Report No.25991-AR Argentina Reforming Policies and Institutions for Efficiency and Equity of Public Expenditures September 10, 2003 Argentina, Chile, Paraguay & Uruguay Country Management Unit Poverty Reduction and Economic Management Unit Latin America and the Caribbean Region Document of the World Bank FISCAL YEAR January 1 - December 31 CURRENCY EQUIVALENTS Currency Unit - Peso($) EXCHANGE RATE 1992-2001 US$1 = $1 US$1 Equivalents 2002 2003 January 1.40 3.24 February 2.15 3.18 March 2.95 2.87 April 2.95 2.90 May 3.60 2.83 June 3.80 2.81 July 3.65 2.67 August 3.62 2.93 September 3.75 October 3.55 November 3.57 December 3.36 Average 3.20 Vice President LCR: David de Ferranti Director LCC7C: Axel van Trotsenburg Director LCSPR: Ernesto May Sector Manager: Mauricio Carrizosa Lead Economist: Paul Levy Task Manager: Mateen Thobani ABBREVIATIONS AND ACRONYMS ADESPA Argentine Association of Utility Asociacion de Empresas de Servicios Companies Publicos Argentinos ADIGAS Association of Gas Distribution Asociacion Distribuidores de Gas Companies AFIP Federal Tax Authority Administracion Federal de Ingresos Publicos AFJP Pension Fund Management Agency Administradora de Fondos de Jubilaciones y Pensiones AIDS Acquired Immunodeficiency Syndrome ANMAT National Administration for Administracion Nacional de Medicamentos, Pharmaceuticals, Food and Medical Alimentos y Tecnologia Medica Technology ANSES National Social Security Administracion Nacional de la Seguridad Administration Social APE Special Programs Administration Administraci6n de Programas Especiales ATN Treasury Contribution Aporte de Tesoro Nacional AYEE State Owned Electricity Company Agua y Energfa Elctrica BODEN State Bonds Bonos del Estado CCFT Corporate Cash Flow Tax CEER Center for the Economic Study of Centro de Estudios Economicos de la Regulation Regulacion CER Consumer Price Index Coeficiente de Estabilizaci6n de Referencia CIER Regional Commission for Electrical Comisi6n de Integraci6n Electrica Regional Integration CIPPEC Center for Implementing Public Centro de Implementacion de Politicas Policies for Equity and Growth Publicas para la Equidad y el Crecimiento CNC National Telecommunications Comisi6n Nacional de Comunicaciones Comrnmission CNRT National Transport Regulation Comisi6n Nacional de Regulacion del Commission Transporte en Argentina COFESA Federal Health Council Consejo Federal de Salud CPI Consumer Price Index CREMA Rehabilitation and Maintenance Contrato de Rehabilitaci6n y Contract Mantenimiento CVAT Compensating Value-Added Tax CVS Wage Price Index Coeficiente de Variaci6n Salarial DNV National Roads Agency Direcci6n Nacional De Vialidad DPV Provincial Roads Agency Direcci6n Provincial De Vialidad EC European Community EGB General Basic Education Educacion General Bdsica EMBI Emerging Markets Bond Index ENARGAS National Gas Regulatory Agency Ente Nacional Regulador del Gas ENRE National Electricity Regulatory Ente Nacional Regulador De La Agency Electricidad ENTEL State Owned Telephone Monopoly Empresa Nacional De Telecomunicaciones EPH Permanent Household Survey Encuesta Permanente de Hogares EPRAC Provincial Water Regulatory Agency Ente Provincial de Regulaci6n de Agua ERESEP C6rdoba Provincial Multisectoral Ente Regulador de Servicios Publicos Regulatory Agency ETOSS Tripartite Water Regulatory Agency Ente Tripartito De Obras Y Servicios Sanitarios FCT Electricity Tariff Compensation Fund Fondo de Compensaci6n Tarifaria FEDEI National Fund for Provincial Electricity Fondo de Desarrollo Electrico del Interior Development FFT Electricity Transmission Trust Fund Fondo Fiduciario de Transporte Electrico FIEL Foundation for Latin American Fundaci6n de Investigaciones Economicas Economic Research Latinoamericanas FLACSO Latin American Faculty of Social Facultad Latinoamericana de Ciencias Sciences Sociales FNEE National Electricity Fund Fondo Nacional de Energia Electrica FONID National Teacher Incentive Fund Fondo Nacional de Incentivo Docente FSR Redistribution Fund Fondo Solidario de Redistribuci6n GBA Metropolitan Area of Buenos Aires Gran Buenos Aires GDP Gross Domestic Product GOA Government of Argentina GRT Gross Receipts Tax Hidronor State Owned Electricity Company Hidroelectrica Norpatagonica HIV Human Immunodeficiency Virus ICL International Constitutional Law ICSID International Center for Settlement of Investment Disputes IDEA Institute for Argentine Enterprise Instituto para el Desarrollo Empresarial de Development la Argentina IDH Human Development Index Indice de Desarrollo Humano IMF International Monetary Fund INARSS National Social Security Resource Instituto Nacional de los Recursos de la Institute Seguridad Social INDEC National Institute of Statistics and Instituto Nacional de Estadisticas y Censos Censuses INSSJyP National Institute of Social Security for Instituto Nacional de Seguridad Social de Pensioners Jubilados y Pensionados (o PAMI) IVA Value-Added Tax Impuesto al Valor Agregado LECOP Federal Quasi-money Letras De Cancelaci6n De Obligaciones Provinciales LPG Liquid Petroleum Gas MECON Ministry of Economy Ministerio de Economia y Producci6n MEyOySP Ministry of Economy, Public Works, Ministerio de Economia y Obras y Servicios and Utilities Pu'blicos MIV Ministry of Housing Ministerio de Vivienda MTEF Medium-term Expenditure Framework NBI Unmet Basic Needs Necesidades Bdsicas Insatisfechas NBO National Budget Office NGOs Non-Governmental Organizations OCCO VI Regulatory Agency for Toll Road 6rgano de Control de Concesiones Viales Concessions OCCRABA Regulatory Agency for Access Road Organo de Control de Concesiones de Concessions for Buenos Aires Carreteras de Acceso a Buenos Aires OCEBA Buenos Aires Provincial Electricity Organismo de Control de Electricidad de Regulatory Agency Buenos Aires OECD Organization for Economic Organizacion de Cooperacion y Desarrollo Cooperation and Development Econ6mico ORAB Buenos Aires Provincial Water Organismo Regulador de Agua Bonaerense Regulatory Agency ORSNA National Airports Regulatory Agency Organo de Regulacion del Sistema Nacional de Aviacion OSN National Social Works Obras Sociales Nacionales OSP Provincial Social Works Obras Sociales Provinciales PAMI Integral Medical Assistance Program Programa de Asistencia Medica Integral (o for Retirees (or INSSJyP) INSSJyP) PAYG Pay-as-you-go Sistema de Reparto PBU Basic Universal Service Package Prestacion Bdsica Universal PEA Food Emergency Program Programa de Emergencia Alimentarfa PFO Orderly Financing Plan Plan de Financiamiento Ordenado PMO Obligatoria Medical Plan Plan Me'dico Obligatorio POF Workforce Deployment Scheme Planta Orgdnica Funcional S&P Standard & Poor's SAFJP Superintendence of Pension Fund Superintendencia de las Administradoras Management Agencies de Fondos de Jubilaciones y Pensiones SEGBA Buenos Aires Electricity Company Servicios Electricos del Gran Buenos Aires SIDIF Integrated Financial Information Sistema Integrado de Informaci6n System Financiera SIFER Integral Railroad System Sistema Integral Ferroviario SLUP Integrated System of Pensions and Sistema Integrado de Jubilaciones y Retirement payments Pensiones SINTyS National Tax and Social Identification Sistema de Identificacion Nacional System Tributario y Social SISFAM System for Identifying and Selecting Sistema de Identificaci6n y Seleccion de Families Benefiting from Social Familias Beneficiarias de Programas y Programs and Services Servicios Sociales SISTAU Integral Public Road Transport System Sistema Integral de Transporte Automotor SISVIAL Integral Roads System Sistema Integral de Vialidad SOE State Owned Enterprise SOPyC Secretary of Public Works and Secretarta de Obras Publicas y Construction Construcci6n SPU Secretariat for University Policies Secretaria de Politicas Universitarias SSS Health Services Superintendence Superintendencia de Servicios de Salud STD Sexually Transmitted Disease SUMA Universal Service and Environment Servicio Universal y Medio Ambiente UCOFIN Coordination Unit for Infrastructure Unidad de Coordinaci6n de Fideicomisos Trust Funds de Infraestructura US CPI United States Consumer Price Index VAT Value-Added Tax Impuesto al Valor Agregado WHO World Health Organization Organizaci6n Mundial de la Salud Acknowledgements This report was authored by Mateen Thobani based on contributions from a team of Bank staff and consultants during February-May 2003 and with the cooperation of government officials. It also draws on material presented by experts at a series of workshops in Buenos Aires during March and April 2003 and from discussions with a wide range of individuals in think tanks, academia, NGOs, and international organizations. A list of the team members who wrote or contributed to the various chapters and related annexes follows below: Chapter 1: Mateen Thobani Chapter 2A: Mateen Thobani and Zeinab Partow with inputs from David Rosenblatt and Marisa Alfiz Chapter 2B: Timothy Irwin with inputs from Cynthia Moskovits Chapter 3: David Rosenblatt and Pablo Serra Chapter 4: Ariel Fiszbein and Marcelo Becerra with contributions from Maria Paula Savanti and Polly Jones. They also drew on material prepared by Juan Pablo Uribe (health), Sergio Espania (education), Rafael Rofman (pensions), and Oscar Cetrangolo (higher education) Chapter 5: Vivien Foster with inputs from Diego Bondorevsky Chapter 6: Allen Schick with inputs from Marisa Alfiz, Paul Sisk and Timothy Irwin. The team would like to thank Gerardo Hita at MECON for having ably conducted government coordination. It would also like to thank many officials of the federal government as well as from the provinces of Buenos Aires, C6rdoba, and Misiones for providing helpful information and engaging in frank discussions. The social sector chapter has benefited greatly from the close collaboration of the team led by Darnian Bonari at MECON. The report also benefited from the competent research assistance of Laura Saenz and the efficient assistance of Patricia Chacon Holt in Washington and Alexandra Tanus in Buenos Aires. Rut Simcovitch and her team provided capable translation and interpretation services. Marisa Alfiz deserves special thanks for providing helpful guidance to the team in a number of areas. Finally, the team would like to thank the peer reviewers David Rosenblatt, Daniel Artana, Monica Pandeiros, and Timothy Irwin for providing useful comments. The report also benefited from helpful comments and discussions from Paul Levy, Alejandro Guerson, Bill Dorotinsky, Carlos Cavalcanti, Martin Petri, and Julio Vinuela. Table of Contents EXECuTriVE SUMMARY ........................... I MATRIX OF POLICY RECOMMENDATIONS .............................. XIv 1. OVERVIEW .............................. 1 A. BACKGROUND AND MOTIVATION .............................. 1 B. OBJECrIVES, SCOPE, AND STRUCTURE ............................... 3 2. FISCAL TRENDS AND ANALYSIS ..............................5 A. THE EVOLUTION OF PUBLIC FLNANCES ............................... 5 Procyclicity of Public Finances ........................... 5 Trends in Expenditures and Revenues ......................... 7 The Problem of the Fiscal Commons ......................... 8 B. CONTINGENT LIABILITES, COMM1TMENTS, AND OTHER FISCAL RISKS ...................... 10 Introduction ..................................................... 10 Off-budget expenditures and growth of debt .................................................... 10 Liabilities created in response to the crisis .................................................... 11 3. FEDERAL-PROVINCIAL FISCAL RELATIONS ............................................... 13 A. FEDERAL TRANSFERS TO PROVINCES ................. .................................... 13 B. PROVINCIAL TAXES .................................................... 1 5 C. IMPROVING PROVINCIAL TAX EFFICiENCY AND ADMINiSTRATION ............... .............. 1 7 Review of Tax Policy Proposals .................................................... 17 Recommended Policy Options ..................................................... 21 D. RATIONALIZING THE FlSCAL REVENUE SHARING AGREEMENT ................................... 22 E. INSTITUTIONAL REFORMS AND "FISCAL RULES" ..................................................... 25 4. IMPROVING THE EFFICIENCY AND EQUITY OF SOCIAL EXPENDITURES ..................................................... 29 BASIC EDUCATION .......................... ; 30 Performance of Provincial Spending .......................... 30 Inefficiencies in Provincial Spending ......................................... 32 Recommendations ......................................... 34 UNIVERSITIES ......................................... 36 Performance of University Spending ......................................... 36 Inefficiency and Inequity in the Public University System ......................................... 37 Reform Proposals and Recommendations ......................................... 38 HEALTH ......................................... 40 Structure and Performance ......................................... 40 Social Health Insurance ......................................... 43 Reforming Health Institutions for Efficiency and Equity ......................................... 44 PENSIONS ......................................... 46 Fiscal Implications of Recent Reforms ......................................... 46 Issues and Prognosis ......................................... 48 Recommendations ......................................... 49 SOCIAL ASSISTANCE ......................................... 50 Experience with the Heads of Household Program .......................................... 51 The Conditional Transfers Program, Familias .......................................... 56 Feeding Programs .......................................... 57 5. ENSURING EFFICIENT INFRASTRUCTURE SERVICES FOR GROWTH.59 BACKGROUND AND MOTIVATION ............................................ 59 Achievements of Infrastructure Reform ........... ................................. 60 Critiques of the reform ............................................ 62 B. RECENT DEVELOPMENTS IN INFRASTRUCTURE .................... ........................ 63 Contract Renegotiation ............................................ 64 Proliferation of trustfunds ............................................ 65 Subsidies and the poor ............................................ 66 C. ENSURING SUSTAINABILrT OF INFRASTRUCTURE SERVICES ...................................... 67 Framework for the Renegotiation Process .............. .............................. 68 Redirecting Subsidies to the Poor ............................................ 69 Reforming the Regulatory Framework .......................................... 70 Reforming Governance of Extra-Budgetary Mechanisms .72 6. MANAGING PUBLIC EXPENDITURES .75 A. INTRODUCTION .75 B. THE FORMAL AND ALTERNATIVE BUDGET PROCESSES .76 The Budget Control Mechanism .76 Escaping Budgetary ControL .77 Budgeting for Fiscal Commitments and Risks .78 B. MODERNIZING PUBLIC ExPENDITURE MANAGEMENT .79 Macro-budgeting .79 Budgeting Allocation Process .83 Managing Public Services .85 REFERENCES .................... 89 ANNEXES .................... 93 STATISTICAL APPENDIX .....................111 List of Boxes Box 1: ROAD TO CRISIS .................................................................. 2 Box 2: WHAT ARE CONTINGENT LiABILrTIES AND COMMITMENTS? ............................................... 10 Box 3: ASYMMETRIC PESIFICATION OF BANKING ASSETS AND LIABILITIES AND GOVERNMENT BONDS .................................................................. 12 Box 4: PRIMARY AND SECONDARY DiSTRiBUTION .................................................................. 14 Box 5: PREVIOUS EXPERIENCE WITH WORKFARE PROGRAMS: TRABAMR ..................................... 51 Box 6: SWEDEN'S TWO-STAGE BUDGET PROCESS .................................. 79 Box 7: ALTERNATIVE WAYS OF BUDGETING FOR CONTINGENT LIABILrES ............ .................... 81 Box 8: CREATING AUTONOMOUS AGENCIES--UNITED KINGDOM AND THAILAND ....................... 88 List of Tables TABLE 1: KEY ECONOMIC INDICATORS .................. ................................................ 1 TABLE 2: THE YIELD OF VAT IN LATIN AMERICA .................................................................. 1 7 TABLE 3: BASIC EDUCATIONAL EXPENDITURES BY LEVEL OF GOVERNMENT, SELECTED YEARS ................... 3 1 TABLE 4: DISTRIBUTION OF UNIVERSITY STUDENTS ACROSS INCOME QUINTILES .......... .............. 37 TABLE 5: COVERAGE LEVELS BY SUB-SECTOR .................................................................. 40 TABLE 6: EXPENDITURES IN HEALTH BY AGENT, 2000 ................................................................. 41 TABLE 7: OSN DISTRIBUTION OF SPENDING BY .................................................................. 44 TABLE 8: FISCAL EFFECTS OF PENSION REFORM AND OTHER POLICIES ................ ........................ 47 TABLE 9: CHARACTERISTICS OF BENEFICIARIES OF HH PROGRAM ..................... .......................... 53 TABLE 10: EVALUATION OF ARGENTINE REGULATORY FRAMEWORK AGAINST INTERNATIONAL BEST PRACTICE .................. ................................................ 71 TABLE 11: EVALUATION OF ARGENTINE TRANSPORT TRUST FUND AGAINST INTERNATIONAL BEST PRACTICE .................................................................. 73 List of Figures FIGURE 1: EEDERAL AND PROVINCIAL FiSCAL BALANCES .......................................................... 6 FIGURE 2: PROVINCIAL FISCAL BALANCE (EXCLUDING PRIVATIZATION REVENUES) .......... ............. 7 FIGURE 3: SPREADS ON ARGENTINE DEBT ..................... ............................................. 8 FIGURE 4: THE CONTRIBUTION OF DEBT RECOGNITION TO THE GROWTH OF DEBT IN ARGENTINA .................................................................. 11 FIGURE 5: PERFORMANCE IMPROVEMENTS FOLLOWING PRIVATIZATION .............. ....................... 61 FIGURE 6: THE PHENOMENON OF EXTRA-BUDGETARY MECHANISMS ................. ......................... 65 FIGURE 7: DISTRIBUTIONAL INCIDENCE OF SUBSIDY SCHEMES IN ENERGY AND TRANSPORT ...... 67 EXECUTIVE SUMMARY Introduction 1. During 2002, Argentina was in the throes of a severe economic and social crisis. GDP fell by 10.9 percent that year, for a cumulative decline of 20 percent since 1998. Unemployment rose to 21.5 percent in May and poverty more than doubled since 1994. The Government stopped servicing most of its debt. The peso, which had been pegged to the dollar at par from 1991 to 2001, fell to 3.8 pesos to the dollar by June. Consumer prices rose by 41 percent even though most utility and transport prices were frozen. Dollar deposits were forcibly converted to pesos and limits placed on withdrawals. Many businesses faced bankruptcy. While a measure of macroeconomic stability and growth returned in 2003, it was within a context of enormous government arrears, commitments to make sizeable compensation payments, and accumulation of large contingent liabilities. 2. The severity of the crisis is even more surprising coming on the heels of strong economic gains through most of the 1990s. Although researchers differ on the causes of the recent crisis, it is clear that without resolution of the public debt default, new investment-critical to sustainable growth-will be slow to come. Even with high growth rates and a favorable renegotiation of its debt, projections suggest that Argentina will need to generate unprecedented primary fiscal surpluses of about 4 percent of GDP. It will need to generate even larger surpluses to reduce its high level of debt in order to reduce vulnerability to future shocks. Doing so while simultaneously reducing its dependence on some taxes that inhibit growth will be particularly challenging. 3. This report seeks to identify policy and institutional reforms to enhance the efficiency, equity, and effectiveness of public expenditures at the federal and provincial levels. These are crucial to generate the kinds of fiscal surpluses needed for fiscal solvency and macroeconomic stability. If not, efforts to stimulate private-sector-led growth and generate employment-which are key for sustained poverty reduction-may not succeed. 4. The report analyzes both sectoral and process-related issues. The sectoral analysis focuses on the major expenditure categories within the social and infrastructure sectors, which accounts for over 70 percent of total expenditures. Effective provision of social and infrastructure services is crucial for poverty reduction, equity, and improved living standards. The process-related issues focus on two areas affecting expenditure efficiency and fiscal solvency: federal-provincial government relations and public expenditure management. While important, federal tax policy and public administration issues are excluded from this report in order to keep its scope manageable. Federal-Provincial Fiscal Relations 5. Over half of provincial government revenues come from taxes imposed and collected by the federal government under a system known as coparticipation. As designed, the system reduces incentives to improve tax collections or rationalize expenditures. It also encourages rent-seeking behavior and reduces transparency and accountability. The system is inequitable with richer provinces receiving larger per capita transfers on average. i Finally, it softens budget constraints. The situation is exacerbated by the presence of some provincial taxes that stifle growth and by inappropriate fiscal rules. 6. To address these issues, action on three fronts is needed: * Provincial tax administration and policy: To enhance revenues and reduce inequities, provinces should consider stopping the use of tax amnesties and discounts, providing better training, investing in information technology, collaborating with AFIP, using withholding taxes, and prosecuting non-compliers. Property tax revenues can increase by improving the roll, cross checking information with municipalities and public utility companies, and obtaining a more realistic assessment of property values, possibly by delegating the valuation function to federal authorities. Provinces may also wish to consider replacing the gross receipts tax (GRT) and stamp tax with either a dual federal-provincial VAT or by piggybacking onto the federal personal income tax. Both these options require a lowering of federal tax rates-something that can be done within the context of the coparticipation reform. In the short-term, it will be difficult to replace the GRT. However, piggybacking onto the personal income tax and decentralizing the wealth tax may still be desirable and feasible, as long as it were accompanied by a commensurate reduction in coparticipation revenues. * Coparticipation: It would be desirable for both the provinces and the nation to become responsible for financing their own budgets on aggregate, with inter- provincial transfers being financed by other provinces' own revenues. This requires (a) transferring some federal taxes to provinces and/or (b) dramatically improving own revenue mobilization from existing taxes or one of the new tax instruments discussed above. If complete separation of finances proves too difficult, simplifying the system by having only one general revenue-sharing pool would be helpful. An allocation formula that compensates provinces for having lower tax capacity per capita will be simpler and more equitable. To smooth the transition and reduce opposition to the reform, provincial revenues could be protected in the short-term by having the new distribution formula apply only to increases in revenue, thereby still providing the right incentives at the margin. * Fiscal rules: Fiscal rules introduced in Argentina in recent years did not work because they were unfortunately timed and badly designed, requiring an overall fiscal surplus despite a huge interest shock and whilst in a recession. An additional credibility issue arose because the rules were framed within an ordinary law, making its reversal relatively easy. To correct this, Argentina may wish to frame the rules within the context of the coparticipation reform and to target the cyclically-adjusted primary fiscal balance. Ideally, this new ley convenio should also require provinces to abide by primary fiscal targets as well as borrowing rules. As in Brazil, penalties for non-compliance could be imposed and mechanisms on returning to compliance could be spelled out. Finally, the creation of a committee of experts at the federal level to set and monitor cyclically-adjusted fiscal targets may be desirable. Social Sectors 7. At about US$1500 per capita, Argentina had one of the highest levels of public social spending in Latin America during the 1990s. However, the spending has not resulted in commensurate social outcomes. Below some of the key issues and recommendations to enhance the efficiency and equity of social services are highlighted in five key areas: basic education, higher education, health, pensions, and social assistance programs. Basic Education 8. , The financing, administration, and provision of basic education in Argentina is the responsibility of provincial governments. Financing by provinces is provided to public schools providing free education as well as to private schools to partially, or totally, cover teacher salaries. Despite a 77 percent increase in real expenditures between 1991 and 2001, serious deficiencies in efficiency, quality, and equity remain. Dropout rates are high, especially in secondary school and especially at lower income levels. Expenditures fell by a colossal 38 percent in real terms during 2002, mainly because of a decline in real wages. Teacher salaries accounted for 91 percent of expenditures on basic education. They are determined by seniority and bonuses for specific duties, with little or no consideration given to teacher quality and effort. 9. Wide variation in spending per capita exists even between provinces of similar incomes. Yet these differences fail to get reflected in academic evaluations, suggesting a potential to improve quality without additional spending. A low student-teacher ratio is the main source of inefficiency. While sensible norms exist for determining the number of teachers in a school, these are not followed for reasons of inertia, political factors, and institutional weaknesses. 10. An improved management and control system, which enjoys the strong support of the provincial leadership, is crucial for reducing inefficiencies. Such a system would establish electronic records matching teachers to teaching positions and provide up-to-date information on the teacher work force and student enrollment. This will facilitate enforcing normns regarding the deployment of teachers and administrative staff. The experience of some provinces with modern management systems has been very positive. For example, between 1996 and 2001, Salta increased enrolments by 16.5 percent while increasing spending by only 5.6 percent. It did this mainly by better allocation of teaching positions, conducting medical checks for teachers on sick leave, and reducing red tape. 11. In order to make the above reforms sustainable, provinces should consider providing greater autonomy to schools in a context of enhanced accountability. Greater participation of parents associations and school boards in school management is a possibility. Overall, these potential changes would enhance the regulatory and funding roles of the provincial governments, which should seek mechanisms to finance schools on transparent criteria such as number of students, population density, or othe'r indicators of need. Such criteria could be applied immediately to private schools that receive public subsidies. The federal government could also improve incentives for enhanced efficiency by tying its support to provinces-for teacher training, didactic materials, and infrastructure maintenance-to improvements in educational efficiencies and quality. liii Higher Education 12. About 1.3 million students are enrolled in Argentina's 41 public and 50 private universities. Between 1991 and 2001, public spending on education rose by 72 percent in real terrns while enrollment rose by 75 percent. With few exceptions, students pay no fees and face no quantitative or qualitative barriers to entry. This combination has resulted in a highly unequal and inefficient public university system. Spending is highly regressive, with 60 percent of students coming from the top two quintiles of the income distribution and less than 6 percent from the bottom quintile. This bias is even larger with respect to students who graduate since low-income students are more prone to dropping out. Graduation rates have fallen from 60 percent in the 1980s to 18 percent currently. Only 11 percent of the students who entered the university in 1990 had graduated by 1999. 13. Competing in the world economy requires an adequate supply of high quality professionals. To meet this challenge in an environment of fiscal constraints, there is a need to enhance efficiency, consider alternative financing mechanisms, and improve opportunities for the poor. Efficiency can be enhanced by (i) enforcing regulations that specify conditions by which one can remain a student, (ii) setting limits to entry-for example via performance at tests when leaving secondary school, (iii) increasing budgets of universities that perform well on indicators such as the share of students graduating or the average tenure of a student in the system, and (iv) introducing or expanding the use of fees. This last measure will also help reduce the regressive nature of university expenditures, especially if accompanied by scholarships for low-income students. 14. Moving in this direction will require a process of consensus building. A useful precedent was the Juri Commission, which considered a number of institutional reforms in the university system even though it did not focus on the controversial issues of financing and efficiency enhancement. Given the success that private universities have had in attracting students and teachers away from public universities, conditions may be ripe for introducing the above policy and institutional reforms. Health 15. With over 8 percent of GDP being spent on health, Argentina's basic health status indicators (e.g. infant and maternal mortality) still fall short of those of comparable countries with lower health spending levels. The institutional arrangements of the health care system and its fragmentation, especially health insurance, is partly responsible. Insurance is mainly done through a network of Obras Sociales Nacionales (OSNs) for formal workers, Obras Sociales Provinciales (OSPs) for provincial government workers, and PAMI, the retirement institute for pensioners. There are also a host of mostly small private health insurers for the better off while over 40 percent of the population lacks insurance. In fact, in 2001, 57 percent of the population received health care exclusively from the public service delivery network. 16. The dispersion of population among too many small insurers reduces the size of risk pools, increases administrative costs, and generates deficits that require fiscal bailouts. It also makes it difficult to develop stable redistributive mechanisms for equity and creates opportunities for cross-subsidies between sub-systems-for example public hospitals iv rarely recover costs of services from health insurers. Moreover, the overlapping responsibilities between levels of government makes a strong and effective regulatory framework difficult. The federal health regulator, SSS, does not have jurisdiction over provinces or even over PAMI, the largest purchaser of health services in the country. The health insurance system excludes millions of informal workers and even some of the lowest-income formal workers. 17. The public health system is excessively biased towards specialized and hospital- based care with insufficient attention paid to health promotion, disease control, and ambulatory care. Allocations for public health services are done on the basis of supply rather than demand and are skewed towards the better off. Hospital beds and doctors are found in higher concentrations in richer jurisdictions where the supply of private practitioners and facilities is also the largest. Furthermore, management of public hospitals tends to be weak and characterized by a lack of performance incentives, minimal autonomy, lack of information, and poor management tools. 18. Improving the health system will require reforms in the public health delivery system, the social health insurance system, and the regulatory system. The public health delivery system could improve by granting public hospitals greater autonomy and then holding them accountable for results. Introducing performance contracts that link budgets to objective processes and results could achieve this. Needier populations could receive larger budget allocations. In addition, regressive cross-subsidies from public hospitals to health insurers could be eliminated, inter alia, by facilitating public hospitals access to updated databases of the OSNs, OSPs and PAMI and by providing technical assistance for proper billing processes. Resulting increases in revenues could be directed to increase support for primary health and provide results-based incentives. 19. Health insurance reforms are best directed to more equitable redistribution mechanisms and to enforcing existing regulations relating to the operation of insurer agents. If complemented by a policy of not providing financial subsidies to insurers, it most likely would lead to the closure of dozens of insurers and encourage the restructuring, merging, and consolidation of others. Larger pool risks would result, with a consequent reduction in administrative and transaction costs. Moreover, monitoring, evaluation, and control will be facilitated, and overall responsiveness to citizens should improve. 20. Regulatory reforms should focus on the national health regulatory agency, SSS. To function transparently, the SSS should not administer resources for the entities it regulates nor perform, directly or indirectly, purchases for them. SSS should be seen as independent, technically competent, and representative of all actors, including the provinces. Finally, it is essential to regulate and reform PAMI, which is notoriously inefficient and has received US$1.7 billion in fiscal bailout during 1995-2001. Via a 2002 legal reform, the national government lost governance control over it, yet remained responsible for its overall financial viability. Legal reforms are urgently needed to align incentives for performance with responsibility and accountability in PAMI. Only then can needed structural reforrns occur to ensure that PAMI is responsive to the public and financially solvent. v Pensions 21. In 1994, Argentina reformed its pay-as-you-go (PAYG) system into a two pillar system: a flat benefit (PBU) to retirees with at least 30 years of contributions who meet minimum age criteria, and a second pillar comprising either a new smaller PAYG system run by the government or a funded scheme with individual accounts run by private pension fund operators called AFJP. The Government finances the transition costs to reflect the years of contributions of those who moved to the AFJPs. It also absorbs the cost of a reduction in employer contributions to encourage formal sector employment and the cost of transferring provincial public pensions to the national scheme. As a result, only 30 percent of the public expenditure on pensions is now recovered by contributions, though this is partly due to weaknesses in collection compliance. INARSS' performance has also been hampered by lack of political support and by rules and practices that effectively exempt certain types of income and workers. 22. Once expenditures related to the old system completely disappear, substantial savings in public expenditures on pensions are projected-by 2050, the Government is projected to spend just 1 percent of GDP on pensions. This result, however, is partially due to a projected decline in coverage. Because of the 30-year coverage requirement and the greater informality in the labor market, more workers will be unable to claim benefits, which have fallen substantially in real terms during 2002. The reduced coverage is both a serious social problem and a threat to longer-term fiscal solvency because of rising social pressures to modify the system. 23. A redesign of the pension system is urgently needed to increase coverage and efficiency, while reducing fiscal costs. Coverage can be extended by integrating and expanding the different benefits aimed to prevent poverty among the elderly. In addition, the requirement of having 30 years of contributions could be replaced with a fiscally neutral scheme that recognizes shorter contribution periods but offers incentives to contribute for longer periods. While these reforms will need wider public debate and discussion, there are several measures that the Government could take right away to enhance transparency and fairness while also increasing collections. They include: * Taking administrative and/or legal steps to make all components of wages taxable and close loopholes in the self-employment regulations. * Undertaking a thorough and independent review of the actuarial and financial status of existing systems to provide both the Government and the general with public information regarding their financial status, benefit levels, and any cross subsidies. * Simplifying the benefit granting process by recognizing as valid information from the SLIP individual accounts, by removing the burden of proof from beneficiaries as much as possible, and by establishing clear rules to adjust benefits over time * Revising investment regulations for pension funds, life insurance, and annuities companies to achieve greater flexibility in asset allocations, ensure a level playing field, and increase risk diversification vi * Improving the regulatory framework for private pension and insurance services providers to promote competition and reduce costs * Reinforcing the political and technical independence of the SAFJP, by ensuring that the senior managers are professionally qualified and that appropriate 'Chinese walls' are set to separate them from other areas of Government. Social Assistance 24. Social assistance programs increased greatly in 2002 in response to the crisis, mainly through the Heads of Household Program (Jefes), a federal program providing monthly grants to eligible unemployed heads of household who agree to participate in work or training activities. Jefes spent $1.7 billion in 2002 and the number of beneficiaries increased from 500,000 in May 2002 to about 2 rmillion in March 2003. The federal government also began a conditional transfer program, Familias, operating in large urban areas and requiring parents to participate in stipulated health and education services. Coverage under this program expanded from 100,000 families by the end of 2002 to 175,000 in March 2003. The goal is to eventually reach 500,000 families. 25. Despite a disorderly and speedy start-up, Jefes has managed to reach the bulk of the needy population. Surveys show that only 13 percent of those eligible to participate did not do so. However, 23 percent of those participating appeared to be ineligible in that the household did not have eligible children or handicapped persons. Also, the program suffers from strong and negative perceptions on transparency and effectiveness and from confusion about its work requirement. While the monthly stipend is low enough on average not to discourage participants from taking on regular jobs, this may not be true in all parts of the country and in all activities. 26. The Government has taken several measures to enhance transparency, including conducting an external audit of the program. The following additional measures could help ensure that it is not a deterrent to work and that eligibility requirements are met: * Keep the monthly stipend below the prevailing market wage for unskilled labor and allow it to differ across regions. * Expand the supervision of municipalities and impose sanctions where consistent irregularities are observed. * Regularly provide municipalities and local councils a list of irregularities for remedial action. * Institute annual checks to control the quality of the registration process; any substantial increase in net registrations need to be scrutinized by the Ministry of Labor. * Enforce the work requirement rigorously and narrow the list of workfare activities to favor higher value ones. 27. Better monitoring efforts are also needed to ensure program goals are met along with a willingness to modify the design, funding, and implementation of the program as appropriate. The longer-term vision for the Jefes program is one of reduced size, higher vii quality, and focused on a counter-cyclical role aimed at reaching the poor. This would imply that economic conditions improve, the size of the Program would tend to diminish, and would only grow again if economic conditions worsened. In that context, the national government should consider a strategic change in the program by devolving administrative and financing responsibility for such a program to the provincial or municipal level, with federal intervention when local unemployment rates exceed, for example, the national average. 28. The Familias program is designed so that the mother in the family receives a per- child subsidy. Beneficiaries must be poor and receive no other benefit from the Government, which effectively excludes those in the formal labor market. Familias is designed to address structural poverty as opposed to cyclical poverty. Its aim is to ensure human capital investment in children when poor households may lack the needed resources, while the workfare program, Jefes, is best suited to offer income support to those willing to work, but unable to obtain full-time employment. 29. Since they address different needs, both types of programs could be part of the Government's social safety net. To be well targeted and effective, the two programs need to be better coordinated at the local level. Families with numerous younger children, particularly those with female heads, would be better directed to Familias instead of Jefes, thereby reducing pressure on the latter. Familias also needs to be coordinated with the existing scholarship program for secondary students and with the system of family allowances. Savings from elimination of duplication of benefits could be used to extend the coverage of family allowances beyond formal sector workers. Finally, a rigorous, impact evaluation is essential for Familias in order to set the "correct" transfer level, to ensure that resources are reaching the poor, and that program objectives such as higher educational attainment and better health indicators are being achieved. Infrastructure 30. Following large-scale privatization of infrastructure services during the 1990s, federal and provincial governments raised substantial revenues and reduced outlays on infrastructure services even while increasing coverage and enhancing quality. However, continued coverage and maintenance have been threatened by the devaluation and recession. Providers were also hit by a government-mandated freeze in prices for most services. A Commission to renegotiate concession contracts was established but progress has been slow and uneven. While utilities received little relief, the transport sector was able to receive subsidies via special financing mechanisms, tax breaks, and price increases. 31. Neither the freezing of tariffs for utilities nor the provision of ad hoc subsidies in transport is desirable. Without a fair and timely renegotiation of infrastructure contracts there is a real danger that the public sector may be forced to take over services that were successfully privatized during the 1990s. Moreover, the tariff freezes have created contingent liabilities that are threatening fiscal solvency. 32. The privatizations were not without problems. Regulators repeatedly gave in to pressures to reduce canon payments and increase subsidy payments to operators. Consumer prices of many services rose in real terms, though this is partly due to the viii increased taxes. In the case of water, for example, Argentina's indirect taxes are higher than in any OECD country, while its energy taxes place it in the top five. Before taxes, Argentine tariffs on gas, electricity, and water are amongst the lowest in Latin America. 33. In recent years, infrastructure subsidies have quadrupled from $422 million to $1.9 billion while the share of infrastructure funds financed through special financing mechanisms have risen from 45 percent to 98 percent. The distribution of subsidies is regressive. About three quarters of the resources channeled to the existing subsidy schemes take the form of supply side subsidies, primarily to transport operators. As these subsidies benefit all customers, their distribution is skewed towards the upper middle-income groups-the predominant users of transport. Similarly, a study finds that the Patagonian gas subsidy excludes 95 .percent of poor gas consumers in Argentina, while 63 percent of its beneficiaries are non-poor. 34. Only the private sector is able to finance Argentina's projected infrastructure investments. For this, investor confidence will need to be restored. This, in turn, will require concluding the renegotiations with private operators and addressing certain second- generation reforms. The alternative of a return to public ownership and control is not desirable given the higher efficiency in the private sector, the other demands on fiscal resources, and problems with public provision of services. As outlined below, what Argentina needs is a fair resolution of renegotiations, better regulatory arrangements, better protection for the poor and better financing mechanisms. 35. Renegotiations will have to facilitate concessionaires' ability to reschedule their debts. Creditors are likely to want a credible agreement within a long-term framework under which tariffs will be determined. Two issues are key: (i) a new mechanism for tariff adjustments that is consistent with the new post-convertibility context and (ii) the method for asset valuation in the post-convertibility regime, with all that this implies towards restoring financial equilibrium (mantener la ecuaci6n econ6mico financiera). Also the normal operation of the regulatory process needs to be restored as early as possible in order to avoid permanent damage to the credibility of the regulatory system. 36. A critical element in the renegotiation process is the incorporation of a mechanism that protects the most vulnerable users from the resulting tariff impact. Argentina is currently devoting $940 million to a variety of social policies related to infrastructure services that, taken jointly, have very little impact on the poorest households. Such resources, properly retargeted, would offer an important financing basis for a social tariff mechanism. In the medium to long term, a comprehensive analysis and dialogue is needed on how best to offer social protection to the poor and vulnerable by asking what mix of social policies-income and food transfers, health, education, and infrastructure subsidies, unemployment and pension schemes-is appropriate for Argentina. 37. In the water and gas sectors, a subsidy on connections could be of greater relevance than a subsidy on consumption. Such a subsidy would be relatively easy to target since it is almost always the poor who are excluded from these services. For bus and rail, an income transfer is a better alternative but the Government may consider providing a per trip subsidy to areas serving the poor rather than maintaining the current per passenger subsidy. While there is little economic justification for a subsidy on electricity, this sector is taxed ix heavily-in part to subsidize low income users and depressed areas. However, the taxes are not used for this purpose. Moreover, pressures to keep the tariffs low for the poor are hampering renegotiations with power operators. Thus there may be merit in directing the electricity surcharges to temporarily freeze rates for those who consume below a minimum threshold even as rates for other users increase. 38. The crisis offers an opportunity for building consensus on 'second generation' regulatory reforms, correcting the deficiencies that have come to light during the last decade, without discarding the considerable positive elements. In particular, it is important that the lessons learnt in those sectors where the reform has worked better, as is the case of electricity and gas, are transferred to the other infrastructure sectors. Following the example of the electricity and gas regulators, which have worked well, it would be desirable to create or reform independent regulators in other sectors. Regulatory bodies could be created by law rather than by decree to provide a clear basis for their authority and limit the extent of ad hoc political interventions. In addition, Commission members of these bodies could be selected by competition and approved by legislatures to help ensure greater political independence. Legislatures could also approve the budgets of the regulatory bodies to help strengthen accountability and avoid the conflicts of interest that can arise when the size of regulators' budgets depend directly on the level of tariffs. Finally, it would be important to ensure participation of civil society in regulatory processes. In this sense, the introduction of a consumers committee that acts as a permanent consultative body for the water regulator in the City of Buenos Aires serves is an interesting model. 39. Finally, while there may be justification for selective special financing mechanisms in infrastructure, the design and implementation of Argentina's mechanisms-especially the large transport trust fund-is not appropriate. It could be improved by having the trust fund: (i) be created by legislation rather than decree, (ii) be an autonomous body with majority user representation on its Board, (iii) employ transparent and established rules for allocation between agencies, (iv) publish financial regulations and procedures, and (v) conduct regular technical and financial audits. Public Expenditure Management 40. A good public expenditure management system provides a government the incentive and means to promote fiscal discipline and efficiency. Argentina's national system falls short. It is not comprehensive-in effect there are at least two budget processes. In law, there is a single process that covers the entire public sector; in practice, there is also a hodge-podge of exceptions and special arrangements that undermine fiscal discipline and efficiency. The coexistence of effective formal controls and escapes from control has led some line ministries to migrate from the formal system to informal arrangements, effectively running parallel budget systems that seek to cope with uncertainties in financing availability. Escape routes include earmarking money for designated purposes, establishing trust funds, giving agencies financial autonomy, removing accounts from the budget, and use of emergency decrees. 41. This dual budget system has adverse side effects. First, it boosts the cost of goods and services purchased by spending units-some suppliers explicitly charge differing x prices according to when they are paid; others simply build expected delays into the prices they charge. Second, it leads budget staff in spending units to spend an inordinate amount of time juggling their books, negotiating with suppliers, and pressuring the center for more money. Third, it gives rise to devious behavior, such as hiding liabilities and building up* arrears when their strict quota gets exhausted. Fourth, the quotas reinforce the tendency to regard budgeting as a means of processing operational expenses rather than for charting the course of government. Fifth, it impairs transparency as the approved budget is not the one implemented by the Government-politicians have an incentive to vote increased spending and managers have an incentive to misclassify transactions. 42. Another deficiency of the budget system is that it fails to adequately consider fiscal risks and multi-year commitments on expenditures. Although a three-year budget is prepared, it does not face public scrutiny. Fiscal risks such as from contingent liabilities are not recorded at all. The budget is typically ignored when these liabilities materialize, and the budget is helpless when the government recognizes the liability. What is especially troublesome is that many of the liabilities are implicit; they are not legal obligations but the Government nevertheless has felt compelled to assume them. Such actions not only expose the Government to moral hazard but also open it to lobbying by powerful interests. 43. Budget planning, formulation, and execution, as well as treasury cash and debt management, are presently carried out by MECON. In addition to the burden this places on the Ministry, the budget ends up being a purely technical exercise, whose approval by elected officials is done without an analysis of policy options. The budget lacks ownership by either elected officials or line ministries and may not reflect national priorities. 44. Improvements to Argentina's system of public expenditure management will need to focus in three broad areas: 45. Macro-budgeting: Instituting a two-stage budget process at the federal level could help improve fiscal discipline. The first stage would establish the macro-economic framework, the overall resource envelope, and broad expenditure categories; the second would prepare annual budgets consistent with the previously agreed totals. To be credible and receive political support, the fiscal targets must be public and voted by Congress. The aggregates could be set for a 3-year timeframe and rolled forward each year as a new budget cycle commences. Modifications to the budget during the year by emergency decree or other means would have to be consistent with the agreed framework. 46. The budget framework could be made more comprehensive by including off- budget accounts, trust funds, arrears, guarantees, fiscal commitments, and other contingent liabilities. It should not only account for liabilities that have already been incurred but be an instrument for regulating the new liabilities to be incurred. The federal government should consider compiling an annual statement of all known explicit contingent liabilities so as to take account of risk both when guarantees are issued and when contingent liabilities are budgeted. However, to avoid moral hazard issues, implicit fiscal risks should only accounted for when there already is high expectation that they will be realized and when there is low probability that affected interests will behave in ways that increase the cost to government. xi 47. Timely review and approval of financial statements is essential. The review is an opportunity to assess the Government's finances, to take stock of the liabilities and risks that have accumulated, to glean whether conditions have improved or worsened over the recent past and why, and to test the reliability of the financial reporting system. Accomplishing these ends depends on building up the role and resources of the Auditor, and using congressional review to call attention to weaknesses in the Government's financial information system and exposure to fiscal risk. The Auditor could produce an annual risk assessment, beginning with financial risk to the State and expanding over time to other vulnerabilities. 48. Provinces should also consider the two-stage budget process. However, the strengthening of financial reporting is a greater priority. It would be highly desirable for each province to produce annual financial statements along the lines of the federal government. This needs to be in a format that facilitates consolidation with the national reports, clearing tracking the inter-governmental flows. For example, if the federal government has an asset in the form of debt owed to it by provinces, this debt should appear in the same amount as a liability on provincial statements. 49. Budget allocation process: Orienting the budget to policy initiatives and national priorities requires that planning responsibility be vested in an agency that has sufficient power to influence budget allocations. Given the difficulty for MECON to handle both the steering and rowing functions, it would be desirable to assign this role to the Office of the Chief of Cabinet (Jefatura). The capacity of core staff with Jefatura to carry out this function effectively will need to be strengthened. To be budget relevant, Jefatura should provide policy guidance to ministries during the framework stage, before ministries prepare detailed budgets. However, the detailed budget formulation and execution should remain in MECON-past efforts to move this to Jefatura were not fruitful. 50. Budgeting could be made into more of a participatory process, which gives sectoral ministries incentives to use resources more effectively. Ministries proposing to shift money to more productive use should be assured of not being penalized by having their budgets cut. To do this, each ministry could be given a resource envelope in the first stage of the budget process and be allowed to alter the mix of programs at the second state of the process, provided that it keeps within the limit specified in the envelope. Each ministry would be encouraged to proposed reallocations with reasonable confidence that its budget would not be cut below the total budgeted for it in the framework. 51. Service efficiency andperformance: Financial management is key to improving service efficiency and performance. The first step to allow for efficient and orderly execution of expenditure plans is for the formally approved budget to determine the resources available to spending units. The annual expenditure decree should serve two purposes: (i) to smooth spending by quarters so that mninistries and agencies do not run out of money before the year has ended; and (ii) to reserve some funds as a hedge against lower revenues and other contingencies. It should not be a means of sequestering appropriated funds, except when it is essential to do so because of macroeconomic instability. The decree should specify provisional quarterly allotments that may be adjusted in final quarterly quotas as the fiscal year progresses. However, monthly limits should be discontinued. In this way, spending units would have reasonable assurance as to the xii resources that will be forthcoming, while the Government will retain short-term macroeconomic capacity. 52. The Government's financial management information system, SIDIF, needs an upgrade, both to incorporate modern technology and to reintegrate subsystems that have been split off from it by various spending units. An increasing number of units have a SIDLF locale, a parallel system for internal management that feeds into SIDLF at some point prior to the processing of payments. Although it is appropriate for ministries and large agencies to have sub-systems for internal management, all should use SIDIF or its successor throughout the financial management process. 53. Provinces can also take several measures to improve the comprehensiveness and timeliness of their financial information systems. First, they should try to bring together the responsibility for budgeting, accounting, and cash management. Some provinces now split the core responsibility for financial management by having a treasurer and comptroller general who are independent from the finance ministry. Second, provinces should accelerate development of modern integrated financial information systems. Some have ambitious projects underway, but none has a fully operational system. Third, basic classifications and reporting requirements should be standardized to permit the aggregation of federal and provincial financial data and to facilitate compliance with fiscal rules and constraints. 54. Finally, clear criteria need to be employed before establishing new autonomous agencies at the federal level. Service provision and government operation can be enhanced by giving greater autonomy to managers to have greater flexibility in using resources. However, recent and ongoing attempts to create organizational enclaves and escape financial and personnel controls may not result in improved performance. This is because the process by which they have been established has not given due consideration to their preparedness for self-management and because inadequate provision has been made for holding them accountable for results. It would be sensible for the Government to make performance contracts a precondition for agency independence. Alternatively, it may adopt an approach introduced by Thailand, which, specified a series of "hurdles" that an agency must surmount before it is granted operational independence. Whatever approach is taken, the important thing is that independence be pursuant to an assessment of the organization's capacity and in exchange for demanding greater accountability for performance. xiii MATRIX OF POLICY RECOMMENDATIONS Area and Objectives Recommendations Timning Federal-Provincial Fiscal Relations Enhance provincial tax efficiency Improve provincial tax administration ST and equity Piggyback onto federal income tax and ST Increase fiscal correspondence decentralize wealth taxes (while reducing Promote fiscal discipline and coparticipation shares) stability Replace GRT and stamp taxes with provincial MT VAT and/or personal income tax in context of coparticipation reform Reform the coparticipation system to reduce or MT eliminate primary distribution and improve equity of secondary distribution Introduce well-designed fiscal rules within the MT context of the coparticipation reform Social Sectors Basic education: Improve management and control system ST Improve efficiency to Provide greater autonomy to schools and hold MT accommodate expansion and them accountable for results quality enhancements Higher education: Enforce rules on remaining a student ST Improve quality, enhance Introduce entrance exams MT efficiency, and reduce regressivity Introduce or expand fees MT Align university budgets with outcomes MT Health: Grant hospitals greater autonomy and hold them MT Improve health outcomes and accountable to results expenditure efficiency Eliminate subsidies from hospitals to health ST insurers and use savings on public health Enforce health insurance regulations ST Stop bailouts to health insurers and reform ST redistribution mechanisms Improve capacity and scope of SSS MT Approve legal reforms for PAMI that align MT performance incentives with accountability Pensions: Replace 30 year contribution requirement with MT Increase coverage and efficiency one that allows contributions for fewer years while reducing fiscal costs with proportionally reduced benefits Integrate and expand benefits aimed at MT preventing poverty among the elderly Make all components of wages taxable and ST close loopholes on self-emnployment Review financial status, benefit levels and ST subsidies in existing pension system Simply benefit-granting process ST Reform investment regulations for greater ST competition and flexibility Reinforce independence of SAFJP ST xiv Social Assistance: Improve Jefes program by: (i) ensuring that ST Improve effectiveness of monthly stipend is appropriate for region, (ii) programs better supervision of registration process and municipalities, and (iii) narrowing scope of workfare activities and enforcing work requirements Devolve social assistance to provincial or MT municipal level and be willing to modify design and funding as appropriate Direct families with young children to Familias ST rather than Jefes and better coordinate this program with scholarship and family allowance programs Infrastructure Ensure effective and efficient long- Conclude fair renegotiations with private ST term provision of infrastructure operators services and expansion of services Improve the targeting of existing subsidies in ST in line with demand the sector Eliminate regressivity of subsidies Reform regulatory bodies to follow example of MT Resolve contingent liabilities in the electricity and gas regulators sector Change the design and implementation of trust MT funds in infrastructure Public Expenditure Management Provide incentives and means to Reform macro-budgeting at federal level by: (i) ST-MT promote fiscal discipline and instituting a two-stage budget process, (ii) efficient budget allocation and making the budget framework expenditure implementation comprehensive, (iii) instituting timely approval of financial statements Reform macro-budgeting at provincial level by ST strengthening financial reporting and making it consistent with federal accounts Improve budget allocation process by (i) vesting MT responsibility for budget planning to Jefatura; however, budget formulation and execution process could remain within MECON; (ii) giving line ministries resource envelopes and allowing them to reallocate resources without the threat of clawback by the Budget Office Improve service efficiency and performance at ST the federal level through: (i) improved financial management; (ii) upgrading SIDIF, (iii) establishing and enforcing clear criteria and procedures for establishing any new autonomous agencies; at the provincial level, responsibility for budgeting, accounting, and cash management could be integrated. Legend: ST = Short-term, measures can be instituted without additional legislation; MT Medium-term, measures are likely to require new legislation and/or consensus-building. xv 1. OVERVIEW A. BACKGROUND AND MOTIVATION 1.1 During 2002, Argentina's economy was in the throes of a severe economic and social crisis. GDP fell by 10.9 percent that year, for a cumulative decline of 20 percent since 1998. Poverty more than doubled since 1994. The Government stopped making payments on most of its debt. The Argentine peso, which had been pegged to the dollar at par from 1991 to 2001, fell to 3.8 pesos to the dollar in June 2002. After three years of successive declines, consumer prices rose by 41 percent during 2002 (though the year-on- year increase was 26 percent). Depositors saw their savings accounts frozen and limitations placed on the amounts they could withdraw from their checking accounts. Their dollar deposits were forcibly converted to pesos at an unfavorable exchange rate. Most utility and transportation tariffs were frozen. Many businesses faced bankruptcy and unemployment rose to 21.5 percent in May, before falling to 17.8 percent in October. 1.2 The severity of the crisis is even more astounding coming on the heels of strong economic gains through most of the 1990s. With the introduction in 1991 of the Convertibility Plan, which guaranteed the conversion of the Argentine peso to the US dollar at par and restricted the ability of the Central Bank to print pesos, the battle against the hyperinflation of the late 1980s was soon won. At the same time, a far-reaching privatization program and some institutional and structural reforms allowed Argentina to attract huge private investment in utilities and infrastructure and to improve public services. Argentina was able to achieve high rates of economic growth during 1992-1998, supported in part by high international capital inflows (Table 1). 1.3 The trigger for the crisis was investor perceptions on a risk of government debt default and of bank failures. This led to reverse capital flows and a large-scale withdrawal of deposits, thereby creating a self-fulfilling financial panic. Efforts to protect a run on Table 1: Key Economic Indicators 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 GDP (real growth, %) 10.3 5.7 5.8 -2.8 5.5 8.1 3.9 -3.4 -0.8 -4.5 -10.9 CPI(y-o-y%change) 24.9 10.6 4.2 3.4 0.2 0.5 0.9 -1.2 -0.9 -1.1 25.9 Consolidated Public Revenues (% of GDP) 23.4 24.6 24.2 23.2 22.2 23.2 23.7 24.3 24.3 23.5 22.4 Consolidated Public Expendit. (% of GDP)* 23.8 24.6 25.6 25.5 25.4 25.3 25.8 28.5 28.1 30.3 33.0 Consolidated Public Sector Bal. (% GDP) -0.5 -0.3 -1.9 -3.1 -3.6 -2.4 -2.1 -4.2 -3.9 -6.8 -10.6 Consolidated Public Debt (% GDP) 32.9 32.8 35.1 39.2 39.8 38.1 41.3 47.4 50.8 64.1 149.1 Unemployment (%)** 7.0 9.3 12.2 16.6 17.3 13.7 12.4 13.8 14.7 18.3 17.8 Intemational Capital Flows (US$ bn) 9.2 13.6 12.7 7.2 12.4 16.8 18.4 13.6 9.7 -4.6 -11.4 1ntemationalReserves(US$bn;endofyear, 11.1 15.1 16.0 16.0 17.7 17.9 19.4 19.0 18.3 14.9 10.5 GDP at current prices (US$ bn) 209 237 257 258 272 293 299 284 284 269 98 *Fedeml and provincial govemnments only: includes off-budget operadons. **Unemployment for October in 28 major urban conglomnerates Sources: IMF for consolidated public sector figures. INDEC for unemployment data. Ministry of Economy for all other variables. banks by limiting withdrawals led to a Box 1: Road to Crisis strangling of liquidity and economic activity. The widespread discontent this generated led I' Dec 1999: De la Rua assumes power in midst to street protests and the resignation of of an economic recession. President de la Rua. He was followed by a I, Jan 2000: Government tries to restore quick succession of new Presidents, a default investor confidence by raising taxes: however, on public sector borrowings, and the recession deepens and doubts about debt abandonment of the Convertibility Plan (Box sustainability increase. ati Dec 2000: Minister Machinea negotiates 1). US$40 billion package to extend debt maturities and ease fears of default. 1.4 While this economic crisis has been 11' Mar 2001: Machinea resigns when growth particularly severe, Argentina is no stranger still fails to materialize; replaced by Lopez to economic crises. During the 1980s alone it Murphy, who resigns two weeks later when a new underwent two bouts of hyperinflation, fiscal austerity package he sent to Congress faces caused mainly by fiscal imbalances. strong opposition. Alth g r Apr 16, 2001: Reinstated Minister Cavallo Although researchers differ on the causes Of proposes amendment to Convertibility Law to peg the recent cfisis, it is clear that without peso to basket of US dollars and euros. resolution of the debt default, new 0' Apr 25, 2001: President of Central Bank Pou investment, which is critical to long-term resigns amid disagreements with Cavallo; growth, will be slow to come. With the replaced by Maccarone. overhang of arrears, potential investors will 11 Jul 10, 2001: After being forced to pay 1410 be uncertain with respect to future prices, basis points over US Treasuries, government announces zero-deficit rule and pushes for IMF interest rates, and tax levels, thereby making support for its implementation. planning decisions difficult and inhibiting 1' Aug-Nov, 2001: Efforts to secure IMF and investment. US Treasury support fail when government fails to get new agreement with provinces on revenue 1.5 While a measure of macroeconomic sharing and fails to achieve its zero-deficit rule. stability and growth returned in 2003, it is 0, Dec 2, 2001: Government announces within a context of enormous government measures to restrict withdrawals on deposits to $250 per week. arrears, commitments to make huge $5 e ck arrpensars,icomm ments, to d make huge Dec 20, 2001: De la Rua resigns following compensation payments, and an street protests; Ramon Puerta assumes presidency. accumulation of contingent liabilities. Many >' Dec 23, 2001: Rodriguez Saa becomes of these fiscal commitments and contingent interim president and declares a formal default. liabilities arose from government laws and P Dec 30,2001: Saa resigns and legislative decrees that asymmetrically converted dollar assembly chooses Duhalde as president. assets and liabilities at arbitrary exchange 1, Jan 2, 2002: Duhalde assumes power, rates. Dollar deposits were converted to officially ends the currency peg, and announces pesos at a rate of 1.4 pesos to the dollar and the floating of the peso. indexed to the consumer price index but ' For example, Shuler (2002), Teijeiro (2001), and Mussa (2002) believe that expansionary fiscal policy was the primary cause of the crisis. Others such as Corden (2002), Calvo et al (2002), Hausmann and Velasco (2002), Powell (2002) and Serven and Perry (2003) believe it was a combination of the hard currency peg, expansionary fiscal stance, and currency mismatches from borrowing in dollars when revenues were in pesos. Stiglitz (2002), Rodrik (2002) and Sachs (2002) focus on the peso overvaluation, depreciation of the real, and financial panic caused by investors and depositors. Cline (2003) and Corrales (2002) give high weight to domestic political shocks. 2 borrowers were allowed to repay loans at par indexed to a wage index. The Government made a commitment to compensate banks for asymmetric pesification; however, the form of payment and its timing were uncertain. The government also declared dollar contracts to be converted to peso contracts at par and to freeze most utility and transport tariffs despite contractual agreements to the contrary. At the same time, in the transport sector at least, sizeable funds were allocated to partially compensate private operators for the tariff freeze. 1.6 Projections carried out by the World Bank and others show that even if Argentina were able to negotiate a substantial reduction in the net present value of its debt and to sustain a high economic growth rate, it will need to generate primary balance surpluses of the order of 4 percent of GDP, as opposed to historical surpluses of under 1 percent of GDP even during the boom years of the 1990s.2 But Argentina will have to generate even larger surpluses to try to reduce its high debt to GDP ratio if it is to avoid vulnerability to future reversals of capital flows. This is all the more challenging if-as recommended-it were to reduce its dependence on some taxes that inhibit growth. 1.7 The achievement of such primary balances will depend to a large extent on addressing certain weaknesses in institutions. One such weakness stems from the system of intergovernmental transfers. Over half of provincial government revenues in aggregate come from taxes imposed and collected by the federal government under a revenue sharing arrangement called coparticipation. Currently this system provides poor incentives for federal and provincial governments to improve tax collections or rationalize expenditures. It reduces transparency and accountability and decreases incentives for improving efficiency. Worst of all, it softens budget constraints, partly because provinces use the shared revenues as collateral to expand debt. Many provinces have managed to pressure the federal government to bail them out via such means as taking over their debts and arrears, assuming their public pension liabilities, and accepting their quasi-monies as payment on federal taxes or exchanging them for pesos. B. OBJECTIVES, SCOPE, AND STRUCTURE 1.8 The main objective of this report is to identify reforms in policies and institutions to enhance the efficiency, equity, and effectiveness of public expenditures at the federal and provincial levels. While these are laudable goals for any country, they are crucial for Argentina in order to generate the kinds of fiscal surpluses needed for fiscal solvency and macroeconomic stability discussed above. Without this, efforts to stimulate private-sector- led growth and generate employment-both of which are key for sustained poverty reduction-may not succeed. Without a sense that services on which public expenditures are spent are efficient, fair, and effective, governments will have difficulty in raising revenues or in implementing their reform agenda. However, important topics such as a detailed analysis of privatization and of the impact of liberalization policies on income distribution and employment are beyond the scope of this report. 2 A detailed macroeconomic and debt sustainability analysis will be addressed in a forthcoming World Bank report. 3 1.9 Two kinds of reforms are discussed: sectoral and process-related. The sectoral emphasis is on expenditures on social and infrastructure services, which in recent years, accounted for over 70 percent of consolidated public expenditures on average (Table B8). Effective provision of these services is necessary for improved living standards, economic growth, and poverty reduction. 1.10 The process-related emphasis is on certain institutions that affect efficiency in expenditures and fiscal solvency. These relate to provincial taxes, federal-provincial fiscal relations, fiscal rules, and public expenditure management. To keep the scope manageable and avoid duplication with other international institutions, reforms in federal tax policy and public administration, while important, are excluded from this report. 1.11 The primary audience for the report is federal and provincial policymakers responsible for approving and implementing most of the reforms. The secondary audience is civil society-NGOs, think tanks, academics, and the press-who can play a useful role in promoting and monitoring the reforms and/or in advising on them. The tertiary audience comprises the Board and staff of the World Bank and other international financial institutions, as well as the broader international financial community, who need to keep informed on Argentina and who have been or may be called upon to support Argentina's reform efforts. 1.12 The report is structured as follows. Chapter 2 analyzes the main fiscal trends over the past decade at the national and provincial government levels in order to set the context. It identifies several themes, most of which are addressed in later chapters. Chapter 3 suggests policy option to improve provincial tax efficiency and administration. It also discusses proposals to improve the current system of inter-governmental transfers and fiscal rules with the aim of aimed at ensuring fiscal solvency. 1.13 Chapters 4 and 5, which form the core of the report, contain analysis and recommendations to improve the efficiency, equity, and efficacy of expenditures in the social and infrastructure sectors respectively. Following an overview of social expenditures, Chapter 4 identifies needed reforms in the important area of education, health, pensions, and social assistance. Chapter 5 reviews the evolution and experience of infrastructure services-electricity, gas, water, roads, rail, urban transport, and ports-over the last decade. It then provides guidelines aimed at ensuring the sustainability, accessibility, and affordability of services in the future while improving the way public funds are allocated for infrastructure. 1.14 Chapter 6 discusses how to manage public expenditures better. It discusses the strengths and weaknesses of the budget formulation and control process and suggests reforms at the macro-budgeting level, including how to deal with the issue of fiscal commitments and risks. This chapter also identifies reforms in the budget allocation process and suggests how certain reforms in the budget process can lead to an improvement in the effectiveness of public services. 4 2. FISCAL TRENDS AND ANALYSIS 2.1 This chapter describes significant developments in federal and provincial finances over the last decade with a view to providing the reader the necessary background and motivation for the analysis and recommendations in later chapters. The themes and conclusions that emerge in this chapter-most of which are further elaborated upon and addressed in subsequent chapters-are as follows: * Argentina followed an expansionary procyclical fiscal stance through most of the 1990s; public debt at the federal and provincial levels increased rapidly and made the economy vulnerable to shocks. * Fiscal performance has varied substantially across provinces; while some provinces succeeded in generating surpluses during the boom years, others saw a significant worsening in their fiscal balances. * Measures taken to adjust to the crisis have dramatically altered the composition and levels of consolidated public expenditures and revenues. * Fiscal solvency has been, and continues to be, thwarted partly because of the problem of fiscal commons that promotes fiscal indiscipline by transferring the costs of certain decisions to others. * Realization of contingent liabilities played a significant role in the debt expansion during the 1990s; as a result of recent government decisions, fiscal risks persist today. * Government commitments on compensations to parties affected by recent decisions have complicated recovery efforts; ways to discourage and make more transparent the costs of such decisions in the future would be useful. A. THE EVOLUTION OF PUBLIC FINANCES Procyclicity of Public Finances 2.2 Despite strong average economic growth, accounts of both the federal and consolidated provincial government showed deficits every year between 1994 and 2001 (Figure 1). While low by international standards, these deficits grew steadily. Moreover, the federal debt skyrocketed, increasing from US$ 70 billion in 1993 to US$ 144 billion in 2001 (Table B2). This was due both to cumulative budget deficits and the settlement of claims arising from old disputes, mainly from contractors and from pensioners whose benefits had been unlawfully cut during the hyperinflation years. Consolidated provincial debt also increased rapidly, from $15.7 billion in 1995 to $32.8 billion in 2001 (Table B7), largely as a result of cumulative fiscal deficits. 5 Figure 1: Federal and Provincial Fiscal Balances 2.3 There was great diversity in fiscal outcomes between provinces. Most provinces performed well during the boom years, successfully privatizing their state- I ffi i ~ owned banks and other enterprises and o § B g - F - t resisting temptations to borrow. Some ,1 a F S 1l [ ! 12 Fprovinces even obtained surpluses. -2 _ ^ P W ~ tf ~ 1 |- l However, the province of Buenos Aires, -3 which had been a source of fiscal strength -4 in first half of the 1990s, increased -5 _ spending sharply during the 1996-1998 -e v)o co°8- > boom years (Figure 2). Initially, the a'i ai o> ' ia, !fi f o o growing deficits were financed by Federal Provincial ---Total privatization proceeds. Later, the province resorted to bond issues in international Source: Tables BI and B5 markets (without a federal guarantee) and, when market conditions deteriorated, it borrowed heavily from its state-owned commercial bank and issued quasi-money to finance its deficit. 2.4 Nine provinces, which exhibited deteriorating fiscal performance between 1997 to 1999, secured a favorable debt restructuring agreement with the Federal Government early in 2000. The expectation of such a bailout by these provinces may have deterred good fiscal governance. In contrast, the "other provinces" group (accounting for about half of provincial expenditures) managed a sharp improvement in fiscal results-only to be swamped later by the dramatic economic downturn (Figure 2). 4 2.5 The procyclicity has been facilitated by: (i) revenue-sharing arrangements by which increased federal revenues result in proportional increases in transfers to provinces, (ii) enhanced ability to borrow in international capital markets during boom times, and (iii) the lack of fiscal rules to smooth out fiscal balances over a business cycle. While many provinces exhibited good fiscal governance, others used the increased revenue and enhanced borrowing capacity to raise expenditures. The Federal Government used it as the opportunity to recognize liabilities that had been implicit or in the distant future and to assume the pension liabilities of the provincial public sector employees. While these actions were efficiency-enhancing, they led to large increases in debt and interest payments and increased Argentina's vulnerability to external shocks. Chapter 3 shows how procylicity could be reduced. 3 The province's wage bill increased from $3.6 billion in 1996 to $5.4 billion in 1999. Part of the increase is explained by teacher costs associated with the expansion of obligatory education. Total education expenditures, both wage and non-wage components, increased from $2.5 billion to $3.8 billion. However, a significant share of this increase was investment in new schools to accommodate increased enrollment. Thus the increased teacher costs probably explained at best half the increase in the wage bill. 4 While this is only one possible grouping of provinces, it reveals the substantial diversity in fiscal performance. In fact, the small deficit shown for 1998 for "other provinces" would actually be a surplus if one deducts the "costs" of Santa Fe province's privatization of its provincial bank. The negative net worth of the bank was realized "above the line" at the time of privatization. 6 Figure 2: Provincial Fiscal Balance (excluding privatization revenues) (% of Revenues) 5% 01% -10% -Y 1 01% -15% -20% -25% -30%- /S -35% -40%Y -Province of Buenos Aires -9 "Original" Debt Restructuring Provinces - Other Provinces Source: MECON Trends in Expenditures and Revenues 2.6 Four developments are striking with respect to the level and composition of public expenditures and revenues-one occurred last year and three over the decade: * the sharp reduction in public expenditures and federal revenues in 2002; * the large increase in interest payments between 1993 and 2001; * the steep decline in expenditures on economic services; and * the high share of expenditures classified as social services in the budget. 2.7 While consolidated public expenditures as a share of GDP grew from 31.7 percent in 1993 to 35.8 percent in 2001, they fell sharply to 29.1 percent of GDP in 2002 (Table B8). The expenditure decline in 2002 is due to a combination of accumulating arrears on interest payments on government debt, which are excluded in published accounts, and the real reduction of public sector salaries, which were largely unchanged even as inflation returned. The share of federal revenues in GDP also fell, from an average of 19.4 percent between 1993 and 2001 to 17.8 percent in 2002. This occurred despite the introduction of two new taxes in 2001 and 2002-the financial transactions or "check" tax and the export tax (Tables B 1 and B3). While provincial revenues also fell, this was entirely due to the reduction in federal transfers as a result of lower federal revenues and the introduction of export taxes that were not shared with provinces. 7 2.8 Interest payments on the consolidated public sector debt, which were 1.8 percent of Figure 3: Spreads on Argentine Debt GDP in 1993, rose to 2.7 percent of GDP in 8000 1998, before jumping to 5.3 percent of GDP in 7000 2001. Their corresponding shares of total public I - expenditures were 5.8, 8.7 and 14.8 percent. + 5(0(1 Thus the increase in interest payments were a > F major contributor to the public sector deficits in X fb) .L- later years. This was due both to the rising debt 20(_0 (see above) and a large increase in interest rates wo0o after the Russian crisis and investor skittishness. hn Spreads on Argentina's debt skyrocketed from 9 o' o o o c o about 400 basis points in 1998 to a whopping ,_ 7000 basis points by the second quarter of 2002 (Figure 3). Soure: JP Morgan 2.9 The third striking feature of fiscal trends over the last decade was the decline in the category known as "economic services"-comprising mainly public expenditures on energy and transport. These fell from 3.4 percent of GDP in 1993 to 1.8 percent of GDP in 2001, before further declining to 1.4 percent in 2002 (Table B8). This was largely due to the privatizations and awarding of private concessions at both the federal and provincial level, which precluded the need to finance infrastructure investments or operational losses. The sector was transformed from a net recipient of annual subsidies of US$1.8 billion to a net contributor of US$4.7 billion, highlighting the need to ensure that the bulk of these services continue to be privately provided (see Chapter 5). 2.10 The largest share of consolidated public expenditures (over 60 percent) are social services, comprising mainly pensions, health, education, and social assistance programs (Table B8). This relatively high level of expenditures (US$1500 per person) has not resulted in commensurate social outcomes in Argentina, suggesting the prevalence of serious efficiency and equity issues. Chapter 4 attempts to identify and address these. The Problem of the Fiscal Commons 2.11 The inability for successive governments in Argentina to achieve fiscal solvency can partly be attributed to the problem of the fiscal commons. Broadly put, this problem occurs when the costs of a decision by one level of government or agency is borne by another. Argentina's system of fiscal federalism offers plenty of opportunities for such behavior to occur. 5 The coparticipation system, with its intertwining of fiscal and provincial finances, provides an opportunity for such types of behavior. Provinces have managed to get discretionary transfers, change the rules for automatic transfers, obtain concessional lending, pass off provincial liabilities to the federal level, and gain federal tax breaks. 5 See Porto and Sanguinetti, 2001 and Tomassi, Saiegh and Sanguinetti, 2001 for political and economic analyses of the Argentina's fiscal federalism. 8 2.12 Despite federal government attempts to control rent-seeking behavior by provinces,. there were several episodes of invasion of the commons in the 1990s. For example, certain provinces were able to get their deficits financed by the federal government at attractive - terms; others managed to negotiate certain fixed transfers.6 Similarly, in the mid-1990s, about half the provinces managed to transfer their public pensions to the federal system, thereby ridding themselves of a large liability. Also, the province of La Rioja was able to obtain extraordinary shares of the discretionary subsidies known as ATNs. In 1999, 12 small provinces managed to strike a debt restructuring deal with the federal government. 2.13 More recently, when provincial governments were unable to repay their debt following the devaluation, the federal government once again stepped in to bail them out. In February 2002, it agreed to undertake a debt swap with provinces at favorable interest rates and exchange rates under a series of bilateral fiscal pacts that were ratified on March 5, 2002. It also agreed to: (a) return to the traditional regime of proportional transfers from the system of fixed transfers that had been in place briefly, (b) put 30 percent of the proceeds of a financial transactions tax into the general revenue-sharing pool, and (c) allow provinces to freely allocate revenues that had been earmarked for specific expenditures. In return, provinces agreed (a) to reduce their fiscal deficits by 60 percent in 2002 and reach fiscal balance in 2003 and (b) to not issue any additional debt, whether directly, via arrears with suppliers, or via issuance of quasi monies.7 2.14 These arrangements were implemented through a so-called Plan of Orderly Financing (PFO). Under the Plan, the federal government was authorized to reduce its transfers under the coparticipation arrangement in the event that a province missed a payment on the swapped debt or added to debt in any form. Provinces also committed to to approving a new coparticipation agreement and to increase transparency and timeliness of data, though such agreements in the past had been ignored. 2.15 The federal government has also taken decisions whose costs are borne by provinces. In the early 1990s, noting that transfers to provinces had increased because of increased federal tax receipts, the federal government transferred the responsibility for secondary education and non-tertiary university education to provinces without providing additional funds for this purpose. Moreover, it provided significant salary increases to teachers before the transfer. Many jurisdictions were faced with the need not only to finance the salaries of the transferred teachers but were obliged to bring their own teachers' salaries up to the level enjoyed by the newly transferred teachers. Finally, autonomous agencies such as PAMI and even private institutions such as the Obras Sociales have - - managed to incur deficits and successfully managed to transfer their accumulated losses to the federal government (see Chapter 4). Chapters 3 and 6 discuss mechanisms aimed at reducing incentives that lead to the problem of the fiscal commons. 6 See World Bank, 2001, which is also available at www.bancomundial.org.ar. 7Ten provinces issued quasi-monies totaling US$7.7 billion. The federal government recently agreed to provide financing to retire these quasi monies. 9 B. CONTINGENT LIABILITIES, COMMITMENTS, AND OTHER FISCAL RISKS Introduction 2.16 As mentioned in Chapter 1, one of the Box 2: What are contingent liabilities and reasons for the sharp increase in public debt commitments? was the realization of contingent liabilities and Contingent liabilities are liabilities (such as fiscal commitments (Box 2). Chapter 1 also bank-deposit insurance) that require cash suggested that the improvement in fiscal outlays only if certain events (such as bank finances during 2002 and 2003 is precarious not default) occur. Commitments are obligations only because it is based on the non-payment of extending beyond the current budget horizon a major part of the debt but also because the that do not show up as debts on the government's balance sheet (such as obligations government made certain commitments on to pay pensions to government employees or future payments and accumulated contingent commitments to compensate banks for liabilities whose consequences will drag on for asymmetric pesification). Both contingent years. This section attempts to document some liabilities and commitments may be explicit of these fiscal risks and commitments with a legal obligations, or they may be implicit view to raising awareness of the problem. practically impossible for governments to resist. Later, in Chapter 6, reforms in the way that Problems can also arise with ordinary "on- such liabilities are recorded are suggested with balance-sheet" assets and liabilities when the a view to reducing incentives to undertake values of those assets and liabilities are volatile. future contingent liabilities and commitments. Borrowing in a foreign-currency under a fixed- exchange rate regime is an example. When 2.17 Argentina is not unique in accumulating interest rates on foreign-currency debt are lower liabilities and commitments. When than on local-currency debt because of concerns continget .about devaluation, governments may be governments are under strong pressures to tempted to borrow in foreign-currency, even reduce reported expenditure, they have an though the total expected cost in the long term incentive to substitute opaque for transparent may be higher. expenditure. They tend, in particular, to replace upfront budgeted cash outlays with decisions to incur contingent liabilities and off-budget commitments and to take other fiscal risks that limit near-term cash outlays. As seen below, such actions have been greater in Argentina than other countries. Moreover, they have increased during the crisis. When a country such as Argentina is unable to meet even its immediate, ordinary obligations, the temptation to take on contingent liabilities and make long-term commitments is even stronger. Off-budget expenditures and growth of debt 2.18 One way to measure the incurrence or realization of contingent liabilities, off- balance-sheet commitments, or other fiscal risks, is to see what part of the growth in debt as a share of GDP is unexplained by the accumulation of fiscal deficits, valuation changes in the stock of debt, privatization receipts and GDP growth. Between 1993 and 2001, such "debt recognition not recorded in expenditures" augmented Argentina's debt on average by 1.5 percent of GDP per year (Figure 4). The problem was particularly large early in the decade when the Government of Argentina issued bonds to settle longstanding claims against it, mainly from suppliers and pensioners. 10 Figure 4: The contribution of debt recognition to the growth of debt in Argentina 14.0- 12.0 - 10.0 8.0 2.0 - 4.0. -20 -6.0- r Contribution of debt recognition Contribution of other factors -icrease in consolidated government debt Source: Adapted from IMF staff data. Other contributions include the primary balance, interest payments, privatization receipts, valuation changes, real GDP growth, and inflation. 2.19 The realization of contingent liabilities, off-balance-sheet commitments, and other fiscal risks can show up in what have been called "hidden deficits"-namely the difference between the growth of public debt and reported budget deficits. One study finds that during the 1980s and 1990s Argentina's hidden deficits averaged about 2.5% of GDP, less than in some countries in the sample, but more than in most (Kharas and Mishra, 2000). Interestingly, the authors of the study find that hidden deficits are associated with currency crises. Liabilities created in response to the crisis 2.20 Facing extremely difficult choices during the crisis, the government took decisions that will impose costs on future governments for many years. The bulk of these are commitments and contingent liabilities stemming from the pesification of financial assets and liabilities and the freezing of utility tariffs. The asymmetric pesification and indexation of banking deposits and loans was associated with a commitment by the government to compensate the banks-the cost of which is estimated at US$10 billion (Box 3). Moreover, a contingent liability was created when depositors protested being paid in pesos at an unfavorable exchange rate. Following a series of judicial injunctions when the courts ruled that part of the deposits needed to be returned to depositors in dollars, the court eventually ruled that all depositors in one case (Province of San Luis vs the Executive Branch) needed to be repaid in dollars. The cost of redollarization, including the judicial injunctions, is estimated at US$ 11 billion. 2.21 Other contingent liabilities in the financial sector arose when the government forcibly converted dollar-denominated bonds held by private pension funds into guaranteed loans at an exchange rate of 1.4 pesos/dollar at below-market interest rates. Most pension fund managers have refused to accept the conversion, and while the Box 3: Asymmetric Pesification of Banldng Assets and Liabilities and Government Bonds Before pesification, some of banks' loans (assets) and deposits (liabilities) were denominated in pesos and others were denominated in dollars. The government's pesification decision converted the domestic dollar loans and deposits into peso loans and deposits, at exchange rates specified by the government. The dollar loans were converted into peso loans at an exchange rate of 1 peso to the dollar: a loan of 100 dollars became a loan of 100 pesos. The deposits, on the other hand, were converted at an exchange rate of 1.4 pesos to the dollar: a deposit of 100 dollars became a deposit of 140 pesos. The asymmetric pesification thus reduced the banks' net worth. Moreover, the government specified that the value of the loans and deposits would be indexed, and the indexation, too, was asymmetric. Deposits were indexed to the consumer price index (the CER in Argentina), while loans were indexed to a wage index (the CVS). Because price inflation has been greater than wage inflation, the asymmetric nature of the indexation worsened the banks' financial position. At the time of the decision, the government agreed to compensate the banks. The cost has been estimated at more than US$10 billion. The government's potential liabilities arising from pesification do not end here, however. New contingent liabilities arose from the forced conversion of dollar deposits into pesos because the market value of the dollar was always above the conversion rate ($1.4 + CER). Some depositors sued the government and in most cases courts found in favor of depositors and required the banks via judicial injunctions (amparos) to pay part of the original deposits in dollars-which again created a liability for the government that is estimated at $13.5 billion. At the end of March 2003 the Supreme Court ruled that all deposits in one case were to be paid in dollars. This led the Government to offer depositors compensation at an exchange rate of $2.98 per dollar to be paid in two parts-in cash at the $1.4/dollar exchange rate plus the CER, and the balance in government bonds. At this exchange rate, the total amount of compensation that would be paid to depositors, including the amparos, would be US$ 11 billion. Finally, there is a contingent liability from the decision to service certain kinds of dollar-denominated bonds-mostly held by institutional investors such as banks, pension funds, and insurance companies-in pesos at an exchange rate of $1.4/dollar and indexing to the CER. If the government were forced to repay bondholders at the market rate of the dollar, the additional liability is of the order of $56 billion. government has not defaulted on this debt, a potential decision to revert the instruments into dollar-denominated assets could result in a maximum fiscal cost of $20 billion, assuming that the pension funds are not forced to take a real reduction in their bonds. 2.22 Further contingent liabilities arose from the government's decision not to allow privatized utilities to increase their rates after the devaluation, even though they were typically indexed to the US dollar. These claims could run into billions of dollars. Nine claims totaling US$ 4 billion were filed in ICSID by April 2003, though some of them are unrelated to the tariff freeze (Table D8). Other large claims are being filed. 2.23 In conclusion, Argentina is still in the midst of a fiscal crisis, in which the government is unable to meet its immediate, ordinary obligations. Commitments and contingent liabilities created in the recent past are now creating very large demands on cash. It is, of course, too late to prevent the problems created by those liabilities now. Yet Argentina can improve its fiscal position, and reputation, by making better decisions about contingent liabilities and extra-budgetary commitments in the future. Making better decisions will be tough, because the temptation to substitute contingent or future payments for certain, immediate payments will remain strong. Nonetheless, other countries are making progress on this matter and Chapter 6 discusses some options. 12 3. FEDERAL-PROVINCIAL FISCAL RELATIONS A. FEDERAL TRANSFERS TO PROVINCES 3.1 The provincial public sector relies upon federal transfers for over half its revenue, with the reliance of some of the smaller provinces being as much as 80 or 90 percent. The vast majority of federal to provincial transfers are automatic (over 90 percent). These are grouped under the formal title of "coparticipation". However, the coparticipation system is an extremely complicated mix of: (i) an untied general fund, which is fed primarily by VAT and income taxes but also by non-fuel excise taxes and, since March 2002, the financial transactions tax, (ii) tax-sharing arrangements for fuels, and (iii) a variety of earmarked taxes. The last two categories-which constitutes about one-third of automatic transfers-have distribution rules across provinces that are distinct from those of the general fund, whose legislation dates back to 1988, though it has undergone several modifications via pacts. The non-automatic transfers are mostly federal programs in the social sectors where the provinces are the executing agencies receiving funds conditional on federal guidelines. 3.2 During the 1990s, provinces received an average of about 35 percent of all federal taxes. With the introduction in 2002 of the export tax, which is not shared with the provinces, the share of federal taxes transferred to the provinces-the so-called primary distribution (Box 4)-declined to 25 percent during 2002. The federal VAT and income tax revenues, which accounted for about 60 percent of federal revenues in recent years also account for the majority of provincial revenues from shared resources. However, not all revenues from these two taxes enter the pool of coparticipation funds. A variety of earmarking arrangements, mostly to finance the federal social security system, siphon away funds. Moreover, certain taxes-most notably federal payroll taxes and export taxes-are not shared at all with the provinces. Also the sharing of certain fuel and excise taxes are legally tied to public investment in particular sectors. However, since 1999 a series of federal agreements virtually untied these transfers. 3.3 The system of distribution between provinces-called secondary distribution and described in greater detail in Annex I-does not redistribute from rich jurisdictions to poorer ones as required by the Constitution. There is positive correlation (0.33) between the transfers per capita and value-added per capita and a negative correlation (-0.20) between transfers per capita and child mortality, suggesting that richer provinces are 8 See Annex I, which uses a figure to explain the existing revenue-sharing arrangement. For greater detail, see World Bank, 1996, 1998, and 1999. The city of Buenos Aires, while treated as province. for most purposes (and is not part of the Province of Buenos Aires), does not participate in the general coparticipation fund. However, it receives a fixed lump-sum transfer, participates in the tied arrangements, and has its police and judiciary expenses paid for and run by the federal government. Federal cash transfers account for roughly 10 percent of city revenues. 13 associated with higher transfers per Box 4: Primary and Secondary Distribution capita and that the needier provinces are Primary distribution refers to the vertical division of associated with lower transfers per coparticipation funds across levels of government. capita. Historically, the division has been based on certain percentages of each individual federal tax. For a brief 3.4 If one were to add natural period-from the end of 1999 to March 2002-fixed resource royalties, the pattern of nominal amounts were put in place according to federal- resoure royaties, he patern ofprovincial joint agreement. Since March 2002, the old redistribution gets worse. The richer system of fixed percentages of particular taxes has been Patagonian provinces with natural back in place. resources benefit in three ways: Secondary distribution, which refers to the distribution royalties from the production of natural of transfers across provinces, is determined by a set of resource-based products, gross receipts laws governing the general "coparticipation" pool and tax collected from the sale of these the specific tax-sharing arrangements. The pool is products, and above-average transfers distributed according to coefficients fixed in place in a 1988 law. These coefficients retained some "memory" per capita from the federal government. of a formula that had been used from the early 1970s to While economies of scale mnight justify the mid- 1980s; however, the law governing the use of higher per capita transfers in some small this formula expired in the mid 1980s, and in a distorted provinces, it is difficult to justify an high inflation environment, the shares entered substantial eight-fold difference (Table B 10). negotiation, with the outcome being based more on political considerations rather than the earlier, more sensible indicators of investment needs. 3.5 While the perverse redistributions caused by the secondary distribution are of concern, the inefficiencies and poor incentives arising from the primary distribution system are an even bigger concern. Provinces have an incentive to treat the system of intergovernmental transfers as a fiscal commons, resources can be extracted to finance expenditures that are politically attractive, but with low social returns. Moreover, federal authorities have less incentive for collecting taxes and provinces have little incentive to assist the federal government in collecting taxes in their province-for every additional peso collected by the federal government, just over half goes to the provinces, and only a very small share to the province from which it was collected. 3.6 The sheer complexity of the current "labyrinth" of revenue sharing and tax-sharing arrangements diminishes transparency and incentives for efficiency. Few Argentines have a full understanding of how and why their province receives its main sources of revenues. Many citizens do not really know who pays for the teacher's salary at their children's school, nor do they really understand how their federal tax monies are spent. As provincial taxes paid by residents are not directly linked to provision of provincial public services, it reduces incentives to monitor provincial government and undermines transparency and accountability, and reduces incentives for efficiency. 3.7 The inequitable system of secondary distribution also results in wasteful spending. One way where this shows up is in public employment levels. Virtually every jurisdiction with above average employment levels also has above average transfers per capita-the correlation coefficient between the two measures is a staggering 0.85. Provinces such as La Rioja and Santa Cruz have roughly three times the number of public employees per capita than provinces such Buenos Aires and C6rdoba (Table B 10). Higher levels of 14 public employment relative to population in low-population density provinces can be expected; however, the dispersion in Argentina is difficult to justify on these grounds. 3.8 Another consequence of the dependence of provinces on federal transfers is that it softens budget constraints. Provinces, especially the politically powerful ones, know that the federal government will bail them out. Although the bailouts have political cost since they are typically accompanied by fiscal conditionality and some loss of autonomy for the provinces, this is not enough of a deterrent to fiscal profligacy. 3.9 In conclusion, the system of inter-government transfers suffers from both efficiency and equity problems and provides poor incentives for improving fiscal discipline, transparency, tax administration, and efficiency of public expenditures. Without reform of the system of transfers-namely the system of coparticipation-it will be difficult to achieve fiscal solvency. However, before discussing a possible reform, it is useful to analyze the way that provinces collect taxes in order to see the potential for greater decentralization of taxes (and hence closer fiscal correspondence) as well as to suggest improvements in the efficiency or equity of the tax system, which can enhance the benefits from the coparticipation reform. B. PRoviNcul TAxEs 3.10 In the year 2000-the last "normal" year-provincial current revenues were 11.5 percent of GDP, comprising federal transfers of 5.8 percent of GDP, own-source taxes of 3.8 percent of GDP and other revenues of 1.8 percent of GDP. Provinces collected only 33 percent of their revenues and 30 percent of their expenditures from taxes (Table B5). Tax effort varies dramatically across provinces. During 1993-2001, average tax collection as a share of provincial GDP varied from a low of 1.5 percent in La Rioja to a high of 4.5 percent in Entre Rios. In per capita terms, it ranged from $57 pesos for Formosa to $879 for Buenos Aires (Table B9). 3.11 Provinces mobilize own tax revenues from four main sources. The largest share- 56 percent of revenues-is derived from a gross receipts tax (GRT). This is a multi-phase tax on turnover with differentiated tax rates. Typical tax rates are 1 percent for primary activities, 1.5 percent for industrial activities, and 2.5 to 3.5 for services, though there is considerable variation between provinces. The second most important tax is a tax on property-real estate and buildings. The other important taxes are a tax on the value of automobiles and a stamp tax, which is a charge on the transfer of property or signing of contracts, based on the value of the asset exchanged. Along with the GRT, these taxes account for over 90 percent of provincial taxes. In addition, a handful of provinces receive significant revenues from natural resource royalties, whose rate and base are determined by federal law (Table B 10). 9 In addition, some of the low population density provinces are characterized by high density zones within the province - i.e., most of the population lives close to the provincial capital or in two or three urban agglomerations. 15 3.12 While taxes, in general, have a negative impact on economic efficiency and output, the GRT is particularly injurious to the country's economy for the following three reasons as mentioned in the literature.'0 First, as the tax is imposed on sales on intermediate stages of production, it creates an incentive for vertical integration even when firms specializing in certain stages of production may be more efficient. In addition, this cascading effect leads to varying tax rates across goods, depending upon the degree of vertical integration of firms producing those goods. Second, it creates a bias against exports and in favor of imports. This is because unlike VAT, there are no export credits for the GRT and to the extent that imports are only taxed at the later stages while domestically produced goods are taxed at intermediate stages, it places local producers at a competitive disadvantage with respect to imports. Third, because of the way the tax is implemented, it acts as a barrier to inter-provincial trade. Although, the provinces started a gradual process of exempting primary and industrial activities from the GRT under the 1992/1993 fiscal pacts, provinces chose to grant tax exemptions only to those activities undertaken within their own territory. Thus inter-provincial sales pay the GRT, while within-province sales are exempted. Even within the province, certain activities are exempt while others are taxed at a surcharge, which further distorts price signals and raises questions of transparency and fairness. 3.13 The stamp duty, which accounts for 9 percent of taxes raised by provinces, is another tax that harms production, in this case by raising transactions costs on the transfer of property and certain contractual agreements, potentially discouraging some efficiency- enhancing transactions. In recent years, some provinces have also been charging stamp taxes on foreign direct investment in natural resource sectors, thereby inhibiting investment in these sectors. 3.14 In addition to efficiency and equity issues relating to tax policy, Argentina has a serious problem with tax administration and with giving tax breaks to selected activities (tax expenditures). In addition to the unfairness of exempting some activities, it is a major efficiency concern. The low yield of taxes forces the government to set high tax rates, which in turn, lowers economic efficiency. " l To illustrate this, consider the yield on the Argentine VAT. The VAT yield is the amount of tax collected as a share of GDP, divided by the tax rate. In 2001, with a 21 percent VAT, the federal government was only able to collect 5.71 percent of GDP; hence its yield was 27.2 percent. In 2002 this dropped further to 23.1 percent. Not only does this show a substantially decline from its yield of 33.3 percent in 1998, Argentina is worse than the majority of countries in Latin America (Table 2). If Argentina could increase its yield to 40 percent, which is still well below those in Honduras, Guatemala, Chile and El Salvador, it could reduce its tax rate to 17.5 percent while collecting 7 percent of GDP from VAT. 3.15 While there is no available information on the extent to which there is non- compliance on provincial taxes, circumstantial information suggests that it is high: tax amnesties yield significant short-term revenues to provincial treasuries; prosecution of tax- 1° For example, World Bank (1996) In some theoretical models, the deadweight loss of the tax system increases with the square of the tax rate. 16 Table 2: The yield of VAT in Latin America offenders is almost nonexistent; and CTryle 2Th Tax rt V iLetin Yiel provincial tax administrations are often Country Tax rate Collection/ Yield % GDP staffed by untrained personnel and starved Argentina 21.0 6.2 29.5 of financial resources to carry out their Bolivia 14.9 5.7 38.3 mandate. Brazil 20.5 8.6 42.0 Chile , 18.0 8.5 47.2 3.16 These conditions reflect a lack of Colombia 15.0 4.9 32.7 political will to reduce evasion. However, Costa Rica . 15.0 6.5 43.3 the tide seems to be changing. Provinces Dominican Rep_ 8.0 3.1 38.8 are taking measures to increase tax Ecuador 12.0 4.4 36.7 compliance. For example, some provinces El Salvador 13.0 5.3 40.8 have introduced tax withholding. For Guatemala 10.0 3.7 37.0 instance, both Buenos Aires and Misiones Haiti 10.0 2.2 22.0.' Honduras 12.0 3.8 31.7 require financial institutions to retain a Mexico 15.0 3.2 21.3 percent of deposits in accounts of GRT Nicaragua 15.0 10.0 66.7 taxpayers, where the income withheld Panama 5.0 2.0 40.0 constitutes a credit against the GRT. Also Paraguay 10.0 4.5 45.0 there are some timid attempts to prosecute Peru 18.0 6.3 35.0 tax delinquents. Some provinces such as Uruguay 23.0 8.4 36.5 Misiones and C6rdoba have even Venezuela 15.5 3.2 20.6 successfully out-sourced collection and/or Average 14.3 5.3 37.1 prosecution of delinquent taxpayers to the Source: Keen, et. al., 2001 private sector. C. IMPROVING PROVINCIAL TAX EFFICIENCY AND ADMINISTRATION 3.17 As discussed earlier, a reform of the system of federal-provincial fiscal relations is needed for fiscal solvency, improved efficiency and equity, and enhanced transparency and accountability. Attainment of these objectives will require reduced provincial dependence on federal transfers. As a first step to this-and regardless of when or whether the coparticipation reform proposal is passed-reforms in provincial tax policy and administration are needed to improve provincial tax efficiency and admninistration while moving closer towards the goal of fiscal correspondence.12 Before suggesting specific recommendations, it is instructive to review five tax reform proposals and the experience of efforts to improve tax administration. Review of Tax Policy Proposals Patching up the current tax system 3.18 Proposals in this area involve eliminating the stamp tax and removing GRT-related internal trade barriers by harmonizing GRT rates and doing away with tax exemptions. The main advantage of this option would be to avoid experimentation with new taxes while achieving some gains in efficiency. The harmonization of GRT rates would eliminate the harmrful use by some provinces of providing tax breaks to attract certain 12 Recall that reforms in federal tax reforms, while very important, are beyond the scope of this report. 17 businesses at the expense of already established firms. It would also simplify monitoring and compliance of firms that sell in more than one province. However, it would leave unchanged many of the negative effects of the GRT. Exempting primary and industrial activities as had been suggested in the 1992/1993 fiscal pacts would reduce these negative effects, but "producer" provinces are likely to oppose extending tax-exemptions to out-of- province firms. Similarly, the harmonization is likely to be opposed by the provinces that have successfully managed to attract industry at the expense of other provinces. Introducing a provincial VAT 3.19 In principle this is an attractive option. The VAT does not suffer from the distortions in the GRT described above. It has the added advantage of encouraging the reporting of activities as sellers seek credits for taxes paid lower down the production chain. There has been successful implernentation of a dual federal-provincial VAT in Quebec, Canada and research from a variety of countries-including Brazil-suggests that a provincial VAT is feasible.13 Administration could be handled in a coordinated fashion or centrally. A provincial surcharge on the federal VAT may be easier to administer. In any case, a single definition of the tax base across provinces would be important for ease of administration. However, the national VAT, at 21 percent, is already high and fraught with evasion. Increasing the VAT rate in order to allow for a provincial VAT would exacerbate the evasion problem and increase the discrimination against those who pay the tax. Thus its implementation is not possible at this stage. Retail sales tax 3.20 A single-phase alternative to the neutral, multi-phase VAT using the destination principle would be a retail sales tax, such as used by most states in the United States. The sales tax is much simpler to implement-it avoids the complexity of dealing with inter- provincial sales and the zero-rating/CVAT issues discussed in the footnote above. The main disadvantage is that it is easier to evade as all the tax is collected in the last transaction and there is an even higher incentive to evade it than the VAT, which allows credit for taxes on inputs and thus encourages reporting. In addition, being a destination- based tax, it would also be opposed by the "producer" provinces. The public may also oppose the tax, which would be perceived as being a tax on the consumer as opposed to the GRT, where the tax is imposed on the producer. Finally, the retail sales tax-as generally implemented-is never purely a retail tax. A variety of businesses -especially small businesses in the service sector-end up paying the retail sales tax on goods that are actually inputs to the production of their particular good or service. For these reasons, it is not a desirable option. 13 See Annex II. In particular, it may be helpful to introduce a special federal compensating VAT (CVAT) to cover inter-provincial transactions as discussed in the annex. This would keep the chain of credits intact and act as a deterrent against falsely declaring local transactions as inter-provincial ones to evade the tax. Another implementation approach for a joint federal-provincial VAT would be the "Viable Integrated VAT" or "VIVAT." This alternative is discussed briefly in Annex II. 18 Corporate cash flow tax 3.21 More recently, some have proposed a corporate cash flow tax (CCFT) to replace the GRT. The base of this tax is the value of cash sales less cash payments for materials and labor, which gives it a closer relationship to profits than the turnover tax and is less distortionary. Like the GRT, it is perceived as a tax on firms rather than a tax on consumption, making it more palatable to the public. A side benefit of this tax is that firms would have to document its expenditures on inputs and wages, which could then be used for monitoring and cross checking to combat tax evasion. One downside is that it is a new tax with more uncertainty on the revenues it will yield. Another downside are the difficulties with allocating the tax base to different provinces when firms carry out business in more than one jurisdiction. For these reasons, this option is not feasible for provincial governments but it could be a consideration for the federal government, where there is no need to assign the tax base across provinces and where it has been suggested as an alternative to the corporate income tax. Use of existingfederal taxes 3.22 One tax reform proposal envisages provinces eliminating the GRT and stamp taxes and using federal taxes to make up the fiscal loss. A simple way to do this would be to "piggyback" onto federal taxes by applying a provincial level surcharge. This would avoid duplicating efforts at tax administration while providing incentives for provincial authorities to cooperate with federal tax authorities. Many countries have subnational personal income taxes that are administered either directly by the subnational entity as in the United States or by employing provincial surcharges as in Canada. The latter is probably more feasible in Argentina, given the tax capacity of most provincial tax authorities given the economies of scale of managing information bases that track the flow of income from a variety of sources. 3.23 The three candidates suggested for piggybacking of taxes in Argentina are the personal assets tax (bienes personales), excise taxes, and personal income tax. Given that the personal assets tax is a natural complement to the real estate and automobile taxes already implemented by provinces, piggybacking would enhance provincial government incentives to cooperate with federal authorities and enhance revenues. 3.24 Piggybacking onto federal excise taxes will be more difficult to implement in Argentina. Excises are generally imposed on goods such as tobacco and alcohol, which have negative externalities and where the tax revenues partially compensate the province . for the higher costs associated with their consumption. However, in Argentina excises are collected at the point of production, not the point of sale, which is fairer in these cases. While this is not an issue for a federal tax, decentralizing it to the provinces would entail having to track inter-provincial sales. The challenge would be compounded if there is no harmonization of excise rates across provinces. It is likely to be unworkable. 3.25 The third option is to piggyback onto the personal income tax, which faces none of the jurisdictional problems discussed above. The main problems relate to the federal tax base and tax rate. The minimum threshold, while it has reduced because of inflation, is still too high, exempting too many individuals and limiting tax proceeds. In addition, 19 adding a surcharge will increase incentives for tax evasion. Thus piggy-backing of this tax makes sense in conjunction with a lowering of federal tax rates and an expansion of the tax base. Efforts to improve tax administration 3.26 As seen earlier, several provinces have taken positive measures on tax administration, both to strengthen enforcement and to improve taxpayer services. By and large, there have resulted in increased revenue collection with further gains projected. Some of the more successful efforts have focused on closer collaboration with the federal tax collection agency, AFIP, and withholding taxes at source. In addition, provinces have endeavored to increase property tax collection by improving the roll, identifying and concentrating monitoring efforts on property owners with more than one property, and crosschecking information with municipalities and public utilities. However, there is still much scope for raising revenues from existing property taxes. 3.27 Other measures include a more forceful monitoring of taxpayers, including sending notices to non-filers immediately after the filing deadline, more active legal prosecution of non-compliers, including seizing and auctioning properties to pay tax debts and closure of shops whose owners have breached the tax law-something that had rarely been done in the past, and increasing the number of audits. Provincial tax authorities are devoting resources to improve the quality of taxpayers services through such means as increasing the number of places to pay taxes, authorizing the use of magnetic means to declare taxes, and promoting the use of internet to interact with the tax administrations. Some provinces also offer a discount of up to 10 percent for those who pay on time. Empirical studies suggest that the taxpayers' level of satisfaction with the way they have been treated by the tax administration might be an important influence on future levels of tax evasion (Thurman, 1991, Wallschutzky, 1984). Especially because honest taxpayers have been discriminated against by frequent tax amnesties, providing a discount to those who pay on time may help raise the level of satisfaction and reduce evasion. 3.28 Other measures may have been counter-productive. During 1999, the provincial legislature of C6rdoba passed a law reducing tax rates together along with a tax amnesty. Property tax rates were reduced by 30 percent; the GRT rate for the car industry and for small and medium taxpayers was also reduced-the betting was that increased economic activity and reduced tax evasion would more than compensate for the reduction in tax rates. In fact, provincial tax revenues fell by 13 percent during 2000 even when federal revenues rose by 3 percent. The province was able to make up for the tax rate reduction with the extra income from the tax amnesty, but this is a short-lived source of revenue. 3.29 Both federal and provincial governments have relied on tax amnesties to raise revenues. Although amnesties increase revenues in the short run, they have the opposite effect on the long run by discouraging compliance as taxpayers await the next amnesty. The province of Buenos Aires even grants a 15 percent discount to those large GRT taxpayers in default who agree to pay in full. The same discount, along with the ability to pay in installments, is extended to small and medium size taxpayers. As GRT taxpayers that pay on time receive no concessions, the incentives not to pay taxes on time are high. 20 Recommended Policy Options 3.30 Improving tax administration is the highest priority. This will also address inequities and inefficiencies arising from the discrimination of those who pay taxes. Stopping the use of tax amnesties and discontinuing the practice of giving discounts to late filers are important first steps. These could be complemented by better training of staff and greater investment in information technology. Better-trained and equipped staff can then direct their attention to gathering information from different sources and crosschecking it. It would allow for a shrewder selection of taxpayers for audit, provide more information to tax examiners, and facilitate identification of non-filers. Better technology can also reduce the scope for corruption and improve taxpayer services through such means as promoting the use of the internet to interact with tax authorities. Also, experience has shown, closer collaboration with AFIP, use of withholding taxes, and more active prosecution of non-compliers can increase revenues. 3.31 Collection efforts could focus first on the less-distortionary taxes, namely real estate and other property taxes. By stepping up efforts to improve the roll, cross-checking information with municipalities and public utility companies, and obtaining more realistic assessment of property values, large increases in revenues should be possible. While provinces could continue to set tax rates, it may be helpful for provinces to delegate responsibility for valuation to the federal government, thereby ensuring uniform methodologies and achieving some economies of scale. 3.32 With respect to tax policy, replacing the gross receipts tax and stamp tax with another broad-based instrument is essential for improving efficiency. However, given that this tax accounts for over half of provincial own-taxes now, eliminating it is unlikely to be feasible until reform of the federal-provincial transfer system. Thus, it is not recommended in the short-term. However, in conjunction with the reform of the coparticipation system discussed later in the chapter, provinces should consider replacing the GRT with one of two candidates: * A dual federal-provincial VAT, with CVAT or VIVAT (see Annex II); however, this would require a reduction in the federal VAT, otherwise the combined federal- provincial VAT rates would be excessive. * Piggy-backing onto the federal income tax, again whilst lowering federal tax rates; one attractive option is for provinces to obtain a certain percentage of each tax slab, except the last one, which should go exclusively to the federal government. 3.33 In the short-term, as a means to improving fiscal correspondence at the margin (more on this later), there may be merit in provinces piggybacking onto the federal personal income tax and for the federal government to agree to lower its personal income taxes as described above. Similarly, the federal government may wish to consider decentralizing the federal wealth tax (bienes personales), as provinces, which already collect property and automobile taxes and have good wealth information on individuals, should have a comparative advantage in collecting the tax. Of course, for the latter two options, a concomitant and commensurate reduction in coparticipation revenues would have to be agreed upon. 21 D. RATIONALIZING THE FISCAL REVENUE SHARING AGREEMENT 3.34 The discussion above highlights the need for reducing transfer dependence in Argentina. Full fiscal decentralization will only come through a much needed-though politically difficult reform-of the coparticipation system. The federal government's recent implementation of the export tax could serve as a warning to provinces: they need to have their own tax instruments over the medium-term. Moreover, given the distortions created by existing provincial tax instruments, new instruments at the provincial level are needed. The goal is to improve the efficiency of the provincial tax system while simultaneously increasing revenues. 3.35 Proposals to reform the revenue-sharing system have been discussed since, at least, 1998, and its most comprehensive version-the establishment of a complete fiscal correspondence-since 2000. The Federal Agreements (Compromisos Federales) of 1999 and 2000 specified a few general guidelines for the reform: simplification of the system and transition rules that would guarantee a nominal minimum for each province in the short-term.14 A variety of excellent and technically well-constructed proposals have been prepared in recent years. However, it has not been possible to build consensus around them because of opposition from provinces that feel that they would lose under the reform.t5 Rather than discuss the details of each of the proposals-since there are points of agreement across proposals-the highlights of some desirable characteristics that are common to several proposals are analyzed below. The reform of the primary distribution is discussed separately from the reform of the secondary distribution, though both are needed and there are advantages to doing them simultaneously. Primary distribution and the degree offiscal correspondence 3.36 The most sweeping proposal for increasing fiscal correspondence is to eliminate the primary distribution, which implies that jurisdictions, as well as the nation, would become responsible for financing their own budgets on aggregate. Eliminating primary distribution will cleanly separate provincial and federal accounts and help improve fiscal discipline as the role of the federal government as a lender of last resort will be greatly diminished. Fiscal correspondence will also increase tax compliance, as both the nation and provinces would have more incentives to collect taxes since they would keep all the revenues they collect. Moreover, if the reform is structured right (more on this in the discussion of the secondary distribution), each province could also keep the marginal revenue it raises. This reform would not eliminate the possibility of fiscal transfers or secondary distribution-it would simply mean that the transfers would be financed by other provinces' own revenues rather than by federal government revenues. Although this reform may prove too difficult to achieve in the medium-term, a simpler reform of the coparticipation system could consider keeping fiscal correspondence at the margin-any additional expenditures would 14 Here reference is made to the clause calling for a new "coparticipation" law. In addition, there was the controversial clause assuring nominal floors for the provincial share; this had to be abandoned at the height of the political and economic crisis in 2001. 15 Most believe that any changes in the current revenue-sharing arrangements would require the approval of all jurisdictions though some believe that a super-majority may be sufficient. 22 need to be financed by increased own revenues. This could be accomplished by such mechanisms as fixing transfers in nominal terms based on history, but allowing increases in revenues to stay with the relevant federal or provincial authority. 3.37 Eliminating the primary distribution requires either (a) transferring to provinces some of the taxes that currently are under the nation's jurisdiction, or (b) dramatically improving own revenue mobilization from existing taxes or one of the new tax instruments discussed in the preceding section, such as the provincial VAT or provincial personal income tax, compensated by a reduction on federal tax rates. A key feature of tax decentralization would be to give provincial authorities the right to set their own tax rates.16 However, for a variety of reasons, it may be more practical to have a unified definition of the base such as would be obtained by piggybacking onto the corresponding federal taxes. 3.38 Even if it proves too difficult politically to eliminate primary distribution, the revenue-sharing system could usefully be simplified by having one general revenue- sharing pool, eliminating the various deductions and earmarks before reaching the pool. Over the long-run, there would be tremendous advantages to eliminating the primary distribution and having equity transfers financed explicitly by the higher revenue provinces as described below. Tax reform for improved provincial tax performance and greater fiscal correspondence is a pre-requisite. Secondary distribution 3.39 Article 75 of the Argentine Constitution establishes the conceptual basis for the secondary distribution: "The distribution between the Nation, the provinces and the city of Buenos Aires and across them, will be made in direct relation to the competencies, services and functions of each one of them contemplating objective criteria of distribution; it will be equitable, solidarity enhancing, and it will give priority to the accomplishment of an equivalent level of development, quality of life and equality of opportunity in all the national territory." 3.40 A variety of criteria can be considered for a simple, transparent formula for distributing transfers across provinces under the guidelines above. The four criteria that have been discussed throughout Argentine history and the academic literature, and that have been practiced by countries, are:17 (a) Equal per capita transfers to provide financing for a basic package of services (in particular with regards to the social sectors) 16 There could be a minimum or maximum imposed, but the key for autonomy and accountability is the setting of rates. 17 See World Bank, 1999 for greater detail on these bullets. 23 (b) Equalization of revenue-raising capacity, through some sort of compensation to jurisdictions with lower per capita ability to generate own tax revenues. The calculation could be made through a variety of measurement techniques'8 (c) Special needs/special costs compensation to provide additional resources for jurisdictions that face unusual demands due to the particular characteristics of their population, or that face higher than average cost of service provision (due to inability to capture economies of scale, or economies of scope, or simply a high local cost of living) (d) Derivation, which returns federally collected revenues to the source of collection, under the understanding that the revenues are provincial revenues that were collected by the central government for cost-efficiency or practical reasons. This approach is the only one that implies no re- distribution. 3.41 While poverty alleviation is not be the main purpose of the coparticipation arrangements-social assistance policies described in Chapter 4 are more appropriate for this-the current system exacerbates regional inequities in contravention of the Constitution. Moreover, simulations show that the fiscal equalization component-even with a provincial VAT or provincial personal income tax-would not consume a large share of existing transfer levels.19 However, it is critical that natural resource royalties should be included when comparing provinces' capacity to raise revenues. Otherwise, the unique benefits derived by a handful of provinces from this substantial revenue source undermine the equity criteria.20 3.42 Simplicity is also key for reform of the secondary distribution. The current complicated formula stems in part from trying to have the secondary distribution meet the many objectives mentioned in Article 75 of the constitution-which it fails to do anyway. Consideration should be given to a very simple compensation formula: the tax-raising capacity of the province divided by its population. Provinces with lower tax capacity per capita will receive larger transfers to compensate them for their more limited ability to raise revenues. This will help assure minimum general service levels while providing incentives for tax effort. While the tax raising capacity is not easy to calculate, it is feasible, in part by analyzing each of the major tax and non-tax (primarily royalties) sources. While provinces should have the main say in determining the new formula, the federal government should also have a voice and could assist by preparing a proposal for the approval of provinces. 18 See Boadway and Hobson (1993) for a complete description of how to calculate this component. 1 See Jefatura de Gabinete (1999) or World Bank (1999). Argafiaraz, Garzon, and Capello (2003) estimate that the compensation necessary for bringing revenues to the national average in below-average jurisdictions would consume 17 percent of federal transfers. 20 For a recent analysis and review of options for natural resource taxation in a federal system, see Ahmad and Mottu, 2002. 24 3.43 Last, but not least, it is crucial to ensure that the efficiency-enhancing changes in the secondary distribution are done gradually to smooth the transition and reduce opposition to the reform. It is important that the revenues received by provinces are protected in the short-term. Having the new distribution formulas apply only to the increases in revenues over the levels of the base year can do this. This would give individual provinces the chance to adjust gradually to the changing circumstances. It would still provide the right incentives at the margin for tax authorities to raise revenues. Although it would delay the benefits from full fiscal correspondence with respect to expenditures, the gradualism is essential to obtain buy-in of the provinces that are currently enjoying more than their fair share of public revenues. E. INSTITUTIONAL REFORMS AND "FISCAL RULES" 3.44 Many Argentine researchers and policy practitioners believe that fiscal rules, such as those requiring governments to meet targets on fiscal deficits and debt, are necessary for fiscal sustainability and to enhance credibility in government policies. They often speak of fiscal rules and reform of the coparticipation arrangements in the same breath.21 A number of coparticipation reform proposals have included the creation of a new federal fiscal entity (organismo fiscal federal) to monitor and enforce fiscal rules.22 3.45 In 1999, the Argentine Congress approved a Fiscal Solvency and Responsibility Law (Ley de Solvencia y Responsibilidad Fiscal) passed a fiscal responsibility law that specified limits on the federal government's deficit and expenditure growth, adoption of a three-year budget, creation of a counter-cyclical fiscal fund, and measures to promote transparency. Although the law did not include rules for subnational governments, it invited provinces to pass similar laws. 3.46 The federal fiscal deficit targets were missed by wide margins during 1999 and 2000. Although the fiscal deficit ceilings were relaxed in 2001 with the issuance of the 2001 Budgetary Law, even the relaxed ceilings were breached in 2001 and 2002. The experience was only slightly better at the provincial level. Fourteen provinces passed fiscal solvency laws. Of the eleven provinces that specified a fiscal deficit ceiling, only five managed to comply. Moreover, even though 16 of the 24 provinces had constitutional limits on debt-service, only ten of them managed to keep within the specified limits. The unfortunate message it has sent investors is that laws can be and are easily broken. Another lesson is that fiscal rules cannot substitute for poor governance or lack of political will to take tough decisions. The province of La Pampa has made impressive fiscal achievements in recent years even without a fiscal solvency law. 3.47 Despite the recent experience with fiscal rules in Argentina, a consensus is emerging that well-designed fiscal rules can indeed help achieve fiscal sustainability and complement the coparticipation reform. The failure of fiscal rules in Argentina is because of a three factors. First, the design was too simple. Specifying an overall fiscal target 21 More details on this approach are presented in World Bank (1999), Frigerio (2003), and CIPPEC (2002) 22 The constutional reform of 1994 called for the creation of a federal fiscal entity. 23 For a summary of the experience of fiscal rules at the national and provincial level, see Braun, 2003. 25 without regard to external shocks (such as occurred with interest rates) or recession meant that the needed adjustments were large, politically costly, and therefore not credible. Also, they were not comprehensive-provinces were not obligated to perform. Second, they were passed via an ordinary law, facilitating its override and failing to generate credibility with creditors. Thirdly, the timing was unfortunate, coming at the beginning of what proved to be a prolonged and deep recession. 3.48 Drawing largely from Brazil's positive experience in this area, it should be possible to formulate a better law. Brazil's law applies to both the federal and provincial governments and has a higher standing than an ordinary law. It was implemented when the federal and state governments had already begun the fiscal adjustments-the law aimed to maintain the momentum rather than spur it. It specified fines-even jail terms-for policy makers and bureaucrats for non-compliance and it contained better-designed rules on provincial indebtedness. 3.49 By having the fiscal rules for both the federal and provincial government embedded within the coparticipation law (ley convenio), almost the entire universe of public expenditures would be covered. Moreover, having it as a ley convenio rather than an ordinary law would make any change difficult, thereby enhancing the credibility of the law-the budget law will not be able to override it. However, at the same time, it means finding the right design and institutions to provide some flexibility in times when strict adherence to targets would be painful and lead to possible breaching of the targets. 3.50 One option at the federal level is for Congress to delegate to a committee of experts-much like the policy-setting committee in Central Banks-the task of setting the annual primary budget balance. 24 This committee would also be responsible for monitoring the execution of the federal budget during the year. In case of slippage, it should have the mandate to propose an adjustment, the details of which (cutting spending or raising taxes) remain in the hands of the government and congress. The members of the committee would be appointed for terms of four or five years on a staggered basis to ensure continuity and institutional memory. Once appointed, they cannot be removed unless they violate the law. The committee would have to report on its decisions, fully providing the details of the analysis that led to them. The larger provinces may also wish to consider creating such committees. 3.51 In designing the fiscal rules, setting targets on the primary balance would be more politically acceptable-and therefore more credible-than setting limits on the overall deficit. Limiting soft borrowing opportunities-via the remaining provincially owned and federally owned banks-as well as from the federal government itself (for the provinces) would be desirable to include in the fiscal rules. While also specifying limits on debt may be desirable, especially for the federal government, at a minimum, approvals for increases in provincial debt should to given by the federal government. 24 Wyplosz (2003) has suggested this approach for Brazil. 26 3.52 In addition, especially at the federal level and possibly for the larger provinces, it would be desirable to institute a multi-year plan that limits the primary balance-and hence, expenditure levels-as a function of expected long-run revenues. Such cyclically adjusted targets are already set in Chile. This would help in avoiding procyclical spending patterns. Publishing the three-year budgets (as opposed to the current practice of preparing them only of internal MECON purposes) would subject them to public scrutiny, thereby ensuring them to be reasonable. A complementary measure, if transfers continue to be an important share of some provinces' revenues in the future, would be to establish a moving- average, fixed nominal amount, or other stabilizing rule to the size of transfers, as opposed to setting the size as a share of actual federal revenues.25 Finally, Argentina should consider introducing individual penalties for non-compliance a la Brazil. 3.53 The timing of Argentina's first attempt at establishing fiscal rules was unfortunate. It came at the beginning of what was to be a three-year recession and when there was low willingness to take difficult decisions and without having cyclically-adjusted targets. Data at the provincial level was of too poor a quality and came with too much of a delay to allow changes in real time. The environment is now more conducive to introducing fiscal rules. The recession appears to have bottomed out and recovery appears to be robust; both the federal and provincial governments have made strong adjustments on their primary balances-the latter within the framework of bilateral agreements between the federal government and almost all the provinces that have fiscal deficits. Controls on subnational borrowing-legally sanctioned in the bilateral agreements-are perhaps the key tool to assuring that deficit targets are met. While provinces still need to do more on the collection and timely reporting of fiscal data, there has been a significant improvement during the last two years, aided again by the bilateral agreements specifying rules on timely reporting. In short, the time is right for introducing well-designed fiscal rules framed within the context of a new ley convenio and also applied to the provinces. 25 See Gonzalez, Rosenblatt, Webb (2002). 27 28 4. IMPROVING THE EFFICIENCY AND EQUITY OF SOCIAL EXPENDITURES 4.1 During the 1990s Argentina's social spending represented roughly 20 percent of GDP and over half of total public spending. Social insurance-pensions, health insurance, unemployment insurance, and family allowances-comprised 62 percent of social expenditures during 2000 and 2001, while education, public health, and social assistance averaged 24, 8, and 6 percent of expenditures respectively (Annex HI). 4.2 This chapter classifies social sectors to include expenditures on education, health, social assistance and social insurance. It thus departs from the traditional definition that also includes expenditures on culture, science and technology, selected spending by the armed forces and security, and infrastructure programs in housing, water, and sewerage. In addition, some programs usually considered as education expenditures are include in this chapter as social assistance expenditures 26. Therefore, the data on social expenditures in Sections B and C of the Statistical Appendix may not be consistent. Methodological details along with an analysis of the evolution of expenditures can be found in Annex III. 4.3 The collapse of the convertibility regime in 2002 substantially changed the composition of social spending in Argentina, basically due to changes in the expenditures of the national government. At the federal level, social assistance expenditures rose by 91 percent in real terms, while education and social insurance expenditures fell by 54 percent and 33 percent respectively. However, the underlying structural characteristics of the various sectors remained unaltered in most cases. As a result, Argentina now faces a series of long-term, structural, policy challenges as well as new ones resulting from the deep social consequences from what is, perhaps, its most serious economic crisis. 4.4 This chapter analyzes the structural features of the social sectors in Argentina, describes the implications of recent changes, and suggests changes in policies and institutions to improve the efficiency and equity of social expenditures. It is divided into five sections: basic education, higher education, pensions, health, and social assistance. These five categories account for the bulk of public expenditures in the social sectors. 26 Typically Scholarships and School Feeding Programs 29 BASIC EDUCATION27 Performance of Provincial Spending 4.5 The financing, administration and provision of basic education in Argentina are now the responsibility of provincial governments. Preschools and primary schools were decentralized during the late 1970s, while the rest of the basic education system was completed in the early 1990s with an ambitious educational reform to improve the relevance, quality, equity, and governance of provincial educational services. The reform aimed to achieve universal coverage for 5 year-olds and reduce the high dropout rates for children aged 12 to 15. The structure of the system was modified from the original seven years of primary school and five years of secondary to nine obligatory years of EGB28 and three years of Polimodal or higher secondary school. The curriculum was revised at all levels and the number of years of obligatory schooling increased from 7 to 10. A decade after the reform was initiated, important progress was made in coverage: between 1994 and 2000 enrollment levels increased-by 19 percent, with nearly universal coverage for children aged 5 and significant improvements in the number of adolescents continuing their studies. 4.6 While the coverage achievements were commendable, serious deficiencies with regard to efficiency, quality and equity remain. Twelve percent of students that start EGB I and II will not complete 6oh grade and 29 percent of students who start 7th grade will not finish the nine years of compulsory schooling. Similarly, 39 percent of the students entering Polimodal level will not graduate. Also, while education spending is relatively progressive at the primary level due to almost universal coverage at that level, there is a sharp bias against children from low income households at the secondary level, especially for those who manage to graduate. Over 75 percent of the children from the highest quintile finish their secondary education while only 24 percent from the lower income quintile do so (Table C.7). At the same time, the educational system has notable weaknesses regarding the quality of student leaming that are manifested most readily in high dropout and repetition rates, as well as in low scores of students on standardized 29 assessments 27 This section is based on Becerra, Espania and Fiszbein (2003). Unless indicated otherwise, the data referred to in this section are drawn from that paper. 28 In turn, the nine years of EGB are divided into three cycles, EGB I and II representing grades 1-6 of primary school and EGB HII representing years 7-9 of middle school. 9 Results from standardized assessments corroborate the existence of a problem regarding educational quality. On the recently released Progress on International Reading Literacy Survey (PIRLS) 2002, Argentine students lagged considerably behind their OECD counterparts. Other analyses show that Argentine students perform relatively well in the LAC region, yet trail behind those from other developed countries. Students tend to do better in areas of more rote learning and worse in analysis and interpretation, read poorly, and score consistently higher in large cities and urban areas than those in rural areas. These findings highlight that improvement of the quality of education is still an unfinished agenda in Argentina. Concerns about quality have deepened as a result of the economic crisis. In some provinces school-days lost represent between 30 to 45% of the regular school year in 2002. This loss of school-days is likely to have an adverse impact on the quality of education provided, and may generate an increase in repetition and/or drop out rates. 30 4.7 The increase in coverage has been costly. During the 1990s consolidated expenditures in education increased by 77 percent in real terms and by 40 percent in terms of its share of GDP. However, the increase in education spending differed greatly among provinces. While the Province of Buenos Aires, which represents 35 percent of total provincial expenditures in education, increased its expenditure levels by 80 percent in real terms, the other provinces did so by 25 percent on average (Table C.2). 4.8 Both the level and composition of basic education within different levels of government have seen experienced important shifts (Table 3).30 With the decentralization of EGB II and mII, national expenditures dropped by $1.1 billion between 1991 and 1993, while provincial expenditures increased by $2.5 billion as provinces took up on added responsibilities and gave hefty teacher salary increases The other major shift in expenditures took place in the late 1990s with the introduction of the National Teacher Incentive Fund (FONID)-an initiative undertaken as an emergency response to strident demands for increases in teacher salaries, in a context of expansive overall government spending. Even after the fiscal situation deteriorated, FONID continued to limp through as an entitlement program. FONID continues to represented a little over $600 million pesos per year starting in 1999.31 It has thus diverted scarce financial resources from programs to improve the quality and pertinence of education and led to a defacto alteration in the division of responsibilities between the Nation and the Provinces established by the Federal Education Law of 1993. Table 3: Basic Educational Expenditures by Level of Government, Selected Years (Millions of 2001 pesos) 1991 1993 1995 1997 1999 2001 2002 National Level 1,208 116 206 394 240 107 181 Provincial Level 4,026 6,500 6,581 7,229 8,641 9,000 5,366 Municipal Level 160 202 284 358 409 438 271 Total 5,394 6,818 7,070 7,981 9,290 9,545 5,818 Source: Table Cl 4.9 With the economic crisis of 2002, consolidated expenditures in education experienced the biggest single-year drop in history, diminishing by 38 percent in real terms and 28 percent as a share of GDP (Table 3) This reduction has been similar at the national and provincial levels and for all educational levels. The reduction in real spending was the result of salaries not being adjusted in the face of inflation and, thus, does not necessarily 30 The column "National" corresponds to spending executed by the federal ministry, excluding transfers to provinces but including social assistance spending executed by the ministry-provincial transfers show up under provincial spending. 31 The actual payment of the FONID was rather erratic given the fiscal constrains faced by the National Government. For example, half of the 2001 payment was actually made in 2002. 31 represent a reduction in the quantity of inputs available to schools.32 At the same time, the deepening of the crisis in the year 2002 forced the government to redirect resources to support school lunches and increase the number of scholarships. As a result, the National Ministry of Education assumed a larger role in the financing of assistance programs, sacrificing the financing of other components that were in keeping with the federal government's quality enhancing role. 4.10 Financing by the provinces is made directly to public schools that provide free education and by means of subsidies to private schools to partially or totally cover the payment of teachers' salaries. The size of the private sector is significant and has been growing in the last decade-a period when private enrollments increased from 21 percent to 25 percent of total enrollments. Public subsidies to private schools represent 13 percent of total public expenditures in education. However, as the allocations are determined by supply side variables (such as number of teachers) rather than by demand side variables (number of students) or by equity concerns (such as the socio-economic condition of students), there is little incentive to improve quality or efficiency or target poorer students. 4.11 Teacher salaries account for 91 percent of provincial education expenditures (Table C.3), though they vary greatly between provinces. Basically, teacher salaries are determined by: (i) years of teaching experience, (ii) bonuses for specific duties such as accepting a position in a remote or unfavorable location, and (iii) promotions within the education hierarchy. In addition, teachers' legal norms are extremely flexible in terms of leaves, much more than in any other labor market in Argentina. Most of the provinces allocate almost null amounts to quality improvement activities or to improve school infrastructure with their own resources. Inefficiencies in Provincial Spending 4.12 The indicative findings on inefficiencies in basic education discussed above are also supported by analysis of data at the aggregate provincial level. While richer provinces are found to spend more on education (Figure C.1), large differences exist within groups of relatively homogenous provinces. For example, although income levels in Salta, Misiones, Corrientes, Tucuman, Jujuy and Chaco are similar (between 3,000 and 3,500 pesos per capita), the levels of spending per student vary between $667 in Salta to $1015 in Chaco. The same is true for the higher income provinces.34 Yet, as Figure C.2 shows, there seems to be no relation between the resources allocated to the system and the results in terms of quality: higher levels of spending per student do not imply better results in academic evaluations. Thus, for example, a province like Salta spends less per student than provinces with a similar level of GDP and obtains better test results. 32 For a detailed analysis of the impact of the 2002 economic crisis on the education sector see Parandekar, Espana and Savanti (2003). 33 For a detailed analysis of teacher salaries see Vegas et. al (1999). 34 The figures exclude the City of Buenos Aires whose level of GDP puts it well outside the range covered by the graphs. 32 4.13 This lack of association between per capita spending levels and results constitutes the greatest manifestation of the inefficiency of provincial spending in education-many provinces could be doing more and better from an educational point of view with the same resources. 4.14 By digging deeper, one finds that the number of teachers is an important determinant of inefficiencies. When the above exercise was repeated using the student/teacher ratio instead of spending/student ratio (which largely reflects the wage bill for teachers), low student/teacher ratios bore little relation to better results in academic evaluations for similar provinces. Thus while differences in average salaries between provinces may contribute to explaining variations in per capita spending, it is the difference in numbers of teachers that constitutes the key factor. 4.15 The way in which teachers are assigned to schools also contributes to inefficiencies the system. While the basic ratios characterizing teacher deployment (students/teachers, students/sections, and teachers/sections) vary greatly among provinces, close analysis of such ratios shows that higher levels of human resources are not necessarily associated with greater needs (Tables C.4-C.6). For example, low student/teacher ratios are neither associated with a larger time load, nor with population density (that is, rural areas do not show particularly lower ratios of students per teacher). Furthermore, there are high standard deviations among homogenous groups of schools within a province.35 The implication is that differences in per-student costs are closely associated with the efficiency with which human resources are allocated and utilized within the schooling system in each province. 4.16 The teaching positions found in each school make up the Planta Orgdnica Funcional (POF) is supposed to be determined by detailed provincial norms. Such norms regulate five main organizational features: (i) the number of sections for each grade, depending on the number of students; (ii) curricular components such as the number of class hours corresponding to every study plan; (iii) the structure of school management (principals, vice principals, etc); (iv) the number and composition of teachers' support staff (secretaries, monitors, etc); and (v) the number of operational staff (janitors, cooks, etc.). 4.17 Despite the guidelines, there are large variations in the POFs of schools within a province. This is the source of the inefficiencies mentioned above. In most provinces, the number of positions at any particular point in time tends to be determined more by inertia than by the existing rules regarding assignment of staff. Except for extraordinary cases, the previous year's POF is usually reproduced without much verification and this is why the relation of student to teachers tends to vary so much among provinces and among schools. The number of teaching positions is usually maintained relatively constant over the years regardless of the trends in enrollment levels. When enrollments fall in a particular school, sections tend not to be closed and teachers not to be reassigned. When enrollments increase, the tendency is to hire new teachers rather than reassigning teachers from within the system. This is typically done by hiring temporary or substitute teachers 35 See Becerra, Espafia and Fiszbein (2003, forthcoming) for the case of Catamarca. 33 for whom the POF norms does not apply. Over time, such behavior raises costs substantially and is the root of the inefficiencies discussed above. 4.18 There are many explanations for why this happens. Decisions to reassign (or fire) personnel are hard, sometimes limited by legal norms, and always politically costly. But the weaknesses of management and control mechanisms within education systems contribute to this pattern. Control mechanisms entail two management functions: pedagogical and administrative. The first one verifies enrollment and class sections and the staff positions to be assigned as a result of these two factors. The administrative function usually focuses on the budgetary resources for the positions. The main problem in most provinces is that there is a great disconnect between these two functions. The pedagogical supervision usually ignores budget availability and the administrative function ignores the enrollment levels in each school. The consequence of this is a weakening of the control system where each area blames the other for the deviations. Recommendations 4.19 Recommendations in this section are made under the assumptions that no important changes would occur in the fiscal relations between the national government and the provinces. Naturally, if such changes were to take place, the range of possible reforms oriented to improve the efficiency and equity of education spending would be significantly different depending on the exact nature of the changes. 4.20 In order to eliminate such inefficiencies in expenditures, it is imperative to improve management and control at the provincial level. This is especially crucial in the current context of increasing demands on the social sectors and scarce fiscal resources. While the required measures will not be easy, the above analysis suggests that a revamping and modernization of the management mechanisms could substantially improve service delivery even if expenses in constant terms do not grow from their reduced 2002 levels. 4.21 A modem management system would include, among other tasks, the establishment of electronic records that match teachers with teaching positions (legajo unico), so that resources are not duplicated and wasted due to administrative laxity. Combining a functioning and up-to-date information system on the teacher work force with a similar information system about student enrollment would be the first step towards ensuring that wage spending for each school is in accordance with the established administrative and pedagogical norms regarding the deployment of teachers and administrative staff. A stricter enforcement of current regulations would make it possible for the existing teacher workforce to be utilized to meet projected increases in enrollment. However, this objective would require a close and constant attention to the actual deployment of human resources across schools, with the information system fundamentally linked to the decision making process of assigning teaching positions and naming teachers to fill those positions. While establishing a modern functioning management system is a necessary condition for controlling inefficiencies, experience indicates that putting them in practice requires strong and sustained political support from the provincial leadership (World Bank, 2001). 34 4.22 Some provinces have made significant advances in their efficiency indicators by adopting modem management systems. For example, between 1996 and 2001, Salta has managed to increase enrolments by 16.5 percent while increasing spending by only 5.6 percent, even as the rest of the country increased spending by 40 percent to accommodate 12.7 percent additional students. This was due to: (i) a more efficient allocation of the teaching positions (POF) by reassigning teachers to cover the growing enrollment; (ii) a reduction in the number of teachers on leave (licencias) by establishing a system of medical checking; and (iii) a simplification and de-bureaucratization in the management of the system. 4.23 While provincial governments are responsible for the financing and management of their education system, the federal government could help by providing incentives for enhanced efficiency. A practical way to do this by tying federal aid to improvements in educational efficiencies. Thus, the limited federal resources for teacher training, didactic materials and infrastructure and maintenance could be utilized to support provincial measures to improve the efficient use of their own educational budgets and to improve education quality. Just as federal transfers could provide a positive benefit to reforming provinces, federal penalties could also be imposed on provinces that systematically fail to' meet adequate minimum service delivery standards, including as a last measure, the recentralization of basic education. 4.24 Improving the quality of education will require an active role on the part of the federal government encouraging and providing incentives to provinces. One way to do that is to link national education aid to provincial performance (adjusted for socioeconomic status of students) and evolution of such indicators. The very large differences in test results among similar schools, provide extremely valuable information that, if adequately analyzed and used, could provide the basis for actions oriented to improving quality indicators. 4.25 If provinces implement the above recommendations they should obtain strong efficiency gains, allowing both the quality and coverage of education to improve without requiring additional public funding. However, some provinces may face strong opposition from vested interest groups that could succeed in thwarting government efforts. The current system relies on enforcement of school-specific criteria by provincial authorities. The main beneficiaries-students and their parents or even local authorities-have little role. In order to better align incentives, it would be useful to implement institutional reforms that provide greater autonomy to schools and make greater use of parents associations and school boards to manage schools. Overall, these potential changes would enhance the role of the provincial governments as regulators and providers of funds to schools based on transparent demand criteria such as number of students and population density. In fact, governments should consider implementing such demand-based allocation criteria immediately for private schools that receive public subsidies. Some province (such as Buenos Aires) have already begun moves to provide greater autonomy to its public schools. These experiments need close attention and support given their potential as means to improve accountability within the education system. 35 UNIVERSITIES36 Performance of University Spending 4.26 The university system in Argentina provides education to approximately 1.3 million students, distributed among 41 public universities (87 percent of enrollments) and 50 private universities (13 percent of enrollments). The public system is characterized by two salient features: gratuity and free entry. Public universities are autonomous entities, fully funded by the National government and with no quantitative (quotas) or qualitative (entrance exams) mechanisms to limit the entry of students.37 4.27 After stagnating during the 1980s, public spending on universities grew by 72 percent in real terms during the 1990s and fluctuated between 16 and 20 percent of total consolidated expenditures in education (Table Cl). Nonetheless, there was no increase in spending per student because enrollments increased by 75 percent. Because of fiscal pressures, real university expenditures during 2000 and 2001 fell slightly, before tumbling by 35 percent in 2002, due to the combination of a constant nominal budget and accelerating inflation. As with basic education, though, this real reduction simply reflected the reduction that took place in real wages during the period. Salaries comprise about three-quarters of university expenditures. About 10 percent is spent on research incentives program, scholarships, and other subsidies to students and teachers, while approximately 4 percent goes to capital expenditures. (Table C.8). 4.28 During the 1990s, public universities expanded their own revenues significantly through the sale of consulting services and fees charged for graduate programs. As a result, in 2001-the latest year for which this information is available-resources from the Treasury, which had represented 99 percent of total spending by the universities, fell to 85 percent (see Table C.9). The bulk of Treasury funding to universities is based on historical trends rather than any kind of objective indicator such as the number of graduates.38 Each year the National Congress decides on the amount of resources to be transferred to each institution based on the previous year's transfer and on the result of political negotiations highly determined by the negotiating strength of each institution. This system makes it impossible to introduce incentives that improve the quality and efficiency of the universities or compensate for regional disparities. 4.29 Private universities receive no public funds and charge tuition fees to their students. Between 1985 and 2000 the number of private universities grew from 20 to 50, while their enrollment grew from about 70,000 to about 166.000. While private enrollment represent 36 This section is based on Becerra, Cetrangolo, Curcio and Jimenez (2003, forthcoming). Unless specified all data comes from that paper. 37 There are some exceptions such as the University of Cordoba that charges an optional tuition fee, the University of Tres de Febrero that has begun to charge tuition fees, and the universities of Buenos Aires and La Plata, which have implemented quotas for the number of students entering Medical School. 38 From 1995 to 2003 there has been a some programs that allocate funds to universities based on objective criteria: AR (Model for allocation of Resources); FOMEC (a Competitive Fund to Quality Improvement); Program of Incentives to Professors - Researchers; Infrastructure Financing Program (PROIN); Scholarships Program and others. However these programs never surpassed 7% of the total public transfers to universities. 36 about 13 percent of total university enrollments in the system, they account for about 27 percent of university graduates. Private universities tend to concentrate in the most developed regions of the country and cater to higher income groups. Inefficiency and Inequity in the Public University System 4.30 The combination of a system with no entry barriers, no access or tuition fees and little external control due to the political and management autonomy that universities possess under the current legal framework leads to a highly inefficient and unequal system. Spending on public universities is highly regressive. About 60 percent of the students in public universities come from families in the top 40 percent of the income distribution, while less than 6 percent belong to the bottom quintile (Table 4). This bias is even larger with respect to students who graduate-low-income students are more prone to dropping out. Table 4: Distribution of University Students Across Income Quintiles (Percent) Income Quintiles 1 2 3 4 5 Distribution of individuals 18-30 years of age by income quintiles 23.0 22.0 19.5 19.8 16.0 Distribution of all university students by quintile 5.3 13.2 18.1 29.5 32.8 Distribution of public university students by quintile 5.9 14.6 20.2 30.7 28.5 Households in the income group with a member that has graduated from a university. 0.6 1.6 3.0 9.0 26.0 Source: Secretariat of University Policies, based on data from the permanent household survey, INDEC 1999 4.31 Although enrollment rates shot up during the 1990s, the number of graduates has shown little increase. This is because students take very long on average to complete their studies and exhibit extremely high dropout rates, features of the system that represent important efficiency concerns. Graduation rates dropped from approximately 60 percent in the early 1980s to 18 percent currently. According to a study made by the Secretariat for University Policies (SPU), only 11.3 percent of the students that entered the university in 199039 had graduated by the 1999, taking an average time of 7.7 years to complete their studies. An additional 2.6 percent are estimated to graduate over the next 10 years. This means that, independently of the time that students take to complete their studies, more than 85 percent of the students that enter a higher education institution do not manage to finish their degree. Argentina graduates approximately 45,000 from its public and private universities per year, while Chile graduates almost 50,000, with much fewer students, and 39 The study considered students in five areas (Accounting, Agronomy, Law, Biochemistry, and Communication) from all national universities except the Universidad de Buenos Aires and the Universidad de La Rioja. 37 Colombia 75,000, with half the number of students. Thus Colombia has roughly four times higher efficiency in terms of output of graduates (Table C. 10). 4.32 Of the students enrolled in the system, 26 percent are entering students (new students) and 74 percent are students who have been in the system for at least one year. Of the latter group, 28 percent had not completed even one class in the previous year. In some universities this group represents as high as 40 percent of the students. When only 'active students' (i.e. those completing at least two courses per year) are considered spending per student increases by 40 percent in the system as a whole, and as much as 70 percent in certain institutions where the percentage of active students is extremely low. According to the study by SPU, the number of students repeating classes in 1999 generated an increase in the cost per graduate of 38 percent. Dropout rates have an even higher impact on costs, implying an increase of 136 percent in the cost of producing a university student. On average, the repetition and dropout rates in 1999 (the year for which this study was conducted) implied a per graduate cost three times as high as that corresponding to a situation with no repetition or dropouts. 4.33 In summary, Argentina spends substantial public resources on a university system in which most students fail to graduate, or take almost double the normal time to complete their studies. This results in very high costs per graduate. Despite university education being largely free, the beneficiaries tend to be the upper income groups in society. In short, university expenditures are extremely inefficient and highly regressive. Reform Proposals and Recommendations 4.34 A well functioning university system is critical for Argentina's long-term economic development. Without an adequate supply of high quality professionals the country will have difficulties competing in the world economy. In addition, the strong relationship between higher education and earnings make university policy a powerful long term instrument to improve income distribution. The challenge is to do this in an environment of fiscal constraints. This will require progress in three key areas: (i) efficiency, (ii) alternative financing mechanisms, and (iii) equality of opportunities. 4.35 To improve efficiency, the number of years to graduate a student will need to fall substantially, thereby reducing the total cost per graduate from its current high level. This will require: * enforcing regulations that specify conditions by which one can remain a student * setting limits to entry via: (i) performance at tests when leaving secondary school40 and (ii) limiting the number of positions available. * providing incentives to universities by increasing their budgets when they perform well on criteria such as the share of students graduating or the average tenure of a student in the system 40 However, this policy should be implemented carefully taking into account the differences that exist between requirements for secondary and higher education. 38 * introducing or expanding the use of fees-initially, at least, one could start modestly by introducing charges for repeating exams or for remaining in the university beyond a minimum number of years 4.36 To improve equity, financial support, such as via scholarships, will need to be provided to low income students to allow them to complete their studies in the normal number of years. To accomplish this within the constrained fiscal environment while simultaneously reducing the regressive nature of university expenditures, fees could be introduced or expanded, coupled with means-tested scholarships and, potentially, student loans. Some proposals for introducing fees have suggested introducing 'graduate taxes' and other forms of indirect and selective cost recovery mechanisms. As mentioned a few universities have already started to introduce fees in their undergraduate programs and tuitions are the norm for graduate programs. 4.37 The current legal framework facing individual public universities allows universities to introduce the efficiency and equity enhancing measures mentioned above. However, the combination of political autonomy, financial autarky, and election of representatives for the university councils creates a complex institutional setting for self- regulation and change. For the larger universities especially, achieving consensus for implementing the above reforms will be difficult. 4.38 One attempt at building consensus for higher education reform was to create a "National Commission for the Improvement of Higher Education" in 2001 (known as the "Juri Commission"). The Commission, which comprised representatives from the university community and experts from various sectors, did not focus on the controversial issues of financing and efficiency enhancement. Rather it proposed the creation of new two-year programs and university colleges under the responsibility of provinces and/or municipalities. It also proposed a system of credits to facilitate greater mobility of students and greater opportunities for those living in remote areas. These reforms have potential benefits on both efficiency and equity. In addition, by reducing the number of students at the universities, it will allow universities to focus on responding to heterogeneous student needs and preferences and to better targeting of public resources-which should result in improved graduation rates. Developing the institutional and financial aspects of this proposal is very desirable. 4.39 Competition from private universities may well be the spur for major reforms of the kind discussed above. Private universities compete with public ones not only for students but also for teaching staff, who are attracted by better remuneration and working conditions in private universities. To the extent that this leads to a loss of prestige and reduced demand for tuition-paying graduate programs in public universities, conditions may be ripe for introducing the policy and institutional reforms discussed abo 39 HEALTH" Structure and Performance 4.40 Before the end of convertibility, Argentina was spending over 8 percent of GDP on health services-more than $650 per capita/year. Two-fifths of this was private spending, with the rest divided between the public and social insurance sectors. According to the World Health Organization (WHO, 2000) only 17 countries in the world spend more on health as a percentage of GDP than Argentina, and only countries in the OECD and a couple of oil-rich middle eastern countries spend more per capita than Argentina. Despite the high expenditures, the performance of Argentina's health system, expressed through health status indicators such as infant and maternal mortality, falls short of that of other comparable countries with lower expenditure levels. 4.41 The health sector in Argentina is characterized by a segmented structure of financing and service provision within a complex federal structure, where health services are ultimately the responsibility of provinces. Health insurance is mainly done through an extensive net of social insurance arrangements including obras sociales nacionales (OSN), covering formal workers of the different productive sectors, 23 independent obras Table 5: Coverage Levels by Sub-sector42 sociales provinciales (OSP), that insure the Individuals Percent workers of the provincial governments, (millions) PAMI, which is the insurance institute for National Obras Sociales 10.8 30 retirees and pensioners, and private health Provincial Obras Sociales 5.1 14 insurers, prepagas, covering mostly high PAMI 3.2 9 income people on a voluntary basis. Those Private Prepagas 3.3 9 Public coverage 13.8 38 without insurance coverage rely on the Total Beneficiaries 36.2 100 (mostly free) health services offered by the neostwo ofree) publc hospvits anfered heathe Source: World Bank (2003 forthcomning), based on information network of public hospitals and health provided by SSS, and the National Ministry of Health. centers. About 40 percent of the population does not have explicit insurance coverage (Table 5), though about 57 percent depended exclusively on public hospitals for health care delivery in 2001. 4.42 The delivery of medical services is done both through public and private providers. A strong and growing number of private providers (doctors, hospitals and clinics, labs, etc.) have become the key deliverers of health services to those with insurance coverage. The percentage of private health facilities has grown from 35 percent in 1969 to 55 percent of the total in 1995, and the percentage of beds from 22 percent to 43 percent during the same period. Due to the process of decentralization that took place during the 1990s, provinces control over 60 percent of the public hospital network with the rest mostly under the control of municipalities. 41 This section considers spending both by the public system and the social health insurance systems. A detailed analysis is provided in World Bank (2003, forthcoming) from which all data is drawn unless SP ecifically indicated. 4The numbers in this table correspond to the administrative records of insurers. Household surveys (e.g. the 2001 ECDV, indicate a smaller proportion of the population as having insurance coverage. 40 4.43 The financing structure of health services in Argentina is also rather complex. About one-fourth of all health spending is financed directly from general revenues at the national, provincial, and municipal levels, while just over one-third is covered by health insurers, financed, in principle, through worker and employer contributions but, as seen below, which often receive bailouts from the government (Table 6). Despite the health insurance system and free public services, households pay for a whopping 42 percent of health costs, allocating on average an additional 7 percent of their income for out-of-pocket health expenditures. This suggests that free public health services are either of poor quality, rationed, do not cover certain expenses (most notably medications), and/or, include significant "informal" copayments. 4.44 The institutional fragmentation with poor articulation of the health Table 6: Expenditures in Health by Agent, 2000 - sector, and the internal arrangements within each sub-system, generate US$ Percent inequities, inefficiencies, and millions governance problems. The dispersion of Public Sector 5,560 23 population among too many small National Health Ministry 805 2 insurers reduces the size of risk pools, Provincial Health Ministries 3,913 12 insurer reduces the size of risk poois, Municipal Health Secretariats 842 3 increases administrative costs, and tends to generate persistent deficits that, often, Social Security 8,243 34 require fiscal bailouts. The OSN 3,861 16 fragmentation makes it extremely PAMI 2,270 9 difficult to develop stable redistributive OSP 2,112 9 mechanisms to ensure equity in Households 10,390 42 financing and creates opportunities for Total 24,193 100 non-transparent (and regressive) cross- Source: PIA-SALUD, El gasto en Salud en Argentina, Tobar F. 2002 subsidies between sub-systems-for example when public hospitals do not recover costs of services from health insurers. 4.45 The health system is also characterized by a weak and incomplete regulatory framework, with overlapping responsibility and unclear jurisdiction between national and provincial governments. There is no framework under which basic service guarantees can be enforced throughout the country and for all citizens. Neither the Federal Health Council (COFESA) nor the National Health Ministry has been able to get provinces to agree upon common health targets. Neither is there agreement on minimum service or quality standards for health providers or the so-called "health networks" that act as intermediaries between payers and providers at the national and provincial levels. The Health Superintendence (SSS) has limited regulatory power over providers to the social health insurance system, and each province establishes its own framework, which sometimes entails a delegation of key functions to professional associations. The SSS does not regulate PAMI, the largest single health purchaser of health services in the country. 4.46 Overlaps and gaps can be found in the regulation of pharmaceuticals as well. De facto, there exist today 16 laboratories run by provinces, 21 by municipalities and others run directly by hospitals, universities and the armed forces, most of which do not respond 41 to the guidelines and norms of ANMAT, the national regulatory agency for pharmaceuticals, food and medical technology. 4.47 There is general consensus that the public health system is badly managed in most provinces, suffering from serious problems of allocative and technical efficiency, efficacy and lack of transparency. There is low correlation between the amount of resources provinces spend on health and the demographic and health characteristics of the provinces (such as its population, percentage of population with unsatisfied basic needs, infant mortality rates and epidemiological profile) suggesting the lack of a spending pattern that responds to the needs of the population.43 For example, per capita spending in health is four times higher in the more advanced provinces than in the poorest ones, in spite of the fact that poorer provinces present infant mortality rates three times as high. 4.48 In fact, the distribution of spending across jurisdictions is strongly associated with the characteristics of the supply (number of doctors, number of medical centers and number of available beds per province) rather than the demand for health services,44 a pattern typical of systems characterized by historical line item public health budgets. Furthermore, installed public sector capacity (e.g. hospital beds and doctors) is unequally distributed in the territory with a relatively high concentration in richer jurisdictions where the supply of private practitioners and facilities is also the largest. 4.49 Argentina, as most countries in the region, has concentrated most of its resources on individual medical health care, especially for high complexity procedures (with lower cost-effectiveness ratios), to the detriment of primary assistance in health and in public health. During 2001, consolidated expenditures on public health programs were just 2.4 percent of total health expenditures while hospitals and health centers received 82 percent (Table 1.5). Although numbers vary between provinces, in general no more than 10 percent of the total health spending in health is dedicated to the primary health. Additionally, this spending is usually "tied" to specific programs with little innovation in their internal management or their distribution. The greater fraction of public spending on health is used for covering the labor costs of higher complexity hospitals, independently of output levels and quality of the services provided. Thus, the system is excessively biased towards specialized and hospital-based care with insufficient attention to health promotion and disease control and ambulatory care. Family medicine and home care is recent and still under-developed. 4.50 Management of public hospitals tends to be weak and characterized by: (i) lack of performance incentives, with budgets being allocated on a historical basis rather than related to output levels and quality of services; (ii) minimal autonomy, governance and reporting; and (iii) poor management tools, partly because of inadequate information on the costs of production. 43 Maceira et al (2002). 44 Correlation coefficients between level of spending and number of doctors is over 90 percent, and the same happens with available beds. For other supply indicators, correlation levels are over 70 percent. 42 4.51 A number of factors also conspire to maintain large regressive cross-subsidies to health insurers. Between 20 percent and 40 percent of those receiving treatments at public facilities carry health insurance but only a fraction of the costs of providing such services is recovered. For example, in 2002 only $2 million per month were recovered by public hospitals while if procedures and norms were applied normally, cost recovery would have been roughly $70 million per month. Only a small fraction of public hospitals make cost recovery efforts and these tend to recover only a very small fraction of their costs. Factors that contribute to these poor results in cost recovery are: (i) the difficulties in identifying beneficiaries given, until 2003, the lack of an updated and consolidated registry; (ii) the high complexity of the cost recovery process; (iii) inconsistencies in the reimbursement rates set up nationally; and (iv) low management capacity in public hospitals that, often, lack incentives for cost recovery.45 Social Health Insurance 4.52 Argentina's large and diverse social health insurance system-comprising obras sociales administered by labor unions, provincial governments and the agency PAMI- suffers from three critical problems: inequity, inefficiency, and weak governance. 4.53 One of the ways the inequity problem manifests itself is in coverage. The system's model of organization, based on formal employment, leaves too many working families out of the system in times of economic growth, let alone in times of economic crisis and. high unemployment. Poor enforcement, lack of innovative and flexible arrangements for including the informal economy, and lack of targeted subsidies for the poor results in just over half the population being covered-a proportion that has been declining and skewed in favor of the better-off. 4.54 Another way the inequity problem manifests itself is in spending by health insurers, which tends to be regressive. For instance, services received by the poorest 20 percent of the population accounted for only 9 percent of total spending by OSN, while the richest 20 percent received the equivalent of 33 percent of total spending by OSN (Table 7). This is because revenues are a function of salaries and there is no effective financial redistribution mechanism. The one redistribution mechanism that existed-the Solidarity Redistribution Fund (FSR) within the OSN-used to guarantee a minimum of $40 monthly per affiliated worker. However, as average family size tends to be lower for richer obras, the actual per- beneficiary guarantee was much lower. This has been recently changed (2003): now, the FSR will guarantee $20 per member and $15 per additional beneficiary, increasing the share of spending going to lower income workers. However, the FSR still does not cover workers with a monthly income below $240 pesos (approximately 4 percent of the beneficiaries). These lower income workers do not fully benefit from the mandatory health benefits package (PMO) nor have automatic family coverage. 45 The absence of effective purchasers at the provincial level reduces incentives for public hospitals to recover costs from insurers. 43 4.55 The second serious problem Table 7: OSN Distribution of Spending by with the social health insurance Income Quintile (1996/97) system relates to inefficiencies stemming from its institutional 35 - arrangements and incentives. The 30 system is fragmented into various 25 subsystems and hundreds of insuring agents, many of them still functioning ,. 20 as quasi-monopolies with captive 15- beneficiaries. Poorly identified _ beneficiaries are often dispersed in @ 10 hundreds of small risk pools, with 5i i high levels of intermediation (and o administrative costs) and persistent 1 2 3 4 5 financial imbalances resulting in Income Quintiles chronic deficits and requiring Source: Ministry of Economy, based on ENGH 96/97 expensive periodic bailouts as described above. Structural reforms to improve the efficiency of mechanisms such as the 'Administraci6n de Programas Especiales' (APE), which reimburses insurers for high complexity pathologies in the national health insurance system, need to further advance; this includes reforming the menu of eligible services, the reimbursement mechanism, and its coverage. Many national and provincial insurers still perform weakly in terms of their purchase functions and are often captured by providers. 4.56 The third serious set of deficiencies of the social health insurance system relates to governance and regulatory matters. Regulations are weakly applied or absent. There is no minimum homogeneity across the various subsystems in the country, with significant differences in content, costs, and quality of services provided across provinces and even across agents of the national health insurance system. Performance information (with financial and service delivery indicators) is rarely made public and accusations on lack of transparency and corruption in financial management and service contracting remain frequent. Reforming Health Institutions for-Efficiency and Equity 4.57 Reducing the efficiency and equity problems in the health system will require changes in institutional structures and incentives for improved governance and performance. Given the federal nature of the sector and entrenched vested interested, serious reforms will be challenging to implement. Past attempts have proven to be divisive and controversial. However, given the large gains that can be obtained, both the provincial and federal government urgently need to begin the reform process. 4.58 Strengthening the public health delivery system will require: a) Creating incentives to improve the quality and efficiency of service delivery b) Eliminating cross-subsidies from public health facilities to health insurers c) Reallocating resources toward primary health care. 44 4.59 Creating incentives to enhance the quality and efficiency of public hospitals, which account for the bulk of public sector expenditures, will require granting them greater autonomy and holding them accountable to results. Introducing performance contracts linking budgets to objective processes and results could help accomplish this. The budget allocations should gradually be reoriented towards the needier populations and be based on transparent demand-based criteria, while the results should stress quality indicators that are easy to monitor. Overall, reducing inefficiencies will require a clearer separation of the functions of regulation, purchase and provision of health services at the provincial level, possibly concentrating Health Ministries' role in the first two functions. In particular, the provincial purchasing function must be strengthened, for example, through the development of provincial insurance initiatives for the poor uninsured. 4.60 Eliminating regressive cross-subsidies from public hospitals to health insurers requires, among other actions, facilitating public hospitals access to consolidated and updated social health insurer's beneficiary databases of the OSNs, PAMI and OSPs. The national government needs to mandate this and to provide technical assistance towards this goal, if necessary. It should also provide technical assistance towards a simplification and enhanced use of billing processes. Increased incomes resulting from collections from health insurers could be directed to increased support for primary health and incentive results-based arrangements. 4.61 Efforts to improve the equity and efficiency of the health insurance system need to focus on better redistribution mechanisms and better regulation of health insurance agents. The 'Fondo Solidario de Redistribuci6n' (FSR) needs to further enhance its capacity to redistribute funds equitably, transparently and with adjustments for risk within the national health insurance system and to cover all low-income formal workers. The development of a national reinsurance scheme for catastrophic events (replacing APE) could complement the mandatory health benefits package (PMO), and help strengthen the financial situation of the insurance sector. While the design of such a reinsurance system is a complex task, key characteristics should include an independent and specialized administration; a sufficiently large risk-pool to ensure sustainability; financing through contributions from all participating insurance funds; and risk-sharing with participating insurers. 4.62 Enormous improvements in the social health insurance system could be made simply by enforcing existing regulations relating to the operation of insurer agents in legal, financial, and service delivery dimensions and by a timely and thorough application of the crisis procedure for non-compliant insurers. If ths effort was complemented by a determination not to provide financial subsidies to insurers, it most likely would lead to the closure of dozens of insurers and encourage the restructuring, merging, and consolidation of remaining insurers, still numerous. Larger pool risks would result, with a consequent reduction in administrative and transaction costs. Moreover, monitoring, evaluation, and control would be facilitated, and overall responsiveness to citizens should improve. 4.63 The reform efforts on social health insurance will require transforming the health regulatory SSS into an entity with greater control capacity and representation in the sector. To allow it to ensure compliance with sector regulations and protecting beneficiary rights, the SSS must not suffer from any conflicts of interest. Thus it should not administer 45 resources for the entities it regulates nor perform direct or indirect purchases for them. Given the externalities the regulation of national social health insurers has on other actors in the health system-particularly on provincial health systems-increasing the accountability of the SSS may require changes in its current organization and functioning to ensure that it is a valid and representative entity for all actors, including the provinces, recognized for its technical capacity and independent from specific interest groups.46 4.64 Regulating PAMI, which is notoriously inefficient, the subject of numerous accusations of lack of transparency, and has received US$1,700 million in fiscal bailout during 1995-2001, poses a particular challenge. Via a 2002 legal reform, the national government lost governance control over the PAMI, yet remained responsible for its overall financial results and viability. This happened precisely when chronic deficits and service delivery problems were exacerbated by the crisis situation and when no structural reforms were being implemented or even discussed. There is an urgent need for legal reforms in PAMI that align incentives for performance with responsibility and accountability. Other structural reform alternatives-such as "provincializing" PAMI under well defined requirements and guarantees-should also be considered, along with measures to open it to competition. Without such type of reforms, the system is likely to remain unresponsive to the public and continue to require significant fiscal bailouts. PENSIONS Fiscal Implications of Recent Reforms 4.65 Argentina's century-old pension system is one of the oldest in the continent. Originally organized as a complex network of independent, funded schemes, it slowly evolved into a pay-as-you-go (PAYG) scheme, run by a government agency and covering most of the formal labor force. However, the aging process, together with lack of planning, poor management, and growing unemployment and informality, produced a fiscal crisis and led to a major reform of the system in 1994. In addition to measures to reduce long term fiscal liabilities, the reforms aimed at providing a source of long-term capital for private investment and promoting formality by creating better incentives in the pension system. 4.66 The new system comprises two pillars: (i) a basic pillar (PBU) that offers a flat benefit to all new retirees with at least 30 years of service who meet the minimum age requirement of 60 years for women and 65 years for men-this benefit is roughly 28 percent of the average wage of current workers and (ii) a second pillar under which workers must chose to join either a new, smaller PAYG scheme run by the government or a funded scheme with individual accounts run by private pension fund operators called AFJPs, whose benefits depend on accumulated contributions and the performance of the fund. The benefits of the new PAYG were constructed to be similar to those in the funded schemes-roughly 30 to 35 percent of the pre-retirement wage for a worker with 35 years of contributions. This combined benefit of about 60 percent of the pre-retirement wage for 46 See World Bank (2003) for a more detailed discussion. 46 Table 8: Fiscal Effects of Pension Reform and Other Policies (Percent of GDP) "Pure" Transition Effects Other policies affecting the pension system (1) (2) (3) = (1- 2) (4) (5) (6) = (4 + 5) (7) = (3+ 6) "Pure" Reduction in Costs of Transfers Savings in Reduction in Transferred Total Other to AFJP Expenditures Cnt Co Provincial Policies Total Effect Cost Contributions Systems 1993 0 0 0 0 0 0 0 1994 0.29 0.14 0.15 0.41 0.23 0.64 0.79 1995 0.83 0.28 0.55 0.37 0.35 0.72 1.28 1996 1.03 0.4 0.63 0.77 0.7 1.47 2.11 1997 1.16 0.5 0.65 0.84 0.83 1.67 2.32 1998 1.27 0.62 0.65 0.77 0.83 1.6 2.25 1999 1.41 0.8 0.61 1.09 0.88 1.97 2.58 2000 1.39 0.95 0.44 1.47 0.86 2.33 2.77 2001 1.37 1.16 0.21 1.47 0.91 2.37 2.58 2002 1.14 1.05 0.09 0.98 0.72 1.7 1.79 Source: Rofman (2003) average workers is significantly below the 70-82% replacement rate promised by the system prior to the 1994 reform, although actual benefits used to be lower. 4.67 ANSES, the government's pension administrator, pays benefits (i) to all workers who retired before the reform, (ii) the first pillar benefit for those retired after the reforn, (iii) the new PAYG benefits, and (iv) some transitional benefits to compensate for the years of contributions before 1994. By 2000, ANSES was able to finance less than 30 percent of its budget with contributions, while the rest was covered with general revenue and other earmarked taxes. This is mainly because of two government policies in the mid- 1990s: (i) the transfer of some provincial public pensions to ANSES, aimed at curtailing fiscally irresponsible behavior on the part of some provinces as well as a desire to make pensions more uniform across the country, and (ii) a number of reductions in employer contributions aimed at stimulating formal labor demand and reduce unemployment. 4.68 The implementation of these two policies resulted in an average fiscal loss of 2.35 percent of GDP during 2000-2001, much larger than the net cost (0.33 percent of GDP) of moving from the PAYG system to the fully funded system (Table 8). This is because the loss of revenue generated by the transfer of contributions to the pension fund (averaging 1.4% of GDP in 2000-2001) was compensated by savings in expenditures, by almost the same amount. During 2002, the real value of wages, contributions and benefits decline sharply in real terms, improving ANSES's financial position. As a result, national pension spending (including ANSES and provincial systems) declined from 6.3 percent of GDP in 2001 to 5.0 percent of GDP in 2002. If benefits are not adjusted during 2003, spending at the national level will decline further to 3.8 percent of GDP while consolidated expenditures on pensions, which averaged 8 percent of GDP during the 1990s, should be below 5 percent of GDP. 47 Issues and Prognosis 4.69 Assuming that the value of pension benefits in real terms will gradually be restored to the levels prior to the crisis, Grushka, 2002 estimates that national spending on pensions will decline to only about 1 percent of GDP by the year 2050, when expenditures linked to the old system (including transitional benefits) will have disappeared and the only relevant component will be the flat benefit (PBU). The projected decline in expenditures is due, in part, to a projected decline in coverage-the 30-year contributory requirement and higher informality in the labor market will result in more workers unable to claim benefits. 4.70 The reduced coverage is both a serious social problem and a threat to longer-term fiscal solvency because of rising social pressures to modify the system. The percent of the population aged 65 or older who had a pension or a survivor benefit declined from 77 percent in 1993 to 67 percent in 2002 and is projected to decline to 40 percent in the medium term. Stopping the decline will be even more challenging as public confidence in institutions in general, and financial institutions in particular, is very weak. Even before the crisis the pension system had difficulties in generating worker interest. For instance, over three-quarters of new workers did not make an explicit choice to join a pension fund or the PAYG, and had to be assigned through a default process. 4.71 Partly in response to the decline in coverage and compensation, the Government created a program called "Plan Mayores" in March 2002, as part of the workfare program "Jefes de Hogares". Under this plan, all residents older than 70 years with no other income receive a monthly benefit of $150. This program is estimated to cover approximately 360,000 beneficiaries, for an annual expenditure of $650 million. However, it is unclear how this program will be linked to the formal pension system, the non- contributory pension programs, or the provincial old age benefits. Furthermore, there is no clarity on how it is going to be financed or whether the program will be permanent or temporary. 4.72 Another fiscal liability facing the government stems from the government decision in early 2002 to unilaterally convert certain dollar-denominated government loans accounting for almost 75 percent of pension fund assets into pesos at an unfavorable exchange rate. Most pension fund managers have refused to accept the conversion, and while the government has continued to deposit interest payments with pension fund custodian banks, a court decision to revert the instruments to dollar-denominated assets, as has been done for dollar deposits, could result in high fiscal costs. 4.73 Some of the coverage and fiscal problems in Argentina's Social Security are caused by the high level of evasion. Collection for all social programs, (including Social Security, Health, Unemployment, and Family allowance) is made by INARSS, a federal revenue collection agency created in 2001. The organization of INARSS has still not completed, and its competence to collect, enforce compliance and distribute social security funds needs reinforcement. At the same time, there are rules and practices regarding labor contracts and payments that effectively exempt certain types of income, creating loopholes and reducing transparency in the system. 48 Recommendations 4.74 The discussion above points to a clear need to redesign the pension system in order, to address problems relating to coverage, efficiency and fiscal costs. Such a major reform will only be politically feasible following a society-wide discussion on the design and financing of the old age income protection system. A piece-meal approach that attempts to solve problems without a systematic orientation is unlikely to work. 4.75 Enhancing coverage is likely to require the integration and expansion of the different benefits aimed to prevent poverty among the elderly, namely the PBU and non- contributive pensions. In order to offer some protection to workers exposed to periods of unemployment or informality, the requirement of having 30 years of contributions could be replaced with a fiscally neutral scheme that recognizes shorter contribution periods but offers incentives to contribute for longer periods. Over the longer-term, one could aspire towards the progressive integration of all national, provincial, professional, and special pension systems into a unique set of rules and supervision procedures. Even though some schemes may continue to have independent administration, they would be under a common structure and provide full and complete information disclosure. 4.76 While the above recommendations will need wider public debate and further discussion before being implemented, several legal, regulatory, and administrative measures to increase transparency could be taken right away. They include: * Taking administrative and/or legal steps to make all components of wages- including special benefits, tickets, and bonuses-taxable for pension purposes and to reduce the abuse of self-employment and other special contributory status classifications. * Undertaking a thorough and independent review of the actuarial and financial status of existing provincial, professional and other special systems to provide clear and transparent information to the Government and the general public regarding their financial status, benefit levels, and the existence of implicit or explicit cross subsidies and inequities among the different schemes. * Simplifying the benefit granting process, both at ANSES and AFJPs, by recognizing as valid information from the SUP individual accounts, by removing the burden of proof from beneficiaries as much as possible, and by establishing clear rules to adjust benefits over time. * Revising investment regulations for pension funds, life insurance and annuities companies to achieve greater flexibility in asset allocations, ensure a level playing field, and increase risk diversification. * Improving the regulatory framework for private pension and insurance services providers to promote competition and reduce costs. * Reinforcing the political autonomy and technical independence of the SAFJP (Superintendence for AFJPs), by ensuring that the senior managers are professionally qualified and that appropriate 'Chinese walls' are set to separate them from other areas of Government. 49 SOCIAL ASSISTANCE 4.77 When the convertibility regime collapsed, Argentina did not have an effective safety net, leaving many of the poor uncovered by any program. Social and economic conditions deteriorated from an already bad position and the number of both the poor and the unemployed rose sharply. During the first few months of 2002, the Government responded quickly to establish a large-scale program to provide social assistance to families at risk, given the fears of social unrest. The Heads ofHousehold Program, which spent $1.7 billion in 2002, constitutes the core of that effort.4 The program is in some aspects similar to a previous workfare program, Trabajar (Box 5), although, as explained below, there are some important differences. 4.78 First, anyone who meets the eligibility criteria (unemployed head of household with children under 18 years of age, or pregnant spouse, or a handicapped person) has the right to participate in the Jefes program. In contrast, Trabajar had no eligibility requirements, but participants entered the program only when selected for a workfare project, which often excluded many. Second, the kinds of work activities that can be undertaken under Jefes were broadened considerably to include community services such as working in a soup kitchen, health center, or school as well as activities such as rehabilitation, maintenance, cleaning, and even sewing circles. Moreover, the work requirement can also be fulfilled through participation in training or basic education courses and number of hours of work required was reduced from 6 to 4 hours/day. 48 Third, although the Ministry of Labor continued to manage the program, most of the responsibility for implementation was transferred to municipalities (registration of beneficiaries, selection and supervision of activities) through councils whose members came from government, NGOs, and the private sector. All participants receive the same monthly payment of $150. 47 Presidential Decree No. 565, dated April 3, 2002, created the Heads of Household Program. It followed a short-lived program with the same title, but a different design and operating procedures. The latter program was gradually phased out during 2002 and by August 2002 all of the beneficiaries that qualified transferred to the program established by the April decree. 48 The program design also includes a requirement for beneficiaries to present school enrollment and health certificates. However, this has not been enforced. Moreover, the program design includes no follow-up of these condition or supporting to their fulfillment. 50 Experience with the Heads of Box 5: Previous Experience with Workfare Household Program Programs: Trabajar 4.79 Given the speed with which the In 1996, the Ministry of Labor consolidated its employment programs into a large-scale program called Program was implemented, the start-up Trabajar. The decision was motivated by an increase in was disorderly. The co-existence for unemployment rates in the latter half of the 1990's and several months of two Jefes Programs the fact that the poor, who were the most affected, were with different designs created confusion. not covered by unemployment insurance. Trabajar was Although the Program eliminated the a workfare program, meaning that participants were problems associated with rationing, required to work in order to obtain benefits. The program used technical criteria to select workfare registration of beneficiaries was not projects and distribute resources and paid beneficiaries always fair. And while two-thirds of the directly through banks, thereby enhancing efficiency beneficiaries registered through their and reducing fraud. municipality, as required by the law, Beneficiaries had to be assigned to a particular others gained entrance to the benefits via approved workfare project in order to enter the local political leaders or piquetero Program. The program often did not receive enough groups. The work requirement was not financing to provide a place for every worker who groups. The wanted to participate. Because of the rationing, there emphasized initially and localities were were accusations of clientilism. given no time to come up with workfare activenotimes bfore peope upwih wores red An evaluation of Trabajar showed that on average the net income gain to beneficiaries after taking into and started to receive benefits. The account their foregone income (the counterfactual of number of beneficiaries expanded what these workers would have been earning had they rapidly from about 500,000 in May to not been participating in Trabajar) was $103, or about 1.7 million in October and approximately half of the average Trabajar wage ($200 pesos/month). 127 million in Octoberil 2003 approximate The targeting was very good by international 2 mnillion by April 2003. standards-over half of the participants came from households in the lowest income decile and about four- 4.80 When the Jefes de Hogar fifths from the lowest income quintile. Program was established in April 2002, In addition to trying to reach poor workers as direct the Government instituted several participants in the workfare program, Trabajar tried to mechanisms to promote transparency and promote work activities in poor areas, so as to benefit social control. In addition to poor residents and ensure that the projects had high decentralizing much responsibility for benefits. Evaluations revealed that: the operation of the Program to local * Gains to the poor varied with total spending-the municipal councils, the General Audit political economy of the program's operation in Agency was to undertake periodic audits, most provinces entailed that cuts were borne an 800 line was set up to respond to mainly by the poor complaints and inquiries, and the * Targeting to poor areas tended to improve over Ministry of Labor said it would make time until 2000, when there were sharp reductions information on implementation available in expenditures on the Program to the public. * Targeting to poor areas was substantially better than in the other two labor programs managed by 4.81 The results of these positive the Ministry partly because of Trabajar's emphasis intentions are mixed. While local on services as opposed to works. councils have been set up in most jurisdictions, they have not been able to exercise fully the role assigned to them. For example, they have no control over the third of beneficiaries who registered outside of the municipalities. Requests to drop beneficiaries from the program because of irregularities, have not consistently been acted on by the Ministry of 51 Labor. The General Audit Agency published on its website reports on Program implementation in 15 provinces and 5 municipalities, but little remedial action has been taken to respond to the weaknesses identified. The 800 line is not adequately staffed, and follow-up is deficient. The Ministry of Labor made a good start in publishing information, but when elections approached, staff stopped providing data on the program. 4.82 Recent trends in beneficiary registration are cause for concern. Most participants should have registered by the end of 2002. Even though the economy started to grow in 2003, the number of beneficiaries increased by 7 percent between the end of 2002 and April 2003. However, the increase varied greatly among provinces. Buenos Aires and Jujuy had increases that were more than double the national average. Also, while the number of beneficiaries declined in 14 provinces by the end of February, by April, only 6 provinces showed a decline as compared to December 2002 even though the performance of the economy continued to improve (Table C 15). 4.83 Reporting requirements are cumbersome. Those wishing to enter the program are required to report on too many personal and household characteristics. Complex, poorly designed, and lengthy forms contributed to difficulties and irregularities. The requirements were well beyond the capabilities of many municipalities. In practice, a lot of the information was not collected in many cases and, even when collected, not entered into the registry. Data from a survey carried out in October 2002 provides some evidence of weaknesses in the registration process in some areas.49 Among the beneficiaries of the program in the survey, 23 percent appeared to be "ineligible" because the household did not contain a child below 18 years of age or a handicapped person. This share varied considerably among regions, from 20 percent in Greater Buenos Aires to slightly over 30 percent in the Northeast and Cuyo Regions.50 Only 40 percent of the program participants appeared to be household heads (the title of the program notwithstanding)-which is consistent with data showing that 70 percent of program beneficiaries were women as opposed to 40 percent in the employed population (Table 9). Among households that were not participating in the program, nearly 80 percent appeared to be ineligible (did not have a child under the age of 18 years of age or a handicapped person). Another 7 percent, although they appeared eligible, were not poor. Thus about 13 percent of people who appear eligible for the program do not participate. 49 Known as the Permanent Household Survey (EPH), it covers large urban areas accounting for about 70 percent of the total population. ° In February 2003 alone, the Ministry of Labor announced that 22,389 beneficiaries were dropped from the Program because they had claimed eligibility based on pregnancy, but had not been able thereafter to document the existence of a child. Although this represents only 1 percent of beneficiaries, if these figures are repeated in the future, it would indicate a defect in the registration procedure. 52 4.84 The participants tended to be Table 9: Characteristics of Beneficiaries of HH younger and less educated than the Program employed. About a third were below 30 years of age, and another 30 SexBeneficiaes Employed percent were between 30 and 39. Male 31 60 Nearly 20 percent had not completed Female 69 40 primary education, while another 40 percent advanced no further than that Age Group level. The participation of single and 20-24 14 12 two parent non-nuclear households 25-29 18 14 was higher than their share in the 30-39 30 24 overall population covered by the 40-49 22 23 EPH-each accounts for about a fifth 50-59 8 17 of participants. For slightly over half 60 or older 3 8 of the beneficiaries, the spouses Education worked. While most of those had Incomplete Primary 18 8 either informal or non-wage earning Primary 37 22 jobs, about 10 percent receive some Incomplete Secondary 27 18 kind of benefits. Between 10-15 Secondary 13 21 percent of households have more than Incomplete Terc/Univ. 5 13 one member participating in the Terc/Univ. 1 19 Program. Household Position Head of Household 43 52 4.85 Perceptions regarding the Spouse 34 20 program tend to be strong and Other 22 28 surprisingly negative. Even among Household Size 5.4 beneficiaries, 40 percent had negative views on the program. About 30 Source: EPH, October 2002 percent considered it to be very transparent, 36 percent only somewhat transparent, and 15 percent do not consider it to be transparent. The rest had no opinion. When asked which institution would be best able to implement the Program fairly, the opinions of beneficiaries were split evenly among national, provincial, and municipal governments, and the church. Beneficiaries were least confident in unions, non-governmental organizations, and piquetero groups and two-thirds believed that the Program was only a temporary solution to the current lack of employment. 4.86 Among the general public, a survey undertaken in October 2002 showed that opinion on the Heads of Household Program was split. About 30 percent of the public thought that the program was good, a similar share thought that the program was average, and 40 percent thought that the Program was bad. The public gave low rankings, particularly on transparency and political independence. While their opinions on efficiency in combating poverty and on administration were better, the share of negative responses outweighed the positive. 4.87 In December 2002 and January 2003, in depth interviews were carried out among mayors, members of Consultative Councils, representatives of Civil Society, national 53 legislators, and representatives of unions and piqueteros. Most respondents thought that the eligibility criteria were necessary and valid and that anyone who met them should be allowed to participate. However, the work requirement was more controversial. Unions, civil society organizations, and opinion leaders believed the stipend was too small to justify the work requirement. Many were confused, saying that when the Government began the program, it was treated as an unemployment subsidy. It only later added the work requirement. Everyone agreed that the amount paid was small in relation to the needs of the family. Most respondents felt the program was well managed, taking into consideration the speed with which it was put into place. Most also thought that in spite of political influence in the program, benefits did reach the poor. Some respondents believed that piquetero groups tried to include people who did not meet the eligibility criteria (without dependents mainly) into the program. They also thought that the piquetero groups linked access to the program to participation in political events. Negative views tend to be based on general perceptions rather than based on directly observed behavior of the program. Many thought that the program has suffered because of the political context. 4.88 During these interviews, respondents mentioned several areas that could be improved. More inforrnation on the registry of beneficiaries needed to be provided at the local level, particularly as to why certain people had been excluded. The registry needed to be computerized and made accessible for consultations at the local level. Information on what beneficiaries are doing for the work requirement needed to be publicized to strengthen public support for the program. More information on the program needed to be made available at the community level. 4.89 In May, the Government prepared a set of actions to respond to the weaknesses and irregularities identified in the Program. These included: closer supervision, the design of a system of sanctions, improving the dissemination of information, obtaining identification numbers for all family members in order to enforce more strictly eligibility requirements, and carry out more cross-checks of beneficiary lists with other data bases. The new administration in the Ministry of Labor is committed to enforce eligibility requirements to reporting periodically on progress in carrying out these remedial measures. Some additional measures to complement these actions are suggested below. Improving the Heads of Household Program 4.90 This program is meant to be an emergency intervention that would gradually wind down as the economy picks up and generates more employment. For this to occur, the transfer level under the program must be below the wage rate for similar work among the poor. Based on the survey data of October 2002, the wage rate is only slightly above the transfer level under the program. The situation needs to be monitored to ensure that the wage stays below the prevailing market wage for unskilled labor.5' Moreover, as wages differ across regions, local authorities should be given the flexibility to reduce the wage 51 In the public works program that was re-established by the Korean Government during the East Asia financial crisis, as wages fell, the public works wage was adjusted downward to ensure that participants have an incentive to take up regular jobs when available 54 rate in their area. Also as employment levels begin to increase in the economy, the Ministry of Labor needs to support beneficiaries to find more stable employment, such as assistance with job search and labor orientation. The Government might also want gradually to tie eligibility to the Program to proof of active job search and lack of alternative job opportunities. 4.91 Initially, the program design incorporated elements of a conditional transfer program-participants were required to provide the school enrollment and health certificate for their children. However, the program did not include the other components such as close links with service providers and family counseling as found in similar programs in Mexico and Colombia. In Argentina, there is little justification for the inclusion of these requirements in the program now that there is a conditional transfer program in Argentina (Familias), which is gradually expanding in urban areas. 4.92 In addition, administration needs to be improved to ensure that beneficiaries meet eligibility requirements. This requires: * Expanding the supervision of municipalities and imposing sanctions where consistent irregularities are observed * Providing to municipalities and local councils on a monthly basis the list of irregularities provided by SINTyS for remedial action * Instituting annual checks to control the quality of the registration process in provinces and some large municipalities-any substantial increase in net registrations in municipalities should be subject to scrutiny by the Ministry of Labor * Enforce the work requirement rigorously. An estimated 20 percent of beneficiaries tend to be found among better-off households. In addition, many are working less than the required minimum of 4 hours/day. Preliminary results show that poor enforcement of the work requirement is adversely affecting the performance of the program in reaching the poor and leading to a regressive distribution of net gains. 4.93 Both the cost-effectiveness of a workfare program and public support for it depend to a large extent on the value of the projects undertaken. The broadening of the types of activities permitted may have been unavoidable initially given the need to expand participation in the Program quickly. However, now that the economy has stabilized, efforts should be made to improve the quality of the workfare activities. First, the national Government should co-finance the materials component of projects in poor areas. Second, as the number of beneficiaries shrinks, the types of activities which can be undertaken to meet the work requirement should be tightened in order to favor higher value ones and eliminate ones with little or no value such sewing circles. 4.94 Over the medium-term, the national government should consider taking into account the shared responsibility for social protection among the different levels of government. For example, the national government could leave responsibility for the implementation and financing of such a program to the provincial or municipal level unless the rate of unemployment in a locality exceeded the national average. There is also a need for better monitoring efforts to see the extent to which the program reaches the poor and a 55 willingness to modify the design, funding, and implementation of the program as appropriate. The longer-term vision for the program is one of reduced size, higher quality, and focused on a counter-cyclical role aimed at reaching the poor. The Conditional Transfers Program, Familias 4.95 During the second half of 2002 the Government began to make payments to beneficiaries of another social assistance program, called Familias. This is a conditional transfer program: subsidies to families are made in exchange for the commitment of the parents to ensure participation in the stipulated health and education services. The program operates in the large urban areas of the country. By the end of 2002, there were about 100,000 families participating and by the end of March 2003, that number had reached nearly 175,000. The goal of the program is to reach 275,000 by the end of 2003 and eventually slightly under 500,000. 4.96 The program is designed so that the mother in the family receives the subsidy, which is calculated per child. Actions defined at the local level are also taken to facilitate the access of the families to the needed health and education services. Eligibility criteria include: (i) having at least one child under 19 years of age or a pregnant women; (ii) having income no higher than the poverty line; and (iii) receiving no other benefit from the Government. As the latter includes family allowances and health insurance, it effectively excludes all those in the formal labor market. Whether families are above or below the poverty line is determined by a proxy means test (SISFAM), which is applied in poor neighborhoods. In addition, any family wishing to participate in the program, but who has not been included in the SISFAM registry, must do so. 4.97 A conditional transfer program such as Familias is designed to address structural poverty, while workfare (Jefes) should be better able to respond to cyclical changes in economic and social conditions. In addition to income redistribution, the aim of conditional transfers is to ensure human capital investment in children when poor households may lack the needed resources (and cannot obtain credit). Workfare is best suited to offer income support to those willing to work, but unable to obtain full-time employment. 4.98 Since they address different needs, both types of programs could be part of the Government's social safety net. However, in order to ensure targeting to the poor and effective support to poor households, the implementation of the two programs needs to be better coordinated at the local level, facilitated by a strategy and framework agreed at the national level. For example, it would be reasonable to direct families with numerous younger children, particularly those with female-heads, to the conditional transfer, instead of the workfare program. These families are more affected by "structural poverty" and are more at risk for insufficient investment in their children. It may also be difficult for them to carry out the work requirement. This also would reduce pressure on the Jefes Program. The geographical focus of Familias (large urban areas) covers exactly those areas of the country that often have trouble coming up with a sufficient number of workfare activities to offer reasonable assignments to all who might meet the eligibility requirements, while 20 percent of the Jefes program are headed by women-the target population of Familias. 56 4.99 The conditional transfer program also needs to be coordinated carefully with the existing scholarship program for secondary students and with the system of family allowances. Savings from elimination of duplication of benefits could be used to extend the coverage of family allowances beyond formal sector workers. Finally, a rigorous, impact evaluation is essential for Familias in order to set the "correct" transfer level, to ensure that resources are reaching the poor, and that program objectives such as higher educational attainment and better health indicators are being achieved. Feeding Programs 4.100 In response to the social and economic crisis in 2002, the Government increased spending on feeding programs. Although there were some improvements from the previous centralized purchase and distribution of in-kind items, the feeding programs continue to be characterized as income support to needy families rather than nutrition interventions. Some of the financing was transferred as block grants to provinces via the Programa Emergencia Alimentaria (PEA). Another line of spending was supporting in the larger, poorer metropolitan areas, existing "soup kitchens" run by non-governmental organizations or community groups for vulnerable groups, mainly children. 4.101 As of now there has been no analysis of how efficiently feeding programs function: how well they are targeted, the extent to which they succeeded in transferring income to families. Beyond their function as an income transfer mechanism, the usefulness of feeding programs to address problems of malnutrition is questionable. Assuring that Argentine children receive adequate nutrition is a task that needs to involve the joint participation of health, education, and social assistance programs.52 52 The primary health care level participates by identifying, through periodic controls, children who are not growing adequately. Those who begin to fall off the standard curve for weight gain require immediate attention (initially parental counseling). Cases of severe malnutrition require immediate hospitalization, followed by intense social work. Close work between Early Childhood Development and health centers, including active community participation, serves the purpose of detecting and following at risk kids. The provision of fortified milk to children under the age of two who are brought to health clinics for immunizations and growth monitoring also plays a useful role: while this extra intake will not prevent acute malnutrition in a situation of acute deprivation, it can help prevent chronic malnutrition and anemia. The education system can also play a role by combining their own in-school feeding programs with information and training to parents and students. 57 58 5. ENSURING EFFICIENT INFRASTRUCTURE SERVICES FOR GROWTH BACKGROUND AND MOTIVATION 5.1 Several studies have shown infrastructure to be an important determinant of economic growth. The elasticity of GDP with respect to infrastructure stocks in Latin America has been estimated to be about 0.15 for Bolivia, Colombia, Mexico, and Venezuela (Baffes and Shah, 1998), 0.23 for Argentina (Fay and Yepes, 2003), and 0.30 for Brazil (Ferreira, 1996). Thus for every one percent increase in the infrastructure stock in Argentina, GDP is estimated to increase by 0.23 percent. Moreover, both in Argentina and Brazil, the geographical distribution of infrastructure stocks has been found to be a significant determinant of convergence between rich and poor regions-infrastructure investments in poorer areas reduce inter-regional inequalities (Estache and Fay, 1995). 5.2 Until the 1980s, infrastructure services throughout the world, especially in developing countries, were predominantly provided by state owned enterprises, most of which were heavily reliant on public subsidies to cover investment needs and sometimes, even operating deficits. This drain on the budget, coupled with growing dissatisfaction with the quality of public services, contributed to a major paradigm shift in the late 1980s and 1990s. The new model for infrastructure provision entailed restructuring national monopolies to open the way for competition wherever possible. Operation of services was delegated to the private sector under a variety of contractual arrangements and the state began to focus its attention on regulating the natural monopoly elements of the sector, as well as overseeing the transition to a competitive market. 5.3 Argentina was at the forefront of the global reform movement in the infrastructure sectors,53 undertaking one of the earliest and most comprehensive privatization programs among developing countries, dramatically changing the role of the public and private sectors in infrastructure provision (Table DI). Following the publication of a State Reform Law (No. 23696 of 1989), the state owned telephone monopoly, ENTEL, the state owned electricity companies, SEGBA, Hidronor, and AYEE, and the state owned natural gas monopoly, Gas del Estado, were restructured and privatized. Over the same period, concessions were awarded for most of the country's railroad network, about a third of the national road network, and for several ports and airports. The provinces rapidly followed suit with concessions awarded to private companies for electricity distribution in 11 provinces, as well as water and sewerage services in the City of Buenos Aires and 13 provinces (Table D2). Overall, Argentina captured 1 1 percent of private sector capital flows to developing countries worldwide during the 1990s, more than any other developing country in the world except Brazil (Izaguirre, 2002). 53 Infrastructure is defined broadly to include electricity, oil and gas, water and sewerage, seaports and airports, telecommunications, roads, and rail. 59 5.4 Views relating to the success of these infrastructure reforms are polarized. This chapter will start by summarizing the achievements of the reforms and provide a critique of their implementation. Against this backdrop, it will discuss the measures taken during the recent recession. Finally, the chapter will suggest ways to ensure the sustainability of infrastructure services. Without efficient provision of infrastructure services, growth is unlikely to be sustainable and living standards unlike to improve. Without sustainable growth, long-term poverty reduction efforts cannot succeed. Achievements of Infrastructure Reform 5.5 The privatization program achieved substantial fiscal savings. First, the 86 infrastructure transactions during the 1990s raised proceeds of US$12.1 billion. Second, about 3.0 billion of public debt was transferred to the private concessionaires. Third, the federal government was able to eliminate about US$1.8 billion worth of subsidies that it had been making to public providers of infrastructure services. Fourth, the newly privatized infrastructure providers generated direct and indirect tax revenues of $4.5 billion in 2001 alone-equivalent to about 1 percent of GDP. Finally, the federal government benefited from the payment of canon for some concessions-primarily in the transport sector-and received dividends on its residual shares in these companies, many of which were making losses before privatization. 5.6 As a result of these reforms, consolidated public expenditures on economic services fell from about 7 percent of GDP in the late 1980s to 1.8 percent of GDP in 2001 and 1.4 percent of GDP in 2002 (Gerchunoff et. al, 2003). Over the same period, economic services' share of total public expenditure fell from 25 percent to around 5 percent, which created the necessary fiscal space for the share of public expenditure allocated to social objectives to rise from around 50 percent in the 1980s to over 60 percent in the 1990s. In the 2003 budget, public expenditure on infrastructure services stood at just of $3 billion, of which two-thirds is allocated to the transport sector for the construction and maintenance of roads, and about half is executed at the provincial level of government (see Table D3 and Figure DI). 5.7 The allocation of public expenditure for infrastructure across the provinces does not seem to bear a very close relationship to the underlying investment needs. For example, federal transfers per kilometer of the provincial road network range from $500 per kilometer in Santiago del Estero to over $6,000 per kilometer in Tierra del Fuego. When provincial expenditures are added in the disparity grows, ranging from less than $10,000 per kilometer in Santiago del Estero, to almost $50,000 per kilometer in La Rioja (Figure D8). Moreover, at the federal level, the amount spent on road maintenance varies hugely according to the contractual regime, ranging from $1,000 per kilometer under simple maintenance contracts to $8,000 per kilometer for rehabilitation and maintenance contracts (CREMA) to $17,000 per kilometer for COT contracts which include improvements and expansions as well as rehabilitation and maintenance (Figure D9). Similarly, federal allocations for investment in rural electrification range from less than $100 per dispersed rural inhabitant per year in the Province of Buenos Aires to over $9,000 per dispersed rural inhabitant per year in the Province of Santa Cruz (Figure D1O). 60 5.8 Beyond these fiscal achievements, the Argentine reforms have enhanced the efficiency, coverage, and quality of service in the infrastructure sectors. On study finds that the Argentine privatization program created efficiency gains whose present value is estimated at 1 percent of GDP (Chisari et al., 1999). Other studies find total factor productivity gains in different infrastructure sectors to range between 2 and 4 percent (Estache, 2003). Labor productivity gains were an astounding 250 percent on average in a sample of the largest private operators (Figure 5(a)). Much of the gains came major cutbacks in the workforce, which fell to 25 percent of pre-privatization levels. Large gains also occurred in the provinces. For example, with private participation in water services, labor productivity almost doubled in both Corrientes and Santa Fe. 5.9 There have also been substantial improvements in coverage and quality of service. In the telecom sector, the average waiting time for a new line fell from nearly 8 months to barely 3 days between 1994 and 1999, while its cost fell from $1,500 to $200. Moreover, the digitalization rate of the network rose from 13 percent to 100 percent, and the number of faults per 100 lines per year fell from 42 to 17. In the City of Buenos Aires, electricity outages have dropped by over 65 percent, while the percentage of customers with inadequate water pressure has fallen from 85 percent to 56 percent (CEER). The improved telecommunication services facilitated the development of the financial sector. 5.10 In the City of Buenos Aires, the number of customers served from the early 1990s to 2000 increased at an average annual rate of 14 percent for telecom, 11 percent for metro-rail, and between 2 percent and 5 percent for electricity, gas, water, and sewerage. As a result, residential service coverage for water and telephony (both fixed and cellular) rose by 15 percentage points since 1993, while coverage of electricity and gas rose by 6 percentage points. Recent household survey evidence indicates that these coverage gains were primarily concentrated in the bottom half of the income distribution, for all services except cellular telephony (Figure 5(b)). Even larger gains in water and sewerage coverage, of the order of 14 to 37 percentage points, resulted from private sector participation in the provinces of Cordoba, Corrientes and Santa Fe. 5.11 Substantial increases in private investments made these gains possible. During the 1980s, Argentina invested an average of $2.5 billion per year from public sector sources Figure 5: Performance Improvements Following Privatization (a) Percentage Gain in Labor Productivity (b) Percentage Gains in Coverage 850Aguas ___ __ _ _ _Water SW - ~~~~~~~~~~~~~Argentinas 35-- Wae 450 -.--EDENOR :2 30 - 400 - 9 1*~~~~~~~ -B- Sewerage - 400 / ~ -, EDESUR 0 25 - - 0 3as0 -c 20 - Electncity 250 -_ BAN 10_ ---Gas loo: CL -a~~~~~~~~~~~~~~~~~~~~~~~~~~~- Fixed "o100 +SUBTE 1 telephony g0 cm D W co r co X 1 2 3 4 5 -e-_Cellular O o) 0 , al l Telecom + Income quinvles telephony …~~~~~~~~~~~~~~Telefonica Source: Adapted from CEER, 2002 and OPSM, 2002. 61 (Calderon et al., 2003). Since the mid-1990s, the private sector alone invested $5.2 billion annually in infrastructure, while public investment fell to $1.5 billion annually. According to Calderon, et al., 2003, Argentina succeeded in financing 75 percent of its infrastructure investment requirements through the private sector during the 1990s-significantly higher than in Brazil (53 percent), Chile (65 percent), Colombia (29 percent), and Mexico (54 percent). The share of private finance in infrastructure ranged substantially across sectors, from about 40 percent for roads and water, to almost 100 percent for electricity, gas, railroads, and telecom (Calderon et al., 2003). Critiques of the reform 5.12 Notwithstanding these substantial achievements, public discontent with the infrastructure reform process has grown throughout the 1990s. Indeed, a regular poll conducted by Latinobarometro found that in the year 2000, 78 percent of Argentine's disagreed with the statement that 'privatization of state companies has been beneficial to the country', up from 50 percent in 1998. 5.13 One of the principal critiques of the Argentine infrastructure reform relates to the equity with which the efficiency gains resulting from privatization have been shared between consumers, investors, and the government (Estache, 2003). This can ultimately be attributed to weaknesses in the regulatory system and concession design, which often failed to translate cost reductions into tariff reductions. In addition, regulators repeatedly gave in to renegotiation pressures from the private sector. Furthermore, the growing tax burden on the infrastructure sectors has sometimes prevented consumers from perceiving real reductions in the underlying tariffs. 5.14 The weakness of regulatory institutions varies by sector. In electricity and gas, where there are independent regulators, the experience has been good. In the transport sector, where there are no independent regulators, most contracts were renegotiated. A recent study finds that 42 percent of infrastructure contracts in Argentina were renegotiated within a two-year period, outside of the normal mechanisms established in the contract (Guasch, 2003). This compares with a rate of 28 percent for Latin America as a whole. Renegotiation rates ranged from 80 percent for water and transport, to 0 percent rate for energy and telecom. This variation is entirely consistent with experience elsewhere in Latin America, except that the renegotiation rate for the transport sector is substantially higher in Argentina than elsewhere in the region. 5.15 Many of the past renegotiations have led to positive outcomes for concessionaires, and primarily take the form of reductions in canon payments, reductions in tolls or tariffs, and increases in subsidy payments (Table D4). The case of the inter-urban toll road concessions is particularly flagrant. The original canon payment was eliminated within a year of signing the original contracts, and replaced by a public subsidy payment. Over the following 12 years, there have been some 20 government interventions outside of the original contractual framework (Table D5), each of which has had the effect of reducing toll rates and raising public subsidies. As a result, the annual volume of subsidy payments has risen from $18 million in 1991 to $363 million in 2003 (Figure D2). A recent decision by the Government to terminate these contracts on their scheduled expiry date of October 62 2003 provides an opportunity to draw a line under this problematic experience and seek a more solid basis for the future of the federal road network. 5.16 A common complaint among Argentine consumers is that prices of services are high by international standards, and that this is attributable to excessive profits by the private sector, as well as indexation to the US CPI, which investors demanded because indexation of contracts to domestic inflation indices was illegal. The US CPI grew by over 15 percent between 1995 and 2001-a period when domestic inflation in Argentina was close to zero (FLACSO, 2002). Although regulators mandated tariff reductions for many sectors, on average these have been significantly lower than the correponding gains in total factor productivity (Table D6). Nonetheless, many of Argentina's tariffs are low by international comparisons. Before devaluation, gas tariffs were among the lowest in the region, electricity and water tariffs were towards the middle of the range observed around the region, and telephone tariffs were among the highest (Figure D3). Nowadays, with tariffs fixed in pesos and at an exchange rate of three to the dollar, Argentina's utility tariffs are among the lowest in Latin America. 5.17 Empirical evidence on the profitability of utilities is scarce. Based on simple calculations, Estache, 2003 finds returns of 7-8 percent for the electricity distribution sector, 6-11 percent for the gas distribution sector, 17-18 percent for the telecom sector, and 22 percent for the water sector. As a point of comparison, the corresponding regulatory agencies have estimated a cost of capital of around 14 percent for the gas distribution sector and 12 percent for the water sector. 5.18 Argentina is an outlier in the indirect tax rates levied on utility bills. These typically amount to about 28 percent of residential bills for water and energy. Owing to wide variations in provincial and municipal taxes, they can be as high as 50 percent in some jurisdictions. Based on comparisons with 20 OECD countries, Argentina has the highest indirect tax rates of all in the water sector, and is in the top five indirect tax rates for the energy sector (Figure D4). These high indirect tax rates go some way towards explaining consumer perceptions of high utility prices. B. RECENT DEVELOPMENTS IN INFRASTRUCTURE 5.19 The devaluation of the peso and subsequent crisis has had very major impacts on the financial situation of the private infrastructure concessionaires. Under the Emergency Law 25,561 of January 2002, infrastructure tariffs (previously expressed in US dollars) were converted to pesos at a rate of 1:1 and frozen in nominal terms throughout 2002, even as the peso fell by over 70 percent and the domestic wholesale price index rose by 90 percent. Despite the freeze in tariffs, consumption of all infrastructure services fell by about 10 percent during 2002 (Foster and CEER, 2003). Disconnections for non-payment of bills grew rapidly and the use of quasi-monies to pay infrastructure bills proliferated. 5.20 As a result, concessionaires faced a serious mismatch between their reduced revenue flows in pesos, and their operating and debt service costs, which are largely dollar- denominated. This led at least eleven concessionaires to default on an aggregate debt stock of US$8 billion, which represented approximately half of existing dollar denominated 63 debts. It also prompted a downgrading of the S&P rating for the Argentine infrastructure sector from BBB (investment grade) to CCC or D, effectively shutting off all new access to capital. In consequence, all investments barring the most critical have been slashed, with aggregate investment levels falling dramatically to an estimated level of $200m in 2002 from US$5.2 billion during the 1990s. Unless the situation is resolved soon, there is a serious danger that some concessionaires will withdraw, while others will remain unwilling to invest until the credibility of a regulatory framework is restored. Contract Renegotiation 5.21 Following the Emergency Law, a Commission for the Renegotiation of Concession Contracts was established under the auspices of the Ministry of Economy and charged with the renegotiation of 59 concessions under Federal jurisdiction. The Commission was originally scheduled to complete the renegotiation process in 120 days, however this deadline was extended, though progress has been limited. Similar renegotiation processes are being repeated at the provincial level. 5.22 The experience at the federal level thus far varies significantly between the utility concessions and the transport concessions. As regards utility concessions, no agreements have yet been reached. One of the major obstacles to the renegotiation has been the social sensitivity attached to any tariff increases, given the income losses that have been suffered by a large proportion of the population, and the absence of any 'social tariff' mechanism to protect the most vulnerable. In the energy sector, there were a number of attempts to provide interim relief through moderate tariff adjustments of the order of 10 percent supported by Executive Decree. However, these were repeatedly overturned in the courts on the grounds that they bypass the renegotiation process managed by the Commission. Another area of legal controversy has been the nature of consumer participation in the renegotiation process. At present, representation is limited to a single consumer advocate on the Commission. Although the Commission has made some attempts to convene public hearings these were subject to legal challenges, and for the time being the channel of engagement has been limited to public consultation documents. 5.23 The transport sector has fared significantly better. In the case of ports and dredging concessions, which have an international customer base, tariffs were simply redollarized, though some of the costs of doing business have fallen in dollar termns. Moreover, although railroad and road concessions are still under renegotiation, they have received significant respite in the form of increased subsidies from the Transport Infrastructure Trust Fund-Table D7 provides a summary of the status of the renegotiations. Although not formally part of the renegotiation process, metropolitan bus operators have also secured sizeable compensatory subsidies from the Transport Infrastructure Trust Fund. This policy is controversial given that there is an estimated 40 percent of excess capacity on the metropolitan bus network. Moreover, the per passenger subsidies result in higher payments made on the more busier routes. 5.24 Growing frustration of the private concessionaires with the delays in the renegotiation process prompted nine of them to file claims totaling US$4 billion with the International Center for the Settlement of Investment Disputes (ICSID), an affiliate of The 64 World Bank Group (Table D8). Three more concessionaires have initiated the process of filing claims, while a further 40 are currently in the prior stage of amicable negotiations. Foreign investors acquired the right to file such claims, even before the local course of justice has been exhausted, under numerous bilateral investment treaties that were signed by the Argentine government between 1992 and 2001. However, at the outset of the renegotiation process, the government unilaterally imposed an obligation to refrain from initiating any legal action against the government outside the framework of the renegotiation process. These actions have created a significant contingent liability that threatens macroeconomic stability and inhibits new investment. Proliferation of trust funds 5.25 Recent years have seen a growing reliance on-extra-budgetary mechanisms for the infrastructure sector financed from a variety of surcharges on petroleum, diesel, natural gas, and electricity. These funds jointly accounted for 40 percent of public expenditure on infrastructure in 2002 (Figure 6 (b)), and are expected to rise to 50 percent in 2003. They are primarily used in the energy and transport sectors to direct investment or operating subsidies to private sector operators under very complex financing arrangements (Figure D5). Their dramatic growth in the last year has largely because of a decision to provide subsidies to inter-urban road and metropolitan passenger transport concessionaires to compensate for the devaluation. 5.26 These extra-budgetary mechanisms take a variety of different forms (Figure 6 and Table D9). First, there are earmarked funds (fondos especificos) within the general budget totaling $638 million in 2002. Important examples include the petroleum tax, which is partially earmarked to the National and Provincial Road Agencies, and the wholesale electricity market surcharge that finances the National Electricity Fund for the development of the electricity sector in the Provinces. Second, there are trust funds (fondosfiduciarios) that build on the framework established by Law 24,441 of 1994. There are currently six infrastructure trust funds, and the creation of a seventh for the water sector is under consideration. The resources channeled to trust funds for the infrastructure sectors doubled from $872 million in 2002 to $1.8 billion 2003, with about three quarters Figure 6: The Phenomenon of Extra-Budgetary Mechanisms (a) Evolution (b) Sectoral Breakdown v 1800 2= 1400 a Special t _ r 1200 Funds 0 60_4 e X - Energy n _ ; 1000 a. % ELa lo-2--Lubwt@39 rdce -d 200 E'~~~~. -3 *Natonat 10% fi fi fi fi A oExagytrutn Budget E * * **Jt* aiY 1999 2000 2001 2002 2003 Energy EneWy Telephone Tmnspcor Water Total *G4r3:bJdel Source: Bank staff elaboration based on MECON data. 65 of these resources going to the transport sector. The most important of these is the Transport Infrastructure Trust Fund, which is financed by a surcharge on diesel, and provides subsidies to inter-urban road concessionaires and metropolitan transport concessionaires. Other significant examples include the Patagonian Gas Trust Fund and the Electricity Transmission Trust Fund. Third, there are fiscal credits (creditosfiscales), of which the most significant is the diesel price differential for vehicles in the passenger transportation sector, amounting to $240m in 2002. Fiscal credits have also been used to finance social tariffs for electricity in the Province of Buenos Aires, as well as the peri- urban slums of the metropolitan area. 5.27 The use of extra-budgetary mechanisms for infrastructure finance has also proliferated at the provincial level. The Province of Buenos Aires alone has three provincial trust funds associated with infrastructure. Both the Electricity Transmission Infrastructure Trust Fund and the Provincial Electricity Tariff Compensation Trust Fund are financed by surcharges on electricity bills, which are used to finance respectively electricity transmission investments, and tariff shortfalls among distribution operators and cooperatives. Similarly, the Province of Misiones has a Public Works Trust Fund financed from 42 percent of the proceeds of the provincial turnover tax, which channels $42 million of resources to the finance and operation of public transport and solid waste collection systems by the private sector. As the next chapter shows, such extra-budgetary financing mechanisms tend to complicate macro-budgeting and undermine transparency. Subsidies and the poor 5.28 Another important trend of the past few years is the growing volume of subsidies to private infrastructure concessionaires (Table DIO), which have quadrupled from $422 million in 1999 to $1.9 billion in 2003 (Figure 7 (a)). This phenomenon is closely linked to the proliferation of trust funds-the proportion of infrastructure subsidies financed from extra-budgetary54 mechanisms has risen from 45 percent in 1999 to 98 percent in 2003. The other striking change is the shift in the relative importance of energy and transport subsidies. Transport subsidies, which accounted for 28 percent of the total in 1999, are projected to account for 69 percent of the total in 2003. 5.29 This growth is primarily attributable to increasing expenditure on existing subsidy schemes, rather than the creation of new subsidy schemes. Indeed, the only new subsidy introduced over this period is the one to metropolitan bus operators, amounting to $306m in 2002. It is striking that existing subsidy schemes have benefited from substantial increases in resources as the funding source was changed from general budget to extra- budgetary trust funds. For example, when financing of transport subsidies switched from budget to off-budget in 2001, average annual subsidy expenditure increased more than nine-fold for road concessions, and more than doubled for rail concessions. 54 Even though most of them are not recorded in the budget, they are outside the budget determination process in that legislator 66 5.30 Although these various subsidy schemes collectively absorb a substantial volume of resources, they are not well-targeted. This is particularly unfortunate as targeted subsidies would mitigate the social concerns raised by the renegotiation process and concomitant need to make tariff adjustments for essential services. About three quarters of the resources channeled to the existing subsidy schemes take the form of supply side subsidies, primarily to transport operators. These are equivalent to an implicit price reduction of $0.09 per bus passenger, $0.14 per metro-rail passenger, and $0.25 per suburban rail passenger. As these subsidies benefit all customers, with no particular attempt to target resources to the poor, the distribution of these resources is skewed towards the upper middle-income groups, who are the predominant users of these modes of transport (Figure 7 (b)). A recent study estimates concentration coefficients for subsidies received by metropolitan commuters to be regressive, ranging from 0.09 for suburban rail, 0.16 for bus, and 0.36 for metro-rail (Foster and CEER, 2003b). Figure 7: Distributional Incidence of Subsidy Schemes in Energy and Transport (a) Patagonian Gas Subsidy (b) Metropolitan Urban Transport Subsidy 35% $35 Average 35% $2.00 Average 30% V S - $30 montly 30°h weeldy 25% $25 subsidy 25% $1.50 (multi-modal) 20% - $20 - Percentage 20% - - _ _ _ _ _ ~~~~~of _ $ 1.00 -0-Percentage 15% $- 15 beneficiaries 15% * of 10% $10 10% - $0550 benefldarles , Percentage 5% $5 of subsidy -- Pementage 0% $0 ~~~~~~~~~~~0% $0.00 of su.bsidy SX l z 3 4 : g *ts_ SX hsomquirh1a2 3 4 1 2 3 4 5 2 3 4 5 ftorre quntas Incme quinhile Source: Adapted from CEER, 2002 and OPSM, 2002 5.31 Most of the remaining subsidy resources take the form of demand-side subsidies for energy consumption targeted to certain geographical areas, based on criteria that have little to do with poverty (Foster and CEER, 2003b). Thus, 92 percent of the resources of the Electricity Tariff Compensation Fund go to the smaller provinces of the interior which have only 44 percent of the nation's poor (Figure D6). Similarly, 100 percent of the resources of the Patagonian gas subsidy go to the far south, which has only 3 percent of the nation's poor. Moreover, the Patagonian gas subsidy was found to have a concentration coefficient of 0.08, indicating that its effect on income distribution is mildly regressive (Figure 7 (a)). Seen from a national perspective, the Patagonian gas subsidy excludes 95 percent of poor gas consumers in Argentina, while 63 percent of its beneficiaries are non- poor households. C. ENSURING SUSTAINABILITY OF INFRASTRUCTURE SERVICES 5.32 Even before the devaluation, IDEA (2001) estimated Argentina's potential to increase exports of cereals, petroleum, automobiles, timber, meat, fish, and produce by 2010 to be between 50 percent and 100 percent (IDEA, 2001). Following the devaluation, even higher export growth should be possible. However, this is dependent on substantial 67 investments throughout the transportation network, from enhancing the availability of all weather rural roads to support the extraction of agricultural produce, to increasing capacity on road, rail and river links along the major inter-urban corridors, to improving multi- modal access to major ports, as well as to enhancing terminal facilities to include cold chain capacity. Fay and Yepes (2003) find that Argentina needs to invest around US$5.2 billion per year during 2005-2010 to support projected GDP growth. This is approximately the level of investment that was achieved during the boom private investment period of 1993-1998. 5.33 Only the private sector will be able to raise such levels of financing. For this to occur, investor confidence in the country and in the infrastructure sector, in particular, needs to be restored. This, in turn, requires satisfactorily concluding the short-term renegotiation process and addressing certain second-generation reforms. The alternative of a return to public ownership and control is not a desirable option, given the higher efficiency in the private sector, the many other demands on fiscal resources, and problems with public provision of services. What Argentina needs is better design of private-public infrastructure provision arrangements, better regulatory arrangements, better governance, better targeting of subsidies, and better institutional arrangements for financing. Some suggestions on how to achieve this are given below. Framework for the Renegotiation Process 5.34 For concessionaires to be able to reschedule their debts they need to have an agreement for a longer-term framework under which tariffs will be determined. The renegotiation process to date has tended to focus mainly on ad hoc short-term measures. However, resolution of two longer-term issues is key. First, a new mechanism is needed for preserving the value of infrastructure tariffs over time. The US Consumer Price Index, whose appropriateness was questionable even in the fixed exchange rate regime, is even less appropriate now. However, the Emergency Law rules out any form of indexation, which is potentially a problem in terms of maintaining the financial sustainability of the service providers. The second-and more complex issue-is how to value the asset base of the regulated companies in a post-devaluation environment, which is a key issue in terms of gradually restoring the financial equilibrium of the concessionaires. 5.35 A positive recent step has been the greater involvement of regulatory agencies in the renegotiation process, albeit in an advisory role. In many-though not all-cases the renegotiation process could have been handled within the existing regulatory framework as an extraordinary tariff review. In fact, in the case of electricity distribution, an ordinary tariff review was already underway in January 2002, but was suspended following the Emergency Law. Although the official renegotiation period has been extended, it will be important to restore the normal functioning of the regulatory system as soon as possible thereafter, in order to avoid permanent damage to the credibility of the regulatory process. Redirecting subsidies to poor consumers can help towards this goal. 68 Redirecting Subsidies to the Poor 5.36 Throughout the 1990s, social concerns about infrastructure pricing have manifested themselves on numerous occasions, prompting the development of a wide variety of ad hoc social policy mechanisms for different sectors in different jurisdictions. This has created a patchwork of federal and provincial social policies for the infrastructure sector and engendered subsidy schemes that collectively absorb $940 million (Table D lI ). Although these resources are substantial, empirical evidence suggests that they are not effectively targeted towards the poor, and thus ultimately benefit only a relatively small set of interest groups (Foster and CEER, 2003b). This is particularly unfortunate given the urgent need to put in place a mechanism to safeguard the most vulnerable consumers from any tariff increases that result from the renegotiation process. The resources already spent on infrastructure social policies, could make a significant contribution to financing such a safety net, if they were appropriately redirected. 5.37 In the medium to long term, a comprehensive analysis and dialogue is needed on how best to offer social protection to the poor and vulnerable by asking what mix of social policies-income and food transfers, health, education, and infrastructure subsidies, unemployment and pension schemes-is appropriate for Argentina. 5.38 Presently, mainly because of the crisis, many infrastructure services have become unaffordable for many households. A household survey conducted in October 2002 found that expenditure on household utilities (namely, water, electricity, natural gas and telephony) has doubled as a share of family budgets relative to the situation that prevailed in 1997, when the last similar national expenditure survey was conducted. In 2002, utility expenditure accounted for 22 percent of the income of the first quintile and 16 percent of the income of the second quintile as opposed to below 15 percent in comparable countries (Foster and Araujo, 2001; Foster and Irusta, 2001). As additional tariff increases would further raise these income shares, the affordability issues takes on even more prominence. 5.39 At the same time, as seen earlier, there are sizeable subsidies in the sector that are presently largely failing to target the poor. It would be far more desirable to use them instead to promote access to water, sewerage, and natural gas services among the lowest income groups, where substantial coverage deficits still remain. In the case of water and sewerage, there is an evident public health case for seeking to achieve universal access in urban areas. In the case of natural gas, those who lack access suffer a significant economic disadvantage, given that the closest substitute, LPG cylinders, cost five times as much per effective unit of energy delivered. In both instances, the substantial connection charges and intra-household investments associated with connecting to the network represent a major barrier to access for low-income households. Therefore connection subsidies appear to be a particularly effective way of reaching genuinely poor households in Argentina. Simulations performed in a recent study, find that the concentration coefficient associated with connection subsidies are strongly progressive, with values ranging from -0.30 to - 0.40, compared with only -.10 to -0.20 for even well-targeted consumption subsidies (Foster and CEER, 2003b). 69 5.40 For bus and rail, an income transfer is a better alternative but the Government may consider providing a per trip subsidy to areas serving the poor rather than maintaining the current per passenger subsidy. In the longer-term, it would be desirable to address the broader issue of taxes and subsidies in the sector. Similarly, while there is little economic justification for a subsidy on electricity, this sector is taxed heavily-in part to subsidize low income users and depressed areas. However, the taxes are not used for this purpose. Moreover, pressures to keep the tariffs low for the poor are hampering renegotiations with power operators. Thus there may be merit in directing the electricity surcharges to temporarily freeze rates for those who consume below a minimum threshold even as rates for other users increase. 5.41 Estimating the cost of such a policy is instructive. Assuming that the policy objective would be to keep the cost of subsistence water, sewerage, electricity, and gas bills for poor households below 15 percent of income, the total cost of this policy would be about $300 million per year at current tariff levels. If all tariffs were to increase by 10 percent, the cost of the policy would escalate to $ 386 million, while a 30 percent tariff increase would generate a cost of $ 559 million. Reforming the Regulatory Framework 5.42 The present economic crisis provides an important opportunity to initiate a thorough public review of the functioning of the regulatory framework, with a view to building consensus around a package of second-generation reforms designed to address the deficiencies identified above. The widely differing performance of current regulatory entities across infrastructure sectors demonstrates that sound legal and institutional design make a real difference to subsequent performance, and that the development of sound regulatory agencies is possible within the Argentine political environment. 5.43 The solidity of the legal and institutional framework for regulation varies greatly across agencies at both the federal and provincial levels (Tables D13 and D14). The two federal energy regulators-ENRE and ENARGAS-stand out as best practice regulatory institutions in a number of respects (Table 11). First, the national regulatory agencies for electricity and gas were created by means of sector laws. This stands in contrast to all of the other Federal regulatory agencies for water, telecom, and transport, which were all of them established by Decree. At the provincial level, there appears to be a greater tendency to establish regulatory agencies by law rather than decree as evidenced by the cases of Buenos Aires, C6rdoba, and Misiones. The advantage of a sector law is that it provides a clear basis for the authority of the regulatory agency, and limits the extent of ad hoc political interventions. 5.44 Second, in common with most other agencies, the energy regulators are governed by a Commission of five to six members. In most cases they are appointed by the highest level of the executive branch of government. However, in the case of the energy sector, the candidates must be selected by competition and approved by Congress, which helps ensure a somewhat greater degree of political independence than in other sectors. 70 Table 10: Evaluation of Argentine Regulatory Framework against International Best Practice International best Argentine electricity and gas Water, telecom and transport practice Regulatory framework Based on electricity and gas sector Established through a variety of established by law laws Executive Decrees Autonomous governing Governing body appointed by Governing body appointed directly body with high technical President, following a competitive by President capacity process and with Congressional approval Financially autonomous Financed by regulatory surcharges, Financed by regulatory surcharges, but publicly accountable with budget subject to Congressional but without budgetary approval accountability Consumer participation in Major decisions are consulted by No formal channels exist regulatory processes public audiences, however not very (except for the water sector in effective in practice GBA) 5.45 Third, while most agencies are financed through sectoral regulatory levies, the energy sector is the only one where agency budgets must nonetheless be formally approved and executed through the budget process. This helps to strengthen accountability and avoids the conflicts of interest that can arise when the size of regulators' budgets depend directly on the level of tariffs which they themselves determine. 5.46 A further institutional problem that affects all of the regulated sectors, both in Argentina and throughout Latin America, has been the very limited involvement of consumer representatives in the regulatory process. Consumer participation is universally problematic due to the free-rider problem. However, in developing countries this problem is exacerbated by the absence of well-developed civil society institutions to represent the consumer perspective. Moreover, the regulatory framework provides only limited channels for this participation to take place. Although some sectors (notably, electricity and gas) provide for public audiences, this mechanism has not functioned very effectively in practice. An interesting counter-example comes from ETOSS, the water regulator in GBA, which has recently introduced a consumers committee that acts as a consultative body for the agency. 5.47 Finally, the focus on regulatory reform in the 1990s privatization program often meant that the strategic policy-making function of the State was debilitated or overlooked. This phenomenon-which is quite commonplace across Latin America-has resulted in a complete vacuum as regards long term planning and the determination of strategic objectives for the infrastructure sectors. The Argentine transport sector provides a particularly clear example of this problem. It would be useful if an official vision for the sector were developed to take advantage of any synergies and complementarities between, for example, federal and provincial infrastructure, or across different modes of transport. 71 Reforming Governance of Extra-Budgetary Mechanisms 5.48 From a macroeconomic perspective, special financing mechanisms are regarded as problematic because they restrict the optimal allocation of scarce public resources among competing social goals. They may also reduce transparency. Nevertheless, there are a number of microeconomic arguments that can sometimes justify their use. First, they provide a means of ensuring that users pay for the services that they receive, for example, from the road network. Second, they help to stabilize the flow of resources for maintenance expenditure, which may be particularly vulnerable to cutbacks during fiscal downturns even though neglected maintenance typically results in much higher rehabilitation costs and inconvenience to users. Third, where there is public-private co- financing of infrastructure investments, a trust fund mechanism can provide additional comfort to private investors that public funds will be forthcoming, thereby reducing the perceived risk of participation. Fourth, the creation of an autonomous source of funding can improve managerial incentives by increasing autonomy from unwarranted political interference. As a result, they have been quite widely advocated and applied in the roads sector with some success. 5.49 Therefore, to the extent that the Argentine special financing mechanisms focus on funding public good components of the infrastructure network (such as roads and electricity transmission), or socially motivated subsidies to privately operated infrastructure services (such as energy subsidies or public transport subsidies), there appears to be a conceptual case for their use. However, the micro-economic case for extra- budgetary mechanisms is contingent on the existence of suitable governance arrangements to ensure that funds are effectively protected and used. The experience of trust funds in Argentina reveals a number of governance problems, which in many cases are so severe that they completely undermine any conceptual advantages that these mechanisms may offer. 5.50 If extra-budgetary mechanisms are to continue to play an important role in the Argentine infrastructure sectors they will need to be radically overhauled. This is particularly true of the Transport Infrastructure Trust Fund, which is by far the largest of the extra-budgetary mechanisms, and also presents the most glaring deficiencies in its design (Table 11). As elaborated upon below, areas for improvement include strengthening governance structures, better safeguarding of resources for their intended purposes, more transparent and better rules governing the allocation of funds, and mechanisms to ensure accountability (Table D15). 5.51 First, there is a need to develop autonomous governing bodies with substantial representation from beneficiary groups, such as the private sector and civil society. At present, the two largest funds (Transport and Hydraulic Infrastructure Trust Funds) have no autonomy whatsoever being administered directly by a unit within the Ministry of Economy. Some of the energy funds have somewhat broader based representation with committees combining members from both Federal and Provincial tiers of government. However, only the governing body for the Regional Infrastructure Development Fund (Fondo Fiduciario de Infraestructura Regional) has some degree of non-governmental 72 Table 11: Evaluation of Argentine Transport Trust Fund Against International Best Practice Design feature International best practice Argentine transport trust fund Legal By separate legislation By Decree 976/01, with subsequent establishment modifications Governance Autonomous entity with majority road Direct government administration without arrangements user representation on Board any road user representation, initially within Ministry of Infrastructure later transferred to Ministry of Economy Relation to Outside Government Budget Officially extra-budgetary, but in practice public budget increasingly integrated within Government Budget Allocation Transparent and established rules for Allocation rules modified five times in the Rules allocation between agencies first year of operation (see Table D17 for further details) Accountability Published financial regulations and Accounts must be submitted to Ministry of procedures supported by regular Economy, National Auditor currently technical and financial audits considering need for further measures. Source: For international best practice, Gwilliam and Shalizi (1999) and Potter (1997) representation, albeit only a single optional member from the Argentine Chamber of Construction. 5.52 Second, there is a need to strengthen the safeguard measures on trust fund resources, both by improving the legal framework and by channeling resources directly to commercial banks. At present, many of the smaller trust funds are legally constituted, however the two largest funds were created by means of Executive Decrees. This combined with their weak governance structure makes them particularly vulnerable to modification and political interference, including the recapture of their resources for the general budget. For example, the 2002 Budget Law responded to a budget surplus in both the Transport and Hydraulic Infrastructure Trust Funds by requiring a non-refundable transfer to the Treasury of $100 million and $70m respectively. Despite appeals, the transfers were completed. A further illustration of this phenomenon is provided by Law 25400 of 2001, which allowed Provinces to divert up to 50 percent of the resources received from the National Electricity Fund to general expenditures but which was arbitrarily raised to 100 percent via Law 25570 of 2002. 5.53 What these incidents reveal is the cyclical political economy dynamics of these special financing mechanisms. These arise initially as a means of providing stable financing for long-term infrastructure needs. However, as fiscal pressures increase, the Treasury finds creative ways of recapturing these resources for general budgetary needs; a process that is made all the easier by the institutional weaknesses in the governance of these special funds, as noted above. Nowhere is the cyclical nature of this phenomenon more evident than in the history of road funds in Argentina, which shows how the principle of earmarking resources for a national road fund was established and successfully applied from the 1930s to 1960s, but gradually eroded and was ultimately abolished over the next three decades, only to be re-established in 2000 (Table DI 6). 73 5.54 Third, there is a need to improve the transparency with which resources are allocated to different purposes. Although all of the funds have clearly defined rules governing the allocation of their resources, in many cases these rules have become subject to discretionary changes over time. The most flagrant example is the Transport Infrastructure Trust Fund, which had its allocation rules modified five times in the space of 12 months, by a series of Decrees and Ministerial Resolutions (Table D17). As a result, this fund-originally intended to finance new road investments-was first converted into a scheme for subsidizing existing road concessions, and then gradually diverted into a source of finance for rail and urban transport services (Figure D7). Similarly, the Patagonian gas subsidy was recently extended to cover consumers in the Puna region of the northwest, even though the original trust fund was designed to subsidize only gas consumers in the extreme south Patagonian region. 5.55 Fourth, accountability needs to be strengthened by creating public oversight mechanisms to monitor the overall need for resources to meet the funds policy objectives, as well as their efficient and appropriate use, and to design specific rules for dealing with financing shortfalls or surpluses. Since their inception in 1994, there have been a number of moves to improve the transparency and accountability of Trust Funds. Law 25152 of 1999 obliged Trust Funds to be incorporated into the General Budget, which was done for the first time in 2001. They are also required to submit accounting data to the Ministry of Economy on a regular basis. The Budget Law of 2002, further modified the legal framework so that Trust Funds are explicitly considered as public sector agencies, akin to SOEs. However, these reallocations and diversions illustrate that the current accountability mechanisms do not appear to be functioning very adequately. 5.56 In conclusion, the performance of Argentina's infrastructure sector improved substantially following its transformation in the 1990s. Nonetheless, equity concerns remain because of weaknesses in the regulatory frameworks put in place to protect the public interest. The devaluation of the peso and subsequent tariff freeze has completely undermined the financial equilibrium of the sector. The way forward will entail laying down clear principles for short and long-term tariff adjustments while developing a social tariff mechanism that protects the most vulnerable. Fiduciary funds provide a financial basis for such a social policy, however their targeting procedures and governance mechanisms need radical overhaul to fulfill this role. Finally, the crisis provides a valuable opportunity to evaluate the functioning of the new framework established in the 1990s, with a view to identifying key second generation reforms, such as improved governance of regulatory agencies (bringing water, telecom and transport up to the level of the energy sectors) and a greater role for consumers in the regulatory process. 74 6. MANAGING PUBLIC EXPENDITURES A. INTRODUCTION 6.1 A good public expenditure management system provides a government the incentive and means to promote fiscal discipline and efficiency. While the system. cannot substitute for good governance, fiscal institutions can be reformed to encourage good governance and to raise the costs of poor governance. Chapter 3 suggested how reforms in the revenue-sharing system and the issuance of fiscal rules could improve. fiscal discipline and generate incentives for good governance and efficiency. Chapters 4 and 5 suggested how reforms in budgetary institutions could promote fiscal discipline and efficiency. This chapter focuses on how changes in the budgetary processes and framework can help achieve these objectives. 6.2 The Federal Government has two budget processes, one more than a well-managed country should have. In law, it has a single process that covers the entire public sector; in practice, it also has a hodge-podge of exceptions and special arrangements that undermine fiscal discipline and efficiency. The formal system has regularized preparation, approval, and execution of the annual budget, and is anchored by an integrated financial management system, SIDIF, that holds spending within authorized amounts. Its internal and external controls have enhanced transparency. 6.3 The formal system does not, however, provide a full account of public expenditure. The discrepancy between the reported budget deficit and the buildup in public debt (see Chapter 2) is powerful evidence that significant fiscal activity occurs outside the formal budget process. Many of these are recognized in debt consolidation laws that retroactively assume pre-existing explicit or implicit obligations. The formal budget system also is impaired by earmarks, off-budget accounts, emergency decrees, and other practices that rigidify budget allocations and weaken financial control. The coexistence of effective formal controls and escapes from control provides a strong incentive to migrate from the formal system to informal arrangements. 6.4 This next section provides an assessment of the formal budget process, concentrating on deficiencies that deter the Government from using the budget to set a prudent fiscal course and to allocate public resources in response to changing national priorities. The subsequent section then suggests various reforms to rationalize public expenditure management. The key is to recreate a single, comprehensive process that fully accounts for all spending and provides a complete and accurate account of deficits and debt. 75 B. THE FORMAL AND ALTERNATIVE BUDGET PROCESSES The Budget Control Mechanism 6.5 Formal budgeting begins with preparation of the annual budget, which is largely a top-down process managed by the National Budget Office within MECON. Once approved by Congress, the budget is implemented through a series of controls designed to ensure that actual spending is within authorized levels. Early in the annual budget cycle, NBO assigns spending targets to each ministry in consultation with the Office of Chief of Cabinet (Jefatura). The targets are imposed unilaterally, without much input from spending units and are based primarily on macroeconomic projections, including preliminary revenue forecasts for the year ahead. Pursuant to the targets, spending units submit budget requests, often in excess of the assigned limits. Negotiations then begin around the amounts spending units need to operate the next year. 6.6 Major policy initiatives tend to be taken outside the confines of the budget process. They may occur anytime during the year, either through Govemment decision or legislation by Congress. In fact, Congress' main influence on spending is through substantive legislation, not through the annual budget law. From the time it receives the budget (in mid September) until final approval (December), Congress makes many adjustments, but most are minor and all are subject to line-item veto by the President. Congress generally influences budgetary policy through substantive legislation, by passing laws dealing with pensions, health cars, taxation, and other issues. In both the Government and Congress, budgeting is regarded more as a means of financing ongoing activities than as an opportunity to define national priorities and policies. 6.7 Budget execution begins with publication of an expenditure distribution decree specifying the amounts allocated to each spending unit. The decree is followed by quarterly quotas limiting the commitments that spending units may incur and by monthly limits on the payments that may be processed through SIDIF, the Government's main integrated financial management system. These quotas take account of revenue collections and the Government's cash position. For most ministries, actual spending in recent fiscal years has ranged between 75-95 percent of budgeted expenditures (Table El). Spending units know their operating budgets will be cut but they do not know the size of the cuts and therefore lack a reasonable measure of certainty as to the resources that will actually be available. 6.8 The cashbox budgeting adverse side effects: (a) It boosts the cost of goods and services purchased by spending units-some suppliers explicitly charge differing prices according to when they are paid; others simply build expected delays into the prices they charge. (b) Budget staff in spending units spend an inordinate amount of time juggling their books, negotiating with suppliers, and pressuring NBO for more money. (c) It gives rise to devious behavior, such as hiding liabilities and building up arrears-in 2002, one ministry had 100 million pesos of unreported liabilities; another had 400 million of valid liabilities that had not been entered into SIDIF because the spending unit has exhausted its quota. (d) The quotas reinforce the tendency to regard budgeting as a means of processing operational expenses rather than as a means of charting the course of government-the 76 payments quota of some ministries is exhausted by the 10th of the month. (e) Transparency is impaired when the approved budget is not the one implemented by the Government-politicians have an incentive to vote more spending than the Government can afford, and managers have an incentive to misclassify transactions. 6.9 The Government's expenditure management process is a minimalist application of the procedures prescribed in law. Ministries carry out the terms of the law, but their actions have little relevance for how public funds are actually allocated and spent. Ministries prepare strategic plans but rarely use them in preparing budget requests. They prepare a 3-year budget, but hardly any consideration is given to the amounts specified for the final two years and these documents are sequestered by NBO. Ministry budgets are classified by programs, but these are sterile classifications that are not connected to government objectives and are not used in trading off among activities serving the same purpose. Spending units compile data on performance, but in most cases these are workload measures; they do not report the results ensuing from government action. Expenditures are supposedly on an accrual basis, but the concept of accruals apparently refers to payments, not to liabilities. Escaping Budgetary Control 6.10 Strict control of discretionary expenditure provides incentives for escape in order to obtain favorable treatment and protect their programs. Escape routes include earmarking money for designated purposes, establishing trust funds, giving agencies financial autonomy, removing accounts from the budget, and use of emergency decrees. These significantly weaken the government's fiscal position, rigidify budget allocation, and force greater stringency on the discretionary portion of the budget. Only about 12 percent of annual expenditures are deemed discretionary (Table E2). 6.11 Despite provisions in the 1992 Law on Financial Administration that discourage earmarking, the practice still thrives, and the portion of national expenditure governed by annual budget decisions has continued to decline. For example, AFIP, the national tax collection agency, -is entitled to retain a fixed percentage of the money it takes in each year. Additional amounts are siphoned to social security funds and other social purposes, for investment in -infrastructure, and for various services provided by the provinces. 6.12 The Government has constitutional power to authorize additional expenditure during the year by emergency decree, often without congressional approval (Table E3). Such decrees are sometimes issued after the year has ended to legitimize spending that has already occurred. Recourse to decrees became more prevalent during the crisis when the peso collapse impelled the Government to funnel additional resources to certain activities. Use of decrees compromises the transparency of the budget and may open the door to abuse. In contrast to the regular budget which receives considerable public notice and is reviewed by Congress, the decrees (although published) get much less attention. The real budget emerges during execution and may be known only after the fiscal year has ended. 77 Budgeting for Fiscal Conunitments and Risks 6.13 The problems discussed earlier in the chapter are within the formal budget system. In terms of fiscal discipline and effective use of public resources, a greater risk arises out of the lack of budgetary capacity to control liabilities stemming from fiscal commitments and risks (See Chapter 2). The budget is typically ignored when these liabilities materialize, and the budget is helpless when the government recognizes the liability. What is especially troublesome is that many of the liabilities are implicit; they are not legal obligations but the Government nevertheless has felt compelled to assume them. For example, the federal government had no legal obligation to finance provincial debt or pay depositors in excess of $30,000 when banks failed, but it was pressured by political circumstances to do so. Such actions not only expose the Government to moral hazard but also open it to lobbying by powerful interests. 6.14 Such liabilities have often been recognized through debt consolidation legislation authorizing the issuance of bonds for specified obligations. In the past, this legislation was a freestanding bill; recently, it has been incorporated into the annual budget law. Doing so gives the Government a more complete view of its debt, but does not bolster the budget's capacity to regulate these liabilities. The reason is that the debt consolidation process recognizes liabilities only when they are to be consolidated, not before then. For example, debt consolidation assumed unfunded liabilities of the pension system that previously were hidden in pay-as-you-go rules. Similarly, the Government acknowledged an obligation for the 13 percent pay cut imposed on federal employees only in the debt consolidation act. 6.15 In addition to its own liabilities, the federal government sometimes finances the liabilities incurred by provinces. Following the devaluation, provinces were unable to pay their dollar-denominated debt and the federal government came under pressure to assume provincial debts but in return demanded provinces to repay the debt, restore fiscal discipline, and improve financial management via bilateral agreements (Chapter 2). The agreements establish a monitoring mechanism to oversee compliance with these terms, but the inadequate state of provincial financial information systems undermines this comnritment. Many provinces lack timely, accurate financial reporting systems, and many do not properly record accrued liabilities. 6.16 The entanglement of federal and provincial finances arises out of a dysfunctional relationship between the two levels of government. Although they have constitutional independence, few provinces are fiscally independent. They are independent in incurring liabilities, dependent in getting the nation to finance their debt. They are independent in spending money, utterly dependent on the nation in generating revenue. They actively lobby the federal government and Congress for more money, finding this an easier path than spending less or taxing more. Only a reform of the federal-provincial revenue sharing arrangement can overcome this imbalance (Chapter 3). 78 B. MODERNIZING PUBLIC EXPENDITURE MANAGEMENT 6.17 In order to improve Argentina's system of public expenditure management, improvements are needed in three broad areas: * Macro-budgeting, which focuses on aggregate expenditure, especially the balance between total revenue and spending, fiscal sustainability, and the government's capacity to set and implement budget constraints * Budget allocation process, which pertains to effective use of the budget to promote the country's economic and social development and the government's capacity to shift resources from less to more effective programs and from lower to higher priorities * Service efficiency and performance, which deals with the cost, quality, and accessibility of public services and efficiency in managing public activities While the reforms are not difficult to define, their implementation is not easy. Also, it is not possible to implement all of them simultaneously. While some can be implemented right away, others will need to wait for strong economic recovery. Macro-budgetary reforms are a priority. Macro-budgeting 6.18 As mentioned in Chapter 3, fiscal rules can help institutionalize the success For Box 6: Sweden's Two-Stage Budget Process Frmany years, Sweden had a chronic budget deficit, of the PFOs. The fiscal rules discussed along with rising tax burdens and spending levels that the need to prepare medium-term fiscal ranked among the highest in the world. In the early targets. This can be done within a two- 1990s, the country's fiscal deficit surged to 13 percent stage budget process. The first stage of GDP, placing it 12h among 13 EC member countries would establish a macro-econoriLc on a scale measuring fiscal discipline. Sweden reformed its budget process in 1996, which allowed it framework; the second would entail the to advance to 3rd place. preparation of annual budgets consistent with the previously agreed totals. The The main element of Sweden's reform is a two-stage process. In the spring, the Government establishes a two stages would be several months medium-term expenditure ceiling for each of the next apart, the first in the April-June period three years. The ceiling is expressed in nominal that is presently the start of the budget amounts and is subdivided among 27 major expenditure preparation cycle, the second in July- categories. The first stage concludes with adoption of September. The concept of a two-stage the spending ceilings by Parliament. budget process was successfully In the second stage, the Government prepares an annual introduced by Sweden in the mid 1990's budget and the legislature votes appropriations. Both in response tofiancathe budget and appropriations must be within the in response to a financial crisis that approved spending ceilings, as well as within the produced a fiscal deficit of 13 percent of amount allocated for each of the 27 spending areas. In GDP (Box 6). Parliament, responsibility for appropriations is assigned to various sectoral committees, each with its own 6.19 The fiscal targets must be made ceiling. Budget amendments that would increase public and voted by Congress. Internal spending must also be within the ceilings. These likely tolakpoot explicit, enforced spending limits contributed greatly to targets are hkely to lack poltical suppor Sweden's fiscal success. and credibility, and may not be able to 79 withstand changes in economic conditions or political sentiment. Public targets would strengthen the linkage of fiscal goals and macroeconomic projections. Ideally, the aggregates should be set for a 3- or 4-year timeframe, and rolled forward each year as a new budget cycle commences. Publishing 3-year frameworks, having them reviewed by Congress, and constraining budgeted and actual expenditure to the amounts accommodated within the framework would give medium-term budgets much greater prominence than they now have and thereby encourage greater care in their preparation. The multi-year budgets could be building blocks for a future medium-term expenditure framework (MTEF), but they may lack critical features that take time to develop, such as reliable forward estimates (baselines), the costing of proposed and adopted policy changes, and capacity in spending units to tradeoff among program options. 6.20 Under this proposal, once the framework has been approved, the Government would be barred from submitting a budget that breached the aggregates and Congress would be precluded from considering any amendment that would cause the aggregates to be exceeded. If, as envisioned here, the framework adopted in the first stage allocates amounts among key sectors, both the Government and Congress' budget would have to be within the sectoral limits. Modifications to the budget during the year by emergency decree or other means also would have to be consistent with the agreed framework. Ideally, the framework should set aside a reserve for contingencies that would be drawn upon to finance emergencies and policy initiatives within the agreed aggregates. Targeting a structural primary surplus will make it easier for the Government to stick to its commitments during an economic downturn or external shocks (Chapter 3). 6.21 A framework is useful only if it is comprehensive. Thus the budget framework should consider having financial and other statements relating to off budget accounts, trust funds, arrears, guarantees, fiscal commitments, and other contingent liabilities that may be issued or called during the next few years. The framework should not only account for liabilities that have already been incurred and that may become due or be financed, but be an instrument for regulating the volume of new liabilities to be incurred. Thus it could include the following entries: * The maximum loss and/or the expected loss stemming from the issuance of new guarantees to be issued, by sector * The total guarantees outstanding, by sector * Floating debt and other arrears * Other fiscal commitments or contingent liabilities * The amount, maturity and terms of types of debt to be consolidated * A reserve for calls on guarantees and other liabilities 6.22 The framework would relate to fiscal commitments and risks in the same way that it does to direct spending. It would have procedural and substantive rules similar to those applied to direct expenditures. In practice, spending units would submit estimates on liabilities at the same time they do so for direct expenditures explaining each liability, its legal basis, the estimated cost to government, and the economic or social justification for 80 the liability, options available to the Government, and so on. The Government could develop standards that discourage the issuance of contingent liabilities as substitutes for direct expenditure and require objective costing of risk. Ideally, these standards would encourage risk sharing by use of co-payments, deductibles, and origination fees, as well as procedures for enabling the Government to recover losses. 6.23 Annex 4 provides more detail on a financial and accounting framework for budgeting for contingent liabilities, fiscal commitments and other fiscal risks and provides examples of what some countries are doing. Rather than considering a sophisticated asset- liability management framework,55 it would be preferable for the Government to consider compiling an annual statement of all known explicit contingent liabilities, specifying its legal obligation with regard to each entry, and estimating the probability of loss and the amounts that may come due in the next 3 years. A simple approach to calculating the loss may be adequate. Under the simple approach the statement would classify the liabilities as high, medium,. or low risks, and to assign a fixed loss estimate to each category. For example, high risks might be given a 90 percent probability of loss, meaning that the expected cost would be that percentage of the maximum loss. Medium risks may be assigned a 50 percent probability of loss, and low risks 10 percent. When maximum losses can be very large, it may also be wise to assign expected losses to these even if the risk is very low. The idea is to ensure that the Government takes account of risk both when it issues guarantees and when it budgets for outstanding contingent liabilities. Several other countries incorporate fiscal risks in their financial statements and decision-making process. Box 7: Alternative Ways of Budgeting for Contingent Liabilities South Africa Government prepares an inventory of fiscal risks and estimates possible budget exposure. New Zealand Contingent liabilities are listed in supplemental statement to audited financial statements. Hungary Budget sets aside funds for projected calls on guarantees during the fiscal year. Canada Proposed loans and guarantees are reviewed during budget preparation. Netherlands Budget treats guarantees as if they were direct expenditure. United States Estimated subsidy cost of loans and loan guarantees is budgeted as an expenditure. 6.24 Some of the fiscal risks are implicit, which makes the matter more difficult. On the one hand, accounting for them would spur moral hazard and boost the cost to government; on the other hand, failure to recognize them would lead the Government to underestimate the liabilities it faces and face fiscal surprises. Perhaps the best counsel is for government to account for implicit liabilities only when (a) there already is high expectation that it will make good on them, and (b) there is low probability that affected interests will behave in ways that increase the cost to government. 6.25 In any case, the Government may wish to exercise due prudence in taking on implicit or explicit contingent liabilities. While guarantees and other contingent liabilities are sometimes appropriate instruments of public policy, overusing them or using them without proper consideration of downstream costs will expose the Government to fiscal turbulence in the future. The Government does not have sufficient resources to serve as 55 See Irwin et. al., 1999. 8 1 payer of last resort for Argentina. It would be wise to regulate its exposure to fiscal risks in the same way it regulates its commitment to direct expenditures. Establishing a parallel budget process for fiscal risks and integrating it with direct expenditures in the "framework" stage would go a long way to bring discipline to this area of public finance. The publication of the 3-year budget framework would also help to take into account the fiscal commitments in a more transparent way. 6.26 In addition to being integrated in the budget framework, liabilities and direct expenditures could be consolidated in financial statements, including a balance sheet that is reconciled to the budget and clearly crosswalks changes in net assets to the actual budget deficit. The statements would explain any variance between the deficit and the change in public debt. They would be supplemented with statements of contingent liabilities, cash flows, and other relevant financial data. Although the Government may have difficulty compiling a balance sheet, this exercise is a vital part of restoring fiscal health, because financial statements provide a fuller account of the Government's financial condition than does the budget. 6.27 At the outset, the balance sheet could be restricted to financial assets and liabilities. The aim should be to provide as true and complete an account of government finance as available data allow. Inasmuch as many of the Government's fiscal difficulties stem from unreported or unfounded liabilities, it is desirable that all such claims on the State be accrued. Thus, even if revenues are accounted on a cash basis, expenditure arrears should be accrued. This double standard may present a slightly more adverse account of government finances than is warranted, but in view of past failures to recognize liabilities, the greater likelihood is that the financial statements will still report a somewhat more favorable picture than is actually the case. 6.28 Because of issues arising out of privatization, approval of financial statements by Congress' Public Accounts Committee has been delayed for a number of years. It is urgent that financial statements not be viewed as simply a technical matter, but as a process that illuminates the Government's financial transactions and condition. Timely review and approval of financial statements is essential. This review should be view an opportunity to assess the Government's finances, to take stock of the liabilities and risks that have accumulated, to glean whether conditions have improved or worsened over the recent past and why, and to test the reliability of the financial reporting system. Accomplishing these ends depends on building up the role and resources of the Auditor, and using congressional review as a means for calling attention to material weaknesses in the Government's financial information system and exposure to risk in public finance. 6.29 At least at the federal level, it would be appropriate for the Auditor to produce an annual risk assessment, beginning with financial risk to the State and expanding over time to other vulnerabilities. In contrast to the Government, which must be circumspect concerning implicit liabilities, the Auditor could include them when there is substantial probability that they may impose costs on the State. Formal financial risk assessment is a useful instrument in sensitizing the Government to hidden liabilities and in spurring it to limit its exposure. 82 6.30 Provinces should also consider the two-stage budget process. However, the strengthening of financial reporting is a greater priority. It would be highly desirable for each province to produce annual financial statements along the lines of the federal government. This needs to be in a format that facilitates consolidation with the national reports, clearing tracking the inter-governmental flows. For example, if the federal government has an asset in the form of debt owed to it by provinces, this debt should appear in the same amount as a liability on provincial statements. Larger provinces may also wish to consider carrying out risk assessments as suggested above. Budgeting Allocation Process 6.31 As in many countries, Argentina's public expenditures are extremely rigid. As a consequence, budget decisions concentrate on the small portion that is regarded as discretionary, which often tends to be investment expenditure and certain discretionary transfers to the provinces. It would be desirable for to make room in its budget for policy initiatives by injecting strategic consideration of development needs and opportunities into spending decisions and by encouraging reallocation from old to new budget uses. This is a difficult task requiring restructuring the budget process. 6.32 Orienting the budget to policy initiatives and national priorities requires that an entity be vested with this responsibility and that it has sufficient power to influence budget allocations. Enlarging MECON's scope would be an option, but it is likely to have difficulty reorienting itself from financial control to policy issues, and doing so may not be an advantage. A more attractive option would be to remake the Office of the Chief of Cabinet (Jefetura) into the lead agency managing the policy process, coordinating programs across ministries, and reviewing budget allocations to assess whether they are in accord with national policies. However, the detailed budget formulation and execution should remain in MECON-past efforts to move this to Jefatura were not fruitful. 6.33 The 1994 constitutional reform that established Jefatura charged it with harmonizing governmental procedures and policies and with preparing and submitting the budget to Congress. However, Jefetura does not have the technical expertise to prepare the budget and relies on MECON to do the technical work. Some have suggested that Jefetura should be entrusted with lead responsibility for overseeing execution of the budget, presumably because it is at this stage that actual expenditures are set, especially in tight fiscal times. However, Jefatura is better suited for policy coordination that for technical budget work. Assigning Jefatura the lead role in policy development will increase the probability that this vital perspective is considered at budget time. 6.34 To be budget relevant, Jefatura could provide policy guidance to ministries during the framework stage, before ministries prepare detailed budgets. Inasmuch as aggregate budget decisions are taken at this stage, Jefatura will have to issue guidance that is consistent with the framework. Equally important, at this stage, Jefatura could actively explore opportunities for reallocation that would make room within the framework for policy initiatives. Working with MECON, Jefatura could estimate the medium-term budget impact of policy changes to assure that they can be accornmodated within the framework. Even without an MTEF, this should make the budget more policy-relevant 83 while maintaining fiscal responsibility by separating the framework stage from detailed budgeting, extending the horizon to 3 years, and calculating the prospective budget impact of policy changes. Proceeding in this manner will enable the Government to move to a full MTEF later in the decade. 6.35 Enabling the government to allocate money in accord with its priorities requires that the budget have uncommitted resources and that spenders have an incentive to take funds from some programs and spend it on others. To make room in the budget for policy initiatives, it is necessary that expenditures become more pliable through reforms that curtail earmarking, trust funds, preferential treatment of some agencies, off-budget accounts, and other practices that remove portions of the budget from effective financial control. It is also necessary to rationalize fiscal relations with the provinces as suggested in Chapter 3 and to curb the practice of retroactive financing of liabilities. 6.36 There cannot be significant new allocation without significant reallocation. There is no prospect of mobilizing resources for major program innovations unless funds are shifted from existing programs to new uses. Managing reallocation has the potential to stir up political conflict, and must be handled sensitively, with due regard to the interests affected. While it may be tempting to impose reallocation from the center-through Jefatura's policy work or MECON's budget constraints-it is highly probable that not much reallocation will occur if it is promoted by the center and resisted by everybody else. Sectoral ministers often prevail against efforts to strip their programs of resources. 6.37 It is sensible, therefore to make budgeting into more of a participatory process that gives sectoral ministries incentives to use resources more effectively. The incentive that matters most is that ministries proposing to shift money to more productive use not penalized by having their budgets cut. One way to encourage reallocation is to give each ministry a resource envelope in the framework stage of the budget process, and allow it to alter the mix of programs when it prepares detailed estimates in the second state of the process, provided that it keeps within the limit specified in the envelope. Each ministry could be encouraged to proposed reallocations with reasonable confidence that its budget would not be cut below the total budgeted for it in the framework. Jefatura and MECON would review proposed reallocations, Jefatura to assure congruence with government priorities, the latter to assure compliance with budget constraints. As this arrangement evolves, Jefatura would improve its capacity in policy development and MECON would improve its capacity in estimating the budget impact of policy changes. These capabilities will pave the way for successful implementation of an MTEF in the years ahead. 6.38 Regardless of the budgetary flexibility offered spending units, it will be difficult to curtail rigid expenditures through ordinary political means. Those who benefit from earmarks, trust funds and other preferential arrangements will fight to protect their interests. Given their superior information and ties to interest groups, legislators and provincial leaders, they often have the upper hand. To ease the political task of reallocation, the Government should consider use of special procedures that facilitate reform. One idea is to have a special commission propose changes in these arrangements, and to package the proposal in an omnibus measure that is considered in a timely manner by Congress. This will not disturb the constitutional structure, but might bypass some 84 roadblocks to reform. Extraordinary measures would have to be negotiated, bilaterally with affected provinces, collegially with congressional leaders. Such measures cannot be applied on an ongoing basis, however. The objective would be to de-rigidify the budget and then revert to ordinary procedures. 6.39 Achieving better fiscal correspondence in federal-provincial fiscal relations, as suggested in Chapter 3, would help control one source of leakage. In the current relationship between the nation and the provinces, the Government's capacity to improve program effectiveness depends on information supplied by provinces on their use of discretionary transfers. The federal government often is inadequately informed on how funds earmarked to provinces for particular purposes have actually been spent. While reforms aimed at better fiscal correspondence on federal-provincial relations will help, this problem will not vanish since the federal government still will prescribe policy for many services delivered by the provinces and it will continue to supply funds for certain activities. Greater uniformity in financial and program reporting by provinces should help reduce the problem. The reporting requirements specified in the bilateral orderly financing agreements should be expanded beyond data on the provinces' fiscal condition to information on their use of resources flowing to them from the national budget 6.40 Budgetary rigidity is not unique to Argentina. Most countries have difficulty shifting resources from less to more effective programs; in most, old priorities have an advantage over new ones. Argentina's situation is similar to many others that have urgent development needs, an inadequate revenue base, and a need to restore confidence in domestic financial institutions. What is unique to Argentina is the extent to which the economic crisis has impoverished large numbers of people who had attained or aspired to a middle class economic status. In improving budgetary allocation, the Government would do well to assess expenditure in terms of its impact in ameliorating poverty, either by redistributing income or by investing in human capital. It could openly assess whether its budget allocations are sufficiently pro-poor; if they are not, it could use the methods recommended here to improve allocations. Poverty reduction is not the only measure of expenditure effectiveness, but it is often ignored as other considerations drive budgetary decisions. Regardless of the reforms introduced in budgeting, allocations will not be significantly more effective if they do not abate poverty. Managing Public Services 6.41 Although they are only a small part of total expenditure, the funds allocated to operating government programs and agencies influence the quality of public services, the efficiency with which they are delivered, relations between citizens and government, and the availability of resources for discretionary purposes. Some of the most innovative reforms adopted in other countries during the past two decades pertain to this aspect of public expenditure management. The central aim has been to reorient public management from control and compliance to performance and results by giving managers broad discretion in using human and financial resources while holding them accountable for outputs and outcomes. The roster of contemporary reforms includes performance-based budgeting, contracting out and other market-like arrangements, creation of autonomous 85 agencies, and contracts between political leaders and managers specifying resources to be supplied and expected results. 6.42 Financial management is key. At present, the manner in which the budget is executed degrades operational efficiency. Throughout the year, managers have ongoing uncertainly as to the resources that will be forthcoming. Just about the only thing that mangers are certain of is that they will not receive all the funds approved in the budget. Operating under quarterly and monthly quotas, much managerial attention is devoted to making it through until the next allotment arrives. As discussed earlier, line ministries haggle with MECON, juggle payments, and build up unrecorded arrears. Under these circumstances, planning and adhering to an orderly work program becomes close to impossible. 6.43 At present, the executed budget determines the resources available to spending units, not the formally approved budget. This inversion of normal budget practice discourages the preparation of realistic budgets and encourages managers to act in ways that damage the Government's financial position but enables them to muddle though the fiscal year. A precondition of improvement in services and operations is that the budget will be implemented as adopted, subject to adjustment due to truly unforeseeable circumstances. The annual expenditure decree should serve two purposes: (i) to smooth spending by quarters so that ministries and agencies do not run out of money before the year has ended; and (ii) to reserve some funds as a hedge against lower revenues and other contingencies. It should not be a means of sequestering appropriated funds, except when it is essential to do so because of macroeconomic instability. The decree could specify provisional quarterly allotments that may be adjusted in final quarterly quotas as the fiscal year progresses. When conditions are favorable, upward adjustments could be made by releasing previously reserved funds; when conditions are unfavorable, it may be appropriate to supplement the reserves by cutting the allotments below the level set in the annual decree. With this arrangement in place, spending units would have reasonable assurance as to the resources that will be forthcoming, and the Government will retain short-term macroeconomic capacity. 6.44 As fiscal conditions ease, the Government could move to quarterly accruals or should cancel this procedure altogether. Once commitments have been authorized and properly entered in SID1F, liabilities could be processed for payment routinely. Ideally, there would be no limitation on the volume of payment claims submitted by spending units, but if a limit still is deemed needed to safeguard the Government's cash position, it could be allocated each quarter, not every month. Operating on a monthly basis may be imperative when circumstances are dire, but this should not become standard operating procedure. 6.45 Misuse of the "accruals" concept breeds confusion and contributes to the accumulation of floating debt and other arrears. SIDIF, or a successor system, should provide for recording accrued liabilities, especially those for which payment may be delayed. There should be no restriction on accruing these liabilities; if there were, spending units simply would not inform the Government of them until it was able to pay. The Government could improve procedures for averting the buildup of floating debt. It 86 appears that problems arise when spending units have ongoing liabilities, such as contracts for leasing office accommodation and for services provided on a continuing basis throughout the year. Resources for these commitments need to be encumbered in advance, not when the spending unit submits an invoice to SIDIF. 6.46 Budget execution through SIDIF-the Government's integrated financial information system-has been a significant accomplishment of public administration. SIDIF was one of the first integrated systems installed in Latin America; it greatly improved accuracy and transparency in public finance and reduced the cost of processing transactions. SIDIF needs an upgrade, however, both to incorporate modem technology and to reintegrate subsystems that have been split off from it by various spending units. An increasing number of units have a SIDIF locale, a parallel system that they use for internal management and that feeds into SIDIF at some point prior to the processing of payments. Although it is appropriate for ministries and large agencies to have sub-systems for internal management, all should consider using SIDIF or its successor throughout the financial management process. 6.47 Provinces can also take several measures to improve the comprehensiveness and timeliness of their financial information systems. First, they could try to bring together the responsibility for budgeting, accounting, and cash management. Some provinces presently split the core responsibility for financial management by having a treasurer and comptroller general who are legally independent from the finance ministry, which makes integrated financial management exceedingly difficult. Second, provinces could accelerate development of modern integration financial information systems. Some have ambitious projects underway, but none has a fully operational system. Third, basic classifications and reporting requirements could be standardized to permit the aggregation of federal and provincial financial data and to facilitate compliance with fiscal rules and constraints. 6.48 Service provision and government operation are widely believed to be enhanced when managers have flexibility in using resources. One of the approaches introduced in the United Kingdom and some other countries is to establish autonomous agencies and give them broad operating freedom in carrying out their missions. In recent years, Argentina has seemingly followed this path by creating organizational enclaves, some with their own source of funding from user charges or other revenue. Given the favored status of these entities, other agencies are petitioning for operating freedom as well. However, there is not much basis for expecting their performance to significantly improve. Two defects in the rush to liberate agencies from financial and personnel controls are perceived. First, the process by which they have been established does not appear to have given due consideration to the agencies' preparedness for self-management. Second, inadequate provision has been made for holding them accountable for results. The process by which an agency is established has considerable bearing on its subsequent performance. 6.49 The Government has a sound basis for reviewing organizational performance and conferring agency status through its pilot test of performance contracts in five agencies. Each participating agency negotiates an agreement with government (and some have internal contracts as well between senior and middle managers) specifying expectations for the next several years. While the contracts are not legally enforceable, they do spur the 87 Government and the affected agency to discuss operational issues, performance targets, and resources. It would be sensible for the Government to make performance contracts a precondition for agency independence. Alternatively, it may adopt an approach introduced by Thailand, which, specified a series of "hurdles" that an agency must surmount before it is granted operational independence (Box 8). Whatever approach is taken, the important thing is that independence be pursuant to an assessment of the organization's capacity and in exchange for demanding greater accountability for performance. Box 8: Creating Autonomous Agencies--United Kingdom and Thailand In the late 1980s, the United Kingdom launched "Next Steps" agencies by separating the service delivery functions of government from policy. Each agency-more than 130 were established-has a broad degree of operating freedom, but each is supposed to be accountable for performance. Before creation, each candidate agency went through an extended review of internal controls, relations with its parent ministry, an assessment of managerial capacity, and objectives and performance targets. A chief executive was selected in open competition and given a term contract specifying performance targets. In addition, each agency negotiated a framework document with Government specifying its mission and operating discretion. Finally, each agency's performance was closely monitored through annual performance targets and periodic evaluation. Thailand launched a "hurdle" approach to giving agencies operating freedom in 1999 as part of a quid pro quo: Agencies that satisfy the hurdles receive more discretion in using budgeted funds; this determination is made on an agency by agency basis. Agencies that meet the hurdles negotiate a resource agreement with the Bureau of the Budget. The hurdles pertain to seven key areas of public management; budget planning, output costing, procurement, budget and funds control, financial and performance reporting, asset management, and internal audit. For each area, a gaps analysis is undertaken, and the affected agency must take steps to fill the gaps before it is given operating discretion. 6.50 The Government is not ready for performance-based budgeting. One reason is that other, more urgent matters demand priority; another is that is has not advanced sufficiently in measuring performance to budget on this basis. But it can improve the quality of public services by adapting some current features of budgeting toward this end. In preparing budget requests, spending units submit high, medium, and low options. The alternative budget scenarios have little utility other than to enable the Government to hold some options open until late in the preparation cycle. These alternatives would have greater value if they were associated with alternative levels of service, provided, of course, that the claimed link of resources and services was reliable. It is useful even if service measures are likely to resemble the physical measures of workload that many spending units now compile. As desirable as better measures are, emphasizing them at this stage may impel the Government to invest inordinate effort in measuring performance. It would be a more productive use of scarce resources to sharpen the Government's understanding of how different spending levels affect services. 88 REFERENCES Agosta, R. 2003. "Aspectos Fiscales del Sector de Transporte en la Argentina". Mimeo, Finance, Private Sector and Infrastructure Department, Latin America and Caribbean Region, World Bank, Washington DC. Ahmad, E., and E. Mottu. 2003. "Oil Revenue Assignments: Country Experiences and Issues." IMF Working Paper WB/02/203, Washington, D.C. Arganiaraz, N., J.M. Garz6n, and M. Capello. 2003. "Federalismo Fiscal Argentino: un Analisis Comparativo con Australia y Canada." IERAL - Fundaci6n Mediterranea, Working Paper, Fiscal Policy Series, Document No. 3. Baffes, J., and Shah; A. 1998. "Productivity of Public Spending, Sectoral Allocation Choices and Economic Growth." Economic Development and Cultural Change 48(2). Becerra, M., 0. Cetrangolo, J. Curcio, J.P. Jimenez. 2003. "El Gasto Publico Universitario en Argentina." Working Paper No. 7, Argentina Country Office, World Bank, Washington, D.C. Becerra, M., S. Espaha, and A. Fiszbein. 2003. "El Gasto en Educaci6n Basica en la Argentina y algunos Aspectos sobre la Eficiencia en la Asignaci6n." Documento de Trabajo No. 6/03, World Bank, Buenos Aires. Bird, R.M.. 1999. "Rethinking Tax Assignment: The Need for Better Subnational Taxes." Seminar presented at the World Bank, April 27, 1999, Washington D.C. Processed. Bird, R.M., and Pierre Paul Gendron. 1998. "Dual VATs and Cross-Border Trade: Two Problems, One Solution?." International Tax and Public Finance 5: 429-442. Boadway, R., and P. Hobson. 1993. "Intergovernmental Fiscal Relations in Canada." Canadian Tax Paper No. 96, Canadian Tax Foundation. Braun, M.. 2003. "Institutional Reform Proposals for Fiscal Solvency in Argentina." CIPPEC. Buenos Aires. Processed. Calder6n, C., W. Easterly, and L. Serven. 2003. "Latin America's Infrastructure in the Era of Macroeconomic Crises." forthcoming. Calvo, G., A. Izquierdo, and E. Talvi. 2002. "Sudden Stops, the Exchange Rate and Fiscal Sustainability: Argentina's Lessons." Mimeo, IADB, Washington, D.C. CEER .2002. "Las Empresas Privadas de Servicio Publico en la Argentina: Analisis de su Contribuci6n a la Competitividad del Pais." Centro de Estudios Econ6micos de la Regulaci6n, Universidad Argentina de la Empresa, Buenos Aires. Chisari, 0., E. Estache, and C. Romero. 1999. "Winners and Losers from Privatization and Regulation of Utilities: Lessons from a General Equilibrium Model of Argentina." World Bank Economic Review 13(2):357-378. CIPPEC. 2002. "Coparticipaci6n con Responsabilidad". Cline, W.R. 2003. "Restoring Economic Growth in Argentina." Center for Global Development and Institute for International Economics. Washington D.C. Corrales, J. 2002. "Killing Me Softly with Each Shock: Explaining Argentina's Depresi6n." Amherst - College, Amherst, Massachusetts. Available online at: http://amherst.edu/-icorrale/ . Corden, W. M. 2002. Too Sensational: on the Choice of Exchange Rate Regimes: MIT Press. De la Torre, A., Levy Yeyati E., and S. L. Schmukler. 2003. "Living and Dying with Hard Pegs: The Rise and Fall of Argentina's Currency Board." World Bank, Washington, D.C. 89 Estache, A. 2003. "Argentina's 1990 Utilities Privatization: A Cure or a Disease?." Mimeo, World Bank Institute, Washington D.C. Estache, A. and M. Fay. 1995. "Regional Growth in Argentina and Brazil: Determinants and Policy Options." World Bank, Washington D.C. Fay, M., and T. Yepes. 2003. "World Infrastructure Investment Needs: 2005/10." Mimeo, Finance, Private Sector and Infrastructure Department, Latin America and Caribbean Region, World Bank, Washington D.C. Fenochietto, R. 1998A. "El IVA Compartido, su Funcionamiento y la Autonomfa Provincial." Chief of Cabinet's Office, Buenos Aires. Processed. -----------------. 1998B. "Proponen la Creaci6n de IVA "Compartido"." Commentary in the newspaper Ambito Financiero, June 2, 1998. ----------------. 1998C. "The Shared Value Added Tax: A Useful Tool to Replace the Gross Receipts Tax and the Decentralization of Taxes." Chief of Cabinet's Office, Buenos Aires. Processed. Ferreira, P.C. 1996. "Investimento em Infra-estrutura no Brasil: Fatos Estilizados e Relacoes de Longo Prazo." Institute for Applied Economic Research, Pesquisa e Planejamento Economico: 26(2): 231-252. FIEL (Fundaci6n de Investigaciones Econ6micas Latinoamericanas). 1998. La Reforma Tributaria en la Argentina. Buenos Aires: Ed. Manantial. FLACSO. 2002. "El Proceso de Privatizaci6n en la Argentina." Bernal: Universidad Nacional de Quilmes Foster, V. and CEER, 2003a. "Impacto Social de la Crisis Argentina en los Sectores de Infraestructura. 6En Que Medida Existe una Polftica Social?", Documento de Trabajo No. 4/04, Oficina del Banco Mundial para Argentina, Chile, Paraguay y Uruguay, Buenos Aires Foster, V., and CEER, 2003b."Hacia una Polftica Social para los Sectores de Infraestructura en Argentina: Evaluando el Pasado y Simulando el Futuro", Mimeo, Departamento de Finanzas, Infraestructura y Sector Privado, Regi6n de America Latina y el Caribe, Banco Mundial, Washington DC. Foster, V., M. Marchioni, M.V. Fazio, D. Petrecolla, 0. Chisari, D. Bondorevsky, J. Cardozo, G. Ferro, C. Romero, and C. Ruzzier. 2003. "Nace una Politica Social para los Sectores de Infraestructura en Argentina." Mimeo, World Bank and Universidad Argentina de la Empresa, Buenos Aires. Foster, V., D. Petrecolla, 0. Chisari, D. Bondorevsky, I. Canay, J. Cardozo, G. Ferro, C. Romero, and C. Ruzzier. 2003. "Impacto Social de la Crisis Argentina en los Sectores de Infraestructura. 6En Qud Medida Existe una Politica Social?." Mimeo. World Bank and Universidad Argentina de la Empresa, Buenos Aires. Frigerio, R. 2003. "Proyecto de Reforma Econ6mica." Presentation at World Bank workshop on Provincial Tax and Coparticipation Reform. Gonzalez, C., D. Rosenblatt, and S. B. Webb. 2002. "Stabilizing Intergovernmental Transfers in Latin America: A Complement to National/Subnational Fiscal Rules?." Policy Research Working Paper 2869, World Bank, Washington, D.C. Gonzalez Cano, H. 1996. "La Reforma Tributaria de Brasil y Posible Aplicaci6n del Nuevo IVA Federal y Estadual para el Reemplazo del Impuesto a los Ingresos Brutos." Boletin de la DGI, No. 513:1391-97, Buenos Aires. Grushka, C. 2002. "Proyecciones Previsionales de Largo Plazo - Argentina 2000-2050." Estudio Especial N°14, Superintendencia de Administradoras de Fondos de Jubilaciones y Pensiones. September. 90 Guasch, J.L. 2003. "Granting and Renegotiating Infrastructure Concessions: Avoiding the Pitfalls." Mimeo, University of California, San Diego. Gwilliam, K., and Shalizi, Z. 1999. "Road Funds, User Charges and Taxes." The World Bank Research Observer 14(2):159-185. Hausmann R., and A. Velasco. 2002. "The Argentine Collapse: Hard Money's SoftUnderbelly." Mimeo, Kennedy School of Government, Harvard University, Cambridge, Massachusetts. IDEA. 2001. "Construyendo Capacidades e Instituciones para el Desarrollo Exportador: Una Agenda Estratdgica para la Argentina del 2010." Instituto de Economfa Aplicada, Buenos Aires. Irwin, T., Klein, M, Perry, G. and Thobani, M., 1997 "Dealing with Public Risk in Private Infrastructure." Chapter 6, World Bank Latin America and Caribbean Studies, Washington D.C. Jefatura de Gabinete. 1999. Draft Coparticipation Law Presented to Congress. Keen, Michael, 2000, "VIVAT, CVAT and All That: New Forms of Value-Added Tax for Federal Systems," IMF Working Paper, WP/00/83. Kharas, Homi, and Deepak Mishra. 2000. "Fiscal Policy, Hidden Deficits, and Currency Crises." Washington,D.C. Keen, Michael, L. Ebrill, J.P. Bodin, and Victoria Summers. 2001. The Modern VAT: Washington, D.C. International Monetary Fund. Maceira, D. 2002. "Financiamiento y Equidad en Salud en un Sistema Descentralizado: El Caso de Argentina." CEDES, Buenos Aires, April 2002. McLure, C.E. 1998.-"Tax Assignment in Argentina: Conceptual and Practical Issues." Paper prepared for the World Bank. Mussa, M. 2002. "Argentina and the Fund: From Triumph to Tragedy." Mimeo, Institute for International Economics, Washington, D.C. Ortega and Zapata. 1999. Draft Coparticipation Law presented to Congress. Parandekar, S., S. Espania, and M. Savanti. 2003. "El Impacto de la Crisis en el Sistema Educativo Argentino." Working Paper No. 3. Argentina Country Office, World Bank, Washington, D.C. Perry, Guillermo, and Luis Serven. 2003. "The Anatomy of a Multiple Crisis: Why was Argentina Special and What can we Learn from It." World Bank, Washington, D.C. Piffano, H. 1997. "El Nuevo Sistema Tributario Provincial en un Escenario de Descentralizaci6n Tributaria." Paper presented at the Second International Seminar on Federalism and Local Governments, National University of La Plata. Poddar, S. 1990. "Options for a VAT at the State Level." in Malcolm Gillis, Carl S. Shoup, and Gerardo P. Sicat, Value Added Taxation in Developing Countries. Washington, D.C:The World Bank. Porto, A., and P. Sanguinetti. 2001. "Political Determinants of Intergovemmental Grants: Evidence from Argentina." Economics and Politics 13(3). Potter, B.H. 1997. "Dedicated Road Funds: A Preliminary View on a World Bank Initiative." IMF Paper on Policy Analysis and Assessment, PPAA/97/7, Fiscal Affairs Department, International Monetary Fund, Washington D.C. Powell A. 2002. "Argentina's Avoidable Crisis: Band Luck, Bad Economist, Bad Politics, Bad Advice." Paper Prepared for Brookings Trade Conference, Washington, D.C. Rofman, R. 2003. "The Pension System and the Crisis in Argentina." World Bank, Washington D.C. Rodrik, D. 2002. "Argentina: Globalization Gone Too Far or not Far Enough."Mimeo, Kennedy School of Government, Harvard University, Cambridge, Massachusetts. 91 Sachs, J. 2002. "Understanding and Responding to Argentina's Economic Crisis." Mimeo, Center for International Development, Harvard University, Cambridge, Massachusetts. Schuler, K. 2002. "Fixing Argentina." Mimeo, Cato Institute, Washington, D.C. Secretaria de Poifticas Universitarias. 2002. "Estudio de los Egresados de Universidades Nacionales." Mimeo. Buenos Aires. Silvani, C. and P. dos Santos. 1996. "Administrative aspects of Brazil's consumption tax reform." International VATMonitor 7:123-32. Stiglitz, J. 2002. "Argentina Shortchanged." Washington Post May 27 2002. Tejeiro, M. 2001. "Una Vez Mas, La Politica Fiscal ... ." Mimeo, Centro de Estudios Publicos. Thurman, Quint C. 1991. "Taxpayer Noncompliance and General Prevention." Public Finance 46(2): 289-298. Tobar, F. 2002. "El gasto en Salud en Argentina y su Metodo de Calculo." PIA-ISALUD, Buenos Aires. Tomassi, M., S. Saiegh, and P. Sanguinetti. 2001. "Fiscal Federalism in Argentina: Policies, Politics and Institutional Reforms." Economia, Spring 2001. Varsano, R. 1999. "Subnational Taxation and the Treatment of Interstate Trade in Brazil: Problems and a Proposed Solutions." Paper to be delivered at the Fifth Annual Bank Conference on Development Economics in Latin America, Valdivia, Chile. Varsano, R. 1995. "A Tributacao do Comercio Interestadual: ICMS versus ICMS Partilhado." Texto para Discussao No. 382, Instituto de Pesquisa Econ6mica Aplicada (IPEA), Rio de Janeiro. Vegas, E., L. Pritchett, and W. Experton. 1999. "C6mo Atraer y Retener Docentes Calificados en la Argentina: Impacto del Nivel y la Estructura de Remuneraci6n." World Bank LCSHD Paper Series, vol. 38S, Washington, D.C. Wallschutzky, I. G. 1984. "Possibly Causes of Tax Evasion." Journal of Economic Psychology 5: 371- 84. Wyplosz, Charles. 2003. "Debt Sustainability in Brazil." Graduate Institute of International Studies, Geneva, and CEPR. World Bank. 1996. "Argentina: Provincial Finances Study." Washington, D.C. 1998. "Argentina: Fiscal Dimension of the Convertibility Plan." Washington, D.C. ------------. 1999. "Argentina: Provincial Tax and Revenue-sharing Reforn" Washington, D.C. ------------. 2001. "Finanzas Publicas Provinciales, Actualizaci6n IV." Washington, D.C. -----------. 2003. "Una estrategia de Salud para Argentina - Recomendaciones para Mejorar el Desempano del Sector Salud." Washington, D.C. World Health Organization. 2000. "The World Health Report." 92 ANNEX I: DESCRIPTION OF REVENUE AND TAX-SHARING ARRANGEMENTS I.1 The figure below provides a roadmap of the complex web of revenue-sharing and tax-sharing arrangements. Triangles represent the various federal taxes, and the size of these triangles is roughly proportional to their shares of total federal revenues during 2002-an atypical year. Because of the devaluation and introduction of export taxes, trade taxes figure prominently here whereas they would have constituted shown up as a very small triangle in the chart during most of the 1991-2001 period. 1.2 The chart shows the vertical division of resources ("primary distribution) across the two levels of government. Below, there is a brief description of the distribution across provinces ("secondary distribution") of the provinces' total share. It is also highly complex, as different rules apply to the various circles/ovals in the attached chart. Some explanations and clarifications to the complicated figure are provided below. 1.3 Roughly two-thirds of the automatic transfers to provinces are through the net coparticipation bubble in the middle of the chart. The distribution of these funds across provinces is according to the fixed coefficients of the general coparticipation law of 1988 (Law 23548).' The fixed amount of $ 549 million deducted from the gross coparticipation pool is distributed according to fixed amounts to a particular set of provinces as agreed in the 1992 Pacto-Fiscal. However, the pre-deduction from the income tax of $440 million follows the secondary distribution of Law 23548. 1.4 The 14 percent of "Net Income Tax" that goes to the provinces started out as a special benefit for the Province of Buenos Aires, whereby 10 percent would go to that province to attend to the urgent social needs of the suburban areas around the city of Buenos Aires, but there was a maximum set of $650 million with any excess of the 10 percent over that amount going to the rest of the provinces, plus another 4 percent going to the rest of the provinces directly. Hence, really 14 percent goes to the provinces with $650 million to the Province of Buenos Aires (note the dotted line in the chart) and the rest of that 14 percent going to rest of the provinces. However, the share of the 10 percent above $650 million is distributed according to an index of "Unsatisfied Basic Needs" (NBI--Necesidades Basicas Insatisfechas), while the straight 4 percent is distributed according to the fixed coefficients of Law 23548. 1.5 The special infrastructure funds that are fed by gasoline taxes are no longer tied to those sectors; hence, the dashed circle that joins them together. However, the separation of these funds is still relevant in terms of the secondary distribution, since each fund has its own rules for the distribution across provinces. ATNs are discretionary funds; however, the fund itself has assured automatic financing from a variety of sources, as shown in the chart. I.6 Although the figure is very complicated, it would be even more so if certain simplifications had not been made. For example, a special regime for the small business "unified tax" (monotributo) has been omitted from the chart as the revenues are quite 93 small. In general, the other smaller taxes-including, for example, the tobacco excise tax-simply feed into the gross revenue-sharing pool. Another omission for the sake of simplification are the $ 3 million that are deducted annually from the personal assets tax and earmarked for the national organ transplant center-prior to the distribution of funds shown in the chart. There are also a few other very small taxes with special rules that are not shown such as the "statistical tax" that is earmarked for the federal government (including a special share for the trade statistics agency) and the emergency tax on sporting events and gambling. There are also some taxes that are more like user fees which are earmarked for particular special agencies. The airport tax is an example of this. These "triangles" would be extremely small in the following chart, and as such, are excluded. 94 ARGENTINA: FEDERAL TAX AND REVENUE-SHARING ARRANGEMENTS GASOLINE & NATURAL GAS N (since May 2002) 7 z OTHER 42% 29% EXCISE & OTHER INCOir/ 29%\ VAT rFINANCIAL TAXE TAX HOUSING I ~~~~10% \ X v | 10% 64 REST /t /(ELEC- 70% "GROSS ARS 20:/ \ROADS /' \ / v = At / ARX i,8/4INFR)~~~~~~~, ~ 30 15% 89toX oo ; ARS 549 \ <\ 100% 440 MAR NA D/ ~~~~~42 NATN\\ \ 11%/ 1 W\ / 57%\ IF I s \| \ARS 120 M / FEDERAL bOV'T \ r PROVMCES . fl *9 / I p TREASURY t ~SECURrTY A \650 MS PENSIONS PO C 100% IOF 1\< !I SA / ARS 157 £10%/\\ 2 -~~~~ TRD/ :1___| FEDERAL T TAXES - OTHER 94% / TXS PERSONAL \ ~ ~~FUELS TAXES ASSETS TAX 95 ANNEX II: FEDERAL/SUBNATIONAL VAT WITH DESTINATION PRINCIPLE-CVAT AND VIVAT ALTERNATIVES 1I.1 Active academic debatea, originating primarily in Latin America, now regards a provincial VAT or VAT surcharge as a promising alternative for broad-based intermediate-level government taxation. The general features of the provincial VAT on which a consensus appears to be emerging is that the provincial VAT should: (a) be destination based; (b) have zero-rating of inter-provincial sales by the province of origin with deferred payment in the province of destination (a la Europe); (c) have provinces set their own rate; and (d) run in parallel to the federal VAT on all transactions (w/ reimbursement of foreign exports). This idea draws upon proposals that have come out of both Brazil and Argentina for an IVA compartido or IVA partilhado, as well as the dual VAT currently functioning in the province of Quebec, Canada. Researchers that have studied implementation issues have suggested a pair of predominant options: (l)the Compensating VAT or "CVAT"; or (2) the Variable Integrated VAT, or "VIVAT." A. CVAT 11.2 On sales within a province, the provincial VAT would be charged along with the traditional federal VAT. On sales across jurisdictions, no provincial VAT would be charged in the province of origin (zero-rating), but when a final sale (or next stage of production sale) occurs in the province of destination, then the destination province's VAT rate would apply to that sale (deferred payment). To eliminate the incentive for households (final consumers) in other provinces to pose as registered traders, a federal "compensating VAT" (CVAT) rate would be applied to all inter-provincial sales - whether they be to households or registered traders. The "compensating VAT" not only would discourage evasion, but it would establish a chain of credits when goods move across boundaries between registered traders, thus maintaining one of the most desirable features of the VAT.b The "compensating VAT" seems to resolve almost all of the problems described in Chapter 3. In Quebec, no CVAT is used, and the power of strong enforcement of the concurrent federal VAT seems to be sufficient to solve this compliance problem. This Canadian institutional feature is probably not in place in Argentina. 11.3 A simple numerical example from McLure, 1998 illustrates the workings of the CVAT (see figure at end of annex 11). The CVAT yields no net revenues on sales between registered traders. Actual collection of revenues only would occur in sales to non-registered traders (e.g., mail-order sales to final consumers in other jurisdictions). In a Varsano (1995, 1999); Fenochietto (1998A,B,C,); FIEL (1998); McLure (1998), Bird (1999), Bird and Gendron (1998); Piffano (1997); Gonzalez Cano (1996); Silvani and dos Santos (1995); and Poddar (1990). b Varsano (1999) refers to this quality of the CVAT as the "little boat" (o barquinho) in that it carries the good across the river (border), allowing it to arrive at the next jurisdiction free from provincial taxes, but with appropriate crediting for the "compenasating" federal tax (CVAT credit). 96 this example, 100 pesos worth of value added is produced in each of three stages. The inter-jurisdictional sale occurs from stages 2 to 3. The federal VAT rate is 20 percent, province A's rate is 4 percent, B's rate is 8 percent, and the CVAT rate is 6 percent. Table I 1: Example of Provicial VAT/CVAT Calculation of Tax Transactions (VAT on sales; credit for input VAT; net VAT liability) (Purchases, Sales, and Value One-province (B) Two Province Example Added) Example Prov'l Federal Tax Prov'l Federal Tax: . Compen-Ordinary Total Tax Tax (A) (B) sating Federal Federal Stage 1: (All stages in B) (This stage occures in Province A) a. Purchases/Credits 0 0 0 0 n.a. n.a. 0 0 b. Sales/Tax 100 8 20 4 n.a. n.a. 20 20 c. Value added/ 100 Net tax this stage 8 20 4 n.a. n.a. 20 20 (c=b-a) Stage 2: a. Purchases/Credits 100 8 20 4 n.a. n.a. 20 20 (d=b) b. SalestTax 200 16 40 (Interstate sale from A to B) 0 n.a. 12 40 52 c. Value added/ 100 Net tax this stage 8 20 -4 n.a. 12 20 32 (f=e-d) Stage 3: (This stage occurs in B) a. Purchases/Credits 200 16 40 n.a. 0 12 40 52 (g=e) b. Sales/Tax 300 24 60 n.a. 24 n.a. 60 60 c. Value addedl 100 Net tax this stage 8 20 n.a. 24 -12 20 8 (i=h-g) Total Tax 24 60 0 24 0 60 60 (j=c+f+i) Source: McLure (1998) H1.4 Another example is presented below. Here a producer of cheap bulk wine purchases grapes from within the province, sells wine to retailers both within the province and in another province, and final consumer sales occur in both jurisdictions. The various assumptions on tax rates appear in the graphic below. These rates are only assumptions used simply to illustrate how the system would work. They should not be viewed as estimates of what the provincial rate would be for the provinces mentioned below. In particular, a combined VAT of over 27-29 percent is probably too high. 97 STAGE 1: Grape farmer in Mendoza sells $100 of grapes to bulk wine producer in Mendoza: Federal VAT 21 Provincial VAT 6 STAGE 2: Bulk wine producer makes $200 of wine, and sells $100 to a local Mendoza store; and $100 to a store in the Province of Buenos Aires To Buenos Aires store: To 100 Mendoza store $100 Tax on Sale: Tax on Sale: Federal VAT 21 Federal VAT 21 Provincial VAT 6 Provicial VAT 0 Credits: Federal CVAT 7 Federal VAT 21 Provincial VAT 6 NET TAX for STAGE 2: Federal VAT 42 - Credits 21 = 21 Provincial VAT 6 - Credit 6 = 0 Federal CVAT 7= 7 Stage 3: Wine sold to Consumer Retail price of wine is $300, f excluding tax (balf sold in Mendoza, t half sold in Buenos Aires) In Mendoza store, wine sold In Buenos Aires, wine sold to large family for $150 to another large family for $150 1 ~~~~~~~~~1 Tax on Final Sale to Consumer. Tax on Final Sale to Consumer: Federal VAT (21%) 31.5 Federal VAT (21%) 31.5 Provincial VAT (8%) 12.0 Provincial VAT (6%) 9.0 Credits: Credits: Federal VAT 21.0 Federal VAT 21.0 Federal CVAT 7.0 Provincial VAT 6.0 Provincial VAT 0.0 Net Tax This Stage: Net Tax This Stage: Federal VAT 10.5 Federal VAT 10.5 Provincial VAT 3.0 Federal CVAT -7.0 Provincial VAT 12.0 98 B. VIVAT 11.5 Another alternative would be the Variable Integrated VAT, or ''VIVAT."'C In an internationational context (for which it was originally conceived), the VIVAT would involve a single rate on all intermediate transactions (i.e., sales to registered traders). On final stage sales (i.e., to non-registered traders), countries could charge a rate at their discretion. In a federal context, the federal government in all likelihood would administer the single-rate tax on intermediate sales, while provinces could set rates on final sales. The latter could, in principle, be administered by provincial tax administrations, or by the federal tax administration on behalf of the provinces. As Keen (2000) points out, "...it is in structural terms equivalent to a common federal VAT levied at the intermediate rate combined with a series of provincial retail sales taxes levied at a rate equal to the difference between the provincial VAT and the common intermediate rate." 11.6 Below is a simple numerical example that shows how the VIVAT would work. The example is adapted from Keen (2000). The intermediate VAT rate is 10 percent, while the final consumption rate is 10 percent in country I and 15 percent in country 11. Firms A and B are located in country I while Firm C is located in country II. The value of sales of goods produced by each firm. are 80, 200, and 480 for Firms A, B and C respectively. Table 1 2: Example using the VIVAT Method Firm A (Country I) Input Tax Paid/Credits: 0 (Gross sales of 80 to Firm B) Output Tax: 8 Firm B (Country I) Input Tax Paid/Credits: 8 (Gross sales of 200 to Firm C) Output Tax: 20 Net Tax: 12 Firm C (Country II) Input Tax Paid/Credits: 20 Output Tax: 72 Net Tax: 52 Total Tax Collected (on 80+120+280=480 of value added) (Sum of Net Tax across 72 stages} {= 15% of Value-Added) cSee Keen, 2000. This article provides a description of the functioning of the VIVAT, plus an interesting comparison of CVAT, VIVAT and straight dual VAT. 99 ANNEX HI: PUBLIC EXPENDITURES IN SOCIAL SECTORS HI. I Table 1I 1 provides a summarized picture of spending in the social sectors before the collapse of the convertibility regime. The table does not include spending by municipal governments which, in those years, represented about 0.8% of GDP. Unfortunately, the information available on municipal spending does not allow the type of disaggregated analysis performed in this chapter. A detailed analysis of municipal social spending remains a challenge for the future. Table III 1: Public Spending in the Social Sectors 2000 2001 Sector In nillions In millions % As % of In millions In mllions % As % of of $ of 2001 $ Structure GDP of $ of 2001 $ Structure GDP Education 12,203.1 12,162.1 23.4 4.3 12,184.9 12,184.9 23.6 4.5 Health 4,398.7 4,325.6 8.3 1.5 4,371.0 4,371.0 8.4 1.6 Social Protection 3,241.4 3,187.6 6.1 1.1 3,252.5 3,252.5 6.3 1.2 Social Security 32,923.0 32,376.2 62.2 11.6 31,922.1 31,922.1 61.7 11.9 Health 8,955.7 8,806.9 16.9 8.4 8,602.7 8,602.7 16.6 8.7 Others (*) 23,967.4 23,569.2 45.3 3.2 23,319.4 23,319.4 45.1 3.2 Total 52,766.3 52,051.5 100.0 18.6 51,730.7 51,730.7 100.0 19.2 (*) Others includes pensions, unemployement assitance and family allowance Source: Ministry of Economy - National Directorate of Consolidated Social Spending M.2 The overall high level of spending is the result of high spending levels on social insurance prograrns representing about 62% of all spending or close to 12% of GDP. Indeed, the other three categories of spending are not particularly high as a percentage of GDP compared to other countries in the region. As it will be discussed in more detail in Sections 3, 4 and 5 this has important implications from an equity point of view as the poorest groups in society tend to be excluded from social insurance programs. H1.3 Table Im 2 provides a rough picture of the roles played by the national and provincial levels in social sector spending. Overall, provinces financed and execute approximately 42% of social sector spending. Not surprisingly, this share is the highest for education and health given the decentralization of basic education and public health services. Perhaps less well known, is the high share (about 60%) of provincial financing of social assistance programs. This is mostly due to the role played by provinces in feeding programs (including school-based ones). This pattern, though, has changed radically after 2001 -as it will be shown below. Table In 2 also provides an initial view of another feature of social policies in Argentina. The national level contributes directly to the financing of decentralized social services through earmarked transfers to provinces. Overall, such federal transfers represent 2.4% of total spending. Such transfers could, in d Non-conditional resources from the revenue sharing system (coparticipaci6n) represent a significant share of those resources for most provinces. 100 theory, play an important compensatory role. And, if badly designed, they could create distortions in the allocation of resources. Table 1I 2: The Federal/Provincial Roles 2000 2001 Federal Provinces Total Federal Provinces Total Transfers Transfers Education 14.1 7.1 78.8 100 13.4 6.7 79.9 100 Health 11.7 2.2 86.1 100 10.6 1.6 87.8 100 Social Assistance 34.8 3.3 61.9 100 36.4 4.5 59 100 Social Insurance 79.5 0.5 19.9 100 78.5 0.5 21 100 Health 75 0.1 25 100 74 0.1 26 100 Others 81.3 0.7 18 100 80.2 0.6 19.2 100 Total 55.9 2.4 41.7 100 54.8 2.3 42.9 100 Source: Ministry of Economy - National Directorate of Consolidated Social Spending A.4 Table m 3 provides a summary of the complex set of financing and executing arrangements that characterizes the social sectors in Argentina. Contributionse by employers and employees represent approximately 40% of social spending in the country. Those resources go to finance services and benefits through social insurance agents at the national and provincial level. Both the Federal and provincial treasuries contribute to spending by social insurance agents (that is, in addition to their direct contributions in their role as employers). These contributions go to finance the operational deficits of the various pension systems as well as some specific social insurance programs that are not self-sufficient (PAMI being the best known). To the extent that beneficiaries of social insurance programs are typically not the lowest segments of the income distribution, these 'subsidies' tend to have a regressive distributional effect. e The revenues of the "national" health insurance system are collected by the Tax Federal Administration (AFIP) and distributed to the Obras Sociales. Both revenues and spending of agents of the national social health insurance system are off budget items, but regulated and supervised by the government through the Superintendencia de Seguros de Salud (SSS). 101 Table 1I 3: Sources of financing and executing agencies (in millions, pesos of 2001) Sources of Financing Executing Agency Federal Provincial Employer & . Employee Total Treasury Treasuries Eoyee Total Contribution Federal Government (Central Adm.) 3,278 3,278 Provincial Governments (Central Adm.) 1,027 15,491 16,518 Social Insurance Agents 11,076 430 20,416 31,923 Federal Insurance Agent 10,511 8,223 18,734 Administration 249 249 Pensions and Retirement 9,599 6,115 15,714 Family Allowances 1,808 1,808 Unemployment Insurance 54 300 354 Subsidies to beneficiaries 609 609 Provincial Pensions Systems 142 430 4,014 4,586 Administration 89 89 Pensions and Retirement Payments 142 341 4,014 4,497 Health Insurance Agents (Obras Sociales) 423 - 8,179 8,602 Provincial Health Insurance Systems 4 2,235 2,239 National System 3,974 3,974 Pensioners (PAMI) 419 1,970 2,389 Total 15,382 15,921 20,416 51,718 Source: Ministry of Economy - National Directorate of Consolidated Social Spending 11.5 The financing share of national and provincial treasuries are very similar (about 30% each). However, while the bulk of provincial resources are executed by the provinces themselves -in the delivery of basic education, health and social assistance services-the central administration of the national level only executes about 20% of the resources it allocates to social spending. The rest are transfers to social security (ANSES) and to provincial governments, as discussed previously. 111.6 Education. Under Argentine law, provincial governments are responsible for the financing, administration and supervision of the provision of all levels of non-university education as well as for the implementation of compensatory and infrastructure programs financed by the national government. The national government is responsible for the financing and regulation of national universities; the coordination, supervision and provision of technical assistance to ensure quality in basic education; and the definition and financing of compensatory policies in basic education to reduce regional and social inequalities. Overall, this distribution of responsibilities is reflected in the distribution of financing of education spending shown in Table HI 4 the national level finances (directly and through transfers to provinces) a little over 20% of total education spending and provinces finance the other 80%. 102 11.7 Consolidated public spending on educationf grew substantially during the 1990s to reach approximately 4.5% of GDP in 2001 - average by international standards, slightly higher that the Latin American average but lower than the OECD average. Basic education (pre-school, primary and secondary) represents approximately 80% of spending, university spending represents approximately 14% with the balance (6%) corresponding to tertiary non-university programs. Salary payments represent over 90% of total education spending. m.8 Once transfers to provinces and national universities are accounted for, spending by the national level represents approximately 1% of total spending. In other words, the national level has a very minor role in the execution of education spending. Transfers to provinces, though, represent a significant addition to provincial education budgets (in the order of 8-9%) suggesting this is the key financing instrument the national level has to influence actions at the sub-national level Table 1m 4 shows, though, that the bulk of those resources go to pay for teacher salaries at the basic and tertiary education levels -in spite of the distribution of functions contemplated in Argentine law. Table EI 4: Education Spending 2000-2001 (In millions, 2001 pesos) 2000 2001 Federal Provinces Total Federal Provinces Total Transfers Transfers Delivery of Education Services 16363 822.5 8914.2 11373.1 1563.5 774.7 9074.2 11412A Payment of Teachers Salaries 14.4 803.5 8425.9 9243.9 11.0 714.3 8805.2 9530.5 Public Schools 649.0 7087.5 7736.5 660.0 7455.2 8115.2 Training Institutes 14.4 154.5 168.9 11.0 54.3 65.3 Private Schools 1337.9 1337.9 1404.3 1404.3 School Maintenance 146.9 146.9 172.7 172.7 Infrastructure and Equipment 55.0 19.0 179.6 253.6 28.7 60.3 96.3 185.3 National Universities 1566.9 1566.9 1523.8 1523.8 Administration and Regulation 51.0 24.2 557.6 632.7 49.8 31.6 588.5 670.0 Central Administration 30.5 541.8 572.3 29.7 569.4 628.4 Information & Evaluation 3.3 6.0 0.7 10.1 3.7 2.4 3.4 9.5 Curriculum & Training 15.0 15.0 6.5 24.8 15.8 22.2 Administrative Reform 6.3 18.2 24.5 3.9 4.4 3.9 Decentralized agencies 10.8 10.8 6.0 6.0 Other Programs 30.3 17.0 108.9 156.3 21.7 5.3 75.5 102.6 Total 1717.6 863.7 9580.7 12162.0 1635.1 811.6 9738.2 12184.9 Source: Ministry of Economy - National Directoriate of Consolidated Social Spending 11.9 Health. Provincial governments are responsible for the financing, management and supervision of health services provided by the public health network consisting of f In this section, spending on scholarships and school lunches by the National Ministry of Education is not classified as education spending but as social assistance. Section 2.1 looks at the budget of the Ministry and in that sense, covers those items as well. 103 hospitals, health centers and other public facilities; and for the implementation of specific programs in public health financed by the national government. The national level is responsible for epidemiological prevention and control; the coordination and financing of maternal child health's policies; and the support to provinces in sanitary emergencies. Overall, as in the case of education, this distribution of responsibilities is reflected in the distribution of financing of health spending shown in Table m 5. The national level finances (directly and through transfers to provinces) less than 15% of total public health spending and provinces finance more than 85%. Table 1I 5: Public spending in health 2000-2001 (In millions, 2001 pesos) 2000 2001 Federal Federal Federal Tr f Provinces Total Federal Transfer Provinces Total Delivery of Health Services 356.1 44.9 3278.7 3679.7 304.5 45.1 3408.8 3758A Providers (hospitals and health centers) 303.6 44.9 3209.6 3558.1 254.8 45.0 3343.0 3642.8 Salaries, supplies and maintenance 218.7 3209.6 3428.3 198.2 3343.0 3541.2 PROFE(*) 84.9 44.9 129.8 56.5 45.0 101.6 Special Drugs Programs (AIDS) 52.5 52.5 49.7 0.1 49.8 Infrastructure and Equipment 69.1 69.1 65.8 65.8 Administration and Regulation 101.8 350.3 452.1 91.1 1.3 363.1 455.5 Central administration 65.6 289.7 355.3 50.7 1.3 304.6 356.6 Training 23.5 47.4 70.9 20.6 46.1 66.7 Planning and Regulation 4.8 4.1 9.0 6.4 4.8 11.2 Research - 5.3 5.3 Health Sector Reform 7.9 9.0 16.9 8.1 7.7 15.7 Public Health Programs 42.0 49.9 20.5 112A 53.9 22.7 27.7 104.3 AIDS and STDs (excludes drugs) 3.2 - 0.6 3.8 3.4 0.8 4.2 Prevention 27.2 6.1 6.1 39.3 38.8 3.7 9.5 51.9 Maternal and Child Health 10.3 43.8 4.0 58.1 10.5 19.0 4.6 34.1 Sanitary Emergencies 1.2 - 9.9 11.1 1.3 12.8 14.1 Other Programs 8A - 73.3 81.6 12.5 0.8 40.1 53.4 Total 508.2 94.7 3722.6 4325.7 462.1 69.9 3839.8 4371.7 (*) Grants to cover the costs of services provided to recipients of non-contributive pensions Source: Ministry of Economy - National Directoriate of Consolidated Social Spending m. 10 Not surprisingly, close to 90% of provincial own resources allocated to health are spent on the costs of running the network of public health facilities. See in Chapter 4 (Health Section) a detailed analysis of the efficiency and equity issues associated with that network. Administration and regulation costs represent close to 10% of the provincial budget (compared to 6% in the case of education) while public health programs (e.g. prevention, control) represent only a small fraction of the provincial health budget. Federal transfers constitute a much less significant contribution to provincial budgets -at least during this period- than in the case of education services, and only half of them (i.e. those related to the maternal and child health program) appear to play a strategic role on the part of the national level. 104 M. 1.1 The correlate of these figures is a relatively small proportion of the national level spending allocated to conventional public health and compensatory programs. Indeed, the bulk of the budget (as shown in Table HI 5) goes to cover the costs of the few remaining national hospitals and decentralized agenciesg and to administrative functions (that represent approximately 20% of spending). m1. 12 Social Assistance. Before the end of the convertibility regime, consolidated spending in social assistance represented a little over 1% of the GDP in Argentina. As shown in Table ff 6 the federal/provincial breakdown is somewhat different to that found for education and health: the national share is approximately 40% of the total. The split between cash and in-kind subsidies, though, is substantially different at the two levels: national spending is mostly (80-85%) in cash and provincial spending is less so (40%). This pattern is not surprising to the extent that provinces would naturally tend to specialize in assistance programs that are more intense in promotion activities while the national level would naturally tend to concentrate on income redistribution activities. Table II 6: Public Spending in Social Assistance 2000-2001 (In millions, 2001 pesos) 2000 2001 Program Federal Transfers Provinces Total Federal Transfers Provinces Total Cash Transfers 905.6 0.7 855.7 1762.0 998.4 87.2 7433 1829.0 Work-fare programs 187.7 0.7 282.4 470.8 177.5 87.2 392.9 657.6 Scholarships 55.9 162.8 218.6 45.2 27.9 73.1 Non-Contributive Pensions 661.8 137.3 799.1 755.7 140.7 896.4 Other cash subsidies 0.3 273.2 273.5 19.9 181.8 201.7 In-kind Transfers 170.3 105.2 1014.0 1289.4 115.1 58.5 1074.0 1247.5 Feeding Programs 45.0 1.5 688.0 734.5 30.1 4.2 670.4 704.7 Others 125.3 103.6 326.0 554.9 85.0 54.3 403.6 542.9 Central Administration . 33.4 101.5 134.9 67.1 0.8 108.0 176.0 Total 1109.2 105.9 1971.3 3186.4 1180.6 146.5 1925.3 3252.4 Source: Ministry of Economy - National Directoriate of Consolidated Social Spending 11. 13 Cash subsidies, though, appeared to be heavily concentrated on the elderly, with the almost complete absence of child-related cash subsidies being perhaps one of the most noticeable aspects. In kind transfers show a very strong concentration on feeding programs (school lunches, community kitchens and provision of food supplies) a rather unusual policy choice for a country of Argentina's characteristics. Previous studies (World Bank [2000], [2001]) analyzed the composition, targeting and adequacy of safety net programs in the convertibility era in Argentina. Overall, programs were found to be fragmented, have relatively good targeting (with the exception of a sub-group of the non- contributory pensions) but suffering from design problems that limited their effectiveness. 9 The hospitals are Garrahan, Posadas and Somnner. The decentralized agencies are ANLIS, ANMAT, INCUCAI, Colonia Montes de Oca, among others. 105 IE. 14 Social Insurance. As mentioned earlier, over 60% of social spending is executed by the social insurance system: old age pensions, social health insurance, unemployment insurance and family allowances. A system that suffers from a variety of problems the most critical one, perhaps, being that it covers a small and falling share of the population -and at a high financial cost to beneficiaries and to the state. Argentina has not been able to make a system based on payroll taxes and contributions work effectively, to a great extent due to the growing levels of informality in labor contracts compounded by lack of enforcement capacity on the part of the state. 1m. 15 As shown in Table m 7 the largest spending item is old age pensions. Chapter 4 (Pensions Section) analyzes in detail the characteristics of Argentina's complex pension system. Here, it suffices to mention a few key facts. First, in spite of a significant (and costly) process whereby some provincial pension systems were transferred to the national level, about 20% of the pensions paid by the public sector are paid at the provincial level. Second, general revenues cover over 60% of the pension payments from the national level. As will be shown in Section 4, only a fraction of this financing needs are related to the transition to a fully funded, privately run, pension system. In that sense, the heavy reliance on general revenues to finance pensions to an increasingly smaller group of elderly, raises equity concerns. Table II 7: Social Insurance Spending 2000-2001 (In millions, 2001 pesos) Program 2000 2001 Federal Transfers Provinces Total Federal Transfers Provinces Total Pensions 16107.8 165.3 4094.3 20367.4 15690.2 142.3 4378.1 20210.6 Subsidies to pensioners 623.2 623.2 609.0 609.0 Family Allowances 1834.0 1834.0 1808.0 1808.0 Unemployment Insurance 297.2 297.2 353.7 353.7 Health Insurance 6603.5 5.0 2192.7 8801.2 6363.0 4.1 2229.8 8596.9 Administration 288.1 87.0 375.1 249.1 89.1 338.2 Total 25753.8 170.3 6374.1 32298.2 25073.0 146.4 6696.9 31916.3 Source: Ministry of Economy - National Directoriate of Consolidated Social Spending 1.16 Health insurance involves three main sub-systems: (i) the national health insurance system that encompasses about 300 union-run insurance funds (obras sociales nacionales) representing 46% of spending; (ii) the insurance programs for provincial public sector employees (obras sociales provinciales) representing 26% of spending and (iii) PAMI, the health insurance fund for pensioners, representing 28% of spending. PAMI is financed through payroll taxes and contributions. However, the national treasury has been repeatedly called upon to cover PAMI's deficits both on and off budget. In principle, both obras sociales nacionales and provinciales are fully funded through payroll taxes/contributions. However, in the case of obras sociales provinciales their autonomy is less than full in many cases and there are many (hidden) subsidies from the provincial treasury. In the case of the obras sociales nacionales, there is also a pattern of bailouts and subsidies that is not recorded through the budget (the basis for a table such as Table II 7). Chapter 4, (Health Section) reviews in some depth the functioning of the social health insurance system. 106 Social sector spending in the post-convertibility period m1. 17 At the national level, the post-convertibility period is characterized by two main changes: (i) a significant increase in the resources allocated to social assistance and to a lesser extent;to health as part of the Government's social emergency program; and (ii) a freeze in nominal terms in practically all other expenditures, particularly pensions and salaries. The consequence of these changes was a radical shift in the composition of social sector spending in the context of a falling envelope (both in real terms and as a percentage of GDP, as shown in Table ImI 8). Namely, a reduction in the share of spending going to education (given the sharp reduction in real wages that followed the collapse of convertibility) and pensions (also due to the inflationary effects of the devaluation). This trend is estimated to continue in 2003. Based on budget figures, federal education spending is expected to fall to 0.44% of GDP and social insurance spending to 5.3% of GDP while health spending is estimated to remain constant and social assistance to experience a small reduction both as a share of GDP.h 11. 18 This spending trend reverses a pattern observed in previous crises in Argentina when social spending was not protected from cuts. Indeed, there was sizable co- movement, with indications that (if anything) social spending was more volatile than other types of spending. During this period social assistance spending was more vulnerable to fiscal contraction. This changed in the period 2002-3, during the sharp economic downturn that accompanied the abandonment of the convertibility plan in early January 2002 as a result of the implementation of the social emergency program which implied an increase of social assistance spending of 90% in real terms and an increase of 35% in real terns in four key priority public health programs (i.e. disease control and prevention, maternal and child health, AIDS/HIV, and health emergencies). 111.19 Table m 9 shows the key changes in social assistance spending'. The sharp expansion in social assistance spending is closely associated with the establishment of Jefes de Hogar, a large and ambitious work-fare program, that represented 60% of total social assistance expenditures in 2002. In addition, the national government created a new feeding program (PEA), a new cash transfer program (Familias, previously known as IDH), expanded its scholarship programn and offered financial support to provinces for their school lunch programs. h The national government's approved budget for 2003 includes an increase of 18% in the total spending in human development in nominal terms. However, taken the expected inflation rate for 2003 implies a new reduction of 15% in real terms. However, official estimates appear to have over-estimated inflation and thus the real reduction is likely to be less. See Tables C. I l-C. 14 for a detailed description of the rest of the Sectors. 107 Table 1I 8: Public spending in the social sectors after Convertibility (a) In millions of 2001 pesos 2001 2002 2002/ Federal Transfers Total Federal Transfers Total 2001 (%) Education 1,635 812 2,447 1,071 56 1,127 (53.92) Health 462 69 531 479 24 504 (5.22) Social Assistance 1,181 147 1,327 2,331 200 2,531 90.72 Social Insurance 25,073 146 25,219 16,673 155 16,828 (33.27) Health 6,363 4 6,367 4,289 4,289 (32.65) Others 18,710 142 18,852 12,385 155 12,540 (33.48) Total 34,714 1,177 35,891 24,843 435 20,990 (41.52) (b) In % of GDP 2001 2002 2002/ Federal Transfers Total Federal Transfers Total 2001 (%) Education 0.61 0.30 0.91 0.52 0.03 0.55 (39.89) Health 0.17 0.05 0.22 0.23 0.01 0.24 9.63 Social Assistance 0.44 0.05 0.49 1.13 0.09 1.22 148.98 Social Insurance 9.33 0.05 9.38 8.10 0.08 8.17 (12.92) Health 2.36 0.00 2.36 2.39 0.00 2.39 1.27 Others 6.07 0.05 7.02 5.71 0.08 5.79 (17.58) Total 10.55 0.46 11.01 9.95 0.24 10.18 (7.49) Source: Ministry of Economy - National Directoriate of Consolidated Social Spending 111.20 As a result of these changes, there have been two significant changes in the structure of social assistance programs. Even though consolidated national and provincial spending figures are not available for 2002, the available evidence on provincial spending (see below) indicates that the national level is now the main source of financing as well as the primary executing actor for social assistance programs. The second change is a radical shift away from in-kind subsidies to cash subsidies. 108 Table m 9: Public spending in social assistance - Federal Government (In millions, 2001 pesos) 2001 2002 2003 (budget) Federal Transfers Total Federal Transfers Total Federal Transfers Total Cash Transfers 998A 87.2 1085.6 2143.3 0.4 2143.8 2245.0 23.5 2268.5 Employment Prograrns 177.5 87.2 177.5 1536.6 0.0 1536.6 1767.9 1767.9 Scholarships 45.2 90.3 132.1 132.1 74.6 74.6 Non Contributive Pensions 755.7 775.7 422.9 422.9 318.9 318.9 Other cash subsidies 19.9 19.9 51.8 0.4 52.2 83.6 23.5 107.1 In-kind Transfers 115.1 58.5 173.6 131.1 195.9. 327.0 123.9 188.7 312.6 Feeding Programs 30.1 4.2 34.3 85.2 194.3 279.5 76.3 185.0 261.3 Others 85.0 54.3 139.3 45.9 1.6 47.4 47.6 3.6 51.2 Administration . 67.1 0.8 67.9 56.9. 3.4 60.2 49.5 5.7 55.2 Total 1180.6 146.5 1327.1 2331.3 199.7 2531.0 2418.4 217.9 2636.3 Source: Ministry of Economy - National Directoriate of Consolidated Social Spending 111.21 We assess changes in social sector spending at the provincial level using preliminary expenditure figures for seven provinces which represent a little over 60% of total provincial spending. The data for these provinces shows a very similar pattern of change to that found for federal spending with constant nominal spending resulting in sharp reductions in real spending. The key difference, though, is that in the case of provinces spending on social assistance followed the same pattern and fell by over 30% in real terms. The reduction in real spending in education and health is directly associated with the fall in real wages and is thus not surprising. In the case of health, though, given the significant increases in the cost of medical inputs, this reduction may in part be resulting in reduced input availability. Table m 10: Provincial Social Sector Spending Buenos Aires, City of Buenos Aires, Cordoba, Chubut, Neuqudn, Rio Negro and San Juan - 2000-2002 (*) (In millions, 2001 pesos) 2000 2001 2002 2002/2001 Education 6,250.14 6,427.06 3,904.70 (39.25) Health 2,458.76 2,474.87 1,633.57 . (33.99) Social Assistance 1,407.66 1,267.73 872.2 (31.20) Social Insurance 4,018.07 4,108.43 2,684.30 (34.66) Health 1,283.98 1,265.01 840.3 (33.57) Others 2,734.09 2,843.42 1,844.00 (35.15) Total 14,134.63 14,278.09 9,094.77 (36.30) (*) Includes federal transfers executed by the provinces Source: Ministry of Economy - National Directoriate of Consolidated Social Spending 111.22 Table III 11 provides more details on provincial spending in social assistance. This is of particular importance given the significant expansion in federal social 109 assistance spending during the period. How did provinces react to the expansion of programs with national financing? In the case of cash transfers, there was a sharp reduction in real spending at the provincial level. In the case of work-fare and scholarships there was a reduction in nominal terms. For work-fare, the new level of spending is similar, in nominal terms, to that observed in 2000. Considering that in 2001 there was a special earmarked transfer to provinces to finance work-fare programs and that the value of subsidies paid to beneficiaries remained constant in nomninal terms, this appears to be evidence that provinces maintained the historical coverage of this programs. In the case of scholarships, the sharp reduction in provincial spending took place already in 2001 when the Province of Buenos Aires discontinued its program. In 2002, the national level incorporated students in Buenos Aires into its own program. Table M 11: Provincial Social Assistance Spending Buenos Aires, City of Buenos Aires, Cordoba, Chubut, Neuqudn, Rio Negro and San Juan - 2000-2002 (*) (In millions, 2001 pesos) Program 2000 2001 2002 2002/2001 Cash Transfers 564.1 469.7 220.7 (53.00) Work-fare Programs 212.9 269.3 140.1 (48.00) Scholarships 152.2 12.6 7.5 (40.40) Non Contributory Pensions (PN 103.4 100.4 66.2 (34.10) Other cash subsidies 95.6 87.4 6.9 (92.10) In-kind transfers 771 722.6 596 (17.50) Feeding programs 513.5 462.4 395.2 (14.50) School Lunches 205.1 182.4 147.9 (18.90) Other feeding programs 308.4 279.9 247.3 (11.70) Others in-kind programs 257.5 260.2 200.8 (22.80) Administration 72.5 75.4 55.5 (26.40) Total 1,407.66 1,267.73 872.2 (31.20) (*) Includes federal transfers executed by the provinces Source: Ministry of Economy - National Directoriate of Consolidated Social Spending m.23 In the case of in-kind programs (particularly feeding programs) there was an increase in nominal spending of a magnitude similar to the transfers received from the national government under the PEA and school lunches program. In other words, although more detailed, provincial-level analysis, is needed there is no evidence of crowding out by the national feeding programs. 110 ANNEX IV: APPROACHES TO THE MANAGEMENT OF CONTINGENT LIABILITIES, COMMITMENTS, AND OTHER FISCAL RISKS IV.1 Improvements in future fiscal policy require, among other things, improvements in government decisions about whether to take on new risks and about how to manage existing risks. Governments can make good decisions even if the surrounding institutions of public-expenditure management are poor and they can make bad decisions even if that framework is good. But perhaps the only reliable way of seeking to improve the average quality of decisions is to improve the framework. Two aspects of the framework are important: first, the rules governing the government's accounting and financial reporting and, second, the rules governing its decision-making, particularly in the context of the annual budget. Before turning to these two issues, however, we need to understand what sort of information can in principle be taken into account in financial reporting and decision making. Measurement of risks IV.2 The government will tend to make better decisions about taking on and managing risks if it can measure the risks. There are several different ways the risks can be measured. IV.3 First, the maximum possible expenditure arising from a contingent liability, commitment, or other fiscal risk can sometimes be estimated. In the case of bank-deposit insurance, the maximum possible loss per depositor (ignoring possible implicit liabilities) is 30,000 pesos. Likewise, the maximum possible loss from a toll-road minimum- revenue guarantee is the amount the government would pay in the (unlikely) event that no vehicles at all used the road. Sometimes, however, the government's potential losses may be uncapped or just indeterminable. IV.4 Second, the government can estimate the expected cost of exposure to the risk, either in sum or in each of several periods (e.g. for each of the next three years). These estimates require the government to take into account the probabilities of different amounts of loss. For example, to calculate expected losses under the minimum-revenue guarantee, the government could estimate the respective probabilities of various amounts of traffic and calculate required guarantee payments, if any, for each amount. Making use of the same information, the government can also estimate the probability distribution of payments-to arrive at statistics such as the 95th percentile of the distribution, i.e. an amount the government can expect its expenditure to be within 95% of the time. IV.5 Third, taking into account the timing and risk-characteristics of possible payments, the government may be able to estimate the present value of the possible losses. For a given expected payment, the present value will be lower if the payment is more distant in time and, in general, higher if it is subject to more risk. An estimate of the present value of the exposure to the risk has the advantage of being commensurable with immediate cash expenditure-for example, a direct subsidy given to the toll-road concessionaire. IV.6 By saying that governments can produce these measurements, we don't mean to imply that it is always sensible to do so. Some estimates will be straightforward and others may be very difficult to make with any accuracy. The government needs to make judgments about what to measure and what to leave unquantified, taking into account the difficulty of measurement and the apparent magnitude of the risk. Sometimes, the best approach may be to rely on ballpark estimates, rather than to develop complex and expensive (and perhaps still not very reliable) models. IV.7 Many governments have made progress in measuring their exposure to risk under contingent liabilities. The New Zealand government calculates and reports, for example, the maximum amounts it could be required to pay to international financial institutions such as the World Bank as a result of the guarantees it provides them as a shareholder (they represent the government's biggest recorded exposure). The United States government has done a lot of work calculating expected losses under loan-guarantee and deposit-insurance schemes. The Colombian and Chilean governments have estimated expected losses and their present values resulting from guarantees given in the context of private infrastructure projects. Accounting for risks IV.8 Governments will tend to make better decisions if they measure the major quantifiable risks to which they are exposed. But what can be done to ensure they perform these measurements? One way of institutionalizing them is to ensure that the standards the government follows in preparing end-of-year financial reports require such measurements to be made and disclosed where appropriate. The challenge in choosing or designing such standards is of course to achieve an appropriate balance between providing useful information and not requiring too much costly estimation. IV.9 One approach to this trade-off is to adopt standards based on those followed by private-sector firms (local or United States "generally accepted accounting principles" or the International Accounting Standards prepared by the International Accounting Standards Board). Governments now have the option of following the "International Public Sector Accounting Standards", produced by the International Federation of Accountants, and derived from International Accounting Standards. IV.10 Adopting such standards has the advantage of automatically subjecting the government to rules designed to require firms-which face similar incentives to disguise expenditure and liabilities-to provide useful information about their financial performance and financial position-all the while taking into account the costs as well as the benefits of making and publishing estimates. The governments of Australia, Canada, New Zealand, the United Kingdom, and the United States are among those that have adopted accounting standards more or less consistent with generally accepted accounting principles for firms. IV. 11 These standards would generally require the government to: * Recognize as a liability any probable expenditure that can be reliably measured, 112 * Disclose the existence of contingent liabilities in a statement of contingent liabilities, c Include some long-term commitments, such as certain ("financial" or "capital") leases, as debt on the government's balance sheet and disclose the existence of others in a statement of commitments, and * Provide quantitative information on the government's exposure to various risks (such as those relating to changes in interest or exchange rates). V. 12 Yet when losses are merely possible or difficult to measure, such standards typically do not require a liability to be recognized. As a result, they typically wouldn't require the government to incorporate in the reported budget deficit the cost of, say, increasing bank-deposit insurance. IV. 13 Thus even if the government followed such standards, it might reasonably want to go beyond compliance with them and to provide-and require itself through legislation to provide-further information, such as an estimate of the expected losses over each of the next three years resulting from all material quantifiable contingent liabilities. 113 STATISTICAL APPENDIX Table of Contents A. SOCIO-ECONOMIC INDICATORS Table Al: Socio-Economic Indicators by Province B. FISCAL Table B I: Federal Government Operations Table B2: Federal Public Sector Debt by Instrument and Duration (end of period) Table B3: National Revenue Table B4: National Expenditure by Sector (a): As Percent of GDP Table B5: Provincial Government Operations Table B6: Provincial Expenditure by Sector (24 Jurisdictions) Table B7: Consolidated Provincial Debt (24 Jurisdictions) Table B8: Consolidated Public Expenditure at Current Prices (a): As Percent of GDP (b): As Percent of Total Table B9: Provincial Tax Revenue in Thousands of Pesos (Average 1993-2001) (a): As Percent of Total (Average 1993-2001) (b): As Share of Provincial GDP Table B 10: Provincial Revenues and Employment C. SOCIAL SECTORS Table Cl: Consolidated Public Expenditures in Education Table C2: Evolution of Education Provincial Spending Table C3: Provincial Education Expenditures- Economic Classification in Year 2000 Table C4: Spending on Education Indicators in Year 2001 Table C5: Elementary Level Indicators in Year 2001 Table C6: Secondary Level Indicators in Year 2001 Table C7: Basic Indicators of Secondary Education by Income levels in Year 1998 Table C8: Financing and Spending of National Universities in Year 2000 Table C9: Financing of National Universities in Year 2000 Table CIO: Enrolled and Graduating Students, National Universities Table C 11: Education Sector - Federal Administration Table C 12: Public Health Sector - Federal Administration Table C13: Social Assistance Sector - Federal Administration Table C14: Social Insurance - Federal Administration Table C15: Jefes Program: Number of Beneficiaries Figure Cl: Relation between per Capita GDP and Spending per Student Figure C2: Public Spending per Student and Results in Academic Evaluations 114 D. INFRASTRUCTURE Table DI: Public Private Allocation of Responsibilities in the Infrastructure Sectors Table D2: Jurisdictional Responsibility for Infrastructure Sectors Table D3: Consolidated Public Expenditure on Infrastructure Sectors Table D4: Summary of Major Renegotiations for Transport and Water Concessions Table D5: Summary of Renegotiation Episodes for Inter-Urban Road Concessions Table D6: Summary of Efficiency Gains and Extent of Pass Through to Tariffs Table D7: Status of Renegotiation of Infrastructure Concessions Table D8: Status of ICSID Cases already Filed against Argentine Government Table D9: Breakdown of Infrastructure Spending by Budgetary Mechanism Table D1O: Evolution of Energy and Transport Subsidies Table DI1: Summary of Social Policies for Infrastructure Services Table Dl 2: Comparativ'e Analysis of Social Tariff Proposals Table Dl 3: Comparison of Design Features of Regulatory Agencies at the Federal Level Table D14: Comparison of Design Features of Regulatory Agencies at the Provincial Level Table Dl 5: Comparison of Special Funds for Infrastructure at the Federal Level Table Dl 6: A Brief History of Road Funds in Argentina Table DI 7: Evolution of Transport Infrastructure Trust Fund Figure Dl (a):Breakdown of Public Expenditure on Infrastructure by Sector Figure Dl(b): Breakdown of Public Expenditure on Infrastructure by Jurisdiction Figure D2: Evolution of Income Stream for Inter-Urban Roads Concessions Figure D3(a): International Comparison of Residential Electricity Prices Figure D3(b): International Comparison of Residential Gas Prices Figure D3(c): International Comparison of Residential Water Prices Figure D3(d): International Comparison of Residential Telephone Prices Figure D4(a): Taxes as a Proportion of Household Electricity Bills Figure D4(b): Taxes as a Proportion of Household Gas Bills Figure D4(c): Taxes as a Proportion of Household Water Bills Figure D5(a): Illustration of flow of funds to electricity sector Figure D5(b): Dlustration of flow of funds to transport sector Figure D6: Allocation of Resources from Electricity Tariff Compensation Fund Figure D7: Allocation of Resources from Transport Infrastructure Trust Fund E. PUBLIC EXPENDITURE MANAGEMENT Table El: Budget Execution: Modifications and Actual Expenditure Table E2: Rigidities in National Administration Expenditure (2003 Budget) Table E3: Emergency Decrees 2002 115 A. SOCIO-ECONOMIC INDICATORS Table Al: Socio-Economic Indicators by Province Share of GDP per Infant Public Household Household Population GDP Capita Mortality Employment Indigence Poverty (thousands) % (pesos) (per thousand) (per thousand) % % 2001 2000 2001 2001 1999 2002 2002 Nation 37,487 100 7,352 16.3 36.0 22.4 49A Buenos Aires - City 3,050 24.2 22,276 9.6 36.9 3.7 14.6 Buenos Aires - Province 14,382 38.4 6,806 15.0 29.8 21.3 50.9 Catamarca 324 0.4 3,899 15.5 73.4 21.8 53.5 C6rdoba 3,122 7.1 6,983 16.2 24.4 19.5 45.2 Corrientes 935 1.0 3,446 23.5 39.6 33.1 63.4 Chaco 963 0.9 3,157 24.0 43.8 32.8 61.3 Chubut 458 1.2 6,438 13.1 51.1 16.8 36.7 Entre Rios 1,122 2.0 5,002 14.9 48.3 29.3 59.2 Formosa 516 0.3 2,440 28.9 65.3 30.0 58.1 Jujuy 614 0.6 3,553 18.4 48.2 28.0 63.2 La Pampa 311 0.9 6,545 12.4 55.3 14.3 39.4 LaRioja 287 0.6 5,231 23.5 81.0 21.0 54.4 Mendoza 1,627 2.5 6,887 12.1 38.9 21.6 48.8 Misiones 1,018 1.6 3,846 19.6 34.9 32.5 59.2 Neuquen 582 1.6 13,356 13.0 62.8 19.0 42.8 Rio Negro 631 1.4 6,461 14.7 46.5 19.2 41.4 Salta 1,090 1.2 3,729 19.1 34.5 32.2 59.4 San Juan 583 1.0 4,636 18.8 48.8 26.4 58.8 San Luis 372 1.7 10,307 17.8 45.0 19.8 51.1 Santa Cruz 212 1.2 9,758 14.5 81.8 7.5 25.7 Santa Fe 3,129 7.6 7,319 14.3 33.4 22.3 50.2 Santiago del Estero 731 0.5 2,321 14.8 47.7 25.3 57.0 Tierra del Fuego 121 0.7 28,331 10.1 67.0 13.3 30.1 Tucuman 1,308 1.5 3,715 24.5 35.2 26.6 61.0 Sources: Federal Council for Investments, MECON, and National Institute of Statistics and Census. 116 B. FISCAL Table Bi: Federal Government Operation 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Millions of Pesos) OVERALL BALANCE 2,730 -286 -1,373 -5,264 4,277 -4,074 4,768 -6,792 -8,719 4,549 Revenue 50,727 51,078 50,294 47,669 55,377 56,726 58,455 56,571 51,319 55,824 Tax Revenue 29,007 31,614 31,035 33,176 38,352 40,363 38,626 40,672 37,165 40,034 Social Security Contributions 13,345 14,084 13,705 10,281 12,202 11,990 10,892 10,684 9,640 9,710 Non-Tax Revenue 2,709 2,910 2,965 2,102 2,241 2,657 6,251 2,487 2,261 2,798 Other 5,665 2,470 2,589 2,110 2,581 1,716 2,687 2,727 2,254 3,281 Expenditure 47,996 51,364 51,667 52,933 59,653 60,800 63,224 63,362 60,038 60,374 Non-Interest Expenditure 45,082 48,214 47,583 48,325 53,908 54,139 55,000 53,706 49,863 53,564 Discretionary (exd. transfers to provinces) 31,247 33,113 32,862 32,219 35,934 35,481 36,869 35,414 33,725 36,911 Wages 7,627 7,642 7,250 7,213 7,554 6,844 7,354 6,899 6,542 6,974 Goods & Services 4,084 2,445 2,482 2,473 2,660 2,477 2,878 2,315 2,243 2,545 Private Transfers + Pensions 15,763 19.809 20,307 19,889 22,900 23,390 23,847 23,587 22,436 25,231 PrivateTransfers 3,250 4,568 4,680 4,445 5,700 5,910 6,411 6,155 5,819 8,696 Pensions 12,513 15,241 15,628 15,444 17,199 17,481 17,436 17,431 16,617 16,535 Other Current Expenditure 2,049 1,895 1,895 1,850 1,822 1,827 2,104 2,179 2,081 1,862 Capital Expenditure 1,723 1,322 927 795 999 943 686 435 423 298 Provindal Transfers 13,836 15,100 14,719 16,105 17,972 18,657 18,131 18,291 16,137 16,653 Primary Balance 5,644 2,864 2,710 -657 1,468 2,587 3,455 2,864 1,455 2,260 Interest 2,914 3,150 4,084 4,608 5,745 6,660 8,224 9,656 10,175 6,810 (Percent of GDP) OVERALL BALANCE 1.2 -0.1 -05 -1.9 -1.5 -1.4 -1.7 -2.4 -3.2 -15 Revenue 21.4 19.8 19.5 175 18.9 19.0 20.6 19.9 19.1 17.8 Tax Revenue 12.3 12.3 12.0 12.2 13.1 13.5 13.6 14.3 13.8 12.8 Social Security Contributions 5.6 5.5 5.3 3.8 4.2 4.0 3.8 3.8 3.6 3.1 Non-Tax Revenue - 1.1 1.1 1.1 0.8 0.8 0.9 2.2 0.9 0.8 0.9 Other 2.4 1.0 1.0 0.8 0.9 0.6 0.9 1.0 0.8 1.0 Expenditure 20.3 20.0 20.0 19.5 20.4 20.3 22.3 22.3 22.3 1923 Non-Interest Expenditure 19.1 18.7 18.4 17.8 18A 18.1 19A 18.9 18.6 17.1 Discretionary (excl. transfers to provinces) 13.2 12.9 12.7 11.8 12.3 11.9 13.0 12.5 12.6 11.8 Wages 3.2 3.0 2.8 2.7 2.6 2.3 2.6 2.4 2.4 2.2 Goods & Services 1.7 0.9 1.0 0.9 0.9 0.8 1.0 0.8 0.8 0.8 PrivateTransfers+Pensions 6.7 7.7 7.9 7.3 7.8 7.8 8.4 8.3 8.3 8.1 Private Transfers 1.4 1.8 1.8 1.6 1.9 2.0 2.3 2.2 2.2 2.8 Pensions 5.3 5.9 6.1 5.7 5.9 5.8 6.1 6.1 6.2 5.3 Other Current Expenditure 0.9 0.7 0.7 0.7 0.6 0.6 0.7 0.8 0.8 0.6 Capital Expenditure . 0.7 0.5 0.4 0.3 0.3 0.3 0.2 0.2 0.2 0.1 Provincial Transfers 5.8 5.9 5.7 5.9 6.1 6.2 6A 6.4 6.0 5.3 Primary Balance 2.4 1.1 1.1 -0.2 0.5 0.9 1.2 1.0 0.5 0.7 Interest 1.2 1.2 1.6 1.7 2.0 2.2 2.9 3A 3.8 2.2 Source: ,MECON 117 Table B2: Federal Public Sector Debt by Instrument and Duration (end of period) 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Millions of Pesos) (1) TOTAL DEBT 69,626 80,678 87,091 97,105 101,101 112,357 121,877 128,018 144,453 486,728 Medium and long term 69,626 80,678 87,091 97,105 101,101 109,062 117,703 122,910 137,707 459,104 Bilateral 9,653 10,966 11,614 10,162 8,104 7,455 5,918 4,561 4,477 16,286 Paris Club 7,433 7,978 8,038 6,725 5,144 4,512 3,235 2,375 1,879 6,790 Other bilateral institutions 2,220 2,988 3,576 3,437 2,960 2,943 2.682 2,186 2,598 9,496 Multilateral 11,005 11,773 15,384 16,367 16,790 19,122 20,311 21,764 32,362 116,494 IDB 3,690 3,687 4,791 4,756 5,050 6,259 7,208 7,622 8,704 31,312 IBRD 3,573 3,760 4,462 5,316 5,816 7,417 8,596 9,054 9,673 32,208 IMF 3,742 4,326 6,120 6,279 5,908 5,420 4,472 5,053 13,952 52,861 Others 0 0 11 15 17 26 36 36 33 114 Other creditors (2) 488 518 437 283 731 628 641 1,045 1,537 77,850 Commercial banks 984 1,064 1,316 1,452 1,423 3,646 5,029 2,461 2,015 4,578 Bonds and public securities 47,496 56,357 58,340 68,841 74,054 78,211 85,804 93,079 55,057 243,895 In domestic currency 5,567 8,400 5,882 8,168 9,500 7,800 6,390 4,410 1,505 15,438 In foreign currency 41,929 47,957 52,458 60,673 64,554 70,411 79,414 88,668 53,552 228,457 Short term (3) 0 0 0 0 0 3,295 4,174 5,108 6,746 27,625 (Percent of GDP) TOTAL DEBT 29.4 31.3 33.8 35.7 34.5 37.6 43.0 45.0 53.8 155.5 Medium and long term 29A 31.3 33.8 35.7 34.5 36.5 41.5 43.2 51.2 146.7 Bilateral 4.1 4.3 4.5 3.7 2.8 2.5 2.1 1.6 1.7 5.2 Paris Club 3.1 3.1 3.1 2.5 1.8 1.5 1.1 0.8 0.7 2.2 Other bilateral institutions 0.9 1.2 1.4 1.3 1.0 1.0 0.9 0.8 1.0 3.0 Multilateral 4.7 4.6 6.0 6.0 5.7 6.4 7.2 7.7 12.0 37.2 IDB 1.6 1.4 1.9 1.7 1.7 2.1 2.5 2.7 3.2 10.0 IBRD 1.5 1.5 1.7 2.0 2.0 2.5 3.0 3.2 3.6 10.3 IMF 1.6 1.7 2.4 2.3 2.0 1.8 1.6 1.8 5.2 16.9 Others 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Other creditors (2) 0.2 0.2 0.2 0.1 0.2 0.2 0.2 0.4 0.6 24.9 Cormercial banks OA 0.4 0.5 0.5 0.5 1.2 1.8 0.9 0.8 15 Bonds and public securities 20.1 21.9 22.6 25.3 25.3 26.2 30.3 32.8 20.5 77.9 In domestic currency 2.4 3.3 2.3 3.0 3.2 2.6 2.3 1.6 0.6 4.9 In foreign currency 17.7 18.6 20.3 22.3 22.0 23.6 28.0 31.2 19.9 73.0 Short term (3) 0.0 0.0 0.0 0.0 0.0 1.1 1.5 1.8 2.5 8.8 (1) Financial and non-Jinancial sectors' internal and external instrumented debt. Thefinancial sector debt includes the debtfrom ihe Central Bank and government banks with the guarantee of the Federal Government. Stock of debt at end of period. (2) Includes loans and open creditsfor the provinces according to Decree N' 1.023/95. (3) Operations conducted according to Law 24.156. Article 82. which authorizes the Treasury Department to issue drafts to payfor seasonalities that have to be reimbursed before the period ends. Source: MECON 118 Table B3: National Revenue 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Millions of Pesos) TOTAL 40,619 44,051 44,029 43,920 48,527 50,037 47,643 49,102 45,402 50,475 Production and consumption taxes 19,743 20,550 18,498 22,153 25,332 25,569 24,005 24,334 23,282 25,090 VAT 16,267 17,366 17,442 18,814 20,474 20,857 18,771 19,009 15,351 15,242 Shared taxes 2,100 2,146 2,060 1,691 1.537 1,482 1,475 1,528 1.621 1,715 Fuels Gasoline 2,059 2,069 1 2,199 2,444 2,325 2,177 2,154 1,821 1,946 Other 0 0 0 139 1,483 1,367 1,411 1,325 1,599 2,454 Interests received 0 0 0 0 0 0 695 855 445 57 Stamp duties 150 99 47 44 47 57 50 46 44 40 Check tax 0 0 0 0 0 0 0 0 2,933 4,857 Refunds (833) (1,129) (1,052) (733) (654) (520) (574) (583) (532) (1,222) Income taxes 4,353 5,904 6,336 6,902 8,440 9,580 10,068 11,137 11,003 9,724 Taxes on personal goods 197 228 1,845 709 558 846 609 1,079 813 570 Trade taxes 2,468 2,800 2,028 2,253 2,834 2,804 2,303 2,008 1,650 6,328 Import duties 1,289 1,563 1,762 1,897 2,435 2,694 2,227 1,938 1,540 1,266 Export taxes 25 31 50 28 6 28 25 32 52 5,022 Other 1,154 1,206 216 329 392 82 51 38 57 41 Social security taxes 12,741 13,431 13,926 10,411 10,551 10,268 9,141 8,998 8,044 8,183 Other taxes 1,117 1,138 1,396 1,491 813 971 1,517 1,546 610 580 (Percent of GDP) TOTAL 17.2 17.1 . 17.1 16.1 16.6 16.7 16.8 17.3 16.9 16.1 Production and consumption taxes 8.3 8.0 7.2 8.1 8.6 8.6 8.5 8.6 8.7 8.0 VAT 6.9 6.7 6.8 6.9 7.0 7.0 6.6 6.7 5.7 4.9 Shared taxes 0.9 0.8 0.8 0.6 0.5 0.5 0.5 0.5 0.6 0.5 Fuels Gasoline 0.9 0.8 0.0 0.8 0.8 0.8 0.8 0.8 0.7 0.6 Others 0.0 0.0 0.0 0.1 0.5 0.5 0.5 0.5 0.6 0.8 Interests received 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.3 0.2 0.0 Stamp duties 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Check tax 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.1 1.6 Refunds -0.4 -0.4 -0.4 -0.3 -0.2 -0.2 -0.2 -0.2 -0.2 -0.4 Income taxes 1.8 2.3 2.5 2.5 2.9 3.2 3.6 3.9 4.1 3.1 Taxes on personal goods 0.1 0.1 0.7 0.3 0.2 0.3 0.2 OA 0.3 0.2 Trade taxes 1.0 1.1 0.8 0.8 1.0 0.9 0.8 0.7 0.6 2.0 Impon duties 0.5 0.6 0.7 0.7 0.8 0.9 0.8 0.7 0.6 0.4 Export taxes 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 Other 0.5 0.5 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 Social security taxes 5A 5.2 5A 3.8 3.6 3A 3.2 3.2 3.0 2.6 Other taxes 0.5 OA 0.5 0,5 03 0.3 05 0.5 0.2 0.2 Source: MECON 119 Table B4: National Expenditure by Sector 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Millions of Pesos) TOTAL 38,676 42,544 44,273 44,127 46,864 47,527 50,011 49,770 49,119 47,299 Functioning of the State 5,993 6,97S 6,944 6,763 6,767 6,906 7,919 6,973 6,260 6,840 General administration 2.444 2.878 2,805 2.672 2.655 2.772 3.645 2,784 2,242 2,531 Jusdce 499 657 688 692 757 782 799 807 790 809 Defense and security 3.050 3,443 3.452 3,399 3,354 3.352 3,475 3.382 3.227 3.500 Social public expenditure 24,165 28,447 29,305 30,260 31,151 31506 31,565 30,983 29,678 32,451 Education. cui and science. and tecnmology 2.184 2.445 2.567 2.734 2.987 3.139 2.923 2.793 2.625 2,760 Basiceducation 109 188 209 294 410 396 240 153 107 275 Tertiaty asd university education 1.387 1.455 1.624 1.738 1,757 1,886 1.876 1.917 1.880 1,850 Other 688 803 734 702 820 857 807 723 637 635 Health 5.460 6A497 6.767 6.533 6.904 7.024 7.266 6.957 6.615 6.935 Publichealthservices 422 642 582 613 720 823 875 805 750 911 Social work - health care 3.093 3.318 3,497 3.464 3.701 3.836 3.893 3.861 3.696 3.621 INSSJyP - Health Care 1.945 2,537 2.689 2,455 2.483 2,364 2,498 2.291 2,169 2.402 Potable Watcr& Sewerage 16 15 29 53 84 96 68 42 29 23 HousingandUrbanDevelopmnct 62 106 12 7 8 8 4 6 5 5 SocialServicesandSocialAssistance 747 896 889 780 847 899 824 793 714 877 Social Security 13.794 16.290 16.569 17.716 17914 18.011 17.950 17.980 17.304 17.349 Labor 1.901 2,197 2,472 2,437 2.407 2,331 2.528 2.412 2.386 4.502 EconoAsic Services 4,649 3,209 3,011 2,172 2,254 2,344 1,939 1,673 1,442 1,344 Primary Production 248 263 255 260 252 290 249 204 247 268 Energyandfuels 2,562 1.335 1.125 610 479 424 371 405 313 331 Industry 354 158 174 109 96 III 99 85 71 81 Services 1.352 1.290 1318 1.045 1.275 1,397 1.118 897 733 577 Transportation 1.161 1.180 1.169 935 1.103 1.244 1.008 808 631 496 Commtounicahions 191 110 148 110 173 152 110 89 102 81 Other 133 162 139 148 151 121 102 83 78 86 Debt Service 3,869 3,911 5,014 4,933 6,693 6,771 8,588 10,142 11,739 6,664 SOURCE: MECON 120 Table B4(a): As Percent of GDP (Percent of GDP) TOTAL 16.4 16.5 17.2 16.2 16.0 15.9 17.6 17.5 18.3 15.1 Functioning of the State 2.5 2.7 2.7 2.5 2.3 2.3 2.8 2.5 2.3 2.2 General admimnstration 10 1 1 1.1 1.0 0.9 09 1.3 1.0 0.8 0.8 Justice 0.2 0 3 0.3 0 3 0.3 0.3 0 3 0.3 0 3 0.3 Defense and security .3 1.3 1.3 1.2 1 1 1.1 1.2 1.2 1.2 L.1 Socialpublicexpenditure 10.2 11.0 IIA 11.1 10.6 10.5 11.1 10.9 11.0 10A Education. culture and science. and technology 0.9 0 9 1.0 1.0 1.0 1.0 I 0 1.0 1.0 0.9 Basiceducation 0.0 01 0 1 0.1 0.1 0.1 0.1 0.1 0.0 01 Tertiary and university education 0 6 0.6 0.6 0.6 0.6 0.6 0.7 0.7 0.7 0.6 Other 0.3 0.3 0 3 0.3 0.3 0.3 0.3 0.3 0.2 0.2 Health 2.3 2.5 2.6 2 4 2 4 2.3 2.6 2.4 2.5 2.2 Public health services 0.2 0 2 0.2 0 2 0 2 0 3 0 3 0.3 0.3 0.3 Social wortk - health care 1.3 13 1.4 1.3 1.3 1.3 14 1.4 1.4 1.2 INSSJyP - Health Care 0.8 1.0 I 0 0.9 0.8 0.8 0.9 0.8 0 8 0.8 Potable Water & Sewerage 0 0 00 00 0 0 0.0 0.0 0.0 0.0 0 0 0.0 Housing and Urban Development 0.0 00 0.0 00 00 0.0 00 0.0 0.0 0.0 Social Services and Social Assistance 0.3 03 0 3 O.3 0.3 0.3 0 3 0.3 0 3 0.3 Social Security 5 8 6.3 6.4 6.5 6.1 6.0 6.3 6.3 6.4 5.5 Labor 0.8 0.9 1.0 0.9 0.8 0.8 0.9 0.8 0.9 1.4 Economic Services 2.0 1.2 1.2 0.8 0.8 0.8 0.7 0.6 0.5 0.4 Pnmary Production 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Energy andfuels 1.1 0.5 0.4 0.2 0.2 0.1 0.1 0.1 0.1 01 Industry 0.1 0.1 0.1 0.0 0.0 0.0 0.0 1.0 0.0 0.0 Services 0.6 0 5 0 5 0 4 0.4 0.5 0.4 0.3 0 3 0.2 Transportation 0.5 0 5 0.5 0.3 0 4 0.4 0.4 0.3 0.2 0 2 Communications 0.1 00 01 0.0 0.1 0.1 0.0 0.0 0.0 0.0 Other 0.1 0.1 0.1 0 1 0.1 0.0 0 0 0.0 0.0 0.0 Debt Service 1.6 1.5 1.9 1.8 2.3 2.3 3.0 3.6 4.4 2.1 Provisional darafor 2002 Source: MECON 121 Table B5: Provincial Government Operations 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Mfillions of Pesos) Revenue 25,507 27,371 26,694 29,081 32,617 33,111 32,274 32,548 30,014 32,179 Transfers from Federal Government 13,400 13,944 13,579 14,607 16,507 16,980 16,653 16,570 15,667 15,217 Provincial taxes 8.797 9,686 9,172 9,802 10,897 11,655 11,050 10,870 9,766 10,604 Other provincial revenue 3,310 3,740 3,944 5,189 5,213 4,476 4,571 5,107 4,580 6,358 Expenditures 27,265 29,561 29,942 30,273 32,730 35,061 36,399 35,860 36,382 33,949 Non-interest expenditure 27,117 29,171 29,556 29,269 31,679 33,862 34,972 33,987 33,956 32,472 Wages 14,159 15,089 15,149 14,752 15,647 16,840 18,048 18,574 18,753 18,150 Goods & services 2,816 3,092 3,192 3,069 3,538 3,795 4,092 3,713 3,889 3,799 Private transfers 1,843 2,164 2,267 2,275 2,789 3,127 3,372 3,190 3.232 3,316 Other 4,511 4,297 4,202 4,511 4,583 4,868 4,804 4,945 4,806 4,899 Capital expenditure 3,788 4,529 4,745 4,662 5,123 5,232 4,656 3,565 3,276 2,307 Primary balance -1,610 -1,800 -2,862 -188 938 -751 -2,698 -1,440 -3,943 -292 Interest 148 391 386 1,004 1,051 1,200 1,427 1,873 2,426 1,478 Overall balance .1,758 -2,191 -3,248 -1,192 .113 -1,951 4,125 -3,312 -6,368 -1,770 (Percent of GDP) Revenue 10.8 10.6 10.3 10.7 11.1 11.1 11.4 11.5 11.2 10.3 Transfers from Federal Governmi 5.7 5.4 5.3 5.4 5.6 5.7 5.9 5.8 5.8 4.9 Provincial taxes 3.7 3.8 3.6 3.6 3.7 3.9 3.9 3.8 3.6 3.4 Other provincial revenue 1.4 1.5 1.5 1.9 1.8 1.5 1.6 1.8 1.7 2.0 Expenditures 11.5 11.5 11.6 11.1 11.2 11.7 12.8 12.6 13.5 10.8 Non-interest expenditure 11.5 11.3 11.5 10.8 10.8 11.3 12.3 12.0 12.6 10.4 Wages 6.0 5.9 5.9 5.4 5.3 5.6 6.4 6.5 7.0 5.8 Goods & services 1.2 1.2 1.2 1.1 1.2 1.3 1.4 1.3 1.4 1.2 Private transfers 0.8 0.8 0.9 0.8 1.0 1.0 1.2 1.1 1.2 1.1 Other 1.9 1.7 1.6 1.7 1.6 1.6 1.7 1.7 1.8 1.6 Capital expenditure 1.6 1.8 1.8 1.7 1.7 1.8 1.6 1.3 1.2 0.7 PrimaryBalance -0.7 -0.7 -1.1 -0.1 0.3 -0.3 -1.0 -0.5 -1.5 -0.1 Interest 0.1 0.2 0.1 0.4 0.4 0.4 0.5 0.7 0.9 0.5 Overall Balance -0.7 -0.9 -1.3 -0.4 0.0 -0.7 -1.5 -1.2 -2.4 -0.6 Source: MECON 122 Table B6: Provincial Expenditure by Sector (24 Jurisdictions) 1993 1994 1995 1996 1997 1998 1999 2000 2001 (Millions of Pesos) TOTAL 27,265 29,561 29,942 30,273 32,730 35,061 36,399 35,800 36,382 Government administration 7,209 7,846 7,549 7,704 8,153 8,880 9,062 8,994 8,969 Security services 2,162 2,319 2,476 2,369 2,676 2,974 3,087 3,087 3,112 Social services 14,454 15,500 15,474 15,407 17,343 18,184 19,409 19,033 19,165 Health 3,372 3,538 3,496 3,508 3,693 3,807 4,116 3,984 4,056 Social assistance 1,000 1,174 965 1,082 1,537 1,657 1,743 1,575 1,501 Social security 1,062 918 853 710 618 638 664 563 605 Education and culture 7,412 8,040 8,280 8,256 9,335 9,769 10,749 11,057 11,049 Science and technology 31 33 33 35 27 32 27 31 31 Labor 0 0 2 4 74 187 216 237 326 Housing and urban development 1,009 1,133 1,104 1,158 1,353 1,463 1,248 1,093 1,140 Potable water & sewerage 377 446 500 351 414 366 328 233 -198 Other urban services 191 217 243 302 293 266 318 260 258 Economic services 2,969 3,326 3,722 3,789 3,506 3,833 3,410 2,815 2,711 Public debt 471 571 721 1,004 1,052 1,190 1,431 1,872 2,426 (Percent of GDP) TOTAL 11.5 11.5 11.6 11.1 11.2 11.7 12.8 12.6 13.5 Government administration 3.0 3.0 2.9 2.8 2.8 3.0 3.2 3.2 3.3 Security services 0.9 0.9 1.0 0.9 0.9 1.0 1.1 1.1 1.2 Social services 6.1 6.0 6.0 5.7 5.9 6.1 6.8 6.7 7.1 Health 1.4 1.4 1.4 1.3 1.3 1.3 1.5 1.4 1.5 Social assistance 0.4 0.5 0.4 0.4 0.5 0.6 0.6 0.6 0.6 Social security 0.4 0.4 0.3 0.3 0.2 0.2 0.2 0.2 0.2 Education and culture 3.1 3.1 3.2 3.0 3.2 3.3 3.8 3.9 4.1 Science and technology 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Labor 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.1 Housing and urban development 0.4 0.4 0.4 0.4 0.5 0.5 0.4 0.4 0.4 Potable water & sewerage 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 Other urban services 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Economic services 1.3 1.3 1.4 1.4 1.2 1.3 1.2 1.0 1.0 Public debt 0.2 0.2 0.3 OA OA OA 0.5 0.7 0.9 Source: MECON 123 Table B7: Consolidated Provincial Debt (24 Jurisdictions) (end of period) As of 1995 1996 1997 1998 1999 2000 2001 June 2002 (Millions of Pesos) TOTAL* 15,664 17,258 14,683 15,718 191913 24,199 32,765 64,290 without floating debt 12,336 13,921 11,802 13,164 16,565 21,260 30,067 n.a. National government 856 2,122 389 233 236 158 144 211 Banks and financial institutions 5,118 6,276 4,469 5,315 6,459 7,578 9,460 16,586 Intemational organizations** 1,439 1,920 2,296 3,071 3,737 4,157 4,706 19,144 Floating debt (arrears) 3,328 3,337 2,881 2,554 3,348 2,939 2,698 n.a. Consolidated debt 1,162 656 561 628 1,223 1,106 981 1,066 Public securities 1,172 2,532 3,909 3,826 4,911 7,121 11,894 27,285 Privatization of provincial bank: 2,588 415 179 91 0 0 0 0 Debt refinancing program 0 0 0 0 0 1,140 2,882 0 (Percent of GDP) TOTAL* 6.1 6.3 5.0 5.3 7.0 8.5 12.2 19.0 without floating debt 4.8 5.1 4.0 4.4 5.8 7.5 11.2 n.a. National government 0.3 0.8 0.1 0.1 0.1 0.1 0.1 0.1 Banks and financial institutions 2.0 2.3 1.5 1.8 2.3 2.7 3.5 4.9 International organizations*" 0.6 0.7 0.8 1.0 1.3 1.5 1.8 5.6 Floating debt (arrears) 1.3 1.2 1.0 0.9 1.2 1.0 1.0 n.a. Consolidated debt 0.5 0.2 0.2 0.2 0.4 0.4 0.4 0.3 Public securities 0.5 0.9 1.3 1.3 1.7 2.5 4.4 8.0 Privatization of provincial bank 1.0 0.2 0.1 0.0 0.0 0.0 0.0 0.0 Debt refinancing program 0.0 0.0 0.0 0.0 0.0 0.4 1.1 0.0 * Excludes $1.458 billion in compensation from the National Government * Also includes the Fiduciary Fundfor Provincial Development and the Fundfor Regional Infrastructure Source: MECON 124 Table B8: Consolidated Public Expenditure at Current Prices 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Millions of Pesos) TOTAL EXPENDITURE 74,882 81,798 83,861 83,156 88,891 92,630 97,090 96,053 96,153 91,111 Administration and defense 14,550 16,228 16,134 15,976 16,610 17,888 19,550 18,388 17,833 17,580 General administration 7,667 8,573 8,188 8,175 8,378 9,223 10,476 9,323 8,902 8,530 Justice 1,672 1,905 2,020 2,033 2,202 2,338 2,515 2,602 2,627 2,536 Defense and security 5,210 5,750 5,926 5,768 6,030 6,327 6,559 6,464 6,304 6,513 Social expenditure 47,905 53,976 54,627 54,602 57,878 59,647 61,354 60,274 59,12 60,962 Education, culture, and science 9,668 10,645 11,088 11,292 12,591 13,181 14,006 14,205 13,951 13,434 Health 10,921 12,526 12,781 12,506 13,198 13,557 14,375 13,934 13,639 13,886 Water 422 458 528 404 501 458 396 285 264 286 Housingandurbandevelopmenlt 1,110 1,274 1,115 1,153 1,361 1,468 1,252 1,095 1,076 968 Social assistance 2,660 3,025 2,714 2,595 3,292 3,679 3,655 3,379 3,287 3,611 Social security 18,642 21,192 21,618 21,775 21,926 22,219 22,330 22,473 21,914 21,989 Unemployment insurance and allowances 1,997 2,337 2,485 2,475 2,521 2,577 2,748 2,701 2,779 4,760 Other 2,486 2,519 2,299 2,402 2,489 2,508 2,592 2,202 2,202 2,028 Economic services 8,055 7,082 7,323 6,566 6,574 7,046 6,080 5,274 4,944 4,351 Primary production 693 789 759 735 761 965 1,112 974 803 779 Energyandfuels 3,090 1,933 1,666 1,194 1,116 1,110 863 905 703 698 Industry 404 236 260 200 212 244 239 198 191 163 Services 2,636 2,896 2,998 2,659 3,375 3,656 3,164 2,704 2,640 2,104 Transport 2,400 2,742 2,805 2,508 3,144 3,444 2,970 2,540 2,431 1,968 Communications 236 154 193 151 231 212 194 165 208 136 Other 1,232 1,228 1,641 1,777 1,110 1,070 701 493 608 608 Debt service 4,372 4,511 5,777 6,012 7,828 8,049 10,107 12,117 14,264 8,218 Source: MECON 125 Table B8(a): As Percent of GDP 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 (Percent of GDP) TOTAL EXPENDITURE 31.7 31.8 32.5 30.6 30.4 31.0 34.2 33.8 35.8 29.1 Administration and defense 6.2 6.3 6.3 5.9 5.7 6.0 6.9 6.5 6.6 5.6 General administration 3.2 3.3 3.2 3.0 2.9 3.1 3.7 3.3 3.3 2.7 Justice 0.7 0.7 0.8 0.7 0.8 0.8 0.9 0.9 1.0 0.8 Defense and security 2.2 2.2 2.3 2.1 2.1 2.1 2.3 2.3 2.3 2.1 Social expenditure 20.3 21.0 21.2 20.1 19.8 20.0 21.6 21.2 22.0 19.5 Education, culture, and science 4.1 4.1 4.3 4.1 4.3 4.4 4.9 5.0 5.2 4.3 Health 4.6 4.9 5.0 4.6 4.5 4.5 5.1 4.9 5.1 4.4 Water 0.2 0.2 0.2 0.1 0.2 0.2 0.1 0.1 0.1 0.1 Housing and urban development 0.5 0.5 0.4 0.4 0.5 0.5 0.4 0.4 0.4 0.3 Social assistance 1.1 1.2 1.1 1.0 1.1 1.2 1.3 1.2 1.2 1.2 Social security 7.9 8.2 8.4 8.0 7.5 7.4 7.9 7.9 8.2 7.0 Unemployment insurance and allowance: 0.8 0.9 1.0 0.9 0.9 0.9 1.0 1.0 1.0 1.5 Other 1.1 1.0 0.9 0.9 0.8 0.8 0.9 0.8 0.8 0.6 Econonmic services 3A 2.8 2.8 2.4 2.2 2.4 2.1 1.9 1.8 1.4 Primary production 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.3 0.3 0.2 Energy and fuels 1.3 0.8 0.6 0.4 0.4 0.4 0.3 0.3 0.3 0.2 Industry 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Services 1.1 1.1 1.2 1.0 1.2 1.2 1.1 1.0 1.0 0.7 Transport 1.0 1.1 1.1 0.9 1.1 1.2 1.0 0.9 0.9 0.6 Communications 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 Other 0.5 0.5 0.6 0.7 0.4 0.4 0.2 0.2 0.2 0.2 Debt service 1.8 1.8 2.2 2.2 2.7 2.7 3.6 4.3 5.3 2.6 Source: MECON 126 Table B8(b): As Percent of Total 1993 1994 1995 1996 l997 1998 1999 2000 2001 2002 (Percent of Total Expenditure) TOTAL EXPENDITURE 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Administration and defense 19A 19.8 19.2 19.2 18.7 193 20.1 19.1 185 193 General admninistration 10.2 10.5 9.8 9.8 9.4 10.0 10.8 9.7 93 9.4 Justice 2.2 2.3 2.4 2.4 2.5 2.5 2.6 2.7 2.7 2.8 Defenseand security 7.0 7.0 7.1 6.9 6.8 6.8 6.8 6.7 6.6 7.1 Sodal expenditure 64.0 66.0 65.1 65.7 65.1 64.4 63.2 62.8 61.5 66.9 Education. culture, and science 12.9 13.0 13.2 13.6 14.2 14.2 14.4 14.8 14.5 14.7 Hcalth 14.6 15.3 15.2 15.0 14.8 14.6 14.8 14.5 14.2 15.2 Water 0.6 0.6 0.6 0.5 0.6 0.5 0.4 0.3 0.3 0.3 Housingandurbandevelopnent 1.5 1.6 1.3 1.4 1.5 1.6 1.3 1.1 1.1 1.1 Social assistance 3.6 3.7 3.2 3.1 3.7 4.0 3.8 3.5 3.4 4.0 - Social security 24.9 25.9 25.8 26.2 24.7 24.0 23.0 23.4 22.8 24.1 Unemnployment insurance and allowances 2.7 2.9 3.0 3.0 2.8 2.8 2.8 2.8 2.9 5.2 Other 33 3.1 2.7 2.9 2.8 2.7 2.7 2.3 2.3 2.2 Economic services 10.8 8.7 8.7 7.9 7.4 7.6 6.3 5.5 5.1 4.8 Primary production 0.9 1.0 0.9 0.9 0.9 1.0 1.1 1.0 0.8 0.9 Energy and fuels 4.1 2.4 2.0 1.4 1.3 1.2 0.9 0.9 0.7 0.8 Industry 0.5 0.3 0.3 0.2 0.2 0.3 0.2 0.2 0.2 0.2 Services 3.5 3.5 3.6 3.2 3.8 3.9 3.3 2.8 2.7 2.3 Transport 3.2 3.4 3.3 3.0 3.5 3.7 3.1 2.6 2.5 2.2. Communications 0.3 0.2 0.2 0.2 0.3 0.2 0.2 0.2 0.2 0.1 Other 1.6 1.5 2.0 2.1 1.2 1.2 0.7 0.5 0.6 0.7 Debt service 5.8 5.5 6.9 7.2 8. 8.7 10A 12.6 14.8 9.0 Source: MECON 127 Table B9: Provincial Tax Revenue in Thousands of Pesos Average 1993-2001 Province Gross Real Estate Stamps Automobiles Other Total lReceipts Property Buenos Aires - City 1,679,109 444,878 0 259,368 119,084 2,502,439 Buenos Aires - Province 1,918,383 590,623 387,690 419,940 628,503 3,945,138 Catamarca 22,564 3,461 4,243 2,966 65 33,300 C6rdoba 455,982 235,799 82,585 18,649 43,328 836,343 Corrientes 49,098 9,801 11,782 0 1,567 72,247 Chaco 71,915 2,294 8,041 0 12,172 94,422 Chubut 45,453 203 12,433 0 3,091 61,180 Entre Rios 119,625 63,344 30,925 29,649 23,672 267,216 Formosa 19,857 1,026 3,316 1 1,121 25,321 Jujuy 35,000 8,292 6,001 114 3,037 52,444 La Pampa 40,162 15,431 17,889 13,135 2,714 89,331 La Rioja 15,896 2,105 2,233 2,447 233 22,914 Mendoza 204,423 44,715 52,683 48,529 17,206 367,556 Misiones 58,484 12,346 9,380 1,982 19,364 101,556 Neuquen 102,133 12,512 27,510 0 6 142,160 Rio Negro 72,836 18,934 21,471 13,831 1,499 128,572 Salta 88,245 4,953 19,188 0 7,037 119,422 San Juan 42,179 7,121 8,733 8,288 14,910 81,230 San Luis 46,422 15,327 10,547 4,076 5,441 81,813 Santa Cruz 46,749 38 10,609 0 460 57,856 Santa Fe 443,864 176,696 147,456 6,434 47,197 821,647 Santiago del Estero 41,144 7,040 6,650 4,588 13,394 72,817 Tierra del Fuego 40,362 76 3,051 0 2,566 46,055 Tucuman 90,681 18,663 29,635 14,189 12,208 165,375 Total 5,750,565 1,695,678 914,051 848,185 979,875 10,188,353 Source: MECON 128 Table B9(a): As Percent of Total Average 1993-2001 Province Gross Real Estate Stamps Automobiles Other Total Receipts Property Buenos Aires - City 67.1 17.8 0.0 10.4 4.8 100.0 Buenos Aires - Province 48.6 15.0 9.8 10.6 15.9 100.0 Catamarca 67.8 10.4 12.7 8.9 0.2 100.0 C6rdoba 54.5 28.2 9.9 2.2 5.2 100.0 Corrientes 68.0 13.6 16.3 0.0 2.2 100.0 Chaco 76.2 2.4 8.5 0.0 12.9 100.0 Chubut 74.3 0.3 20.3 0.0 5.1 100.0 Entre Rfos 44.8 23.7 11.6 11.1 8.9 100.0 Formosa 78.4 4.1 13.1 0.0 4.4 100.0 Jujuy 66.7 15.8 11.4 0.2 5.8 100.0 La Pampa 45.0 17.3 20.0 14.7 3.0 100.0 La Rioja 69.4 9.2 9.7 10.7 1.0 100.0 Mendoza 55.6 12.2 14.3 13.2 4.7 100.0 Misiones 57.6 12.2 9.2 2.0 19.1 100.0 Neuquen 71.8 8.8 19.4 0.0 0.0 100.0 Rfo Negro 56.7 14.7 16.7 10.8 1.2 100.0 Salta 73.9 4.1 16.1 0.0 5.9 100.0 San Juan - 51.9 8.8 10.8 10.2 18.4 100.0 San Luis 56.7 18.7 12.9 5.0 6.7 100.0 Santa Cruz 80.8 0.1 18.3 0.0 0.8 100.0 Santa Fe 54.0 21.5 17.9 0.8 5.7 100.0 Santiago del Estero 56.5 9.7 9.1 6.3 18.4 100.0 Tierra del Fuego 87.6 0.2 6.6 0.0 5.6 100.0 TucumAn 54.8 11.3 17.9 8.6 7.4 100.0 Average of Revenues 56.4 16.6 9.0 8.3 9.6 100.0 Average of Provinces 63.3 11.7 13.0 5.2 6.8 Source: MECON 129 Table B9(b): As Share of Provincial GDP 1993-2001 Provinces Gross Estale Stam Automobil Others Total Average Receipts Poertyt tap es (Pesos per Property capita) Buenos Aires - City 2.7 0.7 0.0 0.4 0.2 4.0 879 Buenos Aires - Province 2.0 0.6 0.4 0.4 0.7 4.1 296 Catamarca 1.7 0.3 0.3 0.2 0.0 2.5 109 C6rdoba 2.1 1.1 0.4 0.1 0.2 3.9 285 Corrientes 1.5 0.3 0.3 0.0 0.1 2.2 83 Chaco 2.3 0.1 0.2 0.0 0.4 3:0 102 Chubut 1.7 0.0 0.4 0.1 0.1 2.3 156 Entre Rios 2.0 1.1 0.5 0.5 0.4 4.5 243 Formosa 1.7 0.1 0.3 0.0 0.1 2.2 57 Jujuy 1.6 0.4 0.2 0.0 0.1 2.3 92 La Pampa 2.0 0.8 0.9 0.7 0.1 4.5 316 La Rioja 1.0 0.1 0.2 0.1 0.0 1.4 89 Mendoza 1.9 0.4 0.5 0.4 0.2 3.4 244 Misiones 1.6 0.3 0.2 0.1 0.5 2.7 114 Neuqu6n 1.6 0.2 0.4 0.0 0.0 2.2 322 Rio Negro 2.0 0.5 0.6 0.4 0.0 3.5 241 Salta 2.2 0.1 0.4 0.0 0.2 2.9 120 San Juan 1.4 0.2 0.3 0.3 0.5 2.7 139 San Luis 1.2 0.4 0.3 0.1 0.1 2.1 245 Santa Cruz 2.4 0.0 0.5 0.0 0.0 2.9 319 Santa Fe 2.0 0.8 0.6 0.1 0.2 3.7 282 Santiago del Estero 2.1 0.4 0.3 0.3 0.7 3.8 97 Tierra del Fuego 1.4 0.0 0.1 0.0 0.1 1.6 536 Tucuman 1.8 0.4 0.5 0.3 0.2 3.2 131 TOTAL 2.1 0.6 0.3 0.3 0.4 3.7 293 Sources: DNCFP. MECON, INDEC Census Data, Bank staff calculations 130 Table B10: Provincial Revenues and Employment in Year 2000 Own Tax Public Employees Transfers per Royalties per Rvn per Per 1,000 capita capita Revenues per per 1,000 capita inhabitants $m $m $rn number Buenos Aires - City 92.0 0.0 985.8 37 Buenos Aires - Province 303.4 0.0 301.3 30 Catamarca 1279.7 0.0 115.7 76 C6rdoba 454.6 0.0 279.0 25 Corrientes 666.3 7.6 83.8 41 Chaco 778.9 137.2 94.1 45 Chubut 828.5 0.0 174.9 53 Entre Rios 686.2 16.6 226.4 49 Formosa 1175.3 8.8 50.2 68 Jujuy 847.7 0.9 100.0 50 LaPampa 1091.1 64.2 306.1 57 LaRioja 1741.4 0.0 87.0 85 Mendoza 436.3 75.1 223.3 40 Misiones 640.4 9.7 117.2 37 Neuquen 740.7 994.9 354.8 68 Rio Negro 804.0 115.5 258.5 48 Salta 613.4 49.0 126.7 36 San Juan 894.2 0.0 144.4 50 San Luis 1017.8 0.0 246.7 47 Santa Cruz 1877.4 843.5 308.9 86 Santa Fe 481.1 0.0 298.4 34 Santiago del Estero 802.5 0.5 100.9 48 Tierra del Fuego 2455.5 568.3 432.2 74 Tucuman 573.0 0.0 135.6 36 Source: MECON 131 CoSOCIAL SECTORS Table Cl: Consolidatedi Public Epennditures an Edlucation Jurisdiccion 1991 1993 1995 1997 1999 2000 2001 2002 National Government 2,227 1,720 1,851 2,126 2,168 2,066 2,014 1,415 Basic Education 1,208 116 206 394 240 150 107 181 Higher Education and University 961 1,473 1,596 1,690 1,872 1,885 1,880 1,217 Other 58 131 49 42 56 31 32 17 Provinces 4,588 7,570 7,883 8,668 10,388 10,620 10,675 6,401 Basic Education 4,026 6,500 6,581 7,229 8,641 8,838 9,000 5,366 Higher Education and University 184 313 516 560 704 680 685 410 Other 378 753 786 877 1,040 1,050 1,019 625 Municipalities 160 202 284 358 409 437 438 271 Basic Education 160 202 284 358 409 437 438 271 Total 6,975 9,487 10,018 11,149 12,961 13,072 13,162 8,087 Basic Education 5,394 6,818 7.070 7,981 9,290 9,426 9,545 5,818 Higher Education and University 1,145 1,786 2,112 2,250 2,576 2,566 2,565 1,627 Other 436 883 836 918 1,095 1,081 1,051 642 Level 1991 1993 1995 1997 1999 2(80 2001 2002 Basic Education 5,394 6,818 7,070 7,981 9,290 9,426 9,545 5,818 Nation 1,208 116 206 394 240 150 107 181 Provinces 4,026 6,500 6,581 7,229 8,641 8,838 9,000 5,366 Municipalities 160 202 284 358 409 437 438 271 ligher Education and Universities 1,145 1,786 2,112 2,250 2,576 2,566 2,565 1,627 Nation 961 1,473 1,596 1,690 1,872 1,885 1,880 1,217 Provinces 184 313 516 560 704 680 685 410 Others 436 883 836 918 1,095 1,081 1,051 642 Nation 58 131 49 42 56 31 32 17 Provinces 378 753 786 877 1,040 1,050 1,019 625 Total Education 6,975 9,492 10,018 11,149 12,961 13,072 13,156 8,087 Nation 2,226 1,720 1,851 2,126 2,168 2,067 2,014 1,415 Provinces 4,588 7,570 7,883 8,666 10,384 10,568 10,704 6,401 Municipalities 160 202 284 358 409 437 438 271 132 Table C2: Evolution of Education Provincial Spending Province 1994 1996 1998 1999 2000 2001 2001/ % ProvJ 1993 Total (millions of 2001 pesos) Buenos Aires 2,328 2,403 3,252 3,711 3,786 3,816 80.1% 35.7% City of Buenos Aires 715 733 782 862 906 873 15.4% 8.2% Catamarca 147 124 141 145 138 139 0.1% 1.3% Chaco 224 217 250 296 287 287 24.5% 2.7% Chubut 134 127 150 155 154 160 20.6% 1.5% C6rdoba 650 539 661 714 823 986 56.5% 9.2% Corrientes 174 167 216 225 245 238 31.5% 2.2% Entre Rfos 303 265 285 370 343 329 13.3% 3.1% Formosa 136 124 158 169 165 162 45.2% 1.5% Jujuy 159 157 171 208 198 201 56.6% 1.9% La Pampa 115 100 117 128 127 127 6.7% 1.2% La Rioja 107 120 118 140 145 138 35.0% 1.3% Mendoza 347 328 361 444 442 434 33.8% 4.1% Misiones 181 176 195 240 229 215 20.7% 2.0% Neuquen 266 212 247 301 289 295 16.0% 2.8% Rio Negro 209 178 176 187 208 200 13.6% 1.9% Salta 201 177 194 218 216 187 4.8% 1.8% San Juan 104 130 155 183 203 193 15.1% 1.8% San Luis 112 81 79 104 116 147 98.7% 1.4% SantaCruz 685 126 157 159 171 151 42.4% 1.4% Santa Fe 208 675 754 834 850 802 17.1% 7.5% Santiago Del Estero 208 174 204 220 230 228 21.6% 2.1% Tierra Del Fuego 76 65 83 83 73 72 -9.1% 0.7% Tucuman 256 266 259 288 277 294 34.0% 2.7% Total 8,043 7,664 9,165 10,388 10,620 10,675 41.0% 100.0% Total Public Expenditure 30,663 29,210 34,083 36,203 35,318 36,382 25.5% Education / Total Expenditures 26.2% 26.2% 26.9% 28.7% 30.1% 29.3% 25.5% Source: Directorate of Consolidated Social Expenditures - Secretariat of Economic Policy 133 Table C3: Provincial Education Expenditures- Econonic Classification in Year 2000 Province Mainten Transfers Other Personnel Current CaitI Personnel ance to Private Transfers Sub Total Expenses E Total Education Sub Total Buenos Aires 76% 3% 14% 5% 89% 98% 2% 100% City of Buenos Aires 76% 3% 15% 3% 92% 98% 2% 100% Catamarca 90% 1% 6% 1% 96% 98% 2% 100% Chaco 88% 2% 6% 2% 94% 98% 2% 100% Chubut 92% 1 % 4% 1 % 96% 99% 1 % 100% Cordoba 67% 2% 22% 1% 89% 92% 8% 100% Corrientes 83% 1% 12% 2% 95% 97% 3% 100% Entre Rios 81% 2% 15% 1% 96% 99% 1% 100% Formosa 93% 1% 3% 2% 96% 99% 1% 100% Jujuy 92% 1% 5% 1% 97% 99% 1% 100% La Pampa 76% 6% 12% 1% 88% 95% 5% 100% La Rioja 88% 3% 2% 1% 91% 95% 5% 100% Mendoza 80% 5% 11% 2% 92% 98% 2% 100% Misiones 73% 5% 14% 5% 87% 97% 3% 100% Neuquen 85% 8% 5% 1% 90% 99% 1% 100% Rio Negro 71% 17% 8% 1% 79% 98% 2% 100% Salta 80% 2% 7% 2% 87% 92% 8% 100% San Juan 81% 3% 13% 1% 94% 98% 2% 100% San Luis 88% 2% 8% 1% 95% 99% 1% 100% Santa Cruz 84% 4% 10% 0% 94% 98% 2% 100% Santa Fe 75% 2% 20% 1% 95% 99% 1% 100% Santiago del Estero 85% 1% 10% 2% 95% 98% 2% 100% Tierra Del Fuego 78% 3% 10% 4% 88% 96% 4% 100% Tucuman 83% 1% 13% 2% 96% 99% 1% 100% TOTAL 78% 3% 13% 3% 91% 97% 3% 100% Source: Education System Costs Program - Ministry of Education 134 Table C4: Spending on Education Indicators in Year 2001 Education Education Per Per Teachers Teachers Spending Spending / Subsidy Pr e Spending Spending/ Per Capita Capita Student Annual Annual Per Public /Private Province Provincial Total GDP Copartici- Copartki- Salary Salary Student Sector Sector * GDP (%) Public (2000) paci6n pacion (Elementa (Secondar (Non Univ. Student Student Expend. Transfers Transfers ry Level) y Level)* Edu.) (Non Univ.) (Non Univ.) (2001 pesos) Buenos Aires 4.3 34.8% 6,161 280 1,040 6,865 9,149 1,239 464 City of Buenos Aire: 1.3 27.4% 23,639 61 273 7,774 9,703 1,470 2,327 483 Catamarca 8.6 25.3% 4,636 1,245 3,988 8,999 10,717 1,353 1,435 733 Chaco 8.5 26.0% 3,265 740 2,408 6,093 8,109 1,013 1,031 788 Chubut 4.4 27.2% 8,392 735 2,535 7,194 7,293 1,215 1,324 430 Cordoba 4.6 33.9% 6,693 430 1,695 9,535 9,363 1,078 1,218 761 Corrientes 7.9 30.1% 3,107 626 1,967 6,595 7,766 875 884 813 Entre Rios 5.4 24.3% 4,994 648 2,343 6,591 7,675 1,120 1,228 756 Formosa 11.3 22.0% 2,835 1,119 3,279 6,130 8,193 963 1,021 308 Jujuy 8.8 27.1% 3,557 737 2,244 6,406 10,007 967 1,032 420 LaPampa 5.5 19.8% 7,424 1,033 4,141 7,340 9,802 1,864 1,881 1,747 LaRioja 9.9 22.7% 4,593 1,117 3,509 10,305 13,153 1,414 1,511 361 Mendoza 4.4 29.6% 6,033 417 1,582 6,883 23,413 1,022 1,085 705 Misiones 5.9 25.1% 3,631 555 1,823 5,641 7,550 783 815 632 Neuquen 5.7 25.7% 10,514 660 2,130 9,886 12,584 1,880 1,960 1,092 Rio Negro 5.1 26.9% 6,837 732 2,435 6,784 8,107 1,465 1,606 738 Salta 4.8 20.6% 3,458 555 1,763 6,041 8,510 667 722 329 San Juan 7.4 24.1% 4,010 838 3,199 7,745 9,446 1,218 1,267 969 San Luis 4.6 22.1% 8,278 996 3,587 9,700 13,809 1,345 1,401 909 Santa Cruz 6.1 19.4% 12,013 1,558 5,177 12,889 14,472 2,565 2,710 1,727 Santa Fe 3.9 32.1% 6,490 455 1,708 7,001 8,094 1,089 1,203 788 Santiago del Estero 9.6 31.7% 2,828 774 2,714 9,209 28,972 1,007 1,050 741 Tierra Del Fuego 4.0 16.4% 17,046 7,151 20,650 12,595 12,503 2,092 2,329 1,084 Tucuman 5.6 25.0% 3,745 175 706 6,271 8,380 856 964 487 Total / Average 6.1 29.3% 7,093 985 3,205 7,936 11,115 1,264 1,385 761 (*) Teachers with 15 years of experience Source: Ministry of Education and National Directoriate for Consolidated Social Expenditures 135 Table C5: Elementary Level Indicators in Year 2001 (Includes Initial, EGB I and II and first year EGB IIJ/Primary) Total Students Students Teachers Spending pStded ig Province E oll t per per per per per capita Teacher Section Section Student GDP GDP (2001 pesos) Buenos Aires 2,477,851 14.3 27.2 1.90 829 13.5 City of Buenos Aires 347,990 11.3 21.9 1.94 1,538 6.5 Catamarca 70,135 13.9 23.6 1.70 1,278 27.6 Chaco 215,894 18.4 24.5 1.33 804 24.6 Chubut 81,803 13.4 23.1 1.72 1,219 14.5 C6rdoba 523,030 18.8 25.8 1.38 1,033 15.4 Conientes 212,794 17.6 24.4 1.39 655 21.1 EntreRfos 211,600 14.1 21.4 1.52 864 17.3 Formosa 120,970 17.3 24.1 1.39 765 27.0 Jujuy 124,980 14.5 24.5 1.69 896 25.2 La Pampa 49,782 12.2 20.3 1.66 1,890 25.5 La Rioja 64,717 15.8 23.8 1.50 1,124 24.5 Mendoza 273,491 17.0 24.1 1.42 813 13.5 Misiones 222,698 19.3 26.7 1.39 524 14.4 Neuquen 100,859 13.7 21.9 1.60 1,551 14.8 Rio Negro 115,824 13.4 22.1 1.65 1,040 15.2 Salta 224,852 17.6 27.9 1.59 419 12.1 San Juan 115,382 16.7 24.5 1.47 963 24.0 San Luis 74,452 14.9 24.5 1.64 1,137 13.7 Santa Cruz 42,533 10.9 23.3 2.13 2,376 19.8 Santa Fd 526,787 15.3 23.7 1.55 863 13.3 Santiago Del Estero 176,030 17.7 24.1 1.36 863 30.5 Tierra Del Fuego 23,707 11.8 22.5 1.91 1,987 11.7 Tucuman 242,340 14.2 25.5 1.79 854 22.8 Total / Average 6,640,501 15.2 23.9 1.7 1,095 203 Source: Ministry of Education and National Directoriate for Consolidated Social Expenditures 136 Table C6: Secondary Level Indicators in Year 2001 (Includes second and third year of EGB III and Polimodal / Secondary) Prove Total Students Students Teachers Spending Sp Provinceota per Studernte Enrollment per per per per per capSta Teacher Section Section Student GDP (2001 pesos) Buenos Aires 1,108,484 7.9 28.7 3.6 1,178.8 15.5 City of Buenos Airn 196,003 7.2 26.8 3.7 1,316.2 5.6 Catamarca 28,142 8.1 24.8 3.1 1,237.2 21.2 Chaco 74,489 10.5 32.9 3.1 1,109.5 32.0 Chubut 35,573 8.3 26.3 3.2 1,346.7 12.8 Cordoba 207,446 9.9 31.1 3.1 1,818.8 21.0 Corrientes 70,535 11.0 32.6 3.0 1,241.5 31.8 Entre Rios 88,094 9.5 26.9 2.8 1,406.0 22.6 Formosa 40,578 12.7 33.4 2.6 1,187.8 32.5 Jujuy 63,228 10.0 33.3 3.3 1,260.1 33.8 La Pampa 23,343 8.3 25.1 .3.0 1,264.8 13.6 La Rioja 22,524 8.0 24.5 3.0 2,802.3 48.0 Mendoza 121,646 9.5 26.9 2.8 1,527.1 20.4 Misiones 63,956 10.3 29.8 2.9 1,045.7 22.1 Neuquen 40,202 7.5 29.7 4.0 2,190.7 20.7 Rio Negro 44,536 9.1 28.7 3.2 1,470.0 21.4 Salta 101,895 13.3 33.0 2.5 841.7 20.0 San Juan 42,896 9.3 28.3 3.0 1,624.4 32.1 San Luis 25,952 6.4 25.7 4.0 1,914.7 18.1 Santa Cruz 15,679 6.6 24.2 3.7 2,362.1 15.3 Santa Fe 223,811 10.0 26.9 2.7 1,260.4 15.6 Santiago del Estero 44,670 9.9 29.5 3.0 1,002.9 29.6 Tierra Del Fuego 9,597 8.3 31.9 3.8 2,296.6 10.5 Tucuman 75,048 10.1 27.8 2.8 975.6 24.4 TOTAL 2,768,327 9.2 28.7 - 3.2 1,486.0 22.5 Source: Ministry of Education and National Directoriate for.Consolidated Social Expenditures 137 Table C7: Basic Indicators of Secondary Education by Income Levels in Year I999 Ilncome Quintiles TotC l 2 3 4 5 6-14 year olds: Late entry in primary school 20,1 32,5 18,6 11,8 8,3 9,3 Late entry in secondary school 40,9 50,2 45,9 42,8 30,0 24,9 Repetition of at least one year 14,1 24,6 12,8 7,0 4,0 3,6 Out of School population (14-18 year olds) 23% 37% 28% 19% 16% 8% Average drop out age 15,7 14,7 15,8 16,5 16,8 16,6 Secondary School Graduation Rate (18-24 year olds) 47,4 23,6 33,9 49,7 64,5 75,8 Source: Siempro - Social Development Secretariat 138 Figure Cl: Relation between per Capita GDP and Spending per Student 2,000 o La Panmpa NeuqueO 1,500 Ri~lo Negro i 1,500 - O i 2~ ~ ~~~~~~c o ~12iojaD Cz ¢Catarnarca rN O Sazn Jnan n Cubut go. Esi?ero CblEc Mecndoza 0 C6rdoba C;1 000 - o_0^-/ UJUm I ool2Ju uyBuenos Aires Corrientes md O~ Msiones o Smite 500 - 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 11,000 per capita GDP Source: Ministry of Education - Program for Costing of the Education System Figure C2: Public Spending per Student and Results in Academic Evaluations 65.0 63.0 o Sca aFi c 1 ffi 61.0 o = 59.0 M2ndozz FOtvwos o la Pampak ~uu 59.0 >0 .Buenos Aires o Co Negro 57.0 C ddb U 3O Salts O Sn LAiS E e °^ 55.0 San Turu :0 = ° ~ o T. Del Fuego Santa Cr X 53.0 Ch Neuque rm 5; o Fonns 51.0 o lMsiones la Rioja 49.0 Tucamin ° o Ctam Sgo. Del Estero t-.. 47.0 45.0 300.00 800.00 1,300.00 1,800.00 2,300.00 Public Spending per Student (elententary level) Source: Ministry of Education- Program for Costing of the Education System and IDECE 139 Table C8: IFinancing and Spending of Natdonal Universities in Year 2000 Personnel Current Non Capital Transfers Total Structure Expenses personnel Goods services (inillions of pesos) National Treasury 1.424,709 17,266 93,604 25,218 121,928 1.682,725 83,6% Own Resources 31,574 22,503 120,606 15,365 27,099 217,147 10,8% Previous Exercises 12,369 4,291 17,495 30,289 21,834 86,278 4.30% Other Sources 4,534 0,106 2,645 12,381 7,528 27,194 1,4% Total 1.473,186 44,166 234,350 83,253 178,389 2.013,344 100,0% Structure 73,2% 2,2% 11,6% 4,1% 8,9% 100,0% Source: Secretariat of University Policies. General Financial- Economi-c Coordination 140 Table C9: Finiancing of National Universities in Year 2000 Sources of Financing Remnants National Own from Other Total Treasury Resources previous Sources exercises (thousands of pesos) Buenos Aires 298,945 130,000 119 2,986 432,050 Catamarca 23,388 952 739 212 25,291 Centro 27,608 1,124 4,335 256 33,323 Comahue 39,948 1,501 2,654 21 44,124 C6rdoba 128,525 22,327 10,909 1,176 162,938 Cuyo 76,279 979 3,415 5,568 86,240 Entre Rfos 23,314 578 1,149 371 25,411 Formosa 10,312 193 154 77 10,735 Gral. San Martfn 12,912 3,201 2,355 1,039 19,507 Gral. Sarmiento 9,035 533 1,576 701 11,845 Instituto Universitario del Arte 12,180 101 10 20 12,311 Jujuy 16,780 124 1,013 36 17,953 La Matanza 25,771 786 2,024 187 28,768 LaPampa 19,958 227 1,110 306 21,601 La Patagonia Austral 12,364 477 2,237 586 15,664 La Patagonia San Juan Bosco 40,089 883 844 311 42,127 La Plata 108,232 7,076 9,860 1,279 126,446 La Rioja 13,757 1,024 640 709 16,130 Lands 5,287 856 704 975 7,822 Litoral 48,510 3,235 2,794 1,250 55,789 Lomas de Zamora 28,470 4,565 3,359 213 36,607 Lujan 22,129 872 1,875 221 25,097 Mar del Plata 41,668 1,721 2,106 498 45,993 Misiones 30,309 1,242 1,422 396 33,370 Nordeste 50,302 5,885 2,928 251 59,366 Quilmes 15,943 1,238 1,001 267 18,448 Rio Cuarto 37,013 1,121 2,485 813 41,431 Rosario 91,464 708 4,195 1,530 97,898 Salta 31,077 550 2,796 389 34,812 San Juan 63,660 521 1,576 114 65,871 San Luis 37,750 563 1,249 269 39,831 Santiago del Estero 17,957 105 727 171 18,960 Sur 37,480 650 3,666 2,093 43,888 Tecnol6gica Nacional 106,574 18,610 4,233 1,164 130,581 Tres de Febrero 4,078 346 1,506 202 6,133 Tucumdn 108,956 1,921 576 401 111,854 Villa Marfa 4,704 352 1,936 138 7,129 Total 1,682,725 217,147 86,279 27,195 2,013,347 Source: Secretariat of University Policies. General Financial- Economic Coordination 141 Table C10: Enrolled and Graduating Students, National Universities Year Total Entering Graduates Graduates Students Students in t t 1980 315,409 49,121 26,254 1981 316,188 49,678 26,98 1982 318,299 57,037 26,166 1983 337,998 65,62 26,888 1984 443,441 132,204 28,158 57.3% 1985 524,590 194,223 30,873 62.1% 1986 581,813 182,306 29,855 52.3% 1987 618,651 170,265 29,393 44.8% 1988 652,997 169,632 31,645 23.9% 1989 661,315 170,619 33,143 17.1% 1990 679,403 172,4 33,465 18.4% 1991 679,495 164,444 32,323 19.0% 1992 698,561 166,02 32,936 19.4% 1993 674,868 172,28 36,964 21.7% 1994 719,671 193,11 35,995 20.9% 1995 766,847 211,581 35,546 21.6% 1996 812,308 228,139 34,675 20.9% 1997 892,759 243,652 34,503 20.0% 1998 948,348 254,693 34,14 17.7% 1999 1,054,014 280,541 38,471 18.2% 2000 1,124,044 289,246 44,522 19.5% 2001 1,187,105 290,613 N/A N/A Source: Secretariat of University Policies. 142 Table ClI: Education Sector - Federal Administration Program 2001 2002 7 2003 Nation Transf. Total Nation Transf. Total Nation Transf. Total (millions of 2001 pesos) Rendering of Educational Services 1,563.5 774.7 2,338.2 1,018.2 42.5 1,060.6 875.9 188.0 1,063.9 Teachers Salaries 11.0 714.3 725.3 6.1 9.0 15.2 7.9 168.3 176.1 Public 660.0 660.0 155.4 155.4 Transfers to Tertiarty Institutes 11.0 54.3 65.3 6.1 9.0 15.2 7.9 12.9 20.8 Transfers to the Private Sector Goods and Services / Maintenance Infrastructure and Equipment 28.7 60.3 89.1 17.9 33.4 51.3 6.0 19.7 25.7 Transfers to Universities 1,523.8 1,523.8 994.2 994.2 862.1 862.1 Policy Actions, Regulation and Coordination 49.8 31.6 81A 44.6 10.4 55.0 27.9 12.2 40.2 Central Administration 29.7 29.7 15.7 15.7 14.3 14.3 Information and Quality Evaluation 3.7 2.4 6.1 1.9 0.7 2.6 2.1 0.8 2.8 Curricular Management and Teacher Training 6.5 24.8 31.3 2.6 6.5 9.1 PREGASE (Reform Program of the Administrative Managemnent of the Provincial Education Systems) 3.9 4.4 8.3 19.3 3.2 22.5 7.3 11.5 18.7 Decentralized organisms (INAP, CONEAU, etc.) 6.0 6.0 5.1 5.1 4.3 4.3 Other Programs 21.7 5.3 27.0 8.2 3.5 11.7 11.3 0.8 12.2 Total 1,635.1 811.6 2,446.7 1,071.0 56.3 1,127.3 915.2 201.0 1,116.2 Source: Ministry of Economy - National Directoriate for Consolidated Social Spending 143 Table C12: Public Health Sector - Federal Administration 2001 2002 2003 Federal Transf. Total Federal Transf. Total Federal Transf. Total (millions of 2001 pesos) Rendering of Health Services 304.49 45.10 349.59 293.09 7.21 300.30 406.00 0.47 406.47 Effectors (hospitals and health centers) 254.75 45.00 299.75 239.32 239.32 219.18 219.18 Salaries, supplies and 198.24 198.24 144.73 144.73 122.61 122.61 Assistance to PNC beneficiaries 56.51 45.00 101.51 94.59 94.59 96.57 96.57 Goods and Services provided by the Nation 49.74 0.08 49.82 53.78 7.21 60.98 186.82 0.47 187.29 AIDS and Sexually Transmitted Diseases 49.74 0.08 49.82 48.96 0.63 49.59 86.98 0.47 87.45 Sanitary Emergency (Proaps Supplies/Remediar) 4.81 6.58 11.39 99.84 99.84 Infrastructure and Equipment 0.00 Policy actions, coordination and regulation 91.10 1.31 92.41 49.60 2.76 52.36 44.51 0.61 45.11 Central administration (mainly salaries) 50.70 1.31 52.01 30.20 2.71 32.91 26.06 0.60 26.65 Formation of Sanitary and Assistencial Human Resources 20.60 20.60 13.00 0.06 13.05 10.73 0.01 10.74 Planification, Control, Regulation and Fiscalization of 6.43 6.43 4.33 4.33 3.70 3.70 Support for Studies and Public Health Investigation 5.29 5.29 1.68 1.68 1.65 1.65 Health Sector Reform 8.07 8.07 0.39 0.39 2.37 2.37 Public Health Progranms 53.93 22.70 76.63 136.47 1437 150.84 102.18 19.73 121.92 AIDS and Sexually Transmitted Diseases 3.41 3.41 1.03 1.03 0.24 0.24 Disease and Specific Risks Prevention and Control (includes inmunizations, VIGI-A, etc.) 38.81 3.65 42.46 100.08 0.43 100.52 64.20 1.66 65.86 Mother and Child Care 10.46 19.00 29.46 26.39 13.93 40.32 23.89 17.94 41.84 Sanitary Emergencies Coverage 1.26 1.26 0.67 0.67 0.88 0.88 Other Programs 12.61 0.10 12.61 8.30 0.01 8.31 12.96 0.13 13.09 Total 462.13 69.21 531.34 479.16 24.35 503.51 552.69 20.81 573.50 144 Table C13: Social Assistance Sector - Federal Administration Program 2001 2002 2003 Federal Transf. Total Federal Transf. Total Federal Transf. Total (millions of 2001 pesos) Targeted Social Programs t Personal Monetary Transfers 998.4 87.2 1,085.6 2,143.3 0.4 2,143.8 2,245.0 23.5 2,268.5 Employment Programs 177.5 87.2 177.5 1,536.6 0.0 1,536.6 1,767.9 1,767.9 Jefes/as de Hogar 42.0 1,479.2 1,479.2 1,438.5 1,438.5 Trabajar 42.0 1.4 177.5 17.9 0.0 17.9 0.0 Labor Emergency Plan 135.5 135.5 38.3 38.3 314.1 314.1 Financial Assistance for provincial social expenditure 85.8 85.8 0.0 0.0 Provincial 45.2 0.0 0.0 Scholarships 45.2 90.3 132.1 132.1 74.6 74.6 National Education Scholarships 45.2 45.2 132.1 132.1 74.6 74.6 Polimodal Education Scholarships -Province of Bs As. 755.7 0.0 0.0 Non Contributory Pensions 755.7 775.7 422.9 422.9 318.9 318.9 Other programs 19.9 19.9 51.8 0.4 52.2 83.6 23.5 107.1 Ingreso Desarrollo Humano 19.9 51.8 0.4 52.2 83.6 23.5 107.1 Solidaridad 19.9 19.9 0.0 0.0 Other National Programs 0.0 0.0 0.0 Provincial Programs 115.1 0.0 0.0 Targeted Social Programs / Non Monetary Transfers 115.1 58.5 173.6 131.1 195.9 327.0 123.9 188.7 312.6 Nutritional 30.1 4.2 34.3 85.2 194.3 279.5 76.3 185.0 261.3 School Lunches 0.0 32.9 16.1 49.0 30.4 16.2 46.6 Nutritional Emergency Program (PEA) 23.2 4.2 27.4 19.3 178.2 197.5 16.4 168.5 184.9 Comedores Comunitarios (FOPAR) 0.0 28.3 0.0 28.3 24.8 0.3 25.1 Provincial nutritional programs 0.0 0.0 0.0 Prohuerta 6.9 6.9 4.7 4.7 4.7 4.7 Others 85.0 54.3 139.3 45.9 1.6 47.4 47.6 3.6 51.2 Central Administration 67.1 0.8 67.9 56.9 3.4 60.2 49.5 5.7 55.2 Ministry of Social Development 51.8 51.8 38.5 38.5 31.0 31.0 Presidency 0.0 6.8 3.4 10.2 10.2 5.7 15.9 Ministry of Labor 15.3 0.8 16.1 11.6 11.6 8.3 8.3 Total 1,180.6 146.5 1,327.1 2,331.3 199.7 2,531.0 2,418.4 217.9 2,636.3 Source: Ministry of Economy - National Directoriate for Consolidated Social Spending 145 Table C14: Social Insurance - Federal Administration Program 2001 2002 2003 Federal Transf. Total I Federal TransE. Total I Federal Transf. Total (mnillions of 2001 pesos) ANSeS and SSS Administration 249.1 249.1 203.3 203.3 151.2 151.2 Pensions and Retirement Payments 15,690.2 142.3 15,832.5 10,430.3 155.1 10,585.4 8,131.1 212.1 8,343.1 National* 15,690.2 142.3 15,832.5 10,430.3 155.1 10,585.4 8,131.1 212.1 8,343.1 Provincial Subsidies 609.0 609.0 355.7 355.7 413.6 413.6 Asignaciones Familiares 1,808.0 1,808.0 1,085.3 1,085.3 833.3 833.3 Unemployment Benefits 353.7 353.7 310.8 310.8 236.4 236.4 Obras Sociales 6,363.0 4.1 6,367.1 4,288.5 4,288.5 3,447.5 3,447.5 PAMI 2,389.2 2,389.2 1,726.5 1,726.5 1,482.5 1,482.5 National 3,973.8 3,973.8 2,562.0 2,562.0 1,965.0 1,965.0 Provincial 4.1 4.1 Total 25,073 146 25,219 16,674 155 16,829 13,213 212 13,425 e Includes ex-provincial pension funds. Source: Ministry of Economy - National Directoriate for Consolidated Social Spending 146 Table C15: Jefes Program: Number of Beneficiaries % Change % pi Province Dec-02 Feb-03 Apr-03 December - 20 TAr bajar ~~~~~~~~~~~~~~April Buenos Aires 658,939 718,166 755,829 14.7 38.0% 30.6 Capital Federal 56,391 53,758 56,248 0.0 2.8% 1.5 Catamarca 36,341 39,227 36,193 0.0 1.8% 1.3 Chaco 106,972 106,665 112,904 5.5 5.7% 5.1 Chubut 14,468 14,096 14,223 -1.7 0.7% 1 Cordoba 117,575 118,232 121,285 3.2 6.1% 7.3 Corrientes 64,732 63,033 66,236 2.3 3.3% 3.9 EntreRios 49,610 49,129 50,467 1.7 2.5% 3.4 Formosa 56,396 53,707 55,809 -1.0 2.8% 2.6 Jujuy 60,202 64,474 70,486 17.0 3.5% 2.9 LaPampa 11,779 11,732 12,079 2.5 0.6% 0.7 LaRioja 19,264 18,888 20,660 7.2 1.0% 0.9 Mendoza 59,816 57,971 58,452 -2.3 2.9% 3.5 Misiones 44,582 45,655 46,880 5.2 2.4% 4.3 Neuquen 20,731 20,190 21,199 2.3 1.1% 1.7 Rio Negro 26,471 25,542 25,492 -3.7 1.3% 2.1 Salta 74,055 73,502 76,047 2.7 3.8% 4.7 SanJuan 37,281 37,425 38,825 4.1 2.0% 1.6 SanLuis 18,647 18,717 18,915 1.4 1.0% 1.1 Santa Cruz 4,195 3,821 3,940 -6.0 0.2% 0.2 Santa Fe 178,192 175,302 180,202 1.1 9.1% 9.9 Santiago del Estero 54,344 54,419 57,348 5.5 2.9% 4.1 Tierra del Fuego 2,750 2,603 2,649 -3.7 0.1 % 0.1 Tucuman 84,924 82,929 85,507 0.7 4.3% 5.7 Total 1,858,657 1,909,185 1,987,875 7.0 100% 100% 147 D. INFRASTRUCTURE Table Dl: Public Private Allocation of Responsibilities in the Infrastructure Sectors Sector Private market share Private sector Cooperatives Public sector Electricity 74% of generation, 100% Generation and Some disperse Nuclear generation, of transmission, and transmission, plus generation (1% of binational hydro distribution to 79% of 18 distribution the market) and schemes, and residential consumers. companies small scale distribution in 11 distribution at the provinces (including provincial and Cordoba and Santa Fe) municipal level Gas Close to 100%. Production and Sub-distribution in transport, plus small cities in some distribution in every provinces region of the country Air 60% of airports, including 33 airports 24 provincial airports all the major ones. Sea Terminals in GBA Some ports transferred and Rosario. by the national government to the provinces Access GBA _ 3 major access routes _ Other access routes under concession Inter-urban 10% of national network 12,200 kms under 204,000kms under roads (50% of Federal and 2% concession, and public sector of Provincial network) 10,000 kms under management CREMA contracts Provincial Largely under public roads management Freight rail 80% of the network 4 lines under I line operated by a concession company where state remains a major shareholder Suburban rail Close to 100% GBA Some provinces such as Buenos Aires and Salta Metro rail 100% GBA Telephone 93% of residential Local, long distance Small provincial and consumers and intemational municipal service services providers Water 62% of residential 12 companies Operating at the Water and sewerage consumers operating at the provincial and companies in certain national and municipal level provinces (Tucuman, provincial level, and Buenos Aires) at least four companies at the municipal level in Buenos Aires Source: Bank staff 148 Table D2: Jurisdictional Responsibility for Infrastructure Sectors Federal Provincial Municipal Electricity Technical control of Electricity distribution and Electricity distribution and generation nationwide, regulation of distribution regulation of distribution in regulation of transmission outside GBA and some cities of the nationwide and distribution generation in disperse provinces. in GBA. systems. Gas Regulation of transport, distribution and sub- distribution of natural gas nationwide. Airports Regulation of the National Administration of System provincial airports (Santa Fe, Neuquen) Seaports Regulation of terminals at Every province either GBA operates or regulates its ports Roads Access roads to GBA, and Local road network inter-urban road network (200,000kms) (30,000kms) Rail Freight railway, suburban Suburban railway in railway and metro in GBA certain provinces Telephone Regulation of local tariffs, and-long distance and international interconnection Water Regulation of water and Regulation and/or Regulation and/or sewerage in GBA operation of water and operation of water and sewerage outside GBA sewerage in small cities Source: Bank staff 149 Table D3: Consolidated Public Expenditure on Infrastructure Sectors 1999 2000 2001 2002 Provisional) (Millions of Pesos) Energy National 371 405 313 331 Provincial 482 460 424 349 Municipal 5 3 3 3 Water National 68 42 29 23 Provincial 327 230 204 277 Municipal 0 0 0 0 Transport National 1,003 801 627 496 Provincial 1,400 1,106 1,132 1,099 Municipal 560 623 614 578 Communications National 110 125 102 81 Provincial 84 76 107 53 Municipal 0 0 0 0 Total National 1,553 1,373 1,072 931 Provincial 2,294 1,872 1,866 1,778 Municipal 565 625 617 580 Source: MECON 150 Table D4: Summary of Major Renegotiations for Transport and Water Concessions Airports The concession of 33 main airports was awarded in 1998 on the basis of an annual canon of $170m. Shortly after the contract was signed, the concessionaire stopped paying the canon. Although a formal renegotiation was not carried out, there have been legal actions and disputes between the Government and the concessionaire. The future of Buenos Aires city airport is a key issue. Ports The concession of six terminals in the Puerto Nuevo section of Buenos Aires port was awarded in 1994 on the basis of an annual canon. Following a substantial shortfall in container traffic from 1998, a merger of two of the terminals was proposed, raising potential concerns about concentration of market power. Dredging and signaling operations for the main waterway-Rfo de la Plata and lower Parana River-were concessioned in 1995 with a government subsidy. After the 2002 devaluation, the rate to be paid by international users was dollarized, without any renegotiation of the subsidy level, in spite of the fact that the devaluation reduced some of the costs of doing business. Postal A concession for postal services was bid out in 1997 to the consonium offering the highest canon. The winning bidder offered $103m per year, while the second highest bid came in at half this value. Almost immediately after the concession contract was signed, the operator stopped paying the canon. Railroads Concessions were awarded for freight railways in 1992, as well as for the Buenos Aires suburban railways and metro in 1994/95. The concessions included both capital and operating subsidies. Freight railways investment obligations have only been partially honored, and the government did not enforce passenger services to pay the established tolls. Suburban and metro rail contracts were first renegotiated in the 1997-99 period, although some agreements could not be implemented because of court actions. These renegotiations were reviewed in 2000-01, and are presently modified again under a railway emergency decree. Most original government financial obligations and new investments are being financed from the Transport Infrastructure Trust Fund through a set of ad hoc decrees and resolutions. Toll roads Since their award in 1990, the 20 inter-urban toll road concessions have been in an almost permanent state of renegotiation. In particular, the government has intervened on some 20 occasions in the space of 13 years to modify toll rates outside of the framework of the original contract. Moreover, within a year of contract signature the canon payments originally scheduled for the concessionaires were completely eliminated and replaced by a growing stream of public subsidies. Water A concession for water and sewerage services in Greater Buenos Aires was awarded in 1993, on the basis of a 27% reduction in the prevailing tariff. * A year later, in 1994, a contract renegotiation led to a 13% increase in the tariff, on the grounds that the costs of the investment program had been under-estimated. However, the corresponding investments were never fully completed. * In 1997 a further contract renegotiation was prompted due to civil opposition to the substantial and compulsory infrastructure charge levied on residents of poor neighborhoods newly connected to the water and sewerage service. As a result, the infrastructure charge was abolished and replaced with a fixed universal access and environmental charge (known as SUMA) on all residential bills designed to finance the coverage expansions. Despite attempts to link the SUMA charge explicitly to the execution of new connections via a trust fund mechanism, today the charge has become part of the concessionaires general revenue base and is not linked to the investment program. Source: Bank staff 151 lable D§: Snummary of Renegotiation Episodes for Inter-Urban Road Concessions Year Month Decision Instrument Type 1990 September Award of road concessions Decree 2039/90 1991 Jan to March Suspension of TOLL collection, later extended Decree 327 and 388/91 Unilateral March (i) Determination of TOLLS, (ii) Elimination of CANON payments, Decree 527/91 Unilateral (iii) Decision to give SUBSIDIES to road concessions 1992 September Approves earlier agreements as a contractual AMENDMENT Decree 1817/92 IST RENEGOTIATION 1993 July Suspension of TOLL increase set for August Ist by Decree 1817/92 Res.SOPyC 168/93 Unilateral October Cancellation of suspension of TOLL increase set by Res.SOPyC 168/93 Res.SOPyC 289/93 Unilateral 1994 July Authorization of TOLL increase as set by Decree 1817/92 Res.SOPyC 388/94 Contractual July Absorption of same TOLL increase via fiscal SUBSIDIES Res.MEyOySP 914/94 Unilateral 1995 April Establishment of terms for contractual AMENDMENTS of concessions Decree 489/95 START OF 2ND RENEGOTIATION July (i) Determination of SUBSIDIES, (ii) Authorization of TOLLS Res.SOP 6/95 Unilateral 1996 August Authorization of new TOLL rate structure starting 9/1/96 Res SOPySP 82/96 Contractual September Approval of contractual AMENDMENT for Concession N° 18 Decree 1019/96 2ND (Caminos del Rio Uruguay) RENEGOTIATION 1997 April Establishment of amount to be paid as SUBSIDY Res.SOP 510/97 Contractual July Agreement to defer TOLL revision set by contract Res.MEyOySP 800/97 Mutual Agreements 1998 August Agreement to continue AMENDMENT of concessions Res.MEyOySP 1089/98 SEQUEL TO 2ND RENEGOTIATION 1999 May Approves SUBSIDY of toll paid by vehicles transporting agricultural products Res.MEyOySP 573/99 Mutual Agreements October Approves SUBSIDY of toll paid by cargo vehicles Res MEyOySP 1250/99 Unilateral December Extends term of transport SUBSIDY scheme (Res MEyOySP 1250/99) Res MIV N' 2/99 Unilateral 2000 January/ March Cooperation Agreements to reduce TOLLS. or maintain/raise SUBSIDIES s/N° Mutual Agreements March to Dec. Extends term of transport SUBSIDY scheme on six separate occasions Res.MIV 91, 182,216, 255,005,064/00 April Modifies TOLLS set by Res. MIV 9 1/00 Res.MIV 118/00 y Unilateral Res.ME 343/00 July Suspends TOLL adjustments according to LIBOR rates Res.SOP 68/00 Unilateral December Letter of Agreement aimed to regularize contractual AMENDMENTS 0N. Mutual Agreement 2001 January Approves agreements to introduce contractual AMENDMENTS Decree 92V01 END OF 2ND RENEGOTIATION January Transfers control of concessions from DNV to OCCOVI. Decree S7/01 Govemment Decision June Creates diesel tax to be used in part to pay-off govemment debt with Decree 802/01 Govemment Decision concessionaires and raise SUBSIDIES to finance further TOLL reductions June Establishes new TOLLS and SUBSIDIES as a result of Decree 802/01 Res.SOP 110/01 Unilateral July Agreement to improve Competitiveness and Employment paid with diesel tax Decree 929/01 Unilateral revenues (refer to Decree 976/01) September Establishes procedures to calculate and verify TOLLS, SUBSIDIES and debt Res. SOP 190/01 Unilateral November Establishes rules for the allocation of diesel tax revenues Decree 1377/01 Government Decision 2002 January Congress declares Public emergency Law 25.561 Govemment Decision February Creation of Commission for the Renegotiation of Concession Contracts Decree 293/02 START OF 3RD I___________________________________________ _____________ RENEGOTIATION 2003 February Announcement of Public Consultation to discuss concessionaires complaints Res. SLA 26/03 RENEGOTIATION Source: Bank staff Notes: DNV-National Road Agency; MEyOySP-Ministry of Economy, Public Works and Public Services; OCCOVI-Toll Road Concessions Regulatory Agency; Res.-Resolution; SOP-Secretary of Public Works; SOPyC-Secretary of Public Works and Roads; SOPySP-Secretary of Public Works and Services. 152 Table D6: Summary of Efficiency Gains and Extent of Pass Through to Tariffs Electricity Gas Water and, Distribution Distribution Sewerage Telecoms Approximate average annual change in costs due to gains in -1.0% -2.9% -1.9% -3.9% Total Factor Productivity Average annual change in tariffs allowed by regulators over -0.% -0.8% -1.8% -0.6% the same period Refers to Greater Buenos Aires only, TEP gains in three Provinces found to be much higher 6.1% on average. Source: Estache, 2003 Table D7: Status of Renegotiation of Infrastructure Concessions Sector No. of Income Debt Status of renegotiations contracts flows obligations Electricity 7 Pesos Foreign In process, attempted interim adjustments by Executive transmission Decree were overturned by judiciary Electricity 3 Pesos Foreign In process, attempted interim adjustments by Executive distribution Decree were overurned by judiciary Gas 2 Pesos and Foreign In process, attempted public audiences and interim transportation dollars adjustments by Executive Decree were overtumed by judiciary Gas 9 Pesos and Foreign In process, attempted public audiences and interim distribution dollars adjustments by Executive Decree were overtumed by judiciary Water and I Pesos Foreign In process, attempted public audiences were ovenurned sewerage by judiciary Telecom 2 Pesos Foreign In process Urban access 4 Pesos Foreign In process toll roads Railroads 11 Pesos Domestic In process, but with increased subsidy from Transport Infrastructure Trust Fund already approved Ports 4 Pesos and Foreign Complete, tariffs have been redollarized at a rate if 1: 1, dollars although some regulated rates were recently pesified at a rate of 1: I Inter-urban 15 Pesos Domestic In process, but with increased subsidy from Transport toll roads Infrastructure Trust Fund already approved Dredging I Dollars Domestic Complete, tariffs have been redollarized at a rate of 1:1. Total 59 Source: Bank staff 153 Table D8: Status of ICSID Cases already Filed against Argentine Government Claimant Sector Date Status Registered Aguas del Aconquija & Vivendi Water & sewerage 02/19/97 In process Enron Corporation & Ponderosa Assets Gas transportation 04/11/01 Tribunal appointed CMS Gas Transmission Company Gas transportation 08/24/01 Tribunal appointed Azurix Corporation Water & sewerage 10/23/01 Tribunal appointed LG&E Energy Corporation consortium Gas distribution 01/31/02 Pending tribunal Siemens A. G. Telecommunications 07/17/02 Tribunal appointed Sempra Energy Intemational Gas supply & distribution 12/06/02 Pending tribunal AES Corporation Electricity generation & distribution 12/19/02 Pending tribunal Camuzzi International S.A. Gas supply & distribution 02/27/03 Pending tribunal Source: Bank staff from ICSID website. Table D9: Breakdown of Infrastructure Spending by Budgetary Mechanism Current General Intra- Extra- Fiscal Total Percentage pesos Budget Budgetary Budgetary Credits Expenditure Outside $millions Earmarked Trust General Funds Funds Budget Energy 430 253 149 0 832 48 Telephone 134 0 115 0 249 46 Transport 1,787 385 578 240 2,990 40 Water 300 0 0 0 300 0 Total 2,651 638 842 240 4,372 39 Source: MECON 154 Table D10: Evolution of Energy and Transport Subsidies Current pesos $millions 1999 2000 2001 2002 2003 1. Energy Subsidies a. Direct subsidy to pensioners for power and gas bills 34.0 38.3 43.6 43.6 43.0 National budget 34.0 38.3 43.6 43.6 43.0 Trust funds 0.0 0.0 0.0 0.0 0.0 b. Patagonian gas subsidy 82.5 90.0 42.5 100.0 100.0 National budget 82.5 90.0 42.5 0.0 0.0 Patagonian gas subsidy trust fund 0.0 0.0 0.0 100.0 100.0 c. Federal electricity transmission subsidy 0.0 0.0 29.1 1.8 83.7 National budget 0.0 0.0 0.0 0.0 0.0 Federal electricity transmission trust fund 0.0 0.0 29.1 1.8 83.7 d. Provincial electricity subsidies 187.8 203.9 235.8 260.0 358.1 National budget 0.0 0.0 0.0 0.0 0.0 Various earmarked funds 187.8 203.9 235.8 260.0 358.1 * Fondo Nacional de Energia Electrica (FNEE) 73.7 66.1 92.7 98.0 132.0 * Fondo de Desarrollo Electrico del Interior (FEDEI) 82.2 92.2 106.4 103.7 130.6 * Fondo Santa Cruz (Ley 63.681) 22.8 26.1 26.3 23.9 26.2 * Fondo Especial Salto Grande (Ley 25.671) 8.1 19.5 10.3 33.3 67.6 * Energia Eolica 1.0 0.0 0.1 1.1 1.7 HI. Transport Subsidies a. Subsidies to metrorail concessionaires 41.2 47.8 25.3 93.7 100.1 National budget 41.2 47.8 25.3 0.0 0.0 Transport infrastructure trust fund 0.0 0.0 0.0 93.7 100.1 b. Subsidies to interurban road concessionaires 30.7 45.1 212.4 358.3 257.9 National budget 30.7 45.1 12.7 0.0 0.0 Transport infrastructure trust fund 0.0 0.0 199.7 358.3 257.9 c. Subsidy to metropolitan bus operators 0.0 0.0 0.0 185.6 306.7 National budget 0.0 0.0 0.0 0.0 0.0 Transport infrastructure trust fund 0.0 0.0 0.0 185.6 306.7 d. Other transport subsidies 0.0 0.0 11.1 150.5 387.5 National budget 0.0 0.0 0.0 0.0 0.0 Transport infrastructure trust fund 0.0 0.0 11.1 150.5 387.5 m. Total subsidies Total subsidies 376.2 425.1 599.8 1,193.5 1,637.0 Total subsidies from national budget 188.4 221.2 124.1 43.6 43.0 Total subsidies from special funds 187.8 203.9 475.7 1,149.9 1,594.0 Total subsidies to energy 304.3 332.2 351.0 405.4 584.8 Total subsidies to transport 71.9 92.9 248.8 788.1 1,052.2 Total subsidies as % of total expenditure 0.7% 0.8% 1.2% 2.3% 2.5% Total subsidies as % of GDP 0.1% 0.2% 0.2% 0.4% 0.4% Source: MECON 155 Table Dll: Summary of Social Policies for Infrastructure Services National schemes Company specific schemes Electricity Electricity Tariff Compensation Fund, Electricity distributors in 18 Provinces apply social financed from $0.0241kWh surcharge on tariff schemes, some funded directly from the ETCF, wholesale electricity market, raises annual others through cross-subsidy. resources of $ lIWm to finance a wide variety of tariff discounts to rural consumers, industrial customers, and social cases in the Provinces. Gas Patagonian gas subsidy, funded from None. surcharge of $0.004/m3 on gas sales, raises annual resources of $100m to finance tariff discounts of around 75% on basic fuel requirements in the Patagonian region. Telecom Universal Service Fund, financed from 1% Major operators offer discounted payment plans to levy on telephone bills, raises $11 5m small consumers and pensioners, as well as annually. These are intended to cover a range discounted public payphone services in low income of social and public telephony programs. areas and isolated communities. flowever, the fund is not yet operational. Transport Transport Infrastructure Trust Fund, financed Most operators practice a number of specific from 18.5% tax on diesel, allocates 40% of its franchises for pensioners, school students, and other resources to subsidize urban rail and bus privileged groups. services in GBA, amounting to $170m annually. Water None. Water utilities in GBA and 7 Provinces apply social tariff schemes, some funded directly from the General Budget, others through cross-subsidy. Source: Adapted from Foster and CEER, 2003 156 Table D12: Comparative Analysis of Social Tariff Proposals .________ __ Source of finance Nature of benefit Beneficiaries ADESPA Cross-subsidy, tax Exemption from Those meeting poverty exemption or subsidy emergency tariff increase criteria as per census or household surveys. ADIGAS Cross-subsidy between Exemption from Jefes y Jefas and low customers. emergency tariff increase income pensiones, with <50% average consumption Electricity sector Off-setting tax reduction in Exemption from Unclear tax burden on the sector emergency tariff increase Gas sector Off-setting tax reduction in Exemption from Unclear tax burden on the sector emergency tariff increase Congress Exemption from VAT, plus Tariff discount of 40% Beneficiaries of other 'concessionaire finance' social programs Metrovias Direct govemment subsidy Tariff discount of 20% Beneficiaries of other (from Fondo Fiduciario) social programs Source: Bank staff 157 Table D13: Comparison of Design Features of Regulatory Agencies at the Federal Level Year PSP Year agency Legal Renit Reporting Governance mechanism Financing Budget 2002 initiated established Basis ministry Mechanism (millions) CNC 1989 1990 Decree Regulates local Secretary of 5 members nominated by Regulatory surcharge of 0.5%, with any $ 19.98 1.185/90 and long distance Communications, President (and one by surplus transferred to Treasury telephony, as well Ministry of Governors' Congress) as the postal Economy CNRT___I service CNRT Longstanding 1996 Decree Regulates road Secretary of 5 members nominated by Regulatory surcharge based on number $ 6.79 for road 1.388/96 and rail transport Transport, the President of passengers, plus charges for vehicle transport, services under Ministry of inspection, and fines 1993 for rail federal Economy sector jurisdiction ENARGAS 1992 1992 Law 24.076 Regulates gas Secretary of 5 members nominated by Regulatory surcharge, plus fines, $ 10.43 transportation and Energy, Ministry the President, with adjusted to budget approved by Congress distribution of Economy Congressional approval and executed through the budget system nationwide ENRE 1992 1992 Law 24.065 Regulates Secretary of 5 members nominated by Regulatory surcharge, plus fines, $ 12.56 electricity Energy, Ministry the President, with adjusted to budget approved by Congress generation and of Economy Congressional approval and executed through the budget system transmission nationwide, and distribution in GBA _ _ _ _ _ __ _ _ _ ET'OSS 1993 1993 Decree Regulates Aguas Subsecretary of Comminee with 2 Regulatory surcharge of 2.67% on water $13 (aprox.) 999/92 Argentinas Water Resources, representatives from City of and sewerage biHs No esta concession in Ministry of BA, 2 from Province of incluido en GBA Economy BA, and 2 from Federation el Presupuesto Nacional ORSNA 1998 1998 Decree Oversees National Secretary of 3 members nominated by Direct transfer of resources from the $ 7.38 375/1997 Airports System Transport, Executive Branch, and the Treasury Ministry of fourth by the Provinces ___ ___ ___ __ ___ ___ _ _ ___ ___ ___ __ Econom y OCCOVI/ 1990 2001 Decree Monitors Presidency Director and Deputy Regulatory surcharge of $ 5.2 (OCCRABA) (1993) 87/2001 compliance with Director nominated by * 0.5% on net toll revenues (Decree toll road Executive Branch * 1 % on service area revenues 1994/93) concessions on * Revenues from toll round-up access to GBA, and inter-urban _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ corridors _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Source: Bank staff 158 Table D14: Comparison of Design Features of Regulatory Agencies at the Provincial Level Year PSP Year agency Legal Remit Reporting Govemance mechanism Financing Budget 2002 initiated established Basis ___ ministry mechanism (millions) EPRAC, 1999 1999 Provincial Regulates provincial water Provincial 4 member commission Regulatory surcharge of 3% Resources Misiones Law 3391 services Executive appointed on sector revenues, however transferred Branch * President and Vice-president resources are channelled via on demand by Executive Branch former water company, since by Executive * 2 Directors appointed by agency has not yet been Branch Congressional majority and formally constituted. primary opposition ERSEP, 1997 2001 Provincial Multisectoral, regulating Provincial 6 member comission appointed Regulatory surcharge on $ 3.5 C6rdoba Law 8835 provincial: Ministry of * 3 by Executive Branch sector revenues: * Water (Aguas Cordobesas Public Works * I by primary opposition * 1.2% for water + 200 coops) * 1 by secondary opposition * 0.4% for electricity * Electricity (EPEC) * I by Consumers Assoc'ns * 1.5% for public transport * Inter-urban transport * 0.5% for road concessions . Road concessions OCEBA, 1997 1997 Provincial Regulates provincial Provincial 5 member commission Regulatory surcharge of $6.4 Buenos Aires Law 11.769 electricity distribution Ministry of appointed by Executdve Branch 0.05% of sector revenues and Decree Public Works 1208/97 ORAB, 1999 1999 Provincial Regulates provincial water Provincial 5 member commission Regulatory surcharged $ 1.5 Buenos Aires Law 11.820 services Ministry of appointed by Executive Branch established by law was and Decree Public Works suspended. currendy 743/99 dependent on surcharge for bulk sale of untreated water. Source: Bank staff 159 Table D115: Comparison of Special Funds for I[nfrastructure at the Federal Level Name Year Type Sector Legal Govemance Source Destination Financing Bud et ($m) of Fund Mechanism Arrangements of Funds of Funds Mechanism 2002 2003 Fondo de 1932 Earmark Transport Law 11658 Managed directly by National Surcharge on petroleum Used to finance Federal Subsidies 385.0 ,? Vialidad ed Roads Agency products and Provincial road Nacional network Fondo 1960 Earmark Energy Laws 15.336 Federal Electricity Council Surcharge of 87% invested in electricity Subsidies 103.5 130.6 Nacional de ed & 24.065 chaired by Energy Secretary $0.96/MWh on infrastructure, 12% lent to and loans Energia with representation from the wholesale electricity cooperatives and Electrica six regions of the country. market, plus 2.19% tax municipalities, 1% to (FEDEI) on liquid fuels and administrative natural gas (Law expenditures _________ ___________ _____________ 23.966) Fondo 1992 Earmark Energy Law 24.065 Federal Electricity Council Surcharge of Tariff subsidies in high Subsidy 97.9 132.0 Nacional de ed chaired by Energy Secretary $1.44/MWh on cost provinces, also Energfa with representation from the wholesale electricity general budget Electrica six regions of the country. market (FCT) Fondo 2000 Trust Energy Law 15.336 Energy Secretary, plus three Surcharge of Finance of electricity Subsidy 49.3 55.0 Fiduciario para & Decree committee members of which $0.60/MWh on transmission projects in el Transporte 657/99 two chosen by Provinces wholesale electricity provinces Electrico market Fondo 2001 Trust Multi- Laws 24.850 Independent committee Shares from sale of 50% to finance Federal Loans 35.9 124.0 Fiduciario de sector & 25.414, & including representation from Banco Hipotecario, plus and 50% to finance [nfraestructura Decrees Ministry of Finance, Provinces loans from Treasury provincial infrastructure Regional 1299/00 and and private sector and multilateral projects _ 965/01 agencies Fondo 2001 Trust Transport Decrees UCOFIN within Ministry of Tax of 18.5% on the 60% to subsidize road Subsidy 578.0 1,204.0 Fiducario del 802/01, Economy price of diesel concessions, and 40% to Sistema de 976/01, subsidize urban Infraestructura 1377/01, & transportation (including de Transporte 652/02 rail and bus) Fondo 2001 Trust Water Decree UCOFIN within Ministry of Surcharge of $0.05/m3 Investments in drainage Subsidy 248.5 239.2 Fiduciario de 1381/01 Economy of GNC and gasoline for flood prone provinces lnfraestructura Hidrica Fondo 2002 Trust Energy Law 25.725 Energy Secretary, Ministry of Surcharge of 7.5% on Subsidy to consumption of Subsidy 100.0 116.3 Fiduciario de Economy the price of natural gas natural gas in the Subsidios al at the wellhead. Patagonia and Puna Consumo regions Residencial del Gas Source: Bank staff 160 Table D16: A Brief History of Road Funds in Argentina 1930 to 1949. The use of road funds in Argentina dates back to Law 11658 of 1932, which established Federal responsibility for the naiional road network, and created a National Road Fund to be financed with the proceeds of a specific tax on petroleum products. The same law also created the National Roads Agency to be responsible for designing, building and maintaining the national road network, as well as a system of federal transfers to support the development of provincial roads. This policy proved successful in promoting the development of the national road network, as well as the creation of a critical mass of human resources in the National Roads Agency. 1950 to 1969. However, by the 1950s the value of the specific tax on petroleum products had been substantially eroded due to inflation. With a view to strengthening the system, Decree 505 was issued in 1958. The new decree replaced the specific tax with an ad valorem tax so as to sustain revenues. It also improved coordination with the provinces by establishing a principle of co-participation, and creating a Federal Roads Council to bring together representatives from the Provincial Roads Agencies. 1970 to 1999. During the 1970s and 1980s growing fiscal pressures meant that the National Road Fund was increasingly diverted by the Treasury to meet other needs. This process was increasingly formalized with Law 23966 of 1991, which allowed changed the name of the tax on petroleum products allowing its proceeds to be allocated outside the road sector, and Law 24699 of 1996, which established that 100% of the revenues of diesel tax and 21 % of the revenues of petroleum taxes would be used to finance state pensions. This last measure, which coincided with financial crisis in the provinces, greatly reduced the resources available for the provincial networks. 2000 onwards. The 1990s saw a shift towards major private participation in construction, rehabilitation and maintenance of significant portions of the Federal road network. By the end of the decade, there was a desire to deepen this approach into a major Build Operate Transfer program for new roads modeled on the British Private Finance Initiative, which led to the creation of the Transport Infrastructure Trust Fund by Decree 1299 of 2000. The fund was originally administered by the Ministry of Infrastructure but later transferred to the Ministry of Economy. The financing mechanism established for the fund was a specific tax of $0.05 per liter of diesel, later modified to an ad valorem tax of 18.5%. In addition, the fund was supposed to receive revenues from a standardized toll of $0.75 per 100 kilometers on the concessioned inter-urban road network. However, this mechanism was never established. Within a year of its creation, the allocation rules for the fund underwent a series of changes that diverted resources away from the original concept of financing new infrastructure and towards a system of subsidies for existing private transport concessionaires (see Table X for.further details). More recently, Budget Law 25,565 of 2002 transferred $100m of resources from this fund back into the general budget. Source: Agosta, 2003 161 Table D17: Evolution of Transport Infrastructure Trust Fund Date 06/15/01 11/21/01 02/26/02 04/22/02 06/05/02 Legal instrument Decrees 802, Res. 531/01 Decree 373/02 Decree 652/02 Joint Res. 11 976,1377/01 (ME) & 61 (MP) Purpose Creation and Establishment Changes to Change of tax Incorporation of allocation of of UCOFIN Res.531/01 rate and new beneficiaries revenues incorporation of new beneficiaries Provisions (A) Diesel tax rate $0.05 per liter 18.5% (=$0.12 per liter) (B) Beneficiaries Road National and Metropolitan concessions and provincial bus Buenos Aires rail priority works operators and metro concessionaires (C) Fund allocations SISVIAL 80% SISVIAL 60% (of which 70% (of which 70% to Provinces) to provinces) SISTAU 0% SISTAU 26% SIFER 20% SIIER 14% (of which 75% (of which 14% to Provinces) to Provinces) (D) Institutional MIV MIV ME ME location Source: UCOFIN Notes: ME-Ministry of Economy; MIV-Ministry of Infrastructure and Housing; MP-Ministry of Planning; UCOFIN-Trust Fund Coordination Unit; SISVIAL-Roads Fund; SISTAU-Public Transport Fund; SIFER-Railroads Fund. 162 Figures Figure D l(a Breakdown of Public Expenditure on Infrastn eture by Sector 6% 21% 0 Energy -9 Water 66% :_- Transport U Communications Source: MECON Figure D1(b): Breakdown of Public Expenditure on Infrastructure by Jurisdiction 30% .J m* National . . o~~~~~~~~ Pro~vincial 53% 0 Municipal Source: MECON Figure D2: Evolution of Income Stream for Inter-Urban Roads Concessions 900 El Fondo Fiduciado E 800 4D700 -oao - Pagos de la Deuda c° 600 - 0 co .2 m ~~~~~~~~~~~~~Otros subsidios, 400 - compensaciones y 0, exenciones 300 200 ; - . ~,.rI n <; Compensacion - 200 -~ -:, -r ,llmdemnizatora co 100 1 ..... z ' , >F;, |Decreto 1.817/92 0 0 Recaudacion por C;1S99NR9zN9N9 69bN99f°o CP4 c E os P P peajes Source: OCCOVI 163 IFigure D3(a): IlnternDationual Comparison off Residen ta Electzricity IPrices Ecuador Uruguay Brasil Peru Chile Argentina (US$1 =$1) Colombia Paraguay Venezuela Argentina (US$1 =$3) 0 20 40 60 80 100 120 Electricity bill (US$/month) Source: CIER Figure D3(b): International Comparfison off ResnidenalR Gas Prices Brasil Uruguay -T I Chile |-i Mexico Bolivia | | Argentina (US$1=$11)' Argentina (US$1 =$3) 0 10 20 30 40 50 60 Gas bill (US$/month) Source: Stone & Webster 164 Figure D3(c): International Comparison of Residential Water Prices (for consumption of 15m3 per month) Chile Brazil ____._ Argentina (US$1=$1) Bolivia . __...__ Peru _ _ -- Argentina (US$1 =$3) Colombia 0 10 20 30 40 50 Water bill (US$ per month) Source: LECG Figure D3(d): International Comparison of Residential Telephone Prices (monthly subscription plus 300 minutes local calls) I II _ Argentina (US$1=$1) Bolivia-- Chile . . *-. .- Peru ___ -_i_. . 1 Mexico C. Uruguay l___l_i - Argentina (US$1=$3) Brasil . . 0 5 10 15 20 25 Local telephone bill (US$/month) Source: Telef6nica 165 Percenitage tax levied on household electncity bills Percentage tax levied on household bills Percentage tax levied on household bilis - f - - - - - -~~~~~~~~~~~~~~~~~ ~UnitedKingdom i Poland I T] United Kingdom Prua Switzerland ILuxembourg PotglIe UK GreceLuxembourg I Czech Republic Switzerland Switzerland - ivkRpbi Japan - p lvkRpbi 0n 0 -, 0 C Portugal -1 Slovak Republic -ugr France 0 ~~~~~~~~~~~I Hungary o ~~~~~~~~~~~~~~~~~~ma tlI reland Belgium ~~~~~~~~~~~~~~~Ireland' Te New Zealand Neltherands 0 OQ laSpain .2 Spain Meic Germany CD ~~~~~~New Zealand CDPln PI Greece 0 . FranceSpi o Italy Turkey Czech Republice Austria C CDC 0Czech Republic o Turk(ey r ti i Finland Hungary ~~~~~~~~~~~~~Poland Turkey n Aretn FiladFinland (IQ Finland ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Austria Norway Austria Denmark to ~ ~~~~~~~~~~~Argentina ry Sweden rANetherlands Netherlands U Argentina Denmark - - - - ~~~~~~~~Denmark -j Figure D5(a): Illustration of flow of funds to electricity sector $24M WhM.pc. . URY, r S24MWh m-~~ELECTRICITi' $ 06 MWh 2,19% 40%L6O% T ItUDGEII. ED FCT _ -i-.45i ND2i,lfill $1 03m $9 m . _ 12% 87% - Regulation % Infrastructure Tariffcompetsations L investments and and general budget _ Others Lans to generalbudget co-operatives Administrative _ Yacyreta ~ ~ ~~~ expenditures i Hydro Schernie _Nuclear_ Ra:r S.Jt.iamnex ~ FTSelf-faaiced Self-fmiwcd . , l rl sxmff > fot ia wdnffs <> vltrfJ > Figure D5(b): Illustration of flow of funds to transport sector 14 ~ ' 60"0 %16r S322mi SXn,14$ Source: Bank staff 167 Figure D6: Allocation of Resources from Electricity Tariff Compensation Fund o 40% 35% Q 30% ;25% 20% D 15%- 10% / *D 5% 0 0% -,---n---- _--i 0. ' &' CO§ o0 E SISTRoAN k ~ 2s0i, -i (SIFER) 0 c e~~~ ~ too c ,p1l., % : SSRN o ETCF resources ; Poor households Source: Adapted from information supplied by FNEE and INDEC Figure D7: Alocation of Resources from Transport Infrastructure Trust Fund 400 U Gerneral Budget 350- A o~~~0Surplus 300 250 a~~~~~~~~~~~E SISTRANS (SIFER) 200- SISTRANS (SISTAU) I ~ ~ ~ ~ ~ : IJ~~1SISVIAL (Public Works) 100- 200 0120 2 00 2 0 1 (Concessiorrn) 0 0 ~~~~~~~~~~~~~Liqtwdiiy Re eive Source: Based on information supplied by UCOFIN 168 E. PUBLIC EXPENDITURE MANAGEMENT Table El: Budget Execution: Modifications and Actual Expenditure Institution Modifications as Expenditure as Modifications as Expenditure as % of Approved % of Final % of Approved % of Findl Budget (2000) Budget Budget (2001) Budget Congress 0.04 95.8 7.3 96.8 National Judiciary 1.6 98.1 2.6 93.4 Public Ministry -1.8 98.5 -1.9 99.1 Presidential Office 11.8 92.1 -19.2 93.9 Chief of Cabinet -2.2 86.6 -4.5 64.9 Ministry of Defense 1.5 98.2 -4.4 102.0 Ministry of Economy 6.6 91.9 3.5 84.0 Ministry of Education -2.2 94.5 -7.7 95.8 Ministry of External Affairs 1.6 97.9 -12.8 98.2 Ministry of Infrastructure 2.8 88.8 -18.4 78.4 Ministry of Internal Affairs 0.4 97.3 -5.4 97.4 Ministry of Health. 0.6 88.3 -19.7 90.0 Ministry of Justice -0.8 92.3 -7.7 93.9 Ministry of Labor 2.0 94.5 -1.5 92.2 Ministry of Social -0.5 93.2 -8.4 102.4 Development Debt Service 14.1 98.6 10.3 95.3 Treasury Obligations 21.2 97.0 9.2 76.5 Total 4.8 95.4 -0.3 92.6 Source: Bank staff estimates 169 Table E2: Rigidities in National Administration Expenditure (2003 Budget) In Millions of Pesos Rigid Expenditure Amount % of Total Budget Expenditure Mandatory Payments: To Households and enterprises (1) 11169 17 To Provinces 3040 5 To Universities 1847 3 To International Organizations 593 1 Capital Transfers 1683 2 Pensions 17911 27 Public Employee Salaries 7127 11 Debt Service 14,582 22 Subtotal Rigid Expenditures 57,952 88 Discretionary Expenditures: Transfers to Public Sector 3743 5 Investment Expenditure 721 1 Operating Expenditure and Other 3757 6 Subtotal Discretional Expenditure 8221 12 Source: Bank staff estimates 170 Table E3: Emergency Decrees 2002 Institution Congress 5,000 Judiciary -2,000 President's Office 154,813 Internal Affairs 634,866 External Affairs 179,500 Justice 88,959 Defense 254,672 Economy 379,527 Production 44,381 Education 176,126 Labor 1,470,000 Health 341,000 Social Development 214,561 Debt Service 10,330,109 Treasury Obligations 13,458 10 emergency decrees were issued in 2002: 8 dealt with program and operating expenditure; two principally with debt service. Debt Service accounted for 72 percent of the expenditures auth6rized by emergency decree. Source: Bank staff estimates 171 Table E4: Consolidation Bonds (In Millions of Pesos) Separate Legislation In Annual Budget Law 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 BOCON 1,500 3,800 2,100 1,100 500 700 900 BOCON 3,000 3,000 1,900 500 400 400 300 Other 2,410 4,000 1 ,400 1,000 100 100 100 Total 6,910 10,800 5,400 2,600 1,000 1,200 1 ,300 1,500 1,310 1 ,350 370 Since 1996, debt consolidation has been authorized in the budget law rather than in separate legislation. But of the total consolidated since 1992, 76 percent was authorized in separate legislation. Source: Bank staff estimates 172 - IMAGING Rpr O. 25 AR Ty a- ER