitious but neither impossible nor improbable. It is aimed at bringing the estimated average family size of slightly over 6 children in 1973-75 down to the slightly more than two-child average, which will be required if the Government current goal of a 1 percent population growth rate by the year 2000 is to be achieved. 12. The prime targets of the population control program for the future are rural areas. Positive signs are shown in the first half of 1978, as over one quarter of the new family planning acceptors recruited by the three main public health agencies during this period were rural women. 13. However, even if the long-term goal of the population decline is achieved, medium-term policy will still have to cope with a flow of new entrants into the labor force, reflecting the past high population growth and the related fact that 46 percent of the total population of Mexico is, at present, 14 years of age or less. And since only 3.4 percent of total popula- tion are aged 65 and over, the dependency rate reaches 100 percent--meaning that half of the population is dependent on the other, theoretically working half. This latter qualification is important, because while 34 million people between 15 and 64 years are in the theoretical working age group, those economically active and paid--the work force--probably number no more than 17-18 million. In turn, in this latter number about 50 percent are underpro- ductive workers--those earning less than the minimum wage because their productivity does not allow them to earn wages sufficient to break the poverty barrier. As therefore the new entrants into the labor market will continue to amount annually to about 700-800 thousand--or, in net terms, about 550 thousand, after adjustments for emigration change in participation and other factors-- Government policy oriented at maximization of employment opportunities in the medium-term will have to retain priority.  - iv - 14. Poverty, widely spread among the Mexican population, has its roots in the duality of the economy. In turn, the sources of this duality are mainly in agriculture. At present, about 40 percent of total population of Mexico still resides in small rural localities, below 2,500 inhabitants each. As much as 34 percent of total labor force is still occupied in primary (preponderantly agricultural) activities. This notwithstanding, the total share of agriculture (including livestock, forestry and fishing) amounts to only about 10 percent of Mexico's GDP. Therefore, the average value added contributed by those employed in agriculture--in other terms, average productivity per worker--is less than one-fifth of the average of sectors other than agriculture, which constrains the level of income those working in agriculture could enjoy to a very low level. This is additionally compounded by internal productivity differences in agriculture, which are considerable, with over one-half of the total number of farms having land whose yield per hectare are equal to only one-third of output and one-seventh of per capita income of that obtained on modern Mexican farms. Average income on subsistence farms is estimated to amount to some 80 dollars per capita (620 dollars per family in 1978 prices) yearly, less than absolute poverty level. 15. Sharp differences are also observed in the industrial sector, where workers' earnings in the largest enterprises exceeded by five-fold the earnings in small establishments. In the latter average wages in the early 70's amounted to only 600 dollars annually. Productivity differences are even larger, with large enterprises producing 8 times as much value added per worker as small ones. Also, while productivity differences clearly militated against any perceptible improvements in incomes of the less privileged citizens of Mexico, slowdown of agricultural production made it even more difficult. Agricultural growth declined from 6 percent per year during 1945-1955 and 4.2 percent per year during 1955-1965 to only 2.1 percent per year during 1965-1975. 16. Comparison of household budget survey results suggests that, in general, income distribution pattern in Mexico might have slightly deteriorated in the past decade. The lowest 20 percent of the population have seen their share of income decline from 3.4 percent of household income in 1968 to 2.9 percent in 1977. As, during this period, average per capita income increased, in real terms, by 17.9 percent over a decade, the real per capita income of the poorest Mexicans remained practically unchanged. However, the lowest 40 percent maintained its share at 10.5 percent of total income, which implies that there has been an improvement, both in relative and absolute terms, in the well-being of those receiving incomes in the range of 2-4 deciles above the bottom range of the income spectrum. 17. Regional imbalance is another important long term problem of the Mexican social and economic situation. The heavy concentration of people, production and public services in the Mexico City area--which has currently a population of about 12 million and is growing at over 5 percent each year-- and the scarcity of productive employment and services for rural Mexicans are challenges to urban management, economic development and social justice. They suggest an imperative diversion of some of Mexico City's future growth to other regions as well as the gradual concentration of scattered rural population into small towns so that basic public services could be provided more economically and productive activities have a better chance of developing.  18. A number of recently taken economic policy decisions are aimed at improving the situation of the poor. Some of those principal decisions are: (a) A steep increase in assistance to rainfed agriculture, the single largest area of Mexican poverty, through agricultural extension and credit. (b) Attempt to slow down the growth of Mexico City and influence the population to move from small rural centers into larger conurban centers, where basic needs could be better satisfied; for this strong regional incentives have been designed and begin to be implemented. Moreover, a better infrastructure and other conditions for development of regional growth poles, mainly in the underutilized South-East region of the country, is now being planned. An ongoing exercise in regional planning should also help to tilt the balance in this direction. (c) Creation of a vast vocational training system, which should provide basic training to some 5 million Mexican workers. (d) A better satisfaction of housing needs of the population through construction of 3 million dwellings by 1982 compared to 1.5 million which would have been constructed if the historical trend were continued. (e) Income tax alleviation for the poor and enforcement of tax collection--using the existing regulations--from the richer population. (f) Modernization of the vast but obsolete, inefficient and costly marketing and distribution system, which accounts for between one-fifth to one-third of the GDP and is at least partly responsible for excessively high prices to consumers of basic goods and for low prices to small scale producers. (g) Special credits given to enterprises engaged in priority industrial activities and located in underdeveloped regions of the country, in proportion to their contribution to increase in employment. (h) Increase in the share of budgetary expenditures on health, welfare and education, which in the new 1979 budget attains 22 percent share of total expenditures, compared to only 8.9 percent in 1974. III. A Social and Economic Strategy for the Future 19. With the discovery of extensive petroleum resources, a foreign exchange constraint on future growth will be greatly relaxed, although it  - vi - will not be completely eliminated. 1/ Foreign exchange is, however, only one of the several constraints on growth and could prove to be of only limited usefulness in the difficult task of elimination of structural rigidities of the economy. Moreover, the existence of such vast oil reserves gives rise to certain imperatives and obligations the country will have to face. The most important of these imperatives is the improvement of the social situation, partly because oil provides a windfall income--no single population group is entitled to a claim to have contributed to, and therefore everybody is entitled to benefit from--partly because it raises the expectations of the poor, which if not met, may lead to grave tensions and social disequilibria. 20. This imperative could be met only through implementation of a cohe- rent program, which would consist of several interrelated parts: (a) elaboration of a comprehensive economic and social strategy, targeted on the achievement of growth, based on efficient production processes and leading to a rapid improvement of the living standards of the poor; (b) incorporation in this strategy of social and economic policies, which were initiated during the past year or two, to assure that the specific directions and targets of these policies are consistent with it; (c) conversion of the public investment program for 1979-82, of the annual current expenditure budgets and of the sectoral and regional plans which are now being elaborated, into vehicles of this strategy. 1/ According to the statement by PEMEX management on December 31, 1978, proven oil and gas reserves amount to the equivalent of 40.2 billion barrels of crude, to which should be added 44.6 billion barrels of probable reserves. A practice is to convert gas reserves into crude oil equivalent at 5,000 cubic feet equal one barrel. Approximately 33 percent of total reserves of Mexico is gas. The true level of Mexican oil and gas reserves has been a matter of considerable speculation, since as recently as the end of 1973 they were stated to be only 5.4 billion barrels. The rapid increase in both reserves and productive capacity has resulted from the discovery of new oil fields in the Southern States of Chiapas and Tabasco, which have production characteristics comparable to those of many Middle-Eastern fields. This has enabled a very rapid build-up of crude oil production and new reserves could be ascertained through drilling of a rather limited number of wells. While the wilder estimate of Mexican oil and gas reserves must be treated with considerable caution, those given above are regarded as reasonable. Moreover, prospects for increasing these reserves in the next few years seem to be good.  - vii - Development Strategy and the Investment Program 21. Some economic policies, when designed and applied without a recog- nition of trade-offs among them, may produce effects contrary to those origin- ally intended. For instance, the gradual elimination of the dual structure of the Mexican economy is the principal measure to raise the incomes of the poor, through increase of their contribution to the economy, which is now marginal because the productivity of the primary units where they are employed, in agriculture, industry and services, is also marginal. However, the modern- ization of these units, if it is achieved without considering trade-offs between the use of capital and labor, may lead, in circumstances of only a moderate growth, to a reduction of overall employment. To the same extent, maintenance or increase in protection against imports may temporarily lead to increases in employment, but the productivity of the new labor force may be minimal or negative, if measured in terms of effective value added calculated in international prices. Both examples are derived from Mexico's experience of the last decade. 22. Likewise, the use of the investment program as a strategy vehicle opens a number of options as to how to resolve the financing of this program in the interest of the poor population. Decisions as to what sources of financing of such programs should be tapped, carry often far reaching impli- cations for the welfare of the poor. Thus, an increase in reliance on oil exports and on public savings derived from oil-generated revenue may entail a weakening in inducement for non-traditional and labor-intensive exports. It may also entail a weakening of the tax effort, the core of the re-distributive mechanism governments use to reduce income inequalities, while such tax effort, especially if concentrated on those with a large ability to pay taxes or a large propensity for consumption expenditures, could become instead one of the most powerful instrLments available to the Mexican authorities to combat inflation. 23. A contribution of foreign savings to financing of such program would be necessary, considering t:hat public savings, even if a maximum feasible effort is assumed, cannot be increased enough to meet the financing require- ments of the investment program of the magnitude now considered as necessary. The one presented in this report assumes an 85 percent increase of public investment in real terms during 1977-82, compared to the 1971-76 period. A substantial increase in expenditures on welfare, social infrastructure, agriculture and rural development are built into this investment program, and these expenditures are directly geared to the improvement of the living stan- dards of the poor population. This, however, still leaves open two problems: that of institutions which will manage social programs and that of the project content of these budgeted amounts. Social projects are normally exceedingly difficult to design, their gestation period is, as a rule, quite long and the knowledge of experience of other countries could be of great use. In this particular area the continuation of foreign borrowing by Mexico, with non- commercial lenders assisting it in building up the necessary institutions and in designing projects, appears not only necessary but also desirable.  - viii - 24. Reduction of the investment program, to adjust it to a weaker resource mobilization effort and to a lower level of financing, could be con- sidered as an alternative. However, a considerable part of this program consists of hard-core investment, which can hardly be postponed. Belongs here the PEMEX investment (25 percent of the total program) which is indispensable for the generation of the expected volume of oil revenue and investment in transport and communication infrastructure, where some marginal bottlenecks are already appearing and where the demand for services imposed by the rapid growth of the economy will have to be met. Consequently, any decision to reduce the investment program will be felt principally, in expenditures either directly related to the standard of living of the poor (expenditures on welfare and social infrastructure) or destined for areas where poverty is concentrated (expenditures on agriculture and rural development). Reduction of budgetary expenditures in these sectors could entail a widespread disappointment of the poorer strata of Mexican population. Growth and Inflation 25. The Government has broken down the six-year period of administration into three two-year periods, of which the first, already completed, was devoted to the stabilization, the second extending over the years 1979-80 would be a consolidation period and the remaining, 1981-82 is planned to become a period of accelerated growth. Such accelerated growth might prove to be difficult to contain already during the consolidation period. Not only the private sector could decide to step up its investment rapidly and therefore stimulate growth, but, even if the public investment efforts remain cautious, the demand for goods and services supplied by the public sector is likely to grow rapidly. This may ultimately lead to growth bottlenecks, but would, in the immediate future also lead to increases, insofar as idle capacity exists, of capital productivity. Given the absence of restrictions on monetary capital movements into and from Mexico, credit restrictions for demand management purposes, in particular those applied to credit for the private sector investment, may not prove to be fully effective. Therefore, rapid growth of both demand and output seems almost certain, but because demand may grow faster than output-- especially in a few key sectors--the inflationary pressures could be substan- tial. 26. In many respects, Mexico can largely afford this growth. Firstly, such growth would be only partly stimulated by perspectives of increased income from oil, with capital expenditures only partly financed from oil. Mexico may never become a truly classical oil country. Neither does it want to become one. The percentage contribution of the oil sector to the GDP was, in 1978, only 4.7 percent. It is expected to increase to about 10 percent by 1982, a much lesser share than the 30-50 percent range share held by oil in the GDPs of other semi-industrialized oil exporting countries. Had the imminent growth rate of the economy been rooted predominantly in the oil income, such dependence might have been, and correctly so, considered as dangerous, for the same reason that, in the past, most economic booms, stim- ulated by and financed from a windfall income contributed by one single commodity, proved to be short-lived and fraught with strongly negative side effects.  - ix - 27. Secondly, such re'Latively rapid growth should prove to be possible in the context of Mexico's capacity--actual or liable to be quickly expanded-- to produce goods and services. One of the principal differences between Mexico and other oil exporting countries is Mexico's excellent infrastructure. While transport bottlenecks have been already signalled, the problem may still be not so much that of an overall inadequacy of the transport network but that of a trade-off between various modes of transportation, which through the use of the price mechanism, couLd be tilted toward the relatively underutilized modes as well as that of the transport management having to arrange for scheduling the sudden surge of incoming imports. Furthermore, in railways, the problem is that of a shortage of locomotives and certain types of rolling stock which could be quickly provided, although in the near future there is also a need for realignment and doubling of the existing tracks; in ports the problem is that of improvement in the scheduling of grain ships and of tilting a choice toward using ocean-going barges rather than ships for grain delivery. All these issues are now being studied. It is clear, however, that the medium-term solution is in an expansion of the transport infrastructure to meet the claims of the early eighties. 28. Thirdly, with the increase in foreign exchange availability, Mexico would be able to meet the increased demand for goods and, to a lesser extent, services also partly from imports. The questions that are usually raised here are which goods are really tradeable, how would the increase in imports affect the willingness of the domestic private sector to invest in production of competitive goods and whether the foreign competition would not attempt to undersell the domestic producers. The answer to the first question is that most goods and some services could become at least temporarily tradeable, and even more so given Mexico's geographic situation which makes it possible, for example, to import cement with lower transport costs than other developing countries normally have to pay, if this is to serve the quick removal of temporary growth bottleneck:s. The answer to the second and also the third question is that the privat:e sector investment plans would be largely guided by considerations of the expected growth and of the expected conditions of the domestic market. This latter notion includes mainly the market signals, and notably the relation of domestic production costs to prices of the foreign competition. This brings t:o the forefront the role of the exchange rate which, if the readiness of the Government to manage it in response to the domestic market consideration becomes evident, could become one of the most powerful market signals, helping to remove the investors' and the producers' apprehensions. Inflation and the Exchange Rate 29. Continuation of the price inflation could become a factor accom- panying the future rapid g-rowth of the Mexican economy. In the recent past, Mexican inflation was mainly a result of a combination of wage adjustments which were in some years more rapid than it was justified by productivity increases, and more importantly, of a strong recourse to deficit-financing of public investment. These factors exerted effect on prices in a general economic environment in which such trends as the increased protection against imports, a decrease in relative capacity to pay for these imports and until the latter part of 1976, the existence of an unchanged nominal exchange rate,  x were tightly intertwined. While there seem to exist little reasons now for these tendencies to re-occur in the near future, other inflationary factors may appear instead. Thus, scarcities of skilled manpower could drive the wages up, inflow of foreign exchange through the capital account--in form of private credits and speculative capital--may not become fully converted into imports of goods and services, which would then swell the reserve accounts, liquidity preference and consumption propensity may shift in directions favoring inflation and, last but not the least, supply bottlenecks may occur. This is also why the decline in price inflation cannot be predicted for Mexico with any acceptable degree of likelihood. To the extent to which most of these inflationary factors could be dampened, some of them thanks to Government foresight and its preventive steps, others through quick application of appro- priate policies, barriers to constrain inflation may prove effective. They may prove even more effective if some of the inflation-constraining barriers are designed for conditions of a continuing rapid growth rather than taken from the arsenal of stabilization measures applicable to other economic circumstances than those expected in Mexico during the coming years. 30. Inflation should not be considered as an unavoidable corollary to a rapid growth of the economy, in Mexico or elsewhere. Thus, in the economic history of countries whose structure resembled that of Mexico today, such as for instance Spain, and very recently Brazil, one finds periods when a rapid increase in the rate of growth of their GDPs was accompanied by a deceleration in their inflation rates. Important, in the case of these countries in the past, and for Mexico in the future, is to be able, while the investment process accelerates, to keep the production capacities fully utilized and the project cycles reasonably short, both of which should prove possible in Mexico, whose managers are capable to run publicly-owned productive capacity efficiently. It is equally important to maintain unrestrained inflow of necessary imports, which would serve a triple purpose--check private entrepre- neurs' tendency to profit from the economic boom through speculative price increases, eliminate supply bottlenecks which would almost unavoidably be created, and convert the foreign exchange into imported goods and services instead of adding it to the foreign exchange reserves of the banking system, over and above the level considered as required for operation of the economy. 31. The real issue is that of the interrelation between the exchange rate and inflation. A number of arguments is now formulated in Mexico against an exchange rate adjustment. Firstly, the depreciation of the dollar in international money markets--somewhat reversed since late 1978--pushed upward the peso rate with respect to other currencies; this helped to preserve the real level of the peso exchange rate at a largely competitive level vis-a-vis the countries other than the USA; secondly, upward exchange rate adjustment may add an additional layer to domestic price inflation especially if wage demands would be triggered by it, and if the psychological attitudes of private entrepreneurs would induce them to use "umbrella prices" which would arm them against the future increases of prices of imported inputs and increases in the labor cost; thirdly, conversion of import licenses into relatively high tariffs--sufficiently high to equalize most import prices with those of domes- tically produced goods except for those where inefficiency of production is blatant--and adequate export incentives would set the level of the effective exchange rate sufficiently high to discourage excessive imports and to encourage further efficient import substitution and increase in non-oil exports.  - xi - 32. Equally strong arguments could be mustered in support of exchange rate adjustments. One is that the decisions to invest in production of goods at least partly destined for exports, are guided by considerations of the future profitability of such exports, compared to the future profitability of production for the domestic market, and the exchange rate is the only policy tool capable to equalize these profitabilities. If Mexican prices increase more rapidly than international prices and exchange rate remains constant in nominal terms, not only exports of non-traditional, often highly labor-intensive, goods will suffer, but investment decisions will also be affected. Tariffs and export incentives may adequately meet current consid- erations, but are not flexible enough to be gradually adjusted upwards over time in response to the changes in economic situation. Second, Mexico started experiencing an inflow of private foreign capital which could become--if it is not already--inflationary. It partly consists of credits to the private sector, and in other part, represents short-term financial investment. The former is stimulated by the interest rate differential, with Mexican rates standing considerably above international rates and with little apparent exchange risk at present. The latter is stimulated by high nominal yields on shares and securities traded in Mexico, which may be attributed both to the inflationary (paper) profits and to anticipations of the future boom. The return on such investment is therefore high and could be transferred back abroad at a nominal--i.e., untouched by the inflation-exchange rate. These trends, whose impact is both distortional and destabilizing, could be fuelled even more, in the future, if the differentials between price trends in Mexico and abroad, persist. In the situation of free international capital movements, which always existed in Mexico and which contributed to financing of its growth in the past, when real interest rates and differential inflation rates were roughly similar, the only way to equalize domestic and foreign interest rates and yields on financial investment would be through the exchange rate adjustments. 33. Consequently, the exchange rate should be used as an active and important tool of the economic policy, rather than merely as an economic indicator or as a variable of an equation. As a tool it should be able to play a double role helping to achieve social change, through employment generation--by stimulating non-traditional exports, and placing competition against imports within a framework where efficient domestic production would be induced, but also acting as a stabilizer of the financial variables of the economy, regulating capitaL outflows and inflows, reducing the present incen- tive to borrow abroad by equalizing the effective (adjusted to the revealed exchange risk) interest rates and ultimately helping to attenuate inflationary trends. It may therefore become necessary to adjust the nominal rate of exchange, so as to keep the real rate of exchange roughly at an unchanged level. To decide this, the differential between Mexico's rate of inflation and the international inflation rate should be taken into account. Productive Sectors 34. Choice of the industrialization path is another important issue for the future. Two main problems will have to be solved--the first is that of a choice of production processes which should help to alleviate the employment  - xii - problem but should also make Mexican industry internationally competitive. The second is the choice of a proper trade-off between imports and domestic production, as sources of supply of the domestic market for industrial goods. The new Industrial Plan for 1979-82 assumes that the manufacturing industry should absorb 30 percent of the new entrants to the labor force. How feasible this target would prove depends largely on the capacity of the public sector and the willingness of the private sector to forge a new network of inter- industry relationships and flows. The problem is that the major investment projects Mexico intends to undertake within the next several years, especially those in energetics (oil and electrical energy), and in engineering and capital goods industry (including steel) will, by definition, have to be highly capital-intensive because they would be based on most modern techno- logies. This does not exclude a possibility, indeed a necessity, to promote a parallel effort in developing a wider use of labor-intensive technologies, although in different enterprises and for a different purpose. More specific- ally, the development of large, capital-intensive enterprises in industry could be closely tied to the creation of a network of subcontractors for these industries, consisting mainly of small-to-medium enterprises where labor- intensive processes are generally applied. Appropriate incentives to stimulate cooperation between these two types of enterprises should be used, without affecting the overall efficiency and the level of production costs in industry. Some of such incentives, which are already being advocated in Mexico, are the reform of fiscal incentives to reverse the past bias which favored machinery use against labor use, the reform of regional development incentives, focusing on industrial development of a few regions with good growth potential, the liberalization of import licensing and rationalization of import tariffs, the encouragement of labor intensive exports, the stimulation of the development of small and medium scale industry, a reform of procurement procedures of public enterprises in the direction of a support to domestic suppliers and increases in the number of jobs. 35. A marginal but nonetheless important trade-off exists between the relative shares of imports and of domestic production in the supply of the domestic market. As an outcome of several past decades of import substitution, Mexico's economy has become largely self-sufficient. Except for some important areas, such as capital goods, and a few basic raw materials, including to some, however small, degree also food, Mexico satisfies almost entirely its demand for goods and services from the domestic sources. Its imports repre- sented hardly 10 percent of its GDP in 1977. To some degree this self- sufficiency is an outcome of import licensing which only until recently applied to some 80 percent of goods. A de-licensing process has begun two years ago, but 45 percent of Mexico's imports are still licensed. This percentage is to be reduced to 35 percent in 1979. A rapid increase in imports should not be expected to affect the structure of domestic production negatively--in the sense of replacing large fraction of it and indiscrimi- nately--when licenses are fully replaced by tariffs, and if the exchange rate-cum-custom duties are maintained at a level which would lead to replace- ment of only the clearly inefficient forms of domestic production. As this latter production is contributing to a high level of domestic prices, it is affecting consumers' welfare negatively, and its disappearance would therefore  - xiii - become beneficial, provided that in the rapid progress of Mexican industrial- ization other, more efficient enterprises would take over, and that an equiva- lent compensatory number of Jobs is generated in the process. Such conversion may be sometimes difficult to achieve in a slowly growing economy, but this will not be the case of Mexico in the near future, when the boom conditions should help to smoothen the "dynamic adjustments" even of the type which leads to deep-reaching, structural changes. 36. Mexico's agricultural development strategy is changing in the right directions. These changes were required by increasing awareness of two different problems. First, the near-stagnation of output growth in the last decade or so shows that major production increases can no longer be expected from a program consisting mostly of large-scale irrigation schemes, targeted on commercial farming in the previously uncultivated areas. The costs of additional water prove to be too high in most areas, and markets for high-value crops were not expanding rapidly enough. Second, the old policy, while successful for some time in increasing production, was of limited use in increasing incomes of the mass of Mexico's rural poor. Allocating more resources to Mexico's new type of rural development and to rainfed agricul- tural programs is the main way to increase output and productivity of the rural poor and of the entire agricultural sector. 37. The present administration has undertaken administrative reform in agriculture which should allow broader and more effective initiatives in the sector. Emphasis is laid on productivity gains and yield increases in both irrigated and rainfed areas. The previous goal of self-sufficiency in agri- cultural goods is being rephrased in terms of the sector's overall capacity to pay for its own imports. Irrigation and drainage works are fostered through programs for rehabilitation of existing irrigation districts to increase effi- ciency of water use. The new emphasis on rainfed agriculture, relatively neglected in the past, should lead to an increase in the productive potential of vast, currently under-exploited areas and to a reversal of past trends towards large income disparities between the modern and traditional agricul- tural subsectors. Policy Scenarios 38. The quantitative dimensions of many of the basic choices Mexico is now facing are often outside the grasp of tools normally used for forecasting or for policy-sensitivity tests, because structural changes of the magnitude expected in Mexico during the next few years could not be predicted within an acceptable probability range. Moreover, the international framework changes quickly. Thus, the unit prices forecasted in this report and used for cal- culation of the value of Mexican exports of petroleum would have been set at a somewhat higher level had this calculation been prepared after the recent events in Iran and the ensuing cascading increases of oil prices. Furthermore, Government oil production goals also seem to be changing. Finally, propensity to consume of the population which determines private savings, is difficult to predict for conditions, like those of Mexico, where the per capita rate of growth of consumption may rise from a negative level experienced during 1976 and 1977 to an annual level of over 6 percent per capita in 1980 and 1981.  - xiv - 39. These imponderables notwithstanding, the authors of this report considered it useful to construct two alternative policy scenarios. The common assumption underlying both scenarios is that a more rapid growth than the Mexican Administration presently expects would materialize and that the policy framework will have therefore to be geared to these high growth rates so as to cushion whatever negative side-effects may appear and also to channel its positive effects where they are most necessary. Consequently, it is assumed in both scenarios presented in Annex I to this report (Policy Options and Development Outlook) that the growth of the GDP would accelerate from 6.6 percent in 1978 to 8.8 percent in 1979 and to 9.5 percent in both 1980 and 1981, and that it might decline slightly thereafter. Besides growth rates, there are three identical aggregates for both scenarios--the investment programs, the market supplies (imports plus non-oil production minus exports) and the foreign savings. However, the combinations of sub-aggregates (imports, oil-exports, domestic productions) are different in each scenario and, accord- ingly, the financing structures and the related policies are also different. The policies--popular or unpopular depending on who are the beneficiaries-- consist, in the first scenario, of high oil exports, openness to imports, and their effect on public savings, which because of the abundant public revenue could attain a required level without a need to maximize strenously the non-oil revenue. In the second scenario, the variables are a more modest growth in oil exports, and maintenance of a strong anti-import protective structure. Their effect on public savings is calculated and it shows that, to match the expenditures, stronger efforts toward mobilization of the public non-oil revenue, to the detriment of private savings are necessary. 40. The first, high oil export scenario, shows the trade ratio (imports plus exports to GDY) increasing quickly, from 20.5 percent in 1978 to 25.1 percent in 1982. Public savings are a residual here, and there is no need either for the increase in real prices (with nominal prices continuing, however, to increase proportionately to inflation) of public sector goods and services or for a major tax reform. Public savings rely mainly on a steep increase in PEMEX savings and on improvement in tax administration. In this situation private savings are less constrained and become a truly dynamic element of the investment financing structure and are growing at about 15 percent real annual rate. In the second scenario oil exports increase by about US$4 billion less than in the preceding one. Consequently, public savings based on oil taxes would also have to decline. Public sector will have to compensate for this shortfall, by increasing its non-oil revenue and savings. The main measure for an increased mobilization of public sector resources is, in the second scenario, the upward adjustment of real prices of public sector goods and services as well as the increase in tax rates. This, in turn, would imply a preemption of the private sector revenue and savings. Another round of consequences are that imports are lower here than in the first scenario--and, with lower oil exports, the current account deficit and the foreign savings remain the same as in the first scenario. A slight increase in the domestic non-oil production will have to occur to compensate for lower imports so as to equalize the domestic market supplies for both scenarios.  - xv Crucial Choices 41. The volume of oil production and oil exports will become therefore a key variable of both scenarios--and of the Mexican economic scene in the near future. This volume for the next few years is largely constrained by the ongoing investment program of the PEMEX. The total Mexican exports of goods and non-factor services are calculated to have to reach between 20-24 billion dollars by 1982. Of this some 35-45 percent, depending on the scenario, would have to be contributed by crude oil, refined products and petrochemicals, and this may prove to be difficult to attain. Even if these export levels are reached, net public sector borrowing will still have to amount to US$4.5 billion and gross disbursements to US$11.6 billion, with debt amortization payments amounting to about: US$7 billion in 1982. This substantial amount of foreign borrowing would still entail, by 1982, a relatively high debt-service ratio. Such ratio would have been otherwise considered as alarming in countries exporting predominantly one product, expecially a product with a high price or income elasticity of international demand, which would therefore be exposed to the world market vagaries. It would have been also alarming for countries where such massive inflow of foreign savings would be used liberally for budgetary support. However, neither of those possibilities is expected to happen in Mexico. Moreover, the existing oil resources could be considered as more than an adequate collateral for foreign borrowing. 42. Another crucial choice is that of finding the most appropriate trade-offs, at least on the margin, between and within the policies' mix, which would satisfy a need for: (i) alleviation of poverty, and maximization of the employment opportunities but also a modernization of Mexico's productive base; (ii) minimization of inflationary pressures, but also achievement of a rapid growth of Mexico's economy; (iii) the optimum use of petroleum resources but also minimization of negative effects normally related to the existence of a petroleum sector, such as (a) appearance of pressure groups staking conflict- ing claims on petroleum income, (b) a general tendency to spend this income rapidly, often in a way contributing little to meeting the country's social and economic needs, and thus deepening the economic dualism instead of removing it, (c) weakening of other important and often fundamental activities such as exports of non-traditional goods or the use of the tax system not only for mobilization of public savings but also with a redistributive purpose; (d) letting the factor prices (labor and capital) and product prices diverge from economically and socially meaningful proportions. Such effects, whose likelihood could be removed only by a timely implementation of a coherent strategy of social and economic development, not only could make the attainment of social and economic objectives difficult, but would also make it difficult to prepare a timely transition from oil to other sources of income which, in the future, should increasingly generate foreign exchange and public revenue.  - xvi - 43. There is no escape from the dilemma. The Mexican Government will have to implement policies which are either unpopular to some "producers," who may feel vulnerable to the opening up of the economy to competitive imports, or unpopular to some "consumers," whose purchasing power would be affected by increases of the real prices of public goods and services. A best solution would probably be an application of a mix of policies, derived from each of the presented scenarios, but in smaller doses, so that the combination is only mildly unpalatable to all social groups and that their final results, while benefitting all, does it in unequal proportion, with the poor benefitting relatively more. If the policy variables are incorporated in an overall strategy, whose aims--such as those specified above--are clear, even unpopular policy changes will be quickly vindicated. 44. While Mexico is facing difficult strategy choices for the next several years, it is fortunate to have acquired a sufficient flexibility of maneuver. Its administration sees now clearly what are the necessary develop- ment goals, both social and economic. With a timely and comprehensive strategy it should be able to attain these goals without any excessive waste of resources and, hopefully, within a foreseeable time period. An important element in the design of this strategy would be to adjust it, downward if necessary, to the existing "delivery mechanism," consisting of a multitude of imponderable although fundamental factors: capacity to design and implement projects, receptivity and flexibility of the regional public administration, quality of private entrepreneurship, etc. Finally, the way towards some of the goals, such as the social goals, may be long. However, it should be less thorny than that of many other countries, which are less blessed by the many gifts of nature and history than Mexico has been.  CHAPTER I CRISIS AND STABILIZATION PERIOD 1. For most of the three decades preceding the mid-seventies, Mexico was outstandingly successfuL in achieving rapid economic growth while maintain- ing stability in prices and the balance of payments. From 1940 to 1970, average GDP growth exceeded 6 percent per year in real terms, inflation averaged less than 5 percent per year from the mid-1960s to 1972, and the dollar value of the peso, fixed in 1954, was maintained until the September 1976 devaluation. This strategy produced rapid growth, but led to a sharpen- ing of contrasts within the Mexican economy. While land redistribution under the reform of 1915 was continued, many peasants remained landless, and most of the peasants who received land could not improve their economic status in the absence of basic infrastructure, credit and technical assistance. Rapid population growth made sociial equity even more difficult to achieve. Popula- tion growth in Mexico accelerated steadily, primarily as a result of reduced death rates and reached 3.5 percent per year by 1970. Despite economic growth, the rapid growth of the labor force has made adequate absorption of Mexico's workers in productive employment difficult. By 1976, some 40 percent of the labor force was either relatively unproductive and poorly paid (earning less than the minimum salary) or openly unemployed. 2. Starting from the early 1970s, Mexico experienced increasing public sector deficits, inflation, large balance of payments deficits, capital flight and a marked slowdown in the real rate of growth of GDP, which dropped to 2 percent in 1976 -- the lowest annual growth rate experienced by Mexico since the mid-thirties. On September 1, 1976 the authorities abandoned the fixed exchange rate of 12.50 pesos per dollar that had remained unchanged since 1954 and let the peso float. Under these circumstances the present Government inherited a near-panic situation and serious economic problems. During the three months following the devaluation of September 1, 1976, inflation was running at an annual rate of about 70 percent, wages had been increased by 23 percent across the board and a new general increase in minimum salaries had to be negotiated before January 1, 1977. The exchange rate market was in disarray and the rate hit a maximum of US$1 = Mex$28 shortly before the inauguration of the new Administration, massive capital flights and financial disintermediation were taking place and the Bank of Mexico had to support the banking system with a special line of credit of about Mex$12.0 billion in order to avoid a collapse of the whole financial system. These financial disequilibria were further aggravated by labor unrest and land invasions with the final result being a sharp drop in economic activity which is estimated to have fallen by about 3-4 percent in real terms during the last quarter of 1976 with respect to the same period of 1975. Origins of the Crisis 3. Wiy did such a situation occur in a country that had been able to sustain a high rate of growth and price stability for three decades? Any serious attempt to answer fully such a complex question calls for a multi- disciplinary analysis and is, therefore, out of the reach of our current  - 2 - knowledge and capabilities and the scope of this report. However, an economic interpretation of the events that led to such situation can cer- tainly be advanced. Such interpretation would also shed light on the constraints the present authorities faced when shaping their economic strategy. 4. As mentioned before (para. 1), the high growth strategy of the past had failed to alleviate the problems of the many Mexicans in need; sharpened contrasts within the society were leading to increasing political and social unrest. The 1968 political crisis is an important landmark in this respect because it brought to national and world attention the gravity of the tensions in Mexico. When President Echeverria took office in December 1970, he was clearly aware of the pressing need for social changes and decided to launch a major effort to bring them about. 5. After its first year in office, a year in which public sector invest- ment dropped by more than 15 percent in real terms and the rate of growth of CDP fell to 3.4 percent -- the lowest level in many years -- the Echeverria Administration shifted to an expansionist policy, in which the public sector was given the primary responsibility for expanding the country's productive base while simultaneously attending to the needs of the many Mexican poor. As Table 1 (page 3) indicates, starting in 1972 public sector expenditures were increased dramatically; they more than doubled in real terms during the sexenio compared to the previous one. 6. Public sector revenues increased too, but not as fast as expendi- tures. The consequence was a rapidly growing public sector deficit and massive borrowing. The public sector deficit (excluding some public enter- prises and financial institutions) went up from 1.8 percent of GDY in 1970 to 8.7 percent of GDY in 1975 (Annex IV, Table 5.2). Both domestic and external borrowings increased dramatically, increasing the country's depen- dence on international financial markets, preempting domestic credit to the private sector and inducing strong inflationary pressures. 7. Public sector aggregate demand grew from 15 percent of GDY in 1970 to 21 percent of GDY in 1976, reflecting the Government's attempt to acquire command of a larger share of real resources in the economy. Such an increase in expenditures in such a short period of time -- the bulk of it took place during 1972-1975 -- would have overheated the economy even if enough resources could have been mobilized to finance it in a non-inflationary way, given tlle profound changes in the structure of demand it would have entailed. 1/ 1/ In aggregate terms any increase in public sector demand can be matched by an equivalent reduction in private sector demand but in practice the adjustment is not so smooth because the composition of public and private demand differ. Similarly, there are differences between the composition of private investment demand and either private consumption demand or public investment demand. Since the structure of supply at a given time reflects the actual and expected structure of demand, any abrupt change in that structure generates excess demand and consequently pressures to increase prices in some sub-sectors of the economy and excess supply in others. The reserve capacity in some sub-sectors of the economy and the access tc foreign markets can help the economy to adjust to such a structural change.  -3- Table 1: MEXICO - SHORT-RUN ECONOMIC INDICATORS, 1965-1976 Average 1965-1970 1971 1972 1973 1974 1975 1976 (Annual. Rate of Growth) GDP (Real) - Mexico 6.8 3.4 7.2 7.7 5.6 4.0 1.6 - USA 3.6 2.9 5.8 5.4 -1.6 -1.6 6.0 Inflation Mexico (wholesale prices) 2.8 3.7 2.9 15.7 22.5 10.5 22.3 USA(wholesale prices) 2.7 3.3 4.5 13.1 18.9 9.2 4.6 Minimum Wages (nominal) 7.7 0 18.3. 5.5 35.9 15.7 29.2 Money Supply (nominal average) 9.7 7.8 14.8 24.7 21.1 22.3 22.8 Percentages of GDY Public Sector Finances Total Current Revenues a/ 11.3 14.8 15.2 16.5 16.6 19.1 20.3 Current Expenditures b/ 6.4 11.1 11.5 14.1 14.8 17.2 17.7 Investment Expenditures D/ 6.2 5.9 8.1 8.2 8.2 10.8 10.2 Deficit c/ 2.6 2.0 4.3 5.5 6.2 8.7 7.4 Balance of Payments Exports (inc. NFS) 8.9 8.3 8.9 9.4 9.3 7.6 8.5 Imports (inc. NFS) 10.0 9.4 10.8 10.6 11.9 10.9 10.4 Current Account 2.4 2.3 2.2 2.9 4.4 5.3 4.3 Inflation Tax dl 0.3 0.5 0.6 1.2 1.8 2.2 2.0 a/ Include Social Security and INFONAVIT contributions and value added by budgetary controlled enterprises b/ Budgetary controlled public sector and INFONAVIT c/ Financial intermediaries and non-budgetary controlled public sector excluded d/ It corresponds to the real quantity of money times the rate of inflation? measured as the relative change in the implicit deflator of GDY. Source: Staff estimates.  - 4 - 8. The Government's attempt to induce a rapid structural change in the economy together with its inability to mobilize enough resources to finance it is at the roots of the recent financial disequilibria. Increasing infla- tion followed by a strong recession were the main outcomes of the evolving crisis and, as usual in these cases, unorganized labor and the self-employed-- the poor in general--turned out to be the losers--although they were supposed to be the beneficiaries of this policy. The evidence available, while flimsy, suggests that both the employment and the income distribution situation might have somewhat deteriorated through 1975-1977, at least in relative terms. Inflation 9. The strong increase in demand brought by the higher public expendi- tures was reflected both in substantial increase in imports and in an accelera- tion of the domestic rate of inflation compared with international trends. Because the increase in public demand was financed increasingly from Central Bank credits and foreign savings, devaluationary expectation developed. This in turn further aggravated the domestic inflation and the balance of payments problems. 10. Prices started to rise in Mexico at a higher rate than the world trends in the beginning of 1973, reflecting the lagged effect of large increases in public sector demand and its structure of financing (Table 1, page 3). In Mexico the rate of increase in wholesale prices exceeded that of the USA 1/ in 1973 and in the first quarter of 1974, but the situation level- led off in the rest of 1974 and 1975. As a consequence, at the end of 1975 the ratio between Mexican and USA wholesale prices had deteriorated by 10 percent compared to 1972. 11. Adjustments of minimum wage are a good proxy of general wage move- ments in Mexico. Until 1972 minimum wages continued to be adjusted, as in the past, every two years and the 1972 adjustment was in line with previous experience. In September 1973 the Government decided to grant an accross- the-board compensatory wage adjustment of about 18 percent, to be followed by the regular January 1, 1974 adjustment (14 percent) and later, in October 1974, by a new general adjustment of about 22 percent, which covered the period October 1974-December 1975. As a consequence, wages in Mexico increased at a substantially faster pace than wages in the US. In 1975 the ratio between Mexican and US wages was 50 percent higher than in 1972. It grew even higher in the first half of 1976. Growth of Output 12. All this affected the pace of economic activity. After two years of strong expansion (1972-1973), GDP growth slowed down and became negative in per- capita terms in 1976. The decline in the rate of growth of GDP was mainly due to the weaker performance of agriculture, manufacturing, construction and com- merce-sectors which accounted for about 70 percent of domestic production. The evidence available strongly suggests that the poor performance of these sectors -- with the exception of agriculture -- is explained by the effect on 1/ US prices were used as as reference because about 65 percent of Mexico's foreign trade is with the USA.  - 5 - the economy of the marked slowdown in external demand in 1975 and in domestic demand, in turn affected by the changing composition of public investment in 1976. 13. The decline in the rate of growth of GDP occurred in spite of a slight increase in the investment coefficient -- from 19 percent of GDY in 1965-1970 to 20.4 percent in 1971-1976 -- which indicates that the change in the composition of fixed investment that took place over the period may be a second important cause of the slowdown in the rate of growth of GDP. The bulk of rapidly growing public sector investments was allocated to petroleum, electricity, steel and large irrigation works, most of them with a long lead time, while private fixed investment slowed down, partly because of the slowdown of external and private final consumption demand and partly because of what much of the business and financial community felt to be a deteriorating investment climate. 14. The slowdown of external demand was attributable to the effects of the world recession and to the overvaluation of the Peso. The slowdown in domestic demand in 1976 is linked to uncertainty as to the financial conditions of the economy, and to a more restrictive public expenditure policy. Thus, savings on current account for the consolidated public sector increased from 1.9 percent of GDP in 1975 to 2.6 percent in 1976, while capital expenditures were reduced from 10.8 to 10.2 percent of GDP in those years. 15. Inflation affects both real final consumption demand and its struc- ture because it tends to reduce the purchasing power of incomes, but not propor- tionately for all income strata. Financially sophisticated economic agents and strong labor unions are in better position to defend themselves, the former by hedging against inflation, the latter by getting their incomes adjusted, while unorganized labor and many self-employed people lose out. There is a strong evidence that this has been the case in Mexico. Even though real wages of organized labor have increased since 1974, people earning a salary represent only about 30 percent of the labor force. The rest, self-employed, under- employed or unemployed people, have in general faced stagnating or even declining real incomes. Open unemployment has also increased, particularly since 1975, adding to the slowdowns in real income and in consumption demand. 16. The effects of inflation on the level and structure of private consumption can be illustrated by the different behavior of output of non- durable and durable consumption goods, which account for the larger share of total consumption of the lower and higher income groups, respectively. Table 2, page 6, shows that the rate of growth of both kinds of goods declined over the period but at a different pace. The reduction in the rate of growth of non-durables started in 1973, reflecting the quick accommodation of supply to the slowdown in consumption due to the increased inflation. 1/ The level of per capita production of non-durable goods was lower at the end of 1976 than at 1/ It could also be argued that this was a reflection of the slow-down of agricultural, particularly crop-production but the large increases in food imports that took place in 1973-1974 suggest that the slow-down in non-durable production cannot be, at least entirely, attributable to supply problems.  Table 2 MEXICO: PRODUCTION OF MANIFACTURED GOODS, 1972-1976 (rates of change over the same period in previous year) Consumption Goods Production Goods Total Total Durables Non-Durables Total Raw Materials Investment Goods 1972 2 2. 11.3 9 2 10.8 1973 v B_4 8.3 6.6 10.7 17.6 19716 14.6 2-5 89 L 19.5 14.o 4.1 u 3.8 3-9 3.3 7. I. Semester 2.6 1.6 7.2 o.6 3.5 2.1 10.9 II. Semester 5.5 6.7 4.3 7.2 4.5 h.5 4.4 1276 L1. 1.7 1.2 1.8 2.2 3.9 -6.6 7 I. Semester 4.2 3.8 4.9 3.5 4.6 5.1 2.2 II. Semester -0.3 -0.3 -2.5 0.0 -o.14 2.7 -15.6 n.a. : not available -- : not applicable Source: Bank of Mexico  the end of 1974. Inflation also affected the demand for durables, but produc- tion growth continued at a decreasing but moderate pace until 1975 -- reflect- ing the different effects of inflation on the structure of consumption (income) and rapid inventory accumulation. In 1976 the rate of growth of durables production declined sharply and became negative by the end of the year, owing to the additional strain on final consumption. Legacy of the Crisis 17. In 1974 an attempt was made to bring the situation under control. Its major results were a mild slowdown in the rate of growth of prices and a recovery of private sector investment. However, these efforts were quickly abandoned and 1975 witnessed a new expansion move, its rationale being the need to counterbalance the effects of the world recession. The results were a record public sector deficit of 8.7 percent of GDY and a record balance of payments current account deficit of 5.3 percent of GDY, accompanied by a slowdown in GDP growth and a drop in real terms of private investment. 18. The corrective actions taken in public sector's expenditures in 1976 were not sufficient to restore private sector's confidence, given the financial desequilibrium that started in 1972. This led to widespread devaluation expectations and huge capital flights--amounting to about US$2.8 million--which forced the Government on August 31 to abandon the fixed exchange rate of US$1=Mex$12.5 that had prevailed since 1954. Al- though the devaluation was announced as part of a comprehensive package of economic measures, the package was never fully implemented and the country was pushed to the brink of financial collapse. 19. When President Lopez Portillo took office, he inherited a country in panic, lacking confidence in its economic leadership, economically stagnant, highly indebted -- foreign public sector debt had climbed to US$19.6 billion by the end of 1976 from US$4.3 billion at the end of 1970 -- and uncertain about its future. However, he also inherited (a) a broader and clearer understanding of the degrees of freedom available to Government for the allocation of real resources within the economy, within the existing political and social framework, and (b) the possibility of making a special effort to address effectively the problems of the poor due to the relaxation of the financial constraints brought about by the discoveries and subsequent develop- ment of vast petroleum resources. Stabilization Period 20. In his most recent State of the Union speech, on September 1, 1978, the President clearly recognized the basic parameters governing economic relations in Mexico as being the existence of (a) "a market economy," (b) "a legal system that guarantees both social and individual rights, including the right to private property," and (c) a [geographic] situation that requires the maintenance of "unrestricted monetary exchange." These parameters are consis- tent with the functioning of a mixed economy in which the roles of the public and private sectors are clearly defined and socially accepted, and economic policies are such that stable and equitable growth is achieved.  - 8 - 21. With this framework in mind and given the grave situation it inherited, the new Administration assigned first priority to regain confidence in economic management and overcome the crisis. For this purpose a stabiliza- tion program was designed, which called on the leadership of the economic authorities and the cooperation of the labor movement, the private sector and the international financial community. 22. The Government committed itself to prudent fiscal and monetary policies and adopted a mixed strategy aimed at reducing lower priority public expenditures while proceeding with high priority investments in petroleum and other sectors. An in-depth administrative reform aiming at rationalizing and improving the overall efficiency of the public sector was implemented, and a special effort to improve the management of public enterprises through more rational pricing and cost control policies was launched. 23. The monetary authorities were to adopt a tight monetary policy and announced their decision to continue floating the Peso. A series of other measures were taken to help the recovery of financial savings and stop and reddress capital flights and to increase credit to the private sector. Early in 1978, the Bank of Mexico started issuing treasury bills thereby gaining additional flexibility for managing monetary policy. 24. The labor movement was asked to exercise restraint in wage demands and accept a 10 percent guideline; minimum wages were adjusted by 10 percent in January 1, 1977. They were again adjusted by about 13 percent on January 1, 1978. The President has clearly stated that he will not allow real wages to deteriorate further but that no automatic price-wage adjustment mechanism will be established. 25. The private sector was asked to exercise restraint on price increases and an agreement was reached between the Government and private producers to ensure an adequate supply of "socially necessary goods and services" at reason- able prices. In return the price controls on other goods were relaxed, special tax and credit support was given to firms in financial difficulties related to the devaluation and, in general, relations between Government and the private sector were greatly improved. 26. The international community was asked to continue supporting Mexico on the basis of the stabilization effort underway and Mexico's potential as a producer and exporter of hydrocarbon resources. Immediately after the devaluation, Mexico obtained major support from the IMF. The Extended Fund Facility (EFF) agreement covering 1977-1979 was ratified by the new Government shortly after taking office three months later. 27. The Government was able to regain confidence in economic management and obtain support for its program. Labor accepted the wage guidelines, the private sector responded slowly but firmly to the challenge and the interna- tional community continued backing Mexico in spite of its high debt service burden. The Government has been able to bring the situation under control, the economy is recovering and future growth seems assured.  - 9 - 28. The devaluation iLn the last quarter of 1976 brought back Mexican prices in line with the world market price levels. In spite of its strong impact on domestic inflation, at the end of 1978 the ratio between Mexican wholesale prices expressed in dollars and US wholesale prices was about the same as that of 1972. The ratio between Mexican wages and US wages both expressed in dollars, which had increased more than 50 percent between 1972 and mid 1976, was only about 15 percent higher in 1978. The public sector deficit, which was reduced from 8.7 to 7.4 percent of GDP from 1975 to 1976, was further reduced to 5.1 percent in 1977. This deficit may have increased somewhat in 1978., but the jump in oil revenues should bring back this deficit in the following years to a manageable proportion. The rate of inflation which soared to annual rates of about 70 percent in the few months after the devaluation, was reduced to less than 18 percent in 1978. 29. There are, howev(er, some aspects in the short run economic picture which call for close surveillance. The rate of exchange has been kept practically constant since early 1977, while during 1977 and 1978, inflation has been higher in Mexico than in the US. Monetary policy has been expansive. While in the years prior to the devaluation the money supply was increasing at about 22 percent per year, since the devaluation that rate has increased to more than 30 percent, and has stayed at those high levels. 30. Some reduction in the real value of the exchange rate may be a necessary consequence of tihe increase in the volume of foreign exchange available. Nevertheless, the question is how to minimize this reduction in order to maintain the rate of exchange as an incentive for increase in the non-oil exports. An alternative is to open wider the economy and thereby match the increase in supply of foreign exchange with a more or less equivalent change in demand for this exchange. Another alternative would be to adjust the finances of the public sector without recurring to steep increases in the oil revenues or of the external credit. If a high level of public expenditures is to be attained, this would require a substantial increase of taxes and of the prices of the goods and services supplied by the Government (fuel, elec- tricity, etc.). If none of these alternatives is taken, it will be very difficult to avoid the acceleration of inflation and/or an important apprecia- tion of the domestic currency, and a resulting decline in non-traditional exports. Furthermore, there is a danger of a cumulative effect as the com- bination of domestic interest rates, external interest rates and the exchange rate make it increasingly attractive to borrow abroad, this adding to reserve accumulation, monetary expansion and inflationary pressure. 31. Public sector's current savings, outside of those provided by PEMEX, deteriorated substantially between 1970 and 1975. Its subsequent improvement was inadequate. Non-PEMEX public's current savings represented 3.1 percent of GDP in 1970 and only 0.1 percent in 1975, increasing to 0.7 and to 1.1 percent of GDP in 1976 and 1977 respectively. On the expenditure side, most of the recent adjustment in public sector finances have taken the form of a slowdown in public sector investments, which dropped in real terms by about 7 percent in 1977 with respect to 1976 -- already a low year -- and even though it increased 19 percent in 1978 reaching Mex$55.4 billion (1972 prices) its level  -10- Table 3: 'LEXICO - SHORT-RUN ECONOMIC INDICATORS, *1976-1978 Average 1971-1976 1976 1977 1978 d (Annual Rate of Growth) GDP (real) Mexico 4.9 1.6 0.9 6.0 USA 2.8 6.0 4.9 4.4 Inflation Mexico (wholesale) 14.5 22.3 41.1 15.8 USA (wholesale) 9.9 4.6 6.1 7.8 MIinimum Wages (nominal) 16.8 29.2 27.9 13.0 Money Supply (nominal average) 21.1 22.8 27.7 31.5 Percentages of GDY Public Sector Finances Current Revenues a/ 17.1 20.3 20.0 21.7 Current Expenditures b/ 14.5 17.7 16.7 18.0 Investment Expenditures _/ 8.6 10.2 8.5 9.2 Deficit cf 5.7 7.4 5.1 5.3 Balance of Payments Exports (inc. NFS) 8.7 8.5 10.0 9.8 Imports (inc. bNTS) 10.7 10.4 10.2 10.4 Current Account Deficit 3.9 4.3 2.7 3.0 Taflation Tax 1.4 2.0 3.3 1.9 a/ Tnclude Social Security anad MMFON&VIT contributions b/ Budgetary controlled public sector and INFONAVIT c/ Financial intermediaries and non-budgetary controlled public sector excluded d/ IBRD estimtes  - 11 - in real terms was still below that of 1975. Since PEMEX investments have been growing rapidly in real terms over the same period, the adjustment has besen borne by the other sectors and particularly by the social sectors. A strong increase in public investment can be achieved without a substantial incresase in the non-PEMEX public savings, because of the magnitude of the oil revesnues. Nevertheless in the final account this may prove to be self defeating, because of the increasing dependency on oil revenues. 32. Mexico's capacity to earn foreign exchange through petroleum exports has greatly relaxed the pressure on the balance of payments and on the exchange rate and may eventually lead to an overvaluation of the domestic currency. Similar development was observed in most other oil exporting LDC's (Venezuela, Algeria, Iran). To fix the exchange rate at its nominal level while inflation is reduced only gradually may in the short run have a negative impact on export, employment and possibly poverty alleviation. It may also prove to be counter-productive in the long run as Mexico's petroleum wealth is gradually used up while its economic structure will be ill-prepared to take over the oil sector's important role of a provider of the foreign exchange and public revenues.  - 12 - CHAPTER II MAJOR DEVELOPMENT ISSUES 33. As President Lopez Portillo said in his State of the Union speech on September 1, 1978, now with Mexico's financial problems under control, efforts should be shifted to tackle the really difficult social problem. As past experience has shown, GDP growth alone is not enough. Growth in production is necessary but not sufficient for adequate progress on the pressing social problems facing Mexico. To achieve such progress, a large share of the benefits of growth must assume the forms of expanded employment, increases in the productivity and the incomes of the poor, and adequate supply of basic goods and services. Growth with equity and stability will, therefore, require a clear commitment and strong actions on the part of the Government to effec- tively attack the basic structural problems facing Mexico. These problems, are: the high rate of growth of population and poverty, the stagnation of agricultural production and the large regional imbalances. We turn now to a detailed discussion of each of these, and of the policies and programs being implemented or under consideration. Poverty and Productivity 34. Mexico's problems of rapid population growth, low productivity workers and poverty are serious, and there are no quick solutions to them. The relevant data, which come from many different sources, are not always in agreement, and some estimates are difficult to accept. Given more time, it should be possible to put together a more accurate and complete picture of trends in productivity and income distribution. But even without the results of such work, the broad outlines of great inequality in productivity and in incomes, and little or no improvement over the last 15 or 20 years, are unmistakable. 35. There are three main causes of the present situation: (a) rapid population growth, especially since about 1950, with a resulting high propor- tion of nonworkers to workers and rapid growth of the labor force; (b) practical difficulties in the implementation of land reform, combined with neglect of agricultural laborers and owners of poor land (including many ejidatarios) who now comprise over 2 million of Mexico's 10 million families and some three- fourths of all Mexicans engaged in agriculture; this problem is in part related to the further one of stagnation of agricultural production over the last 5-10 years; (c) policies that made capital equipment cheaper and labor more expensive than would otherwise have been the case, thus resulting in adoption of a relatively capital rather than labor-intensive technologies. We shall examine each in turn.  - 13 - 36. (a) Rapid population and labor force growth: 1/ During the past 25 years the population of Mexico has grown at about 3.2 percent per year, increasing from 26.5 million in 1950 to about 59 million by 1975. This rapid growth is the result of a substantial decline in mortality and continued high fertility. The extensive economic development that took place in Mexico during the same period, as well as the fall in mortality, prompted many observers to speculate that Mexico was on the verge of the second stage of its demographic transition, namely, a precipitous drop in fertility. Nevertheless, fertility rates remained high through the early 1970s. Birth registration data for recent years, however, indicate that the number of births may be declining. El Colegio de Mexico has estimated that the crude birth rate in 1975 was 40.5 per thousand (down from 43 per thousand in 1970) and preliminary registration data for 1976 put the birth rate at about 38 per thousand. While the estimate by El Colegio de Mexico is used in this report, it should be noted that many authorities remain skeptical, contending that the reported decline in the crude birth rate reflects under-registration rather than any real changes in fertility. 37. The decline in mortality and the persistence of high fertility have produced a markedly young population. Between 1950 and 1970, the median age declined by almost two years to 16.6 years; by 1970 about 47 percent of the population was below 15 years of age. These changes in age structure have brought about a sharp increase in the dependent population. The age dependency ratio rose from 85.6 in 1950 to 100.0 in 1970. The large proportion of the population under age 15 poses serious problems for the country's capacity to educate these young people and to absorb them into the labor force. 38. The population of working age (12 years and over) expanded at a much slower pace than total population in the period 1950-1970, but somewhat more rapidly than it since 1970, because the large cohorts born during the earlier decades are now reaching working age (Table 4, page 14). The labor force itself grew more slowly than the population of working age (and hence even more slowly than total population) from 1950 to 1969, owing to a sharp drop in participation rates during these two decades. 2/ The labor force ex- panded at an annual rate of 2.0 percent from 1950 to 1960, at 2.7 percent: annual pace in 1960-1969, and currently seems to be expanding at over 3 percent per year, even after deducting emigration assumed to be over 200,000 people per year. 1/ For more information, see Mexico Background Paper: Demographic Trends and Prospects, World Bank, December 1978. 2/ Caution is in order in interpreting these rates, which are derived from census data in which the definition of labor force has changed over the period, thus limiting comparability. An unexpectedly sharp drop in participation rates for males reported in 1970 seems especially dubious. While there is little doubt that some decline did take place, the 1970 census seems to have seriously underestimated male labor force partici- pation, especially for 20-24 years old males.  Table 4: MEXICO: POPULATION AND LABOR FORCE, 1950-1969 Annual Growth Rates 1950 1960 1969 1960/1950 1970/60 (thousand people) (percent per year) Total population 26,463 36,003 48,715 3.1 3.4 Population 12 years of age 17,307 22,680 29,713 2.7 3.2 and over Labor force (economically active 8,345 10,213 12,955 2.0 2.7 population) a/ The 1970 Census was taken in January and most of its questions on employment referred to activities performed the previous year. Source: Mission estimates based on Oscar Altimir, "La Medici6n de la Poblaci6n Econ6micamente Activa de Mexico, 1950, 1970," Demogiraifa y Economia, Vol. VIII: 1, 1974; and Victor L. Urquidi, "Ehpleo y Explosi6n Demografica," Demografia y Economla, VIII: 2, 1974.  - 15 - 39. (b) Agriculture, land reform, and rural poverty: Policies applied in agriculture have not succeeded in providing enough remunerative work for Mexico's growing population. Land reform, started after the revolution of 1910, proceeded fastest in the Cardenas period (1935-1941), when massive land redistributions were made, input and credit needs of ejidatarios were recog- nized, and ejidos were organized. After Cardenas, land distribution continued, although at a reduced pace, but most of its beneficiaries were given little in the way of credit and other inputs. Rather, Mexico's development strategy shifted to large-scale irrigated farming and to industrialization, leaving many poor farmers and landless peasants behind. From 1942 to 1969, government investment in the rural sector was concentrated on roads and on regional river basins and other irrigation programs which focused on building large dams and other infrastructure. In the early 1970s, the Mexican government began to reassess strategies for fostering growth and development in rural areas and, with changed financial allocations, encouraged a number of government agencies to step up their efforts to improve the lot of the rural poor. The year 1977, which is the latest for which income distribution data are available, comes too soon after the start of these programs for their effects to be reflected in the data. 40. The stagnation of production in agriculture as a whole over the last ten years or so, analyzed in paras. 71-79 below, only exacerbated these problems, as opportunities for productive employment weakened. The reduced growth of production combined with the increased availability of labor resulted in declines in the average number of days worked and in the average annual incomes of farm workers. 41. (c) Inadequate absorption of labor in manufacturing: The rate of creation of productive jobs was further hindered, especially in the manufactur- ing sector, by economic policies that biased production towards more capital- intensive products and processes. Promoting import substitution of manufac- tures through protection against imports and other special investment incentives, and the negative effects on exports of manufactures, primary products, and services that are inherent in such a strategy, biased the structure of production away from more labor-intensive products. Investment subsidies tied to capital equipment purchases, as well as taxes and other charges related to wages, increased the bias. 42. Mexico achieved much of its promotion of industrial expansion by measures that made capital equipment cheaper, while it used measures that increase the cost of labor to achieve part of its income distribution and other social goals. Import duties on capital goods are low, and are often waived completely. Loans for the purchase of fixed plant and equipment are more readily available than for working capital. These measures lower the cost of equipment by 25 to 50 percent, compared to what it would have been in their absence. On the wage side, the "seventh day" payment, the 5 percent INFONAVIT contribution, the vacation bonus, the one percent education tax and the 9-12 percent social security tax, the year-end bonus, and the 8 percent of profit sharing levy raise the cost of labor by as much as 50 percent.  - 16 - (IBRD, Mexico - Manufacturing Sector: op. cit). For workers earning the minimum salary, these charges are borne completely by the employer, or by his customers if he can pass on the increased costs in the form of higher prices for his products. For workers earning above the minimum salary, however, a part of these charges may be reflected in lower nominal wages, and thus be borne by the workers. The net result of the increased labor charges and the subsidies of equipment purchasing is that the cost of labor, relative to the cost of equipment, is perhaps twice what it would be in the absence of these measures. Such a large distortion must affect the relative amounts of the two factors used, and is especially inappropriate in view of the explosive growth of Mexico's labor force. Eliminating policies that make capital goods arti- ficially cheap, as well as replacing wage-linked taxes and other charges with income or value-added taxes, should promote greater use of labor and thus reduce poverty in Mexico. 43. The result of these three processes (rapid population growth, low productivity of a large part of the agricultural labor force, and a bias in favor of capital-intensive production) was to maintain or enlarge wide differ- ences in labor productivity between different parts of the Mexican economy. The picture here is somewhat clouded by lack of comparability among the population censuses for different years, and conflicting information as between the population censuses and the economic (i.e. industrial and agri- cultural) censuses. However, the view that emerges is as follows. Productivity Differentials 44. Both in 1950 and 1969, value added per worker in Mexico's agricul- ture was less than one-third the national average, which in turn was about one-fourth the level prevailing in mining and petroleum. 1/ (Annex IV, Tables 2.17 and 2.18). Sectoral figures are averages, however, and do not show the differences in productivity within sectors. These differences are recorded for the manufacturing and agricultural sectors in the economic censuses. According to the 1970 Industrial Census, value added per worker in Mexico's manufacturing industry varied within a very wide range, from 15 thousand pesos per year in establishments with gross production below 1.5 million pesos per year, to 124 thousand pesos in establishments producing more than 150 million pesos annually. The average value added was 52 thousand pesos per year in all industries. Workers' earnings also vary widely in Mexico, from an average of 7.5 thousands pesos per year in small establishments, to an average of 39.3 thousand pesos per year in the 200 largest ones. 45. The same duality exists in agriculture. As shown in Table 5, page 17, in 1968 the 7.1 percent of farms classified as modern produced almost 3 times as much output (by value) per hectare, and 7 times as much per capita income, as "traditional" farms. The contrast with "subsistence" farms 1/ If the number of workers engaged in agriculture in 1969 was seriously underestimated, as many believe, then value added per worker in agri- culture was even less in 1969.  - 17 - Table 5: MEXICO: PATTERNS OF LAND TENURE AND PRODUCTION, 1068 Moders'-/ Traditional-/ SubsistenceX/ Total Number of Farms (thousands) 200.1 i,140.0 1,479.4 2,815.9 Ejidos 120.3 676.2 1,062.3 1,858.8 Private 81.2 463.8 412.1 9L7.1 Percent of total 7.1 40.5 52.4 100.0 Average area of land per farm devoted to crops (ha) 33 14 8 12 Total land included in farms devoted to crops (thousand ha)-' 6,650 15,960 11,835 34,445 Incomes Mex$/farm 62,840 9,282 3,472 10,069 Mex$/ha devoted to crops 1,904 663 434 839 Average per capita income/(Mex$) 9,975 1,428 489 1,547 I/ Modern - farmers using modern technology. Traditional - farmers producing cash crops without modern technology. Subsistence - farmers who produce little output for cash market. 2/ Includes some pasture and woodland. Source: Cisneros, Manuel Rodriguez et. al., Caracteristicas de la Agricultura Mexicana, p. 255. Basic data from survey by the Banco de Mexico.  - 18 - is even greater. In 1960, the largest one-half of one percent of all farms produced one-third of the output and accounted for almost half of the increase in production since 1950. The largest 3.3 percent of all holdings produced over 55 percent of output and accounted for 80 percent of the increase in production since 1950 (Annex IV, Table 2.22). From 1960 to 1970, all private farms (i.e., excluding ejidos) taken together reported a decline in the value of production. But the 5 percent of largest private farms showed an increase of 18 percent over the decade, while every other size category reported declines. Thus, concentration in the private sector increased. (Annex IV, Table 7.8). In short, in the 1940-1970 period the government pursued a two-way strategy in agriculture: one part of it, the social justice was addressed through continued redistribution of land, but many peasants received no land at all, other recipients received poor land and nothing else and hence their incomes remained at or near a subsistence level; the other part of it, the agricultural output was increased by bringing new land under irrigation and farming most of it in large, commercial units. More peasants got land and Mexico achieved increased agricultural production, but few of the peasants were able to share in producing the increased output or in enjoying the income derived from it. Most rural families must supplement their income from agriculture because it is not adequate for survival. Income Distribution 46. The further results of these wide variations in productivity are reflected in the highly unequal distribution of personal income in Mexico. Here also the data are incomplete and do not agree in all details (see Annex II), but the general picture is what one would expect as the result of the analysis sketched above. The main sources of data used here are three gene- rally comparable household budget surveys, done in 1963, 1968, and 1975. The results of the 1975 study should be regarded as preliminary, especially because the recently published (April 1977) survey for 1977 shows information which largely undermines that published in 1975. Conclusions as to unequal income distribution are reported in earlier studies, such as those by Navarrete, studies for 1960 and 1970, and others. 47. Income in Mexico is distributed very unequally according to the preliminary Mexican data, the average real income of the poorest 20 percent of Mexican families has remained unchanged during the 1963-1975 period. The sharp decline shown in Annex IV, Table 2.23, is difficult to accept, and requires further study. Such study should be undertaken on the basis of the results of the 1977 survey and of the preceding ones. 48. Estimated inequality in Mexico in 1975 was greater than that predicted by an international cross-section analysis of income distribution in many developing countries, even though the international pattern reflects the fact that most countries at or near Mexico's stage of development have higher inequalities than either more-advanced or less-advanced countries. A further comparison of income distribution in Mexico and that in several similar countries is made in Annex IV, Table 2.24. Given the uncertainty with regard to the 1975 data for Mexico, both 1968 and 1975 estimates are presented. A cross-country comparison using the 1968 data for Mexico indicates a slightly  - 19 - less unequal distribution of income in Mexico than in either Turkey, Brazil or Colombia, where the lowest 40 percent of households receive less than 10.1 percent of income while in each case the richest 20 percent receive more than 60 percent of income. 49. The reported slight increase in relative inequality between 1963 and 1975 can be statistically separated into its urban and rural components (Table 6, page 20). This analysis shows that reported inequality decreased in urban areas between 1963 and 1975, but increased greatly in rural areas. 1/ The marked increase in the index for rural areas, may be overstated; but the direction of change is consistent with the other evidence, reflecting the increasing auality of Mexican agriculture. The figures for urban areas suggest that the large increases in the urban labor force were absorbed without the parallel decline in the share of the poor in urban incomes ancl this in spite of the relatively slow growth of employment in manufacturing. 50. In 1975 most "poor" Mexican families--52 percent--were in the agricultural sector. 2/ Of these, 33 percent are listed as self-employed-- presumably ejidatarios and other small proprietors--while 18.5 percent are listed as salaried--presumably landless agricultural workers. Thus the most important single group of the poor are land-owning peasants and ejidatarios-- 1.5 million families--and the second most important are the even less fortunate landless rural workers--850,000 more families. These 2.4 million families, with average annual incomes estimated preliminarily at approximately US$80 per capita, are the core of Mexican poverty. They include 76 percent of all families in agriculture in Mexico (Annex IV, Table 2.20). The structure of Mexican poverty has not changed much since 1963. The percentage distribution by sector of "poor" families in 1963 was virtually identical to that in 1975; in 1963 agriculture accounted for 54.5 percent of poor families. In 1963 there were proportionally more landless workers and fewer of those who owned land (Annex IV, Table 2.19). 1/ Comparable data are not available for 1968. The criterion for urban in the 1963 and 1975 data is a place with 10,000 or more inhabitants; for the 1968 data it is a place with 2,500 or more. 2/ "Poor" families are defined in the survey as those households with incomes less than half of the estimated national mean--Mex$1,621 per month, equivalent in purchasing power to approximately US$1,315 per year per family, or about US$225 per capita. This cut-off point classifies some 4.6 million Mexican families--45 percent of total--as "poor." An alternate definition, that is commonly used in the World Bank to measure relative poverty, would classify 38 percent of Mexican families as "poor" in 1975. The minimum salary in Mexico, including typical profit sharing and vacation bonus, was Mex$ 2,056 per month in 1975. Thus, the cut--off level used here, for income per family, is approximately 25 percent below the (per worker) minimum wage. Characteristics such as sector, occupation, and education refer to the head of the household.  - 20 - Table 6: MEXICO: THEIL INDICES OF INEQUALITY FOR URBAN AND RURAL AREAS 1963 1975 Urban .488 .402 Tzral .429 .795 Nation-wide-/ .554 .6o8 a/ The nation-wide indices reported here are different from those shown in Table . oecause these were calculated with more d.isaggregated data. The numerical value of the index depends, in general, on the level of disaggregation of the basic data used. Source: Staff estimates from household surveys.  - 21 - 51. Families whose head is not working account for only 12 percent of the "poor"--about 5 percent of all families--and their average incomes, while low, are one-fourth again as much as those of the poor employed in agriculture. Other important sectoral-occupational concentrations of the poor are both self-employed and salaried workers in manufacturing and construction. 52. The uncertainty about the precise shape of income distribution, as well as the need to understand better how it came about, may be a productive topic for researchers. But policy-makers do not need this additional infor- mation to know that the problem is serious. Examination of likely future prospects shows that the labor force will be growing even faster in the next ten years or so than it did in the past (para. 63 below). Urgent and concerted action is therefore necessary, even if only to prevent the problem from getting worse. 53. The Government is aware of the need for improving the income distribu- tion pattern and have made attempts in this direction, such as the efforts to increase the living standards in rural areas by giving high priority to employment oriented rural projects and to achieve a more progressive tax system. However, these efforts have not been fully successful in the past, partially because of the lack of necessary complementary policies, and also due to the adverse impact of increasing inflation in recent years. Population, Employment and Basic Needs Future Population and Labor Force Growth 54. As it became apparent that rapid population growth would adversely affect development goals, the Mexican Government shifted in the early 1970s from a neutral stance regarding the question of contraception to one of active advocacy of family planning. In 1972 Government Health Agencies were instructed to provide information and contraceptive services. Programs to train medical and paramedical workers to dispense family planning services were initiated. With the enactment of the General Law on Population in 1973, the Mexican Government adopted a population policy as part of its program of economic and social development. The legislation underscored the need to strike a balance between demographic increase and the possibility of achieving higher standards of living. It stressed the need to improve the health and social welfare of the Mexican population, particularly that of the neediest groups. 55. Since it launching in 1973, the government family planning program has gained considerable momuentum. By mid-1976, approximately 1.2 million women were protected under public sector programs and 600,000 were purchasing contraceptives through commercial channels. Many claim that the program has already had a demographic impact and that the decline in the birth rate reported in 1976 is largely attributable to the program. 1/ The Government 1/ Victor Urquidi, "Recent Trends in Mexican Population Growth: The Beginning of a Reversal". Paper delivered at Tokyo International Symposium, April, 1977'. The 1.8 million who were covered represented approximately 13 percent of the female population aged 15-49 years.  - 22 - hopes that an intensive program will reduce the birth rate by one point per thousand per year between 1976 and 1982, thus resulting in a rate of 33 births per thousand population. To achieve this target it is estimated that about 4 million additional women must practice contraception. On October 28, 1977, the Government announced a goal of reducing the growth of population to 2.5 percent per year by 1982, and to 1 percent by the year 2000. 56. While family planning efforts in Mexico appear to have made notable progress, even more remains to be done. The majority of acceptors under official programs are age 35 or older. The fertility potential of this age group is only about one-third that of women 15-34 years old. Even though recent data shows that more younger women are starting to participate, con- siderable additional effort must be made to increase the number of younger women covered by the various programs. This will be a formidable task in view of the large numbers of women entering these age groups in the next decades. Program efforts to date have largely been concentrated in the urban areas; if targets are to be met, activities in rural areas will have to be expanded. 57. An explicit population policy is only one element in the effort to reduce the rate of population growth. Many other policies associated with socio-economic development have an inhibiting effect on fertility. The inverse relationship between education and fertility suggests that increasing education for females would have the desirable side effect of reducing the rate of demographic growth, especially if income earning possibilities for women also improve. In Mexico, it should be possible to increase the proportion of the female population enrolled in school significantly, since only 52 percent of the female population 6 - 19 years old was enrolled as of 1970. 58. New projections of the likely ranges of future demographic growth in Mexico have been made by Bank staff, based on the latest information available. Six different basic sets of projections were made, reflecting three different assumptions about changing fertility and two about migration. Mortality was assumed to decline gradually and equally in all six sets. Fertility assumption (1) reflects the very ambitious goals of the Mexican Government which call for a rapid fall in fertility such that the rate of population growth would be 1 percent by the year 2000. Assumption (2) predi- cates a less rapid decline. Assumption (3) is based on a very slow fall in fertility. 59. Since most Mexican emigrants are bound for the United States and their numbers considerably exceed the presently authorized limits, migration assump- tion (A) assumed a constant out-migration to the United States of approxi- mately 240,000 persons annually between now and the year 2000. This flow could be comprised of either undocumented migrants or legally admitted migrants if the immigration quota were to be increased. Assumption (B) corresponds to the present level of authorized Mexican immigration into the United States of (approximately) 40,000 and assumes that this figure will remain unchanged over the projection period. Series (B) thus provides a benchmark; it illustrates the population growth that would occur if only currently legal migration were to take place.  - 23 - Population Growth 60. The population in the year 2000 according to thC six projections ranges from 95 million to 131 million persons (Table 7, page 24). Ea:h of these six scenarios is not equally likely to take place. In view of the de- cline in fertility that apparently has already occurred and the strong commit- ment of the Government to the family planning program, projections 3A and 3B, which assume only a slight decline in fertility, are probably too high. On the other hand, projections IA and IB, based on the Government's targets, seem to be optimistic but not unreasonably so. To reach these targets would imply a very rapid increase in the effective use of contraception, buttressed by other social, economic and cultural policies known to have a negative impact on fertility. As to migratiLon, assumption (A) seems most likely. The popula- tion of Mexico by the year 2000, according to projection 2A, will be 113 million -- more than twice the 1970 level. Out-migration at the currently legal level for emigration to the United States would imply a population in the year 2000 of 120 million (Projection 2B) -- an annual growth rate of 2.9 percent for the period 1970--2000, as opposed to 2.7 percent under the more plausible series, 2A. Even assuming that government targets are achieved, which becomes increasingly probable, given the energetic action and its tangible results in the lowest projection (1A), Mexico's population would grow at an average annual rate ofi 2.1 percent between 1970 and 2000, and would number some 95 million at the beginning of the next century. School Age Population 61. The proportion of the population aged 6 - 14 years will decrease during the projection period; the proportion of the population of secondary school age will increase slightly until 1995 but will decrease by 2000. The absolute numbers in both categories, however, will grow. Over the 20 years between 1970 and 1990, an average of about 405,000 new places per year will be required in primary and intermediate schools (ages 6-14), and 175,000 new places per year in secondary schools, just to provide room for the growth in school age population (Tablie 8, page 25). 62. Thus expenditures for education would have to increase even if only 1970 enrollment ratios (79 percent for primary, 30 percent, secondary) were to be maintained; higher enrollment rates and an upgrading of the quality of education would call for further increases in both investment and current expenditures in the education sector. The growth of the 15-19 year cild age group, coupled with the previous low secondary enrollment implies that a considerable effort will 'have to be mounted to raise the enrollment rates for this group.  Table 7: .wXICO: YOPULATION PROJECTIONS 1970-2000 Mid-Year Population Average Annual Growth Rate Assumptions 1970 1980 1990 2000 1970-80 1980-90 1990-2000 1970-2000 (millions) (percent per year) IA 51.0 67.3 83.7 94.7 2.8 2. 1.2 2.1 1B 51.1 69.6 3o.2 101.3 3.1 2,4 1.4 2.3 2A 51.0 67.6 88.6 112.9 2.9 2.7 2.5 2.7 2B S1.1 69.8 93.2 119.8 3,2 2.9 2.5 2.9 3A 51.0 67.9 92.0 123.8 2.9 3,1. 3.0 3.0 3B 51.1 70.1 96.6 131.0 3.2 3.3 3.1 3.2 Key to Projections: Letters indicate emigration assumptions: A. 267,000 per year from 1970 tlhrough 1976; 240,000 per year thereafter. B. 67,000 per year from 1970 through 1976; 42,000 per year thereafter. (The current legal level is about 42,000 per year.) Numbers indicate fertility assumptions: 1. Consistent with the Governmient tar-et to reduce population growth to 1 percent per year by 2000. The total fertility rate (j2FR) declines co 5.33 by 1980 and to 1.85 by 2000. 2. Less rapid fertility decline;, but still on the optimistic side. TFR declines to 5.57 by 1980 and 3.80 in 2000. 3. MSodest declines in fertility - TlR of 5.88 ir, 1980 and 5.00 in 2000. Source: World BPnk projectlons (AuLust 1976).  - 25 - Table 8: MEXICO: SCHOOL-AGE POPULATION, 1970-2000, ACCORDING TO PROJECTION 2A (millions) 1970 1980 1990 2000 Primary and Intermediate (ages 6-14) 12.6 16.6 20.7 24.9 Secondary (ages 15-18) 4.4 6.1 7.9 9.7 Source: World Bank Projections (Series 2A). Labor Force Growth 63. There is a lag of fifteen to twenty years between changes in fertility and the impact of those changes on the population of working age. Thus the population of working age in the year 1990 has already been born and can only be affected by substantial shifts in mortality or migration. It is unlikely that unforeseen changes in mortality will alter the projected size of the working age cohorts; but fulure changes because of migration are less certain. Labor force growth under projection 2A is shown in Table 9, page 27. The annual growth is projected at about 3.2 percent, or approximately 550,000 workers per year during 1975-1980. The growth rate is projected to accelerate to 3.4 percent per year dur:ing the 1980-1990 period. As already explained, changes in fertility would lhave negligible effects until 1985 or 1990. Cutting emigration, however, could raise the growth to as high as 4 percent per year (750,000 persons), or even higher, depending on how the reduction in out-migration affects labor force participation rates. Employment and Basic Needs 64. Even without considering Mexico's current poverty population, this virtually certain growth constitutes a formidable challenge. When the (roughly) 6.4 million Mexican workers earning less than the minimum wage are also taken into account, the need for a strong and concerted effort to increase the productivity and incomes of the poor becomes even more evident. Even under assumptions of fairly rapid growth in output in the modern sector (around 10 percent during 1979-1982) and fairly slow growth in labor productivity (about 4.5 percent), the number of workers earning less than the minimum wage (40 percent of the labor force or 6.9 million workers, in 1978) is projected to remain roughly constant per year during this period.  - 26 - 65. Faster growth of output is one obvious way to increase productive employment. However, growth assumptions of this report are already mildly optimistic. The estimates show that any GDP growth rates much slower than those projected in Chapter IV would transform what is at best a serious employment problem into a potentially explosive situation. 66. Population policies are also important ways to reduce poverty, even though most of their effects will not be felt for several decades. Unless the growth of population is reduced below its current level of some 3.0 percent per year Mexico may not be able to improve the conditions of the poor in the foreseeable future. The Government's population program (para. 55), as well as other effects of development on fertility, lead to expect a further progress and eventual success in this area. 67. Continued employment problems are inevitable in Mexico, given the rapid growth of the labor force. If these problems are to be ameliorated, the counter-productive policies mentioned in para. 42 must be reversed. The initial measures of an employment program should, therefore, be (a) to foster the expansion of export activities, not only because they are generally more labor intensive than industries established to substitute imports but also as an element to foster total economic growth; (b) to correct, as far as possible, factor price distortions that stimulate the application of capital intensive techniques; and (c) to reallocate Government revenues to programs designed to accelerate the development of agricultural production and labor productivity in small farms and to help small urban enterprises as well. Step (a) has been taken, in large part, by maintenance of a roughly adequate exchange rate and CEDIs. Step (c) and partly step (b) are in progress. 68. Since such policy changes will in any case not be sufficient to solve the problem, the Government should be prepared to supplement them with additional policies such as: (d) to expand labor-intensive activities in social services (mainly in education and health) and in infrastructure (construction and maintenance of low cost housing, waterworks, drainage systems, schools, hospitals, etc.), as well as to increase labor-intensive rural public works designed to raise the standards of living and productivity of small farmers and rural workers (land clearing, soil conservation, rural roads, small irrigation works, etc.); (e) to coordinate education with labor market needs; (f) to improve employment services; and (g) to intensify efforts to slow down labor force growth. 69. Even with large increases in programs to improve productivity of under-productive workers, and to increase the rate of job creation in the modern sector, Mexico's population and labor force are growing so fast that many families will continue to live in poverty. Increased direct assistance to help these families meet basic needs for nutrition, housing, and health care should be considered. The mission has not studied either the specific needs for such programs or how they might be designed. However, the basics of how to design such programs in urban areas are fairly well known, and the main requirements for Mexico are the political will to transfer more resources in this manner and the organization of the effort. In rural areas, low density of population creates further difficulties; the approach under consideration in SAHOP of concentrating such programs in designated small towns (centros rur-urbanos) may be promising.  Table 9: MEXICO: LABOR FORCE PROJECTIONS Labor Force Average Annual Crowth Rates 1970 1975 1980 1985 1990 1995 2000 1970-80 1980-90 1990-20(U (intill ons) (percent per year) 'total 13.5 15.7 18.4 21.8 25.7 30.0 34.7 3.1 3.4 3.1 Males 11.0 12.5 14.3 16.6 19.2 22.3 25.8 2.7 3.0 3.0 Femiiales 2.5 3.2 4.1 5.2 6.5 7.7 8.9 5.1 4.7 3.2 Source; World Bank ProJeccions (based on projection 2A)  - 28 - 70. Although the staff has not attempted to quantify the effects of the various suggestions made above, it appears that under any realistic assumptions the problems of under-productive workers will remain serious in Mexico for some time to come. Because of this, the mission suggests that the Government consider establishing a small group of analysts attached to the Office of the President, with a mandate to study any and all proposed programs with a view to devising modifications that would enhance their employment and poverty-ameliorating effects. The problem is vast enough, and important enough, to merit extraordinary attention and high priority. Agricultural Potential Past Experience 71. Agriculture played an active role in Mexico's economic development during the 1945-1965 period, with a growth rate of more than 5 percent per year. Since the mid-1960's, however, production growth rates of major cereal crops have fallen below the growth in domestic demand. Agriculture's contribution to national production has been declining: the sector's share of GDP declined from 16 percent in 1960 to about 10 percent in 1976-1977 (Table 10, page 30). The Bank of Mexico reports a drop in growth of crop production from almost 6 percent per year during 1945-65 to less than 1 percent per year from 1965-1975, far below the estimated 3.3 percent rate of population growth (Table 11, page 31). The recent trends in maize, beans, sugar and cotton were particularly disappointing, while wheat and rice production increased at acceptable rates, and sorghum and soybean production continued to increase rapidly. 72. In contrast to the poor performance in crop production, livestock production has increased steadily throughout the 1970's at an average annual rate of about 4 percent. 73. To understand Mexican agricultural performance during the last 30 years three distinct stages must be analyzed: 1945-1955 with a rapid annual growth of 6.0 percent; 1955-1965 when the annual growth rate, although still higher than that of population, declined to 4.2 percent; and finally 1965-1975 when the sector growth fell to 2.1 percent, well below the estimated demographic growth of 3.3 percent per year. 74. During the 1945-1955 period substantial increases in both area and yields determined the rapid growth of the sector. In 1946 SRH was created and focussed on development of large irrigation schemes bringing into produc- tion the most favorable lands of the northwestern zone. On the average, about 110,000 ha (new and improved) per year were developed during this period. The total (irrigated and rainfed) area planted in the five major crops increased from 7.2 million ha in 1945 to 9.7 million ha in 1955, i.e. 3.1 percent per year. In terms of area, cotton was the fastest growing crop (from 365,000 ha in 1945 to 905,000 ha in 1955) followed by sugarcane and wheat. Yields also tended to increase appreciably for cotton, wheat and beans (at annual rates of 5 percent, 4.5 percent and 4 percent), less for maize (2 percent) and negli- gibly for sugarcane. The aggregate effect of area and yield increases resulted in annual output growth of: cotton, 14.5 percent; wheat, 9.5 percent; beans,  - 29 - 6 percent; sugarcane, 5.8 percent; and maize, 4.5 percent. Thus, the remark- able rate of growth of crop production of 7.3 percent annually during the 1945-55 decade was mainly due to the expansion of cotton and other crops in new irrigated areas. 75. When compared to the previous decade, annual growth of Mexican agriculture during the 1955-65 period slowed down by about one-third. The expansion of irrigation declined to about 65,000 ha/year. The annual rate of area expansion of the five major crops decreased substantially (1.8 percent in 1955-1965 compared to 3.1 percent in 1945-1955); in addition, there was a significant shift in land use patterns since maize and beans accounted for most of the expansion, wheat area remained virtually unchanged, and cotton area shrunk by 120,000 ha over the period. Thus, although wheat and cotton yields increased respectively by 145 percent and 70 percent over the decade due to improvement of varieties and agricultural practices, crop production grew at an annual rate of 4.4 percent, 40 percent lower than in the preceding period. Opportunities for expanding exports of high value crops narrowed, new irrigation projects were less productive because of increasing physical difficulties, complexity and costs, and a less well prepared population of new beneficiaries, to whom sufficient technical assistance was not provided. Mexican agriculture was approaching the limit of its potential growth within the framework of the development model successfully used in the past. 76. In an attempt to understand better agricultural developments during recent years, available data have been analyzed from a regional standpoint. 1/ Countrywide figures show a substantial increase from 1960 to 1965, a levelling off during 1965-1970, and a decline afterwards (Annex IV, Tables 7.2 and 7.4). Only the state of Tamaulipas and the Northwest region show a steady growth, mostly due to large irrigation schemes, while in virtually all the other regions the expansion of the early sixties arrives at its turning point by 1970 (Annex IV, Table 7.6). 77. Analyzing the distribution of cultivated area by crops, it could be seen that total cropped area stagnated after a sharp increase in the first half of the 1960s; in fact there is a slight decline of 3 percent between 1965 and 1974. 1/ The regional composition is as follows (Federal District and Nuevo Leon excluded): Region/State States Northwest B. California N., B. California S., Sonora, Sinaloa North Chihuahua, Durango, Coahuila Tamaulipas Tamaulipas Pacific Nayarit, Jalisco, Colima, Michoacan, Guerrero, Chiapas, Oaxaca Central Aguascalientes, San Luis Potosi, Zacatecas, Queretaro, Guanajuato, Hidalgo, Mexico, Tlaxcala, Puebla, Morelos Gulf Veracruz, Tabasco, Campeche, Yucatan, Quintana Roo Source: CONACOSA, Desarrollo Agricola Regional: Estrategias de Politica, Mexico, D.F., October 1976.  - 30 - Table 10: MEXICO: AGRICULTURAL PRODUCTION, SHARE IN GDP, AND FOREIGN TRADE (Mex$ Million; 1972 prices) .___________ 1970 1974 1976 1977 Agricultural Production 51,700 55,600 54,300 55,800 Agriculture share in GDP 11.3 10.4 9.7 9.8 (percent) Agricultural Exports 10,418 13,021 9,836 15,021 Agricultural Exports Share in Total Exports (percent) 25.7 22.7 17.4 22.4 Agricultural Imports 3,051 10,104 3,506 7,100 Agricultural Imports Share in Total :mports (percent) 6.3 9.5 2.2 3.3 Agricultural Trade Balance 7,367 2,917 6,330 7,921 a/ All figures deflated by the GDY implicit deflator. Source: Ministry of Programming and Budgeting (P and B) and mission's estimates.  -31- Table 11: MEXICO: AGRICULTURAL PRODUCTION, 1945 - 1975 Year Total Agriculture Agriculture/GDP GDP Crops Livestock C + L Crops Livestock C + L --(Billion pesos at 1960 prices)- -----(Percent) 1945 63.3 6.3 4.3 10.6 10 7 17 1955 111.3 12.8 6.2 19.0 12 6 18 1965 212.9 19.6 8.9 28.5 9 4 13 1975 388.1 21.4 13.7 35.1 6 4 10 Period Annual Rates of Growth GDP Crops Livestock C + L 1945-55 5.8 7.3 3.7 6.0 1955-65 6.7 4.4 3.7 4.2 1965-75 6.2 0.9 4.4 2.1 Source: Banco de Mexico. Production valued at 1960 prices. Values shown are three-year averages.  - 32 - The continued increase of oilseed area (370,000 ha or 88 percent) is offset by losses in both grains (in which beans are included) and industrial crops (cotton and sugarcane) (5 percent and 16 percent respectively). Soya (270,000 ha) and safflower (130,000 ha) are the fastest growing crops; sorghum expan- sion practically compensates the loss (800,000 ha) of maize; beans area shrinks by 570,000 ha and cotton area which during the first half of the decade had fallen by 50 percent (400,000 ha in absolute terms), increased thereafter and ended with an overall 30 percent decline (a net loss of 230,000 ha) over the 1965-1974 period (Annex IV, Table 7.7). 78. The gross value of crop production has grown moderately in real terms (11.5 percent for the 1965-1974 period as a whole) in spite of a contraction in cropped area of 3.2 percent, implying an increase in average yields of 15.2 percent. Average yield increases have therefore also contributed to offset the loss of area. This was mainly accomplished by marked increases in wheat and sorghum yields, which more than compensated for the low and stagnant maize and declining soybean yields. (Annex IV, Table 7.11). 79. If average yields are used to measure relative performance, the best performance was in the Northern region where yields rose by 36 percent, based mainly on grains. Gross value of production in the North increased by 23 percent while cropped area decreased by 12 percent; wheat accounts for most of the growth since its yield doubled from 1.5 t/ha in 1965 to 3 t/ha in 1974. In the Pacific and Central regions, yields grew by 20 percent, mostly due to a substantial decrease in the area devoted to low value crops (maize and beans) which was partly offset by higher yielding crops, mainly sorghum and, in the Pacific region, sugarcane. In the Northwest -- the highest growing region, in terms of gross value of production -- yields grew by 7 percent; the most remarkable feature was the boom in oilseeds, the expansion of which accounted for most of the region's growth. An appreciable proportion of this expansion comes as a result of introducing oilseeds--mainly soya--as second crop in the commercially oriented irrigation districts of the region. In addition to the crops studied here, the Northwest accounts for most of Mexico's fruit and vegetable output for export. The Gulf region and the state of Tamaulipas show the least favorable performance. In the Gulf region only sugarcane shows an upward trend, thus confirming the apparent suitability of the zone's ecological conditions for that crop. In Tamaulipas, the contraction of cotton area as well as the low effective response of the new lands brought into cultivation by the Bajo Rio Bravo, Bajo Rio San Juan, and Soto la Marina irrigation projects have led to a decrease in yields of 30 percent. In those irrigation projects a substantial proportion of the area is used for a single crop of either maize or sorghum and, to a much lesser extent, oilseeds (Annex IV, Table 7.9). Issues and Prospects 80. The increasingly unsatisfactory performance of Mexican agriculture has been caused in large part, by four sets of policies: (a) the overemphasis of large-scale irrigation, while neglecting ways to increase production in the vast under-exploited or unexploited rainfed areas of Mexico; (b) land  - 33 - tenure policies which in some respects not only reduce production potential but also constrain further land redistribution; (c) enormous subsidies through underpricing of irrigation water, which not only were (and still are) drains on the Federal Treasury but also result in misallocation of resources; and (d) an inappropriate structure of crop support prices, which has caused shifts in planting to lower-value crops. 81. Unused Potential of Rainfed Agriculture: Mexican agriculture has always shown a clear-cut division between the irrigated and rainfed subsectors, corresponding to the co-existence (until 1977) of two ministries involved with agricultural development: the Secretaria de Recursos Hidraulicos (SRH), in charge of irrigated agriculture; and the Secretaria de Agricultura y Ganaderia (SAG), responsible for rainfed agriculture and livestock. A large gap between their respective budgets led to an unbalanced situation between the two sub- sectors. While SRH became a strong agency whose efforts and experience in the field of large hydraulic schemes are recognized world-wide, SAG remained relatively weaker. In December 1976, the current Mexican administration took a highly positive step by mkerging the two ministries into a single one, the Secretaria de Agricultura y Recursos Hidraulicos (SARH), with a view to coordinate actions and to inject into the rainfed sub-sector the much needed capability to increase its share in the sectoral output. 82. Mexico's agriculture and livestock land is estimated at about 115 million ha of which only 35 million ha are classified as agriculture land proper. Of the last, only 15-20 million ha are under cultivation. At the end of 1976 Mexico's irrigated agricultural sector consisted of approximately 5 million cropped ha (including double-cropping); the rest was mostly under rain- fed conditions. A large share of the uncultivated agriculture and livestock land is also under rainfed conditions and comprise almost all the tropical low- lands of the Gulf coast and the Pacific coast south of Mazatlan. Studies under- way could further enlarge the estimate of this vast potential by reclassifying as "cultivable" certain areas now deemed restricted to livestock development. 83. Rather than directing their efforts towards improving rainfed agri- culture, past Mexican governments have traditionally chosen to carry out large-scale irrigation schemes, mainly in sparsely populated areas, in order to avoid land tenure and related socio-political problems. This policy was for a time highly successful in increasing output, thanks to (a) the suitabil- ity of natural resources; (b) the relatively low complexity of the projects; (c) the managerial and technical competence of project beneficiaries; and (d) substantial financial support to the producers by means of low prices charged for inputs. Consequently, a relatively small segment of Mexican agriculture -- mostly private but including some ejidos -- achieved a rapid and steady growth. Conversely, the policy led to two adverse consequences: first, the momentum generated by successful development or large irrigation schemes pushed the programs beyond the limit of high return projects; and secondly, most of the basic problems of a large depressed segment of the agricultural sector were left unsolved. This approach has now changed and the emphasis is being laid upon the rehabilitation of irrigation districts.  - 34 - 84. Awareness haS been growing, tnat, Lhrough the neglect of rainfed agriculture, the full production potential of about twc-thirds of the culti- vated area was not being realized, and the income disparities between the modern and traditional sector thereTy aggravated. In tle ecerly mid-1970's several programs were sz.2rted to provide improve-' extension assistance, pri- marily on maize growing. to rainfed crop farmers (PRONDAAT,. as well as complete developmeilt packages to low-:Lncome farmiers In selectcd areas (PIDER, Papaloapan). As these programs were started on.l.y recently, some time must elapse before their results can be ascertained and assessed4; their main thrust is, however, correct in the context of promoting balanced development in Mexican agriculture. Tncreases in the output of rainfed agriculture appear highly promising, but would require major efforts in manpower mobilization and research, as well as capital investments in infrastructure. 85. Land Tenure Problems: Legal and administrative problems related to land tenure have also been serious obstacles to increasing land utilization and agricultural production in some parts of Mexico. Solution of these problems would require tackling delicate issues centering on the interpretation and implementation of the provisions of the federal laws regarding land water resources. These are highly sensitive issues of which Mexican officials are aware and cannot be resolved on solely technical or economic grounds. The two major ones relate to use of land and improvements to land, as they affect the maximum permitted size of holdings. 86. The Agrarian Reform Law sets the maximum size of individual holdings for ranches as the area required to carry 500 animal units, and forbids the growing of commercial crops on ranch land. In many parts of the country there is a considerable potential for commercial crop production on lands classified for ranching. However, under the present law any owner who develops any part of his ranch for crop production would be liable to have his land re-classified as agricultural and thus would be in danger of expropriation of a large part of his holding. The law is also a disincentive to increasing carrying capacity of existing ranch lands to anything beyond the 500 animal limit, since this also would place the owner in danger of expropriation. Ranching areas through- out the country, from the arid highlands to the mid tropics, could be improved through better grazing management, and the use of improved pasture and mineral supplements. Carrying capacities could be increased thereby up to fivefold if incentives were right and complementary inputs were provided. 87. Although the Government has tried to encourage crop production by granting temporary (1 year) certificados de inafectabilidad to ranchers to permit them to grow crops without fear of expropriation, only a handful of ranchers have responded. Until the uncertainties as to land tenure policy are resolved, most commercial ranchers are unlikely to make any serious efforts to increase productivity and large areas of potentially productive land will remain underutilized--a luxury that Mexico cannot afford in the face of its growing population. 88. The other major negative effect of land tenure policy on development of Mexico's agriculture concerns the interpretation of the Federal Water Law as it applies to drainage and flood control. Although the law is not com-  - 3 ; pletely clear on SiAS po4nt, it can be interprerted as limiting the maximum size holdiog iin a drainage: leod c )ntrol ar&;a to 100 Lna when the project is privately financed and 20 hla when it is financed publicly, the same as for irrigationi projects. There is no apparent reason to equate drainage/flood control programs with irrigation programs in a formula for determining the maximum perrmissible size of holdings; the effect of the law is to make private land holders unwilling to improve their properties through drainage and flood control, sinco. they fear that to do so would render them subject to expropriation of a'll lands in excess of the above limits. Holdings of the sizes permitted have not, and almost certainly will not, encourage private investment in drainage facilities or the exploitation of publicly drained lands. 89. These laws, which were enacted to help achieve social justice, not only reduce Mexico's potential agricultural production, but also, paradox- ically, may make it more difficult to increase the incomes of poor farmers. Since the laws inhibit potentially profitabLe investments on and improved management of what are relatively large private landholdings, they effectively reduce the amount of productive land available for redistribution. Techni- cally, the laws could be changed so as to remove restrictions, part of the present landholdings could be redistributed, and the present owners would be left with less land but higher profits while presently landless peasants could receive productive land. Socially and politically, these changes are of course extremely difficult to achieve. 90. The greatest potential for such changes is in the humid tropical zones along the Gulf Coast. The safest and possibly mRost productive method of land exploitation is a imixed farming system combining livestock and crop production. Such a system would help maintain soil structure and fertility at low cost, and would also increase economic stability and productivity in the region. The basic area permitted per holding would have to be adequate to ensure that a viable crop-livestock rotation could be established; this would require more land than that now allowed for intensive cropping under irriga- tion. Investments in drainage, flood control, and perhaps some irrigation works would also be necessary to achieve the increased production, especially in areas that are flooded during the rainy season. Research and extension, to develop and promote use of technologies appropriate to the area, would also be needed to increase productivity. 91. Underpricing of Irrigation Water: In recent years, irrigation has become a growing source of federal outlays and subsidies without bringing about production increases comparable to earlier years. A key issue has been the failure to set and collec:t water charges high enough even to cover oper- ation and maintenance costs of the projects. This has resulted not only in exceedingly high subsidies to the majority of beneficiaries, but in a wasteful and inefficient pattern of water use as well. Subsidies to cover operation and maintenance in large-scale irrigation alone are estimated at Mex$1.0  - 36 - billion in 1977. If capital costs were included, the total subsidy would be about Mex$6.0 billion. Small-scale irrigation, where subsidies are lower, would add another Mex$1.5 billion to the already heavy bill. 1/ 92. Although the fiscal burden of insufficient tariffs is their more visible aspect, their impact on water-use efficiency is at least equally important. CPNH data indicate that the average efficiency of the 20 "distritos principales" covering 60 percent of the harvested area in 1972 was 46 percent during the 1967-1973 period. Furthermore, a study of irrigation water use, comprising actual use with theoretical water requirements, showed a wide disparity of water-use efficiency at farm level, ranging from 18 percent in the Rio Blanco District (Veracruz) to 92 percent in the Rio Yaqui District (Sonora), indicating a great potential for water savings. The average effi- ciency was considerably higher in districts with rates related to water use (71 percent) as compared with districts with fixed charges per hectare (53 percent). It is clear that higher and volume-related water charges would lead to an improvement in water-use efficiency and hence to increases in the crop- ping intensity of irrigated areas. 93. Structure of Support Prices: Self-sufficiency in basic foodstuffs has been a long-standing objective of agricultural policy. Since the mid- 1950's support prices have been used as a tool to achieve it. Considerations of international comparative advantage do not appear to have played a signifi- cant role in setting the absolute or relative levels of support prices for different commodities. Commercial maize growing has been made increasingly unattractive compared to wheat, sorghum, oilseeds and alfalfa, even though for the latter crops yields have risen considerably faster than for maize, and possibilities of double cropping have opened up. As a consequence, maize production has declined. The maize support price, which already typically exceeded the import equivalent in the past, has now been set at a level that may well be excessive in terms of possible economic and social costs (Annex IV, Table 7.10). 94. Maize is admittedly an important component of the popular diet in Mexico, and the white variety favored by consumers is not always readily available in world markets. It seems appropriate to question whether it is in the country's interest to incur the considerable social cost that high support prices that overrule considerations of comparative advantage are likely to entail. Research on improved maize varieties and cultural practices (mainly by CIMMYT) has not yet achieved major breakthroughs, as occurred in the case of wheat; nor have those results that have been achieved been properly disseminated among farmers. It seems worth considering a substantial strengthening of these efforts, together with an expansion in cultivated area, while maintaining relative product prices more in line with real costs, even if this policy implied continued maize imports in the immediate term. 1/ Beneficiaries of small-scale irrigation pay 30 percent of capital costs, while those using large-scale irrigation do not pay any part of capital costs. In the calculations the mission assumed that the capital costs of new irrigation projects would be US$3,000 and US$1,500 per ha, respec- tively, for large- and small-scale projects, and that the amount of land developed in 1977 would be 75,000 ha, and 60,000 ha, respectively.  - 37 - 95. The current administration has recognized the foregoing issues and has undertaken a far-reaching administrative reform that could set the stage for broader and more effective Government action in the sector. The emphasis is being put in productivity gains, yield increases, extension and research; small-scale irrigation is being fostered and large-scale irrigation deempha- sized; and the objective of self-sufficiency is being rephrased in terms of the sector's overall capacity to pay for its own imports. An ambitious agro- industrial program is in its first stage of preparation and it will aim at increasing employment and productivity in the rural areas. 96. To highlight the increasing importance of developing rainfed areas, a Directorate of Rainfed Districts (DT, distritos de temporal) has been estab- lished to parallel the existing Irrigation Districts' Directorate (DR). To date 110 DTs have been defined. Districts are to be administered by a steer- ing committee chaired by a representative of the State government and including delegates of the Directorates of DT, extension, farmer organizations, and water supply works. It appears that the principal role envisaged for the DTs will be to coordinate actions of agencies already operating in the area, with- out any executive powers of their own. Although coordination of ongoing actions would undoubtedly have a positive effect; in time, as human resources become available and goals and actions better defined, the DTs may have to be given leadership and executive functions. 97. Attention is also being given to water pricing. After having increased from Mex$170 million (US$13.6 million) in 1974 to Mex$1,000 million (US$47 million) in 1977, the operational subsidies of the irrigation districts have been frozen at the 1977 current level and the Government has announced its intention to make them financially self-sufficient by 1982. Regional Imbalances 98. In 30 years Mexico has changed from a rural country with only weak economic links among its diverse regions to an urbanized and geographically integrated society. Mexico's population growth has been among the most rapid in the world, and this growing population has migrated to cities apace. The percentage of the nation's people living in cities of 2,500 or more inhabitants grew from 35 percent in 1940 to 59 percent in 1970. 99. In 1975 there were some 11 million people living in and around Mexico City, making it the third largest metropolitan area in the world. Its popula- tion has been growing annually at over 5 percent, which means over half a million additional people each year. UN projections show it growing to 'be the largest city in the world, with a population of over 30 million people, by the year 2000. Within the metropolitan area, in 1975 the Federal District alone had 14 percent of the nation's people, and accounted for over 30 percent of the nation's output in manufacturing, 40 percent of the nation's commercial sales and over 50 percent of the nation's services.  - 38 - 100. To stop the growth of Mexico City is certainly not possible and probably not even advisable. However, to slow it down to some extent by promoting growth in selected other parts of the country woulld be beneficial. The city exhibits all the negative characteristics typical of fast-growing metropolis: tremendous air pollution; over taxed transit, water and sewerage systems; and all the other problems of coping with over 500,000 new inhabitants every year. Mexico City has a further problem in that additio-nal water must be brought great distances at high costs. Thus, although a rigorous estimate of costs and benefits of growth in alternative centers is not possible, an a priori case for trying to divert some of Mexico City's future growth to other areas seems reasonable. 101. Promoting faster growth in selected medium and smaller cities would be far more than a means to reduce growth in Mexico City. More important goals are to provide more productive work in many parts of the country, to bring access to markets, credit, health, and other services nearer to neigh- boring small towns and rural areas (the "hinterlands" of the cities), and thus to improve both economic and personal welfare of people who live in places that are not close to the major cities of the nation. 102. Mexico has large regional disparities in welfare -- as most other semi-industrialized countries do. Per capita income per month in 1969 ranged from Mex$628 in the Federal District to Mex$109 in Oaxaca -- a range of almost 6 to 1. Access to health care, education, potable water, and other basic needs also varies dramatically from one region to another. Although some poor regions present very difficult barriers to attempts to raise incomes, many other parts of Mexico can become more prosperous if appropriate policies are followed. Among the poorest and least-developed regions that are especially promising are (a) parts of the coast of the Gulf of Mexico in the states of Veracruz and Tabasco, (b) the Isthmus of Tehuantepec, in the same states and also in Oaxaca and Chiapas, and (c) the Lazaro Cardenas-Zihuatanejo region of the states of Michoacan and Guerrero. 1/ Developments in petroleum and related activities will cause the first two areas to grow rapidly, regardless of policy; for these areas the challenge is to guide this growth so as to make it beneficial to the local inhabitants as well as the inevitable immigrants. 103. Promoting and guiding growth in promising medium-sized urban areas such as those on the Gulf Coast, the Isthmus, and the Lazaro Cardenas area is therefore in the national interest. Such a strategy can both (a) bring employ- ment and urban services to poor regions, and guide growth in these places so as to be orderly and productive, and (b) shift growth, albeit slightly, to places other than the largest cities. 1/ Recent World Bank studies of regional strategy options in Mexico, based on research done in early 1974, are: Economic Development of the Isthmic Region of Mexico (TIBRD 1080-ME), March 30, 1976; Spatial Development in Mexico (IBRD 1081a-ME), January 31, 1977; and Urban Development in Mexico (IBRD 1449-ME), January 31, 1977. A study on the Tabasco economic development is now in progress.  - 39 - 104. Mexico is now sufficiently developed for a strategy of selective decentralization to succeed. The nation has many cities that are reasonably large and growing, where additional workers can be highly productive. The second and third largest cities, Guadalajara and Monterrey, each have popula- tions of over 1 million, are major industrial and service centers, and haLve been growing even faster than Mexico City. They have the size and economic and social advantages to assure their continued growth. Rapid growth in petroleum production and processing will make urban growth inevitable in and around Coatzacoalcos, Minatitlan, and Villahermosa. The challenge for policy in these cities is to manage the growth so as to reduce congestion, pollution, high infrastructure costs, etc., and to provide basic urban services for all the inhabitants. Forty-four other Mexican cities each had over 100,000 inhabi- tants in 1970. Most of these cities also have the infrastructure, transport and communications links, labor pool, and other advantages that induce future growth. Measures to increase such growth and to manage it in an orderly fashion, paying due attention to the needs of low income residents and immigrants, would have a double pay-off: problems of managing the growth of the capital can be reduced (at least slightly) while efficiency from the national point of view can be increased. 105. Until recently Mexico, like most other nations, did not have a strong explicit spatial policy. Instruments that affected spatial development were not well-coordinated. Among the most important components of past policy (both intentional and unintentional) have been the following: (a) A number of tax exemptions have been available, since 1972, to industries that are located outside the three largest metropolitan areas. Unfortunately, these incentives are too small to affect most decisions, and are available for several high inccome, highly urbanized, rapidly growing areas as well as in poorer and more backward areas. Therefore, the main effect of the incentives has been to increase profits for firms that would have been located in these growing areas in any case. (b) Industrial parks, complexes and cities, and commercial centers have been set up in many places, especially since 1971. A few of these have been quite successful, such as the Queretaro "industrial city." However, most have met with but little success because the main benefit provided -- physical infrastructure -- is not sufficient: to attract most business to the unfavorable locations chosen for many of the sites. (c) A number of programs assist small- and medium-sized industries. Almost all business firms outside the three largest cities are small enough to be eligible, and thus the programs may have helped develop some of Mexico's poorer regions. About one-third of all eligible firms outside the three major cities have been helped by these programs.  - 40 - (d) The maguila program, in which goods are assembled at least in part from inputs imported in bond and then re-exported to the United States, has increased growth in several border cities. (e) Low prices in Mexico City for water, and nationally for gasoline, diesel fuel, and rail transport of raw materials, have all promoted growth in Mexico City. Many of these prices have been raised in the last few years, but to help induce even a little decentralization they would have to be considerably higher. (f) Centralized government decision-making has also tended to increase the growth of Mexico City. The Mexican administrative system, in which many decisions are taken on individual case- by-case basis, makes location in the capital valuable for many businessmen. 106. Within the last three years Mexico has moved towards establishing a stronger, coherent and explicit spatial policy. The Law of Human Settle- ments, adopted in June 1976, provides the legal authority for planning and implementing programs to achieve spatial goals. The stated objectives of the law include improved rural-urban integration, more balanced growth as among cities and regions, promotion of growth in medium-sized cities as alter- natives to the largest cities, greater citizen participation in solving urban problems, land use control, better provision of urban services, and improvement of the housing location-job location-commuting situation within cities. In practice, the most important objectives that seem to be emerging are: (a) National level: (i) Reduce the growth of Mexico City. (ii) Concentrate resources to develop a few promising alternative regions, as opposed to the past practice of generalized incentives described in para. 105. (b) State level: Establish priorities for allocating investment among cities or regions within each state. (c) Local levels: Plan for more orderly urban growth, and implement these plans with public investments and land use control. Where an important urban area is in more than one state, or in more than one municipio, a "conurbation commission" will be established to make these plans (see para. 108). 107. On the national level, both the new Secretariat of Human Settlements and Public Works (SAHOP) and the Secretariat of Programming and Budget (SPP) have potentially important roles. SAHOP has prepared a national urban develop- ment plan, and has continuing responsibilities to revise the plan, do studies, chair the conurbation commissions, advise state and municipal governments on their own relevant laws and plans (at their request), and execute the national plan. The SPP is charged with assuring that federal government investment is consistent with the national, state, and local plans.  - 41 - 108. The Law of Human Settlements provides for new institutions on a new level: the conurbation commissions. Any urban area that is in more than one state, or that simply consists of more than one municipality, can be declared a conurbation (by the national or state governments, respectively, in the two cases mentioned). The commissions for interstate areas will be chaired by the Secretary of SAROP, and composed of relevant state governors, mayors and other officials as appropriate. Their powers are to make a plan for the area (subject to approval by the President), to try to induce local, state and national governments to act: in accordance with the plan (they have no real authority here, except through their members), and (within their powers) to regulate land use within the conurbation area. Conurbation zones will be established mainly in cases of large metropolitan areas (Mexico City was the first), and of smaller areas where either present structure or desired future growth crosses state boundaries (Lazaro Cardenas and the other two conurbation zones established so far, Torreon-Gomez Placio-Lerdo and Tampico-Ciudad Madero-Pueblo Viejo). 109. The existence of a national urban-regional plan, and SPP's respon- sibility to assure that the national government's spending is consistent with that plan, gives the central government the means to implement a strong urban- regional policy. Planning and land use regulation by state or local govern- ments (with conurbation commissions supervising planning in multi-state urban areas) are potentially powerful complements to the federal role. It is still too early to say just how these powers will be used, but the potential for coordinated action exists.  - 42 - CHAPTER III INVESTMENT FINANCING 110. Investment in Mexico grew very fast in the first half of the 70's. After dropping 5 percent in 1971 it increased more than 50 percent in real terms from 1971 to 1975 (see Table 13, page 44). As the relative price of investment goods 1/ increased, the required saving effort had to increase proportionally more than investment. Total savings, which represented about 19 percent of GDP in 1970-1971, increased to 22 percent in 1975. A marked change in the composition of savings in GDP could be attributed to foreign savings, a result of a higher current account deficits in the balance of payments. 111. Mexico's overall national savings performance compares favorably to many countries at a similar stage of development. For the last 10 years Mexico's national savings have averaged around 16-17 percent of GDY, a fairly high savings level (including the inflation tax) (Table 12, page 43). 112. In particular, the share of public savings dropped markedly. Real public savings declined from a peak of Mex$20 billion (4.3 percent of GDY) in 1970 to Mex$11.8 billion (1.9 percent of GDY) in 1975. Excluding PEMEX, real public sector savings shrank from Mex$15.3 billion in 1970 (3.3 percent of GDY) to Mex$1.1 billion in 1976 (0.2 percent of GDY) (Table 13, page 44). 113. In spite of the poor public savings performance, public investment grew at a markedly higher rate than private investment during 1971-1976. The gap between public savings and public investment was financed by large increases in domestic and external borrowing, and inflation. The balance of payments current account deficit increased markedly after 1972, reaching a peak of 5.3 percent of GDY in 1975 and declining to 3.7 percent of GDY in 1976. At the same time, the inflation tax 2/ became an important source of government financing. It represented 1.6 percent of GDY in 1975, 2.2 percent of GDY in 1976 and peaked at 3.3 percent in 1977. In comparison, gross federal government revenues from the commercial receipts tax amounted to 2.4 percent of GDY in those years (Annex IV, Table 5.5). 1/ Defined as the implicit deflator of investments goods divided by the implicit deflator of GDY. 2/ Inflation tax is defined as the rate of inflation multiplied by the real quantity of money. It represents the resources the money holders must spend in order to keep constant the real value of their cash balances. The rate of inflation is defined as the relative change in the implicit deflator of GDY. Because both the nominal quantity of money and the price level are defined as changes from mid year to mid year, the method just mentioned for calculating the inflation tax underestimates the correct figure in years of acceleration of inflation, the opposite happening when inflation decelerates. Conceptually, one should be working with figures calculated for the end of the year, but this poses practical problems.  - 43 - Table 12: MEXIC0 - F`IXED INVESTMENT AND ITS FINANCING, 1965-1976 (Billions of 1972 Mex$)a/ Average 1965-70 Average 1971-76 Average 1965-76 Growt1 Share Growth Share Growth Share Rate of GDY Rate of GDY Rate - of GDY Gross Fixed Investments 9.4 19.0 5.6 21.0 7.3 20.4 Public Sector (13.5) (6.5) (7.8) (8.2) (10.4) (7.7) - Private Sector (7.4) (12.5) (4.2) (12.8) (5.7) (12.7) Gross National Savings 8.0 16.6 5.0 17.1 6.4 17.0 - Public Sector (-1.8) (2.6) - Private Sector (7.3) (14.5) Balance of Payments Current Account Deficit - 2.4 - 3.9 - 3.4 Memorandum Items Savings Channeled through the Banking System - 4.8 - 5.2 - 5.0 - Cash Balances (money) - 0.9 - 1.6 - 1.3 - Quasimoney - 3.9 - 3.6 - 3.7 Inflation Tax -/ - 0.3 - 1.4 - 0.9 a/ All figures deflated by the GDY implicit deflator b/ Annual average growth rate for 1966-70 c/ Annual average growth rate for 1966-76 d/ Defined as money times the inflation rate, see footnote 2, page 42. Source: Staff estimates.  Table 13: MEXICO: NIfESTMENT ANDE ITS FTNMANCING 1/ (billions of 1972 Mex$, units of GDY) 1970 1971 1972 1973 1974 1975 1976 1977 Total Fixed Investment 91.01 86.98 97.80 112.43 125.41 137.24 136.10 126.11 Private Investment 58.71 59;91 64.50 68.09 78.55 77.94 81.25 74.10 Public Investment 32.30 26.17 33.3 44.34 46.86 59.30 54.85 52.01 Consolidated Public Inv. 2/ 19.61 20.58 25.26 33.58 36.64 50.68 46.03 42.53 PEMEX 2.69 4.99 4.72 6.27 6.66 8.62 12.15 14.34 Non PEMEX 16.92 15.59 10.54 27.31 29.98 42.06 33.88 28.19 Non Consoiidated 3/ 12.69 5.59 8.04 10.76 10.22 8.62 8.82 9.48 Total Financing 91.01 86.08 97.80 112.43 125.41 137.24 136.10 126.11 External Savings 14.75 11.03 11.44 15.72 25.95 32.33 26.62 16.94 Public Consolidated Savings 20.04 17.51 18.70 12.76 10.87 11.85 16.03 20.40 PEM'.X 4.72 5.49 7.29 7.81 8.52 10.70 12.02 13.72 Z. Non PEMEX 15.32 12.02 11.41 4.95 2.35 1.15 4.01 6.68 Private Savings 4/ 53.96 55.19 64.76 76.70 75.32 83.00 79.83 68.44 Money 2.46 1.32 4.27 5.76 0.98 2.74 0.08 -3.64 Non Money 51.50 53.87 . 60.49 70.94 . 76.30 80.26 79.75 72.08 Inflation Tax 5/ 2.26 2.35 2.90 7.25 13.28 10.05 13.61 20.33 Memo Item 6/ Relative price of investment 1.010 0.999 1.OQO 0.992 1.019 1.044 1.097 1.100 1/ Current value of investment deflated by the GDY implicit price deflator. 2/ Our consolidation of the public sector does not include the non budgetary controlled Public Enterprises. 3/ Should correspond to the non budgetary controlled Public Enterprises. 4/ It includes the savings of the non budgetary controlled Public Enterprises. 5/ Equals to inflation (increase in the GDY deflator) times the real -uantity of money. 6/ Invest,.ent deflator divided by the GDY deflator. If the values in this Table are divided by this item, the value become expressed in investment units of 1972. Source: Staff estimates.  - 45 - 114. Let us now examine the causes of the decline in public savings and its future prospects, the nature and potential of private savings, and the implications for domestic savings of recent past massive foreign borrowings. Public Sector Finance 115. The relatively low tax ratio, the large increase in general Govern- ment's consumption expenditures and the rapidly deteriorating financial situation of public enterprises are the major factors behind the aggravation of the already weak public savings performance in recent years. Tax Revenues 116. Thie steep decline in overall public sector savings came about in spite of rapidly growing Federal Government revenues, particularly tax revenues. From 1970 to 1976, net tax revenues (net of all transfers to states and other subsidies) rose about four-fold; over this same period, GDY almost tripled: the implied revenlue elasticity was therefore well over 1.3 (Annex IV, Table 5.6). 117. Mexico's tax ratio was about 14 percent of GDY in 1978. Two factors contributed to keeping it at this level: the inequality in the distribution of income, and the very low ratio of imports to GDY (less than 10 percent). The Government has made significant efforts to increase the tax ratio, and succeeded in achieving marginal rates of 25 percent on average for the period 1972-1977. It is expected to reach close to 17 percent of GDY in the next few years partly owing to oil exports (PEMEX's taxes will increase substantially). Furthermore, the increase in foreign exchange availability will allow imports and related import taxes to increase. 118. In spite of the moderate tax ratio, the effective tax rates falling on some sectors and sources of income are very high; indeed, higher than their counterpart rates in tax structures of many developed countries. Personal income tax rates rise to 57 percent, 1/ while the sales tax embodies rates as high as 30 percent on sumptuary and luxury items. The reasons for the low yields of Mexico's tax system lie with the narrow base on which the rates are levied. Anachronistic tax preferences to sectors which no longer require such special treatment, and extraordinarily high rates of avoidance and evasion reduce the de facto tax base, even while the statutory base is adequate. 119. In addition to its inequitable structure and inadequate revenue genera.ing capacity, the Mexican tax system also distorts the relative costs of capital and labor. The impact of fiscal investment incentives and acceler- ated depreciation on the one side, and social security taxes and payroll taxes on the other, have until recently distorted the ratio of the wage bill to the cost of capital. The 10 percent investment credit is another element reducing capital costs relative to wages. Policies exempting or reducing import duties on capital goods are also in this class. The Government has already undertaken steps to modify the incentives system so as to correct some of the bias. 1/ Naominal rates rise to 50 percent; the effective rate, however, goes up to 57 percent.  - 46 - 120. Large number of expected new labor force entrants make employment an increasingly urgent priority in Mexico. Fiscal incentives directed at the creation of new jobs have been, until recently, nonexistent. In fact, the stimuli had the aim of developing Mexico's industrial base by stimulating new investment and new industries. Their effect has uniformly been to dis- courage labor-intensive enterprise. With growing unemployment, these side effect can no longer be ignored. New instruments will have to be designed which encourage labor-intensive production techniques. In this category come multiple-shift and other employment-generating operations, and promotion of firms buying products of small-scale industry, and those providing services-- technical assistance, marketing facilities, and the like. 121. Specific failings in the present fiscal system with respect to resource allocation are detailed below. 122. Payroll Taxes. In the short run, there is probably little that can be done about social security taxes. 1/ Since the system's coverage is so narrow, general revenue financing of all benefits would be unjust. A long run strategy of extending health benefits, financed by tax revenues, to the popula- tion at large, while retaining payroll taxes for pension schemes, has our recommendation. It would be advisable that the use of other payroll taxes, such as the 1 percent and 5 percent (INFONAVIT) levies earmarked for education and worker housing respectively, be discontinued and replaced with financing schemes that do not increase the wage bill. 123. Investment Incentives. Fiscal incentives for "new and necessary industry" (1954) and for "decentralized industrial development" (1972) 4Lave biased capital intensity while their impact on industrial growth is unclear. Their dubious success as incentives, and the revenue sacrifice they imply, alone make them prime candidates for elimination; an even better case to end them can be made in the light of their employment impact. Both the investment tax credit and the accelerated depreciation provisions in the corporate income tax code should also be re-examined with the country's employment objectives in mind. 124. Inflation has generated other distortions: the fiscal system was designed in, and presumably for, a period of relative price stability. For firms, it is the use of historical cost, rather than replacement cost basLs for depreciation and inventory valuation which provides most cause for concern. This leaves firms with inadequate sinking funds, and squeezes replacement investment. The devaluation has also increased the prices of capLiat goods. Some provisions for asset revaluation are necessary if, as is expected, inflation continues in the next few years. 1/ The social security taxes paid by the employer range from 13 percent-22 percent of the wage bill. If to this is added the other payroll levies, the total employer portion may range from 19-28 percent. Total contri- butions would average 25 percent of the wage.  - 47 - 125. On the side of the personal income tax, inflation reinforces the system's regressivity and horizontal inequity, since it is primarily the wage and salary earners, the captive taxpayers, who are affected when their higher nominal wages are taxed at higher bracket rates. For the self-employed, in contrast, the real burden of the income tax during inflation is reduced, since their taxes are due at the end of the fiscal year. Recent legislation has increased the deduction of self employed from 5 percent of gross revenues to 20 percent of net revenues and has revised the limits of brackets to allow for recent high inflation in the country. 126. From the government's point of view, the low provisional or estimated periodic payments of both corporate and self-employed income taxes (with the balance remitted at the end of the year) frustrates the capture of a given level of real resources via the fiscal system. 127. Elasticity. In view of the Government's resource needs, the most important shortcoming of the tax system is its inability to generate the required revenues adequate to meet the needs of rapidly growing expenditure programs, and indeed, much of the increase in revenue which has taken place is due to rate increases rather than an inherently elastic tax design. The main culprits here are the limited progressivity of the personal income tax, and the excise taxes, many of which are specific. The recent change in the personal income tax law broadens and globalizes the income tax base, and should result in increased elasticity: Ad-valorem rates should replace the present specific ones in the excise system. 128. Equity. The overall incidence of the Mexican tax system is diffi- cult to determine. The income tax itself appears broadly regressive (capital income is undertaxed, and "captive taxpayers" bear a disproportionate tax burden). Unearned income gets off lightly in part because of the predominance of capital ownership by means of bearer titles which militates against pro- gressive taxation of capital income, and makes the taxation of capital gains quite impossible. By thwarting the aggregation of all taxpayers' income from all sources, the "anonimato" precludes the introduction of a truly global income tax. 129. Wage and salary earners are "captive" taxpayers, as taxes on employees are withheld monthly. They are subject to a progressive schedule of rates, and unlike their professional colleagues (who can reduce their effective tax burden through evasion) have no means of evading the income tax (inflation also imposes a heavier burden on them than on any other group). The direct tax burden is thus heaviest on the wage and salary class--the middle 50 percent or so of the income distribution. 130. The predominance of indirect taxes in the tax structure need not imply that it is regressive, if the exemptions are such that they leave items of popular consumption untaxed. If surtax rates are applied on luxury consump- tion goods, or on services, as they are in Mexico, some degree of progression can even be achieved (with respect to consumption, rarely with respect to income). The bulk of consumption by "the poor," the lowest 30 percent or so, is non-market consumption: consequently, the federal sales tax does not reach this group. However, a study 1/ shows that the federal excises, especially I/ Reyes Heroles, J., Po:Litica Fiscal y Redistribucion del Ingreso, 197b.  - 48 - those on beverages, beer and tobacco weigh very heavily on them. 1/ The percentage of family income paid in excise taxes by the lowest 5 percent of families is about 3.5 times the percentage paid by the highest. 2/ 131. Evasion. Tax evasion is a problem of very large dimensions: it has been estimated that the Treasury collects less than half of the revenues due it. Poor administration is in part the cause for this, but the failure to write any "internal controls" into the tax system is equally to blame. Evasion occurs on the largest scale among the self-employed, especially professionals, via non-payment of the commercial receipts tax, and within the corporate sector, via inflated deductions. It is the "special tax regimes" (for agriculture, transport and the construction industry) which facilitate the latter. Since these sectors are taxed at much lower (and proportional) rates than those of the corporate income tax, the tax liability on a given increase in income in those sectors is less than the tax saving for a corporate income taxpayer who claims transport or construction expenses as a deduction. The enormous traffic in falsified receipts testified to the ubiquity of this technique. Even quite aside from their role in facilitating tax evasiop-- reason enough to eliminate them--the special regimes have little relevance now that all sectors are modern and commercial and could be brought into the corporate income tax system. 132. Excise Taxes. The system of excise taxes is unsatisfactory because of its administrative complexity: there are simply too many levies. Of the 48 taxes, only 8 have any revenue significance (these 8 generate 87 percent of excise revenues); the remainder are largely anachronistic, sometimes nuisance levies. 133. Revenue Sharing. The Federal Government has preempted revenue sources commonly reserved in other countries for the states or municipalities, leaving these with a rather weak tax base. Both the sales and the excise taxes are federal taxes, and although the states receive a share of these receipts, 3/ there is growing dissatisfaction with these revenue-sharing arrangements. They have plainly exacerbated regional differences, providing more federal funds for richer states than for poorer ones. New sharing agreements will prevail from 1980 for those states willing to join the National System of Fiscal Coordination. However, the new arrangements will not result in an immediate increased in states income, although they will prevent further deterioration. 1/ The excises on the use of automobiles and electric energy are exceptions; both are progressive over most of the income range. 2/ The lowest 5 percent pay 2.0 percent of their income, the highest 0.59 percent. 3/ The states receive 45 percent of sales tax revenues collected in the state, and a variable proportion of excise revenues.  - 49 - 134. The states rely heavily on the property tax, and on other minor taxes, neither of which are fertile, much less elastic, revenue sources. If increased state autonomy and decentralization are indeed goals, state finances will have to be strengthened. Changed revenue sharing provisions are only a first step. 135. The authorities are aware of the many deficiencies of the present tax system and have been working on alternative reform proposals over the past few years. New legislation has been introduced that will be enforced in 1979 and 1980. Income tax has been partially globalized, the tax rate for the different brackets has been reduced to account for the inflationary drag in recent years, and an amount equivalent to the minimum wage is now totally exempted. A consumption type value added tax has been approved starting in 1980 at the initial rate of 10 percent. The tax will not apply to non- processed agricultural production nor to fertilizers and agricultural equip- ment. 136. Strong political opposition together with the new petroleum riches may render an in-depth fiscal reform very difficult. The highly necessary reform of the "special bases" seems to be a particularly sensitive issue because it would affect strong interest groups, i.e., transport, construction and agriculture. However, a strengthening of the administration process and a major revision of indirect taxes will probably take place. Under the circum- stances, this should be considered as a minimum, if the Government wishes to achieve its targets. 137. The introduction of the value-added tax, is expected to eliminate the inefficiency of the cascade aspects of the commercial receipts tax and improve internal controls. The staff thinks that the following modifications also deserve Government's attention. (a) The taxation of the self-employed in the professions is in urgent need of administrative scrutiny, so that evasion rates can be reduced. A crackdown here would also improve the tax-paying climate, since conspicuous evasion increases the cynicism of other groups of taxpayers. Also, the distortions in this tax caused by inflation deserve more attention than they are getting. (b) In the personal income tax, both the self-employed and wage earners should be subject to identical withholding rates, estimated payment schedules and periods. This is far from complete globalization, but would at least reduce the grossest distortions of the present cedular system. (c) The corporate income tax should permit an alternative to historical cost depreciation, allowing assets to be revalued for tax purposes, either automatically or at discrete intervals. 138. Finally, it woulcl be advisable to review the tax treatment of PEMEX in order to reduce the amount of unallocated resources available to PEMEX and to reduce the incentives PEMEX currently has to overinvest in refinery capacity and,petrochemicals plants. According to our projections, PEMEX savings could be as high as 7.2 percent of GDY by 1982, in addition to projected tax payments of about 2.4 percent of GDY. Given the difficulties  - 50 - the Government may have in exercising control over funds internally generated by PEMEX, the Government could secure its command over the major part of these savings by taxing them away. It should be noted that we are already assuming that all petroleum and related product exports would be subject to the same tax treatment as crude oil exports, i.e., that they would be taxed at a 50 percent rate. 139. An alternative to the current system of different tax rates on the returns from PEMEX's sales of different products (50 percent on crude exports, 16 percent on petrochemicals and 12 percent on refined products) would be to tax progressively the differential between the sales price of the different PEMEX products and the respective production cost. This alternative scheme would require the government to fix periodically the production costs that would be used as a basis for the application of the tax, but once the production costs have been agreed upon PEMEX would have an incentive to reduce costs. Additionally, this new tax scheme would assure that a major share of PEMEX potential savings would be transferred to the Federal Government, and it would make PEMEX investments and product mix decisions neutral with respect to price and tax differentials among markets and products. Government Expenditures 140. General Government current expenditures grew at an annual real rate of 16.4 percent during 1971-1976, while its components of consumption, transfer and interest payments grew at 12.3 percent, 28.9 percent and 14.7 percent per year, respectively, during the same period. 141. Table 14, page 51 summarizes the behavior of consumption expendi- tures during the last sexenio. Wages and salaries were the fastest growing component of consumption expenditures. This was the result of a rapid increase in the number of government employees and a somewhat more moderate increase in real wages. The number of public sector employees increased at more than 9 percent per year while real wages and salaries increased by 3.8 percent per year (Annex IV, Tables 5.3 and 5.7). 142. Federal Government current transfers were the fastest growing com- ponent of current expenditures during 1971-1976 and was one bf the mainsprings behind the poor General Government savings performance. Table 15, page 52,, shows the development of transfers over time and breaks them down by major components. 143. The enormous increase of transfers to trust funds in general and to agricultural financial entities in particular appears to be the main cause of the acceleration of extra-system transfers. There are a large number of trust funds but there is little knowledge if any on what the differences are in their functions and, justification for their existence. The staff was told that many of these trust funds would disappear as a consequence of the administ- rative reform now underway. Certainly this is an area in which action is urgently required.  Table 14 MEXICO: GENERAL GOVERNMENT CONSUMPTION, 1971-1976 (real annual rate of growth) 1971-73 1974-76 1971 76 Consumption Expenditures 11.5 13.1 12.3 Wages and Salaries (12.4) (14.3) (13.3) Purchases of Goods and Services ( 8.5) ( 8.1) ( 8.3) Interest Payments 6.7 23.3 14.7 Current Transfers 51.6 9.6 28.9 Other Current Expenditures ].1.0 12.1 11.6 Total Current Expenditures 20.4 12.6 16.4 Source: Staff estimates.  Table 15: MEXICO: FEDERAI, COVERNMENT CURRENT TRANSFERS TO PUBLIC SECTOR ENTITIES, 1970-76 (millions of Mex$) 1970 1971 1972 1973 1974 1975 1976 1977 Current Transfers: 7,420 9,870 13,220 28,380 42,54o 64,620 67,340 94,900 Intrasystem 2,460 2,870 6,210 5,570 12,76o 16,860 13,940 26,410 Extrasystem 4,960 7,000 7,010 22,810 29,780 47,760 53,400 68,490 Agricultural Trust Funds-/ n.a. (453.0)(2,406.o)(2,850.0)(2,729.0)(6,090.0) n.a. n.a. Other Trust Funds n.a. (262.7) (384.9) (873.2)(2,339.3)(2,585.9) n.a. n.a. Agricultural National Banks n.a. (886.0)(1,706.0)(2,639.0)(4,109.0)(6,370.0) n.a. n.a. Sub-total (2+3+4) n.a. (1,601.7)(4,496.9)(6,362.2)(9,177.3)Q5,045.9) - /n.a. n.a. Other Public Entities n.a. (5,398.3)(2,513.-1)6,447.8) o,602.'OG2,714.1) n.a. n.a. UNAM (National Univers.) n.a. 797.0 987.0 1,263.0 1,747.0 2,631.0 n.a. n.a. Other Educational n.a. 333.0 497.0 1,201.0 2,057.0 2,993.0 n.a. n.a. entities Health Entities n.a. 383.0 498.o 609.0 1,092.0 1,837.0 n.a. n.a. a/ Cuenta de la Hacienda Publica Federal 1975, page 19 b/ Mexican 5.5 billion of Rural Bank debt assumed by the Federal Government is not included- Source: Banco de Mexico, Cuenta de la Hacienda Publica Federal and mission's estimates  - 53 - 144. The subsidies to the agricultural national banks (now merged into a single institution, BANRURtAL) have continued to grow and are estimated at Mex$12 billion in 1977. About 40 percent of this amount is accounted for by BANRURAL's operating expenses (which are not covered by revenues), and the rest represents losses on loans in default. It is not clear who these subsidies benefit, or what their impact is on production, if any. There is not enough information as to whether these subsidies are justified or not. There is no evidence as to what criterion they meet or are supposed to meet. 145. The key initiative taken by the new Administration to strengthen its ability to manage the public sector has been a far-reaching administrative reform. The measures adopted include a new law giving the Treasury more effec- tive control over public sector foreign borrowing, another law to improve and modernize the budgetary process and to centralize it under the new Secretariat of Programming and Budgeting, and the centralization of direc- tion of government programs related directly to production activities in two ministries, the Secretariat of Agriculture and Water Resources and the Secretariat of National Property and Industrial Development. Still another element is the intention to strengthen the legal power and the capacity of state governments to plan and undertake expenditures in certain areas where centralized control may be less efficient or less responsive to varying local needs. 146. Yet another important improvement that the Government hopes to achieve with the administrative reform is a better review of priorities in public sector investments. The centralized review of all proposed investments in each broad sector (e.g., industry, agriculture, communications) by one relevant ministry may be the most important change in this direction. As an example, the three public sector owned steel companies have been merged into a large holding company which has been charged with reviewing the supply and demand situation for the entire sector, both public and private, and devising an investment program that would, among other things, reduce temporary excess capacity at individual facilities by means of nation-wide coordination and sharing of production facilities at different stages in the manufacturing process. 147. The emphasis on rationalization and the necessary fiscal restraint called upon by the stabilization program led to a very small real growth of Government consumption in 1977. However, the Government's commitment to launch an effective attack on poverty and to reduce the large gap between demand and supply of basic needs should necessarily lead to a rapid growth of public sector consumption. Public Enterprises 148. Intra-system transfers to budgetary controlled enterprises also increased markedly during 1971-1976. However, subsidies to the "traditional" enterprises, the railroads and CONASUPO, which amounted to more than 95 percent  - 54 - of total intra-system transfers in 1971, have declined to about 70 percent of the total, the difference being mainly accounted for by increasing subsidies to the manufacturing and electric power sectors. 149. The most striking feature of public sector enterprises behavior (excluding PEMEX) during 1971-1976 has been the rapid deterioration of gross operating surplus, which has become increasingly negative during the period. Public sector transfers have not been enough to compensate for these operating losses, the difference being financed by federal government capital participa- tion and borrowing. Lack of flexibility or willingness to adjust prices to inflation, and rapidly increasing wage bills and interest payments explain this trend in almost all the cases (Annex IV, Tables 5.8-5.18). In the case of the railroads and CONASUPO, other factors were also important. 150. More than half of the railroad's deficit is due to the operation of uneconomic passenger services. The rest of the deficit is explained by the above-mentioned reasons plus the fact that the railroads are facing strong competition from an efficient and heavily subsidized road transport industry. The first problem can be resolved by cutting back or eliminating rail passen- ger service which is no longer of any real national importance (6 percent of intercity passenger kilometers in 1973). The second problem can only be solved by reducing the subsidies currently granted to the road industry, by eliminating tax holidays, and setting economic prices for diesel fuel and appropriate road user charges. Implementation of these measures would simultaneously reduce the railroad's deficit and increase other Government revenues. 151. Subsidies to CONASUPO are mostly explained by the Government price support policies. It is estimated that in 1977 maize support prices alone implied a subsidy of about Mex$800 million (1 million tons multiplied by the difference between domestic support prices and international prices, estimated at Mex$800 t/ton in 1977). As discussed before (paras. 92 and 93), the justi- fication of this subsidy is extremely doubtful. Subsidization of "tortillas" and other essential goods and services added an estimated Mex$3.5 billion to this in 1977 and according to President Lopez Portillo, they were likely to amount to Mex$7.4 billion in 1978. Although these type of subsidies can indeed be justified on equity grounds there seems to be ample room for improv- ing its administration and reducing its costs by better targeting the benefi- ciaries. 152. As shown in Table 16, page 55, low energy prices -- of electricity and petroleum products -- accounted for a large share of Government subsidies in 1977. The Government has traditionally regarded concessionary pricing of electricity and petroleum as an instrument of industrial promotion. Electricity tariffs remained constant from 1962 to 1973, except for a 4.5 percent increase in charges to industrial and large commercial consumers in 1966. At the end of 1977, after successive price adjustments, the average price charged by CFE to electricity consumers was Mex$0.63/kwh; in real terms this corresponds to about the 1971 level. According to our staff calculations, the average long-run marginal cost of electricity in Mexico--assuming that the price of fuel oil paid by CFE will remain at its current subsidize level-- would appear to be about Mex$0.79/kwh, i.e., 25 percent higher than the  - 55 - average tariff in 1977 (Annex III). Since demand for electricity was of about 40.4 GWH in 1977, the implicit subsidy amounted to Mex$6.5 billion, or equiva- lent to 30 percent of CFE's investment in 1977. Even this higher rate of Mex$0.79/kwh is low by international standards and would go up to Mex$1.20/ kwh if the fuel oil consumed by CFE were priced at its opportunity cost (assumed to be the international price of bunker C ex Aruba). The current low tariffs, in addition to aggravating the already weak financial position of CFE, lead to an excessive demand for electricity and as a consequence to higher investment requirements. 153. Price differentials between domestic and international markets are even greater in the case of petroleum products. Annex IV, Tables 7.14 and 7.15 gives a comparison of international bulk prices for petroleum fuels in Mexico as compared with prices in the Caribbean area and the estimated subsidy to the Mexican economy implicit in these low fuel prices. The staff has estimated that in 1977 the effective subsidy amounted to about Mex$900 per capita. Table 16: GOVERNMENT SUBSIDIES FOR INDUSTRIAL AND AGRICULTURAL PRODUCTION, 1977 (billions of Mex$) As % of Public Sector Capital Subsidies Account Deficit GDY Energy 72.1 70.2 4.3 - PEMEX 65.6 63.9 3.9 - CFE 6.5 5.2 0.4 Transport 2.5 2.0 0.2 - Railroads 2.5 2.0 0.2 Agriculture 31.4 25.1 1.9 - Large Scale Irrigation 6.0 4.8 0.4 - Small Scale Irrigation 1.5 1.2 0.1 - Price Supports (maize and fertilizers) 2.6 2.1 0.2 - BANRURAL 12.0 9.6 0.7 - Trust Funds 9.3 7.4 0.5 Total 106.0 103.2 6.4 Source: Staff estimates  - 56 - 154. In addition to the price differentials between domestic and inter- national markets, Table 14 in Annex IV also shows that there exists a marked price distortion affecting the relative prices of gascline and diesel fuel, which to a large extent are substitutes in consumption. This price distortion has led to a massive shift from gasoline-engine vehicles to diesel-engine vehicles, to a deterioration of the competitive position of the railroads and to an excess demand for medium distillates. 155. The Government is reviewing the hundreds of semi-autonomus public sector enterprises, from the largest to the smallest, to determine whether any inefficiencies that are found can be remedied or, if not, whether the enter- prises should be dissolved. In his last State of the Union speech, President Lopez-Portillo reported that 12 mergers of public enterprises had taken place and that 70 of these enterprises had been dissolved over the 'ast two years. 156. Perhaps the most important aspect of this review relates to the pricing policies of these enterprises. Although a few public enterprises may have to operate to a loss in order to fulfill their objectives, in recent years many other public firms have allowed -- or have been forced by the Government to allow -- their prices to fall well below costs. The resultant overall deficit has been an important contributor to inflation and has put serious strains on Mexico's tax, credit, and external borrowing systems as these other instruments had to compensate for the effects of the deficits of the enterprises. 157. The authorities are fully aware of this problem but have been reluctant to adjust prices quickly fearing that by so doing they would jeopardize the stabilization effort. Nevertheless, at the end of 1976 and early in 1977 some price increases were approved (for example, 52 percent in electricity, 25 per- cent in petroleum products, 15 percent in steel, 30 percent in telephones), and more recently, a comprehensive package of measures aiming at improving CFE's financial conditions was approved. This package was introduced in July 1978 and included a tariff increase of about 1.5 percent per month to be applied over a period of 24 months, an increase in labor productivity to be obtained by slowing down the rate of growth of CFE's labor force to a maximum of 3.5 percent per year and the assumption by the Government of a large share of CFE's external debt. 158. The CFE approach could be used as an example to be followed in some other cases, as by example, the railroads and FERTIMEX. Prompt attention should also be given to the prices of PEMEX products, not only because of the many distortions they are inducing, but also because this is the single most important source of revenue at hand, revenue that could be used among other things, to pay for subsidies where justifiable. 159. The most appropriate remedy for the highly subsidized price of diesel fuel would be gradually to increase the price, so as to approach a situation in which both regular gasoline and automotive diesel fuel are priced in proportion to their calorific values. Urban bus transport could be subsidized directly, if necessary, to compensate for the increased cost of diesel. Premium gasoline could continue to be treated as a luxury product.  - 57 - 160. The adjustment in domestic relative prices could be combined with a gradual reduction of the price differentials compared to international markets by adjusting the domestic price of middle distillates at a higher rate than the prices of gasoline. This reduction in price differentials is not only necessary for revenue purposes but also because the low domestic prices (a) have inhibited the development of alternative sources of energy other than hydropower and have reduced a once flourishing coal industry to the status of an accessory to the steel industry, and (b) have acted as an stimulus to PEMEX to become involved in more and more manufacturing processes using its own products as raw materials, since the end products can be sold at prices close to import prices, while the inputs are priced consider- ably below their import equivalents. 161. From the preceding discussion, it is clear that one of the causes behind the poor savings performance of the public sector are the massive subsidies that directly or indirectly are granted by the Government. In order to illustrate this point we made a rough attempt to estimate the subsidies granted for industrial and agricultural promotional purposes. Taking only the items shown in Table 16, page 55, subsidies for 1977 amounted to Mex$106 billion. This is equivalent to about 6.4 percent of GDY and slightly lower than the budgetary controlled public sector deficit. 162. By far the largest single item is the sale of petroleum and products at below-world-market levels by PEMEX. Also large are losses in agriculture- related activities. These subsidies not only cost the Government an enormous amount of money, and thus put additional pressures on economic policy instru- ments to reduce the public sector deficit, but the subsidies also may be harmful in the distortions they induce throughout the economy. Cheap fuel induces to consume it in relatively larger quantities in Mexico than in other countries at similar stages of development; the fact that petroleum-based fuels are cheap has resulted in almost total neglect of other sources of energy. In agriculture, cheap water is used to grow low-value crops which could be grown almost as well in rainfed areas. 163. In spite of the amounts involved, the calculation is far from being comprehensive. Tax rebates and import tariff exemptions -- which together amounted to Mex$4.4 billion in 1977 -- special tax treatment for agriculture, transport and construction, preferential credit schemes, low road-user charges, and the popular consumption goods program (Mex$3.5 billion and Mex$7.4 billion in 1977 and 1978, respectively) are other channels currently being used by the government to subsidize the private sector even more. 164. The distortions in the allocation of resources throughout the economy induced by this array of subsidies are enormous, but what may be worse is that it is by no means clear if these subsidies are serving the purpose they were instituted for, if any. It is the staff's belief that it would be worthwhile for the Government to embark upon a comprehensive analysis of the country's current incentive schemes in order to reshape them in the light of the priorities defined by the new Administration, particularly the urgent need to increase employment opportunities and the productivity of the poor.  - 58 - 165. The amounts involved are such that any effort to reduce these subsidies will certainly have a large pay-off in terms of additional public sector savings. Additional resources are necessary not only to compensate for the existing public sector gap but also to pay for the new investmens and poverty and employment oriented programs the current Administration is committed to carry out. As is discussed later, the increase in revenues brought about by the large planned increases in production and sales of hydrocarbon resources may not be enough to finance these additional expendi- ture programs unless a further effort is made to increase revenues and reduce subsidies. Private Savings and the Capital Market 166. Private sector savings have generally amounted to 75 percent of gross national savings in Mexico but only about 10 percent of it is captured by the financial system (Table 17, page 59). Little is known about the composition of these savings but the evidence available suggests that corporations' internal cash generation is its major component. The only data readily available on the flow of funds of Mexican enterprises are from a study on the capital goods industry done by NAFINSA, which shows that 30.3 percent of the financial requirements of these enterprises were financed from internal cash generation and equity during 1971-1975, the other sources of financing being credit from other enterprises (23 percent), foreign credits (16 percent) and credit from the domestic banking system (16 percent). 1/ 167. The limited role of domestic credit in financing Mexican enterprises is a reflection of the relatively small share of total private savings that is being channeled through the financial markets (Table 17). Furthermore, an important share of these are short-run deposits of enterprises. These phenomena together with the almost complete absence of a securities market and other sources of domestic long-term financing, reinforce the hypothesis that most of private sector investments are financed through internal cash generation, mainly undistributed profits and depreciation allowances. 1/ See NAFINSA-UNIDO; Mexico: Una Estrategia para el Desarrollo de la Industria de Bienes de Capital, Mexico, D.F., 1977, Table VI-14, Page 342.  - 59 - Table 17: MEXICO - PRIVATE SECTOR SAVINGS (% of GDY) Average Average 1965-1970 71 72 73 74 75 76 1971-1976 National Savings 16.6 15.7 16.8 17.6 16.9 17.1 17.7 17.0 Private Sector Savings 13.3 12.0 12.6 12.4 13.4 12.7 13.1 12.8 Channeled through Financial markets 3.9 4.5 3.0 1.0 4.4 .7 2.9 - Money* (0.9) (0.8) (1.4) (2.4) (2.4) (2.1) (2.2) (1.9) - Quasi-money (3.1) (3.1) (0.6) (-1.4) (2.3)(-1.5) (1.0) Other 8.1 8.1 9.4 12.4 8.3 12.4 9.9 *It includes the inflation tax. 168. The development strategy of the 60's was based on a combination oE high private sector profits and Government supportive action. The former was supported by the protection available to industry through quantitative restric- tions on imports during all this period, the extremely favorable tax treatment of income from capital and the inequality in income distribution, these factors contributed to abnormally high profit margins in Mexico, the major source of private savings. 169. The commitment of the current Administration to improve income distri- bution, the shift towards more efficient industrial production, the need to open the economy associated with the efficiency drive and the increasing petroleum exports, should all lead to the reduction in abnormally high profit margins and thus may cut into private savings unless new savings channels are created. In this context the strengthening of Mexico's financial market becomes crucial. 170. Contrary to common opinion, Mexico's financial markets are not highly developed; financial interme!diation in Mexico is, in fact, only slightly above the level of the average developing country. Mexico underwent a process of fairly rapid financial deepening through 1972, when the ratio of the combined assets of financial institutions to GDP reached a level somewhat in excess of 50 percent. Since then, this ratio has remained roughly constant. In addition, the volume of securities trading has been equal to only one half of one percent of GDP, a fairly low ratio for an advanced LDC. Table 17 shows that institutional development as measured by the ratio of financial institu- tions' assets to GDP is noticeably higher in many of the more advanced LDCs, and that Mexico's ratio is more characteristic of the relatively less developed LDCs.  - 60 - Table 18: MEXICO - RATIOS OF TOTAL ASSETS OF FINANCIAL INSTITUTIONS TO GDP 1/ Ratio of Total Financial Assets to GNP 1973 1976 South Korea 156 153 Venezuela 97 152 Brazil 107 143 Yugoslavia 153 131 Jordan 103 115 Argentina 93 111 Egypt 67 105 Honduras 83 98 Greece 71 84 Turkey 72 79 Jamaica 64 78 Philippines 63 70 Ecuador 59 69 Thailand 64 68 Indonesia 51 62 Mexico 52 56 Sri Lanka 54 51 Colombia 52 48 Bolivia 51 46 Sudan 40 45 Chile 45 31 1/ There is a certain degree of incomparability between these statistics, given varying definitions of financial institutions and of public sector agencies. Various institutions such as social security and public sector housing agencies are hybrid institutions. This, plus omissions in data collection, make such comparisons difficult. In general, the statistics on financial institutions collected by the Bank of Mexico exclude all such hybrid institutions. Though they are extremely important in many countries in this table, they are less so in Mexico. The combined assets of the Mexican social security system, the most important of these institutions, are equal to only 4 percent of GDP. The combined assets of all other institutions appear to have been less. Overall, we estimate that the inclusion of those institutions would raise these ratios by no more than 10 percentage points at the most. As the other ratios in this table tend to suffer from errors of omission as well, such a modification does not alter the relative position of Mexico vis-a-vis other developing nations.  - 61 - 171. The Mexican financial system is relatively simple and is basically organized around banking type institutions. The securities market is the most underdeveloped sector of the Mexican financial market. Although securities trading in Mexico has increased very rapidly in the last two decades from an almost negligible base, the volume of all securities, both stocks and bonds, traded on the Mexican stock exchange remains quite low. In 1976, the ratio of the value of shares traded to GDP was equal to only one half of one percent compared to ratios of 3 percent or more in some developing countries where significant securities market development has taken place (e.g. Brazil and Korea). 172. Most of the trading activity in Mexican securities occurs in shares rather than in debt securities. Share trading is concentrated in only a handful of the 590 issues registered on the exchange. In 1976, one company, Telefonos de Mexico, accounted for 23 percent of total trading activity in common shares and the 20 most actively traded shares accounted for 68 percent of exchange volume. In the debt securities market, the bonds of Telefonos de Mexico, which are issued on a compulsory basis on installation of a telephone, dominate trading activity. There is some informal or non-institutional securities market activity in Mexico, but it is surprisingly limited in extent. 173. The relative underdevelopment of the Mexican securities market is attributable to several factors: (a) In the past, finance companies and mortgage banks stood ready to repurchase their long-term bonds at par at anytime. This practice allowed these institutions to pay long-term bond rates of interest on instruments which were basically sight deposits and made private long-term corporate bonds relatively non- competitive instruments. It also imposed a significant liquidity risk on these institutions which resulted in serious institu- tional problems in 1976; as a result the redeemable bonds are gradually being eliminated, thereby removing this obstacle to deeper bond market development. (b) A portion of the interest paid to savers on bank deposits (called the sobretasa) is now exempt from tax, whereas all interest paid on debt securities is taxable. This subsidy to bank deposits, together with an unwillingness by the authori- ties to authorize private short-term debt securities distribu- tions, has inhibited the growth of a short-term debt securities market. (c) There is some evidence that the commercial banking system regards the development of a debt securities market as a competitive threat which would reduce the volume of financing conducted through the banking system as well as exert competitive pressures on bank profit margins, Bankers have acted in the past to impede securities market development. The banks also appear to believe that there is a high degree of substitutability over the short run between debt securities issues and bank deposits and that large debt securities offerings could lead to disruptive deposit withdrawals.  - 62 - (d) As to the share market, the main impediments to development are largely institutional. Rates of return on Mexican equity capital are high and for shares registered on the stock exchange there is favorable tax treatment. The unwillingness of closely-held firms to open their capital and to meet minimum disclosure requirements, as well as the absence of a strong securities distribution system, are probably the main impediments to stock exchange development. 174. The reliance of the Mexican financial market on deposit liabilities has made the asset base on which domestic investment financing is built a very liquid one. The Mexican authorities are concerned about the lack of long maturities in the domestic financial market and have begun to stress the need for developing a long-term securities market of significant proportions. In 1977, PEMEX issued bonds (petro-bonos) and more recently there have been several issues of new floating rate medium-term bond issues by major private corporations. 175. Treasury bills began to be issued in January 1978. There were nine separate issues through May, with the last three accompanied by the retirement of the first three. At that time the value of outstanding bills exceeded Mex$ 20 billion. The interest rate is set by the Central Bank and quantities issued are less than those demanded by the market at that price. The majority of the bills have been bought by banks as a substitute for their free liquid funds which were formerly held on deposit at the Banco de Mexico. The Banco de Mexico used to pay interest (6 percent); now it pays no interest on these deposits. Eventually, the Government will move to an auction of bills. Maturities are now for 91 days, and are expected to be increased to 180 days in the future. Government hopes that eventually a wider government securities market will develop and that this will lead to the market for private securities. 176. In addition, to help promote securities market development, the Government in May 1976 instituted a special selective credit scheme whereby 4 percent of all deposit funds mobilized by the savings banks is to be made available to broker-dealers for financing underwritings, secondary market investors, and margin accounts of broker-dealer clients. Efforts to strengthen the securities commission are being made, and this institution is engaged in an active program aimed at regulatory, legal and institutional changes which could further the development of this market subsector. 177. One approach to developing the long-term subsector of the domestic financial market is to encourage the growth of long-term institutional port- folio investment by private and public pension and provident funds, invest- ment trusts, and private insurance companies. However, these institutions require a certain degree of portfolio liquidity and, consequently, usually require a market for domestic securities to properly develop their role as savings mobilizers. This, along with the fact that the Mexican securities market has undergone extremely rapid growth in the last decade, albeit from a small base, indicates that perhaps the most useful and feasible approach to developing a long-term financial market subsector is to emphasize the development of an organized securities market.  - 63 - 178. The development of a market for short-term and medium-term public and private debt securities should progress rapidly, as these instruments do not require the complicated infrastructure of bond markets. Because of their similarity to bank deposits and because of the wide spread between bank deposit and loan rates in Mexico, they could readily displace a portion of existing bank intermediation, and the development of this market should occur at a rapid pace if the authorities authorize this type of instrument. 179. As to long-term securities, these instruments constitute a unique mechanism for generating long-term finance, since secondary market liquidity allows for the immediate disposability of long-term assets, thereby bridging the gap between the short-term time horizons of savers and the long-term time horizons of investors, without requiring any intervention by the monetary authorities. Unfortunately, efficient long-term securities markets are exceptional and, where they exist, they have developed very slowly. In developing countries such as Mexico, accelerating this development requires that the vicious circle of illiquidity in which fledgling country securities markets are caught be decisively broken. Adequate secondary market liquidity in securities market requires that all of the components of a broad, deep and anonymous securities market be already in place. Many developing countries have realized that if a "big push" approach is adopted, whereby extensive institution building efforts are undertaken, thereby putting into place all of the components required for successful securities market operations, these markets would prove to be efficient enough to be self-sustaining. 180. This approach usually requires (a) a legal and institutional framework, including a securities commission and a stock exchange; (b) incen- tives for prospective purchasers of securities and for institutions dealing in securities; and (c) the creation of special institutions to accelerate institutional development. The Mexican Government has embarked on such an attempt; a securities commission was established and the stock exchange is undergoing a major modernization. Technical and financial support for these moves is of great importance, since the quality and reliability of information and the efficiency of trading and distribution facilities are all important for the efficiency and success of the securities market. External Savings 181. The increase in fixed investment that took place during the preceding Administration was entirely financed by additional external savings. The balance of payments current account deficit went up from an average of 2.4 percent of GDY in 1966-1970 to 3.6 percent of GDY in 1971-1976 (reaching a peak of 5.3 percent of GDY in 1975) which together with the massive capital outflows that took place, particularly during 1975-1976, led to a huge increase in foreign borrowing.  - 64 - 182. The total foreign debt of the public sector (including short-term) increased from US$4.3 billion in 1970 to about US$19.6 billion in 1976, with the obvious consequence of a heavy debt service burden. The debt service ratio on medium and long term public sector external debt amounted to 0.31 in 1976 and reached 0.47 in 1978. Similarly, the ratio of interest payments on total public sector external debt (including short-term) to total export (including worker remittances) was equal to 0.19 in 1976 and remained at that level in 1977. 183. Although there is no reliable data on total private sector foreign debt, the best estimates available put it at about US$6.2 billion at the end of 1976 (of which US$2.9 billion were short-term). On the basis of these figures, Mexico's total foreign debt reached US$25.8 billion at the end of 1976, and its overall debt service ratio went up to above 0.40. 184. Even after taking into account the expected large increase in petroleum exports, Mexico's overall debt service ratio will be slightly over 0.50 in 1979-1980 representing a huge claim on export earnings. Notwithstand- ing, our estimates show that Mexico's ample availability of hydrocarbon resources will allow the country to manage its existing debt service burden without jeopardizing its growth prospects; if not for the petroleum outlook the country would have been confronted with a exceedingly difficult situation in which the only way out would have been a drastic slow-down in overall economic activity with consequences that are not easy to visualize. 185. As discussed further, foreign savings could continue playing an important role in helping the country to finance its investment requirements, particularly during the next four to five years. Given PEMEX production and export programs for the sexenio and the Government's intentions to induce a higher pace of economic activity, Mexico will continue running a balance of payments current account deficit during the next four to five years. The size of the deficit -- and therefore the amount of external savings the country can accommodate -- will depend primarily upon the rate of growth of domestic expenditures, the speed at which the economy can be opened to international trade, and the exchange rate policy. 186. Mexico's foreign exchange needs could be financed either by borrowing abroad - using the hydrocarbon resources as collateral -- or by increasing production and exports of petroleum and related products. The decision as to the best mix of these two alternative depends upon many factors, among them Mexico's intentions to pursue a balanced growth strategy, availability of hydrocarbon resources, pace of domestic consumption over time, expectations regarding the future price of oil vis-a-vis the real cost of borrowing and foreign capital market constraints.  - 65 - CHAPTER IV GROWTH PROSPECTS Development Strategy: Past and Future 187. During the 60s Mexico followed an inward-looking, import substitut- ing growth strategy and was able to grow at an annual real average rate of about 7 percent. This strategy, based on a policy of high profits and cheap capital for the private sector, failed to generate enough employment and income opportunities for many Mexicans. The preceding Administration attempted a drastic change of strategy and tried to improve the living conditions of the poor but its excessively expansionary macroeconomic policies led ultimately to increase in unemployment and probably worsened income distribution. 188. The present Administration, aware of these experiences and of the serious social problems the country is facing, is moving toward designing a comprehensive program aimed at strengthening the political system, improving income distribution and the living conditions of the poor, increasing public sector efficiency and accelerating growth. Key pieces of this program would be: (a) the political reform, already underway aimed at broadening the system's political base by allowing the creation of new political parties and its full participation in the country's political life, (b) the administrative reform already implemented to a large extent, which seeks to rationalize the functioning of the Public Administration and increases its efficiency, (c) the Alliance for Production whic:h "would guarantee a minimum supply of socially and nationally necessary goods and services" by combining private and public sector efforts under mutually convenient and jointly agreed conditions; and (d) the economic reform, which would include a tax reform aimed at redistribu- ting income and increasing public sector revenues, a commercial (trade) reform aimed at improving efficiency, better management of public sector expenditures, rational pricing and cost control in public sector enterprises, new policies on wages, prices and profits, and promotion of private sector savings and investments. 189. The Government has defined two overriding medium-term priorities -- energy and agricultural development -- and a three-stage economic strategy. The first stage -- the stabilization period -- is over and has been devoted to overcome the serious economic problems the new Government inherited. The second stage -- 1979-1980 --- would be devoted to laying down the basis for attaining a high, equitable and stable rate of growth beginning from 1981. During this stage efforts would be made to overcome supply bottlenecks, increase agricultural production, improve the marketing system, increase the efficiency and competitiveness of the country's industrial sector and create the conditions necessary for a more egalitarian distribution of the benefits of growth. Accelerated and equitable growth is the characteristic of the third, concluding stage of the period. 190. Mexico has the human institutional and natural resources necessary to successfully implement such a program. The short-run disequilibria that have affected the country during the mid-70s have been brought under control.  - 66 - The improvements, brought about in the past year or so and its effects on public savings and on the balance of payments become clear when 1970, 1975 and 1977 situation is compared. Non-PEMEX public savings represented '.3 percent of GDP in 1970 and only 0.2 percent in 1975. The current account deficit in the balance of payments (excluding oil exports) which was leLs than 3 percent of GDP in the years 1966-1970 exceeded 5.5 percent (excluding oil) in 1975. In turn, in 1977, the non-PEMEX public savings were 1.1 percent of GDP and the current account deficit of the balance of payments, excluding oil, was close to 4 percent of GDP. These improvements did not redress the situation back to its equilibrium of 1970; however, starting from 1980 it is expected that, partly thanks to the effect of oil income, both the public savings and the balance of payments would return to an equilibrium comparable to that of 1970. 191. The relaxation of the foreign exchange and public savings con- straints, is the most important change that took place in the Mexican economy in the recent past. This is a consequence of the country's new oil wealth. With it Mexico not only can export more, but because its credit- worthiness has improved substantially it can also rely on foreign borrowing. Our projections indicate that efficient exploitation of its actual and poten- tial hydrocarbon resources would enable the country to pursue a high and stable growth strategy over the medium-term. A high growth rate of public investment will be one of the propelling forces behind this growth. The corresponding financial requirements should be satisfied by a mixture of continued foreign borrowing and increased production and exports of petroleum and related products (see Annex I). Another alternative for financing high growth strategy is to give less emphasis to oil exports, but to have, instead, the non-PEMEX public savings increased. The policy implication of these two alternatives are quite different, and they are discussed below. 192. It should be noted, moreover, that the relaxation of the two financial constraints (external and internal) on growth does not automatically lead to a higher and more stable growth path. If the higher oil export strategy is followed, it would still be necessary to tcansform the additional foreign exchange into real resources and the only way this can be achieved is through increased imports. The problem here is therefore to allow excess domestic demand to be satisfied by additional imports. The availability of hydrocarbon resources makes that possible, but the problem is, firstly, how to absorb increasing amounts of imports without seriously disrupting domestic markets, i.e. without affecting negatively the domestic production capacity, and, secondly, to prevent, to the extent possi'ile, the entry of unnecessary (non-productive, sumptuary) imports. The former could be prevented by bringing gradually the efficiency of domestic production (and the quality of domestically produced goods) to the level of efficiency and quality norms prevailing in the international market. The litter could be avoided through the imposition of high luxury taxes on imports. 193. Another possibility is, of course, to stem the rapid increase in imports. Less oil exports would be required ii this case. Consequently, the contribution of oil to public savings would be lower than in the other case. To compensate for this, it would be necessary to increase taxes and domestic prices of goods and services produced by the public sector. Therefore, the central issue which determines the choice of strategy variant is that of Mexico's capacity to open the economy as against the capacity to increase  - 67 - taxes and public sector prices. We should discuss the implications of opening the economy, an alternative which might be more within the immediate reach of policy instruments. Open or Protected Economy? 194. Mexico is now far from being an open economy. The past development strategy's response to balance of payments problems led to cyclical attempts to cure them by increased protectionism. However, Mexico's current average level of implicit and effective protection is not excessively high compared with many other countries on a similar or even higher level of development. In view of a gradual disappearance of formerly serious foreign exchange constraint, the economy could be opened without prolonged delays and without painful adjustments. 195. Opening of the economy implies, in the first state, substitution of tariffs for import licenses and other non tariff restrictions. In fact pro- tection may even be increased slightly, if that is considered convenient. The Government had already startLed this process, and in 1979 tariffs will apply to about two-thirds of total imports. 196. The real challenge is to be able to open the economy efficiently, without wasting resources in the process and pursuing a balanced and employ- ment-oriented growth strategy as the primary objective. The experience of numerous countries which were able to increase sharply their foreign and domestic income due to a discovery or increase in export price of one basic commodity, shows that rapid increase in domestic expenditures prior to a development strategy and its implementation, led to under-utilization of capital investments. Availability of managerial capacity and of skilled labor proved to be as important constraints on growth as availability of capital. In many cases, cultivation of protectionism in the presence of rapidly expand- ing domestic demand led to increase of inflation and appreciation of the domestic currency, to changes in relative prices in favor of capital and against labor, and to a deepening -- rather than to elimination -- of these countries' dual economic and social structure. The latter is probably the most serious potential problem that Mexico should take into consideration. 197. Mexico is privileged in being able to benefit from the experience of other countries. The most important criterion in designing economic policies is to use the petroleum resources to promote the development through efficient import substituting and through export promotion. This is not only to avoid a painful adjustment process when the petroleum is exhausted or displaced by alternative energy sources or if the terms-of-trade turn against it, but also because Mexico has pressing need to expand productive employment at the fastest possible pace. The maintenance of an appropriate exchange rate becomes therefore crucial as it is a necessary policy instrument to avoid major structural problems in domestic markets while the economy is becoming increasingly open to international trade.  - 68 - 198. Mexico is now and would be increasingly so during the coming few years in a unique position to open its economy without seriously affecting domestic production, and this for at least two reasons: First, the absence of a binding foreign exchange constraint on growth and the existence of a roughly adequate exchange rate make it unnecessary to have recourse to quantitative controls over the total volume of imports. Second, the existing moderate level of implicit protection should permit to quickly replace direct import controls by import duties without running a risk of affecting imme- diately the level and structure of protectior. which would leave enough time to domestic producers to adjust to the international standards of efficiency and quality. 199. Protection has been high for some products in Mexico, but the estimates at hand indicate that overall it is still at a manageable level, and that overall protection in Mexico is somewhat less than in other large import-substituting LDCs. Apparent nominal protection of the manufacturing sector in Mexico was 16 percent in 1970; and 19 percent in 1975, as measured by difference in the price level in Mexico and in the world market, from which the effect of the overvaluation of peso in this latter year has been netted out. If it were included, the difference would become larger. Average effective protection should be somewhat lower today, owing to the reduction in the number of products subject to direct import controls (licensing) and the reduction in the average import tariff from about 25 percent to about 10 percent that took place after the September 31, 1976 devaluation. 200. The main instrument the Mexican authorities have been using to protect domestic industry and control imports has been a broad-based import licensing system. The number of products subject to licensing has varied over time and reached a maximum in 1975 when practically all imports were subject to licensing. Currently, partly as a consequence of the devaluation but also because the new authorities are consciously trying to reduce protec- tion in order to increase the overall efficiency of the economy, the number of items in the tariff structure still subject to licensing has been reduced to about 35 percent of the total - although they comprise more than 50 percent of private sector imports in value terms. The Government has repeatedly stated its intention to gradually eliminate licenses and replace them by tariffs. Different studies are now underway aiming at that end. 201. The authorities could begin by replacing the licenses by import duties (tariffs) set to grant each product the same or even slightly higher protection it had before (through licensing) so as to maintain the difference between the price in Mexico and abroad. This would require the authorities to accept, partially and temporarily, the structure of apparent nominal protection existent today. This structure which, in spite of its somewhat low average level conceals wide variations with range that goes from a minimum of 37 percent in livestock to 154 percent in beverages. Such procedure would help to mitigate the strong opposition that this measure is likely to generate. 202. It would be useful if the tariff structure and the level of pro- tection toward which the Government would like to move gradually over time are made publicly known, so as to give enough advance notice to present producers and prospective investors.  - 69 - 203. An important by-product of replacing licenses by equivalent or higher import tariffs is the increase in public sector revenues that would be generated. In 1982 this should amount to about Mex$ 39 billion or 0.9 percent of GDY, assuming that average import tariffs would be increased from its present level of about 10 percent to about 20 percent. These additional revenues could be partially used to finance stronger export incentives if those are found to be needed. 204. Although there is no valid reason to think that the suggested import policies would seriously affect domestic production, a strong resistance to them is likely to develop anyway. It could come from within the licensing administration, and also from the beneficiaries of the licenses. 205. Overcoming of such resistance can be taken care of by biasing slightly the tariff calculations in favor of domestic production. Even the possibility of reinstating licenses if imports of certain products increase beyond what is considered reasonable should not be ruled out during the transition period. At this stage the main objective of eliminating licenses would be to permit excess domestic demand to leak abroad rather than reduce or improve the structure of protection. The move toward a lower and more rational protection should be considered as a next largely staged phase. Development Bottlenecks 206. The issue of bottlenecks is relevant for any high growth strategy. However, it is especially important for the one which involves opening of the economy. Its importance depends on the expected increase in the demand for such non-tradeable goods as energy, construction, transport and some other services. Additional demand for energy would not represent a problem in Mexico since PEMEX and CFE should be able to satisfy it without major problems. Construction, particularly increased demand for some materials like cement, could represent a problem but given Mexico's proximity to the USA this problem is not beyond solution. Transport, particularly port facilities, may certainly become an important bottleneck in the face of rapidly expanding imports. How- ever, at the current level of trade there is still idle capacity in the exist- ing ports and this capacity could be further quickly increased if container technology is introduced and grain terminals are built. Unskilled labor is in ample supply but skilled labor less so, and it could become a bottleneck in some sectors and, as importantly, would drive wages up. 207. Altogether, given Mexico's ample energy supply and access to the USA market, shortages of non-tradeable goods which are likely to be in short supply may not become important enough to trigger a cost-push type of inflationary process. It should be noted that during 1972-1975, inflation peaked at about 25 percent in spite of high demand pressures and increasing protectionism. This report is projecting domestic inflation to remain above international inflation in 1979 and 1980 allowing some room for necessary adjustments in relative prices likely to take place during expansion period. Thereafter, domestic inflation should gradually approach international levels and remain roughly at that level in the future.  - 70 - Some Non-alternatives 208. There will be practical difficulties in implementing policy packages related to any of the high growth strategies (for details - see Annex I). It can be argued that, as the Government would have more real resources than ever before, why is it considered necessary to increase domestic taxes and prices, opening the economy and exposing domestic production activities to sometimes dangerous - because pitiless in the Darwinian sense - international competition? Why not use the additional resources within an economic environment similar to that of the past, and let it expand - in the Von Neumann's sense - rather than to change it? 209. The latter alternative is only apparently easy - in a medium term it may prove to be the most difficult, because it would lead to tensions and would leave the country's economy unprepared for the long-term future. Thus, if the economy continues to develop within a relatively closed model, imports would not be permitted to increase as rapidly as needed, but expenditures (private and/or public) would. As there would be more foreign exchange available, it would be exchanged into domestic currency at the given exchange rate, and the newly available domestic currency would push up domestic demand in exactly the same way and with the same consequences as an increase in the money supply which is used to finance, for example, public expenditures. Here ends a first round of interrelated-effects. The increase in domestic demand will fuel domestic inflation and reduce the real rate of exchange, if the nominal rate remains fixed. As imports would become relatively less expensive, the demand for them will increase in the next round. The domestic price level will increase, but the real rate of exchange will have to decline, as much as required for the new foreign exchange to be absorbed by higher imports. This cycle of events thus becomes a "vicious circle" in which the real exchange rate moves increasingly out of line with what is needed to promote the expan- sion of non-petroleum activities. As a result, the dualistic tendencies in the economy are intensified. 210. In Mexico today, this "derived" inflation rate may be significant because tradeable goods represent only a small fraction of total demand, and the excess demand that is leaked abroad in the first round effect will be small, and also because the relative price elasticity of imports is low. Thus, if the present protection system is maintained, a reduction in the real rate of exchange may be required to let imports increase as much as needed. 211. A reduction in the real rate of exchange would have highly negative short, medium and long-term effects. The non-oil exports activities will be discouraged as it will also be the case for the import-substituting activities deprived of any special protection. As a result, the country will become more and more dependent on oil to obtain its foreign exchange requirements, and also, because the discouraged activities are, almost by definition, more labor intensive than oil, their replacement would have adverse effects on employment, and this at a time when creation of employment should be considered as a first priority, in view of the expected rapid growth of the labor force. 212. By letting the domestic currency depreciate and adjusting the nominal rate exchange accordingly, the Government may avoid the deterioration in the  - 71 - real value of the exchange rate. In this case, domestic inflation will be rekindled if imports are not permitted to increase as much as necesary and, instead of transforming the available foreign exchange into real resources, this would lead into transforming it into a higher money supply. 213. The Government may try to avoid some of these adverse effects by letting in imports which were previously prohibited. If it has been previously considered that spending foreign exchange on such goods represented a waste of resources, there may be little reason to change these views now, and such expenditure should still be considered a waste of foreign exchange. Another hypothetical solution may be granting incentives for importing capital equip- ment, but this, in turn, would strengthen incentives for capital-intensive production techniques, discouraging the use of labor. 214. Finally, the Government may decide to hold down public expenditures and oil exports and therefore "control" the oil expansion. Even this policy may not be sufficient to avoid the adverse effects of a high growth-cum-semi- open-economy model because private expenditures would also increase substan- tially, and they would be f:Lnanced from foreign credits. International financial markets consider Mexico an eminently creditworthy country, and for good reasons, and would continue to be ready to extend to it increased credit. In this situation, even if public expenditures and oil exports are controlled within appropriate limits, the adjustment problems brought about by the new oil wealth - even though potential and not fully released in this latter case - would not disappear. 215. It would therefore appear that the only economically and socially sustainable alternatives which would contribute to both employment generation and to only a limited dependence on oil exports are: a) to export oil as needed, and to open the economy; or b) to export somewhat less oil than in variant (a) but instead to increase taxes and prices of Government goods and services, reducing private expenditures, and reducing imports and the needs for foreign exchange. These are two alternatives that are being reviewed in the following pages. It is noteworthy, however, that this report does not pretend to forecast what will happen in the Mexican economy in the years to come. Instead, it tries to illustrate the broad options open to the Mexican policy-makers and the trade-offs among different policy actions. In this sense, the projections that follow should be interpreted as an attempt to illustrate how fast the Mexican economy could grow if certain policies are implemented, and the price the country would have to pay in terms of growth and employment if the policy changes are not implemented. It is realized that some of these policy actions are difficult to take and therefore it is not expected that the economy would behave exactly as projected below. It should be emphasized that in the light of Mexico's pressing social problems, every effort should be made to increase quickly the rate of growth of the economy and to make it efficiently. 216. It is quite clear that there are many difficulties inherent in taking any of the two suggested alternatives and the corresponding policies. Even if there is agreement on them, there will still be lags involving, first, their implementation and then the appropriate adjustments. In the meantime, the economy will behave as suggested in paras. 208-212 and as it is mentioned  - 72 - there, inflation will rise. Nevertheless, if there exists the clear under- standing that the economy will be opened or/and the level of public sector s current revenues increased by an appropriate set of policies, the high rate of inflation will have to be a transitory phenomenon. In order to allow the economy to start with the "right" set of relative prices after this transition, a policy of flexible prices will become indispensable, even if it would contribute to the inflation problem. In particular it may become indispens- able to adjust the nominal rate of exchange, so as not to allow the corres- ponding real rate of exchange to drop below an established target. We would suggest that at least initially, it should be fully adjusted by the differen- tial between Mexico's rate of inflation and the international inflation rate relevant for Mexico. The aim thus is to maintain a real exchange rate that makes a wide range of non-petroleum return competitive on the world market, whether for import substitution or export. A secondary benefit of such a more flexible rate policy would be to reduce the present incentive to borrow abroad which, via additional reserve accumulation and monetary expansion, is 'adding to inflation pressures. Growth and Investment 217. Leaving aside financial considerations, the rate of growth of GDP will depend on the level and growth of the domestic demand and on the level, growth and productivity of investment. Even though demand would be lower in the scenario of high taxes and higher prices for public sector goods and services, it is assumed that this lower demand would be related to a lower volume of imports and that, therefore, the demand for domestic goods would be identical in both scenarios. Investment, its rate of growth and its produc- tivity, will be also assumed to be the same in both scenarios, and thus the rates of growth of the GDP will be the same. 218. GDP is projected to grow at almost 9 percent in real terms per year, during the remainder of the sexenio, while fixed investment is growing at almost 15 percent in real terms per year, during the same period and the average productivity of investment -- as measured by the incremental capital output ratio (ICOR) -- is slightly higher than during 1966-1970 (see Table 19, page 73). Is this investment behavior feasible? The public investment program for the sexenio is summarized in Table 20. Public sector investments are projected to grow at 11.2 percent per year in real terms over the sexenio, but at about 18 percent between 1979 and 1982, reflecting a strong recovery of public investments after the marked slow-down of 1976-1977. The real increase in public investments would be particularly strong in 1979, when it is pro- jected to grow by about 20 percent, declining gradually thereafter to about 11.4 percent in 1982. 219. Past experience shows that during the third year in office Mexican Governments expand public investments at a record pace; it happened in 1967, and again in 1973 when public investment grew at about 31 percent and 34 percent in real terms, respectively. The fourth and the fifth years are also expansionary but the pace is slower. However, during the previous Administration public investments grew at about 20 percent per year in real terms during 1973-1975.  Table 19 : MEXICO - GDP AND INVESTMENT REQUIREMENTS, 1965-1982 (1972 Mex$ billion) Annual Rate of Growth 1977 1978 1979 1980 1981 1982 1966-70 1971-76 1977-82 GDP 620.9 661.3 719.5 787.9 862.7 933.3 6.8 4.9 8.5 (i) Agriculture 56.o 57.6 59.4 61.1 63.0 64.8 2.6 0.8 3.0 (ii) Industry and Mining 226.3 249.9 218.8 319.1 360.5 405.2 7.3 5.8 12.4 - Petroleum (25.0) (30.0) (37.5) (46.1) (55.4) (64.2) (9.5) (8.o) 20.8 (iii) Rest 338.6 353.8 378.3 407.7 439.2 463.2 6.6 5.1 6_5 Fixed Investmenta/ 114.7 130.7 156.1 181.7 202.2 217.4 9.4 5.6 13.6 (i) Public Sector 47.3 55.4 72.7 87.3 100.4 108.3 13.5 7.8 18.0 - PEMEX (11.8) (19.0) (20.0) (23.0) (23.0) (23.0) (5.8) (10.8) (14.3) - Non PFNMX (35.5) (36.4) (52.7) (64.3) (77.4) (85.3) (19.2) (ii) Private Sector 67.4 75.3 83.4 94.4 101.8 109.0 7.4 4.2 10.1 Annual Averages 1966-70 1971-76 1978-82 ICOR - 2.84 2.24 2.29 2.43 2.86 2.71 5.53 2.89 (i) Non-petroleum Economy - 3.22 2.20 2.28 2.42 2.90 2.61 5.87 2.52 a/ As these figures are deflated by the corresponding implicit price indexes, they may not be always equal to the investment figures shown in other text tables where the GDY deflator is used. Source: Staff estimates. * Identical version for both Scenarios.  - 74 - Table 20 MEXICO - PUBLIC SECTOR INVESTKENIS, 1977-1982 (1972 Mex$ billion)&I Rate of 1971-76 1977-82 Change $ % $ % Industry 99.1 39.0 206.0 42.8 107.9 PEMEX (44.5) (17.5) (119.8) (24.9) (169.2) Rest (54.6) (21-5) ( 86.2) (17.9) ( 57.9) Welfare and Social 50.1 19.7 77.9 16.2 55.5 Infrastructure Transport and Communi- 56.3 22.1 79.1 16.4 4o.5 cations Agriculture and Rural 40.3 15.9 97.l 22.2 140.9 Development Other 8.3 3.3 11.3 2.3 36.1 Total 254.1 100.0 471.4 100.0 85.5 Memorandum Item Housing (INFONAVIT) 12.3 - 20.1 - 33.1 at Deflated by the investment deflator Source: Staff estimates.  - 75 - 220. The slow-down of public investment after the third year of each Government's administration has been usually brought about by the emerging financial problems, particularly the need to check the increasing public sector deficit. Leaving aside financial considerations, the only reason for not increasing public investment in 1979 and thereafter at a faster pace than in the past would be the lack of adequate projects and/or the lack of institutional and managerial capacity to handle these projects efficiently. This could indeed become a binding constraint on growth. 221. We cannot, at this stage, form a conclusive judgement on the extent of this problem. However, the experience indicates that Mexico is one of the very few LDC's in which the availability of bankable projects and of good management has never been a problem; on the contrary, the amount of suitable projects was usually in excess of what could be financed. Further- more, Mexico has been a pioneer in successfully implementing more complicated types of projects, such as integrated urban and rural development projects, small scale industry projects, etc. Mexico also has a pool of professionals and experts that could be called upon as consultants to prepare good projects and programs and if needed, also to manage them; the public sector itself is staffed with professional, experienced and highly competent middle and senior management and should be able to administer such a program efficiently. 222. The structure and timing of the proposed investment program could also be contributive in making this issue easier to overcome, since the bulk of the increase in investments until 1980, would take place in the traditional sectors while investments in social sectors and special programs will take off sharply only in 1980-81. This should leave the authorities enough time to set off a special training effort geared toward those areas in which lack of specific skills and managerial abilities have been identified. Different initiatives in this direction are now underway and the Bank is helping Mexico to organize a major effort to improve the public sector's capacity to identify and evaluate investment projects and development programs. 223. On the basis of the above considerations there is a reasonable hope that the Mexican public sector would be capable to manage even the vastly increased investment program quite efficiently. The magnitude of the investment effort is considerable; it would bring the ratio of public investment to GDY from 9.1 percent in 1978 to 13.0 percent in 1982, a large expansion even when compared with the most expansionary period of the recent past, i.e. 1973-1975, during which that ratio went up from 6.5 percent in 1972 to 9.7 percent in 1975. If there are indications that the magnitude of the investment effort is such that it would not be possible for the public sector to manage those resources efficiently, then the private sector might be induced to expand its investment at a faster pace than what is now expected. However, the private sector investment may not look for opportunities in sectors the public in- vestment is planned to expand. In such case, the public sector may have to concentrate on the "core" sectors, where the economic rate of return would considerably exceed the financial rate of return, and vacate for the private capital sectors where these rates are reasonably equal, but still adequately high. The final option, which is to reduce the pace of growth of overall investment (private plus public), should only be considered if it is believed  - 76 - that the private sector will not increase its investments at a higher rate than projected and that by forcing such an expansion of overall investment, resources will be wasted. 224. Private sector investment is projected to grow at an annual real rate of about 10 percent during 1979-1982 which will keep its ratio to GDY at a little over 11 percent. Although this rate of expansion is higher than the extrapolation of past experience would have indicated, it is still quite conservative in the light of projected public sector invest- ments and the experience of other countries experiencing (a few) or having experienced (many) economic booms, particularly in conditions of hardly any financial constraints. 225. It is assumed that, as in the past, private investment will follow with a lag the growth of public investment and that it will increase by 11 percent and 13 percent in 1979 and 1980, respectively, slowing down thereafter to about 7 percent in 1982. It should be noted that the level of private investment projected for 1979 is only 8 percent above its real 1974 level, so that the large increase in 1979 and in 1980 basically reflect the recovery of private investment from the depressed levels it reached in 1976-1977. 226. Private sector investment, being basically a function of expected profitability, depends on confidence in economic management and political stability. Confidence in economic management comprises many factors, ranging from confidence in the individual top decision makers to clear and stable policy signals. This last factor is especially important in market-oriented economies such as Mexico in which Government intervention in the market is exerted indirectly, through economic policies that affect profitability in many different ways. In this field the present Administration achieved a notable success by having been able to generate a positive "business climate" which has already induced a steady recovery of private investment. The Government's ability to manage the severe financial crisis it inherited together with its explicit policy of stimulating cooperation among the different sectors involved in the production process have proved to be key factors in this development. This notwithsdanding, a better and more explicit definition of areas of main activity for the public, private or joint venture would be highly useful, especially in oil related sectors, such as petro- chemicals, and in the capital goods sector. 227. Once confidence has been established, prospects for profits and the availability of financial resources become the overriding factors in deciding how much and where to invest. Demand prospects appear excellent in the light of the expected large increases of public sector demand and disposable income. Furthermore, manufactured exports could be expected to grow at about 10 percent per year in real terms, provided that the exchange rate will be maintained at or near its present level in real terms and export incentives will continued to be strong. 228. Although a revision of current Government incentives to the private sector and an additional effort to increase tax revenues are suggested, this does not imply an across the board elimination of incentives to stimulate selectively private sector investments. Neither does this imply a steep increase in taxation. What could be recommended is a broadening of the tax  - 77 - base, a more effective control of evasion and the elimination of incentives that generate a bias in favor of capital-intensive process and its replacement by incentives aimed at increasing the profitability of labor-intensive forms of productions (e.g. negative wage bill tax) or of using more labor-intensive processes (e.g., multiple shift operations) wherever possible. 229. The rate of growth of private investment projected for the remainder of the sexenio appears rather conservative in the light of the above discussion. It could indeed be higher and could eventually compensate for lower than projected public investments, so as to avoid a slowdown of total investment and, therefore, of GDP growth and employment generation. It should be noted that total fixed investment is projected to grow at about 14 percent per annum in real terms during 1979-1982, a rate that appears reasonable con- sidering the special circumstances of Mexico. 230. This rate of growth, while high when compared with the past perfor- mance of Mexico, does not appear to be overestimated when compared to other LDCs which have experienced an oil-propelled or even general economic boom. Available evidence shows that GDI (Gross Domestic Investment) grew in real terms at about 18 percent per year in Brazil during 1968-1973, at about 14 percent per year in Spain during 1961-1966, at about 11 percent in Venezuela during 1973-1976, at above 17 percent in Algeria during 1966-1977 and at about 25 percent in Iran during 1971-1976. 231. It is assumed that Mexico will invest its resources efficiently and that the ICOR will be slightly lower in 1978-1982 than it was in 1966-1970, the latter period considered as normal. The low ICOR projected for 1979-1980 also reflects the resorption of the slack the economy has been accumulating since 1974, reflected in the abnormally high ICORs observed between 1974-1977. 232. On the basis of these investment and ICOR projections, GDP growth can be expected to fluctuate between 8.5 and 9.5 percent per year in real terms during the remainder of the sexenio. Agriculture is expected to grow at about 3 percent per year, a modest performance that merely reflects the longer lead times of investment in agriculture but also indicates a reversal of past trends; industry and mining are expected to become the most dynamic sectors and are projected to grow at about 12 percent per year, with petroleum and petrochemicals growing at more than 20 percent per year; and finally services are expected to grow at about 7 percent per year. Inflation and the Rate of Exchange 233. During the 50s and 60s when there was no balance of payments problem in Mexico and the domestic rate of inflation converged with that in the world (or rather in the US) the rate of exchange remained constant in nominal terms. Mexico broke this pattern in the mid-70s, and the ensuing financial desequilibrium was solved through devaluation. At the end of 1978 prices in Mexico were roughly comparable with those of the US, and therefore, the rate of exchange appears to have been appropriate. 234. This notwithstanding, Mexico may have to face an adjustment problem in the balance of payments but of a different type than the one experienced in 1972-1977. Imports increased by more than 50 percent in real terms between  - 78 - 1971 and 1974, which resulted in grave problems afterwards: a financial disequilibria in 1975 and 1976, a steep devaluation, and a recession which lasted until 1977. This time the problem is an inverse one: exports of oil which represented less than 10 percent of the other exports of goods and services prior to 1977 may by 1982 become almost equal to the other exports. This implies that the purchasing power of total exports would be almost doubled in a few years, which poses adjustment problems - involving an upward adjustment of imports - and therefore opposite in nature to those of the mid-70s. This is the main reason why, if the supply of foreign exchange would increase by as much as expected, the wider opening the economy to imports and letting the excess domestic demand leak to the rest of the world would become an obvious solution. 235. In the high oil exports scenario, in which the economy is assumed to have been completely opened and the rate of exchange is kept constant in nominal terms Mexico's inflation should come close to the rate of inflation in the US. In the high non-oil public savings scenario, the situation should be similar to the preceding one but only if non-oil savings increase enough to prevent a financial disequilibrium to develop. Under these hypotheses the inflation and exchange rates are assumed to be similar in both scenarios. 236. It is assumed that it should take about three years for Mexico to lower its inflation rate so as to reach the international rate of inflation, which is assumed to hover around a rate of 7 percent per annum. A decisive but gradual decrease of inflation in Mexico is however projected, down to 15 percent in 1979 and 10 percent in 1980. In these two years international inflation is assumed to become 10 and 8 percent per year. As the nominal exchange rate is assumed to remain constant, the real rate of exchange would have to decline slightly -- proportionately to the differential between the world and the Mexican inflation -- in 1979 and 1980. In any event, the implicit deterioration in the real rate of exchange implied by the above price differentials would be less than one-twelfth of its current level, a change so small that it was felt that no different inflation and exchange rate hypothesis should be introduced into the scenario. The assumptions regarding inflation and the nominal rate of exchange are obviously an oversimplification. As suggested in para. 216, inflation is likely to run initially higher than expected, even if the right set of policies is soon applied. The reason for this oversimplification is that, first, it is the real value of the variables that really matter and, second, the mechanism of adjustment could hardly be anticipated. 237. This small decline in the real rate of exchange if it materializes in practice, as it did in all other countries which were blessed with a sudden and steep increase in foreign exchange revenues, would have to be compensated by special incentives to promote increase in non-oil exports. Such policy may be defended on two grounds. Firstly, the non-oil (non- traditional) exports are necessary, indeed imperative, to keep the country's dependence on oil at a reasonable level, practically from the beginning of the export surge. Secondly, "promotional" incentives may be needed to expand these exports and allow them to enter new foreign markets, especially so because Mexico may be interested in diversifying its foreign markets.  - 79 - 238. The strong increase in the rate of economic growth is not in- consistent with a reduction in the rate of inflation if all major traps are avoided and, therefore, if economic policies are designed and implemented with this in mind. This is precisely the thrust of a major part of this report. Examples could be provided to prove that such exercise is not purely academic. Thus, Brazil experienced a strong reduction in the rate of inflation and at the same time succeeded in rapidly increasing its output after 1968. During the preceding period of 1962-1966 inflation rate remained at over 50 percent per year while output growth stayed at less than 4 percent rate per year. In the successive period inflation was reduced to less than 20 percent per year while output climbed to a level close to 10 percent per year. Spain after 1961 also offers a similar case, of output strongly increasing while the inflation declined. Some Alternatives 239. Two alternative growth scenarios have been designed and their outcomes reviewed in Technical Annex I. The variables were calculated as a function of a set of economic policies, different in each scenario. The objective function is the same in both scenarios--a relatively high growth rate of the economy, fixed under the assumption that it may be extremely difficult to maintain the growth rates within the next few years at a lower level commensurate with the Government's intentions to reduce sharply the rate of inflation. Furthermore, it may not even be as advisable to reduce the growth rate, as there exists a possibility to reduce the inflation rate without having to reduce the growth rate, in Mexico's situation--which should persist during a year or two--where there is still some idle capacity which could be better utilized in both the goods' sectors (agriculture and industry) and in infrastructure, through the "switchovers" among modes of transports or sub- sectors of production, possible in cases where the elasticity of substitu- tion is sufficiently high. In such cases a use of a price system reflecting supply scarcities and nudging toward shifts to other modes of transport or other forms of production, supplied by underutilized capacities, should be more amply used. In turn, a wider opening to imports could help in adjusting the price system in the direction, where it would not so much reflect the scarcities of the "closed" system, but the possibilities to eliminate them through imports. Imports should, however, principally help in removing some of the bottlenecks. 240. It has been assumed that the amount of public investment would be the same in both scenarios. This consideration is largely dictated by the Government's concern that substantial public investment will have to be used toward removal of future irLfrastructure bottlenecks--with an average four- year investment cycle it will aim at those expected to appear in 1983--and, as importantly, that a part: of these capital expenditures would help bringing closer the solution of the poverty problem, through an increase of investment in education, traditional agriculture, rural infrastructure and development of backward regions. 241. The two scenarios are meant to illustrate the effect of two different sets of economic policies on the generation of financial resources necessary to finance the investment. Each scenario embodies a mix of popular and  - 80 - unpopular policy measures. The definition of "popularity" depends on how these measures affect "producers" and "consumers," bearing in mind that what is often acceptable to the producer, because it contributes to his profits is, by the same token, hardly acceptable to the consumer, because it might reduce his personal welfare. 242. In the first scenario (A) the level of oil exports has been calculated to reach about 1.6 million barrels of exports per day by 1982. Of this, one- third is contributed by exports of refined products, with about 1.1 million barrels left for exports of crude. In addition, it is expected in this scenario that Mexico will also export some 5.8 tons of petrochemicals per day. These exports, together with exports on non-traditional manufactures and agricultural products, should permit Mexico to attain a total value of exports of about US$24.2 billion by 1982, necessary to pay for US$24.9 billion worth of imports in this year. The current account deficit is projected to be at around US$4.7 billion, mainly because of the high net factor service payments, going mainly to payment of interest on foreign debt. 243. In this scenario, the current revenues of the public sector are to grow at a rate of about 15.5 percent per annum in real terms during 1978-1982, largely thanks to the contribution of PEMEX, whose share in current revenue grows from 24 percent in 1978 to 39 percent in 1982. This volume of current revenue is estimated to be sufficient to generate public savings amounting to Mex$70 billion (in 1972 prices) by 1982, enough to finance 60 percent of total public investment in that year. 244. The principal policy measure which could be considered as unpopular from the producers' view point is that, in this scenario, a wide opening of the economy to imports is the basic condition. This, in turn, could be achieved through a conversion of the import licensing system into a tariff- based system. The growth of imports in this version, set at about 26 percent per annum in nominal terms during 1978-1982 (about 17 percent per annum in real terms) is considered necessary for several reasons. The principal is that a freedom to import would help to eliminate some--although far from all--growth bottlenecks, which may begin to appear in the coming years. It may also help in keeping the rate of inflation closer to the international price trend, by preventing that domestic prices are unduly influenced by speculative price fixing by new and by already entrenched oligopolies and monopolies. Finally, the relatively unrestricted inflow of imports should help in phasing out the most inefficient forms of domestic production, while, at the same time, making the rest of Mexico's domestic production more compe- titive, compared to international standards. 245. Another scenario (B), also designed and reviewed in the Annex, assumes a continuation of the present (medium type of) protectionist policy against imports. Such policy, which could be considered as essentially more popular with the "producers" than that of the preceding scenario, would entail by 1982 a level of imports about US$4 billion lower than the one of the preceding scenario (A). Lower imports call for a lower level of exports and, conse- quently, in this scenario exports of crude oil will amount to only 0.5 million barrels per day by 1982 compared with 1.1 million barrels in the preceding scenario. This would therefore entail a smaller volume of total production of crude than that foreseen in the preceding scenario.  - 81 - 246. Public revenue--and therefore also public savings, at an unchanged public expenditure level--contributed by oil taxes and profits, would there- fore, be lower in the present scenario (B) than in the preceding one. To compensate for such decline, the Government will have to make a strong effort to increase its revenue --and savings-- from non-oil sources. This could be achieved mainly through increases in real prices (i.e. price increases by more than it is justified by the general price index) of goods and services produced by the public sector, consisting mainly of those produced by public enterprises. Contribution of these price increases to the public revenue may not prove to be enough and, to compensate fully for a loss (compared with scenario A) of oil revenue, an increase in tax rates would also be necessary. 247. Such policies could --or rather would-- be considered unpopular by the "consumers". They will also preempt a part of private savings. It is calculated that the shortfall of public: savings would amount to some Mex$24 billion in 1972 prices or some Mex$75 billion ($3.3 billion) in 1978 prices --if no increase in real prices of public goods and services and in tax rates is decided. Conse- quently public savings based on revenue other than those generated by oil will have to increase from about 1 percent of the GDY in 1978 to about 3 percent of GDY in 1982. While not impossible, especially when it is recalled that the latter level was attained in Mexico in the late sixties, this increase might be difficult to achieve, unless a strong action is planned now and is under- taken quickly. 248. The two scenarios show that, within the next several years, there is no escape from the dilemma: either to open the economy wider and, create through this some dissatisfaction among domestic producers, at least among the marginal ones, who may not be able to raise to the challenge of international competition, or to strengthen considerably the domestic resource mobilization effort on the part of the public sector and, as such effort would have to rely on higher public sector prices and higher taxes, confront the dissatisfaction of the private consumers. In not doing either, the risk is that the public investment program would be deprived of adequate financing. The further conse- quence of the latter would be that either the volume of this investment is re- duced or, if it is financed from the banking sources, with credit exceeding what the monetary program would allow, this might raise an inflationary wave. 249. In turn, if it is decided that the public investment program should be reduced, a very real danger is that the social or "soft" programs could fall victims to it. The reason here is in that the other, "hard" projects, in industry or in basic infrastructure, are more difficult to postpone and this for a variety of reasons: industrial investment is normally preceded by contractual arrangements which are costly to rescind, petroleum investment is imperative to assure future foreign exchange income and infrastructure investment would be necessary for elimination of growth bottlenecks. However, a decisions to de- crease the level of investment in social sectors and in social welfare would create an even more damaging, deep reaching long-term effect. This is why all necessary measures to assure that financial resources necessary to support public investment of the planned magnitude are adequate, should be taken as early as possible.  - 82 - 250. There is therefore no way to escape the decision to take measures, relatively unpopular either for the producers or for the consumers or for both these groups. The latter option, i.e. measures which would be relatively unpopular for both groups, could probably be considered as the optimum solution. Calculation of what such measures could yield would require a design of a still another scenario, half-way between the two extreme scenarios reviewed in the attached model. Such "compromise" scenario would imply a slightly higher growth of exports and a somewhat more open economy than that programmed in the second scenario (B) but less exports and a slightly lesser opening of the economy than in the first scenario (A). In terms of other policies, a "compromise" scenario would entail a reasonably strong effort to mobilize public savings, outside those provided by oil production and exports, i.e. somewhat lower than in the second scenario but somewhat higher than in the first under the growth investment constraints. The number of policy solutions is infinite but they are not really radically different from each other. Their mix would depend on the strategic, largely political, choices that the Govern- ment would decide to opt for. This report hopefully provides some material for pondering such options and choices.  - 83 - ANNEX I POLICY OPTIONS AND DEVELOPMENT OUTLOOK 1. Two alternative growth scenarios have been designed and their outcomes reviewed below. The variables were calculated as a function of a set of economic policies, different in each scenario. The objective function is the same in boths scenarios--a relatively high growth rate of the economy, fixed under the assumption that it may not be possible for the Government to reduce the growth rates within the next few years to a level commensurate with the Government's intentions to reduce sharply the rate of inflation. Scenario A: High Oil-Exports Scenario 2. The crucial difference between the two high growth alternatives is in the way investment is financed. In terms of policies, what really matters is their effect on public, private and external savings. The table which follows shows the investment financing structure related to the high oil-exports scenario. (i) Private Savings 3. Private sector savings are projected to grow at about 15 percent per year in real terms during the sexenio, which implies the increase in the average savings ratio (ratio of non-monetary savings to disposable income) from about 0.14 in 1977 to 0.18 in 1980-1982. This is a moderate savings effort which is feasible given the expected high rate of growth of disposable income during the period and the proposed modernization and strenghtening of Mexico's capital market. 4. The strongest private savings effort would, in this scenario take place during 1979-1980, the years when the marginal savings ratio would reach 0.32. This is a result of a momentary conjunction of the effects of the rapid growth of disposable income -- growing by 8.9 percent and 9.8 percent per year respectively in 1979 and 1980 - and the rapid recovery of private savings after its weak performance of 1976-1977, when they dropped in real terms. 5. The average private sector savings ratio projected at 0.18 by 1980 implies a slight improvement with respect to the past performance, when it fluctuated between 15-16 percent. This is not an ambitious savings target since the 1979-1982 period would be unique in terms of the pace at which economic activity would expand and it might well be expected that the level of private savings would receive a strong upward thrust if consumption pro- pensity is not stimulated by unduly high inflation rates. 6. In Mexico, as in other LDCs, most of private sector savings have traditionally come from undistributed profits and depreciation funds. The projected acceleration of investments and the competition brought about by the gradual reduction of protection, necessary to increase the overall efficiency of the economy and foster non-petroleum exports, would force  -84- ANNEX I TABLE 1 MEXICO - HIGH OIL EXPORTS SCENARIO (A) INVESTMENT AND ITS FINANCING (1972 Mex$ billion) a! 1977 1978 1979 1980 1981 1982 Total Fixed Investment 126.11 143.11 171.47 200.19 223.52 241.08 Private Investment 74.10 82.45 91.65 103.99 112.48 120.92 Public Investment 52.01 60.66 79.82 96.20 111.04 120.16 Consolidated 42.53 50.54 66.54 80.16 91.63 100.15 PENEX 14.34 20.80 21.96 25.35 25.44 25.51 Non-PEMEX 28.19 29.74 44.58 54.81 66.19 74.64 Non-Consolidated 9.48 10.12 13.28 16.04 19.41 20.01 Total Financing 126.11 143.11 171.47 200.19 223.52 241.08 External Savings 16.94 22.40 21.13 21.59 22.22 23.21 Public Consolidated 20.40 23.89 35.92 48.36 63.00 70.19 PEMEX 13.72 18.26 28.53 38.54 51.32 60.02 Non-PEMEX 6.68 5.63 7.39 9.82 11.68 10.17 Private Savings 68.44 85.29 103.62 122.28 132.09 140.87 Money -3.64 5.70 7.94 7.57 9.15 8.63 Non- Money 72.08 79.59 95.68 114.71 122.94 132.24 Inflation Tax 20.33 11.53 10.80 7.96 6.21 6.82 Memo Items Relative price of investment 1.097 1.095 1.098 1.102 1.106 1.109 Disposable Income 521.06 566.49 616.68 677.18 739.99 800.03 Oil Exports(US$ million) 915.8 1,874.4 3,809.6 5,924.8 8,695.9 11,065.1 Marginal Private Savings Ratio (out of Non-mone- tary Savings) 0.16 0.32 0.32 0.13 0.16 a! Deflated by the implicit GDY deflator. Source: Staff estimates.  - 85 - ANNEX I private investors to go public and/or to rely increasingly on financial markets for long-term financing. The Government in turn would also need to rely on private savings as complement to its own savings and the external resources it can obtain; the success would depend here upon the effectiveness of monetary authorities in creating a strong and stable financial market. 7. The development of a securities market and the gradual liberaliza- tion of interest rates along the lines suggested earlier in this report would help to attain the saving levels considered as necessary in the above projections. If private sector savings do not respond strongly and reach levels below those projected above, the most likely outcome would be a slow-down of private investment and consequently, of the employment genera- tion and of growth. Another possibility, however, would be an increase in the private sector external borrowing, which may change the leverage struc- ture of private enterprises in Mexico, a structure hitherto rather cautiously observed. Moreover, this would imply a further increase in imports. (ii) External Savings 8. External savings would have to continue, and unavoidably so, to play an important role during the years to come, declining from about 3.0 percent of GDY in 1979 to less than 2.5 in 1982. This amount of external savings will have to correspond to an equivalent current account balance of payments deficit, which, given the large increase in revenue from petroleum and related products implies that imports would have to grow at a substantially faster pace than in the past. 9. The balance of payments projections consistent with the above growth and investment projections are made in the table which follows. The basic assumptions underlying these projections are that: (a) the economy would be opened to import, enough to satisfy excess domestic demand; (b) the nominal rate of exchange will remain unchanged; (c) incentives to stimulate non-petroleum exports would be strengthened. 10. One of the most striking outcomes of these projections is the quick increase of the trade ratio (ratio of imports plus exports to GDY) from 20.5 percent of GDY in 1978 to 25.1 of GDY in 1982. Most of this increase is accounted for by the steep increase in exports -- the export ratio goes up from 9.8 percent of GDY to 12.4 percent during the same period -- but the import ratio increases alsc, substantially, from 10.6 percent of GDY in 1978 to 12.7 percent in 1982. 11. Although most of the increase of merchandise exports is explained by petroleum exports -- which grow from US$1.9 billion in 1978 to US$11.1 billion in 1982 -- manufactured exports (including assembly industry) and petrochemicals are also expected to increase rapidly, at about 10 percent per  TABLE 2 : MEXICO - HIGH OIL EXPORTS SCENARIO (A) BALANCE OF PAYMENTS PROJECTIONS, 1978-1982 (In millions of US$) 1977 1978 1979 1980 1981 1982 Exports 7,454.2 9,160.0 12,530.1 16,117.1 20,283.2 24,244.2 Merchandise 4,278.2 5,387.3 7,884.5 10,661.9 14,041.7 17,132.9 Non-Factor Services 3,175.9 3,772.7 4,645.6 5,455.2 6,241.4 7,111.4 Imports 7,582.8 9,842.1 13,109.5 16,747.9 20,879.1 24,870.0 Merchandise 5,518.5 7,429.4 10,126.2 13,042.0 16,212.5 19,170.2 Non-Factor Services 2,064.3 2,412.8 2,983.2 3,705.9 4,666.6 5,699.8 Factor Service Income (Net) -2,073.7 -2,642.2 -3,029.3 -3,417.6 -3,860.7 -4,348.1 Interest Payments -1,804.3 -2,050.1 -2,343.1 -2,625.5 -2,949.6 -3,303.0 Direct Investment Income - 480.8 - 823.5 - 940.7 -1,072.1 -1,219.1 -1,383.9 WorkersRemittances 211.3 231.4 254.6 280.0 308.0 338.8 Balance of Goods and Services -2,202.4 -3,324.2 -3,608.6 -4,o48.2 -4,456.5 -4,973.8 Transfers (Net) 199.8 218.4 238.9 261.3 285.9 312.9 Current Account Deficit -2,002.6 -3,105.8 -3,369.7 -3,787.0 -4,170.8 -4,660.9 New Direct Investment 437.3 679.7 81.7 875.4 980.5 1,098.2 Net Public Sector Borrowings (M. and LT) 3,896.0 2,890.9 3,454.8 3,721.0 4,122.9 4,510.5 Disbursements 6,365.2 6,778.1 8.426.7 9,821.9 11,135.1 11,685.0 Amortization -2,469.3 -3,887.2 -4,971.9 -6,100.9 -7,012.2 -7,174.5 Other Capital -1,860.0 100.0 - 50.0 100.0 100.0 50.0 Change in Reserves (Gross) - 470.7 - 564.8 - 816.8 - 909.6 -1,032.8 - 997.7 Source: Staff estimates.  - 87 - ANNEX I year in real terms during 1971-1982. The rapid growth of manufactured exports during the remainder of the sexenio, in spite of low growth expecta- tions for the principal potential importers, US and other OECD countries, is predicated upon the assumption above. 12. Imports are projected to grow from US$9.8 billion in 1978 to US$24.9 billion in 1982 -- aL growth rate of about 20 percent per year during 1979-82 (13.3 percent per year in real terms). The implied elasticity with respect to GDP is 1.9 -- more than the double of the elasticity (of 0.9) calculated for 1950-1970. The former is lower than the elasticity calculated for Iran where it exceeded 4.0 during 1971-1976 and for Algeria, where it stood at about 3.0 during 1971-1974. It is noteworthy that both these countries allowed their exchange rates to appreciate during this period. In Brazil the implied import elasticity during the boom period (1968-1973) was about 1.6; the exchange rate was kept roughly constant in real terms, but there were widespread controls on imports. In Spain, during the period 1961-1966 the import elasticity was 2.9 but the real rate of exchange was allowed to decline about 20 percent and balance of payments problems developed. Finally, the short-term import elasticity in Mexico was around 3.0 in 1978, when the GDP grew by 6.6 percent and imports by about 20 percent, both in real terms. This increase was due, however, to the overwhelming extent to an increase in imports of raw materials and capital goods from a relatively depressed level. 13. Since it is suggested in this scenario to open the economy wider to imports while maintaining temporarily the present level and structure of implicit (i.e., resulting from the present price differentials between domestic and international prices) protection, capital goods and miscellaneous inter- mediate goods imports are likely to grow the fastest. This process has already begun in 1978, when imports of these categories grew by 45 percent (about 34 percent on real terms) compared with 1977. Mexico's demand for these imports is predicted to grow rapidly for three reasons: (a) the very large public sector investments, especially by PE4EX and the CFE, (b) the rapid recovery of the private sector investment, and (c) the planned program of import substitution in capital goods. While at first glance it may seem paradoxical that import substitution would raise import requirements, this is certain to occur in the case of capital goods in Mexico. Given the present low level of development of production facilities for both finished capital goods and their parts and components (see IBRD "Mexico: Manufacturing Sector"; op. cit.), any program of import substitution in this sector will itself require large imports of both "machines to produce machines" as well as parts and components, at least during the initial several years. 14. The foreign borrowing necessary to finance the projected balance of payments current account deficit would push upwards the medium- and long-term public sector debt (MLPD) from US$19.4 billion in 1977 to about US$38 billion in 1982 all in current termks (in 1977 prices the amount of 1982 MLPD would be about US$26 billion, a real increase of about 34 percent over five years). As exports would grow even quicker, the debt service ratio would drop from 0.49 in 1977 to 0.42 in 1982. Even though this is still a high ratio, it is not  - 88 - ANNEX I likely to create major problems as Mexico is one of the few countries offering a tangible "collateral". Thus, in 1982, total public debt would be equivalent to less than 2.0 billion barrels of oil, or to less than 5 percent of proven reserves as of early 1979. (iii) Public Savings 15. The model is solved for public savings, resulting from the preceding endogeneous variables. The level of inflation has been assumed, the level of public investment calculated given the needs of the economy, the level of private and external savings has been inferred from both the past and the anticipated behavioral function. As can be seen from the table on the next page, the real value of public savings will triple between 1978 and 1982. In relative terms, public savings in 1978 represented 3.6 percent of GDP but by 1982 they will increase to 7.2 percent. During the period of 1977-1982 the non-PEMEX public savings remain at about 1 percent of GDP, but PEMEX's savings increase dramatically, by more than 300 percent. 16. Six basic assumptions underlie the projections of non-oil public revenues: (i) A modernization of the indirect taxation system will take place although no major tax reform will be introduced. A strong effort to improve the tax administration and reduce evasion will be made. Direct taxes are assumed to represent about 6.1 percent of GDP; indirect taxes are assumed to increase at a somewhat faster pace than GDP so that their ratio to GDP goes up from 5.0 percent in 1978 to 5.3 in 1982, while non-oil export taxes will behave inversely, declining from 0.8 to 0.6 of GDP during the same period; import taxes would almost double their share in GDP, from 0.5 to 0.9 as a combined result of substantial increases in imports and higher tariffs, which are expected to replace import licenses; (ii) the price of goods and services produced by public sector enterprises, are adjusted annually by the inflation index. The demand elasticity for public goods and services with respect to income is estimated at 1.2; (iii) the General Government wage bill will increase by about 12 percent per annum in real terms starting in 1979; (iv) the number of employees of PEMEX and other enterprises will increase by 8 percent and 3.5 percent per year respectively, starting in 1979 and the real salary adjustments will be limited to overall (aggregate) productivity gains; (v) the Government will be able to capture the entire PEMEX savings either directly through profit transfers or indirectly through additional taxes; and  ANNEX I -89- TABLE 3 MEXICO - PUBLIC SECTOR SAVINGS SCENARIO (A) (1972 Mex$ billion) a/ 1977 1978 1979 1980 1981 1982 Current Revenues 123.7 133.7 157.2 183.2 213.7 238.2 Taxes 83.2 91.2 104.7 119.0 138.2 153.6 PEMEX 7.3 9.2 13.2 16.0 22.5 26.4 Other Public Enterprises 1.2 1.3 1.4 1.6 1.7 1.9 Other b/ 74.7 80.7 90.1 101.4 114.0 125.3 Non-tax 24.9 24.9 27.4 30.3 33.4 36.2 Value Added of Budgetary Controlled Public Ent. 24.1 28.1 39.7 51.5 66.3 76.7 PEMEX 17.7 22.7 33.6 44.3 57.9 67.5 Other enterprises 6.4 5.3 6.1 7.2 8.4 9.2 Current Expenditures 103.5 109.8 121.3 134.8 150.7 168.0 Wages and Purchases of Goods and Services 55.0 58.7 65.7 73.6 82.4 92.0 Current Transfers 25.7 26.8 30.0 33.7 37.7 42.3 State Participations 9.9 9.8 11.0 12.3 13.7 15.4 Other 15.8 17.0 19.0 21.4 24.0 26.9 Interest Payments 17.9 18.6 18.9 20.1 21.9 23.7 Other 5.0 5.7 6.6 7.6 8.7 10.0 Public Savings 20.2 23.9 35.9 48.4 63.0 70.2 Memorandum Items (Domestic Sales-PEMEX) (18.0 (17.9) (19.2) (21.0) (23.1) (25.1) (Dotestic Sales-Other Ent.(30.3) (31.7) (35.0) (39.0) (43.4) (47.7) a/ Deflated by the implicit GDY deflator. b/ These are consolidatedtaxes (excluding taxes paid by the budgetary controlled public enterprises). These taxes are also net of development subsidies. Source: Staff estimates.  - 90 - ANNEX I (vi) current transfers from consolidated public sector to the rest of the economy will grow at about 10 percent in real terms. This rate is slightly higher than that of total current expenditures. The state governments' share in Government taxes will increase at the same pace as indirect taxes. (iv) Oil Production and Exports 17. PEMEX revenues will increase sharply during the current sexenio. These revenues would result from large increases in production and exports of crude oil and related products. Table 7.13 of the Statistical Appendix summarizes PEMEX's production and sales programs for 1978-1982. 18. Proven oil and gas proven reserves were stated by PEMEX management on December 31, 1978 to amount to the equivalent of 40.2 billion barrels of crude equivalent. To this should be added 44.6 billion barrels of probable reserves. The true level of Mexican oil and gas reserves has been a matter of considerable speculation. As recently as the end of 1973 they were stated to be only 5.4 billion barrels. The rapid increase in both reserves and productive capacity comes from the discovery of new oil fields in the southern states of Chiapas and Tabasco which have production characteristics comparable to those of many Middle Eastern fields. This has enabled a very rapid build-up of crude oil production and new reserves could be proven by drilling a rela- tively small number of wells. Prospects for increase of these reserves in the coming years are good. Natural gas reserves are converted into crude oil equivalent at the rate of 5,000 cubic feet equals one barrel of crude oil, and at the end of 1978, it has been reported that gas represented 33 percent of total reserves and crude oil 67 percent. 19. PEMEX intends to produce 4,020 million barrels of crude oil and condensates during the period 1977-1982, of which 3,520 million are expected to come from the southern zone. When viewed in the light of the estimates of reserves, and the high productivity of the new fields, there seems to be no reason to suppose that this target cannot be achieved or even exceeded, if surface facilities to handle the production can be constructed on time. Over the longer term, sustained production at high rates of extraction could lead to a rapid production decline, unless new fields are discovered to replace the existing ones. Present indications are that such discoveries are likely to be made. 20. In order to meet these production targets the Government has authorized a massive and ambitious investment program amounting to US$15.4 billion (1977 prices) to be carried out by PEMEX during 1977-1982. The major doubt with regard to the feasibility of the program is whether PEMEX would be capable of carrying out the program without major construc- tion delays. Such doubt is suggested by the delays which have occurred during 1977-1978. 21. The staff has therefore estimated a revised schedule of PEMEX's production and sales programs, at an assumption that PEMEX will not be able to catch on the delays during 1978-80 but that it will almost reach the target  -91- ANNEX I TABLE 4 : MEXICO - GAS BALANCE (millions of cubic feet/day) 1978 1979 1980 1981 1982 u e Csp PEMEX a/ 638 773 950 1031 1153 CFE b7 244 244 265 265 305 Rest c/ 726 794 882 970 1065 Total 1608 1811 2097 2266 2523 Substitution Coefficients d/ PEMEX .50 .70 .90 .90 CFE .50 .70 .90 .90 Rest - .15 .25 .30 Additional Gas Consumption TEM= ~~~~~~~149 258 353 387 CFE 400 570 746 882 Rest - 132 242 319 Total 549 960 1342 1588 Natural Gas, total domestic consumption PEMEX 638 922 1207 1384 1540 CFE 244 644 834 1011 1187 Rest 814 916 1014 1212 1384 Total 1696 2483 3057 3608 4111 Losses e/ 259 310 340 401 457 Total 1955 2793 3397 4009 4568 Natural Gas, production Dry gas ( minimum) 650 600 120 - - Additional dry gas f/ - - - 129 120 Total dry gas 650 600 120 129 120 Associated gas 1939 2505 3408 3880 4448 Total 2589 3105 3528 4009 4568 Gas Flared 634 312 131 - _ a/ Estimated at about 10 percent of production of refined products (see Table 7, p.95) transformed into cubic feet by using a coefficient of 6783 cc to a barrel of fuel oil. b/ Estimated on the basis of CFE projections. c/ Assumed to grow at 1.2 times the rate of growth of non-oil GOP. d/ See text, paragraphs 23 and 24 of this Annex. e/ Estimated as 10 percent of total gas produced. f/ Necessary to meet demand Source: Staff estimates.  -92- ANNEX I TABLE 5 : HEXICO - DOMESTIC CONSUMPTION OF FUEL AND GAS 1977 1978 1979 1980 1981 1982 A. Fuel Oil, unadjusted (000 bbl/day) PEMEX b/ 37 45 54 58 63 CFE c/ 116 118 120 122 144 Other a/ 98 105 115 127 141 154 Total 258 278 301 321 361 B. Natural Gas, unadjusted (000 bbl/day, fuel oil equivalent) a/ PEMEX 94 114 140 152 170 CFE 36 36 39 39 45 Other d/ 100 107 117 130 143 157 Total 237 267 309 334 372 C. Natural Gas: additional consumption (000 bbl/day, fuel oil equivalent) a/ PEMEX 22 38 52 57 CFE 59 84 110 130 Other 20 36 43 Total 81 142 198 230 D. Other Refined (000 bbl/day) PEMEX 73 89 110 118 135 Other d/ 525 562 617 681 754 823 Total 635 706 791 872 958 Summary of Domestic Consumption (000 bbl/day) Fuel Oil e/ 258 197 159 123 131 Other f/ 635 706 791 872 958 Total 893 903 950 995 1089 Gas g/ 237 348 451 532 602 Total(including gas) 1130 1251 1401 1527 1691 a! From Table 4 , transformed into 000 barrels of fuel oil by dividing by 6783. b/ Estimated at about 4 percent of refinery production on the basis of PEMEX's experience. c/ Estimated on the basis of CFE's demand projections. d/ Assumed to grow at 1.2 times the rate of growth of non oil GDP. e/ Total A less total C. fl Total D. / Total B plus total C Source: Staff estimates  -93- ANNEX I TABLE 6: HIGH OIL EXPORTS PROJECTIONS SCENARIO (A) EXPORTS OF PETROLEUM DERIVATES, 1978-82 1978 1979 1980 1981 1982 I - Volume (000'bbl/day Crude Oil 327 430 554 857 1,069 Refined Products 24 248 417 488 571 - Fuel Oil (9) (169) (235) (307) (341) - Others (15) (79) (182) (181) (230) Petrochemicals (Tons/day) 1,156 1,138 2,582 4,912 5,812 III - Value (US$ million) a/ Crude Oil 1,631 2,400.3 3,253.8 5,255.5 6,829.8 Refined Products 121 1,277.1 2,355.4 2,813.9 3,463.2 - Fuel Oil (35) (775.3)(1,134.3)(1,544.0X1,784.8) - Others (86) (501.8)(1,221.1)(1,269.9)L,678.4) Subtotal 1,752 3,677.4 5,609.2 8.069.4 10,293.0 Petrochemicals 122 132.2 315.6 626.6 772.1 Total 1,874 3,809.6 5,924.8 8,696.0 11,065.1 III - Memorandum Item Crude Oil Price b/ (US$ dollars) 13.7 15.3 16.1 16.8 17.5 a/ For projecting the value of exports it has been assumed that the price relationship among crude oil, fuel-oil and other refined products that prevailed in 1977 would remain constant throughout 1977-1982. It was also assumed that natural gas would be priced on the basis of the equivalent caloric content of crude oil (5,000 cubic feet of Mexican natural gas are equivalent in terms of caloric content to one barrel of crude). The equivalences used are the following: 1 bbl refined product (Other) = 1.142 bbl rrude oil 1 bbl fuel-oil = 0.82 crude oil 1 ton petrochemicals = 20.8 bbl crude oil b/ Price of crude oil assumed conservatively to grow by two-thirds of world inflation rate from 1979 onwards. Source: Staff estimates.  _94- ANNEX I TABLE 7 MEXICO - HIGH OIL EXPORT SCENARIO (A) REQUIRED OIL PRODUCTION AND AGGREGATE BALANCES 1978 1979 1980 1981 1982 Crude (Thousand bbl/day) Total Production of Crude 1,263 1,604 1,949 2,370 2,763 Exports of Crude 327 430 554 857 1,069 Required Refinery Input 936 1,174 1,395 1,513 1,694 Domestic Consumption 893 903 950 995 1,089 Exports 24 248 417 488 571 Losses 19 23 28 30 34 Natural Gas (Thousand bbl/day crude oil) Total Production of Natural Gas 518 621 705 802 914 Domestic Consumption 391 559 679 802 914 Gas Flared 127 62 26 - - Total Hydrocarbons 1,781 2,225 2,654 3,172 3,677 Source: Staff estimates.  - 95 - ANNEX I levels by 1982. Also, the PEMEX's domestic consumption estimates were revised by the staff so as to make tlhem consistent with the GDP projections for the sexenio, prepared in the present report which are higher than PEMEX's original estimates. A detailed analysis of petrochemical production and sales was beyond the staff immediate capacity and therefore the PEMEX program in this area has not been evaluated and the most recent - at the time of the last data discussion. PEMEX projections available were accepted at face value. The results of this exercise are shown in Table 7. 22. PEMEX is projecting that the domestic demand for refined products and national gas will grow at about 7-8 percent per year during 1978-1983. These estimates seem to be quite conservative and the staff revised them upwards to make them consistent with its growth projections and with an estimated elasticity of petroleum products consumption with respect to GDP of about 1.2. Table 5 summarizes the resulting demand forecasts. 23. One major unresolved question, which could have a strong impact on the composition of domestic consumption as well as on the composition and value of exports is the perspective of natural gas exports to the US. The new oil fields of Chiapas and Tabasco have a relatively high proportion of asso- ciated gas dissolved in the oil. There is naturally a strong reluctance to flare (burn as waste) natural gas produced with oil as this implies a resource waste. PEMEX originally proposed to construct a 48-inch diameter gas pipeline from the southern oil fields to the US border, following the Caribbean coastline. This solution would also result in linkage with the now isolated northern and southern gas pipeline systems in Mexico. Exports of gas were projected to be the equivalent of 540 thousand bbl/day of crude by end of 1982. The ongoing price discussion poses numerous problems for PEMEX, anxious to reach oil production targets without excessive flaring of gas. The solution now adopted is to construct the gas pipeline to a geographical point close to the Mexican/US border, and to maximize the use of natural gas internally in Mexico as a substitute for fuel oil, considering that the latter can be more readily exported to distant foreign markets. This is technically possible because most industrial installations and power stations are equipped for dual firing. Since PEMEX has retained the option of switching back to fuel oil and exporting gas to the US if the import price solution is reached, it is at present hardly possible to forecast accurately the mix of refined petroleum products and/or gas which would be exported during the period for which forecasts are made here. A tentative projection of domestic consumption and its structure presented in Table 4, includes both alternatives, i.e., with and without gas exports to the US. 24. Staff analysis suggests that the alternative of exporting non- upgraded fuel oil instead of gas would imply some reduction of export revenue the exact amount of which depends on the constantly changing price relation between these two products. Therefore, the task of upgrading the fuel oil destined for export is highly important. The key assumption behind projections is that Mexican industry can easily substitute fuel oil for gas as fuel. This is a realistic assumption since about 50 percent of the combined domestic demand for fuel oil and gas comes from both  - 96 - ANNEX I PEMEX and CFE (stated owned electricity company) and both could convert to gas almost immediately. It is assumed that PEMEX and CFE will replace 50 percent of their current consumption of fuel oil by gas in 1979, 70 percent in 1980 and 90 percent in 1982. It is also assumed that the rest of the economy will convert to gas at a slower pace, substituting 15 percent of fuel oil for gas in 1980, 25 percent in 1981 and 30 percent in 1982. 25. If the above scenario materializes, there would be no significant excess supply of gas in Mexico during the year following 1979, even assuming a high gas-crude production ratio. Some dry gas would even have to be used to meet demand requirements. As this outcome appears to be the most likely one, its variables were used for the balance of payments,projections. 26. The differences between the volume of exports projected here and the one assumed in PEMEX's calculations apply mainly to 1979 and 1980. There is however, hardly any difference toward the end of the 1979-1982 period. Staff projections show exports of oil increasing from US$1.9 billion in 1978 to US$11.1 billion in 1982. About half of the increase would come from higher volume of crude exported and the other half is almost evenly distributed between fuel oil and other refined products. Petrochemical products, while growing substantially, represent only 7 percent of total exports in 1982, about the same proportion as in 1978. 27. The assumption as to oil prices and the non-achievement of gas exports are deliberately conservative. Therefore oil price increases by more than assumed and/or if gas is exported provide an additional margin of flexi- bility offering, if this becomes true, a possibility to adjust the financial equilibria and to change some marginal aspects of the development strategy accordingly. Scenario B: Higher non-oil exports-cum-increased public savings 28. The main difference between this scenario and the previous one is that imports will increase less rapidly, along the pattern described in para. 30. Because it is assumed that foreign savings remain unchanged, exports and more precisely oil exports will have to increase less than in our previous framework. The reduction in imports is brought about through a reduction in private sector's disposable income, which is in turn achieved by increasing both taxes and prices of goods and services provided by the public sector. At a given level of current public expenditures, this rise in revenues will increase public sector savings. Under the assumption that total savings remain at the same level in both scenarios, such increase will be offset, first, by a reduction in public savings derived from lower oil exports, and second, by a reduction in private savings resulting from the lower disposble income. Therefore, in this scenario, both the sources of public savings and the structure of national savings will be different. 29. The main difference between the two scenarios is that the former relies on the opening of the economy to imports, but there is no increase in real prices for goods and services supplied by the Government and no increase in tax rates, while the present (B) involves increase of public sector prices and tax rates but no rapid and wide opening of the economy.  97 ANNEX I TABLE 8 MEXICO - COMPARISON BETWEEN TWO POLICY SCENARIOS A & B 1979 1980 1981 1982 Imports - Scenario A Volume (Constant 1972 US$ million) Capital Goods 2,367.9 2,919.4 3,368.4 3,671.4 Other Goods anid Non-factor 4,289.5 5,011.7 5,890.9 6,660.9 Services Prices (1972 = 100) Capital Goods 198.9 214.8 228.8 243.7 Other Goods and Non-factor Services 192.2 209.0 223.6 239.1 Value (Current prices, US$ million) Capital Goods 4,710.2 6,271.7 7,706.5 8,945.6 Other Goods and Non-factor Services 8,244.4 10,476.2 13,172.6 15,924.4 Imports - Scenario B Volume (Constant 1972 US$ million) Capital Goods 2,285.1 2,658.9 2,959.1 3,181.1 Other Goods and Non-factor Services 4,221.2 4,622.0 5,060.8 5,475.2 Value (Current prices, US$ million) Capital Goods 4,545.5 5,712.1 6,770.1 7,751.2 Other Goods and Non-factor Services 8,113.1 9,661.7 11,316.5 13,090.1 Total 12,658.6 15,373.4 18,086.6 20,841.3 Difference (A-B) Value (Current prices, US$ million) Capital Goods 164.7 559.6 936.4 1,194.4 Other Goods and Non-factor Services 131.3 814.5 1,856.1 2,834.3 Total 296.0 1,374.1 2,792.5 4,028.7 Total Difference converted to billion Mex$ of 1972, GDY units 1.86 7.83 14.88 20.06 Effect on Public Revenues and Oil Exports (billion Mex$ 1972; GDY units) Required Increase in Public Savings 2.74 11.51 17.10 23.90 (lower exports) (1.86) (7.83) (14.88) (20.06) (lower private savings) (0.88) (3.68) (2.22) (3.84) Required Total Public Revenues 158.9 179.0 204.7 226.4 (Required increase in Public Revenues, Z) (1.7) (6.4) (8.4) (10.6) Required Non-Oil Public Savings 9.3 17.6 26.6 30.3 (Required Non-Oil Public Savings as fraction of GDY, Z) (1.3) (2.2) (3.0) (3.1) Memorandum items - Comparison with high oil export Scenario (A) (in thousand bbl/day) Exports of Crude Oil lower by: 53.0 233.8 455.4 630.7 Total Crude Exports Scenario (B) 377 320 402 488 Total Production of Crude Scenario (B) l/ 1,551 1,715 1,915 2,132 1/ This amount of crude will be needed to implement Scenario B. Nowever, export prices of oil and derivates were calculated following unit prices assumptions pre-dating the shortfall in Iranian supply of oil. The adjustment of oil export prices to current (April 1979) higher forecast would necessarily reduce the overall requirement of production of crude on this Scenario. Source: Staff estimates.  - 98 - ANNEX I 30. The economy would not be considered as being "more widely" opened if imports continue to expand at past long-term rates, with imports of consump- tion and intermediate goods and non factor services maintaining their 1.0 elasticity to GDP and imports of capital goods their 1.0 elasticity to fixed investment. The choice of the benchmark period may influence the projections as well. Thus, Mexico's imports were abnormally high in 1973-1975, but were depressed afterwards during the 1976 and 1977 recession, in turn, to an abnormally low level. It is assumed that import elasticity would decline from a very high 1978 level to 1.5 in 1979 for both capital goods and the other imports. Afterwards, both elasticities will go down to 1.0. Lower imports in this scenario (compared to Scenario A) will entail reduction in exports of oil but the ba'lance of payments deficit would remain unchanged, compared to the preceding scenario and interest and external borrowing will therefore be the same in both these scenarios. 31. Level of imports for both capital and non-capital goods would be quite close in both scenarios in 1979, with only slightly lower imports in Scenario B. From there on differences begin to increase. Capital imports are projected to be lower by about 0.5 billion dollars (constant 1972 prices) and other imports lower by US$ 1.2 billion in 1982 in Scenario B compared to Scenario A. In current dollars these differences are more pronounced with capital imports lower by US$1.2 billion and other imports by US$2.8 billion by 1982. As a result total imports add up in Scenario B to about US$20 billion of current prices in 1982, compared with almost US$25 billion in the preceding Scenario A. 32. This will have repercussions on exports of oil, calculated as necessary to pay for imports in Scenario B. Thus, while the corresponding exports of oil would be lower in Scenario B by only 50 thousand bbl/day in 1979, this difference will grow to about 230 thousand bbl/day in 1980 and to 630 thousand bbl/day in 1982. These lower exports notwithstanding, required (derived from the model for Scenario B) total output of crude would have to reach over 2.1 million bbl/day in 1982. Output of 2.1 million is calculated as necessary to generate exports of crude on the order of 440 thousand bbl/day, necessary, in turn, to pay for (relatively lower) imports under Scenario B. 1/ 33. The reduction in imports, and therefore the lower oil exports also entail a reduction in public savings obtained from exports under the assump- tion that PEMEX's marginal cost of producing crude remains low. Thus this reduction in exports is expressed also in Mexican currency (deflated with the GDY implicit deflator) in order to measure the effect on public finance. The reduction in public revenue must therefore be compensated--if the same GDP and investment growth rate are maintained for both scenarios--through the increases in public revenues collected from domestic sources, such as taxes or increased 1/ This calculation is based on unit prices for crude export pre-dating recent declines of supply related to changes in Iran's production plans. Present higher unit price forecasts (May 1979) would bring down the volume of "crude exports required" by at least 10 percent, compared to that projected above.  - 99 - ANNEX I prices of Government goods. In turn, the increase in Government revenues will have to be preempted from the disposable income of the private sector which, in turn, implies lower private savings. At a given marginal propensity to save (Table 1, page 84), the total increase in public savings can be set at a level at which it would compensate for both the reduction in exports revenues to the Government, and for the reduction in private savings. The total increase in public savings which is required is shown in Table 8, page 97. 32. To raise public savings to the level required by assumptions of Scenario B, at unchanged current public expenditures, it is necessary to increase overall public revenues. Total revenues of the public sector, defined as a sum total of consolidated taxes and the value of domestic sales of both the PEMEX and of the other budgetary-controlled public enterprises, are calculated for Scenario A. Increase in public savings calculated as necessary to meet assumptiorns of Scenario B would call for only 2 percent increase in total revenues in 1979, but for as much as 6 percent increase in 1980 and for 11 percent in 1982. Although this rate of increase (Scenario B over Scenario A) do not appear unattainable, they imply more than a doubling of the non-oil public savings in 1980 and their tripling in 1982 compared to their volume in 1979. If this is attained the GDY share of non-oil public savings would increase from their 1979 level of 1 percent to 3 percent in 1981. It is noteworthy to mnention, however, that non-oil public savings were already at this latter level with respect to the GDY prior to the change in public expenditure policies of the early-to-mid seventies. 33. Increase in taxes and public sector prices is largely a political decision, which could normally be easier taken by a new Administration than by one which has traversed already half its term and which holds increasingly abundant foreign exchange resources at hand. This may lead the reader to believe that Scenario A ("opening wider the economy to imports and higher oil exports") is more likely to materialize. 34. It is worthwhile to mention the possible effect the different levels of imports in the two scenarios would have on the domestic production structure. Imports of capital goods in 1982 in Scenario B are 13 percent below those of Scenario A. For imports of other goods and non-factor services the corresponding percentage is 18 percent. At the aggregate level the differences between the two scenarios are small. The reduction in imports of capital goods in Scenario B compared to Scenario A is equal to less than 3 percent of Gross Domestic Fixed Investment planned for 1982. The reduction in other imports is equal to less than 2 percent of domestic consumption. These differences are relatively insignificant, and therefore the production structures would remain broadly similar in both scenarios in 1982. However, if after 1982 exports continue to increase steeply and would entail further increases in imports or alternatively, if lower exports would call for higher taxes, the differential impact on the structure of production of the two scenarios may become substantial. Also the marginal productivity of capital, as perceived by private investors in terms of net return on their new invest- ments, may be lower in Scenario B because of higher taxes and higher prices of public services. Therefore growth would eventually becomes slower in Scenario B, as the private investment would slow down.  - 100 - ANNEX I 35. The two scenarios outlined in this Annex while related to two differ- ent sets of policies, are not mutually exclusive, and can be merged into a "mix" scenario. This is probably what would happen in reality, as the truth-- and therefore the reality--is usually somewhere halfway between the two extremes. Thus there is a multiplicity of "trade-offs" between the scenarios. For example, the Government may realize that the economy is expanding fast and, in case the economy could not be opened fast enough, inflation would accelerate. To meet targets for inflation, an increase of taxes and of prices for public goods and services would therefore become imperative. In an economy whose financial structure is as open as in Mexico, application of restrictive monetary policies may not be of much help in the situation described above and little would be achieved if, to dampen inflation resulting from a combination of high growth rates of the economy and insufficient inflows of imports, monetary authorities would attempt to reduce the rate of expansion of the domestic credit, simply because as external credit would substitute itself for it instead. Here, the solution may be sought rather in a restrictive fiscal policy, along the lines suggested in Scenario B.  - 101 - ANNEX II TECHNICAL NOTE ON DATA AND ADJUSTMENTS MEXICO: INCOME DISTRIBUTION 1963-68-75 1. This note sets forth basic data on Mexican size distribution of income for the three years 1963, 1968, and 1975, and discusses some problems in comparisons among the three years. The 1977 study was published too late (April 1979) to be taken into consideration in this report. Staff is undertaking the analysis of this latter study and will publish the results in early summer of 1979. 2. The three studies used were chosen because (a) they include the most recent available study; prior to 1975 the latest study is of 1970, (b) the questionnaires and methodology are comparable in most respects, and (c) the studies treat family income rather than personal income, which is better for our purposes. 3. We are now using only aggregated data derived from the basic data files for each study. We hope to have access to the files themselves in the future, but for this report we must rely on presently available information. 4. The basic data of the size distribution of income and expendi- tures are shown in Table 1. A comparison between the total of national dis- posable personal income (DPI) reported in the studies, and the same variable estimated from natonal income accounts data, is shown in Table 2. Among the interesting characteristics of these data are the following: (i) At lower income levels, reported expenditure is often much greater than reported income. In many cases the differences seem too large to be accepted as true and due to dis-saving by families whose income has fallen below their "permanent" income level. (ii) The national totals for DPI implied by the studies are considerably less than that implied by national acccounts data. Moreover, the difference increases with time--i.e. is smallest in 1963 and largest in 1975. 5. These characteristics suggest that income may be underestimated by the studies, especially income of the poor, and that the degree of under- estimation increased with each successive study. If this is so, then neither the situation as of one year nor trends during the period are accurately reflected by the reported data. Under-reporting of income, especially by the poor who receive proportionally more income in kind and/or on a less regular basis, would not be surprising. We have tried two dif- ferent methods of adjustments of the data, to try to reach a set of numbers that would more accurately represent what actually happened. 6. The first method of adjustment tried was in two steps. First, we assumed that for each class, income was at least as great as reported  - 102 - ANNEX II expenditures. This assumption may result in a slight over-correction, because in fact some poor families probably did spend somewhat more than their incomes. However, it may be much closer to the truth than the reported data which shows huge dissaving. Moreover, the adjustment is conservative in the sense that it probably errs in the direction of making the poor seem better off than they actually were, and even so the poor seem very badly off. This step still left a gap between DPI as implied by the studies and as implied by national accounts, and the gap still increased over time. The distribution over income size classes of any adjustment for this gap is highly debatable; part or all of the gap may be due to deliberate under-reporting by the well- to-do; part of it may be due to additional under-reporting by the poor. We tentatively decided against adjusting for the entire gap in each year, but rather only enough to make the size of the gap in 1968 and 1975 the same as in 1963 (in proportional terms). Thus, there may well be under-reporting in the adjusted data, but the degree of under-reporting may be the same in each of the three years. This can be seen in the sketch below, in which DPI is shown (a) as reported in the studies, (b) as implied by the national accounts, and (c) after adjusting the study results. The gap after adjustment, (b)-(c), is the same proportion of (b) in each year. / ~~~(b) k(c) (a) 1963 1968 1975 As to the distribution of the adjustment across income classes, we again chose a conservative assumption and distributed it across all income classes proportionately. 7. The results of this (two-step) first adjustment are shown in Tables 3 and 4. It turns out that the first step--increasing incomes of the poor to equal their reported expenditures--was quantitatively most important in 1963. If the adjustment created a larger overestimate of the share of the poor in 1963 than in 1975, it exaggerated the deterioration of the incomes of the poor over time. Since this deterioration is shown in the results to be so large as to be hard to accept, we tried a second type of adjustment.  Table 1: MEXICO: BASIC DATA ON SIZE DISTRIBUTION OF FAMILY INCOME: 1963, 1968 AND 1975 1963 1968 -19 75 Income Size Number of Total Reoorted Income Size Number of Total Renorted lumber of Total Reported Class Families Income Exoenditure Class Families Income Exoenditure Families Inome Eendi (pesos/month) (millions) m7Tillion pesos/month) (nesos/month) (millions) (million pesos/month) (millions) (eillionsos/month) 0-300 1.35 290.2 636,3 0-300 0.69 173. 243.2 1.04L 132a8 213.1 3ol-600 1.8h 805.9 1,11V.J 301-400 0.52 181.8 245.7 1.31 492.7 503.3 601-1,000 1.58 1,231.1 1,460.5 401-530 0.70 323.7 399.4 1.30 1,140.7 1,152.5 1,001-1,500 0.80 990.2 1,137.7 531-700 0.72 437.7 495.9 1.02 1,319.3 1,318.5 1,501-3,000 1.10 2,309.1 2,260.8 701-950 1.16 959.0 1,096.9 1.03 1,830.4 1,785.1 3,001-4,500 0.33 1,218.1 1,110.3 951-1,250 0.94 1,034.8 1,157.5 0.94 2,119.7 2,121.0 H 4,501-6,000 0.1h 729.2 555.B 1,251-1,700 0.92 1,338.9 1,439.3 0.95 2,741.8 2,761.3 6,001-10,000 0.12 924.1 692.1 1,701-2,200 0.75 1,459.3 1,577.3 3.96 3,596.9 3,668.5 10,001 and over 0.07 869.6 469.3 2,201-3,000 o.64 1,636.5 1,580.9 1.02 5,766.1 5,580.3 3,001-4,D00 0.44 1,514.1 1,442.1 0.47 4,107.0 4,033.8 7.33 9,367.4 9,509.8 4,001-5,200 0.29 1,320.8 1,154.6 0.47 9,936.2 9,298.0 5,201 and over 0.49 4,974.5 4,032.9 8.28 15,318.4 14,867.7 10.21 33,283.6 32,435.1 Sources: Encuesta Sobre Ingresos y Gastos Familiares en IMexico - 1963, Banco de Mexico. Estudo de Ingres-os y Gastos de las Familias (1968), Secretaria de la Presidencia, Direccion Gener Economnica .y Social, July 197h Encuesta de Ingresos y Gastos Familiares (1975), Secretaria del Trabajo y Prevision Social, Centro Nacional de Informacion y Estadisticas del Trabajo, 1977.  - 104 - ANNEX II 8. The second adjustment method tried was to correct only for the increase in the DPI gap over time, and to distribute the adjustment propor- tionally over all income classes. It seems likely that this method under- estimates the share of the poor, but may estimate trends over time more accurately. The results are shown in Table 5 and 6. 9. The adjusted data in Tables 5 and 6 probably understate family income for the nation as a whole. However, we see no strong reasons for believing that the degree of understatement differs from year to year. The sharp deterioration at the lower income levels is difficult to accept as accurate, but we have not yet found a valid way to re-estimate the data. The adjustments underlying Tables 3 and 4 seem highly questionable and do not resolve the problem. We shall therefore use the results shown in Tables 5 and 6, until time and access to the original data permit further attempts at valid adjustments.  *-105- ANNEX II Table 2 MEXICO: ESTIMATES OF DISPOSABLE PERSONAL INCOME (million pesos per month) 1963 1968 1975 1. Private Consumption (NIA) 12,154.4 20,643.8 52,111.4 2. Private Savings (IBRD) 1,231.5 1,715.3 1,561.2 3. (1) + (2) = Disposable Personal 13,385.9 22,359.1 53,672.6 Income 4. Reported income (HBS) 9,367 15,318 33,284 5. Reported income plus dis-saving (HBS) 10,471.5 16,103.1 33,406.9 6. (3) / (4) 1.43 1.46 1.61 7. (3) / (5) 1.28 1.39 1.61 Sources: line 1: National Income Accounts 2: Bank staff estimates based on household budget survey data 4 & 5: Household Budget Surveys  -106- Table 3 ANNEX II MEXICO: STIMARY INDICATORS OF INCOME INEOIJALITY (First Adjustment Methodology) a' 1963 1968 1975 Percent of disposable personal income received by: highest 5% of households 25.7 26.5 30.6 highest 20% of households 52.4 55.4 60.2 lowest 20% of households 6.7 4.6 2.1 lowest 40% of households 15.7 12.4 8.4 Ratio of shares, highest 20% /'owest 20% 7.8 12.0 28.7 b/ Gini index b/ .436 .488 .552 Theil index .349 .430 .545 Hlouseholds with real income less than 1975 minimum wage: number of households (million) 4.8 3.9 5.1 percent of all households 65.8 47.7 49.9 Average real income of poorest 20% (1975 dollars per year per household) 812 724 340 a/ Incomes as reported in surveys for each of the three years have been adjusted in two steps: (1) Incomes were increased to equal reported expenditures for groups that reported expenditures higher than incomes; (2) Incomes in 1968 and 1975 were increased so that the national totals for all three years had the same ratio to total personal disposable income as implied by the national income accounts in the three years. The increases were distributed proportionally to incomes, over all income classes. The tremendous deterioration of the situation of the poor shown here is difficult to believe. Further investigation of the data seems to be needed. b/ Calculated from decile distributions. Source: Mission estimates.  Table 4: MEXICO: INCOME DISTRIBUTION BY DECILES: 1963, 68 AND 75 (First Adjustment Methodology) Percent of Income Cumulative % of Income Decile 1963 1968 1975 1963 1968 1975 1 3.30 2.28 0.62 3.30 2.28 0.62 2 3.43 2.29 1.50 6.73 4.57 2.12 3 4.32 3.37 2.59 11.05 7.94 4.71 4 4.66 4.46 3.66 15.71 12.4o 8.37 5 5.62 5.50 5.00 21.33 17.90 13.37 6 6.51 5.30 6.52 27.84 23.20 19.89 7 8.21 10.00 8.51 36.05 33.20 28.40 8 11.54 11.39 11.42 47.59 44.59 39.82 9 15.05 15.37 16.83 62.64 59.96 56.65 9.5 11.68 13.58 12.78 74.32 73.54 69.43 10 25.68 26.46 30.57 100.00 100.00 100.00 Source: Mission estimates from household survey data. I>  -108- ANNEX II Table 5 MEXICO: SUMMARY INDICATORS OF INCOME INEOITALITY (Second Adjustment Methodology)-/ Predicted- 1963 1968 1975 for 1970 Percent of disposable personal income received by: Highest 5% of households 28.7 27.9 30.7 n.a. Highest 20% of households 58.8 58.3 60.4 52.9 Lowest 20% of households 3.5 3.4 1.9 3.5 Lowest 40% of households 10.3 10.5 8.1 11.5 Ratio of shares, highest 20%/Lowest 20% 16.7 17.3 32.5 15.1 Gini index c/ .526 .521 .557 Theil index c/ .492 .488 .555 Households with real income less than 1975 minimum wage: Number of households (millions) 4.6 4.0 5.1 Percent of all households 63.2 48.6 L9.5 Average real income of poorest 20% (1975 dollars per year per household) 381 477 266 a/ Incomes as reported in surveys for 1968 and 1975 were increased so that national totals for all three years had the same ratio to total personal disposable income as implied by national income accounts in the three years. The increases were distributed proportionally to total incomes, across all income classes. b/ Calculated from Ahluwalia (1976), Table 3, equations (a). c/ Calculated from decile distributions. 1963, 1968, and 1975 data are Mission estimates from household survey data.  Table 6: MEXICO : INCOME DISTRIBUTION BY DECILES: 1963, 68 AND 75 (Second Adjustment Methodology) Percent of Income Cumulative percent of Income Decile 1963 1968 1975 1963 1968 1975 1 1.68 1.19 0.39 1.68 1.19 0.39 2 1.85 2.18 1.47 3.53 3.37 1.86 3 3.o6 3.05 2.56 6.59 6.42 4.42 4 3.69 4.09 3.67 10.28 10.51 8.o9 5 4.89 5.12 5.02 15.17 15.63 13.11 6 6.16 6.48 6.54 21.33 22.11 19.65 7 8.00 8.31 8.48 29.33 30.42 28.13 8 11.83 11.32 11.46 41.16 41.74 39.59 9 16.98 16.16 16.87 58.14 57.90 56.46 9.5 13.12 14.24 12.83 71.26 72.14 69.29 10 28.74 27.86 30.71 100.00 100.00 100.00 Source: Mission estimates from household survey data.  - 110 - ANNEX III TECHNICAL NOTE ON THE ENERGY SECTOR: POWER, ANALYSIS OF MARGINAL COST 1. The following is an approximate determination of the long-run marginal cost of supply. It is based upon CFE's cost estimates for the program and the projected growth in demand. The method used is that described in World Bank Staff Working Paper No. 259 as the average incremental cost method. 2. As shown in the attached table, incremental generation, transmission, distribution and other investments for the years 1977 through 1985 were deter- mined from the investment program. Also tabulated are the projected incremental non-coincident maximum demands. All figures in these columns were then dis- counted at 12% to give their present values. The total present worths of the incremental capital costs were then divided by the total "present worth" of the incremental demands to determine the capital cost per kW. These values were then multiplied by the capital recovery factor for 12%/33 years to deter- mine the annual investment cost of meeting one kW of demand (although some assets will have longer lives than 30 years, others will have shorter lives so this average assumed life as accurate enough for this purpose). 3. The last column of the table shows the sector's incremental operating costs (excluding fuel and depreciation) for the same years. These costs were also discounted at 12% and totaled. 4. The long-run marginal cost of an average kWh was then calculated as follows: the capital component was determined by dividing the sum of the annual costs of the marginal capacity by 4580 hours (the ratio of sales to total non-coincident demand in 1976): 2,546 Mex$/kW 55.5J/kWh 4,580 hours 5. The incremental operating cost (other than fuel was determined similarly: 710 Mex$/kW = 15.5J/kWh 4580 Hours 6. Finally the fuel cost of the kWh was calculated on the assumption that all incremental generation would be thermal (there is never any surplus hydro energy in the system). Based on last year's sales and generation figures, 38.2 tWh and 448 tWh respectively, total losses are a out 18X. It was assumed that fuel would continue to cost CFE Mex$260 per m and that the average fuel consumption would be 230 g per kWh. The resulting fuel cost of marginal energy would hence be about 8W/kWh. 7. The total marginal cost would then be 79i/kWh as compared to the average sales price of 42.5i/kWh in 1976 and an estimated selling price of  1 MEXICO: CFE INVESTMENTS: AVERAGE INCREMENTAL COST Incremental Incremental Operating