Indonesia Public Spending in A Time of Change

Public Disclosure Authorized
No. Report 19845-IND
Indonesia in PublicSpending a Timeof Change

30, March 2000 SectorUnit Management and Reduction Economic Poverty Asiaand PacificRegion East
Document of the WortdBank
CURRENCY EQUIVALENTS
(As of March 27, 2000) Currency Unit = Rupiah (Rp.)
US$=1 Rp. 7,475 ABBREVIATIONS AND ACRONYMS
ADB APBN BAKUN BAPPENAS BEPEKA BPKP DAPP

DG
Asian Development Bank State Budget of Income and Spending The Government Accounting Agency National Development Planning Board Indonesia's Supreme Audit Board Indonesia's Internal Audit Agency District Autonomy Pilot Program
Director General
NGOs O&M OECD PER PLN PUMA

REPELITA
- Non-Governmental Organizations - Operations and Maintenance - Organization for Economic Cooperation and Development - Public Expenditure Review - State Electricity Company - OECD Public Management Service
5-Year Development Plan
DM
DPRD
District Magistrate
Dewan Perwakilan Rakyat Daerah
SDO
UK
Subsidy Daerah Otonom

United Kingdom
GFS GTZ GDP IG IPPs IFIs IMF KKN

MoF
- Government Financial Statistics - German Technical Agency - Gross Domestic Product - Inspectorate Generals - Independent Power Producers - International Finance Institutions - International Monetary Fund - Korusi, Kollusi, Nepotasi
Ministry of Finance
UNCITRAL - United Nations Commission on International Law UU PD - Law on Regional Government (No. 22/1999) UU PKPD - Law on Fiscal Balance Between the Central Government and the Regions (No. 25/1999) USAID - United States Agency for International Development
VAT Value Added Tax
MTEF NPC
- Medium-Term Expenditure Framework - National Power Corporation
FISCALYEAR:
Government Bank of Indonesia
-
April 1 to March 31 April 1 to March 31
Regional Vice President: Country Director: Sector Director: Task Team Leader:
Mr. Jemal-ud-din Kassum, EAPVP Mr. Mark Baird, EACIF Mr. Homi Kharas, EASPR Mr. Sudarshan Gooptu, EASPR
ACKNOWLEDGEMENTS
'this reporthas beenpreparedby a core teamof principalauthorsled by Mr. SudarshanGooptu(World Bank)and comprisingof iMessrs. BeTtHofman(fiscal impactand sustainability, fiscal decentralization), AshokaMody(contingent liabilities), VinayaSwaroop(publicexpendituremanagement budgetary and and processes). Overallguidancefor the report was providedby Mr. VikramNehru (Lead Economistfor Indonesia, WoTrdBank). Key written inputs were provided by Messrs./Mme.David Hawes (infrastructure), Jorge Garcia-Garcia (public expenditures environmentalsustainability), on Jacqueline Baptist (education),Fadia Saadah (health), Samuel Lieberman (social sectors), Behdad Nowroozi (governmentaccounting and auditing systems). and Lant Pritchett (social safety nets). Invaluable cornents and suggestionson earlier drafts of this report were also providedby Rana PolackovaBrixi, Piyush Desai(IME), MohammadFarhandi,Ejaz Ghani,RichardNewfarmer,Alan Ruby,AnthonyToft, and JimStevenson (IMF). Themain findingsof this reportare on the basis of the WorldBankmissionthat visitedIndonesia betweenMay4-14, 1999in Jakarta,alongwith ongoingsector-specifio work at the World Bank Office in Jakarta. The mission was joined by Messrs.James Alm, Roy Bahl, Lucky Sondakh, Syahruddin, WerryDarta Taifur(USAID-Consiltants), providcdbackgroundpaperson budgetary and who processesat the sub-national levelsof government.T'hecounterpart team of the Government Indonesia of is headed by Dr. Soekarno Wirokartono(B3APPENAS), representativesfrom the Ministries of with Finance,Home Affairs,and the line ministries(at both Central and Provinciallevels). Invaluabledata I manipulation expertise was provided by Mr. Nunu Hendrawanto, Mmes. Magda Adriani and Barbara
I Ossowicka. Document processing was undertaken by Ms. Muriel Greaves. Peer Reviewers were Prof. Sri
i Mulyani Indrawati (Univ. of Indonesia), Messrs. Malcolm Holmes (PRMPS), William Allan (IMF), and GrahamScott(New Zealand).
INDONESIA PUBLIC SPENDING IN A TIME OF CHANGE TABLE OF CONTENTS
Pages EXECUTIVE SUMMARY................................................................... CHAPTER 1: PUBLIC SPENDINGIN A CRISIS......................................................... Introduction ................................................................... The Unavoidable Build-Up of Fiscal Pressure................................................................... Envisaged Fiscal Stimulus Did Not Materialize.................................................................. Deeper in Debt .................................................................. Fiscal Risks from Off-Budget Claims on Government are on the Rise................................ Domestic Resource Mobilization Efforts Need to be Stepped Up........................................ Policy Implications .................................................................. CHAPTER 2: IMPROVING THE BUDGET ALLOCATION PROCESS....................... A Necessary Response in a Constrained Fiscal Environment ............................................. Reshaping the Roles of Key Budgetary Institutions............................................................ Allocating the Budget in a Democracy.................................................................. Policy Implications .................................................................. CHAPTER 3: MANAGING ACROSS LEVELS OF GOVERNMENT............................ The Process Has Begun................................................................... The Nature of the Decentralization in Indonesia................................................................ Budgetary Impact of Decentralization................................................................... Impact on Inter-Governmental Relations................................................................... Issues in the Implementation of the Reforms................................................................... Decentralization Could Reinforce Civil Society Participation.............................................. Policy Implications ................................................................... REFERENCES ................................................................... i 1 1 2 5 6 10 13 16 18 18 30 31 34 36 36 37 39 39 42 48 50 52
TABLES IN TEXT Table 2.1: Table 2.2: Table 3.1: An Illustrative Performance Budgeting Framework............................................. 32 Rights of Members of Parliament in Budgetary Matters.................................... 34 Impact of the Fiscal Decentralization Law........................................................ 38
FIGURES Figure 1.1: Government subsidies have increased since the crisis . . ...................................... 3 Figure 1.2: Only 73 percent of the development budget in FY98/99 was actually 4 spent................................................................... 5 Figure 1.3: Not stimulating...................................................................
Figure Figure Figure Figure Figure Figure
1.4: 1.5: 2.1: 2.2: 3.1: 3.2:
Indonesia's fiscal developments at a glance ............................................... A long way down ..................................................... A changing environment comes with risks and opportunities.............................. Decision tree for evaluating public programs..................................................... Indonesia is highly fiscally centralized..................................................... Increased inequality.....................................................
7 8 19 24 36 40
BOXES Box Box Box Box Box Box Box 1.1 1.2: 1.3: 2.1: 2.2: 2.3: 2.4 Contingent Government Liabilities: A Hidden Risk to Fiscal Stability. Contingent Liabilities in Malaysia .13 Problems With Contingent Liability Management in the Czech Republic. Malaysia's Public Budget System.20 Medium-Term Planning and Budgeting.23 The impact of Educating Spending in Recent Years .26 Fiscal Transparency in South Africa .34 9 15
ANNEXES Annex 1: Annex 2: Annex 3: Status of Implementation of Recommendations of Public Expenditure Review, 1998.55 A National Risk Management Framework for Contingent Liabilities .59 OECD Experience in Public Expenditure Management Reforms.65
STATISTICALTABLES Table 1: Table 2: Table 3: Table 4: Table 5: Table 6: Table 7: Table 8: Table 9: Table 10: Govemment Revenue in current Rupiah.80 Government Expenditure in current Rupiah.81 Fiscal Summary in current Rupiah.82 Fiscal Summary (as % of GDP).83 Investment Budget (Development Budget) in current Rupiah - By Sector. Investment Budget (Development Budget) in constant 1993/1994 Rupiah. Current Expenditures - Functional Classification in current Rupiah .88 Current Expenditures - Functional Classification in constant 1993/1994 Rupiah Current Expenditures - Economic Classification in constant 1993/1994 Rupiah Current Expenditures - Economic Classification in constant 1993/1994 Rupiah
84 86 89 90 91
EXECUTIVE SUMMARY
1. Indonesia's new government will face a tight fiscal situation throughout its term. The crisis sharply increased debt levels and eroded government revenues, bringing to an end Indonesia's comfortable pre-crisis fiscal position. But not only has the fiscal situation changed, so has the entire environment for policymaking. Gone are the days when a few technocrats could decide on important fiscal issues and have them implemented without demur. Decisions on fiscal policy will now need to be subjected to the healthy rigors of democratic debate and tailored to the demands of a decentralizing political and government system. Calls for greater transparency and less corruption must be heeded throughout government-including in the design and implementation of the government budget. So, fiscal policy in the future will not just be a question of raising revenues and cutting wasteful spending (although both will be important). It will also require fundamental changes in the way Indonesia manages its public finances. 2. This report identifies the strategic priorities for restoring sound public finances. The three key findings of the report Tmphasize keydf three omaindg othen repol emphaiasizethunde to maiainin fiscal sustainability under a constrained budgetary environment and the need to improve the processe for making.buallocations processest forlemakingibudgetary s greater and budget implementation towards greater fica transparency. These fiscal trnprny.hseaesmaie are summaized below: costs of bank restructuring. 'Meanwhile, the revenue base remains weak because of a fragile economic recovery, and a plethora of tax exemptions and weak administration. Fortunately, the recent recovery in international oil prices will provide some room for budgetary maneuver, but much less than meets the eye because higher oil prices abroad also mean higher energy subsidies at home. 4. The report recommends a shift in fiscal policy focus towards maintaining fiscal sustainability in order to ensure a durable and sustained recovery. For this, of course, primary surpluses must be generated. But medium-term fiscal sustainability is at risk from three sources: (i) through macroeconomic fluctuations, since inflexible spending items have gone up and budget is now more vulnerable to variations in interest rates, exchange rates and inflation. Inflexible spending on debt service and civil service salaries take up a large proportion of the budget, so the remaining expenditure items in the budget are squeezed. Simulations show that the budget is more vulnerable to interest rate variability (because o t mato evcn oetcdb)ta exchange rate risk (and its consequences for foreign debt service). The reason is that oil angsrenuspoieteGvnm t and gas revenues provide the GovernTnent with a natural hedge against exchange rate fluctuations. (ii) through contingent liabilities, which pose fiscal risks of becoming actual liabilities suddenly, e.g. banking sector liabilities'; and
' The key contingentliabilitiesinclude:(a) guarantees
* From fiscal stimulus to fiscal sustainability. The fiscal stimulus that wvas envisaged in the past two years failed to emerge. The fiscal deficits for FY98/99 and FY99100have been less than planned.
3. Despite limited budget deficits, coveringdomestic banks' liabilities (excepting equity,of 3. Despite limitedbudget (b) impliedobligations privateinfrastructure course); to governmentdebt explodedin the wake of the providers(mainly in power generation also in toll but banking crisis. As a share of GDP, it is roads);(c) obligations minimum of pensionpaymentsto expectedto quadrupleby the end of FY99/00 civil servants; (d) off-budgetcredit schemes such as comparedto pre-crisislevels; the bulk of the KUT; and (e) a number of guaranteeschemesto the
increase is from domestic debt to cover the privatesector (some of which were institutedsince the crisisto restorethe flowof credit).
(iii) through decentralization. This process would be risky in itself, but the risk is enhanced if the envisaged transfer of resources is not matched by a transfer of expenditure responsibilities. As a result, the center's deficit could rise further. Also, decentralization poses a further fiscal risk through local government borrowing (under Law No. 25). If not carefully managed, local government debt may end up on central govemment books, as in many countries around the world. The report recommends that in order 5. to achieve fiscal sustainability over the medium- term, the Government should seek a cornbination of appropriate domestic revenue generation efforts, spending cuts, accelerated privatization, more aggressive asset recovery from uncooperative loan defaulters, new external finance, and to manage its contingent and other off-budget liabilities as part of an overall national risk management strategy. In addition, efforts to achieve fiscal sustainability should be supported through changes in budget management practices, including more comprehensive budget coverage, preparation of a medium-term budget outlook, and improved debt and risk management. 6. Specifically, the following measures are recommended in this regard: Short-term measures * Cut wasteful spending such as energy subsidies. These subsidies-which will amount to over 2.5 percent of GDP this year-are still poorly targeted, and mainly enjoyed by the better off. Enhance domestic resource mobilization efforts through a combination of measures that include: (a) reducing exemptions on the payment of VAT and certain income taxes; (b) limitingspecial economic zones to truly bonded areas; and ()) improving tax administration, and reducing corruption. c Systematically track all the government's allthea offSystgematil tperakons gviernnto s Off-budgetoperations with a view toward
better monitoring fiscal risks that stem from contingent liabilities. Include all government accounts in the Government's treasury and accounting system. and phase the Carefully plan decentralization process to devolve spending obligations together with revenues. Set up a debt management unit in the Ministry of Finance to manage domestic and foreign debt of Government. Make an inventory of all contingent liabilities to the budget, and include a contingency reserve in the budget to absorb shocks from contingent liabilities.
Medium-term measures
*
Adopt a multi-year rolling budgetary plan through a medium-term expenditure framework to track future budgetary consequences of policy decisions, and to make budgets more predictable for public spending entities. Create room for parliamentary influence over the budgetary framework early in the budget process. Adopt the IMF Code of Good Practices on Fiscal Transparency with a view towards: (a) enhancing clarity in the roles and responsibilities within government; (b) promoting greater information dissemination on government activities to the public; (c) fostering more openbudget preparation, execution, and reporting; and (d) ensuring that fiscal mformation IS subjected to independent assurances of integrity through regular publication of audits of public accounts. liTo limit the growth of contingent liabilities, the Goveement could consider budgeting guarantees and other contingent liabilities. Establish clear and credible rules for regional government borrowing limits. In light of Indonesia's fragile financial markets, direct or rule-based controls are likely to be more prudent than marketbased controls for some time to come.
..
Expand the function of the debt management unit in the Ministry of Finance (see above) to include a government risk management strategy that involves three complementary elements: (a) mitigation of the risk at source through sound microeconomic policies fostering efficient resource allocation and use; (b) transferring risk to third parties with incentives to manage and ability to bear the risk; and (c) monitoring and managing any residual risk that cannot be mitigated or transferred.
II. Improving budget allocation. The tighter fiscal conditions that exist today reinforces the need to achieve the best possible allocation of scarce resources. 7. Clearly, budget formulation is a political process. Experience suggests that a participatory process involving both the Parliament and civil society ensures that public spending choices are continuously evaluated. This tends to improve the quality of public expenditures and the efficiency of public services. This priority could be addressed by facilitating adequate information flows between stakeholders and a budget process that allows for political choices without a loss of fiscal discipline. Public resources will have to be allocated efficiently to those areas that will have the maximum intended impact in terms of public service delivery. To this end, institutional measures are recommended in the report to ensure that adequate audit, evaluation and feedback mechanisms are in place so that the mix of spending allocations in the budget is deemed acceptable to a wider spectrum of the public. This also will go a long way in minimizing the observed divergence between budget estimates and budget outcomes. . ., Indonesia's current budget process is 8. mainly a bureaucratic process. The line ministries submit their bids, and MOF and iii
BAPPENAS cut the bids down to the size required to fulfill the constraint on domestic borrowing. MOF has an elaborate system of expenditure norms, which are used to scrutinize the submitted budgets. Only when the draft budget is submitted to the President, some political decision-making enters the process. Individual ministers do not defend their ministry budget, and appropriations are structured by sector, not by ministry or program. Budgetary decision-making has focused primarily on government investment (i.e. the development budget) and less on decisions regarding the public policy role the government wants to play in any area. Although BAPPENAS has introduced a system for performance evaluation, this has had only limited impact on budget decisionmaking. 9. In this regard, the report recommends that the development and recurrent spending decisions need to be unified into a single budget and then projected out on a three-five year time horizon. In doing so, the budget appropriations could be structured by line ministry. Each line ministry would then have to provide the rationale for public intervention in that activity as part of the budget formulation process at all levels of government. 10. Specifically, the following measures are recommended in order to facilitate better, more transparent budgetary allocation decisions: Short-term measures . Require the line ministries to provide the rationale for public intervention in that activity as part of the budget formulation process. This principle should be applied at all levels of government. Structure budget appropriations by ministry, and discuss each line ministry to budget separately in Parliament minister's enhance individual
~~~~~~~accountability
*
ty Agree with parliament on rules for debates and changes to the budget.
Currently, the Fiscal Outlook document is discussed in Parliament at a plenary session that is held about two months prior to the budget debate. This is an important step towards increasing the "buy-in" from parliament. This fiscal outlook should provide the major budget aggregates and sectoral spending limits before detailed budget preparation starts. * Design earmarked grants for regions to finance national priorities implemented at regional level. * Design a monitoring system for regional government's performance on public service delivery. * With a view towards achieving greater transparency, accountability and fostering anti-corruption measures, make current civil service remuneration transparent by including all salary elements of civil servants in the recurrent budget. The ongoing transition to the adoption of the IMF-GFS Classification System would reinforce this measure. Medium-term measures Unify development l and recurrent spending into a single budget and then project this budget out on a three-five year time horizon. This measure could be tested on a pilot basis in some government agencies or spending entities in the short term. The line ministry appropriations could be ordered by the major programs of the ministry, and the budget documentation should include information on results achieved with spending, and expected results in the future. * Increase civil service salaries to become comparable with private sector wages. At the same time, de-link civil service salaries from development projects. Most of a civil servant's income should be in the form of base salary, rather than through project-related payments. 11. Democracies require full disclosure of public spending information to all stakeholders. This keeps bureaucrats and
politicians honest. It also encourages citizens to pay their taxes, for then they know what their money is buying. Greater transparency also helps communities and NGOs to monitor the quality of public spending programs. There remains the urgent need for greater clarity in the roles of Parliament, Ministry of Finance, BAPPENAS and the line ministries in the budget process. A participatory process involving both the Parliament and civil society ensures that public spending choices are continuously evaluated. This tends to improve the quality of public expenditures and the efficiency of public services. To this end, appropriate guidelines and rules of engagement between the Government and Parliament in the budget formulation process need to be specified. III. Improving budget implementation. When one turns to the implementation of the public spending decisions, as the government rightly believes, decentralization can indeed be a powerful tool to improve public service delivery and accountability in Indonesia. But, if not well managed, it risks fiscal balance at the center. 12. For fiscal decentralization to be successful, a strong central government is needed to lead and coordinate the process, especially in the short run. It must convey clear policy guidelines for ensuring a balance between revenues and expenditures at the local level, a transition plan to achieve this, and hard budget constraints for local governments to instill fiscal discipline including through clear rules on the issuance, reporting and monitoring of guarantees. To this end, the report recommends measures to improve the adequacy, relevance, timeliness and accessibility of government financial reporting and monitoring systems. In the medium-term, strengthening local accountability is recommended through a larger local tax base, increased transparency, and prudent access to financial markets. 13. There remains the urgent need for better coordination of the process towards implementation of the two fiscal and
iv
administrative decentralization laws (Law No. 22/1999 UW PD and Law No. 25/1999 UU Central in the PKPD). Currently, Government, the Ministry of Finance, the Department of Home Affairs and the Office of the State Minister of Regional Autonomy are involved in this effort, but there was a lack of clarity on who was responsible for what aspect of this implementation process. Clearly, this is a complex task and may need attention from the highest levels in government. Specifically, in the open, democratic, 14. and decentralized era that Indonesia is currently embarking on, the following measures should enhance fiscal transparency and improve public service provision: Short-term measures management financial ptrengthen practices in Government by accelerating the pace of computerization in Government. This will improve the adequacy, relevance, timeliness and accessibility of government financial reporting and monitoring systems. To enhance the efficiency and effectiveness of independent audits of public accounts, there is an urgent need to clarify the
mandates, scope, degree of independence and roles of BEPEKA (the Supreme Audit Board), BAKUN (the Government Accounting Agency), B3PKP (the government's internal audit agency) and the Inspectorate Generals (IGs, the internal auditors of each Ministry). Define detailed expenditure assignments across levels of Government. Design an equalization grants scheme, including transition arrangements. Specify a regional borrowing framework.
Medium-term measures
*
. *
Tighten treasury management, expand regular and timely reporting on budget implementation, independent audits of final accounts, and widely disseminate outcomes of those audits. At each level of government, set up a legislative mechanism that ensures follow up action on audit findings. Issue uniform accounting standards for the public sector (including SOEs and sub-national administrations). Draft clear implementing regulations for domestic).
Public Spending in a Time of Change
1.
PUBLIC SPENDING IN A CRISIS
spending has jumped, and revenues have plummneted.Some of the fiscal pressures that have emergedfrom the crisis can be attributed to the jiscal risks due to the government's contingent liabilities. These have stemmed fromn guiarantees previously given that have been realized, and new guarantees that Governmienthas issued in the wake of the crisis. At least some of these liabilities, which are vulnerable to interest rate, foreign exchange rate and price fluctuations, will grow in the short-term with significant potential costs. The fiscal stimulus that was envisaged in the past two years failed to emerge. As the economnyrecovers, Government needs to shift its focus towards maintainingfiscal sustainability in order to ensure a durable and sustained recovery. To this end, the Government should seek a combination of appropriate domestic revenue generation efforts, spending cuts, accelerated privatization, more aggressive asset recovery from uncooperative loan defauilters, new external finance, and to manage its contingent and other off-budget liabilities as part of an overall national risk management strategy.
INTRODUCTION
he Governmentis facing a grimfiscal sitiuation.The crisis has left it deeply in debt,
1.1 The implications of the economic crisis are being played out in an era of fundamental political and social change in Indonesia. The crisis has fueled a growing sentiment among people to change the way business is done in Indonesia, both within and outside government. And it has lowered the tolerance for corruption in government. Responding to these changes poses difficult challenges. Inaction may delay chances of a rapid and sustained recovery. 1.2 The objective of this report is to address key issues in public expenditure management that are emerging under these changing circumstances. It suggests ways of redesigning the budgetary process so as to ensure that public spending will have a significant impact in shielding the poor and the vulnerable from the crisis in the shortrun, and to provide an approach for making fiscally sustainable public expenditure choices, with maximum development impact, over the medium-term.' The fundamental issues that need to be addressed when
Similarly,a more detailedassessment the ongoing of
process of Decentralization in Indonesia is currently underway jointly by the World Bank and the TMF.
deciding where, how much, and how public funds are spent include: (i) the effect of the ongoing crisis on the trade-off between macroeconomic stabilization and public expenditure policy decisions; (ii) the risks to the budget from contingent liabilities of the government (e.g. from the evolving cost of bank recapitalization); (iii) the changing roles of the National Planning Agency (BAPPENAS), the Ministry of Finance and the line ministries in budgetary decisionmaking and their implementation given the ongoing process of fiscal and administrative decentralization in Indonesia; and (iv) the implications of Indonesia's new democratic freedoms and the changing nature of monitoring public expenditures towards achieving better results. In this chapter and the two that follow, these above questions are addressed. The overall theme of this report being-enihancing transparency in the budgetary processes in Government and its management to ensure the best possible use of tax-payers' money. 1.3 This report builds on the previous
Bank work on Indonesia's public finances, along with ongoing work by the ADB, IMF 1
and several bilateral agencies on budgetary
Public Spending In A Crisis processes and decentralization in Indonesia. For instance, an assessment of the institutional arrangements for domestic resource mobilization and tax policy in Indonesia is currently being conducted through technical assistance from the IMF (Fiscal Affairs Department). The ADB's Community and Local Government Support Sector Development Project (March 1999) and the World Bank's Kecamatan Development Project (April 1999) contain components that aim to streamline and strengthen public expenditure management at the sub-national levels of government. Technical assistance activities that address measures to encourage greater accountability and transparency in the use of public funds in the context of the ongoing process of fiscal decentralization in Indonesia are being provided to the Government with support from official bilateral agencies (such as GTZ, USAID) and the international financial institutions (IFIs). THEUNAVOIDABLE BUILD-UPOFFISCAL
PRESSURE
crisis, but they were masked as the economy boomed. For instance, some surveys that depicted investors as bullish on the Indonesian economy, also showed concems about mismanagement of the public sector including corruption, red tape and lack of transparency in public sector decision2 making. The 1998 Public Expenditure Review (undertaken by the World Bank) noted that behind the impressive growth period in the country, the public sector has indulged in favoritism and rent-seeking behavior that has raised the cost of public services. Based on the existing evidence, one could argue that while public sector mismanagement may not have precipitated the crisis in Indonesia, it certainly has been a major constraint in dealing with it' Aggregate Fiscal Situation 1.6 The crisis has sharply reduced revenues. Led by the decline in oil prices,
domestic revenues excluding privatization

fell from almost 16 percent of GDP in FY1996/97 to a projected 10.6 percent this fiscal year. Surprisingly, income tax revenues maintained their level of about 4.5 percent of GDP. The key reason for this was that the increase in interest income as a consequence of tight monetary policy compensated for the revenue loss due to the decline in corporate profits. In contrast, VAT revenues and import taxes dropped sharply in line with domestic absorption. As economic recovery sets in, these taxes could rebound to their former level, although this revenue recovery will most likely be led by oil and gas taxes. The upswing in oil prices since early 1999 is likely to result in higher revenues. In addition, inflation is expected to be lower than expected at the time of budget preparation, and so is the exchange rate.
1.4 Like its East Asian neighbors, Indonesia had demonstrated the virtue of overall fiscal prudence in the pre-crisis years. In early 1990s, the primary fiscal balance (i.e. fiscal deficit net of interest payments) was in surplus for years. The excellent record of sustained high economic growth and continued poverty reduction neglected to highlight the medium-term trade-off between macroeconomic stabilization and public expenditure policy decisions. Indonesia's successes in reducing the poverty rate (from 70 percent in 1970 to 10 percent by 1996) and achieving near universal literacy rates over the past thirty years prior to the crisis clouded any debates about whether public spending was efficiently achieving the government's strategic priorities or not. 1.5 The recent crisis has changed the
See D. Kaufman, G. Mehrez, and S. Schmulkler
environment considerably. The fiscal pressures that have built up as the crisis unfolded brought to the fore fundamental has issues about the way governmentresources
are being spent and what they are buying.
(1998). "The East Asian Crisis: Was it Expected."

(mimeo), World Bank. The perceptions of multinational corporate and industrial staff that Indonesia has a major governance problem has also
beendocumented Bardhan(1997). in
See World Bank. 1998. "Indonesia in Crisis: A Macroeconomic Update", (Washington D.C.: World Bank)
3
These issues were present even prior to the

2
Public Spending in a Time of Change These forces will tend to add to the recovery of revenues. 1.7 Despite efforts to the contrary, spending in terns of GDP remained at around its pre-crisis level. The main change was in the composition of spending, with subsidies showing the largest increase in terms of GDP from virtually nothing in FY1996/97 to 4 percent of GDP in 1998/99. (Figure 1.1) Social safety net spending saw an equally rapid increase to about 1.8 percent of GDP, and interest payments on external debt rose as well, mainly under the influence of the real depreciation of the Rupiah. In
Figure 1.1. Government subsidies have increased since the crisis...

Central Government Routine Expenditures, 1994/95 - 1999/2000 Constant 1993/1994prices (Billion Rupiah)
30,000
40.0O0
t
SOSCO -,iv.>/.
r * . ubsidies S -ii5ti~~~~~~~~~~~~I!
i
0~ DebtServce
M-00
o Tmrasfer to
Regions
* Materiai
LP rsonnel
Source: Government of Indonesia. Note: 1) Preliminary actual figure 2) Revised budget Source
1994709sa 1995/1996 199411997 1907/1968 1940/1999/i1) 19-e-02)
Expendiures
return, other spending items had to be compressed: personnel, material, transfers to the regions all experienced declining spending, while project aid barely kept up with the 2.5 percent of GDP spent before the crisis. 1.8 For FY1999/2000, the costs of bank restructuring will drive up spending. According to current estimates these interest costs on the newly issued bank restructuring bonds add up to some Rp. 34 trillion, although the budget shows only half of that, because asset recovery is subtracted from the interest due. The recent decline in interest rates make the Rp. 34 trillion an increasingly reasonable estimate. Furthermore, increases in civil service salaries and cuts in subsidies are planned. For the latter, although Government has eliminated some petroleum subsidies, others remain, and become more expensive because of the steep rise in oil prices. Moreover, electricity subsidies will 3
have to rise too, if oil prices stay at their current level. In addition, Government may need to clear the arrears on its subsidy payments to SOEs that have accumulated over the last few years4. 1.9 Alongside the increased demands on the public coffers during the crisis, the implementation record has been somewhat slower than budgetary allocations would suggest (Figure 1.2). The key reason for this was lower than expected disbursements of donor funds. Now that government savings have turned negative, these donor funds fully finance the development budget, either through projects (35 percent) or through general budgetary support (65 percent). Project funding was only two-thirds of what was budgeted for, whereas general budget support fell short as well because of policy
Some Rp. 16 trillion in arrears are outstanding to Pertamina and PLN.

4
Public Spending In A Crisis
Figure 1.2. Only 73percent of the development budget in FY98/99 was actually spent...
l ,~ ,I, l0t,l0090
implementation slippage. The shor-tfalls in implementing the budget were also not evenly distributed. Spending by Goverment institutions and on national defense and security was more than 20 percent above budget while the categories Housing and Settlements, and Regional Human Development saw less than 60 percent of their budget being actually realized.' 1.10 Savings have collapsed. Government savings collapsed from 7.8 percent of GDP in 1997/98 to 1.2 percent of GDP in 1998/99. They are expected to tun negative in FY99/00 (i.e. dis-saving). This decline in Govenment savings accounts for the bulk of the decline in national savings from 26 to 17.7 percent of GDP. Unlike most countries that see a sharp decline in Government savings, private savings declined as well. The decline in savings is more pronounced than what the fiscal numbers (in IMF format) show, because it excludes interest payments on the bank restructuring bonds of some 2.8 percent of GDP. In reality, savingsare even lower,
Besides budget finance, the military has revenues from its business operations, which range from business to timber. Most of these businesses are brought under Foundations which are run in cooperation with private sectors firms. Each branch of the military has its own foundation: Karita Eka Pasi for the nmy; Bhumyanca for the Navy; and Adi Upaya for the Air Force. As with the private sector, the revenuesof these foundations have also been adversely affected by the crisis. Typically, these off-budget revenues are used for maintenance, salary stpplements and military equipment purchases.
because a part of the development budget goes towards paying civil servants compensation rather than the acquisition of capital goods.' Sectoral Composition of Expenditures 1.11 The fiscal pressures that have resulted from the crisis have also had their toll on the spending on Indonesia's social sectors, and hence, on public sector service delivery in these sectors. Total realized public spending on education has declined both as a share of total goverment expenditures (7.7 percent in 1996/97 to 3.9 percent in 1997/98) and as a share of GDP (1.4 percent in 1996/97 to 0.7 percent in Similarly, realized health 1997/98). expenditures have suffered as well. Total public sector health spending has fallen by 8 percent in real terms in FY97/98 and a further 12 percent in FY98/99 (declines of 9 percent and 13 percent respectively in real per capita terms). The gap between
6
These are normally due to the honoraria and
transportationcosts that are associated with

development project implementation. The office stationary and supplies related to the implementation of externally-funded development projects are also financed out of the development budget and not by the routine budget. Since the main focus of this report is on budgetary processes, a more detailed assessment of public expenditures in key sectors of the Indonesian economy is provided in Annex I elsewhere. For instance see Health Expenditure Review (t999), Environment Expenditure Review (1998), Review of Social Safety Net budget as reported in President's Report for Social Safety Net Adjustment Loan (SSNAL).
Public Spending in a Time of Change budgetary allocations and actual spending on health increased as well during the crisis (rising more than two-fold between 1997/98 and 1998/99). Social safety nets that were instituted with the objective of shielding the poor and vulnerable groups sromthielding pooe effectsof the able g sou from the ill effects of the crisis, have also grossly under-spent their budgetary allocations. In fact, just before the end of FY98/99 the social safety net program implementing units were granted a special three-months extension of their fiscal year spending authority. This measure was also called for as a result of the need to set up appropriate entities in government, with active civil society involvement, to monitor the use of funds earmarked towards social safety net expenditures. So far no resources from the 1999/2000 budget have been spent on employment and income support programs, pending improvements in targeting and financial monitoring.
Figure 1.3 Budget not stimulating enough
Percent GDP of
2.00
T----
2 00~T
1.000.0
'
Actual balance
-------
-2 00
ei2GOI ll ! g Budget Target
-3.0
l s P
--
QActualFiscalDeficit MCyclically Neutral \ \ , | FiscalImpulse
~~~~~~~~~~~~Balance
-4.00
-5.0
Fiscal
-6.00 jj
-700 -8.f__
GOI ,
impulse
3
96/97
Cyclically neutral Target

97/98 98/99
~~~balance
99/2000
ENVISAGED FISCAL STIMULUS NOT DID

MATERIALIZE
1.12 With the aim towards generating a fiscal stimulus, the government's budgets in the post crisis period envisaged a significant increase in government expenditures relative to pre-crisis levels. The State Budget aimed at achieving budget targets of 7.8% of GDP for FY1998/99 and 5.8% of GDP in FY1999/2000. In practice, the deficit turned out to be substantially less than planned, just
partly because the assumptions made in the budget on the exchange rate and interest rates were too pessimistic, so that the rupiah equivalent of debt service payments from the budget were lower than anticipated. (Figure 1.3). Meanwhile, the fiscal impulse (which is an indicator of change in fiscal policy stance) is estimated to be almost zero (i.e. no budget stimulus) this fiscal year. This result is
2% of GDP last fiscal year and an estimated 8 4 percentof GDPthis fiscalyear . This is
Fiscal impulse is the change in fiscal stance, where the fiscal stance is the difference between actual and cyclically adjusted deficit. The "cyclically adjusted
deficit" is
defined as
[Revenues/GDP] -/-
[Expenditures/GDP]*[Potential GDP/GDP]. Potential GDP was calculated with a Cobb-Douglas Production Function, assuming I percent Total Factor Productivity (TFP) growth (see Armstrong et. al and Sarel). FY96/97 is taken as the base year, for which the fiscal stance is zero by definition.
Public Spending In A Crisis sensitive to assumptions about how potential output growth is computed and the base year used to determine fiscal stance. 1.13 Lower spending was the main cause of the lower than anticipated fiscal stimulus. This in turn can be attributed to three factors: (i) delays in the disbursement of foreign loans; (ii) delays in allocating spending authority to project and program implementing units in government; and (iii) the prohibition on domestic financing of the 9 fiscal deficit. In addition, concerns about how fast the government could increase spending, and in particular, the governance of the social safety net programs, led to a slowdown of actual disbursements of budgeted resources, and hence, a smaller realized fiscal deficit than was anticipated. 1.14 The budget deficit for FY1999/2000 may again be lower than planned. Revenues are likely to outperform the budget estimates by a margin of at least 1 percentage point of GDP because of higher oil revenues lower; development spending; and lower than anticipated interest spending on bank restructuring bonds due to lower interest rate. Also, the implied elasticity of domestic (nonoil) revenue growth to GDP growth in the budget is lower than historically evident. This elasticity is assumed to be almost zero in the budget numbers, while even in the crisis year 1998/99, this elasticity actually reached 0.64. With improved tax administration, these tax revenues could increase further, thereby lowering the fiscal deficit as well. 1.15 The first quarter of FY99/00 seems to confirm these fears: higher revenues and lower spending than planned caused the budget to be in surplus over that period. Revenues have rebounded due to higher oil prices and higher income tax revenues. The latter is in part due to the high interest rates that still prevailed in the first quarter of the calendar year (and that are taxed in the second quarter). But withholding taxes on wages are also up-which is signaling the possibility that wages may now be catching up with inflation. 1.16 Meanwhile, spending this fiscal year is off to a modest start once again. Many project offices have yet to receive their authorization to spend. The fact that the Central Government has introduced a new system for distributing this spending authority-in itself a laudable initiativedoes not necessarily deal with this obstacle: spending authorizations are now stuck at the KPKNs, and so the regional governments are slow in disbursing the funds. For instance, the Social Safety Net funds are not disbursing at all in recent months. The Government is trying to address concerns from civil society and the World Bank about the possible misuse of public funds and to ensure that appropriate safeguards are in place before expenditure authorization are issued. Increased pace of spending must be undertaken without compromising on the quality of the projects and the integrity with which the public funds are spent. DEEPER DEBT IN 1.17 The crisis left the Indonesian Government deeply in debt. As a result, the inflexible public expenditure components (i.e. debt service payments) have increased, thereby reducing the Government's budgetary room to maneuver. Total debt jumped from a pre-crisis level of 24 percent of GDP to more than 90 percent of GDP by the end of the current fiscal year. For the first time since the 1970s, the Government will also be in debt domestically, due to the Rp. 560 trillion (45 percent of GDP) in bonds it had to issue in the course of bank restructuring this fiscal year. Meanwhile the revenue base remains weak, although the recovering oil prices are providing some budgetary relief.
The prohibition on domestic borrowing is part of the MPR instructions to the Government as well as an integral part of the IMF program.
Figure 1.4: Indonesia's fisca developments at a glance

The crisis has driven the budget into deficit
Percent 0f GDP 2D 1 D
But Indonesia usually underbudgets its revenues

5i-00
Rupiah
~ ~~r
.......
--
- - -
- . - ...... .. . ,
1bO,WO~.
Revenues
150' 1 ..
0-
-_'10W
oo/00'
Eopend,tures
Expenditufgs
10 0
~ ~~ ~ ~ ~ ~ ~~ ~ ~ ~ ~~~~~~~~~cta ~~ ~ -~ - , . .. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1200/0 Deficit / .

10.WAtal/
- -W -
Budgeted
94u95
0950
s%/7
97/95
92000
s99/0
999
9O
Note: all 1998/99numbers are preliminary; all 1999/2000numbers are budgeted
..and spends less than planned

8,11-on R.p,ah
000,000 220.00/ --
... keepingfiscal
Per.e.t
stimulus too small.

...
Actual
of GDP
-..--.-.-..--.--...------.-...-.....-.-........-.....--......
.
Budgeted 2WOW
,.
\Bj
60.00/ 140.000 120.WW ^;;/
.-
94t94~~~~~~~~~~~~~~~~0/0 00/00
stU97
07/00
00/00
00/00
Realized
bs.ow > =/
.
T-
~~~~~~~~~ ~ ~~~~~~~~~~Budge ~ , ,.
W..
s2.0w
The crisis has led debt to explode....
If effective interest rates on recapitalization bonds are low, it will tend to keep debt at sustainable levels.
0ID .5
0,,44/4/0l
-10
Less Sustainable
.15.
YVith n.rket / nstes
so.n
- '.''.'' __ _
..
'.'
,:.:.i
.....
,,.
,.
aSSs
9s/99
96l97
97z9,
~~~~~~93199
sD
-30
sMore
O .. ......FY94/5
Sustainable
A
r
rjce
FY9519/
FY96/97
FY97/98
FY99/9
FY99/00
Debt includes bank restructuring bonds from 1999/2000
Note: the graph pictures the standard debt sustainability condition: Ad=[(i-/(I +g)jd,I +pb, where d is debt to GDP, i is the nominal interest rate, g the nominal growth rate, and pb the primary balance as percent of GDP
Public Spending In A Crisis nominal interest rates and that the effective 1.18 The sharp real depreciation of the Rupiah added significantly to Indonesia's interest rate paid on government debt is far 1 debt burden. Of the 24 percentage point of below market rates." Thus, even with small primary deficits-the budget deficit net of GDP increase in external debt between interest payments-the debt to GDP ratio FY96/97 and FY98/99, two-thirds was will tend to fall over time (Figure 1.5). In due to exchange rate variations, one-third to anld only Figure1.5: A longwaydown andisbursements. obv-thios ay (Debt GDP) to disbursements. Obviously, any appreciation of the Rupiah will Govert help reduce the external debt burden as a share of GDP. If this Ol ~~~~~~~No primary surplus \
s
is accompanied by the continued

reduction in interest rates and
07-
EU-Norm
declining inflation, Indonesia's debt burden could be maintained

at fiscally sustainable levels
\
0
With 2 perCent
through
prudent
debt
Pe.rer
FY9s9!0
ys ....... l
FY0ZO03
----------.---- ---- --------------------------FY04CQO FY0700B FY10111
management practices.
3/14
1.19 Prudent debt management, in a fiscally constrained
ANote: chartplotsprospective The debt/GDP ratios underthe assumption gradually of falling inflation interest and rates, a and
rebound of growth to 5 percent in FY2001/02.
environment that Indonesia finds itself in today, will require the Source: MOF, IMFdata;staff estimates authorities to carefully monitor the ratio of debt to government reality, however, primary balances have revenues carefully over time. In come down. The record surplus of 3 order to track fiscal flexibility, the percentage point of GDP in FY97/98 composition of the expenditures will also withered to a mere 0.1 percentage point in have to be closely observed-specifically, 1 FY98/99,2 and is projected to turn slightly the ratio of inflexible components of public negative in FY1999/00. Therefore, achieving spending to total expenditures. One such fiscal sustainability over a medium-term component is the interest bill on the bank horizon will, first and foremost, require restructuring bonds that the government has primary surpluses. With interest payments issued this fiscal year. This will depend on (at market rates) rising to some 4-5 percent of the behavior of domestic market interest rates GDP, attaining primary surpluses that are in Indonesia. Illustrative scenarios using a comparable to pre-crisis levels would limit standard macro-fiscal analytical framework show that at projected rates of interest, GDP the budget deficit to less than 4 percent of growth, and primary balances, the debt-toGDP. GDP ratio is likely to fall over time despite the rapid growth in government debt' 1.20 The low effective interest paid over Government debt is a result of three factors: (Figure 1.4). The reason for this is that (i) the interest rate; (ii) the real appreciation nominal growth is expected to outpace of the Rupiah which reduces the Rupiah value of interest paid in foreign currency; and If one uses the crude definition of fiscal (iii) the low interest rate paid for the bank
sustainability the basisof the "fiscalrules approach" on
(as opposed to the more sophisticated "balance sheet approach"), debt is deemed to be sustainable if the ratio of debt to GDP does not increase over the projection horizon. On this basis, Indonesia's debt burden seems sustainable for now. l l The effective interest rate is calculated as the actual interest paid as a percentage of outstanding debt. 12 This definition excludes privatization revenues, which is considered as financing.
Box 1.1:Contingent Government Liabilities: HiddenRisk to FiscalStability A Recenteventsin Indonesiahave shownthat contingent liabilities rapidlyturn into actualliabilitieswhich can are immediately due-with seriousfiscal implications. extent of the materialized The liabilitiesis typicallya functionof (a) the severity of the crisis;and (b) the underlying structureof the activitiesmost affected.All sourcesof fiscal risksmust be addressedif governments to avoid suddenfiscal instability realizetheir are and long-term policyobjectives. Thereare four types of liabilities: direct and contingent, each of which is either explicit or implicit. Direct liabilitiesare obligationsthat will arise in any event.They are predictableaccordingto specificunderlying economicand policy factors. Contingent liabilitiesare obligations are triggeredby a particulardiscrete that event,whichmay or may notoccur.They aresometimes optimistically deemedunlikelyto occurat the timethe obligation assumed.The probability contingency occurand the magnitudeof the requiredpublicoutlay is of to are exogenous(e.g., the occurrenceof a natural disaster) or endogenous(e.g., implicationsof market institutions of the designof government and programson moralhazardin the markets)to government policies. Explicitliabilitiesare specificgovemment obligations definedby law or contract.The govemmentis legally mandatedto settlesuch an obligation whenit becomesdue. Implicit liabilitiesrepresenta moral obligation or an expectedburdenof the government in the legal sense,but basedon public expectationsand political not pressures. Basedon thesecharacteristics, thereare four typesof fiscalrisk: * Directexplicitliabilitiesare the main subjectconventional fiscal analysis.The repaymentof sovereign debt,expenditures based on budgetlaw in the currentfiscal year,and expenditures the long term for in legallymandateditems,suchas civilservicesalariesand pensions,and in somecountrieseventhe overall socialsecuritysystem. * Direct implicitliabilitiesoften arise as a presumedconsequence public expenditurepoliciesin the of longerterm.Giventheir implicitnature,these obligations not capturedin government are balancesheets. Typically, they are highfor demographically drivenexpenditures. example,in a publicpay-as-you-go For scheme, future pensionsconstitutea direct implicitliability, the size of which reflects the expected generosity andeligibilityfor a benefit,and the futuredemographic economicdevelopments. of and * Contingent explicitliabilitiesare government legalobligation make a paymentonlyif a particular to event occurs. Sincetheir fiscal cost is invisibleuntil they are triggered,they representa hidden subsidy,blur fiscal analysisand drain governmentfinancesin the future. Therefore,state guaranteesand financing throughstate-guaranteed institutions, lookpoliticallymore attractive than budgetarysupportevenif they are more expensivelater. In the markets,contingentgovernmentobligationsmay immediatelycreate moral hazard,particularlyif the govemment guaranteesthe whole rather than a part of the underlying assets,and all risks ratherthan selectedpoliticaland/or commercial risks. State insuranceschemesoften cover uninsurablerisks of infrequentlosses that are enormousin total magnitude.Thus, rather than financing themselves from fees,theyredistribute wealthand rely on government financing. net * Contingentimplicit liabilitiesare not officiallyrecognizeduntil after a failure occurs. The triggering event,the value at risk, and the requiredsize of government outlay are uncertain.In most countries,the financialsystemrepresentsthe most seriouscontingentimplicitgovernmentliability.Experienceshave indicated marketsexpectthe government help financially beyondits legal obligation stability that to far if of the financialsystemis at risk. Fiscalauthorities often compelled to coverthe uncovered are also losses and obligationsof the centralbank,sub-national governments, state-owned large privateenterprises, and budgetaryand extra-budgetary agencies,and any other institutions politicalsignificance.Contingent of liabilities rise with weaknessesin the macroeconomic framework,financial sector, regulatory and supervisorysystemsand information disclosurein the markets.With privatecapital flows,for instance, such weaknesseselevatethe risks in assets valuation,intermediation borrowingbehaviorsin the and markets.
Source: Hana Polackova, World Bank, PREM Note, 1998
restructuring bonds. The Rp. 34 trillion in interest due for bank restructuring this year implies an interest rate of less than 8 percent nominally for the bank restructuring bonds and an assumed asset recovery rate of Rp. 17 trillion. 9
1.21 Under current circumstances, the high levels of Government debt are here to stay. Without primary surpluses on the budget, it will take until FY08/09 until debt is back to 60 percent of GDP-the European Union norm that has become an international
Public Spending In A Crisis standard for a prudent level of debt. But even with primary surpluses similar to those of Indonesia before the crisis, our analysis suggests that debt will not return to its precrisis level until FY2013/14. Increased vulnerability 1.22 The high levels of government debt makes the budget more vulnerable to shocks. Variations in interest rate, exchange rates, and inflation all affect the amount of
government debt service. Specifically, foreign debt is affected by variations in the
1.24 The government's fiscal situation is more vulnerable to interest rates than the exchange rate. This is partly due to the significant domestic indebtedness levels and because the Government's oil revenues provide for a natural hedge on its foreign exchange denominated fiscal obligations. Oil revenues will go up in Rupiah, if the exchange rate appreciates. But interest payments and oil subsidies more than offset this effect.
FISCALRISKS FROMOFF-BUDGETCLAIMS ON GOVERNMENT AREON THERISE
exchange rate, whereas domestic debt is affected by variations in interest rates (due to the market rate linked bonds) and inflation (due to the indexed bonds). 1.23 The lower interest rate and the stronger Rupiah this fiscal year have already saved the Government considerable amounts of debt service. But the level of interest rates in Indonesia also influences the exchange rate, and, through the demand for money, also the inflation rate. Illustrative scenarios that were conducted for this report using a standard macro-fiscal projection model suggests that a 1 percent increase in the domestic market interest rate (in real terms), all other things being equal, will increase the fiscal deficit by 0.3 percent of GDP due to the increased share of debt service payments. Similarly, a 1 percent increase in domestic inflation (as measured by the CPI) will translate itself into a 0.25 percentage point of GDP increase in the fiscal deficit. Based on State Budget numbers for FY 1999/2000, a ten percent depreciation in the Rupiah would increase the deficit by some 0.25 percent of GDP. On a cash flow basis, the effect of a depreciation looks better at first sight: because of the large inflows from foreign borrowing, and fairly small debt repayments, the net cash flow effect on the government's coffers is a positive 0.2 percent of GDP for every 10 percent depreciation of the Rupiah. Unfortunately, this positive effect is nullified, once the debt has to be repaid a similarly depreciated Rupiah levels. Also, for every US dollar drop in the oil price, the deficit would rise by 0.1 percent of GDP that year.
10
1.25 The fiscal risks from contingent liabilities remains high (Box 1 1) In Indonesia these stem from: (1) guarantees to protect depositors in and creditors to the banking system, (2) sovereign guarantees that may have been given on insured credits (through, say, OPIC in the US or MIGA) to private infrastructure providers, for instance, in the provision of toll roads, (3) obligations of minimum pension payments, and (4) a number of guarantee schemes to the private sector (some of which were instituted since the crisis to restore the flow of credit). Of these, the banking sector liabilities have already come due, with annual costs of about 7 percent of GDP over the next few years (assuming a 15 percent interest rate on recapitalization bonds), and there is some possibility that these costs may grow before 3 declining." In addition, future claims on public funds may be called upon due to implicit contingent liabilities with significant potential fiscal costs. In the case of Garuda Indonesia, the state-owned airline, the government has recently decided to take over the repayment of Garuda Indonesia's external debt totalling about US$1.8 billion for the lease of 11 Boeing 737s. This will imply a fiscal cost of US$62 million a year, on average, for the next eight years.14 Of particular concern can be the off-balance sheet obligations of public sector
13 As mentioned earlier in the chapter, this estimate is highly sensitive to the market interest rate on bank restructuring bonds at any point in time. 14 Source: Newspaper article in Antara, Jakarta, September 22, 1999.
Public Spending in a Time of Change corporations, which ultimately are the obligations of the Government.'s The main sources of fiscal risks to the Indonesian government may emerge from the following: Banking sector 1.26 These liabilities are no longer cost." (Vieira da Cunha and Brock 1997; see also Caprio and Klingebiel 1997). Power sector 1.28 As with the banking sector, the
guarantees has resulted in an involuntary cut back of fiscal resources that can be available for discretionary public expenditures since no such provision had been anticipated in the budget prior to the crisis. However, contingent liabilities still exist to the extent that the level of non-performimgloans turns out to be larger than current estimates suggest or if it grows (perhaps due to the emergence of negative interest rate spreads in the bnkingsystm). Te exent o the problem stems largely from the widespread structural problems in the financial sector tha peaetecis. Th.akswr thatprsedatethe sigicarisis. Te bacunk we exposed to significant risks on account of large borrowings in foreign currencies. The 1997 crisis severely aggravated any nonperfonning loans and led to a quantum jump in external debt liabilities, 1.27 Indonesia's experience with the huge banking sector liability mirrors similar .. . . problems in the other cnsis countries in the region. Though the depth of the Indonesian difficultieis greater than that in the other crisis countries, it is not unique. Where deposit insurance exists (and is extended, as in Indonesia, to all creditors) and the asset base is increasingly suspect, incentives to undertake further risky lending are high. A recent review concludes: "In just two or three years excessive risk taking and rapid credit creation can cause a banking crisis that can easily involve a direct fiscal cost of 8 percent eas * 1 of GDP and contingent debt liabilities of 4050 percent of GDP. Deadweight costs of taxation as well as costs associated with bankruptcy proceedings and the interruption of new lending can easily double the fiscal
15Whilethis is a serious problem in some countries, the World Bank's Public Expenditure Review in 1998 concluded that the off-balance sheet activity in public sector corporations is not high. 11
present financial stress in the power sector originates from actions undertaken before the crisis. These include concerns about excess power generation capacity creation and the unsolicited long term take-or-pay contracts that PLN signed with some of the 26 independent power producers (IPPs). 1.29 The crisis has aggravated PLN's debt burden. The Rupiah value of its US$ denominated debt obligations (which incs deb sesrvic t p own olents loans, gas purchases for ltS own plants, and purchases of power from IPPs) skyrocketed as Rupiah depreciated. The gas purchases, through Pertamina, are backed by standby letters of credit on state banks, and were of muhgetripc intal hnIP ic only afew sm allp on steamwe onl a few small plants were on stream when the crisis hit. Moreover, energy sales in Java-Bali (80 percent of PLN's market) are likely to remain flat. PLN's current weak financial situation implies that it will be unable to meet all its contractual obligations. ~~~As IPP's come on stream,' PLN's fiscal more . . . situation will come under increasing pressure. As such, PLN's ability to meet its obligations under the take-or-pay contracts could perhaps force it to defer payments on some of itS other continuing obligations. 1.30 The resulting implications of the crisis are serious. The IPP capacity crare serios The JUSt cacity contracted for by PLN IS iPP coming on stream. During the fiscal year 1999/2000, additional monthly payments of about $50-60
millionl (about 0.4 percent of GDP) come due. These payments will continue to increase for the next 4-5 years as successive
virtually none of the new capacity that comes on stream will be used in the next 4-5 years,
16 Malaysia, on the other hand insulated itself from this type of adverse effect from the crisis since it has a relatively more developed domestic capital market through which most of their IPPs were financed.
Public Spending In A Crisis
given the high level of contracted capacity relative to anticipated levels of demand, these new obligations will present a huge drain on the financial resources of PLN. 1.31 PLN and the Indonesian Government are also exposed to risks associated with legal action which, if they lose, could entail significant contingent claims on them. Roads 1.32 In Indonesia, the Government's obligations for toll road projects are apparently more modest than in the power sector. This is the case for two reasons. First, many of the roads carry intra-urban traffic, which is relatively well-established and steady. These roads, therefore, have not been seriously affected by the crisis. Second, the Government has few explicit guarantees, unlike in Malaysia where default on its loans by the project typically requires the Government to pay the outstanding debt and take over the project (see Box 1.2). 1.33 However, the full extent of the obligations may be larger than is apparent. Much of the financing of the toll roads came from domestic banks. The non-performance of the toll roads would show in the nonperforming portfolio of the banks, which in turn are backed by the Government. Thus, there may actually be more exposure to the Government from the toll road projects than is evident.17 Pension liabilities 1.34 For the public pension funds these are not a contingent but statutory liabilities of the government. But under a pay-as-you-go system, the Government carries a contingent liability to the extent that it has an obligation to meet some minimum pension payments. Though the immediate obligations in Indonesia appear manageable, the long-term trends are serious. Indonesia benefits, in
17 It should be noted that, to the extent this is already part of the overall bad debt portfolio of the troubled banks in Indonesia, it is not an additionalliability for the government.
part, because of a relatively modest-sized pension system. In 1995 (the most recent data that is available), just 12 million workers (less than a fifth of the workforce of 86 employees) were enrolled in pension plans (Leechor 1996). Of the 12 million workers, 8 million were members of the mandatory Jamsostek program and the rest were covered by a civil service program, referred to as Taspen. Though larger in terms of employees, Jamsostek had a much smaller financial asset base. 1.35 Pension enrollments, however, are now growing without a corresponding increase in their funding and asset base. Moreover, as the population grows older over the next 25 years, the issue will become more pressing. Enrollment in Jamsostek grew to almost 14 million employees in February 1999, faster than the growth of the labor force (which in 1998 was about 100 million employees). The Government's obligations vis-a-vis Jamsostek are not clear. The Ministry of Manpower is responsible for enforcing the mandatory enrollment but there are no direct financial obligations to top up pension payments. Those contributing to the plan do not always get the credit for their contributions and administrative expenses have been high and, as such, the rate of compliance is low. The consequence is that Jamsostek has virtually no return on its assets. 1.36 The Government's obligations to the Taspen are clearer, and though smaller in enrollment, the obligations are substantial. Taspen is partially funded from the salaries of civil service employees. Of the 8 percent of salary contributed, 3.25 percent goes towards partially financing lump-sum benefits and a defined-benefit program. The rest, 4.75 percent of the salary, pays 22.5 percent of the pension benefits. Since the Government is not permitted under the current law to make advance funding arrangements, the program is heavily underfunded and hence benefits are paid from the
general revenue. Taspen's assets, which have grown only modestly in recent years, are expected to decline as the benefit payments
12
grow. In Rupiah terms, assets have just about kept pace with inflation and in dollar terms,
they have actually gone down over the past 4 years.
Box 1.2: Contingentliabilities in Malaysia

The recent economic downturn has significantly increased the contingent liabilities and converted some into real liabilities for the Government of Malaysia. These are most important in the transport sector. Especially for the light rail projects, the Government World Bank estimates show that the expected costs of the contingent liabilities may be in the region of 60-80 of project costs (see discussion of expected and unexpected costs in the next section). Road projects have also suffered, but not to the same extent. Also, the vulnerability to further downtums has increased since profit margins have been cut and the probability of project default has, therefore, increased. Four reasons explain the severity of impact: First, in many cases, project revenue forecasts were optimistic. Under the due diligence process (and the absence of competitive bidding), the incentives to realistically estimate project revenues were absent. As such, the project economics appeared better than they really were. Second, government guarantees are more extensive than was perhaps necessary. In effect, the Government has an obligation to take over the project (and repay the outstanding debt) if the project defaults on debts due. This implies that the Government stands to become the owner of poorly performing projects. Third, guarantees provided on maintaining toll rates have come due in some instances as the raising of tolls has proved impractical in the current economic environment. Finally, the severe reduction in traffic volumes has sharply reduced cash inflows. In sum, the problems being faced by the transport sector in Malaysia are only partly due to the current economic crisis. The crisis impacted on a structure that had been rendered vulnerable by the limitations of the due diligence process and extensive guarantees that narrowed the range of risks borne by the private sector.
New off-budget obligations
addition,
guarantee
schemes
to
facilitate
1.37 The government is exposed to additional fiscal risks due to the sovereign guarantees that were issued after the crisis with a view to restoring the flow of credit into the economy and as a mechanism by which external creditors could eventually regain confidence in the ability of Indonesian banks and firms to meet their debt obligations. Guarantees, for example on inter-bank credit, have already come due in part. Further, the external borrowings rescheduled under the Frankfurt Agreement with external private creditors carry the
guarantee of the Indonesian Government.
credit to small producers and exporters are under consideration. While the Central Bank has limited exposure in the international derivatives markets, it does carry some credit risk by rediscounting of letters of credit received by exporters. While not explicitly a guarantee, the process exposes the Central Bank to the credit risks of the borrower. As such, these guarantees are an example of legitimate policy action to revive growth. However, they carry with them the downside risk that, as with past guarantees, some of them may be called.
DOMESTIC RESOURCE MOBILIZATION
Also available is the exchange rate guarantee scheme under the Indonesian Debt 5 Restructuring Authority (INDRA).' In
EFFORTS NEED BESTEPPED TO UP 1.38 The constrained fiscal situation that Indonesia finds itself in today (even before proper accounting of contingent liabilities is done in the budget policy, planning and evaluation process) reinforces the need to find ways of financing any fiscal deficit
through sustainable instruments-both on the
18The government has put in place a risk mitigation scheme to minimize the contingent liability it may be
faced with due to this foreign exchange guarantee.

Specifically, INDRA pays interest to the creditors in dollars at the rate of LIBOR plus margin up to 3% and (after at least 3 years) principal. Payments are made quarterly over the life of the restructured debt. The foreign exchange rate is fixed at the rate prevailing on the date the INDRA scheme was instituted with an adjustment provision in July 2000 that gives the 13
expenditure and revenue sides
of the
government the option to use the prevailing rate at that time if it lowers the cost of the guarantee.
Public Spending In A Crisis government budget. In the short run, a prime candidate for expenditure cuts is energy subsidies. These subsidies-which will amount to over 2.5 percent of GDP this year-are still badly targeted, and mainly enjoyed by the better off. To reduce the subsidies, Government must embark on a program to progressively increase energy prices to world market level-while protecting the poor with properly targeted safety net arrangements. Perhaps some minor cuts in the development budget could be carefully considered. But over time, as tax revenues increase, better budgetary processes are the key to better allocation of public money. 1.39 There is also still hope of further domestic resource mobilization possibilities that could finance public spending. For instance, the Government obtained some assets in exchange for the massive amount of bank restructuring bonds. Some of these assets are typically claims on banks that have already failed, and are likely to be worthless. But some assets have been obtained from settlements with former bank owners. With an economic recovery, these acquired assets could actually be sold at or close to book value. It is too early to tell what in the end the net costs of the banking crisis will be, but a start can be made by looking at the effect on the Government's balance sheet. Before the crisis, this balance sheet had-apart from Government owned enterprises and infrastructure-external debt to the tune of 23 percent of GDP, future revenues, and future spending obligations. The balance was the Government's networth. 1.40 After the crisis, the Government had a large amount of debt, less future revenues, more future spending obligations, and stateowned enterprises that probably lost some of their value in the course of the crisis. But the Government has new assets as well: the recovery potential of the assets transferred to IBRA in the course of Bank restructuring. A rough estimate puts this at Rp. 270 trillion, or almost half of the value of the restructuring bonds issued. However, there are several caveats which must be noted in this regard. 14 First, without further fiscal policy measures, the capitalized surplus of revenues over spending has turned into a large deficitevaluated at 10 percent interest this represents an asset value of over Rp. 500 trillion. Furthermore, the crisis has revealed some key vulnerabilities in the government's fiscal situation-contingent liabilities. The guarantees the Government has handed out before and during the crisis-such as the Central Bank's credit and exchange rate guarantees-will impose an additional cost to the Government, which could be significant. Although the Government is still assessing the exact size of its contingent liabilities, it is clear that the government's net worth will be impaired by them. 1.41 Unfortunately, arrears by the government on subsidy payments to the state-owned enterprises are being used as a cash management instrument. At times of fiscal difficulties, perhaps due to the slow disbursement of donor resources, arrears on subsidy payments have accumulated while an affected SOE has had to borrow from the state-owned banks to meet their operating expenses. In FY1998/1999, this was indeed the case with the its subsidy payments to Pertamina, PLN and Bulog. Over the past few years, arrears to these SOEs have accumulated to the tune of Rpl6 trillion. On the one hand, this is positive news since spending and the fiscal stimulus may have actually been larger than the budget shows. On the other hand, it means that the Government's debt is larger than the accounts show, if these arrears are to be paid off. The Government now plans to pay off these arrears in the current fiscal year. This intention of the government will significantly add to its already tight fiscal situation. 1.42 There are also additional off-budget items that cannot be ignored. The central bank has received indexed bonds from Government to replace the liquidity support the bank gave to commercial banks that faced a bank run at the height of the crisis. These bonds only pay 3 percent in interest, but the face value is indexed with inflation. To pay for this, the Government may need to issue
new bonds. In effect, therefore, Government finances interest payments with the issuance
of new bonds, but neither is show in the budget.
Box 1.3: ProblemsWith ContingentLiability Management the Czech Republic in A review of proceduresand institutionsin the CzechRepublicis illustrativeof the standardstypicallyin place for the management contingentliabilities. The two broad questionsthat require attention are: when should of guaranteesbe issued and how should they be containedwithin manageablelimits. Neither objective is adequatelydealt with in the current policyframeworkfor guarantees. * Absence of a frameworkfor the issutance contingent liabilities. The law authorizingguaranteescontains of no guidance on when guarantees are the appropriate public policy instrument or how they should be comparedto other formns state assistance. "This shortcominghas created an incentive for entities and of spendingministries unable to secure the desired level of state assistance through the budgetaryprocess to seek state assistanceoutside of the budget process throughloan guarantees. The additionalstate assistance providedvia guaranteesis perceivedas cost free and is not includedin the overall financialsupportprovided to a given sector." Also, "...guarantees are often extended as a result of emergency and political pressure rather than as a result of a strategicdecisionto supportoutcomes..." (WorldBank 1999,pp. 10-11). Guaranteesare not budgetedfor, nor are they subject to meaningfullimits. In the absence of systematic budgeting, one way to contain contingent liabilities is to place limits on their issuance. In the Czech Republic,the sum of expected paymentson guaranteedloans in a calendar year is required to be lower than 8 percent of the expected state budget revenues. This limit is too high. In Hungary, the limit is set at I percent of the face value of the guaranteedamount.
Despite these crticisms, it should be notedthat the Czech systemis probably ahead of systems in other countries in as much as proceduresare in place to documentthe contingentliabilitiesand even to categorizethem accordingto risk categories. The paymentsmade under guaranteescalledare also recorded. These are essential first stepstowards a more comprehensivesystem. However,the absence of a frameworkfor guaranteeissuance and the lack of controlshas meant an unrestrictedgrowth,with paymentson some guaranteescoming due and others underhigh risk of default.
This is the classic situation where "hidden 19 deficits" are created. 1.43 The Govemment's systematic under-
However, without domestic financing, and foreign financing driven by donors, strict deficit targeting-although worse for the economy-is the better option for
Government. More external financing may be needed as well. 1.44 Restoring primary surpluses will not happen without policy measures. On the revenue side, Government must raise the tax to GDP ratio by: * Reducing VAT and income tax
prediction of revenues is probably driven by the prohibition (until recently) of domestic financing of a fiscal deficit. True countercyclical policy would realistically predict revenues, and set spending targets in line with structural deficit targets. Disappointing growth would then result in a shortfall of revenues, and larger deficits which automatically stimulate the economy.2 0 19 See Kharas, Homi and Deepak Mishra (1999)
Hidden Deficits and Currency Crises, World Bank
exemptions. Under the old Government, several industries obtained exemption

from VAT and import duties, presumably to stimulate high-tech industrialization. Mo foreign entri ave
(mimeo), April 1999. Also, see Easterly William

(1999), When is Fiscal Adjustmenl an Illusion?,
Economic PolicyJournal,April 1999. 20 Althoughsome of the stimuluscouldbe off-setby increasedprivatesavings. However,during the crisis year (FY98/99)privatesavings droppedas well, and they are projected recoveronlyslightly 99/00. to in 15
Moreover,
foreign
enterprises
have
received exemptions, to stimulate design. To
generous income tax which, however will do little investments because of their raise revenues, Goverment
Public Spending In A Crisis could consider abolishing the exemptions on VAT for hi-tech industries, and replacing the income tax exemptions with investment credits. * Limiting special economic zones to truly bonded areas. While the usefulness of special economic zones can be debates, they should only be limited to truly bonded zones. The current proposals for expanding the special status of BATAM Island to the whole Barelang region is likely to cause much leakage in tax revenues, simply because the zone is too large to be effectively monitored by customs. Improving tax administration. Despite some structural changes over the last few years, including salary increases for tax administrators, the performance of the tax administration remains below par. The Govemment could spearhead its anti-corruption policies through the tax administration, thereby raising buoyancy of tax revenues at the same time. programs could include: subsidies for elementary schools in poor districts and to support school attendance by poor children; subsidies for basic curative care for the poor; and public provision of communicable disease control. As these are likely to be implemented during a period of fiscal pressure, special care will need to be taken when targeting beneficiaries and designing the specifics of these programs. 1.47 Over the medium-term, the authorities need to take appropriate measures to ensure that debt and deficits remain sustainable in the future, and create budgetary room for much needed civil service reformns. addition, as shown earlier In in this chapter, the Government's debt obligations are here to stay for quite a while. Therefore, Government needs to be prepared to manage this debt (domestic and external) more aggressively. On the organizational side, international experience suggests that this requires a dedicated unit in the Ministry of Finance where all the necessary information on the country's external and domestic public and publicly guaranteed debt is brought together, maintained and analyzed 2 on an ongoing basis. 1 Further in-depth analysis of Indonesia's debt burden and debt management practices has become a necessary andurgentimperatve. 1.48 Another urgent priority of the government should be to look more carefully budget items si its fisal oprtionsbudget items in its fiscal operations. Specifically, the management of contingent liabilities should be treated as part of an overall national risk management strategy. (Annex 2 outlines the key steps in this undertaking on the basis of international experiences). Ultimately, one's task should be to reduce both the frequency and the severity of contingencies that create stress on the national economy. An integrated risk
21 See Potter, Barry H., et al (1998), Indonesia: Fiscal Management, Decentralization and Organization of the
1.45 In the short run, the government must shift fiscal policy from fiscal stimulus to fiscal sustainability, by cutting wasteful spending such as energy subsidies, and by closing loopholes in the tax system and improving tax administration. At the same time, Government must embark on a budget refonn program that aims at a budget that includes all Govetment resources, accounts for fiscal risks, and allows for fiscal decisions that ensure fiscal sustainability Overthe medium-runs 1.46 Even in the constrained fiscal environment that the government will continue to face in the near future, the social safety net (SSN) expenditures must remain in some form or another. Contrary to initial intentions, the government has recognized that some of these SSN programs will need to be replaced by more permanent fixtures that support the poor without interfering with the private provision of social services. Such 16
Ministry of Finance, Fiscal Affairs Department, IMF, December 1998 Nvhichalso addressed this issue in the contextof an ongoingreorganization the Ministryof of FinarLce.
Public Spending inta Time of Change management process should perfonrn three principal functions: * Identify the risk exposures under the contingent liabilities implicitly or explicitly undertaken by the government; implement control mechanisms to prevent unintended losses on those risks; and establish systems to continually monitor and reassess how the government's risk exposure changes over time. Measure or quantify those exposures to budget against expected losses and to set aside reserves against unexpectedly high Undertake basic measures, including economic reforms, that reduce the government's exposure by fundamentally reducing risks and also by transferring risks to market participants.
Regulatory and process changes that should be initiated now, but will take longer to implement, in order to better track contingent liabilities include: * Non-budgelaiy control mechanismsfor contingent liabilities. These mechanisms would apply to liabilities grandfathered during a change in budgetary policy, or as a permanent management solution if the government the budget measures infornation failed to enact a change in law. Examples of such are: publishing of on government exposures,
establishing credit quotas (exposure limits), and earmarking of future funds to cover guarantee costs. * In parallel, the basis for measuring cost and "most likely" fiscal "aximutm likely" fiscal costs (also called "value at risk') of various guarantees could be established. This would permit the incorporation of such costs in the budget (either the face value or a proportion thereof) even before an accrual budgeting system is in place.
In the short run, the following measures could be undertaken in order to better monitor and manage fiscal risks:
+ Creating a single cell/department within the Ministry of Finance to take full stock of all guarantees issued, which would require recording the explicit or * budgeting and guarantees, assessing the teGovernment'sCreation of capacity to audit implied impiedguaaneesasessng monitoring of guarantees (in line with likelihood that the Government may be called on to make payments under these Imemfrme of thea over budget obligations, and the monitoring of these the time frame of the overall budget liabilities. processes) would create the appropriate checks within the system. * Creating a list of early warning indicators that signal the likelihood of realization of guarantees (e.g., in the banking system: rate of credit growth tothe private sector and exchange rate movements; in infrastructure: demand growth in relation with existing capacity). * At the same time the Government should begin identification of the contingent liabilities that it should phase out, either because they imply significant risks or because they no longer serve a significant social or economic purpose. 17
Improving the Budget Allocation Process
2.
IMPROVING THE BUDGET ALLOCATION PROCESS

he tighterfiscal situationin thepost crisisperiod hasserved as a wakeup call tofocus
ones thinking towards the task of allocating budgetary resources to better reflect national priorities in the new democratic environment. Budgetary processes must be strengthened to ensure that the scarce public funds are allocated efficiently to those areas that will have the maximum intended impact in terms of public service delivery. To this end, institutional reforms are needed to ensure adequate audit, evaluation and feedback mechanisms are in place so that the mix of spending allocations in the budget is deemed acceptable to a wider spectrum of the public. This will go a long way in minimizing the existing divergence between budget estimates and budget outcomes.
A NECESSARY RESPONSE IN A CONSTRAINED FISCAL ENVIRONMENT
2.1 The buildup of fiscal pressures and increased vulnerabilities due to fiscal risks since the crisis (that was discussed in the previous chapter) has made the assessment of budgetary processes in Indonesia a necessary imperative. Clearly, there are political dimensions to budget decision making. It is important, therefore, to support this political process with adequate information. The increased demands by the Indonesian public for more open and democratic government decision-making makes it an opportune time for the authorities to undertake budget reforms towards enhanced transparency in budgetary processes. The incentives of the stakeholders must be reflected in the way budgetary decisions are made, funds are appropriated and performance monitored. Intemational experience from both developed and developing countries has shown that enhanced transparency in budgetary accounting, planning and allocative decisionmaking will go a long way towards arriving at fiscal budgets that have maximum chances
institutions that are involved in the management of Indonesia's public finances may need change as a result. In undertaking such budgetary reforms, governments can improve their performance at three levels: (1) maintaining macro fiscal discipline; (2) strategic priority setting; and (3) efficient public service delivery.' Several donor agencies, including the World Bank, are currently assisting the government in this 2 regard. Figure 2.1 shows a schematic representation of the complex changes that are taking place in Indonesia today. It highlights the forces driving the need for changes to the institutional framework for fiscal management in Government. It also illustrates the nature of the linkages that any budgetary reform measures must grapple with towards attaining improved results in public service delivery.
of being successfullyimplementedwith the

necessary "buy-in" from the citizens of the country. 2.2
The crisis also provides an opportunity to for Indonesian policy-makers examineand

identify the structural weaknesses within its public sector and to undertakethe necessary budgetary reforms in an appropnate sequenced manner. The roles of the key 18
For details involving public sector performance at the three levels,see the World Bank's Public Expenditure Management Handbook (1998). 2 Besides this review of the public sector, the World Bank did another PER in 1998. In addition a World Bank Civil Service Review is ongoing. Among other donor agencies, the IMF has been working with the government on a number ofpublic sector related issues including budgetary management. The IMF currently has two resident advisors in the Ministry of Finance. The Asian Development Bank is also going to provide some technical assistancein this area later in early
2000.
Public Spending in a Time of Change Maintaining Macro Fiscal Discipline 2.2 Many countries-developed and developing-have found fiscal discipline an elusive goal. The benefits of an increased allocation to a particular ministry accrues to that ministry, while the costs-in tenns of higher taxes or deficits-are borne by society as a whole. Consequently, there is an incentive for each ministry to overstate its budgetary need. The result is an inflated budget, with its attendant problems of economic performance. One possible way to avoid this fiscal problem is to have a strong central government agency, which can enforce fiscal discipline in a transparent and
Figure 2.1-A changing environment comes with risks and opportunities

Globalization
Incre-d
coopl. ,msoour~ > Volatil
Democratization
> lcrrosd rol
Decentralization
IrOm.srd role of
) >
Lo rn> rl Loc I
Fiscal
Transparency
>
&
pressures
bl >gio Largo Debt D: Donor
Accountability
Red-ced cobcWlra...e for cpt oo
ofparhacooO Coni,bro,o
rnIenlc
cpa
-rity
>
Goor-nm-o Po_foro--o_c
Role of Indonesia's Government
Risks
.Political disintegration .Macrooeconomic instazbility .Inequaxlity
Opportulnities
.Better government services .HIf@hergrowth .Increased competitiveness
Institutional Change
Changing Processes > P.h.y

F-rm."..o > > Plannting Legislatiooc
Changing Roles & R,esponsibiites

pba-cog L>gCaabooc Cicil
Changing organizations
>
>
Changing incentives
Dfnilon of
MOF
Bapp-.n L-on Ministi> Local>cI:d >
Changing Legal Framework
R-mnd
accountable manner. Transparency and accountability in the budget process will, however, require appropriate institutions and public reporting requirements. Only then can one assess the true magnitude of the fiscal deficit at any point in time. Ideally, this would include proper accounting of the
19
sources of "hidden deficits" that are due to contingent liabilities of the government. These are not usually incorporated in fiscal policy debates unless a broader concept of deficit-the deficit of the Consolidated General Government-is considered. The ongoing endeavor of the government to adopt
the IMF Government Financial Statistics (GFS) system of classification of the budget items is an essential step towards influencing the incentives of the stakeholders to spend
the tax-payers' money appropriately. Complementary to this, the IMF Code of Good
Fiscal Practices on Fiscal Transparency
provides
good
framework
to
Box 2.1. Malaysia's Public Budget System

Introduced in 1989 to incorporate greater managerial flexibility and accountability, the Malaysian public expenditure management system performs on par with international best practice as exemplified in a few OECD countries. It has the following objectives and features: Main objectives: * * Promote a rational allocation of resources to government programs by imposing fiscal limits upon agencies and forging a link between inputs and outputs; Improve discipline and rationality into the budget process by explicitly quantifying a binding expenditure limit for each agency, shifting from line item budgeting to performance-based budgeting, and producing outputoriented program structures; Improve program management by adopting of better management practices, including encouraging greater delegation of authority from Treasury to agencies and then on to line managers; and Reorient the focus of accountability on issues of program efficiency and effectiveness, by measuring performance against predetermined targets, evaluating programs and activities according to their impact and relevance, and matching accountability and authority by holding managers accountable for performance.
* *
Main features: * The public budget is divided into two parts: the operating budget and the development budget. The Treasury, which is part of the Ministry of Finance, has the primary responsibility for preparing the operating budget. The development budget is prepared by the Economic and Planning Unit that reports to the Prime Minister's Office; A "Generalized Approach to Expenditure Control" is adopted which enables the Treasury to tighten the controls of overall expenditure and delegates the control over details to line ministries; An "Expenditure Target" is provided to each Ministrv/Department at the start of the budget process to which existing policy budget submissions must comply; A "Program Agreement" is signed between the Treasury and the line ministry which determines the level of performance that can be achieved for a budget year with the allocation approved; At the end of the fiscal year, an "Exception Report" is prepared by the line ministries regarding the performance of activities which do not reach the levels specified in the program agreement; and "Program Evaluation" takes place for each activity at least once every five years.
* * * * *
PublicExpenditure Revieiv,1999. Source: World Bank,Malaysia
make an assessment of the existing practices and desirable budget targets (see Box 2.1) for an example of a good practice in public expenditure management. 2.3 Enhanced transparency in the reporting of the government's financial position also requires an appropriate integrated govemment accounting system. This is not in place in Indonesia at this juncture. The govemment accounting system consists of a number of subsystems which are
not integrated. Each agency has its own separate system, and no consolidated report can be produced to show the current 3 Govemment-wide financial position . The present system is also largely manual and does not capture accounting for nonbudgetary transactions. These are accounted for in the DG Budget accounting system. SOEs' finances and local and
does produce, two years after the end of the fiscal year, a consolidated government financial statement for submission to DPR.
3BAKUN
20
Public Spending in a Time of Change provisional government finances are also not captured by the Government accounting system. In addition, there are no uniform accounting standards used in the public sector. The Government plans to issue them in the current fiscal year. There are inefficiencies in the present system that inhibit the effective monitoring of government financial transactions. An appropriately devised comprehensive and integrated accounting and budgeting system 4 can address this current shortcoming. 2.4 Government auditing practices in Indonesia need to be revamped as well, in order to strengthen the overall accountability framework and enforcement mechanisms for compliance with Government regulations and policy. In addition, there is currently a lack of clarity in the mandate, scope, independence, and roles of BEPEKA-the Supreme Audit Board, BAKUN-the Government Accounting Agency (which has the responsibility of recording and preparing the Government's financial reports), BPKP-the Government's internal audit agency, and the Inspectorate Generals (IGs)-the internal auditors of their respective ministries (which have a narrower mandate than does BPKP). The work of IGs at the present time is not well coordinated and is not complementary to the work of BPKP. There are BPKP offices nationwide with the exception of
three provinces.
findings. The enforcement problem, however, extends well beyond the auditor's responsibility. It stems in large part due to weak enforcement and ineffectiveness of the existing sanctions for violations and irregularities. In this context, it is important that auditors use appropriate judgment in deciding when to refer delinquency cases to the courts, instead of reporting them to the police for action. But monitoring and followup of audit recommendations should clearly be the responsibility of the auditors. 2.6 To be consistent with international best practice, one must ensure that BEPEKA's role as the independent audit agency of the Government is maintained. Furthermore, the oversight role of the Parliament over the work of BEPEKA and its audit findings, and implementation of BEPEKA's recommendations, should be properly defined and exercised. Furthermore, any disciplinary measures and sanctions for non-compliance and violations need to be clearly defined, communicated and enforced. A comprehensive review of the internal audit framework, including rationalization of the roles and responsibilities of BPKP, and the IGs, should be conducted. The ADB is currently supporting the efforts by the authorities to clarify and strengthen roles and responsibilities of BPKP.
Strategic Priority Setting
2.5 Currently, the enforcement power and follow-up mechanisms of both BPKP and BEPEKA are poor. For instance, BEPEKA cannot adequately certify government financial accounts since it does not audit tax revenues. Furthermore, due to legal impediments it cannot audit state banks. Audit findings are not made available to the public and there is no systematic follow-up mechanism to ensure implementation of audit
World Bank is assisting BAKUN in developing the necessary computerized accounting system. This project is expected to provide the authorities with the
4The
2.7 Although strategic priority setting in the budget process has political dimensions, this process must be supported by better information in order to enable policy-makers to make better decisions. Such information includes: the reasons for public sector involvement and the appropriate instrument, competing claims on the budget, linking inputs with outputs, and findings of user surveys. Beyond the core public sector
activities (e.g., "pure" public goods like
abilityto produceGovernment financialreportsin the currentfiscal year. However,the issuanceof uniform accountingstandardsfor the publicsectorentitiesas a
whole remains to be accomplished.
national defense, street lighting), the design and implementation of public expenditure priorities requires detailed assessment and careful tailoring to country circumstances. To
economic 21
this end, appropriate determination of the (recurrent vs. capital) and
Improving the Budget Allocation Process functional allocations (among and within sectors) of public expenditures is crucial. The use of IMF-GFS classification standards will facilitate this exercise. 2.8 In its current form, the Indonesian expenditure classification does not properly account for expenditure policies by economic purpose. For example, the development budget includes several items-honorarium to public officials, travel and other allowances-that are recurrent in nature and should ideally be reported under routine expenditure. The existing reporting system in Indonesia partly responds to donor preferences (i.e., donors do not like to finance items in the recurrent budget) but it ends up introducing non-transparency in the compensation of public officials. It also has the perverse incentive among civil servants to "invent" projects (irrespective of need) for possible donor financing, which in turn, increases their honoraria. 2.9 A first step to reforms is to make current civil service remuneration more transparent. If the Government adopts the IMF-GFS classification, this will be automatic. In fact, most information on salary supplements from the budgets is already available in the accounting system. The budget simply needs to reflect this. As important as transparency is to de-link salaries from development projects, this change in the expenditure classification system is a desirable objective in its own right, since it will allow policy-makers to determine the appropriate sector and/or project specific recurrent-capital expenditure mix on the basis of international comparisons. In turn, however, these changes will also have important implications for donor financing. If adopted, the suggested salary reforms would reveal a deficit on the recurrent budget. This deficit already exists, but it is hidden in development budget-paid for by donor money. Thus a go-ahead on the salary reform requires that donors feel comfortable with the fact that, in reality, they finance part of civil servant's salary. This is less of an issue for program financingwhich is general budget support-but could be balked at by donors that thought they were financing projects. 2.10 In Indonesia, the future cost implications of new policies are insufficiently taken into account. This is in part because of the weak link between routine and development expenditures decision making. So, in the short run, this link could be strengthened by ensuring that the MoF is also involved in reviewing the current and future costs of projects when they are first submitted by the line ministries to BAPPENAS. (The IMF has also made this suggestion in its October 1998 report on 5 budget management in Indonesia. ) The ideal way of properly accounting for the future cost implications of current policies is to introduce a multi-year rolling budgetary planning exercise-something that Indonesia should implement over the medium-term (i.e. in the next 2-3 years). This approach unifies the capital and current accounts into a single budget, and then projects this budget out for three to five years, so that the recurrent costs can be evaluated alongside the country's projected revenues. Many developing countries (including Ghana and South Africa) and developed countries (Australia and New Zealand) have successfully adopted this process. 2.11 Budgets in Indonesia are not predictable at the level of the spending entity in government. The adoption at all levels by the government of a medium-term planning and budgeting process would go long way in making budgets more predictable (See Box 2.2). This will make the eventual transition from the current cash management system to an accrual budgeting and accounting system all the more easier.
5 See Indonesia: Budget Management in the Short- and Medium-Term. Fiscal Affairs Department, International Monetary Fund, October 1998.
22
Box 2.2. Medium-Term Planning and Budgeting The medium-term planning and budgetary process consists of three steps: * Macroeconomic and fiscal analysis. To emphasize macro fiscal sustainability, the budget is forrnulated within a three- to five-year horizon and includes macroeconomic constraints, the costs (current and capital) of ongoing policies, and room for new policies. Every year the budget is rolled over the fixed horizon. Fiscal policy formulation. Within the aggregate fiscal constraint, the executive branch of the government decides on new policies and sectoral spending limits, and discusses those with parliament to ensure a general overall agreement. In this phase, the government decides on additional revenue measures, and on which policies to drop in order to create room for new policies. It also detennines the amount of general grants to devolve to local government, and local borrowing limits consistent with macro-economic stability. Detailed budget preparation. Within the broad sectoral limits and the policy decisions made, individual line ministries prepare their budgets. The line mninistry budgets, which cover both capital and recurrent costs, focus on programs rather than projects. The budget proposals are accompanied by result statements detailing what has been achieved in the past budget year(s), and what will be achieved in the coming year. The budget policy agency no longer needs to negotiate down bloated demands for funding, as line ministers need to stay within the broadly agreed limits. Instead, it focuses on the financial and economic analysis underlying the budget proposals.
Process for Evaluating Public Expenditure

Programs 2.12 Budgetary decision-making in primarily on Indonesia has focussed govemment investment (i.e., "the development budget") and less on decisions regarding the public policy role the government wants to play in any area (which manifests itself in public investment, recurrent expenditure, transfers and regulation). Once public finances are recorded and reported transparently as per international standards and conventions, the traegc bdgear proessof akng process of making strategic budgetary decisions will clearly become easier for
policy-makers. The public may also be more willing to accept the strategic priorities the resulting budgetary allocations indicate if the budgetary process was more open. 2.13 The authorties must continuously assess whether the scarce public funds, and charged with, are being put to the "best" use. Cross country experience suggests that such an evaluation of resource allocation within (and across) sectors proceeds by asking three questions: (1) What is the rationale for public intervention? (2) What iS the right rigt intrvent? (2) What iscth instrument? (3) What are the fiscal costs? (eFiue22) (see Figure 2.2).
23
Public Tree Evaluating Programs Figure Decisionfor 2.2.
No rationale
isa There rationale

inpnmary (e.g., externality education)
Outsourcing
Subsidy
Regulation
provision Public
What is the rationale for public a. intervention? 2.14 In Indonesia, a Constitutional provision mandates the govemment to ". assure the supply of essential goods and services and to promote the growth of the economy." Interpreting this provision in its favor the government chose to be involved in any activity that it liked. While in recent years there has been a recognition that the private sector needs to play a more important role in managing the economy, the govemment's budget system still does not incorporate any feature which requires rationalization of the public sector involvement in the proposed activity. 24
2.15 In the current budget system, line ministries submit their recurrent budget proposals to the MoF and proposals of development budget to BAPPENAS. For the line proposals, budget development ministries are required to support their project proposals with statements of objectives and likely benefits, but there is no requirement for either justification for public sector intervention or a systematic costbenefit analysis of proposals. Perhaps an implicit justification for public intervention is that most of these projects are already included in REPELITA. However, new proposals can also be added. Routinely line ministries submit inflated budget proposals knowing in advance that the budget
Public Spending in a Time of Change market for education. The main b ary of education is beneficsary of a uim versity educataon S the student himself. If as a non-poor he is unable to afford the high cost of tertiary education, it could reflect another market failure in the economy, namely, a failure of the credit market. In such a case, there is no market mechanism for a high costcof tenti educatin hGgh cost of tertiary educateon. Govedrment intervention to alleviate the ...credit problem iS perfectly Justified (as iS the case in many OECD countries including the United States). Some the second justifyi their But in most
the university
allocation process is a negotiated game between them and the two central agenciesBAPPENAS and the MoF. 2.16 The govemment appreciates the principle of rationalization of public sector intervention and has prepared a master-plan for a major privatization program that is being implemented. It is also recommended that the rationalization of public sector intervention should be required for all new programs programs prpoe proposed in th bde. in the budget. Failing Failing adequate justification, the proposed program
should not be funded.' This kind of a
governments discipline would help wuld government to discipine the hep thegovemmnt torationale-redistribution-toresort to subsidies to universities. undertake much needed changes in the
composition of its spending. For example, countries almost all
In Malaysia, research showed that an increase in the number of public doctors in a district had no effect on infant mortality-because they were crowding 2.17 Ideally, each program in the budget, out the private doctors-whereas including both its associated current and increases in safe water and sanitation capital expenditures, should be scrutinized (involving large externalities) had a for whether there is a rationale for substantial effect (see Hammer et al, government intervention in that area. Market 1995). failure (externalities and/or public goods) and * The largest item in the Indian concern for social equity (including equal opportunity of access for basic goods and govelrnens ilture bde is a government's agriculture budget is a services) provide the two possible rationales sergovices)mprovden nthertwov possietion awes, subsidy. When the fertilizer was for government intervention. However, afertilizer ainrdcdftyeasgothewsa large number of expenditures, which appear rationale in terms of the information in a majority of government budgets, fail this externalities associated with farmers' adopting the new technology. By now, test. Some examples of these expenditures that externality has vanished, but the and where they have appeared are: fertilizer subsidy lives on. Moreover, the beneficiaries of the subsidy are mostly * In education, the rationale for public better-off farmers. As documented in the provision of higher education is difficult to justify on grounds of failure in the World Bank's 1998 PER, the case of fertilizer subsidy in Indonesia appears to be no different from the Indian case.
*
At times public sector intervention could be justified for other reasons such as political stability or threat of violence. The current electricity and petroleum subsidies in Indonesia are a case in point. It is
6
short- to medium-term policy changes would involve things like spending more on basic health, primary education (to improve quality) and less on direct (untargeted) subsidies. Preventive health care and public health services (vaccinations and vector controls), which could be classified as pure public goods, have consistently received a smal poto bugt
In subsidies accrue to the rich. Indonesia, research has shown that a very high percentages of the public subsidy to the universities has been enjoyed by those in the top decile of the population van de Walle 1998 (
importantto ensure that such subsidies do not get

entrenched and are phased out in a timely fashion.
Some government programs can only be justified in terms of the redistribution rationale. Most transfer schemes (cash and in-kind) fall under this category. The
25
Improving the Budget Allocation Process Box 2.3: The impact of education spending in recent years The crisis in Indonesia has resulted in a cutting back on education spending by 12 percent during the 1997/98fiscal year and afurther 30 percent in the nextyear (See Figure 1). Both development and routine budgets have been cut (in real terms). Development spending fell by 11 percent in 1997/98 followed by an additional 24 percent in 1998/99. Falling development expenditures are partially due to the effect of inflation coupled with the deferral of several activities. The routine budget, which predominantly finances salaries, was also reduced by 13 percent in the first year and followed by a further 30 percent last year. Figurel:The education budget(in real terms)has fallen....
40X
10% :i_ /, _
lE '0019 E 871907!88100&80I9000S00IE9I99/
~~~~~~~~~~~~~~~~~~~~.,""'""-, .. ... .....

.05 1995M910 71971001
21992(931031419
9\9
-20%.
40.,,
-----
--
- -
9-
914
Figure2:...whilerealizededucation spending remains belowbudgetedamounts.... Protecting Basic Education Outcomes during the Crisis. Basic Education services continued to be provided at a satisfactory level, despite declines in budget. Enrollments have been maintained and all teachers have remained employed, although with some eroding impact on their salaries, as has been felt by other civil servants. The falling expenditures are due, partially to the effect of inflation on salaries and partly on declines in development budget expenditures. Some spending cuts are justified (e.g. halting new construction, and overseas training) but some cuts will result in declines in the quality of service delivery (e.g. teacher training). A Crisis Program - Back to SchooL At the beginning of the crisis, GOI estimated that the dropout rate would increase substantially at the primary and junior secondary level from 3 to 5.7 percent and 3.6 to 11.5 percent or 890,000 and 640,000 children would drop out in just one year (Education Sector Strategy). To deter such a high incidence of dropout, the Back-to-School program was launched in early 1998. As shown in the table below, the result has been a slight increase or in some cases the maintenance of Gross and Net Enrollment Rates at all levels. Table1. Schoolenrollment notdeclined has afterthe crisis-Thanks to the "Backto School" Program YEAR 1996/97 1997/98 1998/99 GROSS Primary 112.36 113.50 114.71 JuniorSecondary 69.46 69.88 68.73 SeniorSecondary 37.08 38.02 37.37 NET Primary 94.87 94.06 95.0 JuniorSecondary 51.36 54.43 53.17 SeniorSecondary 28.88 29.99 29.38 (Source:Ministryof Education, 1999) main issue here is whether such programs reach their intended beneficiaries-the poor. 26 In the Indian state of Andhra Pradesh, the rice subsidy scheme-ostensibly targeted for
Public Spending in a Time of Change the poor and needy-reaches more than 65 percent of the population and consumes a sizeable proportion of the state's annual budget, leaving fewer resources for other desirable activities such as basic health and road maintenance. In Jordan, a lower middleincome country, the recently-abandoned cash transfer program reached 76 percent of the population. That it consumed 1.7 percent of GDP made it an excellent candidate for public-sector rationalization according to these criteria. b. What is the right instrument? 2.18 Market failure and distributional consideration may provide the rationale for public sector involvement in an area, but it does not necessarily imply that the public sector should be the service provider in that area. If the rationale is an externality (such as with primary education, for example), the same objective could be achieved by a public subsidy, with the private or nongovernmental sector providing the service. This is the reasoning behind various schemes, such as school vouchers, that seek to improve outcomes in sectors where there to imroveoutcoes were here i secors is a justification for government intervention. In the past, monopoly conditions in infrastructure services usually justified public provision. Over the years, technological advancements have gradually reduced the need for a single supplier and conditions have been created for market competition. Thus, in many infrastructure services, the right instrument for government intervention has become enforcement of efficient and impartial regulation. 2.19 Analyzing the instruments of public sector intervention in Indonesia reveals the following: * The government has designed an emergency assistance programsupported, among others, by the Bankto provide block grants to primary and junior secondary schools in poor areas (as well as scholarships to needy students) to contain school dropouts resulting from lost incomes. (See Box 27 2.3). This is an example of the right kind of public sector intervention. As emphasized in the Bank's 1998 PER, a number o tin and atie of anme foeain n those relating ciiiso PERTAMINA (particularly to production) could be privatized and the agency itself could play the role of an impartial regulator. All subsidies-electricity, fertilizers, food and petroleum-could be gradually phased out and replaced, as appropriate, saporae paetargeted programs to protect the poor. u n elcd by At the same time it is important to plan a communication strategy (with the people) to enhance the chances of success of such a reform program.
*
The government has plans to expand facil at vocational educat facilities at the tertiary level. Given that the benefits of such programs accrue mostly to the better-off students, the government should charge the full costs of the program from the beneficiaries.
c. What are the fiscal costs? 2.20 Amongpublic programs where there is a rationale for those intervention, and where public provision is the appropriate iument, g the r te o de instrument, the governiment needs to decide whichhave thehlighestpriority. Ths again is essentially a political decision, and is typically undertaken at the levels of the cabinet and parliament. Decision-makers will need timely information to evaluate the unavoidable tradeoffs. Although it is often difficult, if not impossible, to quantify the benefits, the fiscal costs (both current and future) can be quantified. The latter information can be used by policy-makers to make judgements on program expansion vs. contraction vs. elimination (or introduction). 2.21 An example of a publicly supported rural credit project in Morocco highlights the The two relevance of this issue. distinguishing features of the rural credit project in Morocco was that: (a) it was justified on grounds of credit-market failure, and (b) it benefited poor rural farmers. But
Improving the Budget Allocation Process the fiscal costs of the project represented 20 percent of the country's educational budget, and 160 percent of its expenditure on preventive health. This information on opportunity cost could be quite useful to decision-makers when evaluating the difficult tradeoffs. Moreover, such information along with the decisions made, be communicated to the citizens in a transparent manner. To apply this instrument in Indonesia in the short-run, decision-makers could look at the subsidy budget for electricity and petroleum, for instance, and deliberate if part of it could be channeled to improved the quality of primary education. The medium-tern planning and budgeting exercise will permit policy-makers to see the total fiscal costs of different programs, permitting a more systematic comparison of competing uses for the public's resources. Efficient Public Service Delivery 2.22 The key to achieving efficiency in the delivery of public services is to have the right incentives in the institutional structures, Public sector agents often face little or Perverse incentives for efficiency. Managers in the public sector are rewarded in terms of the size of their budgets, giving incentives for cost-inefficiency. Non-transparency and inadequate accountability has often contributed to this problem. 2.23 Indonesia's public health system is a case in point where the institutional and budgetary process bottlenecks are giving rise to inefficiencies in service delivery. The public health network has over 7,100 publicly run, mostly rural, health centers and associated sub-clinics and outreach activities It has referral linkages to around 300 government-owned and operated district hospitals with a standardized program for 18 different service areas. Staff and administrative expenditures absorb a significant share of the sectoral budget. An inflexible planning and budgeting system and very limited opportunity to revise budgeted allocations during the fiscal year, along with a proliferation of time consuming but ineffective accountability mechanisms, 28 inhibit proper spending towards efficient public service delivery. 2.24 The existing practiees for public expenditure management in Indonesia do acknowledge the importance of evaluating the perforrnance of public sector service delivery. However, if the performance indicators that are monitored are inappropriate to the objectives that are to be achieved in terms of public service delivery, the policy choices that will emerge from the evaluation will be inappropriate. In this regard the authorities, having realized that past performance evaluation efforts for development projects had been limited to monitoring actual implementation, issued a decree in 1996 to introduce a proper performance evaluation system based on specific indicators. According to this decree (KEP.195/KET/12/1996) each ministry/institution/region shall conduct performance evaluation on its development projects and provide a report to BAPPENAS. While there is little evidence to support whether the system has worked effectively or not, it does not seem to provide any link between a ministry's past performance and future budgetary allocation. When submitting their budget proposals, line ministries (or any other agencies) are neither required nor they have any incentive to report their achievements-that link spending with outputs and outcomes-in the preceding year(s). In turn, budgets are allocated based on inputs and are essentially adjusted incrementally from the previous year. In the absence of any clear incentives and proper accountability, it is not surprising that most public sector studies in Indonesia have documented major inefficiency in public service delivery. 2.25 In addition, the structure of the compensation theseructure of is compensation of civil servants incIndonesia is such that it has created incentives of a perverse kind. Civil service wages in general are quite low. The base pay is calculated as a minimum payment simply for reportmig to work. Supplements or premium for specific assignments are given based on work undertaken, but there is no link between
Public Spending in a Time of Change performance and reward. Over time a system of informal salary supplements (some of which are legal) has evolved. It involves income generated from charges to clients and userof services. I isallgedfrom ovemen It is alleged user of government sevice. that informal payments are built into most government contracts and the money collected is distributed according to a wellGiven these established scheme. circumstances, it is very difficult to imagine how a performance-based reward can be introduced and enforced into the Indonesian civil service unless basic civil service reforms are carried out. Such a task will also involve significant public resources that must be allocated in the fiscal budget, subject to availability of resources.7 2.26 Recognizing these problems, several countries have attempted to reform the incentives in their public sectors. Some of the inctuentives inusheirpubincludectors. Sohave instruments used include: 0 Performance-based budgeting. Rather than evaluating public servants on the size of their budgets, or making their budgets a function of their costs, public managers in New Zealand are given budgets based on a specified set of outputs for which they are held accountable. The manager of a school district, for instance, is given a budget and expected to have a certain number of schooling students complete successfully. Under the current circumstances it is premature to implement any kind of performance based budgeting in Indonesia, although it could be a long term goal that one may want to achieve. In the short run, Indonesia needs to have a more transparent and comprehensive budget as well as improved financial management practices. Enhancing the capability of the civil service to operate under "the new rules of the game" in a more open and democratic environment is more urgent before attempting to launch into performance based budgeting and management in Indonesia. Client surveys. Eliciting information f clientsr has proved tobean clients of proved efficiency important way has improvingto be an and fighting corruption. A field survey of randomly selected 250 government schools in Uganda showed that on average, less than 30 percent of the funds intended for non-salary public spending actually reached schools during 1991-95. In Bangalore, India, the announcement that client surveys were about to be conducted led to an improvement in service delivery. The survey response led to a restructuring of the worst graded public agency in the city and helped create a culture of customer orientation. In several Eastern European countries (Albania, Georgia corruption surveys and Latvia) detailed conducted-among been households, enterprise managers as well as public officials-that have helped policy-makers to design appropriate action plans to improve efficiency and provide better public services. Client surveys are also currently being used in Indonesia monitor the government's core social safety net spending. With support from the World Bank's Social Safety Net Adjustment Loan (SSNAL), the government has formed a team with civil society participation to implement and oversee its key safety net programs. The process includes a publicity and information campaign to reach grass roots level with information about the program, its budget, beneficiaries and what they should receive, decision-making structure, and the channel for complaints. Similar initiatives can be taken elsewhere. For example, a user survey in a randomly selected sample of the 7,200 public health centers would provide useful information to policy-makers regarding the efficacy of spending in that activity. User surveys could also be carried out in public schools or for infrastructure services (water and
electricity). 29
The ongoing Civil Service Review by the World Bank
is examiningthe pay policy of the IndonesianCivil Serviceand the incentivestructureit emanates.
Improving the Budget Allocation Process * Retention of user fees. In addition to the fact that they become a signal of demand and improve resource allocation, user fees for public services-when they can be retained by the service provider-can be an incentive for improved service delivery. Sometimes, the effect can be so dramatic that it compensates for the possibly regressive effect of user fees. However, it is important to ensure that audited accounts of the use of such funds are maintained and are available for inspection by the relevant officials and to those who make the payments. In Indonesia, a number of user fees and charges are currently imposed by government agencies and retained for their own use. This is appropriate from the point of view of giving the service providers an incentive to improve public However, the service delivery. accounting of these funds has not been as transparent as international best practice would suggest. Being off-budget items, they do not go through the regular scrutiny. Even before the crisis, more than 40 percent of the expenditures on education were from off-budget sources, including a variety of user fees and charges. There are also a number of user-fees in the health sector. Given that most of these accounts have not been audited, it is difficult for one to say whether the positive influence of increasing efficiency by retaining user fees outweighs the administrative burden on government of collecting them.
RESHAPING THE ROLES OF KEY BUDGETARY INSTITUTIONS
decentralized form of governance-needs three key framework documents: (a) a statement of its long-term development agenda (in the form of, say, a national development plan); (b) some form of a medium-term budgeting framework to implement that vision (See Box 2.2); and (c) a set of sectoral strategies that assign the responsibility for sector-specific policy development and implementation. 2.28 In Indonesia, the role of BAPPENAS has been crucial in the budget process. It is the key agency for development planning, development analysis, macroeconomic budget compilation, regional development, program evaluation, and donor coordination. BAPPENAS is also involved in development budget implementation and supervision. Through its Deputies (senior technical staff) for Economic Affairs, Infrastructure Human Resources, and Regional and Local Affairs, BAPPENAS offers unique cross-cutting expertise in Government. BAPPENAS has been recognized for its valuable expertise in project preparation and evaluation based in and for Finance Department the Implementation Management. Its most important and visible documents are the Pembangunan Jangka Panjang (PIP) with a 25-year time horizon; the Repelita with a five year horizon; and the annual State Development Budget (APBN). The other key central ministry-the Ministry of Finance has been involved with the formulation of the routine budget, execution of the related transactions (in terms of making the payments) and administration of the tax system. 2.29 Changing budgetary processes in Indonesia will require rethinking the evolving roles and functions of BAPPENAS, Ministry of Finance and the line ministries. How this is undertaken in Indonesia depends largely on tradition, culture, and the Constitution, with outside advisors providing only limited insights. However, what is clear from international experience is that any significant changes in government organizational structures stands the best 30
2.27 Any envisaged reforms in budgetary processes to foster transparency, accountability and effective public service delivery will need strong institutions. This becomes all the more critical when the budgetary reforms are being accompanied by a move towards a more fiscally and administratively decentralized form of government. Cross-country experience has shown that a central government-even in a
Public Spending in a Time of Change chance of succeeding if it is initiated at the beginning of a government's term in office. 2.30 One possible reform process involves long-term strategic planning, participatory budget policy formulation and corresponding planning, budgeting and budget approval procedures. It adapts Indonesia's current processes to the forces of change, and aims for a more transparent, accountable, performance oriented government. It is inspired by OECD experience in the last 25 years (Annex 3), but some elements of the vision have already been introduced in Indonesia, while others are being debated at this moment. this 2.31 Undr pproach,program implementation becomes the responsibility of line ministries, with service delivery being increasingly left to lower levels of agnis,nn goemet spcaie government, specialized agencies, nongovernment organizations, and the private sector. Increased accountability of line ministers for observable and monitorable indicators should provide the incentives that foster evaluation capacity within line ministries. Civil society would play a larger role in evaluation, empowered by increased access to Government documents, and by decentralization which reduces the distance Increasingly, policyto policy-makers. makers would use civil society to improve their services and achieve better results with government spending by conducting user surveys to solicit feedback from the public. Within this approach, BAPPENAS could be: (i) a strategic policy think tank; (ii) a budget policy formulating agency; or (iii) a focused development policy formulating agency. Whatever its role, the nature of BAPPENAS' key outputs will have to change as well. 2.32 Some OECD countries have taken the .. opportunity to redefine their more traditional budget~~~~ maaemn law in a fiscal budget management la.nafsa accountability framework, spelling out roles and responsibility of each agency involved in the budget process. The Public Finance Act the Zealand, of New (1989) the (1986) of Comptabiliteitswet Netherlands, and the Financial Management 31 and Accountability Act (1997) in Australia are such legal instruments (see Table 2.1). In the Indonesian context, such Government reorganization efforts, and the redefinition of roles, responsibilities, and accountabilities would entail a detailed review of the existing budgetary laws and regulations (such as the Presidential Decree No. 44, 1993 and the "Indische Comptabiliteitswet"), (Treasury Law, Staatsblad of 1925, Number 448). The latter is currently under review by a drafting committee headed by MoF.
ALLOCATING DEMOCRACY THE BUDGET IN A
2.33 A critical requirement of fiscal transparency in the context of a democracy iS .cii and fortegture thenopprtnint did tonassess w egoverent sce budgent On asai siwo dther waty it what it said it would do in the budget. One measure of these budget plans that is consistent with budget implementation is the availability of independently audited actual expenditure data. The integrity of the data being reported is also the foundation on which the credibility of the budget rests within the context of macro-economic stability. In general, the link between 2.34 transparency and participation is not an automatic one: they often are prerequisites for the other, yet neither is sufficient in isolation to reap the benefits of better availabilety of information on tough choices c e o ntough oftinfora avaiit and tight constraints a government may be facing in budgeting may give citizens a better the context in which the sense of goverment decisions are being made and build their faith in government. Further,
~meaningfulinput into the debates, and see
when they are given an opportunity to make eventual the rlte teincerns their concerns reflected in the eventual decisions, the reform measures will have the "buy-in" necessary to make them start out on the right footing in the country towards greater chances of success. If, on the other hand, they are denied this opportunity, any faith that may have been
Improving the Budget Allocation Process Table 2.1. An Illustrative Performance Budgeting Framework Process Contents Reports Responsibility
Long term strategic planning Policy formulation Long term view on development Coalition building; Sector strategies, containing goals, instruments and program proposals Detailed analysis of programs and policies, including costbenefit analysis, budget projections, output and outcome indicators Outlook on the economy and the budget under unchanged policies Medium-Tern Policy priorities, new policies, phased out policies, Budget envelopes; general grants to local govermment;changes in tax policy Detailed preparation of sectoral and line ministry budgets within the parameters set in the fiscal policy outlook by Debate and approval parliament 25 year/5 year development outlook Coalition agreement; Discussion papers, white papers, green papers Line Ministries' Corporate plans Strategic planning agency with involvement of universities, civil society, line mninistries Line ministries, cabinet, with discussion in parliament and with civil society Line ministries, under guidance of the planning agency and the budget policy agency
Frequency
Every 3-5 years
Policy planning
After every election a coalition agreement; Ongoing sectoral policy formulation Annually in preparation for the budget
Macroeconomic and fiscal analysis Fiscal Policy Formulation
Macroeconomic and Fiscal Outlook Fiscal Policy Statement
Budget policy agency, endorsed by cabinet, presented to parliament, widely published Budget policy agency; Forward estimates from budget manager, or line ministries
Twice yearly, one fundamental review, one update mid-year Annually, with 3 budget years included
Budget preparation
Draft Budgetincluding functional, program, economic, line rninistry

presentation;
Line ministries, coached by the budget policy agency. Budget policy agency compiles budget documents; President submits Parliament. Budget policy agency presents overall budget; line ministries budget heads. Parliament is supported by budget bureau. Parliament can change budget within agreed
rules
Annually
Budget Approval
Annual budget law, with chapters for each line ministry
Annually. Semiannual revisions in budget will become rare.
Budget implementation
Raising revenue, managing cash, debt, spending money, implementing programs
Intemal and external reports on financial and program implementation
Line ministries for results; Treasury for financial implementation Local Govemment, NGOs, and Private sector for service delivery Line ministries, with participation of civil society through user surveys, hearings, involvement of academia etc. Audit Agency for extemal evaluation
Policy evaluation
Intemal and external evaluation of policies
Intemal: management reports to line ministries Extemal: Audit reports, 'value for money analysis"
Daily, weekly, monthly intemal reports; quarterly reports to parliament; annual financial accounts and result reports to parliament Ongoing, feeds into annual budget discussion
Note: This table is inspired by OECD systems as they have evolved over the last 25 years. Countries such as South Africa, Malaysia, and Thailand have adapted parts of this fiamework to their own needs. For Indonesia, this table should be seen as a vision which could be realized by a series of reforms in the treasury law, state administration law, civil service law, cabinet procedures, financial management regulations, etc.
built by information disclosure alone can even be destroyed. In a recent survey by the South African NGO-IDASA--respondents from all sectors of society in South Africa recognized the link between transparency, participation and better government (see also Box 2.4). The survey was useful in deducing the country specific conditions that may be required to close the circle. The IMF Code of Good Practices for Fiscal Transparency provides a benchmark to measure a country's progress on transparency. It advocates greater 32
openness in the budget process through measures to maximize the effectiveness of civil society participation in the budget. 2.35 Apart from being an intrinsic requirement of democratic governments, the creation of an open and participatory budget process is a worthwhile goal in and of itself. The provision of comprehensive, accurate,
timely and frequent information on a

country's economic conditions and its budget
Public Spending in a Time of Change policies is also beneficial for the following key reasons: * If budget information is not available, it is difficult to discuss it. * It facilitates the identification of weaknesses through public debate, leading to the adoption of needed reforms to foster improved public programs and more efficient resource use. * It goes a long way towards building consensus on and commitment to social tradeoffs and increases faith in government. * It enables legislatures and civil society to hold government accountable, which is only feasible if one has access to reliable information on budget policies, practices, expenditures and outcomes. * It contributes to macroeconomic and fiscal stability as it prevents the buildup of a crisis in secret, giving scope for making smaller adjustments sooner. With a clear understanding of government policies investors may be more likely to invest in a country and can, therefore, facilitate investment and economic growth. budgets that threaten macroeconomic stability. To avoid this, budgetary processes need fundamental reforms towards more medium term planning, -budgeting and performance monitoring (using budget outcomes instead of outputs). However, this will require institutional arrangements and appropriate guidelines that would enable the different stakeholders in this budgeting process to determine, in a transparent manner, whether: (a) the public expenditure is according to plan; (b) this expenditure is reaching the areas or beneficiaries targeted; and (c) this spending is achieving expected outcomes on the basis of pre-specified measurable indicators. 2.38 There remains an urgent need for greater clarity in the roles of Parliament, Ministry of Finance, BAPPENAS and the line ministries in the budget process. Appropriate guidelines and rules of engagement between the Government and Parliament in the budget formulation process need to be finalized. 2.39 A first important step in this collaborative budget allocation process between Parliament and the bureaucracy is to ensure political "buy-in" to aggregate spending limits. This will tend to minimize instances where Parliament changes a budget submission such that fiscal sustainability is threatened. To achieve this, Government could consider presenting a fiscal policy outlook document to Parliament before the 9 actual budget process takes place. Such a report should explain the macroeconomic and fiscal constraints that limit aggregate spending. But beyond information and consent, the Government could consider adopting those budget rules-customary in other countries-that allow Parliament only limited changes in the budget, or only changes that do not affect the fiscal deficit. Table 2.2 summarizes the findings of an analysis of the budget amendment powers of parliamentarians in eighty-two countries by
9 Currently,the Fiscal Outlookdocumentis discussed in Parliament a plenarysessionthat is heldabouttwo at monthspriorto the budgetdebate.
2.36 Accountability differs from direct control. A good audit system combined with a watchful, active parliament, rather than a host of direct controls and regulations, will 8 help promote accountability. 2.37 In a democratic Indonesia, the budget process can no longer be a bureaucratic process. Involvement of Parliament and the Cabinet in the process can ensure that budgetary choices better meet the aspiration of the Indonesian people. At the same time, this involvement could deteriorate in populist
See "ControllingCorruption: A Parliamentarian's Handbook", Parliamnentary The Centre(Canada),1998.
33
Improving the Budget Allocation Process Warren Krafchik and Joachim Wehner (1998). 2.40 Within the aggregate spending limits as prescribed by macroeconomic conditions and fiscal sustainability, Cabinet could decide early on policy priorities. These policy priorities could then be reflected in sectoral spending limits, and individual line ministry limits. Some countries-together with the fiscal policy outlook-present their policy priorities in a "fiscal strategy report," sometimes supplemented - by "corporate plans" of individual line ministries. Within the agreed spending limits, line ministries could be allowed more leeway to allocate money according to what best fits the Government priorities already agreed on.
Table2.2. Rightsof Members Parliament Budgetary of in Matters Number of Members' rights Countries Mayreduceand increaseexpenditure revenue. and 32 May reducebut not increaseexpenditure. 17 4 May reduce expenditure,but only increase it with the pennission of the Government. May reduce and increase expenditureif alternative provisions are made 13 elsewhere. Rightsnot specifiedin detail 15 Not applicable (Nicaragua) 1 82 TOTAL Source.Inter-Parliamentary Union1986:Table38A The central ministries would increasingly refrain from setting detailed expenditure norms, and focus budget discussions with the line ministries on: (i) whether spending remains within the agreed limit; and (ii) whether the proposed spending is in line with cabinet priorities and line ministry policy statements. POLICY IMPLICATIONS 2.41 The current environment of change in Indonesia provides a fertile opportunity for revamping the budgetary processes in Indonesia towards improving public sector performance in service delivery. These efforts should supplement the Government's endeavors to deal with the unavoidable fiscal pressures that have emerged due to the crisis and associated fiscal risks. 2.42 In the short run, aggregate fiscal discipline should be fostered through publication of the government's mediumterm vision and associated budgetary choices. This could be supplemented by published reconciliation of budget estimates and actual expenditures that were incurred in the recent past. To this end, the government's 34 objective of achieving comprehensive coverage of fiscal operations --an endeavor that is currently underway with the transition towards the IMF-GFS classification and reporting of government fiscal accounts-is an essential step in the right direction. In addition, the line ministries should be required to provide the rationale for public intervention in that activity as part of the budget formulation process. This principle should be applied at all levels of government. 2.43 In determining the role of government in a particular new activity, it is recommended that a technical approach that important policy-makers provides information-such as the reason for public sector involvement and the appropriate instrument, competing claims on the budget-should be incorporated in the budgetary process. An appropriate mechanism should be available to the MoF and other key stakeholders to assess the validity of the cost estimates that are provided by the proposing entity of a publicly funded activity before the project or program is approved.
Public Spending in a Time of Change 2.44 Using the lessons from the ongoing social safety net program monitoring efforts in Indonesia, mechanisms to evaluate performance at each level of government should be introduced over the medium term. In a few key areas (e.g., public health centers, public schools, water and electricity) user surveys could be carried out to gauge the quality of public service delivery to target consumer groups. This is a powerful instrument to facilitate greater transparency and accountability in public service delivery. Disseminating the results of a well-designed and executed public service delivery survey can reveal if the Indonesian taxpayers are getting value for their money in that activity. 2.45 In the medium- to long-run, the govemment should introduce a multi-year rolling budgetary planning exercise. To this end, a committee consisting of officials from BAPPENAS, the MoF and other line ministries should be formed to start the exploratory work in this area. 2.46 To manage the - process of institutional change in government prudently, the authorities could consider forming a working group to explore options in government organization that would best match the changing environment. Such a process would: (i) identify forces of change; (ii) redefine the core processes of Government in light of the forces of change; (iii) review comparative advantage and existing roles of organizations; (iv) define a new organization structure of govemment, matching the new roles and responsibilities with the new organizational structure; (v) identify incentives to make these processes work; and (vi) institutionalize the new roles, responsibilities, organizations, and incentives in laws, regulations, and work arrangements.
35
Managing Across Levels of Government
3.
MANAGING ACROSS LEVELS OF GOVERNMENT
ndonesiais makingrapid movestowarddecentralization. Decentralization be a powerfultool can
to improvepublic service delivery and accountability in Indonesia. But if not well managed, it risks fiscal balance at the center. To maximize the benefit and contain the risks of the decentralization, Government should focus on implementation ensure detailed expenditure assignments, decentralize personnel, and design an equalizing grants scheme. In the medium-term, accountability at local level should be strengthened by a larger local tax base, increasedtransparency,and prudent access tofinancial markets. TIHE PROCESS HAS BEGUN 3.1 Until today, fiscal authority in Indonesia has been highly centralized. Revenues of the center constitutealmost 90 percent of all revenues, and sub-national governments depend for almost 60 percent of their spending on central government grants. The central government, including deconcentrated central departments at provincial, district, and village levels, spends over 80 percent of all government spending. The development budget is determined in an elaborate bottom-up process, in which, however, the center has
Figure3.1. Indonesia HighlyFiscallyCentralized is

Central share of Spending
90 60 70 60 50
30 30 40 50 60 70 x0 Central 60 oaenrvne I 00
Source. IMF, GFS.
the final say. This situation, which is at odds with the practice in most large developing countrieswhether unitary or federal-is about to change (Figure 3.1). 3.2 A majEorprogram of govermental decentralizationis underway. Twolawshave been approved by parliament on various aspects of 36
decentralization, Law No. 22/1999 on Regional Governiment (IJU PD) and Law No. 25/1999 on the Fiscal Balance between the Central Government and the Regions (UUI PKPD). If implemented fully, these laws promise to transform intergovermental fiscal relations in Indonesia. However, many details of the program have yet to

be worked out, and questions remain about the ultimate goals and the process to achieve them. This chapter discusses three issues: (a) the 3.3 major features of the new decentralization laws and of their budgetary impact; (b) the impact decentralization on inter-governmental fiscal relations, and potential issues; and (c) specific problems related to the implementation of reforms, especially as the implementation will affect public service delivery and accountability of local governments. Some of the issues that are highlighted here are being addressed more comprehensively in ongoing technical assistance by the IMF and the World Bank. THE NATURE OF THE DECENTRALIZATION IN INDONESIA 3.4 Since the 1950s, Indonesia has been a highly centralized but multi-tier unitary state, with provinces and then local governments as the tiers
1 under the central government. Laws No. 22/1999
assembly (Dewan Perwakilan Rakyat Daerah, or DPRD). In contrast, the provinces will retain a hierarchical relationship with the central government. 3.6 Second, with some limited exceptions, Law No. 22/1999 also makes all deconcentrated central government ministries at the province and the district the responsibility of the respective local government. 2 (The exceptions are for defense and security, foreign policies, monetary and fiscal policies, judiciary affairs, and religious affairs). This change promises a major reorganization in the way in which public services will be delivered in Indonesia. 3.7 Third, Law No. 25/1999 on the Fiscal Balance between the Central Government and the Regions (UU PKPD) alters the transfers received by local governments from the central government. The current routine transfer that is largely used to 2 In the current system, the central governmentand its departments calleda kepala ministries have"deconcentrated" or kanwil,at the provinciallevel;in some kantorwilayah, cases, the kanwil has a sub-branch at the district (or subdistrict) levelcalleda/kantor departemen, kandep. The or and provincehas its ownplanningagency(Bappeda) various autonomous "decentralized" departments (or dinas) under its own control at the provinciallevel,generallyconsistingof departments own revenues (called a dinas pendapatan for daerah,or dipenda), as well as dinas for educationand culture, health,publicworks,trafficmanagement, agriculture, livestock, fishery, forestry, plantations, industry, social welfare, labor, and tourism, all of which have central in kanwils;the government counterparts the deconcentrated provincemayalsohave branchoffices(calledcabangdinas) at the district level, although this is apparentlynot that common. Like provinces,districts have an autonomous "decentralized" department chargeof ownrevenues in (again, calleda dipenda),and they generallyhavedepartments(or dinas)for serviceslikehealthand publicworks,althoughthe rangeof these departments dependsuponsizeand location of the district. In the new system,the deconcentrated central government departments the provincial at level will become theresponsibility the province, thoseat the district of and level will be turnedoverto the district.
and 25/1999 propose to change this in several fundamental ways. 3.5 First, Law No. 22/1999 on Regional Government (UU PD) eliminates the hierarchical relationship between the provincial and the district The district governments governments. previously known as kotamadyalkabupaten and now called kota/kabupaten - will become fully autonomous, so that their heads (the
walikota/bupati) will no longer report to the

governor of the province. Instead, the district heads will be responsible to the locally elected For detailed discussionsof intergovemmental issues in Indonesia priorto theenactment the recentlaws,seeAnwar of Shahand Zia Qureshi, Intergovernmental FiscalRelationsin Indonesia, World Bank Discussion Paper No. 239 (Washington, D.C.: The WorldBank, 1994);and RobertH. Aten, "Why IncreasedLocal DemocraticDecision-making Would Aid IndonesianEconomic Development", USAID Working Paper(Jakarta:December 1997).
37
Managing Across Levels of Government pay the salaries of local civil servants (the Subsidy Daerah Otonom, or SDO) will be eliminated; also eliminated will be general development transfers known as block Instruksi President, or block Inpres. These two transfers are instead combined into a general allocation fund whose total amount is specified as 25 percent of central government domestic revenues and whose distribution among local governments will be determined by formula. Law No. 25/1999 also introduces revenue sharing for provincial and district governments, assigning each level of government its share of revenues from taxes on land and buildings, the transfer of
Table3.1: Impactof the FiscalDecentralization Law

(1999/2000 Budget Data, Rupiah Billion) After
Centralgavernmpnt
Existing Policy
Fiscal Balance Law
Decentralizing Responsibilities
Domesticrevenue Expenditure trnsfer and Own Expenditure Transfers ResourceSharing Deficit

Prmvines
142,204 219,604 190,337 29,267

-
77,400
142,204 231,518 190,337 41,181 3,064 92,378
142,204 216,540 172,294 41,181 3,064 77,400
Revenueand transfer
Resource shares
10,065
-
10,510
613
10,510
613
Transfers Expenditure
Divtriec.and lwpr
5,408 10,065
5,240 10,065
3,555 10,510
Revenueand transfer Resourceshares Transfers Expenditure

Deficit
29,205
-
23,505 29,205
-
43,618 2,451 35,467 29,205

(14,414)
43,618 2451.46 35,787 43,618
Sources MOFstaffestimnates.
land and buildings, forestry, mining, fisheries, oil, and gas. Other local government sources of revenues (e.g., own source revenues, fees and charges, profits from government enterprises, borrowing) are unchanged, as are revenues from specific Inpres grants used to finance development projects in areas like primary schools, health 3 facilities, water supply, and roads.
3.8 The fiscal balance law puts considerable pressure on the central budget. It requires sharing of resource revenues with the provinces and the districts and in addition a transfer of 25 percent of domesticrevenues to sub-national government. Of that 2.5 percentage points is for the provinces, and 22.5 percentage points for the districts.
taxes include the hotel and restaurant tax, the entertainment tax, the advertising tax, the street lighting tax, the mineral tax, and the water use tax.
Provincial taxes consist of a tax on motor vehicles, on the transfer of motor vehicles, and on motor vehicle fuel. District
38
Public Spending in a Time of Chantge 3.9 If this revenue decentralization is not matched with expenditure decentralization at the same time, it risks increasing the central government deficit. In total, the fiscal balance law requires the central government to devolve some 1.2 percentage points of GDP in addition to the transfers already given to provinces and districts (Table 3.1). At the same time, if districts governments-the main beneficiaries of the additional resources-do not receive additional tasks, they could waste resources on non-core functions.Thus, Government needs to decentralize
expenditure responsibilities. BUDGETARY IMPACTOFDECENTRALIZATION
in light of Indonesia's size and diversity, regional governments are likely to spend much more than the 25-30 percent of revenues now assigned to them. Whereas the Fiscal Balance law is specific about the revenues to be decentralized, the Regional Governance law only gives a vague indication of spending responsibilities. Moreover, equalization is defined only in the broadest of terms. And regional borrowing arrangements the achillesneeded in decentralization attempts around the world, needs firming up with urgency.
IMPACTONINTER-GOVERNMENTAL RELATIONS
3.13
The new decentralization laws have the
Unequal Sharing. Resource revenue 3.10 sharing will make equalization among regions more difficult. The resource revenues shared with the regions only adds up to 0.3 percent of GDP because the lucrative offshore oil and gas is excluded, and the sub-national share in the large ReforestationFund is low. Moreover,the oil & gas shares apply to revenues after taxes, not to the corporate income tax received by central government), 3.11 But only five provinces will benefit significantly from the law (Riau, Aceh, Irian Jaya, East and Central Kalimantan). Since these provinces have above average per capita income, the revenue sharing will increase inter-provincial inequality (Figure 3.2). This effect can only be partially undone by equalizationin general revenue allocation, because no negative transfers can be given to provinces. 3.12 Timepressure. Indonesia's new laws only allow 2 years for full implementation of decentralization, and the tasks ahead are massive. Decentralizing 1.2 percent of GDP in spending would not be a problem, if it were not for the time currently imposed pressure. In fact, over time, and
potential to increase significantlythe accountability of local government officials. It is through this accountability that the major advantage of decentralization is obtained - moving government closer to the people. The leaders of the local governments: the governor at the provincial level, the bupati at the kabupaten, and the mayor at the kota - will now be chosen by the respective elected council (although not directly by the voters), rather than appointed from above, so that their responsibility will be directed toward their own elected council and not upward to the central government. The assignment of significant new expenditure responsibilities to provincial and, especially, to kota/kabupatengovernments has the potential to achieve the efficiency gains that come when governmental decisions are more responsive to the wishes of its citizens, so that public services are provided in amounts that correspond more closely to the preferences of the individuals in those jurisdictions, rather than at uniform national levels. Other potential gains include greater revenue mobilization because citizens may be more willing to pay local taxes to provide local public services and because local governments may be more familiar with, and so better able to tap, local tax bases.
39
Managing Across Levels of Government
Figure 3.2: Increased Inequality Rupiah per capita grants and resource revenues per province, and non-oil GDP per capita Transfers + Resource Share (Rp. '000 per capita)
700 800 l 600 < I * 500a 400 300 | 200 100-1 0o 0

w +
After Natural ResurceSharing
Before Natural w Resource Sharing
.. ..... . ..................... . ....... ................................... ..... . ............................................. .......................... . ............... .......

1000 2000 3000 4000 5000 6000 7000 8000
9000
..........J ...
10000
Non-Oil GDP Per Capita (Rp.'000)
3.15 There remains th urgent need for better the poes 3.14 The two laws provide mainly for coriaon f twrd expenditure, but not revenue, decentralization. imlementation of the two fiscal and administrative Despite the enactment of revenue sharing for natural resource revenues, the laws do not give decentralization laws (Law No. 22/1999 TUPD local governments any new, meaningful, and and Law No. 25/1999 UU PKPD). Currently,in the locally controlled tax instruments. This is a major Central Government, the Ministry of Finance, the mii otni aj lDepartment of Home Affairs and the Office of the limitation of the new laws. To establish a link between costs incurred and services demanded by State Minister of Regional Autonomy are involved citizens,.local.govemments have the ability to must in this effort, but there is a lack of clarity on who is citiens,melo algoern ens . sthaei he . responsible for what aspect of this implementation make some real choices in their use of tax instruments (at least at the margin). It is this process. Clearly, this is a complex task and may .. ~~needattention from the highest levels in linkage that is crucial: it makes the citizens aware that there is a connection between the taxes that government.The laws also lack any specific details they pay and the services that they receive (at least that their implementation will require.
if there is transparency), and it establishes The Process and Goals of Decentralization
accountability on the part of the local government officials.This does not require that the government control all of its revenues, only enough to change on the margin the revenues it collects and also the services it provides. However, even this modest amount of local control is not present in the decentralizationreforms.
3.16 The first step in most successful decentralizations is the development of a general framework within which the broad goals of the reforms are articulated and agreed upon, in somethingthat might be embodied in a goverment "white paper". This step seems to have been completely skipped in Indonesia. Instead, the
40
Public Spending in a Time of Change Government has moved directly to drafting decentralization laws, and it is now beginning to think about drafting the implementing regulations that go with these laws. 3.17 There is a lot to be said for "getting on with it". But this kind of shortcut raises some important questions such as: Is the Government clear on what it wants to achieve with decentralization? Is there widespread support within the Government for these objectives? Is the decentralization strategy that is moving forward based on a well thought-out plan, or is it more an idea in its infancywith many unanswered questions about the broad structure of roles and responsibilities that will emerge? 3.18 Countries that have decentralized successfully have always issued a clear statement of the objectives of the reforms. In Indonesia, there has been little advance preparation for decentralization, despite the existence of a pilot program (the "District Autonomy Pilot Program", or DAPP), which was established in 1994/1995in which a number of local governments were given more responsibilities. Further, there seems to be only one common issue that is being addressed by the decentralization program - the need to move government decision-making on taxes and expenditures to the provincial and the district level. It is not yet clear that this means "closer to voters", and it is also not yet clear that central government ministries will resist the effort to impose regulations, mandates, and minimum standards on local government service delivery, all of which would reduce local autonomy. It is certainly clear that the reforms do not extend any new significant revenue raising powers to the local governments. 3.19 International experience shows there is no single best structure for fiscal decentralization. There are many versions, each appropriate to what a country is trying to accomplish. When there is no clearly articulated decentralization policy that commands widespreadsupport and consensus,then 41 there is no road map for designing all features of the program: the laws, the regulations, the transition, the implementation, or the evaluation. Such would seem to be the case in Indonesia. The absence of such a plan tends also to affect implementation: a) The laws defining expenditure assignment and revenue powers are being written in an uncoordinated way by two different groups within government. Without a policy design to guide this drafting, it is virtually certain that local governments will have more expenditure responsibilities than revenue resources, setting the stage for a "softbudget constraint" for local governments. The design of intergovernmental transfers is not being guided by clearly stated government objectives concerning equalization, the desired level of expenditure control by line ministries, local government revenue mobilization, and the like. Without a clearly stated set of objectives and priorities for the decentralization program, it will be difficult to draft the implementing regulations for the various components of the program. Appropriate interactions between government and civil society focus groups will be a valuable instrument for soliciting ideas and fostering citizen acceptance to the envisage reforms. Similarly, the absence of a clear statementof objectives and priorities will make it impossible to evaluate the success for the various components of the program or even to put in place an evaluation effort. Indonesia's decentralization program will grow and develop as the country changes in the coming years, and a strategy for adjusting the structure of the decentralization policy to keep up with this economic
b)
c)
d)
e)
Managing Across Levels of Government development will be essential. This fine tuning must be guided by a clearly articulated set of objectives and priorities. ISSUES THE IN IMPLEMENTATION OFTHE
ISSUS THEIMPEOMENS I IN
3.23 In addition, there is much uncertainty about the revenue-expenditure balance at the local government level. Some local officials believe that there will be no problem in paying the salaries of any new civil servants, in part because they believe that the SDOwill continue, but there could be a problem in funding other routine
FTEmistakenly
3.20 To take the next steps in a fiscal expenditures because some new departments will decentralization program, the Government of be created that do not currently exist and funds Indonesia will need to address a number of must be found for these departments, especially for important issues, most of which affect public maintenance, transportation, utilities, electricity, service delivery and accountability. The key issues papers, water, phones, and the like. Some district in this regard are the following: officials also believe, that their revenues will be in thiIreardarethefolo :lower under the new system. Others, at both the 3.21 Revenue-Expenditure Balance. There province and the district levels, estimate that may well be a mismatch between the revenues that revenues will increase substantially with the are assigned to local governments and their turnover of the deconcentrated central government expenditure responsibilities. On the revenue side, agencies and their budgets. For example, in North the idea is to guarantee a transfer to local Sulawesi revenues and expenditures are Rp. 155.1 governments of 25 percent of domestic revenues billion in the 1999/2000,and officials estimate that plus a share of natural resource revenues, in the revenues will increase to Rp. 1,413.9 billion in form of intergovernmental transfers. On the 2000/2001,given the estimates of their share of the expenditure side, the goal seems to be the general allocation fund, of specific grants, of dexendituraliza of responsibility for a broad range revenue sharing, and of salaries and development decentralization orepniiiyfrabadage budgets of deconcentrated agencies. Clearly, there also beconeater agendierearlities, of expenditure functions. There is little evidence of any analysis of the expenditure budget for subwill also be greater expenditure responsibilities, national governmentsimplied by this assignment of and the department estimates that expenditureswill increase from Rp. 155.1 billion in 1999/2000 to functions, and of the adequacy of the 25 percent revenue transfer. Rp. 1,11 billion in the next year. On balance, l9.1 then, the estimates are that revenues will exceed 3.22 It is not known how the revenue transfer expenditures by Rp. 294.8 billion. Regardless, compares to the targeted level of expenditures, however, some local officials indicated that they were simply unaware as to what will happen to either at the aggregate or at the provincial level, For example, in the 1999/2000 budget the 25 their budgets under the new decentralized system, percent allocation is projected to equal Rp. 35.6 in large part because they have no knowledge of trillion (i.e. 25 percent of Rp. 142.2trillion) which the formulae for grant distribution. Similar is roughly double the projected regional routine uncertainty is present among central government expenditures of Rp. 19.5 trillion. Also, revenue officials. sharing is estimated at about Rp. 6-7 trillion, so that total local government transfers (aside from 3.24 Local GovernmnentCapacity to Deliver the specific allocations) are estimated at Service. The ability of provincial and local approximatelyRp. 43 trillion, governments to absorb the new expenditure responsibilities is critical to the success of decentralization as well. The broad issue here is whether provincial and district governments can 42
Public Spending in a Time of Change absorb the "back office" functions that are now centralized, including personnel management, data processing, procurement, and contracting, and so provide the full range of services that will become their responsibility in the new system. 3.25 One view, often expressed by local officials, is that they will have little difficulty in providing these services, in part because many of these functionshave already been largely shifted to local governments. In North Sulawesi, for example, 20 programs funded by specific Inpres allocations have been channeled directly to district governments by the provincial Bappeda since 1994,including programs for road, health, drinking water, primary schools, agriculture, and marketplace development. Other officials similarly claim that "it is easy to turn over an agency to a district." However, another expressed view is that the funds necessary to provide these support services may not be forthcoming, even if the local skills needed to provide the services are present. For example, the district of Minahasa participated in the District Autonomy Pilot Program (DAPP), during which the transfer of 9 (deconcentrated)departmentswith 500 employees to the district occurred (manpower, manufacturing,trade, social, cooperatives,mining, health, rural development, and registration). According to officials there, the results of the pilot were threefold: there was no problem in absorbing the employees, there was no problem in providing most of the basic services of the departments, but there were financial problems because only funds for routine expenditures (salaries) were transferred to the district and no funds for development expenditures were transferred. This last problem could also arise in the decentralization program, if sufficient funds (aside from salary support) are not provided; if salary support is also not provided and there are very mixed views on the likelihoodof this - then additional problems will be created. 3.26 The magnitude of the tasks to be undertaken by the civil service will be much greater in the new system, so that complete 43 absorption will take some time and will vary greatly by local government. For example, one local government secretary estimated that the extra burden of these functions could be as much as Rp. 15 billion relative to a district budget of Rp. 111.0 billion; another local Secretary estimated that the district civil service would roughly double in size with the reforms, from 12,000employees to 24,000 employees and in another district, the personnel officers said that the number of district dinas would increase from 11 to 21 (adding new ones in fishing, livestock, mining, forestry, land institutes, industry, manpower, education and culture, cooperatives,and trade), with more than a doubling of district civil servants. In the province of North Sulawesi, there are currently 35,000 employees in deconcentrated agencies throughout the province (relative to 45,800 provincial civil servants), and, with decentralization, most of the 35,000 employees in the deconcentrated agencies would become provincial or district civil servants. Also, officials in professional and highly specialized deconcentrated agencies (e.g., education, manpower) worry that they often get extensive central government assistance on matters for which the local governments have little expertise or interest. 3.27 Overall, the views expressed by the officials interviewed suggests that only 10 to 20 percent of the districts can absorb all of their new duties quickly. A BAPPENAS rating indicates that only 3 of the 27 provinces meet appropriate standards at present, especially in the quality of the civil service. Some local government officials are already preparing for the changes. For example, in North Sulawesi, civil servants have been encouraged to improve their educational levels. Currently, there are roughly 100 people in BAPPEDA; 60 percent have the equivalent of a B.A. or B.S. (or 4-year college) degree, 13 have an M.A./M.S. degree, and another 15 are working on an M.A./M.S. However, this appears to be the exception. Instead, most local government officials
Managing Across Levels of Government appear to be waiting for central government assistance. 3.28 This general issue raises a number of specific questions: a) Is the quality of the human capital in the provincial and district governments up to the task of delivering the services to be transferred, or is the idea simply to absorb central employees? How will management be handled? Will former central employees be brought into the provincial and local service, directly under local government managers? Will the managerial personnel from the central government be transferred in? Are there some purely physical limits to the absorption of these functions by subnational governments? For example, is the provincial and local computer system up to the increased record keeping tasks implied? What new budget formats and accountability systems will be required when these new functions are absorbed into the decentralized system? What provisions have been made, or need to be made? Will a new legal framework and a new reporting system need to be established to govern the budget decisions, personnel management, and the like of sub-national governments? Will these governmentsbe able to deal with the compliance costs involved? by central government employees who anticipate being absorbed into the sub-national government service. More than anything else, civil servants are concerned about these issues. 3.30 By most accounts, central government civil servants largely will prefer to stay at the center, rather than be transferred to the province or the district. However, the decentralization will eventually require massive transfers to local governments perhaps as many as one-third of the 1.5 million central government civil servants, according to one estimate. 3.31 Civil servants at all levels are largely in the dark about their future, something that has created much anxiety for them. Surprisingly, some (though clearly not all) officials are unconcerned about salaries, believing that the central government will continue to pay their salaries via the SDO. Officials more often expressed anxiety about guarantees (or lack thereof) for certain positions, especially supervisory ones in departments that will be created or moved to district control. They noted that job rankings could be an issue. For example, one district Secretary has a current ranking of 2B; under the new system, he believes that his position will require a higher ranking of 1B (the same as a provincial-level Secretary), so he may not actually be "qualified" for his position in the new system. More generally, with new required rankings, there may not be sufficient numbers of qualified civil servants for the various positions. Central government civil servants in deconcentrated agencies are also worried about career advancement in local governments, especially since they typically view themselves as "specialists" whose skills may not be appreciated or understood in a local civil service; indeed, a district counterpart agency does not always exist for the deconcentrated agencies, so that regulations need to be written. Issues like seniority (e.g., who will be the head of a deconcentrated department merged with a district counterpart agency), promotion (e.g., will the 44
b)
c)
d)
3.29 Civil Service Issues. The civil service implications could be daunting, and it is not clear how far planning has gone on this set of issues. Issues of seniority, compensation, pensions, fringe benefits, and work rules were raised consistently
Public Spending in a Time of Change criteria for advancement change), and transfer (e.g., who will determine mobility) in the new system are very hazy at present; these will be determined by the district (not the central) governmentpersonnelpeople and their evaluations, and, again, the district may not apply the same standards as currently. 3.32 In the face of these issues, some officials believe that central and local government employees will be placed on the same track in a new system in order to maintain the "unity" of the civil service, as well as to allow the mobility of civil servantsacross provincesand districts; in their view, a unified system will require that the salaries continue to be paid by the central government. However, officials admit that this is largely speculation at this point, and they are awaiting the issuance of regulations. 3.33 Accountability of Elected Local Officials. There is a widespread belief that local accountability will be improved by the electoral process. Voters will now have some say in the determination of the composition of the local councils, and local officials often claimed that "expectations are high". The new procedures for the selection of the local head should also improve accountability. Under the old system, the local head was selected with the heavy involvement of the centralgovernment, so that the responsibilityof the local head was directed mainly upward, to the central government. In contrast, under the new system the responsibility of the local head is to the elected council that will elect the head without approval of the central government. Finally, there will be local approval of budgets, and this leads to increased accountability. Many facets of the reform will therefore move government closer to the people. 3.34 However, there are also some gray areas, where the new reform may not lead to as much accountability to voters. There is no proposal for significant expansion of local revenue raising 45 powers, a point raised by several local officials as Central a limitation on local autonomy. government ministries may well impose a range of regulations, mandates, and minimum standards on the deconcentrated agencies, features that could severely limit local autonomy and therefore local Intergovernmental accountability to voters. transfers may well be conditional, which will limit local government discretion. The electoral prdcess is not direct; that is, the responsibilityof the elected official remains directly to the party rather than to the voters, and the party can select - and remove candidates on its authority. Finally, it may take some time before the accountabilityimplied by free elections actually occurs. 3.35 In fact, local officials have mixed views on the changes in accountability. Some believe that the new laws significantly increase their autonomy, and so their accountability. They point especially to local approval of budgets and local discretion on spending levels and composition. Central government officials generally endorse this view also, fearing that the central government may well lose overall budgetary control but that this is the price of decentralization. 3.36 However, some officials also believe that they already have substantial autonomy in the In particular, the current current system. formulation of the provincial development budget is a bottom-up procedure over which local officials exert substantial influence, and local officials do not believe that the decentralization will enhance (or lessen) this autonomy; the main change with the new laws is that final approval for the budget will come from the provincial assembly, rather than from the central government. On balance, these officials believe that they will have "more room to maneuver" and "less intervention from above", but that these changes will be minor. 3.37 Finally, there are some officials who believe that local autonomy will be largely unaffected by the new laws. There is a strong
Managing Across Levels of Government possibility that central government ministries will resist the transfer of their line ministries to provincial and district governments via regulations and mandates, tendencies that are already apparent, because the ministries will lose much influence, resources, salaries, and the like with any transfer. These local officials also maintain that local autonomy will suffer from a lack of own local revenues; in the words of one district Secretary, there will be "decentralizationin management"but there will be "no decentralization in finance". 3.38 Central Government Assistance and Leadership-A Necessary Imperative. Somewhat paradoxically, successful fiscal decentralization requires a strong central govemment to lead the process. Given that the Indonesian system will have over 300 provincial and district local governments, an important issue is whether the central government has the ability to provide the oversight, guidance, and leadership necessary. 3.39 It is necessary for the central government to carry out analytical work to evaluate and to monitor the decentralized fiscal system on a continuous basis. This involves identifying tax effort performance, tracking local budgets, evaluating any proposed alternative fiscal reforms, and so on. Second, the central government must maintain the intergovernmental transfer system. This will involve a number of activities, like updating the basic equalization formula, and evaluating the equalization features of the grant system. In addition, if there are conditional grants, there needs to be a compliance monitoring system in place and an incentive structure to assist enforcement. Third, and related to the transfer system, one possibility is that the Indonesian government could create a finance commission to oversee and advise on inter-governmental fiscal relations. This body would necessarily have to possess a stronganalytic capability, some modeling support, and an ability to make the system transparent. A finance commission could stand between the central and local governments,thereby 46 becoming the honest broker on intergovernmental fiscal relations. It could also resolve disputes between ministries in terms of fund allocations, thereby de-politicizing intergovernmental fiscal decisions. The central government will play a major role in organizing and supporting this unit. 3.40 With the ability of provincial and district governments to borrow (with the approval of the central government), the central government will need to develop a regulatory framework that can monitor the compliance that goes with this (e.g., disclosure, purpose of borrowing, eligibility, limits). The central government will need a fiscal information system to monitor the progress of decentralization and to serve as the database for research necessary to continue to fine-tune the system. Many large countries with significant intergovernmental fiscal programs have moved to develop a fiscal information system (e.g., Brazil, India, U.S., Canada, Australia). A management information system along these lines is currently being designed in the Ministry of Finance. The role of a centralized debt management unit is critical in this regard. 3.41 The central government, if it imposes the condition of a hard budget constraint on subnational governments, will be charged with determining whether local governments are in compliance. Central governments also must lay down the rules for audit. 3.42 To the extent the central government imposes regulations, mandates, and 'minimum standards (and it almost certainly will because all countries in the world do this), there must be a system to monitor compliance with these requirements. Examples include everything from compensation rates for employees, to environmental regulations, to the adherence with standards for school teachers, to the expenditureof minimum amounts. In addition, the central government may take the leadership in providing technical assistance and training to local
Public Spending in a Time of Change governments. The more technical the training, the more likely is the central government to lead the training and technical assistance. 3.43 In any intergovernmental fiscal system there will be disputes between the central and local governments, among local governments, and even between ministries. Among the possible sources of contention are the specific data used in the formula distribution, compliance with grant conditionality, unclear expenditureassignments, and the like. The resolution of such disputes will require central government involvement. Again, a Finance Commission can assist in this task. Also, if the central government adopts options such as a Financial Control Board to deal with bankruptcy, or if the government acts as a guarantor for local borrowing, central control and monitoring will be required, as will sanctions and remedies for defaulting local governments. 3.44 All of these considerations point to the necessity that fiscal decentralization calls for a strengtheningof the central government'sability to lead and manage the process. A necessary condition for successful decentralization is to have in place a strong central government unit responsible for intergovernmental fiscal relations. This in turn raises some importantquestions: where in government will such a unit be placed, how will its information system be supported, and how will it be staffed? 3.45 Revenue-raising Powers for Local Governments. Under the existing regulatory framework in Indonesia, no provisions have been made for local governments to raise significant amounts of their own revenues. This means that there is presently no plan to devolve significant new taxing sources to local governments. If this is not done, there will not be full accountability of elected officials (or ruling parties) to the votertaxpayers. If all expenditures are financed by intergovernmentaltransfers, the voters will not feel the pain associated with better services, as they 47 would if the local government had some rate setting powers. 3.46 Local governments have some but limited autonomy on the determination of rates and bases. No new taxes for local governments are currently under consideration at the central government. Many local officials would like to take complete control of the land and buildings tax, something that apparently was considered by the central government at one point before being abandoned. Local governments can propose new taxes under some circumstances,with approval necessaryat the central government, and local officials are exploring these possibilities. For example, the district of Bitung has a major port, and possible new taxes include district fees on the entry and exit of ships, on handling of cement, and on handling of asphalt. 3.47 A Hard Budget Constraint for Local Governments. It is not clear that provision has been made for the imposition of a hard budget constraint on sub-national governments. Two conditions are necessary. First, there must be a reasonable balance between expenditure responsibilities assigned and revenue instruments available. Second, the local governmentmust have some access to tax rate setting so that they can tax their constituents to cover any shortfall. As of now, the balance between expenditure assignment and revenues has not been worked out, and, as noted above, there is no provision for significant local taxing power. 3.48 Local government borrowing and onlending arrangements from center to subregional governments need to be clarified. Currently, provincial and district governments can borrow for capital projects, with central government approval, and these loans often come from regional banks. These banks are largely conduits for money from the center, are run by local bureaucrats for the benefit of the local government, and have just been recapitalized.
Managing Across Levels of Government Given these considerations, there is a real concern at the central government that local borrowing will grow out of control, despite the approvalnecessary for any such borrowing as specified in the two decentralizationlaws. In fact, some local officials stated that they believed that local loans would be assumed by the central government in the event that the local govemmentwas unable to service the loan, something that is undoubtedly mistaken but, if widely held, something that would also eliminate a hard budget constraint at the local level by creating a severe "moral hazard" problem for local officials. 3.49 The Laws and their Implementing Regulations. The new decentralization laws in Indonesia are stated in very general terms, and the implementing regulations which, in fact, will define the structure of decentralizationin Indonesia have not yet been issued. There also seems to be some ambiguity about who will issue these implementationregulations. Some important issues that need to be specified in these implementing regulations include: the exact assignment of expenditure responsibility, especially at the provincial level; local tax authority; local user charge authority; borrowing powers; civil service regulations; and the structure of the grants system. 3.50 The needfor a Transition Plan. A fiscal decentralization program in a large and diverse country as Indonesia needs a carefully thought-out transition plan. A major issue is the speed with which a decentralization program can be implemented, even if the plan is carefully thought out to cover all of the issues described above. Even if some governments can assume the responsibilities, there are concerns about the treatment of the remaining provincial governments. Will they be brought slowly into the system, with more limited powers until they prove their ability to take on the new responsibilities? What will be the criteria for graduation to the next class of municipality? What training will their civil servants be provided to assist them in their new 48 responsibilities? These questions must be answered in a transition plan.
DECENTRALIZATION COULDREINFORCE CIVILSOCIETY PARTICIPATION
3.51 Decentralization strengthens citizen participation by bringing governments closer to the people they serve. Its success, however, depends on the strength of the local government institutionsAEmpirical evidence from a sample of 80 countries has shown that citizen participation and public sector accountability go hand in hand with decentralized systems of government.5 Indonesia's move to a fiscally decentralized system of government is a step in the right direction in this context. However, unless the local administrations have sufficient powers, resources and accountability mechanisms (such as: citizens complaint mechanisms; appropriate provisions for recall of elected officials for negligence or misconduct; fostering independent "think tanks" for critical thought and evaluation; independence of judiciary and a free press) in place, the success of decentralization could be curtailed. Also, these changes will take time to implement effectively. 3.52 During the past year, Indonesia has taken
significant steps towards increasing the involvement of civil society in budget formulation and implementation matters in connection with its Social Safety Net Program. Actions being undertaken on a pilot basis in Indonesia under the World Bank's ongoing Social Safety Net
Adjustment Loan (SSNAL), Urban Poverty Project
and Kecamatan Development Project provide valuable lessons regarding alternatives for enhanced civil society involvement in the design, delivery, monitoring and evaluation of development projects. These measures could be customized on the basis of Indonesian conditions and their use eventually expanded to cover the
4 See Heymans, Chris (1996) and Anwar Shah (1998) 5 See Huther, Jeff and Anwar Shah (1998).
Public Spending in a Time of Change entire government budget. Such measures to enhance participatory budget implementation include: provision by the authorities timely and accurate information at all levels of government; empower local communities to monitor public expenditure programs; provide for independent verification of monitoring reports; establish appropriate complaint mechanisms, and make greater use of civil society as independentmonitors at each stage of program delivery. 3.53 To improve the scope for citizen participation in the budget formulation stage, the World Bank Urban Poverty Project utilizes a proposed "KelurahunForum" whereby community groups are to identify and implement the subprojects that are to be financed.6 Provincial and local governments will select the poorest Keluharans (neighborhoods) on the basis of transparent and well publicized criteria. Facilitators will help with targeting and consultants (who may be consulting firms, NGOs, government agencies or universities) will assist the sub-project process. Community groups can select their own facilitators and handle their payment, based on an output and profit sharing basis. Participation by local agencies is encouraged. Mechanisms also need to be in place to review exogenous changes that take place during the implementation of a public project/program. Similar arrangements have been put in place in the Kecamatan Development Project, where a Kecamatan-level council reviews and funds village proposals (one vote per village and one vote per camat). Allocation of funds within the kecamatanis determined by voting at the village level. 3.54 To promote transparency in the use of public funds and to raise the awareness of the rights and responsibilities of stakeholders, to minimize corruption by local elite or beneficiaries
6 See World Bank Urban Poverty Project, PAD (May 1999) and Frida Johansen (1999), "New Indonesia Urban Poverty Project: Harnessing Conmnunity Initiative to Generate Jobs", World Bank.
of the funds, free flow of information will be promoted. To this end, a brief circular about the project/programmay be posted in public places (in pre-selected poor neighborhoods in the case of the Bank's Urban Poverty Project), such as markets, public thoroughfares, schools, places of worship through NGOs and religious organizations and information Door-to-door staff. agency dissemination by facilitators in targeted areas may also be used. In the case of the SSNAL, the BAPPEDA in each of at least 105 Districts has compiled an information folder ("SSN folder") containing the information about the social safety net program (such as description, program start dates by location, plus standard forms, bookkeeping system and environmental and land acquisition guidelines). Each District has made available its SSN folder to the public through providingpublic access during official government hours to said folder at a District office. The certification letter shall include a list identifying each such office, (the "District Information Center"). Each District has provided: (a) at least 2 local NGOs, 2 local media organizations and 1 local university with a copy of its SSN folder and the name and location of the relevant District Information Center; and (b) BAPPENAS with a list identifying such organizations. The certification letter shall include such lists. 3.55 To improve transparency for direct contracting, the executing agency will make public the contract amount. For competitive bid contracting, the executing agency will: (i) make public the shortlist, winning contract, amount, name, and members of the Board of Directors of the winning contractor; and (ii) Implementing agencies at each level will make available upon request to any person or group, at any time during work hours, without undue delay, and free of charge, their monthly reports; lists of eligible households will be publicly available at the village level.
49
Managing Across Levels of Government 3.56 To ensure effective participation, each publicly funded program should contain an adequate specification of the complaint resolution procedure as it relates to the administrativeactions or sanctions against officials after preliminary investigation of complaints, and referral of complaints based on the results of preliminary investigations of complaints to the other appropriate institutions with the Government for investigation and disciplinary/judicial action. The complaints resolution procedure should include a capacity for a "rapid response" unit at the pusat to handle serious allegations (financial misappropriation or political interference). After a complaint is filed and investigated, a public report detailing the results of its investigation would be produced. Such a report should describe: (a) the complaint, including the specific locations and social safety net programs investigated; (b) the composition of the investigating team; (c) the steps undertaken during the investigation (d) the results of the investigation, including the type of infractionsand the number of government officials and non-government personnel involved, and; (e) recommendations for corrective action for each complaint which is found to be substantiated. POLICY IMPLICATIONS 3.57 In many respects, Indonesia's system of fiscal decentralization would seem to make it an outlier when compared with other country experiences. It has many of the characteristics of a country that typically chooses decentralization as an economic policy: a large population and land area, a diverse population, and a reasonable level of economic development. However, until now, Indonesia has chosen to remain a centralized state, perhaps because of a fear of civil unrest. The decentralization reforms that the Government of Indonesia has now embarked upon have the potential to transform intergovernmental relations. However,many things must be done to achieve the gains from decentralization,items that may require external assistance. 3.58 Although, Indonesia has a long way to go in fully and effectively implementing fiscal and administrative decentralization, political sentiment will force the government to move fast. This transition has to be managed carefully to minimize disruptions and uncertainties. Hence, Government needs to develop a broad policy strategy - a "white paper" - for its decentralization policy. This should outline the key issues it must grapple with during this transition process. These include:
* * * *
the minimum,requirements of core staff at the central and sub-national levels of government; the core competencies of the staff; measures to ensure local accountability; fiscal risks that could be borne by the government (e.g. from sub-national borrowing
This "white paper" would be a key ingredient of the public in forming their expectations about what is achievable through decentralization and in what realistic time frame. 3.59 The Government must do some basic quantitative research before there is a decentralization plan ready for implementation. There are at least four areas where this work might be focused: a)
Balance. The plans for the decentralization of revenues via the intergovernmental transfer system, the proposed assignment of expenditures, and the proposals for local taxation and user charge autonomy need to be coordinated. It is necessary to estimate the cost implications of the expenditure assignment under consideration, and to cost out the central mandates that will likely be imposed. Following these calculations, the revenue Revenue/expenditure
50
Public Spending in a Time of Change needs can be estimated, and the feasibility of the 25 percent transfer of domestic revenues and the likelihood of a hard budget constraint can be considered. To do this work, a simulation model needs to be developed and From this, the Indonesian estimated. government could begin the hard work of making choicesas regards the proper "vertical split" between different levels of government. b) Intergovernmental Transfers. Before the government can make decisions about the proper "horizontal split" of revenues among local governments, it must have hard data on the budgetary implications of different formulae for the distribution among local governments. Armed with such quantitative analyses and using simulations to reflect different scenarios, the Government can then begin making its decisions about the proper structure of the intergovernmental transfer system. c) Equalization versus Revenue Mobilization. The government ultimately needs to decide how much it wants to emphasize revenue mobilization and how much it wants to emphasize equalization. There are many pieces to this puzzle: the grant formula, conditionality, borrowing versus grants, expenditure assignment, mandates, local taxing and user charge powers, etc. A thorough quantitative analysis of the options and of their implications needs to be carried out. d) Evaluation. A major flaw in decentralization programs around the world is that there is little analysis of the degree to which the existing program is meeting the objectives set for it. Does the program equalize across local governments? Does it stimulate revenue mobilization? Does it lead to different public expenditure mixes? Are hard budget constraints effective? Do higher income places borrow more? Are budgeting practices improved? A modeling and information system is the basis for a rigorous quantitative evaluation system. Indonesia is at the threshold of implementing a new system. Now is the time to build-in an evaluation system to answer these questions. 3.60 Finally, the Government must develop a detailed implementation plan, accompanied by There are many implementing regulations. implementation issues associated with the transition to a decentralized system in a country as large and as complicated as Indonesia, including such things as exact expenditure assignments, revenue powers and limits, budget preparation and constraints,mandates, borrowing authority, and the like. At present, there does not seem to be a detailed implementation plan. It seems clear that all local governments in Indonesia are not able to absorb their new responsibilities because of their very different management capabilities. However, it is quite easy to hide behind this issue and create a self-fulfilling prophecy that leads to the conclusion that "local governments are not able to manage their own affairs". The right answer is probably that there needs to be a transition in which some local governments participate fully in the decentralization under the new system, while some local governments are classified as not yet ready for decentralized authority, but a clear set of criteria for promotion to full status is specified in detail. A series of perhaps ten case studies of local governments, pointed exactly at what needs to be done in these jurisdictions to develop an implementation program, would be quite valuable in this context. An important part of this effort is the establishment of training programs and institutes to enhance civil service capacities.
51
R:EFERENCES
Anipa, Seth, Felix Kaluma and Elizabeth Muggeridge (1999), The MTEF in Malawi and Ghana, DFID Seminar on Best Practice in Public Expenditure Management, June
1999.
Ahmad, E. et. al., Indonesia: Redesigning Intergovernmental Fiscal Relations, Fiscal Affairs Department, IMF, December 1998. Ahmad, E. et. al., Indonesia: Decentralization - Managing Risks, Fiscal Affairs Department, IMF, June 1999. Bardhan, P. (1997). "Corruption and Development: A Review of Issues", Journal of Economic Literature,Vol. XXXV, pp. 1320-1346. Blejer, Mario and Liliana Schumacher (1998). "Central Bank Vulnerability and the Credibility of Commitments: A Value-at-Risk Approach to Currency Crises." IMF Working Paper WP/98/. ../. Washington D.C.: The International Monetary Fund. Caprio Jr., Gerard and Daniela and Klingbiel (1997). "Bank Insolvency: Bad Luck, Bad Policy, or Bad Banking?" In Michael Bruno and Boris Pleskovic, eds., Annual World Bank Conference on Development economics 1996. Washington D.C.: The World Bank. CIDA/World Bank (1998). Controlling Corruption: A Parliamentarian's Handbook, The Parliamentary Centre, 1998 Laurentian Seminar, Kensington, Canada, July-Aug, 1998. COPLANER (1998). "Local Resources and Decisions Improve Education in Indonesia." Planning Bureau, Ministry of Education and Culture, Indonesia. Easterly William (1999), "When is Fiscal Adjustment an Illusion?", Economic Policy Journal, April 1999. Easterly, W., K. Schmidt-Hebbel, and C. Rodriguez (1994). Public Sector Deficits and Macroeconomic Performance. Oxford University Press. Fadia Saadah, Samuel S. Lieberman, Melanie Juwono, and James Finucane (1999). Indonesian Health Expenditures during the Crisis: Have They Been Protected? World Bank, July
1999.
Frida Johansen (1999). "New Indonesia Urban Poverty Project: Harnessing Community Initiative to Generate Jobs", World Bank. Hammer, J., I. Nabi, and J. Cercone (1995). "Distributional Effects of Social Sector Expenditures in Malaysia, 1974 to 1989." In D. van de Walle and K. Nead, eds., Public Spending and the Poor: Theory and Evidence. Baltimore: Johns Hopkins University Press. Heymans, Chris (1996), "Democratization Through Decentralization: Some South African Observations". World Bank (processed). 52
Honohan, Patrick (1998). "Fiscal Contingency Planning for Financial Crises." Washington D.C.: The World Bank, 1998. Huther, Jeff and Anwar Shah (1998). "A Simple Measure of Good Governance and its Application to the Debate on the Appropriate Level of Fiscal Decentralization"" World Bank, Policy Research Working Paper No. 1894,March 1998. IDASA (1999). Transparency and Participation in the Budget Process. The South African Case, Report presented to the Second International Budget Project Conference, Cape Town South Africa, February 1999. International Monetary Fund (1998). "Indonesia: Budget Management in the Short and Medium Term." Fiscal Affairs Department. Kharas, Homi and Deepak Mishra (1999). "Hidden Deficits and Currency Crises", World Bank (mimeo), April 1999. Kaufman, D. G. Mehrez, and S. Schmulkler (1998). "The East Asian Crisis: Was it Expected." Manuscript, World Bank, Washington, D.C. Krafchik, Warren and Joachim Wehner (1998). The Role Of The NCOP In The Legislative Budget Process, IDASA: Budget Information Service presentation to NCOP workshop, South Africa, 6 August 1998. Leechor, Chad (1996). "Reforming Indonesia's Pension System." Policy Research Working Paper 1677. Washington D.C.: The World Bank. Lewis, Christopher and Ashoka Mody (1997). "The Management of Contingent Liabilities: A Risk Management Framework for National Governments." In Timothy Irwin, Michael Klein, Guillermo Perry, and Mateen Thobani, eds., Dealing with Public Risk in Private Infrastructure. World Bank Latin American and Caribbean Studies. Washington D.C.: The World Bank. Organization for Economic Cooperation and Development (1993). "Accounting for What?: The Value of Accrtial Accounting to the Public Sector." Paris. Potter, Barry H., et. al., Indonesia: Fiscal Management, Decentralization and Organization of the Ministry of Finance, Fiscal Affairs Department, IMF, December 1998. Potter, Barry H. et. al., Indonesia: Budget Management in the Short and Medium Term, Fiscal Affairs Department, IMF, October 1998. Sahgal, Vinod (1999). Financial Accountability in Bangladesh. A Governance Perspective, Operations and Evaluation Department (mimeo), World Bank, July 13, 1999. Shah, Anwar (1998). "Balance, Accountability and Responsiveness: Lessons about Decentralization", World Bank, Policy Research WorkingPaper No. 2021, December 1998. Smetters, Kent (1998). "Privatizing Versus Prefunding Social Security in a Stochastic Economy." Washington D.C.: The Congressional Budget Office, Macroeconomics Analysis Division. 53
Stiglitz, Joseph (1999). On Liberty, the Right to Know, and Public Discourse: The Role of Transparency in Public Life, Oxford Amnesty Lecture, Oxford, U.K., January 27, 1999. van de Walle, D. (1998). "Targeting Revisited." The World Bank Research Observer, Vol. 13, No. 2. Vieira da Cunha, Paulo and Phil Brock (1997). "The Trouble with Insurance-Or, Do High Real Interest Rates Cause Bank Crises?" DEC notes, Research Findings 33. Washington D.C. The World Bank. Warren, Ken (1996). "Implementing Accrual Accounting in Government: The New Zealand Experience." The International Federation of Accountants, Public Sector Committee, Occasional Paper 1. World Bank (1999a). "Czech Republic: Dealing with Contingent Liabilities." Europe and Central Asia Regional Office, Poverty Reduction and Economic Management Unit. Washington D.C. World Bank Urban Poverty Project, PAD (May 1999). World Bank (1999). "Indonesia: Development." Expenditures for the Environment and for Sustainable
World Bank (1998). "Indonesia in Crisis: A Macroeconomic Update." World Bank (1998). "Indonesia Public Expenditure Review: The Budget, Off-Budget Items, State-Owned Enterprises." Report No. 18691-IND. World Bank (1997). Indonesia: "Suggested Priorities for Education." World Bank (1999). Indonesia Civil Service Review, Rewards Assessment Report, March 1999. World Bank (1998). Public Expenditure Management Handbook. Washington, D.C. World Bank (1999b). "Republic of Indonesia: Second Policy Reform Support Loan." (Report No. P-7308-IND). Report of the President to the Board of Executive Directors of the World Bank, Washington D.C.
54
ANNEX 1 - Status of Implementation of Recommendations of Public Expenditure Review, 1998. Remarks Recommendation Revenues Short-term Improve collection Audit accounts of Bulog, Pertamina Medium-term Audit tax collection system Current Expenditures Short-term Abolish in-kind rice subsidy for civil service Medium-term Start civil service reforms Subsidies Short-term Food subsidies limited to rice Phase out subsidies on energy and fuel
IMF TA mission on Revenue Mobilization done Ongoing; results expected by August
IMF TA mission on Revenue Mobilization done
Not yet done, but small share of the budget
World Bank assessment of civil service done
Review food and fuel distribution system
Done. Large residential electricity consumers are charged increased tariffs. Actual spending on subsidies was lower than planned because of lower oil price and recovering Rupiah. Arrears are however accumulating. Comprehensive review of Pertamina done by Government through Boston Consultancy Group and PriceWaterhouse-Coopers. Bulog currently being audited. Ongoing
Medium-term Eliminate all subsidies except for very narrowly targeted food and energy subsidies Transfers to regions Short-term Maintain the real value of current transfers Expand untied INPRES
Run expanded job creation program though existing programs Implement plan to expand local tax base Medium-term Develop local fiscal autonomy
Real value more or less maintained in FY99/00. From next fiscal year, transfers will increase due to the recently approved Fiscal Decentralization Law. Done, but tied "block" grants exist. Fiscal decentralization law will provide more autonomy Done. Fiscal decentralization law assigns resource and real estate revenues to subnational govermment Fiscal equalization law no longer requires central clearing of local budgets, and gives increased revenue assignment. Regional autonomy law gives more spending autonomy
Development spending Agriculture Short-term Stop Kalimantan Rice project
Stopped at the basic investment level. No further expansion planned
55
ANNEX I- Status of Implementation of Recommendations of Public Expenditure Review, 1998. Recommendation Remarks Continue irrigation maintenance Problem identified and addressed as an and rehabilitation institutional issue under Bank's WATSAL Postpone major irrigation works No new projects initiated, but bilateral donor pressure in this area Set up water user associations Part of Bank's WATSAL; water user organizations empowered, and local level will retain water user fees
Medium-term
Eliminate crop planting direction
Phase out government sponsored cooperatives
Conflicting informnation.No official policies anymore (abolished by INPRES No 5, 1998), but at local level still perception of obligation, and guidance through irrigation practices INPRES 18 1998 phases out support for these cooperatives,but marketing cooperatives may threaten market operations, and some forestry concessions non-competitively allocated to cooperatives. "People's economy" initiative could to intensify government support for private cooperatives Government increased allocations, but Parliament in part reversed that decision Experimental program of targeted weighing started No new projects started Government reviewed all toll roads and toll roads projects Budget allocation sustained Strategic Expenditure Planning exercise done under tighter budget constraints Part of Strategic Expenditure Planning
Transport
Short-term
Maintain critical roads Enforce loading limits Postpone new construction of airports, ports Reach agreement on the private toll road to ensure continued operation Maintain essential rail links for Jabotabek to supply PLN
Medium-term
Revise transportation plans to take account of changed circumstances Improve road planning and coordination Energy
Short-term
Reduce subsidies Develop short-term workout programs for the financial crisis, esp. PLN Assure PLN resources for maintenance Facilitate renegotiation of critical contracts for PLN (IPPs and fuel supply) Audit Pertamina
See subsidies section above Part of comprehensive review of Pertamina done by Government through Boston Consultancy Group and PriceWaterhouse-Coopers. PLN restructuring supported by ADB Part of comprehensive review of Pertamina done by Government through Boston Consultancy Group and PriceWaterhouse-Coopers Settlement process started. This is part of a comprehensive review of Pertamina.
56
ANNEX 1 - Status of Implementation ofRecommendations of Public Expenditure Review, 1998. Remarks Recommendation Medium-term Comprehensive review of Pertamina done by Revise and implement PLN Government through Boston Consultancy Group privatization plan and Pricewaterhouse-Coopers Comprehensive review of Pertamina. Includes Review medium-term role of separating government role from Pertamina Pertarnina Health Short-term Emergency allocations to ensure supply of essential medicine Assure funds for basic staffing of primary health care units Direct increase in funds through existing programs Enhance basic public health service Institute training programs to enhance personnel responsiveness Medium-term Major review of the sector Done Done Done Allocation to these programs sustained. Some issues in disbursement Assessed by government to be a medium-term priority The World Bank is preparing a Strategy paper in consultation with Goveinment. Government has established a number of teams to review medium-term strategy. Minister has announced a "New Paradigm" for health Done with World Bank support Done with World Bank support Budget allocations sustained
Education Short-term Implement program of supplementary block grants Implement scholarship program Continue efforts to achieve universal primary and lower secondary education Medium-term Reassess education strategy
World Bank did comprehensive education sector review (Dec. 98) in consultation with Government. MOE initiated higher education strategy, and held major strategy meeting with donors and civil society in February 99 Reforestation fund included in budget, but allocations still intransparent. Audit ongoing, results expected in August Done at Kabutaten level, but not reported to center Reporting on funds improved in budget, but still unclear criteria for allocation Private management of funds not initiated to date
Offbudget accounts
Short-term
Increase accountability and transparency Account for cost recovery charges Make investment fund fully transparent Change management of pension fund Medium-term Review structure and rational of investment funds
57
ANNEXI - Status of Implementation of Recommendations of Public Expenditure Review, 1998. Recommendation Remarks Increase transparency IMF TA on fiscal reporting ongoing Reform Pension Fund SOEs Adopt strategy based on Master Plan Privatize SOEs in commercial markets Sell assets in non-core noncompetitive markets after defining regulatory framework Refrain from investing in major restructuring Strategy and Master Plan adopted Part of the master Plan; Privatization receipts of Rup 3trillion in FY98/99 Telecoms bill submitted to Parliament
Done
58
ANNEX 2 - A NATIONAL RISK MANAGEMENT FRAMEWORK FOR CONTINGENT LIABILITIES
containing risks as a precaution; (b) measuring the extent of the liability at the time it is incurred, even though there is no cash outlay at the time, and budgeting for the liability at the time; and (c) monitoring the liabilities over their life span to take additional precautionary measures.
INTRODUCTION
ontingent Liability Management implies a three-fold strategy: (a) mitigating or
1. Contingent liabilities are those liabilities that become due only in certain circumstances, which are often deemed unlikely to occur at the time the obligation is assumed. These liabilities are used by Governments in the pursuit of several objectives. Governments around the world undertake contingent liabilities to protect the banking system, to promote particular business activities (such as infrastructure projects, exports, small and medium firms, and agricultural production), to protect pension rights, and, on occasion, to bail out large corporate entities that are deemed too big to fail. 2. Contingent liabilities are undertaken for some good and other dubious reasons. They are justified where they prevent systemic risk (i.e., where failure of one part of the economic system can cause distress in other parts); where the government has a social obligation to maintain economic activities and incomes above minimum thresholds; and where, absent a track record, the government's regulatory commitments require reinforcement. At the other extreme, contingent liabilities are questionable, and have proven to be expensive, where they are a substitute- for fundamental economic reform. In an environment of poor recordkeeping, lax regulatory regimes, and misaligned prices, the moral hazard, to which inherently contingent liabilities are susceptible, is exacerbated. In between the two extremes is a gay zone where government contingent liabilities can provide a financing mechanism for transitions, as
59
from a planned to a market economy or from public to private ownership.' 3. Even when potentially valuable, contingent liabilities need to be undertaken with care because they can easily reduce fiscal discipline and lead, eventually, to high budgetary costs and disruption. Discipline is lost, in part, because those in receipt of the guarantees have reduced incentives to act diligently. Discipline is also lost because contingent liabilities typically are not valued for the expected payments that may occur and are sometimes not even recorded systematically. As such, because they do not require an immediate cash payment, they become a mechanism for deferring payments into the future. While it is the case that these future payments may never materialize because the risky events may not occur, experience shows that implied and explicit guarantees are often called and, when that happens, they create budgetary stress. 4. For the management of national debt, it is also important to note that contingent liabilities have a non-linear payment schedule. For example, if the extent of government liabilities depends upon the level of demand in the economy and a 10 percent drop in demand increases contingent liabilities by 7 percent, then a further 10 percent drop in demand will increase contingent liabilities by more than (and often
In the Czech Republic, for example,the so-called mandatedto financethe "transfornationinstitutions," transitionto a marketeconomyhave liberal guarantee
authority. See World Bank (1999).
significantly more than) an additional 7 is The reason for this percent. straightforward. The first drop in 10 percent is cushioned in part by the surpluses in the activity supported by the guarantee (for example in the banking sector, the existing capital and loan loss provisions will take the first hit). However, a further drop does not benefit from the cushion and hence translates into a liability for the guarantor, a liability that grows more rapidly as the shock intensifies. It is for this reason that contingent liabilities can be a treacherous policy tool if appropriate bounds are not 2 placed. 5. As mentioned in Chapter 3, an Intregrated Risk Management process should perform three key functions: (i) It should identify the risk exposures due to the explicit and implicit contingent liabilities that are undertaken by government and implement measures to mitigate losses and continually monitor this exposure to risk; (ii) It should measure those risk exposures to the extent possible and budget against expected losses by setting aside some resources (with strict guuidelinesfor their use); and (iii) It should undertake basic measures to reduce the government's exposure by fundamentally reducing risks and also, whenever feasible, by transferring risks to market participants. 6. Identifying and controlling the risks. The government's exposure to loss can arise from a wide variety of events. Attempting to account for every source of exposure is not feasible. A systematic approach to identifying the principal risks, whether in a specific program or for the economy as a whole, is needed to ensure that all relevant exposures can be classified. 7. One approach to risk assessment is that adopted by federal regulators of financial institutions in the United States and Europe.
Non-linearlity also arises because incentives worsen as the slack declines: "If the (bank's) franchise value has been eroded, even minor deteriorations in the balance sheet can encourage risk-taking strategies that may trigger the government guarantee." (Vicira da Cunha and Brock 1997)
2
With limited staff resources, federal regulators have evolved a top-down approach management risk for conducting examinations of financial institutions. Regulators first examine an enterprise's general categories of risk (financial, business, operational, and event risks) then focus their scarce resources on the highest risk areas. This process yields a risk identification lattice. Using a similar approach a government can assess its own risk exposures (Figure 1). The advantage of this top-down approach is that the government can focus resources on those risk categories, classes, or risk types for which exposure is greatest. This approach economizes on scarce resources. A dedicated cell, within the Ministry 8. of Finance's Public Credit Department, should be mandated to record all explicit guarantees that the Government provides, while also tracking and assessing all contingencies that may increase the Government's debt. Systems of early warning indicators to warn of impending crisis, though imperfect, should be instituted. For example, for the financial sector, the rating agency Standard and Poors' recommends that rapid growth of private sector debt ("lending booms") and the likelihood of exchange rate depreciations are useful indicators of possible stress in the financial system. By linking these indicators to specific areas of vulnerability in the financial system, as identified by the topdown analysis implied by the lattice in figure 1, measures to take early preventive action may be possible. In the power sector, the misalignment between the high level of capacity developed and the falling behind of demand was unexpected: but there was also no strong incentive within PLN to be concerned about the downside risk. It would be the task of a specialized cell such as the one proposed to identify the high cost in the (unlikely) event of sharp slow down in demand. Natural disasters provide another
exampleof the importanceof monitoring.
60
Figure 1 Risk Identification Lattice
Govemment riskexpoure Financialrisk Marketrisk

A _______-_-__
_
Policy risk
Oper,rilriSk
J X
Event=isk I Politicalrisk
_
_
Poor policy
design
Systmsfaiiure-_--
_ _
Liquidityrisk Credit risk
polic ] implementation
Legal risk
Natural disaster
In Peru, early warning indictors of El Nino have partly been instituted with significant economic gains in terms of reduced losses to life and property. While a scientific task such as this would not belong in the Ministry of Finance, it would be appropriate for the proposed cell to identify key natural calamities as priority areas and delegate their monitoring to technical specialists-absent such monitoring, the government ends up with the bill to deal with the havoc in the wake of a natural disaster.
9. Budgeting and reserving for contingent liabilities. The next more sophisticated stage is the budgeting and reserving for contingent liabilities. The budgeting system is not merely a matter of good housekeeping but should be part of a strategy to create the incentives for limiting exposure under contingent liabilities. Thus, for example, the recording of contingent liabilities is required in the Czech Republic, but this is clearly not sufficient when complementary measures to ensure that fiscal discipline are not instituted (see Box 1).
Box 1: The Problems with Cash-Budgeting To see how these incentives under a cash accounting system skew decision making, consider the different ways in which the government could help finance a $100 loan to a private infrastructure provider. If the government provides a 10 percent loan subsidy, the cash budget cost would be $10 in year one. If, instead, the government provided the loan directly, the cash budgetary cost in year one would be $100 the full face value of the loan. If the government agreed to guarantee a loan made by a private bank, the budgetary cost of the guarantee would be zero (or negative if a guarantee fee is collected) the first year. Thus, while the economic and financial value of the three different forms of financial assistance are equal, a legislative body would favor the guarantee option.
10. The first step in the budgeting process is the measurement of contingent liabilities, which is essentially a forward-looking
exercise. Estimates are required of the
actuarial and econometric methods have been used (see Lewis and Mody 1997 for a summary). Gaining favor over these methods
is contingent claims analysis, which
probability that the guarantee will be called

and the payment obligation if that happens. But generating these forecasts of the future relies, at least in part, on past data. Various 61
essentially
simulates
many
different
scenarios that may unfold in the future and determines the payments that would be required under those different scenarios.
While even under contingent claims analysis, some past parameters are required to generate the flow of future outcomes, the consideration of different scenarios reduces the dependence of the results on the past parameters. In the context of developing countries, contingent claims analysis has been used to estimate contingent liabilities for infrastructure projects in Colombia (see Lewis and Mody 1997 for a description) and estimation is ongoing for Malaysian infrastructure projects. Also, models estimating value at risk are an example of contingent claims analysis (for an application of value at risk models to estimating contingent liabilities of developing country central banks, see Blejer and Schumacher 1998). 11. From contingent claims analysis are , ., , . . derived two basic concepts that underlie the management of contingent liabilities. These are the expected costs and the unexpected costs of the contingent liabilities. Expected costs are the average costs (or payments made under the guarantee) over the different scenarios modeled in the contingent claims analysis. Expected costs are thus a measure of the likely payments that will need to be made. As such, the expected costs are also a measure of the subsidy implied by the issuance of the guarantee. From a policy issuanc guarantee. From a policy of the point of view, the expected cost estimate can be used to judge whether the government judg to be usedwhethe the goernment would be willing to support the project through an equivalent cash, up front subsidy. 12. However, a guarantee is not Just a subsidy, and hence the importance of the subsidand ccep,namel, thexmponcted ofsth second concept, namely, unexpected costs. Under certain scenarios, the payment under the guarantee can be much larger than the expected costs. For this reason, risk analysts determine the payments likely in the most adverse circumstances. It is common, for example, to determine the level of payments that would be required at the 95 percentile, i.e., the highest payment in all but the worst five percent of the conditions. This estimate is referred to as the unexpected cost. The unrexpeted coasth couldealeotbe estimaTed a the 99 percentile or any other confidence
level depending on the risk aversion of the guarantor (as discussed below). 13. Having estimated the expected and unexpected costs, each of these needs to be dealt with in its own way. Expected costs, being akin to subsidies, need to be directly budgeted for, and thus be included in the appropriations process. Unexpected costs, in contrast, are dealt with by setting aside capital to deal with the extreme events. 14. Budgeting for expected costs requires accrual or present value accounting, which, in turn, implies a medium-term multiyear budget framework (see OECD 1993). In contrast, most government bodies use a simple cash-based system of budgeting. Under a cash-based system of budgeting, a goeneteutsth equates ugtr oto ~~~~~government the budgetary cost of ossutag financcal assbstance with the cash curren bdet year. Ts, wn a current budget year. Thus, when a g face value of the loan at the date the loan is issued is recorded as a budgetary cost, with loan repayments recorded as cash inflows in subsequent years. Simple cash-based budgeting thus treats the disbursement of a direct loan as a rant equal to the entire face value of the loan, with subsequent repayments representing offsetting receipts for the goverment. Loan guarantee and xrtegvrmn.La urne n insurance programs are not recorded as costs in a simple cash-based budget until a claim is made at some future uncertain date. In fact, since any premium revenue from a government insurance program is recorded up front in exchange for the insurance policy while claims are not recorded until some uncertain date in the future, a simple cashbased budget ma record an insurance g y prga asantevue a budgetaryai.Ts inconsistency creates incentive for policymakers to raise premiums rather than reduce the likelihood or severity of claims insured. Cash-based budgeting thus misrepresents and masks the agegate epsre associand man guaggregand exposure associated with loan guarantees and government insurance programs and creates perverse incentives for selecting one form of financing assistance over another (see Box
1).
62
15. In contrast, the use of a present value system requires the recording of the expense and appropriating against it the expected cost
of the action at the time the transaction is undertaken.
Box 2: The United States Federal Credit Reform Act of 1990 Prompted by the explosion of loan guarantees issued during the 1980s and a recognition of biases created by a simple cash-based system of budgeting, the United States changed the budgetary treatment of direct loans, loan guarantees, and grants in 1992. Under the new budgeting system created by the Federal Credit Reform Act of 1990, each of these forms of credit was valued using a financially equivalent metric-the expected present value of future costs. The budgetary cost of credit is defined as the present value, discounted at Treasury interest rates of comparable maturity, of the expected cash outflows from the government minus the expected cash inflows to the government. The shortfall between borrower fees, repayments, and interest and the amount needed to cover the principal of the loan and the Treasury's cost of borrowing represents a cost to the government. Likewise the difference between the fees borrowers pay to the government and the cost of guaranteed loan defaults (and/or interest subsidies) represents a cost. When agencies seek budget resources (budget authority and budget appropriations) to carry out a credit program in the budget process, they must estimate and request the full expected present value of future costs-including default, interest, and other costs-associated with loan guarantees or direct loans to be issued in the forthcoming budget year. Funding to cover the expected present value of future costs is charged against the appropriation for an agency when the direct loan or loan guarantee is issued and the government's cornmitment is extended. These costs, or subsidies, must compete for budgetary resources on the same basis as other government spending. Credit reform requires more careful record keeping than a simple cash budget. Agencies must identify loans or classes of loans by the appropriation used to fund the transaction, their maturity and date of origination, and their subsequent cash outflows and inflows. In addition, programs are required to develop risk categories based on the characteristics that determine the likelihood of default and other costs. These records are used to re-estimate the value of the subsidies provided for loans or loan guarantees, adjust ex post budgetary expenditures relative to ex ante expectations, and improve the subsidy calculations for new loans or guarantees. This tracking also helps agencies underwrite, service, and control losses on loans or guarantees. The Federal Credit Reform Act does have shortcomings, which provide useful guidance for future budgetary reforms in the United States and elsewhere. First, coverage of unexpected losses is not included as part of the cost of a program. This failure to incorporate some measure of unexpected loss represents a serious shortcoming given that most loss distributions associated with central government guarantees are asymmetrically skewed against the government. Second, incentives remain to use "cheap" insurance structures to cover loan guarantees. Government insurance programs are financially equivalent to guarantee programs and should be treated in a consistent budgetary framework. Third, program agencies must make substantial investments in new information systems technology. In the United States, new investment in information systems placed a strain on many of these agencies. Governments adopting credit reform must recognize at the outset that funds need to be available for this investment. Finally, credit reform requires that agencies re-estimate the subsidy costs of their programs on a regular basis so that the government's exposure can be recalculated and appropriate funding is set aside to cover future costs. Appropriate discipline is required to ensure that agencies do not underestimate subsidy costs with the knowledge that any shortfall will be made up in someone else's watch.
Under this system, then, all actions receive equivalent accounting treatment and thus the choice between various forms of government support can be made on the basis of their intrinsic merits rather than being driven by accounting conventions. Use of a present value system need not affect or distort cashbased estimates of the government's fiscal 63
deficit, since the effect on the deficit is not recorded until actual cash payments are disbursed from the reserve fund. Adoption of a present value method of guarantee budgeting simply forces agencies to set aside resources up front for the expected costs of the guarantee issued (see Box 2 for the recent U.S. experience where certain positive steps
have been taken but shortcomings 3 remain). 16. In addition to budgeting for the full expected present value of costs from credit and insurance programs governments need to set aside reserves against unexpected losses. Preparing for unexpected losses prevents the political backlash associated with redirecting scarce public resources to cover the sudden increase in costs, obviates the need for political battles over additional funding, and eliminates the perception that any sudden increase in costs represents Setting up program mismanagement. reserves to protect against such events can mitigate these problems by reducing the number of events for which the executive branch or administering agency needs to seek additional budgetary resources to cover program costs and by reducing the size of any budgetary requests that are made. Because the United States government did not reserve against unexpected losses, it
incurred high political costs as a result of the $130 billion in losses charged to the United States taxpayer during the thrift crisis of the 1980s. 17. Once a government can assess its risk tolerances and goals, in terms of both which risks and the level of loss it is willing to bear, it can establish reserves against unexpected losses ("risk capital") within its credit and insurance programs. To do so, however, a government Teeds to determine include whether reserves will be set based on the additive unexpected loss exposure of each guarantee or on a portfolio value-at-risk (VaR) approach to account for portfolio diversification, what the investment policy of the reserves will be once they are established, and where the reserves should reside (for a more detailed discussion, see Lewis and Mody 1997).
from the experience in the United States, New Zealand has implemented a similar budgeting approach. Their program covers all contingent liability exposures (including insurance), and the govemment has published a present value budget for both contingentand non-contingent expenditure and revenue flows. (See, for example, Warren 1996.)
3Learning
64
ANNEX 3: OECD EXPERIENCE IN PUBLIC EXPENDITURE MANAGEMENT REFORMS'
ecentreformsin OECD public expenditure management reactedto risingfiscal imbalances and
lagging public sector performance. The reforms have sought to improve macroeconomic and microeconomicaspects of budgeting by better linking policy, planning and budgeting within a multi-year budget framework. Hard and predictable budget constraints, and greater managerial autonomy matched with more accountability for results have been decisive for better public sector performance. The Background of OECD Reforms 1. Growing Governments. OECD reforms began against a background of concem over public expenditure growth, rising deficits and debt burdens, and a general disappointnent with govemment performance. Governments had grown rapidly after World War II, backed by the belief that government could solve many of the macroeconomic ills of the prewar decade, and needed to mend many of the market's perceived failures. By the start of the 1980s OECD governments spent over 40 percent of GDP on average, more than double the share of before the war (Table 1). 2. Disappointing Results. Much of the growth in expenditures resulted from social security, interest on rising debt levels and from govemment consumption, driven by rapidly expanding payrolls. Some therefore concluded that this expenditure growth contributed little to welfare (Schuknecht and Tanzi, 1996). The critics argued that increasing expenditure share did not "buy" more in terms of
Table1: TheGrowth Government of Expenditure, 1870-1990 (percentof GDP)

Pre-World War 11 (about 1870)/a (about 1913)/a (about 1920)/a (about 1937)/a
century War I War I Austria Belgium Canada ... ... ... ...
...
Later 19th
Pre-World
Post-World
Post-World War II (1960)

35.7 30.3 28.6
(1980)
48.1 58.6 38.8
(1990)
48.6 54.8 46.0
...
14.7 ... 13.3
15.2 21.8 18.6
France Germany Italy Japan Netherlands

Norway Spain
12.6 10.0 11.9 8.8 9.1

3.7 ...
17.0 14.8 11.1 8.3 9.0 8.3

8.3
27.6 25.0 22.5 14.8 13.5

13.7 9.3
29.0 42.4 24.5 25.4 19.0

... 18.4
34.6 32.4 30.1 17.5 33.7

29.9 18.8
46.1 47.9 41.9 32.0 55.2

37.5 32.2
49.8 45.1 53.2 31.7 54.0

53.8 42.0
Sweden Switzerland United Kingdom United States

Australia Ireland New Zealand /d
5.7 ... 9.4 3.9

... ... ...
6.3 2.7 12.7 1.8

... ... ...
8.1 4.6 26.2 7.0

...
10.4 6.1 30.0 8.6

...
31.0 17.2 32.2 27.0

21.2 28.0 26.9
60.1 32.8 43.0 31.8

31.6 48.9 38.1
59.1 33.5 39.9 33.3

34.7 41.2 41.3
... ...
... ...
Total Average /a /b /c /d
8.3
9.1
15.4
20.7
27.9
42.6
44.8
Or nearest available year after 1870, before 1913, after 1920 and before 1937. 1992. Average; computed without Germany, Japan and Spain (all undergoing war or war preparations at this time). GFS data, data available for 1960 is 1970, central govemment data only. Source: Schuhknecht and Tanzi (1996). Adapted from: China: Managing Public Expenditures for Better Results, World Bank, July 1998.
65
better outcomes of government policy: most of the economic and social indicators seem unrelated to government size. However,: "Big Governments" perform better in income distribution and containing crime (Table 2).
Higher spending on social security in these countries may also have contributed to lower crime rates in countries with larger governments.
Table 2: The Size and Composition of Government Expenditures and Government Performance Indicators in Different Country Groups
Industrialized Countries "Big" "Medium-sized" Governments /a Governments /b 1960 1990 1960 1990 31.0 55.1 29.3 44.9 2.6 4.5 13.5 ... 3.2 23.4 1.7 2.9 4.9 ...
...
Total Expenditures (percent of GDP) of which Health Education Social Security Research and development Economic and Regulatory Efficiency Indicators Real GDP growth (in percent) /e Gross fixed capital formation (in percent of GDP) Inflation (in percent) Unemploymentrate (in percent) Size of shadow economy (in percent of GDP)ff Patents/10,000 population (inventiveness coefficient) Social Indicators Rankin UNhuman developmentUg Income share oflowest 40 percent Illiterate population as percent of population 15+ Secondary school enrollment (in percent) Life expectancy Infant mortality/'000 births Prisoners/100,000 people Divorces (in percent of marriages contracted, 1987-91) Emigration (in percent of total population)/j
Newly "Small" industrialized Govemments /c countries /d 1960 1990 1990 23.0 34.6 18.2 2.3 3.4 6.2 ... 4.6 19.6 2.3 2.7 3.5
...
6.6 6.4 19.5 2.0 2.6 20.5 5.4 6.1 11.1 2.0 11.0 24.1 2.9 93.0 77.0 6.7 38.0 33.0 0.2
3.0 2.9 9.6 ... 4.0 21.1 1.6 4.6 3.8 ... ... 16.4 13.3 51.0 70.0 29.0 ... ... 0.3
5.9 5.6 13.9 1.6 3.3 21.3 4.3 9.2 8.2 2.3 13.0 21.6 4.6 99.0 77.0 7.1 68.0 33.0 0.8
5.2 5.0 7.9 2.0 3.3 20.7 6.1 4.2 6.2 8.6 6.0 20.8 0.5/h 89.0 77.0 6.4 154.0/i 36.0 0.1
3.3 3.4 1.0 ... 6.2 31.2 15.3 2.9
15.6 9.3 55.0 72.0 23.0 ... ... 0.6
... 17.4 2.2 61.0 71.0 22.4 ... ... 0.2
... 17.0 9.2 81.0 74.0 9.8 ... ... 0.1
/a Belgium, Italy, Netherlands, Norway, Sweden (public expenditure inorethan50percentof GDPin 1990)./b Austnia, Canada, France,Gennany, Ireland, NewZealand, Spain(public expenditure between and50 percentof GDPin 1990). 40 /cAustralia, Japan,Switzerland, UnitedKingdom, UnitedStates (public expenditure than percentof GDPin 1990)./d less 40 Chile,HongKong, Korea,Singapore; 1990or nearestavailable year./eAverage preceding of fiveyears,1956-60 1986-90./lMost or recentdataavailable 1978. is used in 1990column. 1l992. USonly. Othersbelow5 percent,UNESCO /h statistics 1991. Excluding for /i UnitedStates,average 64./jData available 1960is 1970,datafor 1990mayinclude1993in somecountries.Newly is for industrialized countries onlyKoreais available data, (1993). Source: Schuhknecht and Tanzi (1996).
3. In addition, part of the increase in government expenditures can be explained by the lagging productivity inherent in government services, which automatically increases the 2 government's share as income rises. Nevertheless, the growing perception by the end of the 1970s was that government was the problem rather than the solution. 4. Ineffective budgetary procedures. Budgetary procedures and practices in the 1970s contributed to expanding governments,
2 This so-called Baumol effect argues that government
and reinforced the perception that government was incapable of delivering results. Despite earlier reforms (Box 1), at the outset of the most recent reforms. OECD budgeting systems had all or most of the following characteristics:
*
Primary focus on inputs, with performance judged largely by how closely spending matched budget appropriations. Short-term horizon for budget decisionmaking which failed to adequately account for longer-term costs and led to biased choices of policy instruments, for example between capital and current spending and between spending and regulating.
would need an increasingshare of GDP if productivity

increases in government services lags behind that of other
sectors of the economy.
66
Box 1: Earlier Budget Reforms in the OECD Budgetary procedures had evolved from the end of the last century: * Line item budgeting established previously lacking executive control over spending by ministries and agencies. In line item budgets, expenditures are listed according to objects of expenditures or line items, and focused on expenditure control, with the ministry of finance acting as controller. Line item budgets were not compatible with the expanding role of govermnent. They gave no information on what the money was spent on or whether the money was used cost-efficiently or effectively. * Performance budgeting introduced measures of workload and cost of activities into the budget. This increased the focus on cost-efficiency, but still lacked effectiveness information. * Program budgeting introduced focus on tradeoffs among competing goals. While performance budgeting focused on achieving a given goal at least cost, program budgeting treated the goals themselves as variables. Program budgeting requires the budget to be organized in groups of activities towards a certain goal, a program, and in addition required effectiveness and outcome measures. Program budgeting failed to become the main budget tool in part because tradeoffs among competing goals could not be captured in the process, in part because the information required to make the tradeoffs was often not available because information systems did not support the budgeting technique. * Zero Based Budgeting (ZBB), introduced in the 1970s, focused on the process of budgeting rather than the contents. In a pure ZBB system, all programns evaluated from scratch each year. This pure form are has never been introduced, but many govemrnents have used the principles of ZBB, by requiring ministries to propose budgets with only 90 or 80 percent of the existing allocation. The failure of performance, program, and zero-based budgeting to live up to their expectations was, in OECD countries, partially due to the high degree of centralization of the budget process that went along with them. As a consequence, the central ministries became overwhelmed with the paperwork involved in annual budgeting. In developing countries that introduced these or similar budget systems, the failure to address the generally disabling environment for performance in the public sector was key to the limited success. * Bottom-up approach to budgeting, which created strong upward pressure on expenditures. The arbitrary expenditure implementation that cuts in budget followed undermined program and agency level performance. Focus on distributing the fiscal increment across new spending proposals, and an extensive and detailed debate on existing spending, but little discussion of new and existingpolicies. Few incentives for agencies to save budgetary resources, because current year spending was the starting point for the annual budget haggle. Little clarity of purpose, task, and costs of policies, programs and services, and an intertwining in a single agency of policy advice, regulation, service delivery and funding. Recent reforms 5. The OECD countries have adopted varying degrees of reforms over recent years. Some, such as New Zealand, the UK, and Australia, have adopted radical new approaches management and public to expenditure administration. Others, such as Germany and Japan, have stayed much closer to the traditional, strongly bureaucratic model which once dominated budgetary management in OECD countries. These differences are not surprising: budgetary institutions and organizations are deeply embedded in history and culture, and are not easily transferred from one country to the other. Despite the differences, common strands in reforms exist. Most prominently among them is the focus on achieving aggregate fiscal discipline, funding strategic prionties, and delivering services efficiently. Whereas earlier reforms had addressed each of these individually, more recent reforms have taken a more integrated approach, and have recognized that these three levels of performnance are inseparable.
In sum, the budget processes in OECD countries created a disabling environment for performance in the public sector.
67
6. *
The OECD reforms aimed broadly to: better link policy, planning and budgeting by: (i) increasing the scope of the budget to include expenditures for all government activities; (ii) adopting global budgetary targets; (iii) introducing a multi-year perspective to policy and budgeting; and (iv) focusing on strategic policy decisions at the center of government, while delegating implementation to implementing agencies; improve performance by: (i) enforcing hard budget constraints of a comprehensive budget; (ii) creating agencies with more focused tasks; (iii) devolving managerial authority supported by better specification and measurement of performance; (iv) improving information systems and audit; and (v) introducing market mechanisms in the public sector. The main features of
some OECD public expenditure
budget. However, some--such as the UK and France--consider SOE borrowing during budget preparations, and report on public sector deficit and borrowing requirement. 9. Many OECD countries have increased budget coverage to including tax expenditures and loan guarantees. These are reported in the budget presentation, or in mandatory regular reports to parliament, such as the "subsidy report" in Germany which is issued every two years. New Zealand has the fullest treatment of "fiscal risks," including contingent liabilities-claims against the government which will arise if a specific event occurs--and any event which if it occurred, would affect the revenues of the 3 State. In some countries, including the United States there is a regulatory budget so that the costs on the community of regulations are made public.
Global budgetary targets
management systems are summarized in Tables 3 an4.

LINKING POLICY, PLANNING,AND BUDGETING
The scope of the budget 7.. Ofcpudget opEri v crippled financ disci. pInein variocuies in . th 197. iabliiesin Ipliit ocil scurty systems and loan guarantees for (usually) SOEs often became explicit during economic oftessinscam explici dup rin economic recessions that also suppressed government revenues. Offbudget funds also undermined efficient service delivery, as agencies had recourse to alternate means of finance if they ran out of budgetary funds. 8. OECD budgets now include almost all expenses of government departments and their agencies, whatever their source of funding. Exceptions arise only when a law treats a particular agency as off-budget (Chapter 3).
10. Since the later 1970s and early 1980s most OECD countries have adopted global budget targets in response to growing awareness of fiscal problems arising from lack of fiscal restraint. Many countries express these targets on a multi-year time horizon. Targets are usually either ratios (of expenditure, budget balance or debt to a national aggregate such as GDP); a rate of growth for spending (real or nominal); or an absolute value for a target variable (such as cash limits on spending or the size of the deficit). Some sedn rtesz ftedfct.Sm countries, such as the Netherlands, combined deficit targets with limits on Government's
4
share of GDP. 11. Global targets were questioned in the 1990s because they were thought to make the
3
to a fullaccrualsapproachto budgetingd
OECD country performnance relation to targets in has varied substantially due to fiscal shocks (e.g. German reunification), economic recession or change in political direction. However, fiscal balances in the OECD did improve in the decade after 1979, only to deteriorate in the economic slowdown of the early 1990s[Table from OECD] More recently, the European Union members have seen marked improvements in their fiscal balances after adoption of the Maastricht treaty. This treaty restricts membership of a single currency to countries with fiscal deficits below 3 percent of GDP.
4Actual
Social Social security funds may be separately security funds may be separately
managed, but they are normally included in the budget reports, and social security policy is included in annual budget deliberations. Most countries treat state enterprises only on a net basis, reporting taxes and subsidies in the
68
budget more pro-cyclical. Concerns were also raised that an overzealous commitment to targets would lead to the very across-the-board expenditure cuts that budgetary reforms were designed to avoid. Finally, targets often invited creative bookkeeping on the part of government: in the drive to meet the Maastricht criteria several countries have performed such tricks to reach the deficit target of 3 percent. One of the most creative ones came from a government that sold a state enterprise to a state bank, and counted the revenues in its budget. 12. More recent approaches such as New Zealand's Financial Responsibility Act and Australia's Budget Honesty Act define global targets over the medium-term, and in terms that allow for the business cycle and external shocks. The strength of these approaches is the transparency associated with the regular publication of the medium-term fiscal implications of the government's policies. A medium-term approach to budgeting 13. Global targets, whatever their form, are translated into annual budgets via their interaction with some form of baseline estimates of the cost of government policy over the medium-term. In OECD countries in the 1960s and 1970s multi-year budgets were used as planning tools, sometimes in the hands of planning agencies with no responsibility for fiscal management. The total cost of the policies in the plan became the "global target," which was rarely constrained by a realistic assessment of what was affordable in aggregate. The plans and targets raised expectations which could rarely be met, and which lead to many of the dysfunctional budgeting practices referred to above. 14. Beginning in the 1980s, more and more OECD countries adopted a medium-term expenditure framework (MTEF) in some form or another. A MTEF distinguishes between the medium-term costs of existing policies, and a medium-term perspective on the level of sustainable spending. The costs of ongoing policy are determined by multi-year cost forecasts based on a detailed breakdown of expenditures for individual policies. The sustainable spending level is determined by 69
macroeconomic conditions, revenue projections, and the sustainable fiscal deficit. The difference between sustainable spending level and the costs of ongoing projects is available for new policies. If this difference is too small, governments have to take decisions to drop existing policies, to raise additional revenues, or to refrain from starting new policies. The MTEF provides policymakers with 15. the information needed to make decisions on new policies that could be accommodated within the expected resource envelope. The medium-term estimates are linked explicitly to annual budgeting so that annual budget limits are firmly based on the medium-term implications of government policy. 16. Australia's experience with the MTEF shows its potential power to regain fiscal discipline. Introduced in the early eighties, the early projections showed an unsustainable trend in expenditures to finance existing policies. These projections forced policymakers to make the choices necessary to reduce expenditure projections down to realistic medium-term targets (Figure 1). The forward estimates of the costs of ongoing policy came down from an unsustainable 15 percent in 1994 to virtually nil in 1990. Unlike earlier across-the-board expenditure cuts, the expenditure reductions were driven by decisions about dropping specific policies. Recently, several developing countries have adopted a MTEF, including South Africa (Box 2). Strategic policy making at the center of government 17. In a medium-term approach, the challenge for policymakers is to reconcile what is affordable with what is required. What is affordable comes "top down" whereas what is required comes "bottom up." Policymakers need to reconcile the two by considering policy decisions together with their spending implications, rather than looking at policy and spending in isolation. Budgeting must therefore become focused on policy. The MTEF gave policymakers the tools to determine what
Figure I . Australia: Budget and Forward Estimates, Cumulative Real Growth
MA, U4
/
I-/
.- M-y&7
/...~
~10
1982E3
19ff3
1984~ 198584 9S 86
1986. 87
198798
1988 99
1989. 90
199091
199192
17992 1993. 92 94
199495
199596a
policies were affordable. Together with better incentives for public sector organization to spend money efficiently (see below), this allowed policymakers to disengage from the day-to-day operations of government, and to focus on policy. 18. Over the 1980s and 1990s ministers increasingly took control over strategic decision mnakingand, disengaged themselves from dayto-day operational issues. Policymakers moved away from the sequential policy making outside the budget process, which had been an important source of macro instability and poor performance at the strategic and operational levels in the 1970s. Cabinets, often via a powerful sub-committee of key ministers, now focus on the tradeoffs between priority sectors and programs, within the macro framework discussed above. Governments of all OECD countries enforce--at least in principle-collective decisions within the executive. 19. Line ministries bought into the new budget processes because they gained much autonomy in making the policy tradeoffs within their portfolio, as long as they stayed within their budget limits. They no longer needed to negotiate with the Finance Ministry for additional funding, but could concentrate on designing new policies. And they had an
incentive to reconsider existing policies, as savings could be used for new policy. 20. Evaluation. Policy discussions were supported by improved policy evaluation. The increasing interest in outcomes has seen a resurgence of comprehensive approaches to evaluation, the results of which fed into the policy discussion. Canada has had an evaluation system in place for many years. Australia has made evaluation an integral part of its budgeting process. 21. Australia's approach is of particular interest because of its attention to the incentives for evaluation and the associated question of how to link evaluation results into the budget. In the Netherlands the reconsideration process, which involves annual policy reviews, has been in place since 1981 and has played an important role in budgetary control (Box 2). 22. Transparency. OECD Governments differ on how systematic and transparent their strategic decision making process is. Some countries prepare and publish an economic and social strategy which sets out medium-term policy objectives expressed in terms of the desired social and economic impacts of policies, their expected fiscal costs and benefits,
70
Box 2: Excerpts from The 1997 Medium-term
Budget Policy Statement in South Africa
Relationship with the budget Govemment's spending plans for the next three years will be published in March 1998. These plans will give substance to Govemment's reconstruction and development commitments, within an overall level of spending that the naTioncan afford. These are the most important choices any govemment can make. The Budget must reflect Government's social and economic priorities, and its expenditure plans define the nature and scale of the Govemment's ambitions for the nation. What is the Budget Policy Statement? This Medium-term Budget Policy Statement sets out the policy framework for the coming budget. It describes Govemment's goals and objectives. It explains the economic environment within which those objectives are being addressed, and projects the total level of resources that will be available. The Policy Statement analyses the trade-offs and choices that the nation confronts in addressing its reconstruction and development priorities. The Medium-term Budget Policy Statement is an important step forward in the budget process. In keeping with our commitment to open, transparent and cooperative policy-making, it invites the nation to share with Govemment the important choices that must be made. The Medium-term Expenditure Framework Medium-term budgets.The key features of the new medium-term budgeting system are: * publication of three-year forward estimates on Budget Day, consistent with Government's policy priorities and * commitments; * detailed analysis of the policy implications of budget projections; * cooperative teams, composed of national and provincial treasuries and line departments, analyzing key sectors and * reporting to Cabinet and Executive Councils; * quantified, analyzed policy options presented to political office-bearers for decision; and * the publication of a Medium-term Budget Policy Statement, to enable Parliament and the institutions of civil society to * participate meaningfully in the debate. Rolling budgets The Medium-term Expenditure Framework initiates a process of rolling three-year budgets. The MTEF projections published on Budget Day will be considered again in the course of 1998, and revised according to new information and policy priorities. The three-year allocations will represent the starting point for that process, and departments will therefore have agreed spending trajectories within which to plan. Departments will be expected in future to frame their policy proposals within their three-year allocations. The introduction of resource-based planning represents a significant change in theplanning environment and will initiate a major process of program reprioritization and redesign within spending departments. Public debate of plans As in previous years, Parliament will be asked to vote only on the Budget allocations for the coming year, and not on all three years of the spending projections. But the detailed three-year spending plans will provide an opportunity for Parliament, the National Economic Development and Labor Council (NEDLAC) and civil society to evaluate the Budget allocations for the year immediately ahead in the context of the medium-term evolution of Government's expenditure priorities. Parliament will be invited to debate the expenditure plans, and to ensure that they reflect national goals and priorities. This process therefore provides new opportunities for all stakeholders to analyz and discuss the expenditure projections, and to ensure that altemative views are fully taken into account in framing the subsequent MTEF. The MTEF and budget reform Overhaul of the budget process The publication of the Medium-term Budget Policy Statement, and the publication in March of threeyear budget projections, are first steps in a wider overhaul of the budgetary process, emphasizing transparency, output-driven program budgeting and political prioritization. The MTEF provides the bridge between the technical preparation of budgets and the need to reflect political priorities in expenditure plans. A brief account of the new budget process is set out at the end of this chapter. Further steps in budget reform will be taken in 1998.These will include a transition to greater devolution of managerial autonomy, within a framework of improved incentives and greater accountability, accompanied by reforms of financial management. Provincial and local government The policy goals of Govemment will be reflected not only in the national budget, but also in the budgets of provincial and local govemment. Provincial and local govemment will receive their equitable share of nationally raised revenues as well as other transfers. However, they have the responsibility of developing their own budgets, within expenditure allocations consistent with the nation's policy priorities. The national government does not control the details of these budgets, but can influence them indirectly through agreed policies and framework legislation setting norms and standards. The introduction of three-year budgets and their consolidation into resource envelopes for the major provincial services is an important step in the evolution of the institutional framework for intergovemmental policy making and budget planning. The intergovemmental forums of the spending departments will, for the first time, have expenditure projections within which to develop and refine the norms and standards for service delivery. Role ofAccountingOfficers Once Parliament and provincial legislatures have approved their budget proposals, departments must adjus their expenditure to ensure that they stay within their budgets. It is the responsibility of Accounting Officers, appointed by political office-bearers, to ensure that allocations are applied to their intended purposes and spending limits are respected. This principle will be strictly enforced by the proposed Treasury Control Bill which is due to be introduced in Parliament next year, replacing the ten Exchequer Acts which govem provincial and national financial management.
Source: http://www.polity.org.za/govdocs/policy/mediumbudgethtmt
and allocation of implementation responsibility to ministers and agencies. Others are less thorough. All, however, have some process for announcing their policy intentions and their relationship to budgets, and most use the budget speech. Openness and transparency in decisionmaking increases the incentives for policymakers to deliver upon their promises, or pay a political price if they fail to do so.
Budgets in OECD countries are widely published--more and more often on the Internett --and openly discussed in parliament. 23. The role of the legislature. Decision processes in OECD countries are influenced by the nature of representative government. A
http://www.oecd.org/puma/mgmtres/budget/budlinks.htm provides links to MOF and budget links of most OECD countries.
71
Box 3: The Netherlands Reconsideration Procedure
The ReconsiderationProcedure has been the main instrument of policy review in the Netherlands.It was introducedin 1981,and reviewsexistingpoliciesin the public sectorwith the aim to developmore cost effectivealternatives, to abolishthe policy. Policy areas for or reconsideration decided by the Cabinet. The departmentsresponsiblefor the areas of are reviewneed to comeup with savingsoptions,includingone that would reduceoverall funding with 20 percent. The Sectoraldepartment use outsidereviewersin the team to develop can policy alternatives. The reports are made public by means of submissionto Parliament. Savings proposalsare reviewedin the normalbudgetcycle,but reconsideration be initiated can at any point in time. The reconsideration procedure is guided by a small ministerial conmiission by the prime minister,the vice premiers,and the ministerof fmance. MOF led providesadviceandadministrative supportforthe reconsideration.
system with significant distributed power such as the United States emphasizes processes for negotiation between executive and legislature. Others may require mechanisms for compromise because they have minority or coalition executives. The degree to which legislatures changes budget proposals of the executive varies widely in the OECD. In the US, Congress makes many significant changes, whereas in the UK Parliament invariably approves what the government proposes, because the proposal is that of the ruling majority. However, legislatures take an increasingly active role in budgetary policy.
IMPROVING INCENTIVES FORPERFORMANCE
Increased autonomy 25. Managers in line ministries and agencies obtained increasing flexibility in using money to achieve policy objectives. Detailed central controls on line items or expenditure norms have been replaced by more aggregate controls, with greater flexibility at local level to achieve agreed objectives within more global expenditure allocations. To this end, many OECD countries relaxed rules of virement and carryover. The medium-term approach to budgeting gave line agencies greater predictability of funding, which allowed better planning and allocation by public sector
rlmanagers. At the same time, aggregate

spending limits were strictly enforced
by
24. The focus on strategic decisions at the center of government went hand in hand with major, and sometimes radical, reforms in the incentive structure for public sector units, broadly aiming at increased accountability for performance. The reforms granted more autonomy to government agencies and managers in implementing policies, clarified roles and responsibilities of organizations and government employees, specified the required perfornance, and improved monitored of this performance.
improved treasury management systems. 26. Output budgeting. Most OECD governments still defie budgets and control in terms of line items or programs. In contrast, New Zealand and the United Kingdom seek to hold organizations accountable for results in terms ofthe outputs they produce. While these countries acknowledge that outcomes are central, they reason that it is not possible to hold agencies accountable for ultimate impacts, because many other factors affect these outcomes, most of them outside the agency's control.
72
Box 4: Outcomes, Outputs,Programs, Inputs

Increased performance orientation required improved performance definition and measurement. Performance has various dimensions, the usefulness of which depends on their purpose. For policies, and therefore for policy-making budget units outcomes are the most relevant indicators; for service delivery units outputs are usually more appropriate. Outcomes indicate the ultimate goals of Government strategies and policies. For example, education policies could aim for higher productivity and income for people; health policies could aim for higher life expectancy. However, productivity of people and life expectancy of people depends on much more than education and'health policies alone. Policy research can narrow down what effect these policies have on the outcomes, but often not eliminate uncertainty on policy outcomes. The term program is used to describe all the policies which a government has put in place to achieve a desired outcome. Outputs specify the operational results of departments or other agents of government. For example, education could target 9 year compulsory education, and health policies could aim for full immunization of 5 year olds. These indicators are more specific, they often requires quality standards to link the outputs with outcomes. A quality standard for education could be the percentage literacy and numeracy of 12 year olds; and for immunization the range of diseases inoculated for. Significant policy outcomes may be the product of several outputs together with direct budgetary payments and other policy measures such as regulation. Input indicators show resources used in achieving outputs and outcomes. Traditionally, government accounts only registered cash outlays, but increasingly full costs are measured by including accrual elements in the government's accounting system. Other input indicators relate to actions taken to achieve certain goals such as the purchase of textbooks, the construction of hospitals, and the hiring of teachers and doctors.
27. New Zealand went furthest in this approach, and abolished input budgeting altogether, to be replaced by output budgets and appropriation estimates. Within the limits of the appropriations approved by parliament, departmental managers have much freedom to allocate resources the best way possible to achieve the budgeted outputs.
28. Other countries without an outputbased budget, have aspects of performance measurement in their accountability regimes. In Australia, resources are allocated by programs. Budget proposals for programs include a variety of performance indicators which establish efficiency and effectiveness.
Box 5: Examples of output definitions and costs in New Zealand
Provider: Output title

Description
Police Custodial services

Provision of jailing services at police stations and courts for people under arrest or the
Costs Provider Output Title

Output Description
subject of court orders. It includes escorting services to convey remand and sentenced prisoners to penal institutions $NZ 15,503,000 The Treasury Policy Advice: Taxation
Provision of advice on tax policy, including advice on international taxation, personal taxation, indirect taxes, business taxation, tax systems, and compliance
Costs
International Tax NZ$ 1,286,000 Personal tax NZ$ 562,000

Indirect tax Business tax NZ$ 201,000 NZ$ 2,171,000
Tax system Ministerial services
NZ$ 468,000 NZ$ 271,000
Source.-Purchaseagreements between the Ministry of Police and the Commissioner of Police and the Minster of Finance and the Treasury; Quoted in Scott (1996).
These indicators are regularly reported on, and evaluated at regular intervals according to the evaluation plan. The United States' Government Performance and Results Act, passed in 1993, requires that 5 year strategic plans-linked to measurable outcomes-be developed by all federal programs by end 1997
(Box 5). By 1999, every agency is required to craft detailed agency performance plans, and integrating these plans into their operational and budgetary procedures. The Chief Financial Officer Act requires that each agency is externally audited on effectiveness and efficiency.
73
Box. 6: Key provisions of the Government Performance and Result Act of 1993
Agencies must submit to Congress, OMB, and President: Strategic Plans that cover at least five years and updated at least every three years. First Plan due by September 30, 1997. Requirements: * include a mission statement * include general goals and objectives * describe how goals and objectives are to be achieved * describe how the general goals relate to annual performance goals * identify key extemal factors that could affect performance * describe program evaluation Annual performance plans, first plan due to Congress at the FY99 budget request. Requirements: * establish objective, quantifiable perfonrnancegoals; * describe operational processes and resources needed to achieve the goals * establish performance indicators * provide a basis for comparing program results to performance goals * describe the means to be used to verify and validate results Annual performance report, first report due to by March 31, 2000. Requirements * describe the performance indicators, actual results, and how they compare to performance goals * review the success in achieving the performance goals and, if needed, explain why they have not been met, and what should be done * describe any waiver of administrative requirement [an exception from legal requirements on an agency granted by Congress] and their effectiveness * include summary evaluation findings * show results for the preceding year Source: http.//www.npr.govt
29. Measuring Performance. The slow progress in developing reliable and monitorable outcome indicators confirms the inherent difficulties in this area (Box 6). Moreover, measuring performance requires not only measuring outputs, but also costs. Cost information collected on an output basis requires revising the chart of accounts, revising reporting formats and (ideally) adopting accrual accounting. Accrual accounting monitors the full cost of government activities, in contrast to the traditional cash accounting which registers only expenditures, but neglects depreciation, changes in asset position, and accrued expenditures. Countries 30. Personnel management. such as New Zealand and the UK have extended performance contracting beyond the organizational level, and contract performance directly with the permanent secretary or chief executive. In France, a program for concluding
performance agreements with managers of certain well defined activities has been in place for many years. Denmark has been implementing multi-year agreements linking budgets and performance. 31. However, several countries which have experimented with detailed goal-setting for agencies have stopped short of setting individual goals in terms of outputs or performance, often because this does not fit well with the country's culture, or because the internal organizational culture is bureaucratic rather than managerial6 32. Some countries moved away from the strict limits on personnel numbers and salaries.. In the UK and Australia, for instance, departmental personnel limits have been replaced by overall limits on running costs. At the same time, civil service career planning and
See for example PUMA 1995c pp 176-179
74
training was enhanced for current and future managers.

RE-DESIGNING ORGANIZATIONS
36. Privatization. Over the last two decades, many services that were traditionally provided by the public sector were privatized in
OECD countries. This trend started with

utilities, but has expanded to infrastructure and telecommunication services as well. Besides a general backlash against government, the privatization was driven by changes in technology that allowed for private provision, and by the development of more sophisticated financial markets that could better finance large, complex, and long term projects. 37. Competition. The most direct external incentive is to expose government activities directly to competition. Differences among OECD countries' reform strategies are perhaps greatest when it comes to developing competition. Some countries have placed systematic introduction of market elements at the heart of their reforms. Other countries are still studying the extent to which they want to use them. Besides corporatizing or privatizing market activities, OECD countries have used a range of devices to create or simulate markets for more core govemment activities. These include funder-provider models such as used in the UK and New Zealand for health services or in New Zealand also for science activities and some education; or voucher systems such as employed in New Zealand and the United States for low-income housing assistance. 38. OECD countries have subjected internal services to market disciplines by introducing user charging for internally provided services oT exposing internal services to competition. In New Zealand most central services monopolies have been eliminated, and the users of these services are funded directly rather than funding the service. Most other countries however still retain a range of internal service monopolies.
33. Organizational reforms in some OECD countries have emphasized separation of functions into different organizations. These reforms promoted clarity and consistency of mission, and removed conflicts of interest, such as a conflict between responsibility for the provision of service and the provision of policy advice on the same service. Examples of these organizationalreforms include: Sweden, in common with other Nordic countries, has long separated Ministries from agencies with specific purposes. * The United Kingdom in its Next Steps program has so far moved nearly two-thirds of the civil service into Executive Agencies charged with specific delivery functions. * New Zealand has restructured most core government departments to separate policy from operations, purchaser from provider, regulator from operator and inspection from delivery. * The development of Canada's Special Operating Agencies, on the other hand, has been a limited initiative affecting only a few of the government's delivery functions. * Similarly, in France there has been limited use of special administrative agencies outside the basic framework of the Ministries. 34. OECD experience suggest that reorganizing government can yield large efficiency and effectiveness gains, by establishing clear purpose and task; explicit performance expectations; the authority to perform the task and pursue the purpose; and which are accountable for the use of the authority provided. *
39. Contracting out. PUMA reports that experience in the US and the UK with contracting out indicates substantial savings, but experience with contracting out relatively for perfor vance reinforced by increased were simple activities suggests several essential forket pefomnceiltiswerte rebinforcsedbnctreconditions for achieving real benefits, notably: market incentives for the public sector.
7PUMA
1995c p 45
75
* * *
the existence of a competitive market amongst suppliers; open, verifiable procurement procedures; and solid skills in contract management.
40. These pre-conditions set practical limits to contracting out more complex activities in some countries; at least in the shorter term. And, as is the case for all market-type mechanisms, it is evident that potential improvements in performance can be dissipated if implementation issues are not given sufficient 8 attention. 41. The case for contracting out is strongest where the good or service required can be specified with a fair amount of precision., requires little specific capital to produce, and has a strong and competitive private market. Examples are computers, transport, cleaning services, building maintenance, minor capital works, printing and stationery and office accommodation. Some of these conditions may also apply to services such as legal advice and consultancy. Efficient choice between internal and external provision also requires the organization's accounting systems to provide the necessary cost information to set internal prices. 42. The United States has been focusing efforts on improving service delivery at central level through its National Performance Review, aggressively led by Vice President Gore. The NPR has led to a dramatic reform in procurement rules with much greater priority being given to value for money. 43. Competition surrogates. Where services are not contestable, as in the case of most core government services countries have experimented with competition surrogates. Such tools are designed to increase public sector the responsiveness of organization, and include: * Service standards. The Citizens' Charter introduced by the United Kingdom has attracted widespread interest in other
countries. Several countries (Belgium, Italy and Canada) have instituted specific public declarations of departmental standards of service. Others are considering 9 them. In France, a report currently before Ministers speaks of "putting citizens at the center of public service." Benchmarking--comparing the performance of an organization with similar ones elsewhere, or in the private sector--has been used as a tool to increase competitive pressures on public sector organizations. User funding. Rather than funding the public sector organization, its users can be funded. Vouchers are used in some countries' education system, funding the school which parents choose for their children, rather than the school directly. A national health insurance scheme that allows choice of provider works in similar ways.
44. The emphasis on responsiveness to citizens needs to be reconciled with formal organizational accountability upwards. Setting standards of service implies that departments will be accountable for achieving them. This in turn logically implies that budgetary decisions will take account of the outputs to which departments are committed. Further, a balance has to be achieved between client responsiveness and client capture. The public service is not seen to be directly accountable to its clients. Accountability to the public is indirect through the government and the 10 Parliament.
THE CHANGING ROLE OF THE BUDGET 11 OFFICE
45. The role of the budget office in OECD countries has changed significantly over the recent reforms. The traditional role of the budget office (usually MOF, although some OECD countries have budget offices separate from the treasury function) was to function as a central command and control post, specifying the items of expenditure, negotiating budget
5
PUMA 9
199561 ce
OECD 1995 p 49
This section draws heavily from Allan Schick, Thechanging role of the budget office, OECD Document No. GD(97)109, Paris 1997.
76
proposals with line ministries to control overall expenditures, monitoring compliance with regulation, and ensuring that inputs were used as agreed in the budget. 46. This role has changed in those countries that have undergone significant reforms. Managing the budget process is still the central role, but the way this role is operated differs from the traditional role. * MOF is usually in charge of the multiyear forecasts: it either makes them itself (as in Australia) or manages the process by which line ministries make the forecasts. In the latter case, MOF provides the key economic data driving the multiyear forecasts, and guides ministries how to assemble the forecasts. * MOF usually manages the process of strategic policy choice. This no longer implies making allocation decisions during budget negotiations. Rather it means facilitating the process by which line ministries make trade-offs within the overall expenditure limit. MOF usually sets guidelines and procedures by which line ministries propose new policies and review existing policies. * MOF's role in budget implementation has become one of monitoring and informing rather than control. Aggregate spending limits are still rigorously enforced, but line ministries have gained significant leeway in detailed spending decisions. In return, MOF's task is to ensure that resources are productively used. To this purpose, MOF has taken the lead in revising budgeting, accounting and reporting formats to match new concepts of performance, and to ensure that this information is regularly reported. 47. MOF in several OECD countries has taken a leading role in the public expenditure management reforms. However, reforms have only worked in countries in which a broad commitment to reforms existed.
CONCLUSIONS
48. The OECD experience suggest that key elements of a performing public sector include: (i) hard budget constraints based on a multiyear perspective on the budget, and enforced by internalized discipline of budgetary units (ii) results oriented institutional arrangements for decision making, resource allocation, budgeting, personnel management; and (iii) organizations which have clarity of purpose and task, clearly specified performance expectations, authority to perform the task and commensurate accountability mechanisms. 49. Reforms are still ongoing, and some countries have only just begun. It is therefore too early to tell whether the reforms have been successful. However, the growth of government slowed in most OECD countries, and was even reversed in some. Deficits improved throughout the eighties, in particular in the forerunners of reforms, New Zealand and Australia, although a relapse occurred in the recession of the early 1990s. The USA, in part due to an unusually long upswing in the economy, the large deficits of the 1980s have been eliminated altogether. But perhaps the most significant result of public sector reforms is a reassessment of the role of the state. At the end of the 1970, government was seen as the problem rather than the solution. Now, the emerging consensus is that the state has a significant role to play in the economy, and that capable states can do more and with better results.
77
Table 3: Public Expenditure Management in OECD countries Linking Policy, Planning, and Budgeting
Australia
Strategic Goals! Global Budget Targets Coninnitnient to zero budget deficit ovcr the busiiuess cycle. regular publications of fiscal outlook. Forward estitiates (budget year and three otiter years) central to budget process.
France
Medium-temm targets in the "lois d'orientation' Maastrictit criteria
Germany
Finanjcial Plantning Council can set cxpeciditure limits. Currently: Maastricht criteria Five year rolling planis developed by MOF on basis of Finianicial Planning Council agreement, which coordinates policy with subnatiotial Laender; inclided ini the budget documents, but no legal or operationial status Ministries submit proposals to MOF within linmits set iii budget circular. Submissions reviewed by MOF at various levels and ir thieCabinet, which subitrits to Parliatrierit. Two rouaids of review ih Parliamcit.
Japa
Reduction, in bonds issue to below 5 percent of cxpenditures
New Zealand
Move toward surplus in adjusted financial balanice; reduce expenditures to GDP. Targets ii Fiscal Responsibility act Three year projections of "baseline." based on previous year's ttutitbers and Govemment decisions; issucd to liiie ititristries at the start of the budget process. Included in the budget bill. Ouiter year projections becoiire baselitc. Ouitput budgeting. Budget based on baseline estinrate that inicludes policy decisiosis. Miniisters iisake decisions oni output purchases ratlier thani speniding. Speidiiig ministries prepare appropriation estiniates based oti Cabinet oultptit decisionis.
Utnites Kingdom
Expendititre growth lower than GDP growth.
United States
Balanced budget by 2002
Multi-annual budgetitig
Experinsental 3 year rolling platis, annexed to the bridget documents
No fonnal system of nsultiyear estimates. Mediunitemi projections included in Budget Docunsent
Budgetary policy aet withinl Medium-term Financial Strategy, with key focus oai borrowing requiretirerit. Presciited in Budget Documents. New Conitrol Total drives multi-year expenditures.
Office of Manageiicent and Budget nakes five year projectiort, but no operational significance.
Salient Features of Budget Process
Ftudget Year, Budget Calenidar Fxpendituire Review
Cabinet deteniiines overall expenditiure linit. Expenditure review comrriittee(ERC) corisidcrs policy cliairges, based oai policy proposals arid savings optioiis. ERC sets sector ceilings for individual ministries. Ministries prepare scotor budgets Prigraiii riuiaiagetiietit focuses on programs as the basis for resource allocatioui July 1-June 30; Preparation January-June ERC lar reviews available called for in previous budget round. All programs evaluated over a 5-year cycle Responsible for economic, fiscal and nionetary policy. Manages the three year forwaid projectiotns drafts and proposes the budget; evaluates atid reviews government programs aiid expenditure proposals; in charge of finiancial administration No power to propose trew policies; may anmendthe executives' proposal without increasing the tax burden, may refuse approval of budget
Prire minister sets sectoral arid orgaiizatiorial expenditure limits based oni MOF's iintemal projectiois. Budgct proposals sliosild i\ principle not execed liroiits, but Ministries may appeal to PrinmeMinister.
Centially managed by the Finanice ririnistry, Budget Office; Line minfistries submit proposals to MOF oin basis of a cabiiict appmved circular, arid negotiate with MOF. MOF seids proposal to speniding irnniristries, followed by new nsegotiatioins. Cabinet dccides. April I-March 31. Preparation starts in April of previous budget year Efficiency arid effectiveniess reviews ill budget process. Suniset legislation provides autormatic review MOF responsible for budgetary policy, tax administration, bankinig supeivision, Fiscal investment and loan program.
Budget madc within limit of conitrol figures. tIrecasuryinvites "position papers" frrni departnsents, including. spendinig increases that need to be imatched with savinigs proposals. Onz basis of position papers, Cabiniet decides guiidelines for budget. Cabinet cornnriritteereviews ririnisterial proposals, and proposes allocation) of control figiures. Cabinet approves. April I-March 31 Budget process starts April of' previous year "efficiency scrutitnies" to lower costs; budget process forces to propose savings; Ad hoc reviews of major policy areas As in Australia; recenitly sortie monetary policy responsibilities delegated to Bank of Eigliand
Budget cycle divided irs administrative and coiigressional cycle. Adiniiiistrationi prepares on basis of OMB guidelinies. Experiditiures disided ill iiiaridatory and discretionary, the latter subject to legal limits on the deficit. Presideint submits proposal to Conigress, whicil, after a complex process, approves the budget October I-Septernber 30 Bridget pr ocess starts 17 monlthisiii advanice. Evaluation of existing prograiims ittcluded ill ministerial budget requests. 0MB and CBO reviews of specific areas Office of Management and Budget aird Treasiury have responsihilities siitiilar to Australia.
January I-Decenrber 31 Preparation JaniiaryDeceriiber of previouis year No fosmialprocedure. Court des Comptes (atudit office) perforitis ainalysis.
Jarnuary 1-December 31 Preparatioil starts 13 months in advance Ad hoc reviews of existing policies. Biannual report ott subsidies;
July l-Jiriie 31 Process starts July of previouis year Aninlualbuidget cycle rasain vehicle. Ad hoc reviews of specific areas
Powers of Ministry of Finance or treasury in Budget Preparatioe
As in Australia, but with more detailed expenditure colitrol
Respoiisible for fiscal policy. Drafts and proposes the budget; evaluates atid reviews govemmnent progrms and expenditure proposals; in charge of financial arbriirtistration
As in Australia, buit with mitre emphasis on senitinizing buidget proposals arid adniiiistratioir
Role of Parliarnent
No power to propose spending. Parliament may raise expenditures if others are cut.
Unlimited powers to come up with new proposals or propose animiidinetits to the executives' proposal
Both houses of the Diet can miiakeamendments and additions to expeniditures proposed
No power to come uip with new proposal. Executives' proposals may be amended witiout raisinig the tax burden.
No powers to comc up with new sending proposals. taxes and spending may be reduced, but not increased.
Unlimited powers to conic up with new proposals and to anitenid executive proposals
Table 4: Public Expenditure Australia

Performance Specirfication Buidgeting for Operational expenditures Expected Outcomes and associated outputs included in budget documents. tMinistries/agencies receive single running cost allocation, which is formula driven, and requires efficiency dividend. No personnel limits on ministries and agencies. Possibility to shift resources between personnel and running cost, and between program. MOF controls aggregate cash limits; expenditure controls responsibility of line ministry of agency; MOF provides accounting services; Single treasury account. Virement is possible between sublicads auid within running costs, however, this is not regulated by the law. Carryover up to 6 percent of ranning costs; capital expenditure may be caricd over, however, this is not regulated by the law
Management Japan
line item budgets
in OECD countries New Zealand

Output budgets, strategic plan, business plan Allocations for outputs to be produced. Chief executives of ministries, agencies have authority to hire and fire. Limits set on chief executive salaries
Incentives for Performance

France
line item budgets. Ministry and Centers of Responsibility Contracts not available
Germany
Inze item budgets
Unites Kingdom
Output Budgets,
United States
Legislative requirement for strategic plan, agency performance report Total federal employment controlled; limrtitson niumber of senior staff
Unified personnel system for central and local govemmenit. Number of posts approved in budget. actual number determined by appropriation, but reserves can finance staff up to approved number.
Total Staff Number Law determines maximum overall staff. Cabinet order and budget deternsines staffnumbers for ministries. Ministry of Finance involved in salary budget preparation Quarterly commitment and disbursement plants approved by MOF . Govemment consolidated account in Bank of Japan; sub-accounts for line ministries. Virement within one item with approval of MOF. Among iteis with Parliamentary approval. Canyover automatic, if expenditures approved as 'continiues expenditures," If liability already accrued, or expenses already approved, with MOF approval. cash
Personnel costs controlled by running cost system. Pay increases in principle financed fromnefficiency gains.
Expenditure Control
""I
MOF issues warrants, and checks in principle all expenditures, through Inspecteur de Finances. Recently has devolved some authority.; single treasury account at central bank, encompassing regional accounts. Virement is possible within a limit of tO percent of the appropriated amount. No provision for carryover
Approval by the Ministry of Finance is required before disbursement; comnuitmnents can be blocked if a certain percentage of appropriation is spent; MOF and heads of agencies responsible for expenditure control . Single treasury account encompassing regional accounts. Virement possible within dse same chapter, and between salaries and wages Must be approved by the MOF. Investunent expeiiditure and expenditure from earmarked revenue may be carried over
Departmcntal accounts in a commercial bank. Treasury clears the accounts every night, consolidates balances at the Crown Account in the central bank. The law gives the Ministry of Finance or treasury a possibility to require any information, which is used as a base for a warratit system. Transfer allowed up to 5 percent from output to output, if no other transfer to the same appropriatioti lias becii madc during thc year and the total amount is unaltered. Spending authority lapses at the end of the year or as specified in the appropriations act; no authority ovcr five years. Accrual, Generally Accepted Accounting Principles
Warrant issued by the Minister of Finattee. The treasury, and chief financial officers in agencies, ministries responsible for expenditure control.. Single trcasury account; paymaster accounts for ministries. Virement between subheads is possible but needs approval by the treasury. I'he Treasury has no power to authorize virement between votes or to meet additionial expeiiditure with vireinent. Carryover for capital expenditure s, and for meeting expenditures chargeable to that year. Modified cash, moving to accrual
Warrant issued by the treasury. Office of management and budget, treasury, and heads of departmient responsible for expenditure control. Single treasury account; Tax and loan accounts in conunercial batsks, on call for treasury. The law has no provision for virement and carryover.
BIudget accounting and reporting
Audit
Cash with accrual elements. Comsprehensive budget documents and agency annual reports Financial ansdperfomsance audits by the Australian National Audit Office, reporting to Parliament
cash
cash
cash based, with some accrual
Inspecteur des Finances intemal audit. Court des Comptes extemal audit, reported to Parliament
Bundesfinanzhof perfonss financial and efficiency audits, and reports to parliament. relies on pre-audits by auditors in administration N/A.
Key organizational reforms Market-type instruments Main legislation/ documentsfor Public Sector
Reforms
Cut in # of ministries from 28 to It in 1987 increased user chlarging; key agencies have published client charters Financial Management Improvement Program
deconcentration to regional authorities Public service charter
Boars of Audit oversees audits and reports to Cabinet which passes on report to Diet. General Executive Bureau does the actual auditing. No new agencies unless an old one is abolished N/A.
Controller and Auditor General focuses on fittancial audit, control system audit. Regular perfomance audits. reports to parliament
National Audit Officc does financial and efficiency audits. Reports published, and reported to parliament
General Accounting Office performs audits, reports t congress. At state and local level, most audits performed by private fioss. Privatization, deregulation,
N/A.
Separation of implementing agencies ftom ministries Privatization Contract-based chief executive, state secretary; contracting out State services Act Public Finance Act (1989)Financial responsibility act (1994)
"Next Step" agencies Privatization Contract based chief executives, Citizen's charter; contracting out White papers: Financial Management Initiative Next Steps; Initiatve; Citizen's Charter
benchmarking; out
contracting
N/A.
N/A.
N/A.
Chief Financial Officer Act; Government Performance and Result Act.
Sources: OECD 1993, 1995, Garamfalvi and Allan, 1997, World Bank,
STATISTICAL TABLES
Table 1: INDONESIA- GOVERNMENTREVENUE (in billion Rupiah, GOIformat)

Fiscal year: Indonesian 1994/1995 Budget Actual 1995/1996 Budget Actual 1996/97 Budget Actual 1997/1998 Budget Actual 1998/1999 Budget Actual pronsion Budget 1999/2000 BudgetRevision 24/2/99
A. DomesticRevenues 1.OilRevenues 1.CrudeOil 2. Gas II. Non-Oil Revenues 1. Income tax 2. Salestaxn(VAT) 3. Importduties 4. Excises tax 5. Export taxes 6. Property 7. Othertaxes 1/ 8. NonTax Receipts(excl. privn) a. PETransfers (Dividend) o/w NaturalResources b.Other Non-TaxRevenues abroad) i. Immigration(embassies ii. Education iii. Sales of goods iv. Services v. Attorney and court vi . RefundRevenues vii. Transferfrom Investment viii. Self Financing(incl. ReforestationF 9. Privatization 10. Proceeds from petroleumsales Revenues B. Development Programaid ADB IBRD OECF (Japan) Others of whichM/yazawaBank Restr. Mlyazawa Bank Restr. 11. Projectaid
b.
58,952 12,851 9,504 3.347 46,101 18,843 13,239 3,443 2,623 16 1,629 282 3,507 1,550 0 1,957 0 262 0 1,167 0 449 80 0 0 2,519 10,012 0 0 0 0 0 0 0 10,012
66,418 13,537 10,004 3,533 52,881 18,784 16,545 3,900 3,153 131 1,647 302 6,433 1,322 0 5,111 13 283 48 1,878 29 2,461 400 0 0 2,006 9,838 0 0 0 0 0 0 0 9,838
66,199 13,276 9,812 3,463 52,923 19,239 16,655 3,543 3,299 44 1,923 319 6,425 1,695 0 4,730 0 57 0 425 0 502 1,500 2,246 0 1,475 11,759 0 0 0 0 0 0 0 11,759
71,340 16,055 11,964 4,091 55,285 21,012 18,519 3,029 3,593 186 1,894 453 6,111 1,804 0 4,508 24 50 55 741 34 887 2,132 583 0 488 9,009 0 0 0 0 0 0 0 9,009
79,829 14,120 10,316 3,805 65,708 23,708 21,788 3,451 4,033 160 2,277 570 7,202 1,872 0 5,330 0 56 0 512 0 1,148 2,154 1,460 0 2,519 10,012 0 0 0 0 0 0 0 10,012
87,830 20,t37 14,783 5,355 07,493 27,062 20,351 2,579 4,263 81 2,413 591 10,153 2,650 0 7,503 31 58 56 1,364 33 1,368 2,562 2,031 0 0 11,900 0 0 0 0 0 0 0 11,900
88,061 14,871 10,698 4,183 73,189 29,118 24.601 3,322 4,436 100 2,505 633 8,226 1,925 0 6,301 18 61 33 616 18 1,652 2,200 1,703 0 249 13,026 0 0 0 0 0 0 0 13,026
112,275 30,559 22,264 8,295 81,717 34,388 25,199 2,999 5,101 129 2,641 478 10,782 2,341 0 8,442 42 69 66 1,593 30 1,819 3,555 1,267 0 0 14,386 0 0 0 0 0 0 0 14,386
149,302 49,711 32,909 16,803 99,591 25,846 28,940 5,495 7,756 943 3,411 540 11,660 4,000 912 7,660 20 95 38 842 20 2,828 1,472 2,346 15,000 0 114,586 74,045 26,580 21,264 19,776 6,425 0 0 40,541
152,809 41,254 25,829 15,425 111,556 49,714 28,386 2,218 7,974 4,582 3,163 462 11,871 3,524 na 8,347 82 95 94 1,238 32 1,802 2,828 2,284 3,185 0 62,320 36,403 14,705 12,680 5,678 3,340 0 0 25,917
138,380 20,965 12,443 8,522 117,415 40,626 34,697 2,950 9,360 2,595 3,247 565 10,375 4,000 956 6,375 0 118 0 n.a. 0 n.a. 3,111 3,146 13,000 0 77,400 47,400 7,500 7,500 9,900 22,500 11,250 11,250 30,000
142,204 20,965 12,443 8,522 121,239 40,626 34,597 2,950 10,160 2,595 3,247 565 13,499 4,000 1,896 9,499 34 6 38 786 17 1,421 3,111 4,086 13,000 0 77,400 47,400 7,500 7,500 9,900 22,500 11,250 11,250 30,000
00
TOTALREVENUES
68,964
76,256
77,958
80,349
89,841
99,530
101,087
126,661
263,669
215,130
215,780
219,604
Source:VVoard8Bank
Table 2: INDONESIA- GOVERNMENT EXPENDITURE

(in billion Rupiah, IndoneslanFiscalyear 1994/1995 Budget Acual 1995/1996 Budget Atual GOI format) 1999/2000 Budget Acua. Budget Aulualpte.-nMar'09 Budget B2d/t21R-
199691998/1999 Budget Adual
A. RBetineE.penditules I. Peseer.tel EBpeeditetes 1 Wages-. Soues 2.Rc alowatce 3 FoodAlluwrace 4 Others 5 Eute-al II. Matetial EXpndit-ues 1 Dom-stro 2. Euler-al It. .ransfe-tee Begns 1 PersonnelEpenses 2 Nun-p,eronnelExpu-ses
42351 13,011 10,456 1,039 783 392 341 3,751 3,526 225 1,-55 0,605 430 17,969 317 17,652 10,521 7,131 525 0 525 0 0 0 0 0 525
50539 12,596 10,181 973 756 360 311 4,319 4,101 218 7,272 6.919 354 24073 104 24,768 18.298 0,470 1,480 807 793 0 0 0 0 0 793
54,416 15,347 12,416 1,140 035 511 445 4,745 4,457 288 9,400 7,932 477 25,390 319 25,071 17,096 7,175 524 0 524 0 0 0 0 0 524
57.310 13,001 11,048 733 560 370 290 5.175 4,876 300 6,227 7,607 419 20,984 1,020 27.364 20,469 0,075 1,923 0 1,923 0 0 0 0 0 1,923
56.114 18,281 14,763 1,100 1,122 710 492 6,539 6,258 332 10.012 9,496 516 20 227 291 19,936 12,116 7,820 1,005 0 t,005 0 0 0 0 0 1,005
62,561 14,455 13,005 7680 101 400 103 8.109 7,025 204 9,358 8,074 484 27 491 4,509 22,902 13,000 0,902 3,149 1041 1,733 0 0 0 0 0 1,733
B2,159 21192 17,048 1,310 1,234 1,010 551 6,995 6,478 417 11,535 10,960 508 19,571 334 19,237 11,506 7,541 965 0 905 0 0 t) 0 0 965
89.010 17,269 13.698 7a8 1,174 671 939 9.999 8.242 757 11091 10520 541 31,112 1,628 29,485 12150n 10.735 21 109 9 814 11,354 0 10,599 0 0 0 756
171,204 24,791 19,120 1.872 1,484 1.155 1,150 11,425 10,060 1,385 13,289 12,G06 683 66,236 1 940 E4,290 33,202 31,035 55,473 27,534 27,939 6,473 15,709 1,,808 882 130 2,745
147,717 24,480 19,099 1.606 1,687 1.162 937 11,058 10,054 1,004 14,194 13,512 002 55.796 220 55,570 31,404 24094 42,187 27,185 15,002 4,530 7,857 1348 586 27 2,002
134,556 32,037 25,203 2,087 2,107 1,490 1,050 11,039 10,007 1,rl32 18430 17.629 001 44,011 380 44,431 23,905 20,020 28,239 9 089 18,253 7,201 0,836 744 143 153 3.919
137,156 33,569 26.825 2,087 2,107 1,490 1,050 11,039 10,007 1032 19,498 10,697 001 44,011 380 44,431 23,9015 20,520 28,239 9,906 18,253 7,201 6,039 744 143 153 3,919
tV.Debt Senvice 1 ouererc 2 Etemal o Pnec,pol b lntert V. Othet E.eeedrtres V. OtherEspendltates 1 Petr,,ler ob6idies 2 OtherSub,rdlesand Eypendforcs o Elcertl-ty substdy b. Subsidyto BULOG of wh ch rdoro-t -rb,idy for BULOU o Med,, -,esubsidy d OthersuOsdies e OtherCx,Pydores
B. Develepme,t EBpeodit.res 1. RuprahF -ancig a D-eet/rprerptP-u-rae . epartmen,t/GoVlst 0/w, Social Solety Net Program De-en Tasfer to Regions(1PRES) /. Social Safety Net Pgram 0l.Other Pgram b. Others I SoclalSafety Net ii Interest SubsIdies iii. BankRestt,a,ng Cost to,Fcoll/aer Costs o Other DeuelopmentEpenddutes IwO SSN
27,348 17,330 16,760 9,356 0 589 5,340 0 1,482 560 0 0 0 105 303 o 589 10,012 9,447 565
30,692 20,854 10,592 10.566 0 671 5.670 0 1,683 2,262 0 0 0 815 1,447 0 671 9,839 9,231 607
30,734 18,975 10,230 10,285 0 625 5.570 0 1.750 744 0 0 0 143 601 0 724 11,759 11,900 593
28,781 19,772 10,192 10,221 0 759 5,488 0 1,724 1,500 0 0 0 143 1,437 0 759 9,009 B,174 835
34,503 22.089 21,103 11,955 526 768 6,389 2,072 906 0 0 0 137 769 0 899 12,414 11,780 633
35,952 24,052 21,020 11,160 0 999 6,472 0 2,396 3,024 0 0 0 186 2.838 0 1,126 11,900 11,337 583
30,928 25,902 24,25 13,950 2,280 965 7,31 137 2,270 1.077 0 0 0 137 763 0 1,086 13,026 12,385 641
38,359 23.973 21,920 11,160 0 902 7,512 0 2,352 2,048 0 0 0 708 1,340 0 1,022 14,386 13.454 901
92,083 52142 28,203 13,494 9,332 903 10,757 6,947 3,049 23,939 1,702 3,958 15,000 2,125 1,154 15 229 40,541 39.541 1,000
07,869 41,952 26.040 12,482 8,506 922 9,709 1,009 2,B47 15.912 1,624 1,166 10,000 2,125 997 46 318 25.917 24.542 1,375
63,646 53,040 30,652 12,923 8 a 1,264 13.562 3,458 2,902 22.997 0 3,701 18,000 o 1,206 0 0 30,000 28,976 1,024
82,440 52,448 30,152 12,758 2,915 1,264 13,227 2,733 2,902 22,297 0 3.701 17,000 0 1,596 0 0 30,000 28,976 1,024
offMwrresh redforce
2 Pr,jeclald Investment PFjeds Ammed Forces
TOTAL EXPENDITURES S0ur-e WVrdo Bank
09,699
81,231
85.149
86,091
90,610
90,513
101.087
127,969
293,807
215,596
218,204
219,604
Table 3: INDONESIA - FISCAL SUMMARY (in trillion Rupiah, Bank format) Fiscalyear: Indonesian 199411995 Budget Actual 1995/1996 Budget Actual 1996/97 Budget Actual 1997/1998 Budaet Actual | 19981999 audget Actual prusisio Mer'99 1999/2000 Budget Budget Revision
2412109
TotalRevenue& Grants 1.Tax Revenue Oil/LNGtaxes Non-oiltaxes proceeds) (exct. pnvatization 2. Non-taxrevenues 3. Grants
06.4 52.9 12.9 40.1 2.5 0.0
64.4 58.0 13.5 44.4 6.4 0.0
64.7 58.3 13.3 45.0 6.4 0.0
70.9 64.7 16.1 48.7 61 0.0
77.3 70.1 14.1 56.0 7.2 0.0
87.6 77.1 20.1 57.3 10.2 00
87.8 79.6 14.9 64.7 8.2 0.0
112.3 101.5 30.9 70.9 10.8 0.0
134.3 122.6 49.7 72.9 11.7 0.0
149.6 137.8 41.3 96.5 11.9 0.0
125.4 115a0 21.0 94.0 10.4 0.0
129.2 115.7 21.0 94.7 13.5 0.0
rvo
and NetLending TotalExpenditure 1. Cun-entExpenditures a. PersonnelExpenditures b. MatenialExpenditures to c. Transfers Regions debt servic d. Domeslic e. Extemalinterest f. Totalsubsidies(incl. Interestsubsidies) expenditures g Othercunrent (net) h. BankRestnuctunng Banek (on-bdu,t, basis) 2. CapitalExpenditures CurrentSavings Costs excu. Bank Restructuring OverallBaiance Costs(gross) Bank Restructuring Costs(net) BankRestructuring of Recovery BankAssets Proceeds Privatization cost) (incl.Bankrestructuring OverallBudgetBalance discrepancies operationandstatistical off budget & operation stat.disc.) off-budget (including Fiscaldenfcit Financing (net) 1 Extemal Disbursements Amortizations 2. Domestic(netfnancialdtuwdown) 3. Privatization of 4. Recovery BankAssets UnidentifiedFinancing Bane Source: Wuord
59.2 32.6 13.0 3.8 7.1 0.3 7.1 0.2 1.1 0.0 26.6 23.9 -2.7 00 0.0 0.0 0.0 -2.7 00 -2.7 4.3 -0.5 10.0 -105 48 0.0 0.0 1.6
62.9 33.7 12.6 4.3 7.3 0.1 6.5 1.5 1.4 0.0 29.3 30.7 1.5 0.0 0.0 0.0 0.0 1.5 00 1.5 -1.6 -9.5 9.8 -18.3 6.9 0.0 0.0 -0.1
67.1 37.3 15.3 4.7 8.4 0.3 7.2 0.1 1.1 0.0 29.8 27.5 -2.4 0.0 0.0 0.0 0.0 -2.4 0.0 -2.4 33 -6.1 11.8 -17.9 9.4 00 0.0 0.9
6506 37.8 13.0 5.2 8.2 1.6 6.9 0.1 2.8 0.0 27.8 33.1 5.3 00 0.0 0.0 0.0 5.3 0.0 5.3 -0.3 -11.5 9.0 -20.5 6.2 0.0 0.0 -0.1
78.5 44.8 18.3 6.6 10.0 0.3 7.8 0.1 1.6 0.0 337 32.5 -1 2 0.0 0.0 0.0 0.0 -1.2 0.0 -1.2 1.9 -2.1 10.0 -12.1 3.9 0.0 0.0 0.6
85.5 50.3 14.5 8.1 9.4 4.6 9.9 1.6 2.3 0.0 35.2 373 2.1 0o0 0.0 0.0 0.0 2.1 0.0 2.1 -3.5 -1.1 11.9 -13.0 -2.4 0.0 0. -1.4
89.2 51.2 21.2 8.9 115 0.3 75 01 1.6 0.0 38.0 36.6 -1.4 .0 0.0 0.0 0.0 -1.4 0.0 -1.4 2.0 1.3 13.0 -11.7 0.7 0.0 0.0 0.6
115.2 79.5 17.3 9.0 11.1 1.6 16.7 21.1 1.7 0.0 39.7 33.8 -2.9 0.0 0.0 0.0 0 -2.9 0.0 -2.9 3.6 1.6 144 -12.8 1.9 0.0 0.0 0.6
229.9 160.0 24.8 11.4 13.3 1.9 31.0 5898 3.7 15.0 69.9 -25.7 -79.6 15.0 15.0 0.0 15.0 -94.6 0.0 -94.6 96.4 81.3 114.6 -33.3 0.1 19.0 0.0 1.8
182.5 130.9 24.5 11.1 14.2 0.2 241 43.5 3.4 10.0 01.6 18.7 -22.9 10.0 10 0 0.0 3.2 -32.9 10.6 -22.3 23.9 30.8 62.3 -31.5 -10.1 3.2 0.0 -8.9
194.3 133.4 32.0 11.0 1894 0.4 20.5 28.0 4.9 18.0 60.9 -8.0 -50.9 34.0 18.0 t6.0 13.0 -66.9 0.0 -68.9 82.5 53.5 77.4 -23.9 0.0 13.0 16.0 13.6
195.7 135.0 33.6 11.0 19.5 0.4 20.5 28.0 4.9 17.0 607 -5.8 49.5 33.0 17.0 16.0 13.0 -66.5 0.0 -66.5 92.5 53.5 77.4 -23.9 0.0 13.0 16.0 16.0
Table 4: INDONESIA- FISCAL SUMMARY (as percentageof GDP. Bankformat)

IndonesianFiscal year: | 1994/1995 Budget Actual 1995/1996 Budget Actual 1996/97 Budget Actual
1997/1998
Budget Actual
1998/1999 Budget Actual provision Mar 99 14.8 13.6 4.1 9.5 1.2 0.0 18.0 12.9 2.4 1.1 1.4 0.0 2.4 4.3 0.3 1.0 5.1 1.8 -2.3
1999/2000 Budget Budget Revision 24Q2/99 10.6 9.5 1.7 7.7 1.1 0.0 16.0 11.0 2.7 0.9 1.6 0.0 1.7 2.3 0.4 1.4 5.0 -0.5 -4.0
Revenue& Grants exc. Privatization 1.TaxRevenue Oil/LNG taxes Non-oiltaxes 2. Non-taxrevenues 3. Grants Total Expenditures 1. Current Expenditures a. PersonnelExpenditures b. Material Expenditures c. Transfersto Regions d. Intemal debt service e. Extematinterest f. Total subsidies g. Othercurrent expenditures h. Bank Restructuring 2. DevelopmentExpenditures Current Savings Overal balance beforeprivatization proceed and bankrestructurIng cost Bank Restructuring Costs (gross) Bank RestructuringCosts (net) Recovery Bank Assets of Privatization Proceeds Overall Budget Balance(incl.Bank restructuring cost off budget operationand statisticaldiscrepancies Fiscaldeficit (including off-budgetoperation stat. d &
14.1 13.2 3.2 10.0 0.9 0.0 14.8 8.1 3.2 0.9 1.8 0.1 1.8 0.0 0.3 0.0 6.6 6.0 -0.7
16.1 14.5 3.4 11.1 1.6 0.0 15.7 8.4 3.1 1.1 1.8 0.0 1.6 0.4 0.3 0.0 7.3 7.7 0.4
13.8 12.4 2.8 9,6 1.4 0.0 14.3 7.9 3.3 1.0 1.8 0.1 1.5 0.0 0.2 0.0 6.3 5.8 -0.5
15.1 13.9 3.4 10.3 1.3 0.0 13.9 8.0 2.8 1.1 1.7 0.3 1.5 0.0 0.6 0.0 5.9 7.0 1.1
14.0 12.7 2.6 10.1 1.3 0.0 14.2 8.1 3.3 1.2 1.8 0.1 1.4 0.0 0.3 0.0 6.1 5.9 -0.2
115.8 14.0 3.6 10.4 1.8 0.0 15.5 9.1 2.6 1.5 1.7 0.8 1.8 0.3 0.4 0.0 6.4 6.7 0.4
12.6 11.4 2.1 9.3 1.2 0.0 12.8 7.3 3.0 1.3 1.7 0.0 1.1 0.0 0.2 0.0 5.4 5.2 .0.2
16.3 14.7 4.4 10.3 1.6 0.0 16.7 11.4 2.5 1,3 1.6 0.2 2.4 3.1 0.2 0.0 5.3 4.9 0.4 47.5
12.T 11.6 4.7 6.9 1.1 0.0 21.7 15.2 2.3 1.1 1.3 0.2 2.9 5,6 0.4 1.4 6.5 -2.4
10.2 9.4 1.7 7.7 0.8 0.0 15.9 10.9 2.6 0.9 1.5 0.0 1.7 2.3 0.4 1.5 5.0 -0.7 4.2
0.0 0.0 0.0 0.0 0.7 0.0 -0.7 0.0 1.1 -0.1 2.5 -2.6 1.2 0.0 0.0 0.4
0.0 0.0 0.0 0.0 OA 0.0 0.4 0.0 -0.4 -2.1 2.5 -4.6 1.7 0.0 0.0 0.0
0.0 0.0 0.0 0.0 -. 5 0.0 -0.5 0.0 0.7 -1.3 2.5 -3.8 2.0 0.0 0.0 0.2
0.0 0.0 0.0 0.0 1.1 0.0 1.1 0.0 -1.1 -2.4 1.9 -4.4 1.3 0.0 0.0 0.0
0.0 0.0 0.0 0.0 -0.2 0.0 -0.2 0.0 0.3 -0.4 1.6 -2.2 0.7 0.0 0.0 0.1
0.0 0.0 0.0 0.0 OA 0.0 0.4 0.0 -0.6 -0.2 2.2 -2.4 -0.4 0.0 0.0 -0.3
0.0 0.0 0.0 0.0 -0.2 0.0 -0.2 0.0 0.3 0.2 1.9 -1.7 0.1 0.0 0.0 0.1
0.0 0.0 0.0 0.0 -OA 0.0 -0.4 0.0 0.5 0.2 2.1 .1.9 0.3 0.0 0.0 0.1
1.4 1.4 0.0 1.4 -9.0 0.0 -9.0 0.0 9.1 7.7 10,9 -3.2 0.0 1.4 0.0 0.2
1.0 1.0 0.0 0.3 -3.2 1.0 -2.2 0.0 2.4 3.0 6.1 -3.1 -1.0 0.3 0.0 -0.9
2.8 1.5 1.3 1.1 -5.6 0.0 -5.6 0.0 6.7 4.4 6.3 -2.0 0.0 1.1 1.3 1.1
2.7 1.4 1.3 1.1 -5.4 0.0 -5.4 0.0 6.7 4.4 6.3 -2.0 0.0 1.1 1.3 1.3
Financing 1. External(net) Disbursements Amortizations 2. Domestic(netfinancial drawdown) 3. Privatization 4. Recoveryof Bank Assets UnidentifiedFinancing
Table 5: Investment Budget (Development Budget) in billions of Rupiah - By Sector

199411Y95 No Category Plan Abs 450.5 143.8 305.7 989.6 662.4 327.3 1,687.0 809.4 877.6 146.5 119.4 27.1 % 1.6 Actual Abs 565.2 223.6 341.6 1,657.5 1,351.2 306.3 1,932.1 999.5 932.6 109.0 104.8 4.2 % 1.8 Plan Abs 497.3 173.6 323.7 1,103.8 666.5 437.3 2,042.1 874.2 1,167.9 170.6 132.8 37.8 1995/1996 % 1.6 Actual Abss 805.4 619.5 185.9 891.0 676.8 214.2 1,897.0 1,115.4 781.6 139.3 121.6 17.7 % 2.8 Plan Abs 506.6 207.8 298.8 1,294.4 823.3 471.1 2317.4 1,239.8 1,077.6 187.1 160.3 26.8 1996/1997 % 1.5 Actual % bs 1,133.3 1,004.6 128.7 1,308.3 1,027.5 280.0 2,101.6 1,246.8 854.8 197.1 135.6 61.5 3.2 Plan AbS 589.7 245.0 344.8 1,513.0 966.8 546.2 2,616.1 1,472.6 1,143.6 269.4 197.9 71.5 199711998 1.5 Actual Abs 468.7 391.2 77.5 1,837.2 1,422.9 414.2 2,138.9 1,139.2 999.7 250.5 153.2 97.2 % 1.2 Plan Abs 788.2 147.8 640.4 7,484.6 5,450.1 2,034.8 4,774.7 1,254.9 3,519.8 1,304.9 1,118.7 186.2 199811999 % 0.9 Prellminary Actual % Abs 512.8 224.7 288.2 5,840.0 4,886.2 953.8 4,025.3 1,241.1 2,784.1 1,170.7 1,012.4 158.3 0.8 1999/2000 Plan Abs 629.2 239.1 390.1 4,613.3 3,290.1 1,323.1 3,466.2 1,404.7 2,061.6 1,202.1 1,123.5 78.5 % 0.8
I INDUSTRY own Resources Foreign Resources 2 AGRICULTURE AND FORESTRY Own Resources Foreign Resources 3 WATER RESOURCES own Resources Foreign Resources 4 MANPOWER Own Resources Foreign Resources TRADE, DEVELOPMENT OF BUSINESS ENTERPRISES, FINANCE AND COOPERATIVES Own Resources Foreign Resources
3.6
5.4
3.6
3.1
3.8
3.6
3.9
4.8
8.1
8.6
5.6
6.2
6.3
6.6
6.6
8.7
5.8
6.7
5.6
5.2
5.9
4.2
0.5
0.4
0.6
0.5
0.5
0.5
0.7
0.7
1.4
1.7
1.5
736.3 144.8 591.5
2.7
1,439.6 1,138.8 300.8
4.7
533.7 153.7 380.0
1.7
912.4 805.3 307.1
3.2
401.5 181.8 219.6
1.2
1,880.7 1,286.8 393.9
4.7
549.9 226.2 323.7
1.4
1,380.2 633.6 746.6
3.6
16,687.6 15,811.0 876.6
18.0
11,570.9 10,984.9. 586.1
17.0
19,035.6 18,741.9 293.7
23.1
0
I'
& METEOROLOGY TRANSPORT,

6 GEOPHYSICS Own Resources Foreign Resources 7 MINING AND ENERGY own Resources Foreign Resources 5,225.5 3,606.6 1,619.0 3,581.8 13.1 19.1 5,656.8 3,920.4 1,736.4 4,407.3 1,432.3 2,975.0 14.4 18.4
5,897.9 3,822.2 2,075.7 3,894.8 871.5 3,023.3
19.2
5,382.6 3,678.4 1,704.2
18.7
6,771.2 4,228.5 2,542.7 4,101.5 939.4 3,162.2
19.6
5,655.4 4,020.5 1,635.0 3,252.0 883.6 2,368.4
15.7
6,849.9 4,687.9 2,162.0 4,423.0 1,081.0 3,342.0
17.6
6,847.4 3,962.6 2,884.7 4,808.0 1,066.5 3,741.5
17.9
9,642.6 3,412.0 6,230.6 7,059.5 783.3 6,276.1
10.4
7,366.8 3,042.3 4,324.5 5,875.7 707.7 5,168.1
10.9
8,426.6 2,630.8 5,795.8 6,607.7 774.0 5,633.6
10.2
12.7
803.7
2,778.1
3,111.5 825.0 2,286.5
10.8
11.9
9.0
11.4
12.5
7.6
8.7
8.0
AND POSTS TOURISM,

8 TELECOMMUNICATIONS own Resources Foreign Resources REGIONAL DEVELOPMENT AND TRANSMIGRATION own Resources Foreign Resources 10 ENVIRONMENT AND SPATIAL own Resources Foreign Resources EDUCATION, CULTURE, BELIEF IN 11 ALMIGHTY GOD, YOUTH AND SPORTS
721.9 69.5 652.4
2.6
973.3 299.9 673.4
3.2
1,005.6 78.1 927.7
3.3
458.2 69.9 389.3
1.6
1,043.2 94.7 948.5
3.0
570.9 81.9 488.9
1.6
962.7 107.5 855.2
2.5
884.3 78.7 785.6
2.3
1,181.0 75.9 1,105.1
1.3
1,389.6 59.0 1,330.7
2.0
918.1 82.9 835.2
1.1
5,504.2 5,119.5 384.7 452.3 243.3 209.0
20.1
5,461.2 5,222.5 238.7 419.7 252.3 167.4
17.8
6,139.2 5,615.4 523.6 517.3 277.2 240.1
19.9
6,187.5 5,747.3 440.2 439.0 256.5 182.5
21.5
6,509.1 6,222.6 286.5 615.6 322.4 293.2
18.9
7,156.4 6,558.5 597.9 564.5 321.7 242.8
19.9
7,164.1 6,881.9 282.2 685.8 420.6 265.0
18.4
7,136.8 6,619.6 517.2 645.5 354.0 291.5
18.6
19,091.6 9,926.4 9,165.2 780.0 330.3 449.7
20.6
11,321.3 8,831.3 2,490.1 641.7 272.3 368.5
16.7
14,545.9 11,005.6 3,540.1 932.7 579.2 353.6
17.6
1.7
1.4
1.7
1.5
1.8
1.6
1.8
1.7
0.8
0.9
1.1
3,061.1 2,358.6 702.5
11.2
2,988.4 2,409.0 579.4
9.7
3,359.2 2,608.0 751.2
10.9 5
3,130.3 2,450.9 6 79.4
10.9 _
3,970.6 3,057.4 913.2
11.5
3,849.2 2,960.1 889.1
10.7
Own Resources
Foreign Resources
4,676.9 3,735.4 941.6
12.0
4,268.1 3,223.1 1,045.0
11.1 _
8,367.6 4,845.1 3,522.5
9.0
6,150.4 4,361.0 1,788 4
9.1
8,381.3 4,818.7 3,562.6
10.2
Table 5: Investment Budget (Development Budget) in billions of Rupiah - By Sector

199411995 199511996 1996'1997 1997/1998 1998/1999 199912000
No
Cate~~~~~gory
Plan
Abs A o %
Actual
Abs %
Plan
AbS
Actual
%
PlnAtI
Abs % Abs % Abs
Plan
I
Actual
Abs A % Abs
Plan
PreliminaryActual
% Abs
Plan
%
12 POPULATION Own Resources Foreign Resources SOCIAL WELFARE, HEALTH, ROLE 1 OF WOMEN, CHILDREN AND ADOLESCENT Own Resources Forelgn Resources 14 HOUSING AND HUMAN SETTLEMENTS Own Resources Foreign Resources 15 RELIGION Own Resources Foreign Resources 16 SCIENCE ANDTECHNOLOGY Own Resources Foreign Resources
290.2 244 8 45.5
1.1
269.8 221 5 48.3
0.9
300.3 252 5 47 8
1.0
265.6 234.3 31.3
0.9
328.0 277 9 50.1
1.0
332.7 313.2 19 5
0.9
890.9 634.8 56.1
1.8
348.2 343.1 5.1
0.9
582.3 242.7 339.6
0.6
593.7 205 3 388.5
0.9
594.3 2441 350.3
0.7
1,031.0 816 6 214.4
3.8
987.2 839.4 147.8
3.2
1,051.8 860.2 191.6
3.4
987.1 825.3 161.8
3.4
1,364.9 1,090.5 274.5
4.0
1,277.6 1,012.6 265.0
3.6
2,097.2 1,527 6 569 6
5.4
1,703.1 1,414.1 289.1
4.4
4,204.8 2.5910 1,613 8
4.5
3,556.3 2,285.6 1,270.7
5.2
4,786.9 2,908.1 1,878.8
5.8
888.0 469.4 418.6 121.9 112 3 9.5 529.8 423.7 1061 111.4 101.4 9.9
3.2
1,103.4 614.3 489.1 165.6 159 2 6.4 394.1 358.2 36.0 90.8 90 6 0.2
3.6
1,102.1 514.1 588.0 183.3 139.6 43 7 711.2 500 6 210.6 138.7 130.6
3.6
1,037.6 4659.1 588 5 238.4 194.1 44 3 431.6 396.5 351 117.3 117.2 0.1
3.6
1,325.6 596.3 729.2 253.7 177.9 757 805.6 606.8 198.8 172.9 159.g 13.1
3.8
1,365.8 552.1 813 7 282.0 207.2 74.8 554.8 516.3 38.6 150.8 148.9 2.0
3.8
1,533.8 671.3 862.5 304.0 226.2 77.8 881.8 701.4 180.4 195.0 193 6 1.4
3.9
1,333.4 577 0 756.4 213.3 182.4 31.0 878.9 675 2 301.7 153.3 150.8 2.5
3.5
5,615.2 3,615.4 1,999 7 475.9 2957 220.2 1,144.0 462.9 681 2 167.0 153.0 14.0
6.1
2,565.9 1,324.5 1,241.4 327.8 233.9 93.7 922.5 450.8 471.6 137.0 136.8 02
3.8
3,218.4 1,713.3 1,505.1 627.4 312.7 314.7 900.5 568 1 332.4 230.1 220.8 9.3
3.9
0.4
0.5
0.6
0.8
0.7
0.8
0.8
0.6
0.5
0.5
0.8
1.9
1.3
2.3
1.5
2.3
1.5
2.3
2.3
1.2
1.4
1.1
00
r
)
17 LAW Own Resources Foreign Resources is GOV7TAPPARATUS AND SUPERVISION Own Resources Foreign Resources POLITICS, INTERNATIONAL RELATIONS, INFORMATION, COMMUNICATIONS AND MASS MEDIA Own Resources Foreign Resources NATIONAL DEFENCE AND SECURITY Own Resources Foreign Resources TOTAL Own Resources Foreign Resources Source: World Bank
0.4
0.3
0.5
0.4
0.5
0.4
0.5
0.4
0.2
0.2
0.3
7.9
557.0 468.7 88.3
2.0
567.0 417 9 149.1
1.8
664.4 543.5 120.9
2.2
624.0 528.5 95.5
2.2
818.6 663.0 155.5
2.4
838.8 626.1 215.7
2.3
911.0 689.2 221.7
2.3
857.9 571.3 286.6
2.2
786.8 431.2 355.6
0.8
946.6 360.7 585.9
1.4
900.9 428.4 472.5
1.1
157.3 78.7 786
0.6
219.2 123.3 95 9
0.7
152.7 85.9 66.8
0.5
130.6 62.2 48.5
0.5
183.2 140.0 43.2
0.5
184.3 148.5 35.9
0.5
286.1 148.7 137.4
0.7
308.4 118.8 189.6
0.8
421.8 111.9 310 0
0.5
339.6 91.9 247.7
0.5
154.0 105.5 44.8
0.2
1,154.6 589.3 565.3
4.2
1,278.0 671.4 606.6 30,685.2 20,850.2 8,835.0
4.2
1,317.3 724.1 593.2
4.3
1,594.1 759.0 935.1 28,785 1 18,771.8 9,008.8
5.5
1,531.8 898.7 633.1 34,552.7 22,088.1 12,413.6
4.4
3,496.9 998.9 2,498.0 35,951.4 24,051.3 11,900.1
9.7
1,727.5 1,086.2 641.4 38,927.9 25,801.9 13,026.0
4.4
1,918.7 995.7 923.0 38,358.7 23,973.1 14,385.6
5.0
2,122.8 1,122.8 1,000.0 92,683.8 52,142.1 48,540.8
2.3
2,614.8 1.239.7 1.375.2 67,869.3 41,951.9 25.917.3
3.9
2,277.4 1,253.0 1,024.4 82,448.6 52,448.6 30,000.0
2.8
__
27,397.9 100.0 17,386.0 83.5 10,011.8 38,5
10.0 67.9 32.1
30,783.5 100.8 19,824.8 81.6 11,758.9 38.2
180.0 68.7 31.3
108.0 84.0 36.0
100.0 68.9 33.1
10C 66.8 33.5
100.0 62,5 37.5
108. 56.3 43.7
180.0 61.8 38.2
100.0 63.6 36.4
Table 6: Investment Budget (Development Budget) - By Sector

(in billions of Rupiah) - Constant - Base Price 1993/1994
1994/1995
GDP Deflator 1.08
1995/1996
1.18
1996/1997
1.29
1997/1998
1.61
1998/1999
2.76
1999/2000
3.31
No
Category
Plan
Abs
Actual
Abs
Plan
Abs
Actual
Abs
Plan
Abs
Actual
Abs
Plan
Abs
Actual
Abs
Plan
Abs
Preliminary Actual
Abs
Plan
Abs %
1 INDUSTRY
Own Resources Foreign Resources 2 AGRICULTUREAND FORESTRY Own Resources Foreign Resources 3 WATER RESOURCES Own Resources Foreign Resources 4 MANPOWER Own Resources Foreign Resources OF TRADE, DEVELOPMENT BUSINESS ENTERPRISES, FINANCEAND COOPERATIVES Own Resources Foreign Resources oo a7\ 6 TRANSPORT, METEOROLOGY& GEOPHYSICS Own Resources Foreign Resources 7 MININGANDENERGY Own Resources Foreign Resources 8 TOURISM,POSTS AND TELECOMMUNICATIONS Own Resources Foreign Resources
417.1 133.1 284.0 916.3 613.3 303.0 1,562.0 749.4 812.6 135.6 110.6 25.1 681.7 134.1 547.6 4,838.4 3,339.4 1,499.0 3,316.5 744.2 2,572.3 668.4 64.4 604.0
523.3 207.0 316.3 1,534.7 1,251.1 283.6 1,789.0 925.5 863.5 100.9 97.1 3.9 1,333.0 1,054.4 278.5 5,237.8 3,630.0 1,607.8 4,080.8 1,326.2 2,754.6 901.2 277.7 623.5 5,056.7 4,835.6 221.0 388.6 233.6 155.0 2,767.0 2,230.6 536.5 249.8 205.1
44.7 914.0 777.2 136.8 1,021.7 568.8
421.4 147.1 274.3 935.4 564.8 370.6 1,730.6 740.8 989.7 144.6 112.5 32.0 452.3 130.3 322.0 4,998.3 3,239.2 1,759.1 3,300.7 738.6 2,562.1 852.4 66.2 786.2 5,202.7 4,758.8 443.9 438.4 234.9 203.5 2,846.8 2,210.2 636.6 254.5 214.0
40.5 891.4 729.0 162.4 934.0 435.7
682.5 525.0 157.5 755.1 573.6 181.5 1,607.6 945.3 662.4 118.1 103.1 15.0 773.2 513.0 260.3 4,561.5 3,117.3 1,444.2 2,636.9 699.2 1,937.7 388.3 58.4 329.9 5,243.6 4,870.6 373.1 372.0 217.4 154.7 2,652.8 2,077.0 575.8 225.1 198.6
26.5 836.5 699.4 137.1 879.3 397.5
392.7 161.1 231.6 1,003.4 638.2 365.2 1,796.4 961.1 835.3 145.0 124.2 20.8 311.2 140.9 170.3 5,249.0 3,277.9 1,971.1 3,179.5 728.2 2,451.3 808.7 73.4 735.3 5,045.8 4,823.7 222.1 477.2 249.9 227.3 3,078.0 2,370.1 707.9 254.3 215.4
38.9 1,058.1 845.3 212.8 1,027.6 462.3
878.6 778.8 99.8 1,014.2 796.5 217.7 1,629.2 966.5 662.7 152.8 105.1 47.7 1,302.9 997.5 305.4 4,384.1 3,116.6 1,267.4 2,520.9 685.0 1,836.0 442.5 63.5 379.0 5,547.6 5,084.1 463.5 437.6 249.4 188.2 2,983.9 2,294.7 689.2 257.9 242.8
15.1 990.4 785.0 205.4 1,058.8 428.0
366.3 152.1 214.2 939.7 600.5 339.3 1,624.9 914.6 710.3 167.3 122.9 44.4 341.6 140.5 201,1 4,254.6 2,911.7 1,342.8 2,747.2 671.4 2,075.8 597.9 66.8 531.2 4,449.7 4,274.4 175.3 425.9 261.3 164.6 2,904.9 2,320.1 584.8 429.2 394.3
34.8 1,302.6 948.8 353.8 952.6 416.9
291.1 243.0 48.1 1,141.1 883.8 257.3 1,328.5 707.6 620.9 155.6 95.2 60.4 857.3 393.5 463.7 4,253.0 2,461.3 1,791.8 2,986.3 662.4 2,323.9 536.8 48.9 487.9 4,432.8 4,111.6 321.2 400.9 219.9 181.0 2,651.0 2,001.9 649.1 216.2 213.1
3.2 1,057.8 878.3 179.5 828.2 358.4
285.1 53.5 231.6 2,707.0 1,971.2 735.8 1,726.9 453.9 1,273.0 472.0 404.6 67.3 6,035.6 5,718.5 317.0 3,487.5 1,234.0 2,253.5 2,553.3 283.3 2,269.9 427.2 27.5 399.7 6,905.0 3,590.2 3,314.9 282.1 119.5 162.6 3,026.4 1,752.4 1,274.0 210.6 87.8
122.8 1,520.8 937.1 583.7 2,030.9 1,307.6
185.5 81.3 104.2 2,112.2 1,767.2 345.0 1,455.8 448.9 1,007.0 423.4 366.2 57.2 4,185.0 3,973.0 212.0 2,664.4 1,100.3 1,564.1 2,125.1 255.9 1,869.2 502.6 21.3 481.3 4,094.7 3,194.1 900.6 232.1 98.5 133.6 2,224.5 1,577.3 647.2 214.7 74.2
140.5 1,286.2 826.6 459.6 928.0 479.1
190.0 72.2 117.8 1,392.9 993.4 399.5 1,046.6 424.1 622.5 362.9 339.2 23.7 5,747.5 5,658.8 88.7 2,544.3 794.3 1,749.9 1,995.1 233.7 1,761.4 277.2 25.0 252.2 4,391.9 3,323.0 1,068.9 281.6 174.9 106.8 2,530.6 1,454.9 1,075.7 179.4 73.7
105.8 1,445.3 878.0 567.3 971.8 517.3
0.8
5.6
4.2
1.5
23.1
10.2
8.0
1.1
AND TRANSMIGRATION 5,096.5 9 REGIONALDEVELOPMENT 4,740.3 Own Resources 356.2 Foreign Resources AND SPATIAL 10 ENVIRONMENT Own Resources Foreign Resources GOD, CULTURE, BELIEF IN ALMIGHTY 11 EDUCATION, Own Resources Foreign Resources 12 POPULATION Own Resources
Foreign Resources 13 SOCIAL WELFARE, HEALTH, ROLE OF WOMEN, CHI Own Resources Foreign Resources 14 HOUSING AND HUMAN SETTLEMENTS Own Resources
17.6
418.8 225.3 193.5 2,834.4 2,183.9 650.5 268.7 226.6

42.1 954.7 756.1 198.6 822.2 434.6
1.1
10.2
0.7
5.8
3.9
Table 6: Investment Budget (Development Budget) - By Sector

(in billions of Rupiah) - Constant - Base Price 1993/1994
1994/1995 GDP Deflator
No Category Plan Abs
1995/1996 1.18
Plan Abs Actual Abs
1996/1997 1.29
Plan Abs Actual Abs
1997/1998 1.61
Plan Abs Actual Abs Plan Abs
199811999 2.76
Preliminary Actual Abs
1999/2000 3.31
Plan Abs %
1.08
Actual Abs
Foreign Resources 15 RELIGION Own Resources ForeignResources 16 SCIENCEAND TECHNOLOGY Own Resources Foreign Resources 17 LAW Own Resources Foreign Resources
387.6 112.8 104.0 8.8 490.6 392.3 98.2 103.1 93.9 9.2 515.7 434.0 81.7
452.9 153.3 147.4 5.9 364.9 331.6 33.3 84.0 83.9 0.2 525.0 387.0 138.0
498.3 155.3 118.3 37.0 602.7 424.2 178.5 117.5 110.8 6.7 563.0 460.6 102.4
481.8 202.0 164.5 37.5 365.8 336.0 29.7 99.4 99.3 0.1 528.8 447.9 80.9
565.3 196.6 137.9
630.8 218.6 160.6 58.0 430.1 400.2 29.9 116.9 115.4 1.5 648.7 485.4 163.3
535.7 188.8 140.5 48.3 547.7 435.7 112.0 121.1 120.3 0.9 565.8 428.1 137.7
469.8 132.5 113.3 19.2 544.7 357.3 187.4 95.2 93.7 1.5 532.8 354.9 178.0
723.3 172.1 92.5 79.7 413.8 167.4 246.4 60.4 55.3 5.1 284.6 156.0 128.6
449.0 118.5 84.6 33.9 333.6 163.1 170.6 49.5 49.5 0.1 342.4 130.5 211.9
454.4 189.4 94.4 95.0 271.9 171.5 100.4 69.5 66.7 2.8 272.0 129.3 142.7 0.8
58.7
624.5 470.4 154.1 134.0 123.9 10.1 634.6 514.0 120.6
1.1
0.3
18 GOV'TAPPARATUSAND SUPERVISION
Own Resources ForeignResources
POLITICS, INTERNATIONAL RELATIONS, INFORMATION, COMMUNICATIONS AND MASS oo
1.1
19 MEDIA Own Resources ForeignResources
145.7 72.9 72.8 1,069.1 545.6 523.4 25,368.4 16,098.1 9,270.3
203.0 114.2 88.8 1,183.4 621.7 561.7 28,412.2 1 9305.7 9,106.5
129.4 72.8 56.6 1,116.3 613.6 502,7 26,087.7 16,122.5 9,965.2
110.7 69.6 41.1 1,350.9 643.2 707.7 24,390.3 16,755.7 7,634.6
142.0 108.5 33.5 1,187.5 696.7 490.8 26,746.3 17,123.3 9,623.0
142.9 115.1 27.8 2,710.8 774.3 1,936.5 27,869.3 18,644.4 9,224.9
177.7 92.4 85.3 1,073.0 674.7 398.4 24,178.8 16,088.1 8,090.7
191.6 73.8 117.8 1,191.7 618.4 573.3 23,825.3 14,890.1 8,935.1
152.5 40.4 112.1 767.8 406.1 381.7 33,521.4 18,858.7 14,662.8
122.8 33.2 89 6 945.7 448.4 497.4 24,546.8 15,173.1 9,373.7
46.5 33.1 13.4 687.6 378.3 309.3 24,893.9 15,835.9 9,058.0
0.2
20 NATIONALDEFENCEAND SECURITY
Own Resources ForeignResources TOTAL Own Resources Foreign Resources
Source: World Bank
2.8
100.0 63.6 36.4
Table 7: Current Expenditures - Functional Classification

(in millions of Rupiah)
Categorv 1. Industry 2. Agriculture 3. Forestry 4. Water Resources S. Transport 6. Urban 7. Land Use (Env. & Spatial) 8. Health 9. Education & Training 10. Manpower 11. Regional Development 12. Transfer to Regions 13. Mining & Energy 14. Tourism & Posts 1. Science & Technology 16.Trade, Finance & Supports for SME 17. National Defence 18. Government Apparatus & Supervision 19. Politics, Information Mass Media 28. Religion 21. Others5/ Sub Total 22. Debt Service Payments 24. Petroleum Subsidies 25.BuiogandOtherExp.41 TOTAL Of which: Personnel Expenditures Wages & Salaries Supplement (Tice) Materials Pensions
1994/1995 1) Abs % 45,823 91,449 965,587 21,925 169,066 9,845 130,249 554,198 2,697,323 90,946 260,372 7,272,400 81,781 18,470 230,298 3,641,432 0.1 0.2 2.2 0.05 0.4 0.02 0.3 1.3 6.1 0.2 0.6 16.5 0.2 0.04 0.5 8.3
1995/1996 1) Abs % 48,488 104,157 74,729 23,583 186,748 10,714 150,590 643,682 3,102,570 103,936 126,511 8,226,600 103,371 20,634 260,385 4,134,570 0.1 0.2 0.1 0.0 0.4 0.0 0.3 1.3 6.2 0.2 0.3 16.3 0.2 0.04 0.5 8.2
1996/1997 1) Abs % 54,467 125,072 344,807 26,227 254,604 6,420 179,961 757,109 3,692,139 117,411 165,802 9,357,500 114,430 41,529 307,344 4,659,427 0.1 0.2 0.6 0.04 0.4 0.01 0.3 1.2 5.9 0.2 0.3 15.0 0.2 0.08 0.5 7.4
199711998 1) Abs % 71,131 174,028 155,811 30,678 259,295 16,667 209,091 918,074 4,501,563 172,096 177,439 11,060,500 159,092 30,000 350,937 5,925,099 0.1 0.2 0.2 0.03 0.3 0.02 0.2 1.0 5.0 0.2 0.2 12.3 0.2 0.03 0.4 6.6
[1998/1999 2) Abs 98,696 234,875 494,896 42,685 298,816 25,347 252,871 1,084,233 5,013,777 302,817 220,857 14,194,183 292,476 60,205 435,358 7,461,049
% 0.1 0.2 0.3 0.03 0.2 0.02 0.2 0.7 3.4 0.2 0.1 9.6 0.2 0.04 0.3 5.1
1999/2000 3) Abs 108,135 265,884 478,043 50,074 319,922 27,804 424,764 1,269,591 6,045,226 391,589 251,441 19,497,600 341,303 127,590 498,473 12,176,992
% 0.1 0.2 0.3 0.04 0.2 0.02 0.3 0.9 4.4 0.3 0.2 14.2 0.2 0.09 0.4 8.9
3,896,471 1,974,446
8.8 4.5
4,469,865 2,303,622
8.9 4.6
5,249,734 3,809,081
8.4 6.1
6,280,223 3,483,574
7.0 3.9
8,432,394 6,010,420
5.7 4.1
9,909,685 6,423,756
7.2 4.7
800,590
1.8
923,091
1.8
1,099,508
1.8
1,465,035
1.6
2,439,381
1.7
2,710,592
2.0
777,271 456,914 24,186.856 18,402,500 686,800 792,900 44.069,056
1.8 1.0 54.9 41.8 1.6 1.8 100.0
163,299 521,959 26,403,106 22,108,600 0 1,923,300 50,435.006
1.7 1.0 52.4 43.8

-
995,693 550,683 31,915,949 27,491,200 1,416,100 1,732,600 62.555.849
1.6 0.9 51.0 43.9 2.3 2.8 100.0
1,164,336 697,139 37,308.808 31,112,304 9,814,300 11,374,389 89,609,800
1.3 0.8 41.6 34.7 11.0 12.7 100.0
1,443,346 893,789 49.732,470 55,797,503 27,185,015 15,002,163 147,717,151
1.0 0.6 33.7 37.8 18.4 10.2 100.0
1,741,627 1,045,608 64.105.700 44,810,900 9,985,800 18,253,100 137,155,500
1.3 0.8 46.7 32.7 7.3 13.3 100.0
3.8 100.0
8,959,361 973,200 4,311,900 2,662,939
20.3 2.2 9.8 6.0
9,220,115 733,500 5,175,100 3,047,785
18.3 1.5 10.3 6.0
13,004,500 767,700 8,108,500 683,000
20.8 1.2 13.0 1.1
12,256,141 787,900 8,999,300 4,204,967
13.7 0.9 10.0 4.7
17,685,216 1,606,165 11,057,956 5,188,950
12.0 1.1 7.5 3.5
22,695,569 2,087,100 11,039,000 8,786,431
16.5 1.5 8.0 6.4
Notes: 1)PAN=actual (audited) StateBudgetReahiationapproved parliament,2) Prelminary Actual;3)RevisedBudget; by 4/ Other RoutineEpeodisu,esincludesGeneralElectionEnpenses; Girno Post,;Free Porto(BebasPoto); Inspecton ofPre-Shipment; NationalSportCoiritte (Koni); RiceMaintemance (Bulog); Sute Railways (Penrnka) and othens. Co 5/ OthersectoesincludeMeteorology, Geophysics, Searchand Rescue;and La,s Source: WorldBank
88
Table 8: Current Expenditures - Functional Classification

(in millions of Rupiab) - Constant - 1993 price
Category GDP D.eator 1. Industry 2. Agriculture 3. Forestry 4. Water Resources 5. Transport 6. Urban 7. Land Use (Env. & Spatial) 8. Health 9. Education & Training 10. Manpower 11. Regional Development 12. Transfer to Regions 13. Mining & Energy 14. Tourism& Posts 15. Science & Technology 16.Trade,Finance&Supports for SME 17. National Defence 18. Government Apparatus & Supervision 19. Politics, Information Mass Media 20. Religion 21. OthersS/ Sub Total 22. Debt Service Payments 24. Petroleum Subsidies 25. Bulog and Other Exp.4/ TOTAL Of which: Personnel Expenditures Wages & Salaries Supplement (rice) Materials Pensions
N5t.t: I) PAN= actral (audited) - Slete Budget
1994/1995 1) 1.08
Abs 7
1995/1996 1) 1.18
ADS
1996/1997 1) 1.29
AbS
1997/1998 1) 1.61
Abs
1998/1999 2) 2.76
ADs
19992000 3) 3.31
AD
42.429 84,675 894,062 20,301 156,543 9,116 120,601 513,146 2,497,522 84,209 241,085 6,733,704 75,723 17,102 213,239 3,371,696 3,607,843 1.828,191 741,287 719,695 423,069 22,395,237 17,039,352 635,926 734,167 40,804,681 8,295,705 901,111 3,998,981 2,465,684
0.1 0.2 2.2 0.05 0.38 0.02 0.30 1.3 6.1 0.2 0.6 16.5 0.2 0.04 0.5 0.3 8.8 4.5 1.8 1.8 1.0 54.9 41.8 1.6 1.8 100.0 20.3 2.2 9.8 6.0
41,092 88,269 63,330 19,985 158,261 9,080 127,619 545,493 2,629,297 88,081 107,213 6,971,695 87,603 17,486 220,665 3,503,873 3,788,021 1,952,222 782,281 731,610 442,338 22,375,514 18,736,102 0 1,629,915 42.741,531 7,813,657 621,610 4,385,678 2,582,860
0.1 0.2 0.1 0.0 0.4 0.0 0.3 1.3 6.2 0.2 0.3 16.3 0.2 0.04 0.5 8.2 8.9 4.6 1.8 1.7 1.0 52.4 43.8 . 3.8 100.0 18.3 1.5 10.3 6.0
42,223 96,955 267,293 20,331 197,367 4,977 139,505 586,906 2,862,123 91,016 128,529 7,253,876 88,706 37,620 238,251 3,611,959 4,069,561 2,952,776 852,332 771,855 426,886 24,741,046 21,311,008 1,097,752 1,343,101 48492 ,906 10,081,008 595,116 6,285,659 529,457
0.1 0.2 0.6 0.04 0.41 0.01 0.29 1.2 5.9 0.2 0.3 15.0 0.2 0.08 0.5 7.4 8.4 6.1 1.8 1.6 0.9 51.0 43.9 2.3 2.8 100.0 20.8 1.2 13.0 1.1
44,181 108,092 96,777 19,055 161,053 10,352 129,870 570,232 2,800,350 106,892 110.210 6,869,876 98,815 18,634 217,974 3,680,186 3,900,760 2,163,711 909,959 723,190 433,005 23,173,173 19,324,412 6,095,839 7,064,838 55,658,261 7,612,510 489,379 5,589,627 2,611,781
0.079 0.194 0.174 0.034 0.289 0.019 0,233 1.0 5.0 0.2 0.2 12.3 0.2 0.03 0.4 6.6 7.0 3.9 1.6 1,3 0.8 41.6 34.7 11.0 12.7 100.0 13.7 0.9 10.0 4.7
35,759 85,100 179,310 15,465 108,267 9.184 91,620 392,838 1,816,586 109,716 80,021 5,142,820 105,970 21,813 157,738 2,703,278 3,055,215 2,177,688 883,834 522,951 323,837 18.019,011 20,216,487 9,049,643 5,435,566 53.520,707 6,407,687 581,944 4,006,506 1,800,054
0.1 0.2 0.3 0.03 0.20 0.02 0.17 0.7 3.4 0.2 0.1 9.6 0.2 0.04 0.3 5.1 5.7 4.1 1.7 1.0 0.6 33.7 37.8 18.4 10.2 100.0 12.0 1.1 7.5 3.5
32,669 80,327 144,424 15,128 96,653 8,400 128,328 383,562 1,826,352 118,305 75,964 5,890,514 103,113 38,547 150,596 3,678,850 2,993,863 1,940,712 818,910 526,171 315,894 19,367,281 13,538,036 3,016,858 5,514,532 41,436.707 6,856,667 630,544 3,335,045 2,654.511
0.1 0.2 0.3 0.04 0.23 0.02 0.31 0.9 4.4 0.3 0.2 14.2 0.2 0.09 0.4 8.9 7.2 4.7 2.0 1.3 0.8 46.7 32.7 7.3 13.3 100.0 16.5 1.5 8.0 6.4
Realizatio approved parliament; 2) Preliminary Actual; 3) Revised by Budget; 4/ Other Routine Expenditares includes Gener-A Bleckon Expenses; Post;Fee Poeot(Beba Poen); Impoctdon Pe-Shipm.. NationalSportCommitte(Koi); Giro of RicoMaintennanr (Balog); Stati RailwaysCo.(P-nrka) and othe-. S/ Other sectos i=clsudeMeteology, Geophysics, Search and Resc-e; amd Law Soaroe WorldBank
89
Table 9: Current Expenditures - Economic Classification

(in millions of Rupiah)
Category
1994/1995 1) A OS 44,069,000 24,186,800 12,595,500 8,959,361 973,200 2,662,939 4,318,900 7,272,400 6,918,100 354,300 19,882,200 18,402,500 686,800 792,900
7 100.0 54.9 28.6 20.3 2.2 6.0 9.8 16.5 15.7 0.8 45.1 41.8 1.6 1.8
1995/1996 1) Abs8h 08~'i 50,435.000 26,403,100 13,001,400 9,220,115 733,500 3,047,785 5,175,100 8,226,600 7,807,200 419,400 24,031,900 22,108,600 0 1,923,300 100.0 52.4 25.8 18.3 1.5 6.0 10.3 16.3 15.5 0.8 47.6 43.8 , 3.8
1996/1997 1) ttS ' 62,561.100 31,921,200 14,455,200 10,153,007 767,700 3,534,493 8,108,500 9,357,500 8,873,800 483,700 30,639,900 27,491,200 1,416,080 1,732,620 100.0 51.0 23.1 16.2 1.2 5.6 13.0 15.0 14.2 0.8 49.0 43.9 2.3 2.8
1997/1998 1) AOS ve 89.609,800 37,308,808 17,249,008 12,256,141 787,900 4,204,967 8,999,300 11,060,500 10,519,800 540,700 52,300,993 31,112,304 9,814,300 11,374,389 100.0 41.6 19.2 13.7 0.9 4.7 10.0 10.9 11.7 0.6 58.4 35.7 11.3 13.0
-A
1998/1999 2) A %s u 100.0 33.7 16.6 12.0 1.1 3.5 7.5 9.6 9.1 0.5 66.3 55.2 5.1 8.9
1999/2000 3) ADWS 00 137,155.500 64,105,700 33,569,100 22,695,569 2,087,100 8,786,431 11,039,000 19,497,600 18,696,800 800,800 73,049,800 44,810,900 9,985,800 18,253,100 100.0 46.7 24.5 16.5 1.5 6.4 8.0 14.2 13.6 0.6 53.3 55.2 5.1 8.9
TOTAL Ordinary Expenditures Personnel Expenditures Wages & Salaries Supplement (rice) Pensions Materials Transfer to Regions Personnel Non-Personnel Other Debt Service Payments Petroleum Subsidies Bulog and Other Exp.4/
147.7i7.151 49,732,470 24,480,331 17,685.216 1,606,165 5,188,950 11,057,956 14,194,183 13,512,155 682,028 97,984,681 55,797,503 27,185,015 15,002,163
Notes I) PAN= arota (audited) - Stte Budget Realiation approved by paliameot 2) Preliisnasy Acta; 3) Revised Badget; 4/ Other Ron,io Expendirures includes General Election Espeoste; Giro Post; Feen Por (Bnbs Porto); Inspectdon of Pre-Shipent; Rico Maintenance (Bolog); State Railways Co. (Penastk) and othert. Sou-r: World Bank
National Sport Comitne (Koni);
90
Table 10: Current Expenditures - Economic Classification

(in millions of Rupiah) - Constant- 1993/1994 Price
Category GDPDeflator
1994/1995 1) 1.08 Abs % 40w804c630 100.0
199S/1996 1) 1.18 Abs 42,741,525 % 100.0
199619971) 1.29 Abs % 48.496,977 100.0
1997/1998 1) 1.S1 Abs 55,658,261 % 100.0
19981199 2) 2.76 Abs 53.520.707 % 100.0
1sssnooo 3) 3.31 Abs % 41.436.707 100.0
TOTAL Of which: Ordinary Expenditures Personnel Expenditures Wages & Salaries Supplement (rice) Pensions Materials Transfer to Regions Personnel Non-Personnel Other Debt Service Payments Petroleum Subsidies Bulog and Other Exp.4/
22,395,185 11,662,500 8,295,705 901,111 2,465,684 3,998,981 6,733,704 6,405,648 328,056 18,409,444 17,039,352 635,926 734,167
54.9 28.6 20.3 2.2 6.0 9.8 16.5 15.7 0.8 45.1 41.8 1.6 1.8
22,375,508 11,018,136 7,813,657 621,610 2,582,868 4,3S5,67S 6,971,695 6,616,271 355,424 20,366,017 18,736,102 0 1,629,915
52.4 25.8 18.3 1.5 6.0 10.3 16.3 15.5 0.8 47.6 43.8
-
24,745,116 11,205,581 7,870,548 595,116 2,739,917 6,285,659 7,253,876 6,878,915 374,961 23,751,860 21,311,008 1,097.736 1,343,116
51.0 23.1 16.2 1.2 5.6 13.0 15.0 14.2 0.8 49.0 43.9 2.3 2.8
23,173,173 10,713,669 7,612,510 489,379 2,611,751 5,589,627 6,869,876 6,534,037 335,839 32,485,089 19,324.412 6,095,839 7,064,838
41.6 19.2 13.7 0.9 4.7 10.0 12.3 11.7 0.6 58.4 35.7 11.3 13.0
18,019,011 8,869,685 6,407,687 581,944 1,S80,054 4,006,506 5,142,820 4,895,708 247,112 35,501,696 20,216,487 9,849,643 5,435,566
33.7 16.6 12.0 1.1 3.5 7.5 9.6 9.1 0.5 66.3 37.8 18.4 10.2
19,367,281 10,141,722 6,856,667 630,544 2,654.511 3,335,045 5,890,514 5,648,580 241,934 22,069,426 13,538,036 3,016,858 5,514,532
46.7 24.5 16.5 1.5 6.4 S.0 14.2 13.6 0,6 53.3 32.7 7.3 13.3
3.8
Notes: I)PAN=actual (audied)- StatsBudgnRelinationapproved psalianent;2) Preliminary by Acthsl; Revised 3) Budgeti 41OtherRoadneExpenditures includesGencral ElectionEpenses; Gie PostiPree Porto(Bbas Porto); Inspeotin of Pnc-Shipmcnt National Conmittc (Koni); Sport RiseMaitsenance (Bulog); StateRailways (Pernnka)and othe;s. Co. So.=ce:WorldBank
91

Indonesia Public Spending in A Time of Change

Загружено:

Сведения о документе

Исходное описание:

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Indonesia Public Spending in A Time of Change

Загружено:

Авторское право:

Доступные форматы

Public Disclosure Authorized

No. Report 19845-IND

Indonesia in PublicSpending a Timeof Change

Public Disclosure Authorized

Public Disclosure Authorized

Public Disclosure Authorized

Document of the WortdBank

US$=1 Rp. 7,475 ABBREVIATIONS AND ACRONYMS

ADB APBN BAKUN BAPPENAS BEPEKA BPKP DAPP

NGOs O&M OECD PER PLN PUMA

Subsidy Daerah Otonom

GFS GTZ GDP IG IPPs IFIs IMF KKN

- Medium-Term Expenditure Framework - National Power Corporation

April 1 to March 31 April 1 to March 31

INDONESIA PUBLIC SPENDING IN A TIME OF CHANGE TABLE OF CONTENTS

Figure Figure Figure Figure Figure Figure

1.4: 1.5: 2.1: 2.2: 3.1: 3.2:

Public Spending in a Time of Change

PUBLIC SPENDING IN A CRISIS

he Governmentis facing a grimfiscal sitiuation.The crisis has left it deeply in debt,

and several bilateral agencies on budgetary

domestic revenues excluding privatization

See D. Kaufman, G. Mehrez, and S. Schmulkler

(1998). "The East Asian Crisis: Was it Expected."

These issues were present even prior to the

Figure 1.1. Government subsidies have increased since the crisis...

Some Rp. 16 trillion in arrears are outstanding to Pertamina and PLN.

Public Spending In A Crisis

These are normally due to the honoraria and

transportationcosts that are associated with

Figure 1.3 Budget not stimulating enough

ei2GOI ll ! g Budget Target

QActualFiscalDeficit MCyclically Neutral \ \ , | FiscalImpulse

Cyclically neutral Target

ENVISAGED FISCAL STIMULUS NOT DID

Public Spending in a Time of Change

Figure 1.4: Indonesia's fisca developments at a glance

But Indonesia usually underbudgets its revenues

~ ~~ ~ ~ ~ ~ ~~ ~ ~ ~ ~~~~~~~~~cta ~~ ~ -~ - , . .. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1200/0 Deficit / .

Note: all 1998/99numbers are preliminary; all 1999/2000numbers are budgeted

..and spends less than planned

stimulus too small.

60.00/ 140.000 120.WW ^;;/

The crisis has led debt to explode....

Debt includes bank restructuring bonds from 1999/2000

is accompanied by the continued

declining inflation, Indonesia's debt burden could be maintained

----------.---- ---- --------------------------FY04CQO FY0700B FY10111

1.19 Prudent debt management, in a fiscally constrained

Public Spending in a Time of Change

Public Spending In A Crisis

Public Spending in a Time of Change

they have actually gone down over the past 4 years.

Box 1.2: Contingentliabilities in Malaysia

New off-budget obligations

faced with due to this foreign exchange guarantee.

expenditure and revenue sides

Public Spending in a Time of Change

of new bonds, but neither is show in the budget.

exemptions. Under the old Government, several industries obtained exemption

(mimeo), April 1999. Also, see Easterly William

received exemptions, to stimulate design. To

~ ~~ ~ ~ ~ ~ ~~ ~ ~ ~ ~~~~~~~cta ~ -~ - , . .. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1200/0 Deficit / .