Mexico in transition : towards a new role for the public sector

R port No. 8770-ME Mexico in Transition: Towards a New Role for the Public Sector May 22, 1991 Country Operatiorls Division I Country Department 11 Latin America and the Caribbean FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted Jistribution and may be used by recipients . the ,.,sfame o~th~ifigg contents mayTW0*MWse be disclosed without World Bank authorization.  FOR OMFICIAL USE ONLY May 22, 1991 MEXICO IN TRANSITION: TOWARDS A NEW ROLE FOR THE PUBLIC SECTOR This report has been written by Daniela Gressani, John Johnson, Santiago Levy, Daniel Oks, David Pearce and Sweder van Wijnbergen. Robert Kanchuger and Sergio Margulis contributed to the revision and updating of the environmental and public investment chapters. The work took place under direction of Sweder van Wijnbergen. This document has a mestricted distribution and may be used by recipients only in the perfofmance of their official duties. Its contents may not otherwise be disclosed without Wortd Banki authoriztion.  2 EXECUTIVE SUMMARY: PUBLIC POLICY AND PRIVATE SECTOR INCENTiVES . . . . . L I INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 II PUBLIC POLICY AND NACROECONOMIC STABILITY . . . . . . . . . . . . . 5 1 Historlcal B&ckgrnd . . . . .. . . . . . . . . . . . . . . . . . 5 1.1 Macrosconomic Developments ...... . . ..... 5 1.2 The Process of Structural Reform . . . . . . . . . . . . . . 6 1.3 The Debt Restructuring Package of 89/90 . . . . . . . . . . . 6 1.3.1 Rationale . . . . . . . . . . . . . . . . . . . . . 6 1.3.2 Terms of the Agreement . . . . . . . . . . . . . . . 7 1.4 The 89/90 Debt Package and the Recovery of Economic Growth . 8 1.5 Oil Prices and the Macroeconomic Situation . . . . . . . . . 10 1.5.1 The Real Exchange Rate . . . . . . . . . . . . . . . . 12 1.5.2 Oil Prices and the Case for an Oil Stabilization Fund . 13 2 Fiscal Consistency. Debt ffanageMent and the Stabilization Poexm . 14 2.1 Consistency of Fiscal Policy . . . . . . . . . . .. 16 2.2 Price Controls, Exchange Rate Policy and the Success of the Anti-Inflation Plan . . . . . . . . . . 18 2.2.1 The Pacto and the D.e-line in Inflation slnce 1987 . . 18 2.2.2 The PECE inL 1990: where to go from now? . . . . . . . . 23 2.3 Debt Management, Interest Rates and Fiscal Policy . . . . . . 25 2.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . 25 2.3.2 External Shocks, Fiscal Adjustment and Domestic Debt Management. . . . . . . . . . . . . . . . . . . . . . . 27 2.3.3 Policy Credibility, Devaluation Expectations and Default .... . . . . . . . . . . . ..... ... . 29 2.3.4 Domestic Debt Management: Indexation and the Maturity Structure . . . . . . . . . . . . . . . . . . . . . . . 34 2.3.5 Summary and Conclusions . . . . .. . . . . . . . . . . 37 III PUBLIC POLICY FOR WHERE PRIVATE MARKETS FAIL ... . . . . . . . . 39 3 Public Polle towards Poverty Alleviation . . . . . . . . . . . . . 40 3.1 The Extent of Poverty in Mexlco: Concepts and Measurement . . 40 3.1.1 Concepts . . . . ... . . . . . . . . 40 3.1.2 Measurement of Poverty . . . ............. . 42 3.2 A Quantitative Assessment of Poverty in Mexico . . . . . . . 47 3.2.1 Data: the Income-Expenditure Survey of 1984 . . . . . . 47 3.2.2 Quantlfying the Lines of Moderate and Extreme-Poverty 48 3.2.3 Socio-economic Cha:acteristics of Households . . .. . 50 3.2.4 Estimates of Moderate and Extreme-Poverty . . . . . . . 51 3.3 Determinants of Poverty .... . .......... 56 3.3.1 Problems of Rural and Agricultural Development . . . 56 3.3.2 Urban Bias.. . . . ...... .... . 62 3.3.3 Macroeconomic Policy . .. . . . . . .... . 63 3.4 Policies for Poverty Alleviation . . . . . . . . . . . . . . 64 3.4.1 Needs, Behavior, and Policy.. . . . 64 3.4.2 Determinants of Intervention in Poverty Alleviation . 67  3 3.4.3 Objectives in Poverty Alleviation ... . . . 69 3.4.4 Policies for the Extreoaely-Poor ...... . . 71 3.4.5 Development Policies for the Poor . ... . . . . 77 3.5 Government Programs for Poverty . . . . . . . . . . . .... 81 3.5.1 Description of Current Programs . ... . . . . . 81 3.5.2 Preliminary Assessment . . . $. . . . . . . . . . . . 84 3.6 Summing Up . . . . . . . . . . . . . . . . . . . . 88 4 Public Policy and the Enironment inMexico . . . . ..... . . . 90 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . 90 4.2 State of the Environment . . . . . . . . . . . 91 4.2.1 Air Pollution . . . . .* . . . . . . . . . . . . 92 A Emissions and Concentrations .. ..... . . . . 92 B Impacts .......... . .......... . 93 4.2.2 Water Pollution ..94 A Discharges and Water Quality . .94 B Impacts . .95 4.2.3 Solid Waste ..95 A Sources .......... . ......... . 95 B Impacts ..... . . . ...... . . . . . . . 95 4.2.4 Soil Erosion ... . . . . . . . . ...... . . . 96 A Extent of Soil Erosion ... ...... . . . . 96 B Impact ..... . . . . ..... . . . . . . . 96 4.2.5 Deforestation . . . . . . . . . . . . . . . . . . . . . 97 4.3 Instruments for Public Policy on the Environment ..... . 97 4.3.1 The Environment and Property Rights ... ...... . 97 4.3.2 Property Rights Failures in Mexico ... ...... . 98 4.3.3 Intervention Failure . . ...... . 99 4.3.4 Examples of Intervention Failures in Mexico ... 100 A Deforestation ..100 B Water . .100 4.3.5 Choosing Public Policy Instruments . . .100 A Command-and-Control .... . . ...... . . . 100 B Market-Based Incentives ... ...... . . . . 101 C Prices versus Quantities: Tradeable Permits . 102 D Prices versus Quantities: Charges and Taxes . 104 E Experience of Charges in OECD Countries .... . 104 F Market-Based Incentives versus Command and Control ..... ..107 G Market-Based Incentives and Natural Resources . . 108 H Private and Social Cost Pricing . .109 4.4 Choosing Public Policy Instruments .... . ....... . 112 4.5 Institutional Factors in Mexico . . . . . . . . . . . . . . . 114 4.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 118 5 Selected Issues in Public Investment .... . . . . . . . . . . . 120 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . 120 5.2 Infrastructure Priorities ... . 120 5.2.1 Recent Trends in Public Expenditure . . . . . . . 120 5.2.2 Repercussions of the Expenditure Pattern ... 122 5.2.3 Resource Mobilization . . . . .. .122 5.2.4 Expenditure Strategy . . . . . . . . . . . . . . . . . 124  4 5.2.5 Strategic Scenario 1990-94 . . . . . . . . . . . . . . 125 5.2.6 Issues for Future Exploration . . . . . . . . . . . . . 125 5.2.7 Recommendations . ... . . . . . . . . . . . . . . . . . 126 5.3 Environmental Priorities . . . . . . . . . . . . . . . . . . 126 5.3.1 Budgetary Expenditures . . . . . . . . . . . . . . . . 127 5.3.2 Sector Requirements . . . . . . . . . . . . . . . . . 128 5.3.3 Mobilization of Resources for the Environment . . . . 129 5.4 Social Sector Priorities ...... . .. . . .. . . .. . 131 5.5 Conclusions and Recovmendations . . . . . . . . . . . . . . . 133 ANNEXES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 1 Statistical Tables . . . . . . . . . . . . . . . . . . . . . . . . 136 2 Fiscal Consistency .19 3 An Econometric Model of Wage-Price Determination for Mexico . . . . 31 3.1 The model of wage determination ...... . ........ . 31 3.2 Stability of the estimated model of wage determination . . . . 33 3.3 The model of price determination . . . . . . . . . . . . . . . 35 3.4 Data Appendix ............. .... .... ... . 38 4 Macroeconomic Stability, External Balance and Growth: A Quantitative Framework ........ ......... 43 4.1 Purpose and structure of the Model . . . . . . . . . . . . . 43 4.1.1 Model Structure. . . . . 43 4.1.2 The Real Exchange Rate, Interest Rates and the Current Account . . . . . . . . . . . . . . . . . . . . . . . . 44 4.2 An empirical Application to Mexico . . . . . . . . . . . . . 46 5 Fiscal Policy. Real Interest Rates and Debt Managee . . . . . . 51 5.1 Consolidated Domestic Public Debt, Operational Fiscal Deficits, and Finance Sources: Methodology of Calculation . 51 5.2 Risk Premia Breakdown: the Formal Framework . . . . . . . . . 55 5.3 A Model of the Probability of a Devaluation and Exchange Rate Risk. Econometric Results . . . . . . . . . . . . . . . 57 5.4 Explicit Debt Repudiation. Econometric results . . . . . . . 63 5.5 A Model to Evaluate Debt Maturity. Econometric Results . . . 63 6 Poverty: Concepts and Measurement ..67 6.1 Operational Measures of Poverty . . . . . . . . . . . . . . . 67 6.2 Tables . . . . . . . . . . . . . . . . . . . . . . . 69 7 Public Sector Investment Needs . . . . . . . . . . . . . . . . . 81 7.1 Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . 81 7.2 Tables . . . . . . . . . . . . . . . . . . . . . . 81 8.3 Boxes . . . . . . . . . . . . . . . . . . , . . . . . . . . . 109 8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114  EXECUTIVE SUMAORY: PUBLIC POLICY AND PRIVATE SECTOR INCENTIVES I INTRODUCTION 1. Mexico has, over the past five years, made major strides towards a complete restructuring of its economy. Five years ago, international trade was fully controlled via quantitative restrictions, removing any threat of foreign competition to domestic producers. Domestically as well most of the economy operated under restrictive government regulation. Since then, trade has beer liberalized to such an extent that Mexico has become one of the most open economies in the world; and domestically, the two key linkage sectors in the economy, transport and telecommunications, either have been deregulated (transport) or are in the process of being deregulated on a tight time schedule (telecommunications). The financial sector, which had ceased its intermediary functions and was in effect operated as a source of public sector revenues as recently as 1988, has been completely restructured and is about to be privatized. As the private sector incentive structure improved, thk public sector substantially reduced its active involvement in the economy itself. Between 1.982 and 1989, non-financial public expenditure fell from 32.52 of CDP to under 22X of GDP as part of one of the most far-reaching fiscal reform programs implemented anywhere on a sustained basis. Moreover, in a still ongoing process of privatization, more than two thirds of all state enterprises have been shut down or spun off, while many of those that remain are increasingly run on a strict for-profit basis. 2. The process of reform is by no means complete. Some sectors (asxcb as agriculture) are only now beginning to be exposed to the new system or incentives; in others problems of implementation remain, like in industry; while some (for example the laws covering foreign investment) have not yet been reformed in any fundamental way. Moreover, there still is a tension between the ongoing macroeconomic stabilization program (the Pacto de Solidaridad, later rebaptized PECE) and the reform process. The Pacto has introduced considerable relative price rigidity in an attempt to slow down absolute price increases, while the reform program requires relative price flexibility. Thus much remains to be done. 3. However, it is fair to say that the main steps have been taken, and an irrevocable process of private sector oriented reform set in motion. This has changed the rules of the game for the private sector. But by the same token, it has changed the role the public sector has to play in the economy. In many areas, it now has a supportive role to play only, restricted to providing macroeconomic stability and a predictable and rational incentive structure for the private sector. In other areas, such as the envlronment, education, health, poverty alleviation or infrastructure, market forces cannot be relied upon to provide a socially acceptable outcome, or in fact even an economically efficient one. In such areas, a more active role for the public sector is warranted. Thus a comprehensive review of the role of the public sector within this new, private sector based environment is called for, and that is what this report attempts to do. 4. Rationales for public sector policy fall under different categories, and the report is subdivided accordingly. First, benefits of microeconomic reforms  it will fall far short of their potential if there is no macroeconomic stability. There is more to this argument than just the higlh relative price variability that research has now conclusively shown always accompanies high inflation. Expectations of major policy changes in the foreseeable future have an equally disruptive effect on private sector resource allocation, predominantly because of the negative effect of uncertainty on private sector investment. Thus an analysis of macromanagement needs to focus not just on the current success of stabilization efforts but on their sustainability in the future. Moreover, of equal importance are measures the public sector can tale to reduce its vulnerability both to actual shocks and private perceptions of potential shocks. This is the subject of part II of this report. 5. Second, there are a number of areas where private sector activities, even within a competitive environment, lead to inefficiencies and/or undesirable inequities. Some of these areas are conventional and not in need of further scrutiny (the standard public goods arguments for police protection, a judicial system or national defense). Others have either come to the foreground more recently, such as environmental problems, or have taken on a new urgency, such as maintenance of public sector infrastructure, or poverty alleviation in the face of renewed growth. Failure to act in these areas may in the longer term trigger economic and social costs substantially in excess of the cost of policies designed to remedy these problems now. An equally important threat to the long term success of Mexico's new development strategy is non-economic. It is an open issue whether the social consensus behind this strategy can be maintained if the current and future benefits are shared as unequally as they have been in the past. 6. Comprehensive coverage of all such areas of potential public sector policy would be too ambitious an agenda for the current study. We therefore single out what seem to be the three most important issues for discussion in Part III. Note that, given the yet scarce knowledge on these issues, the scope of this part of the report is largely exploratory and, often, only introductory. Further work in these areas is clearly warranted. II PUBLIC POLICY AND MACROECONOMIC STABTLITY 7. A prerequisite for macroeconomic stability is policy consistency: policies that are mutually inconsistent clearly signal future policy change, thus precluding macro stability. On this score, Mexico has done extremely well. The fiscal policy underlying the stabilization program is if anything more restrictive than strictly required to achieve the stated inflation targets at stable or falling debt/output ratios. This is confirmed by the rapid decline in the real value of the domestic debt over 1990 and the improvement in external debt indicators over the past few years. Moreover, at least as important is the manner in which this adjustment has been achieved. Mexico has avoided short-term stopgap measures to achieve its fiscal goals. It has completely streamlined and restructured its tax system, while at the same time improving its administration and collection efforts. Subsidies have been cut dramatically, while even the cutback in public investment is not as unsustainable as it might seem on the surface. The admittedly selective review of the public investment program in this report suggests that Mexico can in fact achieve its aims within the public investment program budgeted for  iii the future. Thus Mexico's reform program has been based on policies that are clearly sustainable and compatible with a restoration of economic growth. Economic growth has in fact taken off, with GDP growth close to 5X in the latter half of 1990, although some caution is called for in interpreting short term data. 4 8. Macroconsistency, e" hen achieved with clearly sustainable measures, could still fail to restore investors' confidence if the program is seen to be vulnerable to internal or external shocks. The dominant sources of volatility in Mexico have been international oil prices and interest rates on domestic debt. On both scores, the GoM has substantially reduced its vulnerability. Most of the 1990 oil windfall windfall gains have been set aside in an Oil Contingency Fund (OCF), to be used only in the event oil prices fall below the budgeted $17 per barrel for 1991. Revenues from privatization have also been set aside in this OCF, with further additions planned from the revenues expected from privatization in 1991. Vulnerability to swings in domestic interest rates has also been reduced substantially as the Government has lengthened the maturity of its domestic debt and is rapidly reducing the amount (at least in real terms). 9. Mexico thus seems poised for growth, with a well designed, stable and sustainable set of macroeconomic policies in place to support this process. That does not mean that there are no risks left. The strong recovery of private investment, and, of more concern, of private consumption, has left Mexico with a widening deficit on the current account, in spite of the decline in public sector deficits. Some of the increased investment comes from abroad through increased DFI and thus in a sense brings its own financing. However much of the rest has been financed out of returned flight capital. The dramatic return of Mexican money held abroad (between US$6 billion and US$10 billion in 1990 on some estimates) is of course a signal of private sector's confidence in Mexico's economic recovery. It does however constitute a very volatile source of financing. Unless private savings recovers strongly in Mexico, finding more stable sources of foreign financing will remain a high priority. III PUBLIC POLICY FOR WHERE PRIVATE MARKETS FAIL (A) Income ineaualitv and uovertY 10. The purpose of this chapter is to: (i) provide a broad overview of the structure of poverty in Mexico; (ii) review, in the light of this survey, the objectives of public sector policy in this respect; (iii) review current -policies judged against those objectives; and (iii) identify important short and medium-term issues that current policies do not yet address adequately. 11. The first section starts with an identification of the poor in Mexico, providing as detailed a classification as possible in terms of regional location, asset ownership, sources of income for the different type of poor people and other characteristics. To sum up the conclusions, poverty is found to be predominantly a rural issue. Most of the extremely poor live in rural areas; the poorest of the poor live in rural areas; the extremely poor have very large families, the highest dependency ratio and the lowest educational  iv levels; most of the extremely poor are engaged in agricultural activities. Thus poverty is inextricably bound to agricultural policy. Second, for each of the relevant categories of poor people, it discusses reasons for low earnings: lack of land, lack of human capital, cropping patterns and ability to bear risk, etc.. A possiblv surprizing finding is that policies that reduce agricultural efficiency and create an urban bias are found to often also have a negative impact on income distribution. Many of the rural poor are poor because market imperfections and, sometimes, policies implemented in the past reduced the rate of return on the assets they own, such as unskilled labor and/or rain-fed land. Thus agricultural reform, while no substitute for an effective poverty alleviation effort, has to be arn integral part of such a program. 12. The next section attempts to outline acceptable objectives of social policy in this area and review the Government's objectives from this perspective, all against the light of the poverty profile sketched before. In particular, the long term objective presumably is to improve the opportunities open to the very poor. But do income transfers and attempts at improving the opportunity set open to the poor work at cross purposes? Could there be a conflict between short term poverty relief and long term attempts to deal with the root causes of poverty? Are there segments of the population that really cannot be reached anymore by programs oriented at long term increases ia earning power, so that there has to be a transfer component even in the long term? How can the incentive costs of such programs be contained? Is there, in addition, a need to deal separately with shorter term increases in poverty due to temporary (but as the recent past has shown, possibly quite drawn out) fluctuations in the economy? The section ends by drawing up a series of objectives of an acceptable medium term anti-poverty program. 13. Finally, policies and instruments that try to work at the relevant margin to increase poor people's earnings potential are discussed against the background of these objectives. This section analyzes a set of economically desirable short and medium-term policies to alleviate poverty. It also provides a brief discussion of administrative and implementation problems, with the goal of identifying a subset of desirable and feasible policies. It then compares these policies with current programs, most of which are under the umbrella of the new National Solidarity Program (Pronasol), the Government's main vehicle for its anti-poverty drive, and draws conclusions for future policies towards poverty alleviation. Conclusions: Poverty Alleviation 14. The government's recent efforts at poverty alleviation are very welcome indeed. In spite of the stringencies imposed by the macroeconomic stabilization program, more resources ore being channeled to the poor. This is a significant achievement. The programs through which resources are being channeled, moreover, appear to be, on the whole, well designed. The replacement of generalized price subsidies by targeted programs, and tho emphasis on social and infrastructure investments in poor regions are all effective means to reach and help the poor. 15. There are, however, improvements that can be made within this general  v effort. First, a key point made in this report is that there Is a substantial group, labeled the extremely-poor, whose wntritional and health status ts such that they are not in a position to function adequately in society and take advantage of the opportunities available to them. These groups cannot be reached effcctively by most of the general programs, which operate through improving the opportunities open to the poor. To make these programs effective for the extremely poor also, the nutritional and health status of the extremely poor needs to be improved first. For this, programs that exploit the complementarities between health, nutrition, education and information provision need more attention. Targeting of direct benefits should, through such programs, focus more on the extremely-poor. 16. Second, there remains an imbalance between urban and rural areas; rural contribution to poverty exceeds by a substantial margin the share of overall resources devoted to poverty alleviation that goes to rural aroas. That does not mean there is no significant urban poverty problem; the report documents the existence of substantial urban poverty too. Thus thls imbalance should preferably boe resolved by channeling more resources to rural areas, not by reducing them in urban areas. In addition, the effectiveness of the resources channeled to the rural areas could probably be enhanced. In particular, programs that exploit the complementarities between health, nutrition, education and information provision deserve more attention, since in rural contexts such complementarities cannot be achieved by simply transfering income to the extremely-poor. And exploiting such complementarities is in particular important in policy design focused on reaching the extremely poor. A step in the right direction has in fact been made: a pilot project focuslng on joint delivery of nutrition, health and education benefits to a particularly vulnerable group, pregnant or lactating mothers, is about to be imlemented. The experience with this pilot project should help assessing the feasibility of the approach recommended in this repurt. 17. Finally, there is a legitimate question whether, given Pronasol's resource and administrative constraints, its attention should be more focused on its key programs than is currently the case. 18. In addition to its own significant merits, a high visibility program can be very valuable in strengthening the social consensus behind other structural changes, and can be a central element in showing that those changes will not be allowed to increase poverty. But the risks of such a program deserve mention as well. A key risk may derive from the impression that Pronasol, by itself, is sufficient for poverty alleviation, and that other policies can be freed from poverty considerations. But given the extent of poverty in Mexico, investing one or one and a half percent of GNP on Pronasol will not have a sufficient impact. Therefore the focus sbould not lie only In allocsting larger resources to Pronasol, but also in directly confronting the roots of poverty. This report has argued that primary poverty in Mexico is the result of institutional structures and interventions in resource allocation mechanisms that generate market outcomes which undervalue the resources owned by the poor. Over the medium term, the greatest impact on poverty alleviation could very well derive from institutional reforms removing the biases that currently tilt market outcomes against the poor. Efforts at reducing  vI generalized price subsidies, at reforming rural regulations, at eliminating urban bias, and at macroeconomic stabilization have been at the center of the government's agenda over the last few years with, quite naturally, varying levels of accomplishment so far. But unless the reforms to these primary determinants of poverty are pursued further, Pronasol's e^tivities, while undoubtedly very valuable, will, so to speak, be like 'swimming against the current'. Only when policies for the poor be