Financial Stability Review
June 2003
This Financial Stability Review (FSR) is one the reports Bank Indonesia
provides to public in order to achieve its mission “to achieve and maintain stability of the Indonesian Rupiah
through maintaining monetary stability and promoting financial system stability for safeguarding long-term and
sustainable national development.”
Published by:
Financial System Stability Bureau
Directorate of Banking Research and Regulation
Bank Indonesia
Jl. MH Thamrin No.2, Jakarta 10010
Indonesia
Information and Order:
This FSR document is also made in pdf format and is accessible at Bank Indonesia’s website at http://www.bi.go.id
All inquiries, comments and advice may be addressed to:
Bank Indonesia
Directorate for Banking Research and Regulation
Financial System Stability Bureau
Jl. MH Thamrin No. 2, Jakarta, Indonesia
Tel: (+62-21) 381 7990, 7353
Fax: (+62-21) 2311 672
Email: [email protected]
FSR is issued biannually and has the following objectives:
• To foster public vision on financial system stability issues, both
domestically and internationally;
• To analyze potential risks to financial system stability; and
• To recommend policies to relevant financial authorities for promoting
a stable financial system
fsrFinancial Stability Review
No. 1, June, 2003
ii
FOREWORD, vii
EXECUTIVE SUMMARY, ix
Chapter 1 INTRODUCTION, 1
Chapter 2 THE IMPORTANCE OF MAINTAINING
FINANCIAL SYSTEM STABILITY, 4
LESSONS LEARNT FROM THE 1997 CRISIS, 4
FINANCIAL SYSTEM STABILITY: WHAT AND WHY IS IT
IMPORTANT?, 4
CORE COMPONENTS OF A STABLE FINANCIAL SYSTEM,5
CENTRAL BANK’S ROLE IN FINANCIAL SYSTEM
STABILITY, 5
CENTRAL BANK’S ROLE IN FINANCIAL SYSTEM
STABILITY , 7
CONCLUSIONS , 8
Chapter 3 EXTERNAL FACTORS, 11
INTERNATIONAL ECONOMY, 11
DOMESTIC ECONOMY, 12
Monetary Conditions, 12
Government’s Finance , 13
Government Bonds, 14
Foreign Debts, 15
Market Confidence, 15
Maturity Profile, 16
REAL SECTOR CONDITION, 17
Small and Medium Enterprises , 17
Pulp and Paper Industry, 19
CONTENTS
Chapter 4 PERFORMANCE AND PROSPECT OF
INDONESIA’S BANKING INDUSTRY, 21
THE STRUCTURE OF BANKING INDUSTRY, 21
ASSETS STRUCTURE, 21
CREDIT RISK, 22
Non-Performing Loans (NPLs) , 23
Loan Restructuring , 25
Lending Growth , 25
LIQUIDITY RISK, 27
Liquidity Assets, 28
Exchange Offer, 29
Core Deposit, 29
Interbank Call Money, 29
Liquid Assets to Cash Outflow (COF) , 30
Corporate Funds , 30
Household Savings Pattern, 30
Maturity Profile, 31
MARKET RISK, 31
Capital Charge for Market Risk, 32
CAPITAL, 32
BANKS’ PERFORMANCE , 34
Profile of Banks at Stock Exchanges , 36
Comparative Performance with Other Selected
Countries, 36
iii
Chapter 5 CAPITAL MARKET, 38
CONFIDENCE TO CAPITAL MARKET, 38
Mutual Funds, 39
Impacts on Financial System Stability, 42
Bond Market, 46
Stocks Market, 46
Chapter 6 PAYMENT SYSTEMS IN INDONESIA, 51
RISKS IN PAYMENT SYSTEMS, 48
Clearing System, 49
Realtime Gross Settlement (RTGS), 49
ROLE OF PAYMENT SYSTEMS IN THE STABILITY OF
FINANCIAL SYSTEM, 49
Payment Systems Oversight, 50
Risks in Clearing System, 50
Risks in RTGS, 50
Chapter 7 CONCLUSION, 54
ARTICLES
1. Redesigning Indonesia’s Crisis Management – S.
Batunanggar
2. Market Risk in Indonesia Banks – Wimboh Santoso
& Enrico Hariantoro
3. An Empirical Analysis of Credit Migration In
Indonesian Banking – Dadang Muljawan
4. New Basel capital Accord : Its likely impacts on
the Indonesian banking industry – Indra Gunawan,
Bambang Arianto, G.A. Indira & Imansyah
iv
Ta b l e s
3.1. Stress Test on Goverment budget (APBN)
2003-04
3.2. Foreign Debt Indicators
3.3. Indonesia Corporate yankee Bonds (Dec 2002)
4.1. Selected Items of Banks Balance Sheets
4.2 Details of Loan
4.3. NPL Stress Test
4.4. Distribution of Loans by Sector
4.5. 14 Large Banks’ Liquidity
4.6. Maturity Profile of Assets and Liabilites of 13
large Banks, December 2002
4.7. Large Exchange Rate Stress Test of Large Bank
to CAR
4.8 Interest Rates Stress Test of large of Large
Bank to CAR
F i g u r e s
3.1. Non-oil and Gas exports by Country Destination
3.2. US and JAPAN : GDP-Inflation
3.3 US and JAPAN : Current Account
3.4. US and Japan : Discount interest Rate and DJIA
& NIKKEI Indices
3.5. Direct and Portfolio Investments
3.6. Domestic Economic Indicators
3.7. Jakarta Composite and Property Sector Indices
3.8. Maturity Profile of Government Bonds
3.9. Fixed Rate Government Bond vs SBI
3.10. Indonesia Government Bonds Rating and Yields
3.11. Maturity Profile of Corporate Foreign Debt
3.12. Loans to SME and Non-SME
3.13. Lending growth to SME By Type of Banks
3.14. SME Loans by Type of Business Uses
3.15. GDP by Sectors to Total GDP
3.16. NPL by Sector
4.1. Total Bank and Asset
4.2 Bank Securities and Loans
4.3. Total Loans and NPL
4.4. NPL and Provisions for Loan Losses
4.5. Non Performing Loan
4.6. NPL Stress Test
4.7. Loan Restructuring
4.8 Loan to Deposit Ratio
4.9. Trends of IDR and Foreign Exchange Loans
4.10. New Lending
4.11. Loans by Business Uses
4.12. Property Loan
4.13. Deposit Growth
4.14. Liquid Assets
4.15. Core & Non Core Deposit
4.16. Stock Liquid Ratio
v
4.17. Maturity Profile of Time Deposit
4.18. Stress Test Exchange Rate
4.19. Interest Rates Stress Test
4.20. Capital ratios
4.21. CAR Evolution
4.22. Source of Interest Income
4.23. Net Earnings and ROA
4.24. Paid-up Capital and ROE
4.25. Interest Income
4.26. Asian Banks’ ROA
4.27. Asian Banks’ NPL
4.28. Asian Banks’ CAR
5.1. Market Liquidity and Jakarta Composite Index
5.2. Development of Mutual Funds and Bank Deposit
5.3. Mutual Funds Growth
5.4. Development of Deposit vs Public Funds in
Mutual Funds
5.5 Development of YTM of Some Fixed-Rate Bonds
and SBI rate
5.6. Government Bonds by Portfolio
5.7. Financial Sector Stock Index
6.1 Real Time Gross Settlements, Clearing and Non-
cash Transactions
Boxes
1. Causes and Process of Financial Crisis
2. Bank Indonesia’s Strategy in Maintaining
Financial Stability
3. Pulp and Paper Industry
4. Market Risks
5. Yield Curve of Government Bonds
6. Mutual Funds
7. Risks in Payment Systems
8. Failure to Settle Scheme
vi
vii
The financial crises that took place in almost all corners of the world, Indonesia included, have
driven growing awareness on the importance of financial system stability. Instability in a financial system
brings in adverse implications such as lower economic growth, loss of domestic productivity and gigantic
fiscal cost. Based on these adverse experiences, it is imperative that financial system stability is maintained
for the interests of the public.
Financial stability is basically avoidance of financial crisis. Maintaining financial system stability
is one of the primary functions of Bank Indonesia, which is not less important compared to maintaining
monetary stability. Financial system stability is a prerequisite for monetary stability. This issue is in line
with Bank Indonesia’s mission “to attain and maintain stability of Rupiah by maintaining monetary stability
and promoting financial system stability to secure sustainable long-term national development.” However,
maintaining financial system stability is not the sole responsibility of a central bank. Rather it is also
mutual responsibility of relevant government authorities including Ministry of Finance, Financial
Supervisory Authorities, Deposit Insurance Corporation beside the central bank.
In accordance with the above, Bank Indonesia assesses and monitors trends and issues surrounding
stability of Indonesia’s financial system and provides recommendations to maintain stability of the financial
system. Results of such assessments and monitoring is laid down in a regularly updated “Financial Stability
Review” (the FSR). Unlike such other reports issued by Bank Indonesia, the FSR focuses on such potential
risks which may weaken stability of national financial system, and is more forward-looking orientation.
Every section of this report also describes the prospects of national financial system.
During the course of 2002, Indonesia’s financial system is relatively stable and is expected to remain
so in the years to come. However, alert needs to be maintained particularly on some pertinent issues
including delays in the recovery of loan quality and performance of the banking sector, as well as external
issues such as low growth in the global economy and the government budget deficit due to the huge
obligations from domestic as well as overseas borrowings.
This FSR is addressed to all stakeholders, Bank Indonesia and relevant financial authorities in
particular, and the public in general. The review and recommendations offered in this FSR are hopefully
useful to the Government as well as all other relevant authorities in the efforts of maintaining stability of
national financial system. In addition, this review will encourage concerns of all stakeholders to the adverse
movements in the financial system within their jurisdictions so that proactive measures can be taken.
F O R E W O R D
viii
The Board of Governor must be grateful and give its appreciation to the DPNP, all relevant units and
personnel for their dedications, contributions and collaboration for the completion of this first edition of
Financial Stability Review. Finally, we will appreciate all advice, commentaries as well as critics from any
and all parties for further improvements of this review in the future.
Jakarta, April 2003
Maman H. Somantri
Deputy Governor
ix
Indonesian financial system during the course of 2002 is stabilized. This is made possible by the
effective policies in stabilizing exchange rates and controlling inflation as well as the progress made
through the micro-prudential policies covering restructuring program of the banking sector as well as
improvement in banking supervisory and regulatory frameworks. However, certain aspects, the endogenous
and exogenous risks, need to be closely observed as they can potentially disturb financial system stability.
The weakening economy of the major trade partners of Indonesia is one of the driving factors
contributing to the slower growth of exports. As the results, exporting companies, particularly those
whose activities are financed by banks confront augmenting financial risks reducing their capacity to
pay their obligations in timely manner. Such condition is the major driving factor leading to decreasing
a quality of earning assets of banks.
Meanwhile, huge domestic fiscal obligations and international debts have prevented higher economic
and real sector growth. The yet to complete corporate debt restructuring also impedes domestic
corporations to expand their businesses and has brought in adverse impacts to the balance of payment
which may potentially prompt debt crisis and eventually jeopardizing stability of financial system.
Indonesia’s banking structure has not yet changed as the results of the banking crisis back in 1997
that led to the recapitalization of hard-hit banks, all of which have significant impacts to the economy.
Indonesia’s banking system is very much concentrated on the 13 large banks with combined assets of
74.9% from the total assets in banking system.
In general, the condition of the banking industry has been improving following the recap program
introduced since 1999. Aggregate ROE stays at 14.8% and CAR at 21.7%. However, the capitalization
capacity of the banks, particularly the recap banks, remains weak as the results of the low loan growth.
Main revenues of the 13 large banks are from bond coupons since their assets are mostly in the form of
recap bonds. Moreover, increased capital at some banks has not been able to absorb the potential
losses, particularly those arising from credit, market and operational risks. With the introduction of the
market risk capital charge, a number of banks will notice a slight capital decrease, although it will
remain above the minimum Capital Adequacy Ratio (8%).
In the course of 2002, the risks surrounding the banking system remain high and with stable trend.
Bank credit risks are high but decreasing. Meanwhile, market risks and banks’ liquidity risks are moderate
with stable trend. The high credit risk is characterized by high percentage of non-performing loans,
EXECUTIVE SUMMARY
x
which is at 8.1% (gross) or 2.1% (net). The decrease in non-performing loans, including those created
during the outbreak of the crisis in 1997, is mainly attributable to the assignment of such non-performing
loans to IBRA, while others are restructured and written off. The primary constraints in lending are: (i)
loan restructuring has been delayed due to non-conducive economic environment; (ii) low absorption by
real sector, particularly corporations, since most of them are still being restructured; (iii) low growth in
new lending, dominated only by small and consumers loans. Monitoring shall be focus on the increasing
possibility that restructured loans as well as non-restructured loans, sold by IBRA to banks, will new
non-performing loans. Stress test indicates that when NPL stays as high as 23.8%, the conditions that
will lead to solvability constraints in some large banks.
The market risks encountered by banking system during 2002 is relatively moderate with stable
trends. This is the result of IDR appreciation against the United States Dollar and the decreasing trend
of interest rates. In general the net open position of 14 large banks is in short position (up to 3 months)
such that the USD depreciation and the lower interest rates have brought positive impacts to their
capital. The short positions reflect bank expectations of declining trends in the interest rate. Banks may
conduct repricing strategy due to the macroeconomic changes. However, future exposure of market
risks, resulting from pressures against the Indonesian Rupiah due to market volatility and political
instability, must be watched.
Indonesian banks have adequate liquidity. This condition is reflected in the sufficiency of liquidity
of the 13 large banks and their independence from interbank call money. However, the funding structure
of some large banks, particularly state-owned banks is mostly in the form of corporate deposits (owned
by state-owned and large companies). In order to address such liquidity risk, the 13 large banks need to
balance their funding structure in terms of concentration type as well as maturities.
Meanwhile, there has been improved efficiency in national payment system, particularly
attributable to the successful implementation of Real Time Gross Settlement (RTGS). Under RTGS, risks
associated to settlement, liquidity and operations have been mitigated and are monitored in compliance
with international standards, i.e. Core Principles For Systematically Important Payment System. In
order to further improve efficiency and security of national payment system, future efforts shall be
focused on two primary strategies, namely: (i) minimizing moral hazards and (ii) optimizing policies
between security and efficiency considerations. Therefore, it is necessary to review the roles of Bank
Indonesia in the operations of the payment system and as lender of last resort. In addition, there are
needs for failure-to-settle mechanism in order to reduce systemic risks.
In order to help promote a stable financial system, Bank Indonesia has improve the effectiveness of
xi
its roles, especially in monitoring and evaluating potential risks that may adversely affect financial
stability. Bank Indonesia also has drafted a blue print on Indonesia’s financial system stability including
policies and framework for Crisis Resolution, which is a prerequisite for the future financial stability.
Now that more defined and comprehensive policies are in place and with the effective coordination
between Bank Indonesia, Government and all stakeholders, a sound and more stable financial system
will be maintained and in order to encourage faster economic growth in Indonesia.
1
Introduction
INTRODUCTION1CHAPTER
The financial crisis that swept over Southeast Asia,
Indonesia included, in 1997 has taught us a very
valuable lesson concerning the importance of
maintaining stability of financial system. During the
past few years, financial system stability has always
been the primary agenda at national and international
levels. The year 1999 saw the establishment of an
international institute and an international forum,
namely the Financial Stability Institute1 and Financial
Stability Forum (FSF)2, intended to assist central banks
and other supervisory authority in strengthening their
financial system. Similar concerns have also been
indicated by IMF and World Bank, who then introduced
a Financial System Assessment Program (FSAP) in order
to strengthen the financial system of the country being
assessed.3
Meanwhile, there has been increasing number
of publications in the forms of books, articles and papers
as well as seminars and conventions discussing financial
crisis and financial system stability. In addition, there
is growing number of central banks creating a unit or
even groups dedicated to addressing financial system
stability issues and financial stability reviews.
Central banks need to maintain financial system
stability based on three primary reasons. Firstly,
financial institutions particularly banks have important
roles -as financial intermediaries and as a transmission
means of monetary policies- in the economy. These
institutions are significantly exposed to high risks
inherent in their operations. Therefore, financial
institutions constitute one of the instability factors most
harmful to the financial system. Secondly, all financial
crises have brought in catastrophic implications to the
economy, lowering economic growth and income. These
eventually create negative impacts to social and
political life if prompt measures fail to address the
crisis rapidly and effectively. Thirdly, financial
instability brings in very expensive fiscal cost in the
course of mitigating the crisis.
In this extent, Bank Indonesia has designated
financial system stability as a complimentary objective
to achieve price stability. Considering the importance
of financial system stability in the course of achieving
the primary objectives, Bank Indonesia is to give more
priority and attention to addressing this issue. In order
to achieve financial system stability, Bank Indonesia
has adopted four major strategies: (i) fostering effective
coordination and cooperation with others; (ii) improving
research and surveillance; (iii) strengthening regulations
and market discipline; and iv) establishing crisis
resolutions and financial safety net. These will be
1. FSI is established by Basel Committee on banking Supervision (BCBS) to
assist supervisory authorities in strengthening their financial system.
For further details visit http://www.bis.org/fsi/index.htm.
2. FSF is meant to improve stability of international financial system
through exchange of information and international cooperation in the
area of research and surveillance. FSF is composed of such members
from relevant authorities (finance ministries, central banks, financial
supervisory authorities) from 11 countries, as well as international
organizations (such as IMF, World Bank, BIS, OECD), international
committees and associations (Basel Committee on Banking Supervision
/ BCBS), International Accounting Standard Board (IASB), International
Association of Insurance Supervisors (IAIS), International Organization
of Securities Commissions (IOSCO), Committee on Payment and
Settlement System (CPSS), Committee on Global Financial System (CGFS)
and European Central Bank. For further details please visit http://
www.fsforum.org/home/home.html.
3. FSAP is a concerted effort of IMF and World Bank which is introduced in
May 1999. This program is intended to increase effectiveness in the
efforts of improving soundness of financial system in member countries.
For further details visit http://www.imf.org/external/np/fsap/
fsap.asp.
2
Chapter 2
described in details in Chapter 2.
Furthermore, the function of maintaining
financial system stability is conducted by Bank Indonesia
through two major activities. First, by assessing and
monitoring any and all aspects affecting financial system
stability. The activities under this category are
attributable to crisis prevention. Second, by coordinating
and cooperating with relevant supervisory authorities,
particularly when dealing with crisis resolution.
Assessment of the financial system stability is
conducted by incorporating an early warning system to
monitor and analyze trends in the macro-prudential
and micro-prudential indicators 4. The economic macro-
prudential indicators include figures associated with
economic growth, balance of payment, inflation,
interest rate and exchange rate ; the contagion effects,
and all other relevant factors. The micro-prudential
indicators include financial indicators such as Capital
Adequacy, Asset Quality, Management, Earnings,
Liquidity and Sensitivity to Market Risk (CAMELS). The
assessment basically contains identification and
evaluation of risks that may adversely affect financial
system stability and recommendations made to the
government and relevant authorities to carry out
actions necessary to address the matter. The analysis
and recommendations are documented and publicized
on regular basis by Bank Indonesia in a “Financial
Stability Review” (FSR).
The FSR has three basic characteristics: (i)
assessment on conditions and current developments in
the financial system; (ii) reviews are based on risks
which may adversely affect financial system stability;
(iii) a more forward-looking approach by presenting
assessments on the prospects of the financial system
for the year to come. With regard to these
characteristics, the format and focus of analysis of this
FSR may change from one edition to the next in line
with the prevailing conditions, issues, and trends
affecting the economic and financial system.
In general, this first edition of the FSR contains
three primary subjects as described below. Firstly, the
concept and practice at maintaining financial system
stability as presented in a short article entitled “The
Importance Of Maintaining Financial System Stability”
in Chapter 2. This concise article discusses the definition
and the importance of achieving and maintaining
financial system stability, prerequisites for stable
financial system, and the role of Bank Indonesia in
promoting financial system stability.
Secondly, external and internal factors that
adversely affect Indonesian financial system stability
are presented in Chapters 3 and 4. Chapter 3 contains
analysis on developments in the international and
national economies that may affect stability of national
financial system. This chapter also discusses in more
detail domestic financial issues covering foreign debts
and fiscal sustainability. Chapter 4 discusses in detail
the conditions, constraints and risks confronting the
banking system in Indonesia. This issue is very important
considering that the banking sector is the dominant
player – with 75% market segment – in national financial
system. This chapter also discusses structural issues
confronting the banking system including the remaining
high credit risks due to the slow pace of loan
restructuring programs, the low lending growth,
liquidity and market risks, and the performance of the
banking industry. Chapter 5 discusses in detail capital
market issues. Chapter 6 identifies recent developments4 Based on such indicators developed by IMF (Evans et al., 2002)
3
Introduction
and risks in payment system with focus on Real Time
Gross Settlement [RTGS] and clearing system. Chapter
7 provides the conclusion.
Thirdly, it contains four articles. The first article
is entitled “Redesigning Indonesia’s Crisis Management:
Lender of Last Resort and Deposit Insurance” (S.
Batunanggar). This article argues fundamental issues
on crisis management: (i) absence of comprehensive
and clearly defined crisis management policies; (ii) the
weakness of the blanket guarantee creating moral
hazards and adding potential to future financial crises;
and (iii) the obscure function of Bank Indonesia as
Lender of Last Resort in the events of systemic crisis.
To redefine Indonesia’s crisis management, two primary
steps are proposed: (i) to gradually replace the blanket
program to limited explicit deposit insurance; and (ii)
to put in place a more transparent policy regarding
lender of last resort for both normal conditions as well
as during systemic crisis. A more transparent LLR policy
will not only function as a more effective instrument
in addressing crisis management but will also put in
place more defined accountability thereby increasing
credibility of central bank, reducing political
interventions and moral hazards, and encouraging
market discipline in order to eventually encouraging
financial system stability.
The second article, “Market Risks In Indonesian
Banks” (by Wimboh Santoso and Enrico Hariantoro)
compares the results of CAR calculation to market risk
between the standard model BIS and the alternative
models, which uses the Exponential Weighted Moving
Average (EMWA) both have been widely used by banking
practitioners. This review is intended to measure as to
how far market risk will adversely affect Indonesia’s
banks in terms of their capital condition. A significant
decline of capital would adversely affect the stability
of Indonesia’s financial system. This review will give
some pictures of how far banks would benefit from
lower capital charge if internal model is applied. This
review proves that the incentive obtained by banks will
be very much dependent on the volatility of the risk
factors. The higher the volatility, the higher capital
charge is. Based on data on volatility of exchange rate
and interest rate, this review concludes that
incorporation of market risk will not reduce a bank’s
CAR to a level below the minimum threshold and
therefore will not create distortions which would
otherwise impair financial system stability. In addition,
application of internal model will generate lower capital
charge considering that volatility of Indonesia’s
exchange rate and interest rate are relatively lower.
The third article, “Empirical Analysis on Loan
Migrations in Indonesia’s Banking Sector” (by Dadang
Muljawan), looks into the relations between industries’
performance and the dynamic lending at certain banks.
From the statistics, two interesting phenomena were
found. Firstly, industrial performance significantly
affects credit migration process. Secondly, there is an
irreversible process in credit migration. This analysis
will provide more analytical information for the
supervisory authority in evaluating banking risks and
efficacy of external oversight.
The last article “New Basel Capital Accord: What
And How It Affects Indonesia’s Banking Sector” (by Indra
Gunawan, Bambang Arianto, Indira & Imansyah),
explores the New Basel Accord and its implications on
Indonesian banking sector.
Chapter 2
4
THE IMPORTANCE OF MAINTAININGFINANCIAL SYSTEM STABILITY2
LESSONS LEARNT FROM THE 1997 CRISIS
There are two most important lessons learned from
the 1997 crisis. Firstly, the crisis was very
complicated to resolve. And secondly, it was very costly.
The fiscal costs borne by the government for
restructuring problem banks is huge, at 51% of annual
GDP. Indonesia’s crisis is the second worst, after
Argentina crisis (1980-1982), which is 55% of annual
GDP. The crisis not only devastated the national
economy but also affected social and political stability
in Indonesia. However, the crisis has also fostered a
realization of the importance of maintaining a sound
financial institutions and a stable financial market.
Basically, the crisis was caused by two factors.
Firstly, the weak fundamentals of Indonesia’s economy
coupled with inconsistent policies (internal factors).
Secondly, the contagion effects of the financial crisis
started in Thailand on July 1997 (external factors). In
general, the financial system fragility was initiated by
huge un-hedged foreign debts by corporations,
imprudent lending activities, violation of the legal
lending limit to affiliated parties, poor risk management
and governance, and weak bank supervision.
FINANCIAL SYSTEM STABILITY: WHAT AND WHY
IS IT IMPORTANT?
Basically, the term financial system stability or
financial stability pertains to the avoidance of financial
crisis (MacFarlane [1999] and Sinclair [2001]). To be
more specific, financial system stability means the
stability of financial institutions and financial markets
in the financial system (Crockett, 1997). Mishkin (1991)
defines financial crisis as disruption to financial markets
where adverse selection and moral hazards worsen so
that financial market is unable to channel funds
efficiently to parties having the best potential
productivity to invest1 . From these definitions, it can
be concluded that a stable financial system will create
stable financial institutions and financial markets
capable of avoiding a financial crisis that may adversely
affect national economic infrastructure.
There are three main reasons as to why this
financial system stability [FSS] is important. Firstly, a
stable financial system will create trusting and enabling
environment favorable to depositors and investors in
investing their money in financial institutions as well
as to secure interests of small depositors. Secondly, a
stable financial system will encourage efficient financial
intermediation which will eventually promote
investment and economic growth. Thirdly, a stable
1 Adverse selection takes place prior to the choosing of a transaction
when a bank would select a potential borrower with greater chances
that the loan is going to become non-performing. Since adverse selection
factor has great potential of becoming non-performing loans, lenders
would not lend to potential borrowers which have low risks. Moral hazard
occurs after the transaction, where lender will be potentially injured
by borrowers which tend not to pay their obligations. Moral hazard
occurs as the result of asymmetrical information in which lenders do
not know much the activities of the borrowers which will allow borrowers
to give rise to moral hazard. Conflicts of interest between borrower
and the lender due to the moral hazard (agency problem) indicate that
most lenders decide not to lend, so that lending and investment
activities fail to be optimized, thus resulting in credit crunch.
CHAPTER
The Importance of Maintaining Financial System Stability
5
financial system will encourage an effective operation
of markets and improve distribution of resources in the
economy.
On the contrary, an unstable financial system
will bring in harmful implications, such as higher fiscal
cost to resolve troubled financial institutions and
decreasing of gross domestic product due to currency
and banking crisis.
A series of developments which took place in the
past few years have placed maintenance of financial
system stability as a top agenda of the central bank,
supervisory authorities as well as the government,
namely: (i) significant growth in financial transactions;
(ii) growing number of non-bank financial institutions
including the products and services they offer; (iii)
increased complexity and risks in banking activities; and
(iv) huge fiscal cost required to remedy the banking crisis.
In addition, there are other constraints such as
changes of policies, financial instruments and others
faced by banking sector as well as real sector, all of
which will make the duty of maintaining financial
system stability to be complicated.
CORE COMPONENTS OF A STABLE FINANCIAL
SYSTEM
The stability of financial system depends on five
components which are associated one with another,
namely: (i) a stable macroeconomic environment; (ii)
well governed financial institutions; (iii) efficient
financial market; (iv) sound prudential oversights; and
(v) safe and reliable payment system (MacFarlane,
1999).
Crisis may be prompted by various risks originating
from the elements in the financial system. The process
leading to a financial crisis is described in Box 1.
Financial system stability can be maintained by
improving resilience of financial institutions and money
market against external volatility. A number of
measures may be taken, such as by applying prudential
standards and good corporate governance within
financial institutions and capital markets, conducive
monetary and fiscal policies, and real sector capable
of promoting economic growth.
Considering that internal weakness within
financial institutions and fragility in capital market,
crisis management policy needs to be put in place.
Therefore, a safety net mechanism and contingency
plan are required to address crisis. For this purpose,
central banks play a very important role in maintaining
stability of financial system, as well as in taking
preventive and corrective actions against crisis. This is
due to the fact that powers to regulate and supervise
as well as to enforce policies of financial institutions
are held by central bank.
CENTRAL BANK’S ROLE IN FINANCIAL SYSTEM
STABILITY
Safeguarding financial stability is a core function
of the modern central bank, no less important than
maintaining monetary stability (Sinclair, 2001). Both
are closely correlated and affected one another.
Effectiveness of financial policies will only manifest
itself in an environment in which there is sound financial
system because financial institutions serve as medium
for monetary policy transmission.
There are two major approaches generally
adopted by central bank in maintaining financial system
stability. Firstly, reliance on market forces and market
Chapter 2
6
Real Sector
Monetary Fiscal
InternationalEconomy
FinancialInstitutions,Markets and
FinancialInfrastructure
Figure:
INTERACTIONS WITHIN A FINANCIAL SYSTEM
Financial crisis may originate from problems
existing within any of the various correlating
components within the financial system such as
financial institutions, banks, non-bank financial
institutions or capital market (first ring); or may
be caused by one or a combination of problems
within the real sector, fiscal or in the payment
system (second ring). Nevertheless, a crisis may also
be spark by some external factor with its contagion
effect that spillover Asia in 1997 (third ring).
Learning from the Asian and Indonesia crisis in
1997, instability of financial system may be occurred
through three major phases (Mishkin, 2001).
Firstly, impaired public confidence in the
financial system. This may be caused by various
problems in the economy or in financial system
Box 1.
CAUSES AND PROCESS OF FINANCIAL CRISIS
such as worsening financial condition of banks,
increased interest rate, decreased share prices and
increased uncertainty.
Then in the second phase, impaired confidence
of customers and investors toward the economy and
the IDR result in the depreciation of the IDR which
then prompts currency crisis.
And finally, such currency crisis would entail
crises of the banking sector prompted by depositors
drawing up their deposits (systemic bank run) which
results in liquidity problem to banks. In addition,
banks would sustain losses from non-performing
loans particularly those of corporations with un-
hedged overseas borrowings. The cost of overseas
loans borne by corporations will skyrocket due to
the depreciation of the IDR against the USD.
The twin crisis (currency and banking crisis) if
not effectively addressed, will result in even wider
complications, as well as social and political
instability.
Consequently, the Government will have to pay
huge of fiscal cost (in the case of Indonesia, 51% of
its Gross Domestic Product) in order to rescue its
banking system. The huge fiscal cost will eventually
be borne by the public, the taxpayers. In addition,
the prolonged financial crisis will bring in adversely
impacts to national economy, such as lower
economic growth and output aggravated by financial
disintermediation.
The Importance of Maintaining Financial System Stability
7
discipline similar to that adopted by Reserve Bank of
New Zealand. Secondly, reliance on regulations. The
latter approach is adopted by wider supervisory
authorities or central banks in both developed and
developing countries. During the past few years, there
has been growing awareness that both approaches need
to be applied more consistently in order to achieve a
better stability in the financial system.
In practice, the definition of financial system
stability [FSS] varies among central banks. Most central
banks state it explicitly in their statutory regulations.
But some rely on joint arrangements such as those
among Bank of England, Financial Services Authority
and HM Treasury.
In general, the role of central banks in stabilizing
financial system covers three primary activities:
a. Research and surveillance
Trends and risks, both internal and external,
affecting the financial system need to be closely
assessed and monitored. Research and surveillance
activity are intended to produce a policy
recommendation for maintaining financial system
stability.
b. Payment systems oversight
Regulation and oversight are required to ensure a
safe and reliable payment systems. The adverse
risks to the payment system, which may lead to
systemic failure and financial crisis, may be
avoided.
c. Crisis resolution
While the latter two activities are related to crisis
prevention, the third activity is taken by the
central bank to address crisis when it actually
occurs. Usually two instruments are used: (i)
providing lending facility to the financial
institutions by the central bank as the lender of
the last resort (LLR); and (ii) to provide deposit
insurance. LLR facilities by central bank may be
given either during normal situation as well as
during systemic crisis. During normal situation,
such facility is provided only to address liquidity
problem for illiquid but solvent banks, and with
sufficient collateral. During systemic crisis, LLR
facility is provided to restructure the banking
system.
CENTRAL BANK’S ROLE IN FINANCIAL SYSTEM
STABILITY
Currently, there is no formal legal basis stating
about Bank Indonesia’s function in maintaining
financial system stability. The function, in fact, is
performed simultaneously with its core tasks of
performing monetary policy, bank supervision and
payment system.
Following the 1997 crisis, there has been growing
awareness in the importance of maintaining financial
system stability. In line with the introduction of Law
No. 23 of 1999, Bank Indonesia incorporates the
financial system stability aspect in its mission: “to
achieve and maintain stability of the Indonesian rupiah
through maintaining financial stability and promoting
of financial system stability for sustainable national
development.” In line with its mission and vision, Bank
Indonesia has formulated a framework that contains
the goals, strategy and instruments required for
maintaining the financial system stability.
The roles of maintaining monetary stability and
promoting financial system stability are closely related.
Chapter 2
8
Both roles are aiming at the same objectives which is
price stability.
In order to achieve a stable financial system,
Bank Indonesia adopts four strategies, namely: (i)
implementing regulation and standards to foster market
discipline; (ii) intensifying research and surveillance;
(iii) improving coordination and cooperation; and (iv)
establishing safety net and crisis resolution framework
(see Box 2).
CONCLUSIONS
Stability of financial system much depends on the
soundness of financial institutions, particularly banks
that dominate the financial system. This will also rely
on the effectiveness bank supervision. Therefore, it is
imperative to have an independent and competent bank
supervisor capable of assessing bank risks and taking
preventive and corrective actions on the problems faced
by banks effectively.
To achieve a stable financial system, effective
coordination must be in place among relevant
authorities. Therefore, there must be a clear division
of roles and responsibilities of each authority. More
importantly is the commitment of the stakeholders to
cooperate in achieving and maintaining financial system
stability. In addition, effective supervision and
consistent law enforcement will foster market players
and the general public to play their roles responsibly.
The Importance of Maintaining Financial System Stability
9
In order to achieve financial system stability,
Bank Indonesia adopts four strategies:
(1) Implementing regulations and
standards. Consistent implementation of
international prudential regulations and standards
are required as a sound basis for both regulator
and the market players in conducting their
business. In addition, consistent discipline of the
market players need to be fostered.
(2) Intensifying research and surveillance.
Development of financial system the relevant
aspects affecting its stability should be assessed
and monitored. Risks which may endanger
Box 2.
BANK INDONESIA’S STRATEGY IN MAINTAINING
FINANCIAL SYSTEM STABILITY
financial system stability are measured and
monitored by incorporating an early warning system
which is composed of micro-prudential and macro-
prudential indicators. Research and surveillance are
aimed at producing a policy recommendation for
maintaining financial system stability.
(3) Establishing safety net and crisis
resolutions framework. Safety net and crisis
resolutions framework and mechanism are required
for resolving financial crisis, once it occurs. These
include policy and procedures of the lender of the
last resort, and the deposit insurance which will
replace the blanket guarantee. Currently, there is no
An active involvement in creating and maintaining a sound andstable national financial system.
Financial System Stability (FSS) Framework
Achieving and maintaining the stability of Rupiah value by maintaining
monetary stability and promoting financial system
stability for sustainable national development.
Implementing
Regulation &
Standards
Intensifying
Research &
Surveillance
Improving
Coordination &
Surveillance
• Regulation & Standard e.g
Basle principles,
CPSIP, IAS,
ISA, dsb.
• Market Discipline
• Early WarningSystems
• Macro prudentialIndicators
• Micro-Prudential Indicators
(aggregate)
- InternalCoordination
– ExternalCoordination&Cooperation
• Lender of last
resort
- Normal
- Systemic Crisis
• Crisis Resolution - Safety Nets
Establishing Financial
Safety Nets & Crises
Resolution
Instruments
FSS Objective
BI ’s Mission
FSS Strategies
Chapter 2
10
a clear legal framework for crisis resolution.
According to Law No. 23/1999, Bank Indonesia is
only allowed to provide lending to address liquidity
problem faced by banks during normal times, but
not for systemic crisis situation. Therefore, there is
an urgent need to formulate this policy in the law
which clearly stipulates the roles of Bank Indonesia
as the lender of the last resort in the events of crisis.
(4) Improving coordination and cooperation.
Coordination and cooperation with related gencies
is very crucial especially in crisis times. Usually, the
coordination was formed in a national committee
which is composed of the Bank Indonesia Governor,
Finance Minister and related agencies including the
Head of Deposit Insurance Agencies to be
established.
11
External Factors
EXTERNAL FACTORS3
INTERNATIONAL ECONOMY
A long with globalization in economics,
Indonesia’s financial system will be affected
by instability in regional and global economies. It occurs
through international trade and money market
channels.
During the last few years, global economy tends
toward a downturn condition. This is provoked by
decreasing economic performances of the industrial
countries in the world, namely the United States and
Japan. Ultimately, this situation will influence
Indonesia’s financial system considering that the United
States and Japan are the largest markets for Indonesia’s
exports. Indonesia’s trade account states with the
United States and Japan reach 17.44% and 22.99%
respectively of total exports. In addition, both countries
are also primary lenders. The slowdown condition of
those industrial countries is expected to continue
following terrorists’ attacks at some places within the
United States in 2001.
The declining economic conditions of these two
major economies was indicated by decreasing Gross
Domestic Products (GDP), increasing in inflation, and
the current account deficit.
FIGURE 3.1:NON-OIL AND GAS EXPORTS
By COUNTRY OF DESTINATION
FIGURE 3.2:US and JAPAN : GDP-INFLATION
FIGURE 3.3:US andJAPAN : CURRENT ACCOUNT
-
10
20
30
40
50
60
70
Percent
U S ASEAN Japan
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Inflation (Percent)GDP (Percent)
1995 1996 1997 1998 1999 2000 2001 2002
(2)
(1)
-
1
2
3
48
(6)
-
(4)
(2)
2
4
6
GDP-US GDP-Japan Inflation-US Inflation-Japan
Miliar USD
(140)
(120)
(100)
(80)
(60)
(40)
(20)
-
20
40
60
1995 1996 1997 1998 1999 2000 2001 2002
U S Japan
CHAPTER
12
Chapter 3
The continuing recession in United States and
Japan also affects their capital markets adversely. This
was illustrated by the fall in composite indices of the
Dow Jones and Nikkei. In fact, such conditions should
have encouraged capital inflows to Indonesia.
Unfortunately, it is not the case, since Indonesia’s
investment environment is not yet conducive, as
evidence by a decision of a restructured corporation in
Japan to close their factories in Indonesia. Such policies
truly bring negative impacts to the money market due
to the decreasing of bank’s lending portfolio in respect
to Japanese corporations.
DOMESTIC ECONOMY
Monetary Conditions
During 2002, monetary condition is quite
conducive as reflected by lower interest rate and
stability of exchange rates. Hopefully, such condition
will prevail so as to stimulate economic growth in 2003.
Unlike the condition in 2000 and 2001, the SBI
interest rate tends to decrease in 2002. This condition
indicates that Bank Indonesia has started to ease its
monetary policy as inflation rate is still in control, while
the rupiah exchange rate remains relatively stable.
However, the lower trend of SBI interest rate is not
immediately followed by a reduction in lending rates.
The declining trend of SBI interest rate will
hopefully encourage more lending to real sectors. In
spite of such increase in lending, the amount is
relatively small and is mostly given to small and medium
enterprises. This reflects banks’ caution in lending and
At the end of 2002, United States and Japan’s
GDP slightly increased by 2.1% and 0.5% respectively.
This was mainly due to the lower discount rate policies
introduced by monetary authorities of both countries.
Compared to 1999-2000, their GDP in 2002 has not fully
recovered and monetary authorities continue their low
interest policies.
The fact that both economies were not improved
in 2002 caused Indonesia’s exports to decrease. This
adversely affect borrower’s financial performance
which will eventually cause a negative impact on banks’
assets quality.
Figure 3.4:US and Japan : Discount interest Rate and
DJIA & NIKKEI Indices
FIGURE 3.5:DIRECT AND PORTFOLIO INVESTMENTS
Percent
10,000
-
5,000
15,000
20,000
25,000
1999 2000 2001 2002
U S Japan DJIA NIKKEI
-
1
2
3
4
5
6
7
2003
Million USD Million USD
0
500
1000
1500
2000
2500
3000
1998 1999 2000 2001 2002
Direct Investment Portfolio Investment
-5000
-4000
-3000
-2000
-1000
0
1000
2000
3000
4000
13
External Factors
the low level of absorption by corporations due to
ongoing restructurings.
The lower interest rate in fact is not followed
with migration of third party capital to capital market
or property sector. However, there are indications that
banks’ third party funds have migrated to mutual funds.
relatively secure.6 However, we can expect to see
further pressures in fiscal during 2003 and 2004,
particularly in connection with budget deficits. Debt
to GDP ratio decreased from 88.4% as of June 2002
to 70.4% as of December 2002. However, Indonesia’s
debt ratio was much lower than that of other
countries such as Argentina (49.4%), Mexico (69.1%)
and Turkey (54.2%) before these countries
descended to financial crisis.
If the debt is not carefully managed, debt
crisis will adversely affect balance of payment and
financial performance of the Government.
Eventually the condition will also adversely affect
financial system stability. One potential issue for
the government is refinancing of government bonds
(refinancing risks), considering the huge amount of
the bonds to mature within a few years (IDR 36.3
trillion in 2004 and IDR 45.8 trillion in 2007).
Maturity dates of government bonds prior to and
after re-profiling is shown in Figure 3.8.
FIGURE 3.6:DOMESTIC ECONOMIC INDICATORS
FIGURE 3.7:JAKARTA COMPOSITE and
PROPERTY SECTOR INDICES
6 Policy Analysis and Planning Division (2002), “Indonesia’s Medium-Term
Fiscal Sustainability.”
Rp/USD Percent
-
2,000
6,000
8,000
10,000
12,000
14,000
4,000
0
10
20
30
40
50
60
70
80
1998 1999 2000 2001 2002
Exchange Rate SBI (%) Interbank (%) Credit /GDP Ratio (%)
2003
0
100
200
300
400
500
600
700
800
0
20
40
60
80
100
120
140
160
180
200
1996 1997 1998 1999 2000 2001 2002
J C S I P S I
2003
Jakarta Composite Stock Index Property Stock Index
Government’s Finance
Bank Indonesia’s review on medium term
fiscal resilience indicates that fiscal condition is
FIGURE 3.8:MATURITY PROFILE OF GOVERNMENT BONDS
Before Reprofiling
After Reprofiling
Trillion Rp
2002 2003 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 20202004
0
10
20
30
40
50
60
70
80
90
14
Chapter 3
Government Bonds
From the Government Budget [APBN] simple stress
test, re-profiling of Government bonds has not fully
taken pressure off the government financial condition.
There are potentials for budget deficit which will
eventually adversely affect the government’s ability in
paying principal and interests of government bonds.
In order to address the obligation to pay principal
and interest of maturing government bonds, issued in
connection with banks re-capitalization program, the
Government restructure of maturities and interest rates
of the government bonds. As for an initial step, the
government re-profile the government bond in 4 State-
Owned Banks portfolio involving a sum of IDR 22.8
trillion.
By considering the process and other fiscal
assumptions, the stress test shows a negative difference
between new debt and maturing debts at 1.37% of GDP
or amounting to around IDR 29 trillion in 2004. The
condition needs to be resolved with another re-profiling
and other strategy such as conducting buy back,
boosting additional income from selling assets etc. On
the other hand, re-capitalization banks should work in
efficient manner and also improve their capital by this
means reducing dependence on government financial
support.
A developed and efficient government bond
market will encourage liquidity. The liquidity is needed
to further improve market confidence and capability
reduce risks if there is negative shock to the market.
Otherwise, market participants will rely on Bank
Indonesia’s liquidity support when crisis occurs. The
role of Bank Indonesia should be limited only in crisis
condition which have systemic impact to the financial
sectors and economy. Sound and liquid government bond
market will help government in reducing refinancing
risks and arranging bonds maturity profile.
Maintaining the government’s ability to pay recap
bonds’ principal and interest at maturing date is critical.
Bonds sold at high discount rate may reflect an
overcrowded of bonds in similar maturity, investors’
FIGURE 3.9:Fixed Rate GOVERNMENT BONDS vs SBI
0
20
40
60
80
100
120
Average Fixed-Rate SBI 1 month
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2 0 0 2
Government Bond Price S B I (%)
0
2
4
6
8
10
12
14
16
18
Table 3.1Stress Test on Goverment budget (APBN) 2003-04
State Revenues 17.76 17.33 15.70State Expenditures 20.11 19.10 15.10Primary Balance 3.04 2.45 4.00Surplus (+) / Deficit (-) -2.35 -1.77 0.60
A. Financing of Government Debentures1. Maturing Government Debentures -1.18 -2.732. Reprofiling of Government Debentures at 4 State-Owned Banks 0.00 1.06Sub Total -1.67
B. Overseas Borrowingsa. Program Loans 0.54 0.53b. Project Loans 1.16 0.97Sub Total 1.70 1.50 1.80
C. Installments of Overseas Loan Principal -0.76 -0.89 -2.10
Financing (A+B+C) 1.77 -1.97
% of GDP
2002 2003 2004
APBN RAPBN RAPBN
15
External Factors
choice and the issuer financial conditions. It is therefore
necessary to maintain sound financial condition to
ensure timely payment of bonds’ principal and interest.
This will increase market confidence and maintain a
more liquid market for government bonds.
Post-crisis financial condition of the Government
is not quite promising. At the moment, the Government
carries huge burden from both domestic and foreign
borrowings. Such situation is worsened by the limited
ability to boost revenues considering the non-conducive
domestic and international economic environments.
Therefore, the future prices of recap bonds will rely
on the Government’s ability to improve its financial
performance as well as performance of the economy
as a whole.
The huge bonds’ principal and interest payment
obligations which will prevail in 2004 through to 2008,
coupled with budget deficits, may give probability of
government debt crisis. The government needs to adopt
more stricter fiscal discipline while striving to increase
revenues. The re-capitalization banks also need to
support government by operating in more sound
governance and obtaining profitable financial condition
to avoid another possibility of government debt and
banking crisis.
Foreign Debts
Foreign debt crisis will adversely affect stability
of financial system. Increasing commercial borrowings
from overseas lenders under binding contracts without
strong repayment capacity, and with uncertainties in
social, political, economic and finance situations, may
impairs international confidence toward Indonesia’s
economy. This situation will damage Indonesia’s rating.
As the implications, lenders will demand higher interest
rate as risk premium raising, thus requiring us to
mobilize more and more US$ to repay the floating
interest obligations as well as for securing new loan
commitments. Consequently, there will be high
demands for US$ funds and US$-denominated deposits
at local banks will be rushed. Such situation will surely
adversely affect financial system stability, similar to
that which swept throughout Asia and in Argentina.
Market Confidence
The confidence level of investors and rating
companies on Indonesia’s financial solvability remains
low, as shown by the rating made by Standards & Poor.
Foreign investors’ perception on Indonesia’s financial
condition is still risky. Yield spread between Indonesian
government’s Yankee bonds and US treasury bonds as
of December 2002 is relatively wide, namely 266.07
base points. Such condition results in relatively higher
risk premium for Indonesia’s government as well as
private foreign borrowings. In addition, (lower) rating
and (higher) risk premium may result in reduced
demands for Indonesian Rupiah, thus adversely affect
Rupiah exchange rate which will eventually increase
market risk.
Debt Service Ratio and total debt ratio against
GDP as of December 2002 are relatively high,
respectively at 30.8% and 70.4%. Despite their
decreasing trend, such rates are still above the normal
levels, namely 20% and 50-80%. Such condition will
indirectly adversely affect financial system stability in
the event of substantial depreciation of the IDR.
Therefore, there shall be concrete efforts to boost
exports by among others securing financing facility from
16
Chapter 3
FIGURE 3.10:Indonesia Government Bonds Rating and Yields
banks, advancing technology in production and focusing
on productive investments particularly on export-
oriented activities. Approval given to the proposed
rescheduling of Indonesia’s debts in the amount of US$
5.4 million during Paris Club III on 12th April 2002 is one
such effort to address the potential risk of debt crisis
in Indonesia.
debts. Although, the private debts ratio to export
account for 30.8% which exceeding the benchmark
level of 20%. The amount of exposure has been
decreasing since quarter 4 2002. Moreover, most of
the debts have been restructured and anticipated. The
projected debt repayment in 1st quarter of 2003 is to
be at US$ 3.4 billion. This will expectedly increase
demand for United States Dollar. However, debt
repayment realization is relatively small due to the
fact that most borrowings have been estimated and
the withdrawal will be made only to meet working
capital needs of the corporations.
Source : Bloomberg
S&P rating convertion : 1=SD, 2=C-, 3=C, 4=C+ etc. 15=BB, 20=A- (under BB is speculative)
S&P Yield
SD
Yield
A-
CCC+
BB
0
2
4
6
8
10
12
14
16
18
20
Jul Apr Oct Dec Jan Jan Mar May Mar Mar Sep Apr Oct May Nov Apr Sep Dec
92 95 97 97 98 98 98 98 99 99 99 00 00 01 01 02 02 02
0
2
4
6
8
10
12
14
16
Debt Service RatioGovernment 15% 11% 10% 10% 7% 11% 11%Private 21% 33% 48% 47% 34% 31% 20%Indonesia 36% 45% 58% 57% 41% 41% 31% 20%
Total Debt to GDP ratio 49% 62% 146% 105% 94% 91% 70% 50%-80%
Ratio 1996 1997 1998 1999 2000 2001 2002 Benchmark
Table 3.2:Foreign Debt Indicators
Maturity Profile
Maturity profile of foreign debts is not yet
reasonable however it will not bring in significant
adverse impacts to financial system stability since the
corporate apply more prudential foreign borrowing
activities. Most of private debts (88%) are corporate
As for Indonesia’s bank foreign borrowing, there
are two banks issue bonds denominated in foreign
currency during 2002. Proceed from such bond issue is
primarily used to repay principal and interest of existing
foreign borrowings (exchange offer). Generally,
Indonesia banks adopt the refinancing pattern to repay
their foreign currency borrowing e.g. issue other short-
term bonds. Learning from 1997 crisis, although such
bonds will not bring much problem in short-term period,
FIGURE 3.11:MATURITY PROFILE OF
CORPORATE FOREIGN DEBT
2 0 0 2
Million USD Million USD
2 0 0 3
Bank
NBFI
Corporate
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
0
200
400
600
800
1000
1200
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
17
External Factors
but in the long run they may adversely affect banking
sector and financial systems.
With respect to that, some factors which might
adversely affect such foreign currency—denominated
bonds issued, must be monitored, such as (1)
uncertainty of international economic condition; (2)
the relatively low international confidence level on
Indonesia’s economy as shown by the low rating; and
(3) the relatively low profitability of banking sector. In
addition, banks need to be cautious of their foreign
currency borrowings by obtaining hedging instruments
in order to reduce market risk, considering the fact
that banks’ revenues are mostly in Indonesian Rupiah.
However, there are constraints such as insufficiency
data regarding private foreign debts. Learning from
1997 crisis, the condition will result in ineffectiveness
of monitoring activity such that the risks and instability
factors against financial system stability, particularly
from foreign debts, cannot be adequately and timely
anticipated. Therefore, foreign debts need to be
managed in prudential manner and monitored carefully.
REAL SECTOR CONDITION
Small and Medium Enterprises
Loan restructuring process faces with significant
obstacles as real sector has not recovered yet. This
condition will repress financial system stability.
After recapitalization process, Indonesian banks
have not found difficulties in obtaining funds to finance
their lending. This is reflected in the increased liquidity
in primary reserve (cash, minimum demand deposit and
SBI), secondary reserve (trade bonds, inter bank call
money) and tertiary reserve (investment bonds).
In fact banks are still reluctant to lend due to the
fact that banks are still facing some constraints, among
BNI Cayman Island B- 145 -
BNI KP CCC 150 728
Medco Energy Int’l B+ 100 766
Indofood B 280 791
Bank Mandiri CCC 125 703
Telkomsel B+ 150 615
Source: Bloomberg
Rating O/S (Million US$) Yield Spread
Table 3.3Indonesia Corporate yankee Bonds (Dec 2002)
There is a significant risk in such corporate bonds
issued overseas against financial stability, due to the
volatility of the exchange rate. In addition, most of
debts are not fully hedged. Such condition might trigger
corporate debt crisis. Eventually, corporate crisis – most
of them were financed by banks – will have contagious
effect to the banking sector. This was what happen in
some east Asia countries, including Indonesia, during
the 1997 crisis.
In order to improve effectiveness in foreign debt
monitoring, Bank Indonesia has put in place prudential
policy and mechanism for monitoring foreign debts.
Figure 3.12Loans to SME and Non-SME
Small - Scale Enterprise Loan Non Small - Scale Enterprise Loan
Trillion Rp
-
50
100
150
200
250
300
350
400
450
2 0 0 1
SepDec Mar Jun
2 0 0 2
18
Chapter 3
7 IBRA Report, September 2002.
others, relatively higher non-performing loans, higher
risks in real sector -particularly corporations with high
debt to equity ratio- and limited information regarding
potential borrowers. In addition, banks’ preference in
portfolio investments has changed to less risky
investment such as placements in SBI, Government
Bonds and inter bank money market.
will adversely affect bank’s performance improvement.
Therefore, providing loans to small and medium
enterprises is one of the options to accelerate economic
recovery and to improve banks’ lending portfolio.
In addition, new loans growth was still low because
most of large companies restructuring at IBRA were
incomplete. The process shows that out of the IDR 369.5
trillion of loans transferred to IBRA, only IDR 19.9 trillion
have been restructured, while IDR 17.1 trillion have
been fully settled.7
Significant growth in new loans may be expected
to occur after completion of the restructuring such
corporations. In fact, the restructuring has not gone
very well and time consuming due to various constraints
particularly uncertainty of business and legal process.
Corporate loans dominate banks’ portfolio. Delays
in the recovery of real sector particularly corporations
However, it must be noted that such strategy poses
risks as banks have insufficient experience in providing
loans to small and medium enterprises and time
consuming.
As of the third quarter of 2002, lending to small
and medium enterprises accounted for IDR 24.6 trillion,
which was 41.8% of the total new lending. For the same
period, private national forex banks were the biggest
lenders to SME, followed by regional development banks
and state owned Banks, contributing 12.9%, 10.1% and
6.2% respectively. This tendency needs to be monitored
mainly because SME debtors need technical or
management assistance as well as marketing training
of which not all banks can provide.
SME’s non-performing loan was still low (4.5%).
Consumption loan dominated SME lending, which might
increase demands for goods and services at local as
well as international market. On one side, increased
demand would generate enlarged goods and services
Q II 2002 Q III 2002
State BankForex Private
Bank
RegionalDevelopment Bank
Joint-venture Bank
Foreign Bank
Percent
-5
0
5
10
15
20
Non-Forex
Private Bank
Figure 3.13LENDING GROWTH TO SME BY TYPE OF BANKS
Figure 3.14SME LOANS BY TYPES OF BUSINESS USES
Working Capital Loan
43%
Investment Loan
11%
Consumer Loan
46%
19
External Factors
fact that the figure was still higher than those of other
sectors. At the end of 2002, non agriculture and mining
sector’s performance showed some improvements.
However, the improvement was still accompanied with
high non-performing loans in non agriculture and mining
sectors, particularly from manufacturing. Considering
the importance of non agriculture and mining sectors
role, particularly manufacturing sector in domestic
economy, the following box 3 illustrates performance
of pulp and paper industry.
inflows from international market which, if not properly
managed, may adversely affect balance of payment.
On the other side, increased demand created business
opportunities for companies in order to improve their
financial performance.
Pulp and Paper Industry
Due to the 1997 crisis, agriculture and mining
sector contribution to GDP decreased, in spite of the
FIGURE 3.16NPL by SECTOR
10
2001
2002
Agriculture
Mining
Industry
Electricity
Construction
Trading
Transportation
Business Services
Social Services
Others
-
20
30
40
50
60
Percent
Figure 3.15GDP BY SECTORS TO TOTAL GDP
Percent
-
2
4
6
8
10
12
14
29
30
31
32
33
34
35
36
Mining & Agriculture Non Mining & Agriculture
Percent
1996 1997 1998 1999 2000 2001 2002
20
Chapter 3
Pulp & paper industry was considered as a risky
business. Such that 7 out of 10 pulp & paper companies
are indebted to banks in the amount of IDR 4,136,577
million. In 2000, 97% of the total outstanding loans
given to this industry were restructured. Unfortunately
however, restructuring program is low because of the
following issues:
• Poor restructure analysis conducted by banks, such
failure to assess borrowers’ cash flow capacity,
individually or group.
• Lack of transparency nor cooperation in disclosing
of their financial conditions;
• Low productivity and decreasing revenues due to
lower prices of their products in markets,
accompanied by increasing cost of goods sold;
• Worsening financial condition due to huge
indebtedness, interest and other expenses caused
by the Rupiah depreciation against the United
States Dollar.
Yet, in the future, pulp & paper industry will face
other challenges, specially low paper consumptions in
Asia (other than Japan) compared to United States,
Box 3.
Pulp and Paper Industry
Japan and Europe. On the other hand, price of raw
material is gradually increase since their HPH (forest
exploration rights) cannot supply adequate raw
material. For illustration, in 1999 group’s own sources
fully supplied raw material requirements, but in 2000
they only contributed 40% of companies demand, and
this generate increases in the log prices.
From 1993 up to June 2002, the average gearing
ratio of 5 pulp & paper companies’ financial statements
had increased. Such increase indicates that this
industry depends more on third party financing instead
of self-financing. On the other side, the deteriorating
of pulp & paper market will weaken most pulp & paper
companies’ revenues.
3,000
2,000
1,000
-
(1,000)
(2,000)
(3,000)
Total Debt/Equity Long Term Debt/Equity Total Debt/Total Asset
Dec Dec Dec Dec Dec Dec Dec Dec Dec Jun
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Percent Percent
120
100
80
40
-
(20)
60
20
21
Performance and Prospect of Indonesia’s BankingPERFORMANCE AND PROSPECT OFINDONESIA’S BANKING INDUSTRY4
THE STRUCTURE OF BANKING INDUSTRY
Indonesia’s financial system stability relies heavily on
the banking industry covering of about 90% of total asset
of financial system. Similarly, the banking system is
dominated by 13 large banks, including 10 recap banks,
represent 74.8% of the total assets of banking industry.
(see Table 4.1)
Therefore, ensuring soundness of these large
banks is the key in maintaining stability of banking
system and financial system. The analysis in this report
is focused on the large banks using data as of December
2002.
ASSETS STRUCTURE
Assets of large banks is largely dominated by
marketable securities accounting for 45.1% while
portion of loans is only 29.1% of total assets of the
large banks as of December 2002. The biggest part
(95.7%) of such marketable securities is recap bonds
(see Figure 4.2).
Figure 4.1.TOTAL BANKS & ASSETS
Trillion Rp
-
50
100
150
200
250
300
Number of Bank
Total AssetNumber of Bank
0
200
400
600
800
1,000
1,200
1995 1996 1997 1998 1999 2000 2001 2002
SELECTED ITEMS
Table 4.1.SELECTED ITEMS OF BANKS BALANCE SHEET
AssetsBank Indonesia 153.8 103.5 67.3 134.3 94.0 70.0Inter-bank Placement 124.6 55.8 44.8 149.4 63.9 42.8Marketable Securities 395.4 374.6 94.8 425.7 406.2 95.4Loans 371.1 241.5 65.1 316.0 190.8 60.4Non-performing loans 33.2 19.7 59.3 43.4 22.2 51.1Total Assets 1112.2 830.6 74.7 1099.7 822.4 74.8
LiabilitiesDeposits 835.8 634.2 75.9 797.4 606.9 76.1Inter bank borrowing 81.3 60.4 74.2 93.6 70.7 75.5Provision for Loan Losses (39.1) (26.4) 67.5 (44.8) (26.7) 59.6Paid-Up Capital 96.4 71.7 74.4 88.1 66.8 75.8Donated Capital 188.9 188.8 99.9 188.9 188.9 100.0
2 0 0 2 2 0 0 1
Total Bank Large Bank Share Large Bank Total Bank Large Bank Share Large Bank
(Trillion Rp) (Trillion Rp) to Total Bank (%) (Trillion Rp) (Trillion Rp) to Total Bank (%)
BALANCE SHEET
CHAPTER
22
Chapter 4
Figure 4.3.Total Loan and NPL
L o a nNPL
0
100
200
300
400
500
600
700
800
1996 1997 1998 1999 2000 2001 2002
Trillion Rp
years maturity. The changed maturities of bonds, forces
banks to adjust their portfolio and lending strategies.
If interest rate decrease to below 12% it will
adversely affect the prices of floating-rate bonds.
Thereby, banks holding floating-rate bonds will have
to reduce their deposit rate in order to adjust their
cost and income structure. As impact they may lose
some of their deposit base, which in turn will hurt their
liquidity due to the migration of funds from these large
banks to such other banks offering higher interest rates
or to other type of investments such as mutual funds.
This will further give pressure to those banks to sell
their floating-rate bonds at big discount rate. However,
such problem can be avoided if the market of recap
bonds is more liquid, so that the trading portfolio bonds
may become an alternative reserves for banks.
CREDIT RISK
During 2002, non-performing loans (NPLs) tend to
decrease, which is mainly attributable to the transfer
of NPLs to the Indonesian Banks Restructuring Agency
[IBRA]. However, there is a potential of increasing NPLs
Thus, the primary source of income of these banks
is interest from recap bonds accounted for 39.0% of
total interest income. Consequently, the fluctuation
of interest rate will pose a high interest rate risk to
the large banks.
Most or 88.5% of total recap bonds held by the
large banks is kept in the investment portfolio, while
the rest is put in the trading portfolio. This strategy is
adopted by banks to minimize market risk.
Prior to reprofiling, most of the recap bonds are
fixed rate at 4 recap banks. Income from fixed rate
bonds, prior to reprofiling, was relatively low due to
low interest rate. Average income from fixed-rate bonds
is at 12.8%, while the average cost of fund of banks is
14.7%. In addition, fixed-rate bonds are traded at quite
huge discount rate. However, after reprofiling of bonds
(November 2002) the average rate for fixed rate bonds
is increased from 12% to 14%.
After reprofiling, bond composition is 35.8% fixed-
rate and 57.1% variable rate. The average market price
for fixed-rate bonds increases and bonds are transacted
at higher prices, particularly bonds with less than three
FIGURE 4.2BANK SECURITIES and LOANS
Trillion Rp
Total Asset L o a n Securities
0
200
400
600
800
1000
1200
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
23
Performance and Prospect of Indonesia’s Banking
Total Asset (trillion Rp) 1,112 831 74.7
IDR (%) 81.2 85.0 78.2
Forex (%) 18.8 14.9 59.2
Loan (trillion Rp) 371 241 65.0
IDR (%) 73.6 76.3 67.3
Forex (%) 26.4 23.7 58.2
NPL (trillion Rp) 33 20 60.6
NPL Gross (%) 8.1 7.1 53.1
Provisions for Earning
Assets Losses (trillion Rp) 31 19 61.3
Loan Restructuring
(trillion Rp) n.a 48
NPL (%) n.a 9
Nominal % of Total Bank
Table 4.2 :Details of Loan
13 Large BankTotalBank
from restructured and un-restructured loans purchased
by banks from IBRA.
The 1997 financial crisis has been so damaging to
banking industry, causing NPLs to soar to 54%. Quality
of bank loans then gradually improves in line with the
banking restructuring program. The gross NPLs
decreased to 8.1% and net NPLs reached 2.1% as of
December 2002 (see Figure 4.3). Meanwhile the NPL of
the 13 large banks is 7.1% (gross) or 1.6% (net). The
decrease in NPLs is mostly attributable to the transfer
of the NPLs to IBRA, while the rest are either
restructured or written off.
average Loan to Deposit Ratio is below 35% since
2000). New loans are mostly extended to small-
scale and consumers loans which explain why bank
lending portfolio is not growing fast.
iv. There are potentials for NPLs to increase out of
those restructured and un-restructured loans
purchased by banks from IBRA. Total ex-IBRA un-
restructured loans as of December 2002 were
approximately 6 trillion or 2.2% of the total lending
of large banks.
Non-Performing Loans (NPLs)
As of December 2002, gross NPLs of banks is 8.1%
(gross), and 4.3% of which are qualified as loss. Most of
them are NPLs from large banks. Total gross NPL at
large banks is at 7.1% of total their loans. Banks and
large banks have provided adequate amount of
provisions for earning assets losses, so that the net NPL
of banks and large banks is 2.1% and 1.6% respectively.
However, four of the large banks have net NPL ratio
higher than 5%.
In order to assess the ability of large banks to
bear such NPL, a more conservative NPL—Equity ratio
Some primary constraints faced by banks in
improving its credit risks are:
i. There are constraints in the process of
restructuring loans due to unfavorable economic
conditions.
ii. The capacity of real sector and corporations to
use credit is relatively low considering the fact
that most of them are still being restructured by
IBRA. New loans is relatively small (indicated by
Figure 4.4. NPL and PROVISIONSFOR LOAN LOSSES
NPLProvisions for Loan Losses
0
50
100
150
200
250
300
350
1996 1997 1998 1999 2000 2001 2002
Trillion Rp
24
Chapter 4
> 8% 9 7 7 6 7 6 5
0% - 8% 4 5 5 4 1 1 2
< 0% 0 1 1 2 5 6 6
5% 8% 10% 12% 15% 18% 20%
Tabel 4.3.NPL Stress Test
Scenario of Increased NPLC A R
Figure 4.5.Non Performing Loan
1997 1998 1999 2001 20022000
Mar Jun SepJun Sep Dec Dec Mar Jun Sep Dec Jun Sep Dec Jun Jul Sep Dec Mar Apr May Jun Jul Aug Oct Nov DecMar Jan Feb SepMar
Gross NPLs
Net NPLs
0
10
20
30
40
50
60
Percent
is used. As of December 2002, the ratio indicates that
29.6% of banks’ capital is to offset the provisions for
earning assets loss, so that the CAR of large banks might
fall from 22.0% to 15.6%. In aggregate this ratio is
adequate for banking industry. Under an even more
conservative measure, i.e. NPL to core equity, indicates
that 39.8% of the banks’ tier one capital will exhaust
to offset losses from NPLs.
Under the scenario that all the NPL are written
off, there will be four of the large banks with CAR less
than 8%, and even 2 of them have their CAR below zero.
As of December 2002, there were un-restructured
loans purchased by large banks from IBRA, which we
projected amounting to 2.2% of their total lending. In
fact this seems returning the problem loans back to
banks. Conditions might even get worse since there
are legal problems associated with the restructuring of
certain large debtors.
large banks is to increase from 7.05% to 9.2%. Such
increase in NPL is relatively high and therefore this
will adversely affect the capital of the 13 large banks.
In order to assess the impacts of lower credit
quality of large banks to their equity, stress test has
been conducted using a number of hypothetical
scenarios, i.e. NPL increase from 5% to 20%. The stress
test results indicate that some of large banks are very
sensitive to changes in NPL such that their equity would
fall down below the minimum capital adequacy. (see
Table 4.3.).
Efforts to restructure such debtors should be
placed as one of top agenda in the banks’ business plan.
Under worst case scenario, where all the loans
purchased from IBRA fail to be restructured or becoming
non-performing, there are potentials that NPL ratio at
Under 5% scenario, there are 4 banks with CAR
falling to 8%, 2 of such banks are stated-owned banks.
If NPL increases by 16%, the CAR of large banks fall
below 8%. However, it should be noted that capital
adequacy of each individual bank is differs.
Figure 4.6.NPL Stress Test
NPL Delta (%)
CAR (Percent)
A B D K Average
-40.0
-30.0
-20.0
-10.0
0.0
10.0
20.0
30.0
0 5 8 10 12 15 18 20
25
Performance and Prospect of Indonesia’s Banking
Loan Restructuring
Total restructures loans of large banks as of
December 2002 is IDR 44.4 trillion, therein inclusive of
IDR 4.1 trillion NPL (9.4%). Internal restructure of NPL
conducted by large banks yields in positive results in
the course of 2001 and 2002. However, there are
chances that such restructured loans would get worse
considering the fact that the real sector and the
economic condition is not yet conducive.
Under worst-case scenario, in which all of pass
and special mentioned restructured loans and deposits
in the amount of IDR 40.2 trillion becoming non-
performing, the NPL of these large banks will get
worsened to 21.5%. The implication is that the banks
need to put aside a provisions for earning assets losses
to cover the potential NPL, and consequently large
banks’ CAR would plunge very significantly from an
average 22.0% to as low as 8.7%, and 3 of such large
banks will have their CAR to fall to below 8%. These
means that the restructured loans carry the potential
of disrupting financial system stability.
In connection with this, there has to be efforts to
closely monitor such restructured loans and there is
urgency for lending expansion. For that matter the
Government needs to create an enabling environment
such as deregulation of the real sector, privatization,
creating more conducive investment climate, good
corporate governance and rule of law. In addition, small
and medium enterprises need to be developed in order
to enable them to absorb more loans. To realize all
that, all relevant parties and stakeholders must give
their commitment to the efforts.
From the above analysis, we can sum up that
banks’ credit risks remain high but stable. However, in
the future there are chances that lending risks will rise
again. This will manifest particularly if there are further
delays for the recovery of the real sector and due
completion of loan restructuring program.
Lending Growth
Banks’ lending has not grown much during the
post-crisis period. Foreign currency denominated loans
as well as corporate loans tend to decrease. However,
retail loans is increase.
Lending growth after the crisis is relatively low.
During 2000—2002, average Loan to Deposit Ratio of
Figure 4.8.Loan to Deposit Ratio
Trillion Rp
0
100
200
300
400
500
600
700
800
900
0
10
20
30
40
50
60
70
80
90
100
Percent
L o a n Deposit LDR
1996 1997 1998 1999 2000 2001 2002 2003
Figure 4.7Loan Restructuring
Apr
Loan Restructuring (LHS)
NPL Restructuring (RHS) NPL (RHS)0
10
20
30
40
50
60
70
80
0
0.5
1
1.5
2
2.5
3
3.5
4
May Jun Jul Aug Sep Oct Nov Dec
2 0 0 2
Trillion Rp Percent
26
Chapter 4
banks stay at a level below 35% compared to that before
the crisis, which is at 72%. In general, the condition is
attributable to lower economic growth and some large
loans portfolios were transferred to IBRA. On one side it
causes lower demand for loans and on the other side
there is increase in deposits. As the results, banks tend
to put their funds in Bank Indonesia Certificates (SBI) as
their source of income. In addition, placements in fixed-
rate recap bonds is also interesting to banks, considering
that this portfolio is exempted from capital charge (zero
risk in calculating risk weighted assets). The falling trend
of SBI interest rate forces banks to extend more loans.
In this regards, attention should be given to avoid lending
to borrowers with high credit risks.
New loans extended during 2002 amounted to IDR
79.4 trillion, of which 38% goes to small and medium
enterprises (see Figure 4.10). After the 1997 financial
crisis, banks tend to concentrate on retails lending
(small and medium enterprises) for the reasons that
they are more resilience to economic crisis and
unaffected by fluctuating exchange rate.
The change of focus from corporate lending to
retail lending gives rise to some implications. On one
side it provides banks with opportunity to diversify their
portfolio and risks. But on the other side, there are
chances for higher operating risk and strategic risk
attributable to insufficient knowledge and expertise
Agriculture 10.5% 7.2% 6.7% 6.1%
Mining 1.6% 1.7% 2.4% 1.7%
Industry 37.1% 39.4% 37.6% 33.1%
Trading 19.1% 16.3% 15.6% 18.1%
Services 19.0% 16.4% 15.8% 16.7%
Others 11.9% 18.3% 21.1% 24.4%
Table 4.4DISTRIBUTION OF LOANS BY SECTOR
Sector 1999 2000 2001 2002
From demand side, the weak economy forces
borrowers and investors to delay their investments.
From the supply side, such situation forces banks to
behave more more conservative in their lending, and
thereby lending growth remains low.
In order the reduce loans concentration, banks
tend to reduce the corporate segment, while increasing
lending to retail segment. This is meant to diversify
their portfolio risks as well as supporting the
government’s call for greater lending to micro and small
enterprises.
Figure 4.9.TRENDS OF IDR & FOREIGN EXCHANGE LOANS
1996 1997 1998 1999 2000 2001 2002
Trillion Rp
IDR Forex
0
50
100
150
200
250
300
350
400
In addition, banks also tend to reduce foreign
currency denominated loans (see Figure 4.9). This is to
reduce exposure to bank credit risks. The crisis has taught
lesson that this type of loans was very risky, since the
foreign currency denominated loans was provided to
borrowers whose revenues are in Indonesian Rupiah.
When the financial crisis strikes, most of such foreign
currency denominated loans become non-performing.
This situation shall not be repeated in the future.
27
Performance and Prospect of Indonesia’s Banking
Figure 4.10.NEW LENDING
Trillion Rp
Total New Loan Small-medium Enterprise New Loan
0
4
8
12
16
20
24
28
32
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
2 0 0 1 2 0 0 2
Figure 4.12.PROPERTY LOANS
Trillion Rp
0
10
20
30
40
50
60
70
80
90
Property
Construction
Real Estate
Housing
1996 2001 20021997 1998 1999 2000 2003
of banks in managing retail lending for they have always
been focusing on corporate lending activities. If not
anticipated well, this will create new non-performing
loans.
The change of lending focus is also reflected in
the increase of consumers loans (see Figure 4.11).
Unlike other type of loans, consumers loans increase
in value even exceeding that during pre-crisis period.
The phenomenon is an indication that the risk of SME
loans is relatively lower than the risk in corporate loans.
One type of consumers loan which is most
attractive to the public is the housing loans (KPR) (see
Trillion Rp
Working Capital Loan
-
50
100
150
200
250
300
350
400
450
Investment Loan
Consumer Loan
1996 1997 1998 1999 2000 2001 2002
Figure 4.11.LOANS BY BUSINESS USES
Figure 4.12). This type of mortgage loan carries
relatively lower risk, since the borrowers are employees
whom have steady income.
KPR loans are mainly provided to finance purchase
of houses or real estates developed by realty companies
before the crisis. Since the beginning of 2001, KPR loans
tend to grow in value in line with the stable interest
rate and banks’ efforts to improve their performance
through, among others, diversifying their lending
portfolio.
Total property loans as of December 2002
amounted to IDR 35 trillion or 9.4% of the total loans
provided by banks. Amount and growth rate of property
loan (see Figure 4.12) indicates that property loans carry
relatively lower risks and insignificant to the aggregate
bank credit risks.
LIQUIDITY RISK
In general, Indonesia’s banks have adequate
liquidity. This is evident by the fact that banks’ liquid
assets account for a quarter of their total assets.
However, Indonesian banks still faces a moderate
liquidity risks due to dependence on short-term
28
Chapter 4
funds and large depositors and foreign debts maturing
in 2003.
The aggregate liquid assets to total assets ratio
of banking industry since December 2000 continue
to increase to 23.30% as of June 2002, in line with
the increase of deposits. Nevertheless, the banks
are still to face some liquidity risks as described
below.
by using all their reserves, including primary, secondary
and tertiary reserves. The total value of reserves is also
adequate to cover the payment for all deposits maturing
within 1 to 3 months (its ratio is 109.0% of deposits
with up to 3 month maturities, or 57% of the total assets
of large banks). Thereby, the large banks shall be able
to meet their current liabilities, thanks particularly to
the relatively huge government bonds– which may be
Up to 1 month maturity 434.0
> 1 - 3 months maturities 67.0
Total 501.0
LiquidityAssets:
Primary Reserves
- Cash 13.0
- Demand deposits at Bank Indonesia 35.0
- SBI 68.0
Total Primary reserves 116.0
Secondary Reserves
- Netting AB (4.0)
- Trade bonds 44.0
Total Secondary reserves 40.0
Total Primary & Secondary reserves 156.0
Tertiary reserves
- Investment bonds 317.0
Total Reserves 473.0
Table 4.5.14 Large Banks’ Liquidity
Deposits Amount
Figure 4.13.DEPOSIT GROWTH
Trillion Rp
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Time Deposit
Checking Accounts
Saving Accounts
0
50
100
150
200
250
300
350
400
450
500
Liquidity Assets
Large banks has adequate liquidity considering
that all their existing liquid assets (primary, secondary
reserves) are sufficient to cover the whole short-term
deposits maturing up to three months. The primary
reserve owned by large banks as of December 2002
account for 26.9% or 14.1% of their total assets. Their
primary and secondary reserve accounted to 35.9% of
their deposits maturing in less than 1 month or 18.8%
of their total assets.
Under a worst-case scenario, if all deposits with
maturity less than 1 month are withdrawn by the
depositors, the large banks will be able to offset that
Figure 4.14.Liquid Assets
Prime Reserve/1 month deposit Secondary Reserve/1 month deposit
Prime Reserve/Total Asset Secondary Reserve/Total Asset
Percent
Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2001 2 0 0 2
0
5
10
15
20
25
30
35
40
45
50
29
Performance and Prospect of Indonesia’s Banking
pledged as tertiary reserve reserves in the event of
emergency.
However, large banks have to manage their
liquidity prudently in order to prevent potential losses
from selling of recap bonds (tertiary reserve).
Exchange Offer
There will be foreign debt obligations (exchange
offer) that must be paid by banks (most of them by
stated-owned banks) of US$ 728.7 million in principal
and US$ 132 million in interest maturing on June and
December 2003. This obligation has the potential
adversely affect to the liquidity of banks if not
anticipated prudently. Usually, banks address this issue
by purchasing foreign currency gradually until maturity
date, and placed in a nostro account. Other way is by
issuing Medium Term Notes (MTN). In order to prevent
shock to the banks and to the financial system as a
whole, the implications of such strategy need to be
anticipated, such as the cost that banks have to pay
and to secure the source of fund available to redeem
the notes, and market risk prompted by exchange rate
fluctuation.
Core Deposit
Banks’ funding is still dominated by non-core
deposits with an increasing trend. The funding gap has
been prevailing for quite awhile due to the higher
interest rate for no-core deposits compared to that of
core deposits (savings and demand deposits).
The high reliance on non-core deposits indicates
that banks face risk of withdrawals of funds in relatively
large amount. This also reflects that banks have
inefficient funding structure, since non-core deposits
carry higher interest rate. As of December 2002, non
core deposits accounted for 56% of the total deposits
or 40% of the total assets of banks.
Trillion Rp
Core Deposit Non Core Deposit
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
0
50
100
150
200
250
300
350
400
450
500
Figure 4.15.Core and Non-Core Deposits
Interbank Call Money
In general, liquidity condition of banking industry
is adequate, as reflected in the sufficient amount of
primary, secondary and tertiary reserves. Banks are not
dependent on interbank call money as their source of
liquidity.
However, in 2002, there are some banks borrowed
from inter-bank market, but then placed them in Bank
Indonesia Certificates or other investment instruments
such as Government Bonds. This is intended to obtain
the spread between interbank call money rate, which
is lower than the rate of Bank Indonesia Certificates.
Such conduct is evident in interbank transactions among
large banks which is short of IDR 4.5 trillion. This is
mainly attributable to the large short position of IDR
20.4 trillion by one bank as of December 2002. The
bank arbitrates considering the wide spread between
interest rate of inter-bank borrowings and SBI’s interest
30
Chapter 4
Grafik 4.17.Maturity Profile of Time
up to 1 month >1 year3 - 6 months
0
50
100
150
200
250
300
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Trillion Rp
Figure 4.16.Stock Liquid Ratio
rate. The bank also to purchase SBI as a secondary
reserve. This practice may be followed by other banks
which have access to the long-term borrowings from
money market. Thereby, with the liquid condition of
banks, interbank call money transaction is not a good
liquidity indicator.
Liquid Assets to Cash Outflow (COF)
Liquid assets held by large banks as of December
2002 is insufficient to cover cash outflow maturing in 1
to 3 months. Primary reserve available is only 19.1% of
COF 1 to 3 months or 25.5% of available primary and
secondary reserve and 77.4% of all the reserve owned
by large banks. Similar condition also applies for COF 1
week ahead, such that all existing reserves will be
adequate to cover all COF for 1 week to come (233.3%).
In addition, with the introduction of Finance
Minister’s policy limiting pension funds placement at a
single bank to maximum of 20% as of mid-2003, will
adversely affect the liquidity of some large banks.
Household Savings Pattern
Public confidence to the banking system is
improving in line with bank restructuring program and
the government’s blanket guarantee.
77.4 %25.5%
19.1%
Primary Reserve/COF 1 - 3 months
Primary &Secondary Reserve/COF 1 - 3 months
All Reserve/COF 1 - 3 months
Corporate Funds
The large banks diversification of source of funding
is very low. The banks are very much dependent on
funds from large corporate depositors. Significant
withdrawals by such corporations will surely affect the
liquidity of the large banks.
This is reflected in the increasing of deposits
consistently after the 1997 crisis. However, such
situation needs to be closely monitored considering that
household deposits still concentrate on short-term
maturity, particularly below 1 month. Such condition
reflects banks’ perception on the likelihood of lower
interest rate in the future. In response, banks offer
more attractive interest rate for short-term deposits.
In order to maintain public confidence, the plan of
withdrawal of government blanket guarantee needs to
be carefully anticipated by putting in place the most
suitable deposit guarantee scheme.
31
Performance and Prospect of Indonesia’s Banking
Maturity Profile
Maturity profile of assets and liabilities of large
banks, both in IDR and foreign currency, indicates a
negative positions. This condition reflects the high
possibility for maturity mismatch experienced by such
banks. In addition, the potential is high for depositors
to withdraw their moneys on maturity dates instead of
to rolled over their money.
considering the rising political tensions as we are
approaching the 2004 general elections. Appreciation
of the United States Dollar against the IDR will cause
interest rate become increasing.
Stress test on exchange rates and interest rates
shows capital resilience of large banks to market
volatility. The stress test adopts a number of scenarios
based on CAR position as of the closing of 2002.
Exchange rate scenario is assumed by
depreciation of IDR against USD amounted from IDR
State-owned banks are having short positions for
up to 3 month maturities in a relatively large amount.
In addition, state-owned banks also face the risk of
withdrawal of corporate deposit, which form the largest
portion of their deposits. However, such condition does
not in fact reflect the condition of large banks’ liquidity
risk, for the reason that the method of calculating and
reporting of maturity profile is not yet fully adopt
withdrawal behavioral method. This reflected from the
accounts which continue to be rolled over is not yet
accounted in calculating the gap of of assets and
liabilities.
MARKET RISK
Future possible depreciation of the IDR toward
the United States Dollar needs to be anticipated
Foreign Currency
Long (bank) 3 5 6 8 8 4
Short (bank) 10 8 7 5 5 9
I D R
Long (bank) 0 4 8 6 11 6
Short (bank) 13 9 5 7 1 7
Table 4.6.Maturity Profile of Assets and Liabilites of
13 large Banks, December 2002
M a t u r i t y
Currency up to >1 up to >3 up to >6 up to > 12 Total
1 month 3 months 6 months 12 months months
> 8% 13 13 12 12 12 12 12
0% - 8% 0 0 1 1 1 1 1
< 0% 0 0 0 0 0 0 0
500 1000 2000 2500 3000 4000 5000
Table 4.7. EXCHANGE RATES STRESS TEST OF LARGE BANKS TO CAR
Appreciation Scenario (IDR/USD)C A R
> 8% 13 13 13 13 12 12 12 12 12 12
0% - 8% 0 0 0 0 1 1 1 1 1 1
< 0% 0 0 0 0 0 0 0 0 0 0
1% 2% 3% 4% 5% 6% 7% 8% 9% 10%
Table 4.8. INTEREST RATES STRESS TESTOF LARGE BANKS TO CAR
Interest Rate IncreaseC A R
Figure 4.18 :EXCHANGE RATES STRESS TEST
Exchange Rate Change
Bank B Average
0 500 1000 2000 2500 3000 4000 5000
CAR (Percent)
0
5
10
15
20
25
30
35
Bank A Bank L
32
Chapter 4
Figure 4.20CAPITAL RATIOS
2001
CAR
Total Capital/Total Asset
Tier 1/Total Capital
Profit /Tier 10
5
10
15
20
25
30
35
Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2 0 0 2
Percent
500 (such that US$ 1 = IDR 9,440) up to IDR 5,000 (US$
1 = IDR 13,940). The stress test results indicate that 5
large banks will face declining CAR. one bank of them
will experience declining CAR to a level below 8% if
the IDR exchange rate depreciates by IDR 2,000 or IDR
10,940 for US$ 1. The small foreign currency exposure
as the impact of restricted transactions between banks
and non-resident parties cause less sensitiveness of
large banks toward exchange rate volatility.
Meanwhile, assuming interest rate of 1 to 3
month deposits increases in range from 1% (become
12.6%) to 10% (become 21.6%), the stress test indicates
that 6 of the large banks will face a falling CAR. If
interest rate increases by 5%, i.e. to 16.6%, only one
bank will fall its CAR under 8% level.
Capital Charge for Market Risk
Implementation of new capital regulation
incorporating capital charge for market risk is not
adversely affecting the CAR of large banks. Based on
simulation of capital charge for market risk as of June
2002, CAR decreases at an average of 1%. This is caused
by the fact that exposure of its portfolio to market risk
is relatively small. However, risk of increasing interest
rate from banking books need to be anticipated.
Although implementation of capital charge for market
risk will not cause instability in the financial system,
however, there are chances of more significant CAR
decreasing if banking books is included in the capital
charge.
CAPITAL
Following the recapitalization of banking industry,
banks capital continue to improve. The average CAR of
banking industry as of December 2002 was 21.7%. In
addition, the Government dominates up to 70% equity
ownership of the banking industry. The relatively high
of banks CAR is mainly attributable to additional capital
injection from recap bonds. This is the implication of
the changes in assets structure of banks after the recap
program by transferring non-performing assets into the
IBRA and substituted them with recap bonds (the risk
or capital charge of recap bond is zero).
In addition, it should be anticipated that capital
base in a number of banks does not fully cover the
Interest Rate Change (%)
CAR (Percent)
0 1 2 3 4 5 7 8 9 10
Bank A Bank B
Bank L Average0
5
10
15
20
25
Figure 4.19 :INTEREST RATES STRESS TEST
33
Performance and Prospect of Indonesia’s Banking
potential loss originating from market risks and
operating risks. The current method in CAR calculation
is imperfect because:(1) it only counts in credit risks
and not yet counting in market and operating risks;
and (2) it fails to reflect the actual credit risks, because
the current method still adopt weighted assets under
Basel 1988. Therefore, there will be some banks may
face inadequate capital if there is no amendment for
capital regulation.
The relatively high CAR of banks is mainly
attributable to the recap program under which earning
assets categorized loss are transferred to IBRA, for
which the banks receive recap bonds. A more
conservative approach indicates that banks’capital is
relatively limited.
This is indicated with core capital to total assets
ratio which is below 8%, and leverage ratio (total
liabilities to total equity) of 9.6 times. A number of
Box 4 :
MARKET RISKS
Currently, calculation of the Capital Adequacy
Ratio for Indonesian banks is essentially based on the
Basel Capital Accord [BCA] 1988. The BCA 1988 assess
the adequacy of capital with the respect to credit risk.
In line with the growing complexity in banking
transactions, this approach needs to implement to
cover various risks. In practice, operations of a bank
will not only entail credit risk but also various other
risks such as market risk and operational risk.
In January 1996, Basel Committee issued an
amendment to BCA 1988 which including market risk
in calculating capital adequacy of a bank. This new
regime is effective since December 1997 in G-10
countries. As for Indonesia, this regulation will be
adopted in near future in line with the improving of
bank performances, the increasing of bank trading
activities, and more integrated domestic money
market with the global money market.
The scope of market risk incorporated in the
amendment BCA 1996 essentially cover interest rate
risk, equity risk, commodities risk, foreign exchange
risk and option risk. Nevertheless, for the time being
banks will be required to allocate it’s capital to cover
risk of loss attributable to volatility of exchange rate
and interest rate, inclusive of the derivative positions.
Application of market risk will technically
increase the RWA in calculating bank’s capital
adequacy. Therefore, in order to maintain a minimum
CAR of 8%, banks in response to the proposed
implementation of this provision will have to manage
the market risk of their portfolio in a more proper
approach.
BIS issue two calculation methods, namely
standard model and internal model. The standard
model can be implemented directly by banks in
calculating additional equity to cover for market risk
to its portfolio. While internal model may be used
depend on the capacity of each individual bank.
In line with the role and function of equity in
bank operations, introduction of regulation about
capital charge for market risk will expectedly
strengthen the bank capital structure. Adoption of
capital charge for market risk will bring a positive
implication to financial system as increasing a sense
of security to investors while improving
competitiveness of Indonesian banks in international
forum.
34
Chapter 4
Sep 98Government Share (percent)
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
5
10
15
20
25
0 10 20 30 40 50 60 70 80 90 100
4%Jun 97
Dec 97
Jun 98
Dec 98
Mar 99
Jun 99
Sep 99
Sep 97Dec 00
Dec 99
Mar 00
Jun 00
Sep 00
Dec 00
Mar 98
Jun 01
Sep 01
Dec 01
Mar 02Jun 02Sep 02
Dec 02 Mar 01
Cap
ital A
deq
uacy R
atio
Percent
Figure 4.21 CAR EVOLUTION
19%
41%
40%
Securities Loan S B I
Trillion Rp Percent
(20)
(15)
(10)
(5)
0
5
Net Income (LHS) ROA (RHS)
-200
-150
-100
-50
0
50
1996 1997 1998 1999 2000 2001 2002
Figure 4.22SOURCE of INTEREST INCOME
Figure 4.23NET EARNINGS and ROA
efforts have been made in order to refine this new
capital regulation. Among them is the plan to adopt
the capital charge for market risk (see Box 4) in 2003
with a transitional period, and provisions adjustments
on credit risk and operational risk as proposed in the
New Basel Accord (Basel II).
In line with bank recap program, the government
is now the dominant owner in banking industry.
However, the ownership by the government is only
temporary and its stake will be divested to private
parties. The divestment will be conducted gradually
until reaching such position prevailing prior to the crisis
(around 6%). Evolution of CAR and government stake in
banking industry is illustrated in Figure 4.20.
BANKS’ PERFORMANCE
In general, Indonesian banking performance
following the recapitalization and restructuring
programs has been improving. This is among others
reflected in operating performance of banks which have
been able to record a net earnings of IDR 16.5 trillion
with ROA at an average of 1.92%; ROE at 14.8% and NII
at 42.9% as of the closing of 2002. The improved
profitability lays the base for banks to improve their
capital structure and increase their performance in the
future.
However, it must be noted that such profitability
still very much dependent on revenues from marketable
securities (particularly from government bonds and SBI)
which account for 60% of banks’ total interest income.
This indicates that bank’s intermediation function is
35
Performance and Prospect of Indonesia’s Banking
not yet fully recovered. The earnings of banks,
particularly large banks (recap banks) does not show
the actual performance of banks with their core
business activities.
In the long run, the dependence of large banks
on income from recap bonds and Bank Indonesia
Certificates will adversely affect the financial system.
If the Government’s financial condition worsens, due
to the absence of multiplier effect from lending, bonds
interests will also adversely affected.
Trillion Rp Percent
(475)(450)(425)(400)(375)(350)(325)(300)(275)(250)(225)(200)(175)(150)(125)(100)(75)(50)(25)-2550
-
20
40
60
80
100
120
1996 1997 1998 1999 2000 2001 2002
Paid-up Capital (LHS) ROE (RHS) NII (RHS)
Figure 4.24PAID-UP CAPITAL and ROE
It is estimated that loan interest income will not
increase in the near future. The huge NPLs experiencing
during the banking crisis and the non-conducive business
environment cause banks to maintain cautions in their
lending activities. As the results, new loans expansion
is limited and banks prefer to place their funds in low
risk instruments such as Bank Indonesia Certificates and
government bonds.
In the near future, if lending opportunity is still
low, local banks will purchase and hold such bonds for
the following purposes: (1) to maintain their revenues
to safeguard their financial condition; (2) to control
circulation of bonds in the market in order to prevent
over supply which would otherwise reduce the prices
of bonds; (3) to maintain liquidity by placing such bonds
as liquidity instruments. Therefore, banks’ earnings will
remain bogus.
Banks still have to face with various constraints.
Large banks’ operating income is very sensitive to
volatility of interest rate in market. Meanwhile, there
are increasing tendency of fund collections by non-bank
financial institutions, such as mutual funds, which poses
different challenges.
In order to improve their performance banks,
particularly recap banks, are required to implement
good corporate governance and risk management and
increase effectiveness of their internal control. All that
is laid down in business plan and actions plan of the
banks in line with the bank restructuring program. In
general, banks have made significant progress in their
corrective measures.
Some malpractice incidents taking place recently
stress the importance of applying the good corporate
governance principle consistently and improved
Figure 4.25Interest Income
Trillion Rp
0
10
20
30
40
50
60
70
80
90
L o a n Securities and SBI
1996 1997 1998 1999 2000 2001 2002
36
Chapter 4
Thailand Indonesia Malaysia Korea
-20
-15
-10
-5
0
5
1997 1998 1999 2000 2001 2002
Percent
Figure 4.26Asian Banks’ ROA
effectiveness of internal control in order to address
operating risks as well as reputation risk. Realization
of good corporate governance in banking sector calls
for commitment and participation from all stakeholders,
i.e. the Government, Bank Indonesia and oversight
authorities, banks’ shareholders, Board of
Commissioners and Board of Directors, internal audit
and audit committee, external auditors and down to
customers. All parties are required to perform their
respective roles and responsibilities consistently.
Profile of Banks at Stock Exchanges
The improved performance of banks unfortunately
is not reflected in the value of their shares listed at
the stock exchanges. From 24 banks (43.1% of the total
assets in banking sector) listed at the stock exchanges,
the transaction volume of their shares does not yet
resemble that of prior to the crisis. As of December
2002, only 3 banks has share price exceeding the IPO
price. This cannot be separated from the bogus earnings
of the banks since most of their revenues come from
bond interests.
Standard & Poor still rates CONSTRAINTS plus and
BB minus for 8 of the large banks which have issued IDR
denominated bonds, with stable tendency. As for foreign
currency denominated bonds the rating is B minus with
stable tendency. Such condition indicates that there
has been increasing confidence on the part of foreign
investors so that the banks are able to utilize
international market as their financing alternatives.
In order to maintain their financial performance,
banks need to increase revenues from lending. This is
important in order to obtain a real performance of the
banks and to eliminate dependence on government
bonds as source of income. Thereby, the banks will have
greater role in encouraging economic growth.
Comparative Performance with Other Selected
Countries
Condition of Indonesian banks after the crisis is
relatively similar to that prevailing in some Asian
countries. The profitability (ROA) of Indonesian banks
scores higher than those of banks in other Asian countries.
Non-performing loans (NPLs) of Indonesia’s banks
is 8.1% (gross) as of December 2002, which is better
than NPLs of banks in other Asian countries. The gradual
1997 1998 1999 2000 2001 20020
10
20
30
40
50
60
Thailand Indonesia Malaysia
Percent
Figure 4.27Asian Banks’ NPL
37
Performance and Prospect of Indonesia’s Banking
improvement in the NPLs is the results of the transfer
of NPLs to IBRA, loans restructuring by the banks
themselves and writen off bad loans.
Figure 4.27 indicates that NPLs of Indonesia’s
banks as of December 2002 is below NPLs of banks in
Thailand and Philippine, ranging from 10—12%.However,
NPL in Indonesia’s banks is still higher that NPL at banks
in South Korea, which stays at 4%.
(Figure 4.28). This is attributable to the government.
recapitalization program for banks, while CAR of non-
recap banks is ranging from 10—15%.
In spite of the fact, to some extent, Indonesian
banks are recovering. However, the condition does not
reflect the actual condition of the banks, considering
the fact that some of the private large banks have
subsidiaries or themselves belonging to a parent or
holding company. Such affiliations - by finance or
ownership - will very much affect the conditions of the
banks. Experience has shown us that the crisis
experienced by some large private banks have in fact
been triggered by financial problem facing by their
affiliated parties. Therefore, this factor may prompt
instability to the banking system as well as financial
system.
Moreover, there have been a number of new
product innovations introduced by banks which add risks
to the banking sector. The rising of products such as
mutual funds with underlying recap bonds,
securitization of assets from non-performing loans,
bancassurance and others must be prudently
anticipated.
-20
-15
-10
-5
0
5
10
15
20
25
1997 1998 1999 2000 2001 2002
Thailand Indonesia Malaysia
Percent
Figure 4.28Asian Banks’ CAR
The aggregate CAR of Indonesian banks which is
22%, is relatively higher than the CAR of banks in
Thailand and Malaysia which ranges from 10—15%.
38
Chapter 5
Box 5 :
YIELD CURVE OF
GOVERNMENT BONDS
One of the important indicators of investor
confidence in the capital market and of investor
economic expectation is a yield curve. The curve is
represented by yield and the remaining year of bonds
life. Based on the market transactions and current
prices in the secondary market during the year 2002,
yield curve of the Government bonds until the year
2008 indicates an upward trend.
The level of public confidence to the prospect
of the national economy in the future shows a rising
movement as seen in the increasing transactions of
bonds. Lower bank deposit interest rates and the
introduction of a variety of derivative products using
government bonds as the basis also give positive impact
to the bond market. However, the change of the
interest rates and the capability of the Government
CAPITAL MARKET5
Negative shocks in the capital markets might
result in a damaging impact to the financial
system. Crisis of confidence or any large unforeseen
fluctuations on the capital market as reflected by stock
prices volatility might create flight to safety
phenomena from stock or bond instrument to cash.
The shock may force investors to sell their stocks
portfolio immediately and accelerate demand for
liquidity, especially in US Dollar. This will spread panic
attack in other financial sector.
Bank customers will eventually withdraw their
fund in rush (bank run), resulting in crisis of liquidity.
As an example is the crisis of the capital market in the
United Governments during the year 1929 and in UK
during the year 1982.
CONFIDENCE TO CAPITAL MARKET
The investors’ confidence to the capital market
in general and banking stocks in particular has not yet
fully recovered and the role of the capital market as
Government Bond Yield to MaturityJanuary - December 2002
to repay bonds will also influence short horizon and
unsophisticated investors.
For financial stability monitoring, this yield curve
analysis is somewhat significant to identify initial signs
of the crisis which might disrupt the stability of the
financial sectors, especially owing to the limitation of
alternative source of financing and investments.
Maturity (Year)
YTM Jan YTM Apr YTM Jul YTM Oct YTM Dec
Percent
11
13
15
17
19
21
23
0 1 2 3 4 5 6 7 8
CHAPTER
39
Capital Market
Apart from that, weak condition of the capital
market leads way to unwillingness of the companies/
financial institutions to resort to cheaper fund in the
capital market. Moreover, scarcity of alternative sources
of financing will cause systemic risk to the companies
including banks, thereby weaken and prolonged the
prevailing crisis.
Mutual Funds
Despite suppressed condition of the capital
market, yet a number of derivative products of the
capital market, especially those related to bonds, have
experienced promising growth. Until December 2002,
products of the Mutual Funds have been increasing
reaching a cumulative value of Rp. 46.6 trillion. This
growth might likely be enhanced by the rise of the
interest of the investors in the Mutual Funds in which
their portfolios comprise of the fixed-rate Government
bonds. This condition can be seen from the declining
trend of the interest rates and the introduction of
policies on the tax exemption for the Capital Gain of
Mutual Funds (for five years from the date of launching).
an alternative source of financing has not developed
appropriately. The condition might result in potential
cause for the rising of systematic risk should banking
crisis prevail.
Some main indicators of the investors’ confidence
such as market liquidity, stock prices index particularly
in financial sector (IHSSK) and spread index, still show
a weak signal. Since 1997 liquidity of the market
remains weak, and volume of the trade remains at the
marginal condition. The average turn over ratio prior
to the crisis era was 0.28%, whereas during the year
2002 it was lowered to stay at 0.08%.
Figure 5.1Market liquidity and JCSI
During 2002, fluctuation of the composite stock
price index (IHSG) and the financial sector stock index
including banking remains at as low as 5% and never
exceeds its psychological limit prior to the economic
crisis in 1997. If this is compared with the liquid stocks
index LQ45, spread index of the financial sector has a
wider trend reflecting the public perception that the
risk in the capital market is not equitable with the
return from stock market investments.
Figure 5.2Development of Mutual Fund and Bank Deposits
Source: Jakarta Stock Exchange
Average Daily Trading to Listed Share Ratio
1997 2002
Liquidity (Percent)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
-
100
200
300
400
500
600
700
800
JCSI
Liquidity
JCSI
1998 1999 2000 2001 2003
Mutual Fund Deposit Mutual Fund Growth (%) Deposit Growth (%)
2 0 0 2
-0.05
0
0.05
0.1
0.15
0.2
0.25
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov0
100
200
300
400
500
600
700
800
900
PercentTrillion Rp
40
Chapter 5
In comparison to the growth of the third party
deposits in banks, the growth of transactions in the
Mutual Funds has shown significant rise. This trend is
expected to encourage the transactions of the other
products of the capital market.
The risk caused by the transactions to the financial
system depends on the role of the bank and the type of
the transactions which have been made. As the banks
in general just act as the agent, the risks which might
arise are might generally in the form of legal risk and
its reputation, especially in connection with purchase
agreement of the products of the Mutual Funds.
Nevertheless, in practice there are banks acting as
standby buyer for guaranteeing the return of fixed
Mutual Funds. These banks are exposed to substantial
price risk in case there is change in the bank interest
rate to the opposite direction against the expectation
of the bank (reverse shock), resulting in high interest
and liquidity risks to the banks. Besides, the supervisory
authorities need to be alert in case there are inter-
transactions among the affiliated parties which violates
legal lending limit regulations.
It is important to avoid the cases in as much to
give protection to the confidence of the public which
likely revives to execute transactions on the capital
market.
Growth in the Mutual Funds gives beneficial
impact to the investors, banks and Government.
However, viewed from the prospect of stability of the
financial system it is necessary to give specific
attention to banks acting as market agent of the Mutual
Funds products. If Investment Manager (MI) defaults,
these banks may be exposed to substantial reputation
damage. The MI’s exposure towards credit risk,
liquidity and stock price is not adequately understood
by the investors. Also, a number of banks acting as
standby buyer might be exposed to substantial price
risk, if they repurchase Government bonds from MI in
case of redemption.
Based on Net Asset Value (NAB), investment in
Mutual Funds (mutual fund or trust unit) has been rising
rapidly during 2002 by 478.50% (y-o-y). By using 1996
as basis year, Mutual Funds have a growth of 175.53%.
Fixed income mutual fund -with Government bonds
as the underlying instruments- has the highest growth
rate. Rapid growth of Mutual Funds is reflected from
the volume of accounts and number of products
offered.
Type of Mutual Fund
Table 5.1.Development of Net Assets of Mutual Fund
(in Billion Rupiah)
Money Market 15.6 25.4 37.7 575.2 1,243.6 2,271.1 7,203.9 6,764.6
Fixed Income 1,897.9 3.635.3 1,968.9 2,839.3 3,062.0 4,660.5 37,328.5 41,988.3
Mixed 349.6 648.5 392.5 598.7 649.4 635.3 1,779.1 1,989.6
Stock 519.3 583.8 535.2 960.8 560.6 490.9 302.3 252.0
Total 2,782.4 4.893.0 2,934.3 4,974.0 5,515.6 8,057.8 46,613.8 50,994.5
Annual Growth (y-o-y) 75.86% -40.03% 69.51% 10.89% 46.09% 478.49% 9.40%
Growth since 1996 56.45% -51.13% 52.78% 10.34% 37.91% 175.53% 8.98%
1996 1997 1998 1999 2000 2001 2002 Jan-03
Source : Bapepam, reprocessed
41
Capital Market
The volume of the investor accounts has been
significantly rising by 143.20% until December 2002
(y-o-y). In 2002, there were 131 types of Mutual Funds
products offered from formerly 25 types in 1996.
The ratio of fund under management of the Mutual
Funds to the third party deposits in banks has been
drastically growing from 1.07% in January 2002 to
5.83% in December 2002. In some banks, Mutual Funds
have attracted investors who previously made the
investment in traditional products such as time
deposit.
Three driving factors of high growth of mutual
funds during 2002 are as following:
- Decrease of SBI rate up to 457 bp which is followed
by decline in time deposit interest rates by 324
bp. Consequently there is conversion of fund from
time deposit to Mutual Funds especially in a
number of banks as this investment instrument
gives more attractive return – at least presently –
in comparison to time deposit, as Mutual Funds
provides fixed income.
- Investment in Mutual Funds is exempted from
income tax especially for the type of Mutual Funds
which due date is less than 5 years. Whereas
investment in time deposit is subject to final
income tax on the interest of the deposit of 20%.
- There is no limitation of the placement of Mutual
Funds in Government bond. Placement in the other
stocks such as Corporate Bond, Bapepam has
introduced a policy to limit the Mutual Funds
placement by maximum 10%.
- Banks have been more active to act as the outlet
of marketing (marketing channel) of the products
of the Mutual Funds which are offered by
Investment Manager (MI).
At present there are 12 foreign exchange banks
that operate as brokers for the sale of Mutual Funds,
and these banks have dominated market share of the
Mutual Funds.
From the bank point of view, the factor which
gives stimulation to drastic growth of the Mutual Funds
is the motivation to maintain market share. By providing
alternative investment of fund especially to prime
customers, banks obtain two major benefits from it,
namely acceleration of fee-based income and
Figure 5.3 :Growth of Mutual Funds
1996 1997 1998 1999 2000 2001 2002
-100
0
100
200
300
400
500
600
Percent
Figure 5.4. Development of Deposit vs PublicFund in Mutual Funds
Jan Feb Mar Apr May Jun Jul Aug Sep Okt Nov Dec
2 0 0 2
Fund Managed by Investment Manager (LHS) Deposit (RHS)
Trillion Rp Trillion Rp
60
50
40
30
20
10
0
840
830
820
810
800
790
780
770
760
750
42
Chapter 5
maintenance of those customers so that they will not
turn to the other banks. At present prime customers
have the main expectation to obtain high return from
their investment in line with the decreasing trend of
interest rate of SBI and time deposit.
Viewed from the interest of the Government, the
growth of the Mutual Funds gives some advantages.
First, tax exemption will help create secondary market
of the Government bond which is liquid and big. Along
with the exemption from tax, Government bond will
become attractive underlying instruments to the
investors and investment Managers, so that these
instruments will be able to trade at prices which reflect
fair market price level. Besides, liquid market will
eventually encourage yield to maturity (YTM) of the
bond to become lower. Secondly, there will be
economization of the Government fund. With lower
YTM, there will be economization of the Government
fund. Graphic 4.30 shows that YTM of the Government
bonds which are active on trade have the tendency of
sharp decline in YTM movement. By the end of February
2003, YTM of bonds of serial number FR 06 and FR 08
(recap bonds) even have been lower than the interest
of SBI for 1 month.
Impacts on Financial System Stability
Viewed from the perspective of monetary system
stability, the impact of drastic growth of Mutual Funds
needs to be given close attention especially to the
banking sector. There are a number of facts which need
to be given the close attention to in relation with the
growth of Mutual Funds as follows :
a. Change of Risk Profile of Bank
Transaction of investment management through
bank has the impact to increase risk of the bank either
who acts as agent or as standby buyer. The bank who
acts as agent for the marketing of the Mutual Funds
products (white label products) will have the risk of
lowering its reputation (reputation risk) in case the
Investment Manager (MI) has default, the more if the
said MI is affiliated with the bank. Public might have
perception that the default will fully become the
responsibility of and can be burdened to the bank. The
bank who acts as standby buyer might have the risk of
significantly higher prices, especially those who make
repurchase agreement of the Government bond with
MI. Some banks have made the repurchase agreement
with MI, in which in case there is withdrawal
(redemption) in big amount, MI will be allowed to resell
the Government bond to the said banks. If the interest
of SBI gets lower, the price of Government bond which
become the underlying assets of fixed income of the
Mutual Funds will increase, so that the bank who makes
the repurchase agreement will face risk exposure of
high price.
Figure 5.5 Development of YTM of Some Fixed Rate Bonds
and SBI RatesPercent
15.0
14.5
14.0
13.5
13,0
12.5
12.0
11.53 10 17 24 31
January 2003 February 2003
FR 06 FR 08 FR 21 1 Month SBI
7 14 21 28
43
Capital Market
b. Liquidity Risk faced by MI
The investors do not yet fully comprehend the
risk of investment in the Mutual Funds, especially which
is related to redemption, in case the investment in the
Mutual Funds is due. MI will face liquidity risk if there
is resale (redemption) in big amount within a short
period by holders of the share units. Besides, liquidity
risk resulted from the redemption will most likely take
place if bank intermediation prevails soon. This liquidity
risk might also be resulted from basic change in the
policy on interest rate or monetary instruments. Also,
liquidity risk faced by MI will become higher if there is
decrease of NAB which generally leads way to
withdrawal. This case might take place especially when
the condition in which the investors in Indonesia in
general do not yet adequately possess knowledge on
the investment in non-bank instruments and that they
generally try to avoid the risk (risk-averse).
c. There is no Market Discipline
Taking up that market discipline in Indonesia is
not yet fully and properly exercised, investment in
Mutual Funds has some risk exposure which might
influence the condition of the stability of the national
monetary system in the long run. At present disclosure
and transparency of the Investment Manager (MI) is still
relatively limited, so that the investors do not yet fully
comprehend the risk of investment in the Mutual Funds,
particularly in connection with credit risk, liquidity risk
and price risk.
d. There is no Uniformity of Valuation
Method.
At present Bapepam does not yet to introduce
any regulation dealing with method for valuation of
the price of bond. The absence of uniformity of method
for determining the price of bond which is decided by
Bapepam is potential to affect the interest of the
investors in the Mutual Funds, especially when they
wish to redeem their investment. Presently, the price
of bond at the time of its redemption by the bank
depends on method of valuation of the price which is
applied by MI, whether it is based on mark to market
or amortization (to be amortized until its due date
(amortization) or by applying the other method of
valuation.
44
Chapter 5
Box 6 :
MUTUAL FUNDS
Mutual Funds is the instrument of investment to
collect fund from the investors to be invested in stock
portfolio by Investment Manager (MI). Each type of
Mutual Funds has its legal entity and NPWP. Besides
providing income (return), Mutual Funds also perform
as the instrument to diversify risk. Mutual Funds in
Indonesia have been rapidly developing since 2002.
This development has been encouraged by active role
of banks in marketing the Mutual Funds and tax
incentive as well as decrease of the interest rates of
time deposits. There are 131 types of Mutual Funds
with combined outlay of Rp. 56,093.90 billion under
its management.
Based on its form, Mutual Funds can be classified
into:
1. Corporate Mutual Funds, in which fund is
collected through selling stocks and then the fund
is invested into a number of types of the stock on
the capital market as well as money market. This
Mutual Fund is further grouped into Closed
Corporate Mutual Funds and Open Corporate
Mutual Funds.
2. Mutual Funds under Collective Contract
(Contractual Type), contract between MI and
custodian banks which binds holders of Share
Units in which MI is given the authority to manage
collective investment portfolios and custodian
bank is given the authority to execute collective
custody.
Based on its investment portfolio, Mutual Funds
can be divided into :
1. Money Market Fund – collection of investment in
stocks which is loan in nature and the due date
is less than one year. The purpose of this
investment is to maintain liquidity and capital.
2. Fixed Income Fund - collection of investment
minimum 80% of the assets is in the form of loan.
This Mutual Funds has relatively higher risk than
money market fund. The purpose is to generate
rate of return which is stable.
3. Equity Fund – collection of investment minimum
80% of its assets is in the form of stocks which are
equity in nature. Its return is in fact higher, yet
the risk is also higher than the previous types of
the Mutual Funds.
4. Discretionary Fund – collection of investment in
the form of stocks which are equity and loan in
nature.
The performance of Mutual Funds is reflected
by its Net Assets Value (NAB), namely fair market value
of the whole stocks and the other assets of the Mutual
Funds deducted by all liabilities of the Mutual Funds
which become burden of the Mutual Funds during the
period. NAB per equity unit on the first day of the
offer of the Mutual Funds is usually Rp. 1,000, and
NAB of per unit of further equity is the total NAB of
the Mutual Funds divided by total amount of the units
of equity which are under circulation on the related
day. Custodian bank shall be obliged to calculate NAB
of per unit equity of everyday, based on fair market
value which is determined by MI and informed to MI
on every business day as the basis of computation of
NAB.
Contractual Type of Mutual Funds is introduced
in Indonesia since 1996. Until the end of December
2002 there are 122 types of Mutual Funds issued by
Securities Company who acts as Investment Manager
(MI). Mutual Funds has grown rapidly during 2002
particularly since banks have been more active to act
as its broker. Owing to the services banks will get
45
Capital Market
compensation of brokerage fee from MI. Therefore,
Mutual Funds has become one of the alternatives of
investment which is attractive to the public beside time
deposits. In Malaysia and the United Governments
investment in Mutual Funds has reached 50% and 60%
respectively of the total fund investment portfolios.
Mechanism of the sale of Mutual Funds through
bank is described in the following mechanism :
a. At the time of purchase of Mutual Funds :
(1b).Customer fills in resale form along with photo copy
of identity to Investment Manager (MI) (T+ 0).
2. Bank sends original form of resale along with
photo copy of identity to Investment Manager
3. Investment Manager submits resale form to
Custodian Bank.
4. Custodian Bank transfers customer’s fund to Bank.
Investor needs to take careful attention to the
overall investment risk in Mutual Funds, to include
the following:
- Credit risk – decline of Net Asset Value (NAB) in
case there is default/ bankruptcy of investment
manager and issuer of promissory notes (issuer).
- Liquidity risk – delay in settlement of portfolio
which is caused by massive resale (redemption)
within a short period by holders of units of equity.
- Price risk – decline of NAB as the result of change
of market price of the portfolio.
The bank who acts as agent of the Mutual Funds
has to face reputation risk which might change to
become legal risk, being resultant to legal claim filed
by the customer to the bank, which is caused by
misperception of the customer that Mutual Funds
which is sold by the bank is the product of the bank
and guaranteed by customer’s deposit guarantee
scheme.
Bank
(3)
Custodian
Bank
Customer
(7)
Investment
Manager(6)
(2) (4) (5)(1)
Description :
(1). Bank provides brochure and prospectus of
Mutual Funds from MI.
(2). Customer fills in order form, deposit form/
transfer to current account along with evidence
of identity.
(3). Bank transfers customer’s fund to MI account.
(4). Bank sends form of order to MI and Custodian
Bank.
(5). Custodian bank sends confirmation letter to MI
(6). MI sends letter of confirmation to customer.
(7). Customer holds certificate of fund investment
from MI.
b. At the time of Resale
Description:
(1a).Customer fills in resale form along with evidence
of identity and equity to bank (T + 0).
Bank
(3)
Custodian
Bank
CustomerInvestment
Manager
(1b)
(4)
(3)(1a)
(2)
46
Chapter 5
Bond Market
Bond market has been developing but remains at
a relatively low level, evidence by the growth of bond
index, number of company issues and bonds market
value.
With the decreasing trend of interest rates, the
bond market is predicted to continue developing in the
immediate future.
As a consequence of falling rates, some
recapitalization banks have started to transfer their
floating rate bonds from investment into trade portfolio.
Since the investment portfolio remains quite large and
the spread of floating rate bond and SBI remains
positive, banks prefer to hold government bond rather
than to channel credit.
It shows on the loan to deposit ratio during 2002
that is relatively low compared to that before crisis at
an average of below 35%. This condition may prevent
the acceleration of Indonesia’s economic recovery and
the improvement of the real sectors.
Stocks Market
The Indonesia’s composite stock price index has
showed a growing trend although still relatively stable
at low level. The financial sector stock price index
dominated by bank stock is fairly low at 40-60.
During the first semester of 2002, stock trend
including banking has shown an optimistic movement.
The January effect and government divestment plans
have given positive signal to the market. Moreover,
market participants expect that the new elected stock
exchange board of directors, Bapepam (Indonesia
capital market supervisory body), Self Regulatory
Organization (SRO) and market players can restructure
and develop the market.
However, on the 2nd half of 2002 the capital market
trend has deteriorated all over again along with the
crisis in the global capital market being affected by
financial scandal in big US’s companies such as Xerox
and Worldcom. On the domestic market, delay in the
improvement of market structure coupled with unsound
practices of some market participants such as bogus
stock scandal of PT. Dharma Samudra Fishing Industries
Tbk (DSFI) and Primarindo Asia Infrastructure Tbk (BIMA)
has contributed to the worsening condition of the
capital market in Indonesia.
The deteriorating stock price index since 1998
doesn’t show any signs to full recovery to the level of
that before crisis as shown in graphic 5.4.
Bank’s stock index primarily influenced by the
performance of recap banks which are sensitive to the
political agenda of the Government, IMF
recommendations and overall condition on the capital
market. Therefore, BPPN and the Government to
Figure 5.6Government Bond By Type of Portfolio
450
400
350
300
250
200
150
100
50
0
-50
1999 2000 2001 2002
Investment Trading
Trillion Rp
2003
47
Capital Market
influence strong investors, such as Pension Fund, does
not yet show any significant result. This failure is due
to historical experience of companies default, bank’s
stock transactions and the introduction of prudential
policy to limit the pension fund to make investment in
the capital markets.
Approaching the end of the year 2002, stock prices
continue declining due to lack of foreign investors
confidence. A positive movement of the price index is
Figure 5.7Financial Sector Stock Index
250
200
150
100
50
0
800
700
600
500
400
0
300
200
100
JCSIFSSI
1997 1998 1999 2000 2001 2002
FSSI JCSI
2003
lead by soaring trading for stock of agricultural and
consumption sector, while the worst performers are in
the stock of financial, property and basic sector.
The offering price of some banks such as Bank
Niaga, Bank International Indonesia, Bank Victoria,
Bank Lippo, Bank Permata and Bank Interpacific are
very low at Rp. 50/share and even the stock price of
Bank International Indonesia and Bank Lippo remain
at below its nominal value. Those cases indicate that
market participants confidence towards banks has yet
recover.
A host of internal problems of the capital market,
uncertainty in the social political condition in the
country particularly prior to the general election in 2004
as well as uncertain of corporate divestment of
government stake in recap banks has caused delay in
the recovery of financial sector. Therefore, it is
necessary to established joint commitment of the
government, stock exchange authority and market
players to revive the condition of the national capital
market.
48
Chapter 5
Box 7 :
RISKS IN PAYMENT SYSTEMS
Credit Risk
This risk that the counter-party failed to comply with
its obligation in full value either at the time of the
due date or thereafter.
Liquidity Risk
This risk that taken place in case the counter party
can not meet its obligation in full value at the due
time but at some time thereafter
Operational Risk
This risk that prevailed because there is problem in
RISKS IN PAYMENT SYSTEMS
Payment system is one of the important aspects
in the stability of the financial system.
Interactions of variety of risks faced by parties through
various form of the payment system prevailing in
Indonesia might result risks to the other parties and
thereby result in systematic risk as well.
Therefore, it will be fair to say that if Bank
Indonesia gives specific attention to monitor the
development and its impact to the stability of financial
system in Indonesia.
Monitoring the stability of the payment system is
one of prerequisite for Bank Indonesia to stabilize the
price and finance system in accordance with the
PAYMENT SYSTEMS IN INDONESIA6
condition as therein stipulated in Law No. 23 year 1999,
especially article 8 on main duty of Bank Indonesia and
article 15 on the implementation of payment system.
Basically, payment system can be properly run if
it meets the following criteria:
- Mutual agreement on the implementation standard
of computation and method of collecting the
payment and liabilities between the parties
participating in the payment system.
- All parties in the payment system agree to mutually
execute settlement for the collection and
liabilities.
- There is regulation on the payment system
mutually agreed.
hardware or software, human error or other causes,
will cause damage or a system fails to function and
therefore causing the increase of financial exposure
and possibility of losses.
Legal Risk
This risk that the occurrence of legal interference or
law uncertainty will cause payment system or
members facing unexpected financial exposures and
possible loss.
Source : Payment System Oversight, Bank of England
CHAPTER
49
Capital Market
Payment system prevailing in Indonesia covers
cash payment and non-cash payment. Settlement of
the transaction between banks is made through the
system of exchanging inter-bank notes (clearing) and
Real Time Gross Settlement (RTGS), which will be
described below.
Clearing System
This payment system has been conducted by BI
since the establishment of BI. The instruments cover
using check, clearing account, debit or credit notes.
Members in this payment system include all foreign
exchange banks operating in Indonesia.
Clearing centers are located in all branch offices
of BI and in a number of cities executed by some state
owned banks. The settlement is made separately for
every clearing center by end of the day. BI may provide
Short Term Loan Facility (FPJP) for banks subject to
availability of liquid collateral and maximum 90 days.
During 2002, average volume of daily clearing
reaches Rp. 6.3 trillion and 72.979 units of clearing
accounts. The development of transactions and total
amount of notes in the daily clearing transactions are
depicted in Graphic 6.1.
Real Time Gross Settlement (RTGS)
RTGS is an automatic settlement between parties
executed by BI. Time lag in the delivery of the payment
institutions and receipt can be avoided. RTGS
settlement has been implemented in 27 cities
throughout Indonesia at the average daily transaction
at the nominal of Rp. 55.7 trillion during 2002. The
development of the breakdown of transaction and
volume in RTGS is presented in Graphic 6.1.
There is also other settlement method which is
executed outside Bank Indonesia through ATM, credit
card, debit card etc, but the portion is still relatively
small in comparison to that executed through clearing
transaction and RTGS.
ROLE OF PAYMENT SYSTEMS IN THE STABILITY
OF FINANCIAL SYSTEM
Financial problem faced by a member in payment
system may spark the problem to other member through
the following reasons:
- Delay in settlement of collection through clearing
will cause problem of liquidity of the beneficiary,
so that it might also result in liquidity risk of the
beneficiary.
- The default of the party in the clearing will cause
credit risk to the beneficiary bank.
- Delay or failure of payment by one a member will
create domino effect. For other member the
condition might cause systematic risk moreover
the stability of financial system.
- Technology malfunction will cause delay of all
Figure 6.1Development of RTGS, clearing and
Non-cash transaction
Non Cash Clearing RTGS
Trillion Rp
Oct Nov Dec DecJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov
2000 2001 2002
0
250
500
750
1,000
1,250
1,500
1,750
2,000
Payment Systems in Indonesia
50
Chapter 5
process of payment system (operational risk), as
information technology has become important
sources in the efficiency of the payment system.
Although its probability remains relatively low, but
the might cause significant loss to the party of the
clearing and user of the finance services. This
condition may result in instability of financial
system.
Payment Systems Oversight
Payment System Oversight is meant to ensure the
efficiency of the payment system in future. The potency
of the problem may then be identified at early stage
and put under control so that its negative impact can
be minimized. Besides, the existing weakness can be
improved for better development ahead. Focus of
oversight is directed more to the capability of the
system to process transaction, procedure of execution,
legal framework, and contingency plan in case problem
arises in the system and last resort in case the condition
is not normal. Oversight is implemented to evaluate
and anticipate in case there is any possibility for risk
to arise in the payment system and how to avoid or
minimize its negative impact in the future.
Risks in Clearing System
Clearing system result is estimated at the end of
day, in which parties being short in the clearing must
be able to cover obligation until the next following
business day. Liquidity risk arises when collection to
the paying bank is not yet paid until the following
business day. This case will cause the lost of opportunity
for the beneficiary bank to have an overnight
investments. On the other hand, the beneficiary bank
in the clearing also has to face credit risk which is the
probability of default arises from the paying bank.
Liquidity risk and credit risk can be reduced in
case there is failure to settle. Failure to settle can be
implemented properly since there is mutual agreement
between parties to be responsible to make settlement.
This method can be executed through a number of ways
which have been mutually agreed upon such as pooling
funds or loss sharing.
Operational risk in clearing system might take place
in case the existing system fails to function properly and
therefore, clearing cannot be proceeding. Disturbance
in clearing system might arise due to some factors among
other include power supply, overload of transaction and
the other technical problem. Until today, there is no the
technical problem which arises and disturbs the process
of the clearing system. Besides, there is contingency plan
if there is some technical problem, so that operational
risk in the clearing system in Indonesia can be regarded
as being low. On the other hand, the total amount of
transactions in the clearing has also been decreasing as
the result of the application of RTGS.
Legal risk in the clearing system might arise due
to unclear mechanism in settlements so that it will be
difficult to make the collection. Clearing system in
Indonesia doesn’t require collateral so participants will
be in the relatively weak position to be able to make
the collection to paying bank.
Risks in RTGS
Liquidity risk in RTGS might take place in case
the process of collection transmission to the other bank
cannot be executed and to some technical problem and/
or liquidity problem in other banks.
Chapter 6
51
Capital Market
In the course of RTGS implementation in Indonesia,
there is no case of gridlock as the volume of RTGS itself
remains low. Besides, RTGS members have enough
liquidity reflected in significant amount at SBI. The
liquidity risk may be very low to disturb stability of
financial system. The rise of recapitalization bond
trading may require larger liquidity in RTGS. The intra-
day facilities provided by Bank Indonesia will play
significant role to avoid gridlock in RTGS.
There is no credit risk in RTGS, as the mechanism
of collection to the other banks will be effective if the
balance is adequate and the system runs well. Legal
risk in this RTGS is relevant taking into consideration
that there is no credit exposure.
Operational risk in RTGS might take place in case
there is trouble in the transmission system, so that all
transactions in RTGS cannot be executed. The main
cause of the trouble is technical problem in RTGS
system or volume of transactions which does not
match with the capacity of the existing system.
Higher gridlock or the deadlocks of the system at
are the nature of operational risk. In this respect,
the comparison between the trend of transaction in
RTGS and the capacity of the system in RTGS must
be considered to be able to measure whether the
actual capacity at present is adequate to
accommodate the development of transactions in
RTGS in future. In Indonesia case, the existing
capacity of the system is still relatively big enough
to cover the future development of RTGS. Therefore,
the capacity of RTGS is regarded to remain under
control in future years.
Payment Systems in Indonesia
52
Chapter 5
Boks 8 :
FAILURE TO SETTLE SCHEME
In relation with Bank Indonesia’s role in clearing,
BI may face credit risk exposure to the banks who suffer
liquidity mismatch in fails to set-off their obligation in
clearing due to short term liquidity mismatch facilities.
Although it is only an overnight, the credit risk
is relatively high. To minimize the risk, Bank Indonesia
conducts in Failure-to-Settle (FTS) mechanism adopted
from Core Principles for Systematically Important
Payment System (CPSIPS). FTS is a mechanism in which
one or more parties in clearing fail to meet their
obligation to make settlement at a given time. Clearing
system has to settle the obligation of illiquid borne by
using funds collected from the clearing members. By
doing so, all risk which might arise resulting from
clearing shall be fully become the responsibility of
the clearing participants. Bank Indonesia is kept
separate from the arising risk.
The CPSIPS obliges every multilateral clearing
system to make settlement at the given time and to
bail-out on the failure of settlement of one of the
parties in the clearing. According to the standard
CPSIPS, the minimum fund for bailing out from the
clearing participants is based on the largest net debit
experienced by the claim participant (or as the balance
of shortfall). The value of bail out is provided in terms
of securities, cash or fund from loss sharing
mechanism.
In accordance with the international practices,
there are a number of alternatives of FTS method
which can also be applied in Indonesia:
1. Defaulter Pay
In this method, failure to settle shall fully
become the responsibility of the defaulter. This method
seems to be more preferable and does not create any
burden the other clearing participants. The weakness
is that in case in the failure of the settlement the
amount is really big enough, while the capability of
the defaulter is limited, so that the defaulter is not
able to cover all of his liabilities and therefore the
payment system fails to become effective.
Consequently, the system needs adequate collateral
provided by participants in the system.
2. Survivor Pay
In this method, failure to settle shall become the
responsibility the whole member. This method is slightly
lower collateral than defaulter pay and probability of
failure of settlement must also be smaller. However,
there must be strong decision to determine portion
(share) of the collateral of each of the party in a fair
manner.
3. Combination of Survivor and Defaulter Pay
In this method, failure to settle must first become
the responsibility of defaulter up to the maximum limit
of the liability of his liquidity. In case the amount is
inadequate it must become mutual responsibility by
the survivors. Under this method, probability of failure
of settlement is lower so that collateral will also be
smaller than that in defaulter pay method.
There are a number of alternatives for fund
sources of FTS:
1. Cash deposit. Members of clearing must deposit a
total sum of fund in Bank Indonesia. The benefit of
applying this approach is to accelerate settlement.
But bank might be doubtful whether the fund
deposited in Bank Indonesia shall become idle.
Chapter 6
53
Capital Market
2. Pool of Collateral. Members of clearing must
place collateral in the form of SBI, Government
Bond or other Government Commercial Notes
which can be used to cover the shortfall of
fund.
3. Loss Sharing. Members of clearing have no
placement of collateral to system. In case there
is a default, the sharing of loss will be based on
the agreed mechanism. System operator will not
responsible for the payment.
4. Pool of Fund. Members of clearing are required
to deposit a certain amount with the minimum
funds is equal by the maximum net debit
extended by a member. The fund may be
managed by the system.
Member of
clearing must
deposit a total
sum of fund in
clearing
system (BI).
Member of
clearing must
deposit
collateral in the
form of SBI/OP
which can be
used to cover
shortfall of fund.
In case there is
default,
member must
be responsible in
accordance with
proportion and
mechanism
which have been
mutually agreed
Member must
collect deposit
money equal to
the largest net
debit extended
by the member.
Cash Deposit Collateral Pool Loss Sharing Pool of Fund
FTS Fund Sources
Payment Systems in Indonesia
54
Chapter 5
CONCLUSION7
1. Financial stability during 2002 can be maintained
in general. However, there are some issues needs
special attention which may spark instability in
the future. Credit risk of banking industry remain
high as was indicated by:
- The aggregate non performing loans (NPLs)
gross of banking industry remain high with
sluggish decline trend, while some banks still
has not met the indicative target NPL (net)
of below 5%.
- There is a tendency of increasing NPLs mainly
due to the increased number of
unrestructured loans bought by banks from
IBRA. Restructured loans will become NPL
since there is unsuccessful recovery in real
sector reflected reflected in uncomplete
corporate restructuring in IBRA and closure
of a number of foreign companies.
- Risk of political instability due to election in
year 2004 may spark speculation in USD which
will create pressure on USD/IDR exchange
rate. This condition may affect commercial
bank liquidity as impact of the volatility of
interest rates.
2. Banking industry has adequate capital with
aggregate capital adequacy ratio (CAR) far above
8%. However, it is necessary to anticipate the
decrease of capitali which is caused by additional
provision required to cover the increase of NPL and
additional capital needed due to implementation
of CAR incorporating market risk in 2003.
3. Currently, there is no formal policy and mechanism
of crisis resolution. This makes crisis resolution
very difficult and could led to worsen the crisis if
it occurs. Bank Indonesia has developed financial
system stability blue-print including policy and
mechanism of crisis resolution. The frame work in
under discussion with related institutions before
it proposed to the parliament to be formally stated
in the Law.
CHAPTER
55
Redesigning Indonesia’s Crisis Management
A r t i c l e s
1. Redesigning Indonesia’s Crisis Management –
S. Batunanggar
2. MARKET RISK IN INDONESIA BANKS –
Wimboh Santoso & Enrico Hariantoro
3. An Empirical Analysis of Credit Migration
In Indonesian Banking –
Dadang Muljawan
4. NEW BASEL CAPITAL ACCORD : Its likely impacts
on the Indonesian banking industry –
Indra Gunawan, Bambang Arianto, G.A. Indira & Imansyah
56
Articles
Redesigning Indonesia’s Crisis Management:
Deposit Insurance and Lender of Last Resort1
A b s t r a ct
There are some fundamental problems in current crisis management (resolution) framework: (i)
the absence of comprehensive and clear policy; (ii) weaknesses in government’s guarantee program
(blanket guarantee) which creates moral hazard and may led to future financial crises; and (iii)
unclear function of Bank of Indonesia as the lender of the last resort in a systemic crisis situation.
Two key steps are suggested to improve the crisis resolution in Indonesia: (i) a gradual replacement
the current blanket guarantee with an explicit and limited deposit insurance scheme; and (ii) a
more well-defined lender of last resort which is more transparent both in normal times and
during systemic crises. A more transparent LLR policy will function as an effective tools for crisis
management, provides a clearer accountability that may increase central bank’s credibility, reduce
politic interference, reduce moral hazard and encourage market discipline which will eventually
improve financial stability.
JEL classification: F34, G18, G21, G28, E44
Keywords: financial crises, banking crisis, crisis management, financial safety nets, lender of
last resort, government guarantee, deposit insurance.
S. Batunanggar2
1 Condensed version of paper entitled ‘Indonesia’s Banking Crisis Resolution: Lessons and the Way Forward’ prepared as part of the financial stability
research project at the Centre for Central Banking Studies (CCBS), Bank of England and presented at the Banking Crisis Resolution Conference, CCBS,
Bank of England, London, 9 December 2002.
2 Senior Bank Researcher at Bank Indonesia. The views expressed in this paper are those of the author and do not necessarily reflect the views of Bank
Indonesia. The author would like to thank Peter Sinclair (Former Director CCBS, Bank of England), Glenn Hoggarth (Bank of England), and colleagues at
Bank Indonesia for comments on earlier draft. All errors are those of the author. E-mail address: [email protected]
57
Redesigning Indonesia’s Crisis Management
INTRODUCTION
The East Asian financial crisis stands out as one of the major crisis of the 20th century. After
enjoying marked economic growth for three decades, Indonesia, together with Thailand and South
Korea, experienced an extraordinarily turbulent “twin crisis” – a currency crisis and a banking crisis.
The crisis resolution suffered from two main problems: (i) a lack of understanding from the IMF and on
the part of the authorities of the crisis which resulted in inappropriate strategies both at the macro and
micro level; and (ii) a lack of government commitment to take consistent and objective measures.
The impact of the crisis has been devastating. Indonesia has suffered particularly badly. There has
been a deep and prolonged recession and the fiscal costs of crisis resolution are so far over 50% of
annual GDP. During the last quarter of the century, only the Argentina crisis in the early 1980s was more
costly to the budget. Although the crisis is now over, but Indonesia will experience its effects for years
to come.
There is now a large literature on the causes of the Asian crisis3 . There are two main polar views of
the causes of the crisis. The first view argues that the main cause of the crisis were weak economic
fundamentals and policy inconsistencies (Krugman (1998), Mishkin (1999). The second view believes
that the root of the crisis was pure contagion and market irrationality ((Radelet and Sachs (1998);
Furman and Stiglitz (1998) and Stiglitz (1999, 2002)). While some other commentators such as Corsetti,
Pesenti and Roubini (1998) and Djiwandono (1999) took the middle ground arguing that both contagion
and poor economic fundamentals caused the Asian crisis. Financial crises do not occur only in the
presence of weak fundamentals, but weak fundamentals can trigger bank run psychology and this in
turn can have disproportionately bad effects on the real economy.
Safeguarding financial stability is a core function of a modern central bank, no less important than
maintaining monetary stability (Sinclair, 2001). Financial stability relies on five interrelated elements:
(i) stable macro-economic environment; (ii) well-managed financial institutions; (iii) efficient financial
markets; (iv) a sound framework of prudential supervision; and (v) a safe and robust payments system
(MacFarlane, 1999).
Fragility in financial institutions particularly in banks, is one source of instability (Crockett, 1997,
2001). Therefore, banking crises should be either prevented or resolved in order to avoid disrupting the
payments system and the flows of credit to the economy. Wherein, crisis prevention involves measures
taken to avoid banking problems occurring, crisis management focuses on how the authorities deal with
3 Note, however, the literature predicting the likelihood of a crisis in Indonesia beforehand is much more parsimonious.
58
Articles
a crisis once it materialises. Facing with a banking crisis, the authorities confronts a trade-off between
maintaining financial stability today – through offering protection to failing banks – and increasing
financial instability in the future through increasing moral hazard (Hoggarth et.al, 2002). In resolving
banking crisis, the authorities should try to minimise the fiscal costs and any moral hazard that may be
induced.
This paper will discusses banking crisis resolution in Indonesia in particular safety nets issues including
blanket guarantees and Bank Indonesia’s liquidity support. Section 2 provides overview of Indonesia’s
banking system. Section 3 discuss Indonesia’s current safety nets. Section 4 proposes some actions that
should be taken to improve financial stability in Indonesia, particularly in the realm of lender of last
resort and deposit insurance. Section 5 concludes.
OVERVIEW OF INDONESIA’S BANKING SYSTEM
The banking system, particularly the commercial banks, dominate Indonesia’s financial system and
play a key role in the financial intermediation process4 . Indonesia’s banking system has two main features.
First, it is a highly concentrated with the 15 largest commercial banks accounting for 75% of total assets
(see Table /Appendix). Second, there is a high degree of government control of over 80% of the banking
system’s assets. Domestic private banks and foreign banks account for only 11% and 8% respectively of
total assets. The latter was a result of the bank restructuring programme in 1999–2000.
Following the bank restructuring programme, the banking system performance improved as indicated
by an increase in profit /and capital as well as deposits. However, the banking system is still vulnerable
and its capacity to support economic growth remains constrained by poor capitalisation and continued
high credit risk in the economy. The main problems facing the banking industry are: (i) a high credit risk
indicated by the continued high level of non performing loans; (ii) slow progress in loan and corporate
restructuring, and; (iii) slow credit growth due to unfavourable economic conditions5 .
4 Indonesia’s financial system comprises banks and non-bank financial institutions (security firms, insurance companies, pension funds and pawn brokers).
The banking system consist of commercial banks with shares up to 99% in terms of total asset to the banking system and rural banks. Commercial banks
are the dominant players consisting of almost 80% of the total assets in the financial system.
5 Empirical study by Agung et al. (2001) found strong evidence of the existence of a credit crunch in Indonesia.
59
Redesigning Indonesia’s Crisis Management
At the heart of the financial crisis in Indonesia in 1997 – 1998 lay a slump in economic growth and
investment. Since 1998, the economy has grown slowly. However, the outlook for the Indonesia economy
is still uncertain. Nasution (2002) observed that the failure of the investment to rebound significantly
suggest that the crisis is having a long-term adverse effect. Moreover, there are critical issues related to
Indonesia’s financial stability including heavy pressures on the government budget deficit, public sector
debts with payment of principal and interests US$9 billion per year, and difficulties in settling private
sector debts.
Figure 1. Asset Quality and
Capital Adequacy
Figure 2.
Profitability Indicators
Table 1. Performance of the Indonesian Banking System: 1997 - 2001 (in trillion rupiah)
Financial Indicators Dec 1997 Dec 1998 Dec 1999 Dec 2000 Dec 2001
Total Assets 715.2 895.5 1,006.7 1,030.5 1,099.7
Credits 378.1 487.4 225.1 269.0 307.6
Deposits 357.6 573.5 625.6 699.1 797.4
Equity 46.7 -98.5 -41.2 52.3 62.3
Capital Adequacy Ratio (CAR) % 9.19 -15.7 -8.12 12.34 19.28
Non Performing Loans (Gross) % 32.18 48.6 32.8 18.8 12.1
Return on Equity-ROE (%) 19.6 -437.23 -110.8 9.65 13.6
Loan to Deposit Ratio-LDR (%) 105.7 85.0 36.0 38.5 38.6
Number of Commercial Banks 222 208 173 164 159
A number of issues need to be resolved in order to restore the banking system to full health, so that
it can perform its vital role as financial intermediary to the economy. Bank loans restructuring and
corporate restructuring should be further accelerated in order to foster the internal growth of the
NPLs CAR Forex Rate
Percent Rp/USD
-80
-60
-40
-20
0
20
40
60
0
2,500
5,000
7,500
10,000
12,500
15,000
17,500
Dec Jun Dec Jun Dec Jun Dec Jun Dec Jun Dec Jun
1996 1997 1998 1999 2000 2001 2002
ROA
Income to Expense Ratio
ROE
ROA, IER (%) ROE (%)
-20
-15
-10
-5
0
5
-500
-400
-300
-200
-100
0
100
Dec Dec Dec Dec Dec Dec Dec Dec Dec Jun
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
60
Articles
banking industry. In addition, the institutional and organisational capabilities should be enhanced toward
the international standards including risk management, good governance and banking supervision.
SAFETY NETS FOR CRISIS RESOLUTION
The Blanket Guarantee Programme
Before the 1997 crisis, none of the East Asian crisis countries except the Philippines, which was
least affected by the crisis, had an explicit deposit insurance scheme. Bank Indonesia provided both
liquidity and capital support to problem banks on an ad-hoc basis and in non transparent ways6 . The
support was also not based on any pre-existing formal guarantee mechanism but rather on a belief that
some of the banks that needed support were too big to fail or the failure of a bank could cause contagion.
A limited deposit guarantee in Indonesia was first applied when the authorities closed down Bank
Summa at the beginning of the 1990s which was considered unsuccessful7 . After then, there were no
bank closures until the authorities closed down 16 banks in November 1997 and introduced a limited
guarantee. However, this failed to prevent systemic bank runs.
To restore domestic and international confidence in the economy and the financial system, the
government signed the second agreement with the IMF on 15 January 1998. However, market perceptions
and reactions to the government commitment and capacity to resolve the crisis were still negative.
There was a huge amount of capital flight of around $600 million to $700 million per day. On 22 January,
the rupiah plummeted to a record low of Rp16.500.
Figure 3. Foreign Debt and
International Reserve
Figure 4.
IBRA’s Restructured Loans
6 However, this was done primarily for domestic rather than foreign banks.
7 The plan for establishing a deposit insurance scheme has been discussed quite intensively since the early 1990s. However, the authorities declined the
proposal because they considered that it would create moral hazard.
Foreign debt (mio USD) Forex rate (IDR/USD)
Int`l reserve (mio USD)
Forex, Reserve Foreign Debt
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
90,000
100,000
110,000
120,000
130,000
140,000
150,000
160,000
1996
QIV QII QIV QII QIV QII QIV QII QIV QII QIV QII
1997 1998 1999 2000 2001 2002
Bad Loans Transf'rd to IBRA Restructured bad loans
Ratio
Trillion Rp Ratio
0
50
100
150
200
250
300
350
0%
5%
10%
15%
20%
25%
30%
Jun Sep Dec Mar Jun Sep Dec Mar Jun Jul
2 0 0 0 2 0 0 1 2 0 0 2
61
Redesigning Indonesia’s Crisis Management
To prevent a further slide and to maintain public confidence in the banking system on 27 January,
the government issued a blanket guarantee. It covered all commercial banks’ liabilities (rupiah and
foreign currency), including both depositors and creditors. It was an interim measure pending the
establishment of the Deposit Insurance Agency8 . Initially, the administration of the blanket guarantee
was a joint task between Bank Indonesia and IBRA. From June 2000 it has been the responsibility of IBRA
alone9 .
The Indonesian case suggests that a very limited deposit insurance scheme was not effective in
preventing bank runs during the 1997 crisis. Deposits denominated of more than Rp20 millions – the
uninsured component – accounted for about 80% of total deposits. Therefore, if a blanket guarantee had
been introduced earlier at the outset of the crisis, the systemic runs might have been reduced.
8 Initially it was to be retained for a minimum of two years, with a provision for an automatic six months extension in the absence of an announcement
of termination of the scheme.
9 BI retains the role of administering the guarantee scheme to trade finance, inter-bank debt exchange and rural banks.
Figure 5.
Interbank Call Money: 1997 - 1999
Figure 6.
Interest Rates: 1997 - 1999
That said, the introduction of the blanket guarantee did not instantly stop bank runs. Although
fewer, there were still bank runs on some insolvent banks (although not on the domestic system as a
whole). This was indicated by the increase in Bank Indonesia’s Liquidity Support from Rp92 trillion in
January to Rp178 trillion in August 1998. These banks were run by depositors for some reasons. The first
possibility is a poor public perception due to unclear policies what was covered and a lack of trust that
the government would stick to their commitment. This led depositors to transfer their money from
perceived bad banks to safer banks. After three highly insolvent banks were taken over and closed down
on 21 August 1998 bank runs, and therefore Bank Indonesia’s liquidity support, decreased markedly.
Billion rupiah
Interbank Call Money Rate (%)
0
10
20
30
40
50
60
70
80
90
100
Percent
Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov
-
50,000
100,000
150,000
200,000
250,000
300,000
1 9 9 91 9 9 81 9 9 7
Percent
0
10
20
30
40
50
60
70
80
90
Inter-bank (o/n) Time deposit (3 mth) BI Certificate (28 days)
Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov
1997 1998 1999
62
Articles
Secondly, depositors anticipated that there would be more bank closures. Even though their deposits
were guaranteed fully by the government, they recognised that they would be transferred to other
banks. This would create a time lag between the claim and the payment of their deposit. To avoid this,
they withdrew their deposits from perceived weak banks and transferred to those banks thought “safer”
(state-owned and foreign banks). There was also a relatively long time lag before the implementation
of the blanket guarantee programme. In fact, IBRA, as administrator of the guarantee programme,
became operational three months after its creation.
Even though it was introduced after some delay, the implementation of a blanket guarantee appears
to have worked. After June 1998 the segmentation in the inter-bank market decreased sharply and
liquidity in the banking system increased. This enabled banks to borrow from the inter-bank market at
lower interest rates (see Figure 5). In addition, there were no significant bank runs despite the change
in government in 1999 and some other policy reversals.
Bank Indonesia Liquidity Support
Mishkin (2001) suggests that central bank could promote recovery from financial crisis by pursuing
lender of last resort role in which it stands ready to lend freely during a financial crisis. There are many
instances of successful lender of last resort operations in industrialised countries (Mishkin, 1991).
Under its old Law of 1968, Bank Indonesia
was authorised to provide emergency loans to
banks facing critical liquidity problems10 .
However, there were no well-defined rules and
procedures on how this function was to be
performed. During the 1997 crisis, Bank Indonesia
provided liquidity support to problem banks in
order to prevent the collapse of the banking
system and to maintain the payments system. The
continuing deterioration of confidence in the
banking system coupled with political
uncertainties and social unrest had caused severe
bank runs from perceived unsound banks to sound
ones.
10 The previous BI’s Law (1967) set BI’s status is a “dependent” agency served as assistance of the Government in carrying out monetary, banking and
payment system policies. The current BI’s Law (1999) sets its the independence from political intervention.
Figure 7.
Bank Indonesia Liquidity Support
Bank Central Asia
Bank Dagang Nasional Indonesia
Bank Danamon
Bank Exim
Bank Umum Nasional
Sub Total for Five Banks
Total for the Banking
System
-
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
Aug Oct Dec Feb Apr Jun AugSep Nov Jan Mar Mei Jul Sep
1 9 9 7 1 9 9 8
Rp Billion
63
Redesigning Indonesia’s Crisis Management
As the crisis intensified, the amount of overdraft facilities increased from Rp31 trillion in December
1997 to Rp170 trillion in December 1998. Liquidity support, however, was concentrated on only a
small number of banks - 80 % of the total was provided to five banks and over 60% to four private
banks. All these banks faced huge deposit withdrawals, except Bank Exim (state-owned) which
faced huge losses on foreign exchange transactions. Bank Central Asia, owned by the Salim group
which close links to Suharto, was the largest borrower of Rp31 trillion following riots in May 1998 (see
Figure 6).
According to conventional wisdom, liquidity support should be only provided to illiquid but solvent
banks. In order to reduce the likely hood of losses the central bank should take adequate collateral.
Enoch (2001) argued that there should be restrictions against protracted use of such lending, since this
is likely to be an indicator of solvency difficulties. After the bank closures of November 1997, and given
there was no intention to close more banks in the near future, Bank Indonesia was locked into providing
liquidity. Since then, Bank Indonesia provided support to all banks without taking any collateral (by
allowing their current accounts with Bank Indonesia to go overdrawn). Instead, Bank Indonesia took
only a personal guarantee from banks’ owners that the borrowings were used to meet liquidity needs,
and their banks were in compliance with all prudential regulations. Unsurprisingly, as illustrated in
Figure 6, liquidity support increased markedly.
The large budgetary cost that this entailed created tension and distrust between Bank Indonesia
and the Government particularly over the accountability and integrity of Bank Indonesia in providing
the emergency liquidity support. Enoch (2001) observed that the main criticism of BI’s LLR practices
related to the lack of control over such lending, for example, whether lending matched a commensurate
loss of deposits. While Bank Indonesia did undertake such matching in the latter part of the crisis – in
particular in May 1998 when BCA was subject to severe withdrawals – there seems to have been less
control during the earlier period. In fact, BI placed 2 or 3 supervisors in each distressed bank to verify
the bank’s transactions. However, it was insufficient to ensure that there was no misuse of the liquidity
support by banks’ management and owners11 .
This case has led to investigations into the operation of the LLR facility and dispute between the
Government and Bank Indonesia about the amount of and who should bear the liquidity support12 . After
a prolonged negotiation there is a commitment to finalise and implement a burden-sharing agreement
11 Later, bank examination found that there were a strong indication of moral hazard indicated by dubious interbank transactions between the closed
banks.
12 An investigation was performed by the Supreme Audit Agency (BPK) reported that there were weaknesses in BI’s internal control and governance in
providing liquidity support.
64
Articles
on Bank Indonesia liquidity support between Bank Indonesia and the government13 . In fact, the decision
to provide the liquidity support to distressed banks has been discussed in the Economic and Financial
Stabilisation Council and approved by the President – ‘that Bank Indonesia could help distressed but
good banks to maintain the banking system and the payment system’14 . However, as argued by Sinclair
(2000) and Goodhart (2002), within the time scale allowed, it is often difficult, if not impossible, for
central banks to distinguish between a solvency and a liquidity problem. Therefore, the basic problem
was unclear criteria for distinguishing a good from bad banks and the absence of LLR guidelines and
procedures to ensure accountability. In addition, there was also a lack of coordination between agencies
in managing the crisis.
In addition, with a large amount of foreign-denominated debt it made difficult for Bank Indonesia
to promote recovery from a financial crisis by using expansionary monetary policy. This policy is likely to
cause domestic currency to depreciate sharply. For similar reasons, as was argued by Mishkin (2001),
LLR activities by a central bank in a emerging market countries with substantial foreign-denominated
debt, may not be as successful as in an industrialized countries. Bank Indonesia lending to the banking
system in the midst of the crisis expands domestic credit. This led to a substantial depreciation of the
rupiah which resulted in the deterioration of banks’ balance sheets which in turn made recovery from
the crisis more difficult likely. Therefore, the use of the LLR by a central bank in countries with a large
amount of foreign-denominated debt is trickier because central bank lending is now a two-edged sword
(Mishkin, 2001).
Based on its current Law of 1999, Bank Indonesia is permitted only a very limited role as lender of
last resort. Bank Indonesia can only provide limited LLR in normal times to banks (for a maximum of 90
days) that have high quality and liquid collateral; and not in exceptional circumstances. The collateral
could be securities or claims issued by the Government or other high-rated legal entities which can be
readily sold to the market. In practice, government recapitalisation bonds and Bank Indonesia Certificate
(SBIs) are the only eligible assets currently available to banks. The facility serves like a discount window,
which the central bank routinely opens at all times to handle normal day to day operational mismatches
which might be experienced by a bank. However, the facility does not constitute a LLR function typically
used to provide emergency liquidity support to the financial system during crisis periods (i.e. when
banks usually do not have high quality collateral).
13 Stated as a performance criterion in the Letter of Intent with the IMF of 31 December 2001.
14 The Council members are the Coordination Minister of the Economy, the Minister of Finance and the Governor of BI.
65
Redesigning Indonesia’s Crisis Management
THE WAY FORWARD
Redesigning the Safety Nets
Two strategic issues are proposed as part of a comprehensive crisis management strategy. First,
replace the blanket guarantee with an explicit deposit insurance scheme and, second redesign the
lender of last resort (LLR) facility.
a. Replacing the blanket guarantee with a deposit insurance scheme
There was a controversy over the adoption of a blanket guarantee. Some commentators such as
Furman and Stiglitz (1998), Stiglitz (1999,2002), Radelet and Sachs (1998) argue that the if the blanket
guarantee had been introduced earlier, before some banks had been liquidated, the damage and costs
of the crisis would have been much less.
In contrast, others criticised the blanket guarantee for being too broad. Goldstein (2000) argued
that had all bad (insolvent) banks been closed at the beginning of the crisis then even with the limited
deposit guarantee scheme in place there would not have been widespread deposit withdrawals because
the remaining banks would have been ‘good’ ones. He believed that with a blanket guarantee, the
government ended-up providing ex-post deposit insurance at a higher fiscal cost and with adverse moral
hazard effects increasing the likelihood of future banking crises. Therefore, he suggested that Indonesian
should develop an incentive-compatible deposit insurance system – along the lines of FDICIA in the
United States – which should be a permanent part of the financial infrastructure.
Honohan and Klingebiel (2000), based on sample of 40 developed and emerging market crises,
found that unlimited deposit guarantees, open-ended liquidity support, repeated recapitalisation, debtors
bail-out and regulatory forbearance significantly and sizeably increase the resolution costs. Moreover,
based on evidence from 61 countries in 1980-97, Demirguc-Kunt and Detragiache (1999), find that that
explicit deposit insurance tends to be detrimental to bank stability, the more so where bank interest
rates are deregulated and the institutional environment is weak. Similarly, Cull et al. (1999) based on a
sample of 58 countries also find that generous deposit insurance leads to financial instability in the
presence of a weak regulatory environment.
However, systemic bank runs in Indonesia at the outset of the 1997 crisis cannot be attributed
solely to the absence of a blanket guarantee. The inconsistent and non transparent bank liquidation
policies applied by the authorities and some political uncertainties during the end of Suharto’s regime
also played their part, as Lindgren et al. (1999) and Scott (2002) document. The introduction of the
blanket guarantee programme at the outset of the crisis might be necessary in order to prevent larger
66
Articles
potential economic and social costs of the systemic crisis (Lindgren et al. 1999). However, the scheme
should be replaced as soon as possible with one that is more appropriate to normal conditions and does
not create moral hazard.
Garcia (1999, 2000), based on surveys in 68 countries, identified the best practices of explicit
systems of deposit insurance principally should have good infrastructure, avoid moral hazard, avoid
adverse selection, reduce agency problems and ensure financial integrity and credibility. Based on a
study of deposit insurance systems in Asian countries, Choi (2001) argues that it is reasonable in Asia to
establish and maintain an explicit and limited deposit insurance system in order to prevent further
possible financial crisis. Pangestu and Habir (2002) suggest that Indonesia’s deposit insurance scheme
should be designed on two key aspects. First, it should provide incentive to better performing banks by
linking the annual premium payment to their risk profile. Second, it should be self-funded in order to
foster market discipline and reduce the fiscal burden.
In order to avoid a disruption to the banking system, Garcia (2000) suggests that a partial guarantee
should not be introduced ideally until: (1) the domestic and international crisis has passed; (2) the
economy has begun to recover; (3) the macro-economic environment is supportive of bank soundness;
(4) the banking system has been restructured successfully; (5) the authorities possess, and are ready to
use, strong remedial and exit policies for bank that in the future are perceived by the public to be
unsound; (6) appropriate accounting, disclosure, and legal systems are in place; (7) a strong prudential
regulatory framework is in operation; and (8) public confidence has been restored. It seems that currently
Indonesia does not meet all these requirements.
Demirguc-Kunt and Kane (2001) suggest that countries should first assess and remedy the weaknesses
of their international and supervisory environments before adopting an explicit deposit insurance system.
In line with this, Wesaratchakit (2002) reported that Thailand decided to adopt a gradual transition
from a blanket guarantee to a limited explicit deposit insurance scheme. It was considered that there
are some preconditions that should be met – particularly the stability of banking system and the economy
as a whole, effectiveness of regulation and supervision as well as public understanding – before shifting
to an explicit limited deposit insurance system.
There is an issue of how depositors will react to the introduction of the limited scheme. In January
2001, Korea replaced its blanket guarantee with a limited deposit insurance system with an insurance
limit of 50 million won per depositor per institution. There was a noticeable migration of funds from
lower rated to sounder banks. Also, large depositors actively split their deposits to several accounts in
banks and non-bank financial institutions. But there has been no bank run on the Korean financial
system as a whole.
67
Redesigning Indonesia’s Crisis Management
It is important to prepare a contingency plan before removing the blanket guarantee in order to
anticipate the worst-case scenarios such as a loss of public confidence. If such conditions occur, the
central bank may have to extend liquidity support to illiquid but solvent banks. In addition, there
should be a clear legal framework for the deposit insurance scheme. To reduce moral hazard and to
induce market discipline, the authorities should set a tough sanctions to the financial institutions and
players which are violate the rules and cause problems into banks and ensure that law enforcement are
in place.
b. Redesigning the Lender of Last Resort (LLR)
Historical experience suggests that successful lender of last resort actions have prevented panics
on numerous occasions (Bordo, 2002). Although there may well be good reasons to maintain ambiguity
over the criteria for providing liquidity assistance, He (2000) argues that properly designed lending
procedures, clearly laid-out authority and accountability, as well as disclosures rules, will promote
financial stability, reduce moral hazard, and protect the lender of last resort from undue political
pressure. There are important advantages for developing and transitional economies to follow a rule-
based approach by setting out ex ante the necessary conditions for support, while maintaining such
conditions is not sufficient for receiving support. In the same vein, Nakaso (2001) suggests that Japan’s
LLR approach has shifted from “constructive ambiguity” towards increasing policy transparency and
accountability.
While individual frameworks differ from country to country, there is a broad consensus on the key
considerations for emergency lending during normal and crisis periods (see Box 2).
In Indonesia, along the line suggested by He (2000), there should be a more clearly defined role for
Bank Indonesia in LLR. In addition, there is also a need for criteria or mechanisms for providing LLR
during systemic crises.
b.1. Lender of Last Resort in Normal Times
In normal times, LLR assistance should be based on clearly-defined rules. Transparent LLR policies
and rules can reduce the probability of self-fulfilling crises, and provide incentives for fostering market
discipline. It may also reduce political intervention and prevent any bias towards forbearance. LLR in
normal times should only be provided for solvent institutions with sufficient acceptable collateral,
while for insolvent banks stricter resolution measures should be applied such as closure. Therefore,
there should be a clear and consistent adoption of a bank exit policy. Once a deposit insurance scheme
68
Articles
has been established, the central bank role in LLR in normal time can be reduced to a minimum since
the deposit insurance company will provide bridging finance in the case where there is a delay in
closure process of a failed institution15 .
1. Have in place clearly laid out lending procedures, authority, and accountability.
2. Maintain close cooperation and exchange of information between the central bank,
the supervisory authority (if it is separate from the central bank), the deposit
insurance fund (if exist), and the ministry of finance.
3. Decision to lend to systemically important institutions at the risk of insolvency or
without sufficient, acceptable collateral should be made jointly by monetary,
supervisory, and the fiscal authority.
4. Lending to non-systemically institutions, if any, should be only to those institutions
that are deemed to be solvent and with sufficient acceptable collateral.
5. Lend speedily
6. Lend in domestic currency
7. Lend at the above average market rates
8. Maintain monetary control by engaging effective sterilization
9. Subject borrowing banks to enhanced supervisory surveillance and restrictions on
activities
10. Lend only for short-term, preferably not exceeding three to six months
11. Have a clear exit strategy
12. Emergency support operations should be disclosed when such disclosure will not be
disruptive to financial stability
13. Repayment terms may be relaxed to accommodate the implementation of a systemic
bank restructuring strategy.
14. Emergency support operation should be disclosed when such disclosure will not be
disruptive to financial stability.
Box 2.
Key Considerations of Emergency Lending
Sumber: Dong He (2000), Dong He (2000) ‘Emergency Liquidity Support Facilities’, IMF Working Paper No. 00/79.
Normal Systemic
Times CrisisKey Considerations of Emergency Lending
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes No
Yes No
Yes No
No Yes
Yes Yes
15 See Nakaso (2001) for a discussion on the Japanese LLR model.
69
Redesigning Indonesia’s Crisis Management
b.2. LLR in Exceptional Circumstances
In systemic crises, LLR should be an integral part of a well-designed crisis management strategy.
There should be a systemic risk exception in providing LLR to the banking system. Repayment terms may
be relaxed to support the implementation of a systemic bank restructuring programme. In systemic
crises the disclosure of the operation of LLR may become an important tool of crisis management. The
criteria of a systemic crisis will depend on the particular circumstances, thus, it is difficult to clearly
state this beforehand in a law. However, the regulations on the LLR facility should clearly set the
guiding principles and specific criteria of a systemic crisis and or a potential bank failure leading to
systemic crisis. To ensure an effective decision making process and accountability, there should be a
clear institutional framework and LLR procedures. Bank Indonesia should be responsible for analysing
the systemic threats to financial stability while the final decision on systemic crises resolution should be
made jointly by Bank Indonesia and the Ministry of Finance. To ensure accountability, an appropriate
documentation audit trail should be maintained.
CONCLUSION
Indonesia’s experience shows that resolving banking crises can be costly, painful and complicated.
However, it is also brought out some important policy lessons. Unlike the other East Asian countries, the
twin currency and banking crisis in Indonesia resulted in a political crisis, which in turn made the
financial crisis more difficult to manage. Social unrest and the volatile political situation limited the
policy options available to policy makers. A lack of a comprehensive strategy both at the micro and
macro level, a lack of commitment to take tough measures together with high level political intervention
make for less effective and costlier resolution of banking crises. To be effective, the resolution process
should be carried out objectively, transparently, and consistently in order to restore the vitality of the
financial system and the economy.
Two key steps are suggested to improve the resolution mechanism in Indonesia: (i) a gradual
replacement the current blanket guarantee with an explicit and limited deposit insurance scheme; and
(ii) a more well-defined and transparent LLR both in normal times and during systemic crises. Deposit
insurance and LLR can be important tools for crisis management, but they are not sufficient to prevent
banking crises. They should be used along with other financial stability tools such as market discipline
and prudential banking supervision.
To strengthen Indonesia’s financial system stability the current long listed of programmes –
enhancement the effectiveness of bank supervision, bank and corporate restructuring, enhancement of
70
Articles
market discipline, improvement of legal and judicial systems, reduction of the share of government in
the banking system, and pursuit of price stability – need to be further accelerated. Central to the first
issue is a call for an independent and competent bank supervision which is able to take prompt corrective
actions and resolve failed institutions at least costs without damaging financial stability. To achieve
this, the organization of bank supervision should be developed consistently to meet the international
standards.
Two other important policy options could be considered. First, restrictions on foreign-denominated
debt both for financial institutions and corporate in order to reduce the vulnerability of the economy to
external shock. Second, limiting too big to fail in order to minimize moral hazard incentives for an
excessive risks taking. This could be done by reducing the government control over banks (or privatization)
and intensifying supervision on systemically important banks. In addition, surveillance should also be
improved to identify, assess and monitor related risks to the financial stability. By no less important is to
promote cooperation between relevant institutions, domestic and international, to manage and to
minimise the costs of crisis.
Cross-country experiences show that banking crises are difficult to predict and, thus, to avoid.
History also tells us that the financial crises are often repeated. That said, the crisis prevention is vital
in order to enhance the resilience of the financial system. It is essential to have a well-devised framework
and strategy, but more important is to make these happen. With a strong commitment and enlightened
leadership from senior policy makers, hopefully we will see a healthier Indonesia’s financial system
which is able to foster higher growth and more stable economy in the future.
REFERENCES
Abrams, Richard K., and Michael W. Taylor (2000) ‘Issues in the Unification of Financial Sector
Supervision, IMF Working Paper, No.00/213, December.
Agenor, Pierre-Richard and Joshua Aizenman (1999) ‘Financial Sector Inefficiencies and Coordination
Failures: Implications for Crisis Management’, World Bank Policy Research Paper No.2185, September.
Agung, Juda et al. (2000) Credit Crunch in Indonesia in the Aftermath of the Crisis: Facts, Causes
and Policy Implications, Bank Indonesia.
71
Redesigning Indonesia’s Crisis Management
Aziz, Iwan J. (1999) ‘Institutional Factors and Politics in the Indonesian Crisis’, paper prepared for
a conference on “The Politics of Reform in Asia”, organised by The Carnegie Endowment for International
Peace, Bangkok, Thailand, June 3-4.
Bank Indonesia, various publications.
Batunanggar, Sukarela (1996) ‘Strategic Management in Action: The Case of Bank Indonesia’,
unpublished MBA Dissertation, University of Nottingham, England.
Boorman, Jack, et al. (2000) Managing Financial Crisis: The Experience in East Asia, Carnegie-
Rochester Conference Series on Public Policy No. 53.
Brealey, Richard et al. (2001) Financial Stability and Central Banks: A Global Perspective, Routledge
and CCBS, Bank of England.
Calomiris, Charles W. (2000) ‘The IMF’s Imprudent Role as Lender of Last Resort’, the Cato Journal
Vol. 17 No. 3.
Caprio, Gerard, Jr., and Daniela Klingebiel (1996) ‘Bank Insolvencies, Cross-country Experience’,
World Bank Policy Research Paper No.1620, July.
Caprio, Gerard, Jr., and Patrick Honohan (2002) ‘Banking Policy and Macroeconomic Stability: An
Exploration’, World Bank Policy Research Paper No.2856, June.
Choi, Jang-Bong (1999) ‘Structuring a Deposit Insurance System from Asian Perspective’, in Rising
to the Challenge in Asia: A Study of Financial Markets, Vol.6, Asian Development Bank.
Claessens, S., and T. Glaessner (1997) ‘Are Financial Sector Weaknesses Undermining the East
Asian Miracle?’, The World Bank, Direction for Development, The International Bank for Reconstruction.
Claessens, S. (1998) Systemic Bank and Corporate Restructuring: Experience and Lessons for East
Asia (Washington: World Bank).
Claessens, S., S. Djankov and L. Klapper (1999) ‘Resolution of Corporate Distress in East: Asia’,
World Bank Policy Research Paper.
Cole David C. and Betty F. Slade (1996) Building a Modern Financial System: The Indonesian
Experience, Cambridge University Press.
_____ (1998) ‘Why has Indonesia’s Financial Crisis been so Bad?, Bulletin of Indonesian Economic
Studies Vol.34 No.2, August pp. 61-6.
72
Articles
Cipriani, Marco and Antonio Guarino (2002) ‘Social Learning and Financial Crisis, BIS CGFS Conference
Vol. No.2 Part 6, October.
Cosetti, G., P. Pesenti and N. Roubini (1999) ‘What Caused the Asian Currency and Financial Crisis?,
Japan and the World Economy No. 11, pp. 305–373.
Crockett, Andrew (1997) ‘Why is Financial Stability a Goal of Public Policy?’, paper presented at
Maintaining Financial Stability in a Global Economy Symposium, the Federal Reserve Bank of Kansas
City, August 28-30.
_____ (1997) ‘The Theory and Practice of Financial Stability’ Essays in International Finance’
International Finance Section, Department of Economics, Princeton University, New Jersey, March.
Cull, Robert, L.W. Senbet, and M. Sorge (2001) ‘Deposit Insurance and Financial Development’,
World Bank Policy Research Paper No. 2682, September.
De Luna-Martinez (2000) ‘Management and Resolution of Banking Crises: Lessons from the Republic
of Korea and Mexico’, World Bank Discussion Paper No. 43, March.
Demiguc-Kunt, Asli, E. Detragiache (1999) ‘Does Deposit Insurance Increase Banking System Stability’,
World Bank Policy Research Paper No.2247, November.
Demirguc–Kunt, Asli, and Harry Guizinga (1999) ‘Market Discipline and Financial Safety Net Design’,
World Bank Policy Research Paper No.2183, September.
Demirguc–Kunt, Asli, Enrica Detragiache, and Poonam Gupta (2000) ‘Inside the Crisis: An Empirical
Analysis of Banking Systems in Distress’, World Bank Policy Research Paper No. 2185, August.
Djiwandono, J. Soedradjat (1997) ‘The Banking Sector in Emerging Markets: The Case of Indonesia’
in Banking Soundness and Monetary Policy”, edited by Charles Enoch and John Green , Washington:
IMF).
_____ (1998) ‘Banking System Soundness and Macroeconomic Management: The Recent Indonesian
Experience’, presented at symposium” ‘Towards the Restoration of Sound Banking Systems in Japan
and its Implication’, July, Kobe.
_____ (2000) ‘Bank Indonesia and the Recent Crisis’, Bulletin of Indonesian Economic Studies,
Vol.36 No.1, April.
Dong, He (2000) ‘Emergency Liquidity Support Facilities’, IMF Working Paper No. 00/79, April.
Enoch, Charles (2000) ‘Interventions in Banks during Banking Crisis, The Experience of Indonesia’,
73
Redesigning Indonesia’s Crisis Management
IMF Policy Discussion Paper No. 00/2, March.
Enoch, Charles et al. (2001) ‘Indonesia: Anatomy of a Banking Crisis, Two Years Living Dangerously
1997–1999’, IMF Working Paper No. 01/52, May.
Enoch, Charles, D. Marston and M. Taylor (Eds.) (2002) Building Strong Banks through Surveillance
and Resolutio, International Monetary Fund, Washington.
Freixas, Xavier, C. Gianini, G. Hoggarth and F. Soussa (2002) ‘Lender of Last Resort: A Review of the
Literature’ in Goodhart, Charles and Gerhard Illing (eds.) (2002) Financial Crises, Contagion, and the
Lender of Last Resort: A Reader, Oxford University Press.
Furman, J. and J.E. Stiglitz (1998) ‘Economic Crisis: Evidence and Insights from East Asia”, Brooking
Papers on Economic Activity.
Garcia, Gillian G.H. (1999) ‘Deposit Insurance: A Survey of Actual and Best Practice,’ IMF Working
Paper No.99/54, April.
Garcia, Gillian G.H. (2000) ‘Deposit Insurance and Crisis Management’, IMF Working Paper No.00/
57, March.
Goodhart, Charles (2001) ‘The Organisational Structure of Banking Supervision’ in Brealey, Richard
et al., Financial Stability and Central Banks : A Global Perspective, Routledge and CCBS, Bank of
England.
Goodhart, Charles and Gerhard Illing (2002), Financial Crises, Contagion, and the Lender of Last
Resort: A Reader, Oxford University Press.
Goldstein, Morris (2000) ‘IMF Structural Programs’, paper prepared for NBER Conference on
“Economic and Financial Crises in Emerging Market Economies”, Vermont, October.
Hal, Hill (1999) The Indonesian Economy in Crisis: Causes, Consequences and Lessons, Institute of
Southeast Asian Studies, Singapore.
Hoggarth, Glenn and F. Soussa (2001) ‘Crisis Management, Lender of Last Resort and the Changing
Nature of the Banking Industry’, in Brealey, Richard et al., Financial Stability and Central Banks : A
Global Perspective, Routledge and CCBS, Bank of England.
Hoggarth, Glenn, Peter Sinclair, and Jack Reidhill (2002) ‘Resolution of Bank Failures: Theory and
Evidence’ paper presented at the Crisis Resolution Conference, Bank of England, December, 9.
Hoggarth, Glenn, Ricardo Reis and Victoria Saporta (2001) ‘Costs of Banking System Instability:
74
Articles
Some Empirical Evidence’, Journal of Banking and Finance No.26 pp. 825–855.
Honohan, Patrick and Daniella Klingebiel (2000) ‘Controlling Fiscal Costs of Banking Crisis’, Policy
Research Paper No.2441, The World Bank, September.
Honohan, Patrick and Daniella Klingebiel (2002) ‘The Fiscal Costs Implications of an Accommodating
Approach to Banking Crisis’, Journal of Banking and Finance (forthcoming).
Hutchison, Michael (2001) ‘A Cure Worse than the Disease? Currency Crises and the Output Costs of
IMF-Supported Stabilization Programs’, National Bureau of Economic Research Working Paper No.8305,
May.
Indrawati, Sri Mulyani (2002) ‘Indonesian Economic Recovery Process and the Role of Government’,
Journal of Asian Economics pp. 577–596.
International Monetary Fund (2000), ‘Indonesia, Selected Issues’, IMF Staff Country Report No. 00/
132, August.
Kenward, Llyoid R. (1999) ‘Assessing Vulnerability to Crisis Evidence from Indonesia’, Bulletin of
Indonesia Economic Studies Vol. 35 No.3, pp. 71–95, December.
Kiggundu, Mosses N. (1996) ‘Integrating Strategic Management Task into Implementing Agencies:
From Firefighting to Prevention’, World Development Vo.24 No.9, pp 1417 – 1430.
Krugman, P. (1998) ‘What Happened to Asia’, mimeo, Massachussets Institute of Technology.
Lindgren, C.J., T.J.T. Balino, C. Enoch, A.-M. Gulde, M. Quintyn and L.Teo (1999) Financial Sector
Crisis and Restructuring: Lessons from Asia, IMF Occasional Paper No.188 (Washington: International
Monetary Fund).
McFarlane, I.J. (1999) ‘The Stability of Financial System’ Reserve Bank of Australia Bulletin, August.
McHale, John (2000) ‘A Report on the Fifth Country Meeting of the National Bureau of Economic
Research (NBER) Project on Exchange Rate Crisis in Emerging Market Countries’, NBER.
McLeod, Ross H. (1998) ‘From Crisis to Cataclysm? The Mismanagement of Indonesia’s Economic
Ailments’, World Economy, Blackwell Publishers, London.
_____ (1999) ‘Indonesia’s Crisis and Future Prospects’, in Karl D. Jackson (ed), Asian Contagion:
The Causes and Consequences of a Financial Crisis, Institute of Southeast Asian Studies Singapore pp.
209–240.
75
Redesigning Indonesia’s Crisis Management
Mishkin, Frederick (1999) ‘Lessons From the Asian Crisis. Journal of International Money and Finance,
No.18 pp 709–723.
_____ (2001) ‘Financial Policies and the Prevention of Financial Crises in Emerging Market Countries’,
NBER Working Paper No. 8087, January.
Nakaso, Hiroshi (2001) ‘The Financial Crisis in Japan during the 1990s: How the Bank of Japan
Responded and the Lessons Learnt’, BIS Papers No.6, October.
Nasution, Anwar (1999) ‘Recent Issues in the Management of Macroeconomic Policies in Indonesia
in Rising to the Challenge in Asia: A Study of Financial Markets, Vol.6, Asian Development Bank.
_____ (2000) ‘The Meltdown of the Indonesian Economy: Causes, Responses and Lessons’, ASEAN
Economic Bulletin, August.
_____ (2002) ‘The Indonesian Economic Recovery from Crisis in 1997–1998, Journal of Asian Economics
No. 13 pp 157–180.
Pangestu, Mari, and Habir Manggi (2002) ‘The Boom, Bust and Restructuring of Indonesian Banks’,
IMF Working Paper No. 02/56, April.
Radelet, S. and J. D. Sachs (1998) ‘The East Asian Financial Crisis: Diagnosis, Remedies, Prospects’,
Brooking Papers on Economic Activity, 1.
_____ (1999) ‘What Have We Learned, So Far, From the Asian Financial Crisis?’, CAER II Project,
Next Steps in the Asian Financial Crisis.
Sabirin, Syahril (2002) ‘Bank Indonesia’s Role in Financial Stability’, Paper presented at the seminar
on Financial Services Authority, Jakarta, February 27.
Sanger, David E. (1998) ‘IMF Now Admits Tactics in Indonesia Deepened the Crisis’, New York Times,
January 14.
Santoso, Wimboh (2000) ‘Indonesia’s Financial and Corporate Sector Reform’, Bank Indonesia Working
Paper, No. 4.
Scott, David (2002) ‘A Practical Guide to Managing Systemic Financial Crises, A Review of Approaches
Taken in Indonesia, the Republic of Korea, and Thailand’, World Bank Policy Research Paper No.2843,
May.
Sinclair, P. J. N. (2000) ‘Central Banks and Financial Stability’, Bank of England Quarterly Bulletin,
Vol.40, No.4, November.
76
Articles
_____ (2002) ‘International Financial Architecture: The Central Bank Governors’ Symposium 2002,
Bank of England Quarterly Bulletin, Vol.42, No.3, Autumn.
Stiglitz, Joseph (1999) ‘Lesson from East Asia’, Journal of Policy Modeling 21(3) pp. 311–330.
_____ (1999) ‘Introduction’, Conference on “The East Asian Crisis: Lessons for Today and for
Tomorrow, Economic Notes No.3.
_____ (2002), Globalization and Its Discontents, W.W. Norton & Co.
Wesaratchakit, Worawut (2002) ‘The Future of Deposit Insurance System in Thailand’, Bank of
Thailand Quarterly Bulletin, March.
World Bank (1997) World Development Report 1997: The State in a Changing World, World Bank,
Washington, DC
_____ (1993) The East Asian Miracle: Economic Growth and Public Policy, New York: Oxford Univeristy
Press
77
Market Risk in Indonesia Banks
Market Risk in Indonesia Banks
A b s t r a c t
This paper is intended to compare the impact of adoption CAR with respect to market risk
using BIS standard model and alternative models. Exponential Moving Average (EWMA), which
widely has been used by banks’ practitioners, will be employed to assess the volatility of exchange
rates. Additionally, the objective of this study is to understand how market risk application in
Indonesian affects to capital base with the rationale that a significant capital decrease will
disrupt financial system stability in Indonesia. The result of this study will also explore what
incentive may be received in the adoption of internal models. If the volatility is very high then
the internal model will provide a higher capital charge, which may be higher than that in standard
model. Based on volatility exchange rate and interest rate data, this study concludes that market
risk application in Indonesia will not push down banks’ CAR to a level at below minimum
requirement and will not disrupt financial system stability. By employing sample volatility of
interest rates and exchange rates from year 2001-2002, internal model gives a lower capital
charge.
JEL classification: G22,G32
Keywords: Capital, Insolvency risk1.
Wimboh Santoso1 dan Enrico Hariantoro2
1 Executive Researcher in Banking Research and Regulatory Directorate (BRRD) of Bank of Indonesia, email: [email protected]
2 Researcher in Banking Research and Regulatory Directorate (BRRD) of Bank of Indonesia, email: [email protected]
78
Article
INTRODUCTION
One of the main indicators to measure banks’ ability in absorbing their risks is the amount of its
Capital Adequacy Ratio (CAR), which is the ratio capital over risks. The capital measure implies that the
higher risks will absorb the higher capital.
In July 1988, BIS Committee on Banking Supervision (the committee) issued CAR regulation proposal
with respect to credit risk as stipulated in Basle Capital Accord 1988 (BCA, 1988). G10 countries had
committed to adopt the proposal by end of 1992. The committee realized that the BCA 1988 contains
deficiencies, among others, only covers credit risk while banks are also exposed to other risks, such as
market risk and operational risk.
In January 1996, the committee issued an amendment of BCA 1988 to incorporate market risk in
CAR. This amendment was agreed to be applied in internationally-active banks by end of 1997. This
proposal must be agenda for supervisory authority, including BI, as all countries try to improve their
compliant to 25 Basel Core Principles (BCP) in banking supervision. This paper examines the possibility
of applying the amendment in Indonesian banks.
This paper will discuss the importance of risks valuation for market risk and inclusion in
capital adequacy framework. It will then be continued with CAR calculation approach to incorporate
market risk. In fact, practitioners have applied internal models for market risk. Therefore, this
paper will also discuss the internal models and the impact to banks’ capital by conducting an
empirical study.
This paper will be organized in the following structure: Section one shows introduction; Section
two discusses risks in banks and risk management system; Section three discusses BIS Standard Method;
Section four outlines Internal Model in market risk, including qualitative and quantitative minimum
requirements; Section five discusses the result of empirical study; and Section 6 makes a summary and
conclusion.
RISKS IN BANKING
Every business, including banking, faces with various risks due to macro and micro condition. Other
factors, such as technology superiority, suppliers’ mistakes, political intervention or natural disasters
are potential risks faced also by all companies. Nevertheless, banks specific roles in intermediary and
payment system may also create specific risks, which are not faced by other business.
79
Market Risk in Indonesia Banks
Risk Identification
Risk can be defined as volatility or standard deviation from a company/business unit’s net cash
flow (Heffernan, 1995). Some economists classify risk associated with bank’s activities, namely market
risk or general economic risk (Flannery and Gutentag, 1979; Guttentag and Herring, 1988), operational
risk, and management risk (Mullin, 1977; Graham and Horner, 1988). In addition, there are also other
risks that may create loss but are difficult to detect in the first phase, such as interest rate risk and
sovereign risk (Stanton, 1994). Gardener (1986) classifies risks as general risk, international risk, and
solvency risk. General risk is fundamental risks occurred in all banks, such as liquidity risk, interest rate
risk, and credit risk. Votja (1973) suggests, risks are classified according to bank’s activities/operations
namely credit risk, investment risk, liquidity risk, operational risk, fraud risk, and fiduciary risk.
In broad outline, risks classifications by the economists are almost the same in description and
coverage. For now, the most significant risk faced by a bank is failure to diversify. This risk may occur
due to high maturity gap or concentration, credit concentration to specific industry, or bank location in
a city without branches. The larger and more modern of a bank, the more complex of the risks they are
dealing with. McNew (1997) suggests that financial risks in modern banks comprises credit risk, market
risk, liquidity risk, operational risk, regulatory risk, and human factor risk.
If we analyze further, the type of risks suggested by those economists are numerous, but they are
similar. However, the committee only proposes market risks in the guidelines. And the guideline for
operational risks is finally included in the Basel II document and still in the process of discussion.
Risk Management
Risk is a probability of experiencing loss caused by volatility of risk factors, which affect the
value of assets and liabilities. Risk management is an activity to manage risk in such away to minimize
loss supported by sufficient systems, such as organization, guidelines, and information system. The
risk management activities includes, among others, risk identification, risk measurement, routine
control, and policy recommendation (shifting/risk hedging, risk absorbent with pricing, insurance,
adding capital, etc.)
Risk identification process is carried out to discover every risk in a transaction, which may
contain more than one risk. These risks will be calculated by aggregating individual risk, taking into
account their correlations. However, not all type of risk can be quantified, such as fraud and legal
risks.
80
Article
Risk measurement procedures are conducted by, among others, determining risk factor, choosing
the approach to be used, running models, and validation. Determining risk factor can be conducted by
identifying source risks (such as credit default, interest rate, exchange rate, and price volatility). Then
measuring risk can be carried out by making forecast on the volatility and trend of risk factors associated
with banks’ position. This step, normally, employs mathematic and econometric models.
Risk management activity can be broken down into 3 (three) category, namely, (1) reduce probability
of risk from happening; (2) limit negative/loss effects to banks; and (3) accept risk by shifting risk
(hedging) or adding capital. However, the management process needs to be accompanied with procedures
and guidelines, organizations and supporting human resources (risk group), and information system.
Efforts to manage risks can be classified in three ways:
- Ex-ante screening
Conduct survey and analysis before a deal to measure the probability of occurring on risk event and
estimate the potential loss as a basis to allocate risk premium in pricing and/or credit rationing.
- Ex-post monitoring
Conduct monitoring and analysis after a deal to update risk level and suggest recommendations to
ensure that risk is still at the acceptable level for management and to take action accordingly with
respect to the internal rules, such as hedging.
As the final step, the bank may have to absorb risks because despite all maximum efforts, there are
parts of risks that cannot be fully removed. As consequences, banks will maintain an appropriate capital
level to cover the risks.
Efforts to maintain the adequacy of capital are the main focus regulatory authority in promoting
financial system stability. Therefore, regulatory authority conducts risk assessment on banks to measure
the solvency and enforces prompt corrective action if necessary.
In performing their business activities, bankers always attempt to maximize return with the
consequence of facing a higher risk (high risk, high return, observe Figure 1).
Banks may be more interested in pursuing high returns business and ignore risks. It may be implicit
moral hazard for banks’ management that should be anticipated by supervisory authorities, because
excessive risk in financial institutions can spark systemic risk, which may threat the entire banking
industry. For that reason, regulatory/supervisory authority obligates to set regulation and enforce
implementation to minimize risks in an appropriate manner. The focus of this paper is to conduct
market risk assessment on large foreign exchange banks.
81
Market Risk in Indonesia Banks
The committee proposes capital charges to cover
BIS credit risk (Basle Accord 1988), market risk
(Amendment – January 1996) and operational risk (Basel
II – 1999). Meanwhile other kinds of risks are not included
in the capital charges due the quantification problem.
Credit risk occurs when a counterparty fails to
fulfill his/her obligation to bank, or sometimes is called
counterparty risk. In managing this risk, bank
management will monitor closely to parties of whom
the bank has claims against; hence, the size of bank
risk can be measured accurately. This type of risk is addressed in Basle Capital Accord 1988.
Market risk is a potential loss occurring either on or off balance sheet position due to volatility in
market price or other commodities. In further details, market risk can be broken down into interest
rate risk, foreign exchange risk, and price risk. Market risk has been incorporated in the CAR in developed
countries in 1997, while regulation of market risk will be issued in Indonesia in 2003 and will be fully
applied in 2005.
Operational risk is a risk related with bank operational activities such as computer system failure,
fraud, wrong doing by internal staff, etc. The committee has issued proposal (Basel II) to incorporate
operational risk in CAR. The proposal has been distributed to various parties for comment and final
version will be released by end of 2003.
The next section will describe standard method to assess market risk as stipulated in amendment
Basel Accord January 1996.
BIS STANDARD METHOD
In January 1996, BIS issued “Amendment to the Capital Accord to incorporate market risks” as
guideline in calculating capital charge for market risk. The purpose this Amendment is to ensure the
adequacy of capital against the volatility of market risk variables such as interest rates, foreign exchange
rates and price of commodities.
The amendment re-emphasizes that capital to cover market risk consists of equity capital and
retained earnings (Tier 1) and supplementary capital (Tier 2). Additionally, the proposal also allows
banks to use Tier 3 capital, which is specifically dedicated to cover market risk. Tier 3 consists of short-
Re t u r n
R i s k
Figure 1
Risk and Return Comparison
82
Article
term subordinate debts with minimum time to maturity of 2 (two) years and subject explicit lock-in
clause. Lock-in-clause means that banks are not allowed to pay interest and/or principal even at the
maturity since the payment causes CAR to fall below minimum level. Eligible Tier 3 is limited to
maximum 250% of Tier 1 allocated to cover market risk.
As anticipation from snowball effect that may take place due to bank failure, BIS views that minimum
CAR is a discipline ought to be applied as an effort to strengthen international banking system stability
and sustainability. Nevertheless, CAR application has to calculate all risks in maximum level. Therefore,
BIS conducts regular studies by requesting inputs for various parties related in international financial
system to obtain the most accurate CAR calculation method. After credit risk was approved and regulated
in Basle Accord 1998, the next step was to add market risk into that Accord through the above-mentioned
Amendment. Meanwhile, the next risk, namely operational risk continues to be scrutinized by BIS.
The Amendment specifies in more details with respect to the calculation of risks:
- Price risk of equities and commodities
- Interest rate risk in interest related instruments, such as swap, forward, and bonds in trading
books.
- Foreign exchange risk in trading book and banking book.
The proposal also provides separate section on the treatment of options.
The proposal allows banks to adopt internal models to assess the adequacy of their capital subject
to verification by supervisory authorities to satisfy the qualitative and qualitative minimum requirements.
Verification will take a long process and close discussion between banks and supervisory authorities to
come-up with an agreed and sound risk management process.
Standard method focuses its measurement on 5 (five) types of market risks, namely interest rate
risk, equity position risk, foreign exchange risk, commodities risk, and price risk (for options). Risk
assessment in interest rate related instrument adopts a “building-block” approach by separating
assessment on capital charge for specific risk and general market risk.
Specific risk is intended to reflect potential loss due to financial instrument issuer default, while
general market risk is intended to reflect potential loss as a result of interest rate volatility. Both risks
are subject to different capital charge. The total capital charge to cover market risk (on top of capital
charge to cover credit risk as mentioned in Basle Accord 1988) is the aggregation of five risks mentioned
above. Detailed risk calculation is outlined in the following section.
83
Market Risk in Indonesia Banks
Interest Rate Risk
Capital charge for interest rate risk is divided into 2 (two) types of risk – specific risk and general
market risk.
Specific risk
In calculating specific risk, offsetting is only possible for identical positions (issuer, rate, maturity,
and the other features are the same). Financial instruments, which are subject to interest rate risk are
all fixed-rate and floating-rate debt securities and other instruments with similar characteristic, but
excluding mortgage securities.
Specific risk is broken down into 5 (five) categories as outlined in the following tables:
Government 0.00% -
Qualifying 0.25% Remaining maturity 6 months or less
1.00% Remaining maturity 6 to 24 months
1.60% Remaining maturity more than 24 months
Others 8.00% -
Instruments Specific Risk Charges Criteria
Table 1 : Specific Risk Capital Charges
“Government” securities include all securities issued by government such as bonds, treasury bills,
and other government instruments. But supervisory authority has discretion to charge certain percentage
for specific risk on securities issued by foreign governments.
“Qualifying” securities include securities issued by state-owned companies, multilateral
development banks, and other securities with rating investment grade issued by good-global rating
agencies and verified by the supervisory authority.
“Other” includes other securities that do not satisfy requirements of “government” and “qualifying”
categories and subject to an 8% charge.
General Market risk
In calculating capital charges for general market risk, there are two options proposed by BIS,
maturity method and duration method. For these two methods, total capital charge is accumulation of
4 (four) components, namely:
84
Article
- Net short or long positions of all trading book.
- Matched positions in every time-band (vertical disallowance).
- Matched positions between time-band (horizontal disallowance).
- Net charge for option positions.
Positions of interest rate related instrument will map into 15 maturity bands based on remaining
maturity for fixed rate instruments and next repricing dates for floating rate instruments. The time
band will be split into 3 (three) zones (Table 2).
Tabel 2 : Time-band and Risk Weights in Maturity Method
Zone 1 : Zone 1 :
0 – 1 month – 1 month 0.00% 1.00
1 – 3 month 1 – 3 month 0.20% 1.00
3 - 6 month 3 - 6 month 0.40% 1.00
6 – 12 month 6 – 12 month 0.70% 1.00
Zone 2 : Zone 2 :
1 – 2 year 1.0–1.9 year 1.25% 0.90
2 – 3 year 1.9–2.8 year 1.75% 0.80
3 – 4 year 2.8–3.6 year 2.25% 0.75
Zone 3 : Zone 3 :
4 - 5 year 3.6-4.3 year 2.75% 0.75
5 – 7 year 4.3–5.7 year 3.25% 0.70
7 – 10 year 5.7–7.3 year 3.75% 0.65
10 – 15 year 7.3–9.3 year 4.50% 0.60
15 – 20 year 9.3–10.6 year 5.25% 0.60
> 20 year 10.6-12 year 6.00% 0.60
12-20 year 8.00% 0.60
> 20 year 12.50% 0.60
Coupon > 3% Coupon < 3% Risk weight Assume Change in Yield
The calculation of general market risk charges will be derived from the following steps:
- First step, multiply all position in time-band with risk weights, which represents price sensitivity to
interest rate changes.
- Second step, offsetting the positions derived from first step in each time band, where the smaller
position, long or short, is subject to 10% capital charge (vertical disallowance) for maturity method
and 5% for duration method.
Sources: BIS Proposal
85
Market Risk in Indonesia Banks
- Third step offsetting residual value (long and short derived from the second step) with the value in
different time-band in the same zone, and with residual value between different zone. For every
offsetting will be subject to a capital charge as illustrated in Table 3:
1 0-1 month
1-3 month
3-6 month 40%
6-12 month 40%
2 1-2 year
2-3 year 30% 100%
3-4 year
3 4-5 year
5-7 year
7-10 year
10-15 year 30%
15-20 year
> 20 year
Zone Time-band Within Intar Zone in Between Zone
zone Sequence Manner 1 and 3
Table 3 : Capital Charge and Horizontal Disallowance
Table 2 and 3 show the time band and capital charges based on maturity method, while duration
method doesn’t separate coupon above and below 3% and the vertical disallowance is only 5%. Total
capital charge is the aggregation of specific and general market risk as specified in the above steps due
to a limited space in this articles.
Interest Rate Derivatives
Interest rate risk measurement also covers risk arising from interest rate derivatives instruments
and off-balance-sheet instruments in trading book, such as forward rate agreements (FRAs), forward
contract, bond futures, interest rate/cross-currency swap, and currency forward. Each of those
instruments has to be converted according to its underlying transaction/instruments and subject to
specific and general market risk calculation. Capital charge calculation for this kind of instrument is
relatively complex so it will not be discussed in detail in this paper.
Sources: BIS Proposal
86
Article
Equity Position Risk
This risk occurs when a bank takes position in equities in trading book. Equities include common
stock, convertible stocks/securities, and commitment to buy/sale those equities. Banking law in Indonesia
prohibits banks to hold stock or other securities in their trading activities. Therefore the risk is irrelevant
to be discussed in this paper.
Foreign Exchange Risk
This risk occurs when a bank takes position in foreign currency, including gold. The BIS standard
method introduces “shorthand” and gross method in calculating capital charge for this risk. This study
only focuses on the standard method with the following rules as stipulated in the proposal.
- Capital charge for foreign exchange risk is 8% from total net open position (local currency) for
foreign exchange and gold.
- Net open position is an accumulation of:
- Net short or long position, which ever is larger, and
- Net (without differentiating long or short) position in gold.
For example, if a bank has position as follow:
- JPY long IDR 50 billion - FFR short IDR 20 billion
- DEM long IDR 100 billion - USD short IDR 180 billion
- GBP long IDR 150 billion
- Gold short IDR 35 billion
The above position implies that long is IDR 300 billion and short is IDR 200 billion, and gold position
is short IDR 35 billion. Capital charge will be = 8% x (300+35) = IDR 26,8 billion.
Commodities Risk
This risk occurs when a bank takes position in commodity such as agriculture products, mineral and
precious metal (other than gold). But, the same as equity position risk, banks in Indonesia are still
prohibited to involve in commodity trading so commodity risk will not be discussed in this paper.
87
Market Risk in Indonesia Banks
Option Price Risk
Option is a contract to provide a right (but not obligation) to the owner/buyer to buy or sell a
certain amount of currency or other financial instruments with a predetermined price in some point in
time in the future. The risk for buyer is only limited to option premium he/she has to pay plus fee to the
broker. But for risk of seller (writer) is unlimited because it will be determined by the difference
between market price and strike price. Therefore, option seller will face higher risk compared to option
buyers.
BIS recognizes the difficulties of measuring option price risk, thus in this standard method there
are some alternatives approaches:
– Option-buyers can employ simplified approach.
– Option-seller bank/write option is expected to use Intermediate Approach or a comprehensive Risk
Management model. BIS claims that the more significant of trading option, the more sophisticated
risk management method it has to use.
In summary, capital charge for options with simplified approach is illustrated in Table 4.
Capital charge = (market price from Underlying
transaction/securities x capital charge amount
for specific and general market risk of the
underlying) – the “in the money” option value
(if any)
Cash Position Option Position Capital Charge Estimation
Table 4 : Simplified Approach – Capital Charge for Option
Long Long put
Same as above
Same as above
Same as above
Capital charge is the smallest between:
– Market price from underlying securities x
capital charge amount for specific and
general market risk from the underlying
securities
– Option market price
Long Long put
Long Long put
Long Long put
Long call
or
Long put
Sources: BIS
88
Article
Long put capital charge = (market value from underlying transaction/securities x accumulation of
capital charge for specific and general market risk from the underlying) – option value “in the money”
(if any)
Simplified Approach: An example
Holding underlying assets of 100 sheets of stock with current market value IDR 1.500 / sheet.
Assuming, the position is put option with strike price IDR 1.600 / sheet.
Underlying assets = (100 x 1.500) x 16% = IDR 24.000
(Note : 8% specific risk + 8% general market risk)
Option in the money = (1.600 – 1.500) x 100 = IDR 10.000
Capital charge = 24.000 – 10.000 = IDR 14.000
The next section will discuss value at risk (VaR) in the internal model, and qualitative and quantitative
requirements for banks, which are going to use internal model.
INTERNAL MODEL (ALTERNATIVE MODEL)
Quantitative and Qualitative Requirements
In measuring capital charge for market risk, the BIS proposal allows banks to use internal model
since risk management in banks satisfies quantitative and qualitative requirements and approved by the
supervisory authority.
The quantitative requirements, among others, are:
– Using value-at-risk (VaR) method with 99% and one-tailed confidence interval.
– The price shock standard used in the model is a minimum of 10 trading days so the minimum holding
period is equal to that period.
– The model employs historical data of a minimum 1-year observation.
– The amount of capital charge for banks using internal model is the higher of:
- VaR of the day before the capital is assessed, or
- 3 (three) times of average daily VaR in the last 60 working days.
While the qualitative requirements covers:
– Fulfill general criteria of adequate risk management system
89
Market Risk in Indonesia Banks
%100*1
+
t
t
y
yLn
– Apply qualitative standard as indicators in case there is a mistake in using the internal model.
– Establish formal guidelines to make adequate market risk factors classification.
– Provide quantitative minimum standard in statistical parameter in measuring risk.
– Provide guidelines for stress testing.
– Standardize validation procedure for external error in modeling.
– Provide clear guidance since banks adopt a combination of internal model and Standard Model.
VaR in Internal Model
In general, the internal model to measure market risk exposure is based on Value-at-Risk (VaR)
concept. VaR is an approach to measure maximum loss may to occur in a portfolio position due to risk
factors volatility such us price, interest rate and exchange rate in a certain period by using certain
confidence level. In the mathematical formula can be described as follows:
ttttt VMVaR /1/1 * ++ =
where,
ttVaR /1+ = Maximum loss of an instrument for time horizon 1+t assessed at time t
tM = Mark to market value of an instrument at time t
ttV /1+ = Volatility of risk factor of the instrument for time horizon 1+t assessed at time t
Application VaR in internal model needs the risk factors volatility to measure the overall risks in
a certain point of time in the future. Therefore, there should be a volatility estimation of risk factor
volatility. There are 2 (two) types of volatility, historical volatility and implied volatility. Historical
volatility is a volatility based on data time series, while implied volatility is derived from option price
into option pricing model such as Black-Scholes formula.
The discussion in the following section will be focused on historical volatility, running model, and
testing procedures.
Data
90
Article
Historical volatility return from time series is calculated with the following equation:
Next, stationary test is required to ensure that there is a proper data included into the model as
the mean-variance analysis assumes that the mean process is constant. The various method can be
applied in stationary test such as Ljung-Box method, Box-Pierce, Dicky-Fuller, etc.
Models
Some volatility models may be used in VaR calculation are:
– Mean-variance analysis, which comprises of:
- Exponential weighted Moving Average (EWMA)
- Generalized Autoregressive Conditional Heteroscedasticity (GARCH)
– Simulation
- Historical simulation
- Monte Carlo simulation
– Neural network
– Algorithmic
In this paper, we will employ mean-variance analysis with univariate estimation models.
Mean-Variance Model
Mean is a projection in calculating average of historical time series. In the projection above, the
average data will always create positive and negative error. Therefore, the errors must be squared to
create variance value to asses the model accuracy. Because the estimation of this average still contains
error, there should be an analysis towards those variance behaviors. Among various variance analysis
methods, this study limits the exercise using exponential weighted moving average (EWMA) and
generalized autoregressive conditional heterocedasticity (GARCH).
Exponential Weighted Moving Average (EWMA)
This approach assumes that today’s projection will be affected by yesterday’s projection and actual
results. The core of EWMA is the application of exponential-smoothing techniques, which previously
used to predict output in marketing and production (operations research) areas.
91
Market Risk in Indonesia Banks
F F Xt t t t t+ −= + −1 1 1| | ( )α α ( )1− =α ρF F X X
F X X
F X X X
F X X X
t t t t t t
t t t t
t t t t t
t t t t t
+ − − −
− − −
− − − −
− − − −
= + +
= + +
= + + +
= + + +
1 1 2 1
21 2 1
22 3 2 1
32 3
22 1
| |
|
|
|
( )
( )
α α ρ ρ
α αρ ρ
α α ρ αρ ρ
α α ρ αρ ρ
F X X X X X Xt t
q
t q
q
t q t t t t+ −−
− − − − −= + + + + +1
1
1
3
3
2
2 1| ( ) .......α ρ α ρ α ρ α ρ αρ ρ
F Xt t
it i
i
q
+ −=
= ∑10
| ρ α
yt = φ
ty
t-1 + ... + φ
py
t-p +e
t - θ
1e
t-1 - ... θe
t-q
In EWMA, the next forecast estimation in a time-series ttF /1+ is the function of previous forecast
1/ −ttF and observation tX .
JP Morgan (1994) has used this model by assuming mean from historical series as 0, so only
forecast variance would be run. Mathematically, EWMA process can be described as follow:
Where,
α = Decay factor )10( <<α
The value of decay factor is very important in EWMA. The lager value of decay factor (assuming
approaching one), the previous forecast will significantly affect to the forecast. Root mean square
errors (RMSE) is a parameter to select the best decay factor in forecasting.
Generalized Autoregressive Conditional Heterosedasticity (GARCH)
GARCH method will employ 2 (two) processes, namely, mean and variance process. The mean
process was first introduced by Box-Jenkin (1976) by making a time series analysis with autoregressive
(AR) and moving average (MA) combination. This method, then, will be integrated into ARMA to get a
stationer time series with the following equation:
92
Article
εt2 = α
0 + α
1ε2
t-1 + α
2ε2
t-2 + ... + α
qε2
t-q +v
t
++= ∑ ∑= =
−−
q
i
p
i
titittv
1 1
2
1
2
10
22 σβεαασ
Because the error variants are not always constant, Engle (1982) improved it with variance process,
namely forecasting method to address error variants. This model is called auto regressive conditional
heteroscedasticity (ARCH). With variability in variants, forecasting becomes:
where vt = white noise (zero mean)
Bollerslev (1986) improved Engle work by considering AR process in heteroscedasticity from variants
into generalised auto regressive conditional heteroscedasticity (GARCH), which will be described in the
following equation:
In the next section, we will discuss the result of empirical study on 13 largest foreign exchange
banks. By using data from those banks, we will assess capital charge for market risk using the BIS’
standard method and Internal Model.
EMPIRICAL STUDY: EXERCISE OF MARKET RISK
CAR of National Banking
Banks in Indonesia are not required to set capital to cover market risk, even though some banks
may expose to this risk. Adoption of CAR market risk may create burden for banks with low capital.
However, market risk regulation is intended to apply just for internationally active banks. In fact, some
Indonesian banks have global operation where the host country supervisory authority also applies market
risk. The absence of market risk in CAR will also create burden for bank to operate globally. Finally,
almost there is no choice for internationally active banks to adopt market risk.
Since financial crisis hit Indonesia in 1997, Indonesian banks have experienced a very significant
capital drain, which causes total capital of the majority banks downed to negative. The government has
taken recovery efforts, among others, bank re-structuring and re-capitalization. However, the small
number of banks of still found difficulty to meet the minimum CAR level of 8% by end 2001. The capital
requirement was only based on CAR calculation according to Basle Accord 1998, thus if the market risk
is also included, they would be in more difficult to achieve 8%.
93
Market Risk in Indonesia Banks
Implementation of BIS Proposal
Implementation of BIS proposal to incorporate market risk either standard approach or internal
model requires some conditions, among others, includes:
- Appropriate risk management system and risk management unit.
- Consistent policy in allocating portfolio in trading and banking for interest rate related instrument.
- Capable human resources specializing in risk measurement area.
- Efficient and accurate information as input in risk management system.
Stringent requirement in applying capital charges for market risk implies that risk measurement
in market risk will be complicated due to technical measurement in risk management. These
requirements are intended to ensure accurately measure banks’ exposure in market risk, which is the
main problem for large banks in Indonesia during crisis. Thus, bank capital had been deteriorating
during crisis.
Empirical study: Exercise of Market Risk Capital Charge
This study assesses the impact of market risk implementation on 13 large banks consisting of 4
state banks and 9 national private forex banks. The exercise employs financial data as of 30 June 2002.
Data gathering was received from survey on those banks. According to the survey, we can summarize
the following information:
- The derivative transactions are still limited to plain hedging transactions, such as swap and forward.
While for a more complex derivative transactions such as forward rate agreement (FRA), futures
and option are rare in Indonesian banking.
- Banks’ security portfolio are only rated by local rating agency
- There was a private bank with negative capital due in the process of re-capitalization by government.
The empirical results is illustrated in Table 5 and Table 6.
The EWMA method is one of approaches in mean-variances forecasting using time-series data. In
general, this study follows the following steps:
1. Collecting interest rate data series from internal BI data base and exchange rate from Bloomberg
for the period of July 2001 to end of June 2002. The exchange rate time series data is the daily mid-
day exchange rate of 41 foreign currency, while the interest rate data is the average of deposit/
94
Article
Bank A 40,825,188 17.50 7,145,988 20,771 320 41,088,826 17.39 - 0.11
Bank B 43,681,668 15.53 6,782,551 60,721 12,287 44,594,268 15.21 - 0.32
Bank C 66,078,882 38.97 25,753,567 600,755 94,009 74,763,428 34.45 - 4.53
Bank D 8,066,948 11.71 944,363 27,262 - 8,407,727 11.23 - 0.47
Bank E 3,927,090 10.09 396,081 4,631 3,559 4,029,455 9.83 - 0.26
Bank F 21,327,084 39.95 8,519,537 23,054 1,494 21,633,934 39.38 - 0.57
Bank G 6,463,829 - 53.90 -3,484,095 112,908 1,880 7,898,684 -63.69 - 9.79
Bank H 15,800,759 32.90 5,197,692 31,928 8,652 16,308,004 31.87 - 1.02
Bank I 8,642,883 26.62 2,301,156 5,937 4,307 8,770,933 26.24 - 0.39
Bank J 8,044,052 19.92 1,602,094 52,161 2,302 8,724,840 18.36 - 1.55
Bank K 9,755,518 33.55 3,272,643 25,700 6,252 10,154,918 32.23 - 1.32
Bank L 4,568,597 22.29 1,018,544 17,855 2,148 4,818,639 21.14 - 1.16
Bank M 5,302,232 15.38 815,221 39,528 1,030 5,809,215 14.03 - 1.34
Table 5 : BIS Standard Methode
Million Rupiah
RWA )* CAR (%) Capital Capital Charge New New CAR +/- CAR
Int. rate risk Forex risk RWA )** (%) (%)
)* Adjustment aftert market risk’s RWA
)** Previous RWA added total of capital charge (interest rate + forex) multiple by 12.5
savings interest rate for 1 month, 3 months, 6 months, 12 months, and above 12 months time
horizon.
Bank A 40,825,188 17.50 7,145,988 3,970 32 40,875,213 17.48 - 0.02
Bank B 43,681,668 15.53 6,782,551 12,911 524 43,849,606 15.47 - 0.06
Bank C 66,078,882 38.97 25,753,567 257,249 8,097 69,395,707 37.11 - 1.86
Bank D 8,066,948 11.71 944,363 9,633 - 8,187,361 11.53 - 0.17
Bank E 3,927,090 10.09 396,081 2,196 284 3,958,090 10.01 - 0.08
Bank F 21,327,084 39.95 8,519,537 10,520 146 21,460,409 39.70 - 0.25
Bank G 6,463,829 - 53.90 - 3,484,095 8,179 155 6,568,004 - 54.75 - 0.85
Bank H 15,800,759 32.90 5,197,692 8,750 638 15,918,109 32.65 - 0.24
Bank I 8,642,883 26.62 2,301,156 2,431 426 8,678,596 26.52 - 0.11
Bank J 8,044,052 19.92 1,602,094 4,546 224 8,103,677 19.77 - 0.15
Bank K 9,755,518 33.55 3,272,643 7,422 430 9,853,668 33.21 - 0.33
Bank L 4,568,597 22.29 1,018,544 5,483 82 4,638,160 21.96 - 0.33
Bank M 5,302,232 15.38 815,221 21,505 97 5,572,257 14.63 - 0.75
Table 6 : E M W A
Million Rupiah
RWA )* CAR (%) Capital Capital Charge New New CAR +/- CAR
Int. rate risk Forex risk RWA )** (%) (%)
)* Adjustment aftert market risk’s RWA
)** Previous RWA added total of capital charge (interest rate + forex) multiple by 12.5
95
Market Risk in Indonesia Banks
Period (250 days)
50 100 150 200 250
-4
-3
-2
-1
0
1
2
Ceiling Floor Actual
Volatility
Figure Result Plot of IDR Estimation
2. Calculating the daily volatility (delta) for those series
3. Treat the data and calculate the decay factor
4. Calculate the volatility by assuming normal distribution with 95% confidence level.
From the exercise, we can derive the following conclusion:
1. CAR decreases in relatively small,
approximately below 2%.
2. In general, EWMA method delivers a smaller
capital charge compared to that in standard
method. However, the BIS proposal still
requires the application of minimum
multiplication factor by 3 on internal model
capital charge as a buffer of shock events. The
plot result of volatility projection for exchange
rates can be illustrated in the following graph.
3. The amount of capital charge is the additional
capital a bank to cover market risk. In the
exercise, we convert market risk charges into risk-weighted-assets by multiplying 12.5 to provide
identical denominator in the capital ratio.
4. The findings show that capital charge result according to standard method is higher than that in
internal model calculation. Inclusion of market risk in CAR will provide more prudent capital, but it
may reduce competitiveness due to additional capital cost. However, in volatile market, the
application of internal model will also create a new burden for banks as it will generate larger
capital charge from that the standard method.
SUMMARY AND CONCLUSION
Since the end of December 1997, internationally active banks in developed have applied market
risk in their CAR based on the BIS proposal 1996. In general, banks in those countries have sophisticated
risk management system, which among others include adoption of internal models. Nevertheless, standard
method is more prudent than internal models.
96
Article
Finally, this study makes the following conclusions and recommendations:
1. Internationally active banks in Indonesia will be subject to market risk if host supervisory authority
applies CAR market risk. If it is the case, market risk regulation in home countries doesn’t create
a new burden for the banks.
2. Application in Indonesia banks will not create disruption to financial stability. Based on this study,
adoption of CAR market risk will not significantly affect to banks’ capital. However, regular exercise
is necessary to ensure that the problem arising from market risk can be identified as early as
possible.
3. Standard method normally provides a higher capital charge than that in internal models because
the prudential principal is the main priority of the proposal.
4. In order to apply market risk management, guidelines from regulatory authority will stimulate
proper implementation of risk management in banks.
BIBLIOGRAPHY
Basle Committee on Banking Supervision (1988), “International Convergence of Capital Measurement
and Capital Standards”, Basle : Bank for International Settlements, July.
Basle Committee on Banking Supervision (1996), “Amendment to the Capital Accord to Incorporate
Market Risk”, Basle : Bank for International Settlements, January.
Bollerslev, Tim (1986), “Generalised Autoregressive Conditional Heteroscedasticity”, Journal of
Econometrics 31, pp. 307-327.
Box, G. and D. Pierce (1970), “Distribution of Autocorrelations in Autoregressive Moving Average
Time Series Models”, Journal of the American Statistical Association 65, pp. 1509-1526.
Box, G. and G. Jenkins (1976), “Time Series Analysis, Forecasting and Control”, San Francisco, CA
: Holden Day.
Engle, R.F. (1982), “Autoregressive Conditional Heteroscedasticity with Estimates of Variance of
United Kingdom Inflation”, Econometrica 50, pp. 987-1007.
97
Market Risk in Indonesia Banks
Flannery, M.J. and J.M. Guttentag (1979), “Identifying Problem Banks”, in proceedings of a
Conference on Bank Structure and Competition, Chicago: The Federal Reserve Bank of Chicago.
Gardener, E.P.M. (1986), “UK Banking Supervision-Evolution : Practice and Issues”, London : Allen
and Unwin.
Graham, F.C. and J.E. Horner (1988), “Bank Failure : An Evaluation of the Factors Contributing to
the Failure of National Banks, in proceedings of a Conference on Bank Structure and Competition,
Chicago: The Federal Reserve Bank of Chicago.
Guttentag, J.M. and R. Herring (1988), “Prudential Supervision to Manage System Vulnerability”,
Chicago: The Federal Reserve Bank of Chicago.
Heffernan, S.A. (1996), “Modern Banking in Theory and Practice”, Chichester, UK : John Wiley &
Sons Ltd.
Jorion, P. (1996), “Value at Risk : The New Benchmark for Controlling Market Risk”, Chicago : Irwin
Professional Publishing.
J.P. Morgan (1995), “RiskMetrics – Technical Document”, New York : Morgan Guaranty Trust Company,
Global Research, 3rd Edition.
J.P. Morgan (1996), “RiskMetrics – Technical Document”, New York : Morgan Guaranty Trust Company,
Global Research, 4th Edition.
Ljung, G. and G. Box (1978), “On a Measure of Lack of Fit in Time Series Models”, Biometrica 65,
pp. 297-303.
McNew, L. (June 1997), “Risk magazine”, pp. 52-57.
Mullin, R. (1977), “The National Bank Surveillance System” in : E.I. Altman and A.W. Sametz, eds.,
Financial Crises : Institutions and Markets in a Fragile Environment, New York : John Wiley & Sons.
Sharpe, W.F. (1970), “Portfolio Theory and Capital Market”, New York : McGraw-Hill, Inc.
Stanton, T.H. (1994), “Non Quantifiable Risk and Financial Institutions : The Mercantilist Legal
Framework of Banks, Thrifts and Government-sponsored Enterprises, in C.A. Stone and A. Zissu, eds.,
Global Risk Based Capital regulations, Illinois : Richard D. Irwin.
Votja, G.J. (1973), “Bank Capital Adequacy”, New York : First National City Bank.
98
Articles
An Empirical Analysis of Credit Migration
In Indonesian Banking
A b s t r a c t
In this paper, we analyze the process of credit migration in an Islamic commercial bank
operating in Indonesia. In particular, we explore the relevance of industrial performance to the
dynamic of loan status in the selected bank. From the statistical results, we find two interesting
phenomena. First, industrial performance is statistically significant in affecting the credit migration
process. And, second, we find irreversibility in the credit migration process. This analysis can be
used to provide additional information to Indonesian banks, as well as their supervisors, to help
improve the accuracy of the risk assessment process, and the efficiency of external oversight
respectively.
JEL classification: C51, C53
Keywords: Credit risk, risk assessment, credit migration
Dadang Muljawan1
1 Research Economist, Bank Indonesia
99
An Empirical Analysis of Credit Migration In Indonesian Banking
INTRODUCTION
Loan quality influences the operational soundness of a bank. Loan quality, in fact, is determined by
many factors, such as the credit granting process, business capacity and the business environment. The
Basel Committee on Banking Supervision has issued a consultative paper on the principles underlying
sound management of credit risk (Basel Committee, 1999a). The main objective of credit risk management
should be to maximize a bank’s risk-adjusted rate of return by maintaining credit risk exposure within
acceptable parameters. The Basel Committee document covers the following areas: (a) establishing an
appropriate credit risk environment; (b) operating under a sound credit granting process; (c) maintaining
an appropriate credit administration and monitoring process; and (d) ensuring adequate controls over
credit risk.
A number of quantitative models have been developed to support more sophisticated credit-risk
assessment. Some models are based on an actuarial approach, while others use statistical approaches.
One of the leading methodologies has been developed by J.P Morgan (J.P Morgan Technical Document,
1997). The approach involves the adoption of a “Mark-To-Market” (MTM) process and the use of a
transitional-probability matrix of the credit migration. The Basel Committee has also issued a document
reviewing current practices and applications of credit risk modeling (Basel Committee, 1999b).
The loan classification process has an important impact on the operational soundness of banking
operations, and can be evaluated using an accounting process (indicators) to assess the financial condition
of the borrowers (using delinquency periods, turn-over ratios and profit/loss figures as the internal data
set). The banks, as well as their supervisors, should be able to conduct ex-post asset valuation and ex-
ante risk assessment in order to obtain accurate estimates of the banks’ real asset values. Apart from
using such an accounting process, loan classification can also be estimated by finding the relevant
external economic indicators, such as industrial performance, that influence asset quality. Such
information can help the banks and their supervisors to reconcile the asset valuation and risk assessment
processes. This paper studies the correlation between industrial performance and the migration of loan
quality for one Islamic commercial bank operating in Indonesia, to help shed light on these issues.
The paper proceeds as follows. The next section provides some background analysis relating the
level of earnings and breakeven analysis to the probability of companies having financial problems.
Section 3 discusses the model used to estimate the correlation between the dynamics of loan classification
and the industrial performance index. Section 4 discusses the empirical results. Section 5 concludes the
paper.
100
Articles
BACKGROUND ANALYSIS
Level of earnings, the break-even point and the probability of bankruptcy
A company can maintain its operational sustainability only if its operation has reached a breakeven
point (BEP)(i.e. revenue earned is at least the same as the costs incurred, including the cost of borrowing2 )
(Brewster, 1997). Breakeven analysis is usually
expressed in terms of volume )( BEPQ , as
illustrated in Exhibit 1. TR , TC and Π represent
total revenues, total costs and profit earned as a
function of business volume Q . Π is basically TR
minus TC . If the revenue earned is less than the
costs incurred (e.g. during a recession), there is
a high probability that the company will face
financial problems in the near future since the
revenue earned will ultimately determine the
level of profitability.
The banks, however, deal with companies
having different levels of business volume and
operational efficiency. To capture this variability,
let us assume that the earnings of the companies
are normally distributed, as shown in Exhibit 2.
)(Qπµ , )(Qπσ and )( BEPQπ represent the average
and variance of profits and profitability at BEP
respectively as a function of Q .
The points lying on the left of the BEP line
(zero profit) indicate those companies having
financial problems (i.e. creating problem loans
for the banks).
Relationship between macroeconomic indicators and business soundness
The aggregate level of business activity in a particular sector of industry can be portrayed using
economic indices. The Gross Domestic Product (GDP) growth rate per sector can be used as an indicator
TC
TR
Π
Break
Point
QBEP
Fixed
costs
£
Q
Figure 1 :
Break Event Analysis
Figure 2 :
Probability of Financial distress as
an Output Function
break event point
x % probability
µπ(Q) lying below
(πQBEP)
µπ(Q)
σπ (Q)
(πQ)
P (Π)
2 The Islamic bank applies a mark-up rate that is invariant to interest rates (10% on average); therefore, the cost of borrowing is relatively fixed and the
BEP is not affected by the fluctuation in interest rates.
101
An Empirical Analysis of Credit Migration In Indonesian Banking
for assessing the business soundness of that particular
sector of industry (Krolzig and Sensier, 1998)3 . A high
rate of GDP growth is indicative of high growth of
business transactions and a high level of revenue
earned; and vice versa. Therefore, different GDP
growth rates indicate different probabilities of the
companies having financial problems. These conditions
are illustratively shown in Exhibit 3.
Mathematically, from Exhibit 3(a) and (b), let us
assume an economic population has the same level of
variability of profitability in two different economic
conditions ( BA ππ σσ = ) and different levels of profitability ( BA ππ µµ > ). The population will face a
lower probability of experiencing financial problems during better economic conditions than during a
recessionary period (i.e. }{}{ BEPQPBEPQP BA <<< or )()()()( BEPBEP Q
B
Q
A
ππ ππ ∞−∞− Φ<Φ ), where
BAiBEPQ
i ,),( )( ∈Φ ∞−ππ represents a cumulative probability distribution function between ∞− and
)( BEPQπ .
This leads to the hypothesis that the level of financial distress (as indicated in the bank’s loan
classification) can also be represented by the variability of the macroeconomic indicator, such as the
GDP growth rate.
THE MODELLING PROCESS
The concept of credit migration
At different points in time, banks may have different proportions of non-performing loans in their
assets. Over time, the loan status may thus change, with loans either being upgraded to a better status,
or remaining in the same status, or being downgraded to a lower status. The probabilities of such
changes occuring in the loan status (credit migration) are illustrated in Exhibit 4. In the analysis, we
also check to see if irreversibility exists in the credit migration process (i.e. the possibility of having
different levels of probability for an individual loan to get downgraded during a recession period and to
get upgraded during a period of growth). In order to capture this phenomenon, the analysis is conducted
Figure 3 :
Level Probabilty of Corporate
Financial Distress
σπB (Q)
ΦΑ (π)
µπA
σπA (Q)
π (Q)µπB
π (QBEP)
ΦΒ (π)(b)
(a)
P (π)
3 Recent research disaggregates economic output to study dynamic developments within different industrial sectors.
102
Articles
individually for each loan granted by the bank for each
interval (quarter). In this paper, we use the level of
GDP growth (i.e. P(X/Y): f(GDP growth)) as the
independent variable explaining the credit migration
process.
P(X/1), P(X/2), P(X/3) and P(X/4) represent the
probabilities of loans of status X migrating to loans of
status 1, 2, 3 and 4 respectively, where the total of
the migration probabilities is equal to 1 (i.e. P(X/1) +
P(X/2) + P(X/3) + P(X/4) = 1).
Data set
Loan status is classified into 4 categories: ‘current’, ‘sub-standard’, ‘doubtful’ and ‘loss’. The
banks conduct loan classification analysis periodically (every month). The loan classification process is
also checked and adjusted during on-site supervision by the supervisory authority (Bank Indonesia)4 .
The data series of individual loans covers the period 1992 to 20005 .
GDP growth rate data is obtained from the Central Bureau of Statistics (Biro Pusat Statistik (BPS)).
Quarterly data is used for the period 1992 to 2000.
Statistical model
We use a logistic function to estimate the probabilities of credit migration6 . The representative
credit migration probability between two different classifications/ categories, is given as follows:
)( 211
1)/1(
iXiie
XYEP ββ +−+===
Exhibit 5 illustrates the methodology used to conduct the regression analysis using the logistic
function.
Step 1 of the logit analysis involves calculating the probability of loan quality migrating to/ remaining
in the loan classification ‘loss’ (coll-4)7 . Step 2 of the logit analysis calculates the probability of loan
Figure 4 :
Probability Migration of Individual loan Status
COLL X
COLL 1
COLL 2
COLL 3
COLL 4
Period t+1Probabi l i tyPer iod t
P ( X / 1)
P ( X / 2)
P ( X / 3)
P ( X / 4)
4 On-site supervision is conducted on an annual basis or otherwise on an ad-hoc basis if there is an indication of fraudulent practices by a particular bank.
5 Individual loans are unidentified.
6 Logistic regression has been widely used in the finance industry to predict corporate failure and assess the performance of consumer credit. Recent
studies include Hand (2001), Westgaard and Van der Wijst (2001), Laitinen (1999) and West (2000).
7 We use measures of collectibility (‘Coll’) to derive the loan classifications. Coll-1, Coll-2, Coll-3 and Coll-4 represent 4 loan classifications: ‘current’,
‘sub-standard’, ‘doubtful’, and ‘loss’.
103
An Empirical Analysis of Credit Migration In Indonesian Banking
quality migrating to/ remaining in the loan classification ‘doubtful’ (coll-3). Step 3 of the logit analysis
calculates the probability of loan quality migrating to/ remaining in the loan classification ‘current’
(coll-1) or ‘substandard’ (coll-2).
STATISTICAL RESULTS
Using the coefficients obtained, we try to simulate the probability of credit migration as a function
of the GDP growth rate8 (see Exhibit 6 and the Appendix for the full regression results9 ). From the
simulation we can observe two interesting phenomena.
First, the external factor (GDP growth rate) is statistically significant in affecting loan status. In
can be seen that during an economic downturn, the probability of loan status being downgraded is
higher than in “normal” economic conditions (especially during the period 1998 to 1999 when economic
turbulence badly affected the Indonesian economy)(Bank Indonesia, 1999). From Exhibit 6(a), it can be
seen that in ‘normal’ conditions, there is only a small possibility of coll-1 type of loans becoming
valueless (coll-1 to coll-4). In a recession, the probability of coll-1 type loans being downgraded (coll-1
to coll-2) is slightly higher. Exhibit 6(b) shows that there is a higher probability of coll-2 type of loans
being downgraded or defaulted on (coll-2 to coll-3/4). Exhibit 6(c) shows very a high probability of coll-
3 type of loans being defaulted on (coll-3 to coll4). In a recessionary period, the probability of loans
being defaulted on jumps from about 30% to 80%. This phenomenon arises because companies with
poorer loan quality most probably perform less well financially. The lower the grade of loans, the higher
the possibility of the loans being downgraded or defaulted on. Mathematically:
8 We have also tried to include other (internal) factors like terms and loan sizes as independent variables in the regression process. Surprisingly, those
factors are not statistically significant in affecting the credit migration process; hence, those factors are excluded.
9 The regression is run under E-Views software package.
Table 1 : Probability Estimation Methodology
Loan
QualityStep 1
logit analysis
Step 2
logit analysisStep 3
logit analysisProbability
1
2
3
4
1-P(4) 1-P(3)1-P(2)
P(2)
P(3)
P(4)
))4(1))(3(1))(2(1()1(ˆ PPPP −−−=
)3())4(1))(3(1()2(ˆ PPPP −−=
)3())4(1()3(ˆ PPP −=
)4()4(ˆ PP =
104
Articles
)/()/( GDPycollGDPxcoll −>− ππ , where yx <
Then, )()( defaultedycollPdefaultedxcollP →−<→−
Second, there is irreversibility in the credit migration process. The loan status can easily be
downgraded during a recessionary period but upgrades during periods of growth are more difficult to
achieve. Exhibit 6(b) shows some possibility of coll-2 type of loans being upgraded (coll2 to coll-1).
From Exhibits 6(c) and 6(d), it can be seen that the lower the quality of loans, the lower the possibility
of the loans being upgraded (coll-3 to coll-2/1 or coll-4 to coll-3/2/1). Exhibit 6(d) shows almost no
possibility of loans in default being upgraded, even in normal conditions. Mathematically:
)()( xcollycollPycollxcollP −→−>−→− , where yx <
Intuitively, it can be understood that once a company falls into financial crisis or bankruptcy, it will
be difficult to rectify the situation since the company will struggle with a heavier financial burden.
SIGNIFICANCE OF THE RESULTS
The analysis conducted of the dynamic of the loan classification provides significant benefits in at
least three areas:
a. Cross checks – The banks sometimes have to optimize two different objectives simultaneously:
efficiency of operations and the accuracy of information on borrowers’ financial soundness. Too
intensive monitoring by the banks could raise monitoring costs to unacceptable levels; yet, at the
same time, the banks need an adequate level of accuracy in respect of the information gathered.
The analysis conducted shows how banks and their supervisors can reconcile ‘on-the-spot’-based
loan classification with a macro-index-based-estimation of loan classification.
b. Anticipatory action – Banks can also use this analysis for achieving a better loan diversification
strategy. Investment quality in a different sector of the economy may have a different level of
sensitivity to the changes in its industrial performance.
c. Understanding the nature of the credit migration process – Finally, the banks, as well as their
supervisors, will be better able to understand the behaviour of credit migration, including the
irreversibility of the migration of loan quality. As a result of this study, it is clear that deterioration
of loan quality in Indonesia cannot be rectified automatically during the period of growth. Therefore,
it is not possible to establish a direct function relating the level of non-performing loans to the level
of industrial performance.
105
An Empirical Analysis of Credit Migration In Indonesian Banking
(d) Probability of migration From coll 4 to 2,3,4
1
0,8
0,6
0,4
0,2
0
1995 1996 1997 1998 1999 2000
(a) Probability of migration From coll 1 to 2,3,4
c1p1 c1p2 c1p3 c1p4
(b) Probability of migration From coll 2 to 2,3,4
Probability of migration
0,5
0,6
0,4
0,2
0
0,3
0,1
1995 1996 1997 1998 1999 2000
c2p1 c2p2 c2p3 c2p4
(c) Probability of migration From coll 1 to 2,3,4
1995 1996 1997 1998 1999 2000
1
0,8
0,6
0,4
0,2
0
c4p1 c4p2 c4p3 c4p4
Probability of migration
c3p1 c3p2 c3p3 c3p4
1
0,8
0,6
0,4
0,2
0
1995 1996 1997 1998 1999 2000
Probability of migration
Probability of migration
CONCLUDING REMARKS
The ability to conduct sound risk assessment analysis is very important for the maintenance of the
sustainability of banking operations. Banks, as well as their supervisors, can use relevant information to
improve the accuracy of their estimates of credit migration and loan quality. One of the relevant
indicators to use, at least within an Indonesian context, in the assessment of asset quality is the GDP
growth rate. The indicator can be used to calculate the probability of migration of loan status for an
individual bank. Using the analysis, there are at least three benefits to be gained. First, the banks, as
well as their supervisors, are better able to assess asset values fairly. Second, the banks can avail
themselves of an additional tool to achieve a better loan diversification strategy. And, finally, the
analysis can be used to provide a better understanding of the nature of the credit migration process in
Indonesian banking.
106
Articles
APPENDIX
Regression Results
CM1
123-4
LOGIT // Dependent Variable is NCOLL
Date: 11/11/00 Time: 21:48
Sample: 1 764
Included observations: 764
Convergence achieved after 6 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C -4.548600 0.355388 -12.79898 0.0000
Log likelihood-44.43098
Obs with Dep=1 8
Obs with Dep=0 756
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
12-3
LOGIT // Dependent Variable is NCOLL
Date: 11/11/00 Time: 21:59
Sample: 1 756
Included observations: 756
Convergence achieved after 5 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C -3.119055 0.180641 -17.26663 0.0000
Log likelihood-132.5068
Obs with Dep=1 32
Obs with Dep=0 724
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
1-2
LOGIT // Dependent Variable is NCOLL
Date: 11/11/00 Time: 22:04
Sample: 1 724
107
An Empirical Analysis of Credit Migration In Indonesian Banking
Included observations: 724
Convergence achieved after 4 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C -1.821809 0.131865 -13.81568 0.0000
INDUSTRY -0.063886 0.014716 -4.341340 0.0000
Log likelihood-239.1804
Obs with Dep=1 78
Obs with Dep=0 646
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
INDUSTRY 5.654972 2.617949 6.021672
CM2
123-4
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 01:35
Sample: 1 112
Included observations: 112
Convergence achieved after 4 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C -1.774090 0.274923 -6.453037 0.0000
INDUSTRY -0.045604 0.027343 -1.667853 0.0982
Log likelihood-44.60844
Obs with Dep=1 16
Obs with Dep=0 96
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
INDUSTRY 1.719643 -1.787500 2.304167
12-3
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 01:42
Sample: 1 96
Included observations: 96
Convergence achieved after 3 iterations
108
Articles
Variable Coefficient Std. Error t-Statistic Prob.
C -0.974394 0.237616 -4.100706 0.0001
INDUSTRY -0.043322 0.025503 -1.698664 0.0927
Log likelihood-53.64330
Obs with Dep=1 25
Obs with Dep=0 71
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
INDUSTRY 2.304167 -0.252000 3.204225
1-2
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 01:48
Sample: 1 71
Included observations: 71
Convergence achieved after 2 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C 0.735707 0.253590 2.901169 0.0050
Log likelihood-44.71626
Obs with Dep=1 48
Obs with Dep=0 23
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
CM3
123-4
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 01:59
Sample: 1 93
Included observations: 93
Convergence achieved after 4 iterations
Variable Coefficient Std. Error t-Statistic Prob.
INDUSTRY -0.101144 0.028721 -3.521616 0.0007
Log likelihood-57.32033
Obs with Dep=1 35
109
An Empirical Analysis of Credit Migration In Indonesian Banking
Obs with Dep=0 58
Variable Mean All Mean D=1 Mean D=0
INDUSTRY 3.788172 0.791429 5.596552
12-3
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 02:04
Sample: 1 58
Included observations: 58
Convergence achieved after 3 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C 1.452252 0.334936 4.335904 0.0001
Log likelihood-28.17191
Obs with Dep=1 47
Obs with Dep=0 11
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
1-2
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 02:08
Sample: 1 11
Included observations: 11
Convergence achieved after 4 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C -2.302585 1.048799 -2.195449 0.0528
Log likelihood-3.350997
Obs with Dep=1 1
Obs with Dep=0 10
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
CM4
123-4
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 02:14
Sample: 1 100
Included observations: 100
110
Articles
Convergence achieved after 4 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C 2.313635 0.349425 6.621267 0.0000
Log likelihood-30.25378
Obs with Dep=1 91
Obs with Dep=0 9
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
12-3
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 08:20
Sample: 1 9
Included observations: 9
Convergence achieved after 3 iterations
Variable Coefficient Std. Error t-Statistic Prob.
INDUSTRY -0.072588 0.132714 -0.546949 0.5993
Log likelihood-6.083913
Obs with Dep=1 5
Obs with Dep=0 4
Variable Mean All Mean D=1 Mean D=0
INDUSTRY 0.777778 -0.160000 1.950000
1-2
LOGIT // Dependent Variable is NCOLL
Date: 11/12/00 Time: 08:25
Sample: 1 4
Included observations: 4
Convergence achieved after 3 iterations
Variable Coefficient Std. Error t-Statistic Prob.
C -1.098612 1.154700 -0.951426 0.4116
Log likelihood-2.249341
Obs with Dep=1 1
Obs with Dep=0 3
Variable Mean All Mean D=1 Mean D=0
C 1.000000 1.000000 1.000000
111
An Empirical Analysis of Credit Migration In Indonesian Banking
REFERENCES
Basel Committee (1999b). Credit Risk Modelling: Current Practices and Applications. Basel, April.
Basel Committee (1999a). Consultative paper on Principles for the management of Credit Risk. Basel, July.
Bank Indonesia (1999) Analisis Perbankan. Available on http://www.bi.go.id/ bank_indonesia2/moneter/
indi_perbankan/analisis_perbankan/
Brewster, D (1997). Business Economics: Decision Making and the Firm. The Dryden Press, Thames Valley University,
London.
Hand, D.J. (2001) ‘Modeling Consumer Credit’. IMA Journal of Management Mathematics, October, Vol.12, No.2.,
pp. 139-155.
JP MORGAN (1997). Credit MetricsTM – Technical Documents. New York: JP Morgan Securities).
Krolzig, H. and Sensier, M. (1998) ‘A Disaggregate Markov-Switching Model of The Business Cycles in UK Manufactur-
ing’. Discussion Paper no. 9812/1999, the School of Economic Studies, University of Manchester. Available on http://
nt2.ec.man.ac.uk/ses/discpap/
Laitinen, E. K. (1999) ‘Predicting a Corporate Credit Analysist’s Risk Estimate by Logistic and Linear Models – The
state of the art’. International Review of Financial Analysis. June, Vol. 8, No. 2, pp. 97-121(25).
West, D (2000) ‘Neural Network Credit Scoring Models’. Computers and Operational Research. September, Vol. 27,
No. 11, pp. 1131-1152.
Westgaard, S and Van der Wijst, N (2001) ‘Default Probabilities in a Corporate Bank Portfolio: A logistic model
approach’. European Journal of Operational Research. December, Vol. 135, No.2, pp. 338-349.
112
Articles
NEW BASEL CAPITAL ACCORD :Its likely impacts
on the Indonesian banking industry
A b s t r a c t
As proposed in the Second Consultative Package document from Basel Committee, the
New Accord has the objective of improving the soundness of financial system by underlining
attention to banks management and internal supervision, supervisory review process and market
discipline. The New Accord implementation requires completion of several conditions which
involves not only banks but supervisory authorities and market players as well. This includes
completing requirements – both qualitative and quantitative – in utilizing the New Accord
approaches. The utilization of some alternative approaches in the New Accord will not only
change banks behavior but will ultimately change perception towards banking business
environment and supervisory activities. Supervision authorities are required to conduct a forward
looking comprehensive analysis by focusing on risks confronting banking industry. Preliminary
analysis on the impact of the New Accord in Indonesian banking industry was conducted by
implementing Quantitative Impact Study (QIS) 3 survey. It was expected that the QIS 3 survey
could give information on the possibilities and impacts of the implementation of the New
Accord in Indonesian banking industry.
Authors:
Indra Gunawan, Bambang Arianto, G.A. Indira, Imansyah
113
New Basel Capital Accord
FOREWORD
Among recent topics that have been discussed heavily among stakeholders in Indonesian banking
industry is the readiness of banking industry to the future implementation of the New Basel Capital
Accord. The discussion arises as the Bank for International Settlement (BIS) released a statement that
they are in the process of establishing and calibrating a proposal to replace the 1988 Basel Capital
Accord. The proposal, known as the New Basel Capital Accord, is aimed to capture a comprehensive
perspective of risk in banking activities and expected to be more risk sensitive than the 1988 Basel
Capital Accord. In this case, the discussion is focused on the steps to be taken, either in banking
industry and supervisory authority, to foresee the future implementation of the New Basel Capital
Accord.
As stated in the January 2001 Consultative Document (CP2), the main objective of the New Basel
Capital Accord is to improve the soundness of financial system by granting more attention to bank
management and internal control, supervisory review process, and market discipline. In line with the
above objective, there are several preconditions that have to be met prior to the implementation of the
New Basel Capital Accord. Analysis of bank’s conditions should address any significant aspect that
affects its performance. In this case, consolidated basis approach would be the best option to capture
the overall condition of a banking institution. Another precondition is the implementation of risk based
analysis in the supervisory methodology. However, the heavily used quantitative aspects in risk based
analysis require supervisors to be adequately supported by statistical and econometric skills and should
also be accompanied by knowledge improvement as a basis to do judgment on risk based analysis.
In order to have some insight of the New Basel Capital Accord proposal, discussion on the condition
and challenges confronting both banking industry and supervisory authority would be beneficial. On top
of that, a quantitative impact study will give us an insight of essential measures to be taken prior to the
implementation of the New Basel Capital Accord.
DEVELOPMENT OF THE CAPITAL ACCORD
In July 1988, the Basel Committee on Banking Supervision issued a report titled International
Convergence of Capital Measurement and Capital Standards (The 1988 Basel Capital Accord) that consists
of two main recommendations; (i) the need for banking institutions (especially for internationally active
banks) to have a minimum capital ratio of 8% to minimize insolvency risk and create a level playing
field, and (ii) capital assessment using forward looking concept, which accommodate credit risk in the
banking book. It uses risk weight bucket (0%, 20%, 50%, and 100%) to fit various assets into certain risk
114
Articles
weight that depends on counterparty classification.
Even though the 1988 Basel Capital Accord had successfully promote stability in banking industry,
it has several weaknesses; (i) vast categorization of risk weighting, (ii) disregard the intention to have
portfolio diversification; (iii) create un-level playing field to non-bank financial institution; and (iv)
does not accommodate increasing risk in the financial and capital market.
Considering those weaknesses, in 1996 the Basel Committee on Banking Supervision released an
amendment of the 1988 Basel Capital Accord to incorporate market risk in the assessment of capital
requirement. The amendment consists of four components to be considered in assessing capital
requirement, e.g. equity risk, commodity risk, foreign exchange risk, and interest rate risk, and allow
bank to develop its own internal model in measuring market risk, subject to approval from supervisory
authority.
On the next step, the Basel Committee on Banking Supervision issued the First Consultative Package
on The New Accord (CP1) in June 1999, which was intended as first proposal to replace the 1988 Basel
Capital Accord. Major improvement in CP1 compared to the 1988 Basel Capital Accord is the methodology
to assess capital charge in a more risk sensitive manner. After receiving comments from various
stakeholders, the committee issued the Second Consultative Package (CP2) in January 2001 with some
refinement and calibration in the formula used to calculate risk weighted assets and capital charge.
According to the agenda setup by the Basel Committee on Banking Supervision, The New Accord will be
finalized and published in late 2003 and its implementation for the G-10 countries will begin in 2006
with three years transition period.
OBJECTIVES OF THE NEW BASEL CAPITAL ACCORD
As stated in previous section, the 1988 Basel Capital Accord uses one size fits all approach in
calculating capital requirement with credit risk factor as the sole element. In line with the dynamic
and a more complex environment in the financial system, banks are now confronting the increasing type
and exposure of risks. As an impact, bank needs a more sophisticated technique to deal with those risks
for the purpose of assessing its capital charge. Considering the vast development in financial system,
Basel Committee on Banking Supervision issued the proposal of the New Basel Capital Accord with
broader and a more complex coverage compared to the 1988 Basel Capital Accord. The proposal
attempts to align the need for adequate capital with risks confronting bank activities by providing
several approaches to measure credit risk, market risk, and operational risk. In addition, the proposal
also incorporates supervisory review process and market discipline as main elements in determining
115
New Basel Capital Accord
minimum capital requirement. Other major substance in the proposal is the flexibility for banks to
develop their own internal model to assess risks, subject to approval from the supervisory authority.
The main objectives of the New Basel Capital Accord proposal are (i) strengthening and improving
the soundness of financial system by maintaining current capital ratio, (ii) improving level playing field,
(iii) creating a more comprehensive approach to incorporate risks, (iv) providing banks with several
approaches to align capital requirement with various level of risks, and (v) focusing on internationally
active banks, even though its basic principles can be applied to all banks.
The New Basel Capital Accord proposal has three main pillars that are interrelated; minimum
capital requirements, supervisory review process, and market discipline. Partial implementation of
those pillars would impair the contribution of the New Basel Capital Accord to the soundness of financial
system. In this case, if a country could not implement the three pillars simultaneously then constructive
measures should be taken.
MAIN CHANGES OF THE CAPITAL ACCORD
The main substances that have not been changed since the implementation of the 1988 Basel
Capital Accord are the definition of capital and minimum capital adequacy ratio of 8%, including the
incorporation of Tier 2 Capital in the assessment of banks’ capital (maximum 100% of Tier 1 Capital). On
the other side, significant changes apply to the calculation of risk weighted assets, in which it now
incorporate credit risk, market risk, operational risk, and the use of consolidated basis principle.
Changes in credit risk include the availability of approaches to use, i.e. standardized approach and
internal ratings-based approach, and the introduction of credit risk mitigation techniques that cover
collateral, guarantee and credit derivatives, netting and asset securitization. Those changes are expected
to be incentives for banks in improving their risk management and administration of risk mitigation
factors. With regard to market risk, the 1996 amendment document of the 1988 Basel Capital Accord
remains unchanged. Therefore, approaches used in the market risk amendment would still be used in
the New Basel Capital Accord.
The newly introduced operational risk covers three approaches, i.e. basic indicator approach,
standardized approach and advanced measurement approach. Basic indicator approach uses one indicator
(referred as alpha) as a factor to obtain capital charge for operational risk, while standardized approach
uses several different indicators (referred as beta) for each business line. In addition, advanced
measurement approach uses banks’ internal loss data to estimate capital charge for operational risk.
116
Articles
Other changes included in the New Basel Capital Accord are the existence of pillar 2 - supervisory
review process and pillar 3 - market discipline. Supervisory review process requires internal control
within a bank to be in place and supervisory authority has to evaluate its effectiveness. In addition,
intensive dialogue between banks and supervisory authority is a compulsory. With regard to pillar 3,
proper disclosure of financial information is the main objective of market discipline. In this case, an
effective disclosure would be a fundamental aspect to ensure that stakeholders could evaluate risk
profile of a bank and its corresponding capital adequacy.
THE SUBSTANCES OF NEW BASEL CAPITAL ACCORD
Pillar 1: Minimum Capital Requirement
In the New Basel Capital Accord proposal, risk weighted assets comprises of three components, i.e.
risk weighted assets for credit risk plus risk weighted assets for market risk and operational risk.
Total capital= CAR (min. 8%)
RWA for Credit risk, Market risk, Operational risk
Credit risk
There are 2 (two) alternative approaches available within the credit risk framework; standardized
approach and internal ratings-based approach (IRB). Basically, credit risk assessment in the standardized
approach is the same as the 1988 Basel Capital Accord. The main difference is on the risk weighting
application, which uses official rating issued by recognized rating agencies. The rating is then converted
into corresponding risk weight of 0%, 20%, 40%, 50%, 75%, 100% and 150%.
Unlike the standardized approach, the internal ratings-based approach has two approaches;
foundation internal ratings-based approach and advanced internal ratings-based approach. In foundation
approach, banks are allowed to have their own estimate only for probability of default while in advanced
approach they are allowed to estimate all factors of risk weighted assets; probability of default, loss
given default, and exposure at default. Implementation of both approaches requires approval from
supervisory authority.
117
New Basel Capital Accord
Market risk
Approaches used to assess market risk capital charge are remains unchanged from those in the
1996 amendment document. The standardized method uses building block approach that calculates
specific risk and general market risk, while internal model enables banks to use their own risk management
model in assessing capital charges. To be able to use internal model, a bank has to fulfill qualitative and
quantitative requirements and also receive written approval from supervisory authority.
Operational risk
There are three approaches used in the calculation of operational risk capital charge in the New
Basel Capital Accord; basic indicator approach, standardized approach, and advanced measurement
approach. Basic indicator approach uses alpha factor as proxy for overall risk exposures and multiply it
with banks’ net operating and non-operating income. In general, alpha factor is approximately 20% of
regulatory capital. Banks that meet minimum requirements can use standardized approach instead of
basic indicator approach. The standardized approach acknowledges eight types of business line with
different beta factor for each business line. In this approach, total capital charge is the summation of
capital charges for each business line.
Another approach, the advanced measurement approach, requires banks to meet stringent
requirements. There are three types of data needed in relation with business lines and type of risks;
operational risk indicator data, probability of event data, and loss given event data. Within the advanced
measurement approach, capital charge for each business line is obtained by multiplying those data with
gamma factor determined by the Basel Committee on Banking Supervision, based on industry aggregation.
Furthermore, total capital charge for operational risk is equal to the summation of capital charges for
each business line.
Pillar 2: Supervisory Review Process
The second pillar of the New Basel Capital Accord has the objective to ensure that all banks have
adequate internal processes to assess their capital adequacy, which is based on comprehensive risks
evaluation. In addition, supervisory authority has to take responsibility of evaluating banks’ internal
processes that includes assessment of measures to be taken to anticipate correlation among risks.
However, supervisory review process is not meant to replace judgment and professionalism of banks’
management, or to handover the responsibility of meeting capital requirement to the supervisory
118
Articles
authority. On the contrary, banks’ management would still be the sole party that has a better
understanding of banks’ risk profile and takes responsibility in managing risks. Moreover, capital adequacy
could not be used as a replacement for lack of supervision or adequate risk management processes of
banks’ management. Supervision of banks’ compliances to existing regulation should also be conducted
by using any necessary means, including on-site and off-site supervision/examination and intensive
discussion with banks’ management.
Pillar 3: Market Discipline
The third pillar of the New Basel Capital Accord can be considered as part of supervision effort to
improve bank and financial system soundness. In this context, transparency is believed to be beneficial
for stakeholders and promote an effective market discipline. Moreover, market discipline could encourage
banks to run their business in a sound and efficient manner. On top of that, adequate public disclosure
regime could perform as supervisory means to encourage banks to measure every risk appropriately,
maintain a safe capital level, and develop and maintain the soundness of risk management. In this
regard, disclosure acts as the main buffer for minimum capital requirement of pillar 1 and improves
supervisory review process of pillar 2. Information to be disclosed should be in timely manner and
sufficient enough for stakeholders to assess risks in banks’ activities. Among several characteristics of
disclosure are materiality, proprietary information, frequency, and comparability.
CREDIT RISK ASSESSMENT
Standardized Approach
The standardized approach is intended to align capital calculation with key element of risks by
providing broader risk weight classification and recognition of credit risk mitigation techniques. The
main difference from the 1988 Basel Capital Accord is the use of rating from recognized rating agencies
in assigning appropriate risk weight to certain exposure. Moreover, the classification of OECD and Non-
OECD countries has been eliminated.
119
New Basel Capital Accord
1988 Basel Capital A c c o r d 19 8 8
Risk Weight Type of Claim
0%
0,10, 20 or 50%
(national discretion)
50%
50%
100%
- Cash.
- Claims to Government and Central Bank in domestic currency and financed with the
same currency.
- Other claims to Government of OECD countries and their Central Banks.
- Claims with cash collateral of securities issued by Government of OECD countries or
guarantee by Government of OECD countries.
- Claims to domestic public-sector entities, other than Central Government, and loans
secured by those entities.
- Claims to multilateral development banks (IBRD, IADB, AsDB, AfDB, EIB) and bills secured
by or own securities collateral issued by those banks.
- Claims to banks in OECD countries and loans guaranteed by bank in OECD countries.
- Claims to banks outside OECD countries with remaining time less than 1 (one) year and
loans with remaining time less than 1 (one) year secured by banks outside OECD countries.
- Claims to non-domestic OECD public-sector entities, other than Central Government,
and loans secured by those entities.
- Cash in the process of collection.
- Loans fully secured by mortgage on residential property which will be used or rented by
debtors.
- Claims to private sector.
- Claims to banks outside OECD countries with remaining time less than 1 year.
- Claims to Central Government of non-OECD countries (exception in domestic currency
and financed with the same currency).
- Claims to commercial companies owned by public.
- Lands, buildings and equipments and other fixed assets.
- Real estate and other investments (including non-consolidated investment participation
of other companies).
- Capital instruments issued by other banks (unless it is issued from capital).
- Other assets.
Ratings issued by rating agencies are becoming more important to assign appropriate risk weight
to certain exposure. In line with the importance of rating agencies, their qualification has to be
recognized by supervisory authority. Some criteria used as reference in determining the eligibility of
a rating agency are objectivity, independency, transparency/international access, disclosure, resources,
and credibility.
120
Articles
TYPES OF CLAIM RATINGPERINGKAT
Claims to Government/Central Bank 0% 20% 50% 100% - 150% 100%
Claims to Bank/State-Owned Entities
- Option I 20% 50% 100% 100% 150% 100%
- Option II 20% 50% 50% 100% - 150% 50%
Claims to Private Sector 20% 50% 100% - 150% - 100%
Residential Mortgage 40%
Retail 75%
Asset securitization and other assets 100%
High risk assets (Rating below BB-/B-, Past due, and loans uncovered by collaterals 150%)
Off-balance sheet: credit exposure calculation method from 1988 Basel Capital Accord is applied
STANDARDIZED APPROACH
AAA to AA- A+ to A- BBB+ to BBB- BB+ to B- Under BB- Under B- Unrated
Compared to the 1988 Basel Capital Accord, the standardized approach has some advantages in
credit risk assessment; (i) provide balance between simplicity and accuracy by giving chances to banks
to calculate its capital requirements based on a sound risk management practices; (ii) more risk sensitive
as shown by wider risk weight classification for each type of claim and endorse the calculation of
economic capital for assessing capital requirements; (iii) a more transparent credit risk assessment by
using rating issued by rating agency to determine risk weight, while at the same time resolving dispute
between bank and market players in assigning proper risk weight for certain exposure; (iv) more emphasis
on economic consideration rather than political influence by eliminating sovereign floor limitation; (v)
process of recognizing rating agency by supervisory authority can be utilized to collect information and
methodology comparison among rating agencies in evaluating creditworthiness of an institution. It may
expand the insights and expertise of supervisory authority in assessing credit risk.
Yet, besides its advantages, the standardized approach has the potential of creating various problems
in its implementation as well. One of them is related with the assignment of risk weight that utilizes
rating from rating agencies. The New Accord implementation will affect the capital requirements,
especially with sovereign and interbank exposures that rated below investment grade. Under 1988
Basel Capital Accord, sovereign and interbank exposures are assigned 0% and 20% risk weight respectively,
while in the standardized approach the risk weight can vary between 0% and 20% to 150%. For example,
if Indonesia has sovereign rating of CCC+, then claims to Indonesian Government have the risk weight of
150%. Considering the present condition with recap bonds form 36,03% of total banking assets, under
the standardized approach banks should hold more capital rather than 1988 Basel Capital Accord.
121
New Basel Capital Accord
Other problems confronting supervisory authority in recognizing rating agency are related to its
competence to assess a qualified rating agency, monitor the fulfillment of those qualifications
continuously, moral obligation in case the acknowledged rating agency do not satisfy those qualifications.
Another dilemma is the different definitions and rating symbols used by different rating agencies. For
instance, S&P uses symbol AAA, AA+, etc. and Moody’s uses symbol Aaa, Aa1, etc., while in fact those
symbols have similar definitions. In this case, the challenge faced by supervisory authority is to make a
map and rating result conversion of all recommended rating agency to avoid differences in the result of
credit risk assessment. Therefore, an objective and consistent “standard assessment framework” is
needed to resolve different assessment standards from all rating agencies.
The other implication is related with solicited and unsolicited rating. In practice, rating agency
may issue a rating for an entity or instrument in financial/capital market based on request (solicited
rating) or without any request from respected parties (unsolicited rating). The issue of solicited/
unsolicited rating has the potential of creating moral hazard to rating agency to force the use of solicited
rating to institutions assessed.
Using rating as references also has the potential to create problems that arises from the impact
of pro-cyclicality. Theoretically, and historically, economic condition of a country will always
experience cycles. As a consequence, the rating result will be affected by the economic cycle (pro-
cyclicality). When the economy is good (boom), rating agency will give a good assessment and it
decrease banks credit risk exposure and increase capital adequacy ratio (overcapitalized). On the
contrary, when the economy is in recession, credit risk exposure will increase as a result of low
rating and capital ratio will decreases (undercapitalized). The problem becomes even worse when
overcapitalization motivate banks to expand their loan aggressively. This is creating another problem
in banks in the event of downturning economy. The capital they hold would be less than required as
an impact of increasing impaired loan. Considering the above impact, an effective supervisory
process is essential to ensure that banks have enough additional capital as reserve (buffer) to
anticipate the movement in economic cycle.
Other potential problem arises from limited recognition for collateral as credit risk mitigation
factor. As implemented in other developing countries, collateral structure in Indonesia is dominated
with lands and buildings, especially collaterals provided by Small and Medium Enterprises (UKM). In this
case, standardized approach will discourage banks to extend loans to UKM sectors because physical
collateral provided by UKM can not be utilized as risk mitigation factor.
122
Articles
Internal Rating-Based Approach
This approach has the advantage of risk sensitivity in calculating capital requirement and encouraging
the completion of bank risk management to obtain a lower amount of capital. However, the approach
requires bank to fulfill certain requirements and consistency principle in its implementation towards
banks’ significant portfolios and business lines. In general, there are five main elements in internal
ratings-based approach; exposure classification, risk component for every exposure, risk weight for
every risk component, minimum requirement, and supervision of the compliance to the minimum
requirements.
Exposure Classification
Bank for International Settlements (BIS) has determined 7 (seven) exposure classifications; corporate
exposure, bank exposures, retail exposures, sovereign exposures, specialized lending exposures, project
finance exposures, and equity exposures. From those exposure classifications, there are five exposure
classifications that have been defined and have the concept of minimum capital allocation. They are
corporate exposure, bank exposures, retail exposures, sovereign exposures, and specialized lending
exposures. Other classifications are still under research and planned to be included in the third
consultative documents which will be released in the first quarter of 2003.
Risk Component and Default Definition in Internal Ratings-Based Approach
There are four risk components in every exposure classification; i.e. probability of default (PD),
loss given default (LGD), exposure at default (EAD), and maturity (M). Probability of default (PD) is the
probability of an obligor and or guarantor to experience default, which is estimated through historical
data. Loss given default (LGD) is the estimated loss that could happen in the event of default. Exposure
at default (EAD) is the estimated outstanding of exposure in the event of default. Maturity (M) is related
with exposure time frame, where the longer period has the bigger maturity risk. In internal ratings-
based approach, an obligor is declared as default if one or more criteria is met: (i) It is confirmed that
the obligor cannot fulfill his debt obligation in full (principal, interests or fee); (ii) There is a credit loss
event related with obligor’s obligation, such as charge-off, specific provision, or is forced to accept
restructuring which includes reduction or postponement of principal, interests or fee; (iii) There is a
past-due of more than 90 days; (iv) The obligor file bankrupt request or similar protection from its
creditors.
123
New Basel Capital Accord
Approaches in Internal Ratings-Based Approach
There are 2 (two) methods in this approach, foundation approach and advanced approach. Foundation
approach is intended for banks that experience difficulties in estimating its valid risk factors but can
obviously determine its obligors’ default risk and meet minimum requirements related to the internal
rating system, risk management process, and risk component estimation. In this case, bank can utilizes
its estimation to determine probability of default and uses the guideline from the supervisory authority
to determine loss given default (LGD), exposure at default (EAD), and maturity (M). The other approach,
advanced internal ratings-based approach, is intended for banks that can estimate obligor default risk
and other risk components consistently, by paying attention to minimum requirements and additional
minimum requirements fulfillment for each of the estimated risk components. In this approach, banks
are allowed to estimate all risk components (PD, LGD, EAD, and M).
The important thing that should be taken into account from the above two approaches is that the
implementation of both approaches should accompanied by validation and written approval from the
supervisory authority. On top of that, it should be tested in parallel with the standardized approach to
have comparison on its results.
Minimum Requirements of Internal Ratings-Based Approach
Implementation of internal ratings-based approach requires banks to comply with minimum
requirements established for every exposure classification and meet additional minimum requirements
for every risk components. Minimum requirements for advanced approach are more complex compared
to foundation approach because the implementation of advanced approach requires banks to have a
more reliable risk management.
The internal ratings-based approach concept has its own strengths and weaknesses. One of the
strengths is the ability to accommodate different risk characteristics in calculating minimum capital
requirements. In this case, higher risk will require more capital allocation. This is in turn encourages
banks to fix and enhance its risk management. Other strength is the wider collateral scope compared to
the standardized approach. Within internal ratings-based approach, banks have more flexibility in utilizing
types of collateral to mitigate risk.
The weaknesses in internal ratings-based approach are mostly related with, among others, the
requirements to have a complete historical data – at least the last 3 to 5 years – and as much as possible
the data should include business cycle in a normal economy condition (not in crisis period). For banks
124
Articles
that do not have a comprehensive database system, this requirement could be a problem and needs
time, efforts and expensive cost to improve the database. Other weakness is its implementation process
that requires a relatively long time horizon to prepare the database, establish structure and rating
criteria, model determination, and rating system supervision and its repeated evaluation with some
adjustment to adopt changing environments.
Considering the tight minimum requirements and the present condition of Indonesian banking
industry, the internal ratings-based approach is unlikely to implement in Indonesia in the near future.
This is based on the fact that most Indonesian banks do not have a comprehensive obligor database with
sufficient data of the latest 3-5 years time frame. It is estimated that compliance to the requirements
in the internal ratings-based approach could take more than 5 (five) years, especially due to the expensive
cost and human resources needed to improve the database. Moreover, post-crisis economy still has not
reached normal condition. In this case, data collected from abnormal condition will lead to bias probability
of default and it may cause errors in determining rating obligor/guarantor and calculation on banks’
capital adequacy.
From supervisory perspective, the implementation of internal ratings-based approach will resulted
in a specific risk characteristic for a bank and therefore one bank will have different characteristic from
other bank. As a consequence, supervisor needs to have in-depth understanding of the bank activities.
It will need a dedicated bank supervision system. Furthermore, the need to have prior validation from
supervisory authority will confront supervisor to have in-depth knowledge and competency on statistics
and econometric analysis.
APPROACHES IN MARKET RISK CALCULATION
Standardized Method
The objective of Amendment to the Capital Accord to Incorporate Market Risks issued by BIS on
January 1996 was to encourage banks in providing enough capital against price changing risk in its
trading activities. The Amendment 1996 re-emphasized that the required capital to cover market risk
should comprise of stockholder capital and retained earning (Tier 1) and supplementary capital (Tier 2).
But banks are allowed to add capital for Tier 3 specified to cover market risk. Tier 3 comprises of short-
term subordinated loan with minimum of 2 (two) years and has lock-in clause which means that they are
not allowed to pay interest or principal (even on due time) if it will cause CAR to drop below the
minimum level. Thus, Tier 3 calculated amount could also be limited to 250% of Tier 1.
125
New Basel Capital Accord
Market risks that should also be calculated as determined in Amendment 1996 are market price
change risk of financial instruments (such as bonds, swap, future, options, etc.) or equities sensitive to
interest rate changes on bank’s trading book; interest rate risk of financial non-derivatives instruments;
foreign exchange risk and risk on holding position in commodities (agriculture product, non-oil mineral,
and non-gold precious metal) on overall bank business activities, both trading book and banking book.
There are 2 (two) approaches in calculating market risk they are standardized method and internal
model. Standardized method focus the measurement to 5 (five) types of market risk component, interest
rate risk, equity position risk, foreign exchange risk, commodities risk, and price risk (for option). The
approach used is called “building-block” by calculating capital charge for specific risk and general
market risk of the traded financial instrument positions (trading book) and all foreign currency and/or
commodity positions (both trading book and banking book). Specific risk is price change risk of financial
instrument due to the issuer factor, while general market risk is price change risk of financial instrument
due to general market fluctuation factor. These two types of risk are subject to different capital charge
calculation. Total additional capital charge to cover market risk (on top of capital charge to cover
credit risk according to 1988 Basel Capital Accord) is arithmetic total of each capital charge subject for
each of the 5 (five) risks mentioned above.
Interest Rate Risk
Capital charge calculation for interest rate risk is divided into 2 (two) type of risks, specific risk
and general market risk. In calculating specific risk, offsetting is only possible for identical positions
(issuer, rate, maturity and other same features). Finance instruments exposed to interest rate risk are
all fixed-rate and floating-rate debt securities and other instruments with similar characteristics but
not classified as securities with KPR-based (mortgage securities). Specific risk application is divided
into 5 (five) general categories with 5 (five) capital charges weight classification: (i) Government (0,00%);
(ii) Qualifying (0,25%, 1,00%, and 1,60%); and (iii) Other (8,00%).
The “government” category includes all securities issued by government such as bonds, treasury
bills, and other short-term instruments. Supervision authorities have the right to impose a certain
percentage to specific risk for securities issued by other foreign government. The “qualifying” category
includes securities issued by state-owned entities, multilateral development banks, and other securities
with rating investment grade issued by rating agency acknowledged by banking supervisor. While “other”
category includes all other securities not included in “government” and “qualifying” categories and are
subject to the highest charge of 8%.
126
Articles
General Market Risk
There are 2 (two) alternative options to calculate capital charge for general market risk, maturity
method and duration method. In these two methods, the capital charge total is a total of 4 (four)
components: (i) Net short or long positions of all trading book; (ii) Matched position in all time-band
(vertical disallowance); (iii) Matched position between time-band (horizontal disallowance); (iv) Net
charge for option position. All bank positions are then sorted (slotting) according to remaining time
frame (fixed-rate instruments) or the next repricing date (floating rate instruments) into 15 time-band
which are divided into 3 (three) zones:
The calculation is conducted according to these following steps:
1. Multiplying all position in every time-band with risk weight reflects price sensitivity caused by
interest rate changes;
2. Offsetting position in every time-band, where the smaller position, both long or short, is subject to
10% capital charge (vertical disallowance);
3. Horizontal disallowance, a residual value (long and short difference from the second step) is offset-
ed with other time band position in the same zone, and then with the residual value between
different zones. Each offset result is subject to capital charge with calculation as follow:
Coupon > 3% Coupon < 3% Risk Weight Asumption Yield Change
Zone 1 :
0 – 1 month
1 – 3 month
3 - 6 month
6 – 12 month
Zone 2 :
1 – 2 month
2 – 3 month
3 – 4 month
Zone 3 :
4 - 5 month
5 – 7 month
7 – 10 month
10 – 15 month
15 – 20 month
> 20 month
Zone 1 :
0 – 1 month
1 – 3 month
3 - 6 month
6 – 12 month
Zone 2 :
1.0–1.9 month
1.9–2.8 month
2.8–3.6 month
Zone 3 :
3.6-4.3 month
4.3–5.7 month
5.7–7.3 month
7.3–9.3 month
9.3–10.6 month
10.6-12 month
12-20 month
> 20 month
0.00%
0.20%
0.40%
0.70%
1.25%
1.75%
2.25%
2.75%
3.25%
3.75%
4.50%
5.25%
6.00%
8.00%
12.50%
1.00
1.00
1.00
1.00
0.90
0.80
0.75
0.75
0.70
0.65
0.60
0.60
0.60
0.60
0.60
Maturity Method : Time-Band and Risk Weights
127
New Basel Capital Accord
Capital Charge for Horizontal Disallowance
Zone Time-Band Within Inter Zone in Between Zone 1
Zone Sequence Manner and 3
1
2
3
40%
30%
30%
40%
40%100%
0-1 month
1-3 month
3-6 month
6-12 month
1-2 month
2-3 month
3-4 month
4-5 month
5-7 month
7-10 month
10-15 month
15-20 month
> 20 month
Both tables above refer to maturity method, while duration method cannot differentiate coupon
above and below 3% and vertical disallowance of only 10%. Total capital charge amount for those three
steps is the capital charge for market risk.
Interest Rate Derivatives
Interest rate risk measurement should also calculate all interest rate derivatives instruments and
off-balance-sheet instruments in trading book, such as forward rate agreements (FRAs), forward contract,
bond futures, interest rate/cross –currency swap, and forward of foreign currency. Each of those
instrument positions has to be converted according to its underlying transaction and is subject to
specific and general market risk calculation. Capital charge calculation for this kind of instrument is
relatively more complex.
Equity Position Risk
This risk occurs if bank has or take position in equities in trading book. The equities meant here
are common stock, convertible stock/securities, and commitment to buy/sell those equities.
Foreign Exchange Risk
This risk occurs if bank has or take position in foreign currency, including gold. Standardized model
introduced shorthand method in calculating capital charge for this risk, that is: (i) Capital charge for
128
Articles
foreign exchange risk is 8% of total net open position (local currency) for foreign currency and gold; (ii)
Net open position is a total or Net short position or long position (which is the greater) and Net (without
differentiating long or short) position in gold.
Commodities Risk
This risk occurs if bank has or take position in commodities such as agriculture products, mineral
and precious metal (besides gold). But the same as equity position risk, banking in Indonesia is still not
allowed to do commodities trading.
Option
Option is a contract that causes right (but not obligation) for its holders/buyers to buy or sell a
certain amount of currency or other finance instruments with a price agreed upon on or before a certain
date in the future. The risk for buyer is limited only to option premium that is paid plus fee to broker.
But for seller (writer) the risk is unlimited because it is determined by the difference of market price
and the agreed price (strike price). Therefore, bank that sells option will face a bigger risk compare to
the option buyer bank. Standardized model give some alternatives they are (i) Bank only as option
buyer can use simplified approach; (ii) Bank that also sell/write option is expected to use intermediate
approach or a comprehensive risk management model. Basel Committee thinks that the more significant
for a bank to do trading option the more it has to use a comprehensive risk assessment method.
Capital charge calculation for option with simplified approach can be seen as follow:
Cash Position Option Position
Long
Long
Short
Short
-
-
Capital Charge Estimation
Long put
Long call
Long call
Long put
Long call
atau
Long put
Capital charge = (Market value of underlying transaction/securities x the
amount of capital charge for specific and general market risk of the
underlying securities) – the value of “in the money” option (if any)
Similiar as above
Similiar as above
Similiar as above
Capital charge is determined between the lowest amount of :
• Market value of underlying securities x the amount of capital charge
for specific and general market risk of underlying securities
• Market value of the option
Simplified Approach – Capital Charge untuk Option
129
New Basel Capital Accord
Internal Model
Quantitative and Qualitative Requirements
The Basel Committee allows bank to use internal model other than the standardized model to
fulfill qualitative and quantitative requirements determined by Basel Committee and after regaining
approval from banking supervision authorities. The quantitative requirements are: (i) Using value-at-
risk (VAR) method which daily calculated with 99% and one-tailed confidence interval; (ii) Price shock
standard used in the model is minimum 10 trading days so the minimum holding period is also the same
with the period; (iii) The model use observe-resulted historical data of a minimum 1 (one) year; (iv) The
amount of capital charge for bank using internal model the bigger one of yesterday’s VAR value or 3
(three) times daily VAR average for the latest 60 working days.
Meanwhile, the qualitative requirements determined by BIS, among others fulfilling general criteria
of adequate risk management system; possess qualitative standard in the event there is a mistake in the
internal model; possess guidelines for a sufficient market risk factor categorization; possess guidelines
for stress testing; possess validation procedure for external error in model utilization; and possess a
clear rule in the even the bank use a combination of internal model and standardized method.
VaR Concept in Internal Model
In general, internal model used by banks is based on Value-at-Risk (VAR) concept. VaR is an approach
to measure loss amount occur on a portfolio position as a result of risk factors changes including price,
interest and exchange rate for certain period by using certain probability level. VaR application method
in internal method require the risk factors changes data to calculate the amount of overall risks faced
by a bank in a certain point of time. Thus, it is also needed to do volatility analysis that is a projection
of risk factor changing in calculating position in portfolio.
There are 2 (two) types of volatility they are historical volatility and implied volatility. Historical
volatility is a volatility based on time series data, while implied volatility is applied for option (non-linear
instruments) calculated by inputting option price into option pricing model like the Black-Scholes formula.
Approaches in Operational Risk Calculation
Basic Indicator Approach
Basic Indicator Approach is a very simple approach and can applied to all banks, but it is more
130
Articles
appropriate for small-scale banks with small business activities variation. This approach uses a certain
financial indicator in calculating risk profile, namely gross income. Next, the amount of a (alpha
factor) is determined which is a multiplication factor to forecast the amount of operational risk, in this
case is the function of gross income. Based on researches conducted by Basle Committee, it is forecasted
the operational risk reach 20% of minimum capital maintenance obligation, so the a factor estimation
is 30% of gross income.
Standardized Approach
It is an approach using combination between financial indicator and bank business line which
determined by supervisor in determining capital charge. This approach divides business units and activities
into several groups and determines indicators for each group to reflect various risk profile of each of
those business activities. The substance of this approach is to calculate b factor (beta factor) for each
business unit which reflects the proxy amount of losses relations from operational risk and each business
line activities which is represented by certain financial indicator. Example of a bank business unit and
activity division can be seen as follow:
Business Level 1 Level 2
Unit
Banking
Others
Activity
Retail Banking
Commercial
Banking
Payment &
Settlement
Agency Services
Trading & Sales
Retail Banking
Card Services
Commercial Banking
Custody
Corporate Agency
Corporate Trust
Sales
Market Making
Proprietary
Positions
Treasury
Lend and Deposit fund services, trustee
Issuing and Managing Credit Card
Provide fund (e.g. loans, letter of guarantee, acceptance
letter, securities negotiations), trade finance, factoring etc
Payment and Collection, Transfer fund, Clearing and
settlements
Escrow, receiving deposits
payment Agent and Issuer
Trustee activities
Selling securities, equity participation, foreign exchange
activities etc.
131
New Basel Capital Accord
Calculation Method
The calculation of operational method is by determining b factor for each business unit determined
by supervision authorities. Next weighting is determined for each business unit.
Retail Banking
Commercial Banking
Payment & Settlement
Agency Services
Trading & sales
Total
Business Unit Weight Interval (%)
21 - 31
16 - 24
15 - 22
10 - 15
18 – 28
80 – 120
Mathematically, the b factor calculation for each business unit is as follow:
βββββ =( 20% x Capital Adequacy Ratio x Weight)
Financial Indicators
[By using 20% ratio of [Kewajiban Penyediaan Modal Minimum] (Minimum Capital Supply Obligation)
as required capital amount standard to anticipate operational risk than the amount will be allocated to
all business units. Capital allocation for each of those business lines is then divided with financial
indicator to get the b factor number.
Retail Banking
Commercial Banking
Payment & Settlement
Agency Services
Trading & sales
Business Unit Financial Indicators
Average Assets/year
Average Assets/year
Number of settlement/year
Gross Income
Gross Income
Advanced Measurement Approach
This approach is applied when a bank has a comprehensive database so it can determine the type
of loss related with operational risk (loss types), make probability estimation of the loss (probability of
loss event), and estimation of proportion amount of a transaction or exposure that may inflict loss (loss
132
Articles
given event). This approach tries to combine the past two approaches they are supervision importance
and loss data of each bank in calculating the bank capital need.
Bank ability to fulfill certain criteria will determine which criteria it will use in calculating the
required capital to cover operational risk. This approach demands bank to have loss event database for
each business line which can be used to determine calculation parameters that is parameter that reflects
the probability of a loss on each business line (Probability of Loss Event/PE) and the loss amount that
may occur in each business line (Loss Given Event/LGE). Based on those parameters we can obtain loss
expectation on each business line (Expected Loss/EL) by multiplying PE, LGE and indicator of each
business line. Next, loss expectation on each of those business lines will be multiplied with certain
percentage (g factor) to determine capital amount a bank need to reserve as anticipation to operational
risk.
Measurement Method
Steps in implementing this approach are:
a. Bank will classify its activities into various business activities, determining operational risk of each
of those business units, also determining financial indicator to become proxy of the amount of each
business unit operational risk exposure (Exposure Indicator/EI). In practice, supervision authorities
will determine business unit standard, risk type and financial indicator for bank references;
b. By using their own loss database, the bank determine a parameter to reflect the probability of loss
for each business unit (Probability of Loss Event/PE) and the loss amount that might occur on each
business unit (Loss Given Event/LGE);
c. Based on those parameters they can obtain loss expectation of each business unit (Expected Loss/
EL) that is EI x PE x LGE;
d. In capital calculation context, authorities will determine g factor (gamma factor) that is a [konstanta]
(constant) used to transform EL value into capital amount a bank has to form (capital charge), g
factor can be define as maximum loss amount in every holding period in a certain confidence
interval.
Although capital charges calculation for operational risk in the New Accord has accommodated
several approaches but there are still some things need to be completed especially in relation with
quantitative criteria. Some of those problems are related with no clear definition that differentiates
credit risk and market risk. It is especially related with a clear limitation with other risks and as a
133
New Basel Capital Accord
foundation for quantification of relevant and credible operational risk. Thus, the impact amount of
operational risk might be very much determined by the bank supervision quality. Operational risk
calculation not entirely related to the bank losses historical data. The main issue is whether the supervision
mechanism conducted is good or not. Management responsibilities cannot be measured by the amount
of capital charge. In contrast, the quality of forward-looking control is the main thing compare to losses
historical data used to calculate operational risk.
Other critic comes from a very different operational risk characteristic if compared with credit
risk and market risk. Operational risk emphasize more on internal bank, context-dependent, highly
multifaceted, interdependent and is not determined or evaluated based on market value. Some of
operational risk components – especially those not related with overall context – can be calculated with
credit risk model. The most important component of operational risk for management – events seldom
happen but has a big impact – is not included in capital calculation. The limitation use in credit risk and
market risk models may not always suitable to calculate those risks.
Case Study: Quantitative Impact Study (QIS) 3 in Indonesia
To do calibration and completion of the New Accord proposal, the Basel Committee compose a
Quantitative Impact Study (QIS) 3 as a comprehensive field test to get required information to for the
calibration process so it can minimize capital charge due to uncertainty factor. The objective of QIS 3
is to collect data which is related with the bank capital calculation (Bank’s capital requirements) and
also fulfilling industry demand for the New Accord to be supported quantitatively. This survey is
conducted by collecting banking data/information of G-10 countries and other countries (± 40 countries
participate with participating bank amount reach ± 200 banks) with different risk profiles background.
In conducting QIS 3 survey in Indonesia, they use a workbook standardized approach based on
consideration that generally Indonesian banking has not fulfilled some pre-requirement for internal
rating-based approach implementation (for instance information technology, human resources,
understanding of the estimation models in IRB, etc.) Thus, Bank of Indonesia as supervision authorities
is considered of not having the ability and expertise to do analysis and validation to the estimation
models in IRB yet.
Summary of the QIS 3 survey is presented in the next table:
134
Articles
A 41,961,945 7,145,988 17.03% 63,906,561 6,676,189 10.12% -6.91%
B 46,151,502 6,782,551 14.70% 53,631,273 4,652,619 11.64% -3.06%
C 71,202 25,753 36.17% 133,733 739 19.15% -17.02%
D 23,463,230 8,066,973 34.38% 42,354,101 3,136,022 17.73% -16.65%
E 20,651,071 5,365,220 25.98% 33,036,296 8,416,231 12.94% -13.04%
F 8,416,231 2,793,321 33.19% 15,567,909 9,687,591 11.06% -22.13%
G 9,906,567 3,272,683 33.04% 15,519,475 965,339 19.85% -13.85%
H 9,582,746 1,600,303 16.70% 16,219,333 4,413,074 7.79% -8.91%
I 4,120,895 396,081 9.61% 6,666,523 2,241,046 4.36% -5.25%
J 2,339,069 300,884 12.86% 1,944,429 945,183 10.41% -2.45%
K 5,178,329 1,038,063 20.05% 6,349,300 558,526 15.03% -5.02%
L 1,820,841 399,165 21.92% 2,229,655 482,008 14.72% -7.20%
M 9,131,799 2,301,156 25.20% 16,993,513 - 13.54% -11.66%
N 14,717,161 3,057,044 20.77% 19,603,039 2,028,453 14.13% -6.64%
O 7,219,213 918,179 12.72% 9,448,912 977,346 8.81% -3.91%
P 5,352,087 772,772 14.44% 8,312,929 385,672 8.88% -5.56%
Q 8,360,559 1,380,795 1 16.52% 10,604,491 1,139,299 11.76% -4.76%
Sample
Bank
Current Accord
RWA
(Credit Risk)Capital CAR
RWA
(Credit Risk)CAR
RWA
(Op. Risk)
Basel II Proposal
Delta CAR
From the QIS 3 survey result it can presents some information as follows:
a. Bank Capital
In the implementation of QIS 3 survey in Indonesia, bank capital need is calculated using standardized
approach and calculating risk mitigation factors and operational risk, but it did not include market
risk component. The survey showed a significant decreasing impact of Capital Adequacy Ratio
compare to 1988 Basel Capital Accord. All banks experience a variety decrease of CAR between
2,45% - 22,13%.
Credit Risk
The CAR decrease related with credit risk was recorded significantly, between 2% - 18%. The
significant CAR decrease was caused by several things, some of them are bank assets portfolios
which are dominated by sovereign claims (Recap bond and SBI / Indonesian securities) subject to
20% risk weight (national discretion). In this survey, the sovereign claims were given 20% risk weight
(different with 1988 Basel Capital Accord that gave 0% risk weight). Because majority of participating
bank investors are in the form of SBI / Indonesian securities and recap bond then the 20% risk
weight to sovereign claims give a direct impact to CAR’s decrease.
135
New Basel Capital Accord
Placement for banks that do not have rating with 50% risk weight (national discretion) – exception
for placement with time frame < 3 months (20%) – also give an impact to bank’s CAR decrease. In
this survey, treatment for bills to banks used option 2 by using national discretion on determining a
certain risk weight. For banks that do not have rating, they are subject to 50% risk weight (bigger
compare to risk weight on 1988 Basel Capital Accord which was only 20%). Besides that, most
debtors do not own rating from acknowledged rating agency (Moody’s, S&P, and Fitch IBCA) so they
are subject to 125% risk weight (national discretion). If in 1988 Basel Capital Accord risk weight for
commercial debtor (other than bank and state-owned institutions) is 100%, in the QIS 3 survey the
risk weight is raised to 125% (national discretion) for debtors who do not have rating. Plus, all
overdue asset portfolios more than 90 days are subject to 150% risk weight.
From collateral side, majority of collateral owned by bank are not eligible so banks could not get
incentive to mitigate risk. Generally, collaterals received by participating bank are in the form of
physical collateral such as property/real estate so they are not eligible to be calculated in risk
mitigation so bank could not get incentive to lessen the exposure risk weight although it has high
collateral value.
Besides those things that inflict financial loss to bank capital, the New Accord also gives incentive
in the form of special treatment to KPR and retail debtors who are subject to risk weight each 40%
and 75%, lower compare to 1988 Basel Capital Accord (50% and 100%). Other incentive is the 0% risk
weight given to guarantee issued by bank which is unconditionally cancelable.
Operational Risk
Additional capital charge placement to anticipate operational risk by using basic indicator approach
and standardized approach that is by using a factor of 15% or b factor of 12% - 18% for every
business lines cause CAR decrease around 0,1% - 6,7%. The relatively insignificant CAR decrease
may be related with some factors such as the use of gross income indicator in approximately the
last 3 (three) years (1999 – 2001) in basic indicator approach resulted in a relatively small capital
charge calculation number for operational risk because profitability performance of some big banks
in 1999 showed a negative net interest margin number as implication of negative spread. Besides,
some big banks could not identify gross income indicators for each business lines according to
standardized approach, so they have no basic in calculating capital charge for operational risk.
136
Articles
b. Risk Management Process
The survey result shows that risk management process conducted by participating bank still has a
lot of weaknesses. Some of those weaknesses include risk identification process where the most
prominent weakness is unavailability/not yet available a standard reference for bank recording
system; unstructured information management system; and a very short database timeframe between
1 – 3 years.
Most of data collected by banks are still qualitative data so they are not informative enough to be
used in capital requirement calculation. One of the reasons is relatively limited information
technology support so bank information management has difficulty in maintaining and supplying
quantitative data. Thus there is no commitment from each organization level also the unavailability
of standard guidelines both internal guidelines and guidelines for supervision authorities which
make risk management initiative really dependent to each bank’s management policy.
Future Challenges for Bank
To anticipate the implementation of the New Accord then banking need to pay attention to some
problems they are facing, among others, that the New Accord implementation requires bank to fulfill
some conditions including improving risk analysis and management abilities, information availability
and documentation. As a result, banks have to increase and equip their knowledge in risk analysis area
or further do changes in organization structure, procedures and decision-making process. A healthy risk
analysis and management also required the use of comprehensive and adequate internal data. This will
need a well-structured information management (database).
All risk mitigation practices really need a good management process (administration and
documentation) especially if it is related with the existing legal aspect. As a consequence, in
implementing risk mitigation a bank has to retain its cautious principal by referring to the available
legal instruments. Therefore, banks have to improve their administration and documentation systems
plus complement their knowledge on legal aspects especially those related with their risk mitigation
practices.
To anticipate potential problems which may occur in the future, banking industry needs to
understand its assets’ structure and capital fulfillment consequences related with the New Accord
implementation so banks could be more careful in conducting their financing policies.
137
New Basel Capital Accord
For Supervision Authorities
The New Accord proposal was organized to anticipate developments and innovations in financial
system by giving various approach alternatives in capital calculation which is more risk sensitive. The
implementation of the New Accord will give implication to supervision authorities in the form of demand
to do adjustments in their assessment methods and means which give emphasis to a forward looking risk
profile of a bank (Risk based supervision) without disregarding compliance audit. Therefore, supervision
authorities needs to realize the importance of knowing and understanding the philosophies o all approaches
used in calculating credit risk, market risk, and even operational risk. Understanding the process requires
supervisors’ adequate competence and expertise so supervisors’ resources needs are not only quantitative
but also qualitative especially those related with statistic and econometric disciplines. Furthermore,
analysis and validation for internal model used by banks made each bank has a unique and specific
nature. This requires specialization in each bank’s supervision. Therefore, supervision authorities is
also required to know overall bank operational activities that need dedicated and specialist supervision.
Indonesian banking condition today is still in recovery period after the crisis so it affected banking
data availability as reference to do analysis or projection to obtain a coefficient of a statistic or
econometric model. Data validity test is also needed considering those data were obtain in not a normal
condition. If the data, which was obtained in not a normal condition, is use for analysis or projection, it
will create bias which will result in inaccurate model.
Rating concept as one of the indicators in creditworthiness evaluation as required in standardized
approach is a relatively new thing. Cost implication occurs from using the rating concept may place
banks and debtors in a difficult position. Bank wants each debtor to have a good rating so bank additional
capital need would not be so big. But, the rating requirement will cause additional cost for debtors.
Based on those conditions, it is important for supervision authorities to take anticipatory steps by
improving competence and expertise in performing risk analysis and developing risk based supervision
that focused to human resources and information technology development. Another important thing is
improving quantitative analysis ability in credit risk, market risk and also operational risk.
Moreover, supervision authorities also needs to encourage banks to do anticipatory steps towards the
New Accord implementation plan through a set of policy and regulation so bank take the initiative to make
changes and improving of their internal infrastructures (such as human resources, decision making process,
information technology, etc.). Also involving market player through a continuous market dialogue to exchange
information and create understanding on New Accord implementation and its impact on financial system in
general also to understand best practices used all these times as reference by market players.
138
Articles
REFERECES
1. International Convergence of Capital Measurement and Capital Standards, Basle Committe on Banking
Supervision, July 1988.
2. Amendment to the Capital Accord to Incorporate Market Risk, Basle Committe on Banking Supervi-
sion, January 1996.
3. The New Basel Capital Accord, Consultative Document, Basel Committe on Banking Supervision,
January 2001.
4. The Standardised Approach to Credit Risk (Supporting Document to the New Basel Capital Accord),
Consultative Document, Basel Committe on Banking Supervision, January 2001.
5. The Internal Ratings-Based Approach (Supporting Document to the New Basel Capital Accord),
Consultative Document, Basel Committe on Banking Supervision, January 2001.
6. Operational Risk (Supporting Document to the New Basel Capital Accord), Consultative Document,
Basel Committe on Banking Supervision, January 2001.
7. Pillar 2 (Supervisory Review Process) (Supporting Document to the New Basel Capital Accord),
Consultative Document, Basel Committe on Banking Supervision, January 2001.
8. Pillar 3 (Market Discipline) (Supporting Document to the New Basel Capital Accord), Consultative
Document, Basel Committe on Banking Supervision, January 2001.
9. Credit Risk Modelling: Current Practices and Applications, Basle Committe on Banking Supervision,
April 1999.
10. Comments on “The New Basel Capital Accord”, The Market Consultative Paper by the Basel Committe
on Banking Supervision, Bank of Japan.
11. Comments on “Consultative Document “The New Basel Capital Accord”, Superintendence of Banks
of Chile,
12. Public Interest Comment on The Basel Committe on Banking Supervision’s Second Consultative on
the New Basel Capital Accord, Regulatory Studies Program, Mercatus Center, George Mason University.
13. Will the Proposed New Basel Capital Accord Have a Net Negative Effect on Developing Countries?,
S. Griffith-Jones & S. Spratt, Institute of Development Studies, University of Sussex, July 2001.
14. A Capital Accord for Emerging Economies?, Andrew Powell, Universidad Torcuato Di Tella and Visiting
Research Fellow, World Bank, September 6th, 2001.
139
New Basel Capital Accord
15. How The Proposed Basel Guidelines on Rating-Agency Assessments Would Addect Developing
Countries?, Giovanni Ferri, Li-Gang Liu, and Giovanni Majnoni, World Bank, University of Bari (Italy).
16. From Basel I to Basel II: Implications and Challenges for Emerging Markets, Liliana Rojas-Suarez,
Presentation.
17. EMEAP Workshop on Macroeconomic Impact of the New Basel Capital Accord, Impact of the New
Basel Capital Accord on Banks in Asia, Simon Topping, Hong Kong Monetary Authority, March 2002.
18. Quantitative Impact Study (Overview, Instruction, Technical Guidance, National Discretion), Basel
Committe on Banking Supervision, October 2002.
Coordinators
Nelson Tampubolon & Muliaman D Hadad
Editor in Chief
Wimboh Santoso & S Batunanggar
Analysts
S Batunanggar Dicky Kartikoyono Ita Rulina
Endang Kurnia Saputra Ricky Satria Yossy Yoswara Dwityapoetra S Besar
Article Contributors
Wimboh Santoso S Batunanggar Enrico Hariantoro
Dadang Muljawan Imansjah G A Indira
Bambang Arianto Indra Gunawan
Compilator & Lay-out
Dwityapoetra S Besar Ricky Satria Sunarto
Design & Production
Ricky Satria
Partners
On-site Supervisory Presence Team
Directorate of Bank Licensing and Banking Information
Directorate of Bank Supervision 1
Directorate of Bank Supervision 2
Directorate of Economic and Monetary Statistics
Directorate of Accounting and Payment System
Directorate of International Affairs
Directorate of Monetary Management
Bureau of Credit
Translation Consultant
Skip Edmonds
JUNE 2003
F I N A N C I A L
S T A B I L I T Y
R E V I E W