Report No. 20236-ME

Mexico:Fiscal Sustainability(In Two Volumes) Volume I: Executive Summary

June 13, 2001

Mexico Country Management UnitPREM Sector Management UnitLatin America and the Caribbean Region

Document of the World Bank

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

CURRENCY EQUIVALENTSCurrency Unit - Mexican Peso (mxp$)

EXCHANGE RATE MARCH 17,20009.35 MXP / I USD

WEIGHTS AND MEASURESMetric System

FISCAL YEARJuly 1 - June 30

ABBREVIATIONS AND ACRONYMSADE Acuerdo de Apoyo Inmediato a Deudores de la BancaADEFAS Adeudos de Ejercicios Fiscales AnterioresASA Aeropuertos y Servicios AuxiliaresBANCOMEXT Banco Nacional de Comercio Exterior, S.N.C.BANJERCITO Banco Nacional del Ejercito, Fuerza Aerea y Armada, S.N.C.BANOBRAS Banco Nacional de Obras y Servicios Publicos, S.N.C.BANRURAL Banca Nacional de Cr6dito Rural, S.N.C.BoM Banco de MexicoCAPUFE Caminos y Puentes Federales de Ingresos y Servicios ConexosCFE Comisi6n Federal de ElectricidadCNBV Comisi6n Nacional Bancaria y de ValoresCONASUPO Compafila Nacional de Subsistencias PopularesEMBI Emerging Market Bond IndexFAMEVAL Fondo de Apoyo al Mercado de ValoresFARAC Fideicomiso de Apoyo al Rescate de Autopistas

FIDEC Fondo para el Desarrollo ComercialFIDELIQ Fideicomiso Liquidario de Instituciones y Organizaciones Auxiliares del CreditoFINA Financiera Nacional AzucareraFINAPE Programa para el Financiamiento del sector Agropecuario y PesqueroFIRA Fideicomisos Instituidos en Relaci6n con la AgriculturaFNM Ferrocanriles Nacionales de MexicoFOBAPROA Fondo Bancario de Protecci6n al AhorroFOPYME Programa de Apoyo Financiero y Fomento a la Micro, Pequefta y Mediana EmpresaFOVI Fondo de Operaci6n y Financiamiento Bancario a la ViviendaGDP Gross Domestic ProductIMF Intemational Monetary FundIMSS Instituo Mexicano del Seguro SocialINEGI Instituto Nacional de Estadistica, Geografia e InformaticaIPAB Instituto de Protecci6n al Ahorro BancarioISSSTE Instituto de Seguridad y Servicios Sociales de los Trabajadores del EstadoLFC Luz y Fuerza del CentroLOTENAL Loteria Nacional para la Asistencia PublicaMXP Mexican PesosNAFINSA Nacional Financiera, S.N.C.NIPA National hicome Products AccountOECD Organization for Economic Co-operation and DevelopmentPEMEX Petr6leos MexicanosPIPSA Productora e Importadora de PapelSCNM Sistema de Cuentas Nacionales de MexicoSCT Secretaria de Comunicaciones y TransporteSHCP Secretaria de Hacienda y Cr6dito PublicoVAT Value Added Tax

The Bank team that produced this report was headed by Stephen Everhart (LCSPE)-task manager, under the guidance ofMarcelo Giugale (LCCIC)-program team leader. Members of the Bank team include: Craig Bumside (DECRG), JoostDraaisma (LCC1C), Robert Duval (LCCIC), Andrew Feltenstein (IMF, Virginia Tech), Russ Murphy (Virginia Tech),Claudia Sepulveda (LCSPR), and Aaron Schwartzman (Ernst & Young-Mexico). Production assistance was provided byMichael Geller and Elizabeth Toxtle (LCCIC).

The Bank appreciates the invaluable support and advice of Eliana Cardoso (LCSPE) and Vicente Fretes-Cibils(LCC4C). This study was undertaken under the general direction of Mr. Olivier Lafourcade (Director, LCCIC). Peer

reviewers are: Messrs. Luis Serven (Lead Specialist - Regional Studies, LCSPR) and Anwar M. Shah (Principal

Evaluation Officer, OEDCR).

ii

VOLUME 1: EXECUTIVE SUMMARYTABLE OF CONTENTS

Fiscal Sustainability,Mexico: A Synthesis

Rationale for the Study ..................................................... IIssues and Focus ..................................................... 2Fiscal Policy, Business Cycles, and Growth in Mexico ...................................................... 2Infrastructure, External Shocks, and Mexico's Fiscal Accounts ..................................................... 3Infrastructure, Private Costs, and Payoffs from Additions to Infrastructure ................................................... 5Fiscal Impact of Contingent Liabilities ..................................................... 6Fiscal Deficit, Public Debt, and Fiscal Sustainability in Mexico .................................................... 10An Extension: Balance Sheet Approach and Quality of Fiscal Adjustments ................................................ 16

The Mexican Case .................................................... 19Implications of the Balance Sheet Approach ..................................................... 23

Conclusions: The Link Between Fiscal Sustainability and Fiscal Reform ................................................... 25References ..................................................... 27

List of Tables

Table E. 1 Estimate of the Overall Cost of the Financial Rescue, June 1999Table E.2 Contingent Liabilities Recognized by the Federal GovernmentTable E.3 Mexico Federal Debt as a Percentage of GDP

List of Figures

Figure E.1 Concentration and Growth of Subnational Debt, 1994-1998: Selected StatesFigure E.2 Gross Federal Debt as Percent of GDP: International BenchmarksFigure E.3 Selected Latin American Eurobond SpreadsFigure E.4 Mexico Budget Indicators: 1980 - 1998Figure E.5 Tax Revenue as Percent of GDP, Selected Countries 1992 - 1998Figure E.6 Primary Deficit vs. Public Investment (percent of GDP)Figure E.7 Primary Deficit vs. Public Investment (percent of GDP) 5 UMI CountriesFigure E.8 Primary Deficit vs. Public Investment (percent of GDP) 6 LMI CountriesFigure E.9 Primary Deficit vs. Privatization Revenues (percent of GDP)Figure E. 10 Primary Deficit vs. Public Investment (percent of GDP) MexicoFigure E. 1 1 Primary Deficit vs. General Government Consumption (percent of GDP) MexicoFigure E.12 Public Investment vs. Oil Prices MexicoFigure E. 13 Programmable Expenditure Decomposition (percent of GDP) Mexico.Figure E.14 Primary Deficit vs. Privatization Revenues (percent of GDP) MexicoFigure E. 15 Deficit Reduction and Oil DependenceFigure E. 16 Components of Tax Revenues as a Percent of GDP, Mexico 1980-1998

iii

VOLUME II: BACKGROUND PAPERSTABLE OF CONTENTS

Chapter 1. Fiscal Policy, Business Cycles, and Growth in Mexico

Perspectives on Mexico's Fiscal Accounts from 1980-98 ............................................. 2The Business Cycle in Mexico ............................................. 3Trends and Cycles in Mexico's Fiscal Accounts ............................................. 6The Cyclically Adjusted Budget Surplus in Mexico ............................................. 16Methods for Computing the Cyclically Adjusted Budget Surplus ................. ............................ 17Budget Surplus Estimates for Mexico ............................................. 19How Fiscal Policy and Output Affect Each Other in Mexico ............................................. 23A Small VAR Model of the Mexican Economy ............................................. 24How Does the Fiscal Surplus Affect Output? ............................................. 25Dynamic Behavior of the Fiscal Surplus ............................................. 27Policy Conclusions ............................................. 29References ............................................. 31Appendix ............................................. 33

Chapter 2. Infrastructure, External Shocks, and Mexico's Fiscal Accounts

Background ............................................. 43Model Structure ............................................. 45Production ............................................. 46Banking ............................................. 48Consumption ............................................. 48The Government ............................................. 49The Foreign Sector and Exchange Rate Determination ............................................. 49Money Supply ............................................. 50Data Sources, Calibration, and Simulation ............................................. 50Simulations ............................................. 55The Benchmark Case ............................................. 55A Shock to Confidence in the Banking System ............................................. 55Trade Shock ............................................. 57Conclusion ............................................. 60References ............................................. 61Appendix ............................................. 62

Chapter 3. Infrastructure, Private Costs, and Payoffsfrom Additions to Infrastructure

Background ............................................. 68The Model ............................................. 72The Data ............................................. 74Sample Horizon ............................................. 75Limitations ............................................. 75Estimation of the Model ............................................. 76Methods ............................................. 76Infrastructure ............................................. 76Estimnation Limitations ............................................. 77Results ............................................. 78Overview ............................................. 79

iv

Payoffs from Additions to Infrastructure ....................................................................... 84Static Costs and Benefits of Increased Infrastructure ................................................ ....................... 85Optimal Infrastructure Stocks ....................................................................... 86Conclusions ....... ................................................................ 87References ....................................................................... 88Appendices ....................................................................... 90

Chapter 4. Fiscal Impact of Contingent Liabilities: The Case of Mexico

Coverage of the Study ....................................................................... 107Government Accounting Issues ....................................................................... 110Methodology ....................................................................... 1 IlllDeposit Insurance Scheme for Private Banks ....................................................................... 112The 1995 Banking Crisis ....................................................................... 113Expected Fiscal Costs of Resolution of the Crisis ........................................................................ 114Govermment Credit Assistance Programns ........................................................................ 120Characteristics of Direct Loans and Loan Guarantees . ...................................................................... 121Expected Fiscal Cost of Government Credit Programs and the Budget ......................... ............................ 126Liabilities Related to Social Security Programs ....................................................................... 127The Financial Condition of IMSS ....................................................................... 127The Financial Condition of ISSSTE ....................................................................... 128Expected Fiscal Costs of Social Security Programs and the Budget .......................................................... 129Private Provision of Infrastructure and Government Guarantees .............................................. ................. 129Power Plants ....... ................................................................. 129Highways ....................................................................... 130The Fiscal Cost of Government Insurance Programs ....................................................................... 130Policy Inplications ....................................................................... 132References ....... ................................................................ 134

Chapter 5. Fiscal Deficit, Public Debt and Fiscal Sustainability in Mexico

The Mexican Fiscal Accounts: Stylized Facts .......................... .............................................. 138Debt Management ....................................................................... 140Budget Indicators ........................................................................ 142Tax System ........................................................................ 143Government Expenditure Composition ........................................................................ 146Is Mexican Fiscal Policy Sustainable? ........................................................................ 148Accounting Approach to Fiscal Solvency ........................................................................ 148Pricing Approach to Fiscal Solvency ....................................................................... 149Intertemporal Approach to Fiscal Solvency: The Medium and Long Term ................... ........................... 151The Short and Medium Term ........................................................................ 152The Long Term: A Time Series Analysis 1980:01-1999:05 ....................................................................... 154Testing the Intertemporal Budget Constraint: Unit Roots ........................................................................ 155The Case of a Stochastic Discount Rate ........................................................................ 156Testing for a Change in Regime ........................................................................ 157Testing the Intertemporal Budget Constraint: A Co-integration Approach ................... ............................ 160Testing Long-Run Relationship Between Government Spending Inclusive of Interest Payments andRevenue ....................................................................... 161Testing Long-Run Relationship Between Government Spending Exclusive of Interest Payments,Interest Payments and Revenue ....................................................................... 163Policy Conclusions ....................................................................... 165References ....................................................................... 167Appendix ....... 169

v

List of Tables

Table 1.1 Summary Budget Figures, 1980-81 1997-98Table 1.2 Cyclical Properties of Public sector Revenue and ExpenditureTable 1.3 Main Components of Public Sector Revenue and Expenditure, 1980-81 and 1997-98Table 1.4 Estunates of Revenue and Expenditure ElasticitiesTable 1.5 Impulse Response Functions from the VARTable 1.6 Variance Decomposition of OutputTable 1.7 Variance Decomposition of the Unadjusted Prinmary Fiscal SurplusTable l.A1 Estimates of a Piecewise Linear Trend in the Logarithm of Seasonally Adjusted Real GDPTable 2.1 Real GDP, 1980-97Table 2.2 Stocks of Infrastructure, 1970-90Table 2.3 Cost Elasticities by Sector and Infrastructure TypeTable 2.4 A Benchmark Simulation, 1995-2000Table 2.5 Reduction in the Interest Elasticity of Money Demand, 1995-2000Table 2.6 Interest Elasticity Decline Combined with an Infrastructure Increase 1995-200Table 2.7 Trade Shock: Real World Income Stagnates, 1995-2000Table 2.8 Trade Shock Combined with an Infrastructure Increase, 1995-2000Table 2.9 Infrastructure Elasticities = 0Table 3.1 Compound Annual Growth Rates 1960-93Table 3.2 Infrastructure Compound Annual Growth Rates, 1960-93 and 1983-93Table 3.3 Physical Infrastructure, Average Annual Growth RatesTable 3.4 Correlations: Physical and Financial Infrastructure MeasuresTable 3.5 Primary Data: Means and Standard DeviationsTable 3.6 Estimated Private Sector Cost Elasticities with Respect to Public Infrastructure StocksTable 3.7 Estimated Private Sector Cost Elasticities with Respect to Public Infrastructure StocksTable 3.8 Estimated Private Sector Cost Elasticities with Respect to Public Infrastructure StocksTable 3.9 Estimated Private Sector Cost Elasticities with Respect to Public Infrastructure StocksTable 3.10 Mean ElasticitiesTable 3.11 Static Cost and Benefits of Increased InfrastructureTable 3.12 Optimal Infrastructure StocksTable 3.13 OLS Panel ElasticitiesTable 3.14 OLS Random Effects ElasticitiesTable 3.15 FGLS Elasticities - Hetero, ARITable 3.16 Growth Rates Used for Norminal Electricity InfrastructureTable 3.17 Coefficient EstimatesTable 3.18 Coefficient Estimates (se's in (s), Dummy Variables ExcludedTable 4.1 Fiscal Risk MatrixTable 4.2 Federal Insurance Programs with Major Fiscal RisksTable 4.3 State Government DebtTable 4.4 State Government Pension Liabilities: 1997Table 4.5 Pro Forma Balance Sheet of FOBAPROA, February 1998Table 4.6 Estimation of the Cost of the Financial Rescue, June 1999Table 4.7 Executed Fiscal Cost on Programns of Financial and Debtors RescueTable 4.8 Estimates of Total Losses of Resolving Major Bank InsolvensesTable 4.9 Government Loans by Major Program Area, Fiscal 1997Table 4.10 IMSS Retirement System Actuarial Deficit, December of 1994Table 4.11 Government Net Liability as a Result of the 1995 Pension ReformTable 4.12 Pro Forma Balance Sheet of FARAC as of November 1998Table 4.13 Contingent Liabilities and Fiscal Deficit AdjustmentsTable 4.14 Contingent Liabilities Recognized by the Federal GovernmentTable 5.1 Accounting Approach Mexico in 1998Table 5.2 Short and Medium-Term Indicators of Fiscal Sustainability as a percentage of GDPTable 5.3 Testing for Nonstationarity in Undiscounted and Discounted Net Public Debt, 1980:01-

1999:05

vi

Table 5.4 The Zivot Andrews Unit Root Test for Undiscounted and Discounted Public DebtTable 5.5 Testing for Nonstationarity in Real Interest Rates, 1980:1-1998:07Table 5.6 Testing for Nonstationarity in Real Government Spending Inclusive Interest Payments

and Government Revenues, 1980:1-1999:05Table 5.7 Results of Co-integration Government Spending Inclusive Interest Payments and

Goverrnent Revenue, 1980:01-1999:05Table 5.8 Testing for Nonstationarity in Real Government Spending, Interest Payments and Government

Revenues, 1980:1-1999:05Table 5.9 Results of Co-integration Noninterest Government Spending, Interest Payments and Government

Revenue, 1980:01-1999:05

List of Figures

Figure 1.1 Real GDP in Mexico, 1980-98Figure 1.2 Seasonally Adjusted Real GDP, 1980-98Figure 1.3 (a) Trends and Cycles in Real GDP: HP TrendFigure 1.3 (b) Trends and Cycles in Real GDP: Deviations from TrendFigure 1.4 Trends in Public Sector Revenues, 1980-98Figure 1.5 Cyclical Components of Revenue, 1980-98Figure 1.6 Trends in Public Sector Expenditure, 1980-98Figure 1.7 Cyclical Components of Expenditure, 1980-98Figure 1.8 The Budget Surplus and the Fiscal Impulse, 1980-98Figure 1.9 Cyclical Fluctuation in Output Caused by Fiscal ShocksFigure l.AI Trends in Real GDPFigure l.A2 Cyclical components of Real GDPFigure 3.1 Changes in Electric, Transport, and Communications InfrastructureFigure 3.2 Education Infrastructure IndexFigure 4.1 Contingent Liabilities related to Potential Crisis of the Banking SectorFigure 5.1 Mexico Public Net Debt and Prirnary Deficit (+) as a percentage of GDP, 1980 - 1998Figure 5.2 Mexico Overall and Primary Deficit (+) as a percentage of GDP, 1980-1998Figure 5.3 Mexico Domestic and Foreign Public Net Debt as a percentage of GDP, 1980-1998Figure 5.4 Public Sector Domestic Debt as a percentage of GDP, 1982-1998Figure 5.5 Mexico Public Foreign Debt: Termn Structure, 1982-1998Figure 5.6 Mexico Budget Indicators, 1980-1998Figure 5.7 Total Tax Revenues as a percentage of GDP-Selected CountriesFigure 5.8 Oil Revenues as a percentage of Total RevenuesFigure 5.9 Seignorage as a Source of Government RevenueFigure 5.10 Federal Govermment Revenue Tax MixFigure 5.11 Total Expenditure of the Central Government as a percentage of GDP-Selected CountriesFigure 5.12 Total Government Expenditure (million p$1994)Figure 5.13 Expenditure CompositionFigure 5.14 Real Annualized Interest Rate CETES 28 daysFigure 5.15 Brady Bonds DiscountsFigure 5.16 EMBI Spread RateFigure 5.17 Credit Rating for Mexico (100 lowest chance of default)Figure 5.18 Mexican Net Public Debt, 1980:1 - 1999:5. Undiscounted at Market Value (in Bill. P$1994)Figure 5.19 Sequential Zivot-Andrews Unit Root Test for the Mexican Undiscounted and Discounted Public

Net Debt, 1980:1-1999:5Figure 5.20 Govermment Spending Inclusive Interest and Government Revenues, 1980:1-1999:5Figure 5.21 Government Spending, Interest Payments and Government Revenues: 1980:01-1999:05

vii

EXECUTIVE SUMMARY

FISCAL SUSTAINABILITY-MEXICO: A SYNTHESIS

Rationale for the Study

1.1 The stabilization efforts and successes that preceded and have underpinnedMexico's sweeping market-oriented structural reforms since the late 1980s have beenanchored in strong fiscal adjustment. Fiscal deficits were drastically reduced (from 15percent of GDP in 1987 to 1.24 percent in 1998), allowing for tighter monetary policyand lessened inflationary pressures (over the same period, inflation fell from 160 percentp.a. to the current level of 13 percent per anum). This created an environment in whichlong-needed structural reform could proceed. The sustainability of those fiscaladjustments is thus central to the Government's macroeconomic policy and, ultimately, tothe country's development future.

1.2 It has become apparent in recent years, however, that potential imbalances mayremain behind the positive results posted in Mexico's fiscal accounts, imbalances that, ifunattended, could bring into question the permanence of the heralded adjustments.Contingent liabilities in the unfunded, pay-as-you-go social security system (that waspartially reformed recently), unlimited bank deposit insurance schemes, off-balance-sheetfinancing of public investments, public guarantees for private sector investments, stateand municipal borrowing/indebtedness and accelerated depletion in public infrastructure,among other issues, have been the focus of increasing attention by practitioners (andacademics) as part of the "quality" analysis of the Mexican fiscal adjustment.'

1.3 The authorities are well aware of the risks embedded in those imbalances andhave been rightly prudent in the implementation of their fiscal policy. This study seeks tosupport their efforts by providing a body of technical analysis that:

(i) sheds light on underlying fiscal trends by correcting them for various business-cycleeffects;

(ii) builds a simulation model to assess the sensitivity of the fiscal budget to exogenousshocks (for example, changes in oil prices and exchange rates) under several structuralscenarios, with particular attention being paid to the depletion of Mexico's infrastructurestock;

' The idea that posted fiscal accounts may not adequately represent the public sector's pressure on aneconomy's resources, or the "quality" of fiscal adjustments, has been addressed in the literature by, amongothers, Towe ("Are all Summary Indicators of the Stance of Fiscal Policy Misleading?", IMF Staff Papers,vol. 36, No. 4, December 1989), Mackenzie ("The Budgetary Control and Fiscal Impact of GovernmentContingent Liabilities", IMF Staff Papers, vol. 38, No. 1, March 1991), and Easterly ("When is FiscalAdjustment an Illusion?", mimeo, World Bank, April 1998).

Executive Summara

(iii) estimates the direct and indirect potential impact on the fiscal accounts of closingpublic infrastructure gaps and funding contingent liabilities; and

(iv) consolidates the financial accounts of the main public sector institutions to assess thesustainability of their aggregate debt path.

In brief, the report is meant to assist Mexican policymakers in the design of sustainablefiscal policy.

Issues and Focus

1.4 In general, fiscal accounts are subject to many sources of instability, explicit andimplicit, realized and contingent, expected and unexpected. Some are within the controlof policymakers, and others are exogenous to the budget process. This spectrum of fiscalpressures also applies to Mexico. The report will not seek to cover the whole of thatspectrum; instead, it focuses on selected sources of instability that, at the margin (that is,given existing analytical work), appear to be the most critical and urgent from the pointof view of future fiscal policy design. More specifically, we first attempt to disentangleunderlying policy stances from business cycle effects. Then the study will address foursources of fiscal instability: external shocks, infrastructure gaps, contingent liabilities,and consolidated public sector debt accumulation.

Chapter One: Fiscal Policy, Business Cycles, and Growth in Mexico

1.5 The identification of Mexico's fiscal budget's cyclical (and seasonal) propertiesis critical to isolating the components of fiscal policy that have an exogenous impact onoutput-that is, to separating the short- and long-run output effects of fiscal shocks fromthe feedback rules implicit in the way output fluctuations, in turn, affect fiscal outcomes.This information is important for policymakers because it allows them to formulate fiscalpolicies that can smooth, rather than exacerbate, real cycles, as well as to predict thegovernment's financing requirements more accurately.

1.6 This section addresses three main questions:

* what is the role of fiscal policy in deterrnining output in the short andmedium termn;

* how does fiscal policy, in turn, respond to the business cycle;* what is the "persistence" of fiscal policy, and how can the authorities use

this persistence to forecast the government's financing needs?

1.7 The chapter looks at these particular issues for a number of reasons Onetraditional role fiscal policy plays in industrial economies is that of a cyclical stabilizer.Fiscal policy is typically designed to "lean against the wind." That is, it is usuallydesigned to stimulate output when the economy moves into recession and to becontractionary when an expansion broadens. This is usually accomplished in two ways.

2

Executive Summarv

The first way is by having components in the budget that respond automatically to thebusiness cycle, such as tax revenues (which respond positively) or unemploymentbenefits (an expenditure item that responds negatively). The second way is by usingdiscretionary components in the budget to provide a stimulus during bad times. A fiscalpolicy designed in this way leads to a strongly procyclical budget surplus.

1.8 Our findings indicate that Mexico's fiscal policy does not lean against the wind.The analysis in this chapter will show that the budget surplus is quite stronglycountercyclical, so that fiscal policy leans with the wind. The automatic stabilizers inplace are weak, and are further weakened by the tendency of another automaticcomponent of the budget, oil-based revenue, which responds sensitively to exogenousworld oil prices, to move countercyclically. Furthermore, the discretionary component ofthe budget surplus also tends to move countercyclically.

1.9 If fiscal policy simply did not matter, then whether or not it leaned with oragainst the wind would be of little consequence. However, in Mexico, as in many othercountries, fiscal policy does matter. The analysis suggests that an increase in thediscretionary surplus of 1 percent of gross domestic product (GDP) causes GDP todecline by 0.6 percent in less than a year. Because in Mexico such increases typicallyoccur during contractions, and these contractions are relatively short-lived (typically lessthan two years), this implies that discretionary policy exacerbates the cycle.

1.10 The results also imply that Mexico's fiscal policy lacks a design that makes it astabilizing feature of the economy. Furthermore, it has not been designed to render itselfmore sustainable. With procyclical fiscal policy (a countercyclical fiscal surplus),deficits cause debt to accumulate during economic expansions, but when the economicexpansion inevitably ends, this debt suddenly becomes extremely costly. To finance it,the government must either take drastic discretionary fiscal measures, or it must financethe debt by borrowing at high real interest rates, or by printing money and inducing rapidinflation. No matter which action the government takes, the implications are similar: aworsening of the economic downturn. This chapter is intended to allow policymakers toformulate fiscal policies that can smooth, rather than exacerbate, real cycles, and that aretherefore more readily sustained in the medium term.

Chapter Two: Infrastructure, External Shocks, and Mexico's Fiscal Accounts

1.11 In the past 25 years Mexico's economy has been subjected to a variety of shocks,both internal and external, including sudden increases and declines in world oil prices,changes in U.S. interest rates, economic collapses in Russia and Asia, and bank panic inMexico. The aim of this chapter is to determine whether changing the provision ofcertain types of infrastructure can mitigate the effects of such shocks. If this is indeed thecase, then choosing alternative infrastructure paths could help stabilize the Mexicaneconomy.

1.12 This chapter investigates the first- and second-order impacts of major exogenousshocks (such as changes in the international price of oil or in the nominal exchange rate,

3

Executive Summarv

or confidence-driven contractions in money demand) on Mexico's fiscal accounts. Howis that impact affected by the choice of the public infrastructure investment path? Thesetwo questions, which are at the forefront of fiscal policy design in Mexico, areinvestigated in this chapter.

1.13 A dynamic general equilibrium model is presented to analyze issues of stabilityin the Mexican economy. It focuses on whether increased provision of infrastructure canreduce the impact of exogenous shocks on the real economy. That is, would the impactof a shock be less if higher levels of infrastructure spending were in place at the time ofthe shock?

1.14 The answer to this question is a qualified yes. All else being equal, higher stocksof infrastructure will tend to reduce the declines in real income caused by certain types ofshocks. We reach this conclusion by carrying out a series of numerical exercises basedon a model that incorporates various types of estimated Mexican data. The modelincorporates four types of infrastructure in the production process: electricity,telecommunications, transportation, and education. In general, the estimates indicate thatin Mexico, increased provision of infrastructure, either by the public or private sector,tends to be cost reducing.

1.15 Using a variety of Mexican data sources, we calibrate our model to the years1995-97 as part of a six year simulation for the years 1995-2000. We subject the modelto two types of shocks. The first is a shock to the interest elasticity of money demandthat causes the absolute value of the elasticity to decline. Such a shock might be causedby a sudden loss of confidence in the banking system, and tends to increase holdings ofmoney and reduce bank deposits. Over time, this shock brings about an increase in thereal interest rate, a deflation, and a reduction in real gross domestic product (GDP)amounting an annual average of about half a percentage points over the six years of thesimulation.

1.16 We then suppose that prior to the shock infrastructure spending was higher, inreal terms, on each of the four types of infrastructure. 7he increased provision ofinfrastructure reduces private sector costs and, as a result, real GDP rises to a somewhathigher level than in the initial pre-shock simulation. We thus conclude that higher levelsof infrastructure stocks can indeed insulate the Mexican economy from certain types ofshocks.

1.17 The second shock is an external shock: stagnation of the world's real income forthe six years of the simulation. This lowers the demand for Mexican exports and,accordingly, the rate of growth of real Mexican income. In this case, as before, a higherlevel of expenditure on infrastructure before the shock and throughout the period of thesimulation tends to neutralize the impact of the shock on real Mexican income. Wetherefore conclude that the positive implications of increased infrastructure outweigh thenegative impact on the budget deficit. Enhanced infrastructure prior to shocks seems tooffer a way to avoid the ex post remedies that have been tried so often, frequently withlittle success.

4

Executive Summary

Chapter Three: Infrastructure, Private Costs, and Payoffs from Additions toInfrastructure

1.18 The role public spending plays in enhancing economic productivity has longbeen a concern for policymakers. In recent decades, public expenditure has primarilybeen evaluated in terms of two roles: enhancing macroeconomic stability and mitigatingmarket failure. An equally important role concerns the ability of public investments ininfrastructure capital to reduce the costs of private firms. Particularly for developingeconomies, this role may be critical because it may allow the private sector to becomemore resilient to external shocks. A major concern related to the recent fiscal adjustmentin Mexico is that it was carried out partly by depleting public infrastructure stocks. Thisdepletion could significantly retard future growth by imposing an additional drag onprivate sector costs and output.

1.19 This chapter provides a macroeconomic estimate of Mexico's publicinfrastructure shortage, or "gap," as well as the fiscal cost of closing that gap. It firstexamines general evidence at the macro level to establish the actual relationship betweencurrent rates of growth and the aggregate stock of public infrastructure. This will besupplemented by an examination of sectoral evidence in education, transport, electricityand telecommunications, considering historical patterns of public spending oninfrastructure across sectors.

1.20 The chapter also provides estimates of the potential (partial equilibrium) payoffsfrom increased investment in public infrastructure and calculates (in a static context) theoptimal infrastructure stocks implied by the elasticity estimates. The chapter alsoconsiders the role that public infrastructure plays in improving the efficiency of theprivate sector. In particular, it focuses on the short-run, static gains that accrue to privatefirms because of government investments in electric, transportation, and communicationsinfrastructure.

1.21 Our findings suggest that public infrastructure in Mexico has generally small,but significant, negative effects on private sector costs. The base case estimates of theelasticity of private sector costs with respect to infrastructure suggest a mean value of -0.106 across 14 sectors of the economy (with a range of -0.563 to 0.355). In general,electric infrastructure appears to have the most beneficial effects on private sector costs(mean base case elasticity of -0.1 71), and transportation infrastructure has the next mostbeneficial effects (mean base case elasticity of -0. 165).

1.22 Using the electricity, transportation, and communications estimates, roughcalculations based on these elasticities suggest that a 1 percent increase in publicinfrastructure stocks would cost approximately mxp 6.6 billion and provide annualbenefits across 14 sectors of the economy of mxp 12.4 billion (both in terms of real 1980pesos). These are static, partial equilibrium estimates, but they suggest that at least in theshort run, additional investment in public infrastructure stocks could be welfareimproving. Given sensible depreciation rates, the present value of these benefits overfuture years might be roughly nine times the single year gross benefits. If the base caseelasticity estimates are correct, static calculations of the optimal size of infrastructure

5

Executive Summarg

stocks suggest that electric and transportation stocks should have been 2.5 and 4 times aslarge as they actually were in 1993.

Chapter Four: Fiscal Impact of Contingent Liabilities

1.23 One result of the financial crises affecting Asia, Latin America and otheremerging economies over the recent past has been a renewed emphasis on the fiscal risks(liabilities) governments' face and means to quantify and mollify such risks. In general,liabilities can be either contingent or direct. Contingent liabilities are defined as"obligations that may or may not come due, depending on whether particular eventsoccur. The probability of their occurrence may be exogenous to government policies (forexample, if they are related to natural disasters) or endogenous (for example, ifgovernment programs create moral hazard)." In contrast, direct liabilities are defined as''obligations whose outcome is predictable."

1.24 Delineating further, "Explicit liabilities are specific obligations, created by law orcontract, that governments must settle. Implicit liabilities represent moral obligations orburdens that, although not legally binding, are likely to be borne by governments becauseof public expectations or political pressures."2

1.25 This chapter addresses the measurement of contingent liabilities for Mexicowithin the government's traditional budget accounting framework. The key researchproblem is to identify, quantify, and understand the future fiscal risks posed by thegovernment's contingent liabilities. The lack of a unified measure of contingentliabilities makes difficult the assessment of the sustainability of the fiscal policy.Therefore, the objective of this chapter is to provide an overview of the problem inMexico and suggest an analytical construct for assessing the magnitude of the problem.

1.26 For many years, federal government insurance programs were a policyinstrument for the Mexican authorities. Those programs were primarily concentrated infour areas: government-guaranteed borrowing; infrastructure franchising; unlimited bankdeposit insurance; and, more recently, coinsurance of the social security system. Thoseprograms were supposedly justified on the basis of observed market failures (includinginformation-based imperfections in credit markets, high risks in the provision ofinfrastructure, systemic risk in the banking system in the case of bank failures, imperfectpooling arrangements in the private insurance sector with respect to certain types ofinsurance, and the like). In effect, before the crisis of 1994, the budgetary implications ofthese contingent liabilities were largely overlooked.

1.27 Since the crisis, however, it has become clear that the financing requirements tocover the realized losses in those insurance programs have the power to destabilize theGovernment's overall fiscal adjustment efforts. More critically, while the latent fiscal

2 Polackova, Hana, "Contingent Government Liabilities: A Hidden Fiscal Risk"; Finance & Development,March 1999.

6

Executive Summarv

cost of the various insurance programs is widely believed to be large, there is no solidestimate available of the eventual cash outlays that the budget will have to afford, mainlybecause those programs were not (and, to a certain extent, are still not) considered in thetraditional budgetary accounts.

1.28 Two important contingent liabilities are omitted in the analysis due to datalimitations: first, the expected effects on the federal fiscal accounts of the debtrestructuring of states and municipalities since the 1995 crisis; and second, the analysis ofthe pensions of sub-national governmental institutions. Assessing these areas is difficultbecause of incomplete and inconsistent data, however these are clearly contingentliabilities at the federal level. The chapter provides estimates of the debt stock at the statelevel only (very limited data exists at the municipal level), where disturbing trends inboth the size and the concentration of the debt at the state level are developing: the debtof the Federal District, State of Mexico, Nuevo Le6n has grown from 33 percent in 1994to 65 percent of the total outstanding in 1998, more than doubling the outstanding sub-national debt stock over the period (see Figure E.1 below). The chapter also providessobering evidence on the health of the 32 sub-national pension systems (31 states plusthe Federal District): 11 of these are either in actuarial deficit now or will be by 2001.

7

Executive Summary

Figure E.1 Concentration and Growth of Subnational Debt, 1994-1998: Selected States

Total Subnational debt more than doubledover the period 1994-1998; some stateswitnessed a ten-fold increase...

Estado de Mexdco

25000 - _ _._ __

.00 and concentration has increased dramatically.20000-

0m 7_Q 15000 -0

10000 1994

50000

E 5000 <_ _ _ _ _ Mexico0 -- -- 20%

1994 1995 1996 1997 1998

44 u Nuevo Leon10%

AU othersFederal District 63% Fed. Distrct

7%

25000

rs 20000-xE

15000

.10000

5000-

0 T 1998

1994 1995 1996 1997 1998

Nuevo Leon Nbxico

All others 26%25000- 35%

20000 -. o LeonCx 10%

E15000 _

10000 _ Fed. Cistrictc10000 29%

5000

1994 1995 1996 1997 1998

8

Executive Summarv

1.29 Table E. 1 presents the most recognized (and highly politicized) contingentliability facing the government, the banking system rescue, with resolution costsestimated to be on the order of 19.3 per cent of GDP. Of this 19.3 percent, 3.1 percenthas already been spent, leaving a government-acknowledged remaining liability of 16.2percent of GDP.

Table E.1 Estimate of the Overall Cost of the Financial Rescue, June 1999

Billion of pesos % of GDPDebtors Aid Programs 174.3 3.9Cost of banking intervention 579 12.8and clean upPurchase of non performing 101.8 2.2assetsToll Roads Program 18 0.4Total 873.1 19.3Source: IPAB

1.30 Since 1998 the Ministry of Finance has made an effort to report what theyrecognize as the contingent liabilities of the government. Table E.2 below presents theirmost recent report, reflecting the increased cost through time.

Table E.2 Contingent Liabilities Recognized by the Federal Government(billions of pesos)

Balance Balance Change Balance ChangeDec, 1998 Mar, 1999 from Jun, 1999 from

Dec 1998 Dec 1998FOBAPROAa 425.03 435.17 10.14 451.79 26.76

FARAC 73.63 79.09 5.46 83.32 9.69

Credit 143.57 153.51 9.94 160.19 16.62AssistancePrograms b

Development 10.08 11.34 1.26 12.02 1.94BanksFAMEVAL 4.07 4.33 0.26 4.48 0.41Others c 5.18 4.72 -0.46 4.66 -0.52

Total d 661.56 688.16 26.60 716.46 54.90a/ Includes only the explicitly guaranteed liabilities by the Federal Government.b/ Includes mainly FIRA, FOVI, FIDEC and FIDELIQ.c/ Includes mainly the Federal Electricity Company, CFE.d/Preliminary data. It excludes guarantees established in the organic laws of the Development Bank.Source: SHCP

9

Executive Summary

1.31 Public estimation of these liabilities by the government is an important steptoward resolution of the problem. However, it must be pointed out that these estimatesexclude all sub-national debt, sub-national pensions, some unrecognized FOBAPROAdebt, and obligations of state-owned enterprises. Our findings suggest the figures abovesubstantially understate the true contingent liabilities facing the government.

Chapter Five: Fiscal Deficit, Public Debt and Fiscal Sustainability in Mexico

1.32 While the numbers shown above are somewhat startling (and the quarter toquarter run-up is quite expensive), it is important to view the problem from anintemational perspective. Even with the addition of the govemment's estimate of thecontingent liabilities to the existing explicit stock of debt, Mexico compares favorably toG-7 nations on a debt to GDP basis. Table E.3 illustrates that even when the contingentliabilities are added to the explicit stock of debt, Mexico would easily qualify forEuropean Union membership, whereas Figure E.2 illustrates that of the G-7 nations, onlyFrance would qualify.3

Table E.3 Mexico Federal Debt as a Percentage of GDP

Gross Public Debt 29.2Cost of Financial Rescue 16.2Other Contingent Liabilities' 5.9Total 51.3

Source: SHCP, Informe sobre la Situaccion Economia, lasFinanzas Publicos y la Deuda Publica, SegundaTrimestre 1999.

Cost of Financial Rescue: IPAB estimates1/ Includes credit assistance programs, FARAC, Development

Banks, FAMEVAL, and CFE.

3 The fiscal criteria under the Maastricht Treaty are set in terms of a general government debt/GDP ratioceiling of 60 percent and a govermnent financial deficit/GDP ratio ceiling of 3 percent.

10

Executive Summary

Figure E.2 Gross Federal Debt as Percent of GDP: International Benchmarks

140118.7 117.9

120

100 95.8

8 8040 ~~~58.2 61.1 62.2 62.160 51.3

4 0 29.2

20

0

CD w~~~~~~~~

Source: October 1999 IMF World Econornic Outlook for G7 Countries, Bank staff estimate for MexicoNote: 1998 estimates for G7 countries, June 1999 for Mexico1/ Includes direct and contingent liabilities.2/ Includes only direct liabilities.

11

Executive Summary

With this information, coupled with one measure of the international investmentcommunity's current diagnosis of Mexico's fiscal health, eurobond spreads (shownbelow in figure E.3), one could easily conclude the need for such stringent analysis isdubious.

Figure E.3 Selected Latin American Eurobond Spreads (basis points)

800 Argentina 900 Brazil700 Soo600 600

0oo 600400 500300 300200 200

IO ' O

' 0 0 i i 1 0 0 o

800 ~Colombia 7,0O0 Ecuador

700 Meic6,2000Vnzul

600 5,000500 4,000400300 3,000200 2,000100 1,000

0 111 1 111 1 ii 0 li 111 11

350 Mexico 1,200 Venezuela300 1,000250 S00

200600

t00 Nt calel 40050 2000

domestic poliis 0 ut, O he can also be exlane by50 MexicssONso's vulonerabiliOty to change

1.33 However, problems remain. Since the early 198Os, Mexico has faced recurrentcrises at approximately six-year intervals. These crises can be attributed largely to poordomestic policies. But, they can also be explained by Mexico's vulnerability to changesin external variables. Examples include the fall in oil prices in 1982 and 1986 and therise in foreign interest rates and sharp decline in foreign capital flows in 1994. Thecontinued dependence on oil as a source of revenue is particularly worrisome.

1.34 The unpredictability of fiscal revenues as a consequence of its dependence on oilhas forced the government to make drastic expenditure cuts. As a fraction of GDP, non-interest expenditures fell from about 25 percent in the 1980s to 19 percent in the 1990s, adrop of 23 percent. By contrast, non-oil revenue remained roughly constant at about16.25 percent of GDP over this period.

12

Executive Summary

Figure E.4 below illustrates the expenditure and revenue cuts over the period 1980 to1998.

Figure E.4 Mexico Budget Indicators: 1980 - 1998

45

40

35

30

25

20

1980 1981 1982 1983 19M 1985 1986 1987 1998 1988 1990 1981 1992 1993 1994 1933 199 1997 1898

|_ TOTAL REVENUE OIL REVENJE TOTAL EXPENDITURE -U-NON INTEREST EXPENDITURES

Source: SHCP

The Mexican tax system has undergone major reforms since 1980. A value added tax(VAT) and indexation to neutralize the effects of inflation were introduced. Personal andcorporate income taxes were integrated, ensuring more neutrality between retained anddistributed profits. Despite these measures, tax revenues as a percent of GDP have notincreased substantially. As figure E.5 illustrates, by regional and international standards,Mexico does not compare favorably. Clearly, fiscal reform is long overdue.

Figure E.5 Tax Revenue as Percent of GDP, Selected Countries 1992 - 1997

45

40

35

30

25

20

15

10

5

0 ..1992 1993 1994 1995 1996 1997

2 Hungary SI Chile * Mexico 0 Uruguay E] Venezuela

Source: WDI-World Bank

13

Executive Summary

Thus, while current conditions appear tenable, further investigation is certainlywarranted. Summarizing the results of chapter five:

* Budgets are affected by oil prices, as total revenues follow the cycle of oil revenuesquite closely. In response to this vulnerability to oil prices and the effect of interestrates, the stance of fiscal policy has remained cautious. The latter has been achievedby running consecutive primary (interest exclusive) surpluses since 1983.

* This primary surplus has been obtained primarily by drastic expenditure cuts. As afraction of GDP, non-interest expenditures fell from about 25 percent in the 1980s to19 percent in the 1990s, a drop of 23 percent. By contrast, non-oil revenue remainedroughly constant at about 16.25 percent of GDP over this period. In particular, themajor adjustment in the 1990s came from capital expenditure.

* Mexico's public debt, relative to GDP, fell from 115 percent in 1986-87 to about 29percent in 1998, due to primary surpluses and a debt management strategy centeredon lengthening maturity and reducing interest payments after the peso crisis.

- The short and medium term projections of fiscal sustainability using an intertemporalapproach for the Mexican economy between 1999-2006 show that the requiredadjustment in the primary surplus will be around 0.5 and 0.8 percent of GDP. Thegovernment can increase government spending or reduce taxes. This under theassumption that government expenditures to GDP remained constant at the 1998 leveland the bailout of FOBAPROA is not taken into account.

- However, results change if FOBAPROA bailout is included. Assuming a discountrate of 5 percent, the government will have to increase taxes or reduce governmentspending in the next years approximately between 0.3 and 0.8 percent of GDP. If thediscount rate is 3 percent this adjustment will be only necessary in 2005 and 2006 foran amount of 0.1 percent of GDP.

* The numbers obtained from this simple exercise do not seem implausible and onlybarely effect the fiscal solvency of the Mexican government. As a cautionary note,this exercise does not include all contingent liabilities.

* The Mexican government has responded to the increase in interest payments on theoutstanding debt by running primary surpluses. Thus, the change in fiscal policy inrecent decades is a signal that the government is trying to meet its intertemporalbudget constraint through fiscal adjustment instead of inflation or default.

* The Mexican government has run a primary surplus since 1983. After each majoreconomic crisis the government has tightened its fiscal policy as part of thestabilization program. In the two years following the debt crisis of 1982, the primarysurplus averaged 4.9 percent of GDP. After the 1986 crisis, the primary surplus rosefurther to 6.5 percent of GDP. Following the December 1994 crisis, tighter fiscalpolicy yielded primary surpluses averaging 4.5 percent of GDP in 1995 - 96. Sincethen, the stance of fiscal policy has been cautious, to consolidate macroeconomicstability.

14

Executive Summarv

* By contrast, Mexico has run an overall budget surplus (including interest payments)only 3 times in the last 19 years. The difference between the primary and overall-deficit is the interest payments on the government domestic and foreign debt (as apercentage of GDP). Thus during the 1980s, the overall deficit was largely explainedby interest payments.

15

Executive Summary

An Extension: Balance Sheet Approach and Quality of Fiscal Adjustments

1.35 Admittedly in many countries there is a need to cut fiscal deficits and spending,yet there are many routes to this end. By providing the authorities a comprehensivereview of their fiscal sustainability, better decisions can be made in the process of fiscaladjustment. In this study we present the conventional approach to fiscal sustainability,emphasizing only the right-hand side of the balance sheet.

1.36 Reasons for this rest in the nature of the available data-it exists for liabilities.Very little information on the assets of the nation are available. Buiter (1983) and Beanand Buiter (1987) suggest that the ideal approach would require evaluating a consolidatedpublic sector balance sheet valued at current prices. In this way fiscal policies could beevaluated in termns of the impact of the government's net worth. Easterly (1999) outlinesnumerous examples of costly fiscal adjustments where the left hand side of the balancesheet is sacrificed in the short-run with deleterious long-term consequences.

1.37 Under this approach, a govermment balance sheet would include financialassets, real capital, land and mineral assets, the present value of fature tax programs andseigniorage on the asset side, with government debt, the stock of high-powered money,and the present value of social security and other entitlement programs, on the liabilityside (Blejer and Cheasty, 1991). While the balance sheet approach to fiscal accounts istheoretically straightforward in giving an idea of a government net worth, it presentsmany empirical problems. In particular, when measuring assets intertemporally, strongassumptions on interest rates, inflation, demographics and natural resource pricing haveto be made.

1.38 Because of these empirical problems, some alternative approaches haveappeared following the theoretical line of government net worth, but employing simplerempirical methods. Easterly (1999) follows this strategy investigating policy aspects offiscal adjustments mandated by the IMF and The World Bank. The author indicates fourlines of action that governments may implement in order to avoid a real fiscal adjustment.These are reduction of public investment, privatization, shifts of revenue and expenditureover time and run up of implicit liabilities.

1.39 Following Easterly's work, we analyze budget deficit reductions in selectedLatin American countries from 1980 to 1997. Specifically, the achievement of fiscaladjustments by means of reductions in public investment and privatization areinvestigated.

1.40 A first glance at a plot of budget deficit versus public investment4 trends in 11Latin American economies (Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica,

4 Where Gross Domestic Fixed Investment of the Public Sector is used as a measure of public investmentand Primary Budget Deficit as a measure of Budget Deficit. The reason for choosing the latter is thatinterest payments play an important role in the composition of budget deficit, but since they are not fullypredicted in advance they may give us a misguided idea of the government efforts in reducing its deficit. Inthe Public Sector we include central government, related institutions and state-owned enterprises, wheneverthe data is available.

16

Executive Summarv

Dominican Republic, Mexico, Panama, Peru and Venezuela) reveals a close relationshipbetween these series (see figure E.6).s The first and largest deficit-investment reductioncomes in 1982 and goes until 1985; this reduction was generated by the 1982 oil crisis.During the middle eighties two other reductions in primary deficit took place (one in1987 and the other in 1990), the latter was again accompanied by reductions in publicinvestment. The correlation coefficient between the two series is 0.78.

Figure E.6 Primary Deficit vs Public Investment (percent of GDP)

91

Selected Counties:8 / Argentina, Brazil, Chile, Mexico, O

Ck / \ \ Venezueta, Eoliia0 7 SX Colombia, Costa Rca, Dorninican

Z1.1 ~~~~ ~ ~~~~~~~Rep.ZPanama & Peru

6 -2

4 -~~~~~~~~~~~~~~~~~~~~4

3 -5c=> c O 00 0 4 t O00 00 00 00 00 (ON ON~ ON ON

- - - Public Investment Primary Deficit

Source: IMF-Government Financial Statistics, LDB- World Bank, SHCP for Mexico.

1.41 Because figure E.6 averages very heterogeneous economies, we split the sampleof countries according to income level. Argentina, Brazil, Chile, Mexico and Venezuelaform the upper-middle income (UMI) group; while the lower-middle income (LMI)group is formed by the rest of the countries in the original sample. Figures E.7 and E.8show the trends in investment and deficit for the sub-samples. As we can see for the twogroups, the series follow each other closely, but a remarkable similarity exists for theLMI countries. While the 1982 fall is stronger for UMI economies, (because oilproduction is essential for countries like Mexico and Venezuela) after that shock thetrends under study seem to follow paths that are more independent. Correlationcoefficients are 0.53 versus. 0.70 for the UMI and LMI countries respectively. There aremany reasons for this divergence in correlation; in particular largest economies in theUMI sample have a more complex economic system that allows them to reduce theirdeficit with different policies like higher taxes, revenues from privatization, etc.

5 Figures are the result of a simple average across countries. No weighting factor was used because all thedata is expressed on a percent of GDP basis.

17

Executive Summarv

Figure E.7 Primary Deficit vs. Public Investment (percent of GDP) 5 UMI Countries

11 0Selected Countries:

10 Argentina, Brazil, Chile, -19 ~~~~~~~Mexico, Venezuela, -2

Wn 7 \ A -4 E

5 ~~~~~~~~~-62S4 ~~~~~~~~~~~~~~~~-7

3 -8

2 -9O ES T O 00 0 N00 00 00 00 00 ON O r ON ON

ON O ON (ON ON~ ON~ ON ON O

- - - Public Investment -Primary Deficit

Source: IMF-Governrment Financial Statistics, LDB-World Bank, SHCP for Mexico.

Figure E.8 Primary Deficit vs. Public Investment (percent of GDP) 6 LMI Countries

8 2

6 o'

4 -2 -3Selected Countries:

9w3Bolivia, Colombia, Costa -3Rica, Dominican Rep.

2 Panama & Peru 4

o es r O 00 0 Cel00 00 00 00 00 ON ON ON ON

ON O ON1 ON ON ON ON ON ONl

- - - Public Investment Primary Deficit

Source: IMF, Government Financial Statistics, & LDB, World Bank.

1.42 Figure E.9 plots the budget deficit trend versus privatization revenues for bothgroups of countries. Except for Chile, which started its structural reforms during the endof the seventies under Pinochet's dictatorship, the wave of privatization in Latin Americacame at the beginning of the nineties. For the UMI countries there is a slight negativerelationship between the two trends (a contemporaneous correlation of -0.04, and a 1-lagcorrelation of -0.29), this would imply that privatization revenues have played a smallrole on budget deficit reductions. Nevertheless, this statement must be taken withprudence because we are only accounting for direct revenues of privatization and notlooking at the expenditure reductions that come from discontinued support for the

18

Executive Summarv

privatized enterprises. The same plot for LMI countries shows also a negative correlationbetween the series (-0.12).

Figure E.9 Primary Deficit vs. Privatization Revenues (percent of GDP)

8w Selected Countries: Selected Countries:n 7 Argentina, Brazil, -2 : 3.5 Bolivia, Colombia, Costa A -0.5

Chile,Mexico, Pfi Rica, Panama&Pesu -1 C

4155 -4 2 ~~~~5-

a4 XC .,2I X

-s~~~~~~~~~~~ ~~~2.5~3 ~~ ~~~-6' 2 1.53 2 -7 A -3.5~~~2

-8~~~~. -4

0:6 / 06O -4.5o -es t m 'Io , O r -r tO8: N ON ; s CN 0y oll>0 8 ECo0_~ a_, .1 _N ON _% _~ ON_ NO NC -

--- Privatization Revenues Primary Deficit --- Privaization Revenues Pimary Deficit

Source: IMF-Government Financial Statistics, LDB-World Bank, SHCP for Mexico.Note: Dominican Republic excluded due to lack of data on privatization revenues.

The Mexican Case

1.43 From the fifties to the beginning of the seventies the Mexican economypresented a remarkable performance in growth with low inflation. These golden yearsare known as the "Desarrollo Estabilizador" (Stabilizer Development), during this periodMexico went through a rapid process of development and industrialization.

1.44 By the first half of the seventies the Mexican economy started to reveal signs ofexhaustion. Growth slowed and inflation picked up due to loose fiscal policies. Duringthis period the State played a central role in the Mexican economy, producing many"private" goods and protecting domestic producers from trade. Expenditure expansionsat the beginning of 1970 were financed mainly by external debt. In 1976 when severaloil reserves where discovered and oil prices were still high, optimistic Mexican policymakers accelerated the process of indebtedness to finance fiscal and current accountbalance deficits. In 1981 declining oil prices resulted in capital flight and the governmentresponse was to finance the exit with higher external debt. By 1982 the situation wasunsustainable and a severe crisis ensued.

1.45 In the years following the crisis, the government implemented a radical fiscaladjustment process, a central condition to receive IMF loans and restructure debt withforeign commercial banks. In order to evaluate to what extent these policies (and theones that followed during the 1980s and 1990s) were effective in increasing governmentnet worth we extend the previous analysis to Mexico. Again our main concern is to see ifdeficit reductions were "real" in terms of augmenting government saving or whether theysimply hide movements in government balances, leaving net worth unchanged.

19

Executive Summary

1.46 Figure E. 10 presents the Mexican budget deficit/public investment relationship.In the beginning of the decade both variables were strongly related, suggesting that the-deficit trend was guided by shifts on public investment. While public investmentcontinued to fall steadily by around 2 percent of GDP, primary deficit reduction from1986 to 1989 doubled that amount and in 1992 deficit started to grow again.Contemporaneous correlation between the series is 0.82, suggesting public deficits werecut via investment reductions. This notion is confirmed by looking at figure E. 11,revealing a stable path of general government consumption around 9 percent of GDP.The remarkable similarity between the two trends from 1989 forward suggests that recentprimary budget deficits were incurred in order to finance government consumption.

1.47 Figure E. 12 relates oil prices to public investment for Mexico. This figureshows that Mexican public investment is strongly related to oil prices. Part of the declinein investment may be explained by the privatization process, taking place from 1988 tothe present, with PEMEX (the Mexican oil state-owned company) remaining as one ofthe few state-owned enterprises.

1.48 Decomposition of Public Programmable Expenditure6 into its components(figure E. 13) shows that a large amount of the reductions in expenditure were due to cutsin fixed investment. From 1981 to 1989 programmable expenditure fell by 11.6 percentof GDP, while the fixed investment component fell by 8.12 per cent of GDP. At thesame time the current expenditure component followed a more or less stable pattern.

1.49 At first glance privatization revenues do not appear to play an important role indeficit reduction over the last decade (see figure E. 14). However, the contemporaneouscorrelation between budget deficit and these revenues is of -0.61. Here it is necessary toclarify the objective of privatization. Even when budget deficit reductions andprivatization policies are desirable, the latter should be undertaken based on efficiencyarguments solely and not in search of fiscal reductions. As Easterly (1999) points out,"(privatization) may allow efficiency gains, but something is amiss when governmentsdevelop a sudden interest in privatization during fiscal austerity."

6 Programmable expenditure is total expenditure minus interest payments, transfers to local goverrnentsand payments for expenditures made in previous fiscal years.

20

Executive Summary

MEXICO

Figure E.10 Primary Deficit vs. Public Investment (percent of GDP)

18 8

16 6

~14 4 P

12 2~~~~~~~10~~~~~~~~~~

6 64

2 -8

o - e. en I* t(N t- 00 ON 0 - ( C o N ^ 0 O oo0000 00000 00 000000 00 C ON~ ~0 ON l 7 oNNCN oE oll Ch oi oo cl o\ Co CN C7 C) CN 0o cl CN ON

-- ~-- - -- - - -- - - --

- - - Public Investiint - Budget Deficit

Source: SHCP and LDB-World Bank.

Figure E.11 Primary Deficit vs. General Government Consumption (percent of GDP)

18 8

0- 16 6

14 4

>12 20 2

8 ~~~~~~~~~~-2

6 -~~~~~~~~~~~~~4

4 ~~~~~~~~~~~~~~~~-62 ~~~~~~~~~~~~~~~~-8

O -10o e0 00 0 oo o C00 00 00 00 00 ON ON ON ONON (ON ON ONl ON ON ON ON ON

- -- Central Govm Consumption Budget Deficit

Source: SHCP and LDB-World Bank.

21

Executive Summary

Figure E.12 Public Investment vs. Oil Prices

35 14

' ~ *~ oil prices/\~~~~~~~~~~~~~~~~~~1

25

30 0

20 rv

public investmn 15

to 2

5 0o - N ~ ~ % '0 0~ 000% 0% - N a, V, N

Source: LDB-World Bank and INEGI.

Figure E-13 Programmable Expenditure Decomposition (percent of GDP)

301

20

15

I10

10 ., . % 0 ,:.,' ,'',,'., ','' '., """" ""- """ '- 00"" ':"l T . . cqtzena Ex o d ~~...... ..

.l t i1:' W-g g0 00 00 00 00 00 0000 0 00 c% 00 a, " 0I 0a ,c

Source: SHCP.

22

Executive Summary

Figure E.14 Primary Deficit versus Privatization Revenues (percent of GDP)

9 0

8

7 ~~~~~~~~~~~~~~~~~-2

6 ~~~~~~~~~~~~~~~~~~~-3 .

5 ~~~~~~~~~~~~~-4

4 ~~~~~~~~~~~~~~~~~-5

3 ~~~~~~~~~-62

.~2 -7

o -900 O 0 -_ e t- 0 r00 00 O\ 0r Oa O o. 05 0, a\ON O. 0. ON O\ ON 0. CN 0. al

- - - Privatization Revenues Primary Deficit

Source: SHCP & LDB, World Bank.

Implications of the Balance Sheet Approach

1.50 In this section, we presented evidence on fiscal adjustment processes for severalLatin American countries during the past 17 years. The main finding is that there is aclear reduction in public investment when going through periods of fiscal adjustment,particularly for LMI countries. Cutting current expenditures is often a perilous politicalproposition and governments did not rely heavily on privatization until 1993; hence, theonly path to their fiscal goals was cutting investment. The evidence remains strong evenwhen some countries in the sample, as in Costa Rica, based a significant amount of their

7deficit reductions on cuts on operations and maintenance spending. The consequencesof such policies on economic performance and development may be dramatic in the longrun.

1.51 On the other hand, we have the upper-middle income Latin Americaneconomies where the process of fiscal adjustment was more complex. There seems to besome relationship between public investment cuts and deficit reductions, but not as directas for the other group of countries. In these economies several factors come into play andmay hide the extent to which governments are relying on public investment cuts andprivatization to avoid real cuts in their current expenditure. For example in the Braziliancase, the large deficit reductions in 1987 and 1990 were due to the structural adjustmentprogram "Plano Cruzado," increasing government revenues by means of higher taxationand controlling inflation with income policies, thereby evading radical expenditurereduction. In the Chilean case, the government based fiscal adjustment post-1982 on

7 See Easterly (1999) p. 6 0 .

23

Executive Summary

current expenditure reductions combined with other macroeconomic policies, providing astimulus to the economy and increasing the government revenue.8

1.52 The most significant cases of deficit reduction by means of public investment-cuts in the UMI group are Mexico and Venezuela. One explanation of this finding is thatboth economies depend heavily on oil production.9 A significant amount of publicinvestment was and still is oil-related. Thus, when a crisis in oil prices occurs, the naturalcandidate to pay for the adjustment is oil investment expenditures.

1.53 Two final remarks are in order: first, the methodology used in this review isquite simple. Plots of figures and correlation coefficients between series are a firstapproach to the subject. The problem is that the information used is generally availableon an annual basis. If we add to these problems the lack of data on different periods fordifferent countries, we find ourselves confronting the impossibility of making a reliablepool regression analysis. Therefore, this exercise must be taken as a first approach whosestrength must be reinforced by a closer look at each country's recent economic history.

1.54 Second, we must keep in mind that a public investment reduction per se is notnecessarily a bad idea. If governments are cutting inefficient investment expenditure thismay be a good step towards economic efficiency and growth. Additionally, there isevidence that in Latin America public investment crowded-out private investment[Easterly, Rodriguez and Schmidt-Hebbel, (1994)1. However, one concern stemmingfrom our analysis is that there are strong reasons to believe that in some cases LatinAmerican governments cut more than simply the inefficient investments. This could slowor even stop the growth process of the underdeveloped countries of the region, hence theneed for persistent investment and expenditure reviews, thereby enhancing the quality offiscal adjustment.

8. Actually, the public investment cuts made during the first fiscal adjustment period occurred in the firstyears of Pinochet's government.

9. Recall oil revenues represent about one third of Mexican government revenues.

24

Executive Summarv

Conclusions: The Link Between Fiscal Sustainability and Fiscal Reform

1.55 Mexico faces a number of challenges on the fiscal front, among them:

* procyclical fiscal policies and the absence of automatic stabilizers

* under-provision of infrastructure stocks, and

* sizable contingent liabilities.

Yet the financial markets confirm the favorable econometric findings presented inchapter five. Narrowing sovereign bond spreads, a strong peso, and the March 2000sovereign upgrade by the ratings agencies endorse the measures taken by the authorities.However, the favorable fiscal sustainability outlook can only be maintained via fiscalreform. There are limits to expenditure cuts.

1.56 Figure E.15 makes clear the authorities have reigned in spending dramaticallyover the past fifteen years. In fact, a review of figures E.10 and E.12 (public investment),indicates authorities may have overcompensated on expenditures. The problem is one ofrevenue.

Figure E.15 Deficit Reduction and Oil Dependence

50

45

40

35

30

25

20

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1998 1997 1998

TOTAL REVEBUE OIL REVENUE TOTAL EXPENBDITURE

Source: SHCP

1.57 As previously stated, the continuing fiscal dependence on oil is worrisome. Oilexports as a percentage of total exports have fallen from 80 percent in 1982 to less than 10percent in 1999, yet oil revenues to total tax revenues still approach 30 percent.Strengthening other areas of the tax system would mollify these concerns.

25

Executive Summary

1.58 There are a number of challenges facing the tax authorities. First, the overly-complex tax system creates incentives for noncompliance among private agents.-Duplication of tax declarations and poor coordination between collectors' agencies arebut two examples of a list of administrative problems.

1.59 Add to this a large informal sector that contributes only marginally to fiscalrevenues and weak law enforcement to motivate possible evaders. Estimates of foregoneVAT revenues in the informal sector approach 0.1 percent of GDP. 1

1.60 The VAT system is in need of reform as well. Numerous exemptions exist (e.g.,zero rate for many domestic transactions) and collection surveillance is weak. As FigureE. 16 shows, the VAT contribution is second only to the income tax, followed byproduction and services taxes.

Figure E.16 Components of Tax Revenues as a Percent of GDP, Mexico 1980-1998

6

5-

0r)

0

o X o o t o o X r- oc o o o - o o o r- 00 00 00 00 00 00 00 00 00 Ol al O~ O7 O\ 0~ ON ON 0~5~ OC 3 ON ON l O~ 0~ aN C% O~ oll a~ ON Ol~ O, ON O~ O~

M Income Tax EB On production and services * Others 1 Imports O VAT

Source: SHCP

1.61 Estimates of lost revenues due to VAT and income tax evasion of are on theorder of 5 percent of GDP."1 Reducing exemptions and stronger surveillance wouldaugment revenues. Undoubtedly, strengthening tax administration would add a few GDPpoints to the tax coffers.

'° CIDE (1999).l2 percent for VAT and the rest for income tax. CIDE (1999).

26

Executive Summary

References

Blanchard, 0. (1990) "Suggestions for a New Set of Fiscal Indicators", OECD WorkingPapers No 79, April.

Blejer M. and A. Cheasty. 1991. "The Measurement of Fiscal Deficits: Analytical andMethodological Issues", Journal of Economic Literature, XXIX, (December): 1644-1678.

Bean C. and W. Buiter. 1987. The Plain Man's Guide to Fiscal and Financial Policy.Employment Institute, London.

Buiter, W. 1983. "Measurement of the Public Sector Deficit and Its Implications forPolicy Evaluation and Design". International Monetary Fund Staff Paper, 30(2),(June):306-349.

CIDE (1999) Los impuestos en Mexico. Quien y c6mo se pagan?, Programa dePresupuesto y Gasto Piublico, CIDE, Mexico.

Cuddington. J.T. (1997). "Analyzing the Sustainability of Fiscal Deficits in DevelopingCountries", The World Bank, Policy Research Working Paper #1784, June

Drudi, F. and A. Prati (1999) "Signaling Fiscal Regime Responsibility", IMF WorkingPapers 86, July.

Easterly W. 1999. "When is Fiscal Adjustment an Illusion?" Economic Policy, (April):57-86.

Easterly W., C. Rodriguez and K. Schmidt-Hebbel (Eds.). 1994. Public Sector Deficitsand Macroeconomic Performance, Oxford University Press, New York.

Hakkio, C.S. and Rush, M. (1991). "Is the Budget Deficit "Too Large"?, EconomicInquiry, Vol. XXIX, July, pp. 429-445.

Hamilton, J. and M. Flavin (1986). "On the Limitations of Government Borrowing: AFramework for Empirical Testing", American Economic Review, Vol. 26, pp. 808-819.

Haug, A. (1991). "Cointegration and Government Borrowing Constraints: Evidence forthe United States", Journal of Business and Economics Statistics, 9, pp.97 -1 0 1 .

Hulten, C. R. 1996. "Infrastructure Capital and Economic Growth: ,How Well You Use ItMay Be More Important Than How Much You Have." Working paper 5847,National Bureau of Economic Research, Cambridge, Mass.

Lewis, Christopher M., and Ashoka Mody. 1997. "The Management of ContingentLiabilities: A Risk Management Framework for National Government." in Timothy

27

Executive Summarv

Irwin et al., eds., Dealing with Public Risk in Private Infrastructure. Washington,D.C.: World Bank.

Lustig, N (1992) "Mexico: The Remaking of an Economy", The Brookings Institution

OECD (1999) "OECD Economic Surveys: Mexico"

Perron, P. (1989). "The Great Crash, the Oil Price Shock, and the Unit RootHypothesis", Econometrica, 57, November, pp. 1,361-1,401

Polackova, Hana. 1999. "Contingent Govemment Liabilities: A Hidden Fiscal Risk";Finance & Development, (March):46-49.

Polackova, Hana. 1998. "Government Contingent Liabilities: A Hidden Risk to FiscalStability." World Bank Policy Research Working Paper No. 1989. Poverty Reductionand Economic Management Unit, Europe and Central Asia Regional Office, WorldBank, Washington, D.C.

Rocha, F (1997). "Long Run Limits on the Brazilian Government Debt", RevistaBrasileira de Eonomia, Out/Dez, pp. 447-470.

Tanner, E. (1994). Balancing the Budget with Implicit Domestic Default: The case ofBrazil in the 1980's, World Development, Vol. 22, No 1, January, pp. 85-98

Tanner, E. (1995). Intertemporal Solvency and Indexed Debt: Evidence from Brazil1976-1991, Journal of International Money and Finance, August, Vol. 14, and No 4,pp. 549-73.

Towe, Christopher, M. 1991. "The Budgetary Control and Fiscal Impact of GovernmentContingent Liabilities." International Monetary Fund Staff Papers 38 (1): 109-34.International Monetary Fund, Washington, D.C.

, 1990. "Government Contingent Liabilities and the Measurement ofFiscal Impact." International Monetary Fund Working Paper, WP/90/57, June.International Monetary Fund, Washington, D.C.

Trehan, B., and C., Walsh (1988). "Common Trends, the Government Budget Constraintand Revenue Smoothing", Journal of Economic Dynamics and Control, Vol. 12, pp.425-444.

Trehan, B., and C. Walsh (1991). "Testing Intertemporal Budget Constraints: Theoryand Applications to U.S. Federal Budget and Current Account Deficits", Joumal ofMoney, Credit and Banking, Vol. 23, May, pp. 206-223.

Uctum, M. and M. Wickens (1997). "Debt and Deficits Ceilings, and Sustainability ofFiscal Policies: An Intertemporal Analysis", CEPR, Discussion Paper #1612, March.

Werner, M. (1992) "La Solvencia del Sector Puiblico: El Caso de Mexico en 1988", ElTrimestre Economico, Octubre-Diciembre, pp. 751-772

28

Executive Summarv

Wilcox, D. (1989). "The Sustainability of Government Deficits: Implications of thePresent-Value Borrowing Constraints", Journal of Money, Credit and Banking, Vol.-21, pp. 291-306

Zivot, E. and Andrews, D. (1992). "Further Evidence of the Great Crash, the Oil-PriceShock, and the Unit Root Hypothesis ", Journal of Business and Economics Statistics,10, pp. 251-70

29