MACROECONOMICS,
TRADE AND
INVESTMENT
MACROECONOMICS, TRADE AND INVESTMENT
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT
Subnational Debt
Financing in Indonesia
Ahya Ihsan, Ratih Dwi Rahmadanti,
Assyifa Szami Ilman
© 2023 International Bank for Reconstruction and Development / The World Bank
1818 H Street NW, Washington DC 20433
Telephone: 202-473-1000; Internet: www.worldbank.org
Some rights reserved.
This work is a product of the staff of The World Bank with external contributions. The findings,
interpretations, and conclusions expressed in this work do not necessarily reflect the views of
The World Bank, its Board of Executive Directors, or the governments they represent. The World
Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors,
denominations, and other information shown on any map in this work do not imply any judgment
on the part of The World Bank concerning the legal status of any territory or the endorsement or
acceptance of such boundaries.
Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges
and immunities of The World Bank, all of which are specifically reserved.
Rights and Permissions
This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO),
http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution
license, you are free to copy, distribute, transmit, and adapt this work, including for commercial
purposes, under the following conditions:
Attribution—Please cite the work as follows: 2023. Subnational Debt Financing in Indonesia. EFI
Insight-Finance. Washington, DC: World Bank.
Translations—If you create a translation of this work, please add the following disclaimer along
with the attribution: This translation was not created by The World Bank and should not be
considered an official World Bank translation. The World Bank shall not be liable for any content
or error in this translation.
Adaptations—If you create an adaptation of this work, please add the following disclaimer along
with the attribution: This is an adaptation of an original work by The World Bank. Views and
opinions expressed in the adaptation are the sole responsibility of the author or authors of the
adaptation and are not endorsed by The World Bank.
Third-party content—The World Bank does not necessarily own each component of the content
contained within the work. The World Bank therefore does not warrant that the use of any third-
party-owned individual component or part contained in the work will not infringe on the rights
of those third parties. The risk of claims resulting from such infringement rests solely with you.
If you wish to reuse a component of the work, it is your responsibility to determine whether
permission is needed for that reuse and to obtain permission from the copyright owner. Examples
of components can include, but are not limited to, tables, figures, or images.
All queries on rights and licenses should be addressed to World Bank Publications, The World
Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org.
Cover design and layout: Diego Catto / www.diegocatto.com. Typset by Arsianti.
Photos: Akhmad Dody Firmansyah/www.shutterstock.com and memet saputro/unsplash.com.
>>>
Acknowledgments
This policy note was prepared by Ahya Ihsan (Senior Economist), Ratih Dwi Rahmadanti (Economist),
Assyifa Szami Ilman (Junior Professional Associate) with the guidance from Lars Christian Moller
(Practice Manager, Macroeconomics, Trade and Investment Global Practice), Alma Kanani (Practice
Manager, Governance Global Practice), and Habib Rab (Lead Economist, World Bank, Indonesia).
The team received inputs from Jurgen Blum (Senior Governance Specialist), Lars Jessen (Lead Debt
Specialist), and Adri Asmoro Laksono Poesoro (Urban Specialist), Andyan Diwangkari (Consultant),
and Neni Lestari (Senior Financial Sector Specialist), Rong Qian (Senior Economist). The team
is grateful for peer-reviewed comments provided by Fernando Andres Blanco Cossio (Principal
Economist, IFC), Griya Rufianne (Urban Specialist), Hakan Yavuz (Senior Debt Specialist), Lilia
Razlog (Senior Debt Specialist), Julie Stalker (First Secretary - Development, DFAT Indonesia),
Ming Zhang (Practice Manager, Urban, Disaster Risk Management, Resilience and Land Global
Practice). The team received valuable inputs and support and cooperation from the Directorate
General Fiscal Balance (DGFB) of the Ministry of Finance, Indonesia, in preparing this note and
implementing subnational debt management capacity survey, including Bhimantara Widyajala
(Director of Capacity and Transfers, DG Fiscal Balance, Ministry of Finance), Dudi Hermawan
(Deputy Director, DG Fiscal Balance, Ministry of Finance), Faisal Fahmi (Head of Section, DG Fiscal
Balance, Ministry of Finance), Irfan Sofi (Analyst, DG Fiscal Balance, Ministry of Finance), Brama
Yudha Kusmara (Analyst, DG Fiscal Balance, Ministry of Finance), Dewi Puspita (Senior Policy
Analyst, Fiscal Policy Agency, Ministry of Finance) and other colleagues from the Ministry of Finance
who contributed to this report. Ariza Nurana (Program Assistant) provided administrative support,
Chris Stewart (Consultant, World Bank) edited the report, and Arsianti (Consultant, World Bank)
designed and formatted the report.
This analysis was produced with financial support from the European Union and the Governments
of Canada and Switzerland under the Public Financial Management Multi-Donor Trust Fund (PFM
MDTF) and from the Government of Australia (Department for Foreign Affairs and Trade) under the
Australia-World Bank Indonesia Partnership (ABIP).
This note serves as one of the background notes for the Decentralization Policy Review (DPR), a
flagship report that examines broader intergovernmental transfer and subnational finance issues in
Indonesia.
>>>
Contents
Acknowledgments 3
Abbreviations and Acronyms 7
Abstract 8
1. Introduction 9
2. Indonesia’s Subnational Debt: Trends and Composition 12
3. Analytical Framework 16
3.1 Which SNGs have borrowing capacity? 18
3.2 Do high-borrowing capacity SNGs have sound
22
debt management capacity?
3.3 Is there a sufficient supply of credits for
26
subnational borrowing?
3.4 Do borrowing rules discourage subnational
28
borrowing?
4. Conclusions and Recommendations 31
4.1 Conclusion 31
4.2 Suggested Reforms 31
References 34
Appendix 1 37
Appendix 2 38
Appendix 3 42
Appendix 4 45
>>>
Figures
Figure 2.1: Indonesia’s SNG debt increased between 2018 and 2021 13
Figure 2.2: …but remains one of the lowest in the world 13
Figure 2.3: Indonesia’s SNG Debt is also lower relative to lower-middle-income country
13
peers
Figure 2.4: Indonesia’s composition of SNG Debt similar to those in low-income countries 13
Figure 3.1: Framework for Identifying Constraints to Subnational Borrowing 17
Figure 3.2: SNGs With High-Borrowing Capacity are Heavily Concentrated in Java-Bali 19
Figure 3.3: Borrowing capacity of districts do not always associate with the level of debt… 20
Figure 3.4: … Similarly for provinces, borrowing capacity do not always associate with the
20
level of debt
Figure 3.5: SNG’s Untapped Borrowing Space Equals an Extra Three to Five Percent of
21
Revenue
Figure 3.6: SNGs Focused on Back-office Functions 23
Figure 3.7: Specialized Knowledge in Debt Management is a Key Constraint 23
Figure 3.8: SNGs have limited understanding of rules and procedures of subnational
25
borrowing
Figure 3.9: Most SNGs do not have project investment list 25
Figure 3.10: Subnational Government Borrowing by Source 26
Figure 3.11: Sectoral Composition of Subnational Borrowing from PT. SMI and BPDs 26
Figure 3.12: Composition of Time Deposits of Third-party Funds of Commercial Banks (%) 27
Figure 3.13: Third-party Funds in Commercial Banks by Province (%) 27
Figure 3.14: Timeline for Subnational Debt Authorization Process 30
Figure A.3.1: Schematic Selection Process of Sample 44
Figure B.1: Relationship Between Subnational Debt and Selected Fiscal Variables 15
Figure B.3.1: Categories of Control Over Subnational Debt 29
>>>
Tables
Table 1.1: Indonesia’s Subnational Government’s Fiscal Structure 2019-20 10
Table 3.1: Statistics (the Average Values) Across Indicators of Borrowing Capacity 18
Table 3.2: Subnational Borrowing Capacity Ranking by Main Island 19
Table 3.3: Fiscal Space is the Strongest Predictor for SNGs Gross Cumulative Loans 20
Table 3.4: SNG Borrowing by Regional Level 26
Table A.1.1: Descriptive Statistics of Size of Subnational Borrowing 2016-21 37
Table A.1.2: Subnational Borrowing Outcomes 2016-21 37
Table A.1.3: Subnational Borrowing Source 2016-21 37
Table A.2.1: List of High-borrowing Capacity SNGs and the Actual Borrowers
39
(in alphabetical order)
Table A.2.2: Statistics (the Average Values) Across Indicators of Borrowing Capacity 40
Table A.3.1: Key Criteria of Sample Selection Process 42
Table A.3.2: List of Survey Respondents 43
Table A.3.3: Core Minimum Requirements for Effective Subnational Debt Management 44
>>>
Boxes
Box 2.1: Country Characteristics and SNG Debt 14
Box 3.1: Previous Attempts to Issue Subnational Bonds in DKI Jakarta and West Java 27
Box 3.2: Categories of Control Over Subnational Debt 29
Box 4.1: Monitoring Subnational Borrowing 32
Box 4.2: Selected Country Experiences with Creditworthiness Enhancement 33
>>>
Abbreviations and Acronyms
APBD Anggaran Pendapatan dan Belanja Daerah MoHA Ministry of Home Affairs
(Subnational Budget)
Bappeda Badan Perencanaan dan Pembangunan Daerah MRT Mass Rapid Transit
(Regional Planning and Development Agency)
Kementerian / Kementerian Perencanaan dan Pembangunan NALAS Norwegian Association of Latin American Studies
Bappenas Nasional (Ministry of National Development and
Planning)
BI Bank Indonesia NBFI NonBank Financial Institution
BPD Bank Pembangunan Daerah (Regional OECD Organisation for Economic Co-operation and
Development Bank) Development
BPK Badan Pemeriksa Keuangan (Supreme Audit OJK Otoritas Jasa Keuangan (Financial Service
Agency) Authority)
BPKAD Badan Pengelola Keuangan dan Aset Daerah OSR Own Source Revenue
(Regional Asset and Finance Management
Agency)
CG Central Government PDF Project Development Fund
DAK Fisik Dana Alokasi Khusus Fisik (Special Allocation PEN Pemulihan Ekonomi Nasional (National Economic
Grants for Physical Infrastructure) Recovery)
DAU Dana Alokasi Umum (General Allocation Funds) PIM Public Investment Management
DBH Dana Bagi Hasil (Revenue Sharing Transfer) PMK Peraturan Menteri Keuangan (Minister of Finance
Regulation)
DGFB Directorate General of Fiscal Balance PPPs Public-Private Partnerships
DKI Jakarta Daerah Khusus Ibukota Jakarta (Special Region Provinsi Province
for Capital City of Jakarta)
DMO Debt Management Office PT. SMI PT. Sarana Multi Infrastruktur
DPRD Dewan Perwakilan Rakyat Daerah (Local RIDF Regional Infrastructure Development Fund
Parliament)
FINDETER Territorial Financing Institution (Colombia) RPJMN Rencana Pembangunan Jangka Menengah
Nasional (National Medium-term Development
Plan)
GDP Gross Domestic Product S&P Standard and Poor’s
GoI Government of Indonesia SILPA Sisa Lebih Pembiayaan Anggaran (Unspent Cash
Balance)
IEP Indonesia Economic Prospects SLA Subsidiary Loan Agreement
IGFS Intergovernmental Fiscal System SN Subnational Debt Management Performance
DeMPA Assessment
InfraSAP Infrastructure Sector Assessment Program SNG Subnational Government
JICA Japan International Cooperation Agency SUSPI Statistik Utang Sektor Publik Indonesia (Public
Sector Debt Statistics)
Kabupaten District (rural local government) UCLG (The World Organization of) United Cities and
Local Governments
Kota City (urban local government) US$ U.S. Dollar
KPPOD Komite Pemantauan Pelaksanaan Otonomi Daerah UU Undang-Undang (Law)
(Regional Autonomy Implementation Monitoring
Committee)
LGUGC Local Government Unit Guarantee Corporation WOFI World Observatory Subnational Government
(Philippines) Finance and Investment
MOF Ministry of Finance
>>>
Abstract
Indonesia’s subnational governments (SNGs) play an important role in delivering infrastructure
services and are responsible for implementing one-half of total government capital expenditure.
Despite significant gaps, infrastructure investments by SNGs in Indonesia remain limited and
rely mostly on central government transfers, while debt financing plays only a small role that
has, until now, been constrained by multiple challenges. This note focuses on debt carrying
and management capacities of SNGs and highlights several issues that could help promote
sustainable local financing. They include: (i) strengthening the framework and data capacity
for determining borrowing capacity and creditworthiness across SNGs; (ii) building debt
management capacity‒especially for high-borrowing capacity SNGs; (iii) improving effectiveness
of financial intermediary institutions and deepening financial markets; and (iv) simplifying
approval processes for borrowing. Addressing some of these issues could help enhance SNGs’
access to local debt markets, although this would have to be done prudently and focused first
on more creditworthy SNGs.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 8
1. >>>
Introduction
SNGs play an increasingly important role in delivering infrastructure services in
developing countries. This is thanks to rapid urbanization and greater fiscal decentralization in
past decades that generates demand for infrastructure investment to support economic growth.
Globally, subnational public investment accounts for 40 percent of total public investment. This
share is higher in OECD countries (57 percent) and varies by government structure. Subnational
public investment accounts for a higher share in countries with a federal system of government
(63 percent) compared to countries with a unitary system (34 percent) (OECD 2016).1 In China,
infrastructure investment as a share of GDP was over 10 percent per year during the period
1990s-2010‒with subnational governments playing a dominant role in urban infrastructure (Liu
and Pradelli 2012).
Debt instruments can serve as an effective means of financing infrastructure services
at the sub-national level. Along with expenditure assignments, revenue assignments, and
intergovernmental transfers, subnational borrowing is a key component of subnational finance.
Borrowing can be a useful tool for funding public investments since infrastructure projects
usually require significant upfront costs that may take several years to complete. By spreading
the associated costs over time, borrowing can help to minimize financial burdens and promote
intergenerational equity by distributing the costs of infrastructure services among different
generations that will benefit from them Martinez-Vazquez and Civelek 2019).
Subnational borrowing must be done prudently, however, to minimize fiscal risks that can
lead to debt distress. Despite its benefits for financing infrastructure investment, irresponsible
and undisciplined subnational borrowing can pose substantial risks to fiscal sustainability that
can lead to macroeconomic instability and disrupt public service delivery. For example, Argentina,
Brazil, Mexico, and the Russian Federation experienced countrywide subnational debt crises
during the 1990s due to excessive debts incurred by sub-central governments (Martinez-
Vazquez and Civelek 2019). In Brazil and Mexico, subnational debt crises occurred frequently
due to a combination of expansionary fiscal policies driven by increased spending obligations
from decentralization and a lack of effective controls on indebtedness (Blanco, 2015).
1. To the best of the authors’ knowledge, this joint project between the OECD and the World Organization of United Cities and Local Governments (UCLG) provided the
most recent and comprehensive statistics on subnational government debt, covering 101 countries.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 9
In Indonesia, SNGs are playing an increasingly important and, to a lesser extent, own source revenue. During 2020-21
role in delivering basic infrastructure services. Following there were few SNGs (less than 10 percent) that borrowed
decentralization reforms in 2001, SNGs have become through intermediary financial institutions and on-lending from
responsible for implementing 43 percent of public expenditure central government for infrastructure investment.
(2015-18), compared to 23 percent pre-decentralization
(1994-2000) (World Bank 2020a). Cities (kota) and districts Increased SNG borrowing could help alleviate fiscal
(kabupaten) are responsible for delivering basic infrastructure constraints to Indonesia’s infrastructure investments.
services and executing one-half of public investment SNGs are highly dependent on transfers (58 percent of total
including housing, drinking water, sewerage, local road revenue) while own-source revenue is small (24 percent of
networks, transportation systems, water and sanitation, and total revenue) (Table 1.1). On the other hand, a sizeable
small irrigation networks. In large cities and metropolitan share of expenditure is earmarked for mandatory spending
areas, investment needs are significant due to decades of (for example, 20 percent for education, 10 percent for health)
underinvestment, rapid urbanization, and increasing demand while spending on personnel accounts for more than one-third
for more complex and modern infrastructure services. A 2015 of expenditure.
market assessment of 14 large Indonesian cities estimated an
overall subnational infrastructure investment gap of US$11.1 This note aims to provide a diagnostic of current
billion (against borrowing capacity of US$1.7 billion) (Joshi et subnational borrowing practices and identify enablers
al. 2015). and impediments of access to debt financing by SNGs in
Indonesia. As discussed in Section 3, this note focuses on
Subnational borrowing currently plays a negligible role access to debt and debt management capacity-related issues
in infrastructure investment in Indonesia. Indonesia’s while broader enabling factors such as intergovernmental
SNGs have limited autonomy over revenue collection and transfers design and public investment management (PIM)
do not have access to debt markets. Between 2018-21, on are beyond the scope of this note. The remaining sections are
average, SNG debt (above one year maturity) accounted for structured as follows: Section 2 reviews recent developments in
only 0.03 percent of GDP or 10 percent of subnational capital subnational debt in Indonesia and other developing countries;
expenditure.2 The bulk of subnational public investment (90 Section 3 analyzes four factors that impact SNGs’ access to
percent) is mostly financed by central government transfers debt financing; and Section 4 discusses policy options.
T A B L E 1 . 1 - Indonesia’s Subnational Government’s Fiscal Structure 2019-20
(% of GDP, % of total)
2019 2020
Subnational fiscal structure (% of GDP)
Province District Total Province District Total
Total revenue 2.3 5.3 7.6 2.2 5.1 7.2
Own-source revenue 1.1 0.8 1.9 1.0 0.7 1.7
Fiscal transfers from central
1.0 3.4 4.4 1.0 3.2 4.2
government
Other revenue 0.2 1.1 1.3 0.2 1.2 1.3
Total expenditure 2.2 5.3 7.5 2.2 5.1 7.3
Primary recurrent expenditure 1.9 4.2 6.1 2.0 4.3 6.2
Capital expenditure 0.3 1.1 1.4 0.2 0.8 1.0
Interest payments 0.0 0.0 0.0 0.0 0.0 0.0
Balance 0.00 0.06 0.07 -0.02 -0.03 -0.04
SNG gross debt stocka n.a n.a 0.33 n.a n.a 0.38
b
SNG annual borrowing n.a 0.02 0.02 0.12 0.04 0.16
SNG gross debt stock a n.a n.a 0.33 n.a n.a 0.38
SNG annual borrowing b n.a 0.02 0.02 0.12 0.04 0.16
2. The Government of Indonesia’s Medium-term Development Plan (RPJMN) for 2020-2024 estimates infrastructure investment needs of Rp 6,421 trillion (or US$452
billion) for the 5-year planning period if Indonesia is to reach infrastructure stock standards of a middle-income country (50 percent of GDP) by 2024. Of this amount
(US$167 billion), 37 percent is expected to come from central and subnational governments.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 10
2019 2020
Subnational fiscal structure (% of total)
Province District Total Province District Total
Total revenue 100.0 100.0 100.0 100.0 100.0 100.0
Own-source revenue 47.4 14.8 24.5 44.6 14.7 23.7
Fiscal transfers from central
44.9 63.9 58.3 47.5 62.4 57.9
government
Other revenue 7.7 21.2 17.2 7.9 22.9 18.4
Total expenditure 100.0 100.0 100.0 100.0 100.0 100.0
Primary recurrent expenditure 85.8 79.7 81.5 89.9 84.2 85.9
Capital expenditure 14.2 20.2 18.4 10.0 15.8 14.0
Interest payments 0.0 0.0 0.0 0.0 0.0 0.0
Source: DG Fiscal Balance Ministry of Finance (DGFB-MoF), Ministry of Home Affairs (MoHA), and Bank of Indonesia SUSPI; World Bank
staff estimates.
Note: (a) SNG gross debt stock data comes from Bank Indonesia Statistik Utang Sektor Publik Indonesia (BI-SUSPI) data; (b) SNG annual
borrowing data comes from the DGFB/Ministry of Home Affairs (MoHA); (c) the effect of rounding may impact on some totals.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 11
2. >>>
Indonesia’s Subnational Debt:
Trends and Composition3
Indonesia’s SNG debt is low and well below the regulated borrowing limit. It has risen
recently from 0.32 percent of GDP (2018) to 0.41 percent of GDP (2021) (Figure 2.1). Much
of the increase over this period was driven by loan facilities from the central government as
part of the fiscal stimulus package to support national economic recovery (Pemulihan Ekonomi
Nasional: PEN) from the COVID-19 shock.4 Nonetheless, Indonesia’s SNG debt is among the
lowest in the world (Figure 2.2). The annual borrowing of 0.02 percent of GDP is also well below
the authorities’ annual borrowing limit of 0.3 percent of GDP. In 2021, SNG debt accounted for
only 1.2 percent of total public debt.
By maturity, nearly two-thirds of Indonesia’s SNG debt is in the form of short-term loans.
Between 2019-21, about 62 percent of total SNG debt was in the form of short-term loans‒
suggesting that borrowing at subnational level is largely used for cash management purposes
including accounts payable to suppliers. Only 38 percent of SNG debt (0.14 of GDP) is classified
as long-term (loan maturity more than one year). This accounts for only 10 percent of subnational
capital expenditure‒suggesting the limited role of debt in financing infrastructure investment.
Indonesia shares similarities with other countries in terms of trend and composition of
SNG debt. SNG debt levels are generally much lower compared to central government debt.
SNG debt accounted for 7.5 percent of global GDP or 11.5 percent of total public debt in 2016
but this varies across countries, income levels, and subnational governance systems (Box 2.1).
SNG debt level ranges from 12.9 percent of GDP in OECD countries to 2.9 percent in non-
OECD countries (Figure 2.3). It is higher in high-income countries (12.1 percent of GDP) but
almost nonexistent in low-income countries. This fits well with the general notion that countries
with strong institutions can carry higher levels of debt. Subnational debt is higher in federal
countries (18.2 percent of GDP) than in unitary countries (4.7 percent of GDP). Indonesia’s SNG
debt at 0.41 percent of GDP (1.2 percent of general government debt) was very low in 2021.5
3. Indonesia’s SNG debt data are derived from two sources: (i) BI-SUSPI comprises a consolidated SNG debt stock data produced quarterly and disaggregated by short
and long-term maturity (> 1 year maturity). This data excludes on-lending Subsidiary Loan Agreements (SLA) from central government to SNGs; and (ii) MoF/MoHA
4 Introduced in 2020, PEN loan facilities were aimed for SNGs.
5. SNG WOFI (World Observatory Subnational Government Finance and Investment) 2016 data of unweighted average based on 76 countries and the WOFI Report 2019
(OECD).
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 12
F I G U R E 2 . 1 - Indonesia’s SNG debt increased F I G U R E 2 . 2 - ... but remains one of the lowest in
between 2018 and 2021 the world
(SNGs debt, % of GDP) (SNGs debt % of GDP, 2019)
0.50% Short term 35
Long term (> 1 year) 31.6
Total 0.41% 30
0.40%
25 21.320.6
19.8
20
0.30%
15 12.5
0.22% 11.1
0.20%
10
0.20% 5.6 4.8
4.3 3.2
5 2.9
1.5 0.6 0.3
0.10% 0
0.00%
2018 2019 2020 2021
Source: BI-SUSPI. Source: SNG WOFI-OECD data and WB staff calculation.
Note: Indonesia uses 2021 data comprise loans with maturity > 1
year.
F I G U R E 2 . 3 - Indonesia’s SNG Debt is also lower F I G U R E 2 . 4 - Indonesia’s composition of SNG Debt
relative to lower-middle-income country peers similar to those in low-income countries
(% of GDP and % of total public debt, 2019) (% of total, 2019)
28 Unitary
24 % GDP % public debt Federal
20
Low income
16 Lower middle income
12 Upper middle income
High Income
8 Indonesia
4 NON-OECD
0 OECD
Average
0 20 40 60 80 100
Loans
Bonds / debt securities
Currency and deposits
Insurance pension and standardised guarantees
Other accounts payable
Source: SNG WOFI-OECD Database and World Bank staff estimates. Source: SNG WOFI-OECD Database and World Bank staff estimates.
Furthermore, as with low-income countries and countries loans from the central government account for about 80 percent
with unitary structures, Indonesia’s subnational borrowing of subnational debt.6 This includes loans channeled through
is dominated almost entirely by other accounts payable financial intermediary institutions (PT. SMI)7 including regional
and direct loans. According to WOFI data, direct lending loan facilities to support economic recovery (regional PEN) (43
represents the highest share of global subnational government percent),8 and the SLA (Subsidiary Loan Agreements)‒a two-
debt stock (57 percent), followed by other accounts payable step lending mechanism where foreign loans are channeled
(29 percent) and bonds/securities (12 percent) (Figure to the SNG via central government‒mainly to the Province
2.4). The share of “other accounts payable” of SNG debt is of DKI Jakarta (37 percent).9 The remainder of subnational
particularly high in low-income countries (nearly 100 percent) borrowing is sourced from regional development banks (Bank
‒ highlighting the potential accumulation of short-term debt Pembangunan Daerah: BPD) where SNGs often own shares.
(commercial debt with suppliers and arrears). In Indonesia,
6 By regulation, Indonesia’s SNGs are allowed to take on debt from diversified sources including: (i) central government; (ii) other SNGs; (iii) banking institutions
(commercial bank/regional bank); (iv) nonbank financial institutions (NBFIs) (such as intermediary financing institutions, pension funds, or insurance funds); and (v) the
public through capital markets such as municipal bonds.
7 PT. Sarana Multi Infrastruktur (PT. SMI) is a State-owned Enterprise (SoE). One of its functions is to provide financing/loans to SNGs for the development of economic
and social infrastructure.
8. Regional PEN (Pemulihan Ekonomi Nasional) was introduced in 2020-21 as part of the COVID-19 fiscal stimulus package.
9 The Provincial Government of DKI Jakarta signed a 40-year loan deal of Rp 27 trillion from Japan International Cooperation Agency (JICA) in 2017 for the construction
of Jakarta’s mass rapid transit (MRT) system.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 13
Despite interest from selected SNGs in Indonesia, none countries. Among developing countries, India and Mexico,
have yet issued bonds. This is in line with trends in lower- which have federal structures, have high shares of bonds/
middle- and low-income countries where bond financing by securities in SNG debt, while bond issuances are higher in
SNGs is insignificant due to less developed capital markets, high- and upper-middle-income countries (14-16 percent) due
however, bonds/debt securities comprised 27 percent of SNG to the depth of financial markets (OECD 2019).
debt stock in federal countries compared to 8 percent in unitary
B O X 2 . 1 - Country Characteristics and SNG Debt
A cross-country data analysis (SNG-WOFI database) using a simple correlation approach offers some insights on how
SNG debt is associated with country characteristics.10
i) Income levels have a positive association with the level of SNG debt. High-income countries tend to have higher SNG
debt (11.9 percent of GDP or 14.9 percent of general government debt) compared to low- and lower-middle-income
countries (1.9 percent of GDP or 6.3 percent of general government public debt). SNGs in upper-middle- and high-
income countries are likely to have higher fiscal capacity and greater access to domestic and external credit markets.
There are, however, some exceptions such as India, a low-income country with a federal system, which has relatively
high level of subnational debt at 21.3 percent of GDP. Indonesia is a lower-middle-income country with shallow
financial markets and significantly low subnational debt.
ii) Federal countries tend to have higher SNG debt than SNGs in the unitary countries.11 SNGs of federal countries
generally have a higher degree of autonomy and less restricted borrowing rules (for example, state governments
in the federal countries are usually not subject to the “golden rule” that restricts borrowing to finance long-term
investment projects). Indonesia is a unitary country with stricter borrowing rules.
iii) Countries with higher vertical fiscal imbalances (VFI) are positively associated with SNG debt albeit only weakly.
Intergovernmental system design (for example, transfers from central government to SNGs to address the imbalances
between expenditure assignment and revenue collection autonomy) also affects SNG debt. Countries with higher VFI
tend to have higher subnational public debt. For example, SNGs in South Africa are responsible for 49.3 percent of
public expenditure while local tax revenue accounts for only 5 percent of total tax revenue. India’s SNGs managed
62 percent of public expenditure but collect only 30 percent of total tax revenue. Similarly, in China, SNGs are
responsible for 63.3 percent of expenditure but receive only 20.6 percent of total tax revenue. In contrast, Indonesia
has relatively high vertical fiscal gaps and a low level of subnational debt.
iv) The degree of local tax revenues is positively associated with subnational debt. Subnational revenue autonomy is
one of the important determinants of borrowing capacity. Subnational tax revenues reflect potential fiscal capacity that
contribute to SNGs creditworthiness and their ability to repay debt. SNGs that have a higher share of tax revenues (for
example, more stable and reliable income sources) are expected to be more creditworthy and have higher capacity to
borrow. In Indonesia, subnational tax revenues are relatively low at 1.7 percent of GDP or 23.7 percent of subnational
revenue.
10 This section drawn from SNG WOFI 2019. The unit of analysis is at country level (consolidated SNGs)
11 Unitary and federal governments provide different opportunities for fiscal decentralization. Unitary countries do not have SNGs that are constitutionally empowered to
make decisions over a specified range of government functions and services; rather, they have multiple subordinate levels of the same government (for example, central,
provincial, district). Federal governments, on the other hand, have constitutionally protected SNGs and, therefore, the possibilities for independent decision making are
clearly stronger under these systems. It is important to note that local governments (as opposed to state or province) may not necessarily enjoy constitutional protection
under federal systems. In practice, however, the extent and nature of decision-making power exercised by lower tiers varies widely from country to country in both federal
and unitary countries and may change (in either direction) from time to time (World Bank 2001).
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 14
v) Capital expenditure is positively associated with subnational debt (Figure B.1). Many countries adopt borrowing rules
that restrict the use of borrowing for financing long-term investment (golden rule). Indonesia’s subnational borrowing
framework also regulates borrowing for infrastructure, but most capital expenditure is financed from transfers from
central government.
F I G U R E B . 1 - Relationship Between Subnational Debt and Selected Fiscal Variables
Income level is positively associated with the level of SNG debt VFI is weakly correlated with the level of SNG debt
50
ZAF AUS
CHN
90,000 DNK
y = 901.83x + 19008 40
80,000 IDN IND
Vertical fiscal imbalance (%)
R² = 0.333 AUT
70,000 CHE NLD BEL
NOR USA 30 PER
GDP pc (USD ppp)
EST
60,000 NLDISL
AUT
DNKSWE DEU GBRSWE CAN
AUS
BEL CAN ROU
LTU
FIN
GBR JPN
20 BOL
50,000 MLTNZLFRA NIC RWA KOR ESP
KOR
ISR ITA ESP KAZ NOR
CYP
SVN HRVGTM ISL CHE
FIN y = 0.1796x + 9.809
40,000 LTU 10 COL R² = 0.0264
HUNPOL
RUS TUR
MAR USA
HRV SLV
30,000 JPN
MNE HND DEU
CHN 0
20,000 MNE
IND 0 BLR 20 40 60 80
10,000 IDN -10 ARG
GRC
0
0 20 40 60 80 -20
SNG debt (% GDP)
SNG debt (% GDP)
Subnational tax revenue is positively Capital expenditure is positively
associated with the level of SNG debt associated with the level of SNG debt
25 5.0
SNG capital expenditure (% GDP)
4.5 y = 0.0482x + 1.1125 CAN
R² = 0.4022
SNG tax revenue (% GDP)
20 y = 0.2655x + 1.9474 4.0
R² = 0.5058 CAN KOR JPN
3.5
IND
CHE
15 ARG SWE PER
3.0 NIC BEL
DNK DEU
CHE ARG FRA AUS DEU
SWE
NOR
FIN
BRA IND 2.5 AUT
FIN
RUSBLR ISLCHN USA
UKR
ZAF ROU ESP
USA
10 ESP 2.0 ITA
BEL JPN DNK
LVAITA AUS
FRA POL
ESTIDNBRA
1.5 NGA LVA
GBR ISL
NZL
HRV
POL KEN
SVN
UKRCOL ISR
5 KOR 1.0 MNEPRT
SRB
KAZ
SVN
0.5
IDN
0 0.0
0 20 40 60 80 0 20 40 60 80
SNG debt (% GDP) SNG debt (% GDP)
Source: WOFI-OECD data and World Bank staff estimates.
Note: a. VFI (share of SNG expenditure of general government expenditure – share of local tax revenues of general government tax revenue).
ARG=Argentina, AUT=Austria, AUS=Australia, BEL=Belgium, BLR=Belarus, CAN=Canada, CHE=Chile, CHN=China, DNK=Denmark,
DEU=Germany, ESP=Spain, GRC=Greece, GBR=Great Britain, IDN=Indonesia, IDN=India, JPN=Japan, NLD=Netherlands, KOR=Republic
of Korea, NOR=Norway, PER=Peru, RUS=Russia, SWE=Sweden, USA=United States of America, ZAF=South Africa.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 15
3.
>>>
Analytical Framework
SNGs’ access to debt financing is influenced by multiple factors. The extent to which
SNGs can access debt financing is affected by overarching conditions such as design of the
intergovernmental fiscal system (IGFS) and PIM capacity‒including the ability to prepare, select,
and implement projects. While this note focuses on assessing key constraints that directly
affect subnational borrowing, the subsequent paragraphs discuss briefly the linkages between
overarching conditions and subnational debt financing.
Subnational revenue autonomy and capital transfers are two key features of IGFS that
may affect SNGs access to debt financing. Subnational fiscal autonomy may contribute to
better creditworthiness of SNGs (for example, debt-carrying capacity) since a higher degree of
revenue autonomy allows SNGs to have better control and access to more stable sources of
revenue. Indonesia’s IGFS has, however, been characterized by a very limited revenue autonomy
of SNGs. In 2019, the share of SNG’s own-source revenue only constituted 14 percent of total
general government revenues across all levels of government. This is low compared with 33
percent in Brazil, or 43 percent in China (IMF Fiscal Decentralization Database 2019). With
this limited autonomy, Indonesia’s SNGs have been heavily reliant on transfers from the central
government. Nonearmarked transfers (general allocation grant) made up over 60 percent of
districts/cities’ revenues in 2019 (World Bank 2020a).
The design of transfers‒particularly transfers/grants for capital spending‒may affect SNGs’
incentive to leverage own revenue or borrowing to finance their priority infrastructure.
Unpredictable capital grants or exclusion of cofunding from SNGs may, in contrast, discourage
SNGs from mobilizing own financing for investment. For example, the earmarked transfers
(Dana Alokasi Khusus Fisik: DAK Fisik or Special Allocation Grants for Physical Infrastructure)
in Indonesia, which is the main capital transfer program, lacks predictability and depends on
lengthy central government approvals. This makes it challenging for Indonesia’s SNGs to plan
for large and complex investments that require more than one year of implementation (World
Bank 2022). Another dimension of IGFS and borrowing is the degree of VFI between central
and subnational governments. Empirical analysis suggests that higher VFI is associated with
detrimental fiscal performance through increased fiscal deficit and higher government debt
accumulation (Aldasoro and Seiferling 2014). Higher VFI tends to encourage local borrowing
since SNGs expect the central government to bail them out. These findings are not evidenced in
Indonesia where, despite relatively high VFI (58 percent), Indonesia’s subnational debt remains
low.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 16
Another key enabler is the PIM capacity of SNGs. more directly related to SNG debt mobilization. In particular,
Weak planning processes, including project selection it does not discuss further broader constraints that may
and preparation, can reduce the effectiveness of local indirectly affect subnational borrowing such as the design of
investment. Together with other factors including project the intergovernmental fiscal system, or PIM capacity. Instead,
appraisal capacity, quality of procurement, budget execution this note looks at four factors that directly affect SNGs’ access
performance, accounting, and reporting affect the credibility to debt financing (Figure 3.1):
of public investment plans and, therefore, creditworthiness. a. Subnational borrowing capacity/creditworthiness:
In addition, appropriate design of a Public Financial Which subnational authorities have borrowing capacity?
Management system is also important for facilitating public Using commonly used criteria, this part provides a
investment. For example, the absence of multiyear project top-down snapshot of SNGs’ potential debt-carrying
planning and budgeting, mostly due to the annual nature and capacities.
high uncertainty of central government transfers presents b. Subnational debt management capacity: Do high-
major challenges in infrastructure development at subnational borrowing capacity SNGs have sound debt management
level. In a survey of 50 SNGs (World Bank 2022), 93 percent capacity? What are the critical gaps? This part discusses
of SNG officials reported that they do not have multiannual the findings from a self-assessment survey of subnational
project planning‒mainly due to the high level of uncertainty/ debt management capacity in Indonesia.
volatility of the earmarked transfers (DAK Fisik) (World Bank c. Borrowing sources: Is there an adequate supply of
2021d). Weaknesses in subnational PIM are also reflected credit (from market or nonmarket sources) to finance
in the high share of DAK Fisik12 proposals rejected by the medium- and long-term infrastructure investment at
central government. In 2021, 81 percent of nearly 8,000 subnational level? This section analyzes the composition
proposed road projects were rejected due to their poor quality of subnational borrowing and identifies gaps.
and noncompliance with guidelines established by central d. Borrowing rules regime: Are regulatory frameworks
government line ministries (World Bank 2021d). and existing rules governing subnational debt conducive
in enabling SNGs access to debt financing? This section
While acknowledging the importance of broader enabling reviews the regulatory framework and borrowing rules
conditions above, this note focuses on factors that are regime in Indonesia relative to its peers.
F I G U R E 3 . 1 - Framework for Identifying Constraints to Subnational Borrowing
Inter-Governmental Public Investment
Fiscal System Management Capacity
Creditworthiness Debt Management Capacity
Borrowing Sources
Commercial credits, Securitized Debts,
Development Bank/Special Financial
Intermediaries, CG on-lending)
Borrowing Rules Regime
Direct controls, Fiscal rules,
Cooperative, Market Discipline
Subnational Debt
Source: Authors
12 The DAK Fisik is the earmarked grant from central government, and it has been the most important fiscal source for SNGs in cofinancing their capital investment. Since
2016, the allocation of DAK Fisik has been proposal-based, in which SNGs submit proposals for specific investment projects that they seek funding for. It has replaced
the previous formula-based allocation.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 17
3.1 Which SNGs have borrowing capacity?13
Indonesia’s SNGs vary widely across regions in terms subnational fiscal management with economic conditions or
of debt capacity due to variation in fiscal capacity and fiscal performance.
economic development, among others. This analysis
provides a top-down snapshot of borrowing capacity across The analysis finds that 55 SNGs (out of 542) or 10 percent
Indonesia’s SNGs, which helps highlight SNGs that may be of all SNGs could potentially have high-borrowing
worth analyzing in more detail to do a thorough creditworthiness capacity, based on the limited economic criteria noted
assessment. above. Notwithstanding wider borrowing constraints, there are
55 SNGs assessed as having relatively strong debt capacity,
Three criteria are used to assess the borrowing capacity comprising 51 districts and four provinces (Appendix 2). Table
of SNGs, drawing on the general methodology adopted 3.1 provides comparisons of characteristics between the high-
by major credit rating agencies. The criteria include: (i) and low-borrowing capacity SNGs. Given the strong correlation
economic conditions; (ii) fiscal performance; and (iii) quality between the subnational economy and fiscal performance,
of fiscal management. Appendix 2 summarizes the criteria it is not surprising that 71 percent of SNGs with potentially
and methodology in measuring SNGs borrowing capacity. high debt-carrying capacity are found in regions with strong
An analysis across these criteria shows that SNGs fiscal economic performance‒especially Java-Bali (Figure 3.2 and
performance is strongly associated with their economic Table 3.2). These two islands constitute nearly 60 percent of
condition. Subnational GDP is a strong predictor of tax receipts the nation’s population and contribute 60 percent of national
for both provincial and district governments so SNGs of larger GDP. Consequently, districts in Java-Bali collect the highest
economic scale are likely to have higher fiscal capacity. There proportion of own-source revenue at 19 percent, or almost four
is, however, a lack of clear association between the quality of times what districts in Maluku-Papua collect (about 5 percent).
T A B L E 3 . 1 - Statistics (the Average Values) Across Indicators of Borrowing Capacity
1. Economic Condition 2. Fiscal Performance
Overall SNGs Econ stability Econ size Population size Interest
Fiscal space
(st-dev econ (% national (% national payment
(Billions of Rp)
growth) GDP) pop) (% revenue)
High-borrowing
(Decile 1) 2.9 1.6 1.6 3,254 0.01
capacity
Low-borrowing
(Decile 10) 3.9 0.1 0.1 634 0.23
capacity
3. Quality of Fiscal Management
Recurrent exp
Overall SNGs OSR outturn Capex outturn
outturn Audit status* on fin. reports
(% planned) (% planned)
(% planned)
High-borrowing
(Decile 1) 98.0 98.4 96.4 3.9
capacity
Low-borrowing
(Decile 10) 170.4 95.8 93.1 3.6
capacity
Source: World Bank staff calculation, using 2016-19 data.
Note: *) Audit result is converted to the following numerical unit: Unqualified (Wajar Tanpa Pengecualian) = 4; Qualified (Wajar Dengan
Pengecualian) = 3; Disclaimer (Tidak Memberikan Pendapat) = 2; Adverse (Tidak Wajar) = 1. Capex refers to capital expenditure, OSR refers
to own-source revenue.
13 Subnational borrowing capacity analysis can also inform the broader design of subnational financing options including specific grants or a mixed instruments model
between loans and grants. Low-borrowing capacity SNGs, that are less likely to meet eligibility criteria from commercial banks and the credit market due to noneconomic/
nonfinancial reasons may need special financing arrangements to address infrastructure gaps such as infrastructure equalization grants.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 18
F I G U R E 3 . 2 - SNGs With High-Borrowing Capacity are Heavily Concentrated in Java-Bali
Source: World Bank staff calculation.
T A B L E 3 . 2 - Subnational Borrowing Capacity Ranking by Main Island
(top and bottom deciles) (number of SNGs-upper table; and share of total-lower table (%)
Maluku-
# SNGs, by decile Java-Bali Kalimantan Papua Sulawesi Sumatra Total
Nusa Tenggara
High-borrowing capacity
40 4 1 1 0 9 55
(Decile 1)
Low-borrowing capacity
6 8 8 12 11 28 73
(Decile 10)
Maluku-
% SNGs, by decile Java-Bali Kalimantan Papua Sulawesi Sumatra Total
Nusa Tenggara
High-borrowing capacity
71.4 7.1 1.8 1.8 0.0 17.9 100.0
(Decile 1)
Low-borrowing capacity
8.2 11.0 11.0 16.4 15.1 38.4 100.0
(Decile 10)
Source: World Bank staff calculation.
Note: Percentage totals may not necessarily equal 100 percent due to the effect of rounding.
One-half of the SNGs with potentially high-borrowing The analysis finds that SNGs borrowing capacity does
capacity are part of metro/urban centers, pointing to not always associate with the level of debt (Figure 3.3 and
potential high demand for infrastructure investment Figure 3.4). High borrowing capacity SNGs do not always
needs. With a high population density, strong commuting borrow while low-borrowing capacity SNGs are able to take
flows, and a high share of formal employment, metropolitan on debt. Of 55 high-borrowing capacity SNGs, only eight
areas increasingly require substantial investment in urban have borrowings/debt (see Appendix 2 for the list of SNGs). In
infrastructure. One-half of SNGs with high-borrowing potential contrast, 18 of the 55 lowest borrowing capacity SNGs have
(27 SNGs) are part of the metro areas that include Jakarta, outstanding debt. In other words, potential borrowing capacity
Surabaya, Bandung, Palembang, and Medan.14 of SNGs (based on three criteria: economic performance, fiscal
capacity, and quality of fiscal management) does not always
14 There are 28 metropolitan areas across Indonesia that includes 75 SNGs, further classified into multidistrict metro areas and single-district metro areas (World Bank
2019d).
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 19
correlate with actual borrowing or restrict SNG’s eligibility to for the provinces (Figure 3.4), even after excluding DKI
borrow, since most subnational borrowing is from nonmarket Jakarta.15 Empirical evidence suggests, however, that fiscal
sources (for example, financial intermediary institutions such space is strongly correlated with SNGs’ gross cumulative debt
as PT. SMI). The variance in the size of loan is more striking (Table 3.3).
F I G U R E 3 . 3 - Borrowing capacity of districts do not F I G U R E 3 . 4 - … Similarly for provinces, borrowing
always associate with the level of debt… capacity do not always associate with the level of debt
400 Districts' average cumulative loan, 2,500 Provinces' average cumulative loan,
by creditworthiness decile by creditworthiness decile
350 (Rp Billion)
(Rp Billion)
2,000
300
250 1,500
200
150 1,000
100
500
50
0 0
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
(Top,
(Top) (Bottom) (Bottom)
except Jakarta)
Source: World Bank staff calculation, based on MoF data. Source: World Bank staff calculation, based on MoF data.
Note: The high- and low-capacity groups refer to the top and bottom Note: The high- and low-capacity groups refer to the top and bottom
decile of borrowing capacity. decile of borrowing capacity. (*) Excludes DKI Jakarta.
T A B L E 3 . 3 - Fiscal Space is the Strongest Predictor for SNGs Gross Cumulative Loans
Dependent: Gross cumulative loans, 2016-21 (% GDP, 2019) (1) (2)
Economic stability (std deviation of real GDP growth 2010-19) 0.001** 0.001**
(0.000) (0.000)
Subnational economy (% national GDP, avg 2016-19) 0.131 0.135
(0.154) (0.150)
Fiscal space (% subnational GDP, 2019) 0.185*** 0.180***
(0.008) (0.008)
OSR outturn (% planned, 2019) 0.013*** 0.012***
(0.003) (0.003)
Capital spending outturn (% planned, avg 2016-19) 0.001 0.007
(0.108) (0.011)
Recurrent spending outturn (% planned, avg 2016-19) -0.050 -0.076
(0.053) (0.052)
Audit result of SNG financial reports (avg 2016-19) 0.021*** 0.020***
Audit result is converted to the following numerical unit: (0.004) (0.004)
Unqualified (WTP) = 4; Qualified (WDP) = 3
Disclaimer (TMP) = 2; Adverse (TW) = 1
Interest payment (% revenue, 2019) 6.638***
(1.250)
Constant -0.057 -0.036
(0.049) (0.048)
Observations 542 542
Prob > F 0.000 0.000
R-squared 0.532 0.556
Adj R-squared 0.526 0.549
Note: *p<0.1; **p<0.05; ***p<0.01
15 DKI Jakarta is considered an outlier with gross cumulative loans reaching Rp 37 trillion or one-half of total subnational borrowings during 2016-21.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 20
The analysis shows that the source of debt varies between The SNG’s untapped borrowing space is estimated at Rp
high- and low-borrowing capacity SNGs. The largest 34 trillion in 2019, or one-half of the capital grants from
proportion of loans (78 percent) in SNGs with high-borrowing central government (DAK Fisik) in the same year. Based
capacity is from PT. SMI. In contrast, the main sources of on the current fiscal rules that limit the size of subnational
borrowing of the low-borrowing capacity SNGs are BPD. debt, how much each SNG can borrow (that is, the borrowing
space) can be estimated. Among the four fiscal rules imposed
Low-capacity SNGs are more likely to bear higher by the MoF18, the maximum limit of an SNG’s budget deficit to
borrowing costs. Among sources of credit, BPDs charge be financed through borrowing (set annually) gives the lowest
the highest interest rate of at least 12 percent annually debt value an SNG can take on. Every year, MoF determines
(World Bank 2021a). On the other hand, the interest rates the maximum limit of deficit for each SNG based on their
on regular loans from PT. SMI are between 8.0-8.5 percent, fiscal capacity.19 Even under this most conservative limit, the
with a maximum five-year tenor. In comparison, the central unused borrowing space is significant, amounting to Rp 34
government’s borrowing costs for similar maturities are 5.8 trillion for all districts in 2019, one-third of which belongs to the
percent.16 For national economic recovery loan facilities for high-borrowing capacity SNGs. The unused borrowing space
SNGs (PEN loan), they were interest free in 2020, then interest is equivalent to an extra 3-5 percent of SNG revenue in 2019,
rates are charged at 5.3 to 6.2 percent in 2021 depending on depending on the category of borrowing capacity (Figure 3.5).
loan maturity.17
F I G U R E 3 . 5 - SNG’s Untapped Borrowing Space
The different composition of borrowing sources between Equals an Extra Three to Five Percent of Revenue
the high- and low-capacity SNGs may also point to SNG’s
6%
ability to prepare infrastructure projects. PT. SMI has District Province
stricter borrowing requirements than BPDs. To borrow from
5%
PT. SMI, SNGs will have to meet the infrastructure projects’
eligibility criteria, on top of adhering to fiscal rules. The eligibility 4%
criteria include: (i) the project’s economic and financial viability;
(ii) technical feasibility (project preparation studies, detailed 3%
engineering designs, and safeguard instruments); and (iii)
social and environmental appraisal. High-capacity SNGs are 2%
likely to have more capability in meeting these requirements.
In comparison, the loan proposals required by BPDs mostly 1%
only cover the project’s technical feasibility and compliance
0%
with fiscal rules. Furthermore, borrowings from BPDs are also 1 2 3 4 5 6 7 8 9 10
allowed for cash management purposes. Finally, the average
size of loans from PT. SMI (Rp 262 billion) is larger than BPDs (High-borrowing capacity) (Low-borrowing capacity)
(Rp 149 billion), indicating that PT. SMI finances a bigger scale Source: World Bank staff calculation, based on MoF 2019 data.
of infrastructure project. Note: The high- and low-capacity groups refer to the top and
bottom decile of borrowing capacity.
16 The five-year sovereign bond yield as of March 2022.
17 See MoF Regulation - PMK No. 43/2021 on PEN Lending for Subnational Governments and recent news coverage.
18 See Part D in this note (on the regime of borrowing rules) for further discussion.
19 The MoF PMK defines fiscal capacity as SNG’s revenue minus earmarked revenue and mandatory expenditure.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 21
3.2 Do high-borrowing capacity SNGs have sound debt management capacity?20
Indonesia’s SNGs vary widely across regions in terms Sound debt management practice is important for
of debt capacity due to variation in fiscal capacity and accessing debt financing, lowering debt servicing costs,
economic development, among others. This analysis and minimizing the risk of default. Sound government debt
provides a top-down snapshot of borrowing capacity across management can contribute to a better credit assessment by
Indonesia’s SNGs, which helps highlight SNGs that may be financial markets, and lower debt-servicing costs by reducing
worth analyzing in more detail to do a thorough creditworthiness the risk premium. Sound debt management is critical because
assessment. a government’s borrowing decisions can significantly impact
their budget. If the cost of the debt increases when the primary
Three criteria are used to assess the borrowing capacity deficit is high or rising, this could force the government to cut
of SNGs, drawing on the general methodology adopted expenditures and/or increase taxes and, in the worst case,
by major credit rating agencies. The criteria include: (i) default on its obligations (World Bank 2017).
economic conditions; (ii) fiscal performance; and (iii) quality
of fiscal management. Appendix 2 summarizes the criteria This section provides an overview of Indonesia’s SNG
and methodology in measuring SNGs borrowing capacity. debt management practices. It draws on the Subnational
An analysis across these criteria shows that SNGs fiscal Debt Management Performance Assessment (SN DeMPA)21
performance is strongly associated with their economic tool. It focuses on initial requirements for debt management
condition. Subnational GDP is a strong predictor of tax receipts rather than all dimensions in the SN DeMPA. Performance is
for both provincial and district governments so SNGs of larger reviewed against minimum conditions that need to be met for
economic scale are likely to have higher fiscal capacity. There SNGs to manage debt within a framework of sound governance
is, however, a lack of clear association between the quality of and minimum operational risk, including: (i) the legal/
subnational fiscal management with economic conditions or regulatory framework; (ii) institutional/staffing arrangement;
fiscal performance. (iii) debt recording, accounting, and debt servicing; and (iv)
debt auditing. The review is based on self-reporting through
The analysis finds that 55 SNGs (out of 542) or 10 percent an online survey of 39 SNGs.22
of all SNGs could potentially have high-borrowing
capacity, based on the limited economic criteria noted More than two-thirds of survey respondents incurred
above. Notwithstanding wider borrowing constraints, there are debt in the past three years (2019-21). Total cumulative debt
55 SNGs assessed as having relatively strong debt capacity, of SNG respondents was Rp 21.3 trillion, accounting for one-
comprising 51 districts and four provinces (Appendix 2). Table half of total borrowings in the same period (Rp 40.9 trillion).
3.1 provides comparisons of characteristics between the high- The purpose of borrowing was mainly to finance infrastructure
and low-borrowing capacity SNGs. Given the strong correlation projects. In line with the introduction of the PEN Program for
between the subnational economy and fiscal performance, SNGs to support economic recovery due to the COVID-19
it is not surprising that 71 percent of SNGs with potentially pandemic, with borrowing objectives expanded to support
high debt-carrying capacity are found in regions with strong national economic recovery. The sources of borrowing are
economic performance‒especially Java-Bali (Figure 3.2 and mostly intermediary financial institutions (PT. SMI) and BPDs.
Table 3.2). These two islands constitute nearly 60 percent of
the nation’s population and contribute 60 percent of national i) Legal/regulatory framework
GDP. Consequently, districts in Java-Bali collect the highest The legal framework for subnational borrowings was
proportion of own-source revenue at 19 percent, or almost four reported to be clear and adequate. Primary legislation
times what districts in Maluku-Papua collect (about 5 percent). includes: (i) Law No. 17/2003 on State Finances (the primary
legislation passed by the National Parliament); and (ii) the
20 This section draws upon: (i) online survey on subnational debt management capacity covering 39 SNGs in Indonesia and conducted in January 2022; (ii) SN DeMPA
assessment tool 2016; and (iii) SN DeMPA Assessment for the Province of DKI Jakarta 2012.
21. SN DeMPA assesses the strengths and weaknesses of subnational debt management practices.
22. The survey targeted a sample of 164 SNGs (or 30 percent of all SNGs) who have borrowed or are planning to borrow, of which 39 SNGs submitted their responses (24
percent). Sample selection was based on a few criteria: (i) SNGs that have outstanding debt; (ii) SNGs that have planned/proposed to borrow but did not get approval
(rejected); (iii) SNGs that have not borrowed but have high-borrowing capacity (based on borrowing capacity assessment); (iv) representation by level of government
(Province, Kabupaten, Kota); and (v) representation by geographical area (Java-Bali, Sumatra, Kalimantan, Sulawesi, Nusa Tenggara, and Maluku-Papua). The main
respondents are units that oversee borrowing/debt management activities within SNGs. This unit can be an existing function within local finance such as the Regional
Asset and Finance Management Agency (Badan Pengelola Keuangan dan Aset Daerah: BPKAD) or a separate agency. The online survey is complemented by a desk
review and in-depth discussions with relevant counterparts (MoF) on the national framework for subnational borrowing. For the detailed approach and methodology of
the subnational debt management online survey, please see Appendix 3.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 22
recently passed Law No.1/2022 on Fiscal Balances. More ii) Institutional and staffing arrangements
detailed guidance and implementation arrangements are There is clear division between political and technical
provided in the secondary/implementing regulations.23 decision making for SNG borrowing. Local parliaments
provide oversight through approval of financing/borrowing
The national legal framework provides clear authorization plans as part of the annual budget process. SNGs, represented
for SNGs to borrow and defines the financing sources, by units within their Finance Department or a committee
purpose, quantitative limits, and approval process. The created by district heads, are given the mandate to take on
mandates are given by the central government and there is debt.
strict adherence to all limits imposed by central government
regulations. The authorization to borrow is with the head Surveyed SNGs do not prepare debt management
of SNGs (Governor, Mayor, or Bupati). The purposes for strategies but do have annual financing/borrowing plans
borrowing are specified in the regulations‒including cash as part of the annual budget process. SNG respondents
management needs (short term or less than a year), and indicate that the information covered in the borrowing plan
infrastructure investment (medium and long term) (Law No. includes: (i) loan characteristics (that is, value, sources, fee,
1/2022). There is a requirement to acquire approval from local tenor/period); (ii) repayment plan; and (iii) portfolio structure,
parliament for medium- and long-term borrowing as part of the however, most SNGs do not publish the document.
annual budget process.
SNGs have not established Debt Management Offices
At the subnational level, most surveyed SNGs (90 percent) (DMO), since SNGs generally have very low or no debt
indicate they do not have local regulations governing and limited borrowing activities. Instead, the borrowing/
subnational debt. Local legislation on borrowing is not debt management function is integrated under the BPKAD,24
required for accessing loans, except for issuing subnational which generally performs back-office functions/activities
bonds/securities. The remaining 10 percent of respondents (that is, debt recording, debt repayment, and debt reporting)
indicated that they had local level regulations covering: (i) (Figure 3.6). SNGs indicated that lack of staff with appropriate
borrowing purposes; (ii) formulation of a debt management knowledge/skills in debt management is the main institutional
strategy; and (iii) reporting obligations to the local parliament challenge. Staff seem to have a relevant educational
and other relevant stakeholders. background: statistics (29 percent), economics (26 percent)
and finance/accounting (19 percent) but lack specialized debt
F I G U R E 3 . 6 - SNGs Focused on Back-office Functions F I G U R E 3 . 7 -Specialized Knowledge in Debt
Share of activities assigned by each SNG, (percent) Management is a Key Constraint
Ranking distribution by type of challenges, (percent)
Responsibilities 1 2 3 4 5
Others
13%
Front Office Changes in regulation
framework
Middle Office
Back Office Frequent staff rotation
50%
Middle Office & Back Staffs does not have
38% Office relevant education
Other background and trainings
Staffs does not have
relevant skills and capacity
to manage debt/loans
0% 50% 100%
Source: MoF-WB subnational debt management capacity survey 2022. Source: MoF-WB subnational debt management capacity survey
2022. Note: measurement scale: 1 = most difficult; 5=least difficult
23. These regulations include: (i) Government Regulation No. 10/2011 on Procedures for Foreign Loans and Acceptance of Grants (regulating SNG’s subsidiary loans
originating from foreign sources); (ii) Government Regulation No. 30/2011 on Local Government Borrowings; (iii) Ministry of Finance Decree No. 147/2006 on the Process
for Municipal Bonds; and (iv) Ministry of Finance Decree No. 127/ 2011 on the Maximum Deficit for Local Government Borrowings.
24 BPKAD stands for Badan Pengelolaan Keuangan dan Aset Daerah. Each SNG might have a different unit name (that is, BPKD) but the objective of the unit in general
is to formulate technical capacity in regional finance and asset management including budgeting, spending, accounting, and verifying regional assets.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 23
management knowledge (Figure 3.7). If SNGs plan to issue The reports are not publicly accessible but available only on
municipal bonds and to establish a debt management unit, demand. No audits on debt management activities have been
they will need skilled staff who are adequately trained in all conducted by independent external auditors since SNGs have
aspects of debt management, including debt data recording, not issued subnational bonds/sukuk (shariah debt financing).
and cost-risk analysis. Only 13 percent of respondents publish the audit report–most
of them through their regional government website.
iii) Debt recording, accounting, and debt servicing
SNGs, led by a unit within BPKAD, prepare monthly cash Most SNGs comply with the fiscal rules when accessing
projections, including for debt repayment, with relatively small regional debt. The risk monitoring and compliance function
deviation (less than 5 percent) between projection and actual. is currently exercised by MoHA and DGFB at the MoF, in
The excess cash holdings are mostly kept as term deposits in ensuring: (i) compliance with deficit thresholds; (ii) borrowings
BPDs and state-owned banks. do not exceed any stated government regulation ceilings;
and (iii) overall risk monitoring. Key indicators to assess
BPKAD is responsible for recording debt-related compliance to fiscal rules are available, but SNGs are not
transactions. Most respondents (62 percent) indicate, obliged to conduct such assessments. The survey found that
however, that they do not have digital debt recording systems most respondents find difficulties in calculating these key
to keep records of all loan contracts and documentation of indicators. Nonetheless, an internal assessment (using 2019
debt service payments, including those related to onlending data) shows that SNGs that have debts comply with the fiscal
from the central government. Most respondents also indicate rules.
that their debt recording systems are not synchronized with
central government systems. v) SNGs’ perception on current borrowing/debt
management practices
iv) Debt reporting and audit In addition to assessing debt management capacity,
SNGs report debt management activities and outstanding the survey asked participants about their borrowing
loans to the local parliament, local inspectorate, and the experience and plans. Most SNG respondents are familiar
MoF. Led by BPKAD, most SNG respondents prepare with borrowing rules and procedures, but this is limited to
annual reports on debt management activities and sources of borrowing they have undertaken previously (Figure
outstanding loans for the local parliament and the MoF. 3.8). Most respondents said that they “understand” rules and
Although Article 59 of Government Regulation No. 30/2011 procedures to apply for loans from PT. SMI (PEN and Regular
requires the local government to report the loan position to Loans), however, almost one-half of respondents stated they
MoF and MoHA each semester, the information presented in “least understand” the borrowing procedures from SLA and
debt reports varies widely; they include information on debt subnational bonds/sukuk.
flows, debt ratio, outstanding debts, and debt realization, but
do not follow a standard format. The evaluation of how debt SNGs are cautious about borrowing and do not perceive
management activities comply with the debt management debt as a primary source for financing infrastructure
strategy has not been prepared or reported. Making the investment. In 2019, most loans were for infrastructure
report publicly available is important for transparency in debt projects. They came mostly from SLAs and loans from PT.
management operations and good governance. SMI. In 2021 and 2022, most borrowing/debt came from
PEN loan facilities managed by PT. SMI to support economic
Financial and compliance audits were conducted by both recovery. Only 30 percent of respondents plan to borrow in
internal and external auditors regularly, but performance 2022 and 2023 while the remaining 70 percent of respondents
audits of debt management policies, activities, and indicated that they do not plan to borrow. Most SNGs indicated
operations have not been undertaken. On an annual basis, that own-source revenue and regional transfers are sufficient
the external audits are conducted by the supreme audit agency to finance expenditure needs/projects. Moreover, budget
(Badan Pemeriksa Keuangan: BPK) and internal audits are execution is low, leaving unspent cash balances (Sisa Lebih
done by the Inspectorate General. The audit reports are Pembiayaan Anggaran: SILPA).25
submitted to the Regional Parliament by the provincial BPK.
25 We found slight differences regarding the number of SNGs who accessed loans in 2019-21 between the survey results and the database collected from DGFB. Despite
the difference, both sources showed a similar increasing trend of SNGs who accessed loans.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 24
F I G U R E 3 . 8 - SNGs have limited understanding of F I G U R E 3 . 9 - Most SNGs do not have project
rules and procedures of subnational borrowing investment list
Degree of understanding on rules and procedures of subnational Share of respondents that have list of projects to be funded
borrowing by source (%) by subnational borrowing in 2022-2023 (%)
1 2 3 4 5 No Answer, 3%
Lainnya
Obligasi Daerah
Yes, 38%
Bank Pembangunan Daerah
PT SMI - Program PEN
Daerah
No, 59%
PT SMI - Pinjaman Reguler
Penerusan Pinjaman
Pemerintah Pusat (On-…
0% 20% 40% 60% 80% 100%
Source: MoF-WB subnational debt management capacity survey 2022. Source: MoF-WB subnational debt management capacity survey
Note: measurement scale: 1 = most difficult; 5=least difficult 2022.
Lack of capacity in project preparation and selection projects has been challenging, however, as only 11 projects
may also contribute to low borrowing. Most SNGs (59 were implemented of a target of 20 projects in 2021 (World
percent) report that they do not have a list of possible Bank 2021c).
projects/investments to be financed through borrowing
(Figure 3.9). About 12.3 percent of the approved loans by SNGs with high-borrowing capacity do not necessarily
the central government were cancelled due to technical perform better in their debt management practices on
reasons (nonbudgetary). Furthermore, most projects (29 out account of their limited borrowing activities. Most high-
of 32 appraised projects) funded by the World Bank-supported borrowing capacity SNGs have basic debt management
Regional Infrastructure Development Fund (RIDF)26 received functions but do not yet meet international good practices. DKI
Project Development Funds (PDF) to support project Jakarta, for example, has the largest amount of outstanding
preparation. debt but it has no debt management strategy. This has
remained unchanged since the 2012 SN DeMPA undertaken
PT. SMI and BPDs remain favorable sources of financing for for DKI Jakarta. DKI Jakarta only prepares annual borrowing
SNGs although there is scope to simplify loan application plans to finance the budget deficit which does not analyze
procedures. SNG respondents rate their experience in the composition of the debt portfolio, nor the implications
processing loans from PT. SMI and SLA as “slightly difficult”. for borrowing costs and risks. As with other SNGs, the debt
Loan processing from preparation of the loan proposal to management function in DKI Jakarta is also integrated under
approval can take between 3-12 months. Nevertheless, about the BPKAD.
75 percent of respondents who were planning to borrow in
2022 stated that they would choose PT. SMI. This is primarily In recent years, Indonesia’s SNG borrowing is largely
due to continuity and familiarity as reported by 35 percent of used to finance roads infrastructure. Between 2016-
respondents. There are no SNGs planning to issue bonds/ 2021, there are 139 SNGs (out of 548) reported to have
sukuk in 2022. borrowing of which their amounts vary largely across
years (Table 3.4). The MoF/MoHA’s subnational borrowing
Loans from PT. SMI are small and implementation has data indicate that about 68 percent of total SNG borrowing
been challenging. Between 2016-21, PT. SMI issued loans of is allocated for financing roads and bridge infrastructure,
Rp 43 trillion to SNGs. Between 2020-21, the bulk of the loans followed by hospitals (17 percent), and market infrastructure
were for regional PEN programs to support the recovery (78 (8 percent). These top three sectors account for 93 percent of
percent) while the remaining 22 percent were for regular loans total loans. Around 3 percent of SNG loans is used to finance
financed through the RIDF facility. Implementation of RIDF government buildings (Figure 3.11).
26 RIDF is a loan facility to increase access to infrastructure financing at subnational level, funded by the GoI (US$200 million), the World Bank (US$100 million), and Asian
Infrastructure Investment Bank (US$100 million).
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 25
T A B L E 3 . 4 - SNG Borrowing by Regional Level
Regional 2016 2017 2018 2019 2020 2021 2016-2021
level Blns of Blns of Blns of Blns of Blns of Blns of Blns of
# SNG # SNG # SNG # SNG # SNG # SNG # SNG
Rp Rp Rp Rp Rp Rp Rp
District/
2 389 4 731 14 2,322 14 2,376 30 4,609 29 4,809 93 15,236
Kabupaten
City/Kota 0 0 1 238 2 240 2 231 7 1,750 9 1,867 21 4,326
Province/
0 0 1 27,101 2 1,000 0 0 15 18,008 7 7,292 25 53,401
Provinsi
Total 2 389 6 28,070 18 3,562 16 2,607 52 24,367 45 13,968 139 72,963
Source: MoF/MoHA and World Bank staff calculations.
F I G U R E 3 . 1 0 - Subnational Government Borrowing F I G U R E 3 . 1 1 - Sectoral Composition of
by Source Subnational Borrowing from PT. SMI and BPDs
(Billions of Rp and percent, 2016-20) (2016-19) (%)
BPD, Roads and bridges
7%
Hospital
Market
SLA, 37% Government buildings
Hospital
PT SMI -
PEN, 43% Water supply
Others
Drainage
PT SMI -
regular, Sport facilities
13%
0 20 40 60 80
Source: World Bank calculations based on data from MoF/MoHA. Source: World Bank staff calculations based on data from MoF/
Note: Data are based on borrowing proposals approved by MoHA/MoF. MoHA. Note: Nonweighted ratio is applied to estimate each sector
loan allocation from multiple sectors data. PEN 2020 and 2021 data
do not have sectoral information.
3.3 Is there a sufficient supply of credits for subnational borrowing?
Indonesia’s credit market for subnational infrastructure domestic market‒in terms of both local banking assets and
investment is still underdeveloped. About 93 percent institutional investors‒is inadequate to keep pace with the
of outstanding subnational debt comes from nonmarket demand for infrastructure financing. It is estimated that local
sources comprising loans from PT. SMI and on-lending by banks have, at most, US$10 to 20 billion of cumulative room
the central government (SLAs) (Figure 3.10). Only 7 percent for infrastructure loan portfolio growth per year before reaching
of subnational loans are sourced from BPDs where SNGs their allocation limit‒compared to the private financing need of
often own shares. Borrowing from commercial banks or bond US$49 billion per year (World Bank 2018).
issuance by SNGs have not taken place. At the aggregate
level, Indonesia’s market capitalization of listed domestic There is a mismatch between infrastructure financing
companies has been growing but remains low at 47 percent needs and commercial bank’s asset liquidity which is not
of GDP‒compared to Thailand (108 percent), Malaysia (105 structured to finance medium- and long-term infrastructure
percent), and the Republic of Korea and Japan (133 percent projects. There are 107 commercial banks in Indonesia,
of GDP each) (World Bank 2021b). comprising four state-owned banks, 27 BPDs, 68 domestic
private banks, and eight branch offices of foreign banks (OJK
There is insufficient supply of financing for long-term 2021). Most banking sector liquidity is in the form of short-term
infrastructure investment at the subnational level. At deposits (less than one year), while only 12.6 percent of liquid
the aggregate level, the amount of capital available in the assets is in the form of term deposits of more than one year.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 26
In addition, most of the loans to SNGs from PT. SMI have a potential source of long-term funding, their investments are
maturity of 3-5 years which is also considered a short term concentrated in less-risky and of higher-liquidity instruments‒
from an infrastructure financing perspective which has long- such as term deposits and government bonds. In addition, the
lasting economic benefits. This significantly constrains banks’ domestic institutional investor base such as pension funds,
ability to lend to longer-term infrastructure projects as it would social security funds, and insurance companies is also small
create maturity mismatches. The regulation to limit SNG loans (totaling approximately US$119 billion or 12 percent of GDP)‒
to such short terms should be reviewed and revised. with negligible growth in recent years (World Bank 2019b).
Most subnational borrowing from the BPDs is for cash Despite interest, no SNGs have been successful in
management purposes. Only a limited number of SNGs issuing municipal bonds. During 2011-17 several SNGs
borrowed for investment purposes. The 27 BPDs hold a total had expressed interest in issuing bonds and conducted
of Rp 674 trillion of third-party funds (4 percent of GDP), creditworthiness assessments through an independent
however, only 14 percent of liquid assets are in the form of agency, however, none of them have been successful in
term deposits of over 12 months. At present, only 12 percent of issuing municipal bonds due to various challenges‒including
total credits in BPDs are allocated for investment. By location, political, regulatory, and capacity. The bond market in
third-party funds in commercial banks are highly concentrated Indonesia is small and costly for issuers. The total market size
in the capital city, Jakarta (Figure 3.13). is only 15.4 percent of GDP‒consisting mostly of government
bonds‒while the corporate bond market is only 2.3 percent
NBFIs can be a potential source of long-term financing of GDP (2015). There are no corporations that have issued
when local debt market is more developed. While NBFIs long-term bonds of 10, 15, or 20 years. The maturity terms
including pension funds (BPJS Employment), financiers of corporate bond issues (2014) ranged from one to seven
(for example, Hajj Fund), and insurance companies are a years‒with a median of only three years (World Bank 2019b).
F I G U R E 3 . 1 2 - Composition of Time Deposits of F I G U R E 3 . 1 3 - Third-party Funds in Commercial
Third-party Funds of Commercial Banks (%) Banks by Province (%)
> 12 month,
12.6
Others,
18.56 Banten, 3.2
6 month, DKI
13.9 1 month, Jakarta,
45.4 51.66 North Sumatera, 4
Central Java,
4.99
3 month,
28.2 East
Java, West Java, 8.12
9.47
Source: OJK (Statistik Perbankan Indonesia September 2021). Source: OJK (Statistik Perbankan Indonesia September 2021).
Note: Percentage totals may not necessarily equal 100 percent due to the Note: “others” refers to provinces with a share of third-party funds
effect of rounding. < 2%.
B O X 3 . 1 - Previous Attempts to Issue Subnational Bonds in DKI Jakarta and West Java
• In 2012, DKI Jakarta cancelled their bond issuance plan due to a change in political leadership and a significant
provincial cash surplus (SiLPA). The issuance value was proposed at Rp 1.2 trillion, but the province had accumulated
a significant cash surplus of Rp 9.5 trillion in the same fiscal year pointing to budget execution challenges.
• West Java discontinued their bond issuance plan for Kertajati Airport after the issuance of Law No. 23/2014 that
prohibits provincial governments from building international airports as it is the responsibility of central government,
although the central government subsequently provided financing for the airport development.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 27
3.4 Do borrowing rules discourage subnational borrowing?
Aligned with international practices, Indonesia has seem to discourage SNGs from borrowing. Indonesia’s
established eligibility criteria related to debt servicing, subnational borrowing regulatory framework is characterized
outstanding debt, and new borrowing (Appendix 4). There as a direct control system considering the multistage loan
are four eligibility criteria imposed by the central government authorization process, annual debt ceilings set by central
for subnational borrowing as stipulated in Government government, and prohibition of external borrowing. The direct
Regulation No. 56/2018: (i) outstanding loans plus proposed control approach is indeed more frequently used by unitary
loans should be less than 75 percent of the previous year’s countries and less often by federal countries (Ter-Minassian
nonearmarked revenue; (ii) the debt service coverage ratio and Craig 1997). The layered requirements create inflexibility
should be at least 2.5;27 (iii) the loan value should not exceed to the overall process. The implementation of subnational
the maximum allowed deficit set by the MoF Regulation each loan approval lacks consistency and transparency. The World
year; and (iv) the annual debt service should not exceed Bank’s team observations suggest that, in practice, affirmative
15 percent of the sum of the general allocation fund (Dana recommendations from MoHA for subnational loans delivered
Alokasi Umum: DAU) and revenue sharing transfer (Dana through PT. SMI could take up to 12 months‒diverging from the
Bagi Hasil: DBH). The central government reserves the right 15 days as stipulated in the regulation. Loan tenor provision is
to redirect the DAU and DBH transfers if the SNG fails to pay not solely based on the risk assessment or recommendation
the interest and principal on loans in a timely manner (intercept by PT. SMI as the lender but could be rebutted by MoHA. In
mechanism to mitigate borrower credit risk and liquidity risk).28 some instances, MoHA could provide a different loan tenor
recommendation to the borrowing proposals to align with the
Despite the presence of fiscal rules, the regulatory term of the mayor/governor. The evaluation against fiscal
framework of subnational borrowing in Indonesia is a de rules is the authority of MoF, but it comes after the DPRD
facto direct control system (Box 3.2). In the past, SNGs had approvals and recommendations from MoHA. Furthermore,
to get approval from their respective DPRD (the Legislative the aggregate limit for SNG debt‒renewed annually by a MoF
Council or the Local Parliament) before submitting their loan Regulation‒has been relatively restrictive. For 2022, total new
proposals to the central government (Figure 3.14). There were SNG borrowing is limited in effect to 0.32 percent of projected
no clear criteria governing the assessment by DPRD and this GDP, or about Rp 57.3 trillion. While total annual subnational
first stage of approval could be highly political. The survey result borrowings have not reached this amount, it is only one-third
on subnational debt management indicated DPRD approvals of the capital grants (the DAK Fisik) from central government
as one of the top-ranked challenges in submitting debt to SNGs in 2021.
proposals. DKI Jakarta also stated that one of their borrowing
proposals was once rejected by their DPRD. In the second Subnational bonds/sukuk issuance are also hampered by
stage, SNGs had to receive an affirmative recommendation layered and possibly overlapping requirements from the
from MoHA for each new debt obligation. According to government. While general requirements for issuing bonds/
Government Regulation No. 56/2018, evaluation by MoHA sukuk by SNGs are almost like borrowings/loans, the specific
includes: (i) proposed activities’ alignment with SNG planning requirements regarding the timing and validity of financial/audit
and budget documents; (ii) proposed activities’ alignment reporting potentially prevents sufficient time for subnational
with local service responsibilities and/or national priorities; bonds issuance. Government Regulation No. 56/2018 (under
and (iii) proposed borrowing’s alignment with other sources revision) requires the submission of audited SNG financial
of funding. Although the criteria applied by MoHA to assess reports while, under OJK regulations, the validity period of
a debt proposal are clearly stipulated under the regulation, such audit reports is a maximum of one year. Considering all
the process for obtaining the MoHA recommendation remains other requirements (recommendation and approval processes
unclear. Evaluation by MoHA had little relationship to the from MoHA and MoF, availability of draft local budget), the
actual SNGs creditworthiness or borrowing capacity.29 stipulation of such a short audit report validity period might
create a very short window for the issuance of SNG bonds/
The application of both direct control and fiscal rules sukuk.
regimes combined with inconsistent implementation
27 The annual debt service should be the equivalent of less than 40 percent of nonearmarked revenue.
28 The intercept rule serves as a security mechanism as revenues and/or assets of SNGs may not be pledged as collateral for loans.
29 In fact, this process was prone to bribery acts. In February 2022, The Jakarta Post reported that the Corruption Eradication Commission detained the former Director
General of Regional Finance in MoHA for allegedly demanding kickbacks from SNG officials derived from PEN loans. The detainee had the authority to approve the PEN
loan proposals submitted by SNGs.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 28
B O X 3 . 2 - Categories of Control Over Subnational Debt
Subnational borrowing regulations and debt supervision can be classified into four categories, according to the degree of
fiscal autonomy left to SNGs (Figure B.3.1). First and on the left, direct or administrative control by the central government
is associated with the lowest degree of fiscal autonomy. This approach gives authority to the central government to directly
control the SNG debt through different instruments, including debt ceilings, prohibition of external borrowing, and prior approval
of conditions for any new debt (Martinez-Vazquez and Civelek 2019). Second, fiscal rules are an intermediate form of control.
Fiscal rules typically “mimic” market discipline by linking limits on the indebtedness of SNGs to indicators of their debt-servicing
capacity‒such as past revenues or the tax base. Fiscal rules differ from direct controls, although the distinction is not always
easy to establish. Whereas fiscal rules are numerical, lasting, and apply to large fiscal aggregate, direct controls tend to be more
“procedural” (some may not even rely on a numerical
threshold), and ad hoc and possibly revised every year
F I G U R E B 3 . 1 - Categories of Control Over
(Eyraud et al. 2020). Third, cooperative arrangements,
Subnational Debt
where SNGs are in an active dialogue with the central
Less Subnational autonomy More Autonomy
government regarding the national deficit target, as
well as debt ceilings for SNGs. Given the intensive
dialogue between SNGs and the central government,
this approach raises general awareness about the
implications of subnational borrowing on macroeconomic Subnational debt to GDP (%) by categories of control, in the latest
stability. Fourth and on the right, market discipline, where year available
constraints are indirectly imposed by investors, provides 20
the highest degree of autonomy.
16 Sweden
There has not, however, been an a priori agreement on
what type of regulations are most effective. Ter-Minassian Finland
12 Brazil
and Craig (1997) concluded that sole reliance on market- France
based regulations is unlikely to be effective and that a rule- South
Poland Africa
based approach is generally preferable to administrative 8
Turkey Czech Portugal
control. Nevertheless, as Balassone et al. (2002) Republic
Slovenia Colombia
examined from the experiences of European countries, 4 Hungary Argentina
Nigeria Russia
the effectiveness of fiscal rules can be compromised if Mexico Romania
only central governments are held accountable. Most Indonesia Bulgaria
0
recently, Martinez-Vazquez and Vulovic (2017) found 0.5 1.5 2.5 3.5 4.5
none of the four broad types of subnational borrowing Direct control Fiscal rules Cooperative Market disciplines
regulations seem to have a distinct significant direct effect
Source: Eyraud et al (2020), adapted from Ter-Minassian and Craig
on fiscal sustainability at the subnational level. Each type (1997); Martinez-Vazquez and Civelek (2019); SNG-WOFI database
of regulation has advantages and disadvantages, with by OECD; and World Bank staff calculation.
varying suitability to a country’s circumstances.
The current subnational borrowing approval process by SNGs to submit their borrowing application and for PT.
is fragmented and lacks a coordinated and centralized SMI (and other key stakeholders in central government) to
approval system. Subnational borrowing approval processes monitor the borrowing implementation. While this system
are fragmented and involve multiple ministries/agencies could be useful to ensure the governance quality of the
including MoHA, the Ministry of National Development and borrowing, unfortunately it has not been utilized to process
Planning (Kementerian Perencanaan dan Pembangunan the recommendation provision/approval, particularly from
Nasional: Bappenas), MoF, and OJK. The system that MoHA. MoHA developed its own system named SIMANDA
facilitates the subnational borrowing approval process is also (Sistem Informasi Pinjaman Daerah/Subnational Borrowing
fragmented. PT. SMI established an online approval system Information System). Both systems are neither connected nor
called ReFiNa (Regional Financing System), which is used interoperated.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 29
The government has taken steps to improve the (MoF, MoHA, and Bappenas); (iii) expanding financing
regulatory framework of subnational borrowing through sources for SNGs to include sukuk and borrowing in the form
the recently approved Intergovernmental Harmonization of program loans/budget support; (iv) including flexibility to
Law (No. 1/2022). The Law includes measures to address key adjust maximum threshold for budget deficit and debt financing
regulatory constraints to SNG borrowing by: (i) streamlining for SNGs in the case of emergency; and (v) strengthening
loan approvals by the DPRD into the annual budget process; administrative sanctions to ensure timely debt repayment
(ii) relaxing the loan tenor beyond the political term of the where the head of the SNG and DPRD members‒such as by
SNG head subject to meeting technical criteria and receiving not providing incentives/salaries for six months if they fail to
consideration from relevant central government agencies allocate a budget for debt repayment.
F I G U R E 3 . 1 4 - Timeline for Subnational Debt Authorization Process
Planning
Local governments plan activities to
be financed from loans
Legislative Council
EIA Preparation (UKL-UPL) Approval
Master Plan Detail Engineering The local government asks
Design (DED) for Legislative Council
Feasibility Study approval for the loan plan
3 months 3 months 3 months 3 months
MoF Consideration MoHA Consideration
Received consideration from the Received consideration from the Proposed
Minister of Finance Minister of Home Affairs Loan Plan
15 working days* 15 working days*
Submission to Regional Head
Approval of MoF is submitted to the
regional head with a copy to the MoHA
Submission of
Loan Agreement Loan Execution Construction
Agreement
1-2 years
The loan agreement depends on The Regional Head submits a Withdrawal and/or distribution
the origin of the loan fund: copy of the agreement to the of central government loans to
• Internal Subsidiary Minister of Home Affairs local government
Loan Agreement
• Foreign Subsidiary
Loan Agreement
• Loan Agreement
Loan will be given a maximum of 40 working days from the time the required
documents are complete and correct.
Source: Handbook for Subnational Government Alternative Financing Modalities (World Bank 2021).
Note: (*) after receipt of the complete and correct Regional Loan plan proposal documents
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 30
4.
>>>
Conclusions and Recommendations
4.1 Conclusion
This note reviews selected areas of SNG debt carrying and management capacities, and
borrowing rules, in Indonesia to explore opportunities for enhanced and prudent local
borrowing for investments. Annual borrowing by SNGs remains well below the maximum
threshold set by the MoF. This is reflective of a combination of shallow financial markets and
weak debt management capacity which, in turn, have called for prudent/conservative fiscal rules
for subnational debt. Efforts to strengthen SNGs access to debt financing should focus around
high-borrowing capacity SNGs. The findings highlight several constraints for SNGs in accessing
debt financing: (i) lack of data and a systematic approach in assessing subnational borrowing
capacity; (ii) lack of debt management capacity at subnational level; (iii) inadequate financial
markets for long-term infrastructure investment; and (iv) stringent and complex subnational
borrowing rules.
4.2 Suggested Reforms
To contribute to a gradual and prudent increase in access to subnational debt markets, the
authorities could consider focusing selected reforms in these areas.
Reform One: Consider further developing a framework to systematically assess
subnational borrowing capacity.
This will entail filling data gaps required for such a framework. The existing approach that focuses
on subnational fiscal capacity to monitor compliance with subnational fiscal rules does not
consider other key parameters such as economic conditions and quality of fiscal performance.
Developing a framework to systematically assess SNG borrowing capacity can inform authorities
in designing: (i) a subnational finance framework based on their creditworthiness characteristics
(for example, loans, grants). Borrowing capacity ranking can also serve as eligibility criteria
for borrowing; and (ii) appropriate and targeted interventions (considering variation in SNGs
characteristics) to promote subnational borrowing by focusing on high-borrowing capacity SNGs.
This will also require concerted efforts to improve data gaps at subnational level. The proposed
framework discussed in this paper could be a starting point.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 31
Reform Two: Improve central government’s monitoring municipal bonds issuance or establishing a single-window
and supervision of subnational borrowing and debt. body/committee for approval and monitoring subnational
borrowing to help expedite the approval processes in a more
Strengthening monitoring and supervision of subnational debt coherent manner. This body/committee could comprise a
by central government will help ensure prudent borrowing joint team from the MoF, MoHA, Bappenas, and other key
practices and help to address concerns over possible stakeholders. The role of such a body can be expanded to
relaxation of ex ante approval processes. In the short provide coordinated technical support to SNGs in project
term, this can include establishing a centralized and online preparation and selection‒for example, a project development
subnational borrowing/debt database (such as amount and facility. The authorities could also initiate the connectivity or
terms of loans). Authorities can also explore good practices interoperability between the ReFiNa and SIMANDA systems.
that have been implemented in other countries such as traffic These systems will be critical in ensuring the governance
light system in Colombia and credit ratings assessment for quality of borrowing and harmonized and coordinated
SNGs in Mexico (Box 4.1). borrowing approval processes.
Reform Three: Leverage the central government’s DMO Reform Five: Finally, on the supply of finance, consider
in strengthening subnational debt management capacity. reviewing the effectiveness and opportunities of
intermediary financial institutions (PT. SMI) as the main
The DMO could play a role in providing capacity building and source of subnational borrowing.
transferring good practices. The central government’s DMO
has relatively good capacity and experience. Sound debt In the short term, authorities can carry out a review to assess
management capacity can help improve SNGs’ understanding effectiveness and opportunities of the existing arrangement
of costs and benefits of various debt instruments. The survey and practices in providing loans for SNGs. This is particularly
findings indicate that most SNGs lack debt management relevant since subnational debt has increased due to regional
capacity and have not met minimum criteria in some core areas PEN programs as part of fiscal stimulus for the economic
of debt management practices. Furthermore, to identify critical recovery. This can include reviewing lessons from other
gaps and targeted technical assistance, the authorities can countries that have set up intermediary institutions or credit
carry out more in-depth assessments for debt management enhancement schemes to mobilize subnational debt financing
capacity (such as DeMPA) by focusing on high-borrowing (Box 4.2). In the medium term, as the subnational debt market
capacity SNGs. becomes more developed, the authorities can carry out further
reviews to identify supply-side constraints for market sourced
Reform Four: To complement these efforts, further financing such as bonds/securities or commercial loans from
streamline and improve the transparency of subnational banking institutions. A deeper analysis of mismatch between
borrowing approval processes at the central level. banking liquidity and long-term infrastructure financing can
help facilitate access to market-sourced credits for SNGs.
This could be done through a more coordinated approach, for
example by empowering the existing joint team for accelerating
B O X 4 . 1 - Monitoring Subnational Borrowing
During 1993-97, Colombia established a “traffic light” system that links each SNG’s debt to its payment capacity and
applies different borrowing discretion, thereby alerting the central government to potentially excessive subnational debt.
The system consists of three layers for local borrowing discretion: (a) red light: if SNGs are highly indebted they are not
allowed to borrow anymore; (b) yellow light: if SNGs have some level of debt, but not in excessive amounts, then it requires
approval from the Ministry of Finance; and (c) green light: all other SNGs with low levels of debt are allowed to borrow at
their will (Ma and Brixi 2002). Mexico requires subnational credit ratings for SNGs who will borrow from banks, along with
a requirement to register subnational loans in the public debt records system of the Ministry of Finance. This reform has
contributed to improved transparency, reduced risks, and lowered borrowing costs, as well as improved investor and rating
agencies confidence that led to an expansion of the municipal bond market in Mexico (World Bank 2019c).
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 32
B O X 4 . 2 - Selected Country Experiences with Creditworthiness Enhancement
Selected countries introduced credit enhancement entities and guarantee programs that facilitate both direct and indirect
market access by sub sovereign borrowers to support the development of sub sovereign infrastructure finance. These include
the following:
• Colombia: Territorial Financing Institution (FINDETER), a second-tier specialized development bank, funded partially in the
market. It acts as a “second-tier” bank that discounts qualifying subnational loans made by commercial banks, providing
the original lending banks with a source of liquidity on loans they have made to subnational units;
• India: Tamil Nadu Urban Development Fund, a state-sponsored municipal development fund transformed into public-
private management/funding and loan pooling scheme;
• Philippines: LGU Guarantee Corporation (LGUGC), a public-private owned municipal bond guarantee company. It provides
bondholders with a guarantee of uninterrupted debt service (principal and interest);
• South Africa: Infrastructure Finance Corporation of South Africa (INCA), a purely private specialized financial institution to
purchase debt obligations and provide loans, funded by donors on a subordinated basis. It involves support for liquidity
and of the secondary market. This type of credit enhancement is, however, seldom used in developing countries but can
be important in middle-income countries with active bond markets where subnational securities present specific liquidity
problems for private sector investors;
• Hungary, China, and Croatia: Local partial credit guarantee programs funded by the Global Environment Facility (GEF) for
local bank loans for energy-efficient investments. The guarantor covers a portion of debt service payments (regardless the
cause of debt service default); and
• United States: Bond banks and state revolving funds, which pool loans, provide adequate reserves, and issue bonds
in the market. In addition to achieving the same economies of scale that are possible with larger issuances of bonds,
this technique offers a reduction in risk through portfolio diversification, and ultimately results in reductions in the cost of
borrowing to local borrowers.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 33
>>>
References
Aldasoro, I., and M. Seiferling. 2014. “Vertical Fiscal Imbalances and the Accumulation of Government Debt.” IMF
Working Papers. Washington, DC: International Monetary Fund.
Balassone, F., D. Franco, and S. Zotteri. 2002. “Fiscal Rules for Sub-National Governments: What Lessons from
EMU Countries?” Paper presented at the IMF and World Bank conference on Rules-Based Macroeconomic
Policies in Emerging Market Economies Conference. Oaxaca, Mexico. 14–16 February.
Blanco, F. 2015. “Subnational Debt Management in Brazil and Mexico.” In Subnational Debt Management and
Restructuring: Lessons from International Experience, edited by S. Kahkonen and S. Gooptu. World Bank.
Bonomo, M., C. Frischtak, and P, Ribeiro. 2021. “Public Investment and Fiscal Crisis in Brazil: Finding Culprits and
Solutions.” Washington, DC: Inter-American Development Bank.
Canuto, O., and L. Liu. 2013. “Until Debt Do Us Part: Subnational Debt, Insolvency, and Markets.” Washington,
DC: World Bank. https://openknowledge.worldbank.org/handle/10986/12597
Diez, T.H. 2018. “How can Indonesia achieve a more sustainable transport system?” World Bank Blogs. https://
blogs.worldbank.org/transport/how-can-indonesia-achieve-more-sustainable-transport-system
Eyraud, L., A. Hodge, J. Ralyea, and J. Reynaud. 2020. “How to Design Subnational Fiscal Rules: A Primer.” IMF
How to Note 20/01. Washington, DC: International Monetary Fund.
Joshi, R., R. Gayathri, A. Permono, B. Ladda, A. Parasher, R. Hutagaol, and R. Prastiwan. 2015. “Consultant
Services for Market Assessment and Operating Framework for the Indonesia Regional Infrastructure Development
Fund: Final Report.” Consultant Report for the World Bank, CRISIL, Mumbai.
Kehew, R., T. Matsukawa, and J. Petersen. 2005. “Local Financing for Sub-Sovereign Infrastructure in Developing
Countries: Case Studies of Innovative Domestic Credit Enhancement Entities and Techniques.” The World Bank.
Köppl–Turyna, M., and H. Pitlik. 2018. “Do equalization payments affect subnational borrowing? Evidence from
regression discontinuity.” European Journal of Political Economy 53: 84-108.
Landon, S. 2003. “Subnational government borrowing in federal systems: evidence from Argentina and Mexico. In
Fiscal Relations in Four Countries: Four Essays, edited by P. Boothe. Ottawa: Forum of Federations. http://www.
forumfed.org/libdocs/FiscalRelBook03/FR-Landon.pdf
Leigland, J., and C. Mandri-Perrott. 2008. “Enhancing the creditworthiness of municipal bonds: Innovations from
Mexico.” Gridlines-PPIAF World Bank. (https://ppiaf.org/documents/3030/download)
Liu, L., M. De Angelis, and S. Torbert. 2017. “Municipal Pooled Financing of Infrastructure in the United States:
Experience and Lessons.” Policy Research Working Paper No. 8212. Washington, DC: World Bank. https://
openknowledge.worldbank.org/handle/10986/28547 License: CC BY 3.0 IGO.”
Liu, L., and J. Pradelli. 2012. “Financing Infrastructure and Monitoring Fiscal Risks at the Subnational Level.”
Research Working Paper 6069. Washington, DC: World Bank.
Ma, J., and H. Polackova. 2002. “Monitoring Fiscal Risks of Subnational Governments.” PREM Note 64.
Washington, DC: World Bank.
Martinez-Vazquez, J., and V. Vulovic. 2017. “How Well Do Subnational Borrowing Regulations Work?” In Central
and Local Government Relations in Asia: Achieving Fiscal Sustainability edited by N. Yoshino and P.J. Morgan.
Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing.
Martinez-Vazquez, J., and Y. Civelek. 2019. “Subnational Government Debt Governance: Lessons from Non-
Asian Emerging Economies.” ICEPP Working Paper 147. Atlanta: Georgia State University.
OECD/UCLG. 2016. “Subnational Governments Around the World: Structure and Finance: A first contribution to
the Global Observatory on Local Finances.”
———. 2019. “2019 Report: World Observatory on Subnational Government Finance and Investment: Key
findings.”
Smoke, P. 2019. “Improving Subnational Government Development Finance in Emerging and Developing
Economies: Toward a Strategic Approach.” ADBI Working Paper 921. Tokyo: Asian Development Bank
Institute. https://www.adb.org/publications/improving-subnational-government-development-finance-emerging-
developing-economies.
Ter-Minassian, T. 2017. “Promoting effective and fiscally sound local investments in infrastructure.” Washington,
DC: Brooking Institute.
Ter-Minassian, T. and J. Craig. 1997. “Control of Subna¬tional Government Borrowing.” In Fiscal Federalism in
Theory and Practice edited by T. Ter-Minassian. Washington, DC: International Monetary Fund.
World Bank. 2001. “Intergovernmental Fiscal Relations.” In Decentralization and Subnational Regional
Economics.”
——— . 2015. “Market Assessment and Operating Framework for the Indonesia Regional Infrastructure
Development Fund.” Unpublished.
——— . 2017. “Government Debt Management: Designing Debt Management Strategies.” Washington, DC:
World Bank.
——— . 2019a. “Cleaning Up Indonesia’s Urban Solid Waste.” World Bank. https://www.worldbank.org/en/
news/press-release/2019/12/05/cleaning-up-indonesias-urban-solid-waste
——— . 2019b. “Indonesia Infrastructure Sector Assessment Program (InfraSAP).” Washington, DC.
———. 2019c. (rep.). “Mexico: Ensuring Subnational Fiscal Sustainability.” World Bank. https://documents1.
worldbank.org/curated/zh/675421554210600315/pdf/Mexico-Ensuring-Subnational-Fiscal-Sustainability.pdf
———. 2019d. “Time to ACT: Realizing Indonesia’s Urban Potential.” Washington, DC: World Bank.
———. 2020a. “Indonesia Public Expenditure Review 2020: Spending for Better Results.” Washington, DC:
World Bank.
———. 2020b. “Indonesia Economic Prospects: The Long Road to Recovery July 2020.” Jakarta: World Bank.
———. 2021a. “Handbook for Subnational Government (SNG) Alternative Financing Modalities.”
———. 2021b. “Market capitalization of listed domestic companies (% of GDP) - Indonesia, Malaysia, Thailand,
China, Japan, Korea, Rep.” https://data.worldbank.org/indicator/CM.MKT.LCAP.GD.ZS?end=2020&locations=ID-
MY-TH-CN-JP-KR&start=2000
World Bank. 2021c. “Regional Infrastructure Development Fund (P154947).” World Bank. https://documents1.
worldbank.org/curated/en/786701621884889674/pdf/Disclosable-Version-of-the-ISR-Regional-Infrastructure-
Development-Fund-P154947-Sequence-No-10.pdf.
———. 2021d. “Strengthening Subnational Public Financial Management for Improved Service Delivery.”
———. 2022. “Dana Alokasi Khusus (DAK) Fisik Implementation Review.” Unpublished
World Bank and Ministry of Finance (China). 2015. “Subnational Debt Management and Restructuring: Lessons
from International Experience.”
Yilmaz, S., and R.D. Ebel. 2020. “Subnational Government, Infrastructure, and the Role of Borrowing and Debt.”
In Development Studies in Regional Science edited by Z. Chen, W.M. Bowen, and D. Whittington.
Zhao, Q., L. Zhang, B. Feng, W. Liu, X. Long, J. Sun, X. Chen, and Q. Xiao. 2020. “Managing the Risks of Public
Infrastructure Financing: Toward Sustainability.” ADBI Working Paper 1155. Tokyo: Asian Development Bank
Institute.
>>>
Appendix 1: Subnational Fiscal and Debt Data
T A B L E A . 1 . 1 - Descriptive Statistics of Size of Subnational Borrowing 2016-21 (Billions of Rp)
Size of Borrowing Province Kabupaten/Kota
Max 27,101 495
Min 33 9
Average 2,175 174
T A B L E A . 1 . 2 - Subnational Borrowing Outcomes 2016-21 (Billions of Rp and Number of SNGs/Loan Contracts)
Outcome Number Percentage
Borrowed 119 81.5
Cancelled - using APBD 9 6.2
Cancelled due to nonAPBD reasons 18 12.3
Total 146 100
T A B L E A . 1 . 3 - Subnational Borrowing Source 2016-21 (Billions of Rp and Number of SNGs/Loan Contracts)
2016 2017 2018 2019 2020 2021
Source Blns of Blns of Blns of Blns of Blns of Blns of
# # # # # #
Rp Rp Rp Rp Rp Rp
PEN 0 0 0 0 0 0 0 0 19,132 28 14,212 17
PT. SMI 389 2 659 3 2,574 11 1,053 7 3,824 13 919 4
BPD 0 0 310 2 988 7 1,360 8 657 8 697 4
BPD syndication with
0 0 0 0 0 0 195 1 213 1 0 0
other BPDs
BPD syndication with
0 0 0 0 0 0 0 0 120 1 0 0
State-owned Bank
BPD syndication with PT.
0 0 0 0 0 0 0 0 421 1 0 0
SMI
SLA 0 0 27,101 1 0 0 0 0 0 0 0 0
Total 389 2 28,070 6 3,562 18 2,608 16 24,367 52 15,828 25
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 37
>>>
Appendix 2: Methodology of Subnational
Borrowing Capacity Assessment
Subnational economic condition, fiscal performance, and quality of fiscal management are indicators used to
approximate the borrowing capacity. These are also the common criteria adopted by the three largest international rating
agencies (S&P, Moody’s, and Fitch) in assessing the SNGs’ creditworthiness.
1. Economic condition. To estimate the condition of the subnational economy, this analysis uses three indicators: (i)
economic stability, as measured by the standard deviation of the historical economic growth during 2011-20; (ii) size of the
local economy, as measured by the SNG’s GDP as a share of national GDP (average share 2016-19); and (iii) size of the
local population as measured by the share of national population (average share 2016-19).
2. Fiscal performance. This criterion includes two main indicators: (i) fiscal space; and (ii) the interest coverage ratio as a
solvency measure. Fiscal space30 measures the flexibility of SNGs revenue and expenditure, and it is defined as follows:
Fiscal space=Total revenue-earmarked revenue-mandatory expenditure
Part of the fiscal transfers from central government are earmarked to specific SNG spending: DAK; DBH of tobacco excise
and reforestation; special autonomy funds (Dana Otsus); and village funds (Dana Desa). Subtracting earmarked revenue
from total revenue yields the discretionary revenue. Finally, fiscal space is estimated by subtracting mandatory expenditure
from the discretionary revenue. Mandatory expenditure includes personnel spending, interest payments, and fiscal transfers
to the lower-level government. Besides fiscal space, this criterion also examines an SNG’s ability to meet its debt obligation,
measured by the interest payment as a share of total revenue. All indicators use actual fiscal outturn data in 2019.
3. Quality of fiscal management. To review the credibility of SNGs fiscal plans and their budgetary discipline, this
component looks at the extent to which actual spending or revenue (spending or revenue outturns) diverge from their
budgeted plans. Spending outturns that consistently exceed budgeted spending potentially reflect lack of credibility in
budgeting and weak budget discipline. Lower spending outturns can also point to potential problems of implementation
capacity. Similarly, consistent deviations between revenue outturns and their budgeted plan may indicate weak capacity in
an SNG’s revenue projection. Three fiscal outturn indicators are used in this assessment: (i) own-source revenue; (ii) capital
spending; and (iii) recurrent spending. In addition, the criterion also looks at the audit result on SNGs financial statements
by BPK. These four indicators use data between 2016-19.
Different measurement scales of indicators need to be standardized to allow comparisons between SNGs. This
analysis uses a simple standardization or normalization to put all the indicators into the same scale. This is achieved by
putting all indicators on a scale from 0 to 100, where 0 is the worst SNG (for a particular indicator), and 100 indicates the
best SNG for the same indicator. This standardized score, called the tz-score,31 eventually becomes the building block for
each SNG’s rank of borrowing capacity.The tz-score for each indicator is calculated as follows:
where x is the indicator’s value for the SNG; x_min is the lowest indicator’s value across all SNGs; and x_max is the highest
indicator’s value across all SNGs. For every indicator, this approach generates a value between 0 and 100 indicating
where the SNG lies on the scale of worst to best SNGs. The tz-score calculations for districts and provinces are conducted
separately.
30 The fiscal space calculation follows the approach of the DGFB (MoF) in estimating SNGs fiscal capacity. The estimation of SNGs fiscal capacity is stipulated in a MoF
Regulation (PMK) and updated annually. The fiscal capacity is then ranked from low to high across all SNGs. Later, it serves as a basis for the ceiling of SNG’s allowable
deficit relative to estimated local revenue. For example, in 2019, the ceiling of the budget deficit is 5 percent and 3 percent of total revenue for SNGs with very high and
very low fiscal capacity, respectively.
31 This analysis adopted the term “tz-score” from a study conducted by Neil McCulloch with KPPOD (Regional Autonomy Watch) and the Asia Foundation in 2007 titled
“Local Economic Governance Index”.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 38
Reversing the tz-score is required for indicators where larger values indicate worse performances. The tz-score is constructed
so that a higher score shows a better performance, however, some indicators have a reversed nature, where a higher value
would indicate a poorer performance. These indicators include: (i) economic stability or the standard deviation of SNG’s
historical real GDP growth; and (ii) interest payment as a share of total revenue. In addition, for the three fiscal outturn
indicators (the outturns of OSR, capital, and recurrent spending), the tz-score calculation is based on the deviation between
100 percent (the perfect outturn) and the outturn rate. The tz-score for fiscal outturn indicators, therefore, also need to be
reversed, as a larger deviation denotes a worse performance. For such indicators, the tz-scores are reversed simply by
calculating as follows.
The final score that represents each SNG’s borrowing capacity is the average tz-score of all creditworthiness
criteria. This analysis does not assign weights to any of the creditworthiness criteria to minimize subjectivity. All criteria and
indicators used in the analysis are treated equally. Finally, SNG’s borrowing capacity is ranked by decile for both district
and provincial governments. Ranking by decile tends to avoid too much heterogeneity within the same group. This analysis
defines SNGs with high- and low-borrowing capacity as those at the top and the bottom decile, respectively.
T A B L E A . 2 . 1 - List of High-borrowing Capacity SNGs and the Actual Borrowers (in alphabetical order)
No Subnational Government Have Borrowing (Yes/No) No Subnational Government Have Borrowing (Yes/No)
1 Kab. Badung No 26 Kab. Magelang No
2 Kab. Bandung No 27 Kab. Malang No
3 Kab. Bandung Barat Yes 28 Kab. Mojokerto No
4 Kab. Banyuasin Yes 29 Kab. Musi Banyuasin Yes
5 Kab. Banyumas Yes 30 Kab. Pasuruan No
6 Kab. Banyuwangi No 31 Kab. Probolinggo Yes
7 Kab. Bekasi No 32 Kab. Purbalingga No
8 Kab. Berau No 33 Kab. Serang No
9 Kab. Bogor No 34 Kab. Sidoarjo No
10 Kab. Bojonegoro No 35 Kab. Sleman No
11 Kab. Buleleng No 36 Kab. Sukabumi No
12 Kab. Cianjur No 37 Kab. Tangerang No
13 Kab. Cilacap No 38 Kab. Tegal No
14 Kab. Deli Serdang No 39 Kab. Teluk Bintuni No
15 Kab. Garut No 40 Kab. Tuban No
16 Kab. Gresik No 41 Kab. Tulungagung No
17 Kab. Indramayu No 42 Kota Bandung No
18 Kab. Jember No 43 Kota Bekasi No
19 Kab. Kampar No 44 Kota Depok No
20 Kab. Karawang No 45 Kota Medan No
21 Kab. Kebumen No 46 Kota Palembang No
22 Kab. Kutai Kartanegara No 47 Kota Samarinda No
23 Kab. Kutai Timur No 48 Kota Semarang No
24 Kab. Lamongan No 49 Kota Surabaya No
25 Kab. Lombok Timur No 50 Kota Tangerang No
51 Kota Tangerang Selatan No
52 Prov. DKI Jakarta Yes
53 Prov. West Java Yes
54 Prov. Central Java No
55 Prov. East Java Yes
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 39
T A B L E A . 2 . 2 - Statistic (Average Values) Across Indicators of Borrowing Capacity
1. Economic Condition
Performance 2. Fiscal 3. Quality of Fiscal
Avg debt Management
# SNGs, size per
Overall SNGs # SNGs actual SNG Econ
Population Fiscal Interest Recurrent
borrower (Billions of stability Econ size OSR Capex Audit
size (% space payment exp
Rp) (stdev (% national outturn (% outturn (% status* on
national (Billions of (% outturn (%
econ GDP) planned) planned) fin reports
pop) Rp) revenue) planned)
growth)
High-
borrowing (Decile 1) 55 8 5,355 2.9 1.6 1.6 3,254 0.01 98.0 98.4 96.4 3.9
capacity
Median (Decile 5) 54 7 239 3.7 0.1 0.2 701 0.01 96.8 99.5 97.2 3.7
Low-
borrowing (Decile 10) 55 18 216 3.9 0.1 0.1 634 0.23 170.4 95.8 93.1 3.6
capacity
1. Economic Condition Performance
2. Fiscal 3. Quality of Fiscal
Avg debt Management
# SNGs, size per Econ
Provinces # SNGs actual SNG Population Fiscal Interest Recurrent
stability Econ size OSR Capex Audit
borrower (Billions of size (% space payment exp
(stdev (% national outturn (% outturn (% status* on
Rp) national (Billions of (% outturn (%
econ GDP) planned) planned) fin reports
pop) Rp) revenue) planned)
growth)
High-
borrowing
capacity,
(Decile 1) 3 2 2,057 2.5 12.5 16.2 9,969 0.01 92.6 92.6 99.0 4.0
except
DKI
Jakarta
High-
borrowing
capacity,
(Decile 1) 1 1 37,032 2.6 16.6 5.9 37,427 0.10 110.8 110.8 85.9 3.8
DKI
Jakarta
only
Median (Decile 5) 3 1 345 2.3 0.9 1.3 1,558 0.00 98.1 98.1 95.8 4.0
Low-
borrowing (Decile 10) 4 2 291 2.6 0.4 0.4 996 0.23 108.8 108.8 90.9 3.9
capacity
1. Economic Condition
Performance 2. Fiscal 3. Quality of Fiscal
Avg debt Management
# SNGs, size per
Districts # SNGs actual SNG Econ
Population Fiscal Interest Recurrent
borrower (Billions of stability Econ size OSR Capex Audit
size (% space payment exp
Rp) (stdev (% national outturn (% outturn (% status* on
national (Billions of (% outturn (%
econ GDP) planned) planned) fin reports
pop) Rp) revenue) planned)
growth)
High-
borrowing (Decile 1) 51 5 339 3.0 0.7 0.7 2,189 0.01 98.1 98.4 96.4 3.9
capacity
Median (Decile 5) 51 6 222 3.8 0.1 0.1 651 0.01 96.7 99.6 97.3 3.7
Low-
borrowing (Decile 10) 51 16 206 4.0 0.1 0.1 606 0.23 175.3 94.8 93.3 3.5
capacity
Source: World Bank staff calculation, using 2016-19 data.
Note: *) Audit result is converted to the following numerical unit: Unqualified (WTP) = 4; Qualified (WDP) = 3; Disclaimer (TMP) = 2; Adverse (TW) = 1.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 40
1. Economic Condition Performance
2. Fiscal 3. Quality of Fiscal
Avg debt Management
# SNGs, size per Econ
Overall SNGs # SNGs actual SNG Population Fiscal Interest Recurrent
stability Econ size OSR Capex Audit
borrower (Billions of size (% space payment exp
(stdev (% national outturn (% outturn (% status* on
Rp) national (Billions of (% outturn (%
econ GDP) planned) planned) fin reports
pop) Rp) revenue) planned)
growth)
High-
borrowing (Decile 1) 55 8 5,355 2.9 1.6 1.6 3,254 0.01 98.0 98.4 96.4 3.9
capacity
Median (Decile 5) 54 7 239 3.7 0.1 0.2 701 0.01 96.8 99.5 97.2 3.7
Low-
borrowing (Decile 10) 55 18 216 3.9 0.1 0.1 634 0.23 170.4 95.8 93.1 3.6
capacity
1. Economic Condition Performance
2. Fiscal 3. Quality of Fiscal
Avg debt Management
# SNGs, size per Econ
Provinces # SNGs actual SNG Population Fiscal Interest Recurrent
stability Econ size OSR Capex Audit
borrower (Billions of size (% space payment exp
(stdev (% national outturn (% outturn (% status* on
Rp) national (Billions of (% outturn (%
econ GDP) planned) planned) fin reports
pop) Rp) revenue) planned)
growth)
High-
borrowing
capacity,
(Decile 1) 3 2 2,057 2.5 12.5 16.2 9,969 0.01 92.6 92.6 99.0 4.0
except
DKI
Jakarta
High-
borrowing
capacity,
(Decile 1) 1 1 37,032 2.6 16.6 5.9 37,427 0.10 110.8 110.8 85.9 3.8
DKI
Jakarta
only
Median (Decile 5) 3 1 345 2.3 0.9 1.3 1,558 0.00 98.1 98.1 95.8 4.0
Low-
borrowing (Decile 10) 4 2 291 2.6 0.4 0.4 996 0.23 108.8 108.8 90.9 3.9
capacity
1. Economic Condition
Performance 2. Fiscal 3. Quality of Fiscal
Avg debt Management
# SNGs, size per
Districts # SNGs actual SNG Econ
Population Fiscal Interest Recurrent
borrower (Billions of stability Econ size OSR Capex Audit
size (% space payment exp
Rp) (stdev (% national outturn (% outturn (% status* on
national (Billions of (% outturn (%
econ GDP) planned) planned) fin reports
pop) Rp) revenue) planned)
growth)
High-
borrowing (Decile 1) 51 5 339 3.0 0.7 0.7 2,189 0.01 98.1 98.4 96.4 3.9
capacity
Median (Decile 5) 51 6 222 3.8 0.1 0.1 651 0.01 96.7 99.6 97.3 3.7
Low-
borrowing (Decile 10) 51 16 206 4.0 0.1 0.1 606 0.23 175.3 94.8 93.3 3.5
capacity
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 41
>>>
Appendix 3: Approach and Methodology of
Subnational Debt Management Capacity
Survey
There are 164 SNGs (30 percent of total SNGs) selected as survey sample. The selection of SNGs as survey sample is
done through purposive sampling approach and takes into account the following criteria: (i) SNGs that have outstanding
debt; (ii) SNGs that have planned/proposed to borrow but did not get approval (rejected); (iii) SNGs that have not borrowed
but have high-borrowing capacity (based on borrowing capacity assessment); (iv) representation by level of government
(Province, Kabupaten, Kota); and (v) representation by geographical area (Java/Bali, Sumatra, Kalimantan, Sulawesi,
Maluku/Papua). The main respondent of the survey is a unit/function that oversees borrowing/debt management activities
within SNGs which can be an existing function within the local finance/planning agency (BPKAD or Bappeda) or in the form
of an independent body (Figure A.3.1 and Table A.3.1).
The online survey was conducted from 27 December 2021 to 31 January 2022. The team also conducted a technical FGD on
19 January 2022 to assist SNGs in filling in the survey questionnaire. The survey covers 10 elements in debt management
practices: (i) Data/Information on Regional Debt/Loans; (ii) Regulatory and Institutional Framework; (iii) Institutional Structure/
Settings; (iv) Debt Planning and Debt Management Strategy; (v) Cashflow Projection and Management; (vi) Debt Recording
and Accounting; (vii) Regional Debt Reporting; (viii) Audit; (ix) Policy Coordination; and (x) Evaluation.
F I G U R E A . 3 . 1 - Schematic Selection Process of Sample
T A B L E A . 3 . 1 - Key Criteria of Sample Selection Process
SNGs have not borrowed
SNGs with borrowing
Province SNGs have borrowed but have high-borrowing Total
proposal rejected
capacity
Java/Bali 24 35 0 59
Kalimantan 9 4 2 15
Maluku/Nusa Tenggara 12 1 0 13
Papua 7 1 0 8
Sulawesi 30 0 1 31
Sumatra 29 7 2 38
Total 111 48 5 164
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 42
Several caveats of the survey results
1. The study has a relatively low participation rate. Only 39 out of 164 invited SNGs returned the questionnaire, or only about
24 percent. The author made sure that higher participation from SNGs can be achieved through extending the questionnaire
submission period and holding technical FGD sessions. Furthermore, a statistics test to ensure a representative sample
was not conducted yet.
2. Submitted data will need further confirmation. We found some differences in debt/loans value–possibly due to differences
in planned vs. actual debt/loans.32 Some respondents also did not complete all the questions‒especially the optional
questions.
T A B L E A . 3 . 2 - List of Survey Respondents
No Kabupaten (Districts) No Kota (Cities) No Provinces
1 Kabupaten Sampang 1 Kota TebingTinggi 1 DKI Jakarta
2 Kabupaten Lampung Barat 2 Kota Bandar Lampung 2 North Kalimantan
Kabupaten Penukal Abab Lematang Kota Mojokerto South Sulawesi
3 3 3
Ilir
4 Kabupaten Tulungagung 4 Kota Palembang 4 South Sumatra
5 Kabupaten Probolinggo 5 Kota Bekasi 5 Lampung
6 Kabupaten Tapanuli Tengah 6 Kota Banda Aceh 6 West Java
7 Kabupaten Lampung Tengah 7 Banten
8 Kabupaten Tapanuli Utara 8 Central Java
9 Kabupaten Grobogan 9 West Sulawesi
10 Kabupaten Bojonegoro
11 Kabupaten Karanganyar
12 Kabupaten Sinjai
13 Kabupaten Sikka
14 Kabupaten Musi Banyuasin
15 Kabupaten Badung
16 Kabupaten Sragen
17 Kabupaten Gresik
18 Kabupaten Bengkayang
19 Kabupaten Ponorogo
20 Kabupaten Lumajang
21 Kabupaten Kepulauan Sangihe
22 Kabupaten Gianyar
23 Kabupaten Trenggalek
24 Kabupaten Tapin
32 Actual debt/loans are based on realization of physical infrastructure.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 43
T A B L E A . 3 . 3 - Core Minimum Requirements for Effective Subnational Debt Management
Dimension Assessment Criteria
Legal/regulatory framework • Within the framework of the national constitution and restrictions imposed by
the central government, the legislation–national and subnational‒provides clear
authorization for the subnational entity to borrow and to issue loan guarantees
(where applicable). In addition, the legislation specifies the purposes for which the
subnational entity can borrow and include reporting requirements.
Institutional and staffing arrangements • There should be a clear division between the political level (local assembly, local
executive branches such as governor/mayor) that sets objectives and decides on
the debt management strategy and the execution level that implements the debt
management strategy.
• Debt management of the SNG is undertaken by a dedicated DMO within the
organizational structure of the Treasury, or similar. Within the unit there is a clear
organizational separation between the entity undertaking transactions (front office)
and the entity responsible for recording (back office). In addition, there should be
staff covering analytical functions (middle office).
• At early stages, where borrowing activities are limited, there is no urgent need to
establish a DMO. The borrowing activities (issuance, recording, and reporting) can
be handled by the Treasury unit. As borrowing activities develop, a specialized
DMO should be established and clear functional separation between entities
responsible for transactions and for debt recording should be introduced to reduce
operational risks.
Debt recording, accounting, and debt • The DMO keeps complete records within a three-month lag for subnational
servicing domestic, external, and guaranteed debt, as well as all debt-related transactions.
Debt records are kept in a safe IT system with controlled access. All loan contracts
and documentation of debt service payments, including those related to onlending
from the central government, are scanned and stored in the debt recording system.
• Payment notifications are prepared by the DMO. All payment notifications are
checked with internal records before payments are made, payment instructions
are subject to a minimum two-person authorization process, and payments are
made by the due date. Forecasts on a rolling three-month basis of debt service
payments are provided to the cash forecasting and cash management entity of the
subnational Treasury.
Debt reporting and audit • Reports on debt size and structure, including basic risk exposure indicators,
and outstanding guarantees, are published at least quarterly on the website of
the subnational entity. No later than four months after the end of the fiscal year,
an annual report containing information on debt and borrowing activities of the
previous fiscal year is presented to the legislature of the subnational entity and
published on the website of the subnational entity.
• An external financial audit of debt management transactions is undertaken
annually. External compliance audits have been conducted in the past two years.
Audit reports are publicly available within six months of completion of the audit.
The relevant decision makers produce a management response to address the
outcomes of the internal and external audits of SNG DeM activities.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 44
>>>
Appendix 4: Fiscal Rules of Subnational
Borrowing Across Countries
Regulation of Local Government Borrowing by Limiting
Country Annual Debt Service Total Annual New
Total Outstanding Debt Other
Borrowing
<77% of recurrent revenues
Albania <20% of average total revenues <72% of operational surplus
for 3 years
<11.5% of net current revenue
Brazil <200% net current revenue < 16% of net current revenue
(without transfers)
<25% of own source Guarantees < 5% of
revenues, equalization own source revenues,
Bulgaria
grants from the previous equalization grants
year (previous year)
<20% of the actual Overall LG’s debt < 3%
Croatia revenues from the of total LG recurrent
previous year revenues
<60% of total planned
Estonia <20% of total budget
budget
Government approval of
Hungary <50% of own revenues
borrowing, guarantees
Budget deficit<3% of State
India <20% of revenue
GDP
≤40% of nonearmarked Varying, depending
revenue, on fiscal capacity, but Aggregate SNGs deficit
<75% of nonearmarked
<15% of the sum of general- maximum of 5.3% of limit at 0.32 percent
Indonesia revenue in the previous
purpose grant (DAU) and planned revenues in of GDP in 2022 and
year
revenue sharing transfer 2022 and renewed renewed annually
(DBH) annually
<20% of annual
<35% of total revenue, <35% of total budget
Lithuania revenue; (<10% for short
excluding specific grant (minus specific grants)
term borrowing)
< 30% of the total current
(Including guarantees)
Macedonia operational budget of the
<100% of total revenues
previous year
Debt service including
Poland <15% forecast annual revenue
guarantees
<30% of average own
Romania
revenues in the past 3 years
<50% of total recurrent
Serbia
revenues
<25% of current revenues in <60% of current revenues
Slovakia
the previous year in the previous year
<8% of revenues of the <20% of total revenues
Slovenia
previous year from previous year
<100% of total budget <10% of realized
Turkey (150% for metropolitan revenues from the
cities) previous year
Source: Dexia 2003, Liu-Pradelli 2012, NALAS 2011, Swianiewicz 2004.
EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT <<< 45
Swiss Confederation
EUROPEAN UNION