05300-9781557755186-en-book.pdf - imf elibrary

OCCASIONAL PAPE R 133

Policy Experiences and Issuesin the Baltics, Russia, and Other Countrie s

of the Forme r Sovie t Unio n

Edited by Daniel A. Citrin and Ashok K. Lahiri

withJonathan Anderson

Eduard BrauAdrienne CheastyRichard Hemming

Ernesto Hernandez-Cat aJeromin Zettelmeyer

INTERNATIONAL MONETAR Y FUN DWashington DCDecember 1995

©International Monetary Fund. Not for Redistribution

© 1995 International Monetary Fund

Cataloging-in-Publication Data

Policy experiences and issues in the Baltics, Russia, and other countriesof the former Soviet Union / edited by Daniel A. Citrin and Ashok K.Lahiri; with Jonathan Anderson . . . [et al.].—Washington, D.C. : In-ternatioanl Monetary Fund, [1995]

p. cm. — (Occasional Paper, ISSN 0251-6365 ; no. 133)

ISBN 1-55775-518-3

1. Former Soviet republics — Economic policy. 2. Economic stabi-lization— Former Soviet republics. 3. Former Soviet republics —Economic conditions. I. Citrin, Daniel. II. Lahiri, Ashok.III. Occasional paper (International Monetary Fund); no. 133.HG336.27.P65 1995

Price: US$15.00(US$12.00 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

Internet: [email protected]

recycled paper


Contents

Preface

I OverviewDaniel A. Citrin

Common Challenges at the Start of TransitionSpecial FactorsResponse Thus Far

Key Common Issues in the Transition ProcessDecline in OutputInflation Developments and Velocity BehaviorEmergence of Interenterprise ArrearsDevelopments and Main Contributing FactorsLessonsRole of the Exchange RateRole of External Financial Assistance, 1992-93Has Financial Assistance Been Adequate?

Concluding RemarksAppendix

Data Issues on Interenterprise Arrears

II The Decline in OutputJonathan Anderson, Daniel A. Citrin, and Ashok K. Lahiri

Output DeclineFactors Related to Economic TransitionFactors Related to the Breakup of the Soviet UnionOther Factors

Determinants of Future GrowthSome Tentative Evidence from Central Europe and the BalticsConclusions and Policy ImplicationsReferences

III The Behavior of Inflation and VelocityJonathan Anderson and Daniel A. Citrin

Behavior of Money, Income, and Prices, 1992-94Explaining Inflation and Velocity Movements

Inflation and Rate of Growth of Broad MoneyFactors Affecting the Underlying Demand for MoneyInfluence of Exogenous FactorsRole of Interenterprise CreditMeasurement Errors

Differences in Velocity PerformanceRecovery of Real Money DemandWage Developments

Conclusions and Implications for Stabilization Policy

ix

1

123669

141920212325262727

28

2829343538384042

44

4448484950565657586162

iii


CONTENTS

IV Interenterprise ArrearsAshok K. Lahiri and Daniel A. Citrin

Evolution of Interenterprise Credit and ArrearsInitial Transition Period, Early 1992 to Mid-1993Mid-1993 to Late 1994Factors Causing ArrearsNormal Factors Operating in a Market EconomyAdditional Factors in the Transition Economies

Policy OptionsThe Stock ProblemPreventing New ArrearsInterstate ArrearsThe Energy Sector

ConclusionsReferences

V The Revenue DeclineRichard Hemming, Adrienne Cheasty, and Ashok K. Lahiri

Revenue Developments Since 1991Movements in Total RevenueMovements in Individual TaxesDevelopments in 1994

Factors Explaining Revenue DevelopmentsChanges in Level and Structure of Economic ActivityDesign of the Tax SystemEfficiency of Tax AdministrationCrisis in Governance

Policy Implications and ResponsesImmediate ChallengesShort- to Medium-Term PrioritiesEstablishing Good Governance

VI Stabilization: Fixed Versus Flexible Exchange RatesJeromin Zettermeyer and Daniel A. Citrin

General Considerations and ExperienceRelative CostsEffectivenessCosts of FailureProbability of SuccessA Currency Board

Exchange Rate Policies in Stabilization ProgramsChoice of Exchange Rate Regime in Fund-Supported ProgramsStabilization Performance Under Alternative Anchors:

Estonia and LatviaExperience with Money-Based Programs Outside the BalticsConclusions for Future Stabilization Policies

References

VII External Financial Assistance: The Record and IssuesEduard Brau

The Record, 1992-93Is Financial Assistance Conditioned and Timed Appropriately?Is Financial Assistance Adequate in Size?Is Right Financial Assistance Being Provided?

63

63646566676871727273746475

76

76777879818183858687889091

93

9393949494959697

98100101102

103

103106108112

iv


Contents

V

Is Assistance Repayable?What Is to Be Done About Financing Problems Between

States in the Region?ConclusionAppendix

How External Financing Helps Economies in TransitionBibliography

VIII Russia and the IMF:The Political Economy of MacrostabilizationErnesto Hernandez-Cata

Inflation: Fables, Tall Tales, and Old TruthsRevolt Against Macrostabilization: Centralized Credits,

Interenterprise Arrears, Budget Subsidies,and Tax Exemptions

Changing the Structure of Deficit Financing as an Alternativeto Fiscal Adjustment: A Fairy Tale

The May 1993 ProgramThe March 1994 Program

Box

1. The IMF and the Ruble Area

Tables

1.1. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Consumer Price Inflation

1.2. Selected Indicators of Reform (as of the end of 1994)1.3. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Changes in Real Gross Domestic Product1.4. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Inflation and Money Growth, Program Performance1.5. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Interenterprise Arrears1.6. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Evolution of Revenue, Excluding Union Grants1.7. Countries in Transition: Stabilization Attempts and Inflation

Performance1.8. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Global Gross Capital Flows, 1992-93, Cumulative

2.1. Real Gross Domestic Product2.2. Gross Fixed Investment and Stockbuilding2.3. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Real Gross Capital Formation2.4. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Gross Capital Formation2.5. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Value of Total Exports to the Rest of the World2.6. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Output by Sector, 19932.7. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Volume of Interstate Trade2.8. The Baltics, Russia, and Other Countries of the Former Soviet

Union: Real Domestic Money Stock2.9. Real Money Stock and GDP at Constant Prices

2.10. Money and Prices

114

114115115115116

117

117

119

120122123

119

34

6

12

18

20

22

24

31

32

32

33

34

36

37383939


CONTENTS

vi

2.11. Eastern European Countries in Transition: Selected Indicators

3.1. The Baltics, Russia, and Other Countries of the Former SovietUnion: Money and Prices, 1992-94

3.2. The Baltics, Russia, and Other Countries of the Former SovietUnion: Quarterly Changes in Macroeconomic Variables

3.3. Real Domestic Money Stocks, as of Third Quarter of 19943.4. Energy Import Prices3.5. Quarterly Changes in Wages, Exchange Rates, and Real

Interest Rates Under Fund-Supported Programs

4.1. Belarus, Latvia, Russia, and Ukraine: Interenterprise Credit4.2. Russia: Interstate Interenterprise Arrears, as of October 1, 19944.3. Russia: Sectoral Composition of Interenterprise Arrears

5.1. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Revenue and Grants, 1991

5.2. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Revenue Developments

5.3. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Change in Real Revenue

5.4. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Decomposition of the Change in theRevenue-to-GDP Ratio, 1991-93

5.5. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Change in Revenue-to-GDP Ratio

5.6. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Change in Tax Bases and Real Revenues, 1991-93

6.1. IMF-Supported Programs in the Baltics, Russia, and Other Countriesof the Former Soviet Union: General Government FiscalBalance, Actual and Program

6.2. The Baltics, Russia, and Other Countries of the Former Soviet Union:Gross Reserve Holdings

6.3. Estonia and Latvia: Disinflation and Output Loss, 1992-946.4. Latvia and Estonia: Interest Rate Differentials

7.1. The Baltics, Russia, and Other States of the FormerSoviet Union: Financing Requirements, 1992-93, Cumulative

7.2. The Baltics, Russia, and Other States of the FormerSoviet Union: Ratios of Capital Inflows to GDP, Importsand Per Capita, 1992-93, Cumulative

7.3. The Baltics, Russia, and Other States of the FormerSoviet Union: Debt and Debt Service, 1992-94

7.4. Commitments and Disbursement to the Baltics, Russia, and OtherCountries of the Former Soviet Union: 1992-June 30, 1994,Cumulative

7.5. Financial Assistance to Russia—Announcements and Outcomes

Charts

1.1. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Inflation and Output Growth: 1992-94

1.2. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Real Gross Domestic Product

1.3. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Growth in Output and Broad Money, 1992-94

1.4. Money Velocity and Inverse Real Money Balances

42

45

464855

60

656768

77

78

79

80

81

82

97

9899

100

104

105

107

109111

5

7

810


The following symbols have been used throughout this paper:

. . . to indicate that data are not available;

— 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 (e.g., 1991-92 or January-June) to indicate the years ormonths covered, including the beginning and ending years or months;

/ between years (e.g., 1991/92) to indicate a crop or fiscal (financial) year.

"Billion" means a thousand million.

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 territor-ial entities that are not states, but for which statistical data are maintained and provided in-ternationally on a separate and independent basis.

Contents

1.5. Difference Between Inflation and Money Growth Rates in SelectedFund-Supported Programs, 1992-94

1.6. Inflation and Lagged Money Growth

2.1. Eastern Europe, the Baltics, and Other Countries of the FormerSoviet Union: Real Gross Domestic Product

2.2. The Baltics, Russia, and Other Countries of the FormerSoviet Union, 1991: Decline in Rest-of-the-WorldExports and Output

2.3. The Baltics, Russia, and Other Countries of the FormerSoviet Union: Real Interest Rates and Real GDP Growth,April 1992-July 1994

3.1. Relative Interest Rates and Real Balances3.2. Real Balances, Real Wages, and the Real Exchange Rate3.3. Real Balances and Interenterprise Credit3.4. Broad Money Velocity, Including Foreign Currency

4.1. Other Central and Eastern European Countries in Transition:Interenterprise Arrears

1416

31

35

40

51535758

66

vii


This page intentionally left blank


Preface

This Occasional Paper is based on papers prepared during late 1994 and early 1995and reflects developments until early 1995. Some of the papers were prepared for anIMF Executive Board discussion of policy experiences and issues in the Baltic coun-tries, Russia, and other countries of the former Soviet Union. The authors would liketo thank John Odling-Smee for his encouragement and support of this project, andStanley Fischer, Susan Schadler, Teresa M. Ter-Minassian, and numerous other col-leagues throughout the IMF for their valuable comments. Thanks are due to LennaGaribian for research assistance, and Joan Campayne and Bonnie Coles for secretar-ial support. Juanita Roushdy of the External Relations Department gave the manu-script a final edit and coordinated production of the publication. The views expressedhere, as well as any errors, are the sole responsibility of the authors and do not neces-sarily reflect the opinions of the Executive Directors of the IMF, or other members ofthe IMF staff.

ix


This page intentionally left blank


I Overvie wDaniel A. Citrin1

T he emergence in late 1991 of 15 independentstates from the territory of the former U.S.S.R.

presented policymakers in these countries, as well asthe international financial community, with unprece-dented economic challenges.2 Although there weresome differences, these countries faced a number ofcomparable problems at the start of their transforma-tion to market economies. Nevertheless, the periodsince the breakup of the former U.S.S.R. has seenwide divergences in progress in making this transfor-mation. This paper begins by reviewing briefly thekey challenges these countries faced at the beginningof the transition process, the progress made to date,and the size and adequacy of external financial assis-tance received.

During the initial years of transition, a number ofcommon developments and issues of concern haveemerged across a wide range of the countries in theregion. Among these, six issues have presented partic-ular difficulty in the design and implementation ofFund-supported financial programs for these coun-tries: (1) the extensive decline in output; (2) the per-sistence of inflation, which in many cases occurredfor some time despite a deceleration in the rate ofmonetary expansion; (3) the emergence of overduepayments among enterprises, or interenterprise ar-rears; (4) the sharp decline in fiscal revenues; (5)questions regarding the appropriate exchange ratestrategy that might be employed in the stabilizationprocess; and (6) the role of external financial assis-tance. The paper reviews developments in each ofthese areas, with a view to drawing lessons for stabi-lization and reform in the period ahead. Many otherissues, such as financial sector reform, privatizationand enterprise restructuring, the functioning of labormarkets, and the structure of government expendi-tures are also critical to the stabilization and reform

1The author wishes to thank Jonathan Anderson, Ashok Lahiri,Jim Wein, and Jeromin Zettelmeyer for their contributions andcomments.

2The process of re-establishing independence had alreadystarted in the Baltic countries in 1989, when working groups wereformed to prepare programs for independent economies, includ-ing separate monetary systems and national currencies.

process, but a detailed analysis of these is beyond thescope of this paper.

Common Challenges at theStart of Transitio n

After the breakup of the U.S.S.R., the Baltics, Rus-sia, and the other states of the former Soviet Unionall faced the immediate need for macroeconomic sta-bilization measures, both to address existing financialimbalances and to contain price pressures in the wakeof price liberalization. In conducting monetary pol-icy, an immediate issue was whether some or allstates would use a common currency as part of aruble area, or whether separate national currencieswould be established.3 On the one hand, the rela-tively limited institutional and administrative capac-ity for conducting independent monetary policies inthe newly formed states pointed to the maintenanceof the ruble area as the means of achieving effectivemonetary control. On the other hand, the ruble areahad an inherent inflationary bias caused by the "freerider" problem, whereby each state with its own cen-tral bank had an incentive to expand credit at a ratefaster than the average. Thus, attaining monetary sta-bility in the context of the ruble area depended on theability to effectively implement a coordinated mone-tary policy for the area as a whole.

During 1992, the Baltic countries and Ukrainemoved to establish independent currencies, not onlybecause of a desire to achieve better monetary con-trol, but also to overcome shortages of ruble bank-notes, to capture income from seigniorage, and tostrengthen their national political and economicidentities. Other non-Russian states, however, pre-ferred to stay in the ruble area. This appears to havebeen motivated by the objective of maintaining tradeand payments relations among the Baltic countries,Russia, and other countries of the former Soviet

3The Baltic countries were exceptions, with all three decidingearly on about the introduction of their own independent national

currencies.

I


Union, and an expectation that remaining in theruble area would result in access to more financingand cheaper energy from Russia.4 It became increas-ingly clear, however, that the prospects for obtainingassistance from Russia were not necessarily linkedwith the ruble area issue. At the same time, manypolicymakers realized that the level of policy coordi-nation among member states needed for the commoncurrency area to function effectively might not beconsistent with their desire to maintain political andeconomic independence. In the event, an effectivecommon monetary policy could not be put in place,initiatives to establish a common currency area wereabandoned within a few months after the Russiancurrency reform in July 1993 that demonetized pre-1993 ruble notes, and since late 1993 monetary pol-icy has been conducted on the basis of independentcurrencies in all countries, with the exception ofTajikistan.5

The early move to establish monetary indepen-dence allowed the early achievement of price stabi-lization in the Baltic countries, first in Estonia andLatvia and later in Lithuania. In Ukraine, the intro-duction of an independent currency was followed byextremely high inflation and a sharply depreciatingexchange rate because of insufficient supporting fi-nancial policies. In the Kyrgyz Republic—the otherstate that introduced its own currency prior to theRussian monetary reform in July 1993—the new cur-rency was also initially accompanied by excessiveinflation and a weak exchange rate, but stabilizationwas subsequently achieved and the currencystrengthened as financial policies were tightened ap-propriately. That said, remaining in the ruble areawas not an attractive option either. Expansionarypolicies in Russia, and the failure to implement a co-ordinated monetary policy among those states con-tinuing to use the ruble, led to a lack of financial sta-bility in the ruble area prior to its demise as well.

4The IMF made it clear to newly independent states that the de-cision as to whether or not to introduce a national currency was asovereign prerogative. However, it stressed (1) that monetaryunion would entail certain economic consequences, particularlythe loss of independence in financial policymaking, and (2) thatthe introduction of a national currency would not improve macro-economic performance unless backed by appropriately prudentfiscal and monetary policies.

5Developments and issues related to monetary arrangements arereviewed in IMF, Common Issues and Interrepublic Relations in theFormer U.S.S.R., IMF Economic Review (Washington, April 1992);Ernesto Hernandez-Cata, "The Introduction of National Currenciesin the Former Soviet Union: Options, Policy Requirements andEarly Experiences," The Economics of New Currencies (London:Centre for Economic Policy Research, 1993), pp. 53-79; andThomas Wolf and others, Financial Relations Among Countries ofthe Former Soviet Union, IMF Economic Review, No. 1 (Washing-ton: International Monetary Fund, February 1994). Tajikistan intro-duced its national currency in May 1995.

The authorities in many states recognized that far-reaching structural changes were needed to establishthe framework for a market-oriented economy. Inmoving successfully to a market-based system, com-prehensive decontrol of prices was essential to guidethe efficient allocation of resources, and in additionas a tool to help absorb excess liquidity at the outsetof reform (i.e., the "monetary overhang"). For the netenergy-importing economies, this meant confrontinga large terms of trade shock that would imply a sub-stantial and permanent reduction in real income, asenergy prices rose markedly toward world-marketlevels. An end to the system of allocation of produc-tion through state orders was also necessary.

In addition to the provision of market incentives toeconomic agents, successful transformation wouldrequire that corporate decisions be taken on the basisof such signals. With the privately owned firm as thebasic element of a market economy, this would entailprivatization of existing state enterprises, as well asfostering the creation of new private firms. At thesame time, recognizing that privatization of large en-terprises in particular would take time, enterprise re-structuring was needed to ensure that state enter-prises would be run on a commercial footing—thatis, on the basis of "hard" budget constraints. Specificmeasures to facilitate restructuring would include re-ducing state subsidies and eliminating the system of"directed credits" at below-market interest rates tar-geted at specific enterprises and sectors.

A key requirement for establishing an efficient andcompetitive productive sector, as well as for providingconsumer choice, was the liberalization of externaltrade, services, and capital transactions. This includedreducing tariff and nontariff restrictions, securing therights of foreign direct investors, and establishing free-dom of access to foreign exchange.

Other structural policies to support transformationincluded institutional and legal reform to provide theappropriate framework for the operation of a marketeconomy, such as the establishment of a well-function-ing (two-tier) banking system that is sufficiently capi-talized and well supervised, of effective treasury oper-ations and tax administration, of a legal frameworkproviding for bankruptcy and the enforcement of con-tractual obligations, and of statistical systems capableof meeting the information requirements of monitor-ing and operating the economy. And, given the in-evitable short-term contraction of output facing thesestates from the dramatic changes in the old structure ofincentives and trading relationships, it was importantto introduce a well-targeted social safety net.

Special Factors

To some extent, significant differences in the situ-ation facing these states at the start of the transition

I OVERVIE W

2


Common Challenges at the Start of Transition

Table I . I. Th e Baltics, Russia, and Other Countries of the Former Soviet Union: Consumer PriceInflation1

(In percent)

EstoniaLatviaLithuaniaMoldovaKyrgyz Republic

TajikistanArmeniaRussiaKazakstanUzbekistan

TurkmenistanAzerbaijanBelarusUkraineGeorgia

1992

1,069951

1,0201,157

855

1,157825

1,3531,516

645

4931,747

9691,210

913

1993

89109410689

1,209

2,1953,732

8961,662

534

2,1021,0471,1884,7353,126

Est.1994

483672

245277

1,5005,169

3021,880

723

2,3971,4932,220

89118,264

I994:Q4Annual Rate2

1824495159

158240388422454

2,0622,2622,7915,3128,479

Sources: Data provided by country authorities; and IMF staff estimates.1Average year-on-year percentage increase. Countries arranged according to annual inflation rate in the fourth quarter of 1994.2Annualized rate of inflation during the fourth quarter of 1994 except for Armenia, Azerbaijan, and Georgia, which are based on data for the three

months through November 1994.

process have also influenced the response of individ-ual states in the transition process. First, geographi-cal proximity and historical links may have made iteasier for some states so inclined, such as theBaltics, to establish new trading relations with mar-ket economies, and vice versa. Second, factor en-dowments varied widely. For example, significantenergy and other natural resources exist in Russia,Turkmenistan, Uzbekistan, Kazakstan, and Azerbai-jan, which, all else equal, could ease the shock oftransition in these states.

Finally, and critically, the political commitment to amarket economy varied from state to state. To an im-portant extent, this reflected differences in the relativepowers of various interest groups, such as the agricul-tural sector and the military-industrial complex. Thesedifferences were reflected in disparities in popularsupport for and political commitment to reforms,which, in turn, had a key impact on the strength andcredibility of the reforms across the region.

Response Thus FarOverall, progress in achieving success in stabiliza-

tion and structural reform has been mixed. Five ofthe 15 countries—the Baltic states, the Kyrgyz Re-public, and Moldova—have gone a long way toward

macroeconomic stabilization (Table 1.1). Elsewhere,however, inflation has remained excessively high. Inthe structural area, the Baltic countries, and to asomewhat lesser extent the Kyrgyz Republic andRussia, have made significant strides (Table 1.2). Inmany of the others, however, structural transforma-tion has advanced little.

The process of stabilization and transformationhas been accompanied by substantial declines in out-put in all countries. There is little to suggest, how-ever, that delays in fighting inflation have been asso-ciated with a more favorable output performance—ifanything, cross-country data for 1992-94 show aninverse relationship between inflation and outputgrowth (Chart 1.1). And in several countries whereinflation has been reduced—that is, the Balticstates—there are now clear indications of economicrecovery (Table 1.3). In most others, however, eco-nomic activity has continued to weaken.

Turning to a brief review of country-specificprogress, it is evident that the Baltic states are clearlyfarthest along in the transition. These states have bene-fited from proximity and past economic relations withmarket economies, particularly in Scandinavia, andprevious experience as independent market economies.Nevertheless, the data clearly bear out the effects ofsustained stabilization efforts in the Baltic states, with

3


Table 1.2 . Selecte d Indicators of Reform (as of the end of 1994) 1

EstoniaLatviaLithuania

Kyrgyz RepublicRussia

KazakstanMoldova

ArmeniaAzerbaijanBelarusGeorgiaTajikistanTurkmenistanUkraineUzbekistan

FiscalConsolidation2

SSS

ML/M

MM

LL

L/MLLL

L/ML/M

Privatization/Land

Restitution

SMS

SS

MM

SL/M

LL/M

LLLL

Government/Institutional

Reform3

SSM

M/SM

ML/M

LLMLLLLL

LegalFramework4

M/SM/SM/S

ML/M

ML

LLLLLLLL

SocialSafetyNet

MMM

LL

LM

LLLLLLLL

TradeLiberalization

SM/S

S

SM/S

MM

L/M

MLLLSL

Source: IMF staff estimates.1L, M, and S stand for little, moderate, and substantial progress, respectively.2Judgment based on a comprehensive view of fiscal and quasi-fiscal developments.3Banking system, treasury, tax administration, foreign exchange control, statistics.4Bankruptcy law, contract law, stock exchange, commercial banking law.

inflation having been reduced to relatively low levelsand positive growth resuming. In addition to adoptingtight fiscal and monetary policies, these countries,which even before the dissolution of the Soviet Union,had begun to deregulate prices, fully liberalized pricesand absorbed the total terms of trade impact of a rise inenergy prices to world market levels with little delay.6

These countries have also increasingly altered their di-rection of trade toward western Europe.

On the structural side, the three Baltic states havemade good progress with privatization, includingthat of large-scale enterprises,7 although land restitu-tion is proceeding slowly and looks likely to be alengthy process. With regard to enterprise restructur-ing, budget subsidies and directed credits havelargely been eliminated, but the implementation of alegal structure for bankruptcies and enforcement ofcontracts is far advanced only in Estonia, with lessprogress in Latvia and Lithuania. Trade regimes inall three Baltic states have been liberalized to a large

6Lithuania lagged behind Estonia and Latvia somewhat in thisrespect. In all three, incomes policy was used as a transitory mea-sure, with a view to absorbing the permanent effects of the energyprice rise without undue inflationary effects.

7Latvia is lagging somewhat with regard to large-scaleprivatization.

degree,8 and both current and capital account trans-actions are virtually fully convertible (all three coun-tries have accepted the obligations of Article VIII,Sections 2, 3, and 4 of the Fund's Articles of Agree-ments). Establishing a well-functioning banking sec-tor has proved to be difficult, however, with supervi-sion remaining weak and insolvency problemssurfacing in some banks. Also, while moderateprogress in the planning of social safety nets9 admin-istered and funded through government (rather thanenterprises) has been made, the Baltic states are ingeneral still not fully prepared to cope with a signifi-cant rise in open unemployment if it were to occur.

The Russian Federation has made significantheadway, although its reform process started laterand less decisively than that of the Baltic states. Forreasons that are explained in Section VIII, efforts atachieving macroeconomic stability have not suc-ceeded, however.10 The monthly rate of inflation fell

8The reduction of agricultural import tariffs in Latvia andLithuania has posed problems.

9Mainly the provision of public housing, the structure of unem-ployment benefits, and reform of the pension system.

10The analysis here incorporates developments until early 1995only. Considerable progress toward stabilization was achieved inRussia around mid-1995.

I OVERVIE W

4


Common Challenges at the Start of Transitio n

to single-digit levels in mid-1994, reflecting a sharptightening of monetary and fiscal policies during1993 and early 1994, but credit expansion began toaccelerate in midyear, and monthly inflation rose todouble digits by the end of 1994. Indeed, by thattime, it became clear that performance on the stabi-lization front had deteriorated. The fiscal deficit hadrisen significantly over 1993, and inflation was on arising—rather than falling—trend. Accordingly, re-ducing the fiscal deficit remained a key challenge,not only because of continuing sectoral pressures toincrease expenditure, but also because of a collapsein revenue owing to a decline in traditional taxbases, numerous exemptions, and problems in taxadministration and compliance.

Russia has made important structural progress. Inparticular, privatization has been rapid, with overhalf of output and employment now in the privatesector. Progress in enterprise restructuring, however,has been slow, in part reflecting lack of effectivecorporate governance, and banking supervision re-mains weak. Prices have been largely freed of con-trols at the federal level with a few local restrictionsremaining, and imports are also more or less free ofrestrictions. A wide range of controls remain, how-ever, on the export side. Russia, like many othercountries in the region, has received technical assis-tance from the Fund for improvements to its socialsafety net, but there has been little reform in this areathus far.

Another country where significant progress to-ward stabilization and reform has been made is theKyrgyz Republic. Fiscal and monetary polices havebeen kept tight, notwithstanding a sharp contractionin revenue collections, and monthly inflation hasaveraged just 3 percent since April 1994. The (som)exchange rate has appreciated considerably againstthe U.S. dollar since that time, and remonetizationof the economy is under way. On the structuralfront, progress has been most notable in privatiza-tion, deregulation of domestic prices, and liberal-ization of the exchange and trade system. Indeed,the Kyrgyz Republic has maintained full currentand capital account convertibility since mid-1993.Key areas where a strengthening of efforts isneeded are enterprise restructuring and banking re-form; indeed, interenterprise arrears are a signifi-cant problem and the financial situation of thebanking sector is weak.

In Moldova, inflation has been dramatically re-duced to low single-digit levels (monthly rate) sinceApril 1994, the exchange rate has stabilized, and pri-vate capital has flowed back into the country. Theprogress in stabilization has not been matched, how-ever, in the structural area. Particular problems arethe poor financial situation of state enterprises(which threatens to undermine price stability), rela-

Chart I . I . The Baltics, Russia, and OtherCountries of the Former Soviet Union:Inflation and Output Growth, 1992-941

Sources: Data provided by country authorities; and IMF staffestimates.

1Average annual percent changes in retail prices and real GDP.

tively little progress in privatization, and a weakbanking sector.

In Kazakstan, a tightening of financial policies inmid-1994 brought about a marked reduction in theinflation rate, but inflation remains too high, andweak revenue performance along with enterpriselosses are, inter alia, exerting considerable inflation-ary pressures. Prices have been largely decontrolled,and trade liberalization has progressed. However,implementing privatization plans has lagged, thestate enterprise sector has not been subject to finan-cial discipline, and important legislation—for exam-ple, in bankruptcy and antimonopoly—remains to beintroduced.

Elsewhere, there was generally little progress to-ward stability or transformation during the 1992-94period as a whole. However, at the end of 1994, stabi-lization and reform programs were initiated inUkraine, Armenia, Georgia, and Uzbekistan, andthere are indications of further significant efforts in1995.ll In Belarus, there was only partial implementa-tion of a Fund-supported program in 1993, followedby policy setbacks, particularly in the structural area.

"Georgia achieved considerable progress toward controllinginflation around mid-1995.

5

Inflation (logarithmic scale)


Table 1.3. Th e Baltics, Russia, and Other Countries of the Former Soviet Union: Changes in RealGross Domestic Product(In percent)

Countries afflicted by armed conflicts

CaucasusArmeniaAzerbaijanGeorgia

Central AsiaTajikistan

OtherMoldova

Countries with a large share of naturalresource exports

Central AsiaTurkmenistanUzbekistan

Others

BalticsEstoniaLatviaLithuania

Central AsiaKazakstanKyrgyz Republic

OtherBelarusRussiaUkraine

Average (weighted)

1991

-10.8-0.7

-20.6

-8.7

-18.0

-4.8-0.9

-7.9-11.1-13.1

-13.0-5.2

-1.2-13.0-11.9

-9.0

1992

-52.4-22.1-45.6

-30.0

-29.1

-5.0-11.0

-17.0-35.2-20.1

-14.0-19.1

-9.6-19.0-17.0

-22.8

1993

-14.8-11.5-30.0

-27.6

-8.8

-9.9-2.4

-2.1-14.8-16.1

-12.0-16.0

-9.5-12.1-14.2

-14.1

Est.1994

-23.1-35.2

-15.1

-22.3

-20.0-2.6

6.01.82.0

-20.1-15.1

-21.5-15.0-23.0

-17.7

Cumulative1991-94

-63.8-47.3-80.4

-60.7

-58.8

-34.8-16.1

-20.7-50.0-40.6

-47.4-45.3

-36.6-47.3-51.7

-49.2

Sources: Data provided by country authorities; and IMF staff estimates.

The commitment to market-oriented reforms wasstrengthened in the second half of 1994, and a numberof reform and stabilization measures were adopted. InAzerbaijan, Tajikistan, and Turkmenistan, there hasbeen little movement to stabilize prices or initiate re-forms, although in Azerbaijan reform efforts are nowunder active consideration.

Key Common Issues in theTransition Process

Among the key common issues shared by the coun-tries in transition were (1) decline in output, (2) infla-tion developments and velocity behavior, (3) emer-gence of interenterprise arrears, (4) revenue decline,

(5) role of exchange rate, and (6) role of external fi-nancial assistance.

Decline in OutputThe significant fall in recorded output throughout

the region since the dissolution of the U.S.S.R. hasraised concerns about the design of the stabilizationand reform strategy being, or to be, pursued. Indeed,real GDP in these states declined by one third on av-erage in 1992-93 and further in 1994, although theextent of deterioration has varied markedly acrosscountries. Even though part of the fall may be attrib-uted to well-known measurement problems—for ex-ample, the overstatement of production under theplanned system followed by incomplete coverage

6

I OVERVIE W


Key Common Issue s in the Transition Proces s

under the new—the actual decline is likely to havebeen large as well.12

Experience and Possible Causes

The output performance has varied significantlyamong the 15 countries. The differences in part re-flect the influence of two special factors. First, fivecountries—Armenia, Azerbaijan, Georgia, Moldova,and Tajikistan—were adversely affected by armedconflicts and in some cases accompanying economicblockades, and thus experienced relatively large out-put declines (Table 1.3 and Chart 1.2). Second,Uzbekistan and Turkmenistan were buoyed by rela-tively stable demand and favorable price movementsfor natural resource exports, which accounted for alarge share of total output in those economies.

Apart from these differences, the weakening ingrowth in the region stemmed from the influence ofa common set of forces. While data limitations pre-vent a precise quantitative analysis for most states inthe region, the available literature and informationpoint to the importance of two types of factors: (1)developments related to the desirable systemic trans-formation, and (2) problems arising from the disinte-gration of the U.S.S.R.13

With regard to the first, the reduction in govern-ment coordination of economic activity led to transi-tional disruptions in input supplies, and marketingdifficulties for enterprises. In addition, domesticprice liberalization, modifications in the structure oftaxes and subsidies, and the end of central planningall led to substantial changes in relative prices andthe structure of demand, with a rise in demand forgoods and services previously in "shortage" and afall in demand for output in traditional sectors. Out-put in some traditionally favored sectors fell becauseof reductions in governmental demand. For exam-ple, the drop in defense spending in Russia con-tributed an estimated 3-4 percentage points to theoverall drop in GDP in 1992 in that country. Simi-larly, for some of the central Asian republics, thecessation of grants from the U.S.S.R. budget from1992, and substantial reductions in official financingfrom Russia amounting to 15 to 20 percent of theircombined GDP from 1993 onwards, exacerbatedoutput losses in these states by constraining theirability to pay for imported inputs.

While a large part of the stock of physical andhuman capital in traditional sectors quickly became

12It should be kept in mind that to the extent that part of the lostoutput may have been unsalable or undesired by consumers andthe public, the decline in output does not necessarily translateone-for-one into a fall in living standards or consumer welfare.

13References to the literature in this area may be found in Sec-tion II.

Chart 1.2 . Th e Baltics, Russia, and OtherCountries of the Former Soviet Union:Real Gross Domestic Product(1991 = 100)

Sources: Wor ld Economic Out look database; and IMF staffestimates.

Note: Group I: (Conflict states) Armenia, Azerbaijan, Georgia,Moldova, Tajikistan. Group II: (Natural resource exporters)Turkmenistan and Uzbekistan. Group IIIA: (Fast reformers)Estonia, Latvia, and Lithuania. Group IIIB: (Moderate reformers)Kazakstan, Kyrgyz Republic, and Russia. Group IIIC: (Slowreformers) Belarus and Ukraine.

obsolete, productive capacity in new activities hasbeen slow to develop because of unavoidable timelags. Meanwhile, the liberalization of external trans-actions has led prices of tradables (especially energyproducts) to rise markedly from previously low andsubsidized levels toward those prevailing on worldmarkets, and has also released previously pent-updemand for foreign goods. Together with the loss oftraditional CMEA (Council for Mutual EconomicAssistance) export markets, these developmentsconstituted a large external shock that had adverseimplications on output and income, particularly inmany net energy-importing countries of the region.For the 14 states apart from Russia, the move inprices of tradables toward world levels is estimatedto have entailed a terms of trade deterioration ofsome 30 percent in 1992-94.14 This worsening in theterms of trade is estimated to have had a direct nega-

14See Wolf and others, Financial Relations Among Countriesof the Former Soviet Union.

7


tive impact on GDP on average in these 14 statesamounting to about 13 percentage points.15 Thecountries in the region in general, and Russia in par-ticular, could not benefit from the terms of trade gainfollowing the abolition of the CMEA because of thecollapse of CMEA trade. For example, from around60 percent of the U.S.S.R.'s total trade in the secondhalf of the 1980s, the former CMEA's share droppedto less than 20 percent of Russia's trade with the areaoutside the Baltic states and other countries of theformer U.S.S.R. in 1992.

As presented in Section II, available data arebroadly consistent with the thesis that the output de-cline in large part reflects the systemic change. Inparticular, a distinct relationship exists between thetiming of reforms and the time path of output. Thosestates (the Baltics) that reformed and stabilizedrapidly recorded larger contractions in activity ini-tially but are now experiencing a renewal of growth,as well as a recovery in investment. In the meantime,countries that have moved more slowly have regis-tered continued declines in output, albeit smallerones early on (see Chart 1.2, Groups IIIB and IIICcountries). Moreover, and consistent with the in-creasing evidence for eastern Europe, it does not ap-pear that the countries in this region that have post-poned action have been able to reduce the cumulativesize of the output decline.

Political changes and disruptions associated withthe demise of the U.S.S.R. have also been important.In addition to armed conflicts, states were affectedby disruptions to exchange and payments arrange-ments for trade with the other states in the regionand with the rest of the world, and by a reduction infinancial transfers from Russia. All these develop-ments contributed to inefficiencies and interruptionsin trade flows and, thereby, to declines in output.That said, it should be emphasized that the decline ininterstate trade must be viewed in the context of sys-temic change and the needed reorientation of tradeonto a market basis; thus, it should not be attributedsolely to the breakup of the U.S.S.R.

There is little evidence that the output decline hasbeen exacerbated by unduly contractionary mone-tary and credit policies. Indeed, in addition to the in-verse relationship between inflation and growthshown in Chart 1.1, the correlation across countriesbetween output growth and money growth is alsonegative during 1992-94 (see Chart 1.3). That said,there is a positive correlation in a number of coun-tries between the cumulative decline in the real

Chart 1.3 . Th e Baltics, Russia, and OtherCountries of the Former Sovie t Union:Growth in Output and Broad Money,1992-941

(In percent)


1Average annual percent change in domestic broad money andreal GDP.

money stock and that in output.16 However, this doesnot constitute convincing evidence in favor of thehypothesis that tight money has exacerbated the cu-mulative output decline, as the decline in real liquid-ity has not necessarily reflected tight credit policies.First, a good portion of the decline in real balancesin 1992 reflects the elimination through price liber-alization of the monetary "overhang" accumulatedin previous years.17 Second, lower real balances arelikely to have partly reflected lower money demandowing to the output decline (and high inflation).Third, substantial declines in real balances were re-ported both in countries that undertook monetarytightening and those that did not.

Moreover, there is little correlation between thedecline in output and the observed level of real in-terest rates: while high positive real rates of returnwere associated with declining output in Ukraine,the achievement of positive real interest rates was

15The calculation is based on 1990 trade intensities estimatedby David A. Tarr in "The Terms of Trade Effects of Moving toWorld Prices on Countries of the Former Soviet Union," Journalof Comparative Economics, Vol. 18 (February 1994), pp. 1-24.

16See Section II, pp. 34-38.17The real money stock in the U.S.S.R. was 35 percent higher

in 1990 than it had been in 1988.

I OVERVIE W

8

Broad money



accompanied by a recovery in output in Armenia,the Baltics, Kazakstan, and Moldova. Meanwhile,strongly negative real rates did not prevent an accel-eration in the rate of decline in output in Azerbaijan,Georgia, Tajikistan, and Turkmenistan. While tight-ening credit policies may have posed a problem fornew private firms, as well as for those in traditionalsectors, its impact is likely to have been quite differ-ent for new and traditional activities, because enter-prises that did well in the new market environmentcould use internal cash flows to finance investment.In addition to stabilizing prices, the primary effectof tighter credit has likely been to accelerate struc-tural change, reducing rather than increasing thetotal output loss associated with transition, as misal-located resources were freed up more quickly andbecame available for use in new activities, and man-agers were found to develop new products and seeknew markets.

The Period Ahead

Looking ahead, the resolution of political disrup-tions could lead to some recovery in activity, includ-ing in traditional sectors. More important, however,the initiation of sustainable growth oriented toward anew structure of production will depend positivelyon the pace of structural change and investment innew activities. Indeed, looking at several central andeastern European countries, as well as the Baltics,(1) all but one country first brought down inflationbefore beginning to recover, and (2) the recovery hasgenerally been accompanied by a strengthening inreal fixed investment, enhancing possibilities forlonger-term growth. This implies that prospects forresumed growth will be better the more decisive arestabilization efforts as well as other actions that havea positive impact on confidence and the investmentclimate, that is, bold liberalization and reform mea-sures, and the development of a stable legal and fi-nancial infrastructure.

Inflation Developments andVelocity Behavio r

An important concern in the transition has been theextremely high and variable rates of inflation ob-served for most economies in the region. Price growthhas been due primarily to rapid monetary expan-sion—indeed, average inflation and money growthrates have been highly correlated during the 1992-94period as a whole for every country. However, therehave also been intervals of striking divergence be-tween rates of inflation and money growth, reflectedin most cases in declining domestic real money bal-ances and increases in the level of velocity.

The Experience

Velocity movements in the Baltics, Russia, andother states of the former Soviet Union have tendedto display the following characteristics (Chart 1.4).First, increases have generally occurred not insmooth trends but rather in large discrete shifts; inextreme cases, such as in Armenia, Georgia, andTurkmenistan in late 1993, velocity rose threefold ormore in the space of one quarter. Second, an acceler-ation in inflation accompanied by a rise in velocityoccurred virtually simultaneously across a numberof countries. During late 1993 and early 1994, veloc-ity rose substantially in every country in the regionexcept for the Baltic states, Russia, and Uzbekistan.Finally, increases in velocity have tended to be pro-tracted. In some cases, velocity has declined onlyslowly following the initial increase, while in othersit has remained at its new, higher level.

A review of Fund-supported program experience,where many of the initial stabilization efforts saw in-flation outcomes exceed program targets by widemargins, also highlights the ability of money growthand inflation paths to diverge substantially in theshort term. Contrary to initial expectations, inflationsubstantially exceeded contemporaneous broadmoney growth rates for two quarters or more in fiveof the seven programs where price objectives werenot attained—Belarus, Kazakstan (1993), KyrgyzRepublic (1993), Lithuania (1992), and Moldova(September, 1993)—and were reflected in markedand unanticipated increases in velocity (see Table1.4 and also Chart 1.5). Only in Russia (1993) wasthe poor inflation outturn explained first and fore-most by excessive monetary expansion.

Four factors appear to have played important rolesin these observed inflation and velocity movements.First, inflation has generally responded with a lag tochanges in monetary conditions, especially in coun-tries where the rate of monetary expansion has beenhighly volatile. Most stabilization efforts, for exam-ple, have only been rewarded with a decline in infla-tion one or two quarters subsequent to decline inmoney growth. Chart 1.6 presents developments ininflation and lagged money growth for the countriesin the region during 1992-94. For several coun-tries—Latvia, Moldova, and Russia—inflation ap-pears to have closely followed broad money growthwith a lag of roughly one quarter. Most countries,however, experienced intervals of strong divergencein the growth rates of money and prices even afteraccounting for such lags.

Second, in some countries, for example, Armenia,Georgia, Kazakstan, and Turkmenistan, inflationaryexpectations and uncertainty about macroeconomicpolicies appear to have led to flight from domesticmonetary assets, particularly during the dismantling

9


Chart 1.4 . Mone y Velocity and Inverse Real Money Balances

of the ruble area and the often ad hoc introduction ofnational currencies. Third, exogenous price pres-sures have given rise to periods of high inflation notaccompanied by corresponding increases in moneygrowth. In particular, exogenous increases in importprices (energy prices and Russian nonenergy exportprices) appear to have had a large impact on infla-tion in many states in late 1993; discrete increases indomestic administered prices have also played a roleat various times.

Finally, there are strong indications that increasesin informal trade credits and arrears have contributedto the rise in velocity. In several countries—including

Azerbaijan, Belarus, the Kyrgyz Republic, andUkraine—a tightening of monetary conditions or ex-ogenous price pressures appears to have been "offset"by the creation of such credits, reducing the need, atleast for a time, to reduce nominal expenditures.

Lessons

Beyond the short term, inflation has been closelylinked to money growth in these countries over themedium term. Nevertheless, the divergence betweenthe two over shorter intervals, especially the sharprise in inflation relative to money growth in many

I OVERVIE W

10



Chart 1. 4 (concluded)

Source: Data provided by country authorities; and IMF staff estimates.Note: Money velocity is domestic broad money velocity, index (April 1992 = 1). Real money stock is real stock of domestic broad money, index

(April 1992= 10).

countries in late 1993, suggests the followinglessons. Regardless of the factors underlying the ini-tial deviations in inflation and money growth and theassociated increases in money velocity, countriesthat have achieved a subsequent reduction in veloc-ity levels have been those that established and main-tained positive rates of return on domestic monetaryassets, as well as other conditions that instilled con-fidence in the currency. Looking at the experience of

Fund-supported programs, the successful anti-infla-tion cases achieved not only low rates of monetaryand credit expansion but also positive real interestrates, which contributed to a stable or appreciatingexchange rate and a recovery in the demand formoney.18 On the other hand, stabilization efforts in

18See Section III.

I I


I OVERVIEW

Tabl

e 1.

4. T

he

Bal

tics,

Rus

sia,

and

Oth

er C

ount

ries

of th

e Fo

rmer

Sov

iet U

nion

: Inf

latio

n a

nd M

oney

Gro

wth

, Pro

gram

Per

form

ance

1

(Per

cent

age

chan

ges)

Th

e B

altic

sE

stonia

(S

BA

, 8/9

2)

Co

ns

um

er

pric

es

Pro

gra

mA

ctual

Bro

ad

mo

ne

y 2

Pro

gra

mA

ctual

Velo

city

Pro

gra

mA

ctual

Latv

ia (

SB

A, 8

/92

)3

Co

ns

um

er

pric

es

Pro

gra

mA

ctual

Bro

ad

mo

ne

yP

rog

ram

Act

ual

Velo

city

Pro

gra

mA

ctual

Lith

uania

(S

BA

, 1

0/9

2)4

Co

nsu

me

r pric

es

Pro

gra

mA

ctual

Bro

ad

mo

ne

yP

rog

ram

Act

ual

Velo

city

Pro

gra

mA

ctual

Q3

18.3

55

.8

96.1

-28

.5

31.0

55.9

26.0

57.9 —

-16

.0

121.

088

.0

65.0

56.5 —

-0.6

1992

Q4

13.2

21.8

18.1 1.7

22.0

43

.7

28.0

14.6

-2.0 3.9

30.0

95.9

25.0

79.1

33.0

-21

.4

Oth

er

cou

ntr

ies o

f th

e f

orm

er S

ovi

et

Un

ion

Bela

rus

(ST

F, 7

/93

)C

on

su

me

r price

sP

rog

ram

Act

ual

Bro

ad

mo

ne

yP

rog

ram

Act

ual

Velo

city

Pro

gra

mA

ctual

57.0

109.

6

51.0

71.3

23.0

4.0

Q1

4.0 9.0

6.1

4.7

10.0

9.8

29.0 6.7

-12

.0-1

1.9 8.0

46.5

10.0

11.7

-6.0

25.4

25.0

201.

3

70.0

61

.7

-7.0

30.0

1993

Q2

1.8

5.4

26

.7

-7.0 5. 0

2.3

14.0

25.2 2.0

-10

.6 9.0

49

.7

10.0

-8.3

-2.0

80.9

Lith

uania

(S

BA

, 1

0/9

3)

Co

ns

um

er

pric

es

Pro

gra

mA

ctual

Bro

ad

mo

ne

yP

rog

ram

Act

ual

Velo

city

Pro

gra

mA

ctual

Rus

sia

(ST

F, 6

/93

)C

on

sum

er

pric

es

Pro

gra

mA

ctual

Bro

ad

mo

ne

y2

Pro

gra

mA

ctual

Velo

city

Pro

gra

mA

ctual

Rus

sia

(ST

F,

4/9

4)

Co

ns

um

er

pric

es

Pro

gra

mA

ctual

Bro

ad

mo

ne

y2

Pro

gra

mA

ctual

Velo

city

Pro

gra

mA

ctual

Mo

ldo

va (

SB

A,

11

/93

)C

on

su

me

r pric

es

Pro

gra

mA

ctual

Bro

ad

mo

ne

yP

rog

ram

Act

ual

Velo

city

Pro

gra

mA

ctual

Q2

60

.268

.2

45.1

81.1 4.9

-5.3

1993

Q3

8.0

8.2

38

.75

4.4

-21

.3-2

6.5

36.8

89

.7

32.5

41

.4

14.1

32.9

Q4

5.5

21

.7

16.1

29

.0

-4.1

-7.7

26

.056

.6

23

.74

4.7

-12

.3 6.6

57

.08

1.9

23

.03

4.8

63

. 0 8.9

Q1 4.5

11.4

10.4

13.5

-1.5 3.1

14.0

67.6 3.0

-3.0

-61

.010

7.3

1994 Q2

4.0

10.8

15.3

-2. 6

48

.222

.9

39

.05

0.7 2.2

-23

.4

10.0

10.7

9.0

65.5

10.0

-32

. 5

Q3

36.8

18.1

43

.727

.5 3.5

6.3

6.0

3.7

19.0

15.7

131.

0-1

.6

12

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion

Key Common Issues in the Transition Process

Kyr

gyz

Rep

ubli

c (S

BA

, 4/9

3)

Con

sum

er p

rice

sPr

ogra

mA

ctua

lB

road

mon

ey

Prog

ram

Act

ual

Vel

ocity

Prog

ram

Act

ual

Kyr

gyz

Rep

ubli

c (E

SAF

, 6/9

4)

Con

sum

er p

rice

sST

F/SB

AA

ctua

lB

road

mon

ey

STF/

SBA

Act

ual

Vel

ocity

STF/

SBA

Act

ual

Mol

dova

(STF

, 9/9

3)C

onsu

mer

pric

es

Prog

ram

Act

ual

Bro

ad m

one

yPr

ogra

mA

ctua

lV

eloc

ityPr

ogra

mA

ctua

l

Q2

33.0

65.8

10.0

48.1

19.0

12.0

1993

Q3

20.0

85.0

15.0

51.4

-5.0

11.0

43.0

81.0

90.0

66.4

24.0

21.5

Q4

14.0

87.7

19.0

-7.0

-2.0

100.

2

28.0

81.9

20.0

34.8 — 8.9

Q1

9.0

67.6

-13.

0-3

.0 —10

7.3

1994 Q2

15.0

12.4

20.0

27.3

-3.6 7.0

10.7 3.0

65.5 —

-32.

5

Q3

15.0 4.8

10.0

19.6 7.0

-12.

2

Kaz

akst

an (S

TF, 7

/93)

Con

sum

er p

rice

sPr

ogra

mA

ctua

lB

road

mon

ey

Prog

ram

Act

ual

Vel

ocity

Prog

ram

Act

ual

Kaz

akst

an (

SB

A, 1

/94)

Con

sum

er p

rice

sPr

ogra

mA

ctua

lB

road

mon

ey

Prog

ram

Act

ual

Vel

ocity

Prog

ram

Act

ual

1993

Q3 48.0

102.

8

26.0

59.7

37.0

10.7

Q4 31.0

188.

8

17.0

22.4

13.0

90.3

Q1

57.0

107.

9

56.0

56.3

-1.0 1.6

1994

Q2

20.0

157.

3

15.0

69.9

24.0

66.5

Q3 7.0

55.9 7.0

46.6 8.0

Sour

ces:

Dat

a pr

ovid

ed b

y co

untry

aut

horit

ies;

and

IMF

staf

f est

imat

es.

Not

e: S

BA

= s

tand

-by

arra

ngem

ent;

STF

= s

yste

mic

tran

sfor

mat

ion

faci

lity;

ESA

F =

enh

ance

d st

ruct

ura

l adj

ustm

ent f

acilit

y.1I

nfla

tion

refle

cts w

ithin

-per

iod

quar

terly

rate

. Bro

ad m

oney

inclu

des

fore

ign

curr

ency

dep

osits

, unl

ess

othe

rwis

e no

ted

; pro

gram

figur

e de

rived

usin

g ac

coun

ting

exch

ange

rate

;act

ual f

igur

e re

flect

s ac

tual

exc

hang

era

te.

2Refle

cts

dom

estic

bro

ad m

oney

.3Fi

rst

two

quar

ters

are

for A

ugus

t sta

nd-b

y ar

rang

emen

t, ne

xt t

wo

quar

ters

are

revi

sed

Mar

ch s

tand

-by

arra

ngem

ent.

4Firs

t thr

ee q

uart

ers

are

for

Oct

ober

sta

nd-b

y ar

rang

emen

t, la

st

one i

s re

vise

d M

arch

.

13

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion

Chart 1.5 . Differenc e Between Inflationand Money Growth Rates in Selected Fund-Supported Programs, 1992-94 1

(In percentage points)

Source: Data provided by country authorities; and IMF staffestimates.

Note: SBB=stand-by arrangement; STF=Systemic trans-formation facility.

1Defined as the within-period quarterly growth rate of retailprices less the within-period quarterly growth rate of broadmoney. The monetary aggregates used are defined in Table 1.4."Q1" refers to the first program quarter for each arrangement.Baltics: Estonia SBA(8/92), Latvia SBA(8/92), Lithuania SBA(8/92).Russia STF(6/93), Russia STF(4/94). Other: Belarus STF(7/93),Kazakstan STF(7/93); SBA(1/94), Kyrgyz SBA(4/93); ESAF(6/94),Moldova STF(9/93); SBA( 11/93).

which real rates of return remained significantly neg-ative, or have only recently turned positive, have hadgreater difficulty in controlling inflation and have yetto bring about a reversal of velocity increases. To theextent that inflationary inertia has played a role, mod-erating wage pressures can also be important—in fact,those program countries (Estonia, Latvia, Lithuania,and Moldova) which were most successful in achiev-ing sustained reductions in inflation and velocity didso during periods of successful wage restraint.

Thus, while the steady and substantial reduction inmonetary expansion—supported by fiscal restraint—is both necessary and sufficient for bringing down in-flation over the medium term, the achievement ofpositive real interest rates can facilitate a more rapidresponse by prices and minimize the effects of ex-ogenous shocks. The desirability of quickly estab-lishing confidence in the currency and favorably af-fecting inflationary expectations would argue formechanisms that clearly signal policy intentions;under appropriate conditions, this could entail a com-

mitment to an exchange rate peg.19 Finally, the abil-ity of enterprises to offset monetary restraint with in-creases in interenterprise credits suggests that poli-cies to discourage an undue rise in such credits—inparticular those that signal the refusal of the govern-ment to "bail out" indebted enterprises—would makestabilization policy more effective.

Emergence of Interenterprise ArrearsAs in many other transition economies, the

process of adjustment from central planning to amarket-oriented system in the Baltics, Russia, andother states of the former Soviet Union has been ac-companied by an increase in interenterprise tradecredits and overruns on the maturity of such credits.Interenterprise trade credits and overdue obligations(or arrears) are common features in all marketeconomies, and some increase in their volumes—from very low initial amounts—was to be expectedin transition economies. It is important to distin-guish, however, between the natural adaptation tomarkets and the special factors present in a transitioneconomy that drive the accumulation of arrears dur-ing the process of transition.20 Interenterprise arrearssignal a lack of financial discipline, can retard in-vestment and the process of privatization, and cancomplicate the conduct of stabilization policies.

Developments and Main Causes

In almost every country of the region, interenter-prise arrears grew rapidly following the dissolutionof the U.S.S.R. (see Table 1.5). In Russia, for exam-ple, they amounted to 21 percent of GDP in grossterms in mid-1992. In an attempt to deal with theproblem of mounting arrears, all the countries in theregion, except the Baltics, carried out netting opera-tions in the second half of 1992 and early 1993. Thenetting operations, however, were accompanied bybailouts of the net debtors. With such action leadingto expectations of further bailouts, interenterprise ar-rears re-emerged quickly.

While interenterprise arrears rose to high levels atvarious times during 1992 and the first half of1993,21 it should be noted that, when measured in re-

19The issue of appropriate exchange rate strategies is discussedbelow.

20In this context, it should be noted that the disruption of theinterrepublican payments system in the immediate aftermath ofthe dissolution of the U.S.S.R. was a special factor that con-tributed to the accumulation of interstate interenterprise arrearsin the region.

21Data on either arrears or total credits are available only for 9of 15 states in the region. The underlying payments situation ofenterprises during this period is obscured by the impact of the

I OVERVIE W

14



lation to GDP, they remained generally low by boththe standards of established market economies andby the levels observed in some central and easternEuropean countries in similar stages of transition.22

Subsequently, that is, from mid-1993 to late-1994,it appears that the experience has been rather varied.During that time, the arrears problem declined inLatvia, Lithuania, and the Kyrgyz Republic.23 At thesame time, arrears increased rather modestly in bothRussia and Turkmenistan, although nonpayment forenergy deliveries by enterprises in these countries tothose in other states have been a significant problem.These arrears, which rose sharply after the increasein energy prices toward world market levels, haveimpeded stabilization efforts by leading to addedpressures for credits and subsidies to energy enter-prises, and complicated interstate relations with en-ergy-importing states in the region.

In contrast, interenterprise arrears increased dra-matically in Azerbaijan, Kazakstan, and Ukraineduring the period, and on the basis of data on inter-enterprise credits, in Belarus as well. In Kazakstan, asharp rise in late 1993 was followed by another jumpin mid-1994, even though the government attemptedto solve the arrears problem in early 1994 by under-taking a netting operation, extending credit to all netdebtor enterprises, and initiating an enterprise re-structuring plan. In the event, financial disciplinewas not established and a new stock of interenter-prise arrears emerged that exceeded by a wide mar-gin the previous amount.

In the initial stages of transition, managers of state-owned enterprises have little experience with the mar-ket, and institutional arrangements and incentivestructures do not change dramatically. The high levelof uncertainty regarding relative prices, the prospec-tive stance of macroeconomic policies, and the futureownership and management structure compound theproblem of inadequate incentives for efficient man-agement. In such circumstances, managers seek toavoid conflicts with the work force on employmentand wage issues, often buying raw materials and in-puts on credit and maintaining a level of productioneven if there is insufficient demand. Such behavioralso may lead to delivering output in exchange for fu-ture promises to pay to buyers with questionable

prospects. Of course, when the promises to pay arenot honored in time, arrears build up. The absence ofa sound banking and payments system, macroeco-nomic stability, and suitable laws and institutions per-petuates a management culture inappropriate for amarket economy and financial discipline.

Policies to Deal with Arrears

Policies to address interenterprise arrears mustfocus on encouraging enterprises in adjusting tomarket signals and on promoting financial disci-pline. The experience with the exceptional measures,such as netting taken in a number of countries,clearly demonstrates the temporary nature of reliefto the arrears problem provided by such measures.Netting and bailout exercises merely address symp-toms rather than underlying causes and, indeed, per-petuate and compound the problem by strengtheningexpectations of future bailouts of debtor enterprisesby the government.

In the short term, the authorities could promote asolution to the "stock" problem of existing inter-enterprise arrears by promoting a secondary marketwhere such claims could be traded. Such a solutioncould lead to a netting out of interenterprise arrearswithout any extension of government or central bankcredit and send a strong signal to enterprises andbanks that the authorities will no longer bail outfirms with arrears or bad debts.24 The involvementof governments and central banks should be limitedto providing an adequate regulatory and institutionalframework to prevent abuses such as insider tradingand manipulative practices.

In addition to avoiding any bailout component inthe measures designed to deal with existing arrears,a few other specific steps could be helpful to preventnew arrears, that is, the flow problem.

First, bankruptcy criteria and judicial proceduresshould be streamlined with a view to providing forthe closing of insolvent enterprises, but also for theirreorganization in an attempt to avoid future liquida-tion and closing. It is unrealistic, however, to expecta bankruptcy system to emerge rapidly just with theenactment of the appropriate laws.25 Until bank-ruptcy procedures work smoothly in providing en-terprises with the incentive to alter their behavior

netting operations. Also, any cross-country comparison of the ar-rears problem is subject to the caveat that the data are not strictlycomparable across countries because of differences in definitionas well as coverage (see appendix, below).

22Interenterprise arrears as proportions of GDP were 16 percentand 22 percent in 1991 in the former Czechoslovakia and Poland,respectively. The corresponding ratios to broad money were 22percent and 69 percent in the two countries.

23For Latvia, this judgment is based on the decline in total in-terenterprise credits.

24This approach played a part in the restructuring of enterprisedebt in Poland.

25Given the relative inexperience of judicial courts in bank-ruptcy cases and the limited availability of trained liquidators, andhence the uncertainty of the judicial awards, creditors are shy ofbecoming "pioneers" and incurring large start-up costs in bank-ruptcy proceedings. This slow start-up is illustrated by the experi-ence of the Czech Republic, where the number of declared bank-ruptcies went from 5 in 1992 to 60 in 1993 and to 254 in 1994.

15


Chart 1.6 . Inflatio n and Lagged Money Growth

and be financially responsible, it is essential for gov-ernments to demonstrate clearly their willingness toallow enterprises to close. The government shouldidentify a narrow set of enterprises with large arrearsproblems. If any of these enterprises are deemed tobe not viable, bankruptcy proceedings should be ini-tiated by the government. In exceptional circum-stances, if bankruptcy is not a viable option for aparticular enterprise because of strategic considera-tions, the flow problem should be dealt with by re-structuring the enterprise and identifying budgetarysupport over the medium term. Furthermore, it is im-portant to make the provision of budgetary support

conditional on, inter alia, the clearance of existingarrears and their avoidance in the future.26

Second, to ensure that managers are accountablefor all payments obligations of their enterprises (in-cluding taxes and wages), governments could im-pose financial and administrative penalties on man-agers of state enterprises, including termination of

26In Poland, more than 1,200 enterprises have been liquidatedby the government because of their financial nonviability. In late1992, the Government of Estonia initiated bankruptcy proceed-ings against a small number of state enterprises because of tax ar-rears and insolvency.

I OVERVIE W

16



Chart 1. 6 (concluded)

Source: Data provided by country authorities; and IMF staff estimates.Note: Inflation is three-month moving average retail price inflation. Money growth is three-month moving average growth of domestic broad money,

lagged one quarter.

service. Moreover, since a significant part of theblame for arrears can be placed on the inappropriatebehavior of managers of creditor enterprises, penal-ties could be applied to them as well.27 Third, forcustomers with arrears in excess of agreed limits,

state-owned enterprises could be required to shipproducts only on the basis of preshipment paymentor on the basis of promissory notes of the receiver ofgoods and services, duly guaranteed by commercialbanks.28 Finally, the timely monitoring of the under-

27In Hungary, for example, the 1991 law on bankruptcy madethe failure on the part of managers to declare bankruptcy, whenpayments were overdue for more than 90 days, punishable ac-cording to the provision of the Civil Code.

28In the former Yugoslavia in the 1980s, promissory notes witha maturity up to 90 days, which were the most important categoryof interenterprise credits, had to be guaranteed by a bank andbacked by a physical transaction.

17


Table 1.5 . The Baltics, Russia, and Other Countries of the Former Soviet Union: InterenterpriseArrears1

(End of period)

AzerbaijanKazakstanKyrgyz Republi cLithuaniaRussia

Industry an d constructio nTurkmenistanUkraine

AzerbaijanKazakstanKyrgyz Republi cLithuaniaRussia

Industry an d constructio nTurkmenistanUkraine

Memorandum itemsBelarusLatvia

BelarusLatvia

1992

1.7

34.29.2

107.852.2

19.3

23.8

204.0

Q1

1.81.99.16.8

2.6

5.812.8

105.074.6

19.9

1.5

38.5

Q2

9.11.98.78.3

2.3

0.8

19.712.8

129.1181.1

17.1

6.9

1993Q3 Q4

(As percent of GDP)1

12.63.49.0

11.5

2.2

0.7

15.05.07.98.83.02.7

1.8

Q\

13.54.0

5.44.03.62.94.6

(As percent of broad money)3

44.928.4

160.5119.726.524.1

6.7

42.881.1

250.750.233.129.663.133.2

57.835.3

248.143.849.544.666.270.3

(Interenterprise payables as percent of GDP)3

3.9 3.5 8.116.2

15.6

1994Q2

. 14. 63.7

2.62.3

56.7

256.445.548.042.981.4

13.9

(Interenterprise payables as percent of broad money)3

36.5 36.2 185.879.5

114.8 141.2

Q3

7.1

214.2227.8

54.745.5

100.8

Sources: Data provided by the country authorities; and IMF staff estimates.1Defined as overdue payables for every country except Turkmenistan, for which the data refer to overdue receivables.2Defined as average during the quarter as a percent of annualized quarterly GDP.3Defined as the ratio of the end-of-period stock of broad money.

lying financial position of enterprises as well as theirarrears to both domestic and foreign banks and en-terprises should be ensured through appropriate re-porting requirements.

Interstate interenterprise arrears related to energydeliveries are largely quasi-governmental, with vol-umes, prices, and means of payment determined onthe basis of intergovernmental agreements. Whilegovernments should take responsibility for regulariz-ing such arrears through the negotiation of reschedul-ing agreements, they should move toward removingthemselves from involvement in interstate trading ac-tivities in the future.

Revenue Decline

Another key concern is the significant deteriora-tion in government revenue that has occurred in thecountries in the region since 1991.29 With the con-traction in economic activity, revenues declined inreal terms in all states except for Uzbekistan. Ofeven greater concern, however, has been a drop inrevenues in relation to GDP. For a number of coun-tries, the cessation of grants from the union govern-

29The decline in revenues is analyzed in depth in Section V.

I OVERVIE W

18



ment following the dissolution of the U.S.S.R.played a significant role. Even excluding suchgrants, however, the revenue-to-GDP ratio is esti-mated to have deteriorated in at least 12 of the 15countries (Table 1.6). Excluding Russia, revenuesfor this group of states fell by an average of 8 per-cent of GDP in 1992-93, and a further sharp fall ofaround 4 percent of GDP is estimated for 1994. InRussia, the revenue-to-GDP ratio is estimated tohave fallen by 5 percent of GDP in 1993-94.30

Developments and MainContributing Factors

It should be stressed at the outset that revenue lev-els and the nature of the fiscal problem differ consid-erably across countries. Revenue declines may beviewed in part as a consequence of market-orientedfiscal reforms and a reduced role of government, andin and of themselves are not necessarily a cause forconcern. Indeed, the fastest reformers—the Balticstates—are among those countries where revenuesare now significantly lower than prior to reforms.Moreover, to the extent that the drop in revenues hascontributed to a widening in fiscal imbalances inmany countries, the problem may be inadequate ef-forts to reduce government spending rather than aneed to enhance receipts. Nevertheless, in a numberof cases, sharp declines in revenues may have ex-ceeded the scope for spending cuts, and thus effortsto address them should be central to correcting thefiscal problem.

The decline in revenues in relation to GDP has itsroots in a number of factors. First is the collapse in ac-tivity in the traditional sectors of the economy. Themain tax bases in the Baltics, Russia, and other coun-tries of the former Soviet Union have been state enter-prise surpluses, wages and salaries paid through en-terprises, and retail sales primarily through statestores. Accordingly, a disproportionate decline in out-put in the traditional sectors of these economies helpsexplain the declining relative yield from majorsources of taxes—that is, enterprise profit taxes, taxeson wages and salaries, and value-added and other in-direct taxes.

Meanwhile, private and informal activities, whichnow account for a significant share of output in mostcountries of the region, have not contributed signifi-cantly to tax receipts.31 Some firms in these sectors

30Due to problems associated with the succession of the Russ-ian Federation to the Soviet Union, reliable 1991 data for Russiaare not available.

31While coverage is incomplete, private and informal activitiesare at least partly covered in the national accounts data for mostcountries. To the extent that they are not fully captured, the revenuedecline relative to GDP would be even larger than indicated above.

have not been profitable in the initial years of opera-tion, while others have received substantial start-upand other exemptions from paying taxes. More gen-erally, many of these entities have not been capturedin the tax net because of weaknesses in tax policyand administration.

Shortcomings in the design of tax systems havealso led to eroding revenue. The switch fromturnover taxes to broad-based value-added taxes(VATs) and excise taxes entailed an unexpected netloss in revenues, in part because VAT liabilities wereallowed to be determined on a cash basis while VATcredits for inputs could be claimed on an accrualbasis. The effects of trade liberalization and ineffec-tive custom administration at newly established bor-ders reduced collections from trade taxes.

Numerous exemptions from all major taxes havebeen a critical factor. Throughout the region, exemp-tions have proliferated as governments have not hadthe will to impose hard budget constraints and havesought to earn the political support of powerful in-terest groups—sectoral lobbies, regional elites, for-eign investors, and vulnerable segments of the popu-lation. Governments have not fully appreciated theimportance of level playing fields in marketeconomies and also have sought to use such exemp-tions as an alternative to a social safety net.

Problems in tax administration in general haveplayed a significant role, by not allowing lost rev-enues to be replaced through the more effective tax-ation of emerging (as well as traditional) sectors. Inplanned economies, tax administration had only alimited function, with assessment essentially deter-mined by the plan and collection taking placethrough the state-owned banking sector. With re-forms, the introduction of a modern tax administra-tion became critical. The countries of the regionhave had serious problems, however, in coping withthe basic functions of tax administration—taxpayeridentification, filing and payment procedures, orga-nization of tax administration, and taxpayer educa-tion. Ineffective tax administration has led to poorcompliance, and together with the low wages of taxofficers has created an environment of corruption.Barter arrangements and multiple exchange rateregimes have also contributed to tax evasion.

The weak legitimacy of new governments has alsohindered tax collection. In some countries, uneasy re-lations with local regions has led to disputes overrevenue-sharing arrangements and delayed or causedinsufficient payments by local governments to thefederal authorities. In addition, political instabilityand the lack of a coherent and committed reformstrategy has led to a culture of misappropriation andwithholding of revenues by ministries, local govern-ments, and state enterprises.

19


Table 1.6 . The Baltics, Russia, and Other Countries of the Former Soviet Union: Evolution ofRevenue, Excluding Union Grants(In percent of GDP)

Countries with an increase in revenue-to-GDP rati o

TajikistanUkraineUzbekistan

Average

Countries with a decrease in revenue-to-GDP rati o

ArmeniaAzerbaijanBelarusEstoniaGeorgiaKazakstanKyrgyz RepublicLatviaLithuaniaMoldovaRussiaTurkmenistan

Average1

Average for all countries

1991

20.236.530.6

29.1

25.637.543.941.033.820.522.437.444.031.6

38.2

34.2

33.1

1992

26.641.531.3

33.1

26.749.042.235.115.322.916.528.133.727.037.642.2

31.4

31.7

1993

26.941.141.0

36.3

23.536.943.636.6

2.922.313.635.827.622.936.719.2

26.8

28.7

Est. 1994

42.8

42.8

18.8

39.234.7

15.713.336.724.118.532.96.2

24.0

25.7


1The calculation of the average for 1991 excludes Russia.

LessonsAddressing the revenue problem will be a com-

plex and formidable task. In the immediate periodahead, the most effective reforms will be those thatimprove the collection of taxes from those sectors ofthe economy that are already theoretically in the taxnet. This would involve, first and foremost, the re-moval or severe restriction of exemptions on alltypes of taxes. Second, every effort should be madeto capture the emerging private sector. In the shortterm, when there are difficulties in assessing tax lia-bility on new taxpayers, some presumptive elementsof taxation could be considered. Third, the agricul-tural sector, which is the largest untaxed sector in theregion, should be brought into the tax net—initiallyby removing exemptions on major taxes but also byintroducing a land tax. Fourth, trade taxes might beenhanced, for example, by applying broad-based anduniform tariffs. Fifth, an increase in petroleum andother energy taxes should be considered. Indeed, forthe region as a whole, energy-related excise taxesare significantly lower than in major countries of the

Organization for Economic Cooperation and Devel-opment (OECD); and for the energy producers in theregion, relatively little is earned from these sectorscompared with international norms. Finally, estab-lishing good governance and a taxpaying culturewill be critical. To this end, governments should es-tablish and commit themselves to a coherent tax sys-tem that is based on objective and transparent crite-ria and not subject to ad hoc adjustments. At thesame time, they should establish on their part, realis-tic budgets and observe their own spending commit-ments and obligations so as to eliminate precaution-ary and retaliatory withholding by taxpayers.

Over the medium term, although with little delay,improvements in the design of the tax system andtax administration will be important. In the designarea, the two main accounting reforms should be theintroduction of accrual accounting to the enterpriseprofit tax and the replacement of the mark-upmethod with the invoice method for calculating VATliability. Desirable reforms to tax administration in-clude improving taxpayer registration and identifica-

I OVERVIE W

20



tion procedures, taxpayer education, and introducingeffective audit procedures and adequate penalties.

Role of the Exchange RateThe question of whether a fixed exchange rate strat-

egy may be more effective in bringing down inflationin transition economies than a strategy based on mone-tary targets with a flexible exchange rate has receivedconsiderable attention in the past year in both the pub-lic debate and discussions within the Fund.

General Considerations and Experience

There are a number of reasons for looking favor-ably at fixed exchange rate approaches, includingmoney-based approaches with exchange rate flexi-bility may be less effective in the face of unstablemoney demand; an exchange rate anchor may bothsignal and help secure the end of an inflationary spi-ral, thereby enhancing confidence and helping inthe re-monetization of transition economies; and anexchange rate peg may induce a greater commit-ment to fiscal adjustment than would otherwise bethe case.

The adoption of a fixed exchange rate, however,as a stabilization tool has a few potential problems.In addition to instability in money demand, coun-tries in the region have been and continue to be ex-posed to large real and external shocks—for exam-ple, changes in the terms of trade—which are betterabsorbed under a flexible exchange rate regime.Second, while an exchange rate anchor may inducegreater fiscal discipline, it also may require greaterfiscal adjustment, since it will require a low infla-tion target. The degree of fiscal adjustment neededto sustain a fixed exchange rate may exceed theadded discipline that is likely to arise with such anarrangement. Third, the failure of an exchange ratepeg will entail significant costs, not only losses inforeign reserves associated with an attempt to savethe peg, but also typically higher inflation than be-fore the program started, mainly because of the gov-ernment's loss of credibility. By contrast, departurefrom a monetary target may be reversible, or theprogram may be adjusted without visibly signalingfailure.

The experience with alternative stabilizationstrategies elsewhere—both in transition economiesand market economies—indicates that exchange rateanchors are an effective and possibly superior ap-proach to stabilization if supporting adjustment mea-sures are adequate. However, the inadequacy of ac-companying adjustment policies—in particular,fiscal restraint—has ultimately led to failure of suchstabilization attempts in many cases. On the otherhand, money-based stabilizations have had a mixed

record. With regard to Central European transitioneconomies, all three attempts to use exchange rateanchors in stabilization were effective in bringingdown inflation quickly (Table 1.7); however, one ofthe three (Yugoslavia) subsequently failed, and infla-tion returned to very high levels. At the same time,four out of six money-based stabilization attempts inCentral Europe were successful, with quarterly infla-tion reduced to single-digit levels within one year.32

In light of the general arguments and experience,the IMF has taken a case-by-case approach in the re-gion of the former Soviet Union, with the choice ofstabilization strategy guided by judgments regardingthe specific fiscal situation, the underlying commit-ment to stabilize, the adequacy of official reserves,and the nature and size of shocks likely to prevail inthe country in question. Thus, the case of Estonia en-tailed an exchange-rate based stabilization approach,while in all other cases stabilization attempts underFund-supported programs involved flexible exchangerates.33 The main factor militating against an ex-change rate peg was an insufficient commitment tostrong restraint in fiscal policies.

The inflation record of cases in the region wheresufficient time has passed to allow an assessment ofthe stabilization attempt is shown in Table 1.7. An ex-change rate anchor was associated with a reduction ininflation in the only case in which it was attemptedduring the main stabilization phase (Estonia), whilefour out of seven money-based stabilizations can alsobe considered successes (Latvia, Lithuania, Moldova,and the Kyrgyz Republic).

The experience with flexible exchange ratearrangements in Fund-supported stabilization pro-grams outside the Baltics, which will be taken upseparately below (Belarus, Kazakstan, Kyrgyz Re-public, Moldova, and Russia) does not indicate thatan exchange rate peg would have been more appro-priate in these countries. In some of these programs,the intended disinflation, as reflected in program tar-gets, was not sufficient to sustain a fixed exchangerate. While most of these economies experiencedlarge swings in velocity that partly undermined theshort-term effectiveness of the monetary anchor,they also experienced large and unforeseen real dis-ruptions, which were probably better dealt withunder flexible rates.

Finally, these countries all failed, by a wide mar-gin in some cases, to meet fiscal targets. Without a

32The October 1993 Croatian stabilization is included as amoney-based approach here.

33Subsequent to the main price stabilization phase, Lithuaniaswitched to a currency board arrangement in March 1994, whileLatvia has maintained a de facto peg to the SDR (without formalcommitment) since February 1994, also after the main stabiliza-tion phase.

21


I OVERVIEW

Tabl

e 1.

7. C

ount

rie

s in

Tra

nsiti

on: S

tabi

lizat

ion

Att

empt

s an

d In

flatio

n P

erfo

rman

ce(A

vera

ge q

uarte

rly p

erce

nt c

hang

es)

Mo

ney

-bas

ed s

tabili

zatio

nA

lba

nia

*

Bulg

aria

Cro

atia

*

Ma

ce

do

nia

, fo

rme

r Y

ug

osl

av

Re

pu

blic

of

Ro

ma

nia

Slo

venia

*

Be

laru

s

Ka

zaks

tan

Kyrg

yz

Re

pu

blic

*

Latv

ia*

Lith

uania

*

Mo

ldo

va

*

Ru

ssia

Exc

han

ge

rate

bas

ed s

tabili

zatio

nC

ze

ch

oslo

va

kia

, fo

rme

r*

Pola

nd*

Yugosl

avi

a

Est

onia

*

Qu

art

er of

Sta

bilizatio

n

Att

em

pt

(Q0

)

III/92

I/91

IV/9

3I/94

I/91

IV/9

1

II/9

3

II/9

3

II/9

3

II/9

2

III/92

IV/9

3II/9

3

I/91

I/90

IV/8

9

II/9

2

Q1

30

.5

109.

9

36

.5

31

.8

78

.6

139.

6

14

3.7

22

2.5

47

.7

70

.5

94

.2 3.8

48

.9

13

5.8

27

9.0

Q0

51

.6

77

.4

44

.6

40

.6

63

.7

90

.0

84

.6

78

.2

69

.5

69

.0

74

.2

70

.9

34

.4

49

.2

22

5.4

61

.2

Q1

33

.3

48

.0

-1.3 8.2

40

.3

48

.5

98

.4

86

.8

72

.4

57

.4

91.1

82

.7

83

.6 9.3

23

.0

11

7.7

55

.0

Q2

12.4

17.7

-2.7 2.5

24

.2

24.1

181.

1

17

0.2

10

2.7

53

.8

57

.6

23

.3

69

.0 1.7

10.4

12.5

27

.0

Q3 1.6

11.6

0.6 6.0

34

.3

10.1

127.

1

13

7.4

50.1

12.9

61

.6 5.1

46

.6 1.5

15.4

5.2

11.3

Q4

10

.2

13.8

0.4

50

.3 8.1

84.1

12

2.8

19.5 3.4

13.5 9.1

25

.4 2.6

26.1

20

.4 7.0

Q5 5.7

18.7

26

.7 7.0

127.

0

92

.6 7. 6 1.7

18.0

18.0 1.7

11.7

17.0 5.6

Q6 3.8

12.7

16.3 3.9

13

0.6

50

.4

12.7

14.8

37.4 2.6

6.1

22.9 9.2

Q7

12.0

15.7

34.5 4.0

13.3

10.1 5.1

9.7

23.9

17.0

Q8

-3.0

18.2

36.7 6.3

5.8

8.3

12.9

58.5

11.9

Sour

ces:

Dat

a pr

ovid

ed b

y co

untr

y au

thor

ities

; and

IMF

staf

f est

imat

es.

*Suc

cess

ful s

tabi

lizat

ions

.

22

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion


counterfactual, it is impossible to say with certaintywhether these fiscal slippages would also have oc-curred under an exchange rate anchor; however, theextent and nature of the slippages suggest that theeven greater fiscal restraint that would have beenneeded under an exchange rate peg could not havebeen delivered. In Kazakstan, the main source offiscal slippage under the program supported by astand-by arrangement in 1994 related to an ill-designed and large bailout of interenterprise arrears,which was a clear indication of the magnitude ofunderlying financial imbalances and lack of com-mitment to financial discipline. In the 1993 programsupported by a stand-by arrangement for the KyrgyzRepublic, the slippage resulted from excessive cred-its to agriculture that were seen as necessary to en-sure oil and gas imports under bilateral barter agree-ments; the size of the problem and the authorities'policy response strongly suggest that a peg wouldnot have been sustained. In Moldova, the slippageson the domestic front were related to external finan-cial shortfalls and a severe drought that were out-side the control of the authorities. The money-basedapproach permitted the program to be modified asneeded without creating the perception that the au-thorities had failed, and thus did little damage toconfidence and the authorities' credibility. Finally,in Belarus and Russia, the political consensus to un-dertake the major adjustments required to supportan exchange rate peg was clearly not present, as wasreflected in failure to take necessary budgetary mea-sures as well as to restrict credit to enterprises andsectors under several Fund-supported programs in1992-94.

The cases of Estonia and Latvia permit a compar-ison of the relative merits of the two approaches incircumstances of similarly appropriate fiscal and in-stitutional preconditions. Estonia and Latvia bothsuccessfully stabilized in 1992-93, Estonia under acurrency board and Latvia under a flexible ex-change rate regime. While the degree of fiscal re-straint was comparable, deposit interest rates weresubstantially lower in Estonia than in Latvia upuntil early 1994, suggesting that the Estonian cur-rency board arrangement commanded greater credi-bility than the Latvian approach. In addition, for asimilar decline in inflation, Estonia's recorded out-put loss during 1992-94 appears to have beensmaller than in Latvia.

It is not clear, however, to what extent the differ-ence in recorded output performance is due to theexchange rate arrangements. Estonia's closer ties tothe Nordic countries may have contributed tohigher foreign direct investment inflows in thatcountry, and Latvia was slower to privatize. More-over, the output figures may be misleading sincethey attempt to account for the new private activi-

ties in Estonia, but not Latvia. On the whole, theexperience in the Baltic countries supports the viewsuggested by the Central European stabilization ex-periences, namely, that pegging the exchange ratemay be useful in stabilization if fiscal conditionsare right, but that pegging is not necessary for sta-bilizing rapidly and successfully.

Conclusions for the Period Ahead

Recent developments in a number of countries inthe region appear to have strengthened the case forrapid disinflation supported by a fixed exchangerate. With progress in price and trade liberalization,a large portion of the real shocks associated withtransition may lie in the past, and the underlyingrecognition of the need for fiscal adjustment and thetaming of inflation has strengthened in many coun-tries. Several countries (including Russia, Kazak-stan, Ukraine, Armenia, and Georgia) may soon beat a point where an exchange rate anchor could be auseful complement to appropriate financial policiesin the stabilization effort. Second, three countries,Latvia, Moldova, and the Kyrgyz Republic, wheremoney-based programs were successful and expo-sure to external shocks (particularly in Moldova)persists, may do better to continue with the prevail-ing exchange rate arrangements. Similarly, there isno case for Estonia or Lithuania to change their cur-rency board arrangements. Finally, there is a groupof countries whose stage in the transition process—and thus the magnitude of prospective real shocks—and whose underlying fiscal and political conditionssuggest that a fixed exchange rate strategy wouldstill entail an unduly high degree of risk. In all this, itshould be emphasized that political and economicuncertainties remain unusually high in this region,warranting a continual reassessment of the appropri-ate policy approach.

Role of External Financial Assistance,1992-93

During 1992-93, the Baltic States, Russia, and theother states of the former Soviet Union received atotal of approximately $69 billion of financing by therest of the world, of which about $61 billion went toRussia (see Table 1.8). In addition to the financingprovided by the IMF and the World Bank, Russia re-ceived bilateral credits—mostly from the seven majorindustrial countries—along with a comprehensivedebt relief package from official creditors under theauspices of the Paris Club, and benefited from debtdeferrals by commercial bank creditors. External fi-nancial assistance was also provided to the othercountries in the context of programs supported by the

23


I OVERVIEW

Tabl

e 1.

8. T

he B

altics

, Rus

sia, a

nd O

ther

Sta

tes

of th

e Fo

rmer

Sov

iet U

nion

: Glo

bal G

ross

Cap

ital F

lows

, 199

2-93

, Cum

ulat

ive

(In

mill

ions

of U

.S. d

olla

rs)

Arm

enia

Azer

baija

nBe

larus

Esto

niaGe

orgi

a

Kaza

ksta

nKy

rgyz

Rep

ublic

Latvi

aLi

thua

nia

Mold

ova

Russ

iaTa

jikist

anTu

rkm

enist

anUk

rain

e

Uzbe

kista

n

Tota

l

Gran

ts

286

285

145

200

198

172

125 98 210 22

5,80

0 25 110

7 14

7,68

8

EU 75 — 113 — 92 2 28 49 55 — 341 — 61 139

... 95

4

IBRD - — — 20 — - — 21 54 29 40

0 — — —..

.

525

Loan

s

EBRD

2 — 4 3 — - — 1 2 — 100 — — —

...

Ill

Bilat

era l

ROW

1 36 190 87 — 567 26 83 86 75

17,90

0 97 109

818

571

20,6

45

FDI

(Gro

ss)

- 20 24 218 11 395 10 82 93 14

2,20

0 21 43 370 57

3,55

8

IMF — 98 58 — 88 61 110

123 87

2,51

4 — — — —

3,13

8

Debt

Reli

efPr

ivate

2

- — — — — - — — — —

11,50

0 — — — —

11,50

0

Offic

ial3

- — 3 — — - — — — —

14,00

0 3 — 7 —

14,01

3

Arre

ars

ROW - — — — — — — — —

6,00

0 3 — 7 —

6,01

0

Othe

r4

- — — 149 — 1 8 — — — 500 — — — 112

770

Tota

lIn

flows 36

334

157

773

530

1

1,225 258

444

623

227

61,2

55 149

214

1,448 754

68,9

12

FSU

Finan

cing5

227

103

934 — 669

1,261 246 14 3 2 81 300

129

603

6,57

7

830

12,00

7

Tota

lIn

flows

,Inc

luding

FSU

590

445

1,511 735

970

2,48

650

445

865

530

8

61,5

55 278

817

8,02

5

1,584

80,9

19

Sour

ce: I

nter

natio

nal M

onet

ary

Fund

.N

ote:

EBR

D =

Eur

opea

n Ba

nk

for

Reco

nstru

ctio

n an

d De

velo

pmen

t; EU

= E

urop

ean

Unio

n; F

DI =

fore

ign

dire

ct in

vest

men

t; F

SU =

the

Balti

cs, R

ussi

a, a

nd o

ther

cou

ntrie

s o

f the

form

er S

ovie

t Uni

on;

IBRD

= In

tern

atio

nal B

ank

for

Reco

nstru

ctio

n an

d De

velo

pmen

t; R

OW =

rest

of t

he w

orld

.1

RO

W in

clud

es

all c

ount

ries

othe

r tha

n th

e Ba

ltics

, Rus

sia,

and

othe

r sta

tes

of th

e fo

rmer

Sov

iet U

nion

.2Pr

ivat

e de

bt re

lief f

or R

ussi

a in

clud

es d

efer

rals

and

resc

hedu

ling

by s

uppl

iers

.3O

ffici

al d

ebt r

elie

f for

Rus

sia

incl

udes

Par

is C

lub

resc

hedu

ling

, def

erra

ls, a

nd re

sche

dulin

g by

oth

er o

ffici

al c

redi

tors

inclu

ding

Kuw

ait a

nd S

aud

i Ara

bia.

4Oth

er fin

ancin

g fo

r Rus

sia i

s en

tirel

y co

mpo

sed

of p

rivat

e m

ediu

m-t

erm

cap

ital f

low

s.5FS

U fin

ancin

g in

clud

es

bila

tera

l loa

ns, a

rrea

rs, a

nd

corr

espo

nden

t acc

ount

s fo

r th

e Ba

ltics

, Rus

sia,

and

oth

er s

tate

s of

the

form

er S

ovie

t Uni

on.

24

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion

Key Common Issues in the Transition Process

IMF and World Bank and under the auspices of theEuropean Union-Group of Twenty-Four creditorcountries and Consultative Groups organized topledge additional financial support.

Many countries in the region—in particular, en-ergy importers—also received substantial financingfrom other countries of the former Soviet Union, pri-marily through two channels: loans from Russia,mostly from the Central Bank of Russia through itscorrespondent accounts during 1992 and early 1993,and a buildup of unpaid claims by Russian and Turk-meni enterprises on companies in other countries.34

Altogether, financing amounted to about $12 billionduring 1992-93.

Has Financial Assistance Been Adequate?External financing has led to a rapid accumulation

of debt both among states of the former Soviet Unionand between them and the rest of the world. Never-theless, the appropriateness of the nature of officialfinancing—in terms of its conditioning, timing, andterms—as well as the adequacy of its size has been ofsome concern. This has been particularly the case be-cause of the limited amount of private capital inflows(see Table 1.8), and the well-known potential of ex-ternal finance to alleviate the substantial short-runadjustment costs associated with transition from cen-tral planning to a market-based system.

Financial support from multilateral agencies wasgenerally correlated with progress in reform, butmuch official bilateral financial assistance was moti-vated by humanitarian assistance, political support,and financing for denuclearization, among otherthings. On balance, however, by the criterion of con-ditionality, the record in 1992-93 may be judged asfavorable. Official support was relatively highest inthe countries with early success in stabilizing cur-rencies and adopting deep market-oriented reforms(Estonia, Latvia, and Lithuania) and relatively low-est in countries at the opposite end of the spectrum(Ukraine, Belarus, and Uzbekistan). The mixed evi-dence comes mainly from Russia; while official sup-port was substantial relative to Russian GDP and rel-ative to other countries, policy success was moreuneven and fragile and not sufficient to prevent mas-sive net private capital outflows in 1992-93.

The timeliness of disbursements of external assis-tance already committed has been unsatisfactory inmany cases. Disbursements by other creditors in sup-port of Fund-supported programs have often beenslower than expected, causing unnecessary budgetary

and balance of payments squeezes. While the reasonsfor delays in disbursements were country-specific, acommon element appears to be unfamiliarity be-tween recipient country authorities and creditors, andadministrative weaknesses on the part of recipientgovernments.

Regarding the type and terms of financial assis-tance provided, the record appears uneven. Balanceof payments and structural adjustment financingboth in the form of official and commercial debt re-lief to Russia and quick disbursing support to Russiaand other countries from the IMF and the WorldBank appear to have been appropriate in the sense ofbeing directed to the right objectives. However, re-payment terms need to allow for a lengthy period oftransition with balance of payments vulnerabilty toshocks. On the part of the IMF, the repayment periodassociated with the systemic transformation facility(STF) appears adequate, but the three-to-five-yearrepayment period of stand-by resources may be tooshort unless followed up by disbursements fromlonger-term facilities, in particular the extended fundfacility arrangement (EFF), and the enhanced struc-tural adjustment facility (ESAF).35

The provision of project financing and of officialfinancing to the emerging private sectors by the In-ternational Bank for Reconstruction and Develop-ment (IBRD), European Bank for Reconstructionand Development (EBRD), and International Fi-nance Corporation (IFC) was relatively small in vol-ume, but it has begun and should, over time, assumea large share of total official financing to the coun-tries in the region.

In the area of bilateral official assistance, tied ex-port credits have been the main vehicle of noncon-cessional financing. Such credits have short repay-ment periods and are inappropriate as sources offinance for prolonged periods. Indeed, there havebeen a number of cases of tied export credit that can-not be regarded as contributing to sound economicrestructuring and productivity enhancement. Evenworse, in 1992 and 1993, short-maturity financingon commercial terms, with bullet repayments thatwill be difficult and sometimes impossible to honor,was provided from some sources, including the Eu-ropean Union, to the poorest among the countries ofthe former Soviet Union.

In the public debate, the question of adequacy offinancing volume to countries in the region has oftenbeen posed with reference to the Marshall Plan forEuropean Recovery between 1948 and 1951. Con-trary to widespread impression, however, the vol-

34For example, arrears to Russia and Turkmenistan on pay-ments for energy deliveries amounted to well over $3 billion inthe third quarter of 1994.

35An ESAF was agreed with the Kyrgyz Republic in mid-1994and an EFF with Lithuania in late 1994; negotiations are underway in several other countries.

25


ume of assistance rendered to Russia and severalother states compares favorably with the scale ofMarshall Plan aid. The $13 billion provided by theUnited States under the Marshall Plan in 1948-51,mostly in the form of foodstuffs and raw materials,was equivalent to an annual average of 2 1/2 percentof European GDP and some 1 1/2 percent of U.S.GDP, certainly a large amount to be provided by onecountry. By comparison, official new money inflowsin 1992-93 (on an average annual basis) accountedfor over 4 percent (nearly 8 percent, if official debtrelief is included) of GDP in 1992 in Russia and forabout 6 percent of GDP in 1992 in the Baltic states.Indeed, Marshall Plan assistance, unlike most offi-cial financing to the Baltic states, Russia, and othercountries of the former Soviet Union in recent times,was mainly on grant terms; nevertheless, the scale ofofficial assistance in the period of 1992-93 was con-siderable relative to its historical precedent. Further-more, none of the stabilization and reform programsagreed between countries and the IMF in 1992-93appears to have failed for lack of foreign financing.

This leaves the question whether there could havebeen more decisive success in stabilization and re-form in Russia and perhaps other countries of theformer Soviet Union if the commitment of officialassistance had been much larger. Russia's policies,however, were largely determined by considerationsthat left little room for any adjustments in responseto the availability of finance.36 As a result, the per-ception of the IMF in 1992 was that the increasedwillingness to stabilize that might be induced bysubstantially increasing financing was too small tojustify the more extensive use of scarce resources,certainly by the criteria of traditional IMF lending.These standards themselves, of course, might havebeen inappropriately strict in the case of transitioneconomies. In recognition of this view, the STF wascreated in early 1993 to allow financial support ofstructural reform measures before policies were inplace to qualify access to the IMF's larger and moreconditional facilities.

On the whole, it is hard to argue—particularlyafter the creation of the STF—that the availabilityof official financing has so far held back the reformprocess in the Baltics, Russia, and other countriesof the former Soviet Union. This said, the availabil-ity of such financing has by now become a chal-lenge. Since late 1994, it has been increasingly dif-ficult to secure adequate official bilateral financingin support of economic reform in several countries,including Ukraine, Armenia, and Georgia, where fi-nancing requirements are particularly large in view

36See also Section VIII.

of accumulated payments arrears and, in the twolatter cases, protracted military and civil conflicts.The availability of external financing is likely topose growing problems as coherent and painfulpolicies requiring financial backing are adopted inmore and more countries after years of chaotic eco-nomic contraction.

Concluding RemarksSuccess in stabilization and reform has been

mixed. The widespread hopes of a quick turnaroundin the countries of the region have been replacedwith the recognition that systemic transformationwill take time. That said, the evidence distinctlypoints to the benefits from an immediate attack oninflation accompanied by the introduction of a com-prehensive and coherent structural reform strategy.The cross-country pattern of output developmentsand the experience with efforts to reduce inflationand to deal with the problem of interenterprise ar-rears all strongly support such an approach. All theeconomies of the region where activity has begun torecover had already achieved a reasonable degree ofprice stability, and nothing suggests that output lev-els can be preserved through more expansionary fi-nancial policies.

The task of reducing inflation from extremelyhigh levels is not easy, not least since monetary man-agement is complicated by unexpected fluctuationsin velocity, or the demand for money. Experiencehas confirmed the essential importance of steadfastmonetary and credit restraint, and of resisting pres-sure for additional credits to accommodate higherprice increases. However, the problem of unstablemoney demand also argues for timely reviews of in-termediate money and credit targets, so that neededadjustments can be made in light of actual inflationdevelopments, taking due account of the normal lagsbetween previous changes in the monetary stanceand price developments. Given the underlying un-certainties, monetary policy formulation in the shortterm may also be facilitated by the use of a marketexchange rate or market-related interest rates, orboth, as intermediate indicators. Under appropriateconditions, credit policies could be governed by afixed exchange rate as the formal nominal anchor.Finally, supporting policy measures that clearly sig-nal the government's policy course, ensure financialdiscipline among enterprises, and promote wage re-straint would facilitate the disinflation process byenhancing credibility and the flexibility of goodsand labor markets. In this vein, it is important thatmeasures to deal with interenterprise arrears by en-suring that enterprises are financially responsible fortheir own actions be implemented without delay.

I OVERVIE W

26


While a fixed exchange rate may be helpful in thedisinflation process, the failure of an exchange ratepeg does entail costs, and the experience of severalcountries has shown that such an approach is not nec-essary for the achievement of stabilization objectives.Decisions on the choice of approach in this area mustcontinue to be guided by case-by-case judgments re-garding the fiscal position, the political commitmentto adjustment, and other relevant factors.

Finally, the discussion of the problem of revenueperformance also confirms the importance of strongcommitment to reform and stabilization.

AppendixData Issues on Interenterprise Arrears

The data on interenterprise credit in the Baltics,Russia, and other countries of the former SovietUnion, where available, are collected on the basis ofenterprise surveys in most of the countries. Only inAzerbaijan are banks the source of the data on inter-enterprise arrears. Interenterprise arrears for Azer-baijan are the sum of enterprises' payment ordersthat could not be executed because of insufficientfunds as of that date.37 The frequency of the data ismonthly in most countries, except in Latvia, where itis annual.

The data typically consist of enterprises' payablesto and receivables from other enterprises, including adecomposition into domestic and foreign compo-nents, as well as by sectors. Total receivables andpayables do not necessarily match because of the in-completeness of the surveys. The figures on inter-enterprise credit and arrears reported in Tables 1.5,4.1, 4.2, and 4.3 do not include the interstate compo-nents, except for Belarus and the Kyrgyz Republic.Furthermore, the figures on credit and arrears includecomponents relating to governments in Kazakstanand the Kyrgyz Republic. For these reasons, the dataon interenterprise credit and arrears are not strictlycomparable across countries.

Some countries distinguish between overdue andother categories of receivables and payables, but noinformation exists on the length of maturity overrunsfor overdues. Furthermore, the definition of when apayment becomes overdue appears to vary fromcountry to country. For example, in Russia the clas-sification of a payment as overdue is left to the dis-cretion of the reporting enterprise, but in Ukrainedebts not paid within 60 days of falling due are de-fined to be in arrears, while in Lithuania any pay-ment overdue by more than 5 days is defined to be inarrears.38 Also, arrears may be somewhat under-stated in the data for several reasons, particularly be-cause accumulated interest and penalties are not in-cluded in trade credit arrears.

37Thus, receivables and payables are the same in the Azerbai-jani data. Furthermore, the figures on arrears for Azerbaijan, re-ported by banks only, have a downward bias compared with simi-lar figures compiled for the other countries through enterprisesurveys.

38In Lithuania, for transactions without any specified paymentdate, the payment is considered to be in arrears if it is not settledwithin five days of the transaction.

Appendix

27


II Th e Declin e in OutputJonathan Anderson, Daniel A. Citrin, and Ashok K. Lahiri

The poor and deteriorating growth performanceof the economy of the U.S.S.R. from the mid-

1970s to late 1980s was a major cause of its demisein 1991.1 The 15 states that emerged from the terri-tory of the former Soviet Union launched their sys-temic transformations from central planning to mar-kets with the primary objective of reversing thisdeclining trend and improving the output perfor-mance of their economies. Since the dissolution ofthe U.S.S.R., however, recorded output has fallensignificantly in all of these states. Following a de-cline for the region as a whole of roughly 10 percentin 1991, real GDP in these states dropped by aboutone third on average in 1992-93, with a further sig-nificant deterioration in 1994 (Table 1.3).

The extent of recorded output decline has variedsubstantially across countries. Through 1994, thelargest cumulative declines were registered in both Ar-menia and Georgia, where output in that year wasroughly one third or less of its 1990 level. In contrast,in Belarus, Estonia, Uzbekistan, and Turkmenistan,real GDP in 1994 was as much as 65 percent or moreof the level prevailing in 1990. Given the poor state ofstatistical reporting in the former Soviet Union, theseestimates of output decline should be interpreted withcaution. While there is no doubt that output has fallensubstantially, it is likely that the decrease has not beenas large as indicated by the official figures. First, statis-tical reporting in the state sector has generally deterio-rated in the transition period owing to administrativedisorder. Second, growth rates may be biased down-ward by the fact that incentives to overstate productionin the planned economy have, under liberalization,given way to strong incentives to report as little rev-enue as possible, or to avoid reporting altogether.Third, production in the newly emerging private or in-formal sectors—where most new production andgrowth has been occurring—is poorly captured by of-ficial statistics.2

This section discusses the major reasons thatcould account for both the overall decline in activityand the observed differences in output performanceacross countries and identifies policy implicationsand determinants of renewed growth. It contains abrief survey of the major factors that have been hy-pothesized to play a role in the observed output de-cline, presents basic evidence across countries andsectors, and suggests some possible conclusions. Italso addresses the prospects for recovery, includingsome observations based on the recent experience inEastern Europe.

Output Declin eThe 15 countries can be placed into three broad

categories in terms of output performance since1990 (Table 1.3). A first group consists of Armenia,Azerbaijan, Georgia, Tajikistan, and Moldova.These countries have been involved in armed con-flicts and in some cases were subject to economicblockades, and have had large declines in output. Asecond group, Uzbekistan and Turkmenistan, arethose with a large share of natural resource exportsrelative to the size of their economies. These twocountries have experienced relatively small outputdeclines, as fairly stable demand and favorable pricemovements for these exports buoyed economic per-formance. A third group, the other eight countries,have recorded a wide range of output performancethat is generally between that of the first two groupsof countries.

Notwithstanding these differences, the underlyingtrend in output has been negative and subject to acommon set of forces in all 15 countries. In that re-gard, a number of explanations have been put for-ward. These include (1) sectoral shifts and aggregateshocks directly related to systemic change from aplanned economy to a market-based system open to

1See Easterly and Fischer (1993) for a description of the eco-nomic decline in the U.S.S.R.

2See Berg and Sachs (1992) and Rajewski (1993) for Poland,and Gavrilenkov and Koen (1994) for Russia. Note that biases inoutput estimation may differ between countries, especially given

the significant variations in the degree to which individual coun-tries have been able to introduce new statistical standards; thesevariations may account for some of the differentiation in individ-ual countries' performance (see, in particular, the discussion onoutput movements in the Baltics, below).

28


Output Decline

world trade; (2) institutional and political problemsarising from the disintegration of the Soviet Union;(3) the monopolistic structure of traditional indus-tries; and (4) overly contractionary monetary andcredit policies. In the following, we argue that thefirst two sets of explanations are broadly consistentwith the cross-country aggregate and sectoral datafor the region, whereas there is little support for thelatter two.

Factors Related to Economic TransitionThe changing role of the state and the liberaliza-

tion of domestic and external sector policies arelikely to have contributed to the decline in output inthese transition economies in a number of ways.3

First, during transition, the reduction in govern-ment coordination of economic activity is not in-stantly and fully replaced by market coordination.The consequent disruptions in input supplies andavailability of credit, and marketing difficulties forenterprises, have contributed to what has beentermed the "transformational recession."4 In thiscontext, the lack of a legal infrastructure and finan-cial institutions appropriate for a market economyhas been a serious impediment to the operation of amarket-based system.

Second, price liberalization implies a large struc-tural shift in production toward goods that were previ-ously in "shortage," and away from those of the tradi-tionally favored industries. The asymmetric responsesof sectors to relative price movements, with cutbacksin production taking place faster than correspondingincreases, led to declines in aggregate output. In Rus-sia, for example, high prices for fodder relative to thatof animal products led nonprivate farms to signifi-cantly reduce herd sizes during 1992. On the otherhand, in spite of large increases in the relative pricesof energy products, output of crude oil in Russia fellby 15 percent in that year. There was little evidence ofsuccessful attempts to rapidly ameliorate the supply-side constraints such as the depletion of low-costhigh-yield fields and the deterioration in the quality ofexisting wells to take advantage of higher prices. Fur-thermore, some of the benefits from radical price lib-eralization in improving the efficiency of resource al-location were postponed owing to the considerableuncertainty and confusion about relative prices at the

3A general discussion of the following issues in the context ofCentral Europe is provided in Berg (1994), Bruno (1992), Com-mander and Coricelli (1992) and Williamson (1993a and 1993b).

4See Kornai (1994). According to the survey data for Hungaryreported by Kornai, more than one fourth of the respondents con-sidered the insufficient supply of raw materials and spare parts ofdomestic origin an impediment to production until early 1990. Fi-nancing problems were also mentioned as an impediment to pro-duction until early 1992 by more than half of the respondents.

early stages following liberalization. Adjustments inrelative prices were protracted over a lengthy period,and the geographic dispersion of prices was large.5

Third, the shift of production away from tradition-ally favored sectors was reinforced by the reductionin direct or indirect government demand in a numberof industrial and military sectors. For example, his-torically, the U.S.S.R. had extended sizable exportcredit to developing countries, mainly to financearms exports. Arms exports and output of the militaryindustrial complex were adversely affected by thesharp reduction in the provision of new export creditsfrom the late 1980s.6 Output in this sector was alsoaffected by a decline in domestic demand—for ex-ample, defense spending in Russia reportedlydropped from over 8 percent of GDP in 1991 to 5 1/2percent of GDP in 1992, with sizable cuts in militaryprocurements. As a result, the share of industrial out-put in defense-related industries oriented toward mil-itary purposes reportedly fell from 40 percent in 1991to 30 percent in 1992.

In the short term, the decline of traditional sectorsis likely to be more rapid than the growth of newsectors for several reasons. Investment in new sec-tors may be particularly slow to respond because ofuncertainty regarding relative prices, property rights,legal infrastructure, and the general macroeconomicenvironment.7 In addition, even if investment in newactivities takes off quickly, this may only translateslowly into new productive capacity because of thetime required to build new plants and train labor.One would thus expect a net reduction in overalloutput during the transition period.

Fourth, the transition process has typically en-tailed modifying the tax and subsidy system—par-ticularly trade taxes, variable profit taxes, and subsi-dies—to substantially reduce, if not eliminate, thelarge variations in net tax incidence that existedunder central planning. This has implied potentiallylarge increases in profitability for some sectors andlarge decreases for others, which may also have

5For example, in Russia, the price of food relative to overallconsumer prices rose by 12 percent during 1991, a further 16 per-cent in January 1992, fell somewhat during the next few months,and rose again from June onward and ended the year 25 percentabove its December 1990 base. Similarly, prices sometimes var-ied by a factor of ten or more in state stores. See Koen andPhillips (1993) for a detailed analysis of price liberalization inRussia.

6See Christensen (1994), pp. 5-6.7Indeed, the process of privatization in most states of the region

has been delayed until long after the beginning of economic re-forms (the Baltics and Russia are clear exceptions), particularlywith regard to medium- and large-scale enterprises. Also, tradi-tional forms of organization under the plan have generally notbeen replaced with a comprehensive set of commercial codes orcourt procedures for contract enforcement of claims (see Kornai(1994) for a discussion of these issues in the case of Hungary).

29


II TH E DECLINE IN OUTPUT

translated to lower aggregate output through anasymmetric growth response as described above.8

Such changes have been manifest not only across in-dustries but across regions as well, as reflected in amarked decline in financial transfers to the poorerstates. In the former U.S.S.R., some of the centralAsian republics received grants from the union bud-get of 15 percent to 20 percent of their GDP. Trans-fers also took place through low prices for energyand raw materials, relative to world market stan-dards. In 1992, while the transfers from the unionbudget ceased, following significant upward adjust-ments in energy prices, Russia continued to provideofficial financing to the Baltics and other states ofthe former Soviet Union equivalent to around 20percent of these countries' combined GDP, and insome cases (Georgia, Tajikistan, Turkmenistan, andUzbekistan) amounting to one third or more of esti-mated GDP.9 Official financing from Russia, how-ever, was substantially reduced from 1993 onward.This exacerbated the output losses in these states byreducing their ability to pay for imported inputs.10

Finally, the liberalization of external transactionshas had important sectoral and aggregate implica-tions. Prices of tradable have moved toward worldlevels, notably for energy and other raw materials.11

According to one estimate, for the 14 states apartfrom Russia, this entailed a terms of trade deteriora-tion of some 23 percent in 1992 followed by a fur-ther worsening of over 10 percent in 1993-94.12 At1990 levels of trade, a 30 percent deterioration in theterms of trade is estimated to have had a 13 percentnegative impact on GDP on average in the 14 statesexcluding Russia.13 On the other hand, the sharp de-cline in the volume of interrepublican trade pre-vented Russia from taking advantage of the terms oftrade improvement.

At the same time, and especially for the countriesin this region, traditional Council for Mutual Eco-

8The traditional system of taxation and subsidization, and thechanges in this system under transition, is described in Holzmann(1991), Kopits (1991), and IMF and others (1991).

9See IMF (1994a).10The exact magnitude of the net impact of the cessation of

transfers on these economies is hard to estimate without detailedinformation on the composition of imports precluded by the ces-sation of transfers, and import intensities of production in the var-ious sectors.

11See IMF (1994a) for an analysis of the magnitude of relativeenergy price increases in the Baltic countries, Russia, and othercountries of the former Soviet Union. In their study of eastern Eu-ropean transition economies, Borensztein and Ostry (1993) findthat rising energy prices were particularly important in explainingoutput decline. A more general discussion of movements in theterms of trade under liberalization is provided in Tarr (1994).

12See IMF (1994a).13Trade intensities in 1990 estimated by Tarr (1994) have been

used in this calculation.

nomic Assistance (CMEA) export markets havebeen lost. The abolition of CMEA and the revisionof prices of raw materials, such as energy products,implied an estimated 4 0 - 5 percent improvement inthe terms of trade with the other CMEA countries.14

However, CMEA trade volumes collapsed, prevent-ing the Baltic countries, Russia, and other countriesof the former Soviet Union from taking full advan-tage of this terms of trade improvement. Thus, forexample, from around 60 percent of the U.S.S.R.'stotal trade in the second half of the 1980s, the formerCMEA's share dropped to less than 20 percent ofRussia's trade with the area outside the Baltic statesand other countries of the former Soviet Union in1992. Finally, pent-up demand for foreign goods hasled to substantial real depreciation as exchange mar-kets were decontrolled. This, together with the re-moval of implicit import price subsidies, has re-sulted in rising imported input costs, lowering theviability of enterprises that depend heavily on theseinputs.

In terms of systemic change, broadly speaking,the Baltic states are farthest along in the transition,followed by the Kyrgyz Republic, Russia, Moldova,and Kazakstan. The other countries have made var-ied but generally less progress. Once country-spe-cific factors, such as armed conflicts and relative im-portance of natural resource exports, are taken intoaccount, the cross-country experience regarding ag-gregate output, investment, and trade appears to bebroadly consistent with the above arguments thatstress the relationship between output decline andsystemic change.

Among the nine countries of the third group, thespeed of output decline varied directly with the speedof systemic change during the first two years of transi-tion. Indeed the output decline was almost twice asfast in the Baltics compared with Belarus and Ukrainein 1992 (Table 2.1 and Chart 1.2). The positive re-wards in output from systemic change subsequentlyaccrue with a lag, and thus, in the Baltic states, al-though greater declines in output were recorded at thebeginning of the transition period, subsequently out-put stabilized or increased. Countries, such as Belarusand Ukraine, which were slow to both stabilize andundertake structural reforms, registered smaller de-clines early on, but the fall in output has recently ac-celerated, reflecting the progressive collapse of thetraditional sectors and the lack of significant growth innew activities. Countries that liberalized relativelyearly but did not immediately stabilize (Russia and the

14Kornai (1994), Rodrik (1992), and Rosati (1993) discuss theeffects of the dismantling of the CMEA in Eastern Europe; Chris-tensen (1994) presents estimates of the loss in CMEA trade vol-ume in Russia.

30


Output Decline

Table 2.1. Rea l Gross Domestic Product 1

(Average annual percentage change)

Fast reformersEstonia, Latvia, and Lithuania

Moderate reformersKazakstan, Kyrgyz Republic, and Russia

Slow reformersBelarus and Ukraine

1991

-10.7

-10.4

-6.5

1992

-24.1

-17.4

-13.3

1993

-11.0

-13.4

-11.9

1994

3.3

-16.7

-22.3

Cumulative1991-94

-37.1

-46.7

-44.2

1Excludes Armenia, Azerbaijan, Georgia, Moldova, and Tajikistan because of armed conflicts, and Turkmenistan and Uzbekistan with large natural resource

exports relative to the size of their domestic economy.

Kyrgyz Republic) have recorded continued declinesthroughout the period. In these countries, continuedquasi-fiscal support of unprofitable enterprises in theinitial period may have slowed adjustment and thuspostponed the decline in output compared with the

Chart 2.1 . Easter n Europe, the Baltics, andOther Countries of the Former SovietUnion: Rea l Gross Domestic Product 1

(1990 = 100)

Sources: Wor ld Economic Out look database; and IMF staffest imates.

Note: Eastern Europe I = the Czech Republic, Hungary, Poland;Eastern Europe II = Bulgaria, Romania.

1Excluding Russia.

Baltics, while relatively strong reform efforts led tolarger initial output losses than for the slow reformers.

Moreover, it does not appear that countries thathave delayed liberalization and stabilization mea-sures have been able to reduce the magnitude of cu-mulative output decline. In fact, the opposite mayhave been true. Early Eastern European reformers,such as Poland, Hungary, and the Czech Republic,have recorded lower total declines in output thancountries such as Bulgaria and Romania, which havehad higher inflation and are still undertaking liberal-ization measures (Chart 2.1); the same is true whenthe Baltic states are compared with Russia and othercountries of the former Soviet Union.

In the countries of the region, systemic reformpromoted improvements in inventory management,and this in turn also had a dampening effect on out-put in the short run. Inventory levels were exces-sively high under the command system with its em-phasis on forced growth, a production structureunadjusted to user requirements, and rates of interestadministratively held at artificially low levels. Thereare problems of measuring inventories under highinflation in the countries of the region, but it appearsthat there were downward adjustments in the levelsof inventories in the fast-reforming countries of theBaltics (Table 2.2).15 For example, in Estonia, stocks

15Direct evidence for structural change—including moves fromlarge to small firms, state to private sector, and industry to ser-vices—and the consequent improvements in productivity and ef-ficiency have to rely on micro-level disaggregated data. Thus, acomparison of the pace of structural change with the pace of re-form is beyond the scope of this paper. There is evidence that asubstantial share of recorded stockbuilding represents revaluationof existing stocks rather than the physical accumulation of inven-tories, leading to serious overestimation—particularly in periodsof high inflation.

31



EstoniaFixed investmentChanges in stocks

LatviaFixed investmentChanges in stocks

LithuaniaFixed investmentChanges in stocks

RussiaFixed investmentChanges in stocks

1990

19.6-0.6

31.86.2

30.11.5

1991

22.1-8.5

29.05.3

25.114.0

1992

22.7-6.8

12.614.4

16.16.1

19.115.8

1993

25.1-0.3

13.2-0.2

17.5-1.5

21.35.5

Table 2.3. Th e Baltics, Russia, and OtherCountries of the Former Sovie t Union: RealGross Capital Formation1

(1990 = 100)

1The precision of these estimates is subject to considerable

uncertainty because of valuation problems under high inflation.

declined by more than 6 percent of GDP in 1992. Al-though the downward adjustment in the historicallyhigh level of inventories could be expected to con-tribute to improvements in the efficiency of resourceallocation in the medium to longer run, its effect onaggregate output in the short run was unambigu-ously negative. Thus, in the three Baltic countries onaverage, stock decumulation is estimated to havecontributed about 7 percentage points to the 11 per-cent decline in GDP in 1993. By contrast, stockscontinued to be accumulated in slowly reformingcountries, where enterprises were reluctant to adjusttheir work force and continued to produce and hoardgoods for which demand had disappeared.

The available data on investment are consistentwith the view that the aggregate output decline hasbeen exacerbated by sluggish investment in new sec-tors.16 Virtually every country has recorded largefalls in real gross capital expenditure, both in levelsand as a percentage of GDP (Tables 2.3 and 2.4).17

Moreover, these figures appear to mask an evenmore significant deterioration in fixed capital invest-ment, because available data indicate a sharp in-crease of about 30 percentage points on average be-

16At the same time, as with production, investment activity innontraditional sectors has likely been underrecorded in officialstatistics.

17While a contraction in real investment expenditure has con-tributed to the decline in output, it is likely that—in line with thearguments made above—much of the contraction representedlower investment in unprofitable sectors.



Central Asi aKazakstanKyrgyz Republic2

TajikistanTurkmenistanUzbekistan

OtherBelarusMoldovaRussiaUkraine

Average(unweighted)

1991

65.9

78.4

78.488.694.5

85.939.1

70.0

109.8

107.898.4

83.3

1992

63.9

40.6

18.264.439.9

60.540.7

84.0

87.8

79.588.4

60.7

1993

97.6

24.5

15.557.531.5

35.544.8

65.8

72.0

48.564.1

50.7

Est.1994

141.5

28.1

45.0

Sources: Data provided by country authorities; and IMF staff

estimates.1Except where shown, figures include changes in stocks.2Figures shown are net of stockbuilding.

tween 1990 and 1992 in the share of stockbuilding ingross capital expenditure in a number of countries.18

It should also be noted that substantial private capi-tal flight has accompanied liberalization. From virtu-ally zero in 1990 and 1991 for the U.S.S.R. as awhole, nonofficial capital outflows increased to anaverage of more than 9 percent of GDP for the re-gion in 1992, before subsiding to 5 percent of GDPin 1993.19 This acceleration may reflect the samefactors that explain low investment, namely, a lackof confidence in domestic policies and the risks im-plied by the political and economic transformation.

With regard to the relationship between develop-ments in trade and output, declines in exports to trad-

18Belarus, Moldova, Russia, Tajikistan, and Ukraine.19This measure is merely suggestive, since it includes "errors

and omissions," which can reflect both unrecorded current andcapital transactions; moreover, private capital flows in individualcountries varied strongly, from outflows of over 50 percent to in-flows of 40 percent.

32

Table 2.2. Gros s Fixed Investment andStockbuilding1

(Percent of GDP)


Output Decline

Table 2.4. Th e Baltics, Russia, and Other Countries of the Former Soviet Union: GrossCapital Formation1

(In percent of GDP)

1990

BalticsEstonia 19.0LatviaLithuania 38.0

CaucasusArmenia 23.3Azerbaijan 25.0Georgia 32.5

Central AsiaKazakstan 33.3Kyrgyz Republic2 8.5Tajikistan ...Turkmenistan ...Uzbekistan 26.6

OtherBelarus 32.5MoldovaRussia 31.6Ukraine 23.0

Average (unweighted) 26.1

1991

13.6

34.3

20.722.338.6

32.93.5

...54.418.8

36.1

39.125.7

25.5

1992

15.927.022.2

10.020.830.0

26.94.5

...57.524.9

32.057.634.927.8

22.5

1993

24.813.016.0

10.021.033.8

18.05.9

...

20.0

29.736.826.823.5

20.7

Est.1994

33.921.018.0

...

...

...

...

...

...

...

...

...

...27.0

...

...

Sources: Data provided by country authorities; and IMF staff estimates.1Except where shown, figures include changes in stocks.2Figures are net of stockbuilding.

ing partners other than the Baltics, Russia, and othercountries of the former Soviet Union (Table 2.5) ap-pear to have been positively correlated with outputdeclines across countries in 1991 (Chart 2.2), reflect-ing the loss of traditional CMEA export markets. Ex-port trends in 1992 are much more difficult to inter-pret, in part because movements in export valuesmay have been governed by relative price move-ments, and in part because of statistical breaks in theseries. Nonetheless, from the data on export volumes,the relationship is clear: countries that sufferedsmaller output declines and those even beginning torecover (Estonia, Latvia, Lithuania, and the KyrgyzRepublic) are characterized by strong export volumegrowth. The strongest export performers over the pe-riod from 1992 to 1994 have been those economiesthat pursued a relatively rapid reform strategy (Esto-nia, the Kyrgyz Republic, and Lithuania)—whichmay have promoted a greater supply response—andUzbekistan, whose favorable resource endowmentfacilitated transition to Western markets.20

20Two exceptions are Armenia and Tajikistan. While Armeniamaintained its exports by directing scarce energy supplies to in-

The impact of systemic change in the countries ofthe region is evident in the differential pattern ofoutput developments (measured by net materialproduct (NMP)) across sectors (Table 2.6). Thelargest recorded declines have been in construction,transportation, communications, and retail trade.21

The relative weakness in construction activity haslikely reflected the strong declines in investment de-mand discussed above. The declining shares of retailtrade and transportation, on the other hand, appear toreflect disruptions due to the breakup of the SovietUnion, as well as widespread movement into privateactivity in these areas. NMP statistics do not report

dustry and the production of exportables, Tajikistan's perfor-mance was related to strong market conditions in aluminum andcotton.

21Within the industrial and agricultural sectors, detailed pro-duction data also reveal substantial variations in output perfor-mance between product types (the coverage, however, is usuallyvery spotty). Moreover, sectors that have seen strong increases indemand (financial services is one widely quoted example) tend tobe outside the traditional state-owned sphere; thus, those areaswith the best output performance and potential for new invest-ment may not be captured in official statistics.

33



Table 2.5. Th e Baltics, Russia, and Other Countries of the Former Soviet Union; Valu e of TotalExports to the Rest of the World(1990 = 100)

ArmeniaAzerbaijanBelarusEstoniaGeorgia

KazakstanKyrgyz RepublicLatviaLithuaniaMoldova

RussiaTajikistanTurkmenistanUkraineUzbekistan

Average (unweighted)

1991

64.267.448.327.1

5.8

61.938.234.849.444.4

66.569.674.963.590.4

53.8

1992

11.0104.431.5

126.031.3

118.286.0

107.780.545.7

51.415.7

587.244.862.5

100.2

1993

27.048.024.2

244.843.1

124.3115.8113.6100.643.0

54.952.2

592.747.2

103.5

115.7

Est.1994

27.050.630.2

304.2

117.6124.996.3

114.645.2

60.266.5

46.2116.7

92.3


value added in other services sectors; in the Balticstates, which report sectoral breakdowns of GDP in-stead of NMP, activity in the services sector hasfallen far less on average than that in other sectors.Indeed, differences in coverage may help explainvariations in recorded output performance. In a num-ber of countries in the region, GDP is calculated onthe basis of NMP estimates, with a mark-up to ac-count for services and other activities not containedin the latter indicator. If, however, the underlyingperformance of services and other areas is morebuoyant than in basic industrial and agricultural sec-tors, GDP estimates based on NMP figures mayoverstate the decline in output. Thus, the Balticstates—and in particular Estonia—record somewhatlower GDP declines than other countries in the re-gion, while value added in agriculture and industry(which make up the bulk of NMP estimates) hasfallen by much more than in other states.

Factors Related to the Breakup of theSoviet Union

Output in these countries has also been adverselyaffected by disruptions associated with the dissolu-tion of the Soviet Union.

The breakup of the U.S.S.R. into independentstates contributed to a disruption of interstate com-

merce, which in turn is likely to have played a rolein the output decline. Most states have been slow toestablish full current account convertibility for ex-ternal transactions, preferring initially to limit accessto foreign exchange markets to official organiza-tions, often at administratively determined exchangerates (the Baltic states, Russia, and the Kyrgyz Re-public were among the earliest to liberalize currentaccount transactions); this approach was reflected inthe widespread preservation of quota and licensingarrangements for imports and exports.22

The lack of effective payments systems alsoplayed a role in hampering interstate, and in manycases also domestic, trade. Under the "ruble area"arrangements in 1992 and into 1993, interstate pay-ments were forced to take place through centralizedcorrespondent accounts, which often entailed delaysin clearing of three months or more; because of dis-array in exchange and financing arrangements, somepayments were never carried out at all.23 This situa-tion led in some cases to informal payments mecha-nisms (including reportedly large physical move-ments of cash between the Baltic countries, Russia,and other countries of the former Soviet Union),

22See IMF (1994b) for a detailed discussion.23IMF (1994a) provides a detailed description of payments

developments.

34


Output Decline

Chart 2.2. Th e Baltics, Russia, and OtherCountries of the Former Soviet Union,1991: Declin e in Rest-of-the-World Export sand Output(In percent)


barter, or outright autarky, with corresponding out-put losses.

The dramatic reduction in the volume of interstatetrade is shown in Table 2.7.24 Countries with largeoutput declines have typically also suffered fromlarge reductions in interstate trade, the most extremeexamples being Armenia, Georgia, and Tajikistan;indeed, the magnitude of decline in interstate tradehas significantly exceeded that of output for virtu-ally every country in the region.25

Given the faster decline in interstate trade relativeto output, it is tempting to attribute the output de-cline in the region mainly to the breakup of theU.S.S.R. However, the decline in interstate tradeshould be seen in the context of the economic transi-tion itself. The main factor determining the patternof interstate trade during the U.S.S.R. era was theclose specialization and integration of the republican

24Weights for interstate trade volume indices were calculatedusing 1990 relative prices; thus, the decline in interstate trade vol-ume may be overstated due to the undervaluation of raw materialstrade (which generally fell by less than other product categories).

25Armed conflicts in Armenia, Georgia, and Tajikistan were amajor reason for the large decline in trade in these states.

economies developed for strategic and self-suffi-ciency considerations in a command regime. The re-orientation of trade on the basis of economic costand benefits was a fundamental component of the re-form process in the countries of the region. For ex-ample, the sharp drop in trade with Russia and othercountries of the former Soviet Union observed in theBaltics, in part, reflects the rapid redirection of theirtrade toward the West. Thus, while the breakup ofthe U.S.S.R. itself probably did contribute indepen-dently to the fall in interstate trade and the output de-cline in the 15 countries, much of the fall must beviewed as part and parcel of the process of systemictransformation.

Other Factors

Factors other than those related to the economictransition may have also played a role in the declinein output, particularly monopolistic market struc-tures and contracting monetary and credit policies.

Monopolistic Market Structures

It has been argued, particularly in the initial yearsof transition, that the "monopolized" productionstructure in transition economies may have exacer-bated the output decline. Central planners favoredlarge-scale production concentrated in a few enter-prises. To capture monopoly rents, these enterprisesmay have lowered production and raised prices afterliberalization. While a comprehensive review of theevidence is beyond the scope of this paper, it shouldbe noted that empirical findings as to the concentra-tion of industry have yielded mixed results. Recentstudies, in particular, have contested the stylizedcharacterization of a strongly monopolized produc-tion structure at the national level.26 In general, mo-nopoly market power is unlikely to have played asignificant role in the observed output declines, par-ticularly since trade liberalization would have under-mined the ability of firms to exercise such power.

At the same time, the regional concentration of in-dustry may have contributed to output declines. Tothe extent that there were disruptions to interstatetrade, as discussed above, their impact on activitywould have been exacerbated if viable industrieswere unable to trade in inputs and components.

Contractionary Monetary andCredit Policies

Some observers have asked whether the fall inoutput has been exacerbated by "unduly tight" credit

26See, in particular, Brown, Ickes, and Ryterman (1994).

35



Table 2.6. Th e Baltics, Russia, and Other Countries of the Former Sovie t Union: Output bySector, 1993 1

(1990 = 100)

ArmeniaAzerbaijanBelarusGeorgiaKazakstan2

Kyrgyz Republic2

MoldovaRussiaTajikistanTurkmenistan

UkraineUzbekistan

Average (unweighted )Average sectoral weight in net

material product (NMP )

Memorandum itemsEstonia3

LatviaLithuania

Average sectoral weight in GDP

Agriculture

88.060.480.334.483.2

77.468.983.1

72.487.2

73.5

34.9

Agriculture

42.359.8

20.0

Industry

38.271.683.533.980.9

93.662.063.2

66.1104.4

69.7

35.9

Manufacturing

50.234.7

36.3

Construction

4.526.781.4

3.128.4

39.030.242.3

45.979.0

38.0

13.1

Construction

76.412.2

8.2

Transport

12.926.667.719.265.2

57.940.953.0

48.987.7

47.4

5.5

Services

129.067.0

33.3

Retail Trade

21.110.858.6

5.229.3

38.725.560.04

50.2

29.9

5.0

Other

30.628.183.023.915.2

136.434.3

36.9103.5

54.7

5.9

Sources: Data provided by country authorities; and IMF staff estimates.1Sectoral data are taken from NMP statistics, except for the Baltics, where the data come from GDP statistics.2Data are for 1992.31991 = 100.4Refers to activities of officially registered enterprises only.

policies, resulting in low levels of real liquidity, de-clining real working capital holdings of enterprises,and high real interest rates.27 The discussion hasbeen motivated by the fact that real domestic moneybalances have generally been observed to fall intransition economies, and as discussed in Section III,this decline has been particularly large for a numberof states in the region. The simultaneous decline inoutput and money stock in real terms (Table 2.8) hasgiven the hypothesis of unduly tight credit and mon-etary policies a priori plausibility. This plausibilityhas been strengthened by a casual examination of afew country experiences. As of 1993, for example,Uzbekistan and Turkmenistan, which saw the lowestdeclines in output, also reported the highest levels of

27See, for example, Bruno (1992), Calvo and Coricelli (1992aand 1992b), and Vienna Institute for Comparative EconomicStudies (1993).

real balances. Armenia and Georgia, on the otherhand, recorded by far the largest falls in both outputand real money balances. However, even at the levelof ordinary correlation the hypothesis fares ratherpoorly. For example, in the five countries of centralAsia, between 1992 and 1993, a significant slow-down in the rate of decline in real money stock wasnot accompanied by a corresponding improvementin output performance (Table 2.9). Similarly, in thesame five countries as well as in the three countriesin the Caucasus, between 1993 and 1994, a markedacceleration in the rate of decline in real moneystock has not been associated with a noticeable ac-celeration in the decline in real GDP.

Furthermore, declining real liquidity has not nec-essarily reflected tight credit policies. First, a goodportion of the decline in real balances in 1992 re-flects the elimination through price liberalization ofthe monetary "overhang" accumulated in previousyears. Desired money holdings at the beginning of

36


Output Declin e

Table 2.7. Th e Baltics, Russia, and OtherCountries of the Former Soviet Union:Volume of Interstate Trade1

(1990 = 100)

ArmeniaExportsImports

AzerbaijanExportsImports

BelarusExportsImports

EstoniaExportsImports

GeorgiaExportsImports

KazakstanExportsImports

Kyrgyz RepublicExportsImports

LatviaExportsImports

LithuaniaExportsImports

MoldovaExportsImports

RussiaExportsImports

TajikistanExportsImports

TurkmenistanExportsImports

UkraineExportsImports

UzbekistanExportsImports

Average (unweighted)ExportsImports

1991

5469

7587

7275

7857

4851

8664

10771

6250

7253

5159

7964

6868

10665

7185

8162

7465

1992

3824

3840

5657

3022

1219

8270

4940

4941

3537

2736

5755

1822

10175

4667

3731

4542

1993

1618

1820

4346

1710

I I17

5546

2422

1513

2115

I I22

3735

10I I

5865

2848

3527

2728

liberalization were much lower than observed levels;the real money stock in the U.S.S.R. was 35 percenthigher in 1990 than it had been in 1988, despite thefact that output had already begun to fall.28 Second,the lines of causality between output and real bal-ances run in both directions. Output could havefallen because of an ex ante decline in real liquidity;on the other hand, the demand for real balances islikely to have fallen as a result of the output decline.

Third, substantial declines in real money balanceswere reported both in countries that undertook mon-etary tightening and those that did not. In fact, thelargest declines were in countries with very highrates of growth of money in nominal terms (e.g., Ar-menia and Georgia), while the only countries wherereal balances rose were those that successfullyachieved monetary stabilization (i.e., the Balticstates), suggesting that declining real money bal-ances were the result of loose, rather than tight,monetary and financial policies (Table 2.10).

Fourth, there has been little correlation acrosscountries between the decline in output and the ob-served level of real interest rates (Chart 2.3); whilehigh positive rates of return were associated withstrong declines in output in Moldova, Russia, andUkraine, the decline in output accelerated understrongly negative real rates in Georgia, Belarus, andTurkmenistan. In the Baltic states, moreover, highreal interest rates were accompanied by relativelystable or even increasing output.29

Finally, there is little evidence in favor of the per-ceived trade-off between output growth and disinfla-tion in the transition economies of the Baltics, Rus-sia, and other countries of the former Soviet Union(Chart l.l).30 Ukraine has experienced much higherinflation than Russia, while output declines in thetwo countries have been roughly of the same order

Source: The World Bank.1In constant 1990 prices.

28The subsequent decline in the real money stock beginning in1992, however, decreased real holdings not only relative to 1990but also to 1988 or even 1982 levels; thus, the decline was proba-bly greater than could be explained by the "money overhang"alone (see Cottarelli and Blejer (1992) for estimates of the mone-tary overhang and forced saving in the Soviet Union). On theother hand, while there is no agreed theory as to whether the de-mand for real transactions balances (for a given level of output)should be higher or lower in a market economy compared with aplanned regime, it is almost certain that the real demand formoney as an asset fell substantially after 1991, as increased ac-cess to both foreign and nonmonetary domestic assets caused ashift out of domestic money balances—particularly as inflationaccelerated through the end of 1992 and beyond.

29Blanchard and Berg (1993) provide an additional argument;empirical evidence in Poland shows that inventories accumulatedin the industries where output was declining; if output werefalling owing to decreases in real credit, one would expect to seeoutput declines in those industries where inventories were beingrun down.

30See Koen and Marrese (1995) for a discussion of the econo-metric evidence for Russia.

37



Table 2.8. Th e Baltics, Russia and Other Countries of the Former Soviet Union: Rea l DomesticMoney Stock1

(1988 average = 100)



Central AsiaKazakstanKyrgyz RepublicTajikistanTurkmenistanUzbekistan

OtherBelarusMoldovaRussiaUkraine

Average (unweighted)

1990

...

...

...

...

...

...

...

...

...

...

...

...

...

134.6

1991

136.0155.6109.7

109.9152.1129.3

137.7129.3108.8140.5129.7

121.0154.8152.4129.0

133.1

1992

23.234.821.8

23.550.931.4

26.436.233.491.755.0

32.532.039.643.4

38.4

1993

26.627.211.7

10.342.017.6

17.817.229.1

129.165.3

28.219.628.726.2

33.1

Est.1994

26.434.415.0

1.414.23.3

4.37.4

13.116.1

12.712.824.212.3

14.1

Sources: Data provided by country authorities; and IMF staff estimates.1Annual average level of real balances.

of magnitude. Furthermore, growth has resumed inthe three Baltic states, which have successfullybrought inflation down.

Determinants of Future GrowthThe above discussion suggests that several factors

are likely to play an important role in determiningfuture growth prospects:

(l)To the extent that the declines in output havebeen driven by the unavoidable transitional costs ofsystemic change, recovery in aggregate output will,of course, begin as soon as growth in new indus-tries and services and in the private sector exceedsthe decline in traditional industries and the statesector. In this sense, the speed of recovery will de-pend closely on the speed of sectoral adjustmentand change. In particular, the factors governing thepace of investment in the new sectors will be cru-cial to the recovery of aggregate output. Ceterisparibus, we would thus expect recovery to be en-hanced by stabilization, privatization, and improve-

ments in the legal infrastructure as well as in finan-cial intermediation.

(2) At the same time, to the extent that the declinein output is due to political disruptions (including in-terstate trade and payments problems) and macro-economic factors, there may be scope for recoveryacross all sectors, including the traditional sectors.The removal of trade restrictions and regularizationof payments mechanisms will induce some revivalin output, as will the end of conflicts and the renewalof economic ties in countries now at war.

Some Tentative Evidenc e fromCentral Europe and the Baltics

It is still early for there to be solid evidence on theresumption of growth in transition economies. Thepath of output in both central and eastern Europeantransition economies and the Baltic economies hasnot been characterized by a strong rebound in pro-duction in traditional sectors; rather, initial declineshave been followed by relative stabilization and fi-

38


Some Tentative Evidence from Central Europe and the Baltics

Table 2.9. Rea l Money Stock and GDP atConstant Prices(Average annual percentage changes)

Table 2. 10. Mone y and Prices(Percent changes)

1All three countries were affected by armed conflicts.

nally by moderate growth after a lag of about a cou-ple of years (Chart 2.1; see also Tables 2.3 and 2.11).In those countries that have achieved an output turn-around, growth rates have been on the order of 2 per-cent to 6 percent a year. Nevertheless, the followingobservations can be made.

First, a striking feature of the output performanceof transition economies is that few countries haveachieved a resumption of output growth without firstbringing down inflation to low single-digit levels ona monthly basis. Among the five central Europeancountries included here, all the three that have suc-cessfully stabilized—Poland, Hungary, and theCzech Republic—registered positive growth in1994, with Poland experiencing sustained growth forthree years in a row. The problem of relatively highinflation has persisted in Bulgaria, and output hascontinued to stagnate. Only Romania has experi-enced some growth while maintaining high inflation.Similarly, the only states in the territory of the for-mer Soviet Union that resumed growing in 1994, thethree Baltic countries, all have undergone successfulstabilizations. On the other hand, it is also clear thatstabilization is not sufficient for the resumption ofgrowth, and that output recovery may take time even

after inflation has been brought under control. Forinstance, Hungary and the Czech Republic experi-enced relatively low levels of inflation for about twoyears before the resumption of growth. Among thecountries in the region of the former Soviet Union,Moldova and the Kyrgyz Republic may by now beconsidered as successful stabilizers, but as yetgrowth remains elusive.

Second, where new growth has been observed incentral and eastern Europe, the recovery has gener-ally been accompanied by a rise in real gross capitalformation (Table 2.11). The observed increases havetypically been undramatic, but they may mask amore decisive recovery in fixed capital investment tothe extent that the share of stockbuilding in recordedinvestment tends to fall with a reduction in inflation.Nevertheless, higher capital formation may be morea symptom of recovery and of improvement in theinvestment climate, rather than an explanation ofoutput growth in the short run. That said, it would bea prerequisite for sustainable growth, and hence thetransition, over the longer run. Apart from the prob-lem of inadequate incentives, one of the major

39

1All three countries were affected by armed conflicts.

Baltics(Estonia, Latvia, and Lithuania)

Real money stockReal GDP

Caucasus(Armenia, Azerbaijan, and

Georgia)1


Central Asia(Kazakstan, Kyrgyz Republic,

Tajikistan, Turkmenistan,and Uzbekistan)


Others(Belarus, Moldova, Russia,

and Ukraine)Real money stockReal GDP

1992

-80.2-24.1

-73.6-40.0

-62.9-15.8

-73.2-18.7

1993

-17.8-11.0

-39.2-18.8

-7.7-13.6

-29.8-11.1

1994

18.03.3

-77.9-19.4

-75.1-14.6

-39.6-20.5

Baltics(Estonia, Latvia, and Lithuania)

Real money stockNominal money stockInflation

Caucasus(Armenia, Azerbaijan, and

Georgia)1


Central Asia(Kazakstan, Kyrgyz Republic,

Tajikistan, Turkmenistan,and Uzbekistan)


Others(Belarus, Moldova, Russia,

and Ukraine)Real money stockNominal money stockInflation

1992

-80100

1,013

-74118

1,162

-63255933

-73266

1,172

1993

-18153203

-39967

2,635

-8911

1,540

-30808

1,877

1994

187152

-781,1788,309

-75220

1,355

-40362915



Chart 2.3. Th e Baltics, Russia, and Other Countries of the Former Soviet Union: Rea l InterestRates and Real GDP Growth, April 1992-Jul y 199 4

problems behind the economic decline in the lastdecade or so of the former socialist economies, in-cluding the U.S.S.R., was low productivity stemmingin part from the obsolescence of capital and technol-ogy. While depreciation of the old capital stock in-herited from the command regime may continue todampen the effect of new investment on net capitalformation in a statistical sense, technologicalprogress embodied in the new investments—particu-larly in the emerging private sector—should provideone of the major growth stimuli to these transitioneconomies. The pickup in gross capital formationseems to have not only begun in the leading countries

of the region, namely, the Baltics, but also appears tohave been more pronounced than in the central Euro-pean countries (Tables 2.3 and 2.11).

Conclusions and Polic yImplications

In addition to region-specific problems, such astrade disruptions and war, the output declines in theBaltics, Russia, and other states of the former SovietUnion appear to be primarily associated with struc-tural changes and sectoral adjustments that accom-

40


S o m e Tentative Evidence from Central Europe and the Baltics

Chart 2.3 (concluded)

Source: IM F staf f estimates.

pany the transition to a market economy and as suchare an inevitable part of the transition process. Theevidence suggests that countries attempting a grad-ual strategy have not been able to reduce the c u m u -lative output cost of transition; instead, the principaleffect of such a strategy appears to be a delay in theresumption of growth.

Output recovery has so far almost always been pre-ceded by stabilization, suggesting that controlling in-flation is an important precondition for recovery. O n

the other hand, there have been lags between stabiliza-tion and the resumption of growth of up to two years.

These findings suggest that bold and rapid stabi-lization, liberalization, and reform measures are in-deed conducive to a sustained recovery of eco-nomic activity. In particular, improvements in thelegal infrastructure and the development of finan-cial markets are likely to contribute to the n e w in-vestments that are needed for growth over thelonger term.

41©International Monetary Fund. Not for Redistribution


Table 2.11. Easter n European Countries in Transition: Selecte d Indicators

BulgariaReal GDP (1990 = 100)Gross capital formation (percent of GDP)Inflation (percent changes)1

Czech RepublicReal GDP (1990 = 100)Gross capital formation (percent of GDP)Inflation (percent changes)1

HungaryReal GDP (1990 = 100)Gross capital formation (percent of GDP)Inflation (percent changes)1

PolandReal GDP (1990 = 100)Gross capital formation (percent of GDP)Inflation (percent changes)1

RomaniaReal GDP (1990 = 100)Gross capital formation (percent of GDP)Inflation (percent changes)1

1990

100.030.414.8

100.028.6

9.6

100.025.428.9

100.029.1

585.9

100.030.24.7

1991

88.327.9

333.5

85.829.856.5

88.522.635.0

92.419.970.3

87.128.0

161.1

1992

83.319.782.0

80.124.011.1

84.719.523.0

94.815.242.9

78.332.0

210.3

1993

79.815.872.8

79.917.020.8

82.722.622.0

98.415.635.4

79.230.2

256.0

Est.1994

79.820.196.0

81.921.310.0

85.222.219.0

104.115.333.2

81.132.2

136.0

Sources: Data provided by country authorities; and IMF staff estimates.1Reflects change in consumer prices.

ReferencesBerg, Andrew, "Does Macroeconomic Reform Cause

Structural Adjustment? Lessons From Poland," Jour-nal of Comparative Economics, Vol. 18 (June 1994),pp. 376-409.

, and Jeffrey Sachs, "Structural Adjustment andInternational Trade in Eastern Europe: The Case ofPoland," Economic Policy, Vol. 14 (April 1992),pp. 117-73.

Blanchard, Olivier Jen, and A. Berg, "Stabilization andTransition: Poland 1990-1991," in NBER Conferenceon Transition in Eastern Europe, ed. by J. Sachs, O.Blanchard, and K. Froot (Chicago: University ofChicago Press, 1993).

Borensztein, Eduardo, and Jonathan Ostry, "An EmpiricalAnalysis of the Output Declines in Three Eastern Eu-ropean Countries," Staff Papers, International Mone-tary Fund, Vol. 40 (March 1993), pp. 1-31.

Brown, A., B. Ickes, and R. Ryterman, "The Myth of Mo-nopoly: A New View of Industrial Structure in Rus-sia," World Bank Policy Research Working PaperNo. 1331 (Washington: World Bank, August 1994).

Bruno, M., "Stabilization and Reform in Eastern Europe: APreliminary Evaluation," Staff Papers, InternationalMonetary Fund, Vol. 39 (December 1992), pp. 741-77.

Calvo, Guillermo, and Fabrizio Coricelli (1992a), "Stabi-lizing a Previously Centrally Planned Economy:

Poland 1990," Economic Policy, Vol. 14 (April 1992),pp. 176-226.

(1992b), "Output Collapse in Eastern Europe:The Role of Credit," IMF Working Paper, Staff Pa-pers, International Monetary Fund, Vol. 40 (March1993), pp. 32-52.

Christensen, Benedicte Vibe, The Russian Federation inTransition: External Developments, IMF OccasionalPaper, No. 111 (Washington: International MonetaryFund, 1994).

Commander, Simon, and Fabrizio Coricelli, "Output De-cline in Hungary and Poland in 1990-91," WorldBank Working Paper, WPS 1036 (Washington: WorldBank, November 1992).

Cottarelli, C , and M. Blejer, "Forced Saving and Re-pressed Inflation in the Soviet Union, 1986-1990:Some Empirical Results," Staff Papers, InternationalMonetary Fund, Vol. 39 (June 1992), pp. 256-86.

Easterly, William, and Paulo Vieira da Cunha, "Financingthe Storm: Macroeconomic Crisis in Russia," Econom-ics of Transition, Vol. 2 (December 1994), pp. 443-65.

Easterly, W., and Stanley Fischer, "The Soviet EconomicDecline: Historical and Republican Data" (mimeo-graph; Washington: World Bank, March 1993).

Gavrilenkov, G., and V. Koen, "How Large Was OutputCollapse in Russia? Alternative Estimates and WelfareImplications," IMF Working Paper, WP/94/154 (Wash-ington: International Monetary Fund, December 1994).

42


Holzmann, R., "Budgetary Subsidies in Centrally PlannedEconomies," IMF Working Paper, WP/91/11 (Wash-ington: International Monetary Fund, February1991).

International Monetary Fund, the World Bank, Organiza-tion for Economic Cooperation and Development,and European Bank for Reconstruction and Develop-ment, A Study of the Soviet Economy, Vols. 1, 2, and 3(Washington, February 1991).

International Monetary Fund (1994a), Financial RelationsAmong Countries of the Former Soviet Union, IMFEconomic Reviews, No. 1 (Washington: InternationalMonetary Fund, 1994).

(1994b), Trade Policy Reform in the Countries of theFormer Soviet Union, IMF Economic Reviews, No. 2(Washington: International Monetary Fund, 1994).

Koen, Vincent, "Measuring the Transition: A User's Viewon National Accounts in Russia," IMF WorkingPaper, WP/94/6 (Washington: International MonetaryFund, January 1994).

, and Michael Marrese, "Stabilization and Struc-tural Change in Russia, 1992-94" (mimeograph;Washington: International Monetary Fund, January1995).

Koen, Vincent, and S. Phillips, Price Liberalization in Rus-sia: Behavior of Prices, Household Incomes and Con-sumption in the First Year, IMF Occasional Paper,No. 104 (Washington: International Monetary Fund,June 1993).

Kopits, George, "Fiscal Reform in European Economiesin Transition," IMF Working Paper, WP/91/43(Washington, International Monetary Fund, April1991).

Kornai, Janos, "Transformational Recession: The MainCauses," Journal of Comparative Economics, Vol. 19(August 1994), pp. 39-63.

Rajewski, Z., "National Income," in Results of the PolishEconomic Transformation, by L. Ziekowski (War-saw: GUS/Polish Academy of Sciences, 1993).

Rodrik, D., "Making Sense of the Soviet Trade Shock inEastern Europe: A Framework and Some Estimates,"paper presented at the IMF-World Bank Conferenceon The Macroeconomic Situation in Eastern Europe,Washington, June 1992.

Rosati, D., "The Impact of the Soviet Trade Shock on Out-put Levels in Central and East European Economies,"paper presented at Conference on Output Decline inEastern Europe—Prospects for Recovery, Interna-tional Institute for Applied Systems Analysis, Vienna,November 1993.

Tanzi, Vito, "Fiscal Policy and the Economic Restructur-ing of Economies in Transition," IMF Working Paper,WP/93/22 (Washington: International MonetaryFund, March 1993).

Tarr, David G., "The Terms of Trade Effects of Moving toWorld Prices on Countries of the Former SovietUnion," Journal of Comparative Economics, Vol. 18(February 1994), pp. 1-24.

Williamson, John (1993a), "On Diagnosing the Causes ofthe Slump in Eastern Europe," in Poverty, Prosperity,and the World Economy: Essays in Memory of SidneyDell, ed. by Gerry Helleiner and others (London:Macmillan Press, 1995).

(1993b), "Output Decline in Eastern Europe: ASummary of the Debate," paper presented at Confer-ence on Output Decline in Eastern Europe—Prospects for Recovery, International Institute for Ap-plied Systems Analysis, Vienna, November 1993.

Vienna Institute for Comparative Economic Studies,"Transition from the Command to the Market Sys-tem: What Went Wrong and What to Do Now?"(mimeograph; Vienna, March 1993).

References

43


Ill Th e Behavio r of Inflation and VelocityJonathan Anderson and Daniel A. Citrin

This section examines the behavior of inflationand broad money velocity in the Baltic coun-

tries, Russia, and other states of the former SovietUnion during the transition period to early 1995.1

Over the 1992-94 period as a whole, the path of in-flation has closely tracked the growth rate of mone-tary aggregates. In many states, however, concernhas been growing over the substantial increases invelocity and declining levels of real money balances,which have been reflected in intervals of strong di-vergence between rates of inflation and monetaryexpansion. Proposed explanations for these inflationand velocity movements have included "crises ofconfidence" in the domestic currency as well as in-dependent inflationary pressures due to import priceincreases or domestic price liberalization.

Of particular interest here is whether movementsin real money balances and velocity have been driv-en by shifts in the underlying demand for real moneyholdings or by other factors. A review of the avail-able data suggests that money demand movementshave been important in some cases, but, for manycountries, independent factors such as exogenousimport price increases or adjustments in adminis-tered prices may have played a key role. Moreover,the response of inflation to changes in the stance ofmonetary policy has often been slow.

For those countries that experienced large increasesin velocity and declines in real money balances, sub-sequent reversals have been achieved through firmtightening of monetary policies and establishing posi-tive real rates of return on domestic monetary assets.These policies reduced inflation and led to an even-tual increase in confidence and rising demand for realbalances. Containing wage pressures may also havefacilitated reductions in velocity, as countries whereinitial inflationary shocks fed into a cycle of highwage growth experienced greater difficulty in control-ling inflation and restoring confidence.

The behavior of inflation and velocity in the re-gion since 1992 is reviewed, with particular empha-

sis on the role of velocity movements in stabilizationefforts under Fund-supported programs. Possible ex-planations are then given for the observed increasesin velocity as well as the diverging performance invarious countries. A discussion on policy responsesto inflationary and money demand shocks follows,with a review of country experiences in achievingsubsequent recovery in real money balances. Finally,lessons and implications for future stabilization ef-forts are discussed.

Behavior of Money, Income, andPrices, 1992-9 4

The inflation experience of the Baltics, Russia,and other states of the former Soviet Union for theApril 1992 to September 1994 period is presented inTables 3.1 and 3.2.2 Countries fall into three basiccategories: the first, comprising the Baltic states,suffered high inflation early on but then succeededin stabilizing over the course of 1993. The KyrgyzRepublic, Moldova, and Russia form a second groupwhere inflation was brought under (at least) moder-ate control during the first three quarters of 1994.3

The remaining countries have been characterized byvery high and often unstable rates of price increases,and stabilization continues to be a priority.

The behavior of inflation is related to that of theother macroeconomic variables in Table 3.1 on thebasis of a standard Fisher equation, whereby,4

P = M + V - Y , (1)

1The analysis in this section draws heavily on numerous staff re-ports on individual countries in the region over the period covered.

2April 1992 is chosen as the starting point in order to abstractfrom the large decline in real money balances associated with theelimination of monetary overhang at the beginning of that year.

3In Russia, inflation after accelerating in late 1994 started tocome down around mid-1995.

4For rates of growth over discrete time periods, the relationshipwill not be exact; thus, if A represents the percentage change overa given discrete time period, then

AP « AM + AV - AY.

44


Behavior o f Money, Income, and Prices , 1992-9 4

Table 3.1. Th e Baltics, Russia, and OtherCountries of the Former Soviet Union:Money and Prices, 1992-941

(Average monthly percent changes)


Average

Kyrgyz RepublicMoldovaRussia

Average

ArmeniaAzerbaijanBelarusGeorgiaKazakstanTajikistanTurkmenistanUkraineUzbekistan

Average

DomesticBroadMoney

6.75.79.37.2

11.313.315.213.3

24.117.320.428.618.719.222.522.119.421.4

RetailPrice

Inflation

5.45.3

11.67.4

17.015.615.916.2

38.523.924.740.326.824.630.026.616.928.0

BroadMoney

Velocity2

-1.3-0.7

1.3-0.2

3.90.7

-0.11.5

8.95.63.39.15.54.66.22.40.45.1


1Inflation, money, and velocity growth are calculated for theperiod from the second quarter of 1992 through the third quar-ter of 1994; velocity is defined on the basis of real GDP and retailprice movements.

2For Azerbaijan, Georgia, Tajikistan, and Turkmenistan, becausequarterly data on real GDP are not fully available, changes in ve-locity are proxied by rates of decline of real money balances.

with P equal to the consumer price index, M to do-mestic broad money (i.e., exclusive of foreign cur-rency holdings; the role of the latter is discussed fur-ther below), V to velocity of broad money, Y to realGDP, and a "." over a variable indicating its rate ofchange.5

Data for the two-and-a-half year period as awhole confirm that inflation in the region has beenprimarily a monetary phenomenon; inflation hasbeen strongly correlated with the rate of monetaryexpansion in every country. As a group, the Balticstates registered both the lowest rates of broadmoney growth over the period (average monthlyrate of 7 percent) and the lowest inflation (7 per-cent). The intermediate group, which experiencedsomewhat higher monthly inflation (16 percent),also recorded higher monetary expansion on aver-age (13 percent). The remaining countries in the re-gion on average recorded significantly highermonthly money growth (21 percent) and inflation(28 percent).

Over the period as a whole, changes in the veloc-ity of broad money have, in general, been muchsmaller than those in prices and money (Table 3.1).Notably, velocity tended to be more stable the lowerthe inflation rates that were recorded. In the Balticstates, velocity was virtually unchanged on average,while in the intermediate inflation group, it rose atan average monthly rate of 1 1/2 percent during theperiod. In most high-inflation countries, velocityrose significantly—at an average monthly rate of 5percent. In a few cases, the increases correspondedto extreme declines in the level of domestic realmoney balances. In Armenia and Georgia, for exam-ple, an average monthly increase of broad money ve-locity of 9 percent implied a decline in the real stockof broad money to a small fraction of the level inApril 1992 by the end of the period (Table 3.3).Within this group, only in Uzbekistan was velocityroughly unchanged over the period.

An examination of quarterly data reveals a consid-erably greater degree of variability in the short-termrelationship between money and prices. Differencesof from 15 to 30 percentage points between averagemonthly rates of inflation and money growth may beobserved in a number of countries.

Divergences in inflation and rates of monetary ex-pansion have almost always been associated withcorresponding movements in velocity (rather thanlarge swings in output).6 Indeed, on a quarterlybasis, changes in velocity have sometimes over-

5In this study, unless otherwise noted, estimates for velocity arederived as an implied index from the path of money, prices, andreal output, as in the above equation, instead of using direct esti-mates for money velocity derived from nominal GDP estimates.This is done primarily to avoid inconsistencies that arise from er-rors and uncertainties in nominal GDP figures, because (1) esti-mates of the level of nominal GDP—and thus of the level of ve-locity—can vary by wide margins and are unlikely to becomparable across countries; (2) the path of the implicit GDP de-flator underlying estimates of nominal GDP growth is in somecases markedly different from that of retail and wholesale priceindices. Real GDP trends are not shown in Table 3.1.

6Changes in velocity have accompanied not only movements ininflation and money growth but also short-term fluctuations inoutput. Some countries (Armenia, Belarus, Tajikistan, Turk-menistan, Uzbekistan) have registered large seasonal fluctuationsin real output, while in others the recorded path of output hasbeen smoother—although it should be noted that estimates of realoutput movements are tentative and subject to significant mea-surement errors. In any case, for those countries where recordedoutput exhibited seasonal behavior, there is little evidence thatthis seasonality affected the observed paths of broad money or re-tail prices; instead, there were corresponding seasonal move-ments in velocity.

45


Ill BEHAVIO R OF INFLATION AND VELOCITY

Table 3.2. Th e Baltics, Russia, and Other Countries of the Former Soviet Union: Quarterl yChanges in Macroeconomic Variables1

(Percentage changes within the quarter)

EstoniaBroad moneyRetail pricesVelocity

LatviaBroad moneyRetail pricesVelocity

LithuaniaBroad moneyRetail pricesVelocity

Kyrgyz RepublicBroad moneyRetail pricesVelocity

MoldovaBroad moneyRetail pricesVelocity

RussiaBroad moneyRetail pricesVelocity

ArmeniaBroad moneyRetail pricesVelocity

AzerbaijanBroad moneyRetail pricesVelocity

Q2

16.729.630.7

16.644.827.5

30.532.34.8

28.129.3

6.8

46.040.5

4.3

26.347.523.6

11.494.587.5

34.042.9

1992

Q3

96.155.8

-25.5

59.355.9-4.9

18.988.031.4

77.243.9

-30.3

63.324.7

-38.6

129.334.0

-43.8

59.221.343.4

33.536.6

Q4

18.121.8

1.9

1.943.745.2

84.895.9-6.9

85.289.5-2.9

90.8111.9

4.3

57.7108.8

9.6

34.7101.8-74.1

62.2130.5

Q\

6.19.06.0

5.29.8

-7.8

13.846.535.1

42.8154.663.1

17.095.270.1

47.388.418.2

73.0120.444.7

108.093.9

Q2

26.75.4

-10.3

32.82.3

-16.0

-7.949.772.9

40.965.822.6

59.745.6

-14.3

76.968.2

1.7

69.081.959.7

157.478.3

1993

Q3

19.16.4

-8.0

21.6I.I

-15.6

94.98.2

-41.8

39.685.020.9

66.481.024.7

38.189.735.6

83.058.9

175.2

28.943.2

Q4

22.011.1

-13.9

23.218.7-3.7

38.221.7

-13.8

—87.798.9

29.081.9

4.4

46.156.611.6

183.61,626.2

43.7

39.9176.0

Q1

I.I20.9

9.4

4.39.2

-3.0

13.211.43.3

19.541.615.9

-2.667.685.4

22.140.3

0.8

240.1212.5-23.9

51.2111.6

1994

Q2

2.75.0

19.8

14.05.0

-1.3

15.910.8-3.1

28.612.4

-10.2

72.210.7

-34.7

50.722.9

-13.7

65.3185.8190.4

51.2105.4

Q3

2.57.28.2

7.94.11.0

15.16.70.8

19.74.8

-12.4

15.73.7

-1.6

30.518.1

-11.5

20.28.2

63.937.7

whelmed movements in money in their "explanatorypower" over prices.

Of particular note is the fact that movements in realmoney balances and velocity have generally occurrednot in smooth trends, but rather in large discreteshifts; moreover, bouts of high inflation and velocitygrowth occurred at approximately the same timeacross a number of countries. During late 1993 andearly 1994, velocity rose substantially (often two- tothreefold or more) in the space of one to two quartersin ten countries in the region, that is, all except for theBaltic states, Russia, and Uzbekistan (Chart 1.4).7 In

7Movements in velocity have, by and large, been mirrored bythe behavior of real money balances.

most cases, these increases in velocity were associ-ated with a substantial acceleration in inflation, forexample, in Armenia, Georgia, Tajikistan, and Turk-menistan. Only in a few instances (e.g., the KyrgyzRepublic and Moldova) was the rise associated withstable inflation and a decline in the rate of monetaryexpansion.

Also of interest is the protracted nature of move-ments in velocity and real balances in many countries;in some, velocity has declined only slowly followingthe initial increase, while in others it has remained atits new, higher level. Only in Lithuania and Moldovahave sharp swings in velocity and real balances beenfollowed by a quick return to previous levels.

A review of the experience of Fund-supported pro-grams further underscores the ability of money growth

46


Behavior o f Money, Income, and Prices , 1992-9 4

Table 3.2 (concluded)

BelarusBroad moneyRetail pricesVelocity

GeorgiaBroad moneyRetail pricesVelocity

KazakstanBroad moneyRetail pricesVelocity

TajikistanBroad moneyRetail pricesVelocity

TurkmenistanBroad moneyRetail pricesVelocity

UkraineBroad moneyRetail pricesVelocity

UzbekistanBroad moneyRetail pricesVelocity

Q2

84.849.6-8.3

17.241.6

59.698.753.7

79.825.0

103.255.7

-15.1

1992

Q3

41.033.915.2

121.941.9

98.645.1

-24.1

65.146.6

61.738.1

56.246.3-5.9

Q4

85.678.113.0

46.472.2

87.0106.5-4.6

39.631.0

178.931.2

91.685.3

-17.9

Q1

73.676.0

-45.2

62.0104.8

69.2133.130.0

88.999.3

184.4218.2

106.1172.4

17.2

23.461.1

-70.6

Q2

83.088.4

5.8

139.4133.8

109.266.2-6.5

87.9157.1

82.8122.8

80.8170.823.0

116.662.1-9.4

1993

Q3

51.0109.668.8

99.9132.8

59.5102.821.6

166.5157.1

76.462.9

206.6201.9

19.2

94.943.659.0

Q4

40.1201.3166.8

230.5581.0

21.0188.894.8

61.7465.3

1.2752.0

45.0360.5131.8

69.9162.777.7

Q1

115.684.2

-59.1

88.4337.0

56.3107.9

1.6

-8.9-42.3

104.497.7

51.441.9-3.4

77.676.2

-77.6

1994

Q2

80.197.815.6

238.1325.4

69.9157.366.5

64.675.0

61.415.9-9.7

36.385.7

108.3

Q3

91.2143.771.1

52.7340.9

46.655.9

46.4109.8

72.012.4

-25.9

90.84.5

53.7

Sources: Data provided by country authorities; and IMF staff estimates.1Broad money refers to domestic currency M2; velocity is defined as an index based on the methodology described on page 45, footnote 5.

and inflation paths to diverge substantially in the shortterm. Table 1.4 shows the behavior of money, prices,and velocity under 13 arrangements (and several Fundfacilities) in eight different countries in the region.8 Inmost cases, inflation outcomes under stabilization pro-grams exceeded program targets by wide margins,with the rate of inflation either remaining high or evenaccelerating during the program period. The only ini-tial stabilization programs whose inflation targets weremet or close to being met were the stand-by arrange-

8While successor or second stand-by arrangements are in-cluded in the sample for four countries where the reduction of ex-tremely high inflation remained a program priority, successorstand-by arrangements for Estonia and Latvia are not includedbecause these programs aimed at consolidating progress already

ments for Estonia and Latvia (both introduced in Au-gust 1992); in the Kyrgyz Republic, Moldova, andLithuania, inflation was reduced as envisaged underfollow-up arrangements, but the first attempts at stabi-lization failed.9

In five of the seven stabilization programs whereprice objectives were not attained—Belarus, Kazakh-stan (1993), Kyrgyz Republic, Lithuania, and

made in stabilization, rather than achieving a marked reduction ininflation. Also, to ensure comparability with program definitions,figures for broad money velocity used when reviewing programoutcomes and shown in Table 1.4 are defined on the basis of nomi-nal GDP.

9Under the 1994 program in Russia supported by a systemictransformation facility, inflation objectives were achieved initially,but end-1994 inflation targets were missed by a substantial margin.

47



Table 3.3. Rea l Domestic Money Stocks, asof Third Quarter of 1994(1992 : Second quarter - 100)

1As of the first quarter of 1994.

Moldova—higher-than-expected inflation occurred inthe face of much lower contemporaneous rates ofbroad money growth and reflected marked and unan-ticipated increases in velocity. In Kazakstan (1994),the poor inflation outcome was associated with bothexcessive monetary expansion and an unanticipatedrise in velocity. Only in Russia (1993) was the poorinflation outturn mainly associated with similarlyhigh levels of monetary expansion.

Indeed, the behavior of inflation and velocity con-trasted sharply to program projections. While the pro-grams generally recognized that inflation would notrespond immediately to monetary tightening, and thusprojected a rise in the velocity of broad money in thefirst quarter of the program, velocity was subsequentlyexpected to remain broadly stable or to decline.10 Theprojections for real money balances beyond the initialperiod were even smoother, as velocity was generallyprojected to offset seasonal fluctuations in output. Inthe event, velocity rose and real balances fell sharply

10When data have been unavailable and extreme changes in fi-nancial and economic structures have occurred, projections ofmoney demand or velocity have been subject to unusually high de-grees of uncertainty. Particularly at the outset of the transitionperiod for the Baltic states, Russia, and other countries of the for-mer Soviet Union, empirical work on the demand for money intransition economies was extremely limited and did not exist forthe newly independent states. Thus, program projections werebased on an analysis of recent trends and contained a significantjudgmental element.

for several quarters in almost all of the cases wherestabilization was not achieved.

Explaining Inflation andVelocity Movement s

We now turn to examining the possible factors un-derlying the observed behavior of inflation and ve-locity. Here the focus is on those periods when largeshifts in velocity and real balances and velocity oc-curred, then the subsequent path of velocity andmoney in the region is examined. The analysis isbased on a standard aggregate demand and supplyframework that provides a general depiction of theshort- and long-term determinants of the rate of in-flation.11 Under such a framework, the rate of mone-tary expansion determines the rate of inflation inlong-run equilibrium. In the short run, however, in-flation can depend on a number of other factors, giv-ing rise to fluctuations in real money balances andthe velocity of broad money. First, inflation maysimply have an inertial component, resulting in alagged response of prices to changes in monetary ex-pansion. Second, changes in the demand for moneydue, for example, to changes in its underlying deter-minants (e.g., a rise in inflationary expectations) willbe reflected in movements in the velocity of broadmoney (i.e., short-run deviations in inflation relativeto the rate of monetary expansion). Third, factorssuch as lax wage policies, exogenous import priceincreases and changes in domestic relative pricesowing to changes in administered prices can put up-ward pressure on prices and lead to diverging pathsof inflation and money growth in the short run. Fi-nally, observed differences in the paths of inflationand monetary expansion can reflect improper mea-surement of one or both of the variables.

Inflation and Rate of Growth ofBroad Money

The lack of contemporaneous correlation betweenmoney and price movements is likely to reflect—atleast in part—the lagged response of inflation tochanges in the rate of monetary expansion. Lags inthe response of inflation to changes in financial poli-cies are common in all economies; a particular con-tributing factor in the Baltics, Russia, and otherstates of the former Soviet Union has almost cer-

11The usefulness of standard models is, of course, questionablein transition economies, where nominal variables, such as prices,interest rates and exchange rates, may not clear markets—that is,where "disequilibrium" conditions may apply. Where possible,these factors are accounted for below.

48

EstoniaLatviaLithuaniaKyrgyz RepublicMoldova

RussiaArmeniaAzerbaijanBelarusGeorgia

KazakstanTajikistanTurkmenistanUkraineUzbekistan

Index

135110652658

837

233910

17391

2038

167


Explaining Inflation and Velocity Movements

tainly been the high volatility of money growth. Inmany countries, growth rates of the monetary aggre-gates have varied at times by up to 50 percentagepoints or more on a month-to-month basis. Undersuch conditions, the short-term behavior of mone-tary aggregates may provide little information as tounderlying trends. Thus, not only will price andwage expectations be likely to be based, at leastpartly, on the past behavior of inflation, but expecta-tions of the future stance of monetary policy arelikely to react slowly to shifts in the observed ratesof money growth.

Accordingly, changes in the rate of monetary ex-pansion would generally result first in short-termfluctuations in real money balances and velocity, andonly subsequently in changes in the rate of inflation.This has likely been particularly true under stabiliza-tion efforts, which envision often dramatic changesin the path of money growth; a significant monetarytightening might not be perceived (or be consideredcredible) until well into the stabilization program.Allowing for lags in the response of inflation tomoney growth does seem to yield a "good fit" in theobserved relationship for a few countries—in partic-ular, for Latvia, Moldova, and Russia (Chart 1.6). Inthese countries, inflation appears to have closely fol-lowed broad money growth with a lag of roughlyone quarter.12

For most other countries, however—in particular,Armenia, Azerbaijan, Belarus, Georgia, Kazakstan,and Turkmenistan—intervals of strong divergence inthe growth rates of money and prices remain evenafter accounting for such lags. Moreover, althoughthe growth rates of prices and broad money havetended to converge following such periods of sharpdifferences, the effects of the previous deviations onvelocity have not been reversed.13 Instead, as notedabove, the sharp increases in velocity registered inthe latter half of 1993 generally were not followedby corresponding declines in 1994, with the excep-tion of Moldova and to a lesser extent Ukraine.

Thus, for many of the countries in the region,prices appear to have responded to factors other than

12For Russia, simple regressions of monthly inflation againstcurrent and lagged values of ruble broad money growth indicatethat the bulk of the impact of money growth on prices was feltwith a two-to-four month lag. See Vincent Koen and MichaelMarrese, "Stabilization and Structural Change in Russia,1992-94," IMF Working Paper, WP/95/13 (Washington: Interna-tional Monetary Fund, January 1995).

13To the extent that a lagged price response led to inflation wellin excess of contemporaneous rates of monetary expansion, onewould normally expect that the resulting rise in velocity wouldonly be temporary, and that, ceteris paribus, real money balanceswould be restored in subsequent months; indeed, inflation wouldfor some time decline to a level below the rate of growth ofmoney.

changes in monetary policy in the short run. In thatregard, possible explanations are changes in the de-mand for real money balances, as well as exogenousmovements in the exchange rate, wages, the terms oftrade, and administered prices. These factors are ex-amined below.

Factors Affecting the Underlyin gDemand for Money

Inflationary pressure not related to changes inmonetary policy may be caused by an exogenous de-crease in the underlying demand for real money bal-ances (i.e., a shift in the demand curve)—related, forexample, to a loss of confidence in the domestic cur-rency. The demand for real money balances is speci-fied as a standard function of the relative rate of re-turn to domestic monetary assets as compared withother assets p, the level of real activity (Y), and ashift parameter (y/) that acts as a proxy for such fac-tors as changes in the level of uncertainty regardingthe future relative return to monetary assets, orchanges in the internal or external convertibility ofdomestic money. The relative rate of return to do-mestic money is defined as the nominal rate of re-turn (i)—equal to the deposit rate of interest for non-cash holdings and zero for cash holdings—lesseither the return to domestic real assets (proxied bythe rate of inflation, n) or the return to foreign mone-tary assets (proxied by the rate of nominal exchangerate depreciation, e). With the relative underdevelop-ment of domestic bond markets in these countries, itis assumed that there are no alternative domestic fi-nancial assets. Thus,

[y ] =M(pd,pf,Y,¥),

where pd = i - n, pf = i - e.

(2)

From this money demand function and the Fisherequation shown above, an equation for velocity as afunction of the determinants of money demand maybe derived.

Movement out of domestic monetary assets canhave an impact on the price level both indirectly, ascapital flight leads to exchange rate depreciation,and directly through precautionary goods hoarding.

The observed relationship between real balancesand real interest rates for countries in the regiontends to support the role of money demand shifts inexplaining inflation and velocity movements. An ex-ogenous decline in the demand for real money bal-ances would be expected to put upward pressure on

49



the rate of return on domestic monetary assets.14

And, indeed, by plotting movements of the realmoney stock together with those of the real interestrate, it would appear that, for the region as a whole,large swings in real balances have been accompa-nied by inverse movements in the rate of return onmonetary assets (Chart 3.1).15 In particular, for eachof the ten countries that recorded large increases invelocity in late 1993 and early 1994, declines in realmoney balances were associated with increases inthe observed real interest rate.

Other evidence, however, is less supportive of therole of money-demand shocks. First, while for a fewcountries it is easy to point to direct factors that couldhave caused a decline in real money demand, for mostothers there is no obvious explanation. For example,the introduction of national currencies could easilyhave led to a crisis of confidence, particularly if therewere no strong signals as to the future stance of mon-etary policy. In fact, in Armenia and Turkmenistan,and to some extent in Kazakstan, the acceleration ininflation and rise in velocity at the end of 1993 oc-curred at roughly the same time as the introduction ofnew national currencies. Other states, however, intro-duced new currencies with no effect on velocity (Es-tonia, Latvia, Uzbekistan), or registered significantvelocity movements either much earlier or much laterthan the currency reform (Azerbaijan, Belarus, Kyr-gyz Republic, Lithuania, Moldova, Ukraine).16

In addition, declining demand for domestic moneydoes not appear to have reflected increasing infla-tionary expectations caused by earlier accelerations

14By contrast, if real balances declined owing to supply-sidefactors such as exogenous price pressures—that is, a shift in thesupply of real money balances and a movement along the demandcurve—the relative rate of return to monetary assets would be ex-pected to fall.

15The relative rate of return in Chart 3.1 is defined as pd thenominal interest rate on refinance credit less the contemporane-ous rate of inflation. It should be noted that focusing on interestrates of the banking system may be improper when a substantialportion of broad money consists of cash holdings (as is so inmany of the economies in this study); under such circumstances,pd should be redefined as the negative of the inflation rate. Ameasure of relative return that included the return on foreign as-sets (pf) would be equally valid. However, analysis of the rela-tionship between real money balances and these alternative mea-sures of the rate of return does not lead to qualitatively differentfindings from those contained in Chart 3.1.

16A number of countries (Azerbaijan, Belarus, Georgia,Moldova) issued new currencies or coupons that initially circu-lated in parallel (and at a fixed par) with the Russian ruble; inthese instances, a "crisis" in confidence in the domestic currencymay have occurred later, after July 1993, when Russia demone-tized pre-1993 ruble notes and at the same time significantly re-duced the provision of new interstate credits—causing the abovestates to declare their currencies sole legal tender and leadingmany countries to finally delink the national currency from theruble.

in monetary expansion.17 In fact, most of the tencountries that experienced bouts of strong inflationaccompanied by velocity increases (Armenia, Azer-baijan, Belarus, Georgia, Kazakstan, Turkmenistan,and, to some extent, Ukraine) did so during periodswhen the rate of money growth had been roughlystable or falling for months (Chart 1.6).18

Finally, the relationship between inflation and ex-change rate behavior in these countries does not sug-gest that shifts in money demand in favor of foreigncurrency were significant in explaining increases ininflation and velocity. As mentioned, money demandmovements in favor of foreign currency would beexpected to affect the rate of inflation throughchanges in the nominal exchange rate. If flight fromdomestic currency were the primary cause of infla-tion, however, then the rate of nominal exchangerate depreciation would be expected to be as high, orhigher, than the inflation rate. In fact, an examina-tion of the period from late 1993 to early 1994, whenthe rate of inflation in these countries was muchhigher than that of broad money growth, reveals thatdeclining real money balances were associated withan appreciation of the real exchange rate (Chart3.2).19 The main exception seems to be Georgia,where the real exchange rate depreciated dramati-cally as real balances fell in late 1993. (The evidenceis very different for periods of decreasing velocityand rising real balances; these cases are discussedfurther below).

Thus, the broad evidence on the role of money de-mand shocks is mixed. While there are strong argu-ments in favor of money demand movements in afew country cases (Armenia, Georgia, Kazakstan,Turkmenistan), for the remaining countries no firmconclusions can be drawn.

Influence of Exogenous FactorsA third explanation for velocity movements re-

lates to the role of exogenous price pressures, which

17Periods of strong monetary expansion would fuel both actualinflation and inflationary expectations, leading to decreased con-fidence and lower money demand.

18It also does not appear that money growth in excess of pro-gram targets led to declining confidence in the currency in thosecountries with Fund-supported programs. An examination ofTable 1.4 and Chart 1.6 shows that programs where monetary tar-gets were missed still achieved a relative tightening comparedwith preprogram levels of monetary expansion.

19Care must be taken when interpreting movements in recordedexchange rates. In some cases, official exchange rates were fixedadministratively at unrealistically low levels, while domestic cur-rency prices of imports were based on black market rates. Thus,for example, observed exchange rate movements in Azerbaijan,Tajikistan, Turkmenistan, and Uzbekistan have had little effect oninflation.

50


Explaining Inflation and Velocit y Movement s

Chart 3.1 . Relativ e Interest Rates and Real Balances(Left scale: real interest rate; right scale: real balances)

can give rise to periods of high inflation not accom-panied by corresponding increases in moneygrowth.20 Exogenous changes in nominal wages—owing, for example, to adjustments in minimumwages or public sector wages that may not be di-rectly linked to labor market conditions—could alsohave an independent influence on inflation. Hereagain, however, the aggregate data suggest that ex-ogenous wage pressures generally were not an inde-

20In contrast to money demand shocks, exogenous inflationarypressures would affect the supply of real balances.

pendent factor pushing up prices. As Chart 3.2clearly indicates, when real balances fell markedlyin the second half of 1993 and early 1994, realwages also declined in each of the ten relevant cases.

In contrast, exogenous increases in import pricesappear to have a large impact on inflation in most ofthe ten states. For those that were importers of en-ergy products, the primary external price shock hasbeen the large increase in energy prices in intrare-gional trade between 1992 and 1994. In some cases,prices for energy products increased tenfold or morein U.S. dollar terms, and with the notable exceptionof the Baltic states, most of this jump appears to

(Continued on next page.)

51




Sources: Data provided by country authorities; and IMF staff estimates.Note: Real interest rate is the central bank refinance rate minus three-month moving average consumer price inflation; real money stock is the real

domestic stock of broad money (April 1992 = 1).1The real interest rate is calculated using commercial three-month interest rates.

have occurred in the latter half of 1993 (Table 3.4).21

Another important factor for the region was im-ported inflation from Russia: the substantial rise inthe real exchange rate of the Russian ruble againstthe U.S. dollar implied estimated increases of

21For the four countries (Belarus, Kazakstan, Moldova, andUzbekistan) for which quarterly price data are available, naturalgas prices tripled and oil prices rose by 2 1/2 times during the sec-ond half of 1993.

around 200 percent in the dollar prices of importsfrom Russia during the second half of 1993.22

Accordingly, these increases in external prices arelikely to have made an important contribution to the

22Close trading partners such as Belarus and Kazakstan, in par-ticular, may have suffered from large increases in the prices ofnon-energy imports from Russia, while countries that were moresuccessful in redirecting trade toward western markets (such asthe Baltic states and Moldova) would have been less affected.

52


Explaining Inflation an d Velocit y Movements

Chart 3.2. Rea l Balances, Real Wages, and the Real Exchange Rate

rise in broad money velocity across much of the re-gion in the latter part of 1993. For Estonia andLatvia, where energy import prices had alreadyreached world levels by the middle of 1992, strongincreases in velocity were not registered during theperiod covered in this paper. In Lithuania, energyprices rose dramatically during 1992 to reach worldmarket levels; at the same time, the velocity of broadmoney rose by about three fourths.

In contrast, in Russia, the system of export quotasallowed domestic prices of energy to remain low inthe face of marked increases in prices of energy ex-ports. Furthermore, the real appreciation of the Rus-

sian ruble worked to dampen rather than add to under-lying domestic inflationary pressures. This could helpexplain why in the case of Russia there was only a rel-atively small unanticipated increase in velocity.

Another independent influence on the short-termrate of inflation and velocity has likely been adjust-ments in administered prices. Available data indicatethat in late 1993 Azerbaijan, Belarus, Georgia, Kazak-stan, Moldova, Turkmenistan, and Ukraine all regis-tered significant increases in domestic administeredprices for food and energy products, and transport andcommunications services. In the absence of fulldownward flexibility of nonregulated prices, these

(Continued on next page.)

53




Sources: Data provided by country authorities; and IMF staff estimates.Note: Real wage is the average wage deflated by retail price index; the real exchange rate is the nominal U.S. dollar exchange rate deflated by retail

price index (+=depreciation).1Real wage calculated using minimum wage.

increases in regulated prices—which ranged from twoto ten times in a single month—tended to push up theoverall price level beyond what would be implied bythe rate of monetary expansion.23

23It should be noted that in many cases the adjustments in ad-ministered prices were associated with the increases in externalenergy prices.

Exogenous price movements, like those men-tioned above, would be reflected only in temporaryincreases in inflation and velocity. If velocity rosebecause of exogenous short-term increases in infla-tion, then one would normally expect subsequentpressures for realignment in money markets and arecovery in real money balances. Thus, indepen-dent price shocks cannot of themselves explain the

54


Table 3.4. Energ y Import Prices(In U.S. dollars) 1

Explaining Inflatio n and Velocity Movement s

1992 1993 1994

Sources: Data provided by country authorities; and IMF staff estimates.1Unless otherwise noted.2ln thousands of cubic meters.3The data shown for the first quarter are for the first half of the year.4Domestic price.5Price of imports from Russia.6Price of imports from Kazakstan.

ArmeniaNatural gas2

Oil and oil products (per ton)

AzerbaijanNatural gas2


BelarusNatural gas2

Oil (per ton)Gasoline (per ton)

EstoniaNatural gas2

Gasoline (per ton)Diesel fuel (per ton)

GeorgiaNatural gas2


KazakstanNatural gas2,4

Oil (per ton)5

Gasoline (per ton)4

Diesel fuel (per ton)4

LatviaEnergy price index

LithuaniaNatural gas2


MoldovaNatural gas2


UkraineNatural gas2


UzbekistanNatural gas2,4

Oil (per ton)5

Gasoline (per ton)5

Diesel fuel (per ton)6

Memorandum itemDollar index of Russian

non-energy goods prices

1stQ1

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

100

57

19

...

...

2781

...

2ndQ2

...

...

...

...

...

...

...

193I553

1033

...

...

...

...

...

...

176

2937

1646

...

...

22343

8

100

3rdQ3

...

...

...

...

...

...

...

69253205

...

...

...

...

...

...

256

5825

1454

...

...

24366

8

101

4thQ4

...

...

...

...

...

...

...

85208166

...

...

...

...

...

...

307

111120

344

...

...

173

12227

86

1stQ1

...

...

...

...

195497

79218164

...

...

2829

10561

290

97104

1787

...

...

210013758

116

2ndQ2

...

...

...

...

175884

76229161

...

...

1720

10868

268

82103

15100

...

...

767

157102

122

3rdQ3

...

...

...

...

3459

112

...

...

...

...

...

3248

16991

274

88114

1088

...

...

23103205221

202

4thQ4

...

...

...

...

5873

140

...

...

...

...

...

59106229177

299

80114

44131

...

...

20116177379

291

1stQ1

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

284

7886

...

...

...

...

...

...

...

...

325

1992

828

109

91820

58205158

194

8.........

210

5147

847

944

23660II

...

1993

8094

2625

3261

109

11114162

4781

3451

15399

283

87109

22109

5085

1396

169190

183

55



relative permanence of velocity movements in manystates. However, independent price pressures couldlead to more prolonged velocity movements if (1)initial shocks gave impetus to a cycle of further in-flationary pressures, for example, through domesticwage movements, or (2) money market pressureswere accommodated through the creation of infor-mal liquidity. The possible role of wages in prolong-ing velocity swings is discussed later.

Role of Interenterprise Credi tIf informal claims, such as interenterprise credits or

arrears, are a sufficiently close substitute for bankingsystem liquidity, then price movements will reflectnot only the growth of monetary aggregates but alsothe creation of informal claims. In particular, periodsof significant expansion in informal credits could bereflected in diverging rates of inflation and moneygrowth and thus increases in observed velocity.24 Thewidespread presence of informal credit, moreover,could sustain a high level of nominal expendituresand thus extend the adjustment period, helping to ex-plain the protracted nature of velocity movements inmany countries in the region.

Indeed, there are strong indications that suchcredits have contributed to upward movements invelocity. For a number of countries for which a con-sistent time series exists—Azerbaijan, Belarus, theKyrgyz Republic, Russia, and Ukraine—there is arather striking relationship between the fall in realbalances and the rise in interenterprise credits inlate 1993 (Chart 3.3). Other countries (e.g.,Moldova, Kazakstan, and Turkmenistan) have alsoreported large run-ups of informal credits and ar-rears during periods of falling real balances and ris-ing money velocity.25

Measurement Error sA final explanation for the observed velocity

movements in the region would be errors in measur-ing both money and price movements. With regardto the money stock, prior to the introduction of na-

24As opposed to the underlying level of "liquidity velocity"—conceptually measured to include both informal credits and broadmoney—which would tend to be more stable.

25See Section IV for additional details on arrears developments. Amore rigorous approach would be to directly construct a series for"alternative liquidity," containing both standard monetary aggre-gates and informal credits, and compare movements in this liquidityindicator with observed price changes. The available literature,however, provides no clear guidelines as to what definition of inter-enterprise credits (i.e., the outstanding stock or the current flow;overall receivables or a definition that excludes overdue arrears)would correspond conceptually to standard monetary aggregates.

tional currencies, official monetary statistics werebased on officially reported shipments of cashrubles from Russia and excluded unrecorded inter-state flows of pre-1993 cash rubles that may havealtered the true magnitude of cash in circulation. Inthe event, a number of states that still accepted pre-1993 cash rubles as legal tender in the fall of 1993experienced inflows of these notes as neighboringcountries moved to national currencies. In mostcases, these inflows are not likely to have had a sig-nificant impact. In a few instances, however, theymay have been quite substantial. Armenia andTajikistan, in particular, reported large inflows dur-ing the last two months of 1993, as most of theirneighbors introduced their own currencies; these in-flows are likely to have contributed significantly toa 500 percent increase in the price level in Novem-ber in Armenia and a 200 percent rise in Decemberin Tajikistan.

Measuring velocity movements using the domes-tic money stock would also create distortions if therewere significant currency substitution, that is, flightto foreign currency holdings for use both as an assetand for transactions purposes. While data on foreigncurrency holdings are notoriously sketchy,26 their in-clusion in the definition of broad money does reducesomewhat the magnitude of observed velocityswings for a few countries (such as Belarus andGeorgia; see Chart 3.4). However, the effect is notnearly large enough to explain the entire change invelocity. And for most other countries, the inclusionof foreign currency provides little additional ex-planatory power.

On the price side, a contributing factor to the mea-sured rise in velocity may be the upward bias tomeasured inflation that is implied by traditional So-viet price indices, that is, the so-called Sauerbeckproblem.27 This measurement error is estimated tohave had a significant impact on the measured rise in

26Available data on domestically held foreign currency gener-ally include foreign exchange deposits but do not include cashholdings; in addition, there is little information as to foreign cur-rency held abroad.

27This problem arises when price indices are calculated usingweighted averages of individual price changes in a "chained" fash-ion (rather than computed in reference to a fixed base period); seefor example, Francois Lequiller and Kimberly D. Zieschang, "Driftin Producer Price Measurement in FSU Countries," IMF WorkingPaper, WP/94/35 (Washington: International Monetary Fund,1994). In addition to methodological issues in calculation, price in-dices in these economies are also subject to inadequacies in thearea of coverage; in particular, goods traded in the illegal or semile-gal sectors—or even, for some countries, the nonstate sector ingeneral—are not included in the indices. It is not clear, however,whether these deficiencies in coverage would necessarily lead tobiases that could help explain the observed velocity trends.

56


Differences i n Velocit y Performanc e

Chart 3.3. Rea l Balances and Interenterprise Credi t

Sources: Data provided by country authorities; and IMF staff estimates.Note: The real money stock is the three-month moving average growth rate of real domestic broad money stock; informal credit is the three-

month moving average growth rate of interenterprise credit (series smoothed to exclude effects of netting operations).

velocity in Moldova, for example, prior to the con-struction of a corrected price index.28

Differences in Velocity Performanc eThe above section discussed possible explanations

for the observed shocks to real balances and veloc-ity. Here, policy responses and the subsequent be-havior of velocity are analyzed. While two of the tencountries (Moldova and Ukraine) that experiencedsharp increases in velocity in late 1993 to early 1994

28See J. Haley and G. Shabsigh, "Monitoring Financial Stabi-lization in Moldova: The Role of Monetary Policy, InstitutionalFactors and Statistical Anomalies," IMF Paper on Policy Analysisand Assessment, PPAA/94/25 (Washington: International Mone-tary Fund, December 1994).

have seen velocity decline close to previous levels,in others (especially Armenia, Azerbaijan, Belarus,Georgia, Kazakstan, and Turkmenistan), velocityhas remained high. In Lithuania, velocity fell backfollowing a rise that occurred in late 1992-early1993.

The following points emerge from the discussion:first, regardless of the source of the initial inflationand velocity increases, countries where velocity sub-sequently fell achieved that result by raising the de-mand for real balances through establishing high ratesof return on monetary assets and other conditions thatinstilled confidence in the currency. Second, the con-tainment of wage pressures may also have facilitatedreductions in velocity, as countries where initial infla-tionary shocks fed into a cycle of high wage growthwere those that experienced more difficulty in con-trolling inflation and restoring confidence.

57



Chart 3.4. Broa d Honey Velocity Including Foreign Currency

Recovery of Real Money Deman dWhile it has been suggested earlier that nominal

exchange rate movements did not appear to play adominant role in explaining initial surges in inflationand velocity, strong evidence exists that increasedconfidence in the currency has played a key role insubsequent reductions in velocity and accompanyingincreases in real balances. As indicated in Chart 3.4,recent periods of increasing real money balances inthe Kyrgyz Republic, Moldova, and Russia, as wellas the earlier experience of the Baltic states, featureda stable or appreciating real exchange rate, which

suggests that reductions in velocity were associatedwith strengthened demand for domestic currencyand a containment of capital flight.

Looking specifically at the experience of Fund-supported programs, it is clear that the four success-ful anti-inflation programs presented in Table 1.4featured relatively stable or even appreciating nomi-nal exchange rates together with declining velocity(Table 3.5). In Estonia and Latvia, exchange ratesstabilized at the beginning of the 1992 stabilizationprograms, contributing to a quick decline in the rateof increase in traded goods and overall consumer

58


Differences in Velocity Performanc e


Sources: Data provided by country authorities; and IMF staff estimates.Note: Velocity is domestic broad money velocity, index (April 1992 = 1). Velocity including foreign exchange is total broad money velocity including

foreign currency deposits, index (April 1992 = 1).1Velocity is defined as the inverse of the real money stock.

prices.29 In the Kyrgyz Republic, Lithuania, andMoldova, the sharp reductions in inflation undertheir second stabilization programs have been ac-

29In addition to the favorable direct impact on prices, the fixedexchange rate regime in Estonia is likely to have had favorable in-direct effects on stabilization. See Section VI for an analysis of therelationship between exchange rate management and stabilization.

companied by a halt to the substantial exchange ratedepreciations that were registered during the initialprograms.

At the same time, in the remaining programs,where inflation was not reduced as envisaged, unan-ticipated increases in velocity were accompanied bynominal exchange rate depreciation at a rate that wasin general significantly higher than assumed under the

59



Table 3.5. Quarterl y Changes in Wages, Exchange Rates, and Real Interest Rates UnderFund-Supported Programs1

(Percent change)

EstoniaAverage wageMinimum wageExchange rateReal interest rate (level)2

LatviaAverage wageMinimum wageExchange rateReal interest rate (level)2

LithuaniaAverage wageMinimum wageExchange rateReal interest rate (level)2

BelarusAverage wageMinimum wageExchange rateReal interest rate (level)2

KazakstanAverage wageMinimum wageExchange rateReal interest rate (level)2

Kyrgyz RepublicAverage wageMinimum wageExchange rateReal interest rate (level)2

MoldovaAverage wageMinimum wageExchange rateReal interest rate (level)2

RussiaAverage wageMinimum wageExchange rateReal interest rate (level)2

Q2

58.8

3.4

107.8123.9

6.5-55.0

79.750.519.1

1992

Q3

50.5

-7.4-26.3

21.026.723.1

-38.3

36.524.188.0

Q4

18.2—

14.3-6.2

56.457.9-6.3-7.0

10.6—

55.1-47.1

Q1

20.7——1.3

7.3—

-17.618.5

54.838.228.9

-35.9

50.3177.858.7

-95.8

1993

Q2

7.6—4.64.0

8.8—

-5.712.2

33.033.2

-11.6-13.3

94.080.063.6

-69.4

46.3—

25.7-53.1

101.190.055.0

-A8.9

Q3

15.721.4-A.2

3.5

70.936.499.4

-99.8

88.8—

46.8-93.5

69.275.051.2

-51.1

113.2150.085.6

-32.3

70.881.110.3

-32.9

Q4

40.226.3-6.5

4.0

145.0122.283.7

-104.1

159.6233.3

99.4-105.4

68.014.323.5

-11.3

61.133.378.3

-22.6

74.688.9

6.7-5.3

Q1

19.24.22.6

44.650.0

160.6-53.1

23.933.3

216.0-62.6

6.640.645.725.0

37.935.09.3

22.1

16.7—

40.616.5

1994

Q2

1.35.6—

122.9100.051.2

-53.0

185.3150.0117.1-32.7

25.128.9-3.421.0

33.033.3

1.828.6

25.9—

13.525.0

Q3

98.066.7

104.7-74.5

83.050.010.9

-20.1

4.117.2-6.231.1

8.4—3.5

13.5

22.040.232.416.6

Sources: Data provided by country authorities; and IMF staff estimates.1Changes in average wages, minimum wages, and the exchange rate (an increase in the latter indicates depreciation) reflect within-period quarterly rates.

Except where indicated, the real interest rate is defined as the average central bank refinance rate (simple quarterly basis) less the contemporaneous moving

average quarterly rate of inflation.2Defined using commercial three-month lending rate (quarterly basis).

60


Differences i n Velocit y Performanc e

program.30 This weakening in exchange rates is likelyto have contributed to the poor inflation outcomes inthese countries by increasing the prices of tradablesand adversely affecting inflationary expectations.

Exchange rate and velocity movements, in turn,have been strongly influenced by domestic financialpolicies—in particular, by the achievement of (orfailure to achieve) positive real rates of return to do-mestic monetary assets. This point is brought out inChart 3.1; those countries where rates of return wereallowed to reach positive levels for a significant pe-riod of time (Estonia, the Kyrgyz Republic, Latvia,Lithuania, Moldova, Russia, Ukraine) stopped ve-locity increases and initiated a subsequent reversaland recovery in money demand. Countries wherereal rates of return have remained significantly nega-tive (Azerbaijan, Belarus, Georgia, Tajikistan, Turk-menistan)—or have only recently turned positive(Armenia, Kazakstan)—have yet to bring inflationand velocity increases under control.31 Thus, whilethe achievement of positive real interest rates is not asufficient condition for reducing inflation, it cansubstantially strengthen stabilization efforts by stem-ming capital flight and reducing pressure on the ex-change rate, as well as on domestic prices.

The role of financial policies is underscored bythe experience of Fund-supported programs. Wheninflation was brought down successfully, interestrates reached levels that were positive in real terms.In Estonia and Latvia in 1992-93, and in the succes-sor arrangements for the Kyrgyz Republic, Lithua-nia, and Moldova in 1994, central bank refinancerates became positive in real terms, at least by thethird program quarter. In all of these cases, exchangerates stabilized or even appreciated in nominalterms, and domestic inflation declined substantially.

In the remaining cases, when inflation was not re-duced as targeted, interest rates remained negative inreal terms in all countries except Russia.32 In thatcountry, under the 1993 program supported by a sys-temic transformation facility inflation remainedhigher than targeted in the fourth quarter of the year,but monetary policy was tightened significantly, therefinance rate became positive in real terms during

30Programs generally assumed either a constant real exchangerate (i.e., nominal depreciation equal to the rate of inflation) or anappreciation in real terms (reflecting the expected effects of do-mestic financial policies).

31The analysis incorporated data until the end of 1994. In somecountries, for example, Georgia, there was a sharp deceleration ininflation around mid-1995.

32It may be noted that while actual performance was mixed, allprograms did aim at domestic interest rates reaching positive reallevels, through (1) increases in the central bank refinancing rate;(2) the introduction or widening of central bank credit auctions;and, in a few cases, (3) the liberalization of commercial bank loaninterest rates.

that quarter, and inflation fell during the first half of1994. In Belarus, both Kazakstan programs, andLithuania's October 1992 program supported by astand-by arrangement, real interest rates remainednegative throughout the program. While refinancerates were raised significantly in nominal terms, theseincreases were insufficient relative to the extremelyhigh rates of inflation, contributing to weakness in thecurrency and rising velocity. Under the first programsin the Kyrgyz Republic and Moldova in 1993, refi-nance rates were negative in real terms throughoutmost of the program and inflation objectives were notattained. However, interest rate increases during theprogram served to set the stage for reduced inflationand a stabilizing of the currency in the period immedi-ately following the end of those programs (i.e., ratherquickly under the successor arrangements).

Wage Development sWhile it was suggested earlier that exogenous

wage movements cannot be considered a primarycause of the increases in inflation and velocity thatbegan in the second half of 1993, there are some in-dications that high increases in nominal wages in re-action to the initial inflation shock sustained and fu-eled inflationary pressure in a number of countries,undermining efforts to reduce inflation and increasereal money balances.

This possibility is brought out in the experienceof Fund-supported programs in the region, wherethe failures in achieving inflation targets were ac-companied by excessive wage increases. All pro-grams sought to achieve wage restraint, and gener-ally envisaged wages in real terms to either remainconstant or decline over the program period.33 Onlyin the successful stabilization efforts, however (theinitial programs in Estonia and Latvia, and the suc-cessor programs in the Kyrgyz Republic, Lithuania,and Moldova)—as well as in the second Russia pro-gram supported by the STF— were actual wage in-creases roughly in line with targeted rates of infla-tion and monetary expansion. In the remainingprograms, rates of wage growth were significantlyhigher than the corresponding inflation targets—de-spite more ambitious monetary efforts in somecases. Moreover, in most countries, average wagesrose broadly in line with minimum wages, suggest-ing that the authorities may have been attempting to

33Proposed measures included (1) direct control of the wagebill in the budgetary sector; (2) limits on increases in the mini-mum wage; and (3) tax-based incomes policies governing wagesin state enterprises, usually taking the form of a maximum admis-sible average wage that was a certain multiple of the minimumwage. The exception was Russia, where programs contained noexplicit objectives or measures with respect to the path of wages.

61



maintain real wages in the face of generalized pres-sures on prices. Those program countries that weremost successful in achieving sustained reductions invelocity (Estonia, Latvia, Lithuania, Moldova) didso during periods when wage increases were rela-tively restrained.

Conclusions and Implications forStabilization Policy

The findings of this study suggest the followinglessons for stabilization policy. Sustained reductionin broad money growth is fundamental to inflationstabilization; the experience of the Baltics, Russia,and other countries of the former Soviet Union con-firms the primary role of monetary policy in deter-mining the medium-term rate of inflation. In theshort run, however, the relationship between infla-tion and money growth can be weaker. This wasespecially so in late 1993-early 1994 when infla-tion rose sharply in a number of countries, withoutthere having been an acceleration of money growth.First, there are normal lags in the transmissionprocess; second, exogenous factors such as risingimport prices or strong wage or administered priceincreases can have a significant influence on theprice level. And third, changes in the underlying de-mand for money, particularly in conditions of insta-bility and uncertainty, can manifest themselves inexchange rate and price behavior. While determin-ing the exact cause of price movements is difficult,this study has found evidence that all three factors

have, at various times, had a strong influence on thebehavior of inflation velocity. Moreover, it hasproved difficult to subsequently restore domesticreal balances.

While a reduction in the rate of monetary expan-sion is both a necessary and sufficient condition forbringing down inflation, there may be other factorsthat can assist in speeding the response of prices andminimizing the effects of exogenous shocks. The ex-perience of successful stabilizers suggests that risingdemand for domestic money balances has been par-ticularly important. Thus, the achievement of posi-tive real interest rates can stem capital flight and ex-change rate depreciation; increasing domesticinterest rates or committing to exchange rate stabil-ity can also have a signaling effect on expectationsthat will support monetary tightening. At the sametime, avoiding unwarranted nominal wage growthcan prevent overadjustment to external price in-creases or domestic price liberalization.

Finally, the ability in many of these countries tooffset declining real money balances with increasedcreation of informal credit has also weakened thelinkages between money and prices, often leading torun-ups of interenterprise arrears in the short run. Alarge overhang of arrears can give rise to pressuresto reverse monetary policy and may have associatedreal costs as well; thus, policies to prevent an en-dogenous response of informal credit creation (andin particular those that signal the refusal of the gov-ernment to "bail out" indebted enterprises) wouldmake stabilization policy more effective.

62


IV Interenterpris e ArrearsAshok K. Lahiri and Daniel A. Citrin

During the process of adjustment from centralplanning to a market-oriented system, many

transition economies have experienced the problem ofenterprises not receiving payments from each other ontime for goods and services delivered.1 Such nonpay-ments—or interenterprise arrears—signal a lack of fi-nancial discipline, can retard investment and theprocess of privatization, and can undermine the con-duct of prudent macroeconomic policies. This paperexamines developments in interenterprise arrears inthe Baltics, Russia, and other countries of the formerSoviet Union, identifies its possible sources, and setsout policy options toward resolving the problem.

The potential adverse effect of interenterprise ar-rears can be much wider than its direct impact on en-terprises that did not get paid for goods and servicesdelivered. First, these enterprises themselves mayincur arrears on their payables because of insufficientliquidity. Enterprises can quickly become entangledin multiple layers of arrears, with one enterprise'spayment to a second contingent on receipt of paymentfrom a third, and so on. Such a snowballing of arrearscan threaten the viability of fundamentally profitablefirms as well as that of lossmaking enterprises. Sec-ond, massive amounts of receivables and payables onenterprise balance sheets, much of dubious value, canconfuse and hinder a realistic assessment of the finan-cial situation of enterprises by their own managers,suppliers, banks, and potential investors. As a result,investment, credit, and privatization decisions may beimpeded. Third, interenterprise arrears (and tradecredits, more generally) constitute a form of financialdisintermediation and can complicate monetary man-agement, in particular, by leading to a decline in thedemand for bank credit and an increase in the velocityof money (see Section III).

In the Baltics, Russia, and other countries of theformer Soviet Union, interenterprise arrears grewrapidly during the first six months of 1992, from rela-

tively modest initial amounts. The subsequent behav-ior of interenterprise arrears has varied across coun-tries. Although a cross-country comparison is com-plicated by the lack of reliable data on a consistentbasis,2 the problem, by the end of 1994, appeared tohave subsided considerably in countries that madesignificant progress toward macroeconomic stabilityand structural reform—for example, in the threeBaltic states. In contrast, some other countries—mostnotably, Azerbaijan and Ukraine—which were notwell advanced in stabilization and reform, saw an in-tensification of the arrears problem.

It should be noted that while the interenterprise ar-rears problem has been accompanied by a buildup inwage and tax arrears, the discussion here focuses onoverdue payments obligations between enterprises.Overdue payments between enterprises across na-tional boundaries are also an important aspect of thearrears problem. The interstate component of arrearsmainly relate to the energy sector in most countries.Such arrears are often large in volume and compli-cate not only the balance of payments but also inter-state financial relations.

The evolution of interenterprise credits and ar-rears in the Baltics, Russia, and other countries ofthe former Soviet Union is discussed here first, fol-lowed by some conceptual issues regarding inter-enterprise credit and arrears and some of their proba-ble causes. Policy options to deal with the existingstock of debt and to prevent the emergence of newarrears are then considered.

Evolution of InterenterpriseCredit and Arrears

Under central planning, access to bank credit wasvirtually automatic for activities approved by theplan, and the need for interenterprise credit was ex-tremely limited. Interenterprise transactions were car-ried out in "account" or "deposit" money by making

1In the Baltics, Russia, and other countries of the former SovietUnion, the disruption of the interrepublican payment system inthe immediate aftermath of the U.S.S.R. was a special factor thatcontributed to the initial buildup of arrears. 2See the appendix of Section I for a discussion of data issues.

63


IV INTERENTERPRIS E ARREARS

money transfers through the old interbranch settle-ment system. A supplier usually received payment bysubmitting a payment demand order to its own bankrequesting payment from the buyer.3

Under this system, overdue payments were only aminor problem. Unpaid bills of enterprises wererecorded under File No. 2 (kartoteka nomer dva) ofcommercial banks, and any arrears at the end of thefinancial year were written off by the extension ofautomatic credit.4 At the end of 1990, for the entireU.S.S.R., total interenterprise receivables for goodsand services were equivalent to 13 percent of GDP,and at the end of 1991, the outstanding balance inFile No. 2 for the region stood at an estimated 7 per-cent of GDP.

Initial Transition Period , Early 199 2 toMid-1993

All this changed with the phasing out of strict cen-tral planning in all the countries by early 1992.5 Thebanking sector ceased to function as an administra-tive arm of the planning mechanism, and bank creditto enterprises ceased to be automatic. Furthermore,the phasing out of state orders, which had providedan assured market for a large portion of enterprises'output, implied that enterprises had to look for cus-tomers, start to engage in marketing activities, andalso perhaps coax buyers by providing trade credit.

Interenterprise credit, which had been mostly con-fined to the File No. 2 of banks, began to appear out-side the purview of the banking system, and the FileNo. 2 system was abolished in many states by themiddle of 1992.6 Enterprises were free to instructtheir banks which creditors to pay, if at all, and notnecessarily in order of chronological seniority. Atthe same time, the outstanding amounts in the FileNo. 2 were no longer guaranteed, and the quality ofsuch claims suddenly became highly questionable.

Total interenterprise credits (including those notconsidered to be overdue) increased rapidly in almost

3A detailed description of the payments system prior to mid-1992 is available in Bigman and Leite (1993).

4In years of poor economic performance, the write-offs couldbe large enough to have a macroeconomic impact (as with theagricultural debt write-offs of 1986).

5The initial transition period begins with the dissolution of theU.S.S.R. and ends just prior to Russia's demonetization of pre-1993 Russian rubles.

6It has not been abolished in Azerbaijan, Turkmenistan, andUzbekistan. In July 1992, the Central Bank of Russia changed theregulations regarding File No. 2: in case of insufficient funds,payment demands in most cases were not to be queued anymorebut returned to the originator. Thus, the coverage of the data thatcontinued to be compiled under File No. 2, clearly changed fromJuly 1992. Regulations regarding File No. 2 were changed againin October 1993. (After this paper was written, File No. 2 wasabolished in Uzbekistan in mid-1995. Eds.)

all countries during 1992 and the first half of 1993. In-deed, these credits rose substantially in relation to thestock of broad money, contributing to inflationarypressure. Nevertheless, at the end of the period, thelevels of interenterprise credits were modest (Table4.1) compared with the standards of both establishedmarket economies, and those of other reformingeconomies in eastern Europe such as Poland, Hun-gary, and the former Czechoslovakia, at a comparablestage of transition. As of mid-1993, total interenter-prise receivables ranged from 5 percent to 19 percentof GDP in Belarus, Russia, and Ukraine—the onlythree countries for which data on total interenterprisecredits exist at higher-than-annual frequencies.7 In the1980s, in the United Kingdom, interenterprise creditamong industrial and commercial companies aloneranged from 17 percent to 20 percent of GDP,8 whilein the United States, trade credit of nonfinancial cor-porations stood at 17 percent of GDP in the first quar-ter of 1992. The levels of interenterprise receivableswere equivalent to 20-60 percent of GDP in Polandand the former Czechoslovakia during 1988-91(Chart 4.1).

Turning to the overdue portion (arrears), their lev-els—when measured in relation to GDP or in termsof total interenterprise credit—also remained rela-tively low (Table 1.5).9 For example, in the CzechRepublic, just under 50 percent of total receivableswere overdue at the end of 1991, and in many west-ern European countries nearly half the firms waituntil bills are two weeks or more overdue beforepaying them.10

Nevertheless, by mid-1992, the rapid growth of in-terenterprise arrears became a major source of concernamong policymakers, particularly in the countries ofthe Commonwealth of Independent States (CIS).While the Baltic states followed a policy of benign ne-glect, in the CIS meetings in Bishkek and Tashkent inMay 1992, broad agreement was reached among offi-cials on a four-pronged plan of action to (1) increasethe responsibility of enterprises for assuring promptpayment for their sales; (2) raise the interest rate thatenterprises were advised to charge on arrears; (3) net

7In Russia, prior to the third quarter of 1993, data on interenter-prise credit and arrears cover only the industrial and constructionsectors. However, for the subsequent period, data indicate thattotal interenterprise credits are typically 10-20 percent higherthan the corresponding figures for industry and construction only.The corresponding levels of interenterprise payables in mid-1993were in the range of 4 percent to 23 percent of GDP in these threecountries (Table 4.1).

8See Begg and Portes (1992), Table 4.9As explained in the appendix to Section I, cross-country com-

parisons of interenterprise arrears are subject to the caveat that thedefinition of when a payment becomes overdue and hence when aninterenterprise credit becomes an interenterprise arrear varies acrosscountries.

10See Fan and Schaffer (1994), p. 158.

64


Evolution of Interenterprise Credit and Arrears

Table 4.1. Belarus, Latvia, Russia, and Ukraine: Interenterprise Credi t(Payables, end of period)

BelarusLatvia2

RussiaIndustry and construction

Ukraine

BelarusLatvia2

RussiaIndustry and construction

Ukraine

1992

Q4

23.8

204.0

46.3

Q1

6.7

38.5

53.190.8

Q2

3.9

6.322.6

36.5

47.6166.4

1993

Q3 Q4

(As percent of GDP)1

3.5

6.216.2

8.116.28.06.9

20.9

(As percent of broad money)

36.2

76.466.6

140.7

185.879.584.873.6

334.1

Q1

15.6

9.98.3

30.5

114.8

121.499.1

389.8

1994

Q2

13.9

6.04.9

33.5

141.2

107.988.3

361.6

Q3

105.1

83.2

Sources: Data provided by country authorities; and IMF staff estimates.1Defined as average during the quarter as a percent of annualized quarterly GDP.2Includes payables to the government as well as banks.

out arrears first within each state and subsequently atthe interstate level; and (4) replenish the working capi-tal of state enterprises. While action on the first twocomponents of the policy package was not immedi-ately obvious, all countries of the CIS carried out net-ting out exercises for interenterprise arrears and fol-lowed these up with additional bank credit to clear thenet outstanding claims as well as to replenish theworking capital of enterprises. In the event, these ac-tions added to inflationary pressures, and the reprieveto the arrears problem achieved by the exercise provedto be temporary.

Russia provides a clear example of these develop-ments. By mid-1992, interenterprise arrears had risento 21 percent of GDP,11 and deliveries started to be-come sensitive to payments, with reports of enter-prises delivering cash by truck or plane against deliv-ery of goods. In response, the authorities carried out amultilateral netting out of arrears in the second half of1992, through the establishment and operation of spe-cial accounts in commercial banks. Furthermore, netpositive balances of creditors remaining after that op-eration—which amounted to about 3 1/2 percent ofGDP—were then monetized by making them usablefor tax payments and for paying off liabilities tobanks.12 While a significant arrears problem did not

11The interenterprise payables of Russian enterprises on File

No. 2 of banks amounted to Rub 3 trillion by the end of June

1992. See Fan and Schaffer (1994).12See Rostowski (1994), pp. 19-21.

re-emerge in Russia until the end of 1993, with thelevel of overdues averaging around 2 1/2 percent ofGDP in the first half of 1993, the clearing and bailoutoperation clearly had expectational and inflationaryimplications.

Mid-1993 to Late 1994For the period from mid-1993 to late-1994, the

countries for which information is available can beplaced in three distinct groups: (1) Latvia, Lithuania,and the Kyrgyz Republic, where the arrears problemdiminished; (2) Russia and Turkmenistan, where itslowly increased; and (3) Azerbaijan, Belarus, Ka-zakstan, and Ukraine, where it escalated rapidly.

In Latvia, total interenterprise credits fell below 18percent of GDP by the end of 1993, implying evenlower values for overdues. In Lithuania, overduepayables, after a modest increase in the third quarterof 1993, declined steadily not only in relation toGDP, but also as a proportion of broad money (Table1.5). The evolution of arrears in the Kyrgyz Republicwas similar to that in Lithuania with one exception:although the stock declined as a proportion of GDP,the ratio to broad money remained stubbornly high ataround 2 1/2 until the third quarter of 1994.

There was a rather modest rise in interenterprisearrears—as measured against GDP—in Russia andTurkmenistan during this period. Overdue payablesdid increase more significantly in relation to broadmoney in both countries, however, with implications

65



Chart 4.1 . Othe r Centra l and EasternEuropean Countries in Transition:Interenterprise Arrears

Source: IMF staff estimates.1End-year stock in relation to annual GDP.2End-year stock in relation to end-year stock broad money.

for the effective conduct of monetary policy. Also,the interstate component of interenterprise arrearswas a problematic aspect of the arrears problem inboth Russia (Table 4.2) and Turkmenistan, reflectingin particular debts on energy deliveries. In Russia,arrears involving the fuel and electrical energy sec-tors amounted to almost one half of total interenter-prise arrears in 1994 (Table 4.3). In Turkmenistan,overdue receivables of gas companies from enter-prises in other states of the region rose by about $0.5billion during the first nine months of 1994 to about60 percent of GDP at the end of September, 1994.These nonpayments significantly complicated infla-tion control in Turkmenistan, with the authoritiesgranting additional credits to enterprises in lieu ofthe receipts from abroad.

In contrast, interenterprise payables and arrearsrose to relatively high levels in Azerbaijan, Belarus,Kazakstan, and Ukraine during this period.13 In

13For Belarus, separate data on overdues are not available, butone might presume that the rise in total payables was accompa-nied by a rise in arrears.

Ukraine, arrears began to re-emerge relatively soonafter an arrears clearing and bailout operation con-ducted in February 1993, and increased sharply inlate 1993 and early 1994, particularly in relation tobroad money. Two further netting out operationswere carried out in Ukraine in 1994. In Kazakstan,arrears rose sharply during the second half of 1993.In response, in early 1994, the Government ofKazakstan carried out a netting operation, followingwhich net creditors were paid the full amount oftheir claims. Part of these payments were in the formof government bonds, where the debtors became li-able to the government and their payments obliga-tions were to be enforced through the threat of bank-ruptcy. In the event, most of the debt to thegovernment was not repaid, and liquidation proceed-ings have only been initiated for a small proportionof the debtors.14 Moreover, arrears once again rosesharply in the third quarter of 1994.15 Finally, inAzerbaijan and Belarus as well, interenterprise ar-rears escalated in spite of clearing operations carriedout in the third quarter of 1993.

Although little information is readily available re-garding the volume of interenterprise arrears inUzbekistan, it is likely it also belongs to this thirdgroup of countries. And indeed, the Uzbek authori-ties undertook, under the auspices of the centralbank, a netting out of gross overdue interenterpriseclaims and liabilities between August 15 and Sep-tember 5, 1994.

Thus, indications are that by late 1994, while ar-rears ceased to be a serious problem in some coun-tries, they appeared to be of significant concern in anumber of others. Moreover, in Russia (Table 4.2)and Turkmenistan, while the aggregate level of do-mestic arrears appeared to be modest, they werehighly concentrated in a few sectors. And it wouldseem that the higher the concentration, the more thepressure for an ultimate government bailout of suchoverdue obligations. Indeed, the one of Turk-menistan serves as a vivid example of this.

Factors Causing ArrearsInterenterprise credit and arrears are common fea-

tures in all market economies, and it can be argued

14Provisional results showed that there was a reduction in grossarrears of around 38 billion Tenge, or 8 percent of GDP. Claimsof net creditor enterprises were settled by a payment by the gov-ernment of T 1.1 billion (mainly to the energy sector) and T 8 bil-lion of one-year, U.S. dollar indexed treasury bills, bearing a 3percent interest rate.

15Kazakstan is a major producer of energy, like Russia andTurkmenistan. While no information is readily available on over-due interenterprise receivables in Kazakstan, a large part of it alsoappears to be concentrated in the energy sector.

66


Evolution of Interenterprise Credi t and Arrears

Table 4.2. Russia : Interstate Interenterprise Arrears, as of October 1, 1994(In billions of rubles)

Receivables of Russian Enterprises fromEnterprises in the CIS 1 and Baltic Countries

Payables of Russian Enterprises toEnterprises in CIS and Baltic Countries

State

ArmeniaAzerbaijanBelarusGeorgiaKazakstanKyrgyzstanUzbekistanUkraineTajikistanTurkmenistanMoldova

Subtotal

LatviaLithuaniaEstonia

Total

Total

8.533.9

267.721.8

404.833.9

147.2769.842.430.442.1

1,802.9

36.928.346.2

1,914.5

Overdue

4.328.3

158.811.1

274.026.992.7

513.432.524.935.1

1,202.4

18.711.017.5

1,249.6

Share ofoverdue

(in percent)

50.283.559.351.367.779.663.066.776.782.083.266.7

50.738.837.865.3

Total

9.816.4

136.09.2

341.527.864.8

354.510.28.8

14.0993.2

9.718.226.0

1,047.1

Overdue

2.78.4

66.87.0

229.919.236.4

209.33.56.17.9

591.2

4.24.04.7

604.1

Share ofoverdue

(in percent)

27.850.949.276.467.369.156.257.333.770.056.259.5

43.222.018.357.7

Net Receivable sTotal Overdu e

-1.417.5

131.812.663.3

6.182.5

415.432.121.728.2

809.7

27.310.220.3

867.4

1.520.092.04.1

44.17.7

56.3310.329.118.827.2

611.3

14.57.0

12.8645.6

Source: Goskomstat.1Commonwealth of Independent States consists of Russia and the other states of the former Soviet Union, except the Baltic states.

that some increase in the volume of such credit andarrears—from very low initial amounts—can be ex-pected as these economies move toward market-based systems.16 However, it is important to distin-guish between the natural adaptation to markets andspecial factors present during transition that drivethe accumulation of arrears. A brief discussion ofthese different factors follows.

Normal Factor s Operating in aMarket Econom y

Interfirm trade credit is quite substantial and nor-mal in market economies.17 The various explana-tions of trade credit include (1) transactions costs as-sociated with cash payments; (2) informational costsassociated with bank lending or credit market imper-fections (e.g., interest rate spreads) that raise the rel-ative cost of bank loans; and (3) goods market im-

perfections that force firms to provide credit as amarketing activity.18

Market imperfections can be hypothesized to haveplayed a significant role in the rapid growth of inter-enterprise credit in the Baltics, Russia, and othercountries of the former Soviet Union from insignifi-cant initial amounts. The process of replacing theprevious system of planning and state orders by a setof perfectly functioning product markets will begradual and tortuous. Furthermore, the system ofdistributing a large part of the available limitedamount of credit to selected enterprises at subsidizedrates continued well into 1994 in many of thesecountries and was a source of credit market imper-fections.19 The spreads between lending and depositrates in many of the countries—for example, morethan 50 percentage points on an annual basis in Rus-sia in the second half of 1994—continued to be veryhigh through 1994.

16See Begg and Portes (1992) and Fan and Schaffer (1994).17For example, stocks and flows of trade credit, on average, are

about twice the corresponding amounts of bank credit in theUnited States and the United Kingdom, see Crawford (1992).

18An overview of the literature of trade credit is available inCrawford (1992).

19According to enterprise surveys in Russia, part ownership ofbanks by enterprises made borrowing easier—more evidence ofcredit market imperfection. See Fan and Schaffer (1994).

67



Table 4.3. Russia: Sectoral Composition of Interenterprise Arrears(Overdue payables, end of period)

IndustryConstructionAgricultureTransport

IndustryConstructionAgricultureTransport

Total industryElectrical energyFuelFerrous metallurgyNonferrous metallurgyChemicalsMachineryTimber processing and paperBuilding materialsGlasses industryLight industryFood industryOther industry

Total industryElectrical energyFuelFerrous metallurgyNonferrous metallurgyChemicalsMachineryTimber processing and paperBuilding materialsGlasses industryLight industryFood industryOther industry

1992

Q4

41.641.8

100.03.2

21.015.16.6

12.920.3

5.34.00.23.66.41.4

41.627.933.457.645.248.643.752.959.837.346.427.636.5

Q1

37.636.8

100.04.3

23.413.16.3

15.117.64.73.70.22.67.9I.I

37.622.534.550.834.250.037.844.952.026.034.729.222.3

1993

Q2 Q3 Q4 Q1

(Sector's share in total overdue payables in the economy)

83.18.13.15.7

80.98.64.85.8

81.08.94.75.3

(Share of overdue payables in total payables in the sector)

36.432.8

37.030.322.424.5

40.537.534.129.4

45.143.740.715.9

(Subsector's share in total overdue payables in industry)

100.03.2

28.410.87.8

12.816.14.32.80.12.5

10.0I.I

100.05.1

30.512.65.7

13.215.54.32.50.21.78.00.8

100.07.1

26.211.07.0

14.715.74.52.9

1.87.9I.I

100.013.030.59.74.7

12.913.03.42.70.21.77.20.9

(Share of overdue payables in total payables in the subsector)

36.421.235.942.336.945.236.845.145.824.633.930.320.5

37.024.239.445.431.150.137.446.438.429.029.424.522.4

40.527.739.443.949.554.841.750.245.6

34.631.422.1

45.146.146.445.240.355.842.149.246.534.837.337.030.7

1994

Q2

80.98.64.36.1

49.342.942.120.0

100.015.229.9

8.94.9

12.214.13.52.50.31.76.00.9

49.348.653.044.146.560.647.155.849.548.945.337.830.5

Q3

75.18.14.6

12.2

55.945.750.039.5

100.016.230.39.14.3

10.916.03.22.30.21.55.10.8

55.952.464.352.945.363.155.460.252.046.449.941.434.2


Nonetheless, and although overruns on informalcredits between firms are also common in marketeconomies,20 the explanation of maturity overrunson the scale registered in some of these countriescannot rely on these imperfections alone. This leads

20For example, according to a survey of more than seven hun-dred firms in the United Kingdom, the maturity overrun on tradecredit is about one month on the average. See Fan and Schaffer(1994).

to possible explanations of the arrears problem interms of structural factors, including the expectedpolicy stance of the governments.

Additional Factors in theTransition Economie s

In the initial stages of transition, the managers ofstate-owned enterprises have little experience withthe rules of the market, and the institutional arrange-

68


Evolution of Interenterprise Credit and Arrears

ments and incentives structure do not change in adramatic way toward encouraging market behavior.The high level of uncertainty regarding relativeprices and prospective stance of macroeconomicpolicies, and regarding the future ownership andmanagement structure, compound the problem of in-adequate incentives for efficient management. Insuch circumstances, managers seek to avoid con-flicts with the work force on employment and wageissues, often buying raw materials and inputs oncredit and maintaining a level of production even ifthere is insufficient demand. Such behavior maylead to either the accumulation of inventories or de-livering output in exchange for future promises topay to buyers with questionable prospects. When thepromises to pay are not honored in time, arrearsbuild up. The absence of a sound banking and pay-ments system, of a stable macroeconomic frame-work, and of suitable laws and institutions perpetu-ate a management culture inappropriate for a marketeconomy and financial discipline. The role of theserelevant factors in the evolution of interenterprise ar-rears is briefly discussed below.

Undeveloped Financial Systems

Underdeveloped financial systems are likely tohave contributed to the growth of interenterprisecredit and arrears in the region. The services providedby financial intermediaries in market economies, forexample, making payments and facilitating transac-tions and risk management, monitoring firm man-agers, and credit assessment, were either irrelevant orof minor importance under central planning.21 Duringthe initial stages of transition, the supply of these fi-nancial services expanded but generally fell far shortof demand. Enterprises were led to organize financialarrangements among themselves to sustain trade andoutput. In the process, suppliers assumed high risksthat at least partly ended up as overdue receivables.

Delays in the settlements system also were a factor.Under the newly emerging market system, paymentsorders could be executed only after clearing, and lackof adequate computing and communication facilitiesled to inordinate delays in both domestic and inter-state payments.22 The disintegration of large enter-prises into many autonomous units, often across stateboundaries, also may have contributed, by increasingoverall credit needs of enterprises and the volume offinancial transactions in the economy.

Credit market imperfections also may have exacer-bated the arrears problem, particularly when restric-

tive monetary policies were pursued by some coun-tries to combat high inflation. In imperfect creditmarkets characterized by widespread use of artifi-cially low interest rates and credit rationing, a decel-eration in the growth of credit expansion may haveled to the substitution of bank lending by forced lend-ing by enterprises to each other, and hence interenter-prise arrears.

Thus, inadequate progress in financial sector re-form is likely to have contributed to the buildup ofarrears. Indeed, the containment of the interenter-prise arrears problem in Latvia and Lithuania maypartly be attributed to the progress made toward es-tablishing sound and efficient banking and paymentssystems.

Uncertainties, Macroeconomic Instability,and Collusion

Transition in the Baltics, Russia, and other coun-tries of the former Soviet Union has been character-ized by an unpredictable economic environment.Uncertainties have prevailed in relative prices, inmacroeconomic policies, and in the availability ofsupplies of inputs and markets for output because ofdisruption in traditional interstate trade and financiallinkages. In such circumstances, enterprises havehad extreme difficulties evaluating credit risks ofdifferent buyers and may have found it optimal toseek buyers indiscriminately but at a price adjustedupward to take account of higher credit risk.23 Aconsequence of such behavior has been the accumu-lation of arrears, as a high proportion of credits haveproved to be of bad quality.

Dramatic changes in relative prices followingprice liberalization, particularly with regard to en-ergy products, has severely affected the profitabilityof enterprises—particularly those intensive in en-ergy use. Thus, the asset quality of receivables fromsuch enterprises has suffered heavily in the processand has contributed to the arrears problem.24

The failure to address large macroeconomic imbal-ances and associated balance of payments problemshas also had implications for the accumulation of ar-rears in Russia and some of the other countries of theformer Soviet Union. After the dissolution of theU.S.S.R., the volumes of interstate interenterprisetrade—historically high in energy, raw materials, andintermediate products—remained substantial. Asidefrom the same factors influencing domestic inter-

21See Caprio and Levine (1994), pp. 2-3.22Reportedly, a payment order could take as long as three

months to be executed.

23See Kim and Kwon (1995).24Residential and industrial customers were in arrears to utilities

producing electricity and heating, which in turn were in arrears tothe gas and petroleum companies. In energy-importing countries,these led to large overdue payables to foreign suppliers.

69



enterprise arrears, shortages of foreign exchange orthe escalating domestic costs of imports because ofrapidly depreciating currencies, or both, are likely tohave contributed to the rise in interstate arrears.

The government's determination in pursuing sta-bilization policies is often tested by enterprises intransition economies by collusive behavior. Suchbehavior, often only "implicit" and without a well-coordinated scheme, may arise from the enterprises'rational beliefs about each other's likely response topolicies in light of management's and workers' en-trenched interests.25 Frequently, the initial attemptat stabilization started off with the authorities tryingto control inflation by curbing the supply of creditfrom the central banks. One of the aims of such atight monetary policy was to promote financial dis-cipline among enterprises, induce them to restruc-ture, and encourage them to substitute internal fi-nance for bank credit. In response to a tightening infinancial policies, enterprises—anticipating thelikely response of other enterprises—may have de-layed any adjustments in their behavior, guessingthat if every enterprise built up arrears, the govern-ment would be unwilling to risk the consequencesof a systemic collapse and would be forced to easepolicies. In other words, the government's policystance was not seen to be credible, and enterprisesstopped paying each other, knowing that a generalbailout was inevitable down the road.26

Developments in Russia during the first half of 1992are a case in point: the uncertainties regarding the fu-ture policy stance—including the tenure of the centralbank governor—are likely to have contributed signifi-cantly to the expectation of a bailout, a rapid buildupof arrears, replacement of the central bank's manage-ment, and, indeed, an eventual bailout in the latter partof the year.27 Similarly, the rise in arrears that accom-panied the tightening of monetary policy in Russiaduring the first quarter of 1994 may indeed partly re-flect collusive behavior by some selected and big en-terprises. The arrears were concentrated in the energy,ferrous metallurgy, chemicals, machinery, and con-struction sectors, which are dominated by large state-owned enterprises from the Soviet era. The enterprisesinvolved are typically those that are commonly per-ceived to be "too big to fail," and it is likely that ex-pectations—or at least hopes—of an eventual bailoutby the government encouraged the accumulation of

arrears.28 The lack of credibility in the authorities'tight policy stance was likely reinforced by the earlierbailout in 1992, and also the departure of key reform-minded ministers from the government in early 1994.

Any demonstration of the lack of steadfastness ofgovernment policies—in the form of a temporaryloosening of monetary policy, a selective doling outof subsidies to state enterprises, or netting out ofgross interenterprise arrears, and clearing the re-maining claims by central bank credit—only adds tothe arrears problem in the long run (see below).Thus, the determined pursuit of stabilization policiesby Latvia and Lithuania resulted in a rapid decelera-tion of inflation, a stable macroeconomic frame-work, and containment of the arrears problem. Onthe other hand, the stop-go nature of policies in anumber of countries, and the bailout of enterprisesthrough multilateral clearing of interenterprise ar-rears, seems only to have led to a quick re-emer-gence of arrears and contributed to the persistence ofthe protracted problem of high inflation and macro-economic instability.

It should be noted that the sharp rise in the ratio ofarrears to broad money in some of the countries ofthe region has strengthened the appeal of the "exces-sive credit contraction hypothesis"—a hypothesisthat has its root in the simultaneous drastic falls inreal credit and output in those countries.29 But as al-ready discussed above, the substitution of bank lend-ing by interenterprise arrears reflects imperfectionsin the credit market and structural inadequacies. Anydecrease in arrears brought about by a one-shot in-crease in credit is likely to be transient at best, withthe gain quickly dissipated through a fast transmis-sion of higher credits into even higher prices.30

Structural Inadequacies

In the Baltics, Russia, and other countries of theformer Soviet Union, like in other transition coun-tries, the transformation toward market-based sys-tems began without the laws and institutions neces-sary to support the functioning of markets. Accurateand meaningful financial statements, clearly definedproperty rights, and a legal system with full powers

25SeePerotti(1994).26For example, according to an enterprise survey in July 1993

in Bulgaria, even when no clearing of interenterprise arrears wasbeing actively debated, a fifth of the managers of state enterprisescovered by the survey admitted that suppliers provide goods oncredit to uncreditworthy enterprises in the expectation of abailout. See Perotti (1994).

27See Section VIII, pp. 119-20.

28The prevalence of arrears may also differ between upstreamand downstream industries. Producers and retailers of final goodsmay get paid in cash, but "pay" their suppliers with arrears, whileproducers of intermediate goods may "pay" and get "paid" witharrears. But it is unlikely that the upstream and downstream dif-ferences alone explain the sectoral concentration of arrears.

29See Section II for a discussion of the general plausibility ofthis hypothesis. Also see Calvo and Coricelli (1994).

30Containing of the inflationary impulse and establishing credi-bility through price and wage controls is not an appealing optionbecause of the high level of initial distortions in relative pricesand wages in transition economies.

70


of enforcing contracts are necessary for the properfunctioning of the loan market and limiting loandelinquencies. In a functioning market economy,failure to service a loan obligation on time affectsfuture access to credit, and ultimately involves thethreat of bankruptcy. Enterprises do not incur arrearswhen the costs of incurring them are high,31 and thelegal and institutional framework is crucial for deter-mining these costs. Of course, structural reformsneed to be reinforced by a clear "no bailout policy,"as any bailouts strengthen the expectation of yet an-other bailout in the future.

In the event, in most of the region, accounting lawswere inappropriate for a market economy, and enter-prises did not have accurate and meaningful balancesheets or profit-and-loss statements. Property rightswere not clearly defined and there was no bankruptcylaw with the power to fully enforce contracts. Theemerging private sector was dominated by an over-whelmingly large state-owned sector where corporategovernance left much to be desired. The dismantlingof planning along with the weakening of central con-trols led to the transformation of state-owned enter-prises and banks into organizations controlled by in-siders (managers and workers), particularly prior toprivatization. In such circumstances, extremely weakbalance sheet positions encouraged risk taking by en-terprise managers and actions were taken to protectshort-term output and employment at all costs.

The interaction between inadequate bankruptcylaws and weak balance sheet positions of enterprisesled to "creditor passivity" and contributed to the per-petuation of the interenterprise arrears problem.32

First, creditors may not have initiated bankruptcyproceedings because the costs of enforcing bank-ruptcy exceeded the expected value of the debtor'sassets, or because of the option value of waiting.Second, initiating action against the debtor enter-prise may have signaled the existence of financialproblems in the creditor enterprise itself, and henceundermined the confidence of the creditor's credi-tors. Third, with multiple creditors, the incentive forindividual creditors to wait for others to initiate pro-ceedings may have been acute, particularly when thenet worth of the debtor was in doubt.

Thus, together with better bankruptcy laws, restruc-turing—including recapitalization—and privatizationof the state-owned banks and enterprises are crucialfor establishing a framework of adequate incentives

and efficient operation of economic activity in thesetransition economies. Indeed, the evolution of inter-enterprise arrears in the Baltics, Russia, and othercountries of the former Soviet Union clearly demon-strate the importance of structural reform for contain-ing arrears. Latvia and Lithuania, which have madesignificant progress in the structural front, have alsobeen successful in managing the arrears problem.

Policy OptionsInterenterprise arrears can only be contained on a

sustained basis by ensuring that enterprises adjust tothe emerging market system and observe financialdiscipline. A stable macroeconomic framework anda structure of appropriate incentives are essential inthis regard and can be achieved only by the steadfastpursuit of strong stabilization policies and acceler-ated structural reform, including financial and enter-prise sector restructuring. Such policies provide thebest antidote to the arrears problem over the mediumterm.

Some transition countries, such as the formerCzechoslovakia, Poland, and the Baltic states ofLatvia and Lithuania have followed a hands-off pol-icy toward interenterprise arrears. They have left theresolution of the problem to the market; so far theapproach is successful. On the other hand, the grid-lock created by interenterprise arrears sometimesposes a serious threat—or a perception thereof—ofan imminent collapse of the payments system, andhence production, in the process of transition. Thus,countries in the region have taken exceptional mea-sures-—some more than once—to deal with inter-enterprise arrears.

Romania took exceptional measures on severaloccasions in 1991-92 to deal with interenterprise ar-rears.33 The measures included government-spon-sored netting out exercises, the extension of centralcredits and government transfers and, on one occa-sion, a scheme under which banks were required toextend credit to enterprises, under government guar-antees, for the sole purpose of clearing arrears.34 Thecreation of additional liquidity as a result of these

31For example, in a simple model, Calvo and Coricelli (1994)have shown that an enterprise will not incur arrears, if it reckonsthat arrears incurred this period increase into costs next period ata rate higher than the rate of interest that wage earners can earnon their deposits.

32See Begg and Portes (1992) and Mitchell (1993) for a discus-sion of creditor passivity.

33Albania and Bulgaria also took some exceptional measuresinvolving public sector financing to address interenterprise ar-rears. In January 1993, the Albanian authorities adopted a decreeto move enterprises from a system of automatic postshipmentpayments administered through banks to a system of preshipmentpayments, and also set aside funds to provide selective partialcompensation to net creditor firms with good performance in thepast six months. In Bulgaria, a debt write-off was organized inDecember 1992, mainly to enable the energy sector severely af-fected by the problem of unpaid bills to partially clear off its ar-rears to enterprises and banks.

34See Khan and Clifton (1992).

Policy Option s

71



operations was clearly inflationary; while there wasa temporary reprieve to the arrears problem, ulti-mately the government moved to enact a stringentbankruptcy law and launch a public campaign tocombat the moral hazard problem inherent in the im-plementation of such schemes and convince enter-prises that no further bailout was likely.

The experience with exceptional policy measuresto deal with interenterprise arrears indicates the needfor extreme caution in applying such measures in theBaltics, Russia, and other countries of the former So-viet Union. Thus, in designing any such measures toaddress the "stock" problem of existing arrears, it iscritical to ensure that the measures include steps toavoid the emergence of new arrears (i.e., the "flow"problem), and also are governed by the following twokey considerations. First, the creation of a market en-vironment where enterprises must accept responsibil-ity for their own indebtedness and must behave in a fi-nancially responsible way or face the consequences.Second, at a time when the governments are facingthe urgent and difficult task of reducing inflation, themeasures must avoid additional credit creation thatcould lead to new inflationary pressures.

The Stock Problem

The resolution of the stock problem should beachieved by encouraging debtors to reschedule theiroverdue payables through voluntary bilateral con-tacts with creditors. In addition, enterprises should beencouraged to securitize their interenterprise claimsto make them tradable: an enterprise in need of liq-uidity could sell an interenterprise claim at a discountto another firm; the buyer could then either hold theclaim or set it off against its own liabilities to theoriginal debtor firm (on whom it now owns a claim),if such liabilities exist. Banks could also become in-volved in this activity, purchasing and selling securi-ties subject to normal prudential regulations. The op-portunity for profitable trade in interenterprise claimsin a market environment should help resolve theproblem of arrears. Development of requisite safe-guards to prevent insider trading and other manipula-tive practices in the securities market should proceedin parallel with the securitization of claims.

In the short term, governments and central bankscan promote a solution to the "stock" problem of ex-isting interenterprise arrears by promoting decentral-ized bank-enterprise debt workouts with a menu ofoptions, as well as a secondary market where arrearscould be traded. Such an approach could lead to a net-ting out of interenterprise arrears without any exten-sion of government or central bank credit, and send astrong signal to enterprises and banks that the authori-ties will no longer bail out firms with arrears or bad

debts, thereby promoting the restructuring processthrough the adoption of hard budget constraints.35

The involvement of governments and centralbanks should be limited to providing an adequateregulatory and institutional framework, for example,regulations that appropriately govern trading in in-terenterprise credits. Rules allowing limited forms ofmultilateral netting, whereby a group of firms forman arrangement to net out debts within the group,should also be considered. In any event, market-based solutions are unlikely to work without somecredible threat of bankruptcy for notorious delin-quencies, and there should be no government guar-antees for the securities.

Public concern over the arrears problem leads topressure on governments to organize clearingschemes to net out gross arrears. Such explicit net-ting out exercises need not directly involve creditcreation. However, an important argument againstcompulsory netting out operations—whether or notthey involve the creation of additional bank credit—is that the expectation of even these operations cre-ates perverse incentives. Firms will be more willing,on the one hand, to withhold payment for their in-puts and, on the other, to ship their output to cus-tomers they strongly suspect will not pay—calculat-ing that these arrears will cancel out in the nextnetting out operation. Moreover, there is always astrong risk, particularly because of uncertainties re-garding the valuation of claims, that the govern-ment's involvement creates expectations of a bailoutof the net claims remaining after the implementationof a clearing scheme. Furthermore, such netting outexercises take time to complete and create consider-able uncertainty for macroeconomic management.These considerations argue strongly against a gov-ernment-sponsored netting out exercise.

Preventing New ArrearsApart from measures to improve the payments

system, so as to shorten the period of uncertaintyabout the creditworthiness of the payer, policies toprevent new arrears should focus on convincing en-terprises that they must accept responsibility for col-lecting their own claims on other enterprises, andthat they should not expect government interventionof any sort. In addition to avoiding any bailout com-ponent in the measures designed to deal with thestock of existing arrears, a few other specific stepscould be helpful.

First, bankruptcy criteria and judicial proceduresshould be streamlined with a view to providing not

35The approach played a part in the restructuring of enterprisedebt in Poland.

72


only for the closing of insolvent enterprises, but alsofor their reorganization in an attempt to avoid futureliquidation and closing. Creditors should be empow-ered to sell their debtors' assets if debts are overduefor more than a certain period of time. Any judicialprotection debtors receive to delay such asset salesshould be accompanied by appropriate interest andpenalty payments on the overdue claims.

It is unrealistic, however, to expect a bankruptcysystem to emerge rapidly just with the enactment ofthe appropriate laws. Given the relative inexperienceof judicial courts in bankruptcy cases and the limitedavailability of trained liquidators, and hence the un-certainty of the judicial awards, creditors tend to fightshy of becoming "pioneers" and incurring large start-up costs in bankruptcy proceedings.36 Until bank-ruptcy procedures work smoothly and convince enter-prises to alter their behavior and be financiallyresponsible, governments must clearly demonstratetheir willingness to allow enterprises to close. Thegovernment should identify a narrow set of enter-prises responsible for a large proportion of the arrears.After monitoring these enterprises for a clearly de-fined period, the government should decide on theirlong-term viability. If they are deemed to be not vi-able, then bankruptcy proceedings should be initiatedby the government. In exceptional circumstances, ifbankruptcy is not a viable option for a particular en-terprise because of strategic considerations, the flowproblem should be dealt with by restructuring the en-terprise and identifying budgetary support over themedium term. Furthermore, it is important to makethe provision of budgetary support conditional on,inter alia, the clearance of existing arrears and theiravoidance in the future.37

Second, to ensure that managers are accountablefor all payments obligations of their enterprises (in-cluding taxes and wages), governments should im-pose financial and administrative penalties on man-agers of state enterprises, including termination ofservice, and should acquire the legal authority toplace restrictions on the wage bill of state enterprisesthat have arrears in excess of agreed limits.38

36According to Rostowski (1993), p. 155, in Poland in mid-1991, that is, 18 months into the stabilization program, there wereonly six trained liquidators for the courts to hire. The slow start-up in bankruptcies is illustrated by the experience of the CzechRepublic, where according to Hrncir and Klacek (1995), the num-ber of declared bankruptcies went from 5 in 1992 to 60 in 1993and to 254 in 1994.

37In Poland, more than 1,200 enterprises have been liquidated bythe government because of their financial nonviability. In late1992, the Government of Estonia initiated bankruptcy proceedingsagainst two state enterprises because of tax arrears and insolvency.

38In Hungary, for example, the 1991 law on bankruptcy madethe failure on the part of managers to declare bankruptcy, whenpayments were overdue for more than 90 days, punishable ac-

Third, for customers with arrears in excess ofagreed limits, state-owned enterprises should be re-quired by law to ship products only on the basis ofpreshipment payment or on the basis of promissorynotes of the receiver of goods and services, dulyguaranteed by commercial banks.39

Fourth, central banks should impose prudentiallimits on the amount of guarantees that a bank canextend in relation to its total assets. Promissory notesissued by enterprises should never carry governmentor central bank guarantees.

Finally, the timely and regular monitoring of theunderlying financial position of state enterprises aswell as their arrears should be ensured. A standarddefinition of overdue claims should be introducedand firms should be required to observe well-docu-mented and transparent reporting requirements on aperiodic basis.

Interstate ArrearsInterstate interenterprise arrears among the Baltics,

Russia, and other states of the former Soviet Unionare an important problem, contributing significantlyto interenterprise arrears in a number of sectors (par-ticularly energy). With regard to these arrears, it is im-portant to distinguish between quasi-governmental ar-rears incurred by state enterprises operating underintergovernmental trade agreements, and others.Much of the interstate interenterprise arrears pertain-ing to the energy sector are of the quasi-governmentalvariety, as in most cases trade in energy products was,and still is, carried out on the basis of intergovern-mental agreements on trade volumes, prices andmeans of payments.40

Governments should take responsibility for inter-state interenterprise arrears of the quasi-governmen-tal variety. The existing stock of such arrears shouldbe regularized through the negotiation of reschedul-ing agreements with a realistic mechanism for settle-ment. In working to prevent the emergence of newarrears of this type, governments should removethemselves from intergovernmental trading activities.In the process, they should make it clear that theywill not undertake any future responsibility for pay-ments related to this trade; in so doing, expectations

cording to the provisions of the Civil Code. See Hungary (1991),Section 9, p. 19.

39In the former Yugoslavia in the 1980s, promissory notes witha maturity up to 90 days, which were the most important categoryof interenterprise credits, had to be guaranteed by a bank andbacked by a physical transaction.

40Some of these agreements call for parallel deliveries ofgoods, or essentially barter trade. This frequently complicatesboth the measurement of the arrears problem and the determina-tion of who bears the ultimate responsibility for the arrears.

Policy Option s

73



of financial assistance to suppliers in the future willbe avoided. Should the government feel compelled inthe immediate future to remain involved in interstatetrade to some degree, any payments obligationsshould be clearly defined at the outset and reflectedin the budget. Moreover, all parallel trade featuresunder intergovernmental agreements should be elimi-nated so as to make trade transactions and associatedpayments fully transparent. All this would also fosterthe development of a transparent and market-orientedtrading environment within the region.

Enterprises with overdue claims on others acrossstate boundaries, without any involvement of thegovernment, should be left on their own to take re-sponsibility for their decision to supply goods oncredit. However, any resolution of the past stock ofinterstate interenterprise arrears of the purely non-governmental variety is crucially dependent ondebtor enterprises having full access to foreign ex-change to settle their debts. Governments should en-sure availability of foreign exchange for settlementof interstate interenterprise arrears.

The Energy Sector

It is clear that the energy sector (including elec-tricity, oil, gas, coal, and peat) plays a critical role inthe arrears situation in Russia and Turkmenistan.Most of Turkmenistan's, and also Russia's, exportsof energy products to other states of the former So-viet Union are still carried out on the basis of inter-governmental or quasi-governmental agreements ontrade volumes, prices, and means of payment. Asnoted above, intergovernmental trading activitiesshould be eliminated as soon as possible. In the in-terim, they should be regarded as quasi-fiscal opera-tions and be brought on budget, and the energy en-terprises should be paid market prices on theirsupplies to other states of the former Soviet Union.

Conclusions

Interenterprise credit and arrears have increased inthe Baltics, Russia, and other states of the former So-viet Union during the transition process. Such an in-crease is to be in part expected as a natural response tothe transformation from central planning to a market-oriented economy. Furthermore, in a number of coun-tries the levels of these credits and arrears appearmodest by the standards of other east European coun-

tries during comparable periods of transition. In anumber of countries, however, there were attempts tocheck the rapid increases in arrears by official nettingout and clearing operations, although these had infla-tionary consequences and sustained the lack of finan-cial discipline in the enterprise sector. Also, in somecountries, these credits and arrears have been heavilyconcentrated in a limited number of sectors, raisingconcerns that governments will not succeed in impos-ing discipline because enterprises in those sectors willbe perceived as "too important to fail."

The experience of Latvia, Lithuania, and severaleast European countries clearly demonstrates the im-portance of moving toward an efficient financial andpayments system, achieving macroeconomic stabil-ity, and furthering structural reform in promoting theability and willingness of enterprises to honor theirfinancial obligations on time. The accumulation ofinterenterprise arrears in a number of countries (e.g.,Azerbaijan, Belarus, and Ukraine), on the otherhand, is one of the many manifestations of slippagesin stabilization and reform efforts during the transi-tion process.

The experience with the exceptional measurestaken in a number of countries clearly demonstratesthe temporary nature of relief to the arrears problemprovided by such measures. Netting out and bailoutexercises merely address symptoms rather than un-derlying causes, and indeed perpetuate and com-pound the problem by strengthening expectations offuture bailouts of debtor enterprises by the govern-ment. Instead, governments should maintain ahands-off policy and deal with arrears by promotingthe financial health and discipline of enterprisesthrough the maintenance of a stable macroeconomicenvironment, ensuring a sound banking system, andestablishing the legal and institutional prerequisitesof a market system through accelerated structural re-form. Among the latter are bankruptcy procedures. Itshould be recognized, however, that improvementsin these procedures are likely to add to financial dis-cipline only gradually because of the need not onlyto pass the relevant legislation but also to developcourts and other institutions and to improve the ben-efit-cost ratio of enforcing bankruptcies. Thus, untilbankruptcy procedures are working smoothly, gov-ernments—as the formal owners of state-owned en-terprises—should take a highly visible role by iden-tifying state enterprises that are major offenders offinancial discipline through the accumulation of ar-rears, and initiating steps to restructure, privatize, orclose them down.

74


ReferencesBigman, David, and Servio Pereira Leite, "Enterprise Ar-

rears in Russia: Causes and Policy Options," IMFWorking Paper, WP/93/61 (Washington: InternationalMonetary Fund, August 1993).

Begg, David, and Richard Portes, Enterprise Debt andEconomic Transformation: Financial Restructuringof the State Sector in Central and Eastern Europe,CEPR Discussion Paper, No. 695 (London: Centrefor Economic Policy Research, June 1992), pp. 1-43.

Calvo, Guillermo A., and Fabrizio Coricelli, "MonetaryPolicy and Interenterprise Arrears in Post-Commu-nist Economies: Theory and Evidence," Center forInternational Economics, Department of Economics,University of Maryland at College Park, WorkingPaper Series, No. 6, October 1994.

Caprio, Jr., Gerard, and Ross Levine, "Reforming Fi-nance in Transitional Socialist Economies," TheWorld Bank Research Observer, Vol. 9 (January1994), pp. 1-24.

Crawford, Paul, "A Survey of the Trade Credit Litera-ture," University of Bristol, Department of Econom-ics, Discussion Paper No. 92/324, March 1992.

Fan, Qimiao, and Mark E. Schaffer, "Government Finan-cial Transfers and Enterprise Adjustments in Russia,with Comparisons to Central and Eastern Europe,"Economics of Transition, Vol. 2 (June 1994), pp.151-87.

Hrncir, Miroslav, and Jan Klacek, "Interenterprise Indebt-edness and Performance of Banking System," paper

presented to the Bucharest Workshop organized withinthe ACE Research Project-920379-R, February 1995.

Hungary, Ministry of Finance, "Public Finance in Hungary,No. 87: Act on Bankruptcy, Winding-Up Proceduresand Final Accounting" (Budapest:, 1991).

Ickes, Barry W., and Randi Ryterman, "The InterenterpriseArrears Crisis in Russia," Post-Soviet Affairs, Vol. 8(October-December 1992), pp. 331-61.

Khan, Mohsin, and Eric Clifton, "Interenterprise Arrearsin Transforming Economies: The Case of Romania,"IMF Paper on Policy Analysis and Assessment, No. 1(Washington: International Monetary Fund, 1992).

Kim, Se-Jik, and Goohoon Kwon, "Equilibrium Inter-enterprise Arrears in Transition Economies: A Gen-eral Equilibrium Approach to Russian Experiences"(mimeograph; Washington, International MonetaryFund, January 1995).

Mitchell, Janet, "Creditor Passivity and Bankruptcy: Impli-cations for Economic Reform," in Capital Markets andFinancial Intermediation, ed. by Colin Mayer andXavier Vives (Cambridge: Cambridge University Press,1993).

Perotti, Enrico C., "Collusive Arrears in TransitionEconomies" (mimeograph; Washington, June 23, 1994).

Rostowski, Jacek, "The Interenterprise Debt Explosion inthe Former Soviet Union: Causes, Consequences,Cures," Communist Economies and Economic Trans-formation, Vol. 5, No. 2 (1993), pp. 131-59.

,"Interenterprise Arrears in Post-CommunistEconomies," IMF Working Paper WP/94/43 (Wash-ington: International Monetary Fund, April 1994).

References

75


V Th e Revenu e Declin eRichard Hemming, Adrienne Cheasty, and Ashok K. Lahiri

Since the dissolution of the U.S.S.R. and the begin-ning of the transition to market-based economies,

a significant deterioration in fiscal revenue perfor-mance has occurred in the Baltic countries, Russia,and other countries of the former Soviet Union.1 Ex-cluding Russia, revenue fell by an average of 5 percentof GDP between 1991 and 1993 and is estimated tohave fallen farther, by about 4 percent of GDP, in1994. Revenue-to-GDP ratios have fallen in mostcountries, by wide-ranging degrees. Data for 1991 arenot available for Russia, but the revenue-to-GDP ratiois estimated to have fallen by about 5 percent of GDPbetween 1992 and 1994.

This revenue decline has to be viewed in the contextof the broader reforms in the region. In moving fromcentral planning to the market, one objective is to re-duce the all-encompassing role of government. If gov-ernment expenditure was falling as a result, the rev-enue decline would be a concern mainly insofar as itexceeded government policy objectives for expendi-ture reduction and threatened stabilization objectives.Throughout the region, however, a wide range of im-plicit fiscal activities previously undertaken by enter-prises and quasi-governmental institutions have beentransferred to the budget and made explicit. Hence,government spending has in most countries been ris-ing as the role of government has been reduced. Thishas resulted in widening fiscal imbalances, which havemade inflation difficult to control. Where monetarypolicy has been tightened as a consequence, growingdeficits are stifling new private activity and crowdingout potentially viable state enterprises.

Note: Guzel Anulova, Anthony Pellechio, Volker Treichel, andAsegedech Woldemariam collaborated as coauthors in the pro-duction of this paper. The authors wish to thank Teresa M. Ter-Minassian, Milka Casanegra de Jantscher, and Howell Zee fortheir valuable comments.

1The revenue decline represents an acceleration of the trend al-ready observed in the U.S.S.R. during 1985-89 when state budgetrevenue as a percentage of GDP fell from 47 percent to 41 per-cent. This process can therefore be viewed as having started withperestroika. See International Monetary Fund, World Bank, Orga-nization for Economic Cooperation and Development, and Euro-pean Bank for Reconstruction and Development, A Study of theSoviet Economy (Paris: OECD, 1991).

This section discusses the factors underlying therevenue decline and lays out policy options for cop-ing with it. The sharp fall in output throughout thecountries of the region is the immediate reason whyrevenues have fallen. It is not, however, sufficient toexplain the alarming fall in revenue relative to GDPthat has occurred in many countries. Part of the ex-planation can be found in the changing compositionof economic activity and the erosion of traditionaltax bases. But difficulties experienced by the newlyindependent countries in integrating the emergingprivate sector into the tax net, developing an appro-priate framework for tax policy, and establishing aneffective tax administration have also contributedsignificantly to the revenue decline.

For the immediate future, an improvement in rev-enue performance is critical for stabilization. Thiscan be achieved through policy and administrativemeasures to strengthen existing taxes and throughthe introduction of more effective new taxes. Overthe medium term, tax reform should not only be de-signed to support efficient growth, but also be di-rected toward ensuring that recovering economic ac-tivity is effectively taxed, so that the budget capturessufficient resources to finance the legitimate func-tions of government. A central aim, both for the im-mediate future and the medium term, is to imposetaxes more fully on the private sector, with a viewnot only to raising revenue but also to securing amore equitable distribution of the tax burden.

Revenue Developments Since 199 1Revenue developments in the Baltic countries,

Russia, and other countries of the former SovietUnion have not been uniform across countries. Theway in which total revenue and its major tax compo-nents have behaved during the early years of transi-tion are described here. While the reasons for therevenue trend in certain countries (particularly themore extreme cases) and for the behavior of sometax categories are commented upon, a full discussionof the factors explaining revenue developments ap-pears later.

76


Revenue Developments Since 1991

Movements in Total Revenue

Between 1991 and 1993, revenue declined relativeto GDP in most countries of the region.2 For a numberof countries, a contributing factor to this decline wasthe cessation of grants from the union governmentfollowing the dissolution of the U.S.S.R. (Table 5.1).Union grants—ranging from around 5-20 percent ofGDP—were extended to less developed republics andto some energy-exporting republics to compensate fortaxes implicit in the U.S.S.R.'s low administered en-ergy prices. They were discontinued in 1992, and sev-eral countries (Azerbaijan, Kazakstan, the Kyrgyz Re-public, Tajikistan, Turkmenistan, and Uzbekistan)were faced with sudden cuts in revenue.3 The energyexporters, Azerbaijan and Turkmenistan, offset theirlosses through foreign exchange taxes on energy; andKazakstan received payments from foreign oil com-panies, which partly covered its shortfall. The remain-ing countries, however, were unable to recover lostgrants.4

Revenue excluding grants declined in ten coun-tries (excluding Russia) by an average of 10.3 per-cent of GDP between 1991 and 1993, while therewas an increase in four countries averaging 5.9 per-cent of GDP (Table 5.2). Owing to problems associ-ated with the succession of the Soviet Union to theRussian Federation, no reliable 1991 data are avail-able for Russia. Without such data, it is impossibleto provide a definitive accounting of revenue trendsfor the entire territory of what used to be theU.S.S.R. However, the revenue-to-GDP ratio in Rus-sia has fallen since mid-1992.

Among the four countries where the revenue-to-GDP ratio increased over the two-year period, therelatively favorable outcome can be attributed tospecial factors. In Uzbekistan, for example, whilethe government raised some revenue through gold-related transactions, excise and export taxes on cot-ton, together with a good cotton crop, accounted fora large part of the higher revenue in 1992 and 1993.Extraordinary factors also led to some of the moredrastic declines in revenue. An important sequel tothe dissolution of the U.S.S.R. has been the eruption

Table 5.1. Baltic Countries, Russia, andOther Countries of the Former SovietUnion: Revenue and Grants, 1991(In percent of GDP)

2Revenue in this paper refers to revenue of the general govern-ment, including social security contributions and other payrolltaxes. It does not take into account quasi-fiscal revenues otherthan profit transfers from the central bank. It should be noted thatGDP estimates are of limited reliability. See Section II for furtherdiscussion.

3Donor countries—chiefly Russia—gained correspondingly.4For a discussion of the tax policy response in Central Asia in

1992, see Parthasarathi Shome and Julio Escolano, "The State ofTax Policy in the Central Asian and Transcaucasian Newly Inde-pendent States (NIS)," IMF Paper on Policy Analysis and Assess-ment, PPAA/93/8 (Washington: International Monetary Fund,July 1993).


KazakstanKyrgyz RepublicLatviaLithuaniaMoldova


Total

25.651.143.941.033.8

25.034.837.444.031.6

39.246.536.549.1

Of Which:Union Grants

13.6

—

4.512.4

——

19.08.3

18.5

Source: Data provided by country authorities.

of regional wars. Armed conflicts involving Arme-nia, Georgia, and Moldova led to particularly seriousdifficulties in revenue collection. In two energy-ex-porting countries, Azerbaijan and Turkmenistan,revenue performance has fluctuated sharply—rally-ing in 1992 but then collapsing in 1993—reflectingthe instability of the energy sector to respond tochanging relative prices and real exchange rates andpayment difficulties of trading partners.

With real output dropping sharply across the re-gion, the deterioration in revenue is even more strik-ing when measured in real terms. Revenue in realterms, derived using the GDP deflator, declined inall but one country—Uzbekistan—between 1991and 1993 (Table 5.3).5 In six countries—Armenia,Georgia, Kyrgyz Republic, Lithuania, Moldova, andTurkmenistan—a decline in excess of 50 percentwas registered.

By 1993, the (unweighted) average revenue-to-GDPratio for the Baltic countries, Russia, and other coun-tries of the former Soviet Union had fallen to 29 per-cent, with a range of 3—44 percent. It is instructive tocompare these with the comparable ratios in western

5The consumer price index (CPI) could have been used to de-flate revenue. While it is in many cases a better indicator of un-derlying inflation, the GDP deflator more closely relates to theoverall resource base from which revenue is derived.

77


V REVENU E DECLIN E

Table 5.2. Th e Baltics, Russia, and Other Countries of the Forme rSoviet Union: Revenu e Developments(In percent of GDP)

Countries with an increase inrevenue-to-GDP ratio

KazakstanTajikistanUkraineUzbekistan

Average

Countries with a decrease inrevenue-to-GDP ratio

ArmeniaAzerbaijanBelarusEstoniaGeorgiaKyrgyz RepublicLatviaLithuaniaMoldovaRussiaTurkmenistan

Average


1991

20.520.236.530.626.9

25.637.543.941.033.822.437.444.031.6

38.2

35.51

33.11

1992

22.926.641.531.330.8

26.749.042.235.115.316.528.133.727.037.642.2

32.1

31.7

1993

22.326.941.141.032.8

23.536.943.636.6

2.913.635.827.622.936.719.2

27.2

28.7

Change1991-93

1.76.74.6

10.45.9

-2.1-0.6-0.4-4.4

-30.9-8.8-1.6

-16.4-8.7

-19.0

-10.31

-5.0'

Sources: Data provided by country authorities; and IMF staff estimates.1Excluding Russia.

Europe and eastern Europe. In the former, the averagetax-to-GDP ratio alone was over 40 percent in 1993,with a range of 23-50 percent. In the latter, the averagerevenue-to-GDP ratio was 41 percent, with a range of2872-54 percent. It is also instructive to compare therevenue decline in the Baltic countries, Russia, andother countries of the former Soviet Union with that ineastern Europe. Between 1989 and 1991, the revenue-to-GDP ratio fell by over 10 percent in the latter, com-pared with about 5 percent of GDP in the former duringthe first two years of transition. By 1993, the ratio foreastern Europe had fallen by a further 3 1/2 percent. Therevenue decline in the Baltic countries, Russia, andother countries of the former Soviet Union has there-fore not yet been as sharp as in eastern Europe; how-ever, except in a few countries, the revenue-to-GDPratio is already comparatively low.

(comprising personal income tax and payroll taxes);value-added tax (VAT) and other domestic indirecttaxes; trade taxes; and a residual category consistingof other taxes and nontax items.6 Table 5.4 presentsa decomposition of the cumulative decline in therevenue-to-GDP ratio between 1991 and 1993 interms of the contributions of these five components.

Generally speaking, each of the major taxes excepttrade taxes contributed to the revenue decline. Tradetaxes were buoyant because tariffs replaced the wide-ranging quantitative controls of the Soviet regime.VAT and other indirect taxes were the major factor be-hind the decline in the revenue-to-GDP ratio in coun-tries where the ratio declined. With the onset of re-forms, all countries except Lithuania replaced theturnover tax with a VAT in 1991-92.7 However, as dis-

Movements in Individual TaxesTotal revenue can be split into five broad compo-

nents: enterprise taxes; taxes on wages and salaries

6Enterprise taxes cover all direct taxes paid by enterprises.Trade taxes include both customs duties and export taxes.

7Lithuania replaced the General Excise Tax by a VAT in May1994.

78


Revenue Developments Since 1991

Table 5.3. Th e Baltics, Russia, and Other Countries of the Forme rSoviet Union: Change in Real Revenue1

(In percent)

Countries with no decline2

Uzbekistan

Countries with a less than 50 percent decline2

AzerbaijanBelarusEstoniaKazakstanLatviaRussiaTajikistanUkraine

Average

Countries with a more than 50 percent decline2

ArmeniaGeorgiaKyrgyz RepublicLithuaniaMoldovaTurkmenistan

Average


1991-92

-8.9

1.3-13.2-29.0

-3.9-51.3

1.4-5.6

- I4 .3 3

-50.0-74.2-40.4-52.5-36.4

4.6

-41.5

-25.63

1992-93

27.4

-34.9-8.8

2.2-4.4

8.7-15.8-20.7-15.0

-11.1

-43.1-88.9-30.9-31.3-27.0-58.7

-46.6

-22.0

1991-93

16.1

-34.0-20.8-27.5

-8.2-47.1

-19.6-19.8

-25.33

-71.5-97.1-58.8-67.4-53.6-56.7

-67.5

-40.43

Sources: Data provided by country authorities; and IMF staff estimates.1Derived using the GDP deflator.2Over two years.

Excluding Russia.

cussed below, its introduction was fraught with com-plications, many of which have still not been resolved.

Taxes on wages and salaries also made an impor-tant contribution to the revenue decline. The fall inreal wages observed throughout the region caused re-ceipts from payroll and individual income taxes to besignificantly eroded. But this drop was not compen-sated by a rise in profit taxes, because of a shrinkingenterprise tax base, reductions in tax rates, and otherfactors discussed below. Hence, enterprise taxes alsocontributed to the overall decline, but they were notthe overriding factor they appear to have been in east-ern Europe.8.The remainder of the revenue decline re-flects changes in other tax and nontax items, whichinclude natural resource taxes (the largest compo-

8See International Monetary Fund, "Eastern Europe—FactorsUnderlying the Weakening Performance of Tax Revenue," IMFWorking Paper, WP/94/104 (Washington: International MonetaryFund, September 1994).

nent), property taxes, privatization receipts, explo-ration fees, and transfers of central bank profit. Thiscategory made an unusually large contribution inAzerbaijan and Turkmenistan, where energy tax re-ceipts collapsed in 1993.

Developments in 1994Limited data on revenue developments during the

first half of 1994 are available for nine countries.9

They indicate that the declining trend in revenuecontinued into 1994, and estimates for the full yearassume no significant recovery in the second half ofthe year (Table 5.5).

There were some notable changes in the revenuetrend in 1994. In Kazakstan, following a relatively

9This paper was written in early 1995, and the data analyzedare those available until the end of 1994. Eds.

79


V REVENU E DECLIN E

Table 5.4. The Baltics, Russia, and Other Countries of the Former Soviet Union:Decomposition of the Change in the Revenue-to-GDP Ratio, 1991-93(In percent of GDP, over two years)

Countries with an increase inrevenue-to-GDP ratio

KazakstanTajikistanUkraineUzbekistan

Average

Countries with a decrease inrevenue-to-GDP ratio

Armenia1

AzerbaijanBelarusEstoniaGeorgiaKyrgyz RepublicLatviaLithuaniaMoldovaRussia1

Turkmenistan

Average


Totalrevenue

1.76.74.6

10.4

5.9

-3.1-0.6-0.4-4.4

-30.9-8.8-1.6

-16.4-8.7-0.9

-19.0

-8.6

-4.7

Enterprisetaxes

-3.31.3

-0.23.1

0.2

-3.61.7

-2.2-4.0-5.7-0.6

0.8-1.6-3.1

1.61.0

-1.4

-1.0

Of

Taxes onwages and

salaries

-0.9-0.3

0.52.0

0.3

-3.9-0.2

2.42.5

-9.2-1.3

0.9-3.7-4.6

0.1-1.0

-1.6

- I . I

Which: Contribution

VAT and otherindirect taxes

-2.63.73.22.0

1.6

-3.311.4-0.3-0.9-9.1-4.0-2.8-7.3-0.5-4.7-1.3

-2.1

- I . I

from

Tradetaxes

1.50.91.01.4

1.2

0.22.14.1

-1.8-0.4

—1.40.80.91.07.2

1.4

1.4

Others

7.1I.I0.11.9

2.6

7.5-15.6

-4.3-0.2-6.6-2.9-1.9-4.7-1.4 I

1.0 |-24.8 |

-A3

-2.9

Sources: Data provided by country authorities; and IMF staff estimates.11992-93.

modest decline in 1993, the revenue-to-GDP ratiofell sharply in 1994. The decline in 1993 reflected areduction in the VAT rate, growing tax arrears, and adrop in foreign trade. The arrears problem worsenedin 1994 and was compounded by a failure to imple-ment budgeted revenue measures. To a lesser degree,a similar pattern has emerged in Armenia, althoughthe deterioration in the first half of 1994—mainlydue to a precipitous drop in real wages—moderatedin the second half.

In Belarus and Estonia, an improvement in rev-enue performance in 1993 was reversed in 1994. InBelarus, this reflected mainly the impact of discre-tionary tax exemptions and a fall in real wages,while in Estonia income tax rates were lowered,mainly with longer-term supply-side considerationsin mind. For Turkmenistan, the 13 percent of GDPfall in the revenue-to-GDP ratio in 1994 has matchedthat in 1993, mainly reflecting a worsening problemof nonpayment for gas exports by Ukraine and Geor-

gia. In Russia, the revenue decline of 1993 intensi-fied in 1994, giving an overall decline of 4.7 percentof GDP between 1992 and 1994. Elsewhere, thetrend observed in 1992 and 1993 has continued(Lithuania and Moldova) or the revenue-to-GDPratio has changed only slightly (the Kyrgyz Repub-lic, Latvia, and Ukraine).

The continuation of the revenue decline into 1994raises the issue of whether it is likely to continueinto 1995 and beyond. Eastern Europe's experiencesuggests that it might, although any such judgment isnot independent of the sources of the revenue de-cline and the policy response to it. Because of datalimitations, revenue developments in 1994 cannot beanalyzed in as much detail as those in 1992 and1993. However, the discussion above has pointed tosome of the influences on developments in 1994.This chapter now turns to a more systematic analysisof the determinants of developments in the two pre-ceding years.

80


Factors Explaining Revenue Developments

Table 5.5. The Baltics, Russia, and OtherCountries of the Former Soviet Union:Change in Revenue-to-GDP Ratio1

(In percent of GDP)

ArmeniaBelarusEstoniaKazakstanKyrgyz Republic

LatviaLithuaniaMoldovaRussiaTurkmenistan

Ukraine

FirstHalf

-8.5

-3.2-8.00.7

-0.3-3.9-5.0-5.0-11.4

1994

SecondHalf2

-2.9

-0.8-6.2- I . I

2.0-3.2-4.0-3.1

-14.6

FullYear2

-4.7-4.4-1.9-6.6-0.3

0.9-3.5-4.4-3.8

-13.0

1.7


1Compared with the ratio for full-year 1993.2Estimates.

Factors Explaining RevenueDevelopments

Output in the Baltic countries, Russia, and othercountries of the former Soviet Union declined sharplyduring 1991-93 in response to the disruption of inter-state trade in the wake of the disintegration of theU.S.S.R., the massive adjustment problems associatedwith the transition from central planning to market,and, for many countries, the adverse terms of tradeshock arising from the switch to world prices for im-ported energy and raw materials. Regional conflictsalso contributed to the fall in output.

The way in which an output decline affects rev-enue reflects a number of factors. How the structureof economic activity changes is especially impor-tant. The bases of the main taxes derive from majorincome and expenditure components of GDP (prof-its, wages, and consumption). If an output decline iswidespread and the tax system is broad based, realrevenue will fall, but the revenue-to-GDP ratioshould not be significantly affected. But if an outputdecline is uneven, and if revenue derives more fromtax bases that have fallen relatively fast, the revenue-to-GDP ratio will decline. As will be shown below,the shrinkage of the major tax bases relative to GDPis one of the main explanations for the decline in therevenue-to-GDP ratio.

The design of the tax system is also important. Ifthe tax system is elastic, in the sense that tax revenueincreases or decreases faster than the underlying taxbases, the revenue-to-GDP ratio will tend to fall asGDP falls.10 Unfortunately, it is difficult to formjudgments about the elasticity of a tax system whenstructural reform is ongoing. In addition, as the taxstructure is changing, new demands are placed ontax administration. Outmoded administrative struc-tures have come under increasing strain. However,new systems and procedures are being put in place,but they are yet to have their full impact on adminis-trative efficiency. Finally, and of considerable im-portance, the tax system is having difficulty copingwith the growth of private sector activity.

The remainder of this section analyzes the maincauses of the revenue decline in terms of factorsmentioned above—changes in the level and struc-ture of economic activity, the design of the tax sys-tem, and the efficiency of tax administration. Theimpact of growing private sector activity is assessed,and the crisis in governance is also discussed.

Changes in Level and Structureof Economic Activity

The traditional tax bases in the Baltic countries,Russia, and other countries of the former SovietUnion are state enterprise surpluses, wages andsalaries paid mainly by enterprises, and retail salesprimarily through state stores. One of the explana-tions for the declining yield from the major taxes isthe disproportionate shrinkage in their bases relativeto GDP.

Declining Tax Bases

Recorded industrial production by the enterprisesector dropped sharply between 1991 and 1993(Table 5.6). For the countries where data are avail-able, the fall in real profits was even larger. How-ever, except in Kazakstan, the enterprise tax base in-cludes some or all of the wage bill, in which casechanges in industrial production probably provide abetter guide to the evolution of the tax base. The fallin industrial production and profits puts downwardpressure on real wages and on consumer demand.Unemployment also emerged. Thus, the bases ofother taxes—taxes on wages and salaries, and VATand other indirect taxes—also fell sharply.

The fall in tax bases was in many cases (i.e., for anumber of tax categories and countries) larger than

10This elasticity can result from several sources, including pro-gressivity of rates and differentiation of rates across various in-come classes or groups of commodities.

81


V REVENU E DECLIN E

Table 5.6. Th e Baltics, Russia, and Other Countries of the Former Soviet Union: Change in Tax Bases and RealRevenue, 1991-93 1

(Percentage change, over two years)


KazakstanKyrgyz Republi cLatviaLithuaniaMoldova


Industrialproduction

-54-29 /-20 * /-68 * /-60

-28 *-44 * /-54 * /-65 * /-36 *

-31-39 /-10-16-17

Tax Bases

Realprofits

-98

-52 * /

-90 * /-99 * /

-82 *

-14

Realwages

-92-492

-22-55 * •-80 *

72-62 *-23-61 * /

- 52 2

-31

-29 *-61 * /

72

Real retai lturnover

-83-79-38-56-93

-50-67-52-70-66

-35-81-46-47-40

* /* /*

* /* /*** /

* /* /* /

Enterprisetax

-64 2

-14-36-58-96

-54-42-38-60-61

2 2

-299

-3030

Real Revenu e

Taxes on wagesand salaries

-662-34

3- 6

-97

-40-69-41-61-61

-13 2

-46-62-27

46

VAT and otherindirect taxes

-58 2

-15 2

-22-26-97

-47-61-60-73-40

-46 2

-14-34

-54

MemorandumItem

RealGDP

-69-33-20-19-66

-15-32-45-48-36

-30-40-12-29-13

Sources: CIS Goskomstat; IMF Economic Reviews; and IMF staff estimates.1For countries where revenue has declined, a tax base decline greater than or equal to the fall in real GDP is indicated by an *, while a tax base decline greater than or

equal to the fall in real revenue is indicated by a / . Only cases where data are available for the same period for tax bases, real revenue, and real GDP are marked.21992-93.

the fall in real GDP and the corresponding revenuedecline. For example, shrinking tax bases are morethan sufficient to explain an overall drop in the rev-enue-to-GDP ratio in Estonia. To varying and gener-ally lesser degrees, declining tax bases have had asimilar impact in other countries. However, the Esto-nia case also illustrates that there is more going on.For example, while real wages fell by 55 percent,revenue from taxes on wages and salaries droppedby only 6 percent. Clear anomalies (most notably inUzbekistan, where falling tax bases were accompa-nied by increasing revenue) also highlight that thedecline in traditional tax bases is only part of theoverall story. So do the many cases where the fall inrevenue cannot be explained by declining tax basesalone. In Estonia, taxes on wages and salaries wererelatively buoyant because the tax system was capa-ble of taxing incomes outside the enterprise sector.But this has not proved to be the rule.

The Impact of an Emerging Private Sector

To the extent that profits and wages have fallenfaster than output, other incomes must be falling lessfast. National accounts data reveal that agricultural

incomes have increased as a share of GDP across theregion. But there has also been a large and increas-ing residual, which can be largely attributed to pri-vate or informal sector incomes. A significant shareof output in most countries of the region has nowshifted to the private sector, because new firms havebeen established or have emerged through privatiza-tion, and because an atomistic trade and servicessector is developing rapidly. Such firms (and indi-viduals) have not contributed significantly to tax re-ceipts, because of weaknesses in tax policy and taxadministration.

One implication of such weaknesses—which aredescribed below—is that the shift in production to theemerging private sector would probably have reducedenterprise tax revenue even if output had not fallen.11

Moreover, to the extent that private firms are not cap-tured in the tax net, both taxes on wages and salaries,and VAT and other indirect taxes, may not be fully col-lected from them. It is also worth noting that while pri-

11This conclusion is also supported by the revenue experienceof the last few years in China, where the revenue-to-GDP ratiohas declined despite rapid growth in GDP.

82


Factors Explaining Revenue Developments

vate sector activity is at least partly reflected in GDPdata, to the extent that it is not fully captured, the rev-enue-to-GDP ratio is in fact overestimated by the cur-rently available data. In other words, the revenue de-cline may be larger than indicated above.

Design of the Tax SystemWhile the decline in the major tax bases can ex-

plain many instances of the revenue decline, revenueshave also been eroded because of shortcomings in taxdesign. Some of these are a legacy of the plannedeconomy tax system; others have been the unintendedconsequence of the reform effort.

Some general features of reforms in the Balticcountries, Russia, and other countries of the formerSoviet Union are worth noting at the outset. First,the most significant reforms of regional tax struc-tures took place earlier than, or at the beginning of,the period covered in this study—notably, the enter-prise tax reform and the replacement of the turnovertax by the VAT. Consequently, formal tax structuresin the region looked quite like those of countries inwestern Europe before the onset of the revenue de-cline. Second, the fact that faster reformers—like theBaltic countries—have experienced revenue lossesbroadly similar in magnitude to those of someslower reformers in the region suggests that the paceof tax reform has not been critical. The major prob-lem in the area of tax policy has been that recentchanges to the design of taxes in some countries,rather than furthering the transformation of the taxstructure, have been retrogressive.

Changes in Enterprise Taxes

The enterprise tax provides the clearest exampleof this retrogression. The U.S.S.R. introduced an or-thodox profits tax at the beginning of 1991. It had acentral rate of 45 percent and a base defined more orless in accordance with industrial country account-ing principles.12 Since then, the base has beenchanged and the rates have drifted downward.

First, and most notably, almost every country (asalready noted, the exception is Kazakstan) has in oneway or another reintroduced wages into the tax base.In some cases, this almost doubled the base. Second,and working in the opposite direction, the base hasbeen eroded as exemptions and other tax preferenceshave proliferated. Third, all countries have cut the

basic tax rate and many have returned to a differenti-ated rate system. In some countries, most notablyRussia, this began in 1991 when they started to enterinto tax competition with the U.S.S.R. government;in others it came later, to compensate for the expan-sion of the base. In no country is the basic rate nowgreater than 35 percent. In some countries (Ukraine,Uzbekistan), it fell to 18 percent in 1993.13 Finally,six countries (Armenia, the Kyrgyz Republic,Moldova, Turkmenistan, Ukraine, and Uzbekistan)reintroduced excess profit taxes or asset taxes. Theseare essentially taxes on capital, justified as proxydividends paid by the state enterprises to theirowner, the government.

The impact of these reforms on revenue is hard togauge. The capital taxes and the reinclusion ofwages in the profit tax base clearly boosted revenuebeyond the yield of a pure profit tax. However, forthose countries where the cuts in enterprise tax rateswere large (such as Belarus, Kazakstan, Turk-menistan, Ukraine, and Uzbekistan), these causedsignificant offsetting revenue losses.

From Turnover Tax to VAT and Excises

An important reason for the underperformance ofthe VAT has been the contraction in consumption.However, there have also been a large number offlaws in VAT design, legislation, and implementation.The choice of rates has been perhaps the single majordifficulty. At the outset, it was difficult to estimaterevenue-neutral VAT rates, given the many turnovertax rates and the large and staggered impact that priceliberalization was having on relative prices. Hence, in1992, half the countries of the region lost revenuewhen they introduced the VAT. Despite this loss, anumber of countries (Armenia, Azerbaijan, Georgia,Kazakstan, Moldova, and Russia) cut their rates or in-troduced differentiated rates in 1993, or both, with aresulting adverse effect on VAT revenue. By contrast,Estonia increased its VAT rate, thus ameliorating therevenue decline.14 In addition, some particular fea-tures of the VAT system as adopted hinder timely andeffective revenue collection.

(1) Cash basis versus accrual basis. The generalrule is to allow taxpayers to use a cash rather than anaccrual basis for establishing VAT liability.15 In gen-eral, this procedure delays VAT receipts, since pay-ments are slow. Levying VAT on a cash basis, and al-

12This reform was estimated to be revenue-losing. According toIMF, World Bank, OECD, and EBRD, A Study of the Soviet Econ-omy, a 55 percent rate would have been revenue neutral. The actualrevision of accounting procedures at the enterprise level, however,has been protracted and is still continuing in most countries.

13By comparison, the OECD average basic rate is about 40 per-cent. Basic rates are also higher in eastern Europe. In 1993, it was40 percent in Bulgaria, 45 percent in the Czech Republic, 36 per-cent in Hungary, and 40 percent in Poland.

14To increase VAT revenue, Russia introduced a 3 percent sur-charge for 1994.

15The exceptions are Estonia, Latvia, and Lithuania.

83


V REVENU E DECLIN E

lowing manufacturers to claim VAT credits for mate-rials at the time they are used in production, even ifthey have not paid their suppliers, is particularlycostly in revenue terms when there is an accumula-tion of interenterprise arrears. Since suppliers are notliable until they are paid, VAT credits for materialsare being taken or refunded by the government be-fore VAT liability has arisen on purchases of inputs.Moreover, since tax becomes due when arrears arepaid, the incentive for enterprises to collect receiv-ables is correspondingly diminished. The practice ofallowing VAT credit only when inputs enter into pro-duction is also a deterrent to collection because it re-quires a complex accounting system to monitor theuse of inventories.16

(2) Markup method versus invoice method. Thenonmanufacturing sector does not use the usual in-voice method to determine VAT liabilities, but rathera markup system whereby VAT is paid not on the ac-tual price but on an estimated fixed markup overcost. Accounting methods for calculating taxablemarkup (especially the treatment of inventories) aresubject to abuse, and their complexity makes auditdifficult. There are also adverse consequences wheninflation is high (see below).

(3) VAT on international trade. Turnover tax wasnot levied on imports or exports. Since the introduc-tion of the VAT, trade among members of the Com-monwealth of Independent States (CIS)—whichconsists of all the countries in the region other thanthe three Baltic states—has been taxed essentially onthe origin principle, that is, VAT is collected in theproducing country. Initially, trade with other coun-tries was not taxed, but provision was soon made totax this trade on the destination principle, that is,with imports taxed and exports zero rated.17 Net im-porters—from outside the CIS—should have bene-fited from this provision. However, there are numer-ous exemptions for imports, and the administrationof the destination principle has proved complex.Consequently, trade is taxed haphazardly and yieldsless VAT revenue than it should.

The introduction of excises has also been disap-pointing. In contrast to market economies, exciseshave not been a stable high-yield revenue source.Most notable is the absence of important commodi-ties, such as gasoline for cars, car imports, and sometypes of alcohol, from the excise base in most coun-tries. Also, tax rates reflect pressure from producer

lobbies seeking to keep the prices of excisable goodslow. At present, rates on alcohol and cigarettes areabout 20 percentage points lower than in countriesof the Organization for Economic Cooperation andDevelopment (OECD).

Trade Liberalization

In the former U.S.S.R., revenue from foreign trademainly came from net export profits.18 These have ei-ther disappeared or diminished sharply with the liber-alization of trade.19 It was envisaged that they wouldbe more than made up for by the introduction of ex-plicit import tariffs and by temporary taxes on wind-fall gains on exports as the ruble was being depreci-ated in the move to convertibility. However, thoughrevenue from trade taxes has been growing rapidly, itstarted from a very low base and generally accountsfor less than 5 percent of total revenue. Moreover, itspotential is limited by a variety of factors.

First, the Baltics, Russia, and other countries of theformer Soviet Union have been reluctant to imposesignificant tariffs on imports from outside the CIS be-cause they see foreign competition as a central ele-ment in domestic demonopolization. Second, mostcountries do not levy import duties on imports fromCIS countries or on personal property imported byCIS citizens. In some cases, these provisions have ledto imports being almost fully relieved of duty.20 Third,it has also been necessary to introduce a customs ser-vice from scratch and to patrol borders that are manythousands of miles longer than those prior to 1991.Apart from the intrinsic teething problems of a newadministration, financial problems have also hinderedthe introduction of customs. Finally, to the extent thatcountries were willing to tax trade, they often im-posed export taxes rather than import tariffs to maxi-mize the domestic availability of goods.

Finally, in net energy-exporting countries—no-tably Russia and Turkmenistan—it had been ex-pected that the windfall gains from temporary exporttaxes would provide a cushion for the budget. Thesepotential gains have, however, been undermined. InRussia, pressure for tax relief came not only fromthe oil industry but also from political tensions be-tween the Moscow government and oil-rich regions

16See Victoria P. Summers and Emil M. Sunley, "An Analysisof Value-Added Taxes in Russia and Other Countries of the For-mer Soviet Union," IMF Working Paper, WP/95/1 (Washington:International Monetary Fund, January 1995).

17However, a number of central Asian countries (e.g., the Kyr-gyz Republic and Tajikistan) did not have legal provisions forlevying VAT on imports until the end of 1994.

18The state budget of the U.S.S.R. used to fully absorb thegains or fully cover the losses of state foreign trade organizationsthat resulted from the insulation of domestic prices from those inthe world market.

19Budget revenue from profits on foreign trade in Russia, forexample, did not disappear but fell significantly. A centralized ex-port scheme continued to be in place into 1994.

20In the Kyrgyz Republic, the origin principle was interpretedto encompass any import arriving via another CIS country and thedeclaration of nonpersonal property was solely at the discretionof the importer.

84


Factors Explaining Revenue Development s

that threatened to secede if they were taxed tooheavily. In Turkmenistan, the inability of tradingpartners to pay for their gas imports caused a mas-sive decline in revenue from the gas sector.

InflationBesides eroding real tax bases, the failure of most

countries to stabilize during 1992-93 left tax systemsvulnerable to distortions caused by high inflation.Average inflation in the region exceeded 1,000 per-cent during 1991-93, though it came down substan-tially in 1994 (see Section III). The net impact ofhigh inflation on revenues, however, is uncertain.

On the one hand, the absence of proper inflation ac-counting has led enterprises to book raw materials andinputs at historical costs and some to report abnor-mally high profits. Similarly, depreciation allowanceswere not indexed—at least until 1993—and the enter-prise tax base expanded on account of decapitalizingtaxes on depreciation. Because, in the Soviet system,replenishing the depreciation fund traditionally pre-empted around 25 percent of total enterprise income,the short-run effect of high inflation could have beento increase enterprise tax receipts by as much as onefourth. A certain amount of bracket creep also oc-curred, particularly in 1992, although the revenuegain from this was probably minimal.

On the other hand, high inflation depresses rev-enue in real terms because of collection lags and themarkup method of computing VAT liabilities. De-spite collection lags that tended to be relatively shortby international standards, reflecting the automaticcollection of taxes from enterprises through the statebanking sector, near hyperinflation in some countriesand an inadequate payments system probably led tosignificant losses in real revenue. A second seriouscost to the budget arises from the use of the markupmethod to compute VAT liabilities. When large priceincreases occur between the time of purchase and thetime of sale, the application of a standard fixed VATmarkup on the historic purchase price implies thatmuch value added may go untaxed.

Exemptions

In most countries, revenue has also been erodedby numerous exemptions from all major taxes.21

Much new activity benefits from exemptions, andthe range of exemptions has increased as govern-ments have tried to earn the support of powerfulpressure groups. The main groups lobbying for ex-emptions are the following.

21In Ukraine, for instance, the government estimates that theremoval of VAT exemptions would raise VAT yield by more thanone third.

(1) Sectoral lobbies. The influence of sectorallobbies is manifest in exemptions and preferencesthat benefit a whole sector (most notably agricul-ture), specific subsectors, or individual large enter-prises. In Moldova, for example, the agriculturalsector—which has the largest representation in Par-liament—benefited from exemptions on the enter-prise tax that amounted to an estimated 1 percent ofGDP in 1993. And in the Kyrgyz Republic, the prof-itable gold mining sector was exempt from the enter-prise tax and VAT until 1994.

(2) Regions. Pressure from regional elites has led to"special rights" being accorded to individual regionsby presidential decrees, government resolutions, andparliamentary decisions. This has been a serious prob-lem in Russia, which has experienced a proliferationof free ports and environmentally distressed emer-gency zones. Territories embracing about a quarter ofits area and a third of its productive capacity werefunctioning under highly concessional special regimesin the middle of 1993. They received, and continue toreceive, exemptions from taxes and duties and fromthe obligatory conversion of hard currency export pro-ceeds, as well as various other fiscal preferences.

(3) Foreign investors. Tax preferences for foreigninvestors abound, especially for enterprise tax andcustoms duties. For example, in Azerbaijan, joint ven-tures with at least 30 percent foreign participation aresubject to a reduced profit tax rate of 25 percent; athree-year tax holiday is granted to foreign enterprisescommencing activity in the country, which may alsobe excluded from VAT and customs duties.

Efficiency of Tax Administratio nTaken together, the collapse of traditional tax

bases and shortcomings in the design of existingtaxes go a long way toward explaining the revenuedecline. However, it has been compounded by seri-ous weaknesses in tax administration. Under stateownership and central planning, tax collection wasmainly a matter of state enterprises transferringturnover taxes and specified proportions of theirprofits to the budget. The tax service did not have toengage in any assessment, collection, or enforce-ment of taxes. Tax payments were determined by theplan and were effected through automatic transfersbetween accounts held with the state banking sector.The role of the Soviet tax administration was thuslimited to ensuring that stipulated accounting con-ventions were followed and to verifying the arith-metical correctness of the bank transfers. Being di-rected only toward these objectives, such a systemcame under immediate pressure as economic activitymoved significantly outside the state enterprise sec-tor. Moreover, tax compliance by state enterprisesweakened, and they have accumulated sizable tax

85


V REVENU E DECLIN E

arrears. While the elements of a modern tax adminis-tration are beginning to be put in place, shortcom-ings in basic administrative functions remain.

Taxpayer Identification

In most countries, there is no system to identify tax-payers by a single taxpayer identification number(TIN). Without such a system, the tax administrationhas no common identifier that ties together all of a tax-payer's records. Hence, it is impossible to correlatedata with a view to detecting underreporting of incomeor other forms of evasion. Further, only a few coun-tries have implemented an effective system of tax-payer registration.

Filing and Payment of Taxes

Administrative practices are fraught with duplica-tion and inefficiency. For example, in most countriestaxpayers separately submit tax returns to the localtax office and payment orders to the bank. The needto match tax returns and payment orders complicatesthe task of detecting taxpayers who fail to submit re-turns and to make timely or full tax payments. Thismatching process is particularly slow because of thelack of a TIN. In addition, too much of tax officials'time is devoted to manually checking tax returns,rather than encouraging taxpayer compliance andcontrolling the collection process.

Organization

In most countries, a single tax officer is usuallyassigned to handle most administrative functions fora specific group of taxpayers. Consequently, tax offi-cers see the same taxpayers repeatedly and are re-sponsible for several taxes. Regular direct interac-tion between the same tax officer and taxpayer raisesthe potential for collusion and consequent tax eva-sion. Tax officers' responsibility for a wide range ofadministrative functions also limits their knowledgein specific technical areas, especially accounting andaudit techniques. In particular, they lack the inves-tigative skills to conduct effective audits.

Audit

Audits are often conducted in a perfunctory manner,without any analysis or investigation of enterprises'business activities. An important reason for this, in ad-dition to the lack of specialized audit skills among taxofficers, is the requirement in most countries that allenterprises be audited every one or two years. The de-velopment of the private sector is making this imprac-tical and has led tax officers to often take an exces-sively formalistic approach to auditing taxpayers.Audits frequently amount to no more than determining

whether the taxpayer has the correct records with theappropriate signatures and seals. Tax audits are alsohindered by the requirement that they cover all taxes indepth. This made sense under central planning, whenthe major taxes were all tied to a single accountingsystem, but it now precludes effective audit.

Taxpayer Education

Across the region, taxpayer awareness remains low,and budgets for taxpayer education are almost nonex-istent. In many countries, tax returns and other keyforms necessary for compliance are not provided totaxpayers. Even when forms are available, there are noinstructions on the form or separate document to ex-plain how to complete it.

Customs

Most countries do not have adequate customs con-trols, either for assessing and imposing tariffs or forapplying VAT and excise taxes to imports. To copewith this problem, some countries (such as Lithuaniaand Moldova) attempt to bypass the customs admin-istration and have their internal revenue service col-lect domestic indirect taxes on imports. However,both because it is more difficult to follow up on im-porters once imports have passed across the borderand because procedures for information sharing be-tween customs and the internal revenue services arepoorly defined, collection has been weak. Further-more, as already mentioned, the application of thedestination principle for VAT on international tradehas posed difficulties for customs administration.

Crisis i n Governanc eThe legitimacy of government is also a factor in

tax collection. In an unsettled political environment,many of the new governments remain weak. Theirfinancial powers are especially limited by an inabil-ity to control local governments and a volatile policyenvironment.

Fiscal Federalism

In some countries in the region, uneasy relationsbetween various levels of government appear tohave been a significant factor affecting overall rev-enue performance. In Armenia, Georgia, Moldova,Russia, and Ukraine, collection problems havearisen from disputes over revenue sharing and taxassignment provisions among the central, regional,and local governments. In an ideal situation, fiscalfederalism should not influence the revenue positionof the general government after consolidation, butseveral factors have tended to make the overall out-come weaker than expected.

86


Policy Implications and Response s

First, legislative confusion and conflicting tax reg-ulations at the central and regional levels in Russia,connected with the struggle of constituent republicsfor greater economic independence, gave enterprisesan opportunity to obey "softer" tax laws or to delaypayments without explicitly breaking the law.

Second, revenue underreporting by lower levels ofgovernment appears to be widespread and significantacross the region. For instance, in 1992-93 com-modity-producing, resource-rich regions in Russia—especially those that had been net interregionaldonors—tended not to transfer revenue to the centralbudget. While some territories withheld revenue in-formally, several autonomous republics (Tatarstan,Chechnya, and Bashkortostan) formally declaredtheir unilateral introduction of a "one-channel taxa-tion system." This meant that all their taxes were tobe collected by the regional authorities, and revenuesharing with the central budget was to be negotiatedafterward. Under these circumstances, the regionalauthorities had little incentive to report their revenuecollections accurately to the central government.

And third, tax collections at the local level may notbe as effective as with a unified tax administration.When Ukraine's central government tried to transferrevenue responsibilities to local governments in 1994,overall collection dropped. Officials suggested thatthe local governments were more vulnerable to pres-sure from enterprises than was the central tax service,particularly when an enterprise was a large local em-ployer and provider of social services. In Belarus, en-terprise tax collections were reported to be chaotic in1992, because local authorities could not handle theirrevenue collection responsibility; now all collectionhas been unified under the State Tax Administration.

Policy Environment

The abandonment of the Soviet regime was associ-ated in most countries with political instability, fre-quent changes of ministers, and ambiguous policycourses. This had several implications for tax collec-tion. First, the lack of a coherent reform strategy wasreflected in frequent amendments to tax legislation,with negative repercussions for taxpayer complianceas taxpayers either delayed payments in the hope of areversal of disadvantageous changes or were simplyunfamiliar with new requirements. More generally, thelow quality of tax legislation has likely been a con-tributing factor to the revenue decline. With the excep-tion of the Baltic countries and Russia, tax legislationtends to be badly drafted, is difficult to understand, andcontains inconsistent and poorly thought out rules.Second, some governments, in part because of theirimpermanence and in part because of falling bud-getary wages, have had a low degree of legitimacywith their civil service. Such disaffection has con-

tributed to corruption and the misappropriation of rev-enue. In Azerbaijan in 1993, for instance, "bonus de-velopment payments" by foreign oil companies for ex-ploration leases were never received by the budget butwere diverted to the State Oil Company (SOCAR).Third, the heavy reliance on sequestration as an instru-ment of expenditure control has led ministries, localgovernments, and enterprises to protect themselvesagainst unexpected expenditure cutbacks by withhold-ing revenue from the central authorities.

Policy Implication s an d Response sThe revenue decline experienced by the Baltic

countries, Russia, and other countries of the formerSoviet Union since 1991 is a complex phenomenon.It is clearly related to the collapse in output acrossthe region. This has affected established tax bases—that is, profits, wages, and retail sales in the tradi-tional sectors of the economy. In many countries,these bases have fallen faster than output. Conse-quently, falling real revenue has been accompaniedby a declining revenue-to-GDP ratio. However, thepoor revenue performance cannot be blamed on theoutput collapse alone. It also reflects problems in thedesign and administration of taxes, and other factors(related to the crisis in governance).

The major tax bases traditionally derive from theactivities of state enterprises. While the enterprisesector has declined sharply, agricultural output hasbeen fairly robust, and with price reform, the impor-tance of agricultural incomes has increased. However,such incomes are in most instances exempt from taxa-tion. A private sector is also emerging, both throughthe start-up of new activity and the privatization ofexisting activity. Yet the private sector is not effec-tively taxed. It benefits from the pervasive exemp-tions that characterize the major taxes. A significantpart of it also remains outside the reach of an out-moded tax administration. More rapid progress withtax reform might have arrested part of the revenue de-cline. It is necessary for its reversal, and urgently so ifa further revenue decline is in prospect.

An output recovery will generate more revenue.But the heavily taxed sectors of the economy thathave declined so sharply will not be the ones to leada recovery. If economic transformation in the Balticcountries, Russia, and other countries of the formerSoviet Union is a success, the obsolete part of thestate enterprise sector will be replaced by a vibrant,but difficult-to-tax, private sector.22 To reverse the

22This observation is made in the context of economic transfor-mation more generally by Vito Tanzi, "Reforming Public Fi-nances in Economies in Transition," International Tax and PublicFinance, Vol. 1 (Boston: Kluwer Academic Publishers, 1994).

87


V REVENU E DECLIN E

revenue decline, and especially to restore revenue-to-GDP ratios to where priority spending can be un-dertaken without compromising macroeconomic sta-bilization objectives, tax systems must be reorientedto encompass emerging activity. Existing activity,and especially agriculture, also must be taxed moreeffectively.

The rest of this section summarizes the prioritiesfor the reform of tax policy and tax administration.These priorities reflect the advice given by IMF tech-nical assistance missions and experts to the countriesof the region.23 In the area of tax policy, many of thereforms can be implemented immediately, with thepromise of significant short-term revenue gains. In-deed, most could have been in place already and con-tributed to a moderation of the revenue decline. How-ever, for a variety of reasons—mainly related tofailure to secure the necessary political consensus—much of the IMF's advice has often not been fol-lowed. The most notable headway in the region as awhole has been the simplification and rationalizationof customs duties and some broadening of the base ofVAT and excises. Across the range of taxes, there hasbeen progress in Estonia and Kazakstan. In the area oftax administration, advice has been more influential,especially in connection with major projects in Russiaand Ukraine. But there remains much to be done.Moreover, as with all institution building, putting inplace a modern tax administration will yield resultsonly over the medium term.

Immediate Challenge sThe most reliable short-term benefits from tax

policy reform will come from improving the effec-tiveness of taxes on the traditional branches of theeconomy that are already theoretically in the taxnet. These include, but are not confined to, sectorsdominated by state enterprises. Since a primarymedium-term reform goal is to promote growth,higher tax yields from already taxed sectors must beachieved by broadening bases and blocking loop-holes—mainly through the removal of exemp-tions—rather than by raising tax rates. Indeed, ra-tionalizing tax rates and reducing rate-relateddistortions may improve the legitimacy and effec-tiveness of the tax system and, by minimizing disin-centives to work, save, and invest, improve growthprospects.

23Alternative sources of technical assistance include The WorldBank, Organization for Economic Cooperation and Development,European Bank for Reconstruction and Development, the Euro-pean Union, the U.S. Treasury, United States Agency for Interna-tional Development, bilateral agencies, and private consultants.They provide advice across the region, both on tax policy and taxadministration.

Removal of Exemptions

Exemptions can usually be removed, or at leastsubstantially reduced, immediately. While it maytake previously exempt activities some time to ad-just to the imposition of taxes, the response beingasked of them is no more painful than for other sec-tors exposed suddenly to price liberalization. Experi-ence in the region since 1991 suggests that little isgained by delaying price shocks.

There are numerous exemptions that should be re-moved or restricted. As regards enterprise taxes, elim-inating pervasive exemptions by source of revenueand type of activity, apart from raising revenue, willcreate a level playing field for production. State andcollective farms, small businesses, manufacturing forexport, and enterprises with foreign investment unjus-tifiably benefit from such exemptions. Given the im-portance of the emerging private sector, there is an es-pecially strong case for eliminating tax holidays fornewly established businesses and replacing them bygenerous carry-forward of start-up losses, or acceler-ated depreciation of capital assets, or both.

With regard to taxes on wages and salaries, thedefinition of taxable income under the personal in-come tax should be broadened to cover all forms ofincome, including social security benefits and pay-ments in kind (most notably free housing). Also, taxconcessions for families with children and other de-pendents should be replaced by family allowancespaid through the social security system, to lowerbudgetary costs through improved targeting.

VAT exemptions beyond those justified on socialor administrative grounds (education, health, finan-cial services, and so forth) should be eliminated.This would involve extending VAT to most services.In the countries of the CIS, VAT should be applied toall imports from outside the CIS at the same rate asdomestic goods. It should also be levied on productprices inclusive of other taxes. Specifically, the VATbase should include any applicable excises. For im-ports, VAT should be based on their landed value,plus customs tariff and excise duty, converted at themarket exchange rate. Zero-rating should be re-stricted to exports.

Rationalization of Tax RatesHigh tax rates are an obvious disincentive to eco-

nomic revival, while a multiplicity of rates not onlyresults in distortions and inequities but also compli-cates tax administration and creates opportunities fortax evasion. The immediate harmonization of ratesmight produce some revenue gains from improve-ments in efficiency. But, in all likelihood, revenueneeds will not permit an immediate reduction in taxrates, which will instead have to be gradually re-duced as revenue collections improve.

88


Policy Implication s an d Response s

For multiple-rate enterprise taxes, rates should beconsolidated into a single standard rate, whichwould be aligned with the top personal income taxrate. The rate structure of taxes on wages andsalaries should be made less distortionary by rais-ing the threshold and reducing top marginal rates,while increasing the basic rate, which tends to below by international standards. VAT should alsohave a single standard rate. Excises should apply atthe same rate regardless of origin. In particular, thefailure to impose excises on imports discriminatesagainst all domestic production. To the extent pos-sible, the rates of excises and other duties (includ-ing import tariffs) should be converted to an ad val-orem basis.

Disguised High Tax Rates

Some tax rates in the region are higher than theyseem. In countries where wages are included in theenterprise tax base, the effective rate on profits ismuch higher than the statutory rate. Likewise, depre-ciation allowances based on historic cost boost theeffective tax rate on true profits in countries whereinflation is high. For instance, in Ukraine in 1994,the enterprise tax rate was 22 percent; adjusting forthe wage component in the base, the effective profittax was 36 percent. To remove the hidden disincen-tives, tax bases should be rationalized as quickly aspossible. Any such adjustments can be made rev-enue-neutral by a corresponding adjustment to thetax rate. Most crucially, costs should not be taxed asprofits. Also, the taxation of inflationary gainsshould be eliminated, because it causes the decapi-talization of enterprises, thus making them morelikely to fail.

Taxing the Emerging Private Sector

It is hoped that a sizable share of the emerging pri-vate sector will be taxed as a consequence of eliminat-ing exemptions. Some activities may remain elusive,however, and will have to be taxed by alternativemeans.

One of the main challenges faced by the region isto tax private business, much of which is concen-trated in small-scale activity. Even when new tax-payers have been identified, operational means oftaxing them have to be found. For this to be possiblein the short term, incorporating some presumptiveelements of taxation into existing income taxesneeds to be considered, as a way of short-circuitingdifficulties in assessing tax liability. For instance, aminimum levy or low flat-rate tax based on pre-sumed turnover of individual proprietorships couldbe introduced, in combination with a basic license

fee.24 A presumptive tax could also be levied in lieuof VAT on small business. The advantage of pre-sumptive taxes is that they broaden the tax base atrelatively low administrative cost, encourage tax-payers to move from the informal to the formal sec-tor, and are fairly efficient (because marginal taxrates are zero) and equitable (because they focus onthose who would not otherwise pay tax).

Taxing Agriculture

Agriculture is the largest untaxed sector in manycountries of the region and has declined less thanother sectors because it has benefited significantlyfrom the liberalization of prices of basic foodstuffs.For this reason, it is particularly important that agri-culture should be brought into the tax net. The re-moval of agricultural exemptions from the majortaxes is the most desirable way of taxing agriculture.When such an approach is not adequate, considera-tion should be given to finding alternative means oftaxing agriculture, at least in the short run. For in-stance, payment of a land tax (which already existsin some countries) could be made a secondary proofof ownership or a prerequisite for the right to useland. For the immediate future, the land tax can beseen as a means of bringing farmers into the tax net,and difficulties in land valuation can be dealt withby a certain tolerance for underassessment.25 A landtax can also help reach new small private farmers.

Of particular concern to farmers is an appropriatetreatment of losses and carryovers, since their in-comes fluctuate from year to year. Hence, if agricul-tural profits and earnings are to be subject to tax, itmay be important—not least for political acceptabil-ity—to provide for income averaging (if tax rates areprogressive) and loss carryover.26

Taxing Trade

A further likely source of short-term revenue isthrough enhanced attention to the taxation of trade.Significant revenue can be raised in the CIS countriesby applying of a broad-based simple tariff with a stan-dard uniform rate of 10-15 percent on all ex-CIS im-ports. As some countries have already found, an

24There are various ways of determining presumed turnover,such as the number of employees, the area of the businesspremises, and the consumption of electricity, which to apply hasto be assessed case by case.

25In the long run, cadastres should be improved, and land taxrates should be adjusted automatically each year in line with pricechanges.

26Income averaging, however, might be a complexity that foradministrative reasons should not be considered, in which casetax rates should be as uniform as possible.

89


V REVENU E DECLIN E

across-the-board tariff is relatively simple to adminis-ter and minimizes distortions. Some countries couldimprove their revenue yield from trade taxes if im-ports were valued for tax purposes at a market ex-change rate rather than at an official rate. Indeed, theimplied loss of revenue is one reason for unifying of-ficial and market exchange rates. Export taxation—including implicit taxes imposed through hard cur-rency surrender requirements—should be avoided.

Other Quick-Yield Taxes

Where enterprises prove difficult to tax in theshort run, there may be some justification for apply-ing an enterprise property tax of 1-3 percent on allassets of state enterprises other than financial assetsand land (net of depreciation). This charge, like anobligatory dividend, is designed to compensate gov-ernment for its investment, to prepare enterprises forprivatization, and, most important, to act as a pre-sumptive tax prior to the introduction of a satisfac-tory bookkeeping system.27

Petroleum and other energy excises are an impor-tant source of revenue in many countries and can bequickly developed. Downstream taxes are particularlylow in the Baltic countries, Russia, and other coun-tries of the former Soviet Union. The Russian gaso-line excise rate, for example, is 15 1/2 percent com-pared with tax-inclusive excise rates on gasoline ofbetween 25 percent and 71 percent in major countriesof the Organization for Economic Cooperation andDevelopment (OECD). Also, many countries of theregion are major producers of petroleum, gas, andminerals but earn little revenue from these sectorscompared with international norms. More attention totaxing these products could well generate sizable rev-enue, as well as encourage energy conservation. Insome cases, added revenue can be earned from jointventure arrangements, including up-front paymentson the signing of agreements (as in Azerbaijan).

Short- to Medium-Term Prioritie sThe policy measures described above can, to a

large extent, be introduced rapidly through changes inlegislation. However, their potential revenue impactwill not be fully realized unless they are supported bymodern accounting practices and the tax administra-tion is able to collect revenue from activities broughtinto the tax net by the removal of exemptions, the in-troduction of new taxes, and other reforms.

Accounting Practices

Inappropriate accounting practices have alreadyproved to be an obstacle to increasing revenue insome countries. Their reform is a necessary comple-ment to tax reform.

(1) Invoice-method VAT accounting. The replace-ment of the markup method for calculating VAT lia-bility with the invoice method would ensure that VATwas paid on the full price of goods sold and preventany erosion of VAT returns on account of inflation.28

A shift to the invoice method would be more feasibleif it were accompanied by simplified VAT account-ing. In particular, enterprises should be permitted toclaim credits in the period they have incurred VAT ontheir inputs, and not only when these inputs are usedin production. Symmetry of treatment for inputs andoutputs should be restored for taxpayers using eitherthe cash or accrual basis of accounting for taxes.Specifically, taxpayers who use the cash basis shouldnot take VAT credit for goods or services purchaseduntil they are fully paid for.

(2) Accrual accounting. The use of cash account-ing rather than accrual accounting has been costlyfor the budget. As discussed earlier, if tax falls due atthe time payment takes place rather than at the timeit accrues, an enterprise's incentive to liquidate re-ceivables is reduced. A shift to accrual accountingwould thus reduce the incentive to incur interenter-prise arrears. However, it may lead to greater tax ar-rears, since cash-strapped enterprises would have topay accrued VAT on sales to late or nonpaying cus-tomers. This is probably an unavoidable conse-quence of tightening the budget constraint on enter-prises. A shift to accrual accounting would alsoreduce (though not eliminate) incentives for ineffi-cient barter.

Tax Administration

An unmeasurable but significant reason for therevenue decline has been the diminishing effective-ness of planned economy tax collection methods asthe transition has progressed. Existing tax adminis-tration systems and procedures need to be over-hauled and many new ones introduced if the tax ad-ministration is to handle a larger and more diversetaxpaying population. These reforms need to beginimmediately, or to be accelerated where they havealready started, in tandem with the policy changesdescribed above.

27It should be noted, however, that an enterprise property tax—like a land tax—will need to deal tolerantly with valuation diffi-culties because the fair market value of most property will not yetbe well established.

28In some countries, manufacturers use the invoice system,while the liability of the nonmanufacturing sector is calculated bymarkup. The shift to an invoice system would in this case also re-move a sectoral distortion, namely, that nonmanufacturers cannotclaim VAT credit for goods purchased for resale.

90


Policy Implications and Response s

(1) Registration. For emerging sectors to be cap-tured, new taxpayers will have to be identified andregistered. Local tax offices, with assistance and su-pervision from the central tax administration, shoulddevote considerable resources to ensuring completeregistration of all firms and individuals covered by thetax law. Moreover, upon registration, taxpayersshould be given a unique TIN. The adoption of a na-tionwide TIN will vastly facilitate the cross-checkingof information regarding taxpayer compliance (e.g.,income reported by third parties, such as governmentinstitutions against income reported by taxpayers).

(2) Returns (and other documents). More gener-ally, moving away from a tax system based on cen-tral planning accounting techniques will require theoverhaul of return forms and other tax documents sothat they reflect newly appropriate accounting pro-cedures, can be used by more diverse types of tax-payers, are easy for the taxpayer to understand, andare quick to process. Shortcomings in the design ofdocumentation are considered to have been particu-larly damaging to the productivity of the VAT. A sin-gle standard document should be adopted for thesupplier and purchaser to record a sale and its asso-ciated VAT liability.

(3) Taxpayer education. In a system with risingnumbers of taxpayers, there are likely to be impor-tant gains to revenue collection if taxpayers canpay their liabilities without involving tax adminis-trators. In the first instance, taxpayers should begiven clear and concise information describingwhat is taxable, when taxes apply, how to calculatetheir tax liabilities, and exactly what they are re-quired to do to pay their taxes. In the long run, thegoal will be for tax collection to be based largelyon self assessment.

(4) Organization. The tax administrations of theBaltic countries, Russia, and other countries of theformer Soviet Union should organize their opera-tions according to basic functions—registration, tax-payer information, returns processing, collection en-forcement, and audit. Assigning specific taxpayersto tax officers is no longer appropriate and createsundesirable opportunities for evasion and corrup-tion. Greater specialization and expertise among taxofficials should be achieved by assigning staff to de-fined and limited functions.

A large-taxpayers unit should be created.29 Itwould monitor collection of taxes from importanttaxpayers who, although not numerous, account forthe major part of tax revenue.30 In energy-export-

ing economies, a few large taxpayers can accountfor 90 percent of total revenue. In some countries,large taxpayers engaged in cash cropping and pro-cessing account for a substantial part of revenues.A large-taxpayers unit properly focuses initial ef-forts of tax administration reform on the adoptionof more efficient procedures for the more importanttaxpayers, rather than on the increasing number ofsmall taxpayers whose revenue contribution is notsubstantial.

(5) Audit. Taxpayers must believe that if they fail tocomply with the tax law they will be caught and penal-ized. They also need to be assured that their competi-tors are paying the same taxes. Hence, an effectiveaudit program is an integral part of tax collection andrequires the setting up of an audit unit. To enhancecompliance, the tax administration should not make itsaudit strategy public, nor should it disclose the numberof taxpayers to be audited during a given period. In-stead, it should promote the view that a variety of fac-tors may lead to an audit and the belief that the proba-bility of an audit is higher than it really is.

(6) Penalties. Adequate but not excessive penaltiesmust be imposed, not only for tax evasion but also forfailure to file returns, issue and maintain appropriatetax documents, and pay full tax on time. Interest pro-visions for late tax payment should compensate thegovernment for the time that the taxpayer has use ofthe government's revenue. To ensure that penalties re-main realistic under conditions of high inflation, theyshould to be periodically revised.

(7) Arrears. A specific strategy for collecting out-standing tax arrears is needed. Each local tax officeshould carry out an analysis of its arrears, identify-ing them by debtor, size, time outstanding, and typeof tax. This will allow administrators to set prioritiesfor pursuing the largest and most recent debts and toeliminate those that are considered uncollectible orinsignificant.

(8) Customs. For trade to be taxed effectively,countries need to strengthen their customs adminis-tration and should allocate realistic budgetary fundsto do so. Besides valuing and taxing imports, cus-toms administrations should develop and implementreliable procedures for validating exports of goodsfor zero-rating under the VAT, and should installsafeguards against the reimporting or "round-trip-ping" of goods to and from the region. Eventually, tominimize leakage of revenue, customs administra-tions will have to assume responsibility for collect-ing VAT and excise taxes on imports.

29This could be done quite quickly.30If a large-taxpayers unit is created quickly, and the economic

structure is changing rapidly, the possibility that tomorrow's andtoday's large taxpayers need not be the same should be taken intoaccount.

Establishing Good Governance

Even the best-intended reform efforts can be un-done by volatile legislation, which deters investment

91


V REVENU E DECLIN E

and encourages short-term speculative activity at theexpense of production. Therefore, governmentsshould work toward putting in place a comprehen-sive permanent tax code.31 From then on, theyshould minimize piecemeal revisions to the code.More generally, successful tax reform will require an

31A tax code combines all the general rules for taxation in oneplace, thus avoiding the inconsistencies typical of separate lawsand decrees.

end to arbitrary governance. Two aspects of this areespecially important. First, tax assignments to otherlevels of government, and revenue-sharing arrange-ments, should be objective and transparent and notsubject to bilateral negotiations. Second, govern-ments should endeavor to draw up realistic budgetsthat would provide government suppliers and otheragents with financial blueprints of at least one year'sduration. They would also eliminate the need for ex-penditure sequestration, and the precautionary with-holding of tax payments.

92


VI Stabilization : Fixed Versus FlexibleExchange Rates

Jeromin Zettermeyer and Daniel A. Citrin

T his section discusses issues regarding the choiceof exchange rate regime in achieving price stabil-

ity in the Baltics, Russia, and other states of the formerSoviet Union: in particular, the relative merits of ex-change rate based stabilization versus the targeting ofmoney (or credit) growth in conjunction with a degreeof exchange rate flexibility are examined.1 The mainarguments for and against the alternative exchangerate arrangements in the context of stabilizing fromhigh or moderate levels of inflation are reviewed first.Then, the recent policy strategy in the region in light ofthese arguments is discussed using two benchmarkcases of fixed and flexible exchange rates. Through-out, the objective of the section is limited to discussingalternative nominal anchors as tools for achieving asubstantial reduction in inflation in the region. No at-tempt is made to discuss the relative merits of differentexchange rate regimes as permanent policy choices,that is, beyond the disinflation phase.

General Consideration sand Experience

In reviewing the general arguments affecting thechoice of exchange rate strategy for stabilization,one may focus on four key considerations: (1) thecosts of stabilization while the program is in place;(2) the effectiveness of the policy approach in bring-ing down inflation; (3) the costs resulting from pro-gram failure; and (4) the chances of failure and theconditions that would minimize the risk of failure.

Relative CostsIn assessing the relative costs of disinflation of

money versus exchange rate based programs with

1IMF-supported programs involving flexible exchange rates inthe Baltics, Russia and other countries of the former Soviet Unionhave typically entailed binding credit targets established on thebasis of indicative monetary targets or projections. While, strictlyspeaking, such programs did not entail a nominal money anchor,with credit expansion the primary factor determining moneygrowth in these countries we follow the literature (e.g., Sahay andVegh (1995)) in referring to such programs as "money based."

comparable stabilization objectives, two distinct con-siderations apply. First, the process of disinflation it-self will usually have consequences for output, al-though probably less so in transition economies than inmore developed market systems (see Section II). Stan-dard arguments suggest that the output costs of a pre-announced disinflation will depend on its credibility.2

Thus, the question is which type of nominal anchor ismore likely to be viewed as sustainable—an issue tobe taken up separately below. Second, since stabiliza-tion from high levels of inflation in transitioneconomies is not instantaneous (see Table 1.7), thetotal output cost of each approach will also depend onthe nature of economic shocks, other than the disinfla-tionary shock itself, that affect the economy during thestabilization period.3 Money demand shocks will havesmaller consequences for output under a fixed ex-change rate regime, because they can be absorbed byendogenous responses in the money supply. In con-trast, under flexible exchange rates such shocks willlead to swings in domestic interest rates and the ex-change rate (or, in the absence of a well-developedcapital market, rationing, and cash shortages) and con-sequently fluctuations in output and velocity. On theother hand, unless goods and labor markets were per-fectly flexible, real shocks affecting the demand forgoods or the terms of trade would tend to be amplifiedunder a fixed exchange rate regime, because the ex-change rate and interest rates are not available as ad-justment mechanisms.4

Unfortunately, during the initial stabilizationphase, both types of shocks are likely to be present

2For example, Sargent (1982). See also Calvo (1989) andChadha, Masson and Meredith (1992).

3See Fischer (1986), pp. 257-59 for a discussion of the effectsof shocks under alternative regimes.

4These arguments can be appreciated most simply in an openeconomy IS-LM framework, where a credible fixed exchangerate regime amounts to fixing the interest rate to world levels, im-plying a complete isolation of the real sector of the economy frommoney demand shocks, but at the same time eliminating interestrate adjustment to goods demand shocks. Fixing money has theopposite effect, because all short-run adjustment to money de-mand shifts takes place through the interest rate and exchangerate and is thus passed on to the real sector, while goods demandshocks are mitigated by interest rate adjustment.

93


VI STABILIZATION : FIXED VERSUS FLEXIBLE EXCHANGE RATE S

in the economies in the region. On the one hand, asdiscussed in Section III, a number of factors havecontributed to substantial volatility in money de-mand over the past several years. On the other hand,the initial stabilization period has coincided with ex-ternal opening and domestic and other structuralchanges that have led to substantial real and externalshocks.5 In practice, it is difficult to argue a prioriwhich type of shock, in addition to the impact of dis-inflation itself, is more likely to impose significantoutput costs during stabilization. However, it islikely that relative to the huge output declines in theBaltics, Russia, and the remaining countries of for-mer Soviet Union that are an immediate conse-quence of structural change and political develop-ments, the added output losses from choosing themore "costly" stabilization strategy will be minor.This suggests that the relative costs of stabilizationshould be a less important consideration in guidingthe choice of nominal anchor than the effectivenessof the anchor in reducing inflation quickly and thechances and consequences of failure.6

EffectivenessThere are reasons for believing that exchange rate

anchors may be more effective in controling infla-tion than monetary anchors in the transitioneconomies.7 First, the effectiveness of a disinflation-ary program will depend on how tightly the interme-diate target — that is, either the exchange rate ormoney—is linked to the ultimate target, namely, theprice level. If the situation of transition economies isone of shifts in money demand and unstable veloc-ity—a view corroborated by the findings in SectionIII—then this will reduce the effectiveness of moneyas an anchor, while an exchange rate peg continuesto anchor the price level through its impact on trad-ables prices. Second, the control of the money sup-ply may itself require more sophisticated policymechanisms than the establishment of an exchangerate peg, which merely involves a decision to pas-sively buy or sell foreign exchange at a given price.

While these arguments are valid and important,one should bear in mind that if the objective is to re-duce inflation from very high to moderate levels,rather than achieving narrowly defined low inflationtargets, then the looseness in inflation control associ-ated with monetary anchors might be of secondaryimportance. Also, as to the control of the monetary

5For example, see Tarr (1994) and Section I for estimates ofterms of trade shocks affecting the Baltic states, Russia, and othercountries of former Soviet Union since 1992.

6The question of which considerations are important beyondthe disinflation phase are a separate matter.

7See, for instance, Sahay and Vegh (1995).

aggregates themselves, bank-by-bank credit ceil-ings—even though undesirable as permanent policytools—may in certain circumstances be effective inreigning in monetary expansion during the initialstabilization effort.

Costs of FailureIn general, the costs of failure may be expected to

be higher in exchange rate based stabilizations. First,the demise of an exchange rate peg typically in-volves a speculative attack on the currency, oftenwith a large loss in foreign official reserves. In con-trast, failure to attain monetary targets is likely tolead to an exchange rate depreciation broadly in linewith excessive money growth, which does not im-pose a comparable direct cost on the governmentand may be reversible without requiring the programto be called off immediately. Second, the reputa-tional cost to the government will in general behigher after the failure of an exchange rate basedprogram. The government will be faced with havingfailed to sustain a highly visible economic objectiveand the ensuing loss in credibility is likely to make itmore difficult to stabilize in the future. This notion issupported by evidence suggesting that failures of ex-change rate based stabilizations have typically re-sulted in a return to inflation levels higher than at theoutset of the program (see Vegh (1992)).

The failure to observe monetary objectives undera money-based stabilization, accompanied by down-ward pressures on the exchange rate, would alsodamage credibility. Missing monetary targets, how-ever, may be perceived less explicitly as a failure ofthe government than the forced floating of the cur-rency or a substantial devaluation, and it is easier forthe target path to be subsequently adjusted to recoupslippages and preserve the essential goals of the sta-bilization program.

Probability of SuccessIt has been argued that the public observability of

the exchange rate at any point in time—as opposedto monetary or credit aggregates, which are only ob-servable at substantial intervals based on data thatare usually supplied by the government—will en-hance the credibility of an exchange rate based ap-proach.8 In addition, since the unpleasant conse-quences of failure should act as a deterrent toabandoning the program, an exchange rate based sta-bilization may induce a higher commitment to un-dertake the necessary accompanying stabilizationmeasures—in particular, fiscal adjustment—and

8See Sachs (1994) and references quoted therein.

94


General Considerations and Experience

thus a higher probability of success. On the otherhand, while losses in confidence and private capitaloutflows may be harmful under both exchange ratebased and money-based approaches, they will pose adirect threat to the program's nominal anchor onlyunder an exchange rate based stabilization. Thus,success in stabilization under an exchange rate basedapproach will also require greater confidence in thegovernment's ability to stick to its target path of fis-cal adjustment.9 In addition, the magnitude of neces-sary fiscal adjustment needed under a fixed ex-change rate regime will most likely be higher thanthat under a money-based program, since peggingone's exchange rate will dictate a low inflation targetthat a money-based stabilization would not necessar-ily need to observe.10

These arguments seem to leave an open question,but one might argue as follows. Exchange rate basedstabilization may be preferable whenever the under-lying commitment of policymakers to fiscal disci-pline is high11 and the risk of adverse shocks that arebeyond their control is deemed reasonably small—that is, when it is likely that policymakers will be ina position to undertake the necessary fiscal adjust-ment, thereby increasing success rate of the peg. Onthe other hand, if there are serious questions regard-ing the government's underlying commitment to fis-cal restraint or large uncertainties that may renderadherence to fiscal targets difficult to achieve, thenan exchange rate peg would hardly be credible. Iftried, it would soon fail, with high costs to reputationand to reserve holdings.

In short, the commitment to an exchange rate pegis only one of the factors likely to influence fiscaldiscipline. It may be that the degree of disciplinerequired for the peg to be sustainable simply ex-ceeds the discipline that is likely to arise because ofa desire to avoid the costs of failure. A fixed ex-change rate regime may enhance the determinationto adjust fiscally, but it is unlikely to work politicalmiracles.

9This fiscal adjustment would include reductions in quasi-fis-cal expenditures, such as central bank directed credits and interestrate subsidies, where they are significant.

10Note that this argument need not apply to a crawling peg pol-icy, since under a crawling peg a less stringent path of fiscal ad-justment could be accommodated by designing the preannouncedexchange rate depreciation schedule accordingly. However, inconditions of uncertainty regarding future economic shocks andthe political feasibility of achieving fiscal targets, the relativelyhigh costs of failure associated with an exchange rate based ap-proach may lead to a less ambitious inflation objective under acrawling peg than under a money-based program.

11The "underlying commitment of policymakers to fiscal disci-pline" refers to the exogenous preferences of policymakers, as op-posed to their actual behavior, which will most likely be endoge-nous to the choice of nominal anchor.

A Currency BoardBefore turning to a discussion of the experience

with alternative anchors, it may be useful to brieflydiscuss the case of a currency board, which has beenestablished in two countries in the region. A currencyboard may be viewed as an extreme form of peggingin the sense that it leaves even less room for discre-tionary monetary policy.12 Thus, the arguments pre-sented so far regarding the standard case of a fixed ex-change rate—the potential superiority of the exchangerate as a nominal anchor, the problems associatedwith adjusting to real shocks, the higher costs of fail-ure and the more stringent implications for fiscal ad-justment—will also apply to the currency boardarrangement. However, both because of the impliedhard currency backing and the institutional shieldingof monetary policy from credit demands, a currencyboard has the advantage of being more credible than asimple peg, implying a lower risk of failure and lowercost of disinflation. This advantage must be traded offagainst the lack of flexibility that follows from therigidity of the arrangement, in particular with regardto adjusting the exchange rate peg and supplyingshort-term liquidity to the banking system. In a situa-tion of financial fragility and external shocks, thisrigidity could imply significant costs. A currencyboard might thus be accompanied with stipulationsfor the emergency situations that would allow liquid-ity support without undermining the institution's in-tegrity in normal circumstances.13

The above arguments on both the relative effective-ness and the risks of exchange rate and money-basedstabilization strategies are broadly corroborated bythe experience of countries outside the Baltics, Rus-sia, and other countries of the former Soviet Union(the latter will be discussed separately below). First,as shown in Table 1.7 and emphasized by Sahay andVegh (1995), all three attempts at exchange ratebased stabilization in Central Europe (Yugoslavia,Poland, and Czechoslovakia)14 were highly effectivein the sense that inflation was reduced to single-digitquarterly levels within less than one year. However,one of the three—Yugoslavia—subsequently failed asthe peg soon became unsustainable in the absence of

12See Bennett (1993, 1994).13On the problem of introducing some elements of discretion in

the currency board arrangement, see Camard (1994).14Note that Hungary has not been included in Table 1.7 as ei-

ther a case of money-based or exchange rate based stabilizationfor two reasons. First, inflation in Hungary was never very high,hovering between 5 percent and 15 percent a quarter in 1990-91,and stabilization attempts in 1991 did not make much of a differ-ence. Second, Hungary's approach to stabilization is difficult toclassify, because its policy of "adjustable pegging" can be viewedas either a managed float or an extremely loose exchange rate"anchor."

95



adequate supporting adjustment measures and infla-tion returned to very high levels. The recent record ofmoney-based stabilizations in Central Europe is alsorelatively favorable, in the sense that stabilization waseffective within a year and there were no major re-lapses, in four out of six cases—Albania, Slovenia,Croatia, and the Federal Yugoslav Republic of Mace-donia (three out of five if the Croatian approach is notinterpreted as a money-based stabilization).15 Be-cause this type of comparison does not control forother differences among the countries and circum-stances of stabilization, it does not constitute veryhard evidence; moreover, the sample sizes involvedare small. Nevertheless, it illustrates (1) that ex-change rate based stabilization can be very effectivein reducing inflation quickly from high levels and (2)that money-based stabilizations can also be both ef-fective and ultimately successful when monetary andcredit policies are consistently tight.

With regard to evidence from a broader set ofcountries, two major studies are noteworthy. Vegh(1992) reviews ten major exchange rate based plansto stop high chronic inflation in market economies(all in Latin American countries, except for the 1985Israeli stabilization). Seven of these are classified asfailures, in the sense that the peg could not be sus-tained and initial reductions in inflation were subse-quently reversed.16 In two cases the failure is attrib-uted to a real appreciation of the currency followingslow convergence of inflation, in spite of achievingfiscal balance (Chile and Uruguay, 1978). In the re-maining five, however, failure to adjust fiscally isthe main culprit. This experience shows that the dis-cipline induced by the exchange rate anchor may notin itself be sufficient to ensure the fiscal adjustmentnecessary to sustain the peg.

An analysis of a broad set of stabilization experi-ences between mid-1988 and mid-1991 suggeststhat—measured by the mean reduction in inflation inthe first year after stabilization, as compared withtarget inflation—exchange rate based stabilizationshave been generally more successful than stabiliza-tion attempts that did not use the exchange rate as anominal anchor. At the same time, in the group of

15Croatia pursued a money-based stabilization strategy in thesense that during the stabilization phase, in which inflation wasspectacularly brought down to practically zero in only four weeks(October 3, 1993 to early November 1993), the National Bank ofCroatia (NBC) relied on tight and publicly announced basemoney targets and there was no exchange rate peg. Because ofthe fast ensuing remonetization of the economy, the original basemoney targets were allowed to be exceeded, and from mid-No-vember onward the NBC began to focus much more on the ex-change rate as the nominal anchor. However, price stability hadalready been achieved at that point.

16See Vegh (1992), pp. 644-48. The successful exceptions areBrazil (1964), Israel (1985), and Mexico (1987).

five countries (Yugoslavia, Poland, Argentina, Mex-ico, and the former Czechoslovakia) that used ex-change rate anchors to stabilize from high levels ofinflation (in excess of 50 percent a year), two failedafter the first year (Argentina 1989 and Yugoslavia1989). Exchange rate anchors should be consideredsuperior in a first best world, but when programs arenot as strong as the ideal and indexing is not ad-dressed, exchange rate anchors are more likely toend in crises than to work. Thus, decisions to use anexchange rate anchor in programs should placegreatest priority on the prospects for resolute fiscaladjustment. In the context of transition economies,indexation is as yet not a major problem; however,the prospects for fiscal adjustment are indeed criticalin deciding whether or not an exchange rate anchoris appropriate.

To sum up, both the arguments and the experi-ences reviewed suggest that exchange rate anchorsare an effective and possibly superior approach tostabilization if supporting adjustment measures areadequate. On the other hand, stabilization attemptsin Central Europe that were money-based (or at leastdid not involve pegging) have in most cases been ef-fective as well. Whether or not a peg can be sus-tained will largely depend on whether it is accompa-nied by resolute fiscal adjustment. The latter needsto be assessed explicitly, as the evidence shows thatadequate fiscal adjustment is not automatically in-duced by the commitment to peg. Given the greatdisparities in the ability and willingness to adjust fis-cally across countries in the region that are the sub-ject of this paper, one cannot say in general whichapproach is likely to be best overall. Careful consid-eration of the individual circumstances is required.

Exchange Rate Policies inStabilization Program s

To date, most financial programs supported byFund resources in the region have entailed a degreeof exchange rate flexibility. Of the eleven countriesin the region for which programs were approved be-tween August 1992 and December 1994, ten pursuedflexible exchange rate strategies during the main sta-bilization attempt (Armenia, Belarus, Georgia, Ka-zakstan, Kyrgyz Republic, Latvia, Lithuania,17

Moldova, Russia, and Ukraine). In contrast, Estoniastabilized under a currency board arrangement.

17Subsequent to the main price stabilization phase, Lithuaniaintroduced a currency board in March 1994, while Latvia hasmaintained a de facto peg against the SDR (without formallycommitting to a fixed exchange rate) since February 1994, alsoafter the main stabilization phase.

96


Exchange Rate Policies in Stabilization Programs

In Armenia, Georgia, and Ukraine, the stabiliza-tion effort has only just begun, and it is too early foran assessment.18 For all other countries, inflationdata is reproduced in Table 1.7. According to the cri-teria used above, stabilization has been effective infour out of the seven countries that pursued amoney-based approach (Latvia, Lithuania, KyrgyzRepublic, and Moldova). Stabilization has also beeneffective in Estonia, the only country that stabilizedunder a fixed exchange rate strategy.

This section first describes the considerations thatled to the choice of a given anchor and then reviewsthe performance of both types of anchors in programcountries, beginning with a comparison between Es-tonia and Latvia and then turning to the experiencewith monetary anchors in the remaining programcountries. Finally, some conclusions regarding fu-ture stabilization policies in the region are presented.

Choice of Exchange Rate Regime inFund-Supported Program s

The choice of a nominal anchor in these programcountries has reflected the general arguments andtrade-offs described above. While the IMF has en-couraged the authorities to adopt ambitious disinfla-tion programs, the key requirements to underpin thesuccess of a fixed exchange rate strategy were notpresent in most cases. First, the political commitmentand implementation capacity required to achieve thefiscal adjustment needed to support an exchange ratepeg was generally not apparent. With the exception ofthe Baltics, prevailing fiscal deficits were extremelylarge when the first stabilization programs were intro-duced in these countries (Table 6.1). These underlyingimbalances cast serious doubt on the feasibility ofachieving the required fiscal adjustment.

Moreover, in most of these countries, there was noclear consensus on the desirability of radical stabi-lization and reform, implying that any program pred-icated on a sharp fiscal contraction that left littleroom for maneuver carried great risks. Second, mostof these countries lacked access to foreign exchangereserves that would be large enough both to with-stand normal temporary swings in the net supply offoreign exchange and to inspire market confidencein the sustainability of the peg (Table 6.2). Indeed, inthe seven program countries that have operated withflexible exchange rate regimes, foreign official re-serves averaged less than one month of imports atthe time of the first Fund-supported program, and itwas not clear that most of these countries would

Table 6.1. IMF - Supported Programs in theBaltics, Russia, and Other Countries of theFormer Soviet Union: General GovernmentFiscal Balance, Actual and Program1

(In percent of GDP)

Estonia3

Actual

SBA, 8/92

Latvia3

Actual

SBA, 8/92

Lithuania3

Actual

SBA, 9/92

SBA, 10/93

Kyrgyz Republic

Actual

SBA, 4/93

ESAF, 6/94

Moldova

Actual

SBA, 11/93

Russia4

Actual

STF, 6/93

STF, 4/94

Belarus

Actual

STF, 7/93

Kazakstan

Actual

STF, 7/93

SBA, 1/94

Ukraine

Actual

1991

5.2

6.3

7.2

4.8

-16.0

1.2

-7.9

-15.8

1992

0.9

0.6

—

-1.8

1.0

-14.8

-23.4

-18.8

-4.5

-7.4

-29.3

1993

2.3

1.0

0.9

-0.7

-8.2

-6.9

-8.9

-8.0

-10.0

-8.3

-9.7

-1.2

-13.2

19942

0.9

-1.7

-1.3

-1.0

-3.8

-4.1

-7.7

-3.5

-10.3

-6.5

-1.5

-6.0

-4.0

-8.0

18It is important to remember that this paper was written inearly 1995. Considerable progress has been achieved in somecountries since then. Eds.

Source: IMF staff estimates.

Note: EASF = enhanced structural adjustment facility; SBA =

standy-by arrangement; and STF = systemic transformation facility.1/ All program projections refer to original program targets.2Actual 1994 data reflect current estimates.3For the Baltic countries, both actuals and program numbers

refer to the financial balance (i.e., they exclude net lending).4Russian numbers refer to enlarged government deficit (i.e.

they include unbudgeted import subsidies).

have access to borrowed reserves, even in the face ofan ex ante strong program.

In contrast, a number of conditions prevailed in Es-tonia that supported the pegging of the exchange rate.Estonia registered budget surpluses in 1991 and thefirst half of 1992. A deterioration in the fiscal positionat the beginning of 1992 was quickly correctedthrough a strong fiscal package that included a set ofrevenue-enhancing measures in June. Thus, there waslittle doubt in Estonia's ability to undertake the fiscal

97



Table 6.2. Th e Baltics, Russia, and Other Countries of the FormerSoviet Union: Gross Reserve Holdings(In months of imports)


Kyrgyz Republi cMoldovaRussia

BelarusKazakstanUkraineArmeniaAzerbaijanGeorgiaTurkmenistanUzbekistan

June

1.41.60.4

...

...0.5

...

...

...

...

...

...

...

...

1992Dec.

4.51.41.2

0.9—0.9

...0.30.4

...0.6—

0.5

1993June

3.43.00.9

1.40.21.4

...

...

...

...—

...0.3

Dec.

4.85.52.3

1.50.91.8

0.21.30.70.10.3—5.92.2

June

2.84.22.6

1.52.51.9

0.6......

0.10.10.16.62.6

1994Sept.

3.04.62.8

2.32.11.2

0.62.60.21

0.22

—6.64.9

Sources: Data provided by country authorities; and IMF staff estimates.1October 1994.2August 1994.

restraint necessary to sustain the exchange rate peg.Moreover, the return of pre-war Estonian gold fromSweden, the Bank of England, and the Bank for Inter-national Settlements (BIS) provided sufficient re-serves for the full backing of domestic base moneyunder a currency board arrangement.

As for Lithuania, the currency board was intro-duced in April 1994 only after the Lithuanian govern-ment had demonstrated its capacity to adjust fiscally,reduce inflation, and stabilize the exchange rate.With monetary policy tightened significantly fromearly 1993, the monthly rate of inflation had beenstabilized at low single-digit levels for some time,and the budgetary position was strong. The litas hadbeen stable for some time, and by the time the cur-rency board was established, reserves were sufficientto provide more than 100 percent backing of basemoney at the then market exchange rate. Thus, whilethe new currency board arrangement was deemed im-portant to buttress the credibility of stabilizationthrough an institutional anchoring of fiscal and mon-etary restraint, the authorities' basic commitment tostabilization policies was clear.

In Latvia, the necessary conditions for peggingwere also broadly satisfied, with widespread politi-cal support for strong stabilization policies. How-ever, the Latvian authorities opted for a strong inde-pendent central bank in conjunction with a flexible

exchange rate arrangement. In terms of credibilityand the likelihood of fiscal restraint, the Latvianarrangement thus promised to be as effective as acredible exchange rate peg. In view of this, and theadvantages of the Latvian arrangement in copingwith external and real shocks, the IMF staff sup-ported the authorities' desire to employ a money-based approach in stabilizing.

Stabilization Performanc e Unde rAlternative Anchors: Estonia and Latvia

Estonia and Latvia have been broadly similar inmany of the circumstances that have influenced eco-nomic performance: location, size, factor endow-ments, external relations with Russia, and other coun-tries of the former Soviet Union, and the timing of thestabilization effort. Moreover, as Saavalainen (1995)points out, the overall monetary and fiscal stances ofthe two countries since mid-1992 have been compara-ble. Thus, Estonia and Latvia present a useful pair ofcountries for a more detailed comparison of the ef-fects of alternative stabilization approaches.19

19This was forcefully argued by Hansson and Sachs (1994) in afirst systematic comparison of the Baltic stabilization experiences.

98



Table 6.3. Estoni a and Latvia: Disinflation and Output Loss 1992-9 4

EstoniaReal GDP Inde x (1991Annualized inflation (in

LatviaReal GDP Inde x (199 1Annualized inflation (in

= 100 )percent)

= 100 )percent)

Q2

851,028

64722

1992Q3

801,167

621,097

Q4

761,122

601,272

Q1

77293

51398

Q2

82135

55186

1993Q3

8561

5685

Q4

8238

5934

Q1

7744

5234

1994Q2

8550

5637

PercentagePoint Change

I992:Q2-1994: Q2

-0.5

-8.6

Sources: Saavalainen (1995); and IMF staff estimates.

In comparing the behavior of macroeconomicvariables during the disinflation process in the twocountries, the main official indicators would suggestthe following:

• The initial disinflation paths were very similar forboth. Over the last year, however, inflation hasbeen lower in Latvia than in Estonia (Table 6.3);

• The estimated output decline during disinflationwas larger for Latvia than for Estonia;

• A comparison of domestic deposit rates betweenLatvia and Estonia for 1993 and 1994 (Table6.4) reveals large nominal interest rate differen-tials in favor of Latvia. Most important, a sub-stantial differential remains for 1993 and thefirst half of 1994 even after controlling for dif-ferences in domestic credit risk by taking intoaccount differentials between the Latvian andEstonian foreign exchange deposit rates.20 Thissuggests that a large part of the differential be-tween Estonian and Latvian deposit rates in1993 and early 1994 can indeed be attributed tothe different exchange rate arrangements.

20The Latvian deposit rate refers to government securities,whereas the Estonian rate refers to central bank securities. Thus,"some residual risk" may apply to the former relative to the latter,"since only the latter are guaranteed to be honored in cash (whichis also a central bank liability), (Saavalainen (1995) pp. 15-16).

For the first half of 1994, there is a substantial differential be-tween the Latvian and Estonian deposit rates, even after subtractingthe corresponding foreign exchange deposit rates (which capturedifferences in credit risk but not in the exchange arrangements—see Table 6.4, last row). For 1993, there is no data on Estonian for-eign exchange deposits, since such deposits only became legal in1994. However, even if we assume that Estonian interest rates onforeign exchange deposits would have been zero at that time, thus

The Latvian experience confirms that inflationcan be effectively and rapidly reduced under amoney-based stabilization and that the exchange ratepeg is not a precondition for fiscal discipline andquick stabilization. At the same time, interest ratedifferentials suggest that Estonia's exchange ratestrategy may have commanded greater credibilitythan Latvia's, which may have reduced the real costof stabilization. And indeed, the path of recordedoutput appears to favor Estonia.

Substantial care, however, should be exercised inconcluding that Estonia had a more favorable outputperformance because of its fixed exchange rate strat-egy. Indeed, in spite of their many similarities, thereare important differences in the conditions underwhich Latvia and Estonia stabilized that worked toEstonia's advantage. In particular, Estonia's closerties to Finland and Sweden may have contributed torelatively high foreign investment. In addition,Latvia was slow to privatize relative to the otherBaltic countries.21 Finally, and perhaps most impor-tant, the data on output for the two countries are notcomparable. The official national accounts data forEstonia incorporate an estimate of private sector ac-tivity—which is said to have grown rapidly to ac-count for a large share of total value added—whereas the Latvian figures do not include any suchestimate. Thus, the recorded difference in outputperformance almost certainly overstates the actualdifference; indeed, the actual situation may not havebeen significantly different.

maximizing the implicit differences in credit risk between Latviaand Estonia, a substantial differential would have remained be-tween the credit-risk-adjusted Latvian and Estonian rates.

21See Saavalainen (1995).

99



Table 6.4. Latvia and Estonia: Interest Rate Differentials1

(In percentage points)

1. LatviaDeposit rate (A)Forex deposit rate (B)Difference A-B (C)

2. EstoniaDeposit rate (A)Forex deposit rate (B)Difference A-B (C)

Differential (IA-2A)Differential (IB-2B)Differential (IC-2C)

January

86.426.360.1

15.5

70.9

April

65.818.447.4

18.6

47.2

1993

July

61.017.443.6

17.7

43.3

October

50.125.424.7

14.2

35.9

January

42.526.316.2

11.57.54.0

31.018.812.2

1994

April

32.321.211.1

11.511.40.1

20.89.8

11.0

July

34.022.012.0

12.15.26.9

21.916.85.1

October2

23.719.44.3

11.25.55.7

12.513.9-1.4

Sources: Data provided by country authorities; and IMF staff estimates.1End-of-month data. All rates refer to three-to-six-month time deposits.2End-September data for Estonia.

Experience with Money-Based Program sOutside the Baltics

We now turn to the stabilization experience in Be-larus, Russia, Kazakstan, the Kyrgyz Republic, andMoldova, the five countries that, in addition toLatvia and Lithuania, used money-based strategiesto combat high inflation. Belarus and Russia havehad programs supported under the Systemic Trans-formation Facility, while the remaining three coun-tries presently have programs supported by stand-byarrangements (followed by an enhanced structuraladjustment facility for the Kyrgyz Republic).

In the absence of a benchmark country operatingwith a fixed exchange rate but otherwise similareconomic conditions, it is impossible to disentanglethe effect of the exchange rate arrangement fromother factors affecting inflation and output perfor-mance. Our approach is thus to ask whether the con-ditions under which a monetary anchor ought to bethe better choice—based on the general arguments—in fact turned out to be present in the five countries.The difficulty with this approach is that it is notstraightforward to infer these conditions from the re-alized paths of economic variables. Our conclusionswill thus necessarily remain tentative and will needto consider information about the political environ-ment and the motivation of certain government ac-tions as well as economic conditions.

The experience indicates that, with the exceptionof Russia, all of these countries experienced large

swings in velocity following the adoption of theirfirst stabilization programs (Table 3.2). The instabil-ity in velocity has been particularly pronouncedsince mid-1993, with a large rise at the end of 1993and the first quarter of 1994. This implies that, atleast until the spring of 1994, money was a rather in-effective nominal anchor. Abstracting from all otherconsiderations for the time being, an exchange ratepeg might have been warranted for two distinct rea-sons. First, in the Kyrgyz Republic, Kazakstan, andMoldova, the initial rise in velocity in part may havereflected low confidence in the new national curren-cies; a credible exchange rate peg may have servedto enhance confidence. Second, apart from any con-fidence effects, an exchange rate anchor may havehad a direct moderating impact on inflation throughits effect on prices of tradables.

On the other hand, it is also true that during thesame period, the five economies (particularly the en-ergy importers) were hit by large real shocks, bothinternally—because of structural reforms andchanges in profitability in the traded and nontradedsectors—and externally—owing to sharp rises in en-ergy import prices (see Table 3.3). As a result, theequilibrium real exchange rate must have been sub-ject to large swings during the period, which mighthave undermined any given exchange rate peg, eventhough a significantly undervalued rate at the outsetwould have guarded against this risk to some extent.

In addition, all countries in this group failed—bya wide margin in certain cases—to meet fiscal and

100



credit targets. Surges in credit to the agricultural sec-tor occurred in the Kyrgyz Republic (Sep-tember-October 1993) and Kazakstan (March-May1994); similar credits contributed to excessive mon-etary expansion in Russia in the fall and winter of1993, and again from the summer of 1994 onward.In Belarus, fiscal targets were nominally met, butmonetary targets were missed by wide margins fromlate 1993 onward as the government effectivelytransferred budgetary operations—including creditsto agriculture—to the banking system. Slippages inMoldova primarily reflected delays in foreign fi-nancing and the effects of natural catastrophes.

Under an exchange rate based program, these slip-pages would have forced an abandonment of the pegor at least a devaluation. The crucial issue thus be-comes whether the commitment induced by a pegwould have had a substantial effect in preventing orreducing the slippages. In the absence of a counter-factual, it is impossible to know this with certainty,however, we would argue that the answer is likely tobe "no" in all cases. Indeed, some of these targetsthemselves did not reflect a sufficiently ambitiousdisinflation that would have been consistent with afixed exchange rate; and the size and nature of the fi-nancial slippages indicate that the even greater fiscaladjustment needed to sustain an exchange rate pegcould not have been achieved. In Russia and Belarus,the underlying political willingness and consensus toundertake the required adjustment policies was lack-ing, in spite of the efforts of some reform-mindedgovernment officials.22 This absence of sufficientcommitment to disinflation was reflected in the con-siderable policy failures under various Fund-sup-ported programs in 1992-94 some of which havebeen highlighted above. In Kazakstan, the mainsource of fiscal slippage under the 1994 stand-byarrangement was a large, ill-designed bailout of inter-enterprise arrears, which reflected the magnitude ofunderlying imbalances in the enterprise sector andlack of commitment to financial discipline at thattime (see Section IV). For the Kyrgyz Republic, theexpansion in credit in late 1993 primarily resultedfrom the perceived need to sustain agricultural pro-duction to ensure oil and gas deliveries under barteragreements with Russia and Uzbekistan. The size ofthe problem and the authorities' policy responsestrongly suggest that a peg would not have been sus-tained. In Moldova, the domestic slippages were re-sponses to exogenous shocks that were outside thecontrol of the authorities.

The Moldovan case is an example of a situationwhere a monetary anchor could be the preferablestabilization regime in a transition economy even if a

22See Section VIII for the case of Russia.

strong underlying commitment to adjust exists. If atransition economy is affected by exogenous shocksof the type experienced in Moldova, a temporary de-parture from fiscal consolidation may well be war-ranted. Without large reserves, this would generallyrequire a devaluation under an exchange rate basedprogram that, even though justified by the change inunderlying fundamentals, could be seen as a policyfailure by the public. A money-based program, how-ever, provides the opportunity of making such ad-justments while keeping the reputational losses in-volved small and ensuring that the deviation isindeed temporary. Moreover, in Moldova, the per-ception appears to have been that the Fund had mod-ified the program, not that the authorities had failed.Thus, the commitment of the authorities to the stabi-lization goal and its credibility in the eyes of thepublic was preserved in a way that may not havebeen possible under a devaluation.

In short, a review of the shocks that affected thefive transition economies during 1993-94 does notpoint to one or the other exchange rate approach asthe preferred choice. It is possible that exchange ratebased stabilizations—if they had been successful—would have provided a tighter control of the pricelevel in these countries than was possible under themoney-based programs. However, the politicaleconomy in these countries during the period of re-view suggests that exchange rate based programswould have carried unduly high risks and, givenchanging circumstances, may not have allowed asufficiently flexible response. In Russia and Belarus,money-based programs appear to have been the onlyfeasible alternative.

Conclusions for Future Stabilizatio nPolicies

Experience so far indicates that money-based pro-grams have succeeded in bringing inflation down ina number of countries in the region. Nevertheless,recent developments in many states appear to havestrengthened the case for adopting a fixed exchangerate strategy. Specifically, with progress in price andtrade liberalization, many of the required real shockshave already been registered in several countries.Moreover, political support for reform and stabiliza-tion has recently grown in some countries, perhapsreflecting the increasing realization that the gradual-ist approach has not yielded favorable results. Thus,the most important precondition for pegging—astrong underlying commitment to fiscal adjust-ment—may increasingly be present.

In assessing the appropriateness of alternative ex-change rate arrangements in the region during theperiod ahead, the different situations of individualcountries need to be borne in mind:

101



(1) Certain countries (including Armenia, Geor-gia, Kazakstan, Russia, and Ukraine) are at a stagewhen fixing the exchange rate could make a valu-able contribution to macroeconomic stabilization.Whether or not a peg would ultimately be warrantedwould depend crucially on whether the fiscal adjust-ment set out in these programs was sufficientlydeep—and the likelihood that it would be followedsufficiently high—to give to the exchange rate peg areasonable chance of success. In addition, the feasi-bility of pegging would, of course, depend on suffi-cient access to reserves from external sources, in-cluding possibly a currency stabilization fund.

(2) In other cases (e.g., Latvia, Moldova, and theKyrgyz Republic), consideration of an exchange ratepeg might not be actively pursued, even though thefiscal preconditions for such an approach might besatisfied. In Moldova and the Kyrgyz Republic,money-based programs have been successful in sub-stantially bringing down inflation, and re-monetiza-tion of the economy is well under way. Thus, thereseems little reason to change these arrangementsnow. Indeed, the authorities in both countriesstrongly advocate retaining the current money-basedapproach. As for Latvia, it has maintained an unan-nounced peg since February 1994, and there is an

issue as to whether this state should be formalizedthrough the introduction of a formally fixed ex-change rate or through a currency board arrange-ment. Given that the monetary authorities in Latviaenjoy high credibility, that inflation is the lowest ofany country of the former Soviet Union, and thatoutput recovery is under way, there are good reasonsfor retaining the current arrangement. The credibilitygain from switching to a formal peg would seemminor in this case and may be offset by the loss inflexibility.

(3) In the remaining countries, it would appear atthe moment that the preconditions for adopting anexchange rate peg are unlikely to prevail in the nearfuture: either the fiscal adjustment required under apeg will not yet be feasible, or the countries will beundergoing severe structural changes that imply largechanges in the equilibrium exchange rate, or both.

The appropriate exchange rate regime is likely toremain an important issue as reform in the regionprogresses. It is likely that political and economicuncertainties will remain unusually large, and under-lying circumstances highly fluid. A continual re-assessment of what may be the appropriate policyapproach will be required.

ReferencesBennett, Adam G.G., "The Operation of the Estonian Cur-

rency Board", Staff Papers, International MonetaryFund, Vol. 40 (June 1993), pp. 451-70.

, "Currency Boards: Issues and Experiences," IMFPapers on Policy Analysis and Assessment, PPAA94/18 (Washington: International Monetary Fund,September 1994).

Calvo, Guillermo, "Incredible Reforms", in Debt, Stabi-lization and Development, ed. by Guillermo Calvoand others, (Oxford: Basil Blackwell, 1989).

Camard, Wayne W., "The Design of Currency BoardArrangements: Discretion with Rules?" (unpublished;Washington: International Monetary Fund, October1994).

Chadha, Bankim, Paul R. Masson, and Guy Meredith,"Models of Inflation and the Costs of Disinflation,"Staff Papers, International Monetary Fund, Vol. 39(June 1992), pp. 395-431.

Fischer, Stanley, "Exchange Rate versus Money Targets inDisinflation", in Indexing, Inflation and EconomicPolicy, by Stanley Fischer, (Cambridge, Massachu-setts: MIT Press, 1986), pp. 247-62.

Hansson, Ardo, and Jeffrey Sachs, "Monetary Institutionsand Credible Stabilization: A Comparison of Experience

in the Baltics" (mimeograph; Stockholm: StockholmInstitute of East European Economics, June 1994).

Saavalainen, Tapio, "Stabilization in the Baltic Countries: AComparative Analysis," IMF Working Paper, WP/95/44(Washington: International Monetary Fund, April1995).

Sachs, Jeffrey, "Russia's Struggle with Stabilization: Con-ceptual Issues and Evidence," paper prepared for theWorld Bank's Annual Conference on DevelopmentEconomics, held in Washington, April 1994.

Sahay, Ratna and Carlos Vegh, "Inflation and Stabiliza-tion in Transition Economies: A Comparison withMarket Economies", IMF Working Paper, WP/95/8,(Washington: International Monetary Fund, January1995).

Sargent, Thomas, "The End of Four Big Inflations," in In-flation: Causes and Effects, ed. by Robert E. Hall(Chicago: Chicago University Press, 1982).

Tarr, David G., "The Terms of Trade Effects of Moving toWorld Prices on Countries of the Former SovietUnion," Journal of Comparative Economics, Vol. 18(February 1994), pp. 1-24.

Vegh, Carlos A. (1992), "Stopping High Inflation. An An-alytical Overview," Staff Papers, International Mone-tary Fund, Vol. 39 (September 1992), pp. 626-95.

102


VII Externa l Financia l Assistance:The Recor d and Issues

Eduard Brau1

T he subject of external financial assistance to theBaltic states, Russia, and other states of the for-

mer Soviet Union raises many questions that are ofcontinuing policy and operational interest and that aredifficult to do justice to within a reasonably con-densed format. This chapter makes an attempt, how-ever, by addressing, in turn, the following issues aftera presentation of the record in the two years 1992 and1993: Is financial assistance appropriately condi-tioned and timed? Is it adequate in size? Is the right fi-nancial assistance being provided? Is the assistancerepayable? And, last, what is to be done about the fi-nancing problems between the states in the region?

The Record , 1992-9 3Since the dissolution of the U.S.S.R. at the end of

1991, three influences have dominated balance ofpayments developments and financing flows of theBaltic states, Russia, and other states of the formerSoviet Union. The first is the assumption by Russia ofall external assets and liabilities of the U.S.S.R., withthe result that Russia has acquired a large stock of in-debtedness on which it has received debt relief fromofficial and commercial bank creditors. It has also in-herited sizable claims mainly on developing countriesthat, for the most part, are not being realized. A furtherconsequence is that all the other states began indepen-dence free of external liabilities (and claims)2 and allassistance to them has taken the form of new creditflows. The second major influence is the differentia-tion in trade balance positions and financing require-ments among these states, principally as a result of themovement of prices of energy products in their mu-tual trade to world levels. Net energy importers

1The author wishes to thank Patrick Lenain and MichaelBuchanan for their assistance in the preparation of this chapter.The paper was presented at the Conference on "Marketization ofthe former Soviet Union: The Banking and Finance Sector," orga-nized by the Adam Smith Institute, London, October 26—27,1994.

2The events leading to this result are well described in Chris-tensen(1994).

(Ukraine, Belarus, the Baltic states, Armenia, Geor-gia, the Kyrgyz Republic, Moldova, and Tajikistan)experienced a very sharp terms of trade deteriorationand trade deficits, while Russia and Turkmenistanbenefited from the change in relative prices. For themost part, however, the trade surpluses of the lattertwo with other countries in the region have been fi-nanced in a disorderly manner by the accumulation ofexternal payments arrears—by Ukraine, Belarus,Georgia, and Armenia, in particular—although theyhave been settled in an orderly fashion in the Balticstates, since 1993. The third major influence is thedominance in most countries of official financingamong net capital inflows and the large size of net pri-vate capital outflows, including capital flight. Excep-tions have been the Baltic states, where foreign directinvestment has been significant and capital flight neg-ligible. These influences are reflected in the estimatespresented in Tables 1.8, 7.1, and 7.2 for each of thecountries in the region.

As has been clear for some time, the transition ofthese countries toward market-based economies isgoing to be long and difficult. All have to adjust tothe impact of the disintegration of the Soviet Union;they also have to face a rapidly changing economicenvironment, confronted with large declines in out-put, high inflation rates in most instances and bal-ance of payments problems, and eroding living stan-dards. To help alleviate these economic difficulties,external financing was provided generally in supportof efforts to stabilize and reform. In addition to thefinancing provided by the IMF and the World Bank,bilateral credits were extended to Russia mostlyfrom seven major industrial countries, along with acomprehensive debt relief package for Russia by of-ficial creditors under the auspices of the Paris Club,and debt deferrals by commercial bank creditors.

Altogether, during 1992-93 the Russian govern-ment benefited from external financing estimated atabout $61 billion. Although a current accountdeficit (excluding grants) of nearly $6 billion wasrecorded in 1992, it is noteworthy that external as-sistance did not continue to finance a deficit in1993; rather, it is estimated that Russia generated asurplus of over $2 billion in that year. The external

103


VII EXTERNA L FINANCIAL ASSISTANCE: RECORD AND ISSUE STa

ble

7.1

. Th

e B

altic

s, R

ussi

a, a

nd O

ther

Sta

tes

of th

e Fo

rme

r Sov

iet U

nion

: Fin

anci

ng R

equi

rem

ents

,1992-9

3, C

um

ula

tive

(In m

illio

ns

of

U.S

. dollars

)

Arm

en

ia

Azerb

aija

n

Bela

rus

Est

onia

Ge

org

ia

Kaza

ksta

n

Kyrg

yz R

epublic

Latv

ia

Lith

uania

Mo

ldo

va

Russ

ia

Tajik

ista

n

Tu

rkm

en

ista

n

Ukra

ine

Uzbekis

tan

Tota

l

Cu

rre

nt

Acco

un

ts,

Exclu

din

g G

ran

ts

-33

9

372

-31

1

-10

6

-63

7

-1,8

47

-36

3

142

-12

1

-16

9

-3,4

00

-28

2

1,95

4

-80

0

-66

9

-6,5

76

Offic

ial

Re

serv

es

-5-2

6-3

3-4

35 -2 638

-17

-40

7

-33

9

-11

-6,9

00 -3

-83

6

-19

4

-81

6

-9,4

51

Am

ortiz

ation

to A

bro

ad

- — -2-1

2 -9 —

-22 — —

-23

,00

0

— —-1

30

-18

6

-23

,36

1

Oth

er

FS

U1

Arr

ears

- —

-15

0 — —

-65

7 — — — —

-2,5

00

—-1

,05

9 —

-4,3

61

Corr

espondent

accounts

- — —

-28 — —

-28

-41 —

-6,7

00

-78 — —

-6,8

75

Rest

of W

orl

d2

Com

merc

ial

banks

-

-47 —

-75 -7

-15

4

-17

-98

-24 —

-6,3

00 — —

-3,0

00

-9,7

22

Dis

burs

em

ents

to a

bro

ad

- -6 — — —

-10

0 — — — —

-1,3

00 — — —

-1,4

06

Oth

er

short-term

-72 —

-10

4 -4 —

-80 — — 30 2

-5,2

00 — — —

-14

-5,4

42

Err

ors

and

om

issio

ns

-17

4

-73

8

-91

1

-74

-31

5

-28

6

-10

7

-45

-16

1

-64

-6,2

41 86

-87

5

-3,9

01

102

-13

,70

5

Tota

l

-59

0

-445

-1,5

11

-73

5

-97

0

-2,4

86

-50

4

-45

8-6

55

-30

8

-61

,54

1

-27

8

-81

7

-8,0

25

-1,5

84

-80

,90

5

Sour

ce: I

nter

natio

nal M

onet

ary

Fund

.1F

inan

cing

for

coun

tries

of t

he fo

rmer

Sov

iet U

nion

incl

ude

s bi

late

ral l

oans

, arr

ears

an

d co

rres

pond

ent a

ccou

nts

for

the

Balti

cs, R

ussi

a, a

nd o

ther

sta

tes

of t

he fo

rmer

Sov

iet U

nion

.2Re

st o

f wor

ld in

clud

es al

l cou

ntrie

s ot

her t

han

the

Balti

cs, R

ussi

a, a

nd o

ther

stat

es o

f the

form

er S

ovie

t Uni

on.

104

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion

The Record, 1992-93Ta

ble

7.2.

Th

e B

altic

s, R

ussi

a, a

nd O

ther

Sta

tes

of th

e Fo

rmer

Sov

iet U

nion

: Rat

ios

of C

apita

l Inf

low

s to

GDP

,Im

port

s an

d Pe

r Cap

ita, 1

992-

93, C

umul

ativ

e

Arm

en

ia

Azerb

aija

n

Bela

rus

Esto

nia

Ge

org

ia

Kazaksta

n

Kyrg

yz R

epublic

Latv

ia

Lith

uania

Mo

ldo

va

Russ

ia

Tajik

ista

n

Turk

menis

tan

Ukra

ine

Uzbekis

tan

Tota

l

GD

P1

99

2 O

nly

1

(In m

illio

ns

of

U.S

. d

olla

rs)

2,7

18

5,4

32

30

,12

5

4,2

89

4,6

60

28

,58

0

3,6

55

5,0

81

4,9

22

5,6

37

38

7,4

76

3,7

93

4,7

44

94

,83

1

14

,87

5

60

0,8

28

Popula

tion

En

d-1

99

22

(In

thousa

nds)

3,6

77

7,3

70

10

,29

6

1,5

54

5,4

66

17

,03

8

4,4

93

2,6

40

3,7

56

4,3

60

14

8,9

86

5,5

52

3,8

57

52

,12

6

21

,45

8

29

2,6

29

GD

PP

er

Capita

(1

99

2)2

(In

U.S

. do

llars

)

739

737

2,9

26

2,7

60

853

1,6

77

816

1,92

5

1,3

10

1,29

3

2,6

01

683

1,2

30

1,8

19

693

1,4

71

Imports

(19

92

-3)

(In m

illio

ns

of

U.S

. d

olla

rs)

741

1,39

0

6,5

23

1,48

8

1,36

7

9,9

21

883

2,1

24

3,1

09

1,44

8

90

,06

7

900

2,6

02

28

,52

8

4,9

09

156,0

00

Inflo

ws fro

m R

est

of W

orl

d (

19

92

-93

)3

In p

erc

ent of

1992

G

DP

4

6.7

3.1 1.0

8.6

3.2

2.1

3.5

4.4 6.3

2.0

7.9

2.0 2.3

0.8 2.5

5.7

In p

erc

ent of

19

92

-3 im

ports

49

.0

24

.5 8.8

49

.4

22

.0

12.3

29

. 2

20

.9

20

.0

15.6

68

.0

16.5

8.2 5.1

15.4

44

.2

Per

capita

(19

92

-3)

(In

U.S

. do

llars

)

161

60 147

473

177

146

11

2

173

174

71

413 50

212

154

74

276

Sou

rce:

Inte

rnatio

nal

Mo

ne

tary

Fun

d.

1W

orl

d B

ank

data

; GN

P fo

r T

urk

me

nis

tan a

nd E

ston

ia.

2W

orl

d B

ank

data

.3R

est

of w

orl

d incl

udes

all

countr

ies o

the

r th

an t

he B

altics

, R

ussi

a, a

nd o

the

r st

ate

s o

f the f

orm

er

Sovi

et

Un

ion

.4A

vera

ge

annu

al in

flow

s in

19

92

-93

as p

erc

ent

of

GD

P in

1

99

2.

105

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion

VII EXTERNA L FINANCIAL ASSISTANCE: RECORD AND ISSUE S

financing was accompanied by a replenishment ofreserves of nearly $7 billion (from insignificant lev-els at the end of 1991), the granting of financing toother countries of the former Soviet Union andother countries in transition amounting to over $10billion (including arrears), refinancing of amortiza-tion payments to creditors of around $23 billion, re-financing of short-term obligations to banks and re-payments to suppliers for a total of $11 billion andthe flight of capital as recorded by the "errors andomissions" item of the balance of payments (over$6 billion).

External financial assistance was also provided tothe other countries in the context of programs sup-ported by the IMF and World Bank and under the aus-pices of the European Union and Group of Twenty-Four creditor countries and Consultative Groupsorganized to pledge additional financial support.Moldova, Kazakstan, and the Kyrgyz Republic bene-fited from the pledges made in the context of Consul-tative Groups, and Estonia, Latvia, and Lithuania re-ceived European Union and Group of Twenty-Fourfinancing for their programs. Expressed in relation toGDP, the Baltic states received the largest relativeamounts of foreign assistance—consistent with theirgreater commitment to and more rapid progress in re-form and stabilization. Ukraine received among thesmallest relative amounts of foreign financing—alsoconsistent with the lack of economic reform and stabi-lization in 1992-93. Kazakstan received pledges ofover $1 billion in early 1994 a substantial part waspledged by Japan as financing complementary to thestand-by arrangement of Kazakstan with the IMF andrehabilitation loan from the World Bank.

The economic hardship of many countries was ex-acerbated by the decline in the large explicit and im-plicit transfers that they had been receiving fromRussia. These included fiscal transfers to the poorerstates from the former union budget, which disap-peared in late 1991, and the subsidy for net energyimporters implicit in the underpricing of energy andraw material exports (relative to world prices),which was reduced significantly as interstate pricesfor these goods were raised. The effect of thesechanges was partially alleviated by new financialtransfers extended by Russia, mostly by the CentralBank of Russia through its correspondent accountsduring 1992 and early 1993, and by a buildup of un-paid claims by Russian and Turkmen enterprises oncompanies in other countries of the former SovietUnion. Altogether, the recipient countries receivedabout $12 billion during 1992-93.

These disbursements and arrears have led to arapid accumulation of debt both among states ofthe former Soviet Union and between them and therest of the world. Obligations between the statesare not documented comprehensively at this time,

but arrears to Russia and Turkmenistan on pay-ments for energy deliveries amounted to well over$3 billion in the third quarter of 1994, for example.The trend in debt to the rest of the world is clearfrom Table 7.3.

Is Financial Assistance Conditionedand Timed Appropriately?

The most important issue concerning official fi-nancial assistance is the effectiveness of the condi-tionality attached to it. If debt-creating official assis-tance is not conditioned on implementing policies toachieve sustained growth in the recipient economy,its purpose will be lost and its repayment willquickly become a burden. In particular, official fi-nancial support over an extended period will be nei-ther appropriate nor available on the scale thathelped finance the 1992-93 magnitude of capitalflight in many countries in the region. Thus, such as-sistance makes economic sense only to the extentthat it supports policies that lead to sustainedgrowth, minimizes net private capital outflows, andcreates conditions for net private inflows and equitycapital. Evidence in central and eastern Europe sug-gests that where a bold reform program is adopted,private capital inflows may begin to pick up substan-tially within two to three years.

By the criterion of conditionality, the record in1992-93 may be judged to be favorable, on balance.The clearly favorable evidence is that official sup-port was relatively highest in the countries achievingearly success in stabilizing currencies and adoptingdeep market-oriented reforms (Estonia, Latvia,Lithuania) and relatively lowest in countries at theopposite end of the spectrum (Ukraine, Belarus,Uzbekistan). The mixed evidence comes mainlyfrom Russia; while official support was substantialrelative to Russian GDP and relative to other coun-tries, policy success was more uneven and fragileand not sufficient to prevent massive net private cap-ital outflows in 1992-93. Reflecting the less ambi-tious policy commitment during that period, supportfor Russia was not as high as it could have been, in-cluding from the IMF.

Clearly, judgments, such as the above, raise theissue of whether greater assistance would have cre-ated greater policy success (see below). It is also dif-ficult to make fair judgments because much of theofficial bilateral financial assistance is motivated byconsiderations not related to support of economic re-forms in a direct sense. Humanitarian assistance, po-litical support, financing for denuclearization and avariety of other purposes are of varying importancein each recipient country. The support from multilat-

106


Is Financial Assistance Conditione d an d Timed Appropriately ?

Table 7.3. Th e Baltic States, Russia, and Other States of the Former Soviet Union : Debt andDebt Service, 1992-9 4(In millions of US. dollars)

Armenia

Azerbaijan

Belarus

Estonia

Georgia

Kazakstan

Kyrgyz Republic

Latvia

Lithuania

Moldova

Russia

Tajikistan

Turkmenistan

Ukraine

Uzbekistan

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 419921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

19921993

(est.) 199 4

External Deb t Due :(Stocks, end of period)

Rest of world1

...66.7

...—

52.0239.0

398.0961.0

1,530.0

64.0143.0288.0

...178.9264.3

1,700.02,747.0

5.0138.0277.0

69.0241.0469.0

99.0309.0580.0

60.0168.0360.0

78,700.083,700.084,800.0

43.296.7

270.6

...143.2375.0

386.0834.0

1,428.0

167.0990.0

1,341.0

Total

Interest

0.71.5

...

...

...

...

...12.0

...0.16.6

...

...10.0

...

...12.0

...

...0.4

...0.27.0

...

...13.4

...

...2.5

...5,500.04,900.0

...

...(4.2)(9.4)

...

...

...14.057.0

...——

...

Debt Service to Res t of

Amortization

—...............

2.0...

4.27.8

...

...3.9

...

...—

...

...—

...17.05.0

...

...—

...

...—

...8,900.0

14,700.0......

(1.7)

...

...

...—

123.0...

20.0190.0

...

World, Scheduled

Total

0.71.5

...

...

...

...

...14.0

...4.3

16.4......

13.9......

12.0......

0.4...

17.212.0

...

...13.4

...

...2.5

...14,400.019,600.0

...

...(4.2)

(11.1)

...

...

...14.0

180.0...

20.0190.0

...

Percent ofexports to

rest of world

5.85.1

...

...

...

...

...1.6

...1.83.5

...

...6.2

...

...0.8

...

...0.3

...3.92.6

...

...1.9

...

...1.4

...34.642.3

...

...1.32.7

...

...

...0.22.8

...2.3

13.2...

Source: IMF, World Economic Outlook, May 1994.1Rest of world includes all countries other than the Baltics, Russia, and other states of the former Soviet Union.

107



eral agencies is more directly correlated with reformprogress (Table 7.4).

In countries where conditions were not conduciveto private credit inflows or to larger official bilateralsupport, official export credit agencies have sought toestablish security arrangements in the form of off-shore escrow accounts. To allow the establishment ofthese arrangements for the financing of public sectorinvestments, the World Bank and the European Bankfor Reconstruction and Development (EBRD)changed their negative pledge clause policies in 1993to permit waivers of the clause in certain circum-stances. To date, waivers have been granted forKazakstan, the Russian Federation, and Uzbekistan.Escrow-secured lending has not yet become a majorelement in the financing of these economies largelybecause of resistance by the debtor country govern-ments to authorize the setting up of these arrange-ments, though some large deals have recently beenconcluded on this basis. The export credit agencies ofItaly, Japan, and the United States have reachedagreements with Russia on financing packages total-ing some $7 billion for the oil and gas sectors, to besecured by exports channeled through offshore es-crow accounts. Disbursements from these financingpackages are expected to be made over several years.Further deals are under discussion.

Escrow-secured financing is problematical. Whileit has the potential to generate additional foreign ex-change if targeted closely toward the rehabilitationof enterprises in the export sectors and the develop-ment of new sources of exports, a proliferation of es-crow accounts can encumber a large part of foreignexchange earnings. Extensive reliance on escrow-secured financing could reduce access to nonsecuri-tized lending over an extended period, even with im-provements in policies in the recipient country. Sucha proliferation would also raise serious concerns forthe multilateral institutions, including the Fund, at atime when these institutions are being called on toprovide a growing share of the financing for thecountries in this region.

As far as the timing of official assistance is con-cerned, two issues have attracted attention. One iswhether a major opportunity for success was missedin Russia by not making available in 1992 a cur-rency stabilization fund of $6 billion to help buildconfidence in a fixed ruble exchange rate, alongsidemore substantial balance of payments and budgetarysupport. The participants in the General Arrange-ments to Borrow (GAB), the Group of Ten countries,and Switzerland stated in 1992 that they were pre-pared to make such a fund available to Russia,through the IMF, once fiscal and monetary policieswere in prospect that would make feasible and sus-tainable a pegging of the ruble exchange rate to helpbring down high inflation. These conditions did not

exist in 1992-93. It is useful to recall that a currencystabilization fund would not make available balanceof payments financing, but it would provide re-sources for strictly temporary intervention by thecentral bank. The central bank would then be able tosupport the currency at the chosen exchange rateagainst speculative pressures, which arise eventhough monetary and fiscal policies remain broadlyappropriate. The offer to make available a currencystabilization fund in support of a bold stabilizationprogram has not been withdrawn.

The other issue concerns the timeliness of disburse-ments once external assistance is committed. The fi-nancing provided by other creditors in support ofFund-supported programs has very often been dis-bursed more slowly than expected. This has causedtighter-than-needed budgetary and balance of pay-ments conditions under some programs. The delays indisbursements have country-specific causes, but acommon element appears to be unfamiliarity betweenrecipient country authorities and creditors, and ad-ministrative weaknesses on the part of recipient gov-ernments. Experience suggests by now that disburse-ments will occur considerably slower in many of thecountries in this region than is customary elsewhere.

Is Financial Assistance Adequatein Size?

Whether multilateral and bilateral official financ-ing for reform in the transition economies is ade-quate in size requires answers to political questionsand to questions of political economy. Among thepolitical questions is whether the countries of theOrganization for Economic Cooperation and Devel-opment as a group have made, and are making, pub-lic budgetary resources available to aid the reform ofthe former communist countries in an amount com-mensurate with the historic opportunity and chal-lenge. Some observers have answered in the nega-tive, citing the resources previously employed in themilitary confrontation, and the gains to all concernedfrom the success of transition. Among the politicaleconomy questions is whether the available externalofficial resources have allowed individual reformingcountries to achieve a tolerable balance between re-duced macroeconomic imbalances and temporaryexceptional financing of remaining imbalances.

These issues of adequacy may be examined fromseveral perspectives. The approaches popular in thepublic debate have been to compare the scale of as-sistance to the Baltic countries, Russia, and otherstates of the former Soviet Union with that of theMarshall Plan for European Recovery provided bythe United States between 1948 and 1951, and tocompare actual assistance to Russia with announce-

108


Is Financia l Assistance Adequate in Size ?

Tabl

e 7.

4. C

om

mit

men

ts

and

Dis

burs

emen

t to

the

Bal

tics,

Rus

sia,

and

Oth

er

Sta

tes

of t

he F

orm

er S

ovie

t U

nion

: 199

2-Ju

ne

30,

1994

, Cum

ulat

ive

Arm

en

ia

Bela

rus

STF

Esto

ni a

SB

A(1

)SB

A (2

)ST

FG

eo

rgia

Ka

zaks

tan

ST

F(1

)ST

F (2

)SB

AK

yrg

yz

Republic

ST

F(1

)SB

AST

F (2

)ES

AF

Latv

ia

SB

A(1

)S

TF

(1)

SBA

(2)

STF

(2)

Lith

uani

aS

BA

(1)

ST

F(1

)SB

A (2

)ST

F (2

)M

old

ova

CC

FFS

TF

(1)

STF

(2)

SBA

Russ

ia

SBA

ST

F(1

)ST

F (2

)U

krai

neU

zbek

ista

n

Tota

l3

Qu

ota

(In m

illio

ns

of U

.S.

dolla

rs)

98.6

409.

4

67.9

162.

136

1.4

94.2

133.

6

151.

1

131.

4

6,29

7.1

1,45

6.1

291.

3

9,65

4.0

Effe

ctiv

e

Da

te2

07/2

8/93

09/1

6/92

10/2

7/93

10/2

7/93

07/2

3/93

01/2

6/94

01/2

6/94

05/1

2/93

05/1

2/93

09/2

0/93

07/2

0/93

09/1

4/92

12/1

5/93

12/1

5/93

07/1

5/94

10/2

1/92

10/2

2/93

10/2

2/93

04/0

8/94

02/0

5/93

09/1

6/93

12/1

7/93

12/1

7/93

08/0

5/92

06/3

0/93

04/2

0/94

Inte

rnatio

nal

Mo

ne

tary

Fun

d

Exp

iratio

n

Da

te

09/1

5/93

03/2

6/95

05/3

1/95

04/1

1/94

07/1

9/97

09/1

3/93

03/1

4/95

09/2

0/93

03/2

1/95

03/1

6/95

01/0

4/93

Co

mm

itm

en

ts

(In m

illio

ns

of U

.S.

dolla

rs)

102.

3

40.7

17.0

17.0

90.3

90.3

180.

7

23.5

39.6

23.5

103.

6

80.2

33.4

33.4

33.4

83.1

37.8

37.8

37.8

19.7

32.9

32.9

75.6

1,04

9.7

1,57

4.3

1,57

4.3

5,46

4.7

(In p

erc

ent

of

qu

ota

)

25.0

60.0

25.0

25.0

25.0

25.0

50.0

25.0

42.0

25.0

110.

0

60.0

25.0

25.0

25.0

55.0

25.0

25.0

25.0

15.0

25.0

25.0

57.5

16. 7

25.0

25.0

56.6

Dis

bu

rse

me

nts

(In m

illio

ns

of U

.S.

dolla

rs)

102.

3

40. 7 3.4 17.0

90.3

90.3

18.0

23.5

17.0

23.5

13.8

80.2

33.4

14.5

33.4

83.1

37.8 7.6 37.8

19.7

32.9

32.9

31.8

1,04

9.7

1,57

4.3

1,57

4.3

5,08

2.9

(In p

erc

ent

of q

uo

ta)

25. 0

60.0 5.0 25.0

25.0

25.0 5.0 25.0

18.0

25.0

14.7

60.0

25.0

10.8

25.0

55.0

25.0 5.0

25.0

15.0

25.0

25.0

24.2

1,04

9.7

25.0

25.0

52.7

IBRD

/IDA

Co

mm

itm

en

ts

(In m

illio

ns

of U

.S.

dolla

rs)

40.0

170.

280

.4

10.1

273.

7

155.

0

70.0

86.4

86.0

2,99

0.0

27.0

21.0

4,00

9.8

(In p

erc

ent

of

qu

ota

)

40.6

41.6

118.

4 6.2 75.7

164.

6

52.4

57.2

65.4

47.5 1.9 1.4

41.5

Dis

bu

rse

me

nts

(In m

illio

ns

of U

.S.

dolla

rs)

4.0

37.4

26.2 —

117.

3

39.3

34.9

45.9

76.2

586.

7

0.8 0.5

969.

2

(In p

erc

ent

of q

uo

ta)

4.1

9.1

38. 6 —

32.5

41.7

26.1

30.4

58.0 9.3

0.1 —

10. 0

EBRD

1

Co

mm

itm

en

ts

(In m

illio

ns

of U

.S.

dolla

rs)

57.4

145.

088

.6

120.

0

9.4 52.3

98.5 —

950.

0

62.6

112.

6

1,69

6.4

(In p

erc

ent

of

qu

ota

)

58.2

35.4

130.

5

33.2

10.0

39.1

65.2 —

15.1

4.3

7.7

17.6

Dis

burs

em

ents

(In

mill

ions

of

U.S

.

dolla

rs)

4.1

7.4

29.0 1.2

—

10.8

48.0 —

119.

4

1.9

50.3

27

2.1

(In p

erc

ent

of quota

)

4.2 1.8

42.7 0.3 — 8.1

31.8 — 1.9 0.1 3.5 2.8

No

te:

CC

FF

= c

om

pe

nsa

tory

and

contin

gency

fin

anci

ng fa

cilty;

EB

RD =

Euro

pean

Ban

k fo

r R

eco

nst

ruct

ion

and

De

velo

pm

en

t; E

SA

F =

en

ha

nce

d s

tructu

ral adju

stm

ent fa

cility;

IBR

D =

Inte

rnatio

nal

Ban

k fo

r

Reco

nst

ruct

ion

and

De

velo

pm

en

t; I

DA =

Inte

rnatio

nal

De

velo

pm

en

t Ass

oci

atio

n; S

BA =

sta

nd-b

y arr

angem

ent;

ST

F =

sys

tem

ic tra

nsf

orm

atio

n facility.

1EB

RD

co

mm

itm

en

ts a

nd d

isbu

rsem

ents

ar

e c

onve

rted a

t th

e E

uro

pe

an c

urr

ency

unit

(EC

U)

rate

on

Jun

e 3

0,

19

94

.2T

he

date

app

rove

d b

y th

e I

MF

's E

xecu

tive

Bo

ard

.3R

atio

s of to

tal d

isbu

rsem

ents

and

co

mm

itm

en

ts to

IM

F q

uota

are

bas

ed o

nly

on c

ountrie

s in

cluded in

the table

.

109

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion


ments made by the Group of Seven. Contrary towidespread impressions, the assistance rendered toRussia and several other states compares favorablywith the scale of Marshall Plan aid, even though as-sistance to Russia was significantly lower than itcould have been, or than it can be in support of boldpolicies in the future, especially from the IMF. A keydifference is that Marshall Plan assistance wasmainly on grant terms, whereas most assistance tothese states is not.

By modern standards, the Marshall Plan was a far-reaching adjustment program that required recipientcountries to liberalize their markets, open theireconomies to trade, restore internal financial stabil-ity, and stabilize exchange rates. Each expendituretranche of the Plan was negotiated with the U.S. au-thorities, and each dollar of aid was required to bematched by domestic counterpart funds that were tobe used for approved purposes, such as retiring pub-lic debt or funding specific public investments. The$13 billion provided by the United States in1948-51, mostly in the form of foodstuffs and rawmaterials, was equivalent to an annual average of 2 1/2percent of European GDP and some 1 1/2 percent ofU.S. GDP, certainly a large amount to be providedby one country mainly on grant terms.3 By compari-son, official new money inflows in 1992-93 (on anaverage annual basis) accounted for over 4 percentof 1992 GDP in Russia (nearly 8 percent of GDP in-clusive of official debt relief) and for about 6 percentof 1992 GDP in the Baltic states. Because of signifi-cant measurement problems concerning the GDP es-timates, these comparisons must be used with cau-tion; nevertheless, the scale of official assistancewas considerable in the period.

Assistance announced by the governments of theseven major industrial countries for the transforma-tion and stabilization of the Russian economy totaled$55 billion in 1992-93, subject to the fulfillment ofconditions necessary for using this assistance. Thetotal new financing actually provided to Russia in1992-93 fell short of the amount announced by theGroup of Seven by $13 billion, because of lower dis-bursements by the IMF and World Bank (Table 7.5).Official debt relief actually granted amounted to$12.4 billion, and when deferrals of pre-cutoff datematurities of $1.6 billion are considered, total debt re-lief closely approximates the Group of Seven esti-mates of $15 billion, and bilateral sources providedalmost $3 billion more than the proposed $21 billion.Disbursements by the IMF of $2.5 billion and byother multilateral creditors of only $0.5 billion weresubstantially below the amounts assumed by theGroup of Seven, because the Russian authorities were

unable to implement appropriate stabilization andstructural reform policies necessary to qualify forgreater multilateral support. It should also be notedthat in addition to official financing, Russia benefitedfrom $11.5 billion in commercial financing throughcommercial bank debt-service deferrals.

Another perspective on the question of whetherthe external assistance to these countries has beenadequate in size is to ask whether any of the stabi-lization programs agreed between countries and theIMF failed for lack of foreign financing. In 1992-93,this does not appear to have been the case. The pro-grams with Estonia, Latvia, and Lithuania receivedsubstantial external support and broadly achievedtheir objectives, albeit under very difficult condi-tions and with deeper-than-expected declines in out-put and real incomes. However, as mentioned ear-lier, delays in the disbursement of balance ofpayments assistance committed by creditors otherthan the IMF temporarily made the adjustmentsomewhat more difficult than it need have been,though some recipient countries also experiencedproblems in absorbing available assistance in a pro-ductive manner. In the Kyrgyz Republic, substantialexternal support was also available but supportunder the 1993 stand-by arrangement with the IMFwas curtailed when Kyrgyz credit policy was inap-propriately eased in late 1993. Substantial externalsupport became again available in conjunction withthe ESAF arrangement concluded in July 1994. InRussia, the authorities committed themselves tomodest stabilization objectives under programs for1992 and 1993 supported by a first credit trancheand a first systemic transformation facility purchase,respectively, from the IMF, and by substantial otherfinancing. Progress was made in several areas ofsystemic reform, but stabilization objectives werenot realized. There is no suggestion from any quarterthat these programs failed in their stabilization ob-jectives because of nondelivery of committed for-eign financing.

Of course, such a statement begs the more inter-esting question of whether there could have beenmore decisive success in stabilization and reform inRussia if the commitment of official assistance hadbeen much larger. It will never be possible to settlethis counterfactual question satisfactorily. In gen-eral, one might expect larger official assistance tocreate a willingness to undertake tougher anti-infla-tion policies, as well as providing money that couldbe used to ease the adjustment burden. However,Russia's policies have usually been determined byother considerations, leaving little room for any ad-justments in response to the availability of finance.4

3See Eichengreen and Uzan (1992). 4See Section VIII.

110


Is Financial Assistance Adequate in Size?

Tabl

e 7.

5. F

inan

cia

l Ass

ista

nce

to R

ussi

a—An

noun

cem

ents

and

Out

com

es

Bila

tera

l cre

dito

rs a

nd

Euro

pea

n U

nio

nG

rant

s3

Loan

s

Mul

tilat

eral

cre

dito

rsW

orld

Ban

kEu

rope

an B

ank

for

Rec

onst

ruct

ion

and

Dev

elop

men

t

IMF Sy

stem

ic tr

ansf

orm

atio

n fa

cility

Stan

d-by

arr

ange

men

tS

tabi

lizat

ion

fund

Deb

t rel

ief

gran

ted

Offi

cial

, bila

tera

l cre

dito

rs7

Non

offic

ial c

redi

tors

Am

ount

s pe

ndin

g se

ttle

men

tO

ffici

al, b

ilate

ral c

redi

tors

Def

erra

l of p

re-c

utof

f mat

uriti

esA

rrea

rsN

onof

ficia

l cre

dito

rsD

efer

ral f

rom

com

mer

cia

l ban

ks

and

supp

liers

Arr

ears

Tota

l (ex

clud

ing

deb

t rel

ief

and

amou

nts

pend

ing

from

non

offic

ial c

redi

tors

)To

tal (

incl

udin

g d

ebt r

elie

f an

d am

ount

s pe

ndin

gfr

om n

onof

ficia

l cre

dito

rs)

$24

billio

npa

ckag

e1

11.0 1.5 9.0

3.0

6.0

2.5

2.5

24.0

24.0

1992

Firs

t C

red

ittr

anch

e pr

ogra

m

12.2 2.7

9.5 1.0 0.8

0.2 1.0 1.0 — 7.8l0

1.6 1.6 6.2

4.2 1.9 15. 8

22.0

Out

com

e2

15.8 3.0

12.8

0.01 —

0.01 1.0 1.0 — 14.1 5.4 1.6 3.8

8.7

5.6 3.1

22.2

30.9

$43

billio

npa

ckag

e1

1993

STF

prog

ram

Pre

limin

ary

outc

ome2

(In

billio

ns o

f U

.S.

do

llars

)

10. 0

10.0 5.3 5.04

0.35

13.1 3.0

4.16

6.0

15.0

15.0

43.4

43.4

10.0 3.0

7.0 1.3 1.1

0.2

3.0

3.0

28.5

13.5

8

15.0

9

27.8

42.8

8.1

2.9

5.2

0.5

0.4

0.1 1.5 1.5 12.4

12.4 5.0

-2.7

-2.7 7.7

5.9 1.8 19.8

27.5

1992

-93

Cum

ulat

ive

tota

l

Pack

ages

21.0 5.3 14.1 3.0

5.1

6.0

15.0

15.0

55.4

55.4

Prog

ram

22.2 5.7

16.5 1.3 1.1

0.2

4.0 3.0 1.0

30.1

15.1

15.0

42.6

57.6

Out

com

es2

23.9 5.9 18.0 0.5

0.4

0.1

2.5 1.5 1.0 12.4

12.4

19.1

11

2.7 1.6 1.1

16.4

11.5 4.9

42.0

58.4

Sour

ces:

Rus

sian

Fed

erat

ion

Min

istr

y of

Fin

ance

; Vne

shek

onom

bank

; IM

F st

aff e

stim

ates

; an

d va

rious

pre

ss re

leas

es o

f Gro

up o

f Sev

en m

inis

teria

l pac

kage

s.1I

nclu

des

tech

nica

l ass

ista

nce.

2Cove

rs c

alen

dar y

ears

and

exc

lude

s te

chni

cal a

ssis

tanc

e.3I

nclu

des

hum

anita

rian

assi

stan

ce a

nd G

erm

an ai

d to

reho

use

Russ

ian

troo

ps.

4Inc

lude

s es

timat

ed in

itia

l dis

burs

emen

ts o

f $1.

1 bi

llion

und

er a

$0.

6 bi

llion

impo

rt re

habi

litat

ion

loan

app

rove

d in

Augu

st 1

992

and

an $

0.5

billio

n o

il se

ctor

loa

n ap

prov

ed in

Jun

e 19

93.

5N

ew c

omm

itmen

ts, in

clud

ing

$0.5

billi

on in

cof

inan

cing

of o

il se

ctor

loan

s.6Th

e fig

ure

refe

rred

to

max

imum

pot

entia

l com

mitm

ent b

egin

nin

g fr

om O

ctob

er 1

993

.7D

ebt r

elie

f is

brok

en d

own

by

relie

f act

ually

gra

nted

and

def

erra

l of p

re-c

utof

f mat

uriti

es; t

o co

mpa

re w

ith

Tabl

e 1.

3, d

efer

rals

mus

t be

adde

d to

deb

t rel

ief a

ctua

lly g

rant

ed.

8Exc

ludi

ng $

1.6

billio

n o

n th

e pr

inci

pal c

over

ed b

y th

e 19

92

debt

def

erra

l fro

m o

ffici

al c

redi

tors

, whi

ch is

incl

uded

in th

e $1

5 bi

llion

that

was

par

t of t

he $

43 b

illio

n p

acka

ge.

9Inc

lude

d as

sum

ed d

ebt r

elie

f of $

8.7

billio

n w

ith re

spec

t to

the

199

2 de

ferr

al a

nd a

rrear

s.10

Prog

ram

ass

umed

that

cre

dito

rs w

ould

def

er R

ussia

's 6

1 pe

rcen

t sha

re o

f the

pay

men

ts o

f the

cou

ntrie

s o

f the

form

er S

ovie

t Uni

on

of p

re-c

utof

f prin

cipa

l and

inte

rest

due

in 1

992

as w

ell a

s pr

e-19

91 c

omm

er-

cial a

rrear

s.11

Amou

nts

due

to o

ffici

al c

redi

tors

with

resp

ect t

o 1

992

mat

uriti

es o

f $5.

4 bi

llion

are

incl

uded

in d

ebt r

elie

f gra

nted

in 1

993.

111

©In

tern

atio

nal M

onet

ary

Fund

. Not

for R

edis

tribu

tion


In 1992, the needed ambitious policies meritinglarge-scale IMF support could not be agreed, partlybecause the ruble area issue could not be settled, andRussia requested Fund support under the first credittranche.5 In 1993, and after substantial difficulties inpolicy implementation in Russia in 1992, it was pre-cisely the absence of the prospect of sufficiently am-bitious and credible policy objectives in Russia andseveral other states that contributed to the develop-ment by the IMF in early 1993 of the systemic trans-formation facility (STF), a facility that permits dis-bursements before policies are in place that wouldqualify for support by the IMF's larger and moreconditional facilities. The STF is a bridge facility toallow more ambitious policies to be prepared; Rus-sia used it in June 1993 and again in April 1994. Inthe fall of 1994, the IMF again urged Russia to adoptbold stabilization policies for which significant IMFsupport would be available.

Another perspective on the issue of the adequacyof financial assistance to states of the former SovietUnion is a forward-looking one. Here, a key chal-lenge is posed by the size of the financing needs ofUkraine.6 Now that coherent and determined poli-cies are beginning to be adopted—after the period ofdisastrous economic drift since 1992—relativelylarge external financing requirements will need to bemet initially by official assistance and regularizationof accumulated payments arrears in order to allowUkraine to escape from continued chaotic economiccontraction. As soon as appropriate economic pro-grams are ready for implementation, the same chal-lenge will be faced in Belarus, and also in Armenia,Georgia, and to some extent Azerbaijan, which, ad-ditionally, have to cope with the aftermath of pro-tracted military and civil conflicts.

A final perspective on the adequacy of financial as-sistance is provided by an assessment of net privatecapital flows to Russia and neighboring transitioneconomies. As noted earlier, the record in 1992-93 inRussia and other countries in the region other than theBaltics is poor. It is clear that conditions must be cre-ated that allow the transition economies to raise do-mestic savings and to achieve the transfer of foreignsavings and of technology essential to the public andprivate investment necessary for growth. The ade-

5Purchases under the first credit tranche raised the Fund's hold-ings of the member's currency in the credit tranches to not morethan 25 percent of quota. Subsequent purchases in three succes-sive tranches, each equivalent to 25 percent of quota, are known asupper credit tranches. Conditionality rises through the tranches,and the Fund adopts a liberal attitude in making resources avail-able in the first credit tranche. For more information on this andthe Fund's facilities, see IMF (1995).

6The reader should keep in mind that this paper was completedin October 1994 and incorporates information until that time.Eds.

quacy of foreign assistance should to a large extent bejudged by the extent to which countries using officialassistance create such conditions and achieve accessto private foreign financing. Foreign assistanceshould not only be judged by the size of official flows,though a critical mass of assistance is, of course, im-portant when reforms are begun and sizable officialsupport can foster the needed policies. The conditionsnecessary to attract sizable private financing includereasonable stability of the currency, modernized com-mercial banking systems, a liberal exchange and tradeand foreign investment regime, and greater pre-dictability in the enforcement of private contracts. Akey purpose of official financial assistance must be tobring down decisively and on a sustained basis thenoncommercial risks for committing external and do-mestic private capital.

Such conditions are increasingly being establishedin the countries of central and eastern Europe andhave contributed to gross private inflows of privateforeign direct investment to the region of some $17billion in 1991-93, as against a mere $3.6 billion inthe Baltics, Russia, and other countries of the formerSoviet Union in 1992-93. To be sure, reforms startedearlier in central and eastern Europe and the difficul-ties facing the Baltics, Russia, and other countries ofthe former Soviet Union are more extensive, but soare the opportunities—especially in Russia with itsvast resource base, technically skilled workforce,and low wage rates.

Is Right Financia l AssistanceBeing Provided ?

Balance of payments financing in the form of offi-cial and commercial debt relief to Russia and quickdisbursing support to Russia and other countriesfrom the IMF and the World Bank has played an im-portant role, alongside the provision of export cred-its from major industrial countries. While the provi-sion of balance of payments financing is appropriate,the very low project and sectoral financing disburse-ments to date are noteworthy, including those frommultilateral development banks. This mainly reflectsthe natural delay in building a project pipeline anddeveloping sectoral policies in the unsettled condi-tions in many countries and the need for carefulpreparation. Indeed, a significant increase in projectand sectoral financing commitments by multilateralbanks occurred in late 1994 in several states of theregion. It is essential that this trend accelerate andspread to all these countries as structural reformsdeepen and need impetus from, and support by, ex-ternal financing and know-how. Unfortunately, in1992-93, manifold administrative problems in Rus-sia obstructed the International Bank for Recon-

112


Is Right Financia l Assistance Bein g Provided ?

struction and Development's (IBRD) proceedingwith project preparation and disbursements in atimely manner.

The provision of official financing to the emerg-ing private sectors in the Baltics, Russia, and othercountries of the former Soviet Union by the IBRD,the EBRD, and the International Finance Corpora-tion (IFC) has begun and should, over time, assumea large share of total official financing to the coun-tries in the region.

Concerning the terms on which financing is pro-vided, the large differences in the debt-carrying ca-pacities among these countries require careful creditorpractices. In 1992 and 1993, short-maturity financingon commercial terms was provided from somesources, including the European Union (EU), to thepoorest among the countries of the former SovietUnion. This inappropriate activity has been comingunder somewhat better control through the limitsplaced on nonconcessional external borrowing underIMF-supported programs. Among the countries ofthe former Soviet Union, Armenia, Georgia, theKyrgyz Republic, and Tajikistan are eligible for con-cessional International Development Association(IDA) funds from the World Bank, and for conces-sional enhanced structural adjustment facility(ESAF) resources from the IMF.7 In the future, thisfinancing from multilateral sources will need to bematched by bilateral concessional financing andgrants.

In general, financing should be provided on termsappropriate to the circumstance of a lengthy periodof transition during which the balance of paymentspositions would be vulnerable to shocks. On the partof the IMF, this means that its financing, wheneverpossible, should be provided on terms requiring re-payment of loans over a longer period than the three-to-five-year repayment of stand-by resources, partic-ularly in view of the decision of the Executive Boardin October 1994 to increase maximum annual accessto regular Fund resources to 100 percent of quota.Repayment periods of four to ten years apply to pur-chases under the STF and under the extended Fundfacility (EFF). A three-year extended arrangementwas agreed with Lithuania to begin in October 1994,and other countries have expressed an interest inputting in place policies suitable for support underthe EFF.

For bilateral official assistance, tied export creditsare the main vehicle of nonconcessional financingand, given the specific features of this financingsource, need to be handled with care. Such creditsare tied to new imports of specific products and are

expected to be repaid quite quickly. This suggeststhat they should not be relied on as a major source ofnet financing for prolonged periods. Countries thathave used export credit for the bulk of their financ-ing needs have often run into debt- servicing diffi-culties; of the 20 largest recipients of exports creditsthrough 1992, only 6 have avoided substantial pay-ments arrears or recent debt rescheduling.8

These limitations of export credit financing under-score the urgency of diversifying financing to tapnonofficial sources; they also caused Russia to seekto curtail and scrutinize carefully the contracting oftied export credits in 1994 and beyond. There are in-stances, including in some of the smaller countries inthe region, where official credits have been extendedon relatively short maturities and with bullet repay-ments that will be difficult for borrowers to honor;indeed, in one or two cases, they may well be impos-sible to honor. Similarly, there are a number of casesof tied export credit financing for purposes that can-not be regarded as contributing to sound economicrestructuring and productivity enhancement.

A different aspect of diversification of financingsources is diversification among providers of officialfinancing. In the near term, the share of financingfrom multilateral sources is bound to rise signifi-cantly as IMF and World Bank support for economicprograms increases, both in countries already receiv-ing multilateral assistance and in new ones, includ-ing Ukraine. Bilateral donors and creditors, how-ever, will find it increasingly difficult to maintain thescale of their financing provided in 1992 and 1993without changes in burden sharing among them. Inthe period from September 1990 to mid-1994, newfinancing commitments from the EU, its memberstates, and other European countries accounted fortwo thirds of all bilateral commitments to theBaltics, Russia, and other states of the former SovietUnion; of the European commitments in turn, morethan two thirds was accounted for by Germany. Theshare of the EU and its member states is estimated at49 percent of humanitarian grants, 42 percent oftechnical assistance, 59 percent of credits and creditguarantees, and 90 percent of assistance in connec-tion with the withdrawal of military forces.

As mentioned, much financing has been extendedfor balance of payments purposes. Some observershave, in addition, called for larger amounts of directbudgetary financing. All balance of payments financingextends an economy's absorption possibilities, includ-ing that of the central government. The governments ofmany countries of the former Soviet Union receive of-ficial external financing directly benefiting the budget,as well as higher recourse to domestic credit financing

7Azerbaijan became eligible for using ESAF resources fromMay, 1995, after this paper was written, Eds.

8See Kuhn, Horvath, and Jarvis (1995).

113



made possible by the balance of payments support. Be-yond this, the issue of budget support and, more gener-ally, support of a political nature is one for bilateralcreditors and could obviously be helpful if available forimportant expenditures such as on social safety netsand retraining, enterprise restructuring, infrastructureimprovement, and others. Such external financing ofbudgetary expenditures helps the government achieveinflation objectives with less curtailment of those ex-penditures that can foster quick economic transition asmay be seen from the appendix. However, the structureof budget expenditures must be conducive to reformand the need for budgetary support must be temporary,such as relating to enterprise restructuring. By contrast,support for financing subsidies and credits to existingenterprises, for example, without incentives to change,is usually counterproductive.

A large amount of financing is in support of techni-cal assistance provided principally by the EuropeanUnion, the United States, Germany, other govern-ments, and the World Bank, EBRD, OECD, the UnitedNations Development Program (UNDP), and the IMF,a share of whose technical assistance is financed bygrants from Japan. It is estimated that commitmentsfor technical assistance financing in all fieldsamounted to as much as $5 billion by late 1994. Whenwell coordinated by providers and effectively imple-mented by the recipient countries, this assistance hasvery high benefits to the transition economies.

Is Assistance Repayable ?Except in Russia, external debt and debt-service

obligations to countries other than those of the for-mer Soviet Union are low at present relative to GDPor exports (Table 7.3). From this perspective, mostcountries appeared to pose limited repayment risk atthis stage. The uptake of debt, however, has alreadyrisen sharply in some of these countries that beganindependence in 1992 with no debt obligations. Al-ready, particularly acute balance of payments diffi-culties in countries such as Georgia and Ukraine willmake the servicing of post-1992 debt, as scheduled,not possible, in part as a result of lumpy repaymentsfalling due on short maturity credits. In these coun-tries, financing was provided on inappropriate termsand was not in step with progress in reform, and spe-cial help will be needed on repayments of credits tothe OECD area as well as regularizing arrears withother states in the region (see below) when programsof stabilization and reform are launched.

Medium-term balance of payments projectionsprepared in conjunction with Fund-supported adjust-ment programs suggest difficult balance of paymentspositions in the next few years for many countries inthe region and in particular for the net energy im-

porters—other than the Baltic states, where the out-look is more favorable as a result of the rapid eco-nomic transition. The projections invariably suggestthat two factors will be critical in containing the dif-ficulties to manageable proportions. The first is thesuccess of policies in stabilizing monetary condi-tions in the countries concerned, which should helpreverse capital flight and start sustained inflows ofprivate financing. The second critical factor is thesuccess of policies in switching exports to worldmarkets and increasing domestic supply capacity.The experience of central and eastern Europe andthe Baltic states suggests that dramatic gains in ex-port earnings are possible. However, with a rela-tively rapid accumulation of debt obligations in mostnet energy-importing countries, their balance of pay-ments prospects will remain vulnerable to adversedevelopments in the pricing of energy imports andother shocks, notwithstanding the large increase inenergy efficiency that should occur.

The rapid buildup in debt obligations in somecountries is occurring in conditions in which debtmonitoring and management systems are unsatisfac-tory. This also accounts for a relatively common ten-dency among countries of the region for debt-servic-ing payments not to be made on time—even wheredebt obligations are low in relative and absoluteamounts—with obvious consequences for percep-tions of creditworthiness. It is important that themanagement of external credit operations bestrengthened quickly with technical assistance fromthe World Bank and the UNDP.

What I s to Be Done AboutFinancing Problems Betwee nStates in the Region?

As noted earlier, Russia and Turkmenistan havebeen accumulating large unpaid current net claimson several countries in the region, notably Ukraineand Belarus, largely from deliveries of energy sup-plies. it is estimated that arrears on energy paymentsto Russia and Turkmenistan amounted to a total of$2.7 billion for Ukraine and $0.5 billion for Belarusby the third quarter of 1994. Periodically, mutualclaims are offset and the net amounts are consoli-dated into state credits from Russia and Turk-menistan. Typically, however, the debtors promptlyfall into arrears again, given their macroeconomicimbalances. These interstate arrears are a threat tothe development of normal trade and financial rela-tions among these countries and give rise to otherfrictions.9 Moreover, unless the arrears problem is

9See also Section IV.

114


solved, it could complicate attempts to arrange pro-grams with the IMF and to gain necessary financingsupport from other creditors.

The main types of arrears for which these govern-ments must take responsibility are those relating tononsettlement of payments due under intergovern-mental agreements (credits and intergovernmentalprocurement) and under intergovernmental tradeagreements carried out by state enterprises, notablyin energy. To the extent that arrears are being builtup owing to restrictions on access to foreign ex-change, the governments will also have to eliminatethose restrictions. In addition, enterprises not tradingunder intergovernmental agreements may incur ar-rears due to their own illiquidity, insolvency, or out-right refusal to pay; these arrears, however, are partof the broader problem of interenterprise arrears thatexist in many of these countries and they will haveto be resolved by enterprises themselves.

IMF staff have offered to extend their good offices,if requested by all parties, to help them resolve ques-tions over the magnitude of interstate arrears and to as-sist in other ways. IMF-supported programs withcountries in interstate arrears require that existing ar-rears be regularized on terms that are sustainable andthat new arrears due to the actions of the authorities beavoided. This will entail, in addition to the debtor'sstabilization and adjustment effort, new balance ofpayments financing mobilized in the context of Con-sultative Groups. Regional creditors such as Russiaand Turkmenistan would participate in such new fi-nancing, along with OECD countries and multilateralinstitutions. Any renewed failure by debtors to honorthe new payments schedules will endanger theprospects of stabilizing their economies and mobiliz-ing external financing. At the same time, all govern-ments should take steps to remove themselves asquickly as possible from negotiating energy and otherproduct deliveries in order to avoid the multiple moralhazards present in current arrangements.

ConclusionOf the differentiated and complex external financ-

ing issues facing the Baltics, Russia, and other coun-tries of the former Soviet Union, four concluding re-marks may be made. First, bilateral and multilateralfinancing that is not linked to rapid stabilization andreform is inappropriate; this applies to all nonhu-manitarian official bilateral assistance that is not ex-tended on grant terms, and especially to export creditsupport. Second, the size of official support avail-able to Russia and other reforming states appears notso far to have constrained reform success, but raisingsufficient official support for countries only now be-ginning stabilization and reform will be a significant

challenge. Third, the adequacy of bilateral and mul-tilateral external assistance is critically affected bythe extent to which that assistance supports policiesthat curtail the large net private capital outflowsfrom the region that occurred in 1992-93 and, in-stead, establishes conditions for sustained privatecapital inflows. Finally, financial relations betweenmany of these states remain disorderly and need tobe normalized as soon as possible.

Appendix

How Externa l Financing Helps Economiesin Transition

From the balance of payments identity, and ex-pressing the current account balance in terms of sav-ings and investment balances:

= AR-GS-(PS-PI) = (PI-PS),

where KA denotes the external capital account bal-ance, AR the change in official international reserves(excluding foreign borrowing by the central bank),GS the fiscal balance, GD the fiscal deficit, PS pri-vate savings, and PI private investment. As the fiscaldeficit must be financed through net credit from thebanking system (ADCg), net foreign borrowing(FBg), or net domestic sales of government bonds(GB), and as the capital account balance may bewritten as the sum of net foreign government bor-rowing, net foreign private borrowing (FBp), and netforeign lending to the central bank (FBc), the aboveequation may be rewritten as:

FBg + FBp + FBc = AR + (ADCg + FBg + GB)+ (PI-PS).

It is obvious that external financing to the budgetcan be used to finance a larger fiscal deficit withoutrecourse to higher domestic bank credit (with its infla-tionary consequences), and that external private fi-nancing would allow greater noninflationary spend-ing by the private sector. Alternatively, for a givenfiscal deficit, external financing to the budget can beused to reduce bank credit to government and in-crease the international reserves.

Balance of payments support to the central bankalso allows for higher noninflationary governmentexpenditures since, from the following depiction ofthe operations of the banking system, it can be seenthat it allows for higher credit to the governmentfrom the banking system without an increase inmonetary expansion:

AR + ADCg = FBc + AM,

where AM equals the change in monetary liabilitiesof the banking system.

115

Appendix



BibliographyCasella, Alessandra, and Barry Eichengreen, Can Foreign

Aid Accelerate Stabilization? CEPR DiscussionPaper, No. 961 (London: Centre for Economic PolicyResearch, May 1994).

Christensen, Benedicte Vibe, The Russian Federation inTransition: External Developments, IMF OccasionalPaper, No. 111 (Washington: International MonetaryFund, February 1994).

Collins, Susan M., "Capital Flows to Developing Countries:Implications from the Economies in Transition?" Pro-ceedings of the World Bank Annual Conference on De-velopment Economics, 1992, Supplement to The WorldBank Economic Review and The World Bank Re-search Observer (Washington: The World Bank, 1992).

Eichengreen, Barry, and Marc Uzan, "The Marshall Plan:Economic Effects and Implications for Eastern Eu-rope and the Former U.S.S.R.," Economic Policy: AEuropean Forum, No. 14 (April 1992), pp. 14-75.

Fischer, Stanley, "Economic Reform in the USSR and theRole of Aid," Brookings Papers on Economic Activ-ity:2 (Washington), pp. 209-303.

International Monetary Fund, Financial Organization andOperations of the IMF, IMF Pamphlet Series, No. 45(Washington: International Monetary Fund, 4th. ed.1995).

Kuhn, Michael, Balazs Horvath, and Christopher Jarvis,Officially Supported Export Credits: Recent Develop-ments and Prospects, World Economic and FinancialSurveys (Washington: International Monetary Fund,March 1995).

Sachs, Jeffrey, "Russia's Struggle with Stabilization:Conceptual Issues and Evidence," paper presented atthe World Bank Annual Conference on Develop-ment Economics, Washington, April 28 and 29,1994.

United Nations, Economic Commission for Europe, Eco-nomic Survey of Europe in 1992-93 (New York,1993).

116


VIII Russi a and the IMF: The Politica lEconomy of Macrostabilization

Ernesto Hernandez-Cata 1

F or the past three and a half years, Russian re-formers have been struggling with the gigantic

task of stabilizing the Russian economy while trans-forming it into a modern and efficient structure ableto compete in international markets and respond tothe needs of the Russian people. They have traveled along distance. Since January 1992, they have abol-ished central planning, sharply cut the system of stateorders, decontrolled most prices, at least at the fed-eral level, unified and liberalized the foreign ex-change market, made the ruble convertible, and pri-vatized two thirds of the economy. They haveaccomplished all this in spite of the crushing burdensand the historically unprecedented problems they in-herited from the communist system: huge industrialenterprises that are unprofitable under the new struc-ture of relative prices and require extensive reorgani-zation or liquidation; an obsolete capital stock; acrumbling energy sector; armies that must be repatri-ated from distant countries and resettled at home; anda huge external debt. They have accomplished all thatdespite constant resistance from political groups op-posed to reform, including, until its dissolution in thefall of 1993, the Supreme Soviet.

Most of these impressive achievements have beenin the area of structural reform. In contrast, successin the area of macro economic stabilization has re-mained elusive. From mid-1992 to the end of 1993,consumer price inflation fluctuated around 20 per-cent a month (nearly 800 percent at an annual rate),the ruble dropped from Rub 144 to almost Rub 1,250per U.S. dollar, and capital flight may have beensomewhere around $20 billion. Monthly inflation de-clined to an average of less than 10 percent in thefirst six months of 1994 and bottomed out at less than5 percent in August. But it rose after that, and by De-cember 1994 it had reached 16 percent, by which

time the ruble had dropped to around Rub 3,500 perU.S. dollar.

Why have Russian reformers failed, so far, toadopt and implement a successful stabilization pro-gram? Is it now possible to consolidate the limitedprogress that has been made and hope that a renewedeffort will succeed?2 This paper attempts to answerthese questions.

Inflation: Fables, Tall Tales, andOld Truths

From the outset, the fight for stabilization encoun-tered several difficulties that were certainly notunique to Russia but were exacerbated by the particu-lar conditions prevailing after the breakdown of theSoviet Union. Some reflected the misconceptionsthat were to be expected in a country where eco-nomic analysis had been clouded by Marxist ideol-ogy for several decades; others were clearly political.

The first set of difficulties was of an intellectualnature. When the reforms started, most Russiansknew very little about the nature, the causes or theconsequences of inflation, or about how to deal withit. After all, prior to 1991 the most recent experiencewith high, open inflation dated to World War II andbefore that to Lenin's days, and few had any clearrecollection of what had happened then. Numerousmisconceptions plagued the debate about monetarypolicy and inflation, and there was a great deal ofconvincing to be done.

In scores of discussions with officials and stafffrom the government, the Central Bank of Russia andthe Parliament, IMF missions argued that high infla-tion was bound to damage the Russian economy inmany ways: by generating uncertainty about keyprices, including the real interest rate and the real ex-change rate, thus deterring long-term credit, invest-ment, and growth; by encouraging unproductive ac-

1This paper was first published in Problems of Post-Commu-nism, Vol. 41 (May/June 1995), pp. 20-27. Unless otherwisenoted, references to the "IMF" relate to the staff working on Rus-sia, not to the Executive Board or to management.

2This paper was completed in January 1995 before the start ofthe current economic program with Russia supported by the April1995 stand-by arrangement with the IMF. Eds.

117


VIII RUSSI A AND THE IMF

tivities aimed solely at hedging against inflation; byhurting those social groups that lack the politicalstrength to protect their real incomes against risingprices; and by contributing to a general climate of un-certainty and lack of confidence in government poli-cies, thus encouraging one-way speculation againstthe ruble and capital flight. And, of course, IMFteams often referred to the extensive statistical evi-dence suggesting that, over the medium to long term,low inflation tends to be associated with high growthamong both developing and industrial countries.3

Another source of controversy related to the causesof inflation. There was a widespread view, also withinthe central bank, that inflation in Russia resulted fromthe high degree of monopolization of the economy.But whereas monopolies can lead to an excessivelyhigh price level for certain goods, they cannot explaincontinuously rising prices. Besides, serious researchindicates that the Russian economy is not particularlymonopolistic by international standards.4 Others,claimed that inflation resulted from price liberaliza-tion. The IMF replied that a comprehensive price liber-alization could lead to a once-and-for-all jump in thegeneral price level—possibly a very large one if, as inthe case of Russia, it occurred against the backgroundof a sizable monetary overhang—but not to a contin-ued inflationary process. The Russian experienceclearly supports that view: consumer prices surged byalmost 300 percent in January 1992, when most priceswere decontrolled; but afterward, the monthly rate ofprice increase fell sharply, to 27 percent in Februaryand to 7 percent in July. Inflation began to acceleratein the summer of 1992 as monetary policy became ex-pansionary. By then, however, price decontrol was nolonger a significant factor.

So, the IMF argued that inflation does not resultfrom price liberalization, market imperfections,greedy workers, or monopolists, but that, in the end,it results from excessive growth in the money sup-ply, which in turn reflects excessive credit expansionby the central bank. We also maintained that capitalflight and the weakness of the ruble did not reflect aconspiracy by speculators at the Moscow InterbankForeign Exchange market, as a popular tale wouldhave it. To be sure, capital flight reflected in part taxevasion by wealthy individuals and enterprises. But

3See, for example, International Monetary Fund, World Eco-nomic Outlook, World Economic and Financial Surveys (Wash-ington: International Monetary Fund, May 1993); Robert E.Lucas, Jr., "On the Welfare Costs of Inflation" (Chicago: Univer-sity of Chicago, 1994); and Brian Motley, "Growth and Inflation:A Cross Country Study," Working Paper 93-11 (San Francisco:Federal Reserve Bank of San Francisco, 1993).

4See Annette Brown, Barry Ickes, and Randi Ryterman, "TheMyth of Monopoly: A New View of Industrial Structure in Rus-sia," World Bank Policy Research Working Paper, No. 1331(Washington: World Bank, August 1994).

it resulted mainly from a rational assessment of rela-tive rates of return by wealth holders in Russia: atleast through mid-1993, the rate of interest on ruble-denominated assets was much too low to offset thedepreciation of the ruble that was to be expected aslong as domestic inflation exceeded inflation abroadby a huge margin.

The implication, of course, was that credit had to betightened and interest rates raised. But many in Rus-sia—and more than a few Western "experts" as well—vehemently disagreed. Some, because they were per-suaded that higher interest rates would lead to higherprices—an interesting, but not entirely new misconcep-tion. Others, more sophisticated, disagreed becausethey believed that an anti-inflationary monetary policywould exacerbate the fall in production—a sort ofRussian version of the Phillips curve. But there is noevidence of an inverse relation between inflation andgrowth in Russia, even in the short run.5 Output hasbeen falling in Russia (as well as in other countries ofthe former Soviet Union) primarily because state ordershave been slashed, particularly for defense industries,and because large segments of the old productive struc-ture are unprofitable under the new, much freer, struc-ture of relative prices. There are other reasons, nodoubt, for the recorded fall in production: the collapseof trade with countries of the former Council for Mu-tual Economic Assistance and with other states of theformer U.S.S.R.; and the fact that the contribution ofoutput in the new private sector, which is growing veryrapidly albeit from a very small base, is not adequatelycaptured by the official statistics. But none of this hasanything to do with monetary policy being too tight.

Some did not need to be convinced that Russianeeded macroeconomic stabilization, and that stabi-lization required a much tighter monetary policy. Infact, the IMF's message on this point was very sim-ilar to the one delivered consistently and to a muchwider audience by men like Boris Fedorov, EgorGaidar, and Anatoly Chubais, who have played aleading role in the Russian government since the re-birth of the country in 1991. And while some re-main unconvinced, the debate about inflation andmacroeconomic policy has helped to persuademany, including at the Central Bank of Russia, thatthe IMF's recommendations are in the best interestof the country. Indeed, a number of officials at thecentral bank, led by Acting Chairman Tatiana Para-monova, are today among the most serious andcourageous advocates of financial discipline.

5While it is too early to test for long-term relationships, it isclear that those economies in transition that have succeeded inbringing down inflation rapidly (like the Czech Republic, Esto-nia, Latvia, and Poland) have also experienced a relatively earlyresumption of output growth.

118


Revolt Agains t Macrostabilizatio n

Revolt Against Macrostabilization:Centralized Credits , InterenterpriseArrears, Budget Subsidies, andTax Exemptions

The fundamental root of the opposition tomacrostabilization was, unfortunately, politicalmore than intellectual. Within Russia, it came fromthose who represented ailing, but still powerful,sectors and regions such as the Northern Territories,agriculture and agro-industry, and, to a lesser ex-tent, the military-industrial complex. Outside Rus-sia, it came from many of the other states of the for-mer Soviet Union, where newly created centralbanks—supported by their governments and some-times by accomplices in Russia—were eager to ex-pand domestic credit, largely by borrowing from theCentral Bank of Russia. This indirect source of in-flationary pressures apparently disappeared with thecollapse of the ruble area in late 1993 (Box 1).

Within Russia, those who lobbied for special in-terest groups knew that a tighter monetary policy—particularly one that operated indirectly through fi-nancial markets and not through the centralizeddistribution of credit—would mean that their shareof the pie would be considerably diminished. Sothey fought to retain the existing system in which thebulk of financial resources was administratively di-rected to them, at heavily subsidized interest rates—often well below the central bank refinance rate,which until the summer of 1993 was itself wellbelow market rates—and with as much flexibility aspossible regarding the repayment of principal. As toother borrowers, including many in the incipient pri-vate sector, they would, if they were lucky, obtainsome credit from commercial banks at market inter-est rates—otherwise they would simply be rationedout. So the system turned into a process of politicalcompetition for credit, a political scramble in whicheach special interest group would use its politicalclout with the government, with the Supreme Soviet,the Credit Policy Commission, or even the Presi-dent, in a race to maximize its share of directed cred-its and stay ahead of inflation.

The IMF, on the contrary, fought to eliminate thecentralization and the subsidization of credit, and itdid so with the support of then-Finance MinisterFedorov. The existing system needed to be over-hauled for both macro- and microeconomic reasons:because the process of political competition it in-volved made it extremely difficult to limit aggregatecredit expansion to anything consistent with the goalof reducing inflation; and because in the absence ofrestraint on monetary expansion, and as long as mar-ket conditions did not play a proper role in the allo-cation of financial resources, Russian enterprises

Box 1. The IMF and the Ruble Area

The evolution of the IMF's attitude toward theruble area, which could be the subject of a separatepaper, has often been misrepresented. To make along and complicated story short, the Fund staff didnot try to "save" the ruble area, as is sometimes al-leged. Rather, it tried to eliminate the inflationarybias built into the ruble area by proposing at the in-terstate conference of Tashkent, in May 1992, a setof rules for a coordinated monetary policy. The pro-posal failed, largely because it was torpedoed by theRussian delegation. In hindsight, however, it is fairto say it might not have worked had it been ap-proved because several countries in the area proba-bly would not have followed its relatively strictrules. Convinced that the ruble area could not be re-formed, the IMF sought to persuade countries that arapid choice had to be made between (1) remainingin the ruble zone with a single monetary authority;and (2) issuing a separate currency. Russia's deci-sion on July 1, 1992 to discontinue the automatic ex-tension of central bank credit to other central banksled to the de facto fragmentation of the ruble areaand to the appearance of separate deposit currenciesin other countries of the former Soviet Union. Thequest for a clear-cut solution remained elusive, how-ever, as supporters and adversaries of the ruble areacontinued to fight, even within Russia. The CentralBank of Russia's sudden withdrawal of pre-1993ruble notes in July 1993 and the collapse of negotia-tions on a monetary union between Russia andKazakstan in November of that year sealed the fateof the old ruble area. Subsequent attempts to estab-lish a currency union between Russia and Belaruswere unsuccessful.

would continue to operate without the discipline ofhard budget constraints, and one of the key objec-tives of reforms—to encourage a competitive andprofitable enterprise sector—would be jeopardized.

For the same reasons, IMF teams opposed any planto monetize interenterprise arrears. The buildup of ar-rears in the first half of 1992 was, in part, a revolt ofstate enterprise managers against the tightening ofmonetary policy sponsored by Acting Prime MinisterGaidar and carried out, albeit not always in good co-operation with the government, by then-central bankChairman Georgy Matiukhin.6 But in July of 1992,

6To some extent, the buildup of interenterprise receivables,which is sometimes confused with the accumulation of arrears,has represented the normal growth of interenterprise credit. Thisis a welcome development that calls for government action in theform of market supervision and regulation, but not for credit ex-pansion (see Section IV).

119



Matiukhin was forced by the Supreme Soviet to re-sign—perhaps because he was effectively resistingexorbitant demands for credit—and was replaced byViktor Gerashchenko. A few months later the centralbank cleared the net stock of arrears by extendingcredit to the net creditor enterprises. The action tookmany by surprise, including officials at the GKI (thegovernment privatization agency) who had elabo-rated specific plans to deal with the problem withoutresorting to credit expansion. In 1993, interenterprisearrears started to grow again, reaching 7 percent ofGDP by June 1994 and prompting the creation of aGovernment Commission under First Deputy PrimeMinister Oleg Soskovets to explore ways to resolvethe problem. The IMF continues to believe that a sec-ond bailout by the central bank would be a disaster.Not only would it boost the money supply and raiseinflationary expectations, but it would give enterprisemanagers a clear signal that in the future they can re-spond to a tightening of credit conditions simply bynot paying their bills.

One of the key elements of IMF-supported pro-grams, the need to reduce Russia's huge budgetdeficit, was fully accepted by Acting Prime MinisterGaidar and then by Fedorov who became FinanceMinister in December 1992. However, in his efforts totackle the fiscal problem, Fedorov quickly met withthe opposition of sectoral lobbyists who wished tomaintain the generous transfers and subsidies they re-ceived from the budget. A higher budget deficit wouldhave to be financed largely by credit expansion, thusfueling inflation and eroding the real incomes of theentire population, but this did not bother them; as longas they received a disproportionate share of the bud-getary pie, they thought they would come out ahead.The IMF, of course, argued for cuts in budgetarytransfers and subsidies, not only to reduce the budgetdeficit and the government's need to borrow from thecentral bank and thus bring down inflation, and notonly to do away with the misallocation of resourcesimplied by those subsidies, but also to make room foran adequate social safety net and in particular for awell-funded unemployment insurance system. Hereagain macrostabilization and microefficiency objec-tives pointed in the same direction, and so did socialconsiderations.

But the budget deficit was large, and cutting sub-sidies would not be enough.7 A strategy to achieve a

7The deficit of the federal government was 12 1/2 percent ofGDP in 1992. It fell to 7 1/2 percent in 1993 before rising to some-where between 11 percent and 12 percent of GDP in 1994. Thesenumbers exclude unbudgeted import subsidies financed by for-eign export credits, which amounted to 12 percent and 2 1/2 percentof GDP, respectively, in 1992 and 1993.

sufficient reduction in the deficit had to involveacross-the-board restraint on expenditure and higherrevenue, notably from the value-added tax (VAT)and from the energy sector; this would require im-provements in tax administration and, at least forsome time, higher tax rates. None of this was politi-cally easy, of course. At various times in 1992-93,the Supreme Soviet opposed increased rates on theVAT and reduced the effective yield of various taxes,including the VAT and import duties, by granting alarge number of exemptions. Energy companiesclaimed that they could not pay existing taxes (letalone higher taxes) because Russian households andenterprises were not paying their energy bills andbecause continued shipments of natural gas toUkraine and Belarus had resulted in a huge accumu-lation of arrears to Gazprom, the Russian natural gasgiant. All this was true, but it did not prevent Rus-sian energy companies from holding large bank de-posits—in rubles as well as in foreign currency—that would have allowed them to pay their taxes, hadthe will been there and the threat of enforcementbeen credible.

In the end, efforts to increase revenue failed. Indeed,federal revenue fell in relation to GDP by almost 3 per-centage points from 1992 to 1993. The proliferation oftax exemptions granted to sectoral lobbyists was amajor cause of this deterioration but other factors alsoplayed a role, including the shift in production toward aservices-intensive private sector, which often escapedtaxation, and disputes between federal and regionalgovernments over revenue sharing. In particular, theautonomous republics of Tatarstan, Bashkortostan, andChechnya unilaterally declared that all taxes would becollected by the regional authorities and that revenuesharing would be the subject of subsequent negotia-tions with the federal authorities.

Changing the Structure of DeficitFinancing as an Alternative toFiscal Adjustment: A Fair y Tale

As if domestic political opposition to budget cutshad not been enough, the IMF drew some unex-pected fire from Professor Jeffrey Sachs. On variousoccasions Professor Sachs argued that the IMF waswrong in asking for deep budget cuts; this, he said,was not necessary to reduce inflation.8 The IMF hadfailed to understand, and therefore had failed to ex-plain to the Russians, that monetary financing of thebudget deficit (and therefore inflation) could be re-

8See, for example, Professor Sachs's comments in Testimonyto the Committee on Banking, Housing and Urban Affairs of theU.S. Senate, February 5, 1994.

120


Changing the Structure of Deficit Financin g

duced quite simply by increasing the proportion ofthe deficit financed by borrowing from abroad or byissuing domestic interest-bearing debt. A remarkablysimple proposition that lead to an obvious question:why would any country that has benefited from suchgood advice continue to suffer from high inflation?Perhaps because things are not quite that simple.

First, foreign financing does not grow on trees. Tobe sure, and contrary to Professor Sachs's allegations,Russia did receive considerable external financing—altogether around $60 billion in the two-year period1992-93.9 It is true that most of it was in the form ofdebt rescheduling or tied export credits, and it can beargued that these did not provide cash financing to thebudget.10 It would certainly have been much better ifexternal financing had been predominantly in the formof long-term credits to the government or, even better,in the form of grants. But large-scale financing of thistype is not available.11 One can hope for a world inwhich parliaments and public opinion in the wealthycountries recognize that the international communityis facing a challenge of historical significance—that ofensuring the transition from a totalitarian, centrallyplanned, and bellicose Soviet state to a new, peaceful,democratic, and free Russia—and that this extraordi-nary challenge requires a measure of sacrifice in theform of a temporary increase in taxes to finance tem-porary assistance to Russia and the other countries inthe former U.S.S.R. But that, unfortunately, is not theworld in which we live today. Foreign financing on fa-vorable terms is, and will continue to be, limited.

There is another unfortunate complication that Pro-fessor Sachs chose to ignore in his analysis. Externaldebt—unless it is entirely in grant form—must be ser-viced. And, of course, this applies also to domesticdebt. It is fine to say that bond financing can substi-tute for monetary financing; in fact, the IMF has en-couraged the Russian authorities to create a marketfor treasury bills and provided technical assistance tospeed up the process. But the IMF recognizes, asmany Russians do, that there are risks in going too farand too fast in this direction. If, in present circum-stances, the Russian government attempted to financea large share of its budget deficit by issuing treasurybills, the public would probably insist on very shortmaturities and very high real interest rates, and the re-

9See Section VII.10This is probably correct in the case of debt rescheduling, as

cash payments by Russia probably would not have been muchlarger in the absence of the agreements. As for export credits,they should have provided cash financing to the budget had theRussian authorities collected counterpart funds from domestic en-terprises. Instead, the foreign export credits were used predomi-nantly to finance domestic import subsidies.

11A notable exception has been the grants provided by Ger-many to finance the withdrawal of Russian troops from Germanterritory.

suiting debt burden could well mean a higher deficitin subsequent years, even in percent of GDP.12 Thiswould be particularly the case given the less-than-per-fect record of Russian (and Soviet) governments inmeeting their financial obligations, of which the 1993demonetization of "old rubles" and continued arrearson internal commodity debt are only two examples.

The more general point here is that Russia's fiscalproblem is a structural problem that must be re-solved over time. The deficit cannot be eliminated inone year (this would be socially and politically im-possible), but deficit reduction cannot be postponedindefinitely, it must begin now. This means that a se-rious fiscal program must involve from the outset acombination of financing and adjustment: there mustbe some domestic financing—hopefully with a ris-ing share of debt financing and a diminishing shareof monetary financing over time—and there shouldbe some external financing—if possible on betterterms than has been seen so far. But the program willnot be credible unless it involves a substantial doseof fiscal adjustment, so that structural deficits are re-duced year after year and there is a reasonable ex-pectation that Russia will be able to avoid excessivereliance on both indebtedness and inflationary fi-nancing in the foreseeable future.

Incidentally, some of the points made above alsoapply to financing by the international organizations,including the IMF, which contributed a total of $4billion in 1992-94 in support of adjustment pro-grams that did not work very well. This included a$1 billion first credit tranche arrangement in 1992,and two drawings of $1 1/2 billion each in 1993 and1994, respectively, under the systemic transforma-tion facility (STF)—a facility that was created espe-cially to assist economic reform in countries likeRussia. And the Fund's contribution could increasesubstantially: negotiations are currently under wayon a program that could be supported by a full stand-by arrangement under which Russia could receiveadditional credits of several billion dollars.13

But the IMF's money belongs to the internationalcommunity. It is there to help member countries dealwith temporary balance of payments difficulties,and, therefore, it can only be made available subjectto certain conditions. The rationale for these condi-tions is well known: to provide reasonable assurance

12The pitfalls of excessive reliance on interest-bearing govern-ment debt as a tool of deficit financing in a context of high infla-tion are illustrated by the Brazilian case. See, for example, Mar-cio G. P. Garcia "Avoiding Some of the Costs of Iinflation andCrawling Towards Hyperinflation: The Case of the Brazilian Do-mestic Currency Substitute" (Rio de Janeiro: Pontifical CatholicUniversity of Rio de Janeiro, April 1994).

13Such an arrangement was approved by the Executive Boardon April 11, 1995. Eds.

121



that existing imbalances will be corrected so that thecountry will be in a position to repay its liabilities tothe Fund and, more important, so that its economicsituation will show a lasting improvement. Here,too, financing can complement adjustment; but thereis no substitute for fiscal adjustment.

The May 199 3 Progra mThe difficulties that have been faced in the strug-

gle for macroeconomic stabilization in Russia can beillustrated with reference to the history of two recentIMF programs. In May 1993, the representatives ofthe Russian Government and the central bank,headed by Finance Minister Fedorov, agreed with anIMF staff team on a program that was subsequentlyapproved by Prime Minister Chernomyrdin andCentral Bank of Russia's Chairman Gerashchenko inthe form of a Statement of Economic Policies of theRussian Federation. On July 1, 1993, the programwas approved by the Fund's Executive Board andsupported by a credit of $1.5 billion under the sys-temic transformation facility. Two aspects of thisprogram are worth noting.

First, the program specified that the $1.5 billion tobe obtained from the Fund would not have to beadded to the stock of Russia's official internationalreserves but would be available to provide additionalcredit to the economy (including to the budget) with-out relying on monetary expansion. In other words,IMF financing was to facilitate a gradual adjustmentin the fiscal position while beginning the process ofreducing credit growth and inflation. (To some ex-tent, the program was doing what Professor Sachsaccused it of not having done.) Logically, financingwas to be coupled with adjustment: it was expectedthat the deficit of the enlarged government—includ-ing federal and local budgets, extrabudgetary fundsand unbudgeted import subsidies—would fall from11 percent of GDP in the second quarter of 1993 to8 1/2 percent in the fourth quarter, which would havemade it possible for inflation to drop from a monthlyaverage of 17 percent in the second quarter to 8 per-cent in the fourth quarter. This, of course, was still150 percent at an annual rate, but given the failuresof the past, it seemed preferable to make sustained,albeit undramatic progress.

Second, the financial program envisaged not onlya decline in the rate of growth of money and credit,but also a number of measures aimed at breaking thesystem of politically motivated, administratively di-rected and heavily subsidized credits, and replacingit by a system in which financial markets would playan increasing role and a restrained monetary policywould have a chance of success. To achieve that ob-jective, the program included five key elements: (1)

upward pressure on market-determined interest ratesthrough a reduction in the growth of net domesticcredit and base money; (2) a rule tying the centralbank lending rate to the market-determined,Moscow interbank rate; (3) a commitment by thecentral bank not to extend credit to enterprises (di-rectly or through commercial banks) at interest ratesbelow the central bank's normal lending rate; (4) acommitment to limit budgetary interest rate sub-sides; and (5) a commitment of both government andcentral bank not to repeat the bailout of interenter-prise arrears that had occurred in 1992.

The program went on to a good start. In June-July1993, the government liberalized the price of coal andtook several measures to reduce the budget deficit, in-cluding cuts in import subsidies and in subsidies tograin producers. The Central Bank of Russia, stillunder the chairmanship of Gerashchenko, increasedits lending rate from 100 percent in May 1993 to 170percent on July 1 and then honored its commitmentunder the program by allowing its lending rate to risealong with the interbank rate.14 The results came verysoon: the ruble, which had been losing value almostcontinuously against the U.S. dollar for about oneyear, appreciated by more than 10 percent from mid-June to the end of July 1993, and the central bankfound itself intervening in the foreign exchange mar-ket to moderate the appreciation of the ruble, to thetune of $1.2 billion. The Russian economy appearedto be on the way to stabilization.

Unfortunately, soon after that the economic pro-gram went off track, and its monetary and fiscal tar-gets for the third and fourth quarters of 1993 wereexceeded by wide margins. Monthly inflation aver-aged around 20 percent in the second half of 1993,and by the end of the year the ruble had lost all theground it had gained in June-July, in spite of sub-stantial intervention by the central bank.

What went wrong? The answer is simple: the pro-gram was working. But once again, tight financialpolicies met with political resistance and sabotagefrom those special interest groups that were deprivedof the subsidies and the cheap credits to which theyhad become accustomed and to which they thoughtthey were entitled. These groups pressured and lob-bied through the Supreme Soviet and within the gov-ernment; and in the end, in spite of staunch resis-tance by Fedorov and his associates, they obtainedthe support they were seeking—not all of it, butenough to derail the program.

A few examples by way of illustration include thefollowing. First, grain subsidies were cut as had

14The central bank finance rate peaked at 210 percent on Octo-ber 15, 1993 and remained at that level until April 29, 1994. Itwas then allowed to drop in stages following, albeit with a sub-stantial lag, the fall in the interbank rate.

122


The March 199 4 Program

been expected, but this was more than offset by alarge increase in budgetary spending on agriculture,including a rise in the procurement price of grainover and above what could be justified by marketconditions, and large subsidies for cattle raising.Second, the scheduled reimbursement to the federalgovernment of a Rub 1 trillion loan (roughly 1 per-cent of GDP in those days) by Rosskhleboprodukt,the state grain procurement agency, failed to materi-alize. Third, in a courageous move, the governmentliberalized the price of coal, but this was followed bya substantial rise in government support to the coalsector. Fourth, the Supreme Soviet refused to in-crease excise taxes on energy and to raise the VATby 7 percentage points, as had been proposed by thegovernment. Fifth, under pressure, the governmentand the central bank extended large-scale, unsched-uled credits to the agricultural sector and the North-ern Territories, breaking the commitment under theprogram to phase out the direct allocation of subsi-dized credits to specific sectors and regions.

This is the story of the May 1993 program. Itwent off track because it was not adhered to; and itwas not adhered to because major political forceswithin the country refused to live by its provisionsand because the government, in spite of the effortsof Fedorov and his associates, was not able to pro-tect the program against those who wanted to derailit. In the last months of 1993, Fedorov used the onlyweapon that remained in his arsenal: he simply re-fused to pay. The policy of aggressive sequestrationcontinued in the first half of 1994 and, coupled witha restrained monetary policy by the central bank, ithelped to achieve a significant, albeit temporary re-duction in inflation. But the other result of seques-tration was a buildup in government arrears, someof which would have to be repaid, thus raising thebase on which sequestration would have to be ap-plied in the future.

The March 199 4 ProgramFedorov resigned on January 16, 1994, following

by a few weeks the resignation of Gaidar as DeputyPrime Minister and Minister of the Economy.Chubais, another key figure among the reformers,remained in the Cabinet; but he could not be ex-pected to play a leading role in the stabilization ef-fort, as he would have to devote all his energies tothe continuation of the privatization plan that hisGKI had implemented so successfully. But many ofFedorov's former associates in the Ministry of Fi-nance remained in place, including his formerDeputy, Sergei Dubinin, who was appointed ActingFinance Minister. In January-March of 1994, aRussian negotiating team headed by Dubinin and in-

cluding senior Finance Ministry and central bank of-ficials worked out with the IMF an economic pro-gram on which final agreement was reached onMarch 20 between Prime Minister Chernomyrdinand the Managing Director of the IMF. The programwas approved by the Fund's Executive Board onApril 20 and supported by a second credit of $1.5billion under the STF.

The major goals of the program were to safeguardthe achievements of reform in the areas of price andexchange market liberalization, extend privatization,make some progress in liberalizing external trade,and reduce the monthly rate of inflation to 7 percentat the end of 1994, on the way to further reductionsin 1995. The fiscal program included revenue mea-sures expected to yield 2 1/2 percent of GDP in 1994and a tight expenditure plan. The monetary programenvisaged a gradual reduction in the growth ofmoney and credit, a gradual increase in the share ofcentral bank credit extended through credit auctionsand other market-related mechanisms, and a gradualphasing out of directed credits. For all the fablesabout "shock therapy" and "big bang," gradualismwas the key operating concept—not out of design,but because a more audacious approach was not po-litically acceptable.

In the first quarter of 1994, the government man-aged to keep its borrowing from the central bankunder control in spite of a collapse in revenue,mainly through aggressive sequestration of expendi-ture. Although government borrowing picked up inthe second quarter, by the end of June all the fiscaland credit targets of the STF program had been met;official international reserves (including holdings offoreign exchange and gold by both the central bankand the Ministry of Finance) were above the mini-mum levels specified in the program; and themonthly rate of inflation fell to 6 percent in June,compared with 18 percent in January and an STFprogram projection of 10 percent. The budget for1994 was finally approved by both the Duma and theFederation Council in June 1994. During the preced-ing parliamentary debate the government managedto deflect pressure for a large increase in defensespending, but it did have to accept a rise in expendi-ture, particularly in the form of higher transfers toagriculture.

The trouble, as had been the case in previousyears, came in the summer. Net central bank creditto the government surged in the third quarter of 1994as government revenue continued to drop in relationto GDP and subsidies to the agricultural sector, theFar North, and other customary recipients of govern-ment handouts rose sharply while the government'sability to use sequestration diminished. (Given thefailure to obtain Duma approval for most of the rev-enue measures specified in the second STF program,

123



sequestration had become the only instrument ofbudgetary control.) Yet, inflation remained relativelysubdued in the third quarter: the monthly rate of in-crease in consumer prices fell to 4 1/2 percent in Au-gust, the lowest monthly rate registered in the period1992-94, before rising to just under 8 percent inSeptember—still below the STF program's projec-tion for that month.

But this combination of rapid growth in credit tothe government and relatively moderate inflation wasunsustainable, for two reasons. First, the average lagof three to four months between money growth andinflation in Russia meant that the acceleration ofmoney and credit in the spring of 1994 would showup eventually in the inflation numbers—as it did inthe fourth quarter. Second, beginning in July 1994the authorities attempted to offset the impact on themonetary base and inflation of a surge in net credit tothe government by running down official interna-tional reserves. But the game could not last. From theend of June to the end of September 1994, interna-tional reserves dropped by almost $4 billion, and par-ticipants in the foreign exchange market, recognizingthat the underlying fiscal position was out of controland that international reserves were finite, took aimat the ruble. On October 11, 1994, the ruble tumbledin the Moscow interbank market by more than 20percent against the U.S. dollar.

"Black Tuesday," as October 11, 1994 was subse-quently referred to, was the first full-fledged finan-cial crisis in the history of post-communist Russia. Itreflected a rational reaction of market participants,by now well informed, quick to move, and aware ofthe lagged but predictable effects of monetary ex-pansion on inflation and exchange rates. It was a ra-tional reaction to bad macroeconomic policies, andparticularly to an unsustainable fiscal deficit, and theappropriate response should have been to correctthose policies. Instead, and in spite of the fact thatthe ruble recovered its previous losses on October12-13, Acting Finance Minister Dubinin and CentralBank of Russia's Chairman Gerashchenko werefired; and federal agents were sent to the centralbank and the Moscow foreign exchange market in afutile attempt to identify the authors of an alleged"financial coup."

In the fourth quarter, the central bank limitedcredit expansion to the government. The Ministry of

Finance reacted by resisting strong expenditure pres-sures, but also by stepping up issues of governmentsecurities including, despite IMF objections, securi-ties bearing interest rates well below market levels.The central bank also reduced net lending to com-mercial banks, which led to a sharp increase in inter-est rates. Nevertheless, inflation continued to rise,and in December 1994 it reached a monthly rate of16 percent, more than double the STF target. By theend of 1994, the fiscal targets under the STF pro-gram had been exceeded by wide margins and thestock of international reserves had dropped wellbelow the program floor.

While the sequence of events had been different,the March 1994 program seemingly had failed ascompletely as its predecessor. The central bank hadkept credit expansion to the commercial banks withinthe program targets and had managed its lending rateprudently, keeping it above the Moscow interbankrate—sometimes by a wide margin—and signifi-cantly above the rate of inflation. But a tight monetarypolicy could not undo the damage caused by a federalfiscal deficit in excess of 10 percent of GDP.

The rise in inflation, the accumulation of govern-ment arrears, and the exchange market crisis of Oc-tober 11 all gave ammunition to those who distrustedmarket mechanisms and wished to reinstate adminis-trative controls. But, just when things looked thedarkest, a number of key appointments strengthenedthe position of those in the government and the cen-tral bank that continued to fight for reform and stabi-lization. Chubais was appointed Deputy Prime Min-ister in charge of macroeconomic policy; YevgeniYasin became Minister of the Economy; and Para-monova was made Acting Chairman of the CentralBank of Russia. Negotiations with the IMF resumed,this time on a program that could be supported by astand-by credit of up to $6 billion, and the govern-ment submitted to the Duma a budget for 1995 thatenvisaged substantial cuts in subsidies and in lend-ing to agriculture and the coal sector, and a reductionin federal transfers to regional governments and theNorthern Territories.15 The struggle for macroeco-nomic stabilization continued.

15These negotiations culminated in a stand-by arrangementwith the IMF in April 1995. Eds.

124


Occasional Paper s

Recent Occasional Papers of the Internationa l Monetar y Fund133. Policy Experiences and Issues in the Baltics, Russia, and Other Countries of the Former Soviet Union,

edited by Daniel A. Citrin and Ashok K. Lahiri. 1995.

132. Financial Fragilities in Latin America: The 1980s and 1990s, by Liliana Rojas-Suarez and Steven R.Weisbrod. 1995.

131. Capital Account Convertibility: Review of Experience and Implications for IMF Policies, by staff teamsheaded by Peter J. Quirk and Owen Evans. 1995.

130. Challenges to the Swedish Welfare State, by Desmond Lachman, Adam Bennett, John H. Green, RobertHagemann, and Ramana Ramaswamy. 1995.

129. IMF Conditionality: Experience Under Stand-By and Extended Arrangements. Part II: Background Pa-pers. Susan Schadler, Editor, with Adam Bennett, Maria Carkovic, Louis Dicks-Mireaux, MauroMecagni, James H.J. Morsink, and Miguel A. Savastano. 1995.

128. IMF Conditionality: Experience Under Stand-By and Extended Arrangements. Part I: Key Issues andFindings, by Susan Schadler, Adam Bennett, Maria Carkovic, Louis Dicks-Mireaux, Mauro Mecagni,James H.J. Morsink, and Miguel A. Savastano. 1995.

127. Road Maps of the Transition: The Baltics, the Czech Republic, Hungary, and Russia, by Biswajit Banerjee,Vincent Koen, Thomas Krueger, Mark S. Lutz, Michael Marrese, and Tapio O. Saavalainen. 1995.

126. The Adoption of Indirect Instruments of Monetary Policy, by a Staff Team headed by William E. Alexan-der, Tomas J.T. Balino, and Charles Enoch and comprising Francesco Caramazza, George Iden, DavidMarston, Johannes Mueller, Ceyla Pazarbasioglu, Marc Quintyn, Matthew Saal, and Gabriel Sensen-brenner. 1995.

125. United Germany: The First Five Years—Performance and Policy Issues, by Robert Corker, Robert A.Feldman, Karl Habermeier, Hari Vittas, and Tessa van der Willigen. 1995.

124. Saving Behavior and the Asset Price "Bubble" in Japan: Analytical Studies, edited by Ulrich Baumgart-ner and Guy Meredith. 1995.

123. Comprehensive Tax Reform: The Colombian Experience, edited by Parthasarathi Shome. 1995.

122. Capital Flows in the APEC Region, edited by Mohsin S. Khan and Carmen M. Reinhart. 1995.

121. Uganda: Adjustment with Growth, 1987-94, by Robert L. Sharer, Hema R. De Zoysa, and Calvin A.McDonald. 1995.

120. Economic Dislocation and Recovery in Lebanon, by Sena Eken, Paul Cashin, S. Nuri Erbas, JoseMartelino, and Adnan Mazarei. 1995.

119. Singapore: A Case Study in Rapid Development, edited by Kenneth Bercuson with a staff team compris-ing Robert G. Carling, Aasim M. Husain, Thomas Rumbaugh, and Rachel van Elkan. 1995.

118. Sub-Saharan Africa: Growth, Savings, and Investment, by Michael T. Hadjimichael, Dhaneshwar Ghura,Martin Mlihleisen, Roger Nord, and E. Murat Ucer. 1995.

117. Resilience and Growth Through Sustained Adjustment: The Moroccan Experience, by Saleh M. Nsouli, SenaEken, Klaus Enders, Van-Can Thai, Jorg Decressin, and Filippo Cartiglia, with Janet Bungay. 1995.

116. Improving the International Monetary System: Constraints and Possibilities, by Michael Mussa, MorrisGoldstein, Peter B. Clark, Donald J. Mathieson, and Tamim Bayoumi. 1994.

115. Exchange Rates and Economic Fundamentals: A Framework for Analysis, by Peter B. Clark, LeonardoBartolini, Tamim Bayoumi, and Steven Symansky. 1994.

114. Economic Reform in China: A New Phase, by Wanda Tseng, Hoe Ee Khor, Kalpana Kochhar, DubravkoMihaljek, and David Burton. 1994.

113. Poland: The Path to a Market Economy, by Liam P. Ebrill, Ajai Chopra, Charalambos Christofides, PaulMylonas, Inci Otker, and Gerd Schwartz. 1994.

112. The Behavior of Non-Oil Commodity Prices, by Eduardo Borensztein, Mohsin S. Khan, Carmen M.Reinhart, and Peter Wickham. 1994.

111. The Russian Federation in Transition: External Developments, by Benedicte Vibe Christensen. 1994.

125


OCCASIONAL PAPER S

110. Limiting Central Bank Credit to the Government: Theory and Practice, by Carlo Cottarelli. 1993.

109. The Path to Convertibility and Growth: The Tunisian Experience, by Saleh M. Nsouli, Sena Eken, PaulDuran, Gerwin Bell, and Ziihtu Yiicelik. 1993.

108. Recent Experiences with Surges in Capital Inflows, by Susan Schadler, Maria Carkovic, Adam Bennett,and Robert Kahn. 1993.

107. China at the Threshold of a Market Economy, by Michael W. Bell, Hoe Ee Khor, and Kalpana Kochharwith Jun Ma, Simon N'guiamba, and Rajiv Lall. 1993.

106. Economic Adjustment in Low-Income Countries: Experience Under the Enhanced Structural AdjustmentFacility, by Susan Schadler, Franek Rozwadowski, Siddharth Tiwari, and David O. Robinson. 1993.

105. The Structure and Operation of the World Gold Market, by Gary O'Callaghan. 1993.

104. Price Liberalization in Russia: Behavior of Prices, Household Incomes, and Consumption During theFirst Year, by Vincent Koen and Steven Phillips. 1993.

103. Liberalization of the Capital Account: Experiences and Issues, by Donald J. Mathieson and Liliana Rojas-Suarez. 1993.

102. Financial Sector Reforms and Exchange Arrangements in Eastern Europe. Part I: Financial Markets andIntermediation, by Guillermo A. Calvo and Manmohan S. Kumar. Part II: Exchange Arrangements ofPreviously Centrally Planned Economies, by Eduardo Borensztein and Paul R. Masson. 1993.

101. Spain: Converging with the European Community, by Michel Galy, Gonzalo Pastor, and Thierry Pujol.1993.

100. The Gambia: Economic Adjustment in a Small Open Economy, by Michael T. Hadjimichael, ThomasRumbaugh, and Eric Verreydt. 1992.

99. Mexico: The Strategy to Achieve Sustained Economic Growth, edited by Claudio Loser and Eliot Kalter. 1992.

98. Albania: From Isolation Toward Reform, by Mario I. Blejer, Mauro Mecagni, Ratna Sahay, RichardHides, Barry Johnston, Piroska Nagy, and Roy Pepper. 1992.

97. Rules and Discretion in International Economic Policy, by Manuel Guitian. 1992.

96. Policy Issues in the Evolving International Monetary System, by Morris Goldstein, Peter Isard, Paul R.Masson, and Mark P. Taylor. 1992.

95. The Fiscal Dimensions of Adjustment in Low-Income Countries, by Karim Nashashibi, Sanjeev Gupta,Claire Liuksila, Henri Lorie, and Walter Mahler. 1992.

94. Tax Harmonization in the European Community: Policy Issues and Analysis, edited by George Kopits. 1992.

93. Regional Trade Arrangements, by Augusto de la Torre and Margaret R. Kelly. 1992.

92. Stabilization and Structural Reform in the Czech and Slovak Federal Republic: First Stage, by Bijan B.Aghevli, Eduardo Borensztein, and Tessa van der Willigen. 1992.

91. Economic Policies for a New South Africa, edited by Desmond Lachman and Kenneth Bercuson with astaff team comprising Daudi Ballali, Robert Corker, Charalambos Christofides, and James Wein. 1992.

90. The Internationalization of Currencies: An Appraisal of the Japanese Yen, by George S. Tavlas andYuzuru Ozeki. 1992.

89. The Romanian Economic Reform Program, by Dimitri G. Demekas and Mohsin S. Khan. 1991.

88. Value-Added Tax: Administrative and Policy Issues, edited by Alan A. Tait. 1991.

87. Financial Assistance from Arab Countries and Arab Regional Institutions, by Pierre van den Boogaerde. 1991.

86. Ghana: Adjustment and Growth, 1983-91, by Ishan Kapur, Michael T. Hadjimichael, Paul Hilbers, JeraldSchiff, and Philippe Szymczak. 1991.

85. Thailand: Adjusting to Success—Current Policy Issues, by David Robinson, Yangho Byeon, and RanjitTeja with Wanda Tseng. 1991.

84. Financial Liberalization, Money Demand, and Monetary Policy in Asian Countries, by Wanda Tseng andRobert Corker. 1991.

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

126


05300-9781557755186-en-book.pdf - imf elibrary

Documents