OCCASIONAL PAPER - IMF eLibrary · Summary of Good Practices in Institutional Transparency6 2. Summary of Good Practices in Accounting Transparency8 3. Summary of Good Practices in

O C C A S I O N A L PA P E R 158

Transparency inGovernment Operations

George Kopits and Jon Craig

INTERNATIONAL MONETARY FUND

Washington DC

January 1998

©International Monetary Fund. Not for Redistribution

© 1998 International Monetary Fund

Cataloging-in-Publication Data

Kopits, George.Transparency in government operations / George Kopits and Jon

Craig.

p. cm.—(Occasional paper, ISSN 0251-6365 ; 158)

“February 1998.”Includes bibliographical references.

ISBN 1-55775-697-X

1. Finance, Public—Accounting. 2. Fiscal policy. I. Craig, J.D. (JonD.) II. International Monetary Fund. III. Title. IV. Series: Occasionalpaper (International Monetary Fund) ; no. 158.HJ9733.K66 1998

336.3—dc21 97-51353CIP

Price: US$18.00(US$15.00 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

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

recycled paper


http://www.imf.org

Preface v

I Introduction 1

Conceptual Setting 1Arguments For and Against Fiscal Transparency 2

II Issues in Fiscal Transparency 5

Institutional Framework 5Public Accounts 7Indicators and Projections 9

III Role of the IMF in Promoting Fiscal Transparency 11

Overall Trends and Priorities 11Role of the IMF 11

IV Summary and Conclusions 13

Boxes

Section

II 1. Summary of Good Practices in Institutional Transparency 62. Summary of Good Practices in Accounting Transparency 83. Summary of Good Practices in Transparency of Indicators

and Projections 9

Appendices

I Transparency in Institutions and Behavior 15

II Transparency in Public Accounts 21

III Transparency in Indicators and Projections 27

IV Selected Country Experiences 33

References 38

Contents

iii


CONTENTS

iv

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., 1994–95 or January–June) to indicate the years ormonths covered, including the beginning and ending years or months;

/ between years (e.g., 1994/95) 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.


This paper was prepared by George Kopits, Assistant Director, and Jon Craig, Se-nior Economist, Fiscal Affairs Department. They wish to acknowledge contributionsby Hugh Young and other members of the department’s Fiscal Analysis Division.Also, many colleagues in different departments reviewed the factual accuracy andreasoning at various stages from inception to completion. In particular, the paper hasbenefited from comments and encouragement from Stanley Fischer, Peter Heller,Flemming Larsen, Thomas McLoughlin, A. Premchand, Vito Tanzi, and Teresa Ter-Minassian. The authors are indebted to Diana Ellyn and Beulah David for secretarialassistance, to Diane Cross for editorial support, to Elisa Diehl of the External Rela-tions Department for editing the final version and coordinating production, and to Alicia Etchebarne-Bourdin for composition.

An earlier draft of the paper was discussed by the Executive Board in October1997. The present version incorporates the suggestions received on that occasion.However, the opinions expressed are those of the authors and do not necessarily re-flect the views of the Executive Directors, the management, or the staff of the Interna-tional Monetary Fund.

Preface

v


This page intentionally left blank


T ransparency in government operations is widelyregarded as an important precondition for

macroeconomic fiscal sustainability, good gover-nance, and overall fiscal rectitude. Notably, the In-terim Committee, at its April and September 1996meetings, stressed the need for greater fiscal trans-parency. Specifically, in the Declaration on Partner-ship for Sustainable Global Growth, the Committeestated that “it is essential to enhance the transparencyof fiscal policy by persevering with efforts to reduceoff-budget transactions and quasi-fiscal deficits”(International Monetary Fund, 1996c, p. xii).Prompted by these concerns, this paper represents afirst attempt to address many of the aspects of trans-parency in government operations. It provides anoverview of major issues in fiscal transparency andexamines the IMF’s role in promoting transparencyin government operations. It is, however, not to beviewed as a comprehensive compendium of transpar-ent practices.

Conceptual Setting

Fiscal transparency is defined in this paper asopenness toward the public at large about govern-ment structure and functions, fiscal policy inten-tions, public sector accounts, and projections. It in-volves ready access to reliable, comprehensive,timely, understandable, and internationally compara-ble information on government activities—whetherundertaken inside or outside the government sec-tor—so that the electorate and financial markets canaccurately assess the government’s financial positionand the true costs and benefits of government activi-ties, including their present and future economic andsocial implications.

Transparency in government operations has sev-eral dimensions. First, at an aggregate level, trans-parency requires the provision of reliable informa-tion on the government’s fiscal policy intentions andforecasts. Second, detailed data and information arerequired on government operations, including thepublication of comprehensive budget documents thatcontain properly classified accounts for the general

government and estimates of quasi-fiscal activitiesconducted outside the government. The third dimen-sion consists of mainly behavioral aspects, includingclearly established conflict-of-interest rules forelected and appointed officials, freedom-of-information requirements, a transparent regulatoryframework, open public procurement and employ-ment practices, a code of conduct for tax officials,and published performance audits. In all three di-mensions, fiscal transparency is closely associatedwith the successful implementation of good gover-nance.1 While this paper provides an overview ofmeasures to strengthen behavioral aspects, it chieflyfocuses on the first two dimensions of fiscal trans-parency, which are generally more amenable to IMFsurveillance (Section III). In particular, it seeks toidentify specific practices (e.g., transparency in fis-cal statistics and targets as well as budget and taxcollection procedures) that enhance good gover-nance through greater visibility to the public.

From a practical standpoint, it is necessary to dis-tinguish between deliberate secrecy, or misreporting,and a technical inability to provide certain informa-tion (e.g., owing to inadequate data collection sys-tems). The latter is usually attributable to the slowpace of technical or administrative development,which can be corrected through training and institu-tion building, supported by technical assistance. Adeliberate lack of fiscal transparency is often attrib-utable to a government’s attempt to escape publicscrutiny of its behavior2—especially in the run-up toelections—to avoid or postpone possible adverse re-action from the electorate and from financial mar-kets, on which it depends for political support anddeficit financing, respectively.

Pressures to engage in nontransparent practicesare likely to mount during periods of fiscal stress.Rather than take unpopular corrective action, gov-

I Introduction

1

1The linkage between fiscal transparency and good governanceis discussed, for example, in World Bank (1994).

2Public choice theory emphasizes the incentive of governmentsto understate the actual costs and overstate the benefits associatedwith specific policy decisions, exploiting the fiscal illusion of vot-ers. See, for example, the discussion in Buchanan and Wagner(1976) and an early treatment in Puviani (1903).


I INTRODUCTION

ernments may resort to such practices when facingdifficulties in meeting near-term budget targets.3Such budget targets may be either self-imposed orpursuant to an outside commitment—including inthe context of an IMF-supported program. A currentillustration is provided by the creative accountingpractices adopted in some member countries of theEuropean Union (EU) to meet the fiscal criteriaspecified under the Maastricht Treaty.4

Arguments For and Against Fiscal Transparency

Fiscal transparency, in each of its three dimen-sions, is a necessary condition for sound economicpolicy. Timely publication of a clearly presented bud-get document makes it easier for the market to evalu-ate the government’s intentions and allows the mar-ket itself to impose a constructive discipline on thegovernment. Transparency increases the political riskof unsustainable policies, whereas the lack thereofmeans that fiscal profligacy can go undetected longerthan it otherwise would. Similarly, a transparent pub-lic financial accounting system makes it possible forthe market to determine what the government has ac-tually done and to compare budgeted and actual fi-nancial operations. Fiscal transparency—including,for example, open procurement practices—not onlyfacilitates the achievement of the basic macroeco-nomic policy objectives, but also increases the pro-ductivity of public expenditure. More generally,transparency, by increasing the trust that the popula-tion reposes in the government, has a salutary effecton society and the economy. There are thus strong apriori reasons for arguing that transparency improvesthe performance of the economy.

It may, however, be difficult to prove that fiscaltransparency always leads to better policy settingsthat are, in turn, translated into improved fiscal andeconomic outcomes. The difficulty stems partlyfrom the fact that transparent practices are only oneinfluence on the overall outcome and partly from theproblems encountered in specifying the counterfac-tual experience.

In general, countries characterized by a relativelyhigh degree of fiscal transparency have exhibitedgreater fiscal discipline and, in many instances, havebeen able to achieve a more robust economic perfor-mance than other broadly comparable—in terms ofresource endowment and cultural characteristics—countries with less transparent practices within thesame region.5 There are countries where secrecy ingovernment operations and a lack of discipline con-tributed to a disastrous economic performance, de-spite an ample resource endowment.6 In a broadercontext, the recent crisis in Southeast and East Asiaillustrates that transparency in the financial sector—including the extent of government-directed or -guaranteed lending—is a prerequisite for sustainedgrowth. Also, cross-country studies suggest a posi-tive relationship between broadly defined transpar-ent budget practices and fiscal discipline.7 Althoughevidence points to a positive linkage between fiscaltransparency and performance, corroboration of therelationship would require a more systematic docu-mentation of country practices.

Nontransparent fiscal practices tend to be destabi-lizing, to create allocative distortions, and to exacer-bate inequities. These adverse repercussions may notbe apparent in the near term,8 but may surface laterin the form of a severe financial crisis, requiringmuch costlier remedial action. For instance, non-transparent tax concessions, quasi-fiscal subsidies,and off-budget spending all contribute to fiscal im-balances. The destabilizing consequences of an ac-cumulation of payment arrears and of unfunded con-tingent liabilities are usually felt with longer lags.More immediately, governments that do not disclose

2

3According to Goodhart’s law—named after Professor CharlesGoodhart’s observation about targeting monetary aggregates—astatistical indicator ceases to be reliable once it is declared an of-ficial target for policy purposes.

4Although creative accounting practices associated with thesefiscal criteria have been given considerable attention in the media,it is particularly their future implications—including the addi-tional measures needed to compensate for the insufficient fiscaladjustment at the outset—that lack transparency. The broaderissue is that creative accounting damages the credibility of the fis-cal criteria, and thus their effectiveness, when the countries en-gaging in such practices are seen as not adhering to the criteria.

5The country experiences presented in Appendix IV providesome support for this view. Indeed, Botswana, Chile, Denmark,and New Zealand, which display considerable fiscal transparencyand discipline, have experienced macroeconomic stability andstrong growth in recent years. Both Hungary and Jordan, despitemajor handicaps relative to other countries in their respective re-gions (i.e., high indebtedness and a lack of a significant resourceendowment), have seen an improvement in economic perfor-mance while implementing fiscal adjustment in an increasinglytransparent context.

6The former members of the Council for Mutual Economic As-sistance (CMEA) operated, under socialist central planning, withopaque budget documents and budget process, a large number ofnonparametric turnover taxes and subsidies, and a proliferation offiscal and quasi-fiscal activities through branch ministries and thestate-owned enterprise and banking sectors. See InternationalMonetary Fund and others (1991).

7See Wagner (1976) for the United States, von Hagen andHarden (1994) for EU member countries, and Alesina and others(1995) for Latin America.

8Although fiscal transparency cannot guarantee consensus,there have been episodes (including recent ones) where a failureto prepare the population, through adequate and candid explana-tion, for the removal of a critical subsidy or of a labor market reg-ulation has led to major unrest and jeopardized the improved eco-nomic performance sought by those measures.


Arguments For and Against Fiscal Transparency

sufficient information to financial markets are likelyto incur an increased risk premium over time. Alltaxes and subsidies, as well as economic regulations,alter relative prices and factor returns and cause dis-tortions in resource allocation. In addition, however,when they are not transparent, the harmful impact ofthese practices, including the benefits they may pro-vide to influential interest groups at the expense ofpoorer and less vocal groups, is hidden from publicview and debate.

Fiscal transparency can also impose costs. Obvi-ously, up-front costs are incurred in creating thetechnical capacity and institutions to establish a cen-tralized information system, develop reliable fore-casting tools, implement appropriate accountingtechniques, and simplify regulatory practices ormake their cost visible. Moreover, there are recur-rent, albeit often declining, costs in maintainingthese practices and disseminating the generated in-formation. The costs of transforming a culture of se-crecy into one of transparency may be at leastequally large.

Whereas institutional arrangements and account-ing practices in the public sector must always betransparent to reap the known benefits, the timing ofpublic disclosure as to the formulation of govern-ment decisions may require some judgment. In somecases, broad policy goals, including the fiscal targetsand the measures to achieve them, are announced assoon as they are decided; following the announce-ment, a debate may ensue, and the initial targets andmeasures may be altered in the light of that debate.However, alternative approaches may work equallywell. For example, some political leaders maychoose to air policy targets and intentions informallyin an attempt to test public opinion and generate thenecessary consensus to carry out, or modify, thosepolicy intentions. In any event, deliberations withineach branch of government on specific features andtiming of the measures being considered may haveto be closed to the public to avoid undue influencefrom more powerful and active lobby groups.9

Generally speaking, once the decision is reachedon a given measure, it should be publicly announcedunless it is a component of a broader policy package.In that event, a delay may be justified until the entirepackage is ready to be unveiled, to preserve the in-tegrity of the package. For example, a real wage cutfor civil servants in the context of the draft budget

that is announced before agreement is reached withthe relevant labor representatives may undermine thegovernment’s position at the negotiating table andunravel the coherence of the budget.10

Transparency entails risks when the resulting be-havior of some groups may be inimical to the gen-eral welfare or may erode the effectiveness of a spe-cific policy instrument. For example, in conductingmonetary policy, the authorities refrain from reveal-ing their intentions about a future exchange rate orinterest rate action to maximize its effectiveness andminimize windfall gains to certain economicagents.11 In the fiscal area, premature announcementof the introduction of a subsidy or tax incentive mayalso weaken its intended effect and result in a wind-fall gain for some agents and an unnecessary bud-getary cost. Announcement of, say, an investmenttax credit prior to its effective date may induce enter-prises to postpone investment expenditures theywould have undertaken without the tax credit. Sym-metrically, announcing the repeal of an existing taxcredit before the effective date may induce an accel-eration of capital formation that would take placeanyway. Likewise, announcement of future tariffcuts (increases) may induce a postponement (an ac-celeration) of imports prior to the effective date ofthe policy action.12 The foregoing considerationsare, to an extent, comparable to basic principles oftransparency, namely, simultaneous release of, andequal access to, statistical information to avoid un-fair rent seeking. These principles do not, of course,preclude administrative practices that restrict prere-lease access to sensitive information, for a shorttime, to key government officials.13

In general, provision for contingencies should bemade explicit in policy statements and budget docu-ments. However, there are situations where for tacticalreasons governments may choose to adopt implicitly asomewhat conservative set of underlying macroeco-nomic assumptions or fiscal parameters in formulat-ing and submitting a draft budget so as to reduce the

3

9In the United States, for example, according to Reese (1979),it was observed that since 1971, when legislative markup sessionsbecame open to the public, congressmen have relied increasinglyon political aides and have been exposed to greater pressuresfrom special interest groups, whereas previously they resorted toa largely apolitical technical staff and were somewhat protectedfrom such pressures.

10Further, it could be argued that the premature disclosure of in-formation about the likely adverse effects of pending discretionarymeasures on certain groups may be counterproductive—risking apolicy reversal once the information becomes available—espe-cially where the more general benefits of the measures may not beunderstood or quantifiable.

11On the reaction of financial markets to secrecy in monetarypolicy, see Tabellini (1987).

12By the same token, tax preferences intended to promote agiven activity, when granted retroactively—say, to the beginningof the fiscal year—confer a windfall gain on economic agentswho would have engaged in it without the preferences. Retroac-tive elimination of the tax preference is tantamount to imposing atax and may be seen as unfair to those who had counted on thepreference when taking their decisions.

13These practices are explicitly recognized under the IMF’sSpecial Data Dissemination Standard (SDDS).


I INTRODUCTION

downside risk of a subsequent weakening of the fiscalstance by the legislature. Along similar lines, in somecountries, the authorities may be reluctant to revealthe accumulation of fiscal surpluses—stemming, forexample, from an unexpected rise in the world priceof a key export commodity or from a windfall gainfrom privatization—for fear of pressure to ease thefiscal policy stance.14 The merits of other, time-incon-sistent, arguments for secrecy or ambiguity—benefi-cial to elected officials15—cannot be viewed as a pre-scription for best practice.

In summary, the case for fiscal transparency restson the fundamental principles of public finance: sta-bility, efficiency, and fairness. Overall, fiscal trans-parency tends to be associated with fiscal disciplineand enhances good governance, thus contributing to

improved economic performance. Because trans-parency leads to government accountability—whereas, conversely, statutory accountability require-ments can bring about transparent practices—andcredibility, the beneficial effects are reflected inlower risk premiums in financial markets andstronger support of government policies by a well-informed electorate.16 Nonetheless, the arguments infavor of transparency are subject to certain caveats;under certain well-defined circumstances and for alimited time, access to sensitive budget data or infor-mation on specific policy measures that may conferunfair benefits on some groups may have to be re-stricted to a few key government officials.

4

14The less-than-full public disclosure of large foreign exchangereserve holdings or the earmarking of externally generated fiscalsurpluses to hidden off-budget accounts, for prudential reasons,has been a hallmark of some countries in Southeast Asia and theMiddle East.

15See Alesina and Cukierman (1990) for examples of institu-tional arrangements and practices that maximize politicians’chances of being reelected while concealing their true ideologicalpreferences: relying on legislative subcommittees and task forces,avoiding roll-call voting, and secretly influencing monetary au-thorities and then using them as scapegoats.

16The central role of the government budget in this process hasbeen acknowledged, for example, in the United States (President’sCommission on Budget Concepts, 1967): the budget should pro-vide “the public with information ... essential for private business,labor, agriculture, and other groups, and for an informed assess-ment by citizens of governmental stewardship of the public’smoney and resources” (p. 2). And, in turn, the public “must be ableto participate intelligently in the big decisions that come to focusthere: the overall size of government; the relative emphasis on dif-ferent government programs and activities intended to benefit theNation; the efficiency and effectiveness of major government pro-grams in the light of their intended purposes; the need for tax in-creases or the opportunities for tax cuts; and fiscal policies de-signed to promote national prosperity” (p. 11).


T his section reviews the major facets of fiscaltransparency in three overlapping and interre-

lated areas.17 The first encompasses transparency ingovernment institutions and behavior. To secure sup-port for fiscal policy and its implementation, the au-thorities must inform the public about the overallstructure of government, as reflected in the relationsbetween the public sector and private agents and inthe interactions within the public sector. These in-clude openness in the budget process, tax policystatutes and administration, the government’s financ-ing operations, and the nature and costs of the regu-latory framework. The second area consists of trans-parency in public accounts—that is, themeasurement of government transactions, owner-ship, and obligations—required for sound fiscal pol-icymaking. For this purpose, it is necessary to focuson the coverage, recording basis, valuation, recogni-tion, and classification of relevant flows and stocks.The third area deals with the transparency of sum-mary indicators used to assess fiscal policy stanceand sustainability as well as of projections of fiscalaggregates—all dependent on the quality and trans-parency of public accounts.

Institutional Framework

A basic requirement for transparency in the over-all structure and functions of government is a cleardemarcation of the boundaries between the publicand private sectors (Box 1). Apart from some activ-ities (e.g., education, health care, and public utili-ties) where both sectors may operate concurrently,in many countries, the government is solely respon-sible for public administration, internal safety,defense, foreign relations, and macroeconomic pol-icy. As part of the delineation of areas of compe-tence, it is essential to limit private rent seekingby officials in the public domain through enforce-ment of conflict-of-interest legislation. Meanwhile,freedom-of-information legislation helps ensure

government transparency and accountability bygiving citizens access to public documents and as-signing to government the burden of justifyingnondisclosure.

Within the public sector, it is necessary to main-tain transparency in the relations between the gov-ernment and the state-owned enterprise sector. Al-though there are sound reasons for limiting theextent to which this sector is engaged in activitiesmore legitimately carried out by the government,where such practices nevertheless occur, it is im-portant that they be well documented in publiclyavailable reports. Similarly, in spite of the generaltrend toward managing and operating state-ownedenterprises along commercial lines, many enter-prises still incur losses from nontransparent quasi-fiscal activities (e.g., in providing social benefits).Equally, the cost of quasi-fiscal activities con-ducted by public or private financial institutions—through multiple exchange rates, preferential cred-its, and guarantees—is frequently not made explicitor even calculated. Regardless of the actual bud-getary treatment of quasi-fiscal activities,18 infor-mation should be provided—preferably in annualbudget documents—on the associated cost (orgain). Conversely, documentation should be madeavailable when government ministries perform ac-tivities that would normally be performed in thestate-owned enterprise sector.

A related issue of major concern is the provisionof adequate information on the fiscal costs and theterms of assistance or of restructuring of public (or, in some cases, private) financial institutionsand nonfinancial enterprises.19 Similarly, privatiza-tion of such entities must be conducted with asmuch openness as permitted by sound marketingconsiderations.

Responsibility for expenditures on basic publicservices and functions needs to be clearly divided

II Issues in Fiscal Transparency

5

17For a more detailed discussion of these issues, see Appen-dices I through III.

18Namely, regardless of whether enterprises are compensatedor not with explicit transfers from the budget for the cost of quasi-fiscal activities.

19For a survey of bank rescue operations, see Alexander andothers (1997).


II ISSUES IN FISCAL TRANSPARENCY

between national and various subnational levels ofgovernment. Accordingly, the revenue base of eachlevel of government should be defined unambigu-ously, possibly accompanied by formula-basedarrangements for revenue sharing and intergovern-mental transfers, thus limiting the room for ad hocbargaining among jurisdictions.20

In principle, extrabudgetary operations can be effi-cient for pursuing certain tasks for which the spendingobligation transcends the annual budget appropriationprocess.21 Cases in point are social insurance andcommodity stabilization funds. By contrast, in a num-ber of countries, extrabudgetary funds—for instance,for military spending—have been created mainly toavoid legislative scrutiny. More often, funds originallyestablished for valid reasons have become highly dys-functional. For example, in some countries, easy ac-cess to old-age and disability pension funds is usednontransparently to alleviate unemployment. In other

countries, reserves accumulated in commodity stabi-lization funds—or, in general, reserve funds estab-lished from the sale of nonrenewable resources—havebeen diverted to finance consumer subsidies or pres-tige projects. Whatever the actual practices adopted,an important task facing governments is to ensure thatadequate and timely information is provided on suchactivities. This should be done in a format that permitseasy consolidation with that provided on other publicsector activities.

A potentially important instrument for ensuringtransparency in government operations is an inde-pendent review agency responsible for conductingperformance audits and studies on selected fiscal is-sues. To be effective, such an agency must be ac-countable to the legislature and the public at largeand must be endowed with wide investigative and re-porting authority over government operations.

The tasks involved at each stage of the budgetprocess are usually specified in some detail in thebudget framework law or, less frequently, are basedon past conventions and rulings. The draft budget—preferably incorporating broad fiscal targets and strat-egy in a multiyear context—should be disclosed insufficient detail to the public, although the timing ofthe disclosure may need to be carefully controlled toensure its coherence (Section I). The subsequent leg-

6

Overall structure and functions

Clear demarcation of functions between public andprivate sectors.

Delineation of the boundaries of the operations ofstate-owned nonfinancial enterprises and financial in-stitutions from those of the general government; andprovision of information on the costs of quasi-fiscalactivities performed by such enterprises and institu-tions, as well as any financial rescue operations fundedby the government.

Clear assignment of responsibilities and resourcesamong national and subnational levels of government(limiting the scope for case-by-case negotiation).

Clear statement of the rationale for, and extent of,extrabudgetary fund operations.

Establishment of an independent review agencywith wide investigative authority over governmentoperations.

Budget process

Detailed public explanation of fiscal targets and pri-orities in the draft budget.

Open legislative debate and authorization.

Transparent execution and control (including pro-curement, contracting, and employment).

Public disclosure of results of performance and fi-nancial audits.

Tax treatment

Explicit statutory basis (instead of discretionary taxconcessions or negotiated tax liabilities).

Clear administrative procedures, information re-quirements, taxpayers’ rights and obligations, and taxofficials’ code of conduct.

Estimates of tax expenditure budget.

Financing operations

Disclosure of terms (interest yield and maturity) andsources of government deficit financing.

Specification of policy criteria as well as terms andconditions of government lending decisions.

Regulation

Open legislative and administrative process (e.g.,hearings, approval).

Clear and simple statutes and implementation.

Estimates of regulatory costs.

Box 1. Summary of Good Practices in Institutional Transparency

20Increasingly, in a number of countries, authorities at differentlevels of government are engaged in an open dialogue to set fiscaltargets, delineate responsibilities, and prevent fiscal off-loadingfrom one jurisdiction to another.

21Prompted largely by the need to prepare some governmentagencies for privatization, a recent practice of some EU membercountries has been to move off-budget agencies that derive mostof their revenue from user fees.


Public Accounts

islative debate and approval should normally be openand the outcome published. At the execution stage,the government should periodically inform both thepublic and the legislature about the budgetary out-come and how it compares with the objectives. A fur-ther test of transparency in budget execution and con-trol involves open public procurement, contracting,and employment practices. Finally, adequate informa-tion is necessary for conducting both financial andperformance audits, and the results of such audits, in-cluding recommendations for and implementation ofcorrective steps, should be made public.

Transparency in tax treatment entails a well-defined statutory basis, as well as clear and simpleadministration—assisted by dissemination of all thenecessary information, including filing instructions.Discretionary tax relief provided to particular indi-viduals or enterprises or case-by-case negotiation oftax liabilities between tax officials and taxpayers, al-though unavoidable in some circumstances, impairsthe transparency and credibility of the tax system.For greater transparency in tax administration, manycountries have adopted statutes on taxpayers’ rightsand obligations, as well as rules of conduct for taxofficials. As a justified departure from transparency(Section I), the announcement of changes in tax in-centives should be timed to avoid eroding their cost-effectiveness.

Tax preferences can be viewed as tantamount tobudgetary outlays benefiting certain households, en-terprises, sectors, or activities. Estimates of tax ex-penditures—that is, revenue forgone because of taxpreferences—are an important input for the debateof the annual draft budget or tax reform. Similarly,incidence studies can provide useful informationabout the distributional implications of the tax sub-sidy system. Although, admittedly, only some ad-vanced economies have the technical capacity toprepare and publish such estimates on a regularbasis, even approximate estimates of tax expendi-tures can facilitate the evaluation of the cost of taxpreferences, including alternative proposals.

Transparency in government financing operationshas been enhanced by financial deregulation. An in-creasing number of countries are relying on open,market-based financing (often from nonbanksources) of deficits while confirming the indepen-dence of the central bank and other public financialinstitutions. Such arrangements require provision ofadequate data to market participants on such mattersas timing of tenders, security issues, coupons of-fered, prices (and effective interest rate yields), andbids and offers accepted. Moreover, governmentswishing to access international and domestic finan-cial markets must furnish rating agencies, underwrit-ers, and supervisory agencies with considerable dataon the magnitude, terms (interest yield and matu-

rity), and holders of the public debt and on the gov-ernment’s debt-service capacity (often with a state-ment on its debt-management strategy). Of course,transparency in debt operations assumes evengreater importance in a regulated market because thedistortions introduced by constraints, such as re-quired asset ratios for government security holdings,understate the true cost of financing the budgetdeficit at the expense of other lenders (which mayinclude other components of general government,such as public pension funds). Transparency in gov-ernment lending (directly or through financial insti-tutions) entails adopting clearly specified policy cri-teria for loans made, including risk assessment, aswell as releasing information on the terms and con-ditions of the loans.

Notwithstanding the justification for regulation ina number of areas (mainly public health, consumersafety, environmental protection, labor protection,and competition policy), a lack of clarity in the regu-latory framework can lead to governance problems.Also, regulation is sometimes used as a nontranspar-ent substitute for budget transfers or taxes. Unlikepublic services financed from the budget, the costsof compliance, whether borne directly by the regu-lated entities or indirectly by the rest of the econ-omy, are not visible. The need to quantify regulatorycosts is underscored by the temptation to substituteregulation for budgetary spending when govern-ments attempt to compensate for pressures to cutgovernment expenditures. In general, however, themovement toward deregulation in commodity, labor,and financial markets has contributed to greatertransparency in government operations in this area.

Public Accounts

In principle, the general government is universallyregarded as providing the most comprehensive cov-erage of the noncommercial public sector, consistingof the budget as well as social security and other ex-trabudgetary accounts, consolidated across all levelsof government (Box 2).22 However, even the conceptof general government may fall short of providingfor a full coverage of fiscal operations to the extentthat it excludes the quasi-fiscal activities of state-owned financial institutions and nonfinancial enter-prises. To account for direct or indirect government

7

22Relatively few IMF member countries can achieve this cover-age on a timely basis; see International Monetary Fund (1995).Incomplete coverage is a major impediment for timely reportingunder the IMF’s SDDS. Of the 42 members that have so far sub-scribed to the SDDS, most have indicated that they will meet therequest for up-to-date summary data on central government oper-ations and debt, but about one-half may not be able to supplytimely data on general government.



ownership of nonfinancial public enterprises, a num-ber of countries report data for the nonfinancial pub-lic sector; others have expanded coverage to includeofficial financial institutions, thereby encompassingthe entire public sector. As a preferable alternative,the concept of general government activity23 is in-tended to accurately capture the cost of all govern-ment functions, including quasi-fiscal activities con-ducted outside the general government. In the eventthat the cost of quasi-fiscal activities proves impossi-ble to quantify, transparency would be served by atleast listing such activities.

The recording basis of government transactions—namely, cash or accrual—has important implicationsfor the transparency of fiscal performance. Reliancesolely on the cash-based approach, although helpfulfor assessing the first-order impact of governmentborrowing on inflation and the external balance, can

result in a misstatement of the magnitude and timingof fiscal operations. By comparison, the accrual-based approach is indispensable for gauging themacroeconomic resource repercussions of fiscal pol-icy, especially over the medium to long term. Majordistortions under cash-based recording stem fromthe exclusion of information on accumulation of ar-rears (especially expenditure arrears), transactions inkind (including issuance of government obligationsto suppliers or tax refunds or in connection withbank restructuring), and the cost of borrowing at adiscount. Although, increasingly, the accrual-basedapproach is seen as best practice, the presently morewidespread cash-based government accounts shouldalso be provided where both methods are possible.

The valuation and recognition of assets and liabili-ties can be critical for the transparency and consis-tency of government financial statements. In this re-gard, investment expenditure may be subject tovarying treatment (full expensing or depreciation) de-pending on the analytical purpose at hand. Given thebroader difficulties of valuing public assets, gross(rather than net) measurement of public debt is oftenpreferable. Although attempts to estimate governmentnet worth may ultimately prove successful, most coun-tries are not yet in a position to prepare such estimates.

In addition, it is necessary to compile and dis-close information on commitments and contingentliabilities. Examples of these liabilities are guaran-tees for credits extended by financial institutionsand for deposits in those institutions, many ofwhich are not quantifiable because they are contin-gent on the realization of the insured occurrence.24

More important, they include obligations to futurebeneficiaries of social insurance for old-age, unem-ployment, and health care programs. Under ac-counting conventions followed in most countries,contingent liabilities for such benefit programs arenot to be added to actual government debt. Thus,rather than being included in the government bal-ance sheet, estimates of the net unfunded portion ofthese benefit-related contingent liabilities are gen-erally published separately.25

Fiscal transparency requires classifying data ongovernment operations, ownership, and liabilities intoanalytically useful categories of flows and stocks.

8

Box 2. Summary of Good Practices in Accounting Transparency

Coverage

General government, with sufficient detail on lev-els of government, and extrabudgetary funds (in-cluding social security institutions).

Quasi-fiscal activities of state-owned nonfinancialenterprises and financial institutions.

Recording basis

Accrual-based recording (including transactions inkind, cost of borrowing at a discount, and changes inarrears).

Supplementary cash-based recording.

Valuation and recognition

Measurement of government assets (including in-vestment and depreciation).

Measurement of government liabilities outstanding.

Information on commitments and contingent(funded and unfunded) liabilities.

Estimates of net worth.

Classification

Breakdown of revenue categories.

Disaggregation of expenditure on an economicand functional basis.

Breakdown of debt and financing (by type, matu-rity, and creditor).

23This definition has been proposed for the revised GovernmentFinance Statistics Manual, in International Monetary Fund(1996a).

24Where these contingent liabilities can be reliably estimated,they may be included in published balance sheets. For example,in the balance sheets prepared for the United States, the authori-ties have sought to overcome these problems by developing tech-niques that permit estimates to be made of the accruing cost ofgovernment contingent liabilities arising from loan guaranteesand insurance programs. See United States, Office of Manage-ment and Budget (1997).

25For example, in the United States, the budget documents in-clude a separate actuarial calculation of the financial balance ofthe Social Security and the Medicare Trust Funds.


Indicators and Projections

Revenue should be broken down into major tax andnontax categories and unrequited transfers, while fi-nancing, including privatization receipts, should beshown below the line.26 Expenditure must be classi-fied into major functional and economic categories,separating debt amortization from interest paymentsand placing the former as a below-the-line negative fi-nancing item. Finally, financing flows, as well as thecorresponding debt stock, should be disaggregated bycurrency denomination, maturity, and source.

Indicators and Projections

The most commonly available direct indicator ofthe fiscal balance, namely, the overall balance of gov-ernment operations, is transparent only to the extentthat it is free of distortions in data coverage, record-ing, and classification. Although generally useful, theoverall balance must be supplemented by alternativemeasures subject to the same transparency require-ments (Box 3). Calculating the current balance, as anindicator of the government’s contribution to nationalsaving, requires a clear separation of current and capi-tal transactions; this separation, in turn, depends on asatisfactory measurement of government investmentexpenditure. The primary balance, instrumental fordetermining the fiscal policy effort needed to stabilizeor reduce public debt, requires exclusion of net inter-est payments from the overall balance. Measurementof the operational balance—in countries that have ex-perienced high inflation and large levels of indebted-ness—entails identifying interest payments and de-ducting the inflationary component of such interestoutlays from the overall balance.

In addition to these flow indicators, most coun-tries compile some direct indicator of the change inthe stock of assets and liabilities. It is often arguedthat the net worth—calculated as the difference be-tween the total stock of assets and liabilities—of thegeneral government is a superior measure.27 Giventhat most IMF members are unlikely to be able tocompile such indicators in the short run, the nextpreferred alternative may be financial net worth, thatis, the difference between the stock of financial as-sets and liabilities of the general government. How-ever, in the absence of satisfactory data on financialassets, it is necessary to rely on gross debt mea-

sures—usually the focus of financial markets (Inter-national Monetary Fund, 1996b).

To understand the direction of fiscal policy, thepublic and financial markets often find it useful toexamine analytical indicators of the short-run fiscalstance. Perhaps the best known are indicators that re-move the effect of cyclical fluctuations or exogenousshocks from direct measures of the budget balance.Given the underlying information required onmacroeconomic developments and key fiscal param-eters, indicators such as the structural balance areused predominantly in the advanced economies.

Another major question is whether persistent bud-get deficits result in a growth path for debt that is notsustainable. Long-run debt sustainability can be deter-mined on the basis of an examination of current in-debtedness levels, the interest rate, the GDP growthrate, and the ratio of the primary budget balance toGDP. In addition to explicit debt measures, it is usefulto gauge the sustainability of the implicit governmentdebt stemming from public pensions, health care

9

Box 3. Summary of Good Practicesin Transparency of Indicators

and Projections

Direct indicators

Overall balance, complemented by current, pri-mary, and operational balances.

Gross and net government debt, accompanied byan estimate of net worth.

Analytical indicators

Structural or cyclically adjusted balance.

Sustainability calculations (showing primary bal-ance required to stabilize debt, given realistic as-sumptions about future interest rates and GDPgrowth rates).

Present value of net unfunded liabilities (undercontributory, defined-benefit social security pro-grams); calculation of contribution gap.

Generational accounts.

Short- to medium-term forecasts

Separate baseline forecast and policy forecasts.

Clear and realistic underlying macroeconomicforecasts and parameters.

Long-term scenarios

Separate baseline scenario and adjustment scenar-ios (especially for old-age and health care programs).

Clear and realistic range of underlying macroeco-nomic parameters and demographic trends.

26Under the proposed revision of the Government Finance Sta-tistics (GFS) classification, privatization receipts are to be shown“below the line” as financing—thus removing the impact of suchtransactions from the measurement of the thrust of fiscal policy.This practice is in conformity with the System of National Ac-counts (SNA) definitions (Inter-Secretariat Working Group onNational Accounts, 1993).

27For a summary of the arguments for and against the net worthconcept, see Blejer and Cheasty (1993).



benefits, and other defined-benefit programs. This assessment can be made by estimating the presentvalue of net unfunded liabilities of these programs asa ratio of GDP. Another approach consists in calculat-ing the contribution gap that would have to be met tomake these programs sustainable.28 Publication ofsuch measures can help develop a widespread under-standing of the need for structural reform to preserveor restore the sustainability of these schemes.

Increasing concern with the long-run distribu-tional implications of the existing fiscal structure hasalso spawned the construction of generational ac-counts. One benefit of such accounts is that theyoften highlight the fact that policy changes can shiftresources across generations without affecting thepresent fiscal deficit at all; however, largely becauseof their computational complexity, these accountsare available for only a few advanced economieswith rapidly aging populations.29

Formulation of fiscal policy, as well as the ac-companying public debate on targets and strategy,is usually predicated on projections of future trendsin government finances and in overall economicperformance. Transparency in projections, whethermodel-based or judgmental, requires explicit, real-istic, and timely assumptions, along with internalconsistency. In the publication of short-run fiscalforecasts and the underlying macroeconomic fore-casts, governments should provide baseline projec-tions, which assume unchanged policies, and alter-native projections, which incorporate the impact ofmajor policy changes. Fiscal projections must bebased on realistic and explicitly documentedmacroeconomic assumptions and parameters (e.g.,effective tax rates, tax bases, compliance coeffi-cients, or collection lags). Transparency is furtherassisted when the actual outcome is evaluatedagainst projections.

Similarly, assessment of fiscal sustainability—inview of rapidly aging populations, the rising cost ofhealth care, and the rigidity of most social entitle-ments—should be based on long-term scenarios,showing the evolution of the fiscal balance over sev-eral decades. Such scenarios can be particularly use-ful for ascertaining the sustainability of social secu-rity systems and illustrating, in a transparent manner,the effect of specific reform options.

10

28The contribution gap is the difference between a constant sus-tainable contribution that, over a long period of time, would notlead to a buildup of liabilities under public pension and other de-fined-benefit programs above an initial level and the expected av-erage contribution rate likely to prevail under current law. See In-ternational Monetary Fund (1996b).

29In many cases, generational accounts are prepared by privatesector analysts rather than governments, partly reflecting the judg-mental assumptions often required to compile these estimates.


Major progress has been made throughout virtu-ally the entire IMF membership toward fiscal

transparency. The IMF has played a role in this en-deavor through various channels: surveillance in thecontext of Article IV consultation discussions andWorld Economic Outlook (WEO) exercises; condi-tionality in IMF-supported programs; technical assis-tance, including training covering a wide range ofpublic finance and statistical issues; development andpublication of the Government Finance Statistics(GFS) Database; development of standards for datadissemination; publication of research on selectedfiscal policy issues; and the recent initiative to pro-mote good governance. For the most part, with theexception of technical assistance, IMF involvementhas focused on fiscal transparency at the macroeco-nomic level, as well as on microeconomic issues(e.g., quasi-fiscal operations of public enterprises)that have direct macroeconomic repercussions.

Overall Trends and Priorities

In recent years, most IMF member countries havemade important steps toward greater fiscal trans-parency.30 To be sure, however, even the advancedeconomies exhibit some nontransparent arrangementsand practices. The economies in transition—whichuntil the beginning of the decade had conducted fiscalpolicy in virtual secrecy—have made the greatest leaptoward transparency, although some of them still haveamong the least transparent fiscal systems. Consider-able scope remains for eliminating nontransparentpractices in many countries; nevertheless, it is impor-tant to keep in mind the distinction between unin-tended nontransparency attributable to slow technicaland institutional development and deliberate misrepre-sentation or suppression of information. The latter, inprinciple, can be remedied over a shorter time horizon.

In the future, for the advanced economies, the mainpriority appears to lie in making public comprehen-sive information to facilitate the debate over fiscal dis-

cipline and debt sustainability issues. To this end,more precise specification of policy targets and subse-quent monitoring of progress toward achieving themare required. These steps, in turn, presuppose in-creased attention to most measurement issues raisedin the previous section, including those that bear onthe intertemporal implications of fiscal policy—espe-cially entitlement programs associated with aging. Inparticular, progress on these issues involves estimatesof net unfunded liabilities and quasi-fiscal operations,costing of tax expenditures and regulation, and im-proved documentation of fiscal projections.

In the developing economies, attention should befocused on institutional reforms—enhancing trans-parency mainly in the budget process, taxation, andquasi-fiscal activities of public or private nonfinancialenterprises and financial institutions—and on thecompilation and dissemination of essential fiscal dataand projections. The economies in transition sharemany of the needs of developing countries, but theirhistory of secrecy—including widespread quasi-fiscalactivities and data systems oriented to planning ratherthan to market needs—may make it more difficult forthem to move rapidly toward transparent fiscal prac-tices. In many developing and transition economies,efforts in this area are particularly important for pro-moting good governance.

Role of the IMF

Article IV consultations provide the main vehiclefor exercising surveillance over member countrypolicies. In this context, the IMF must assess the in-tent and viability of fiscal adjustment programslargely on the basis of the underlying data and fore-casts. The quality of the assessment therefore de-pends on the transparency of available information.Conversely, as part of this assessment, the IMF hasoften urged the authorities to increase transparencyby, among other measures, eliminating or making ex-plicit government expenditure and tax payment ar-rears, off-budget operations, and quasi-fiscal activi-ties; removing nontransparent subsidies fromcommodity stabilization funds and public pension

III Role of the IMF in Promoting Fiscal Transparency

11

30For a selection of country examples, see Appendix IV.


III ROLE OF THE IMF IN PROMOTING FISCAL TRANSPARENCY

funds; and trimming or making apparent the cost ofgovernment regulation of labor, commodity, and fi-nancial markets. Fiscal policy advice at the individ-ual country level has been supported by analysis in amultilateral context through the WEO exercises—drawing, in part, on internationally comparable mea-sures of fiscal stance and sustainability for major ad-vanced economies.

Conditionality in IMF-supported programs oftenrequires implementation of fiscal policy advice inthe form of prior actions, performance criteria, andstructural benchmarks. Quantitative performancecriteria for fiscal aggregates, including the budgetbalance, need to be transparent and based on theprinciples discussed above (Section II). In addition,the programs are normally to be cast in the frame-work of consistent medium-term fiscal adjustmentscenarios.

At the request of member countries, the IMF’sFiscal Affairs Department has provided technical as-sistance over a wide range of public finance issues.Some of the recommendations offered through suchassistance—with a view to promoting trans-parency—have included eliminating discretionary ornegotiated tax preferences and hidden subsidyschemes, formulating budget framework laws,unifiying treasury accounts, establishing timely andcomprehensive budget reporting systems31 and pub-lic enterprise monitoring systems, improving fiscalforecasting, and developing long-term policy scenar-ios to simulate various pension and health care re-form options. Often, these recommendations havebeen incorporated in IMF-supported programs orhave served as inputs in Article IV consultation dis-cussions. In addition, the IMF’s Statistics Departmenthas extended considerable assistance to improve gov-ernment statistics on the basis of well-established accounting conventions.

A major effort in the promotion of fiscal trans-parency has been the development of the GFSmethodology, its adoption by IMF member coun-tries, and the construction and publication of theGFS Database. The current revision of the GFSguidelines, in line with the 1993 System of NationalAccounts (SNA) standards, represents a further steptoward improving the transparency and consistency

of fiscal statistics.32 The recent creation of the IMF’sSpecial Data Dissemination Standard (SDDS),aimed at countries accessing international capitalmarkets, represents another important effort to pro-mote transparency in the dissemination of macro-economic data; fiscal databases are included, albeiton a highly aggregated basis.33

In recent years, transparency has been furtherserved by the publication of research on selected fis-cal policy issues, in the form of background papersfor Article IV consultations and the semiannualWorld Economic Outlook reports. In addition, the re-cent publication of staff studies on the quasi-fiscalactivities of public financial institutions, the fiscalimplications of aging, and the fiscal costs of bank re-structuring has contributed to the public’s under-standing of such issues. Most recently, the IMF haslaunched a major initiative to promote good gover-nance in member countries, which involves, ofcourse, increased fiscal transparency.

Overall, IMF advice and assistance in public fi-nances, encompassing institutional arrangements,policy measures, administrative practices, statisticalstandards, and forecasting techniques, have made amajor contribution to the current trend towardgreater fiscal transparency.

Although considerable headway has been made to-ward transparency in government operations, there isample scope for further progress in most membercountries. Against this background, consideration isbeing given to extending IMF involvement—in coop-eration with other multilateral institutions. In manycountries, the IMF can build on the willingness of theauthorities to move ahead rapidly with reforms. Withsome members providing examples of good practice,the IMF may play an increasing role as a catalyst, en-gendering interest in and information on good prac-tice techniques to governments that show a desire forimprovement.

12

31The World Bank (1994) has also provided assistance to in-crease transparency in the budget process in Africa, Asia, andLatin America.

32The revised GFS guidelines formally adopt an accrual-basedrecording method to supplement, rather than replace, the cash-based recording method currently used in many member coun-tries (Appendix II). Operationally, accrual-based recording underthe new guidelines can be achieved in these countries by compil-ing the additional information (on arrears and various noncashtransactions) needed to complement existing cash data in in-stances where the differences between cash and accrual data aresignificant.

33The General Data Dissemination System, expected to be established soon, will promote transparency throughout the membership.


F iscal transparency—defined as public opennessin government institutions, fiscal policy inten-

tions, public sector accounts, indicators, and fore-casts—is fundamental to sound economic policy.Transparency allows the market to evaluate, and im-pose discipline on, government policy and increasesthe political risk of unsustainable policies. The po-tential role of transparency in promoting good gov-ernance has been widely recognized.

Despite the inherent difficulty of fully corroborat-ing the link between transparency, on the one hand,and fiscal discipline and economic performance, onthe other, it can be shown that the better-performingcountries in the various major regions of IMF mem-bership generally follow more transparent fiscal prac-tices. There are good reasons to believe that fiscaltransparency contributes to macroeconomic stability,allocative efficiency, and fairness. Furthermore, fiscaltransparency leads to increased credibility, which, inturn, helps reduce risk premiums in financial marketsand strengthens support by the electorate. Notwith-standing this general presumption, a temporary de-parture from transparency may sometimes be justi-fied—notably, when the premature announcement ofsensitive statistical data or policy measures wouldweaken their effectiveness and confer unintendedwindfall gains on some groups.

A review of issues in fiscal transparency points tothe importance of adopting practices and providinginformation that ensure a clear demarcation betweenthe public and private sectors. This includes docu-mentation and quantification of the extent of govern-ment intervention in such areas as bank rescue andenterprise restructuring operations and privatization.Also, information should be made available on thecost of any quasi-fiscal activities of state-owned orprivate nonfinancial enterprises and financial institu-tions. Unless there is a clear economic rationale forplacing certain activities off-budget, governmentsshould refrain from doing so. In any event, sufficientinformation on such activities should be provided toensure that they are subject to the same public andlegislative scrutiny as on-budget operations. Thebudget process should involve, above all, public dis-closure of budget documents—including clear enun-

ciation of fiscal targets and strategies—and openlegislative debate. Transparency in budget executionand control includes open procurement, contracting,and hiring practices. Financial and performance au-dits of budget operations should be subject to publicscrutiny. In the tax area, transparency entails clearlydefined statutes, with no recourse to discretionarytax concessions or negotiated tax liabilities. Pub-lished tax expenditure estimates should underpin de-bates on the draft budget or on tax reform. Informa-tion should be made available on the terms andsources of government deficit financing. Similarly,the criteria and terms of government lending shouldbe clearly disclosed. Recent gains from deregulationshould be consolidated with the implementation of asimple regulatory framework that can be costedtransparently.

Sound fiscal policymaking requires clarity in themeasurement of government transactions, ownership,and obligations. To this end, coverage of fiscal ac-counts should extend to the entire general govern-ment, supplemented with information on quasi-fiscalactivities, preferably applying accrual-based record-ing. In practice, cash-based recording—useful forshort-run fiscal analysis—can be enhanced with dataon payment arrears, transactions in kind, and discountsecurities to approximate an accrual-based presenta-tion, which is necessary to determine the medium-term implications of budget policy. Other issues thathave a bearing on the transparency of government ac-counts are the appropriate valuation of assets and lia-bilities—including recognition of commitments andcontingent liabilities—and a meaningful classifica-tion of revenue, expenditure, and debt.

Transparency is also necessary for indicators offiscal stance and sustainability. Various measures ofgovernment balance and public indebtedness mustbe disclosed. Also, calculation of the actuarial valueof net unfunded entitlement liabilities and estimatesof generational accounts are useful. Publication offiscal projections should be documented with suffi-cient information on the underlying methodologyand key macroeconomic assumptions. Explicitmedium-term fiscal projections need to be incorpo-rated in annual budget documents. In countries with

IV Summary and Conclusions

13


IV SUMMARY AND CONCLUSIONS

rapidly aging populations, long-term scenarios forpublic pension and health care programs should bepublished regularly. Moreover, the governmentshould make public a retrospective evaluation of theactual outcome against projections.

Although, in recent years, many IMF membercountries have made substantial progress towardtransparency in virtually all these areas, there is con-siderable scope for improvement. Advancedeconomies should concentrate on developing moretransparent measures of fiscal sustainability, alongwith an open debate of reform options in the face ofaging populations. In developing economies, empha-sis should be on promoting transparency in govern-ment institutions and on disseminating essential fis-cal data and projections. In the economies intransition, adoption of new attitudes is essential tosupport the institution-building process.

The case for eliminating the remaining nontrans-parent practices is underscored by the vulnerabilityof the world economy to the increasing integration offinancial markets across country boundaries. Trans-parency in public finances can contribute to policy

credibility and thus help prevent a potential crisisfrom materializing and spreading across countries. Atthe same time, it is important to recognize thatgreater transparency may involve costs, particularlywhere information systems and reporting practiceshave to be developed or improved; and that efforts topromote fiscal openness must enjoy strong support inmember countries if they are to succeed.

The IMF has contributed to the improvement infiscal transparency in most areas through surveil-lance, program design and implementation, techni-cal assistance, development and application of statis-tical standards, and publication of research activities.However, at least until recently, theseefforts did not always translate into enhanced fiscaltransparency by member governments toward theirown electorate and financial markets. The recentestablishment of the Special Data DisseminationStandard—to promote transparency mainly inmacroeconomic databases—is a potentially impor-tant initiative in this regard. In addition, further stepsare being considered to strengthen the role of theIMF in promoting fiscal transparency.

14


T ransparency in government behavior is re-flected mainly in the structure and functions of

the public sector, and particularly in the budgetprocess, tax treatment, financing operations, andregulatory mechanism. This appendix focuses on thebroad range of fiscal arrangements that determinethe frontier between the public and private sectors inthese areas. In addition, it examines the functions of,and interactions among, various components of thegovernment as well as their relationship with the restof the public sector.

Overall Structure and Functions

Transparency in the boundary between the publicsector and private economic agents entails develop-ment of a well-defined and publicly available pol-icy statement that sets the relative areas of compe-tence and provides limited, if any, scope for privaterent seeking in the public domain. Apart from someareas (e.g., education, health care, and public utili-ties) where public and private providers may oper-ate concurrently, in most industrial countries andsome developing countries, the separation betweenpublic and private activity has been clarified overthe past two decades. While the government has re-tained responsibility for public administration, in-ternal safety, defense, foreign relations, and macro-economic policy, the private sector has become theprincipal actor in commercial and profit-makingactivities and, in some cases, infrastructure.

Recently, certain industrial countries have takensteps to conduct the remaining government opera-tions by applying more explicit cost and performanceindicators and management techniques used by pri-vate enterprises.34 But the marked increase in the out-sourcing of government operations to large-scale pri-vate suppliers or contractors (in areas such asdefense, health care, and certain administrative ser-vices), whose principal or sole client is the govern-ment, may offer scope for nontransparent behavior,even if the transactions are formally subject to open

tender.35 Whereas the boundaries between public andprivate spheres are generally weaker in developingthan in industrial countries,36 the demarcation is per-haps least transparent in the economies in transition.In these economies, for the most part, the boundarieshave not yet been defined—lacking even conflict-of-interest legislation—to the point where, in somecountries, public officials are allowed to engage intransactions for personal profit using public re-sources, including internal government information.

Freedom-of-information legislation is anotherfundamental instrument of government transparencyand accountability.37 It gives citizens access to gov-ernment documents without their first having toprove a special interest, and the burden of justifyingnondisclosure falls on the government. Althoughsuch legislation imposes administrative costs, andexempting highly sensitive issues in certain areas(e.g., defense and foreign affairs) can be justified,experience suggests that it can change the attitude ofelected officials and civil servants toward fiscaltransparency.

Within the public sector, publicly released infor-mation on the assignment of functions among vari-ous components of the sector is not always availableand, when available, is not always clear, even insome advanced economies. The nexus between thegovernment and public financial institutions or non-financial public enterprises is rather opaque in manycases.38 The paucity of publicly available informa-

Appendix I Transparency in Institutions and Behavior

15

34See the examples cited in Gore (1993, p. 76).

35See, for example, Gansler (1995) on U.S. defense expenditures.36Sometimes, the boundaries between the activities of the state

and those of its political leaders are by no means transparent. Forexample, in some countries, the differences between public re-sources and the personal finances of the ruling family are oftennot explicitly articulated.

37The earliest legislation governing the opening of governmentrecords to the public dates back to 1776 in Sweden. The presentlaw is unique in that it is one of the four laws that together makeup Sweden’s Constitution. Similar, but less rigorous, systemswere introduced in the 1970s in Denmark, Finland, Norway, andthe United States. Since then, legislation on open records hasspread to most industrial countries; see Pope (1996).

38In France, for example, Morin and Dupuy (1993) have arguedthat both the extent of subsidization and the degree of state in-volvement in public enterprises lack transparency.


tion on the fiscal costs and the terms of bank rescueoperations (Alexander and others, 1997) and of en-terprise restructuring has been notable, not only invirtually all economies in transition but also in someindustrial and developing countries. In some cases,financial injections were provided in an ad hoc man-ner, often from off-budget sources.39 The privatiza-tion of state-owned enterprises has been conductedwith varying degrees of openness across countries.In some countries (e.g., Australia, most EU coun-tries, New Zealand, and United States), there hasbeen nearly full public disclosure on all aspects ofenterprise sale.40

Notwithstanding the general trend toward manag-ing and operating state-owned enterprises in an in-creasingly open way along commercial lines (e.g.,Australia and New Zealand)—in some cases, inpreparation for eventual privatization—state-ownedenterprises in a number of countries are still engagedin sizable quasi-fiscal activities, mainly in the formof social benefits, overemployment, and price subsi-dies without compensation from the budget.41 Suchquasi-fiscal activities are remnants of the widespreadprovision of implicit subsidies for consumption ofloosely defined “merit” goods and services and foremployment, not only in the former centrallyplanned economies but also in many developingcountries.42 Although the ideal solution would be tophase out such activities over time, transparency re-quires that estimates of their costs be published inbudget documents.

Perhaps less transparent are the quasi-fiscal activ-ities conducted by public financial institutions.These activities take many different forms: multipleexchange rates, exchange rate guarantees, importdeposits, preferential credits and loan guarantees(directed, e.g., to exports, and investment in specificactivities), deposit guarantees, and selective reserverequirements (Mackenzie and Stella, 1996). Suchimplicit subsidies and taxes, not explicitly remuner-ated by the government, can be found especially indeveloping and transition economies. Again, as afirst step, it is necessary to improve the transparencyof these operations through the publication of esti-mates of the costs involved.

A lack of clarity in the division of responsibilitiesbetween national and subnational levels of govern-ment is a major problem in many developing andtransition economies and, to a lesser degree, in someindustrial countries. Although revenue sharing hasbeen clarified in some countries, including thosewith clear formula-based arrangements (e.g., Canadaand Denmark), in others, bargaining and ad hoc ne-gotiations persist for intergovernmental transfers.The least transparent distribution of functions andresources can be found, of course, in someeconomies in transition that have recently evolvedfrom a highly centralized system of government to afederation or confederation.

In principle, extrabudgetary operations can beuseful for pursuing certain tasks more efficiently (orfor ensuring greater support from users) on a fidu-ciary basis, in the form of separate funds—thoughoften administered in a manner similar to budget en-tities—particularly if their spending obligation tran-scends the annual budget appropriation process.Such is the case for highway funds (established forroad construction and maintenance and financedfrom earmarked user fees and capital transfers),commodity stabilization funds (set up to mitigatethe effect of export price fluctuations), or social in-surance funds (to protect against unemployment,old age, or disability).43 In a number of countries,however, extrabudgetary funds—for example, to fi-nance export promotion, tourism, housing, and, inparticular, military expenditures—have apparentlybeen created, at least in part, to avoid legislativescrutiny. Often, extrabudgetary funds that wereoriginally created for valid reasons have becomehighly dysfunctional. In some African countries, re-serves accumulated in commodity stabilizationfunds have been diverted to finance consumer subsi-dies or prestige projects. Instead of relying on activelabor-market programs or deregulation, a number ofEuropean and Latin American countries (exceptunder defined-contribution schemes) use social se-curity funds as a nontransparent means for alleviat-ing structural unemployment, through easy eligibil-ity for old-age and disability pensions. In somecountries, reform efforts are under way to limitearly retirement and excessive use of disability ben-efits and sick pay. Although efforts to do away withsuch dysfunctional practices should be stepped up,the immediate task is to make the activities of thesefunds transparent through publication of their oper-ations on a timely basis and in a manner that per-mits consolidation with budget operations.

16

APPENDIX I

39For example, in the United States, until 1992, the rescue offailed savings and loan institutions was treated off-budget.

40Other examples where transparency prevailed in certain as-pects of privatization include mass distribution of state-owned en-terprises (Czech Republic); earmarking of the bulk of proceedsfor debt reduction (Hungary); and rapid and open marketing ofassets held (former German Democratic Republic).

41In practice, such uncompensated quasi-fiscal operations tendto go hand in hand with tax exemptions, deferments of tax pay-ments, and other tax concessions.

42See the discussion of such practices in former centrallyplanned economies in Kopits (1991) and Tanzi (1993).

43Examples of well-functioning funds are the U.S. HighwayTrust Fund; Chile’s Copper Compensatory Fund; and the U.S.Old-Age, Survivors’, and Disability Insurance trust funds. Seealso Potter (1997).


One potentially important instrument for ensuringtransparency in government operations is a perma-nent independent review agency that is responsiblefor conducting studies on selected fiscal issues, per-formance audits—focusing on the efficiency and ef-fectiveness of government programs—and ap-praisals of the activities of specific public sectorinstitutions, including state-owned enterprises. Suchan agency, accountable to the legislature and thepublic at large, can be found in most industrialcountries, some developing countries, and a feweconomies in transition.44 Occasionally, some gov-ernments commission outside experts to preparespecial studies—such as white papers on selectedpolicy issues in Commonwealth countries that con-tain detailed factual assessments and nonbindingrecommendations for reform.

Budgetary Practices

The budget process, undertaken under varying de-grees of transparency, normally comprises four mainstages: formulation and preparation of the draft bud-get, legislation and approval, budget execution, andaudit and review. In some countries (e.g., Germanyand Switzerland), including an increasing number ofeconomies in transition, the tasks involved at eachstage are detailed in the budget framework law. Inothers (e.g., United Kingdom), the process has beenset by past conventions and rulings. In either case,the process can be said to be transparent if it followsclearly established guidelines rather than an impro-vised approach, is fully explained, openly debatedand enacted, and applied preferably to the entiregeneral government, including extrabudgetary oper-ations and local governments.

The first stage—preparation of the draft budget—involves internal debate and resolution of prioritieswithin the government. Although the actual formula-tion of budget proposals is an internal matter, thetimetable and decision-making process used by thegovernment should be publicly known. Of course,once completed, the budget and the supporting docu-ments (such as the budget speech and various spe-cialized analyses) need to be published and madewidely available in an understandable and meaning-ful format. Clear announcement of the fiscal targetembodied in the budget documents is seen as one ofthe most important aspects of transparency because,

by analogy with the inflation targets used in mone-tary policy, it conditions the expectations of privateand public sector participants.45

Following formal presentation of the budget, gov-ernment representatives are responsible for explain-ing—as openly and in as much detail as necessary—its contents and the underlying rationale andpolicies, often at legislative hearings. By their na-ture, the actual legislative debate and authorizationare normally open to the public, and the outcome isrecorded in public documents that become the foun-dation for the next stage. The execution stage re-quires both internal transparency (timely monitoringof information on actual spending and cash move-ments through the treasury) and external trans-parency (prompt reporting to the public). Finally, ad-equate information is to be made available forconducting both financial and performance audits,and the results, including any actions recommendedand implemented, are to be disclosed to the public.46

Whereas the foregoing transparency criteria are for-mally met in most countries that have separate execu-tive and legislative branches, the quality of informa-tion is often inadequate. Moreover, the specificationof fiscal targets, particularly medium-term targets,47

incorporated in the annual budget document is inmany cases vague. The linkage between these targetsand actual policy measures, on the one hand, and ulti-mate economic goals, on the other, is often unclear. Amajor problem stems from the reporting on the real-ization (or nonrealization) of fiscal targets. Many gov-ernments, especially in nonindustrial countries, fail toexplain fiscal developments in the course of the year;thus, public knowledge of slippages is delayed. Addedweaknesses in budget execution transparency stemfrom information lags and incomplete coverage,

17

Appendix I

44Examples are Australia’s Auditor General, the United King-dom’s National Audit Office, the United States General Account-ing Office, and Hungary’s Office of State Audit. In France, re-viewing and monitoring are carried out under the authority of theInterministerial Evaluation Committee. However, in a number ofLatin American countries, the comptroller general is responsibleonly for legal and financial audits.

45The innovations introduced in New Zealand (Appendix IV)instructing the government to declare fiscal targets consistentwith a set of prudent principles, enshrined in the Fiscal Responsi-bility Act, is an example of a high level of transparency. Recently,Australia adopted the Charter of Budget Honesty, requiring,among other things, clear announcement of fiscal targets and in-tergenerational analysis of fiscal policies. Although all membercountries of the Organization for Economic Cooperation and De-velopment (OECD) and a number of other countries have adoptedfiscal targets in one form or another, many are poorly specifiedand lack policy relevance in that they cannot be easily related togeneral economic goals (such as debt stabilization and desired na-tional savings levels). See Organization for Economic Coopera-tion and Development (1995a) for a critique.

46Although performance reviews (including quantitative perfor-mance indicators) of specific agency programs may also be pub-lished, for many governments this is largely an internal matter thatfeeds into the preparation of the subsequent budget. Thus, the re-view reports may be attached to the budget documents and canjustify the proposed expansion or reduction in specific allocations.

47For an overview of multiyear budgets and targets in OECDcountries, see Organization for Economic Cooperation and De-velopment (1995a) and Appendix III.


mostly in developing countries and economies in tran-sition. It is perhaps most difficult to achieve trans-parency in budget audit and review. Although a num-ber of Asian countries (e.g., India and Korea) havemade considerable strides toward openness, someother countries in the region and in Africa still exhibitnontransparency at various stages of the budgetprocess.48 Despite taking some important steps toshed a tradition of secrecy, most economies in transi-tion (with the exception of a few Central Europeancountries) need to make further progress toward trans-parency in the budget process as a whole.

In response to unanticipated exogenous develop-ments, most industrial countries follow clear guide-lines for using contingency reserves (e.g., Australiaand United Kingdom), resorting to cash limits (e.g.,formerly in the United Kingdom), or preparing a sup-plementary budget. By contrast, many economies intransition rely widely on discretionary steps, includ-ing sequestration of entitlements, often to contain anunbudgeted widening of the deficit in the course ofthe fiscal year.

A further test of transparency in budget executionand control involves public procurement, contracting,and employment practices. In the majority of OECDmember countries—at the national rather than at sub-national government levels—public tenders are is-sued for major government contracts, procurement ofsupplies above a certain threshold are subject to openbidding, and awards are publicly announced.49 Over-all, there is increasing reliance in these countries onthe contracting out of basic government services.Similarly, a number of countries that previously oper-ated closed career systems have introduced open re-cruitment. Although some progress has also beenmade in Latin America (e.g., Chile and Venezuela)toward increased transparency in this area, in manydeveloping countries, there is considerable room fordiscretionary purchasing practices and patronage inthe civil service.50 The lack of transparency in the

economies in transition51 is not surprising, given abackground of completely internal procurement prac-tices—especially during the shortages that prevailedunder central planning—and of hiring under thenomenklatura system.

Tax Treatment

In general, a transparent tax system is characterizedby relatively small differences between statutory andeffective tax rates on transactions, income, and prop-erty. The extent of such differences is determined bytax evasion, administrative shortcomings, or tax con-cessions. Hence, to reduce the differences, the taxbase should be well defined, and modificationsthrough preferential treatment should be based onclear statutory criteria of eligibility. Also, tax adminis-tration should be conducted in a clear and simplemanner, assisted by dissemination of all the necessaryinformation, including instructions for completing taxreturns. Discretionary tax relief provided to particularindividuals or enterprises (including state-owned en-terprises), as practiced in some developing countriesand virtually all economies in transition, constitutes adeparture from transparency. In these countries aswell as in a few industrial countries, taxpayers areable to accumulate substantial tax or social securitycontribution arrears, or overdue obligations, subject toalmost endless litigation—in some cases, even withtacit government approval. Least transparent, how-ever, is the case-by-case negotiation of tax liabilitiesbetween tax officials and corporate taxpayers, whichhas been made available mainly to foreign investors inmany developing countries. In some of these coun-tries, however, as well as in economies in transition,this approach has proliferated.

As a step toward transparent tax administration,most of the advanced economies have promulgatedlaws or statutes specifying taxpayers’ rights andobligations (Organization for Economic Cooperationand Development, 1990). Obligations include filingtax returns and making tax payments by the duedate; and filing information returns on wages, inter-est income, or payments for merchandise made bythird parties. Likewise, the authority to imposepenalties, including the search and seizure of assets,is legally defined. In many countries, tax officials areobliged to adhere to strictly enforced rules of con-duct; taxpayers who have the right to privacy mayturn to the administration or to the courts for protec-tion from arbitrary treatment. In some countries, tax-payers may appeal to a tax referee; the legal costs

18

APPENDIX I

48Although progress is being achieved in a number of Africancountries, where budgets are required to be openly debated in na-tional assemblies.

49In countries such as Australia, Canada, New Zealand, theUnited Kingdom, and the United States, these practices are highlyvisible. Recently, EU members have been required to adopt uni-form procurement standards. The development of open-tenderpractices across country borders has been stimulated by standardsset by the World Trade Organization. The OECD has also playedan important role in propagating such practices.

50The World Bank and other multilateral agencies have activelypromoted improvements in these areas. One potentially promis-ing initiative in the procurement area is the “Islands of Integrity”approach pursued by Transparency International, a nonprofit or-ganization. Under this scheme, endorsed by the World Bank,Transparency International monitors the bidding for public con-tracts; if either party—the national authorities or a bidder—isfound to engage in corrupt practices, that party is excluded fromany subsequent bidding.

51As an exception, effective in 1996, Hungary began applyingprocurement standards compatible with those adopted by the EU.


are covered if the taxpayer’s appeal is successful.Such a clear definition and enforcement of rules andregulations are absent in many developing countriesand most economies in transition.

Tax preferences can be viewed as tantamount tobudgetary outlays benefiting households or specificeconomic sectors or activities. For this reason, esti-mates of tax expenditures—that is, loss of revenuecompared with the level that would obtain in the ab-sence of tax allowances—can provide valuable infor-mation to the legislature and the public, particularlyat the time of the annual budget debate or during taxreform, despite the difficulty of defining the refer-ence value for each tax expenditure category. At pres-ent, one dozen OECD member countries regularlypublish expenditure accounts that show the estimatedbudgetary cost of tax exemptions, deductions, cred-its, deferrals, and reduced rates.52

Similarly, useful information about the distribu-tional implications of a country’s tax subsidy systemcan be gleaned from incidence studies, albeit withsome methodological and measurement limitations,including those associated with the underlying house-hold budget surveys. In some industrial countries(e.g., Australia), the government periodically pub-lishes tax subsidy incidence calculations, often pre-pared under contract by nongovernmental organiza-tions. In addition, of course, it would be informativeto occasionally estimate differences among statutory,effective, and realized tax rates—especially for taxeson company income and foreign trade.53

Financing Operations

Besides its adverse monetary and macroeconomicconsequences, nonmarket bank financing of govern-ment deficits also conceals the true cost of debt ser-vicing. For this reason, in the past two decades, anincreasing number of countries have resorted to mar-ket-based, nonbank financing of deficits, accompa-nied by a parallel effort to create an arm’s-length re-lationship between governments on the one hand andcentral banks and other financial institutions on theother. Thus, apart from a few economies in transitionand some developing countries, public financial in-

stitutions are no longer a captive source of deficit fi-nancing at below-market interest rates. However,such practices still account at least partially for thedomestic borrowing requirement of the public sectorin certain large countries.

The trend toward market-based financing opera-tions has enhanced transparency and received thesupport of financial markets. In fact, governmentswishing to access international markets must furnishrating agencies, underwriters, and security marketsupervisory agencies with considerable data on theirdebt stock, including guarantees, debt-service capac-ity, and overall debt-management strategy—as wellas information on macroeconomic developments andoutlook. For successful domestic financing, govern-ments need to provide information on marketingtechniques, the mix of domestic and external matu-rity structure and other characteristics of securitiesto be issued, and rules governing public tenders andopen market operations.

Nonetheless, certain subtle nontransparent debt-management practices can be found in some indus-trial and developing countries. For instance, strongmoral suasion by the authorities on institutional in-vestors and discretionary selection of underwriters,financing consortiums, and primary dealers are ex-amples of less visible practices. In addition, insuffi-cient or excessively complex public information onthe terms of acquisition or sale of various types ofgovernment securities, coupled with a lack of ade-quate retail outlets or secondary markets, may con-strain proper risk evaluation and, hence, the develop-ment of markets for such instruments. In countrieswith regulated markets, transparency would beserved by providing estimates of the cost of such re-straints in budget documents to show the degree ofsubsidization and consequent resource misallocationintroduced by such practices.

In government lending, the least transparent prac-tices prevail in some economies in transition and partsof Asia and the Middle East, where the authorities al-locate credit directly.54 On-lending by governments orby public (or, in some cases, private) financial institu-tions, often through government guarantee, can beequally nontransparent.

Regulation

Government regulation is often justified for a vari-ety of objectives: public health, consumer protection,environmental protection, labor protection, and moralhazard prevention. In many countries, however, the

19

Appendix I

52In Austria, Belgium, France, Portugal, Spain, and the UnitedStates, the government is legally obliged to prepare an annual taxexpenditure report, whereas in Australia and Germany, reports areprepared twice a year; see Organization for Economic Coopera-tion and Development (1996c).

53See, for example, Kyrouz (1975) and King and Fullerton(1984) on company income taxation. Whereas realized (average)rates are calculated from actual tax revenue as a proportion of ac-tual income or transaction flows, effective (marginal) tax rates areestimated on an economically meaningful base (e.g., user cost ofcapital or required rate of return).

54In China, for example, the consolidated budget deficit for1995 is to be adjusted upward by about 5 percent of GDP whenpolicy lending by public financial institutions is included.


20

APPENDIX I

complexity, overlapping jurisdictions (between na-tional and subnational levels), frequent changes, andsheer volume of regulations all contribute to a lack ofclarity in the regulatory framework. Moreover, regula-tions are sometimes used as a nontransparent substi-tute for budget transfers or taxes. For example, rentcontrols can be viewed as a substitute for transfers tolow-income households; zoning regulations, obliga-tory military service, and training requirements foremployers are substitutes for various forms of taxa-tion (Tanzi, 1995). Similar substitution can also arisewhen a government department or agency (e.g., de-partment of forestry) is responsible for both regula-tion and budget spending, given that managers mayface conflicting pressures to achieve fiscal savingsthrough the use of (less transparent) regulation.

Unlike public services financed from the budget,the costs of complying with regulations, some bornedirectly by the regulated households and enterprisesand some borne indirectly by the rest of the economy,are not transparent. An attempt to assess these costssuggests that the extent of nontransparency in regula-tion can be considerable, especially in industrial

countries.55 In developing and transition economies,with greater latitude for evading regulations throughthe informal sector, these direct costs tend to be off-set, at least in part, by costs associated with payoffsto the regulators themselves and to intermediarieswho assist in evasion.

In general, the recent—practically worldwide—trend toward deregulation in commodity, labor, andfinancial markets has contributed to greater trans-parency in government operations in these areas. Al-though difficult to quantify, considerable progresstoward deregulation has been attained by a numberof advanced economies (e.g., Denmark, NewZealand, United Kingdom, and United States).56 Insome of these countries, however, there is a risk thatthese gains may be undone by renewed substitutionof regulation for fiscal measures. While some devel-oping countries in East Asia and Latin America (e.g.,Chile) have embarked on major deregulation efforts,most economies in transition have made an impres-sive evolution from central planning—arguably themost overregulated system.

55According to Hopkins (1996), in the United States, the directcosts of federal regulation since 1992 have been estimated atabout 9 percent of GDP, equivalent to nearly one-half of federalbudget outlays.

56See Organization for Economic Cooperation and Devel-opment (1992) and Koedjik and Kremers (1996). Within the EUthere has been considerable progress toward harmoniz-ing regulations, but not always toward achieving greater transparency.


T ransparency in the measurement of governmenttransactions, ownership, and obligations rests

on a set of basic attributes: institutional coverage,recording basis, valuation, and classification offlows and stocks. This appendix highlights thosecountry practices that are consistent with, or deviatefrom, acceptable standards. Although in many re-spects they are indistinguishable from those appliedin the private sector, these standards must conformprimarily to the requirements of fiscal policy analy-sis. In assessing transparency in measurement, onemust keep in mind the limitations of human capitalin agencies charged with compiling government fi-nance statistics, especially in some developingeconomies and economies in transition.

CoverageIn principle, the general government is universally

regarded as the most comprehensive definition of thesector that “performs primarily the functions of sup-plying certain public goods and services and fulfill-ing certain public purposes not for commercial or fi-nancial reasons, or, if of a commercial or financialnature, not on a major basis or not primarily for aprofit” (International Monetary Fund, 1986, p. 7). Itscoverage encompasses the national (central) govern-ment consolidated with all subnational (provincial,municipal, or local) governments, consisting of thebudget, as well as social security and other extrabud-getary accounts, at each level of government. Al-though nearly all countries can supply up-to-date in-formation on the central budget—which for many isthe main focus of policy—only in about two-thirdsof the member countries of the Organization forEconomic Cooperation and Development (OECD)are timely data for the consolidated general govern-ment available on a yearly basis.57

Many countries, including some major ones, do notprovide comprehensive and timely information on ex-

trabudgetary activities. Particular shortcomings in-clude the inconsistency and lags in social securitydata in countries where such institutions enjoy a highdegree of autonomy. Moreover, because the data onthe operations of subnational governments have notbeen compiled and consolidated with those of thecentral government in a timely fashion, full coverageof the general government has not been possible. Forsome countries, large and small, the sheer number ofsubnational governments poses a major problem. Forothers, consolidation is made difficult by the need toattribute a large number of hard-to-identify intergov-ernmental fiscal transfers to grantor or recipient levelsof governments. Also, data collection may be inhib-ited by legal restraints or by the complexities associ-ated with the interposition of an additional regional orprovincial tier of government in the subnational struc-ture. However, in some developing countries andeconomies in transition, the main difficulty stemsfrom the lack of consistent recording and from delaysin information, which, in turn, can be traced largely tothe shortage of well-developed financial managementsystems at the subnational level.

Further evidence of the difficulty of providingfull and timely coverage is that in 1995 only 40 IMFmember countries supplied internationally compa-rable information on consolidated central govern-ment accounts for the most recent year; an addi-tional 30 members provided data with a one-yearlag; another 32, with a lag of two or more years; and13 provided recent data covering only the centralbudget. Of these 115 countries, 45 report timelydata on social security institutions, and fewer than20 provide up-to-date information on subnationalgovernment operations.58

In practice, even the general government fallsshort of full coverage of fiscal operations because it

Appendix II Transparency in Public Accounts

21

57Such data are published in Organization for Economic Coop-eration and Development (1996b) for 14 countries for the mostrecent year, for 3 countries with a one-year lag, and for 4 coun-tries with a lag of two or more years.

58See International Monetary Fund (1995). Incomplete cover-age of general government is a major impediment to timely re-porting under the IMF’s Special Data Dissemination Standard. Ofthe 42 members that intend to participate in the Special Data Dis-semination Standard, most have indicated that they will meet therequest for up-to-date summary data on central government oper-ations and debt, but about one-half may not be able to supplytimely data on general government.


excludes quasi-fiscal activities (including the costof regulations) of public financial institutions andnonfinancial enterprises. To account fully for director indirect government ownership of nonfinancialpublic enterprises, some countries, particularly inSouth America (e.g., Brazil, Peru, and Venezuela),publicly report data for the nonfinancial public sec-tor; others (e.g., Costa Rica, New Zealand, Portu-gal, and United Kingdom) have expanded coverageto official financial institutions as well, that is, en-compassing the entire public sector.59 Albeit a stepin the right direction, this broad coverage may beviewed as a rough approximation of the true magni-tude of government functions, whether undertakeninside or outside the general government sector. Infact, such coverage cannot be used to identify thesource of losses incurred by public enterprises orfinancial institutions, which may reflect a cyclicaldownturn or poor business decisions, or, alterna-tively, the cost of quasi-fiscal activities. However,in a significant number of developing countries andeconomies in transition, there is little, if any, publicreporting of quasi-fiscal activities by the state-owned enterprise sector or financial sector.

The concept of general government activity, pro-posed in the revised Government Finance StatisticsManual, is intended to accurately capture the cost ofall government functions, including quasi-fiscal ac-tivities conducted outside the general government.60

Although still at a conceptual stage, this coveragewould provide a useful complement to institutionallybased measures of general government and deserveswidespread application in the future. However, themeasure does not remove the need to separately costquasi-fiscal activities whenever possible. If these ac-tivities prove impossible to quantify,61 listing themwould aid transparency.

Recording Basis

How government transactions are recorded,namely, on a cash or an accrual basis, is critical forthe transparency of fiscal performance. On a cash

basis, transactions are recorded when financial re-ceipts or payments actually take place, even if thereis no associated real economic flow.62 This approachcan be useful for two reasons. First, it is preferredfor short-run budgetary control and analysis.Whereas it may be difficult to alter budget commit-ments or payment obligations in the short run, it maybe relatively easy to record and monitor cash pay-ments through treasury operations or, for example,to enhance revenue through an accelerated sale ofassets. Second, cash-based accounts have been usedtraditionally to assess the first-order financial impactof the government borrowing requirement, if mone-tized, on inflation and on the external balance. Bycontrast, on an accrual basis, all economic transac-tions are recorded as they occur, regardless ofwhether the transaction is an in-kind or a for-cashpayment.63 Because transactions are recorded whenthey reflect changes in government assets and/or lia-bilities, including depreciation of plant and equip-ment, this approach permits full consistency be-tween the government’s income statement andbalance sheet.64 Accordingly, accrual-based ac-counts are indispensable for gauging the macroeco-nomic repercussions of fiscal policy—as well as theresource implications for the management of spe-cific government programs—especially over themedium to long term.

In practice, neither the cash nor the accrual methodneeds to be adopted in a pure form; indeed, somecountries find it attractive to pursue a modified cashor a modified accrual approach.65 In countries thatrely on the cash approach alone, the recording of cer-tain important government transactions can be eitherdistorted or nonexistent. Major areas where nontrans-parency may arise would be the exclusion of infor-mation on the accumulation of arrears, transactionsin kind, and the cost of borrowing at a discount.

Measurement of revenue and expenditure arrearshas become a problem, especially in countries wherethe tax or social security administration may facelegal or political impediments in collecting taxes orpayroll contributions, or where cash limits or se-questration of funds is used to contain expendi-

22

APPENDIX II

59Among these countries, only New Zealand has applied theformal consolidation rules required for “whole of government”reporting.

60According to International Monetary Fund (1996a, p. 12), theconcept “includes all government functions carried out by generalgovernment units and units outside the general government sectorand excludes nongovernment functions carried out by generalgovernment units.”

61For example, the cost of sectoral credit ceilings used for direct-ing credit to specific activities would require calculating implicittaxes and subsidies relative to an unconstrained outcome. Accord-ing to estimates in Tanzi (1995), the cost of financial suppressionthrough regulation in past years amounted to as much as 40 percentof total revenue in Mexico and 20 percent in India and Pakistan.

62A variant of the accrual basis is the due-for-payment basis,whereby the transaction is recorded when receipt or payment fallsdue.

63Specifically, recording takes place at the time economic valueis created, transformed, exchanged, transferred, or extinguished.

64For a discussion of the usefulness of accrual-based govern-ment accounting, see Efford (1996) and International Federationof Accountants (1997).

65For instance, under a modified cash approach, changes infloating debt can be shown as a memorandum item. In another ex-ample, under a modified accrual approach, only payment or taxarrears within a certain time limit are recorded in the financialstatement. See the general treatment in International Federationof Accountants (1995, 1997).


tures.66 Exclusion of expenditure (tax) arrears leadsto an understatement (overstatement) of governmentdeficits.67 Cash-based recording can, of course, besupplemented with memorandum items showingchanges in floating debt that account for the differ-ence between payment orders (tax liabilities in-curred) and cash or checks issued (collected). How-ever, payment orders issued often do not cover thefull range of commitments made, and it may not befeasible to eliminate normal lags in processingchecks; hence, floating debt cannot be regarded as ameasure of genuine arrears.

Transactions in kind include certain grants andloans extended by one country to another and, some-times, by one level of government to another; grantsare more important in developing countries, andloans are more important in transition economies.Through such operations, a donor country may pur-chase goods and services and then pass them directlyto the recipient or pay an outside commercial sup-plier to provide goods and services on its behalf.68 Inaddition, barter transactions, which were frequentamong former centrally planned economies (and vis-à-vis their free market trading partners), still occuroccasionally in some economies in transition. Othernoncash transactions include the issuance of govern-ment obligations—often in the form of securities—as payment for services received from suppliers,which, in turn, can be used to offset tax arrears or fortax refunds due to taxpayers; in a different context,tax arrears were written off with the transfer of eq-uity to the government. In each case, the immediateeffect of the transaction is an understatement of thetrue deficit.

A nontransparent accounting practice may arise inconnection with noncash issuance of governmentdebt, particularly in bank restructuring. Regardlessof the recording basis, all interest payments involvedin the bank recapitalization are to be shown as gov-ernment expenditures. But under the accrual ap-proach, the principal component of transferred gov-

ernment paper (less the market value of any nonper-forming loans assumed by the government), if it isnot accompanied by an increase in government eq-uity, should also be recorded as expenditure. In ei-ther case, the net value of the transfer would con-tribute to an increase in government net debt. Whilealways recording interest payments as expenditure, anumber of countries have failed to show the princi-pal component of the net value of bond transfers asexpenditure. (For a review of country experiences,see Alexander and others, 1997.) Under this princi-ple, any government debt issued in exchange for pre-vious outstanding doubtful liabilities, including ar-rears, should be recorded as an expenditure in theamount of the difference in value between the newdebt and the old debt.69

Several governments issue discount securities,with a zero or a low coupon rate of interest. Thetransparency implications of these securities undercash recording depend on the exact formulation of in-terest and repayment terms. In most cases, govern-ments can permanently reduce the cash measure ofinterest payments. The cost of the borrowing is in-stead placed “below the line” as amortization pay-ments. In some formulations, amortization can alsobe delayed by requiring a “bullet” repayment at ma-turity. A similar understatement of interest cost maytake place with bonds that carry a fixed interest rate,although the principal is indexed to inflation with theeffect of understating cash expenditures.70 In all thesesituations, under accrual-based recording, all interestcharges, including the indexation component,whether paid at maturity or not, would be shown con-tinuously during the life of the instrument.

Accrual-based recording has some potential short-comings, above and beyond the limitations men-tioned above. An often-cited deficiency involves themeasurement of accrued tax liabilities so overduethat they can no longer be regarded as collectible be-cause the liable enterprise may already have beenliquidated or declared bankrupt. Such overdue taxesor payroll contributions for social security, oftensubject to unending litigation, cannot reasonablyqualify as arrears. The preferred solution on a modi-fied accrual basis is to measure all tax and contribu-tion revenue on a cash basis.

Although cash is the predominant basis of record-ing worldwide, a number of countries have sought tocomplement it with, or shift altogether to, a modifiedaccrual basis, broadly consistent with international

23

Appendix II

66Advance tax payments (or other methods of accelerating rev-enue) have, of course, an effect similar to that of accumulatingexpenditure arrears, in that they temporarily improve the budgetbalance under cash-based recording. An analogous case of non-transparency is the recording of extraordinary—though re-quited—transfers from state-owned enterprises as revenue in thebudget, without an offset for the assumption of (e.g., pension) lia-bilities from those enterprises by the government in an equivalentamount.

67A similar practice, found, for example, in Italy and Sri Lanka,involves recording expenditures that have not been incurred be-cause of suspended appropriations. The unused cash is depositedby the spending agencies in government accounts to boost thecash available to meet the public sector borrowing requirement;see Premchand (1994).

68See Premchand (1996) for a critique of the problems createdfor fiscal accounting.

69According to Teijeiro (1996), this has been an importantsource of understatement of the officially recorded budget deficitin Argentina in the early 1990s.

70Whereas it may be desirable to exclude the inflation compo-nent of interest payments for calculating an operational measureof the budget balance (Appendix III), such an adjustment shouldbe executed uniformly, not just for certain indexed securities.


accounting standards.71 So far, only New Zealandhas moved to a comprehensive system of accrual-based budgeting, management, and recording; a fewothers have adopted at least partial accrual recording(e.g., France, Greece, Iceland, Mali, Spain, and Swe-den) or intend to move to a full or partial accrual-based approach (e.g., Canada and United King-dom).72 More generally, there is increased interest inaccrual-based recording, including steps to reconcileexisting cash accounts to an accrual basis, partly forgreater transparency.73

Valuation and Recognition

Accrual-based recording prescribes certain stan-dards for valuation and recognition of assets and lia-bilities, with implications for the transparency ofgovernment financial statements. In this context, thetreatment of capital assets and contingent liabilitiesdeserves discussion.

The valuation and recognition of capital assets maybe subject to varying treatment depending on the ana-lytical purpose at hand. Accounting fully for invest-ment spending when it is incurred captures the impacton aggregate demand, the focus of macroeconomicstabilization. By contrast, from the standpoint of re-source use, only capital depreciation during the periodshould be shown, regardless of when the acquisitionof fixed assets takes place.74 However, this perspec-tive—in line with the so-called generally accepted ac-counting practices applicable in the private sector,with full consistency in the calculation of the operat-ing balance and changes in net worth—is limited for

the government because of the difficulty of measuringthe depreciation of fixed assets in an economicallymeaningful way.75 Indeed, reliable measurement ofthe value of government assets in infrastructure,given, at best, a rather thin secondary market for suchassets, is likely to pose a major difficulty in mostcountries. Hence, in some cases, it may be necessaryto resort to somewhat arbitrary solutions.76

In view of the unsolved valuation difficulties—be-sides the obvious distinction between governmentand private enterprises, whose objective function isto maximize net worth—the reluctance to preparefull government balance sheets is not surprising. Theadvanced economies, some developing economies inAsia and Latin America, plus a few economies intransition (Czech Republic, Hungary, and Poland)prepare timely data on gross public debt. Althoughpractically all IMF member countries report data ontheir official international reserves, only 42 countriesreport up-to-date information on gross public debt.Most of these countries limit the compilation of gov-ernment debt to financial obligations servicedthrough the central budget.77 (Net government liabil-ities can then be calculated by deducting financialassets, in the form of securities and foreign assets,from gross debt.) In a few countries (New Zealandand United States), an effort is under way to publisha comprehensive balance sheet for the public sectoras a whole. For these and other countries, the mea-surement of balance sheet information has been in-fluenced by both the generally accepted accountingpractices issued by private accounting bodies78 andby the national accounting standards.79

Besides correcting the market value of actual lia-bilities for inflation, public sector balance sheetsare affected by the recognition and valuation ofcommitments and contingent liabilities. The impor-tance of accumulating such liabilities as a vehiclefor spending is underscored when they are substi-tuted for actual government expenditures but notreflected in the fiscal accounts. Most governmentsundertake commitments (e.g., to multilateral insti-tutions) and provide guarantees for credits ex-

24

APPENDIX II

71See the revised version in Inter-Secretariat Working Group onNational Accounts (1993).

72In Australia, several subnational governments employ ac-crual-based recording, and the Australian National Commissionof Audit (1996) has recommended this approach for the nationallevel. In the United States, the Federal Accounting Standards Ad-visory Board (FASAB) has been responsible for examining gov-ernment accounting practices and issuing recommendations. Foran early recommendation to adopt accrual-based recording, seeUnited States, President’s Commission on Budget Concepts(1967).

73For example, member countries of the European Union (EU)are required to show their financial statements for the generalgovernment on a modified accrual basis for determining compli-ance with the fiscal criteria under Economic and Monetary Union(EMU); see European Commission (1995). In a recent survey ofGFS data users, undertaken by the IMF’s Statistics Department,87 percent of respondents expressed a need for cash-based infor-mation, whereas 47 percent indicated that accrual-based data arealso required; see Efford (1996).

74An advantage of this approach may be apparent in budget-cutting exercises. Under a cash approach, there is an incentive tocut expenditures that will affect the current year’s budget,whereas accrual-based recording removes the bias of cutting in-vestment outlays because only the capital used in the year isshown.

75See Comiez (1966) for an early discussion of depreciation ofgovernment assets, and Coen (1975) for an analysis of economicdepreciation in general.

76For example, the guidelines issued recently by FASAB havebeen adopted by the U.S. government.

77In some countries, government liabilities are consolidatedwith those of the central bank, as the latter acts as the govern-ment’s fiscal agent in external debt management.

78At the international level, the Public Sector Committee of theInternational Federation of Accountants (1996) has proposedguidelines on public sector accounting.

79That is, the standards embodied in the System of NationalAccounts issued by the Inter-Secretariat Working Group on Na-tional Accounts (1993), which encompass public sector balancesheets.


tended by financial institutions (e.g., for businessinvestment, commodity sales or purchases, and ex-ports) for direct investment overseas and for bankdeposits. Often, however, the motive for extendinginvestment or loan guarantees—mainly to indus-tries in difficulty—is the nontransparent postpone-ment of budget outlays. In addition, governmentsprovide social insurance for old age, unemploy-ment, and health care—in most countries, againstpayroll-based contribution payments.

Although these liabilities represent a potential fu-ture obligation for the government, contingent on therealization of the insured occurrence—and thus theneed to estimate their unfunded portion—they arenot comparable or additive to actual governmentdebt. In general, neither the funded nor the unfundedportions of contingent liabilities relating to social en-titlement schemes should be included in the govern-ment balance sheet; likewise, changes in such liabili-ties should not be entered in an accrual-basedincome statement of the government. However, de-spite some measurement problems, the actuarialvalue of net unfunded liabilities for government em-ployees’ pensions should normally be included in thecalculation of government liabilities in the balancesheet—much like for private enterprises—becausethey represent a strictly enforceable contractualobligation. This practice, however, is not always fol-lowed, even in some major industrial countries.Leaving the arguments for estimating net unfundedgovernment liabilities to Appendix III, it is worthnoting here that the measurement of such liabilities issubject to certain technical limitations.80

Many countries that compile government debt sta-tistics show separate data on the unfunded portion ofcontingent liabilities and commitments other thanthose related to entitlements, specifying the nature ofthe commitments and contingent liabilities. How-ever, these data are not formally included in a state-ment of government liabilities outstanding unlessthey can be quantified and definitely identified as re-quiring future expenditure. Although most commit-ments can be quantified, contingent liabilities aremore difficult to measure, especially when they arenot incurred with the expectation that they will haveto be met at some future date. Rather, they are a formof indemnity for the entity receiving the guarantee,which may not be realized. Including the value ofsuch guarantees in the balance sheet requires an as-sessment of the likelihood of default or some otherevent likely to require realization of the liability,which in many cases can be done with acceptableaccuracy.81

Classification

Fiscal transparency requires that data on govern-ment operations be classified into analytically usefulcategories of flows and stocks. Short of an exhaustivediscussion of classification issues, some areas wherenontransparency is most prevalent are examined.

First, the breakdown of debt service between inter-est and amortization payments—and less frequentlyof certain nontax revenue flows between unrequitedtransfers and borrowing—is not always available inpublished government financial statements. Giventhe indistinguishable nature of such transactions, es-pecially between unrequited transfers and financing,and the ambiguous nature of debt servicing under theformer system of central planning, it is still difficultto draw the distinction between “below the line” and“above the line” in a number of countries in transi-tion. There are also some developing countries thattraditionally classify certain financing items abovethe line. An important distortion in measuring fiscalpolicy, in countries engaged in large-scale privatiza-tion of state-owned enterprises, is the inclusion ofgross receipts from privatization with other nontaxrevenue in national presentations.82

Second, separation of tax and nontax revenue canbe a contentious issue, as is the occasional misclassi-fication of revenue from capital taxation as a form ofcapital revenue, especially in some economies intransition. Misclassification of revenue by type oftax is rare, except for various types of income taxa-tion, which, at times, is ambiguous. More frequent,however, is the unintended failure to disaggregaterevenue at the subnational level of government. Of83 countries reporting up-to-date revenue data to theIMF, only one-third provided any disaggregation atthe subnational level.

Third, deficiencies abound in the classification ofgovernment expenditure. Without an adequate break-

25

Appendix II

80For a review of specific technical difficulties encountered inprojecting pension obligations, see Van den Noord and Herd(1995) and Franco (1995).

81For the United States, for example, in 1997, the governmentbalance sheet contains an assessment of the accrued cost of de-posit insurance and private pension scheme guarantees. Whileknown losses on loan guarantees or other contingencies are in-cluded in most countries, Italy includes estimates of likely losseson loans, legal claims, and the like. If such estimates cannot bemade, some countries (e.g., Netherlands) list contingencies in aseparate statement. In New Zealand, for example, all quantifiableand nonquantifiable contingencies are so listed. Quantifiableclaims include loans guaranteed for public corporations, callablecapital to international and other organizations, prospective costsof contract disputes, and legal claims. Nonquantifiable claims in-clude natural disasters, exchange rate losses incurred by the Re-serve Bank of New Zealand, losses incurred as a result of court de-cisions, and indemnities arising from various regulatory practices.In Canada, the matter is handled through formal provisioning offunds against identified contingent losses likely to be realized.

82Under the modified GFS classification, privatization receiptsare to be shown below the line to remove the impact of suchtransactions from the measurement of the thrust of fiscal policy.


26

APPENDIX II

down of spending, it is difficult to undertake a mean-ingful fiscal adjustment plan in terms of the composi-tion of adjustment. Nearly all countries rely primarilyon an institutional classification of spending83 for pur-poses of budget preparation, legislation, and execu-tion. Only for broad policy planning and analysis domost industrial countries and some developing andtransition economies use an economic classification—largely subject to the shortcomings of cash-based ac-counting. Again, of the countries that reported up-to-date information on government expenditure byeconomic category, only about one-third did so at thesubnational level. Deficiencies in this classificationwere especially evident in some economies in transi-tion.84 However, finer distinctions between currentand capital outlays tend to be arbitrary even in themost advanced countries, some of which have tradi-tionally failed to separate out capital expenditures inbudget statements. Difficulties can also be found inthe classification of the wage bill as distinct fromother payments for services.

Far more problematic for most countries is tomove from an institutional classification to a trulyfunctional disaggregation of outlays. As an example,although education and cultural spending may begrouped under the same ministry for legislative pur-poses, these two categories should be shown underseparate functions. Relatively few countries are able

to approximate satisfactory international standardsin this regard, especially at the subnational level.Only 66 countries were able to supply the IMF withsuch a classification at the central budget or govern-ment level, and about 20 countries at the subnationallevel. Particularly helpful are ongoing reforms insome member countries (e.g., Italy and United King-dom) to align the management and control of gov-ernment operations on the basis of a functional clas-sification of expenditures. For analytical purposes, itis useful to separate discretionary spending from ex-penditure under mandatory programs, insofar as thelatter may be altered only through structural reformover the medium to long term.

Finally, disaggregation of financing flows and thecorresponding debt stocks by currency denomina-tion, maturity of (securitized) instrument, and source(domestic bank, nonbank, external) is available foronly a limited number of countries. Of particular dif-ficulty is establishing consistency between financingflows and the stock of outstanding debt classified ona uniform basis and between financing and revenueand expenditure above the line. Part of the difficultystems from the diversity of data sources. Whereasbelow-the-line data on financing are derived mainlyfrom monetary and balance of payments accounts,compiled (for the calendar year) by the central bankor the statistical office, the above-the-line data ongovernment operations are usually recorded (for thefiscal year)—sometimes using a different recordingbase—by the responsible fiscal authority. Only 40countries were able to provide timely data on financ-ing flows and corresponding debt stock by type ofinstrument and creditor at the central governmentlevel; of these countries, only a few did so at the sub-national level.

83For an informative critique of the traditional administrativebudget in the United States, see United States, President’s Com-mission on Budget Concepts (1967).

84The lack of sufficient disaggregation in the statia codes, for-merly used to compile economic information in the countries ofthe former Soviet Union, led to a large share of expendituresbeing lumped into a very large “other expenditure” category; thecodes also often failed to distinguish between nonrepayable andrepayable payments, that is, between expenditure and net lending,social security contributions, and transfers to various economicsectors. See International Monetary Fund and others (1991) andMontanjees (1995).


Building on the discussion of measurement is-sues, this appendix focuses on the transparency

of summary indicators of fiscal policy and perfor-mance, as well as of projections of fiscal aggregates.It does not provide an in-depth review or a compara-tive assessment of fiscal indicators and projections,but examines the transparency of various countrypractices in this area.

Direct Indicators

The most commonly available fiscal indicator isthe overall balance of government operations, pub-lished by nearly all IMF member countries on a rou-tine (at least yearly) basis. As a first approximation ofthe impact of the budget on domestic demand, this in-dicator is subject to all the distortions in coverage,recording, and classification found in each country’sfiscal accounts discussed in Appendix II. Amongthese distortions, the most prevalent and significant is the exclusion of extrabudgetary operations—particularly social security accounts—and of quasi-fiscal activities. Apart from the above nontransparen-cies in measurement, some countries present the bud-get balance by excluding or misclassifying some orall of the economic flows associated with the externalsector. Others tend to significantly understate the truedeficit and fail to report data on deficit financing. Forpurposes of policy formulation, certain countries(e.g., Australia and European Union (EU) members)intend to follow the more transparent definition of an“underlying balance,” which excludes one-off trans-actions, especially privatization, even if they other-wise rely on cash-based recording.85

Measurement of subsets of the overall balance de-pends mainly on the adequacy of economic classifi-cation, as discussed above. Calculation of the currentbalance as an indicator of the government’s contri-

bution to national saving requires a clear separationof capital and current transactions, which is difficultto achieve without a satisfactory measurement ofgovernment investment expenditure. Furthermore, ifconventionally measured, government investmentmay not provide a meaningful gauge of the long-term effect of public spending insofar as it includessome nonproductive capital structures (e.g., execu-tive palaces and monuments) yet excludes currentexpenditures on human capital through educationand health care. Notwithstanding these caveats,many developing countries show separately a capitalor development budget balance in the context oftheir development plans.

Similarly, the primary balance, calculated to deter-mine the fiscal policy effort necessary to stabilize orreduce public debt, requires that net interest paymentsbe excluded from the overall balance. In some coun-tries, interest payments cannot be readily separatedfrom capital transactions or from financing flows. Anumber of countries (e.g., Argentina, Italy, and UnitedStates) include calculations of the primary balance inbudget documents. Measurement of the operationalbalance entails both identification of interest pay-ments and deduction of the corresponding inflationarycomponent of those payments from the overall bal-ance. Notwithstanding some potential measurementproblems,86 the operational balance is routinely calcu-lated, as a supplement to the overall balance, in coun-tries that have experienced high inflation and largedomestic indebtedness (e.g., Brazil and Israel). How-ever, some advocates of the operational balance con-cept argue that even countries with moderate inflationshould apply this concept to avoid adopting an inap-propriate fiscal policy stance.87

Appendix III Transparency in Indicators and Projections

27

85To determine compliance with the fiscal criteria for participa-tion in European Economic and Monetary Union (EMU), EUmembers are required to exclude proceeds from privatization—whether undertaken directly or through a state holding com-pany—from government revenue; see European Commission(1995).

86As nominal interest rates reflect expected inflation—to com-pensate the lender for the loss in capital value—the inflationarycomponent should be calculated in reference to the expected rateof inflation rather than the recorded rate of price inflation.

87In the United States, during 1977–80, on the basis of an esti-mate that the real value of net government debt declined by $82 bil-lion, generating a cumulative surplus in that amount (whereas theofficially recorded deficits totaled $96 billion, at constant prices),Eisner (1986) argued that the authorities aggravated the ensuing re-cession by adopting a restrictive monetary and fiscal stance.


Some countries with a traditionally large publicenterprise sector (e.g., Italy, Portugal, and UnitedKingdom) report the public sector borrowing re-quirement on a cash basis to assess the net impact ofthe public sector on financial markets, implicitly tak-ing into account the net effect of quasi-fiscal activi-ties of state-owned enterprises and financial institu-tions, but without revealing the underlying grossflows. Although occasionally subject to debate andto varying treatment, for consistency, privatizationreceipts are often removed from the public sectorborrowing requirement.

Given the difficulties of valuing government assetswith precision, gross debt is widely regarded as amore reliable and internationally comparable indica-tor.88 Alternatively, by subtracting only negotiable fi-nancial assets—mainly domestic securities, gold, andforeign exchange reserves—the net debt ratio couldbe an acceptable variant.89 The most ambitious re-finement, of course, is the calculation of governmentnet worth. Perhaps because of the ambiguities in val-uation and recognition of certain assets and liabili-ties, only New Zealand has begun to publish a com-prehensive and consistent set of statements of theoperating budget balance and net worth for the publicsector, following the generally accepted accountingpractices. (See New Zealand Treasury, 1996.)

Analytical Indicators

Measures of actual government balance and debt,or their variants, as a proportion of GDP may be use-ful as a first approximation of ex ante fiscal stanceand sustainability or of ex post fiscal performance.However, given the inherent limitations of such di-rect indicators, some countries have adopted a num-ber of more focused analytical indicators, either as apermanent feature or at least on an experimentalbasis.

To ascertain whether fiscal policy is expansion-ary, contractionary, or neutral, various summary in-dicators of fiscal policy stance have been developedby removing the effect of cyclical fluctuations orexogenous shocks from the direct measures of bud-get balance. The simplest analytical measures offiscal stance are those based on the full employ-ment surplus (developed in the United States) orthe cyclically neutral balance (Germany) and, alter-

natively, of fiscal impulse, measured by the changein the cyclically adjusted effect of the budget (asoriginally published by the Netherlands). All arecompatible with virtually any macroeconomicmodel and impose rather modest additional data re-quirements. They require information on the nat-ural rate of unemployment or on a satisfactorymeasure of the level of potential output. Calcula-tion of the structural balance requires, in addition,estimates of the elasticity of each major tax andtransfer category with respect to GDP, which underthe cyclically neutral balance are constrained tounity (see Heller, Haas, and Mansur, 1986). Be-cause of the uncertainties surrounding these param-eters and estimates of potential output underlyingthe level of the structural balance, it is preferable tolook at the yearly change in the structural balanceas indicative of the discretionary thrust of fiscalpolicy. When the influence of inflation is removed,the operational structural balance provides a usefulindicator of discretionary policy in times of infla-tion. Meanwhile, measures of the primary struc-tural balance abstract the combined influence of in-terest rates and the debt stock, on the one hand, andof cyclical factors, on the other.

Selection of the appropriate indicator for a givencountry depends on its stage of development, depthof financial intermediation, tradition, and, perhapsmost important, computational convenience. A fewgovernments, or their advisory agencies or councils,publish regularly (Canada, Denmark, Germany, andUnited States) or occasionally (France) an analyticalindicator of fiscal stance, whereas others prefer tocalculate such an indicator only for internal pur-poses.90 No comparable official calculations can befound in developing countries, because structuralshifts preclude reliable time-series estimates of po-tential output and GDP elasticities of tax revenue ortransfer expenditure. Of course, this problem is com-pounded for economies in transition, where the ex-traordinary contraction experienced in the first halfof the 1990s reflects the radical shift from a centrallyplanned level of output to a market-determined levelof output.

Beyond the short term, it is of much interest to pol-icymakers and financial markets to ascertain whetherfiscal policies or, rather, fiscal institutions—includingpublic pension and health care programs—are sus-tainable well into the future. There are only generaldefinitions of what constitutes sustainable fiscal pol-icy, and economic theory provides little guidance

28

APPENDIX III

88This largely explains the choice of general government grossdebt ratio to define the public debt criterion for participation inthe EMU.

89Along these lines, Canada subtracts all fixed assets valuedsymbolically at one dollar. Despite widespread agreement aboutnetting out social security reserves, Japan disregards such re-serves for precautionary reasons.

90In a number of countries, various government agencies mayconstruct their own indicators of fiscal stance. The U.S. Congres-sional Budget Office calculates the standardized employmentdeficit. The New Zealand Treasury is experimenting with an“economic fiscal indicator” that allows for a sectoral breakdown.


about the optimal or desirable level and path of theratio of public debt to GDP.91 Only the actual prospectof insolvency and debt default signals a clear indica-tion of unsustainability of the public sector. It is, how-ever, useful to determine for any country, even withrelatively limited data availability, whether its indebt-edness is stable or explosive over the medium term.92

A comprehensive indicator of sustainabilityshould include the present value of net unfunded lia-bilities as a ratio of GDP, particularly liabilities con-sisting of the future stream of defined benefits forpublic pension and health care programs, net of ex-isting assets and future contributions to these pro-grams. However, calculating the actuarial value ofunfunded liabilities requires long-term projections ofsocial security benefits and contributions, as dis-cussed below. At present, few countries (Canada,United Kingdom, and United States) regularly publisha transparent statement of all actual, as well as fundedand unfunded, contingent government liabilities forsocial entitlement schemes. Recently, the presentvalue of net unfunded public pension liabilities ratiohas been calculated for most member countries of theOrganization for Economic Cooperation and Devel-opment (OECD).93 The ratio of contingent liabilitiesto GDP is neither comparable nor additive to the ac-tual debt ratio of the public sector for purposes of de-termining sustainability. Rather, the ratio of net un-funded government liabilities to GDP should beinterpreted as an indicator of the magnitude of thenecessary structural reform of the public pension,health care, and other programs—by raising taxes orcontributions or by rationalizing benefits—to restorefinancial viability to the social security system.94 An

alternative approach to evaluating the necessary adjustment involves calculating the contributiongap.95

Whereas the foregoing indicators transparentlyshow whether fiscal policy meets the intertemporalbudget constraint, they convey limited informationon the intergenerational equity of fiscal benefits andburdens. Increasing concern with not only fiscal sus-tainability, but also the intergenerational distribu-tional implications, of the existing fiscal structurehas spawned the construction of generational ac-counts, especially in countries with rapidly agingpopulations. In essence, generational accounts areaimed at calculating the difference between thetransfers and services received and the tax payments(in net present value terms and adjusted for produc-tivity growth) of a generation, given the eventualneed to satisfy an intertemporal budget constraint.

Notwithstanding some proposals to substitutegenerational accounts for conventional budget bal-ance measures,96 the empirical implementation ofthe former imposes heavy data demands and relieson specific behavioral assumptions derived from theone-good, two-period, life-cycle model. The calcula-tions are highly sensitive to the assumed determina-tion of private savings (with most calculations deny-ing liquidity constraints or a bequest motive); to theselected discount rate and productivity growth; tostylized assumptions about a range of taxes, includ-ing those on capital; and to the assumed absence ofintergenerational benefit from public consumptionor capital spending.

So far, only two countries (Norway and UnitedStates) are known to have published official calcula-tions of generational accounts, whereas the OECDhas prepared such accounts for five member countries(Germany, Italy, Norway, Sweden, and United States).These calculations suggest that, with unchanged fiscalpolicies, generations born in the mid-1990s can ex-pect to face significantly higher tax burdens than pre-sent generations in most of these countries.97 Al-though, clearly, it would be difficult to prepare such

29

Appendix III

91For example, according to a general definition in Bean andBuiter (1987, p. 27), “a government is solvent if its spending pro-gramme, its tax-transfer programme, and its planned future use ofseigniorage [that is, its ability to issue currency at a face value inexcess of its cost of production] are consistent with its outstand-ing, initial financial and real assets and liabilities (in the sensethat the present value of its spending programme is equal to itscomprehensive net worth).” For a theoretical treatment of sustain-ability, see Horne (1991). See also proposals by Parker and Kast-ner (1993) for a practical framework for assessing fiscal sustain-ability in IMF-supported programs.

92The stability condition is given by the equation Δd =[(r–g)/(1 + g)]d–1–pb, where d is the debt-to-GDP ratio, pb is theratio of the primary balance to GDP, r is the nominal interest rateon public debt, and g is the growth rate of nominal GDP. See, forexample, the calculations for major industrial countries in Inter-national Monetary Fund (1996b, pp. 50–51).

93For 10 OECD member countries, the estimated present valueof net unfunded pension liabilities reaches or exceeds the level ofGDP, assuming a 5 percent discount rate and 1.5 percent yearlyproductivity growth. See Roseveare and others (1996) and Orga-nization for Economic Cooperation and Development (1996a, pp.36–37).

94To illustrate the point, in the United States, as a result of the1983 amendments to the Social Security Act, the present value ofnet unfunded liabilities of the old-age, survivors’, disability, and

hospital insurance programs fell from 70 percent of GNP at end-1982 to less than 10 percent by end-1983.

95The contribution gap is the difference between a constant sus-tainable contribution rate that, over a long time, would lead to nobuildup of pension debt above an initial level and the expected av-erage contribution rate under existing budget law. See, for exam-ple, Chand and Jaeger (1996).

96For an overview of generational accounts, see Auerbach,Gokhale, and Kotlikoff (1994), and for a critical assessment, seeHaveman (1994).

97See United States, Office of Management and Budget (1995),Hagemann and John (1995), and Organization for Economic Co-operation and Development (1995b). Italy shows the most in-equitable outcome; there, future generations will have to pay fivetimes larger net taxes than the present generation, assuming a 5percent discount rate and 1.5 percent yearly productivity growth.


estimates over a wide range of countries, private re-searchers have attempted similar calculations for anumber of other industrial countries (Australia, Den-mark, New Zealand, and United Kingdom). As withthe estimates of unfunded liabilities, generational ac-counts should be regarded merely as a broad indicatorof the structural adjustment required to avoid aggra-vating the inequities among generations.

Projections

The formulation of fiscal policy as well as the con-comitant public debate are usually predicated on pro-jections of future trends in government finances andin overall economic performance. The more transpar-ent the projections, the more informed the public de-bate and the more credible and sustainable the result-ing fiscal policy decisions. Broadly speaking,explicit, realistic, timely, and internally consistentprojections—whether model-based or judgmental—can be regarded as transparent. But transparency inthe underlying macroeconomic projections does notnecessarily ensure transparent fiscal projections.Transparency needs to be examined separately underdifferent time frames: short- to medium-term fore-casts, as distinct from long-term scenarios.

Short- to Medium-Term Forecasts

With the exception of the economies in transitionand the least developed market economies, most IMFmember countries have come a long way in develop-ing and improving their ability to prepare macroeco-nomic forecasts through some combination of econo-metric or computational models and judgment.98 Theresulting macroeconomic forecasts provide theframework for forecasting fiscal aggregates. Al-though virtually all governments disclose forecasts ofkey macroeconomic indicators (growth, inflation, un-employment, interest rates, and key commodityprices), they may be reluctant to reveal the methodsthey use to derive them, including behavioral andtechnological parameters.99 However, most advanced

economies and some developing economies disclosethe detailed macroeconomic forecasts they use as thebasis of fiscal forecasting.100

In the past, governments often adopted assump-tions that were difficult to justify, especially aboutinterest rates, unemployment rates, and output andincome growth, so as to understate government ex-penditure forecasts and overstate revenue forecasts.The propensity for such practices—often based onassertions rather than on transparent measurementusing empirical relationships—was particularly highwhen governments felt compelled to meet short- ormedium-term fiscal targets. In recent years, how-ever, many countries have become more realistic,open, and consistent in their forecasts, realizing thatthey are likely to be evaluated and challenged bytheir legislatures, the public, and financial markets.Under one approach (e.g., Canada), the governmentrelies on consensual macroeconomic forecasts forthe preparation of fiscal projections.101 Followinganother approach, country authorities deliberatelyuse conservative macroeconomic assumptions (e.g.,Chile and Netherlands).102 In a number of countries(e.g., Austria, Germany, Sweden, and United King-dom), transparency has been enhanced by open de-bate with independent forecasting agencies or pan-els, created to assist governments in selectingrealistic forecast values.

In preparing fiscal forecasts and the accompanyingmacroeconomic forecasts, policymakers should pro-vide, separately, baseline projections that assume unchanged policies and alternative projections incor-porating the impact of major (anticipated or recom-mended) policy changes. In most countries, officialprojections usually include policy measures, with thenet budgetary impact of the measures—in terms of thevalue of revenue and expenditure measures—shownseparately. Under the circumstances, the scope fornontransparent practices is rather ample. For example,poorly documented, pessimistic baseline projectionsmay give a misleading impression of the size of the

30

APPENDIX III

98A government has the prerogative to choose the most appro-priate framework—emphasizing, for example, Keynesian, Ricar-dian, or various eclectic features—and to supplement model-based results with judgment as to elements—particularly,credibility effects—that are not always captured satisfactorily bythose results. Alternatively, lacking adequate data or statisticallysignificant estimates, it may justifiably decide to rely purely oncomputational or judgmental methods. Transparency simply re-quires explicit and timely disclosure and, preferably, the publica-tion of methods to permit public assessment of the quality, in-cluding consistency, and track record of the forecasts.

99In Australia, Sweden, and the United Kingdom, for example,the macroeconomic model used for official fiscal forecasting is

100Although they still face severe data constraints, a number oftransition economies are also pushing ahead with developingtheir macroeconomic forecasting capability.

101Canada’s Department of Finance adopts assumptions on thebasis of an average of forecasts prepared by private institutions,with a prudential margin added to interest rates. The impact ofthese upward-adjusted rates is reflected in output and inflation as-sumptions, with the private sector averages being revised accord-ingly. Thus, for example, the short-term interest rate assumptionfor fiscal year 1997/98 was raised by 80 basis points above theprivate sector consensus.

102In France, for example, the budget deficit for 2001 was ini-tially based on a 2.1 percent annual medium-term GDP growthrate—at the time seen as a conservative assumption—to signaladherence to the target even under less favorable cyclical condi-tions. In Chile, for the macroeconomic and budget forecasts, arelatively low price is assumed for copper exports.


required fiscal adjustment.103 A less obvious distor-tion involves reliance on rather opaque implicit tax orexpenditure parameters. Instead, a fully documentedset of baseline and policy projections, including feed-back effects, should be published.104

Preparation of comprehensive forecasts is, ofcourse, beyond the reach of many nonindustrial coun-tries, particularly economies in transition that are un-dergoing rapid structural change and lack adequatedatabases. In the interest of transparency, these coun-tries should offer a public explanation of even rudi-mentary forecasting methods and assumptions used,so that users can gain some perception of the risks—both upside and downside—inherent in the forecasts.Although a significant number of member countriespossess a modest forecasting capacity, the authoritiesseldom publish a note explaining the forecasts.

Quite apart from a rather uneven disclosure of themethodology and basic assumptions, nearly all IMFmember governments provide fairly detailed fore-casts for the annual central budget, subject to defi-ciencies in classification (Appendix II), as part of thedraft budget bill they submit for legislative enact-ment for the upcoming fiscal year. Particularly en-lightening for users of fiscal forecasts—lacking suf-ficient information on methods—is the publicationof (1) a sensitivity analysis showing the response offiscal aggregates to a marginal change in key macro-economic variables, (2) a breakdown between dis-cretionary and mandatory spending, and (3) support-ing forecasts of “cost drivers,” such as the number ofgovernment employees and the number of beneficia-ries of entitlements (e.g., Australia, Canada, andUnited States).

Detailed and open documentation of microeco-nomic linkages and assumptions (e.g., effective taxrates, tax bases, and parameters, including compli-ance coefficients or collection lags, used for eachrevenue category) underlying the fiscal forecasts isnot readily available in most countries. As a result, itis difficult to assess any deviations of outcome fromforecasts for specific revenue or expenditure cate-gories.105 Yet, logically, initial budget forecasts

should be followed with periodic updates and anevaluation of how outcomes deviate from forecasts.In fact, several OECD member countries (Australia,New Zealand, United Kingdom, and United States)publish within-year updates, even in the absence ofsupplementary budget legislation. Such updatesshould be accompanied, if possible, by a public ex-planation of the reasons for significant deviationsfrom forecasts.106

With increasing concern for sustainability, in the1980s, some industrial countries (Italy, Nether-lands, Sweden, and United States) began develop-ing multiyear budget projections—often contrastedwith baseline projections—in the context ofmedium-term fiscal consolidation plans. At pres-ent, in some countries, these forecasts are simplymeant to provide top-down guidance on the needfor discretionary fiscal action beyond the currentone-year budget period, that is, the baseline for for-mulating medium-term budget priorities. In somecountries (France and Japan), the multiyear projec-tions provide the context for the medium-term out-look for only the broadest fiscal aggregates, with-out necessarily spelling out the actual policymeasures to be used for attaining the fiscal tar-get.107 Nevertheless, a number of countries (Aus-tralia, Germany, Italy, Netherlands, New Zealand,Switzerland, United Kingdom, and United States)provide detailed medium-term fiscal forecasts in-corporating the authorities’ policy intentions.108

Regardless of the approach used, the message ofsuch exercises is that the built-in momentum of ex-isting programs—especially mandatory entitlementprograms—measured against the revenue yield of

31

Appendix III

103In Italy, for example, the baseline budget was occasionallyinflated because spending ministries were asked to calculate thecost of maintaining existing policies with no regard for potentialcost savings.

104In the United States, both the Office of Management andBudget, which is responsible for forecasts incorporating all pol-icy proposals of the administration, and the Congressional Bud-get Office, directly responsible to the Congress, prepare separateprojections of baseline “current services” and of the effect ofnew policy measures. Differences between the two sets of pro-jections reflect primarily differences in economic and technicalassumptions.

105This is illustrated, for example, by the significant unantici-pated shortfall in the value-added tax and corporate income taxrevenue recorded recently in the United Kingdom.

106In Australia, for example, the authorities provided the fol-lowing summary reconciliation for fiscal year 1995/96 (in bil-lions of Australian dollars):

Revenue forecast 124.4

Effect of policy changes –0.2

Effect of adjustment in parameters and other changes –2.5

Revenue outcome 121.7

Expenditure forecast 123.7

Effect of policy changes 0.8

Effect of wage and price adjustments 0.4

Effect of adjustment in parameters 0.2

Other adjustments 1.7

Expenditure outcome 126.7

Source: Statement 1 to Budget Speech 1996/97 (Australia,Department of the Treasury, 1997).

107In Japan, the Ministry of Finance has prepared both baselineand policy-adjusted projections covering the period through2006; see Okamura (1996).

108In several countries (e.g., Germany, New Zealand, andSwitzerland), these forecasts are limited to the central government.


32

APPENDIX III

existing tax policies leaves a narrow, if any, marginfor new program initiatives.109

Long-Run Scenarios

A growing concern with long-term financial sus-tainability in view of rapidly aging populations, therising cost of health care, and the rigidity of most so-cial entitlements prompted some governments in in-dustrial countries to prepare long-term scenariosshowing the annual fiscal balance over a period of upto seven decades. They have carried out these calcula-tions, mostly on an occasional basis, to examine thesustainability of existing public pension systems.110

There are, however, a few countries where social secu-rity agencies are legally required to publish each yearlong-term projections separately for each retirementfund (Canada and United States) and each health carefund (United States)111 under alternative sets of gen-eral assumptions regarding economic growth, wagegrowth, interest rates, inflation, unemployment, fertil-ity, net migration, and mortality, as well as under spe-cific assumptions about disability, hospital, and med-ical costs. On the basis of these baseline projections, itis possible to test the effect of alternative reform op-

tions, such as tightening eligibility criteria for specificpension benefits or various cost-containment measuresin health care programs.112

Overall, these scenarios contain considerable in-formation, particularly regarding the path of finan-cial balance over a prolonged period, and can beused to simulate in a transparent manner the effectof corrective reform measures.113 Furthermore,these scenario results can be summarized, using thepresent value of net unfunded government liabilitiesto choose an appropriate discount rate, as discussedabove. Expressed as a percentage of GDP, the for-mer can be interpreted as an indicator of fiscal sus-tainability, comparable across countries and overtime. In view of the acknowledged uncertaintiesabout the underlying macroeconomic and demo-graphic projections and assumptions, presentationof scenario results should always be accompaniedby a sensitivity test with respect to changes in keyassumptions.

109In the United Kingdom, this process was for a time strength-ened by a new emphasis on imposition of cash limits on manybudget aggregates—a warning to managers that initial estimateshad to be prepared more carefully and that the source of potentialerror, including the interaction of cost factors with prospectivemacroeconomic forecasts, had to be carefully reconciled. Al-though in the United States the projections are highly disaggre-gated for mandatory and discretionary programs, projections ofdiscretionary expenditures depend on yearly appropriations andcannot be made effective through current legislative actions.

110See the review of national projections for EU member coun-tries in Franco and Munzi (1996).

111In the United States, this requirement encompasses moving10-year and 75-year projections separately for the old-age andsurvivors’ insurance, disability insurance, and hospital insurancetrust funds. For the 10-year projection, the trust fund ratio (trustfund reserves as a percentage of annual benefit payments) is cal-culated; the 75-year projection consists of a summary actuarial

balance (annual revenue less payments as a ratio of taxable pay-roll, adjusted to include the beginning fund reserves and the costof ending the projection period with reserves equivalent to yearlybenefit payments). The purpose of the exercise is to ascertain theprofile of these indicators over time and to determine the perioduntil exhaustion of the reserves in each fund. As of end-1995, theremaining period under intermediate assumptions was 35 yearsfor the old-age and survivors’ insurance, 19 years for disabilityinsurance, and 5 years for hospital insurance; see Social Securityand Medicare Boards of Trustees (1995). Canada also preparesdetailed long-run projections for public pension programs; seeCanadian Pension Plan Secretariat (1996).

112See, for example, the recent long-term simulations, incorpo-rating hypothetical reform options for the old-age and survivors’insurance and the hospital insurance programs, in United States,Congressional Budget Office (1996).

113Ideally, of course, policy simulations should be performedwith an appropriate economywide model that allows fully for en-dogenous macroeconomic repercussions of the hypothesized pol-icy changes. See, for instance, the application of such a model inthe United States in Aaron, Bosworth, and Burtless (1989). Themost recent set of model-based policy scenarios can be found inUnited States, Congressional Budget Office (1996).


T his appendix summarizes the experiences of se-lected countries in several regions of the IMF’s

membership. Although none of the countries selectedis necessarily to be regarded as a model of fiscal trans-parency, each has made significant progress towardtransparency within its own region in recent years. Theimpetus for improvement came largely from within, insome cases with support from the IMF.

Botswana

Within Africa, Botswana has achieved a substan-tial degree of fiscal transparency. Factors that havecontributed to transparency include a prudent andopen budget process and a high level of competencein the civil service.

The government’s fiscal policy strategy is setdown in published five-year development plans.However, these plans are revised periodically to takeinto account the evolving economic situation. Thus,in the published midterm review of the current plan,expected revenue shortfalls were identified, alongwith intended measures to reduce current expendi-ture and improve project implementation. Succes-sive budget surpluses, driven by strong growth inmining revenues, contributed to a large accumula-tion of government deposits with the banking systemand associated foreign exchange reserves.

The Seventh Development Plan, covering the pe-riod 1991/92 to 1996/97, announced publicly that thecountry intended to run down cash deposits to fund arevenue shortfall as mineral revenues slowed and cur-rent expenditures increased. During the preparationof the Eighth Development Plan, the Ministry of Fi-nance and Planning released to the public a policypaper setting down the thrust of fiscal policies overthe period to 2002/2003. Given the anticipated slow-down in revenues, particularly from the mining sec-tor, the paper prudently highlighted the options tocurtail the growth of government expenditure.

Tax legislation clearly specifies the rates and tax-able bases. The authorities have resisted pressure toexpand the extensive investment and other conces-sions available under the tax system. Tax administra-

tors have limited discretion, and appeals are possi-ble. At this time, the main problem in tax administra-tion is the buildup of tax arrears, on which only lim-ited information is available.

Operations of the central budget are recorded on acash basis, with a clear breakdown of tax and nontaxrevenues and a functional and economic classifica-tion of expenditures, which accord with internationalstandards. Although no balance sheet is prepared, ac-curate up-to-date records of gross internal and exter-nal government debt are published. Detailed annualdata on the central budget are normally publishedwith a lag of two to three months. Nonfinancial pub-lic enterprises and public financial institutions are re-quired to publish annual reports. No comprehensivedata on local government operations are available.

Although Botswana has an extensive range of fi-nancial and nonfinancial public enterprises, the au-thorities have largely avoided using these enterprisesto engage in quasi-fiscal activities.

Chile

In the mid-1970s, Chile formulated an economicstrategy to facilitate private sector development by re-ducing the size and role of government. It has main-tained this strategy over time, and the scope of gov-ernment activities has shrunk significantly.Subsequently, priority has shifted to improving effi-ciency in government activities, to ensuring that fiscalmanagement is consistent with broader macroeco-nomic and social objectives, and to enhancing theoverall transparency of public finances.

The role of the government has been reshapedthrough privatization of state-owned enterprises; de-centralization of government activities (e.g., in pri-mary and secondary education); shifting of traditionalgovernment services to the private sector (includinghealth care insurance and management of pensionfunds); increasing of private sector ownership or par-ticipation in the construction, operation, and mainte-nance of public infrastructure (including electricitygeneration and distribution, harbors, and transportprojects); and effective operation of the copper fund.

Appendix IV Selected Country Experiences

33


The government budget has been in surplus over thelast decade, partly as a result of these reforms.

The budget process has been designed to achievetransparent control over public finances and to en-sure consistency of fiscal policy with the announcedmacroeconomic objectives. Each year, the draft bud-get submitted to the legislature contains policy state-ments on macroeconomic targets, supported by rela-tively conservative assumptions for key macrovariables (including the export price of copper). Itincludes detailed operating plans for each spendingagency and detailed revenue estimates by tax cate-gory. The draft budget is subject to review by spe-cialized legislative committees; since 1995, budgetexecution has been monitored on the basis of quar-terly statements submitted by the Budget Office ofthe Ministry of Finance with a lag of 60–90 days.The quarterly reports are widely available throughpublications and the electronic media (Internet).

Although operated outside the budget, the copperstabilization fund has played an important role in thebudget process since its creation in 1981. The fund isrequired to deposit the revenues that exceed an es-tablished benchmark in a special account at the cen-tral bank. Use of the fund was initially restricted todebt service and subsequently to prepayment of pub-lic debt. In 1986, legislation formally establishingthe Copper Compensation Fund provided clear rulesfor deposits and withdrawals on the basis of the dif-ference between a medium-term reference price andthe actual export price of copper. Changes in operat-ing procedures for this fund must be jointly agreedto by four institutions: the central bank, CODELCO(the state-owned copper corporation), the treasury,and the budget office. Prudent management of theCopper Compensation Fund under transparent ruleshas lessened the impact of external shocks and al-lowed for a buildup of foreign exchange reserves.However, less transparent is the allocation of 10 per-cent of gross copper exports by CODELCO—equiv-alent to 0.6 percent of GDP during the 1990s—formilitary expenditures to a fund operating outside thebudget. This allocation is in addition to the regularbudgetary allocations to the Ministry of Defense.

Tax reform has resulted in a simple system, by in-ternational standards, with comparatively low ratesand transparent bases. Tax administration has be-come broadly transparent, although some past ex-emptions have been grandfathered under variousstatutes, making it difficult to assess their full impli-cations; also, these exemptions follow a variety ofless than clearly defined objectives. The authoritieshave yet to develop a tax expenditure budget.

Reports on budgetary accounts, recorded on acash basis, include consolidated central governmentoperations and the operations of state-owned enter-prises, but exclude those of the local governments.

Government expenditures are classified by eco-nomic, functional, and institutional categories, andrevenues are classified by tax categories, consistentwith international standards. Chile subscribes to theIMF’s Special Data Dissemination Standard.

For a limited number of decentralized agencies(such as institutions of higher education), only bud-getary transfers are recorded, whereas for the public fi-nancial institutions, comprehensive reports are pub-lished on a regular basis. The central bank’s balancesheet contains aggregate data on quasi-fiscal opera-tions associated with schemes to support the bankingsystem during the crisis of the early 1980s and with theplacement of promissory notes for sterilization pur-poses. State-owned enterprises are operated transpar-ently without hidden subsidies or preferential credits.

As part of the process of increasing the efficiencyand transparency of government operations, the au-thorities are developing selective performance indi-cators for spending agencies. Recently, they adopteda pilot project and are expanding it to 65 specializedagencies by 1997, covering areas such as publicworks, mining, education, health care, and finance.

Denmark

Significant efforts have been made to improve thetransparency of government operations in Denmarkto strengthen financial discipline and to enhanceoversight by the electorate. The driving force forthese reforms has been the large size of the generalgovernment (total expenditures amount to 60 percentof GDP) and the high degree of decentralization(more than half of government expenditures are atthe subnational level). The reforms included majorlegislative changes and deregulation in variousareas, including financial and labor markets.

One of the key measures enhancing fiscal trans-parency was the harmonization of the budgetary sys-tems of the central and local governments, which fa-cilitated the publication of the annual consolidatedgeneral government budget. This entailed prescrib-ing common budget standards, adopting a meaning-ful system of classification, and laying down a stricttimetable for the preparation and legislative approvalof the budgets at all levels. Current legislative provi-sions require all governments to disseminate clearinformation on their operations (in the form of pam-phlets) prior to the budget debates in the nationaland subnational legislatures.

The government budget has been transformed froma mere accounting and monitoring device to an infor-mation vehicle to assist political decision making. Theintroduction of compulsory multiyear budgeting at allgovernment levels has strengthened the budget as aninstrument for medium-term policy planning by shift-

34

APPENDIX IV


ing the emphasis from short-term fiscal managementto setting strategic priorities and longer-term objec-tives. Its introduction has been underpinned by com-prehensive analyses of consolidated multiyear publicinvestment programs for all levels of government.Government operations are now reviewed regularlyby specialized bodies of fiscal audit, which publishtheir findings to further increase the accountability ofgovernment entities. The central and local govern-ments are also required to monitor budgetary develop-ments closely during the budget year and to preparesupplementary budgets if necessary.

In recent years, considerable efforts have beenmade to simplify the tax system. As a result, a largenumber of personal income tax payers need not filetax returns; tax authorities rely instead on adminis-trative records for information on incomes, whichare sent to taxpayers for verification. A system of ad-vance rulings has been introduced to facilitate tax-payer awareness of compliance requirements, whilea comprehensive statute protects taxpayers from dis-cretionary action.

Denmark subscribes to the Special Data Dissemi-nation Standard. The fiscal accounts, prepared annu-ally, have a broad coverage that extends to the con-solidated general government. This presentationtypically forms the basis for the budget process andthe determination of intergovernmental fiscal trans-fers for the forthcoming year (e.g., grants for newtasks to be performed at the subnational level). Al-though the accounts follow a cash-based recording,this approach has been supplemented in recent yearsby accrual-based recording for some items. The clas-sification of expenditures and revenues closely fol-lows international standards. Although no balancesheets are published, aggregate information on thevalue of public sector assets and liabilities (includinga detailed breakdown of gross central governmentdebt) is published on a regular basis.

Denmark regularly publishes official estimates ofthe fiscal impulse and the structural deficit for thegeneral government sector, which the Ministry of Fi-nance updates at least four times a year. Quasi-fiscalactivities are virtually nonexistent. There has beenconsiderable privatization; the public enterprises thatremain are required to meet commercial criteria.

Hungary

Hungary has evolved from a country with secre-tive and nontransparent fiscal practices—includingmisreporting to the IMF—under socialist centralplanning to the one with probably the highest levelof transparency among economies in transition as itbecomes an emerging financial market. This trans-formation has been assisted by an active legislative

and institution-building process. However, like othereconomies in transition, Hungary has yet to reach alevel comparable to that of most industrial countries.

Since the early 1990s, Hungary has operated ahighly decentralized and complex budget system,with the central budget receiving the revenue fromthe major tax categories and then making transfersto autonomous central budgetary institutions, extra-budgetary funds, and local governments. Althoughthe Ministry of Finance exercised overall controlover budget transfers, the system conferred consid-erable freedom of action on the various partici-pants, many of whom retained some own-sourcerevenue (mainly user fees) and were not subject todetailed monitoring of expenditure. The implemen-tation of a centralized treasury system in 1996—with a single bank account through which all cen-tral budget revenues and expenditures are effectedunder the scrutiny of the Ministry of Finance—hascontributed to both improved central budget controland transparency. However, the social securityfunds retain considerable independence, and re-porting deficiencies make monitoring the funds’ ac-tivities difficult.

Timely data on general government operations arenot yet available. For operational purposes, the au-thorities rely on information on the consolidatedcentral government, which encompasses the centralbudget (including the operations of the central bud-getary institutions partly financed by own revenues)and extrabudgetary funds (including social securityfunds). Data on over 3,000 subnational governmentsare available only with a significant time lag. Al-though subject to legislative approval, the operationsof the privatization agency are excluded from thecoverage of the consolidated central government(e.g., as in Germany). In 1996, the bulk of privatiza-tion receipts were earmarked in a transparent man-ner for debt reduction, although some allocations(partly in the form of loan guarantees) were made di-rectly to state-owned enterprises (notably, the Bor-sod steel complex) and local governments.

Although all transactions are recorded on a cashbasis, up-to-date information is made available forsocial security contribution arrears and contingent li-abilities. Institutional and economic classification ofexpenditures is published for the consolidated cen-tral government, and a comprehensive functionalclassification is under development within the re-cently formed treasury.

Considerable progress has been made in compil-ing comprehensive data on financing flows for theconsolidated central government, largely from bal-ance sheet information prepared by the NationalBank of Hungary. Indeed, gross debt data for theconsolidated central government are published annu-ally. Until recently, the task of compiling such infor-

35

Appendix IV


mation had been complicated by the unique publicsector debt-management arrangements, whereby theNational Bank of Hungary acted as the sovereignborrower abroad while the Ministry of Finance man-aged domestic debt. This arrangement was discon-tinued in 1997, and the Ministry of Finance assumedresponsibility for managing all government debt andpublishing comprehensive data on government lia-bilities. Information on new guaranteed debt to non-financial public enterprises is now published in thebudget each year.

Like all former centrally planned economies, in thepast, Hungary conducted a large number of quasi-fiscal activities through state-owned enterprises andfinancial institutions, although not to the extent ofsome of its neighbor countries, and the ongoing re-forms and large-scale privatization have reduced theextent of quasi-fiscal activities still further. Althoughsome nontransparent subsidies remain, the authori-ties have replaced most with explicit transfers fromthe budget. A legacy in this area is the large nonper-forming loans—mainly to bankrupt state-owned en-terprises—held by several state-owned banks. By1995, the issuance of government paper to financecapital injections to these banks had been completed,with the servicing of the debt subsequently assumedby, and shown transparently in, the central budget.The remaining tasks of rationalization (includingdebt workouts and downsizing of a redundant work-force) and privatization are under way.

Although all the major taxes have been codified intax laws—Hungary was the first former centrallyplanned economy to introduce fairly transparent taxrates and bases in the late 1980s—some remain exces-sively complex and are impaired by frequent regula-tory changes. During the transition, a number oflargely ad hoc exemptions and deductions weregranted to private, especially foreign-owned, compa-nies. Although the effect of past arrangements still per-sists, the ad hoc granting of new tax concessions wasdiscontinued in 1994. The budget documents provideonly fragmentary information on tax expenditures.

The State Audit Office, established in 1992, is ful-filling a mandate to prepare and publish reports on theperformance of budgetary institutions, social securityagencies, and selected state-owned enterprises.

Jordan

In recent years, Jordan has made substantialprogress toward transparency. Historically, Jordandid not formulate a fiscal strategy. The need for amore concerted approach to fiscal restraint wasclearly underscored by the explosion of governmentdeficits and debt in the late 1980s. Since then, Jor-dan has publicly announced a medium-term fiscal

strategy, supported by an Extended Arrangementwith the IMF.

The budget law is enacted by the end of each yearin accordance with comprehensive arrangements forthe conduct of fiscal policy: the Ministry of Financehas the primary responsibility for preparing and exe-cuting the budget. The draft budget is reviewed by aparliamentary finance committee before it is submit-ted for legislative action.

Comprehensive information is available monthlyand annually for the central government budget,which covers all departments and the Jordan ValleyAuthority. Budget coverage has improved substan-tially since 1990; it now includes within its ambitall extrabudgetary operations, except defense. Aportion of defense spending is handled through anextrabudgetary fund that is financed through for-eign loans. Military wages and operating costs areon-budget, but from published data it is difficult todetermine the split between the two components.Earmarked revenues (in particular, a portion of im-port duties) are channeled directly to municipalitiesand universities and do not appear in the budget.Limited information is available on a number of de-centralized government agencies (such as universi-ties, institutes, marketing authorities, and otheragencies) that, besides their own revenue sources,are also supported to varying degrees by budgettransfers and state-owned financial and nonfinan-cial entities.

Government accounts are recorded on a cash basisand published with a two-month lag. The classifica-tion used in the annual budget follows a largely insti-tutional format, but the authorities compile data forthe Fund’s Government Finance Statistics Yearbook,reclassifying revenues, expenditures (both by func-tional and by economic type), and financing transac-tions according to international standards. Compre-hensive data are also compiled on both external andinternal gross public debt. Although no separate in-formation is available on the large and accumulatingunfunded liabilities of the public pension system,data on loan guarantees are published by the CentralBank of Jordan.

Jordan has a large state-owned enterprise sector,and the government owns shares in some private en-terprises through its holding company, the Jordan In-vestment Corporation. Although these enterprises canborrow domestically, guarantees must be approvedby the legislature. Overall, it is difficult to assess therelationship between the enterprises and the govern-ment. An annual report on public enterprises includesa consolidated balance sheet; however, no informa-tion is available on the extent of their quasi-fiscal ac-tivities. To improve the monitoring of public financesand increase transparency, public enterprises will berequired to submit monthly financial statements and

36

APPENDIX IV


quarterly projections of revenue and expenditure tothe Ministry of Finance.

The tax system has been simplified and modern-ized. The principal objective of the reforms has beento lower marginal rates, reduce the number of rates,and broaden tax bases. In principle, the degree ofdiscretion in tax administration has been reduced.Tariff rates have been reduced and unified and cus-toms procedures simplified. However, multiple ratesand extensive exemptions and deductions remain,and the determination of tax liabilities is still oftensubject to considerable negotiation and administra-tive interpretation. Taxpayer rights and responsibili-ties are clearly specified.

New Zealand

New Zealand represents a benchmark for publicsector transparency. The Fiscal Responsibility Actof 1994, which contains a set of principles for fis-cal management and transparency, is the culmina-tion of a decade of reform designed to improve theefficiency, effectiveness, and accountability ofwhat had been a large and interventionist publicsector.114 Until the mid-1980s, economic perfor-mance in New Zealand was disappointing becauseof inadequate adjustment to changes in the worldeconomy. Since then, the government has madesolid progress in downsizing the public sector,moving the budget into surplus, and reducing pub-lic debt. The measures to increase fiscal trans-parency may be seen partly as a means of assuringdomestic and overseas investors that the govern-ment, regardless of political orientation, is commit-ted to behaving responsibly.

The government is required to follow the fiscalmanagement principles established in the Fiscal Re-sponsibility Act, whose goals consist in reducing andmaintaining debt at prudent levels; aiming for a bal-anced operating budget within a reasonable timeframe, with allowance for cyclically induced deficitsor surpluses; providing a buffer against possible ad-verse developments by maintaining adequate levelsof net worth in the government balance sheet; andensuring a reasonable degree of predictability aboutthe level and stability of taxes. (The government isallowed to depart temporarily from these principlesprovided it explains its reasons and indicates how,and within what time frame, it plans to return tothem.) In addition, the legislation obliges the gov-ernment to publish the following: a budget policystatement setting down its strategic priorities for the

coming fiscal year as well as its long-term objectivesregarding revenue, expenditure, deficit and debt lev-els, and net worth; an economic and fiscal updatecontaining economic and fiscal forecasts for the cur-rent budget year and the following two fiscal years;and a fiscal strategy report that explains whether theupdate is consistent with the government’s previ-ously stated intentions and that projects the path ofall relevant fiscal variables for at least 10 years. Aseparate preelection economic and fiscal updatemust be published two to six weeks before an elec-tion. The government is required to disclose allitems involving fiscal costs or gains and to identifypotential risks.115

The tax system is simple and contains clearlyspecified tax rates and bases. The Inland RevenueDepartment has limited administrative discretion inthe application of the tax laws. The department as-sists taxpayers by publishing technical informationon specific points of interpretation. Since 1992, thedepartment can also issue legally binding rulings.Estimates of tax expenditures are not published—thesubstantial curtailment of tax deductions and con-cessions has rendered this task unnecessary. The lawspecifies taxpayer rights and obligations.

Public accounts are reported on a whole-of-gov-ernment basis, which covers the consolidated centralgovernment (including decentralized entities) as wellas nonfinancial public enterprises and public finan-cial institutions (including the central bank). Subna-tional governments are not included in these ac-counts, but information on this relatively small sectoris now collected in a timely fashion by the statisticalagency. Quasi-fiscal activities are not separatelyidentified, but their net impact is implicitly capturedin the whole-of-government data. New Zealand’s ac-counting statements are accrual-based, but cash ac-counts continue to be published. Scope for creativeaccounting is constrained by the requirement to con-form to standards set by the independent AccountingStandards Review Board. Detailed classifications ofrevenues and expenditures follow international stan-dards. Comprehensive national balance sheets areproduced semiannually. The statements attached tothe balance sheets explain revaluation changes andcontain comprehensive listings and estimates (wherequantifiable) of all measurable commitments andcontingent liabilities.

37

Appendix IV

114For a more complete account of these reforms, see Cangiano(1996) and Scott (1996).

115The Fiscal Responsibility Act requires that all governmentfinancial statements (including projections) be drawn up in accor-dance with generally accepted accounting practices and thus meetthe same standards (set by the Accounting Standards ReviewBoard) as private sector financial reports. The Minister of Financemust formally accept overall responsibility for the integrity of thedisclosures and their consistency with the act, and must also statethat all relevant information has been transmitted to the Secretaryof the Treasury.


Aaron, Henry J., Barry P. Bosworth, and Gary T. Burt-less, 1989, Can America Afford to Grow Old? Paying for Social Security (Washington: Brookings Institution).

Alesina, Alberto, and Alex Cukierman, 1990, “The Poli-tics of Ambiguity,” Quarterly Journal of Economics,Vol. 105 (November), pp. 829–50.

Alesina, Alberto, and others, 1995, “Budget Institutionsand Fiscal Performance in Latin America,” IDBWorking Paper WP-004 (Washington: Inter-AmericanDevelopment Bank, Office of the Chief Economist).

Alexander, William E., and others, 1997, Systemic BankRestructuring and Macroeconomic Policy (Washing-ton: International Monetary Fund).

Auerbach, Alan J., Jagadeesh Gokhale, and Laurence J.Kotlikoff, 1994, “Generational Accounting: A Mean-ingful Way to Evaluate Fiscal Policy,” Journal of Eco-nomic Perspectives, Vol. 8 (Winter), pp. 73–94.

Australia, National Commission of Audit, 1996, Report tothe Commonwealth Government (Canberra: Aus-tralian Government Publishing Service).

, Department of the Treasury, 1997, BudgetSpeech 1996/97 (Canberra).

Bean, Charles R., and Willem H. Buiter, 1987, The PlainMan’s Guide to Fiscal and Financial Policy (London:Employment Institute).

Blejer, Mario I., and Adrienne Cheasty, eds., 1993, How toMeasure the Fiscal Deficit: Analytical and Method-ological Issues (Washington: International MonetaryFund).

Buchanan, James M., and Richard E. Wagner, 1977,Democracy in Deficit: The Political Legacy of LordKeynes (New York: Academic Press).

Canada, Canadian Pension Plan Secretariat, 1996, “An In-formation Paper for Consultations on the CanadianPension Plan,” condensed version, released by theFederal, Provincial, and Territorial Governments ofCanada (Ottawa: Department of Finance).

Cangiano, Marco, 1996, “Accountability and Trans-parency in the Public Sector: The New Zealand Expe-rience,” IMF Working Paper 96/122 (Washington: In-ternational Monetary Fund).

Chand, Sheetal K., and Albert Jaeger, 1996, Aging Popula-tions and Public Pension Schemes, IMF OccasionalPaper No. 147 (Washington: International MonetaryFund).

Coen, Robert M., 1975, “Investment Behavior, the Mea-surement of Depreciation, and Tax Policy,” AmericanEconomic Review, Vol. 65 (March), pp. 59–74.

Comiez, Maynard S., 1966, A Capital Budget Statementfor the U.S. Government (Washington: BrookingsInstitution).

Efford, Don, 1996, “The Case for Accrual Recording inthe IMF’s Government Finance Statistics System,”IMF Working Paper 96/73 (Washington: InternationalMonetary Fund).

Eisner, Robert, 1986, How Real Is the Federal Deficit?(New York: Free Press).

European Commission, 1995, “Methodological and Oper-ational Aspects of the Reporting of GovernmentDeficits and Debt Levels in the Context of the Exces-sive Deficit Procedure” (Brussels: Directorate Gen-eral II).

Franco, Daniele, 1995, “Pension Liabilities: Their Use andMisuse in the Assessment of Fiscal Policies,” Euro-pean Commission Economic Papers No. 110 (Brus-sels: Directorate-General for Economic and FinancialAffairs).

, and Teresa Munzi, 1996, “Public Pension Ex-penditure Prospects in the European Union: A Surveyof National Projections,” in Ageing and Pension Ex-penditure Prospects in the Western World, EuropeanEconomy, Reports and Studies, No. 3 (Luxembourg:European Commission, Directorate-General for Eco-nomic and Financial Affairs).

Gansler, Jacques S., 1995, Defense Conversion: Trans-forming the Arsenal of Democracy (Cambridge: MITPress).

Gore, Albert, 1993, From Red Tape to Results: Creating aGovernment That Works Better & Costs Less: Reportof the National Performance Review (Washington:Government Printing Office).

Hagemann, Robert P., and Christoph John, 1995, “The Fis-cal Stance in Sweden: A Generational AccountingPerspective,” IMF Working Paper 95/105 (Washing-ton: International Monetary Fund).

Haveman, Robert, 1994, “Should Generational AccountsReplace Public Budgets and Deficits?” Journal of Economic Perspectives, Vol. 8 (Winter), pp.95–111.

Heller, Peter S., Richard D. Haas, and Ahsan S. Mansur,1986, A Review of the Fiscal Impulse Measure, IMFOccasional Paper No. 44 (Washington: InternationalMonetary Fund).

Hopkins, Thomas D., 1996, “Regulatory Costs in Pro-file,” Center for the Study of American Business,Policy Study No. 132 (St. Louis, Missouri: Washing-ton University).

References

38


Horne, Jocelyn, 1991, “Indicators of Fiscal Sustainability,”IMF Working Paper 91/5 (Washington: InternationalMonetary Fund).

International Federation of Accountants, 1995, “Account-ing for and Reporting Liabilities,” Public SectorCommittee Study 6 (New York: IFAC).

, 1996, Handbook 1996: Technical Pronounce-ments (New York: IFAC).

, 1997, Perspectives on Accrual Accounting, Pub-lic Sector Committee, Occasional Paper 3 (New York:IFAC).

International Monetary Fund, l986, A Manual on Govern-ment Finance Statistics (Washington).

, 1995, Government Finance Statistics Yearbook(Washington).

, 1996a, Government Finance Statistics Manual:Annotated Outline (Washington).

, 1996b, World Economic Outlook, May 1996: ASurvey by the Staff of the International Monetary Fund,World Economic and Financial Surveys (Washington).

, 1996c, World Economic Outlook, October1996: A Survey by the Staff of the InternationalMonetary Fund, World Economic and Financial Sur-veys (Washington).

, and others, 1991, A Study of the Soviet Econ-omy, Vol. 1 (Washington: IMF).

Inter-Secretariat Working Group on National Accounts,1993, System of National Accounts 1993 (Brussels:Commission of the European Communities).

King, Mervyn A., and Don Fullerton, eds., 1984, The Taxa-tion of Income from Capital: A Comparative Study inthe United States, the United Kingdom, Sweden, andWest Germany (Chicago: University of Chicago Press).

Koedijk, Kees, and Jeroen Kremers, 1996, “Market Open-ing, Regulation and Growth in Europe,” EconomicPolicy: A European Forum, No. 23 (October), pp.445–85.

Kopits, George, 1991, “Fiscal Reform in European Econ-omies in Transition,” in The Transition to a MarketEconomy, Vol. 2, ed. by Salvatore Zecchini and PaulMarer (Paris: Organization for Economic Cooperationand Development), pp. 359–88.

Kotlikoff, Laurence J., 1989, “From Deficit Delusion tothe Fiscal Balance Rule: Looking for an Economi-cally Meaningful Way to Assess Fiscal Policy,” IMFWorking Paper 89/50 (Washington: InternationalMonetary Fund).

Kyrouz, M.E., 1975, “Foreign Tax Rates and Tax Bases,”National Tax Journal, Vol. 28 (March), pp. 61–80.

Mackenzie, G.A., and Peter Stella, 1996, Quasi-FiscalOperations of Public Financial Institutions, IMF Oc-casional Paper No. 142 (Washington: InternationalMonetary Fund).

Montanjees, Marie, 1995, “Government Finance Statisticsin the Countries of the Former Soviet Union: Compila-tion and Methodological Issues,” IMF Working Paper95/2 (Washington: International Monetary Fund).

Morin, François, and Claude Dupuy, 1993, Le Coeur Fi-nancier Européen (Paris: Economica).

New Zealand Treasury, 1996, Financial Statements of theGovernment of New Zealand for the Year Ended 30June 1996 (Wellington).

Okamura, Kenji, 1996, “Japan’s Medium- and Long-TermFiscal Challenges,” IMF Working Paper 96/113(Washington: International Monetary Fund).

Organization for Economic Cooperation and Development,1990, Taxpayers’ Rights and Obligations: A Survey ofthe Legal Situation in OECD Countries (Paris: OECD).

, 1992, Regulatory Reform, Privatisation andCompetition Policy (Paris: OECD).

, 1995a, Budgeting for Results: Perspectives onPublic Expenditure Management (Paris: OECD).

, 1995b, OECD Economic Outlook, No. 57(Paris: OECD).

, 1996a, Ageing in OECD Countries: A CriticalPolicy Challenge (Paris: OECD).

, 1996b, National Accounts: Detailed Tables,1982–94, Vol. 2 (Paris: OECD).

, 1996c, Tax Expenditures: Recent Experiences(Paris: OECD).

Parker, Karen Elizabeth, and Steffen Kastner, 1993, “AFramework for Assessing Fiscal Sustainability andExternal Viability, with an Application to India,” IMFWorking Paper 93/78 (Washington: InternationalMonetary Fund).

Pope, Jeremy, ed., 1996, National Integrity Systems: TheTI Source Book (Berlin: Transparency International).

Potter, Barry H., 1997, “Dedicated Road Funds—A Pre-liminary View on a World Bank Initiative,” IMF Paperon Policy Analysis and Assessment 97/7 (Washing-ton: International Monetary Fund).

Premchand, A., 1994, “Changing Patterns in Public Ex-penditure Management: An Overview,” IMF WorkingPaper 94/28 (Washington: International MonetaryFund).

, 1996, “Erosion of Expenditure ManagementSystems: An Unintended Consequence of Donor Ap-proaches,” IMF Working Paper 96/102 (Washington:International Monetary Fund).

Puviani, Amilcare, 1903, Teoria della Illusione Fi-nanziaria (Palermo: Laterza); reprinted as Puviani,Amilcare, and Franco Volpi, 1973, “Teoria della Illu-sione Finanziaria,” Classici dell’economia politica 13(Milan: ISEDI).

Reese, Thomas J., 1979, “The Politics of Tax Reform,” Na-tional Tax Journal, Vol. 32 (September), pp. 248–54.

Roseveare, Deborah, and others, 1996, “Ageing Popula-tions, Pension Systems and Government Budgets:Simulations for 20 OECD Countries,” Economics De-partment Working Paper No. 168 (Paris: Organizationfor Economic Cooperation and Development).

Scott, Graham C., 1996, Government Reform in NewZealand, IMF Occasional Paper No. 140 (Washing-ton: International Monetary Fund).

Social Security and Medicare Boards of Trustees, 1995,“Status of the Social Security and Medicare Pro-grams: A Summary of the 1995 Annual Reports,” TaxNotes, Vol. 67 (May), pp. 837–43.

Tabellini, Guido, 1987, “Secrecy of Monetary Policy andthe Variability of Interest Rates,” Journal of Money,Credit and Banking, Vol. 19 (November), pp. 425–36.

Tanzi, Vito, 1993, “The Budget Deficit in Transition:A Cautionary Note,” Staff Papers, InternationalMonetary Fund, Vol. 40 (September), pp. 697–707.

39

References


40

REFERENCES

, 1995, “Government Role and the Efficiency ofPolicy Instruments,” IMF Working Paper 95/100(Washington: International Monetary Fund).

Teijeiro, Mario, 1996, La Política Fiscal Durante la Con-vertibilidad, 1991–95 (Buenos Aires: Centro de Estu-dios Públicos).

United States, Congressional Budget Office, 1996, The Eco-nomic and Budget Outlook: Fiscal Years 1997–2006(Washington: Government Printing Office).

United States, Department of the Treasury, 1994, Consoli-dated Financial Statements (Washington: Govern-ment Printing Office).

, 1996, Treasury Bulletin (Washington: Govern-ment Printing Office).

United States, Office of Management and Budget, 1995,Analytical Perspectives: Budget of United States Gov-ernment, Fiscal Year 1996 (Washington: GovernmentPrinting Office).

, 1996, Analytical Perspectives: Budget of theUnited States Government, Fiscal Year 1997 (Wash-ington: Government Printing Office).

, 1997, Analytical Perspectives: Budget of theUnited States Government, Fiscal Year 1998 (Washing-ton: Government Printing Office).

United States, President’s Commission on Budget Con-cepts, 1967, Report of the President’s Commission onBudget Concepts (Washington: Government PrintingOffice).

Van den Noord, Paul, and Richard Herd, 1995, “Estimat-ing Pension Liabilities: A Methodological Frame-work,” OECD Economic Studies, No. 23 (Paris:Organization for Economic Cooperation and Devel-opment, February).

von Hagen, Jürgen, and Ian Harden, 1994, “NationalBudget Processes and Fiscal Performance,” in Euro-pean Economy: Towards Greater Fiscal Discipline,Reports and Studies, No. 3 (Brussels: EuropeanCommission, Directorate-General for Economic andFinancial Affairs).

, 1996, “Budget Processes and Commitment toFiscal Discipline,” IMF Working Paper 96/78 (Wash-ington: International Monetary Fund).

Wagner, Richard E., 1976, “Revenue Structure, Fiscal Illu-sion, and Budgetary Choice,” Public Choice, Vol. 25(Spring), pp. 45–61.

World Bank, 1994, Governance: The World Bank’s Expe-rience (Washington).


41

Occasional Papers

Recent Occasional Papers of the International Monetary Fund

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

157. Central Bank Reforms in the Baltics, Russia, and the Other Countries of the Former Soviet Union, by aStaff Team led by Malcolm Knight and comprising Susana Almuiña, John Dalton, Inci Otker, CeylaPazarba

’s1oglu, Arne B. Petersen, Peter Quirk, Nicholas M. Roberts, Gabriel Sensenbrenner, and Jan

Willem van der Vossen. 1997.

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

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

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

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

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

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

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

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

148. Nigeria: Experience with Structural Adjustment, by Gary Moser, Scott Rogers, and Reinold van Til, withRobin Kibuka and Inutu Lukonga. 1997.

147. Aging Populations and Public Pension Schemes, by Sheetal K. Chand and Albert Jaeger. 1996.

146. Thailand: The Road to Sustained Growth, by Kalpana Kochhar, Louis Dicks-Mireaux, Balazs Horvath,Mauro Mecagni, Erik Offerdal, and Jianping Zhou. 1996.

145. Exchange Rate Movements and Their Impact on Trade and Investment in the APEC Region, by TakatoshiIto, Peter Isard, Steven Symansky, and Tamim Bayoumi. 1996.

144. National Bank of Poland: The Road to Indirect Instruments, by Piero Ugolini. 1996.

143. Adjustment for Growth: The African Experience, by Michael T. Hadjimichael, Michael Nowak, RobertSharer, and Amor Tahari. 1996.

142. Quasi-Fiscal Operations of Public Financial Institutions, by G.A. Mackenzie and Peter Stella. 1996.

141. Monetary and Exchange System Reforms in China: An Experiment in Gradualism, by Hassanali Mehran,Marc Quintyn, Tom Nordman, and Bernard Laurens. 1996.

140. Government Reform in New Zealand, by Graham C. Scott. 1996.

139. Reinvigorating Growth in Developing Countries: Lessons from Adjustment Policies in Eight Economies,by David Goldsbrough, Sharmini Coorey, Louis Dicks-Mireaux, Balazs Horvath, Kalpana Kochhar,Mauro Mecagni, Erik Offerdal, and Jianping Zhou. 1996.

138. Aftermath of the CFA Franc Devaluation, by Jean A.P. Clément, with Johannes Mueller, Stéphane Cossé,and Jean Le Dem. 1996.

137. The Lao People’s Democratic Republic: Systemic Transformation and Adjustment, edited by Ichiro Otaniand Chi Do Pham. 1996.

136. Jordan: Strategy for Adjustment and Growth, edited by Edouard Maciejewski and Ahsan Mansur. 1996.

135. Vietnam: Transition to a Market Economy, by John R. Dodsworth, Erich Spitäller, Michael Braulke, KeonHyok Lee, Kenneth Miranda, Christian Mulder, Hisanobu Shishido, and Krishna Srinivasan. 1996.

134. India: Economic Reform and Growth, by Ajai Chopra, Charles Collyns, Richard Hemming, and KarenParker with Woosik Chu and Oliver Fratzscher. 1995.

133. 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.


42

OCCASIONAL PAPERS

132. Financial Fragilities in Latin America: The 1980s and 1990s, by Liliana Rojas-Suárez 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 Baner-jee, 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, Tomás J.T. Baliño, and Charles Enoch. 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 Baumgartnerand 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, Jose

Martelino, 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 Mühleisen, Roger Nord, and E. Murat Uçer. 1995.117. Resilience and Growth Through Sustained Adjustment: The Moroccan Experience, by Saleh M. Nsouli, Sena

Eken, Klaus Enders, Van-Can Thai, Jörg Decressin, and Filippo Cartiglia, with Janet Bungay. 1995.116. Improving the International Monetary System: Constraints and Possibilities, by Michael Mussa, Morris

Goldstein, Peter B. Clark, Donald J. Mathieson, and Tamim Bayoumi. 1994.115. Exchange Rates and Economic Fundamentals: A Framework for Analysis, by Peter B. Clark, Leonardo

Bartolini, Tamim Bayoumi, and Steven Symansky. 1994.114. Economic Reform in China: A New Phase, by Wanda Tseng, Hoe Ee Khor, Kalpana Kochhar, Dubravko

Mihaljek, and David Burton. 1994.113. Poland: The Path to a Market Economy, by Liam P. Ebrill, Ajai Chopra, Charalambos Christofides, Paul

Mylonas, 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. 1994.110. 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, Paul

Duran, Gerwin Bell, and Zühtü Yücelik. 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 Kochhar

with Jun Ma, Simon N’guiamba, and Rajiv Lall. 1993.Note: For information on the title and availability of Occasional Papers not listed, please consult the IMF Publications Catalog or contact IMF

Publication Services.


Documents

OCCASIONAL PAPER - IMF eLibrary · Summary of Good Practices in Institutional Transparency6 2. Summary of Good Practices in Accounting Transparency8 3. Summary of Good Practices in