Fiscal and Macroeconomic Impact of Privatization · economic impact of privatization.1 Among the inter-national financial institutions, the World Bank has had the lead role in advising

Fiscal and MacroeconomicImpact of Privatization

Jeffrey Davis, Rolando Ossowski,Thomas Richardson, and Steven Barnett

INTERNATIONAL MONETARY FUNDWashington DC

2000

©International Monetary Fund. Not for Redistribution

© 2000 International Monetary Fund

Production: IMF Graphics SectionTypesetting: Julio R. PregoFigures: Sanaa Elaroussi

Library of Congress Cataloging-in-Publication Data

Fiscal and macroeconomic impact of privatization / Jeffrey Davis... [et al.].p. cm—(Occasional paper, ISSN 0251-6365 ; no. 194)

Includes bibliographical references.ISBN 1-55775-888-3

1. Privatization—Case studies. 2. Government business enterprises—Case studies. 3. Finance, Public—Case studies. I. Davis, Jeffrey M.,1946 II. Occasional paper (International Monetary Fund); 194.HD3845.6 .F57 2000338.9'75—dc21 00-0031909

Price: US$20.00(US$17.50 to full-time faculty members and

students at universities and colleges)

Please send orders to:International Monetary Fund, Publication Services

700 19th Street, N.W., Washington, D.C. 20431, U.S.A.Tel: (202) 623-7430 Telefax: (202) 623-7201

E-mail: [email protected]: http://www.imf.org

©recycled paper


http://www.imf.org

Contents

Preface

Page

V I I

I Overview

Privatization ProceedsUses of Privatization ProceedsFiscal Impact of PrivatizationMacroeconomic Impact of PrivatizationIssues for IMF-Supported Programs

II Privatization Proceeds

Scale of Privatization Proceeds in the Case Study CountriesFactors Affecting Budgetary ProceedsFiscal Reporting of Privatization Proceeds

III Use of Privatization Proceeds

Privatization and Government Net WorthShort-run Macroeconomic Effects of PrivatizationUse of Privatization Proceeds: Choices and Experience

IV Fiscal Impact of Privatization

Contemporaneous Use of Privatization ProceedsEffects of Privatization on the Fiscal Accounts Over Time

V Macroeconomic Impact of Privatization

Effects on Growth and InvestmentImpact on Labor Markets

VI Issues for IMF-Supported Programs

Privatization in IMF-Supported ProgramsPrivatization and Program Design

Appendices

12223

447

91012

16

1617

23

2324

27

2729

I Factors Affecting the Sale Price of State Assets 32

II Privatization and Fiscal and Macroeconomic Developments 37

References 40


Boxes

Section

III 1.IV 2.

3.V 4.

Appendix

I 5.6.

Tables

Section

II

III

IV

V

VI

1.

2.

3.4.5.6.7.8.

9.

Appendix

I 10.11.

12.

II

Figures

Section

IV 1.

VI

2.

3.

Sterilization of Privatization Revenues in Hungary 11The Impact of Economic Transition on Tax Administration 18Fiscal Issues in the Privatization of Profitable Public Enterprises 20Mitigating the Social Impact of Privatization 25

Governance Issues in Privatization 34Loans-for-Shares Privatization in Russia 35

Gross and Budgetary Privatization Proceeds inCase Study Countries 5

Privatization Proceeds Accruing to the Budget inCase Study Countries 6

Budgetary Treatment and Classification of Privatization Proceeds 8Use of Privatization Proceeds in Case Study Countries 12Earmarking of Privatization Receipts in Selected Countries 14Impact of Privatization on Domestic Financing 17Reduction in Debt Stock and Privatization 22Summary of Three Studies of Firm-Level Efficiency Gains from

Privatization 24Design of Adjusters Related to Privatization Receipts in

Programs with Case Study Countries 30

Factors Affecting the Sale Price: Country Illustrations 33Contemporaneous Impact of Budgetary Privatization Proceeds on

Domestic Financing 38Structural Relationship Between Total Privatization Proceeds and

Selected Variables 39

Gross Budgetary Transfers and Subsidies to Public Enterprisesfor Selected Countries 19

Operations of the Public Enterprise Sector Before and AfterPrivatization for Selected Countries 21

Monitoring Privatization in IMF Arrangements Approvedin 1994-98 28


The following symbols have been used throughout this paper:

. . . to indicate that data are not available;

n.a. to indicate not applicable;

— to indicate that the figure is zero or less than half the final digit shown, or that the itemdoes not exist;

between years or months (i.e., 1997-98 or January-June) to indicate the years or monthscovered, including the beginning and ending years or months;

/ between years or months (i.e., 1997/98) to indicate a crop or fiscal (financial) year.

"Billion" means a thousand million; "trillion" means a thousand billion.

Minor discrepancies between constituent figures and totals are due to rounding.

The term "country," as used in this paper, does not in all cases refer to a territorial entity thatis a state as understood by international law and practice; the term also covers some territorialentities that are not states, but for which statistical data are maintained and provided interna-tionally on a separate and independent basis.



Preface

The paper reflects the contributions of several members of staff in the IMF's FiscalAffairs Department. In particular, the authors are grateful to Rosa Alonso, BenedictBingham, Isaias Coelho, Louis Kuijs, Jun Ma, Eric Mottu, Martin Petri, ArnimSchwidrowski, Istvan Szekely, George Tsibouris, and Timo Valila for their researchinto the privatization experience of the case study countries. They would also like tothank John Nellis, as well as others at the World Bank, for their advice and support.Helpful and insightful comments on the paper were provided by Vito Tanzi, and col-leagues in the Fiscal Affairs Department and in other departments in the IMF.

The authors appreciate as well the efficient research assistance provided byWilliam Riordan and the document preparation assistance provided by HeatherHuckstep. Gail Berre of the External Relations Department edited the paper and coor-dinated production of the publication.

The views expressed in the paper, as well as any errors, are the sole responsibilityof the authors and do not necessarily represent the opinions of the Executive Board ofthe IMF or other members of the IMF staff.


This page intentionally left blank


I Overview

Privatization has been a key element of structuralreform in many developing and transition

economies during the last decade. Governments un-dertaking privatization have pursued a variety of ob-jectives: achieving gains in economic efficiency,given the extensive prevalence of poor economicperformance of public enterprises in many countriesand limited success with their reform; and improv-ing the fiscal position, particularly in cases wheregovernments have been unwilling or unable to con-tinue to finance deficits in the public enterprise sec-tor. In addition, liquidity-constrained governmentsfacing fiscal pressures have sometimes privatizedwith a view to financing fiscal deficits with the pro-ceeds. Other objectives have included the develop-ment of domestic capital markets.

These efforts have been reviewed in a large litera-ture on the microeconomic aspects of privatizationthat has emphasized the potential efficiency gains.However, there has been less work, particularly ofan empirical nature, done on the fiscal and macro-economic impact of privatization.1 Among the inter-national financial institutions, the World Bank hashad the lead role in advising on the design and im-plementation of the reform of public enterprises, in-cluding divestiture. Privatization, however, has im-portant fiscal and macroeconomic implications andis therefore also of interest to the International Mon-etary Fund (IMF). Indeed, privatization has becomean important component of programs in a large num-ber of countries.

This paper reviews fiscal and macroeconomic is-sues in the privatization of nonfinancial public enter-prises in developing and transition economies.2 Suc-

1An excellent survey of the microeconomic literature is pro-vided in Megginson and Netter (1999). Heller (1990) and Hem-ming and Mansoor (1987) provide examples of earlier discus-sions of fiscal and macroeconomic issues, and Hachette andLuders (1993), Pinheiro and Schneider (1995), and Larrain andWinograd (1996) offer examples of more recent empirical workin this area. World Bank (1995) discusses many aspects of inter-national privatization experience.

2For a discussion of issues in the privatization of financial in-stitutions, see Beyer, Dziobek, and Garrett (1999); and Ver-brugge, Megginson, and Owens (1999).

cessive sections of the paper consider the followingissues: proceeds from privatization and the factorsdetermining the amount accruing to the budget; usesof the proceeds; empirical evidence on the impact ofprivatization on the budget and macroeconomic ag-gregates; and the privatization component of IMF-supported programs. The empirical evidence drawson a series of case study countries selected to be rep-resentative of a range of privatization experience indeveloping and transition economies and to reflectgeographical diversity.3

Privatization Proceeds

Proceeds from privatization have been substantialin a number of developing and transition economies.Gross receipts that can be transferred to the budgetare affected by actions prior to sale, the salesprocess, and the postprivatization regime. Amountsaccruing to the budget are found to be less than thesales proceeds as a result of extrabudgetary manage-ment and the wide divergence between gross and netreceipts. Among the major conclusions concerningprivatization proceeds are the following:

First, off-budget placement of privatization pro-ceeds can lead to limited control and lack of trans-parency in their use. Extrabudgetary funds should beregulated, with accounts publicly reported, audited,and subject to parliamentary oversight.

Second, privatization transactions should be trans-parently reported on a gross basis. Costs for restruc-turing, recapitalization, or writing off public enter-prise debt should be recorded as spending financedby the gross proceeds of sale.

Third, privatization is an exchange of assets; thereceipts are lumpy, one-off, and uncertain. Thus, pri-vatization proceeds should be treated as a financingitem in the fiscal accounts.

3The sample comprises 18 countries and includes 10 nontransi-tion economies (Argentina, Bolivia, Cote d'Ivoire, Egypt, Mex-ico, Morocco, Mozambique, Peru, the Philippines, and Uganda)and 8 transition economies (the Czech Republic, Estonia, Hun-gary, Kazakhstan, Mongolia, Russia, Ukraine, and Vietnam).


Uses of Privatization Proceeds

An evaluation of the potential uses of privatiza-tion receipts should reflect the implications for gov-ernment net worth and their macroeconomicimpact.

Insofar as government net worth is concerned, re-ceipts from privatization do not of themselves indi-cate that the government is better off. Privatizationhas longer-term implications in terms of revenuesforegone and/or expenditures that will not be madein the future, and government decisions on the use ofproceeds should reflect these intertemporal effects.Government net worth will rise to the extent that pri-vate sector ownership leads to an increase in effi-ciency and the government shares in this gain.

The macroeconomic effects of privatization de-pend, in part, on whether receipts are from domesticor foreign sources, the degree of capital mobility,and the exchange regime. Broadly, the effects of anincrease in the deficit financed by privatization re-ceipts would be similar to those resulting from adebt-financed fiscal expansion. Use of proceeds toreduce external debt provides for an automatic steril-ization of what may be substantial capital inflowsassociated with privatization. Reduction of domesticdebt may impact domestic liquidity.

Assessment of governments' stated intentionswith respect to the use of privatization receipts needsto allow for the fungibility of resources. Uses com-monly considered are

• Higher expenditure. Privatization receipts aretemporary and often uncertain, thus it is not ad-visable to rely on them for current spending.Targeted use to help cushion the short-term so-cial impact of privatization can be appropriate.Use of the proceeds to finance additional capitalspending need not reduce government networth, though it often raises concerns as to thequality of the projects.

• Reduction in net debt. This may be achieved byretiring debt (or settling arrears), or building upassets, with the choice determined by debt man-agement considerations. Advantages includemaintenance of government net worth and pos-sibly favorable signaling effects that could im-pact the cost and availability of debt.

• Earmarking of privatization receipts. Doing sofor particular expenditures complicates fiscalmanagement and makes it difficult to reallocatespending in response to changes in circum-stances and priorities. As such it should bediscouraged.

• Relaxing the fiscal constraint. Privatization pro-ceeds might serve a limited role in providing atemporary cushion for countries pursuing ag-gressive adjustment and reform programs.

Fiscal Impact of Privatization

Privatization has a contemporaneous impact onthe budget and budgetary effects over time. Econo-metric results suggest that, for the case study coun-tries, privatization receipts are saved rather thanspent. This result applies for receipts channeledthrough the budget, which does not support argu-ments that extrabudgetary treatment is necessarilyrequired for prudent management. Most of the coun-tries covered had an IMF program in place, and con-sequent limitations on the deficit may have substan-tially influenced this finding.

Piecemeal evidence suggests that, over time, thefiscal situation tends to benefit from privatization. Inparticular, both the firm level and more aggregatedata support positive impacts on revenue; transfersdecline markedly following periods of privatization;and broader indicators of consolidated public enter-prise accounts for some countries indicate a largedecline in deficits, and probably also in quasi-fiscaloperations.

Data on public enterprise operations and on the fi-nancial flows between enterprises and the govern-ment (including quasi-fiscal flows) are often inade-quate. There is a need for improvements in this areain many countries.

Macroeconomic Impact ofPrivatization

Both the microeconomic and case study data aresupportive of the positive effects of privatizationover time on growth and employment. These resultshold for the full sample and the transition countries,though they are less pronounced for the latter.

Growth

The microeconomic evidence indicates that privatefirms are operationally more efficient than those heldby the state, particularly in competitive industries. Astrong correlation is also found for the case studycountries between privatization and growth. How-ever, and consistent with the growth literature, privat-ization is likely serving as a proxy in the regressionsfor one or more missing variables that may broadlybe characterized as a favorable regime change.

Labor Markets

Public enterprises often seek to maintain employ-ment, and benefit from soft budget constraints. Con-sequently, there is concern that privatization maylead to increased unemployment. Although empiri-cal evidence suggests that aggregate unemployment


tends to decrease following privatization, particulargroups of workers may still be adversely affected.This lends importance to measures that mitigate itssocial impact.

Issues for IMF-Supported Programs

The World Bank takes the lead in privatization,but the IMF has cooperated closely with it in thisarea. Drawing on the Bank's experience and recom-mendations, a majority of IMF-supported programsin recent years have included some form of condi-tionality on privatization.

Monitoring of privatization in IMF-supportedprograms has emphasized conditionality on process

and targets. Consistent with the recent emphasis inthe World Bank, there is scope for IMF conditional-ity to give more weight to privatization procedureswhere these have important fiscal and macroeco-nomic impact. Similarly, programs should, in somecases, give greater importance to the establishmentof an appropriate regulatory environment withinwhich privatized firms operate.

The design of financial programs should includeas broad a definition of privatization receipts as pos-sible in the fiscal targets and quantitative perfor-mance criteria, and consider the macroeconomic ef-fects in assessing use. Adjusters should address theuncertainty attached to the amount of receipts; ingeneral, higher-than-anticipated receipts should besaved.


II Privatization Proceeds

This section presents data on the scale of grossprivatization proceeds and the amounts accruing

to the budget for the case study countries. It thenconsiders the factors affecting budgetary proceedsand their treatment in the fiscal accounts.

Scale of Privatization Proceeds in theCase Study Countries

Table 1 presents data on gross privatization re-ceipts and the amounts accruing to the budget. Dataon cumulative gross privatization proceeds collectedduring the years of active privatization through 1997are derived from a privatization database preparedby the World Bank. Privatization proceeds were onaverage 1 3/4 percent of GDP a year during the activeprivatization period. For the transition case studycountries, this figure was 2 percent of GDP a year,while for nontransition countries it was 1 1/2 percentof GDP.4

A somewhat different picture is provided, how-ever, by data on privatization proceeds accruing tothe budget as recorded in the IMF's fiscal accounts.These proceeds were on average 3/4 percent of GDP ayear during the active privatization period (see Table1 and Table 2). It is striking that in many countriesthe privatization proceeds actually received by thebudget were less, and in some cases significantly so,than the gross privatization receipts. The institu-tional, operational, and accounting reasons for thesedifferences are discussed in the next section.

Factors Affecting Budgetary Proceeds

Gross receipts that can be transferred to the bud-get are affected by actions prior to sale, the salesprocess, and the postprivatization regime. In somecountries, these receipts have been limited by invest-

ments to physically restructure public enterprisesprior to privatization, restrictions on potential bid-ders, and postprivatization commitments on the newowners.5 Actions in those various areas may alsogive rise to governance issues. Factors affecting thesale price of assets are reviewed in Appendix I.

As noted above, in many countries the budgetaryproceeds from privatization have been less than thegross sales value of the assets divested. This reflectsextrabudgetary management of the privatization pro-ceeds and the fact that budgets have tended to re-ceive the net value of divested assets after certaincosts have been subtracted.

Extrabudgetary Management of PrivatizationProceeds

From an institutional point of view, privatizationrevenues may accrue in the first instance to the bud-get or to off-budget public institutions. The casestudy countries illustrate the prevalence of off-budget treatment of privatization transactions (seeTable 3). In several cases, the proceeds initially ac-crued to an extrabudgetary fund, which then madetransfers to the budget of a portion of the receipts. Ina few instances, the proceeds are recorded entirelyoff budget. These institutional and operationalarrangements may have reflected two concerns onthe part of governments: first, the notion that privat-ization proceeds should be held off budget as ameans of protecting them from parliaments thatmight be inclined to spend them inappropriately; andsecond, the one-off nature of privatization proceedsand the perception that they constitute an unex-pected windfall.

The accrual of privatization receipts to off-budgetagencies and/or their exclusion from the budget mayhamper fiscal policy control and reduce trans-parency and oversight. Although in some countriesthe financial transactions related to extrabudgetary

4These data underestimate the extent of state divestiture of as-sets to the extent that noncash methods of privatization—such asvouchers in transition countries—were used.

5Appropriate regulatory frameworks should, however, be setup prior to the privatization of public enterprises with substantialmonopoly power.


•

Table I. Gross and Budgetary Privatization Proceeds in Case Study Countries

Argentina

Bolivia

Cote d'lvoire

Czech Republic

EgyptEstonia

Hungary

Kazakhstan

MexicoMongolia

Morocco

Mozambique

PeruPhilippines

Russia

UgandaUkraine

Vietnam

Average of all countries

with dataAverage of transition

countries with data6

Average of nontransition


Years of ActivePrivatization

1990-951995-981994-971991-971993-981992-981991-981993-981989-941994-981993-971992-981994-971987-971992-981991-981993-981993-98

Gross Privatization ProceedsDuring Period

of Active Privatization1

Millions ofU.S. dollars

22,885

863459

2,369

2,778

46711,841

5,798

25,249

1,489

1017,1363,8106,569

152243

n.a.

n.a.

n.a.

Percent ofGDP4

2.04.2

1.10.9

0.82.9

4.0

5.51.3

1.01.13.20.70.30.4——

1.7

1.9

1.6

Net Privatization ProceedsAccruing to Budget During Period

of Active Privatization2

Millions ofU.S. dollars5

11,452

160448

1,7362,194

—6,672

2,425

6,31045

1,307—

5,821

3,407

8,632—

518—

n.a.

n.a.

n.a.

Percent ofGDP4

0.9

0.5

1.1

0.8

0.5—

2.2

2.2

0.30.9

0.9—

2.6

0.50.4—

0.2—

0.8

0.8

0.7

Ratio of Net ProceedsAccruing to Budgetto Gross Proceeds3

Percent

5012987379—562525

88—8288

—

—

45

31!

52

Sources:World Bank; and IMF staff estimates.

1World Bank Privatization database. Data are available for 1988-97.2 Executive Board documents and staff estimates. Data through latest available observation.3 Ratio calculated for the period of active privatization for which information is available in both sets of data. Differences in coverage in the two series

may account for the higher proceeds recorded in IMF data for Russia and Ukraine.4 Average of annual ratios of privatization proceeds to GDP during the period of active privatization. For the Philippines, ratios to GNP.5 Annual data in national currency converted to U.S. dollars using annual average exchange rates.6 Transition case study countries comprise the Czech Republic, Estonia, Hungary, Kazakhstan, Mongolia, Russia, Ukraine, and Vietnam.7 Nontransition case study countries comprise Argentina, Bolivia, Cote d'lvoire, Egypt, Mexico, Morocco, Mozambique, Peru, the Philippines, and

Uganda.

funds—including those with responsibility forprivatization and state enterprise management—aretransparent and subject to parliamentary or otheroversight (for instance, in Hungary), this is not al-ways the case. The privatization receipts that do notflow through the budget may then be subject to lesspublic scrutiny on their accrual and use than regularbudgetary spending.6 In addition, the use of the pro-ceeds may not be subject to priority-setting withinthe budget process.

6Section 2.1.1 of the Code of Good Practices on Fiscal Trans-parency—Declaration of Principles states: "The annual budgetshould cover all central government operations in detail andshould also provide information on central government extrabud-getary activities."

In Uganda, for example, privatization receiptswere placed off budget in a "divestiture account,"which was charged with paying off the debts of pub-lic enterprises, compensating workers who were laidoff as a result of divestiture, and other unspecifiedpurposes meant to promote successful privatization.The legislation setting up the divestiture accountwas vague on accounting and audit procedures, andallegations of misuse of the funds that passedthrough it—as well as widespread asset-stripping—have led to frequent complaints by parliament,among others, that governance issues in the privat-ization process needed to be addressed.

Privatization proceeds should be transparentlyrecorded and subject to effective oversight. This


Table 2. Privatization Proceeds Accruing to the Budget in Case Study Countries1

(In percent of GDP)

ArgentinaBoliviaCote d'lvoireCzech RepublicEgypt3

EstoniaHungaryKazakhstanMexicoMongoliaMorocco3

MozambiquePeruPhilippines4

RussiaUganda3

UkraineVietnamAverage of all

countries with dataAverage of all transition


Average of all nontransitioncountries with data6

1990

1.00.1

——

0.3

0.3

0.3

1991

1.20.1

—0.4

0.3

———0.3

—

0.2

0.2

0.2

1992

0.80.20.1

—1.5

0.8

——0.10.10.5—

0.3

0.7

0.2

ISiliiitliiSli

1993

1.5—0.10.7——1.02.70.4—0.9—0.40.10.2—0.1—

0.5

0.6

0.3

1994

0.30.20.21.4——1.70.30.20.20.7.—4.61.70.1—0.2—

0.7

0.5

0.8

1995

0.40.20.71.2——4.00.7—0.90.4—1.61.20.3—0.1—

0.7

0.9

0.5

1996

0.10.31.00.20.2—4.22.2—0.41.0—3.30.30.1—0.2—

0.8

0.9

0.6

1997

0.92.40.42.5—2.63.30.61.41.4—0.90.40.9—0.1—

1.0

1.1

0.9

1998

0.60.5—

—

4.30.41.70.4

0.40.10.7—0.4—

0.6

1.0

0.3

Annual AverageDuring Active

Privatization Period2

0.90.51.10.80.5—2.22.20.30.90.9—2.60.50.4—0.2—

0.8

0.8

0.7

Source: IMF staff estimates.1Shaded years indicate periods of active privatization.2Average of annual ratios of privatization proceeds to GDP during the period of active privatization.3Data for Egypt, Morocco, and Uganda are on a fiscal year basis, shown in the table as the year in which the fiscal year begins. In 1996, Morocco

switched from a January—December fiscal year to a July—June fiscal year. As a result, 1996 data are for the first half of 1996 only, 1997 data are for July1996-June 1997, and 1998 data are for July 1997-June 1998.

4ln percent of GNP.5Transition case study countries comprise the Czech Republic, Estonia, Hungary, Kazakhstan, Mongolia, Russia, Ukraine, and Vietnam.6Nontransition case study countries comprise Argentina, Bolivia, Cote d'lvoire, Egypt, Mexico, Morocco, Mozambique, Peru, the Philippines, and

Uganda.

could be achieved through full consolidation of theprivatization accounts in the budget that is approvedby parliament to ensure that the claim on govern-ment resources arising from privatization expenses,as well as the revenues, are adequately accountedfor; that financing decisions are taken with due re-gard to all government assets and liabilities; and thatall operations affecting the government's balancesheet are reported to parliament. If, however, extra-budgetary institutions are in charge of privatization,their operations must be subject to clear rules andregulations, and they should be publicly reported,audited, and subject to parliamentary oversight.Consolidated general government fiscal perfor-mance, inclusive of privatization transactions,should also be reported.

Reporting Gross Versus Net PrivatizationProceeds

In virtually all of the case study countries, privat-ization proceeds were reported on a net basis (inboth the IMF and the authorities' accounts), aftertaking account of the overhead costs of the privatiza-tion agency, and usually after subtracting the costs ofany preprivatization restructuring. Moreover, it isdifficult to form an accurate picture of the overallgross amount of sales proceeds, and relatively littleinformation is available on the actual disposition ofthe assets that were allocated to the costs ofprivatization.

In many cases, only cash privatization revenuesare included in the budgetary or fiscal accounts. Pay-


ments made by buyers in the form of debt instru-ments previously issued by the government sellingthe assets have often not been included in the priva-tization revenues reported in the fiscal accounts.Since payment in the form of government debt in-struments represents amortization of public debt, aclearer picture is conveyed by including in theprivatization receipts the market value of the debtextinguished, and by showing debt amortization ascounterpart.

These considerations suggest that a transparentapproach to recording the transactions related to dis-posing of state assets would be on a gross basis. Thiswould involve reporting expenditures for restructur-ing, recapitalization, or for writing off public enter-prise debt as expenditures on the budget that are fi-nanced by the gross proceeds of the sales.7

Fiscal Reporting of PrivatizationProceeds

The accounting treatment of privatization pro-ceeds does not itself predetermine use. However, theway that privatization receipts are presented in thefiscal accounts may have a bearing on decisions re-garding their use, as well as on public perceptions ofthe fiscal stance.

This issue involves the question of whether suchproceeds should be regarded as government revenueor as financing. The current version of A Manual onGovernment Finance Statistics (IMF, 1986) takesthe position that transactions involving nonfinancialassets are part of capital revenue or expenditure, andthat financial asset transactions undertaken for pol-icy purposes are part of net lending. Hence, it rec-ommends that privatization receipts be treated as adeficit-determining ("above-the-line") item andrecorded as capital revenue in the case of sales ofequipment, and (negative) net lending in the case ofshares. The suggested classification ensures consis-tency in the intertemporal treatment of the assets in

7If the amount of public enterprise liabilities assumed by thebudget is large, it might be appropriate to show a deficit includingand excluding this amount.

question. If the acquisition of the productive asset ororiginal investment in the public enterprise were un-dertaken for policy purposes and classified asdeficit-determining items, on grounds of symmetry asimilar treatment would be required in the case ofprivatization proceeds.

This treatment of receipts from the sale of assetshas been regarded as an unsatisfactory basis for fis-cal analysis for some time. Privatization receiptshave characteristics that warrant differentiating themfrom other government revenues in the design andassessment of fiscal policy: privatization is an ex-change of assets, and spending the proceeds affectsthe government's intertemporal budget constraint;these receipts represent one-off resources and shouldnot be counted upon to support the fiscal positionpermanently; their lumpy nature can distort analysisof the underlying deficit and provide a misleadingview of the sustainability of the fiscal position; and,in general they are more uncertain in timing and sizethan other government revenues. Because of theseconsiderations, it is preferable to treat privatizationproceeds as a financing ("below-the-line") item inthe fiscal accounts (Mackenzie, 1998).

A revision of A Manual on Government FinanceStatistics is under way, which will address the prob-lems in the treatment of privatization proceeds byclearly separating asset/liability transactions fromgovernment operations that change net worth. Also,a distinction will be made between asset/liabilitytransactions that are undertaken for policy purposesand those intended to manage the liquidity positionof the government. In this new treatment, privatiza-tion proceeds will have no impact on the balance ofgovernment operations that determines the change innet worth, but the proceeds will be a determinant ofthe volume of transactions required for managingthe government's liquidity position.

In most of the case study countries, the proceedsfrom privatization that go through the budget wereclassified above the line by the authorities, either asrevenue or as negative net lending. The fiscal ac-counts compiled by IMF staff for the countries in thesample, however, increasingly record privatizationreceipts as an element of deficit financing; currently,this is done in about two-thirds of the cases in thesample (see Table 3).


Table 3. Budgetary Treatment and Classification of Privatization Proceeds

Country Placement of Proceeds Official Budgetary Recordingand Classification

IMF Budget Classification

Argentina

Bolivia

Cote d'lvoire

Czech Republic

EgyptEstonia

Hungary

Kazakhstan

Mexico

Mongolia

Morocco

Mozambique

Peru

Philippines

Russia

Uganda

Ukraine

Vietnam

Budget

Capitalization: enterprise

Cash receipts: budget

Budget

Agency3

Agency3

Agency (partly)6

Budget (partly)

Agency3

Budget

Budget (partly)

Budget

Budget

Agency6

Budget (partly)

Budget

Budget

Agency6

Budget

Agency6

On budget: revenue

Capitalization part: off budget

Cash receipts: on budget: revenue

On budget: revenue2

On budget: lending minus repayment4

Only transfers from agency: financing

Agency part: off budget

Budget part: revenue

Only transfers from agency: revenue

On budget: lending minus repayment

Revenue8

On budget: revenue

On budget: revenue

Off budget

On budget: revenue10

On budget: revenue

On budget: revenue

Off budget

On budget: revenue

Off budget

Financing

Cash receipts: revenue1

Financing

Revenue4

Financing5

Financing (since 1999)7

Not included (through 1998)

Financing (since 1998)5

Revenue (through I997)5

Financing (since 1998)Revenue (through 1997)

Revenue8

Revenue

Financing9

Revenue (since 1999)Not included (through 1998)



Financing

Not included

Financing

Not included

Sources: Data provided by country authorities; and IMF staff.1Under Bolivia's capitalization scheme, investors bid for a 50 percent stake in the public enterprise in exchange for a commitment to undertake invest-

ment. The remaining shareholding is retained by the government and used to finance a minimum pension scheme. The capitalization program is treatedas off-budget investment in the firms partly divested.

2ln the official classification, all financing is treated as "revenue."3The privatization agency makes transfers to the budget.4The revenue accruing to the privatization agency is consolidated in the budget and in the IMF budget accounts since 1998. The fiscal balance is shown

including and excluding privatization receipts.5Only includes the privatization proceeds transferred to the government by the privatization agencies.6The privatization agency does not make transfers to the budget.7Only the part accruing to the budget.8Proceeds from the privatization of Telmex and commercial banks were kept off budget.9Recorded on a separate line in the IMF presentation between the overall balance (excluding privatization receipts) and financing.10Proceeds used for social expenditure and to recapitalize the pension fund were kept off budget. The Ministry of Finance shows budgetary proceeds

as revenue, and the central bank as financing.


Ill Use of Privatization Proceeds

The appropriate size of the fiscal deficit is largelydetermined by the overall macroeconomic ob-

jectives and fiscal sustainability. Viewed as a sourceof financing, akin to a bond sale, the amount of pri-vatization proceeds generally should not itself deter-mine the size of the deficit, and, moreover, themacroeconomic consequences are also similar toconventional bond financing. Nonetheless, privat-ization proceeds are distinct in certain ways. First,privatization may impact government net worth,which in turn has consequences for fiscal sustain-ability. Second, the privatization program mightpose specific risks to macroeconomic stability—partly due to the size, lumpiness, and uncertain tim-ing of privatization receipts—that enhance the needfor monetary and fiscal policy coordination. Andthird, the discrete nature of privatization proceedsleads to questions regarding their appropriate use.

Privatization and GovernmentNet Worth

The privatization of government productive assetsyields immediate financial proceeds. But the receiptsfrom privatization do not of themselves indicate thatthe government is better off or that its spending con-straints today and over time are relaxed. Privatiza-tion has longer-term implications in terms of rev-enues foregone and/or expenditures that will not bemade in the future, and government decisions on theuse of privatization proceeds should reflect these in-tertemporal effects. Assessing them requires ananalysis of the impact of privatization on the incomeflows and on government net worth.

The financial relationship between the govern-ment and public enterprises includes budgetary re-ceipts of taxes and dividends, and other transfersfrom, and current or capital transfers to, public en-terprises, including in the form of subsidies. In addi-tion, there are often quasi-fiscal costs that are notrecorded in the budget, such as directed and/or sub-sidized lending, as well as contingent liabilities asso-ciated with implicit or explicit government guaran-tees. The analysis of intrapublic sector financial

flows should also consider any cross arrears, for in-stance, arrears on utility bills by the government andother public sector entities, which are common insome countries.

In a broader context, therefore, the change in gov-ernment net worth due to privatization would equalthe sum of the privatization proceeds and the netpresent value of the taxes of the privatized firm,minus the sum of the net present value of the netsubsidies and transfers (including dividends) to orfrom the public enterprise (including quasi-fiscalcosts)8 and the net present value of the lost taxes ofthe public enterprise.9

In the case where the rate of return on the asset inthe public sector equals that of the asset in privatehands, the government receives financial assets andgives up a stream of net future earnings on the assetsof equal value. Privatization in this case would sim-ply involve a change in the composition of the gov-ernment's assets without effects on its net worth, andthe government's intertemporal budget constraintwould not be affected. However, the conditions forthis to be true are restrictive: markets should be effi-cient—meaning, among other things, that informa-tional asymmetries should not be important and thatthe government can privatize efficiently; the publicand private sectors must use the same discount rate;the firm should be no more or less profitable afterprivatization; and it should face an identical externalenvironment before and after privatization (Heller,1990, and Hemming and Mansoor, 1987).

If the private sector is expected to run the enter-prise more efficiently than would the state, govern-ment net worth would increase provided the govern-ment can privatize and tax efficiently. In such a case,the government's intertemporal budget constraint

8This assumes that such transfers cease after privatization.9The sale of a productive asset, which would have yielded a

certain rate of return in the form of dividends or social benefitshad it remained in public hands, may be compared to that of agovernment bond. In either case, the state receives liquid assetsthat, were they to be used to finance current expenditures, wouldrequire higher taxes or lower spending in the future. However, theone-off nature of privatization is an important feature that distin-guishes it from the issuance of bonds.


would be loosened, and privatization would have apermanent effect on public finances. Conversely, thesale price of a public enterprise may be below thenet present value of the expected flow of net earn-ings in private hands: the state, which may manage afirm less efficiently than would the private sector,may not be capable of selling it well either; it maychoose to sell the firm at a price less than the maxi-mum in recognition of externalities associated withthe creation and development of capital markets; andthe private and the public sector may value a firmdifferently due to different tolerances for risk, accessto financial markets, and rates of time preference.Moreover, a government facing liquidity constraintsmay be willing to sell state assets at less than theireconomic value to finance higher expenditure.

Short-run Macroeconomic Effects ofPrivatization

Decisions on the use of privatization receiptsshould reflect the consequent impact on the fiscalpolicy stance and on macroeconomic aggregates andobjectives. These implications may differ accordingto such factors as the source of privatization receipts(domestic or external), the degree of capital mobil-ity, and the exchange rate regime. It is also importantto consider the fiscal and monetary policy implica-tions of the substantial capital inflows that may belinked to privatization.

Fiscal Stance

The effects of an increase in the deficit throughhigher spending or lower taxes that is financed byprivatization proceeds would be similar to those re-sulting from a fiscal expansion financed by an in-crease in public debt (Mackenzie, 1998). It wouldput pressure on domestic demand with consequencesfor activity, inflation, and the external current ac-count. The relative significance of these effects willdepend in part on the initial macroeconomic positionand the composition of the increase in spending interms of imported and domestic goods and services.

The effects of a fiscal expansion financed withprivatization proceeds will, in many developingcountries, depend on the source of the proceeds,though this need not be the case with perfect capitalmobility.10 If the receipts are from abroad, the ef-

fects would be similar to a foreign-financed increasein the fiscal deficit and are likely to include pressuretoward the real appreciation of the currency. For do-mestic receipts, the impact on aggregate demandwould be similar to that of an expansion financedwith domestic bond placements. In this case, theneed for the buyer to finance the purchase could putupward pressure on interest rates, and the effect onaggregate demand may be more limited.11

The government may use privatization proceedsto reduce its net indebtedness by retiring external ordomestic debt, settling domestic or external arrears(if any), building up financial assets at home orabroad, or depositing the proceeds with the centralbank. These operations would generally be associ-ated with reductions in the country's risk premium.An automatic sterilization of foreign privatizationreceipts is achieved by parallel reductions in the netpublic external debt, including domestic currencydebt held by nonresidents and external arrears. Theexternal current account benefits from such opera-tions and the reduction of country risk through lowernet public interest payments. This, however, will beoffset, at least in part, by remittances of profits anddividends to foreign investors holding shares in pri-vatized firms.

In some countries, privatization proceeds may, inpart, be used to repay domestic debt, particularlywhen this debt is expensive and the external debt ismainly on concessional terms. Consideration needsto be given, however, to the effects on domestic li-quidity and interest rates of a net redemption of gov-ernment domestic debt, including foreign currencydebt held by residents and domestic arrears, underconditions of less-than-perfect capital mobility. Thiscould be expansionary, including through the associ-ated reduction in interest rates, and might imply theneed for offsetting monetary policy measures.

Privatization and Capital Inflows

Large-scale privatization programs have some-times been associated with capital inflows that havecreated complications for macroeconomic policy.Such inflows can pose challenges to financial poli-cies if the government spends them or repays do-mestic debt. There is a direct link between privatiza-tion and capital inflows when asset sales arefinanced through foreign direct investment.12 Addi-tional capital inflows may occur in the form of for-eign direct investment if the new owners of privat-

10If the source of the privatization receipts is domestic, the ex-cess demand for money resulting from the purchase of the assetby the private sector would be expected to lead to capital inflowsthat would be similar to an inflow of privatization proceeds fromabroad. However, this requires well-functioning domestic finan-cial markets.

11The actual impact on interest rates will depend on the subse-quent use of the proceeds and the degree of openness of the capi-tal market.

12Estimates based on World Bank data suggest that, on average,close to 40 percent of privatization receipts were paid in foreignexchange in the case study countries.


Box I. Sterilization of Privatization Revenues in Hungary

The Hungarian privatization process generated sig-nificant revenues in the years 1991-98, when gross pri-vatization proceeds averaged 4 percent of GDP a year,largely from abroad. In the early 1990s, Hungary fol-lowed a fixed, but frequently adjusted, exchange rate,and in early 1995 the government switched to a crawl-ing peg system. Under both regimes, the spending bythe government of part of the privatization revenues, aswell as capital flows associated with privatization,posed challenges to monetary policy. Capital inflowsled to increases in the money supply, thereby poten-tially compromising the objectives for inflation (to theextent that the liquidity injection exceeded the growthin base money demand), and the real exchange rate. Atvarious times, the authorities attempted to sterilize theeffect of capital flows on monetary aggregates.

The challenges for monetary policy posed by privat-ization-related capital inflows, and government deficitspartly financed with privatization receipts, first ap-peared in 1991, when significant foreign direct invest-ment was received, while a decline of real GDP bymore than 10 percent reduced money demand. Thiscreated pressure for the National Bank of Hungary toremove liquidity from the system through sterilization.In the next three years, as foreign direct investmentflows continued—most of them related to privatiza-tion—while output remained subdued and inflation re-mained high, some sterilization took place through theissuance of interest-bearing debt. At the same time, theNational Bank of Hungary used changes in reserve re-quirements to influence liquidity.

In 1995, in the aftermath of financial turmoil trig-gered by concerns over Hungary's capacity to serviceits foreign debt, the privatization process was speededup as part of a stabilization package. The new govern-ment made the reduction of the foreign debt a corner-stone of the economic strategy. Consequently, the bulk

of the privatization revenues from abroad was used di-rectly to reduce the foreign debt and, hence, did notneed to be sterilized. Of the US$3.8 billion in privatiza-tion revenue received in 1995 (out of a total of US$4.4billion in foreign direct investment), some US$3 billionwas used to repay foreign debt. In later years, a simi-larly high share of privatization revenues was used torepay foreign debt, and although the absolute amountsbecame more modest, privatization revenues allowedfor a substantial improvement in Hungary's foreigndebt position.

External debt reduction has important cost advan-tages as an instrument to offset the domestic impact ofprivatization receipts from abroad. Sterilization impliesa cost to the central bank proportional to the differentialbetween the interest rate on the debt issued and the re-turn on foreign assets. External debt reduction does notimply such costs, and it reduces the debt-related riskpremium. In the case of Hungary, however, the verysuccess of the privatization program, and the use of thereceipts for debt reduction, triggered other capital in-flows and the consequent need for sterilization.

In late 1995 and early 1996, the National Bank ofHungary intervened heavily in the foreign exchangemarket to keep the forint from appreciating, and in thefirst half of 1996 sterilization took place through re-verse repo operations and the sale of governmentbonds. As international confidence increased, and withsubstantial foreign investment from 1995 to mid-1998,the forint has typically been at the most appreciatededge of the band, requiring sustained foreign exchangepurchases by the National Bank of Hungary and there-fore sterilization operations. From March 1995 to end-1997, the liabilities of the National Bank of Hungaryarising from sterilization surged from Ft 18 billion toFt 681 billion (more than currency in the hands of thepublic).

ized enterprises decide to finance investment pro-grams with foreign capital, as illustrated by the caseof Hungary (see Box 1).13 Substantial privatizationprograms may also signal favorable changes in thepolicy regime and in growth prospects that triggerexogenous capital inflows not directly related to theprivatized enterprises. In Argentina, for example, therapid and large-scale privatization program of theearly 1990s and the use of part of the receipts to re-duce public debt conveyed a signal to markets aboutthe new policy regime that contributed to concertedflows.

13Foreign-financed investment programs in privatized enter-prises may contain a large import component—particularly if ob-solete technologies need to be modernized—and, to that extent,the inflows would be "sterilized" through a deterioration in theexternal current account.

Sustained capital inflows of the type discussed inthe previous paragraph will contribute to the appre-ciation of the real effective exchange rate. Under aflexible exchange rate regime, they would result inan appreciation of the exchange rate and a reduc-tion of inflationary pressures. Under a fixed ex-change rate, or managed float, and in the absenceof sterilization, such inflows would cause an ex-pansion of base money through the increase in for-eign exchange reserves, as well as inflationarypressures.

Several policy responses are available to deal withthe potential destabilizing effects of concerted pri-vatization-related capital inflows (Ariyoshi and oth-ers, 2000). Some sterilization may be necessary tosmooth out and limit the effects on liquidity; how-ever, depending on interest rate differentials, the


Table 4. Use of Privatization Proceeds in Case Study Countries

Country Use of Proceeds

Argentina Partly debt retirement, partly unearmarked budget financing.

Bolivia Off-budget part was fully earmarked for reinvestment in the privatized public enterprises; on-budget part was

unearmarked.1

Cote d'lvoire Unearmarked.

Czech Republic Of the proceeds received by the budget, part was used for debt reduction.2

Egypt During 1996-98, about one-fourth of gross proceeds went to severance pay and one-third to financialrestructuring of the enterprises. The remainder was used for debt reduction.

Estonia Initially to cover costs of currency reform (1992-93); then restitution, privatization costs, enterpriserestructuring, and several extrabudgetary funds.3

Hungary The on-budget share was partly used for debt retirement, while the off-budget share was used for restitution

•(compensation coupons), overhead of the privatization process, capital transfers to firms unrelated to the

privatization process, and financing of extrabudgetary funds.

Kazakhstan Mostly unearmarked.45

Mexico Proceeds kept off budget were used to finance a contingency fund for debt retirement.

Mongolia Unearmarked.4

Morocco Unearmarked.6

Mozambique Although most of the proceeds were used to settle the liabilities of the divested firms, in 1996 a special fund tosupport small-scale entrepreneurs using these resources was established.

Peru The on-budget share was used for social spending, while the off-budget part was used to recapitalize the public

pension system.7

Philippines Some proceeds have been earmarked.

Russia Unearmarked.4

Uganda Severance pay to workers retrenched in the process, settlement of debts and costs of the public enterprises

before sale, and promotion of entrepreneurship.

Ukraine Unearmarked.6

Vietnam The limited proceeds received so far have been reinvested in the divested firms.Source: IMF staff.

1Proceeds from equity sales in the capitalization scheme were effectively reinvested in the firms divested.2 Vouchers, including for restitution purposes, were used for most assets divested in the early period.3Vouchers were used for restitution.4Vouchers were used.5Some of the proceeds were earmarked for social spending.

6Privatization certificates were issued.7The remaining proceeds have been held in a special account by the Ministry of Finance as reserves.

quasi-fiscal costs of sterilization could be significant Use of Privatization Proceeds: Choicesand, in any case, must be taken fully into account and Experiencewhen formulating policies. Real effective exchangerate appreciation may take place, depending on the Countries have indicated their intentions to allo-exchange rate regime, through combinations of cate privatization proceeds to a wide variety of usesnominal exchange rate appreciation and increases in (see Table 4).14 Below are some of the fiscal factorsthe prices of nontradables, but the use of exchange to be taken into account when considering the possi-rate appreciation may be constrained by competitive ble uses of privatization receipts.considerations. Fiscal policy may need to be tight-ened, particularly to contain inflation and prevent anexcessive appreciation of the real effective exchange 14Assessment of whether Privatization receipts were actually

used in additive fashion tor the intended purpose requires consid-rate, and especially if the flows are large relative to eration of the counterfactual situation, for example, the amount ofthe absorptive capacity of the economy. spending in the absence of such receipts.

Ill USE OF PRIVATIZATION PROCEEDS


Fiscal Expansion Reduction of the Net Public Debt

A fiscal expansion financed through privatization The government may use privatization proceedsreceipts would generally tighten the intertemporal to reduce the net public debt by retiring debt or set-budget constraint. The more specific implications of tling arrears, or building up financial assets, includ-higher expenditure would, however, depend on the ing official reserves. This occurred in several of thetype of spending increases that might be case study countries, including Argentina, Egypt,considered.15 Hungary, Mexico, and Peru, with the debt retired

The temporary and uncertain nature of privatiza- being predominantly external. A reduction in the nettion receipts suggests caution in relying on them as a public debt permanently lowers the fiscal deficitsource of financing of increases in current expendi- through a decline in net interest payments and mayture. Such increases are often difficult to reverse at help reduce the exposure of the fiscal position tosome later point, and therefore there is a danger that changes in market sentiment. In addition, the im-spending may become entrenched at levels not con- provement in fiscal sustainability, signaled by thesistent with the revenue-raising capacity of the gov- saving of the proceeds, could contribute to fosteringernment. If increases in current spending are market confidence and lead to reductions in the ratenonetheless considered, their likely quality would of interest on the public debt. The choice of whetherneed to be carefully evaluated. to reduce gross indebtedness or build financial as-

In the short run, privatization can result in job sets, and how to effect it, are debt-management is-losses and wage cuts for workers and, in those cases sues that require appropriate consideration of rela-where price subsidies are removed, higher prices for tive yields; risk; the currency composition ofconsumers. When designing expenditure policies in budgetary flows, assets, and liabilities; and liquiditya setting of large-scale privatization, it is important preference.to address these concerns through policies to cush- Some countries, including Bolivia, Estonia, andion the short-term social impact of privatization (see Hungary, have explicitly linked the use of privatiza-Section IV). Such policies can also increase public tion proceeds to the transition costs associated with asupport for the reform process. pension reform. In effect, this recognizes the implicit

The fiscal stance may also be loosened through in- pension debt of the government and makes provi-creases in domestic public investment. Particularly sions for it with assets held by the state. If a govern-in countries with pressing infrastructure needs, a ment is also making efforts at maintaining or im-case could be made for implementing high-quality proving fiscal sustainability, the recognition ofinvestment programs. Such an investment policy pension debt could lend credibility and help garnercould aim at enhancing competitiveness by focusing political support for privatization.16

on the development of infrastructure and humancapital, thereby leading to higher growth. The impli- E a r m a r k i n g o f privatization Proceedscations of additional investment for recurrent gov-ernment spending would need to be taken into A number of countries have formally earmarkedaccount. privatization revenues (see Table 5). To some extent,

Government net worth would not decline as a re- the earmarking of privatization receipts may be seensult of capital spending if the expected return to the as an attempt to deal explicitly with the temporarygovernment on the new assets were comparable to nature of privatization receipts. Under certain cir-that on financial assets. However, the conditions for cumstances, such treatment of the receipts mightthis to hold are restrictive: even if public investment help avoid permanent increases in spending. An ele-were to have a high return, the government would ment of earmarking may also be put forward tostill need to capture the additional returns from the make privatization more politically acceptable, asinvestment for it to be self-sustaining (Fischer and was the case in the early stages of the Peruvian pri-Easterly, 1990). Just as important, experience shows vatization process. The impact of earmarking is,that in policy settings characterized by institutionalweaknesses and constraints there is a risk that publicsector projects will be poorly conceived and imple- 16In china, the implicit pension debt has been estimated at justmented, and that many low-return projects will end under 50 percent of current GDP, and settling it has complicatedup being incorporated in investment programs. e f f o r t s t o reform or divest public enterprises that have substantial

pension obligations to current workers or pensioners. The WorldBank has suggested that one method of financing these obliga-

15The impact of using privatization proceeds to reduce taxes tions could be to use the assets of the state enterprises themselves,would be comparable to that of raising current spending, though arguing that a portion of the implicit pension debt was accumu-clearly the distribution of the benefits would be different. If over lated as a way of financing the creation of these enterprise assets.time privatization leads to higher tax receipts, this might, how- This practice has already been implemented in a few Chinese lo-ever, provide an opportunity to reduce rates (see Section IV). calities (World Bank, 1997).


Table 5. Earmarking of Privatization Receipts in Selected Countries

Country Form of Earmarking

Bolivia Under the capitalization system, the proceeds remain in the privatized enterprises and are pledged forinvestment.

El Salvador Early in 1999, legislation was passed establishing a fund to manage the proceeds obtained from the privatization ofthe state telephone company, under which only the interest generated by these proceeds may be spent and onlyin the social sectors and for the development of information centers at the local level.

Estonia After settling debt obligations and tax arrears that were accepted by the successful bidders, some 45-50 percentof the remaining privatization proceeds were transferred to the Compensation Fund to cover its liabilities (bonds

I that were issued in exchange for privatization vouchers), while the remainder was earmarked to compensate

previous owners of real estate.Kazakhstan Some of the cash proceeds from the auction of small enterprises were earmarked for social safety net purposes.Peru During the early stages of the privatization process, the government earmarked part of the receipts from

privatization for social expenditure and structural reform programs designed in cooperation with the World Bankand the Inter-American Development Bank; inter alia, to foster public support for the privatization and reformprocess.

Uganda By law, privatization proceeds are to be used to settle enterprise debts, finance retrenchment programs, meet anyother costs of enterprises prior to divestiture, and promote entrepreneurship.

Vietnam Revenues from equitization may only be used for enterprise and labor force development.1

Source: IMF staff.1In Vietnam, equitization is a form of partial privatization of nonstrategic small- and medium-sized public enterprises. Although the mechanism was in-

troduced in 1992, the scope and depth of equitization to date remain limited.

however, uncertain insofar as budgetary resources and the proceeds are consumed immediately or overare fungible. a longer period. This has been advanced as a hypoth-

As a result of earmarking, budgetary resources, esis to explain privatization activity, with countrieswhose size ex ante may be quite difficult to predict, more likely to sell stakes in public enterprises if theyare placed outside the allocative budget process. face higher borrowing costs (Yarrow, 1999).17

Also, earmarking complicates fiscal management There have been recent cases where privatizationand makes it more difficult to reallocate spending in programs appear to have been accelerated—involv-response to changes in circumstances or priorities, ing in some cases high-quality assets—because ofpotentially forcing governments to devote more re- the perceived need to finance deficits and supportsources to an activity than they would otherwise do. current spending in the face of financing restrictions.Conversely, necessary spending may be constrained In Kazakhstan, the significant increase in the generalby shortfalls in privatization receipts. If, in addition, government deficit in recent years—to 8 percent ofprivatization revenues accrue to off-budget agencies GDP in 1998—was financed to an important extentand are partly disposed of outside the budget, these from privatization receipts, which rose to more thanproblems may be compounded by a reduction in fis- 4 percent of GDP that year, boosted by privatizationcal control and transparency. Therefore, in general, activity in the energy sector. Similarly, the fiscal pol-the earmarking of privatization revenues for specific icy stance in Lithuania was loosened considerably inuses is not an efficient spending allocation 1998. The fiscal deficit rose by 4 percentage pointsmechanism. to close to 6 percent of GDP, reflecting the largely

off-budget spending of a surge in privatization re-Privatization, Fiscal Pressure, and Liquidity ceipt equivalent to 5 Percentage Points of GDP, of

which the bulk was generated by the sale of theConstraints

telecommunications company.Fiscal pressures may lead a government to privat-

ize partly with a view to relaxing a financing con-straint so as to achieve a certain level of public 17However, s o m e analysts c i t ing e v i d e n c e m a i n l y f rom La t in

spending. Privatization in this case is Viewed as an- America, have argued that privatization has done little to alleviateOther source of revenue—it is regarded as a windfall, fiscal crises (Pinheiro and Schneider, 1995).

14

Ill USE OF PRIVATIZATION PROCEEDS


In cases where governments embarking on adjust- gaps while fiscal consolidation is effected and re-ment and reform programs initially face financing forms take hold, and until confidence is restored andconstraints, some limited use of privatization re- market access is reestablished. In the absence of anceipts may perform a useful role. They can provide effective program of fiscal consolidation, however,some financial cushion and fill short-term financing such a strategy will not be sustainable.


IV Fiscal Impact of Privatization

The fiscal impact of privatization will reflect the ings of the data, the findings on the fiscal effects of

amount and use of the proceeds and the subse- privatization need to be considered as preliminaryquent changes in financial flows—taxes, transfers, and interpreted with particular care.and dividends—to and from the budget.18 In abroader context, consideration should also be givento the impact of privatization on quasi-fiscal costs, C o n t e m p o r a n e o u s U s e ofincluding subsidized credit to public enterprises. Privatization ProceedsThere may also be impacts on the budget from theassumption of quasi-fiscal costs that were previously Privatization proceeds accruing to the budget mayimposed on public enterprises by the government be saved or spent, the former implying reductions inarising from the pursuit of public policies, such as domestic or external financing and the latter higheruncompensated provision of subsidized goods and spending or reduced taxes (see Section III). Theseservices. Privatization often takes place as part of a hypotheses were tested using the case study data andpolicy package that involves a substantial change in panel econometric techniques. To test the robustnessthe macroeconomic environment, which makes it of the findings, the effects on saving and spendingdifficult to distinguish its impact from the other ef- were tested independently and considered for differ-fects of a regime change. Moreover, data on public ent samples and coverage of explanatory variables.enterprise operations is often scanty, and the infor- The methodology and results are reported in Appen-mation available on the financial flows between en- dix II, with summary findings on the impact of pri-terprises and the government, including quasi-fiscal vatization on domestic financing presented inflows, is less than complete.19 Table 6.

This section examines the contemporaneous im- The results suggest that privatization proceedspact of privatization on the budget, and its effects transferred to the budget are saved. More specifi-over time, for the case study countries. To a substan- cally, they are consistent with these proceeds beingtial extent, the short-run impact depends on how the substantially used to substitute for other sources ofprivatization receipts are used, while long-run gains domestic financing; the coefficient on privatizationreflect the microeconomic benefits from exposing is always close to minus one and statistically signifi-the enterprises to market discipline. The empirical cant. There is some weaker evidence for the nontran-evidence draws on the panel econometric estimates sition economies that about one-fifth of the privat-reported in Appendix II and on a less formal review ization receipts are used to reduce externalof the case study experience.20 Given the shortcom- financing, with the rest substituting for domestic fi-

nancing.21 Regressions with the overall deficit, total18Given the limitations of data and the problem of specifying a expenditure, and taxation as dependent variables did

counterfactual, only a few studies have attemted to measure the not provide support for the suggestiont that proceedsthe firm level. For the most part these have found that govern- were generally Spent.ment net worth was increased by privatization, particularly when The interpretation of this econometric evidenceefficiency gains are factored in (Hachette and Luders, 1993, and needs to be carefully qualified, and it is difficult toWhite and Kelegama 1994). While studying the concept of social a s s e s s the general applicability o f t h e s e results. First,welfare, Galal and others (1994) found that in 11 of the 12 firmsconsidered, the impact on net wealth was positive. t h e r e s u l t s a r e b a s e d o n a s e l e c t s a m p l e o f countries

19The inadequacy of data on public enterprise operations in and for a limited number of years for which datamany countries has been documented in other studies by IMFstaff as hampering analysis and policy advice. See Schadler andothers (1995), and Bredenkamp and Schadler (1999). 21This reflects results from a regression with external financing

20Evidence based on comparisons between the situation before a s the dependent variable and is consistent with the size of the co-and after privatization is subject to the important qualification efficient in the regression for nontransition countries in the equa-that other events might affect the relevant variables. tion for domestic financing.

16©International Monetary Fund. Not for Redistribution

Table 6. Impact of Privatization on Domestic Financing

Explanatory Variables1

Budgetary Overall External Number ofRegression privatization2,3 balance2 financing2 R-squared Observations

Full sample -0.97 (.13) -0.74 (.15) — 0.54 83

Nontransition countries -0.79 (.12) -0.90 (.22) — 0.50 52

Full sample -1.19 (.19) — -0.65 (.20) 0.39 82

Nontransition countries -1.03 (.11) — -0.96 (.19) 0.58 52

Source: Table II.1Dependent variable is the first difference of the ratio of domestic financing to GDP.2Variables are significant at the 1 percent level; standard errors of estimates are in parentheses.3Privatization proceeds accruing to the budget.

were available. Second, the evidence applies to the public debt. The partial and piecemeal evidence isbudget narrowly defined, because the sample only consistent with the public sector in the case studycovers privatization proceeds identified as flowing countries benefiting from privatization in a longerto the budget; as discussed in Section III, there are time perspective.examples of countries subject to liquidity constraintsspending proceeds that accrue to extrabudgetary Tax Revenuefunds. Third, most of the countries in the sample hadan IMF program for at least part of the period when Taxes paid by privatized enterprises will reflectprivatization receipts were at their height. changes in efficiency and differences in the tax

If, however, the resources that are channeled regime. Increased profitability would benefit bud-through the budget tend to be saved, this would not getary revenue, as would any intensification of re-support arguments that the placement of the privat- strictions to competition that favor the new owners.ization proceeds in extrabudgetary funds is neces- The impact of differences in the tax regime when ansary to prevent the misuse of the resources. Further- enterprise is privatized is less certain. Revenue maymore, the results would be consistent with the increase to the extent that public enterprises havehypothesis that the formulation of fiscal plans under sometimes been subject to less rigorous auditing andIMF programs is made on the basis of a deficit con- collection efforts than private firms. However, a taxsistent with overall program objectives, and that pri- system may be more difficult to administer in anvatization receipts are not a determining factor in the economy characterized by many smaller privatesetting of fiscal targets, except at the margin (see firms than in one dominated by large state firmsSection VI). whose financial operations are, in principle, trans-

parently available to the tax authorities. Moreover,private firms may have stronger incentives to evade

Effects of Privatization on the Fiscal o r a v o i d t a x e s a n d may a l s o P r o v e m o r e skillful inAccounts Over Time doing so. In transition economies in particular, pri-

vate enterprises have often proved difficult forPrivatization may affect the fiscal accounts over nascent tax administrations to capture in the tax net

time in several ways: directly, through its effect on ( s e e Box 2).22

financial flows to and from the privatized enterprise; A t the microeconomic level, there is evidence thatindirectly, insofar as it influences the macroeco- privatized firms have paid higher taxes compared tonomic environment (see Section V); and as a result the preprivatization period (Galal and others, 1994).of decisions as to the initial use of the proceeds. This Evidence for some nontransition case study coun-section presents evidence on developments in taxrevenue, net transfers to public enterprises, more 22For a discussion of tax issues arising from privatization inbroadly defined public enterprise sector deficits, and transition economies, see Kodrzycki and Zolt (1994).


Box 2. The Impact of Economic Transition on Tax Administration

The Previous system. State ownership of the pro- Changes associated with transition to a marketductive sector significantly determined the basic prac- economy. Many transition countries have continued totices and procedures of tax administrations in the tran- apply effectively the same tax assessment and collectionsition countries. Under the previous system, the state procedures, despite the privatization of state-owned enter-was entitled to all of the profits from state-owned en- prises, development of a private sector, and the establish-terprises. Whether the state received this profit as tax ment of privately owned banks. The growing number ofrevenue or as dividend payment, the total resources small enterprises, often operating outside the formal econ-available to the state were the same. Moreover, because omy has increased the burden on local tax offices consid-both the tax assessor (auditor) and the state-owned en- erably. For example, in Bulgaria the number of small en-terprise's tax accountant were both effectively em- terprises registered by the tax administration hasployed by the state, their relationship was less adver- increased by about 100,000 each year between 1993 andsarial than that between a tax auditor and the 1 9 9 9 . In Moldova, the number of taxpayers increasedrepresentative of a private business in a market- from about 34 000 in 1991 to more than 350,000 in 1999,oriented economy. Ownership also allowed the state to with m a n y ° f t h e n e w r e g i s t r a n t s b e i n g s m a 1 1 - a n d

make discretionary ex post changes to an enterprise's medium-sized businesses.tax liabilities. In China, for example, tax authorities can In all of these countries, transtion has b e e n a

difficult challenge for tax administration. In the absencestill impose an adjustment tax on top of the enterpriseof a tax-paying culture and a willingness on the part of the

income tax. This tax is intended to be an equalizer that tax a d m i n i t r a t i o n t o m o d e m i z e procedures and systems,compensates for differential profitability, so its rates tax e v a s i o n a n d n o n c o m p l i a n c e are s e r i o u s problems. Sig-may vary depending on the particular circumstances of n i f i c a n t d o l l a r i z a t i o n , continuing barter arrangements,each enterprise. "cash-economy" business operations, and the nascent de-

Another feature of state ownership is that the number velopment of the banking system have all exacerbated taxof taxpayers is much smaller than in market-oriented compliance problems. A number of laws, such as account-economies. Most tax revenue is obtained from a limited i n g l a w s , are rudimentary, complicate the work of tax ad-number of large state-owned enterprises or collectively ministration, and increase the cost of compliance to tax-owned enterprises through turnover tax and profits tax. payers. Even where tax and other laws have been updated,State-owned banks played a major role in monitoring tax officials are often poorly trained. Also, the typicallythe tax payments of these state-owned enterprises; for low salary levels of officials in these countries may in-example, in some transition countries banks will still crease the risk of corruption. Finally, moribund and fail-not release funds to the state-owned enterprises to pay ing business enterprises have also created major problemswages until taxes due (including wage withholding) for tax administration, especially when their managershave been paid. give little, if any, priority to payment of taxes.

tries also suggests a similar conclusion.23 In vate sector. The econometric results offer some lim-Argentina, for example, taxes paid by five large ited evidence that privatization leads to a positiveprivatized firms increased significantly following and ongoing increase in tax revenue as a share ofprivatization (Shaikh and others, 1996; see also GDP in the nontransition countries (see AppendixLarrain and Winograd, 1996), while in Mexico pri- II). This increase could be due to several factors:vatized firms became significant tax contributors higher collection rates from the privatized firms, ei-after having received, on average, a small net trans- ther from improved compliance or enhanced admin-fer prior to privatization (La Porta and L6pez-de- istrative scrutiny; privatization leading to a shift inSilanes, 1997). World Bank studies for Cote d'Ivoire the structure of GDP toward sectors paying moreand Mozambique also provide evidence of substan- taxes; or the privatization process coinciding with atial increases in tax revenue from privatized firms. general strengthening in macroeconomic manage-

Overall government tax revenue benefits from the ment, including possible improvements in tax policylikely greater efficiency of the privatized firms and and administration. The fact that a majority of thethe concomitant impact on growth, but it may also sample observations coincide with the presence ofsuffer from the increased difficulty of taxing the pri- an IMF program may be of importance in the con-

text of the last factor.

23Several factors make it difficult to directly measure the extentto which privatized firms have actually paid higher taxes once di- Net Transfers to Public Enterprisesvested: firms often change form—merge, restructure, disap-pear—after privatization; and few tax services collect informa- I n m a n y countries, the public enterprise sector hastion specifically on taxes paid by previously state-owned firms. required substantial net resource flows from the bud-


Figure I. Gross Budgetary Transfers and Subsidies to Public Enterprisesfor Selected Countries(In percent of GDP)

19931984-86

H2 H H 1992-93 H

1985-89 1990-92 1987-91

i JH H _1990-92__H —M—JB —I B —0 • • • • • | | I ' " ' HI998 |Sources: Data provided by the authorities; and IMF staff estimates.1Excludes Petroleos Mexicanos. Includes some decentralized government agencies.

get over extended periods, suggesting that often the In the case study countries for which data arepolicy goals pursued through their operations are available, dividends paid to the budget by the publicbeing achieved at high fiscal cost. Viewed in this enterprise sector declined following privatization.way, privatization provides an opportunity to en- There is often concern as to the potential negativehance the efficiency of public expenditure. impact on the budget of the privatization of highly

For several case study countries, gross budgetary profitable public enterprises. Box 3 suggests, how-transfers to the public enterprise sector have tended ever, that such privatization need not adversely im-to decline with privatization (see Figure 1).24 The re- pact the budget.ductions have been particularly significant in coun- In many cases, the fiscal savings from privatiza-tries such as Argentina, the Czech Republic, Mexico, tion are larger than those arising from the elimina-Mongolia, Mozambique, and the Philippines. In ad- tion of budgetary transfers to public enterprises anddition, there may be quasi-fiscal support to public above-the-line quasi-fiscal support. This is particu-enterprises that is not captured in the budgetary data. larly likely to be the case if public enterprises recordPublic enterprises may run deficits after transfers fi- sustained deficits after such transfers, which are fi-nanced by various forms of nonmarket credit, obtain nanced through various forms of voluntary and in-"financing" in the form of insufficient investment voluntary credit.26 Figure 2 presents data for fourand the corresponding deterioration of the capital countries on the overall fiscal balance of the publicstock, and benefit from government-guaranteed lia- enterprise sector before transfers to and from thebilities and tax relief.25 government, the amounts of net transfers, and the

24Transfers made to divested enterprises would be classified as to 3 percent of GDP even though direct budgetary subsidies to en-transfers to the private sector in the budgetary accounts. There is terprises have been limited.a presumption that the scope of such transfers, if they continue to 26Public enterprise deficits may occur in the context of normalbe made at all, is generally likely to be limited. operations and need not be indicative of a cost to the government.25For example, in Uganda, the Ministry of Finance estimated However, chronic deficits after transfers likely imply a large ele-that the value of subsidies to public enterprises in 1996 amounted ment of quasi-fiscal costs.


Box 3. Fiscal Issues in the Privatization of Profitable Public Enterprises

Sometimes there are concerns about possible nega- Large dividends and transfers from profitabletive effects on the fiscal accounts from the privatization public enterprises prior to divestiture. Sometimes,of profitable public enterprises. In particular, the per- when the budget gets large revenues from certain pub-ception that it could result in revenue losses has some- lic enterprises in the form of dividends or transfers,times deterred more rapid progress in privatization. there is a perception that as these come to an end fol-The analysis of this case involves the consideration of lowing privatization, budget revenue will fall resultingseveral issues. in a deterioration of the fiscal position. However, this is

Profitability and efficiency. These are distinct con- a partial analysis focused solely on the budget that failscepts that are sometimes confused. If the private sector t 0 t ak e into account the financial position of the consol-can increase the efficiency of a profitable public enter- idated public sector as a whole.prise, and the government shares in these gains through One possibility is that in order to make large trans-the sale price and subsequent taxation, government net f e r s to t h e budget, the profitable public enterprise isworth will increase even though profitable enterprises h a v i n g t o b o r r o w o r i s decapitalizing because of insuf-are divested. ficient investment. These preprivatization factors,

Efficiency of privatization procedures. If prof- w h i c h affect t h e fiscal p o s i t i o n o f t h e p u b l i c sector as aitable public enterprises are sold for less than their true w h o l e , are n o t c a p t u r e d i f t h e a n a l y s i s f o c u s e s o n thevalue, then concerns about possible detrimental effects b u d g e t . F o l l o w i n g privatiZation, the "borrowing" hith-of privatization on the fiscal position would be war- erto d o n e by the public e n t e r p r i s e ( i n t h e form o f c r e d i t

ranted, but these considerations would apply equally to or d e c a p i t a l i z a t i o n ) w o u l d j u s t s h i f t t o t h e b u d g e t to

loss-making public enterprises. m a k e up for the l o s s in d i v i d e n d s , n e t o f t h e i n t e r e s t

Ability to tax. Profitable public enterprises may be athat will be earned on the financial assets generated by

source of sizable tax receipts to the budget, but the gov- . . .privatization revenue.

ernment may be unable to tax these enterprises efrec- privatization revenue.tively following privatization. However, if the sales A n o t h e r possibility is that a very profitable publicprocess is competitive, in principle this should be re- enterprise is able to finance large dividend and transferflected in the sale price, and the tax regime should have payments to the budget without having to borrow orno bearing on the impact of privatization on govern- decapitalize; in the absence of such transfers, it wouldment net worth. Potential buyers will incorporate infor- b e building up assets. In this case, under efficientmation about the tax regime in their offers, bidding up privatization mechanisms, the sale price should also re-the price of the enterprise up to the discounted value of flect t h e enterprise's profitability, and the loss of divi-net earnings after tax. In effect, the government would d e n d s a n d transfers to the budget should be made upget the net present value of the taxes forgone due to the through the return on financial assets generated by pri-inability to tax effectively in the lump-sum form of a vatization revenue.higher sale price, which in turn should contribute to a T a x a n d interest factors to take into account inlower net interest bill in the future. fiscal projections. The perception of a negative fiscal

Postprivatization investment and taxation. The effect from the privatization of profitable public enter-new buyers may need to undertake investments, which, prises could arise from the failure to include appropri-depending on the corporate income tax regime, may ate estimates of the taxes expected to be paid by theentail temporary effects on income tax revenue. How- privatized public enterprises in the fiscal projections, asever, if the public enterprise remained in public hands, well as the interest that will accrue on financial assetspresumably investments would also need to be under- purchased with privatization proceeds, or the reductiontaken at some stage to forestall decapitalization, with in interest due following debt amortization financedeffects on the overall public sector balance. with the proceeds.

balance that remained to be financed through vari- Public Debt and Interest Paymentsous forms of credit. These data are for the years be-fore the peak privatization period and for the latest A number of case study countries instituted ex-year for which data are available. For these coun- plicit o r i m P l i c i t rules requiring that a part of privat-tries, both the deficit before and after transfers de- i z a t l o n p r o c e e d s b e d i r e c t e d t o d e b t reduction. F ° u r

clined markedly following privatization, which case study countries that expressed an explicit inten-would suggest a lessening of the fiscal burden from tion t o u s e Privatization proceeds for debt reductionthese enterprises.27 (Argentina, Egypt, Hungary, and Mexico) had initial

stock of registered public debt ranging between 40

27The balance of the public enterprise sector as a whole also re- Also, it is not possible on the basis of the available data to includeflects any changes that may have taken place in the financial per- above-the-line quasi-fiscal costs, such as subsidized interest fromformance of public enterprises that remain in the public sector. the banking system, explicitly into the analysis.

•©International Monetary Fund. Not for Redistribution

Figure 2. Operations of the Public Enterprise Sector Before and AfterPrivatization for Selected Countries(In percent of GDP)

Overall Balance of Public Enterprises Before TransfersI

0 1998 1998 f 1997 •, _ • • • • • 1989-93

_l _— _ ^ g — _ _1990-94

1985-89

1984-88

_4 I I I I II Net Transfers from the Rest of the Nonfinancial Public Sector21984-88_ ^ g I 19981990-94 1997 1989-930 __••_ 1998 • • • •HflHi ••L^^^.1998_ | | | | | |Argentina Bolivia Mexico1 PeruOverall Balance of Public Enterprises After Transfers1997 19981998 • • • IB0 — _ _1998_ _ _ J M M M _ ^ ^ --«*» _ _ _ -S^BBW—19981984-88 1989-931985-89 1990-94_2 i I I I IArgentina Bolivia Mexico1 PeruSources: Data provided by the authorities; and IMF staff estimates.1Excludes Petroleos Mexicanos. Includes some decentralized government agencies.2Gross transfers less dividends.percent and 130 percent of GDP. In each of these In Argentina, Egypt, and Mexico, interest pay-countries, the debt stock fell sharply between the ments as a share of GDP fell significantly from theiryear before the period of most active privatization levels just prior to the initiation of their privatizationand the last year of active privatization, though programs. In general, where privatization proceedsclearly this involved many other factors in addition were used to reduce public indebtedness, privatiza-to the use of the privatization proceeds (see tion contributed to the strengthening and stabiliza-Table 7). tion of the economy. Thus, reductions in the interest©International Monetary Fund. Not for Redistribution

Table 7. Reduction in Debt Stock and Privatization(In percent of GDP)

Start of Active End of ActiveCountry Privatization Period Privatization Period Initial Debt Stock1 End-year Debt Stock2

Argentina 1990 1995 40.1 32.0

Egypt 1993 1998 129.4 89.6

Hungary 1991 1998 66.3 60.4

Mexico 1989 1994 64.4 33.3

Source: IMF staff estimates.1 Stock at end of previous year, except for Argentina, where 1990 data are used because of discrepancies in the 1989 data.2Stock at end of fiscal year of the active privatization period, except for Egypt, where debt data for 1997 are used.

burden are likely also to have reflected lower inter- assets they own because of data deficiencies andest rates arising from changes in the policy regime problems with adequate reporting. Governmentsassociated with privatization. In Argentina and Mex- should identify clearly in their accounts the revenuesico, significant declines in inflation rates during the collected from public enterprises, including taxesperiod of intensive privatization also resulted in paid and dividends transferred, as well as any cur-lower nominal interest rates on the domestic current or capital transfers made to the enterprises.rency debt. Quasi-fiscal support to enterprises should be bud-

getized. Efforts are also needed to improve data per-Data Issues taining to the public sector fiscal position, including

more adequate reporting—along with the budgetaryMany countries have experienced difficulties even accounts—of the operations and financing of

forming a clear picture of the value of the productive deficits or surpluses of the sector as a whole.

22


V Macroeconomic Impact of Privatization

This section examines evidence of the impact of to be more rapid where the share of the private sec-privatization on growth, aggregate investment, tor in GDP was higher,29 a number of the structural

and labor markets and unemployment. The micro- measures noted above tend to substitute for one an-economic evidence in each area is first summarized, other as predictors of growth. Similar results havethen followed by econometric results from the case been found in explaining growth in transitionstudy countries. The measured impact of privatiza- economies (Havrylyshyn, Izvorski, and van Rooden,tion on macroeconomic performance should be in- 1998).terpreted with caution, given the association of pri- The data for the case study countries support thesevatization with a broader regime change. findings inasmuch as they show a significant and

positive relationship between privatization andgrowth rates (see Appendix II). This relationship is

Effects on Growth and Investment more pronounced in the nontransition countries, butit holds for the full sample as well.30 Privatizationalone is not the suggested cause of the large in-

Economic Growth creases in growth rates shown in the regressions.There is substantial and growing microeconomic R a t h e r , it i s l i k e l y t h a t privatization is serving as a

literature that strongly supports the notion that pri- P r o x y i n t h e s e regressions for a range of structuralvate firms are operationally more efficient than those measures that may be characterized as a change inheld by the state. This conclusion holds for firms in economic regime. The results, however, are at leastcompetitive industries and for enterprises in less consistent with those on the microeconomic effi-competitive settings as well, although in the latter c i e n c y g a i n s associated with privatization.case the conclusions may be drawn less sharply(Megginson and Netter, 1999). A wide range of stud- Investmenties of firm-level performance in both developed anddeveloping countries supports this result (see Table T h e privatization literature suggests conflicting8), as does a recent survey of evidence for transition implications for the impact of privatization on in-economies (Havrylyshyn and McGettigan, 1999).28 vestment. Privatization should stimulate investment

The literature on growth in developing and transi- insofar as the management of public enterprises hastion countries suggests that policy variables—partic- b e e n associated with significant episodes of decapi-ularly fiscal discipline, price and trade liberalization, talization. The authorities, perhaps facing severe fi-deregulation, privatization, and the clarification and nancing constraints, may have elected to forgoprotection of property rights—are extremely impor- needed investment in public enterprises, effectivelytant in determining a country's growth performance. consuming a portion of the capital stock. PrivateAziz and Wescott (1997) argue, moreover, that there purchasers of such an enterprise would need to in-may be important policy complementarities among v e s t significant sums to modernize the firm, drivingthese measures: taken individually they may have up g r o s s investment in the postprivatization periodonly a limited effect on growth, while conjointly a s a r e s u l t . M o r e generally, and over time, a positivethey are strongly associated with rapid expansion ofeconomic activity. Consistent with this argument, 29Perotti and van Oijen (1999), moreover, present evidence thatSala-i-Martin (1997) finds that, while growth tended privatization serves as a positive signal for investors and that it

reduces investor uncertainty over the willingness of a country'sauthorities to pursue prudent macroeconomic policies and to cre-

28The emphasis placed on speed early in the transition, particu- ate a stable set of microeconomic incentives for investors.larly in the form of voucher privatization, may have come at a 30Havrylyshyn, Izvorski, and van Rooden (1998) discuss a va-cost in terms of weaker corporate governance and slower enter- riety of secular reasons why growth tended to drop sharply in theprise restructuring (Nellis, 1999, and Stiglitz, 1999). early years of transition.


Table 8. Summary of Three Studies of Firm-Level Efficiency Gains from Privatization1

Developed Countries Developing Countries Developed and Developing(1980s-early 1990s) (1980s-early 1990s) Countries (1990s)

Profitability (net income/sales)Mean before privatization 0.06 0.05 0.14Mean after privatization 0.08* 0.11* 0.17*Share of firms with improved performance 69 63 71

Efficiency (real sales per employee)2

Mean before privatization 0.96 0.92 1.02Mean after privatization 1.06* 1.17** 1.23*Share of firms with improved performance 86 80 79

Output (real sales)2

Mean before privatization 0.90 0.97 0.93Mean after privatization 1.14* 1.22* 2.70*Share of firms with improved performance 75 76 88

Leverage (total debt/total assets)Mean before privatization 0.66 0.55 0.29Mean after privatization 0.63** 0.50** 0.23*Share of firms with improved performance 72 63 67

Dividends (cash dividends/sales)Mean before privatization 0.01 0.03 0.02Mean after privatization 0.03* 0.05* 0.04*Share of firms with improved performance 90 76 79

Sources: Megginson, Nash, and van Randenborgh (1994); Boubakri and Cosset (1998); D'Souza and Megginson (1999); and Megginson and Netter

(1999).

1Statistical significance of the differences in means at the 1 percent and 5 percent levels is indicated by * and ** , respectively.2Real sales and real sales per employee are normalized at 1 in the year of privatization.

• - • - - " • • - - — » » - - - " - " - - - - ^ ^ !!M^:tim3:iiLaiiLaiit&^intfuinwii'iiii'^^ttv^mi^f M^.Mi^f^'^i'-^^t'imwii win .nniiiiniibiimtr;^1 "•' m • • ,»»• , •»• •»•»»»»•»•• •»• • •» ,»—B

impact of privatization on growth should be linked crease in investment of 126 percent in competitiveto an increase in investment. firms across 62 percent of the firms studied

Privatization could, however, lead to a reduction (Boubakri and Cosset, 1998).31

in investment to the extent that the authorities ini- At the macroeconomic level, however, no strongtially nationalized, or founded, public enterprises as relationship between privatization and investmenta means of stimulating investment in domestic pro- emerges in the case study countries (see Appendixductive capacity. Furthermore, if these enterprises II). This evidence would be consistent with the posi-were able to borrow at subsidized interest rates, ei- tive effect of privatization on growth being driventher explicitly or via an implicit government guaran- largely by efficiency gains.tee, their investment could exceed that of privatefirms, although this would not necessarily increaseoverall investment. Impact on Labor Markets

Megginson, Nash, and van Randenborgh (1994)found that firm-level capital expenditures, as a pro- State-owned firms that occupy noncompetitiveportion of sales, rose significantly (by an average 45 markets, or are protected through soft budget con-percent) in a sample of 61 privatized firms within a straints, may be overstaffed and pay excessiveset of 18 developed and developing countries. In-vestment increased in 67 percent of the firms theyStudied, with a much more significant impact on 31D'Souza and Megginson (1999) find a much weaker relation-firms in competitive, rather than noncompetitive, in- ship between privatization and investment for firms studies afterdustries. This test, when replicated for a set of devel- c a u s e m a n y o f t h e f i r m s divested in the later sample were utilitiesoping countries, showed an even larger average in- that were extremely capital intensive before privatization.


Box 4. Mitigating the Social Impact of Privatization

A recent study (Gupta, Schiller, and Ma, 1999) re- flexible labor markets, tend to produce strong growthviewed options for dealing with the social impact of in output and in employment. Public works programsprivatization. can directly create jobs, but these are usually of a tern-

Cushioning job losses. Employment guarantees fol- porary nature, and care must be taken to ensure that thelowing privatization can spread the adverse impact wages on offer do not discourage private sector jobover a longer period, so as to allow time for the job search.market to become more buoyant. The drawbacks of Providing income support. Severance pay, early re-such guarantees, however, are the likely lower sale tirement packages, preferential allocations of shares,price, which has implications for the government's and unemployment insurance are possible means ofability to fund other social spending, and the more support. Early retirement schemes, severance pay, andslowly realized efficiency gains, because restructuring preferential share allocations can, however, imposeby new owners is delayed. large costs, ultimately borne by the government either

Facilitating the transfer of labor to new uses. Ac- directly or indirectly (i.e., through a reduced saletive labor market policies can help reduce unemploy- price). Unemployment benefits presume the existencement duration and shift the skill mix toward occupa- of an effective unemployment insurance scheme,tions in demand. Specific policies include job which may not be the case, especially in transition andcounseling, job search assistance, assistance and train- developing economies. Moreover, coverage, the mini-ing for self-employment, and retraining. mum contribution period, and the duration and level of

Job creation. Sound macroeconomic and structural benefits must be chosen carefully to minimize disin-policies, which encourage a dynamic private sector and centives to employment search.

wages and benefits. For example, in India and fects of the privatization and regulation of utilities inTurkey in the early 1990s, overstaffing at state enter- Argentina suggested that privatization was not aprises was estimated as high as 35 percent, while in major contributor to the large rise in unemploymentAfrica, Asia, and Latin America in the 1980s, non- between 1993 and 1995 (Chisari, Estache, andwage benefits averaged 20-35 percent of the wage Romero, 1999).33

bill (Banerji and Sabot, 1994). Privatization of such Privatization, particularly when accompanied byenterprises can lead to large adjustments in employ- deregulation, can lead to enough new business gen-ment conditions. For instance, in four Mexican steel eration that the overall level of employment in theplants 50 percent of the original labor force was sector rises even if employment and wages in theeliminated during the process of privatization former state firm fall. In Zambia, for example, the(La Porta and L6pez-de-Silanes, 1997). In transition liquidation of the state airline and the bus firm led toeconomies, nonwage benefits, such as schools and two new airlines and several new bus firms, and inmedical care, may also need to be scaled back both cases sectoral employment rose (Kikeri, 1998).substantially. Even if aggregate employment increases in a sector,

At the level of the firm, a growing number of em- it is possible that this reflects new entrants into thatpirical studies suggest that privatization is not asso- labor force, with some previous employees from theciated with large-scale job losses. Megginson, Nash, privatized enterprises remaining unemployed.and van Randenborgh (1994) found that, for 61 The evidence from the case study countries sug-firms in 18 predominantly industrial countries, em- gests that privatization tends to be associated with aployment tended to increase after privatization. Em- reduction in both the contemporaneous and laggedployment levels rose in about 64 percent of the firms unemployment rate (see Appendix II). As suggestedthey studied, although in some of the cases in their earlier, it is likely that the strength of these resultssample employment had been substantially reduced stems from the combined effect of many policiesprior to privatization. In a sample of developing that are felicitous for growth and unemployment. Atcountries, Boubakri and Cosset (1998) found a simi- the same time, these results are consistent with thelar result, with about 60 percent of the firms in their microeconomic evidence and do not support con-studies experiencing an increase in employment fol-lowing privatization.32 A broader study of themacroeconomic, distributional, and employment ef- eluded a much higher proportion of firms in regulated industries,

which were therefore less competitive and more labor intensive,than did the earlier studies.

32D'Souza and Megginson (1999), however, found a signifi- 33This study also concluded that infrastructure privatizationcant decline in employment for 78 privatized firms in 25 coun- and effective regulation yield significant macroeconomic benefitstries (of which 10 were developing countries), in a sample that in- and gains for all income classes.


cerns as to general adverse effects of privatization and, for those whose skills are more specific or whoon employment, at least at the aggregate level. may be closer to retirement, possibly permanent ef-

Even if over time privatization does not have ad- fects. This lends importance to measures that miti-verse effects on sectoral or total employment, there gate the social impact of privatization, as discussedwill be workers who lose jobs on a temporary basis in Box 4 on page 25.


VI Issues for IMF-Supported Programs

Privatization is a reform area where the World under the SBA, EFF, SAF, and ESAF approved byBank has the lead role. The privatization compo- the Executive Board during 1994-98.

nent of IMF-supported programs has often drawnfrom and supported the strategy and measures con-tained in World Bank-supported reform packages. Privatization in IMF-SupportedIMF conditionality has reinforced public enterprise Programsreforms, including privatization, planned under theaegis of the World Bank, where these reforms are The incorporation of privatization and privatiz-important to the objectives of the program (Bre- ation-related measures in IMF-supported programsdenkamp and Schadler, 1999). IMF staff have relied has been flexible. This has reflected the underlyingon World Bank expertise for evaluating the appropri- diversity of approaches across countries in the designateness and feasibility of the measures and timeta- of divestiture programs and supporting measures.bles, as well as for monitoring the implementation of A large number of programs in recent years havespecific measures. included commitments associated with privatization

Reflecting the significant burden that public enter- (see Figure 3).35 Monitoring has relied primarily onprises have frequently imposed on the budget and structural benchmarks, prior actions, structural per-the economy, IMF-supported programs have placed formance criteria, and review clauses, as well asgrowing emphasis on privatization (World Bank, qualitative assessments of progress. The measures1995, and Bredenkamp and Schadler, 1999). This associated with privatization, which have been cov-trend has followed from the lack of success in hard- ered by IMF structural conditionality, can be classi-ening the budget constraint of public enterprises in a fied according to the various stages of the privatiza-number of developing and transition countries and tion process to which they apply.the modest results of attempts at restructuring large Structural conditionality associated with the pri-enterprises and improving their operational effi- vatization process has aimed at establishing theciency, including through performance contracts. As groundwork for privatization. This has often in-a result, there has been a growing recognition that volved measures such as the preparation, approval,privatization may be the only way to sever inappro- and announcement of privatization plans, the sub-priate financial links between public enterprises and mission or enactment of legislation to enable orthe government. In addition, privatization has been a facilitate privatization, and the setting up of institu-key element of structural change in IMF-supported tional structures and the strengthening of manage-programs with countries in transition from a cen- ment needed for the privatization process. It has alsotrally planned to a market-driven economy. involved the preparation and/or restructuring of en-

This section examines aspects of the privatization terprises prior to being put up for sale and actionscomponent of IMF-supported programs approved aimed at bringing specific enterprises or groups ofduring the period 1994-98 through a review of the enterprises to the point of sale.case study countries, as well as the IMF's MONA Conditionality attached to the pace of privatiza-database.34 The study will be limited to programs tion, as measured by privatization targets, has also

been used frequently in programs. Such targets havetaken several forms. Most frequently, they have been

34MONA (database for monitoring IMF arrangements) hasbeen compiled by the Policy Development and Review Depart-ment since 1993. It contains information on all IMF-supported 35The World Bank has supported privatization programs as partprograms under Stand-By (SBA) and extended arrangements of public enterprise reform strategies under structural and sectoral(Extended Fund Facility (EFF), Structural Adjustment Facility operations in many countries. Examples of this support are pro-(SAF), and Enhanced Structural Adjustment Facility (ESAF)) ap- vided at the World Bank website: http://www.worldbank.org. Seeproved since January 1, 1993. also Bredenkamp and Schadler (1999).


http://www.worldbank.org

Figure 3. Monitoring Privatization in IMF Arrangements Approved in1994-981

IMF-supported programs approved in 1994-98:132 programs in 81 countries

iOf which programs containing monitoring of

conditionality related to privatization:23

98 programs in 71 countries(74 percent of all programs)

Of which programs containing monitoring of conditionality related to the following:4

• |, ^ ii ^ ii ii iPrivatization Privatization Use of Privatization Postprivatization Other5

Process Targets Receipts Regulation

79 programs 74 programs 3 programs 8 programs 30 programs60 56 2 6 2381 || 76 || 3 II 8 || 31

Sources: MONA; and IMF staff estimates.1Types of IMF arrangements: Extended Fund Facility (EFF), Structural Adjustment Facility (SAF), Enhanced

Structural Adjustment Facility (ESAF), and Stand-By Arrangement (SBA). In 1999, the ESAF was replaced bythe Poverty Reduction and Growth Facility (PRGF).

2Excludes the privatization of financial institutions.3Monitoring in the form of structural performance criteria, prior actions, and structural benchmarks.4The first line of figures shows the number of programs containing monitoring of conditionality in each

category. The second line shows these figures as a percent of all IMF-supported programs approved in1994—98.The third line shows them as a percent of the IMF-supported programs approved in 1994—98 thatcontained monitoring of conditionality related to privatization.

5Mainly measures under mass privatization in transition economies, and the privatization of land, housing,and real estate.

quantitative—including general targets that involve ations vis-a-vis potential buyers. To guard againsta given number of public enterprises to be divested such an eventuality, some programs attach condi-by certain dates, and enterprise-specific targets that tionality to the bringing of public enterprises to theinvolve certain enterprises or groups of enterprises. point of sale, or to particular elements of the processTargets for fiscal receipts have also been used, albeit short of final sale. In such cases, it is important tomuch less frequently. ensure, in consultation with the World Bank, that

The widespread conditionality on privatization any reservation prices set by the authorities are real-targets reflects the fundamental objective that pri- istic and, if possible, supported by independentvatization should actually take place and the often valuations.substantial drain of public enterprises on the econ- In contrast to the process and target stages in theomy and the budget. Moreover, privatization targets privatization process, conditionality in the form ofare more specific and more readily monitored than structural performance criteria, prior actions, orother aspects of the divestiture process. The need to structural benchmarks has been attached much lessobserve announced deadlines for specific privatiza- frequently to other aspects of the privatizationtions when structural performance criteria or prior process. Conditionality on the specific use of privat-actions are attached to the actual sales has the merit ization proceeds has been applied only in a fewof emphasizing the importance attached to the sales cases. Program ceilings, however, may be estab-in the program. However, this can also undermine lished so as to limit in effect the use of proceeds. In-the government's negotiating position in these oper- deed, restrictions on the use of privatization pro-


ceeds and on the spending of proceeds in excess of on process and regulation should also seek to en-programmed amounts have been included in a large hance transparency and promote good governance innumber of programs through the use of quantitative privatization procedures.performance criteria and adjusters (see below). Con- Conditionality on actual privatization will oftenditionality on postprivatization regulation through remain important. This might, in some cases, be bet-structural performance criteria, prior actions, and ter placed in the form of structural benchmarksstructural benchmarks has not been used often. rather than structural performance criteria and prior

Privatization conditionality in IMF-supported pro- actions to limit adverse bargaining incentives. Quali-grams should continue to reinforce the divestiture tative assessments, with input from the World Bank,strategies supported by the World Bank. Within the would then be required where there are delays inframework of Bank-IMF collaboration, the IMF specific privatizations.should focus on those aspects of policies—the rela-tionship between privatization and macroeconomicpolicies—that are in its area of responsibility. In this Privatization and Program Designregard, IMF staff should help guide reform prioritieson the basis of the financial impact of public enter- The determination of fiscal performance criteriaprises, as well as monitor budget constraints.36 should reflect the macroeconomic impact of privat-

There is room for somewhat greater selectivity in ization proceeds and their use. From the viewpointthe use of structural performance criteria, prior ac- of fiscal control and transparency, it is important totions, and structural benchmarks to strengthen their include as broad a definition of privatization receiptscontribution to the success of reform. The focus, as possible in the fiscal targets and quantitative per-while supportive of World Bank privatization strate- formance criteria established in the program. In par-gies, should be on privatization measures that have a ticular, programs should give due consideration tosignificant fiscal and macroeconomic impact and the extrabudgetary activities that may be associatedthat are deemed critical to the achievement of the with privatization, and these should be consolidatedobjectives of the financial program. in the fiscal accounts. Moreover, privatization re-

The World Bank's last formal review of its assis- ceipts should be classified as financing in the fiscaltance to privatization in developing countries em- programs included in IMF arrangements. The designphasized the importance of focusing on the entire of programs should also take into account the ex-privatization process, rather than mainly on the num- pected sources of the privatization proceeds and theber of privatized enterprises. World Bank condition- degree of capital mobility, particularly if the privat-ality for tranche release has subsequently been ization program is large relative to income and mon-linked more strongly to the process aspects of pri- etary aggregates. As these issues were discussedvatization, as well as to setting up appropriate legal substantially in Sections II and III, this section fo-frameworks and regulatory institutions for the post- cuses on the more specific issue of the combinationsprivatization period. of fiscal performance criteria and automatic ad-

There is scope for IMF conditionality to selec- justers that indicate how fiscal ceilings shouldtively reinforce this emphasis on process and regula- change when, as commonly occurs, privatization re-tion where measures in these areas are considered ceipts deviate from programmed levels.central to the financial program.37 This might in- The effect on the program of deviations from pro-clude consideration of institutional processes; opera- grammed levels of privatization receipts differs ac-tional mechanisms for asset sales; legal and proce- cording to whether privatization proceeds are classi-dural changes; and the establishment of regulatory fied as revenue or financing and the nature offrameworks within which privatized firms, espe- performance criteria and adjusters. In practice, adaily utilities, will operate. An increased emphasis wide variety of combinations of performance criteria

and adjusters associated with deviations from pro-36Bredenkamp and Schadier (1999) discuss cases where the j e c t e d receipts resulting from privatization has been

key aspects of public enterprise reform are not covered by ongo- used in programs with the case study countries (seeing World Bank activities. In those cases, it is recommended that Table 9). Depending on the particular program, pri-me IMF take an active role in advising the authorities on those as- vatization-related adjusters (symmetric or asymmet-pects of policies toward public enterprises that are particularlyimportant to the financial program, in consultation with World ric) h a v e b e e n attached to One or several quantitativeBank staff, until the World Bank can be more actively engaged. fiscal performance criteria and to the net interna-

37In Uganda, for example, the government recently revised its tional reserves and net domestic assets of the centralprivatization strategy in consultation with the World Bank, and bank or of the banking system. In some cases, theconditionality in the IMF program switched from structural applicability of these adjusters has been linked to thebenchmarks attached to quantitative privatization targets to prior

actions on processes designed to bring a certain number of enter- origin (foreign or domestic) of privatization pro-prises to the point of sale. ceeds or to their currency composition.



Most programs were designed with the aim of pre- The need to save excess privatization receipts,venting the use of excess privatization proceeds over seen frequently in programs, is based on the princi-the program baseline. This was achieved in a variety ple that the fiscal program has been designed withof ways. The fiscal deficit was capped where privat- certain macroeconomic objectives in mind and thatization receipts were classified as financing (Ar- additional nonrecurrent financial resources shouldgentina, Cote d'Ivoire, Egypt, Kazakhstan (since be used in a manner consistent with those objectives,1998), the Philippines, Russia, and Ukraine), while which typically would entail saving them. This ap-in Peru the fiscal performance criterion was set on proach is generally consistent with the discussion innet domestic financing (including privatization re- Section III of the likely impact of spending addi-ceipts). In other programs, where privatization re- tional privatization receipts on government netceipts were classified as revenue, adjusters limited worth and aggregate demand.the spending of excess privatization receipts (Bo- Choices regarding program design in the presencelivia, Mongolia, and Mozambique), or fiscal perfor- of projected privatization receipts should be mademance criteria were specified excluding these re- on the basis of the fiscal and macroeconomic cir-ceipts (Hungary and Mexico).38 However, where cumstances of the country, the objectives of the pro-privatization receipts were classified as revenue, the gram, and the availability and timeliness of data.absence of adjusters to the fiscal performance crite- Judgments about types of risks to the programria implied that excess receipts could be spent by the should also influence choices. Particular care needsgovernment (Kazakhstan (1996-97)). to be exercised to ensure consistency between fiscal

Programs differed somewhat more in their treat- and other performance criteria. For instance, if thement of shortfalls in privatization receipts relative to government must save excess privatization re-the program baseline. In some programs, such short- ceipts—to the extent that this is achieved throughfalls implied the need for fiscal adjustment (Ar- additional deposits at the central bank—net interna-gentina, Bolivia, Hungary, Kazakhstan (1996-97), tional reserves and net domestic assets of the centralMongolia, Mozambique, and the Philippines), while bank should also have privatization-related ad-in others privatization shortfalls could be financed, justers. Otherwise, government deposits of the ex-though usually up to certain caps (Cote d'Ivoire, cess receipts with the central bank would createKazakhstan (since 1998), Peru, Russia, and room for credit to the private sector to expand be-Ukraine).39 yond the original program targets.

38Privatization-related adjusters were attached to the net inter-national reserves and net domestic assets of the central bank inthe case of excess privatization receipts in some of these pro-grams (Bolivia, Hungary, Mexico, and Peru). In Hungary, the ad-justers applied only to the extent that cash privatization receiptsoriginated abroad, while in Mexico the adjusters were triggered 39In Egypt and Mexico, the program baselines assumed no pri-only if the receipts were made in foreign currency. vatization receipts.


Appendix I Factors Affecting the Sale Priceof State Assets

This appendix discusses elements of the privatiza-tion process that have a bearing on the sale price

of assets and consequently on the resources thatmight be available to the budget. The material drawson the World Bank's experience and recommenda-tions regarding privatization procedures (WorldBank, 1995, and Lieberman and Kirkness, 1998).The factors that influence the sale price of assetsmay be classified into three broad categories accord-ing to the stage of the privatization process in whichthey may arise: actions taken prior to the sale of theasset, the sale process itself, and the postprivatiza-tion regime (see Table 10). Box 5 discusses relatedgovernance issues that may occur at each of thesestages of the privatization process.

Actions Prior to Sale

The preparatory phase prior to privatization mayinclude the restructuring of public enterprises. Thismay involve legal restructuring (e.g., clarification oftitles and legal changes that may be required to per-mit private investors to acquire shares in a public en-terprise); organizational restructuring (demonopo-lization or breakup); financial restructuring (e.g.,assumption of the enterprise's debts by the govern-ment and cleanup of balance sheets); operational re-structuring (e.g., increased investment to revamp thecapital stock or closure of certain activities); andlabor restructuring.

Legal, organizational, and financial restructuringmay in some cases be essential to ensure that privat-ization takes place or to forestall the emergence ofprivate monopolies. For example, the clarification ofenterprise debts and legal obligations includinglabor issues will increase the chance of sale. Thebreakup of very large enterprises is sometimesneeded to promote greater competition. In a numberof transition economies, divestiture of social over-head functions, such as schools, clinics, and hous-ing, was required before sale. Whether public enter-prises are sold with or without their liabilities neednot change government net worth. However, thiscould have implications for the composition and ma-

turity profile of the public debt and for the cash flowgenerated by the privatization and, therefore, forgovernment liquidity.

Decisions on whether to operationally restructurepublic enterprises prior to privatization need to takeinto account the likely effect on the liquidity posi-tion of the government and on government networth. The private sector may well be able to re-structure public enterprises more efficiently than thegovernment. Thus, preprivatization restructuring,while conceivably raising the sale price, may actu-ally reduce government net worth.40 Hence, addi-tional investments to undertake the physical restruc-turing of public enterprises should, in general, be leftto private owners once a decision has been made toprivatize the enterprise (World Bank, 1995, andLieberman and Kirkness, 1998). To the extent that agovernment decides to restructure firms before sale,it is important to ensure that the funds invested aresubject to the same standard as for the use of anyother public funds.

The issue of whether labor restructuring should becarried out by the government or left to the newowners involves difficult tradeoffs. While cases in-volving the need for limited restructuring are bestleft to the private sector, governments may considerhandling large-scale redundancies prior to sale to re-duce labor resistance and enhance the likelihood thata social safety net will be provided, and possibly in-crease the sale value. Government restructuring,however, is likely to entail fiscal costs because theremay be a tendency for the state to be generous in theterms offered to the labor force to make restructur-ing politically and socially more acceptable (Kikeri,1998).

The Sale Process

The state, as seller of an asset, is most likely tomaximize the sale price when there are no restric-

40There is evidence that operational restructuring of public en-terprises prior to privatization in Mexico was not reflected in thenet sale price (L6pez-de-Silanes, 1997).


Table 10. Factors Affecting the Sale Price: Country Illustrations

Actions prior to saleLegal restructuring

EgyptRussia

UkraineOrganizational restructuring

Argentina, BoliviaTransition countries (various)

Financial restructuringArgentinaEgypt

Operational and labor restructuringArgentina1

Mexico, MozambiquePeru

The sale processEgypt, Mongolia, Russia, UkraineKazakhstanMongolia

The postprivatization regimeRegulation

ArgentinaBoliviaCote d'lvoireCzech Republic

Postprivatization commitmentsArgentina, Bolivia, Hungary, PeruHungaryMongoliaVietnam

Legislation passed to authorize the privatization of public insurance companies.Presidential decree issued reducing the list of "strategic" joint-stock companies thatwere banned from privatization.Parliamentary approval required to privatize certain monopolies.

Breaking up of some public utilities into smaller units.Divestiture of social overhead functions.

Treasury assumed debt of enterprises to be privatized.About one-third of the proceeds of privatization in FY 1996/97 and FY 1997/98 wasdevoted to financial restructuring costs.

Rationalization of employment of the state oil company.Substantial investment and modernization drives.Reduction of payroll of public enterprises and tight control of wages.

Limitations on the legal or effective participation of nonresidents.High auction floor prices for small enterprises (eliminated 1996).Legal restrictions established high reservation prices (abolished 1997).

The telecom monopoly was largely preserved.Entry to provision of long-distance and international services prohibited.Privatized telecom company granted a seven-year monopoly on some services.A 27 percent stake in SPT Telekom was sold to a consortium. SPT Telekom willmaintain its monopoly status until end-2000.

Requirements on the amount of investment to be undertaken by the buyers.Purchasers required to take responsibility for environmental costs.New owners required to maintain staffing levels (lifted in 1999).Involuntary layoffs not permitted in first year after privatization.

Sources: Azpiazu and Vispo (1994), Lopez-de-Silanes (1997); IMF staff.

1In the initial stages of the privatization process, the government placed emphasis on the speed of divestiture, and operational and labor restructuring

was limited.

tions on the number of potential bidders. This wouldinvolve permitting nonresidents to participate onequal terms with residents, particularly becausequite often in developing and transition countries thenumber of potential domestic purchasers, or the fi-nancing available to them, is limited. However,countries have sometimes placed limitations on theparticipation of nonresidents in the privatizationprocess (see Table 10). In some cases, nonresidentshave not been allowed to bid, or have been allowedto do so only for minority stakes in firms, while inothers equity participation by foreigners has been re-stricted to "nonstrategic" firms. Yet "strategic" hasoften been defined fairly broadly, effectively servingto limit the set of potential buyers to domestic resi-dents. Whether nonresidents are allowed to bid or

not, the process should be sufficiently open andnondiscriminatory so as to prevent noncompetitivearrangements, as occurred, for example, in the Russ-ian loans-for-shares scheme (see Box 6).

Governments may sometimes privatize throughmechanisms, such as retail offerings that may begenerously priced, aimed at generating strong de-mand from domestic retail investors, in order to fos-ter domestic participation in, and support for, the pri-vatization process; create a new class of savers; andcontribute to the development of the domestic capi-tal market. In some cases, this has also involved giv-ing preferential subscription rights to employees andother stakeholders of the public enterprises to be pri-vatized. The pursuit of these wider objectives entailsfiscal costs that need to be considered; it should also


Box 5. Governance Issues in PrivatizationPrivatization has been invoked as a means to combat

corruption. However, the privatization process itselfhas often proved to be a significant source of corrup-tion. Corrupt systems have tended to engender corruptprivatization processes and, unless accompanied bysignificant institutional reform, transparency has usu-ally not benefited from privatization.

Governance issues in public ownership. Corrup-tion, defined as the abuse of public power for privatebenefit, has been argued to increase with the extent ofgovernment intervention in the economy, the degree ofdiscretion of government officials in applying regula-tions, the weakness of institutions and rule of law, andpoor and unsystematic accountability of public officials(Rose-Ackerman, 1997). Public enterprises have beena major source of corrupt activities in many developingand transition countries (Tanzi, 1998). Such activitieshave included overinvoicing, accepting bribes for con-tracts, rationing below-market-priced goods and ser-vices through bribes, and awarding workers and man-agers excessively high wages and fringe benefits.

Privatization and corruption. Countries with a tra-dition of strong institutions, rule of law, and judicial ac-countability, as in western Europe, have generally en-gendered transparent privatization processes. Incountries where privatization has been part of a com-prehensive change in regime, institution building, andreorientation of the economy toward the market, suchas a number of Latin American countries in the 1990s,the change in the rules of the game has, on the whole,been credible and the privatization process reasonablytransparent. However, corrupt practices associated withprivatization have been reported, particularly wherethere has been limited oversight from other branches ofgovernment (Manzetti, 1998). Where the institutionalframework and rule of law have been weak, strong andwell-organized interest groups have tended to "hijack"the privatization process to their advantage. This hasoccurred in several of the transition economies, per-haps most notoriously in the case of Russia (Aslund,1999). The proposition that it does not matter muchwho gets the assets during privatization because themarket soon reallocates them to efficient owners does

not seem to work in practice, partly because of the lackof capital market development in these countries.

Privatization and regulation. Since the early1990s, a growing number of countries have started toprivatize "strategic" or "core" public enterprises, suchas utilities, telecommunications, transport, and energyenterprises. Problems have arisen, however, where is-sues related to the structure of postprivatization mar-kets and the creation of sound regulatory regimes havenot been adequately addressed prior to privatization.The major problems fall broadly into the following cat-egories: inadequate competition, or the preservation ofmonopolies with no economic justification and insuffi-cient attention to antitrust issues both before and afterprivatization; poorly designed regulations, or ambigui-ties in tariff-setting mechanisms and in the regulatoryframework; and weak regulatory institutions, or thefailure to prevent abuse of market power by dominantenterprises, to foster competition through the entry ofnew operators, and to create a favorable investmentclimate.

Possible lessons. First, privatization is no alternativeto regulation, nor has it proven to be an impediment torent seeking. Second, the policy of "privatizing nowand regulating later" has often failed because early pri-vatization has created strong vested interests to blockthe later attempts at regulation. Third, corruption sig-nificantly endangers the legitimacy of the privatizationprocess and, more generally, weakens support formarket-oriented reforms.

Therefore, privatization would be most efficient if itwere preceded by institution building and the establish-ment of an appropriate regulatory framework and therule of law. If the institutional underpinnings are miss-ing but the government is making progress toward es-tablishing them, delaying privatization until they bearfruit may be a desirable strategy. In those difficult butfrequent cases where the government is unwilling or in-capable of taking the necessary prior steps, the bestcourse of action may be a cautious case-by-case, tender-based privatization with the assistance of internation-ally recognized financial advisors (Nellis, 1999).

be consistent with the key objective of efficiencygains.41

The choice of sales method, for example, auc-tions, initial public offerings, and trade sales tostrategic investors, and its transparency can have im-portant implications for the number of bidders andfor the sale price. In general, auctions and initial

41 The issue of mobilization of political support for privatiza-tion and the role of domestic investors have been, and continue tobe, researched by the World Bank. See, for example, World Bank(1995), Lieberman and Kirkness (1998), and Shafik (1996).

public offerings have served to generate higher re-turns than bilateral trade sales (Berg and Berg,1997). Initial public offerings are the more commonsales mechanism in industrialized countries, whereastrade sales have been more frequently used in devel-oping countries. However, the latter are less trans-parent and less open to public scrutiny than morecompetitive processes. Moreover, they are morelikely to involve significant elements of restructur-ing at the expense of the state as seller. On balance,therefore, it would seem preferable to rely on openauction mechanisms subject to public oversight


Box 6. Loans-for-Shares Privatization in Russia

Following a mass privatization campaign during1995, significant share holdings in some of Russia'slargest companies—Norilsk Nickel, Yukos (oil),LUKoil, Surgutneftegaz (oil), Novolipetsk Iron andSteel, and Novorossiisk Shipping—were assigned tocommercial banks as collateral for a loan to the federalgovernment. The banks had to compete for the right tomanage these assets by bidding at auction on the size ofa loan to the government, but in practice only a fewbanks won: Uneximbank, Menatep, Stolichny, and Im-perial, all of whom had close relations to the govern-ment. In most cases, the banks bid for the assets as partof a consortium, of which there were two, with thegroup that was politically less well connected winningnone of the auctions. Foreigners were technically al-lowed to bid, but in practice the auctions were often setso as to ensure that the favored banks won.1 The winnerhad the right to run the company until it would be sold,

1In the case of Surgutneftegaz, the auction was announcedfor the distant city of Surgut, where the airport was closed fortwo days before the auction, purportedly for weather-relatedreasons. Similarly, a large stake in the Norilsk Nickel con-glomerate was won by a subsidiary of Uneximbank, which

again at auction, though in practice the banks them-selves were generally able to acquire the assets theywere managing through insider means. Six of thetwelve companies were bought by the banks, whichacted as auction managers, and four more were won bycorporate affiliates of the firms being sold. Many of thebanks that won the loans-for-shares auctions were po-litically active in supporting the government throughthe election campaigns of 1995 and 1996, while thefirms that they acquired—particularly those in the en-ergy sector—soon came to head the list of Russia'slargest tax debtors.

A number of large stakes in Russian industry, partic-ularly in the oil sector, were sold at prices that seembelow their market value and in auctions that were nottruly competitive. The perception that the loans-for-shares program was corrupt has weakened popular sup-port for the privatization program overall, as well as forother market-oriented reforms.

also managed the auction, and which, on a technicality, hadexcluded a key competitor offering twice the winning bid(Lieberman and Veimetra, 1996).

whenever possible. Also, the reservation prices forassets should not be set so high that privatization iseffectively foreclosed.42

A number of transition countries attempted to ac-celerate the privatization process and overcome theshortage of domestic capital by mass privatization.This mechanism involved the use of one or anotherform of vouchers in lieu of money. Assets were di-vested at little or no cost to the population, some orall of whom received a form of privatization moneywith which to bid on the assets for sale(Havrylyshyn and McGettigan, 1999). This form ofmass privatization did not generate significant cashreceipts.

The Postprivatization Regime

The key importance of setting up appropriate reg-ulatory frameworks and institutions prior to the pri-vatization of public enterprises with substantial mo-

42The government may retain some "golden" share, that is, ashare with special voting rights, in privatized firms—often utili-ties and natural monopolies. If it does so, this may lead to similarmanagerial and governance issues as under privatization (WorldBank, 1995). However, the government's aim may be to retain thepower to veto certain actions, such as takeovers, that could havethe effect of asset stripping (Megginson and Netter, 1999). Fromthe investors' perspective, the new owners may also want thegovernment to keep a share as a form of insurance against ad hocregulation.

nopoly power, including enterprises with network ornatural monopoly characteristics, is widely recog-nized (World Bank, 1995). Expectations as to the ex-tent of such postprivatization regulation can be im-portant in determining the sale price. Firms withmonopoly power that are likely to be regulated onlylightly should sell for a better price than those thatwill be more heavily regulated. However, artificiallyinflating the sale price by precommitting to a laxregulatory regime would lead to a higher price atsale at the expense of potentially large efficiencycosts in the rest of the economy and lower social re-turns over time. Moreover, countries have some-times found it difficult to implement adequate regu-latory restrictions not put in place before the firmswere sold.

There is evidence in some case study countries ofweak regulatory frameworks and the granting of mo-nopoly rights in the telecommunications sector (seeTable 10). These actions might have been aimed inpart at enhancing privatization proceeds.

The placement of requirements on the new ownersat the time of sale in the form of regulatory commit-ments, particularly in the case of public utilities andmonopolies (for example, minimum operating stan-dards, service levels, safety requirements, and mea-sures to foster competition), is often warranted.Governments have sometimes placed additional re-quirements on buyers related to investment condi-tions, environmental clean up, purchase from certainvendors, and labor hoarding. Such additional re-


quirements may reduce the sale price and couldmake the process less transparent. Generally, theymay not be a cost-effective means of pursuing publicpolicy goals.

An important factor affecting the amount of pro-ceeds is the terms of payment, which may involve anextended settlement period and raise liquidity andrisk management issues. In some cases, governmentshave not obtained payment for assets privatized in atimely fashion from buyers. This may be related to an

inadequate degree of risk transfer from the govern-ment to the private sector because of the provision ofgenerous credit terms. For example, in Ugandaamendments to the Public Enterprise Reform and Di-vestiture statute introduced in 1997 sought to restrictextended terms of payment for enterprises privatizedbecause of problems of nonpayment. Failure to im-pose discipline in this area, besides raising gover-nance concerns, affects the effective price eventuallyreceived for the assets sold.


Appendix II Privatization and Fiscal andMacroeconomic Developments

This appendix uses data from the case study coun-tries to investigate the empirical relationship

between privatization and various fiscal and macro-economic variables.43 The issues examined arewhether privatization proceeds transferred to thebudget are used to finance a larger deficit (spent) orto reduce other sources of financing (saved), andwhether the total amount of privatization receipts iscorrelated with changes in macroeconomic or fiscalperformance.

The sample is comprised of annual data from the18 case study countries, using the period of activeprivatization for which the necessary data are avail-able. The sample, therefore, varies between regres-sions due to data availability. All variables are ex-pressed as a percent of GDP, with the exception ofreal GDP growth and the unemployment rate. Unlessotherwise noted, the data are from the correspondingcountry authorities and staff estimates; the unem-ployment rate is taken from the IMF World Eco-nomic Outlook database. Finally, the country dataare pooled to form the unbalanced panels that areused in the regressions.

Proceeds Transferred to the Budget

The contemporaneous relationship between privat-ization proceeds transferred to the budget and dif-ferent fiscal variables is examined using regressionsof the following form:

Ay,,, = 111 + 8Apu + pAxu + uitU (1)

where A is the first difference operator, yt>t, pi>t, Xjtt,and ui,t are, respectively, the dependent variable, pri-vatization proceeds, an additional explanatory vari-able (if included), and the residual; subscripts referto the value for country i in period t. The parametersto be estimated are jui,, which is the country-specificor fixed effect, S, and /3. The hypothesis that privat-ization proceeds transferred to the budget are spentis tested by examining the statistical significance of

<5 using the overall balance, total expenditure and netlending, and total revenue as the dependent vari-ables. The saving hypothesis is tested using domes-tic financing, external financing, and governmentdebt as the dependent variables.

The empirical results are consistent with privat-ization proceeds being saved, specifically that theysubstitute one-for-one with domestic financing (seeTable 11). This conclusion is robust and is not funda-mentally altered by changing the sample or addingexplanatory variables. Moreover, it is supported bythe findings that privatization proceeds are not usedto increase the deficit, increase spending, or lowerrevenue.44 For the nontransition sample, there issome evidence that about one-fifth of privatizationreceipts are used to reduce external financing, withthe rest offsetting domestic financing. These resultsneed to be qualified by recognizing that the regres-sions are based on a limited sample, largely com-prised of observations that coincide with periodswhen the country had an IMF-supported program;45

and, by design, only budgetary privatization pro-ceeds are included, leaving open the question of whathappens to amounts not transferred to the budget.

Total Amount of Privatization

The correlation between the total amount of pri-vatization receipts, which better indicates the switchfrom private to public ownership, and variables,such as growth, unemployment, and tax revenue, isexamined using regressions of the following form:

43A fuller description of methodology and results is availableon request from the authors.

44Results from the spending and revenue regressions are notshown. Also, regressions using the debt stock as the dependentvariable suggest that it is independent of the amount of budgetaryprivatization proceeds. This likely reflects noise in the debt-to-GDP ratio due to movements in nominal GDP growth rates andfinancing operations that affect the debt stock without impactingthe recorded deficit.

45A formal test of the proposition that budgetary privatizationproceeds are only used to reduce domestic financing when thereis an IMF-supported program is rejected. There are, however,limited observations without an IMF-supported program, and insome such cases a program may have been under discussion.


Table 11. Contemporaneous Impact of Budgetary Privatization Proceeds on Domestic Financing

A Budgetary privatization (t)

A Overall balance (t)

A External financing (t)

Observations:R-Squared

A Budgetary privatization (t)

A Real GDP growth (t)

A Unemployment (t)


-1.14*(.19)

n.a.

n.a.

83.19

.25(.19)

n.a.

n.a.

89.12

Full Sample Nontransition Short Sample 2

(Dependent variable: first difference of domestic financing)

-.97*(.13)

-.74*(.15)

n.a.

83.54

-1.19*(.19)

n.a.

-.65*(.20)

82.39

-.85*(.13)

n.a.

n.a.

52.17

-.79*(.12)

-.90*(.22)

n.a.

52.50

-1.03*(.11)

n.a.

-.96*(.19)

52.58

-1.20*(.23)

n.a.

n.a.

26.41

(Dependent variable: first difference of the overall balance)

.22(.20)

.03(.04)

n.a.

88.12

.31(.20)

n.a.

-.24***(.12)

81.16

.09(.11)

n.a.

n.a.

58.13

.10(.11)

-.02(.02)

n.a.

57.14

.09(.12)

n.a.

-.20(.14)

51.15

.22(.28)

n.a.

n.a.

28.01

-1.12*(.15)

-.46*(.17)

n.a.

26.64

.15(.36)

.09(.25)

n.a.

28.03

-1.21*(.23)

n.a.

-.10(.26)

25.46

.32(.26)

n.a.

-.44(.46)

24.09

Sources: Data provided by country authorities; and IMF staff estimates.1Standard errors are in parentheses and based on White's (1980) Heteroskedasticity-consistent covariance matrix. Asterisks indicate significance lev-

els:* is 1 percent level;** is 5 percent level;*** is 10 percent level. Except for the short sample regressions, in which a constant is included, the regressionsinclude a complete set of country-specific dummies for which the estimates are not reported. All variables are expressed as a share of GDP.

2Comprises observations corresponding to the two largest movements in privatization proceeds for each country.

Ayu = Hi pAxi,t + ui,t, (2)

where the notation is the same as before. In equation(2), the first difference of the dependent variable isrun on the level of privatization.46 The dependentvariable, therefore, is assumed to follow a randomwalk with drift during the sample period, and pri-vatization is now allowed to have either a lasting orone-period (<5=-y) effect on the dependent variable.

There is some evidence of a positive and lastingimpact of privatization on tax revenue for the non-transition sample, but not for the full sample (see

real GDP growth is the dependent variable, it is in-cluded in levels, and its lagged value is added as an additional ex-planatory variable.

Table 12). This may reflect higher collection ratesfrom the privatized firms, a shift in the structure ofGDP toward sectors paying more taxes, or a generalimprovement in macroeconomic management. Pri-vatization receipts are also found to be strongly cor-related with a lasting improvement in macroeco-nomic performance, as manifested in higher realgrowth and lower unemployment. Given the simplespecification that is used, the results should be inter-preted cautiously and not construed to imply causa-tion. Moreover, the privatization variable is likelycapturing the positive impact of a general regimechange toward better economic policies. Finally, theevidence from regressions (results not shown) usingfixed investment as the dependent variable suggeststhat it is not correlated with privatization.


Table 12. Structural Relationship Between Total Privatization Proceeds and Selected Variables1

Privatization (t)

Privatization (t - 1)

A Privatization (t)

A Unemployment (t)


Privatization (t)


Real GDP growth (t - 1)

Observations:

Privatization (t)


A Privatization (t)

Real GDP growth (t - 1)


.16(.17)

n.a.

n.a.

n.a.

104.26

Full

.16(.17)

-0.15(.15)

n.a.

n.a.

104.27

LSDV

1.07**(.49)

n.a.

.05(.11)

107

-.27***(.15)

n.a.

n.a.

n.a.

86.15

1.01**(.46)

.71**(.36)

.01(.11)

107

Sample Nontransition

(Dependent variable: first difference of tax revenue)

.15(.17)

-.23(.16)

n.a.

-.11(.07)

83.28

n.a.

n.a.

.16(.11)

n.a.

104.27

n.a.

n.a.

.18(.12)

-.10(.06)

83.28

.28**(.12)

n.a.

n.a.

n.a.

69.17

.27**(.13)

.05(.11)

n.a.

n.a.

69.18

(Dependent variable: real GDP growth, in percent)2

Anderson-Hsiao

.37*(.13)

n.a.

.15*(.03)

90

.55*(.12)

.35*(.12)

.13*(.03)

90

LSDV

1.96*(.53)

n.a.

-.35**(.14)

70

1.82*(.57)

1.09**(.50)

- .41*(.14)

70

.26**(.11)

-.02(.09)

n.a.

-.24**(.10)

49.27

(Dependent variable: first difference of the unemployment rate)

-.25***(.13)

n.a.

n.a.

-.03(.06)

86.16

-.21**(.10)

-.50*(.19)

n.a.

.02(.05)

86.24

n.a.

n.a.

.14(.12)

n.a.

86.14

n.a.

n.a.

.13(.11)

-.04(.06)

86.15

-.12(.10)

n.a.

n.a.

n.a.

50.18

-.28**(.14)

n.a.

n.a.

.13**(.06)

50.25

-.27**(.13)

-.16(.18)

n.a.

.14**(.07)

50.26

n.a.

n.a.

.11(.10)

n.a.

69.14

n.a.

n.a.

.14***(.08)

-.24**(.10)

49.24

Anderson-Hsiao

.72*(.21)

n.a.

-.25*(.04)

60

n.a.

n.a.

-.08(.07)

n.a.

50.17

M l *(.20)

1.12*(.20)

-.26*(.04)

60

n.a.

n.a.

-.06(.10)

.10***(.06)

50.23

Sources: Data provided by country authorities; and IMF staff estimates.1Standard errors are in parentheses and based on White's (1980) Heteroskedasticity-consistent covariance matrix. Asterisks indicate significance lev-

els: * is 1 percent level; ** is 5 percent level; *** is 10 percent level. The regressions include a complete set of country-specific dummies for which the esti-mates are not reported. The Anderson-Hsiao estimator, however, takes first differences to remove the country dummies prior to estimation. Except forreal GDP growth and the unemployment rate, all variables are expressed as a share of GDP.

2The combination of a lagged dependent variable and country-specific dummy (i.e., fixed effect) may lead to estimates that are biased using ordinaryleast squares (LSDV). Although the Anderson-Hsiao estimator avoids this problem, such alternative estimators may not provide better estimates of thecoefficients on the privatization terms, and thus both results are reported (Judson and Owen, 1997).


References

Anderson, T.W., and Cheng Hsiao, 1982, "Formulationand Estimation of Dynamic Models Using PanelData," Journal of Econometrics, Vol. 18 (January),pp. 47-82.

Ariyoshi, Akira, and others, 2000, Country Experienceswith the Use and Liberalization of Capital Controls,IMF Occasional Paper No. 190 (Washington: Interna-tional Monetary Fund).Aslund, Anders, 1999, Why Has Russia's EconomicTransformation Been So Arduous? paper prepared forthe Annual World Bank Conference on DevelopmentEconomics, Washington, April.

Aziz, Jahangir, and Robert F. Wescott, 1997, "Policy Com-plementarities and the Washington Consensus," IMFWorking Paper 97/118 (Washington: InternationalMonetary Fund).

Azpiazu, Daniel, and Adolfo Vispo, 1994, "Some Lessonsof the Argentine Privatization Process," CEPAL Re-view, Vol. 54 (December), pp. 129-47.

Banerji, Arup, and Richard H. Sabot, 1994, "Wage Distor-tions, Overmanning and Reform in DevelopingCountry Public Enterprises," Vice Presidency for Fi-nance and Private Sector Development (Washington:World Bank).

Berg, Andrew, and Elliot Berg, 1997, "Methods of Privat-ization," Journal of International Affairs, Vol. 50(October), pp. 357-90.

Beyer, Hans-Joachim, Claudia Dziobek, and John Garrett,1999, "Economic and Legal Considerations of Opti-mal Privatization—Case Studies of Mortgage Firms(DePfa Group and Fannie Mae)," IMF Working Paper99/69 (Washington: International Monetary Fund).

Boubakri, Narjess, and Jean-Claude Cosset, 1998, "TheFinancial and Operating Performance of Newly Pri-vatized Firms: Evidence from Developing Coun-tries," Journal of Finance, Vol. 53 (June), pp.1081-110.

Bredenkamp, Hugh, and Susan Schadler, eds., 1999, Eco-nomic Adjustment and Reform in Low-Income Coun-tries (Washington: International Monetary Fund).

Chisari, Omar, Antonio Estache, and Carlos Romero,1999, "Winners and Losers from the Privatization andRegulation of Utilities: Lessons from a General Equi-librium Model of Argentina," World Bank EconomicReview, Vol. 13 (May), pp. 357-78.

D'Souza, Juliet, and William L. Megginson, 1999, "TheFinancial and Operating Performance of PrivatizedFirms During the 1990's," Journal of Finance, Vol.54 (August), pp. 1397-438.

Fischer, Stanley, and William Easterly, 1990, "The Eco-nomic of the Government Budget Constraint," TheWorld Bank Research Observer, Vol. 5 (July), pp.127-42.

Galal, Ahmed, and others, 1994, Welfare Consequences ofSelling Public Enterprises: An Empirical Analysis(New York: Oxford University Press).

Gupta, Sanjeev, Christian Schiller, and Henry Ma, 1999,"Privatization, Social Impact, and Social SafetyNets," IMF Working Paper 99/68 (Washington: Inter-national Monetary Fund).

Hachette, Dominique, and Rolf Luders, 1993, Privatiza-tion in Chile: An Economic Appraisal (San Francisco:International Center for Economic Growth).

Havrylyshyn, Oleh, Ivailo Izvorski, and Ron van Rooden,1998, "Recovery and Growth in TransitionEconomies 1990-97: A Stylized Regression Analy-sis," IMF Working Paper 98/141 (Washington: Inter-national Monetary Fund).

Havrylyshyn, Oleh, and Donal McGettigan, 1999, "Pri-vatization in Transition Countries: A Sampling of theLiterature," IMF Working Paper 99/6 (Washington:International Monetary Fund).

Heller, Peter S., 1990, "The Budgetary Impact of Privat-ization: Examples from New Zealand and Prospectsfor Ireland," in Privatization: Issues of Principle andImplementation in Ireland, ed. by Frank J. Converyand Moore McDowell (Dublin: Gill and Macmillan)., and Christian Schiller, 1989, "The Fiscal Impact ofPrivatization, with Some Examples from Arab Coun-tries," World Development, Vol. 17 (May), pp.757-67.

Heller, Peter S., Richard Hemming, and Ruba Chakrabarti,1997, "Macroeconomic Constraints and the Modali-ties of Privatization," in Fiscal Policy and EconomicReform, ed. by Mario I. Blejer and Teresa Ter-Minassian (London: Routledge).

Hemming, Richard, and Ali M. Mansoor, 1987, "Privat-ization and Public Enterprises," IMF Working Paper87/9 (Washington: International Monetary Fund).

International Monetary Fund, 1986, A Manual on Govern-ment Finance Statistics (Washington: InternationalMonetary Fund).

Jones, Leroy P., Yahya Jammal, and Nilgun Gokgur, 1998,"Impact of Privatization in Cote d'lvoire," report pre-pared for the Privatization Committee, Boston Insti-tute for Developing Economies, Boston, November.

Judson, Ruth A., and Ann L. Owen, 1997, "Estimating Dy-namic Panel Data Models: A Practical Guide for


Macroeconomists," Finance and Economics Discus-sion Paper Series 97/3 (Washington: Federal ReserveBoard of Governors).

Kikeri, Sunita, 1998, "Privatization and Labor: What Hap-pens to Workers When Governments Divest?" WorldBank Technical Paper No. 396 (Washington: WorldBank).

Kikeri, Sunita, John Nellis, and Mary Shirley, 1992, Pri-vatization: The Lessons of Experience (Washington:World Bank).

Kodrzycki, Yolanda K., and Eric M. Zolt, 1994, "Tax Is-sues Arising from Privatization in the Formerly So-cialist Countries," Law and Policy in InternationalBusiness, Vol. 25, pp. 609-37.

La Porta, Rafael, and Florencio L6pez-de-Silanes, 1997,"The Benefits of Privatization: Evidence from Mex-ico," NBER Working Paper 6215 (Cambridge, Mass-achusetts: National Bureau of Economic Research).

Larrain R., Guillermo, and Carlos D. Winograd, 1996,"Privatisation massive, finances publiques et macro-economie: Le cas de l'Argentine et du Chili," Revueeconomique, Vol. 47 (November), pp. 1373-408.

Lieberman, Ira W., and Rogi Veimetra, 1996, "The Rushfor State Shares in the 'Klondike' of Wild East Capi-talism: Loans-for-Shares Transactions in Russia," TheGeorge Washington Journal of International Law andEconomics, Vol. 29, pp. 737-68.

Lieberman, Ira W., and Christopher D. Kirkness, eds.,1998, Privatization and Emerging Equity Markets(Washington: World Bank).

L6pez-de-Silanes, Florencio, 1997, "Determinants of Pri-vatization Prices," Quarterly Journal of Economics,Vol. 112 (November), pp. 965-1025.

Mackenzie, George A., 1998, "The Macroeconomic Im-pact of Privatization,"Staff Papers, InternationalMonetary Fund, Vol. 45 (June), pp. 363-73.

Mansoor, Ali M., 1993, "Budgetary Impact of Privatiza-tion," in How to Measure the Fiscal Deficit: Analyti-cal and Methodological Issues, ed. by Mario I. Blejerand Adrienne Cheasty (Washington: InternationalMonetary Fund).

Manzetti, Luigi, 1998, "Are Market Reforms Ending Cor-ruption in Latin America?" Mondes en developpe-ment, Vol. 26, pp. 69-81.

Megginson, William L., Robert C. Nash, and Matthias vanRandenborgh, 1994, "The Financial and OperatingPerformance of Newly Privatized Firms: An Interna-tional Empirical Analysis," Journal of Finance, Vol.49 (June), pp. 403-52.

Megginson, William L., and Jeffry M. Netter, 1999, "FromState to Market: A Survey of Empirical Studies onPrivatization," draft paper prepared for joint confer-ence of SBF Bourse de Paris and the New York StockExchange, Paris.

Nellis, John R., 1999, "Time to Rethink Privatization inTransition Economies?" Finance and Development,Vol. 38 (June), pp. 16-19.

Newbery, David M., 1997, "The Budgetary Impact of Pri-vatization," in Fiscal Policy and Economic Reform,

ed. by Mario I. Blejer and Teresa M. Ter-Minassian(London: Routledge).

Perotti, Enrico C, and Pieter van Oijen, 1999, "Privatiza-tion, Political Risk and Stock Market Development,"CEPR Discussion Paper no. 2243 (London: Centrefor Economic Policy Research).

Pinheiro, Armando Castelar, and Ben Ross Schneider,1995, "Fiscal Impact of Privatization in Latin Amer-ica," Journal of Development Studies, Vol. 31 (June),pp. 751-85.

Rose-Ackerman, Susan, 1997, "The Political Economy ofCorruption," in Corruption and the Global Economy,ed. by Kimberly A. Elliott (Washington: Institute forInternational Economics).

Sala-i-Martin, Xavier, 1997, "I Just Ran Four Million Re-gressions," NBER Working Paper 6252 (Cambridge,Massachusetts: National Bureau of EconomicResearch).

Schadler, Susan, and others, 1995, IMF Conditionality:Experience Under Stand-By and Extended Arrange-ments, Part I, Key Issues and Findings, IMF Occa-sional Paper No. 128 (Washington: InternationalMonetary Fund).

Shafik, Nemat, 1996, "Selling Privatization Politically,"Columbia Journal of World Business, Vol. 31, pp.20-29.

Shaikh, Hafeez, and others, 1996, Argentina PrivatizationProgram: A Review of Five Cases (Washington:World Bank).

Stiglitz, Joseph E., 1999, Whither Reform? Ten Years ofthe Transition, paper prepared for the World BankAnnual Bank Conference on Development Econom-ics, Washington, April.

Tanzi, Vito, 1998, Corruption Around the World: Causes,Consequences, Scope, and Cures, IMF WorkingPaper 98/63 (Washington: International MonetaryFund).

Verbrugge, James A., William L. Megginson, and WandaL. Owens, 1999, "State Ownership and the FinancialPerformance of Privatized Banks: An EmpiricalAnalysis," paper presented at the World Bank/FederalReserve Bank of Dallas Conference on Banking Pri-vatization, Washington, March.

White, Halbert, 1980, "A Heteroskedasticity-ConsistentCovariance Matrix Estimator and a Direct Test forHeteroskedasticity," Econometrica, Vol. 48 (May),pp. 817-38.

White, Howard, and Saman Kelegama, 1994, "The FiscalImplications of Privatisation in Developing Coun-tries: The Sri Lankan Experience," Institute of SocialStudies Working Paper No. 179 (August), pp. 1-38.

World Bank, 1995, Bureaucrats in Business: The Econom-ics and Politics of Government Ownership (Washing-ton: Oxford University Press).

, 1997, Old Age Security: Pension Reform inChina, China 2020 series (Washington: World Bank).

Yarrow, George, 1999, "A Theory of Privatization, or WhyBureaucrats Are Still in Business," World Develop-ment, Vol. 27 (January), pp. 157-68.


Recent Occasional Papers of the International Monetary Fund

194. Fiscal and Macroeconomic Impact of Privatization, by Jeffrey Davis, Rolando Ossowski, ThomasRichardson, and Steven Barnett. 2000.

193. Exchange Rate Regimes in an Increasingly Integrated World Economy, by Michael Mussa, Paul Masson,Alexander Swoboda, Esteban Jadresic, Paolo Mauro, and Andy Berg. 2000.

192. Macroprudential Indicators of Financial System Soundness, by a staff team led by Owen Evans, AlfredoM. Leone, Mahinder Gill, and Paul Hilbers. 2000.

191. Social Issues in IMF-Supported Programs, by Sanjeev Gupta, Louis Dicks-Mireaux, Ritha Khemani,Calvin McDonald, and Marijn Verhoeven. 2000.

190. Capital Controls: Country Experiences with Their Use and Liberalization, by Akira Ariyoshi, Karl Haber-meier, Bernard Laurens, Inci Otker-Robe, Jorge Ivan Canales Kriljenko, and Andrei Kirilenko. 2000.

189. Current Account and External Sustainability in the Baltics, Russia, and Other Countries of the Former So-viet Union, by Donal McGettigan. 2000.

188. Financial Sector Crisis and Restructuring: Lessons from Asia, by Carl-Johan Lindgren, Tomas J.T. Bal-ifio, Charles Enoch, Anne-Marie Guide, Marc Quintyn, and Leslie Teo. 1999.

187. Philippines: Toward Sustainable and Rapid Growth, Recent Developments and the Agenda Ahead, byMarkus Rodlauer, Prakash Loungani, Vivek Arora, Charalambos Christofides, Enrique G. De la Piedra,Piyabha Kongsamut, Kristina Kostial, Victoria Summers, and Athanasios Vamvakidis. 2000.

186. Anticipating Balance of Payments Crises: The Role of Early Warning Systems, by Andrew Berg, Ed-uardo Borensztein, Gian Maria Milesi-Ferretti, and Catherine Pattillo. 1999.

185. Oman Beyond the Oil Horizon: Policies Toward Sustainable Growth, edited by Ahsan Mansur and VolkerTreichel. 1999.

184. Growth Experience in Transition Countries, 1990-98, by Oleh Havrylyshyn, Thomas Wolf, Julian Beren-gaut, Marta Castello-Branco, Ron van Rooden, and Valerie Mercer-Blackman. 1999.

183. Economic Reforms in Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan, and Uzbekistan, byEmine Gurgen, Harry Snoek, Jon Craig, Jimmy McHugh, Ivailo Izvorski, and Ron van Rooden. 1999.

182. Tax Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union, by a Staff Team Ledby Liam Ebrill and Oleh Havrylyshyn. 1999.

181. The Netherlands: Transforming a Market Economy, by C. Maxwell Watson, Bas B. Bakker, Jan KeesMartijn, and Ioannis Halikias. 1999.

180. Revenue Implications of Trade Liberalization, by Liam Ebrill, Janet Stotsky, and Reint Gropp. 1999.179. Dinsinflation in Transition: 1993-97, by Carlo Cottarelli and Peter Doyle. 1999.178. IMF-Supported Programs in Indonesia, Korea, and Thailand: A Preliminary Assessment, by Timothy Lane,

Atish Ghosh, Javier Hamann, Steven Phillips, Marianne Schulze-Ghattas, and Tsidi Tsikata. 1999.177. Perspectives on Regional Unemployment in Europe, by Paolo Mauro, Eswar Prasad, and Antonio Spilim-

bergo. 1999.176. Back to the Future: Postwar Reconstruction and Stabilization in Lebanon, edited by Sena Eken and

Thomas Helbling. 1999.175. Macroeconomic Developments in the Baltics, Russia, and Other Countries of the Former Soviet Union,

1992-97, by Luis M. Valdivieso. 1998.174. Impact of EMU on Selected Non-European Union Countries, by R. Feldman, K. Nashashibi, R. Nord,

P. Allum, D. Desruelle, K. Enders, R. Kahn, and H. Temprano-Arroyo. 1998.173. The Baltic Countries: From Economic Stabilization to EU Accession, by Julian Berengaut, Augusto

Lopez-Claros, Francoise Le Gall, Dennis Jones, Richard Stern, Ann-Margret Westin, Effie Psalida,Pietro Garibaldi. 1998.

172. Capital Account Liberalization: Theoretical and Practical Aspects, by a staff team led by Barry Eichen-green and Michael Mussa, with Giovanni DeH'Ariccia, Enrica Detragiache, Gian Maria Milesi-Ferretti,and Andrew Tweedie. 1998.


171. Monetary Policy in Dollarized Economies, by Tomas Baliño, Adam Bennett, and Eduardo Borensztein.1998.

170. The West African Economic and Monetary Union: Recent Developments and Policy Issues, by a staffteam led by Ernesto Hernandez-Cata and comprising Christian A. Francois, Paul Masson, Pascal Bou-vier, Patrick Peroz, Dominique Desruelle, and Athanasios Vamvakidis. 1998.

169. Financial Sector Development in Sub-Saharan African Countries, by Hassanali Mehran, Piero Ugolini,Jean Phillipe Briffaux, George Iden, Tonny Lybek, Stephen Swaray, and Peter Hayward. 1998.

168. Exit Strategies: Policy Options for Countries Seeking Greater Exchange Rate Flexibility, by a staff teamled by Barry Eichengreen and Paul Masson with Hugh Bredenkamp, Barry Johnston, Javier Hamann,Esteban Jadresic, and Inci Otker. 1998.

167. Exchange Rate Assessment: Extensions of the Macroeconomic Balance Approach, edited by Peter Isardand Hamid Faruqee. 1998

166. Hedge Funds and Financial Market Dynamics, by a staff team led by Barry Eichengreen and DonaldMathieson with Bankim Chadha, Anne Jansen, Laura Kodres, and Sunil Sharma. 1998.

165. Algeria: Stabilization and Transition to the Market, by Karim Nashashibi, Patricia Alonso-Gamo, StefaniaBazzoni, Alain Feler, Nicole Laframboise, and Sebastian Paris Horvitz. 1998.

164. MULTIMOD Mark III: The Core Dynamic and Steady-State Model, by Douglas Laxton, Peter Isard,Hamid Faruqee, Eswar Prasad, and Bart Turtelboom. 1998.

163. Egypt: Beyond Stabilization, Toward a Dynamic Market Economy, by a staff team led by Howard Handy.1998.

162. Fiscal Policy Rules, by George Kopits and Steven Symansky. 1998.

161. The Nordic Banking Crises: Pitfalls in Financial Liberalization? by Burkhard Dress and CeylaPazarba§ioglu. 1998.

160. Fiscal Reform in Low-Income Countries: Experience Under IMF-Supported Programs, by a staff teamled by George T. Abed and comprising Liam Ebrill, Sanjeev Gupta, Benedict Clements, Ronald Mc-Morran, Anthony Pellechio, Jerald Schiff, and Marijn Verhoeven. 1998.

159. Hungary: Economic Policies for Sustainable Growth, Carlo Cottarelli, Thomas Krueger, RezaMoghadam, Perry Perone, Edgardo Ruggiero, and Rachel van Elkan. 1998.

158. Transparency in Government Operations, by George Kopits and Jon Craig. 1998.

157. Central Bank Reforms in the Baltics, Russia, and the Other Countries of the Former Soviet Union, by astaff team led by Malcolm Knight and comprising Susana Almuina, John Dalton, Inci Otker, CeylaPazarba§ioglu, Arne B. Petersen, Peter Quirk, Nicholas M. Roberts, Gabriel Sensenbrenner, and JanWillem van der Vossen. 1997.

156. The ESAF at Ten Years: Economic Adjustment and Reform in Low-Income Countries, by the staff of theInternational Monetary Fund. 1997.

155. Fiscal Policy Issues During the Transition in Russia, by Augusto Lopez-Claros and Sergei V. Alexas-henko. 1998.

154. Credibility Without Rules? Monetary Frameworks in the Post-Bretton Woods Era, by Carlo Cottarelli andCurzio Giannini. 1997.

153. Pension Regimes and Saving, by G.A. Mackenzie, Philip Gerson, and Alfredo Cuevas. 1997.

152. Hong Kong, China: Growth, Structural Change, and Economic Stability During the Transition, by JohnDodsworth and Dubravko Mihaljek. 1997.

151. Currency Board Arrangements: Issues and Experiences, by a staff team led by Tomas J.T. Baliño andCharles Enoch. 1997.

150. Kuwait: From Reconstruction to Accumulation for Future Generations, by Nigel Andrew Chalk, Mo-hamed A. El-Erian, Susan J. Fennell, Alexei P. Kireyev, and John F. Wilson. 1997.

149. The Composition of Fiscal Adjustment and Growth: Lessons from Fiscal Reforms in Eight Economies, byG.A. Mackenzie, David W.H. Orsmond, and Philip R. Gerson. 1997.

Note: For information on the title and availability of Occasional Papers not listed, please consult the IMF Publications Catalog or contact IMFPublication Services.


Documents

Fiscal and Macroeconomic Impact of Privatization · economic impact of privatization.1 Among the inter-national financial institutions, the World Bank has had the lead role in advising