bulletin of monetary economics and banking · prof. dr. iwan jaya azis prof. iftekhar hasan dr. m....
TRANSCRIPT
1ANALISIS TRIWULANAN: Perkembangan Moneter, Perbankan dan Sistem Pembayaran, Triwulan II - 2007
BULLETIN OF MONETARY ECONOMICS AND BANKING
Directorate of Economic Research and Monetary PolicyBank Indonesia
PatronPatronPatronPatronPatronBoard of Governor Bank Indonesia
Editorial BoardEditorial BoardEditorial BoardEditorial BoardEditorial BoardProf. Dr. Anwar Nasution
Prof. Dr. Miranda S. GoeltomProf. Dr. Insukindro
Prof. Dr. Iwan Jaya AzisProf. Iftekhar HasanDr. M. Syamsuddin
Dr. Perry WarjiyoDr. Halim Alamsyah
Dr. Iskandar SimorangkirDr. Solikin M. JuhroDr. Haris Munandar
Dr. Andi M. Alfian Parewangi
Editorial ChairmanEditorial ChairmanEditorial ChairmanEditorial ChairmanEditorial ChairmanDr. Perry Warjiyo
Dr. Iskandar Simorangkir
Executive DirectorExecutive DirectorExecutive DirectorExecutive DirectorExecutive DirectorDr. Andi M. Alfian Parewangi
SecretariatSecretariatSecretariatSecretariatSecretariatToto Zurianto, MBA
MS. Artiningsih, MBA
The Bulletin of Monetary Economics and Banking (BEMP) is a quarterly accreditedjournal published by Directorate of Economic Research and Monetary Policy-BankIndonesia. The views expressed in this publication are those of the author(s) anddo not necessarily reflect those of Bank Indonesia.
We invite academician and practitioners to write on this journal. Please submityour paper and send it via mail to: [email protected]. See the writing guideon the back of this book.
This journal is published quarterly; January √ April √ August √ October. The digitalversions including all back issues are available online; please visit our stable link:√http://www.bi.go.id/web/id/Publikasi/Jurnal+Ekonomi/. If you are interested to sub-scribe for printed version, please contact our distribution department: Publicationand Administration Section √ Directorate of Economy and Monetary Statistics,Bank Indonesia, Building Sjafruddin Prawiranegara, 2nd Floor - Jl. M. H. ThamrinNo.2 Central Jakarta, Indonesia, Ph. +62-21-3818202, Fax. +62-21-3802283,Email: [email protected].
BULLETIN OF MONETARY ECONOMICSAND BANKING
Volume 14, Nomor 2, October 2011
QUARTERLY ANALYSIS: The Progress of Monetary, Banking and Payment System,
Quarter III - 2011
Author Team of Quarterly Report, Bank Indonesia
Does Financial Development Absorb or Amplify the Shock?
Meily Ika Permata, Ibrahim, Hidayah Dhini Ari
Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly -
Traded Banks in Asia
Wahyoe Soedarmono
Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation,
Competition and Concentration in the Indonesia Banking Industry between 2001
and 2009
Tri Mulyaningsih, Anne Daly
Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
Tumpak Silalahi, Tevy Chawwa
107
141
103
127
177
103QUARTERLY ANALYSIS: The Progress of Monetary, Banking and Payment System, Quarter III, 2011
QUARTERLY ANALYSIS:The Progress of Monetary, Banking and Payment System
Quarter III - 2011
Author Team of Quarterly Report, Bank Indonesia
The Board of Governor Meeting of Bank Indonesia on October 11, 2011 decided to
lower the BI rate by 25 bps to the level of 6.5%. Bank Indonesia will also maintain the
stabilization of Rupiah particularly from the impact of global financial market shock. The decision
is in line with the inflation expectation of below 5% on current and next year. Furthermore,
these policies are meant to anticipate and to mitigate the negative impact of the global economic
and financial slowdown on Indonesian economy. Looking ahead, the Board of Governor will
continue to evaluate the global economic and financial performance and use the interest rate
as well as the mix of monetary and the other micro prudential policies to mitigate the possible
slowing down of Indonesian economic performance, especially on achieving the inflation target
of 5% + 1% in 2011 and 4.5% + 1% in 2012.
The Board of Governor put a high concern on the high risk and uncertainties in global
financial market and the trend of global economic slowdown due to the debt and fiscal problem
in Europe and US. Main attention is addressed to the short term impact through the financial
channel in the form of the stock market weakening, the increase of debt risk indicator, and
the capital reversal pressure by global investor from emerging countries, including Indonesia.
Meanwhile, the global economic performances also weaken as indicated on the slowing
production activities and retail sale, accompanied by the decrease of consumer»s confident in
developed countries and also numbers of correction on international commodity prices. On
the other side, the inflation pressure has decreased, although the inflation of emerging countries
remains high hence shifts the monetary policy response to neutral or accommodative.
Looking ahead, on general the Board of Governor foresee a slower output growth in
developed countries, the lower international trade volume, and lower global commodity prices.
Meanwhile, the large excess liquidity and risk perception of investor will still drive high capital
inflow into emerging countries, including Indonesia, both in foreign direct investment (FDI) or
in portfolio investment.
104 Bulletin of Monetary Economics and Banking, October 2011
The Board of Governor considers the fundamental of national economic and banking in
Indonesia are still strong among higher concern about the world economic prospect. The
output growth on 2011Q4 is predicted to be higher, especially driven by consumption and
investment that the output will reach 6.6% in 2011. So far, the impact of global economic
shock is limited to the financial sector, while the real sector is relatively not affected. However,
the slowing global economic is predicted to influence the domestic economic performance in
2012, both on financial market and international trade.
The domestic economic growth in 2012 is predicted to be in the range of 6.2% - 6.7%.
The source of this growth is driven by strong consumption and increasing investment.
OnSectorallevel, all sectors are predicted to have positive growth. The main growth contributors
are Manufacture, Trade Hotel and Restaurant, and Transport and Communication.
The Balance of Payment (BOP) in 2011Q4 is predicted to gain surplus following the high
pressure due to the capital outflow in previous quarter. In aggregate for 2011, BOP is predicted
to have a large surplus. The main sources of this surplus are the increasing surplus of capital
and financial transaction, both in the form of portfolio and direct investment. In line with this
the reserve on the end of September 2011 was USD114.5 milliard or equivalent to 6.5 months
of import and government debt service payment. This reserve is more than enough to support
the stabilization of Rupiah.
The exchange rate of Rupiah in 2011Q3 experience a pressure, especially on September
2011 where the Rupiah depreciate by 2.42% (ptp) to the level of Rp8,790 per USD with
increasing volatilities. However, this depreciation is in line with other currencies on regional
peer»s member. Among several factors, the pressure on Rupiah is caused by the increase of
global risk due to the increasing concern on the global economic prospect. In addition, the
increase of foreign exchange demand to pay import also contributed to the pressure of Rupiah.
Looking ahead, Bank Indonesia will continue to keep the stability of Rupiah»s exchange rate to
support the macroeconomic stability.
The inflation pressure continues to decrease. CPI inflation on 2011Q3 was 1.89% (qtq)
or 4.61% (yoy), lower than the previous year. The lower inflation pressure came from the
volatile food and administered price, along with the supply improvement, the lower international
food prices and the minimum government intervention concerning strategic commodities.
Meanwhile, the core inflation pressure excluding the hike of gold price is relatively well managed,
because of the appreciation of Rupiah on previous period and the sufficient supply. With
these facts, in 2011 the inflation will be lower than 5%.
105QUARTERLY ANALYSIS: The Progress of Monetary, Banking and Payment System, Quarter III, 2011
In 2012, the inflation will be under control and is expected to be lower than 5%, along
with the correction on international commodity prices and the weakening of world economy.
The stability of banking system will be also maintained with better intermediary function,
despite of the global financial shock. Banking industry stability are also well maintained as
reflected on the Capital Adequacy Ratio (CAR) which is far above the requirement 8% and
Non-Performing Loan which is lower than 5% gross. In addition, the distribution of credit
continues to finance the real economic activities where its growth reached 23.8% (yoy) until
the end of September 2011. Bank Indonesia maintains the banking system stability and
encourages its intermediary function while still keeping the prudent banking practice by directing
the credit allocation for productive sectors in order to optimize the economic growth among
the uncertain of global economic.
The reliability and efficiency of payment system also contributes on macroeconomic
performance in Indonesia. The support of the payment system is indicated on the availability
of the Bank Indonesia Real Time Gross Settlement (BIRTGS), Bank Indonesia-Scrip less Security
Settlement System (BI-SSSS), and National Clearing System that reach 100%. Besides, the
reliability of card payment system and electronic money, arranged and processedoutside Bank
Indonesia are also well maintained. Bank Indonesia can also supply the fiat money even though
there wassignificant increase of money demand during the Holy Month of Ramadhan and
«Idul Fithri.
106 Bulletin of Monetary Economics and Banking, October 2011
This page is intentionally left blank
107Does Financial Development Absorb or Amplify the Shock?
DOES FINANCIAL DEVELOPMENTABSORB OR AMPLIFY THE SHOCK?
Meily Ika PermataIbrahim
Hidayah Dhini Ari 1
This paper analyzes the role of financial development on economic output in Indonesia. Using
vector autoregressive method, the results confirm the positive impact of financial development on output
growth. The interaction between the financial development and the shock either in financial or real sector
shows that the financial development has a positive role to dampen the negative impact of the shock on
the output growth, while strengthen the positive one. Another variable on the model, which significantly
affect the output growth are excess money, term of trade, and the price. Compare to these variables, the
marginal effect of financial development on output is smaller.
Keywords : Financial development, shock, output volatility, VAR.
JEL Classification : E44, O16
1 Meily Ika Pertama ([email protected]), Ibrahim ([email protected]) and Hidayah Dhini Ari ([email protected]) are researchers in EconomicResearch Bureau, Directorate of Economic and Monetary Policy Research. The author thanks to Dr. Noer Azzam, Dr. Perry Warjiyo, Dr.Iskandar Simorangkir and other colleagues in DKM for the excellent input and suggestion.
Abstract
108 Bulletin of Monetary Economics and Banking, October 2011
1. INTRODUCTION
Macroeconomics stability is a necessary condition for sustainable economic development.
Therefore, efforts to understand the sources of macroeconomic instability condition is one of
the challanges of the important attention in economics.
Role of the financial sector is associated with economic growth that has long been a
subject research, especially after Schumpeter (1912)2. Since then, continues studies have been
conducted to examine the interaction between financial development and output growth.
Based on review of various conducted studies , the relationship between financial development
and output growth tend to be inconclusive.
Some views believe that the financial development will drive the growth (initiated by
Schumpeter, 1911 and Gurley and Shaw, 1995) because financial sector can overcome the
problem of financial constraint, contribute to the allocation of resources more efficiently, channel
funding and other activities related to risk sharing and financial innovation, be the medium for
the monetary policy transmission (Cecchetti and Krause, 2001 and Krause and Rioja, 2006), and
can overcome the problem of imperfect capital market) (Bernanke and Gertler (1989), Greenwald
and Stiglitz (2003) and Kiyotaki and Moore (1997). Financial development is also believed to
have positive role in reducing the volatility of macroeconomic variables,(Dynan, Elmendorf and
Sichel (2006), Denizer, Iyigun, and Owen (2002), Harvey and Lundblad (2006), Aghion et. al
(2005), Aghion et. al (2009), Cecchetti, Lagunes and Krause (2006), and Mendicino (2007)).
However, Bacchetta and Caminal (2000), Easterly et al. (2002) and Kuneida (2008), show
that the financial constraint can be a factor that restrain or even exacerbate the shock impact
that occured in the economy. Meanwhile, Lopez and Spiegel (2002), denizer et al. (2002), Silva
(2002), and Tharavanji (2007) actually found negative relationship between financial development
and output volatility. Other researchers such as Tiryaki (2003), Beck (2006) & Guryal et al.
(2007) did not find significant relationship between financial development and growth volatility.
The output volatility shown on the figure, suggest a need to conduct empirical testing
about the role of financial sector in macroeconomic stability, especially GDP growth in Indonesia.
As one of the developing countries with a growing financial sector, the study about the role of
financial sector in the context of macroeconomic stability is expected to assist the formulation
of a more appropriate monetary policy. When compared to some developed countries, the
output volatility in emerging countries is generally higher, except the period of global crisis.
Among emerging countries, Indonesia»s output volatility post-crisis in 1997 looks relatively smaller.
2 Schumpeter, J.A., 1991. The Theory of Economic Development. Cambridge, Mass: Harvard University Press
109Does Financial Development Absorb or Amplify the Shock?
Figure 1. Output Volatilityin Developing and Emerging Countries
Explicitly, this paper identifies the relationship between the role of financial sector and
output volatility in Indonesia. To measure the financial development we use the ratio of credit
to private sector towards GDP. This measure commonly used in previous studies. The conclusive
relationship between financial developmentsto macroeconomic stability is expected to be a
valuable input in formulating better monetary policy.
The second part of this paper reviews the theory and literature, the third part review the
methodology and empirical models are used, while the results and analysis are described in the
fourth. Conclusions, implication and policy recommendation will close the presentation.
II. THEORY
The view that financial development will drive the growth (started by Schumpeter, 1911
and Gurley and Shaw, 1955) is based on the view that the financial sector can overcome
financial constraint. In addition, the existence of financial sector contributes to the allocation of
resources, both financial and non-financialmore efficiently. Financial deepening and development
of can reduce volatility in economy, through its ability to channel financing and other activities
related to risk sharing and financial innovation.
Financial sector contributes in stabilizing macroeconomic condition in general through
its ability to be the medium for the monetary policy transmission (Cecchetti and Krause, 2001
and Krause and Rioja, 2006). Bernanke and Gertler (1989), Greenwald and Stiglitz (1993) and
0
1
2
3
4
5
6
7
8
1994
Q4
1995
Q3
1996
Q2
1997
Q1
1997
Q4
1998
Q3
1999
Q2
2000
Q1
2000
Q4
2001
Q3
2002
Q2
2003
Q1
2003
Q4
2004
Q3
2005
Q2
2006
Q1
2006
Q4
2007
Q3
2008
Q2
2009
Q1
2009
Q4
China Indonesia BrazilThailand EU USA
110 Bulletin of Monetary Economics and Banking, October 2011
Kiyotaki and Moore (1997) carry out a study and confirm the view above, by showing that
financial sector contribute to overcome the problem of imperfect capital market, and there by
reduce output volatility in economy.
Dynan, Elmendorf and Sichel (2006) and Denizer, Iyigun, and Owen (2002) and Beakaert,
Harvey and Lundblad (2006) present the results of study showing that financial development
have a positive role in reducing the volatility of macroeconomic variables. Aghion et al. (2005)
and Aghion et al. (2009) confirm that credit constraint contribute to enlarge the shock impact
in economy through the choice of type investment made by businesses. Cecchetti, Lagunes
and Krause (2006) and Mendicino (2007) prove that the consumption credit have a positive
role in overcoming liquidity constraint at the household level so that it can reduce the volatility
of economic growth.
However, Bacchetta and Caminal (2000) show that financial constraint can be a factor
that restrain or even exacerbate the shock impact that occured in the economy, depending on
the type of shock that occured in the economy. In line with Bacchetta and Caminal, Easterly et
al. (2002) and Kuneida (2008) indicate the characteristic of relationship between financial
development and economic growth tends to be nonlinear. In this case, financial development
will reduce macroeconomic volatility to a certain point, beyond that point, more credit to private
sector will increase the volatility.
Meanwhile, Lopez and Spiegel (2002), Denizer et al. (2002), Silva (2002), and Tharavanji
(2007) actually found a negative relationship between financial development and the volatility
of economic growth. More interestingly, Tiryaki (2003), Beck (2006) and Guryay et al. (2007)
did not find significant relationship between financial development and volatility growth.
Based on several conclusions from various research above, the relationship between
financial development and output volatility is still ambiguous or inconclusive. That characteristic
of relationship depends on the source of the shock on economy, whether it comes from the
real or monetary sector, and how an economy responds to the shocks .
One simple model about financial development was built by Bacchetta and Caminal
(2000), and then developed and modified by Beck (2006). The model assumethe economy
consists of consumers and entrepreneurs. Every entrepreneur has the same access to production
technology, represented by f (k), with f (0) = 0, f ‘(k) > 0, and f “(k) < 0. Although each
entrepreneur has same access to production technology, entrepreneur basically has a different
level of wealth b. As much as β ratio of the entrepreneur is entrepreneur with high wealth. And
as much as (1 - β) ratio of the entrepreneurs are those with low wealth. High group is assumed
to be able to meet the financing needs of its investment and has excess funds in the bank and
111Does Financial Development Absorb or Amplify the Shock?
Meanwhile, Low has limited financing so that had to borrow funds with interest rate rL.
The existence of assymetric information and moral hazard potentially cause the Low group to
bear an agency cost ϕ. With this condition, Low profit maximization is achieved at:
(1)
(2)
(3)
get the interest rate rD. In addition, High is assumed to have no financing constraint in financial
market, and thereby High profit maximization is achieved at :
Thus, the relative marginal productivity between High and Low is :
The greater the difference between rL and rD and ϕ, the higher ratio kH / kL and the
greater the difference of marginal productivity between Low and High. The higher ϕ, the greater
effect of real location funds between High and Low. This real location funds will influence
aggregate productivity and ultimately economy»s output as a whole.
The financial intermediaries, in this case is bank, is assumed to work in a perfect competition
condition, without cost, and only holds the asset in the form of credit. However, the bank
deposit is subject to reserve requirement of the monetary authority, that is τ. Thus, the credit
can be allocated by bank to Low (1 − τ ) is multiplied by the deposits of High. Thus, the increase
of τ will reduce funds available for loan forthe Low, and vice versa..... Total aggregate of loans
from bank is:
(4)
where bH is internal funds of High and kH is the desired level of investment of group High.
While kL and bL, respectively, is the level of investment desired by Low and their internal funds.
Low internal funds areassumed to be very small so that it cannot meet the level of their desired
investment.
Furthermore, assuming there is no possibility of default, then the ratio of rL and rD is only
influenced by the size of reserve requirement, τ , hence:
112 Bulletin of Monetary Economics and Banking, October 2011
With the existence of assymetric information that generates agency cost, then Low
naturally will always face sub-optimal investment condition. This means the realized level of
investment will always below the desired level. In this case, agency cost stated as:
(5)
(6)
Where ω is a function of several technological parameters, which are exogenous. This
means that agency cost is a negative linear function of ratio between Low»s internal funds
owned and their desired level of investment. The greater internal funds owned, the smaller the
agency cost charged by bank to Low. Furthermore, in this case, is:
Equation (6) shows that Low will be exposed to credit-constrained condition, so that the
level of investment that might be achieved will lower than the desired one. Low»s level of
investment will decrease with the increaseof rL, ω, and the leverage ratio .
Based on the equations above, then the marginal productivity of High and Low can be
elaborated in following equation:
(7)
Thus, the market clearing condition in financial market is:
(8)
Based on equation (7) and (8), we can seethat the relative investment kL / kH will increase
along with the increase of ratio bL / kL and the relative ratio of internal funds bL / bH and
declining agency cost ω, and reserve requirement τ.
113Does Financial Development Absorb or Amplify the Shock?
xt = A
0 + A
1 x
t-1 + A
2 x
t-2 + .... + A
p x
t-p + e
t
Based on the equations above, we can obtain some interpretation below:
1) The impact of shocks that occur in the real sector will be largerin the presence of assymetric
information condition on capital market. The impact was even greater in line with the
increase in agency cost ω. This is shown in equation (7), where the investments of the High
will be greater than the Low, so that causing the Low marginal productivity is higher than in
High. The difference of marginal productivity will be greater with higher agency cost; a
condition found in the capital market conditions that characterized assymetric information.
With in creasing magnitude of the difference of marginal productivity, then the impact of
shocks to the economy»s output will be greater.
2) The impact of shocks coming from monetary policy affects the supply of loan able funds will
have smaller impact on the conditions where there is assymetric information on the capital
market. Relaxing the monetary policy in the form of reduction of level reserve requirement,
would increase the amount of loan able loan that will eventually degrade. Never the less,
the reduction in interest rates will increase the leverage and eventually agency cost for the
Low. This in turn will reduce some of the positive impact of reserve requirement reduction.
Mean while, the decline in agency cost as a result of the growing of financial development
will further streng then the positive impact of reduction in reserve requirements to the
output.
These should be the the base of our hypothes is to be tested on this paper.
III. METHODOLOGY
This study will use an econometric approach to test the testable hypothesis by utilizing
the method of vector autoregression (VAR). We use this VAR method to test the case of Indonesia.
VAR model is a linear function analysis of past data movement of a set of variables
(endogenous variable) in the same period (t=1,...,T). A VAR model of order (p) can be represented
in the following equation:
Which et is a vector from error term that meet the standard condition; E (e
t) = 0, and E (e
t e’
t)
= Ω.
VAR model is often used in macroeconomic analysis. However, although the VAR
approach has the advantage in modeling dynamic behavior of economic variable and
114 Bulletin of Monetary Economics and Banking, October 2011
forecasting, many criticism directed against VAR approach which tends to be a-theory, where
the lack of restriction in the VAR model lag structure is associated with the lack of a structure
underlying the relationship among variables in the VAR system.This would create difficulties
in interpreting the results.
The theoretical model described previously has provided a hypothesisto be tested. But
the test to perform is not solely based on a reduced form derived explicitly from the model
above, but rather we use empirical ad hoc model in order to accommodate some other control
variables that are not captured explicitly in theoretical model.
Some of the variables to be included in the model are real GDP growth and price
movements, in order to illustrate the volatility of macroeconomic variables. Also included are
variables in real and monetary sectors. For the real sector, we can use terms of trade variable for
instance, while for monetary sector we may use excess money. Financial development ( FD ) is
measured with the credit ratio to GDP. Meanwhile, to capture a market imperfection/assymetric
information, we can usea variable that can explain the existence of risks such as interest spread
between lending and SBI rate.
The data used in this study are quarterly data, covering first quarter period of 1997 to
second quarter of 2010. The variables involved are, (i) Real GDP, representing the output,
(ii) Consumer Price Index, (iii) Term of trade, representing the shock from real sector, (iv)
Excess money, represents shock from monetary sector, (v) Ratio of credit to GDP, measuring
the financial development, (vi) Interest rate spread between lending rate and the SBI rate,
measuring the magnitude of risk due to the existence of market imperfection/assymetric
information.
IV. RESULTS AND ANALYSIS
In general, the macroeconomic indicator in Indonesia at the beginning of the observation
period had a relatively high volatility movement. During the period ofcrisis in 1998, the economic
growth contracted by more than 13.1% and inflation jumped to 69.8% followed by a variety
of other indicators such as the ratio of credit to GDP, the excess money, and the risk indicators
represented by the spread of lending rates to SBI (Figure 5). These volatilities were getting
smaller along with the improvement of monetary management.
By looking at the more detail movement on standard deviation of the macroeconomic
and the financial indicators, we can obtain a clearer picture. In general, the volatility of
GDP with other variables has same direction. The correlation of growth volatility with the
115Does Financial Development Absorb or Amplify the Shock?
average of credit to GDP ratio reached 81% and the correlation between growth and
inflation volatility by 78% (Figure 6). Volatilityin the figure is calculated by using the three
years moving standard deviation for each indicator, then we calculate the pair correlations.
Based on the initial hypothesis on the co-movement among these variables, we proceed to
use the VAR method.
Prior the VAR estimates we run the stationary tests on each variable by using the
Augmented Dickey Fuller (ADF) unit root tests. The test results show that GDP and IHK variables
are non-stationary in levels (Table 1). Based on these unit root tests, the variables selected for
inclusion in the VAR model is dLnPDB, dLnIHK, dLnTOT, dLnFD, risk growth and excess money
growth.
Figure 2.The Main Indicators Development
Figure 3.The Financial Development (dln FD)
Figure 4.Excess Money
Figure 5.Spread Credit Interest Rate-SBI
-20
0
20
40
60
80
PDB
Inflasi
1997Mar
1998Mar
1999Mar
2000Mar
2001Mar
2002Mar
2003Mar
2004Mar
2005Mar
2006Mar
2007Mar
2008Mar
2009Mar
2010Mar
1997Mar
1998Mar
1999Mar
2000Mar
2001Mar
2002Mar
2003Mar
2004Mar
2005Mar
2006Mar
2007Mar
2008Mar
2009Mar
2010Mar
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-0.10
-0.12
1997Mar
1998Mei
1999 2000 2001Jul
2003Sep
2004Mar
2005Mei
2006Jul
2007Sep
2008Nov
2010Jan
40
30
20
10
0
-10
-20
-30Nov Jan
1997Mar
1998Mar
1999Mar
2000Mar
2001Mar
2002Mar
2003Mar
2004Mar
2005Mar
2006Mar
2007Mar
2008Mar
2009Mar
2010Mar
20
10
0
-10
-20
-30
-40
-50
116 Bulletin of Monetary Economics and Banking, October 2011
Optimum lag order for VAR procedure showed mixed results. Based on the Schwartz
Information Criteria, the optimal lag is 1, while the Akaike Information Criteria and Hannan
Quinn Information Criteria generate optimal lag is 6. However, the lag order 6 is not selected,
since the number of observation is only 54. This study also did not follow the Schwart optimum
lag with lag order 1, but sets the use of lag order 2 to better capture the variables dynamics.
Based on lag structure test, it is known that the estimated VAR with lag order 2 is stable with all
roots smaller than1 and is in the unit circle (Figure 7).
Figure 6.The Connection Development of Main Indicator
0.00
0.50
1.00
1.50
2.00
2.50
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
1990Mar
1990May
1992Jul
1993Sep
1994Nov Mar May Jul Sep NovJan
1996 1997 1998 1999 2000 2001Mar May Jul Sep NovJan
2003 2004 2005 2006 2007 2008Jan
2010
Average of Credit/GDP (3 years)
Rolling StDev of GDP Growth(3 years -yoy) (RHS)
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00Rolling StDev of Inflation(3 years - yoy)
Rolling StDev of GDP Growth(3 years -yoy) (RHS)
1990 1991 1992 1994 1995 1997 1999 2000 2001 2002 2004 2005 2006 2007 2009 20101996Apr Jul Oct Jan AprJanApr Jul OctJanApr Jul OctJanJun SepMar
Table 1.The Test Result of Unit Root Test
Variable Level (P-value)
GDP 1.0000dLnPDB 0.0967GDP Growth 0.0943CPI 0.8050dLnIHK 0.0006Inflation 0.3043TOT 0.0258dLnTOT 0.0020TOT Growth 0.0057FD 0.0163dLnFD 0.0024FD Growth 0.2049Risk 0.0000Growth Risk 0.0000Excess Money Growth 0.0002
117Does Financial Development Absorb or Amplify the Shock?
The results of Granger Causality/block exogeneity test showed that the simultaneous
movement of inflation, TOT, excess money, financial development and the risk are the
explanatory variable of the GDP movement (Table 2). However, individually the risk can not
explain the GDP movement, but we keep it on the model since the risk is the control variables
that describe the magnitude of risk due to the existence of market imperfection/assymetric
information.
Figure 7.AR Roots Graph
Table 2.VAR Granger Causality/Block Exogeneity Wald Tests
ExcludedExcludedExcludedExcludedExcluded Chi-sqChi-sqChi-sqChi-sqChi-sq dfdfdfdfdf ProbabilityProbabilityProbabilityProbabilityProbability
DLNIHK 17.48 2 0.000DLNTOT 11.80 2 0.003EXCMON 8.52 2 0.014DLNFD 9.53 2 0.009GRISK 1.09 2 0.580AllAllAllAllAll 56.7356.7356.7356.7356.73 1010101010 0.0000.0000.0000.0000.000
Inverse Roots of AR Characteristic Polynomial
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
4.1. Impact to GDP Growth
The results ofimpulse response shows that financial development have a positive impact
in improving GDP growth and significant in quarter 2 (Figure 8). Meanwhile, an increase of
inflation will reduce GDP growth significantly during quarter 2 to 5. The increase of TOT will
118 Bulletin of Monetary Economics and Banking, October 2011
2 4 6 8 10 12 14 16 18 20 22 24
Accumulated Response of DLNPDB to NonfactorizedOne Unit EXCMON Innovation
-.024
-.020
-.016
-.012
-.008
-.004
.000
.004
2 4 6 8 10 12 14 16 18 20 22 24
Response of DLNPDB to NonfactorizedOne Unit EXCMON Innovation
-.004
-.003
-.002
-.001
.000
.001
.002
2 4 6 8 10 12 14 16 18 20 22 24
Accumulated Response of DLNPDB to NonfactorizedOne Unit DLNTOT Innovation
-1,2
-0,8
-0,4
0,0
0,4
2 4 6 8 10 12 14 16 18 20 22 24-.5
-.4
-.3
-.2
-.1
.0
.1
.2
.3
Response of DLNPDB to NonfactorizedOne Unit DLNTOT Innovation
2 4 6 8 10 12 14 16 18 20 22 24
Accumulated Response of DLNPDB to NonfactorizedOne Unit DLNIHK Innovation
-3
-2
-1
0
1
Response of DLNPDB to NonfactorizedOne Unit DLNIHK Innovation
-.8
-.6
-.4
-.2
.0
.2
.4
.6
2 4 6 8 10 12 14 16 18 20 22 24
also significantly reduce the GDP during these cond to third quarter. In addition, the increase
of excess money will also reduce GDP significantly in quarter 2. However, the impact of
increased risk due to market imperfection/assymetric information to GDP tends to be not
significant.
Figure 8. The Shock Impact of Inflation, TOT development, Excess Money Growth,Financial Development and Risk Changes against GDP Growth
119Does Financial Development Absorb or Amplify the Shock?
2 4 6 8 10 12 14 16 18 20 22 24-.00002
-.00001
.00000
.00001
.00002
.00003
Accumulated Response of DLNPDB to NonfactorizedOne Unit GRISK Innovation
2 4 6 8 10 12 14 16 18 20 22 24-.000008
-.000006
-.000004
-.000002
.000000
.000002
.000004
.000006
Response of DLNPDB to NonfactorizedOne Unit GRISK Innovation
2 4 6 8 10 12 14 16 18 20 22 24
Accumulated Response of DLNPDB to NonfactorizedOne Unit DLNFD Innovation
-.4
-.2
.0
.2
.4
.6
.8
2 4 6 8 10 12 14 16 18 20 22 24-.08
-.04
.00
.04
.08
.12
.16
Response of DLNPDB to NonfactorizedOne Unit DLNFD Innovation
Figure 8. The Shock Impact of Inflation, TOT development, Excess Money Growth,Financial Development and Risk Changes against GDP Growth (continued)
In aggregate, the rise of inflation, TOT and the excess money growth will reduce GDP
growth, while financial development will increase it. Cumulatively, an increase of financial
development of 1% will give additional increase in GDP of 2.4% in 2.5 years (Figure 9). In
contrast, a decrease of financial development of 1% will cause accumulation of GDP decline of
2.4% in 2.5 years.
Furthermore, using the variance decomposition analysis shows that excess money
growth, TOT development and price movement are variables that contribute most in
explaining the GDP movements in long-term, respectively by 27%, 11% and 9% (Table 3).
While financial development contributes only about 2.6% as well the risk factor only
contribute less than 1%.
120 Bulletin of Monetary Economics and Banking, October 2011
4.2. Impact to Inflation
Financial development and the increase of risk due to market imperfection/assymetric
information do not influence inflation significantly. A significant factor in influencing inflation
is the increase in TOT, which will raise inflation for 5 quarters, from quarter 2 to quarter 6. In
addition, the increase in excess money will also reduce GDP significantly for 11quarters, from
quarter 2 to quarter 12. The results of impulse response showing that increase in TOT and
excess money growth accumulatively will increase inflation.
Figure 9. The Impact ofFinancial Development Against GDP Growth
Cumulative Increase in GDP %0.30
0.20
0.10
0.00
-0.10
-0.20
-0.30
Shock Increased 1% Financial Development
Shock Decline 1% Financial Development
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 0,026 100,000 0,00 0,000 0,000 0,000 0,000 2 0,033 81,750 5,535 11,579 0,623 0,438 0,075 3 0,040 70,384 8,575 13,504 6,495 0,374 0,669 4 0,044 62,215 9,940 12,844 13,955 0,423 0,622 5 0,046 56,466 10,170 11,625 20,609 0,574 0,555 6 0,048 52,796 9,954 10,869 24,969 0,876 0,536 7 0,049 50,784 9,659 10,685 27,004 1,308 0,560 8 0,049 49,859 9,440 10,822 27,481 1,800 0,598 9 0,050 49,475 9,314 10,998 27,326 2,262 0,624 10 0,050 49,281 9,249 11,078 27,129 2,629 0,633
Period S.E. DLNPDB DLNIHK DLNTOT EXCMON DLNFD GRISK
Table 3.Variance Decomposition of DLNPDB
121Does Financial Development Absorb or Amplify the Shock?
2 4 6 8 10 12 14 16 18 20 22 24
Response of DLNIHK to NonfactorizedOne Unit EXCMON Innovation
-.008
-.004
.000
.004
.008
.012
2 4 6 8 10 12 14 16 18 20 22 24-.01
.00
.01
.02
.03
.04
.05
.06
Accumulated Response of DLNIHK to NonfactorizedOne Unit EXCMON Innovation
Cumulated Response of DLNIHK to NonfactorizedOne Unit DLNTOT Innovation
2 4 6 8 10 12 14 16 18 20 22 24-2
-1
0
1
2
3
2 4 6 8 10 12 14 16 18 20 22 24
Response of DLNIHK to NonfactorizedOne Unit DLNTOT Innovation
-0.8
-0.4
0.0
0.4
0.8
1.2
2 4 6 8 10 12 14 16 18 20 22 24-8
-4
0
4
8
12
Accumulated Response of DLNIHK to NonfactorizedOne Unit DLNPDB Innovation
-2
-1
0
1
2
3
2 4 6 8 10 12 14 16 18 20 22 24
Response of DLNIHK to NonfactorizedOne Unit DLNPDB Innovation
Figure 10.The Impact of Changes in GDP Growth, TOT, Excess Money Growth,
Financial Development and Changes in Risk on Inflation
122 Bulletin of Monetary Economics and Banking, October 2011
2 4 6 8 10 12 14 16 18 20 22 24-.3
-.2
-.1
.0
.1
.2
Response of DLNIHK to NonfactorizedOne Unit DLNFD Innovation
2 4 6 8 10 12 14 16 18 20 22 24-.00008
-.00006
-.00004
-.00002
.00000
.00002
.00004
.00006
Accumulated Response of DLNIHK to NonfactorizedOne Unit GRISK Innovation
2 4 6 8 10 12 14 16 18 20 22 24-.000012
-.000008
-.000004
.000000
.000004
.000008
.000012
.000016
Response of DLNIHK to NonfactorizedOne Unit GRISK Innovation
2 4 6 8 10 12 14 16 18 20 22 24-1.6
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
Accumulated Response of DLNIHK to NonfactorizedOne Unit DLNFD Innovation
1 0,021 22,875 77,125 0,000 0,000 0,000 0,000 2 0,032 42,434 38,577 17,416 1,221 0,058 0,293 3 0,040 39,753 33,418 18,054 8,582 0,047 0,146 4 0,045 29,168 32,137 22,651 15,769 0,164 0,111 5 0,048 23,780 31,123 21,890 22,814 0,171 0,202 6 0,049 21,424 28,777 19,244 30,189 0,189 0,177 7 0,051 20,988 26,287 18,068 34,240 0,197 0,221 8 0,052 20,345 25,022 18,978 35,073 0,197 0,385 9 0,053 19,593 25,173 20,268 34,112 0,209 0,646 10 0,053 19,030 25,862 20,954 33,133 0,216 0,804
Period S.E. DLNPDB DLNIHK DLNTOT EXCMON DLNFD GRISK
Table 4.Variance Decomposition of DLNIHK
Figure 10.The Impact of Changes in GDP Growth, TOT, Excess Money Growth,Financial Development and Changes in Risk on Inflation (continued)
123Does Financial Development Absorb or Amplify the Shock?
Based on the variance decomposition analysis for inflation, the increase of excess money,
TOT and GDP movement are variables that contribute most in explaining the movement of
inflation in the long run respectively by 33%, 21%and 19% (Table 4). While financial
development and risk factor only contribute less than 1%.
4.3 The Impact of Interaction between Financial Development with the Shockin Real Sector (TOT) and Monetary (Excess Money)
The result of impulse response that incorporates interaction between the shock in the
real sector and monetary with financial development shows that financial development has a
positive role in dampening the negative impact of the shock on GDP growth (Fig.11 and 12). In
contrast, financial development will help increasing (amplify) the positive impact of the shock
on economic growth.
Cumulatively, increase of TOT by 1% would reduce GDP by 0.4% within 4 years, but if at
the same time there was also an increase of financial development by 1%, the cumulative
impact of the decline in GDP in the period of 4 years will tend to be smaller, only amounting to
0.17%. In contrast, if there is a TOT decreased of 1%, then cumulatively will provide an additiona
lincrease in GDP in the period of 4 years of 0.4%. Interestingly, if the TOT decline by 1% is
accompanied by increased financial development by 1%, the cumulative impact of increase in
GDP during the period of 4 years will be higher, amounting to 0.64%.
Figure 11. Impacts of Shock Interaction betweenTOT Development and Financial Development on GDP Growth
Cumulative Increase in GDP %
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Increased 1% TOT shock is accompanied by increase of 1% Financial DevelopmentIncreased 1% TOT shockThe decrease of 1% TOT shock is accompanied by increase of 1% Financial DevelopmentShock decline 1% TOT
124 Bulletin of Monetary Economics and Banking, October 2011
Cumulatively, the growth of 1% excess money would reduce GDP by 1% within 2 years,
but if at the same time there was an increase financial development of 1%, the cumulative
impact of the decline in GDPin the period of 2 years will be smaller, only amounting to 0.75%.
In contrast, if there is a decrease of 1% excess money, then cumulatively would raise GDP in
the period of 2 year sat 1%. However, if the decrease of 1% excess money is accompanied by
increased financial development by 1% then the cumulative impact of increase in GDP during
the period of two years will be higher, amounting to 1.25%.
Figure 12. Impact of Shock Interaction betweenExcess Money and Financial Development on GDP Growth
Cumulative Increase in GDP %
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20-1.50
-1.00
-0.50
0.00
0.50
1.00
1.50
Shock Increased 1% Excess Money which is accompanied by an increase of 1% Financial DEvelopment
Shock Increased 1% Excess Money
Shock Decline 1% Excess Money which is accompanied by an increase of 1% Financial DEvelopment
Shock Decline 1% Excess Money
V. CONCLUSION
Financial development and economic growth hasa positive relationship, where the
increasing financial development will have a positive impact on economic growth. However,
the impact of risk increase due to market imperfection/assymetric information is not significant
to GDP.
The interaction between the shock in real and monetary sector with financial development
shows that financial development has a positive role dampening the negative shock impact on
GDP growth, while the positive shock impact on economic growth will be amplified.
125Does Financial Development Absorb or Amplify the Shock?
Some factors that influence the movement of long-term growth is the increase of excess
money, TOT and the price movement. Meanwhile, although the financial development has a
positive role in the development of economic growth but its contribution is smaller compared
to these factors.
Related to the price movement, financial development and increased risk due to market
imperfection/assymetric information are not significant in influencing the inflation. This is in
line with the finding that financial development and risk factors do not contribute greatly in
explaining inflation dynamics in the long run.
126 Bulletin of Monetary Economics and Banking, October 2011
REFERENCES
Bacchetta, Philippe & Ramon Caminal (2000), ≈Do capital market imperfections exacerbate
output fluctuations?∆. European Economic Review, No. 44, pp. 449-468.
Beck, Thorsten, Mattias Lundberg & Giovanni Majnoni (2006), ≈Financial Intermediary
Development and Growth Volatility : Do Intermediaries Dampen or Magnify Shocks?∆. Journal
of International Money and Finance, Volume 25, Issue 7, pp. 1146-1167.
Enders, Walter (2004), Applied Econometric Time Series, Wiley Series in Probability and Statistics.
John Wiley & Sons, Inc.
Greene, William H. (2008), Econometric Analysis. Prentice Hall.
Guryay, Erdal, okan Veli Safakli & Behiye Tuzel. (2007). ≈Financial Development and Economic
Growth: Evidence from Nothern Cyprus∆. International Research Journal of Finance and
Economics, Issue 8.
Kuneida, Takuma (2008). ≈Financial Development and Volatility of Growth Rates : New Evidence∆.
MPRA Paper No. 11341.
Schumpeter, J.A. (1911),The Theory of Economic Development, Cambridge, Mass. Harvard
University Press.
Stock, J.H. and M.W. Watson (2001), ≈Vector Autoregression∆.Journal of Economic Perspectives,
15, 4.
127Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly - Traded Banks in Asia
BANK CAPITAL INFLOWS,INSTITUTIONAL DEVELOPMENT AND RISK:
EVIDENCE FROM PUBLICLY - TRADED BANKS IN ASIA
Wahyoe Soedarmono1
This paper examines the relationship between bank capital inflows and financial stability. Using a
sample of publicly-traded commercial banks in Asia over the 2002-2008 period, our empirical results
show that higher banking inflows measured by the share of foreign liabilities in banking reduces systematic
risk, but increases bank-specific risk and total risk. A deeper investigation further suggests that an increase
in total risk and bank-specific risk is driven by strong institutional development. Specifically, higher foreign
liabilities in banking exacerbate bank-specific risk and total risk in countries with greater economic freedom.
Hence, the reinforcement of prudential regulations is necessary to overcome bank-specific risk and total
risk, particularly when the countries move to a more liberal economic environment.
1 The author is a PhD candidate in Money, Banking and Finance at the University of Limoges, France. The views expressed in this paperis the author»s and do not reflect those of the author»s affiliation. The author can be contacted through the following email:[email protected] or [email protected]
Abstract
JEL Classification : : : : : G21, G28, G38
Keywords : Banking Globalization, Economic Freedom, Capital Market Measures of Risk
128 Bulletin of Monetary Economics and Banking, October 2011
I. INTRODUCTION
Despite a large number of studies examining the impact of foreign participation on bank
risk, the current wave of financial globalization and crises again highlight the increasing needs
for better understanding how foreign participation affects financial stability through channels
other than banking ownership or penetration per se. For instance, foreign participation in
banking can be through the presence of foreign managers, foreign debts, or even foreign
customers that demand services from domestic banking institutions.
This paper is the first to analyze the link between foreign participation and bank stability
in the Asian context through financial globalization channel by which foreign counterparts can
play their role in influencing banks» behavior. In this context, we focus on the role of banks»
foreign liabilities, where higher foreign liabilities in banking mean that foreign counterparts
have higher incentives to monitor banks» behavior. Higher foreign liabilities can also be associated
with higher technological innovation and better risk management in banking, so that banks
are able to access financing from international market beforehand. However, higher foreign
liabilities may induce banks to become prone to exchange rate depreciation as observed in the
1997 Asian crisis.
As the wave of financial globalization in Asian banks emerges over the last decades,
foreign participation in banking can also come from foreign debts. Sahminan (2007) works on
the Indonesian banking industry and shows that banks with higher ratio of foreign currency
assets to foreign liabilities are less exposed to exchange rate depreciation which in turn reduces
banks» insolvency risk. Our paper is close to Sahminan (2007) who examines the impact of
exchange rate depreciation on bank stability, but we focus on the role of foreign liabilities in
banking regardless of the exchange rate depreciation aspects. More specifically, our contribution
is threefold.
First, we work on a cross-country setting and focus on the post-1997 Asian crisis period,
while Sahminan (2007) focuses on the Indonesian banking industry in the pre-1997 Asian crisis
period. Second, it is admitted that banks» activities are now increasingly linked to financial
market and thus, contemporary banking crises are due to risks related to financial market
activities. In this regard, the present paper considers various risk measures based financial market
data instead of focusing on banks» balance-sheet indicators. Third, this paper augments the
analysis by examining the role of institutional development. Institutional development has indeed
become an important dimension in attracting foreign participation through the reinforcement
of shareholders» protection and the level of international playing field2. In the Asian context,
institutional development also played a critical role during the 1997 Asian crisis, where Furman
129Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly - Traded Banks in Asia
and Stiglitz (1998) point out that liberalized countries with weak institutional quality are the
ones hardest hit by the 1997 Asian crisis.
In order to investigate the issues raised in this paper, we work on a sample of publicly-
traded banks in seven Asian countries during the 2002-2008 period3. We choose Asian countries
whose data on banking globalization is available, and that are likely to acquire a particular
attention because of their capacity in attracting foreign capital during the last decade, as well
as their openness in permitting foreign participation in banking. These include India, Indonesia,
Hong Kong, Japan, South Korea, Thailand, and the Philippines. The rest of this paper is organized
as follows. Section 2 presents the theory,section 3 describes hypotheses and econometric
methodology. Section 4 discusses empirical findings and presents some robustness checks.
Section 5 concludes and highlights policy implications.
II. THEORY
Regarding the link between foreign participation and bank stability, the first strand of
literature focuses on the direct link with foreign entry or foreign ownership. From a broad
sample of emerging economies, Demirgüc-Kunt and Detregiache (1998) find that the presence
of foreign banks reduces the likelihood of banking crisis. In a similar vein, Detregiache and
Gupta (2004) document that the presence of foreign banks has a stabilizing effect before and
during financial crisis. Levy-Yeyati and Micco (2007) further suggest a positive relationship
between foreign bank entry and bank stability in Latin America. In the emergent economies of
Central and Eastern Europe, Dinger (2009) finds that the presence of foreign banks reduces the
risk of aggregate liquidity shortages. Foreign banks are also shown to have better position than
domestic banks without foreign participation, since foreign banks can provide modern banking
activities with better technological innovation and risk management, as well as access to
international financial markets (Berger et al., 2001; Bonin et al., 2005). On the contrary, some
have argued that domestic banks are better than foreign banks because domestic banks are
unlikely to suffer from a home-bias risk that exacerbates agency problems due to cultural
differences and regulatory environments (Berger et al., 2001; Lensink and Naaborg, 2007).
2 Institutional development also affects the link between financial development and economic growth, so that financial sector doesnot always boost economic growth. Specifically, threshold effects in the finance-growth nexus are likely to exist in developingcountries, which in most cases, lack of institutional development (Augier and Soedarmono, 2011).
3 The reason why we focus on the Asian banking industry is that Asian banking has experienced substantial changes in terms offoreign participation, particularly after the 1997 Asian crisis period. This includes a rapid growth of foreign ownership in bankingbecause of bank mergers and acquisitions (M&As), as well as an increase in foreign direct investment entering Asian banks. SeeSoedarmono et al. (2011a) for further discussion.
130 Bulletin of Monetary Economics and Banking, October 2011
The second strand of literature is associated with the role of bank competition. Jeon et al.
(2011) highlight from a sample of Asian and Latin American banks during 1997-2008that
higher foreign participation in terms of bank ownership enhances competition in the banking
market through spillover effects from foreign partners to domestic counterparts.However,
empirical results on the implications of foreign participation through the link between bank
competition and financial stability are mixed. Working on a broad set of commercial banks
across developing countries over the 1999-2005 period, Ariss (2010) finds that the higher bank
market power, the lower risk and the higher profit efficiency of banks, although higher market
power deteriorates cost efficiency. On the other hand, Soedarmono et al (2011a) focus on the
Asian banking industry and find that banks in less competitive market tend to have higher
insolvency risk due to moral hazard effects.
This paper employs a set of dependent variables to assess financial stability. Specifically,
we use three types of risk measure coming from financial market data. These include total risk
(TRISK), systematic risk (BETA) and idiosyncratic risk or bank-specific risk (SRISK). To calculate
these risk measures, we initially construct the standard market model as follows4.
where ri,j,t
is bank i stock return at country j and at day t, while r mj,t
is the daily market
returns computed on the basis of domestic market index at country j. In the meantime, ri,j,t
and
r mj,t
are calculated as follows
and
where ri,t
and rm (j),t
are daily banks» stock price and total market index, respectively. In
constructing Equation (1), we impose a criterion to deal with low bank trading volume.
Specifically, we eliminate banks whose trading days are less than 70% of the total number of
trading days. For each year during the 2002-2008 period, moreover, Equation (1) is estimated
by applying the Panel Least Squares method. Hence, we obtain Equation (1) for each bank i
year by year.
TRISK is the annual standard deviation of daily bank stock returns during the 2002-2008
period, where the measure of daily bank stock returns is expressed by ri,j,t
. BETA is the annual
(1)
(2)
4 The model specification follows Bautista et al (2009).
131Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly - Traded Banks in Asia
systematic risk or the beta, β, coefficient in Equation (1). Systematic risk is a risk linked to
financial market activities and thus, is often referred to as market risk or non-diversifiable risk.
SRISK is the annual bank-specific risk or idiosyncratic risk which is diversifiable through several
risk managements at the bank level. SRISK is represented by the annual residual terms of the
standard market model, ε, as presented in Equation (1).
Variable of interest in this study is the level of globalization in banking activities and the
degree of economic freedom. We use country-specific data from the International Investment
Position section of the International Financial Statistics as a proxy of banking globalization.
Specifically, we use the share of international liabilities in banking sector over total international
liabilities (BLIAB). Greater BLIAB is associated with higher foreign participation in banking activities
and thus, representing greater financial globalization in banking. In parallel, the degree of
economic freedom is assessed by the Economic Freedom index (FREEDOM) coming from Heritage
Foundation. FREEDOM is a composite index of 10 indicators ranking policies in the areas of
trade, government finances, government interventions, monetary policy, capital flows and foreign
investment, banking and finance, wages and prices, property rights, regulation and black market
activity. The index scores from 0 to 1 with higher scores indicating policies being more conducive
to competition and economic openness.
Several control variables are also included in this study. First, we include the ratio of total
loans to total assets (LOAN) to account for bank opacity. Greater LOAN is likely to exacerbate
asymmetric information between banks and borrowers, while LOAN is the major source of
bank risk. Furthermore, we incorporate the ratio of total deposits to total assets (DEPO), as
bank deposits are also the major source of risk, particularly when market discipline is weak. We
also incorporate the ratio of loan loss reserves to total gross loans (LLR) as a proxy of credit risk,
where higher LLR is expected to increase total risk, systematic risk or bank-specific risk (Agusman
et al, 2008). Following Agusmanet. al. (2008) as well, we incorporate the ratio of equity to
total assets (EQTA) and the ratio of liquid assets to total assets (LIQUIDITY) to account for
leverage risk andliquidity risk, respectively. Higher EQTA and LIQUIDITY are expected to enhance
financial stability.
III. METHODOLOGY
There are two hypotheses tested in this paper. First, we aim to test if there is a relationship
between globalization in banking and financial stability as measured by total risk, systematic
risk and idiosyncratic risk. Second, we examine whether or not institutional development affects
the impact of banking globalization on financial stability. Let i, j, and t be bank index, country
132 Bulletin of Monetary Economics and Banking, October 2011
index, and time index, respectively, then both steps are respectively presented in the following
equations.
RISK consists of TRISK, BETA and SRISK, while Control represents a set of control variables.
Initially, we calculate risk indicators coming from financial market data. We then run the
regressions of banking globalization on these indicators. Moreover, we also investigate the
impact of the interaction term between banking globalization and institutional development
on financial stability. To estimate such relationships, we use the Fixed Effect regressions with
heteroscedasticity-consistent standard errors.
In the present paper, we use bank-specific and country-specific data. For bank-specific
data, we retrieve banks» financial indicators from BankScope Fitch IBCA over the 2002-2008
period. Our initial sample consists of 189 publicly-traded commercial banks in seven Asian
countries. These include India, Indonesia, Hong Kong, Japan, Philippines, South Korea and
Thailand. In the meantime, we retrieve banking globalization data as country-specific indicator
from International Financial Statistics provided by the IMF. Other country-specific indicator such
Table 1.Descriptive statistics
TRISK 0.024158 0.022083 0.135767 0.00037 0.01215 1181BETA 0.832851 0.876176 1.967998 -0.28694 0.403701 1181SRISK 0.020088 0.01783 0.134859 0.000369 0.011475 1181BLIAB 0.191862 0.182324 0.396951 0.021616 0.111458 1323LOAN 0.58088 0.600438 0.886473 0.032652 0.1294 1219DEPO 0.867947 0.889310 0.970079 0.073737 0.096575 1225LLR 0.034881 0.02017 0.80149 5.00E-05 0.05337 1135EQTA 0.068095 0.05796 0.57868 0.00009 5.089816 1226LIQUIDITY 0.042605 0.001 0.57 0.0001 0.101373 916FREEDOM 0.629607 0.637 0.90 0.512 0.094772 1820
Variables Mean Median Maximum Minimum Std.Dev. Obs.
Source : Author»s calculationNote : TRISK is the indicator of total risk measured by the annual standard deviation of daily bank stock returns. BETA is the annualbeta coefficient from the standard market model, in which BETA represents market risk or systematic risk. SRISK is bank-specific riskmeasured by the annual residual term of the standard market model. BLIAB is the aggregate share of foreign liabilities in banking overtotal foreign liabilities. LOAN is the ratio of total loans to total assets. DEPO is the ratio of total deposits to total assets. LLR is the ratioof loan loss reserves to total gross loan. EQTA is the ratio of total equity to total assets. LIQUIDITY is the ratio of liquid assets to totalassets. FREEDOM is the Economic Freedom index retrieved from Heritage Foundation.
(3)
(4)
133Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly - Traded Banks in Asia
as institutional development comes from Heritage Foundation, where we assess institutional
development by using the Economic Freedom index. In order to assess financial stability based
on financial market data, we retrieve daily bank stock prices and daily total market index during
2002-2008 from Thomson Datastream International.
Before we run regressions, we impose restrictions to our dataset to deal with outliers and
missing values. We exclude all values that are less than zero for LLR, EQTA and LIQUIDITY. In
Table 1, we present the descriptive statistics of all clean variables used in this study.
IV. RESULT AND ANALYSIS
4.1. Empirical results
Table 2 shows the relationship between foreign liabilities in banking (BLIAB) and total risk
(TRISK). It can be seen that higher foreign liabilities in banking tend to exacerbate total risk.
Table 2.The relationship between banking globalization, economic freedom and total
BLIAB 0.0129*** 0.0131*** 0.0135*** 0.0101*** 0.0081** 0.0143*** -0.3622***(2.083) (2.152) (2.215) (2.606) (2.142) (3.499) (-4.369)
LOAN -0.0275*** -0.0258*** -0.0195 -0.0194 -0.0212 -0.0321***(-5.048) (-4.643) (-1.443) (-1.451) (-1.242) (-4.421)
DEPO -0.0125* 0.0061 -0.0148 -0.0093 -0.0126(-1.655) (0.6927) (-1.431) (-0.7805) (-1.179)
LLR 0.0559*** 0.0562*** 0.0541*** 0.0505***(3.901) (4.504) (4.316) (5.147)
EQTA -0.0627*** -0.0667*** -0.0698***(-2.909) (-2.738) (-3.735)
LIQUIDITY 0.0161 0.0172**(1.565) (2.464)
FREEDOM -0.0354(-1.079)
BLIAB*FREEDOM 0.5964***(4.944)
Obsevation 1181 1118 1118 1043 1043 859 859Adj R-square 0.42 0.42 0.42 0.44 0.45 0.44 0.51
ExplanatoryVar.
Source: Author»s calculation.Note: Dependent variable is TRISK which represents the indicator of total risk and measured by the annual standard deviation of dailybank stock returns. BLIAB is the aggregate share of foreign liabilities in banking over total foreign liabilities. LOAN is the ratio of totalloans to total assets. DEPO is the ratio of total deposits to total assets. LLR is the ratio of loan loss reserves to total gross loan. EQTAis the ratio of total equity to total assets. LIQUIDITY is the ratio of liquid assets to total assets. FREEDOM is the Economic Freedomindex retrieved from Heritage Foundation. Estimations are carried out using the Panel Fixed Effect regressions by considering White»sheteroscedasticity-consistent standard errors. Constants are included but not reported. ***,**,* indicate significant at the 1%, 5%and 1% levels, respectively.
Dependent Var. : TRISK
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
134 Bulletin of Monetary Economics and Banking, October 2011
This finding is robust to various control variable modifications as presented in Model 1 to
Model 6. Moreover, we examine how the interaction term between institutional development
and foreign liabilities in banking affects total risk, as shown in Model 7. It shows that the
positive impact of banks» foreign liabilities on total risk is dependent on institutional development.
More precisely, only in countries with higher economic freedom, the positive relationship between
banking globalization and total risk holds. This further suggests that in countries with lower
economic freedom, higher foreign liabilities in banking reduce total risk.
Meanwhile, Table 3 presents the impact of banks» foreign liabilities (BLIAB) on systematic
risk (BETA). By estimating different model specifications as presented in Model 1 to Model 6, it
shows that banks» foreign liabilities have a stabilizing effect in terms of reducing systematic
risk. But, the link between banks» foreign liabilities and systematic risk is no longer significant
when we take into account the role of institutional development as shown in Model 7. Economic
freedom does not seem to influence the relationship between bank globalization and systematic
risk.
Table 3.The relationship between banking globalization, economic freedom and systematic risk
BLIAB -1.984*** -2.032*** -2.041*** -2.129*** -2.163*** -2.161*** -3.105(-11.689) (-11.924) (-11.972) (-12.319) (-12.493) (-11.786) (-1.335)
LOAN -0.6573*** -0.6977*** -0.5690*** -0.5684*** -0.4779** -0.4672**(-4.303) (-4.485) (-3.161) (-3.164) (-2.381) (-2.297)
DEPO 0.2874 0.1649 -0.1927 -0.2190 -0.2168(1.361) (0.6918) (-0.6684) (-0.7301) (-0.7216)
LLR -0.2086 -0.2051 -0.1372 -0.0799(-0.7807) (-0.7693) (-0.515) (-0.2903)
EQTA -1.073** -1.109** -1.153**(-2.197) (-2.134) (-2.198)
LIQUIDITY -0.1841 -0.1932(-0.9408) (-0.9849)
FREEDOM -0.6407(-0.6964)
BLIAB*FREEDOM 1.245(0.3679)
Obsevation 1181 1118 1118 1043 1043 859 859Adj R-square 0.61 0.61 0.61 0.62 0.62 0.64 0.64
ExplanatoryVar.
Source: Author»s calculation.Note: Dependent variable is BETA which represents the indicator of systematic risk and measured by the annual beta coefficient ofthe standard market model. BLIAB is the aggregate share of foreign liabilities in banking over total foreign liabilities. LOAN is the ratioof total loans to total assets. DEPO is the ratio of total deposits to total assets. LLR is the ratio of loan loss reserves to total gross loan.EQTA is the ratio of total equity to total assets. LIQUIDITY is the ratio of liquid assets to total assets. FREEDOM is the EconomicFreedom index retrieved from Heritage Foundation. Estimations are carried out using the Panel Fixed Effect regressions by consideringWhite»s heteroscedasticity-consistent standard errors. Constants are included but not reported. ***,**,* indicate significant at the1%, 5% and 1% levels, respectively.
Dependent Var. : BETA
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
135Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly - Traded Banks in Asia
In terms of the link between banks» foreign liabilities (BLIAB) and bank-specific risk (SRISK),
Table 4 presents our empirical findings for a number of model specifications. Higher foreign
liabilities in banking increase idiosyncratic risk, but again this relationship is dependent on
economic freedom. Only in countries with greater economic freedom, higher foreign liabilities
in banking worsen bank-specific risk. This may be due to the fact that under higher economic
freedom, foreign liabilities held by banks may be excessive, as banks are free to raise funding
from international market. Hence, the countries with greater economic freedom can be more
exposed to exchange rate depreciation that in turn exacerbates bank-specific risk.
Table 4. The relationship between banking globalization,economic freedom and idiosyncratic risk (bank-specific risk)
BLIAB 0.0221*** 0.0212*** 0.0216*** 0.0174*** 0.0157*** 0.0201*** -0.1618**(3.969) (3.945) (4.019) (3.204) (2.896) (3.228) (-2.098)
LOAN -0.0362*** -0.0344*** -0.0295*** -0.0294*** -0.0331*** -0.0386***(-7.515) (-7.027) (-5.197) (-5.227) (-4.876) (-5.719)
DEPO -0.0125 0.0054 -0.0128 -0.0082 -0.0099(-1.885) (0.7135) (-1.416) (-0.810) (-0.9901)
LLR 0.0546*** 0.0548*** 0.0524*** 0.0503***(6.489) (6.555) (5.817) (5.519)
EQTA -0.0545*** -0.0589*** -0.0602***(-3.558) (-3.347) (-3.462)
LIQUIDITY 0.0131** 0.0136**(1.969) (2.099)
FREEDOM -0.0141(-0.4608)
BLIAB*FREEDOM 0.2894**(2.579)
Obsevation 1181 1118 1118 1043 1043 859 859Adj R-square 0.47 0.49 0.49 0.52 0.52 0.51 0.53
ExplanatoryVar.
Source: Author»s calculation.Note: Dependent variable is SRISK which represents the indicator of idiosyncratic risk and measured by the annual residual term of thestandard market model. BLIAB is the aggregate share of foreign liabilities in banking over total foreign liabilities. LOAN is the ratio oftotal loans to total assets. DEPO is the ratio of total deposits to total assets. LLR is the ratio of loan loss reserves to total gross loan.EQTA is the ratio of total equity to total assets. LIQUIDITY is the ratio of liquid assets to total assets. FREEDOM is the EconomicFreedom index retrieved from Heritage Foundation. Estimations are carried out using the Panel Fixed Effect regressions by consideringWhite»s heteroscedasticity-consistent standard errors. Constants are included but not reported. ***,**,* indicate significant at the1%, 5% and 1% levels, respectively.
Dependent Var. : SRISK
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
This finding is somehow consistent to Sahminan (2007), where there is a positive
relationship between exchange rate depreciation and insolvency risk for banks with higher
share of foreign liabilities. To the extent that only idiosyncratic risk plays a significant role in
136 Bulletin of Monetary Economics and Banking, October 2011
capturing bank instability in Asia (Agusman et al, 2008), our findings are also consistent with
Agusman et al (2008). Specifically, systematic risk in the present paper is not an important
source of instability due to higher banks» foreign liabilities, regardless of the level of countries»
institutional development. Overall, the empirical findings suggest that investors in countries
with higher economic freedom are more concerned with total risk and bank-specific risk in this
regard.
With regards to control variables, higher banks» loan portfolio (LOAN) is associated with
lower total risk, lower systematic risk and lower bank-specific risk. This suggests that bank
lending activities are not source of instability. This result contradicts with the nature of bank
loan portfolio which is opaque. Presumably, Asian banks suffer from a managerial self-interest
problem in which managers in banks with higher asymmetric information might be driving
banks to become safer by holding less risky loan portfolios (Bris and Cantale, 2004; Soedarmono
et al, 2011b). Bank deposits (DEPO) also does not seem to be the source of instability. In line
with Agusman et al (2008), higher loan loss reserves ratio (LLR) is associated with higher total
risk and bank-specific risk. Meanwhile, the link between bank capitalization (EQTA) and RISK
fulfills our expected sign. Higher bank capital ratio reduces total risk, systematic risk and bank-
specific risk.
4.2. Robustness checks
In order to further ensure for robustness linked to variable omission issues, we also perform
some sensitivity analyses5. First, it is shown that Model 6 and 7 from Table 2, 3 and 4 suffer
from observation loss when we incorporate LIQUDITY as control variable. To ensure that the
empirical results obtained are not due to observation bias, we exclude LIQUIDITY and re-estimate
Model 6 and 7 in all three cases. However, this consideration does not change our empirical
results discussed in Section 4.1. Second, due to the fact that our bank sample comes from
countries with different level of macroeconomic environment, we also consider the influence
of economic development and inflation rate to account for this dimension. In other words, we
include the real gross domestic product (GDP) and inflation rate (INF) as control variable. On the
whole, our empirical results are not altered with regards to the link between banking
globalization, economic freedom and financial stability as measured by total risk, systematic
risk and idiosyncratic risk.
5 The results are not presented in the paper, but are available upon request.
137Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly - Traded Banks in Asia
V. CONCLUSION
In the aftermath of the 1997 Asian crisis, financial globalization in Asian banking in the
form of FDI and foreign ownership, as well as a rapid growth of foreign capital inflows entering
Asian countries due to the 2008 credit crisis and the 2010 European debt crisis, urgently require
better understanding whether or not greater foreign participations enhance financial stability.
Nowadays, such an increasing trend of foreign participations also show that Asian countries
are already in a better position in terms of institutional development, so that they have been
successful in attracting foreign participations. However, there is no attempt to investigate the
impact of foreign participations and institutional development on financial stability in the Asian
context.
This paper attempts to fulfill such a gap by assessing the impact of foreign participation
on financial stability through channel other than foreign participation commonly used in the
previous literature, such as bank foreign ownership, foreign bank entry, or bank competition.
In the present paper, we consider the indicator of foreign participation measured by the aggregate
share of foreign liabilities in banking over total foreign liabilities. Using a sample of publicly-
traded commercial banks in seven Asian countries during 2002-2008, the empirical results
from the Fixed Effect regressions show that higher foreign liabilities in banking reduces systematic
risk, but exacerbates bank-specific risk and total risk. However, we further shows that such
findings only hold for countries with greater economic freedom, suggesting the needs for
enhancing prudential regulations when banking in a more liberal environment are more
globalized in terms of their abilities to raise financing from international financial market. To
this end, bank-specific risk and total risk can be reduced, bank failure can be prevented, and
the systemic risk of bank failure can be avoided.
138 Bulletin of Monetary Economics and Banking, October 2011
REFERENCES
Agusman, A., Monroe, G.S., Gasbarro, D., Zumwalt, J.K., 2008. Accounting and capital market
measures of risk: Evidence from Asian banks during 1998-2003. Journal of Banking and
Finance 32, 480-488
Ariss, R.T., 2010. On the implications of market power in banking: Evidence from developing
coutnries. Journal of Banking and Finance 34 (4), 765-775
Augier, L., Soedarmono, W., 2011. Intermédiation financière, croissance et effet de seuil. Revue
Economique (forthcoming)
Bautista, C., Rous, P., Tarazi, A., 2009. The determinants of bank stock return»s co-movements
in East Asia. Economics Bulletin 29(3), 1596-1601
Berger, A.N., DeYoung, R., Genay, H., Udell, G.F., 2001. Globalization of financial institutions:
Evidence from cross-border banking performance. Federal Reserve Bank of Chicago, Working
paper 1999-25
Bonin, J.P., Hasan,I.,Wachtel, P., 2005. Bank performance, efficiency and ownership in transition
countries.Journal of Banking and Finance29, 31-53
Bris, A., Cantale, S., 2004. Bank capital requirement and managerial self-interest. Quarterly
Review of Economics and Finance 44, 77-101.
Demirgüc-Kunt, A., Detregiache, E., 1998.The determinants of banking crises in developing
and developed countries.IMF Staff Papers 45 (1).
Detregiache, E., Gupta, P., 2004. Foreign banks in emerging market crises: Evidence from
Malaysia, IMF Working Paper 04/129
Dinger, V., 2009. Do foreign-owned banks affect banking system liquidity risk? Journal of
Comparative Economics 37, 647-657
Jeon, B.N., Olivero, M.P., Wu, J., 2011. Do foreign banks increase competition? Evidence from
emerging Asian and Latin American banking markets. Journal of Banking and Finance 35,
856-875
Lensink, R.,Naaborg, L., 2007. Does foreign ownership foster bank performance.Applied Financial
Economics17, 881-885
Levy-Yeyati, E., Micco, A., 2007. Concentration and foreign penetration in Latin American
banking sectors: Impact on competition and risk. Journal of Banking and Finance 31, 1633-
1647
139Bank Capital Inflows, Institutional Development and Risk: Evidence from Publicly - Traded Banks in Asia
Sahminan, S., 2007.Effects of exchange rate depreciation on commercial bank failures in
Indonesia.Journal of Financial Stability 3, 175-193
Soedarmono, W., Machrouh, F., Tarazi, A., 2011a. Bank market power, economic growth and
financial stability: Evidence from Asian banks. Journal of Asian Economicsdoi: 10.1016/
j.asieco.2011.08.003.
Soedarmono, W., Rous, P., Tarazi, A., 2011b. Bank capital and self-interest managers: Evidence
from Indonesia. LAPE Working Paper, Université de Limoges
140 Bulletin of Monetary Economics and Banking, October 2011
This page is intentionally left blank
141Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
COMPETITIVE CONDITIONS IN BANKING INDUSTRY:AN EMPIRICAL ANALYSIS OF THE CONSOLIDATION,
COMPETITION AND CONCENTRATION INTHE INDONESIA BANKING INDUSTRY BETWEEN 2001 AND 2009
Tri Mulyaningsih 1
Anne Daly 2
Few large banks dominate the Indonesia banking industri. Furthermore, in the past ten years, there
were a series of mergers and acquisitions in the banking market. The facts cause implications on competition.
In this paper, we examine these issues exploiting an unconsolidated annual financial report of all commercial
banks between 2001 and 2009. The Panzar-Rose method is employed to examine the banks behavior in
competition. Estimates indicate that banks in all three subsamples, large; medium-sized and small are
working in a monopolistically competitive market. The analysis of market concentration supports the
conventional view that concentration impairs competition. The study shows that the most competitive
market was the medium-sized banks because it was least concentrated. In contrast, the large market was
more concentrated thus it was less competitive. The consolidation policies driven by the Central Bank
reduced market concentration because mergers and acquisitions were mostly conducted by the medium-
sized and small banks. Further the improvement of market share distribution and the increasing capacity
of the merging banks enhanced competition in the Indonesia banking industry.
Keywords: Banking, market competition, market structure
JEL Classification: D43, G21
1 Doctoral Student, Faculty of Business and Government, University of Canberra Corresponding author: [email protected] Professor in Economics, Faculty of Business and Government, University of Canberra Corresponding author: [email protected]
Abstract
142 Bulletin of Monetary Economics and Banking, October 2011
I. INTRODUCTION
The Indonesia banking industry experienced a structural change in 1990 where the number
of banks increased significantly. The change in market structure was promoted by the
deregulation policy in the banking sector in 1980s. Through the second deregulation in October
1988 the government removed barriers to entry to the banking sector. The minimum capital
requirement was lowered and it was easier for banks to obtain licenses to undertake foreign
exchange operations (McLeod, 1999). In addition, if the license was obtained, all branches
were automatically eligible to provide foreign exchange services.
After a series of deregulation policies during 1980s, the authorities decided to reduce
the speed of the banking expansion. During 1995 and 1996, the Central Bank introduced
some policies related to the prudential aspect of banks. McLeod (1999, p. 281-285) explained
that during those periods the Central Bank re-regulated the industry by re-imposing controls of
Table 1.The Microeconomic Policies in Indonesia Banking 1983 - 1997
Y e a r Policy Details
Removing the control on the interest rates of time deposits of State Owned Banksand the lending of all banks.
1. Opening the banking industry to the new private banks and the new joint venturebanks by lowering minimum capital requirement.
2. Removing business restrictions such as easing requirement of the existing banksto expand their branch easily; limit on interbank borrowing and permitting banksto introduce savings deposits product of their own design.
1. Opening the market by permitting foreign investors to purchase shares indomestic banks listed on the stock exchange
2. Partial privatizing by permitting State Owned Banks to list on capital market
Re-RegulationRe-RegulationRe-RegulationRe-RegulationRe-Regulation
1. Re-imposing the controls of bank lending;2. Extending the control to bank involvement with commercial paper issues;3. Extending the supervisory to non-bank finance company; increased the required
reserve ratio;4. Tightening of licensing of new bank branches;5. Imposing the fines on banks which expand more rapidly than permitted;6. Increasing the required reserve ratio and tightening prudential regulation
DeregulationDeregulationDeregulationDeregulationDeregulation
June 1983
October 1988
February 1992
1995-1997
Source: McLeod (1999, p. 293-295) and Chua, BH (2003)
143Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
bank lending (in 1995); extending the control to bank involvement with commercial paper
issues (in August 1995); extending supervision to non-bank finance companies (December
1995); increasing the required reserve ratio (February 1996); tightening of licensing of new
bank branches (June 1996); increasing the required reserve ratio and tightening prudential
regulation (in April 1997). Further, the 1997 economic crisis had heightened the importance of
re-structuring the industry as well as increasing prudential concern.
The consolidation was started on December 1997. The 1997 economic crisis generated
distress in the banking industry. In order to improve the performance of the state owned banks,
the Central Bank decided to merge four State Owned Banks. The authorities also closed the
operation of twenty three banks in 1997. The liquidation policies contributed to the reduction
on the number of banks. The consolidation process was continued by the introduction of the
Indonesia Banking Architecture (API) on January 2004. The policy essentially encourages banks
to reach their economies of scale as well as fosters the creation of a healthy banking system.
Table 2.The Microeconomic Policies in Indonesia Banking 1997 - 2010
Y e a r Policy Details
1. Banks liquidation (23 banks).
2. Banks recapitalization
3. Merger of 4 State Owned Banks became Bank Mandiri
Privatization of the bailout banks under the Indonesia Banking Restructuring Agency
(IBRA)
The introduction of the Indonesia Banking Architecture (API)
A series of mergers and consolidation conducted in banking to comply with Single
Presence Policy and Minimum Capital Requirement.
ConsolidationConsolidationConsolidationConsolidationConsolidation
1997
2003
2004
2004 - 2010
Source: Chua, BH (2003) and Central Bank of Indonesia (2010)
The banking architecture provides direction of the Indonesia banking system development
in the next ten years. The policy aims to create a strong, healthy and efficient banking structure.
In order to achieve its objectives, the Central Bank developed six pillars which are creating a
healthy domestic banking industry; inventing an effective system for regulating the banking
sector based on international standards; increasing the monitoring function of central banks
based on international standards; creating a strong banking industry with highly competitive
144 Bulletin of Monetary Economics and Banking, October 2011
banks and fostering good corporate governance; realizing the proper infrastructure to support
the creation of a healthy banking system; and lastly increasing consumers» empowerment and
protection (Bank of Indonesia, 2010).
We suspect that at least two policies under the Indonesian Banking Architecture will
directly affect the market structure and competition in the banking industry. First is the minimum
capital requirement as regulated under the Central Bank Regulation number 10/15/PBI/2005.
The second policy is the single presence policy as explained in the Central Bank Regulation
number 8/16/PBI/2006.
Under Indonesian Banking Architecture, banks must increase their capital which is aligned
with their business scale (Bank of Indonesia, 2010). Bigger capital will enable banks to maintain
their business and risk develop the infrastructure such as information technology and increase
the scale of support of the expansion of credit capacity. Capital consists of in-paid capital and
disclosed reserves. According to the regulation, all existing banks, including the banks established
by regional governments, should attain minimum capital of Rp100 billion by 31 of December
2010 otherwise the Central Bank will impose several restrictions to the banks. In order to
increase capital, banks are allowed to receive additional capital injections from existing owners,
Figure 1.The Vision of Indonesian Banking Architecture Policy in 2014
Capital(Trillion)
Micro Bank Bank with Limited Business
Bank with Certain Focus
Corporate Retail Others
NationalBank
50
0.1
10
International
Local
Source: The Central Bank of Indonesia, (2010), Indonesian Banking Architecture, www.bi.go.id
145Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
merge with other banks, be acquired by bigger banks, or sell their shares on the capital market
(Bank of Indonesia, 2010). In the next ten or fifteen years, the Central Bank plans to reduce the
number of banks to 60, which consisting of 2-3 international banks, 3-5 national banks, and
30-50 specialized banks (Bank of Indonesia, 2010).
The single presence policy was introduced to re-arrange the structure of bank ownership
(the regulation of Bank of Indonesia, Number 8/16/PBI/2006). Single ownership refers to the
condition where one party is just eligible to be a main shareholder in one bank. The regulation
applies to shareholders who owned more than 25 per cent and those with ownership less than
25 percent but can control the banks. According to the rule, there should be an adjustment of
ownership structure by transferring the ownership (some or all) to one bank. Therefore they
only become a main shareholder in one bank. In addition, banks under the same ownership are
encouraged to merge. At last, they can set up a Bank Holding Company.
The study aims to assess the impact of consolidation policies in the banking industry on
market condition. The consolidation policy was started from the 1997 economic crisis and was
further enhanced by the introduction of the Indonesia Banking Architecture on January 2004.
Referring to Table 3, a large number of mergers and acquisitions took place in the Indonesia
banking industry between 1997 and 2010. One merger occurred in large bank market conducted
by the Bank Niaga and Bank Lippo in order to comply with the single presence policy. In addition,
there were 7 mergers in medium-sized banks and 7 mergers conducted by small banks.
A series of mergers and acquisitions reduces the number of banks. This raises some
important issues such as, whether a smaller number of banks reduced or increased the degree
of concentration in the banking industry; whether a smaller number of banks created a more
concentrated market or improved the market share distribution. Further we are also interested
in understanding the competition in the banking industry during the implementation of
consolidation policy, and whether consolidation generates a significant change in the
competition. Lastly, this paper would like to examine the link between the degree of market
concentration and the nature of competition during the consolidation process in the Indonesia
banking industry.
146 Bulletin of Monetary Economics and Banking, October 2011
Table 3.The list of mergers and acquisition between 2000 and 2010
Bank Category Merging Bank
PT Bank Niaga
PT Bank Lihalo
Bank Dai-Ichi Kanggo
Bank IBJ Indonesia
Bank Bali
Bank Artha Media
Bank Universal
Bank Prima Express
Bank Patriot
PT Bank Sumitomo Mitsuo
Indonesia
Sakura Swadarma Bank
UFJ Indonesia Bank
Tokai Lippo Bank
UFJ Indonesia
PT Bank of Tokyo
Mitsubishi
Bank Hagakita
Bank Haga
Bank Rabobank Duta
Bank Buana
Bank UOB Indonesia
Bank Pikko
Bank CIC
Bank Danpac
Bank Artha Graha
Bank Inter-pacific Tbk.
Commonwealth Indonesia
Artha Niaga Kencana
Bank Multicor
Bank Windu Kentjana
Bank Harmoni
International
Bank Index Selindo
Bank Haga
Bank Hagakita
Bank OCBC
Bank NISP
Large Banks
Medium-Sized
Banks
Small Banks
Source: Banks» Annual Financial Report Published by the Central Bank of Indonesia
No Year Nama of New Bank
1
1
2
3
4
5
6
7
1
2
3
4
5
6
7
2008
2000
2001
2001
2001
2006
2008
2010
2001
2001
2004
2005
2007
2007
2008
2008
2009
PT Bank CIMB Niaga Tbk
PT Bank Mizuho Indonesia
PT Bank Permata Tbk
PT Bank Sumitomo Mitsuo Indonesia
UFJ Indonesia Bank
PT Bank of Tokyo Mitsubishi UFJ Ltd.
PT Bank Rabobank International
Indonesia Bank
PT Bank UOB Buana Tbk
PT Bank Mutiara Tbk
PT Bank Artha Graha International
Tbk
PT Bank Commonwealth
PT Bank Windu Kentjana
International Tbk.
PT Bank Index Selindo
Rabobank Duta Bank
PT Bank OCBC-NISP Tbk
147Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
II. THEORY
Many studies were interested to assess the impact of consolidation on concentration and
competition. Further, some journal articles also discuss the relationship between the degree of
concentration and competition. There are two prominent approaches dealing with this issue,
which is structural and non-structural (Bikker and Haaf, 2002).
According to the first approach, there is a direct relationship between market structure,
firms behavior and industry performance. This approach is based on traditional structure-conduct-
performance hypotheses (SCP). The second approach is known as the non-structural approach.
It has a different view on the relationship between those three elements. As explained by
Shaffer (1994a), competitive outcomes such as efficient prices can be established in un-
concentrated as well as highly concentrated markets therefore the link between market structure
and performance is not linear. As the relationship among market structure and performance is
not linear, they suggested focusing on the competitive conduct of banks instead of relying on
the market structure (Bikker and Haaf, 2002).
2.1. Structural Approach
The first foundation of traditional structure-conduct-performance approach was built by
Manson (1939). He concluded that fewer firms in the market will generally lead to less competitive
conduct, in terms of higher prices and reduced output levels, simply likes a monopolist model.
Furthermore, a concentrated market will produce less competitive performance, where the
price ratio to cost will be higher at the expense of lower consumer welfare. The small number
of firms might also facilitate firms to collude with their competitors. The collusion aims to boost
price, so it will be much higher than marginal cost (Yeyati and Micco, 2003b).
On the other hand, the competitive market will produce an efficient outcome as price
equals marginal cost. Thus, an increase in firm numbers will lead to more competitive conduct
by lowering price and reducing firms» profitability. In addition, the structure-conduct performance
(SCP) approach believed that the competitive market, which is produced by a low concentration
in the market, will deliver higher consumer welfare (Shaffer, 1994a). Efficient firms are able to
produce a higher output with lower prices.
Bain (1951) tested the structure-performance hypotheses empirically on American
manufacturing industry during period 1936 to 1940. His research confirmed the structure-
performance hypotheses. By using a Z score test, he compared the profit rates of a group of
firms that worked in a high concentrated market and those that worked in less concentrated
148 Bulletin of Monetary Economics and Banking, October 2011
market. He found that the profits rates of firms in industries of high seller concentration should
on average be larger than those firms in industries of lower concentration.
Calem and Carlino (1991) found a correlation between market concentration and
performance in banking. Bain employed profit to assess market performance, Calem and Carlino
tried to use price to reflect market performance. Their conclusion supported traditional structure-
performance hypotheses that the concentrated market might contribute to collusive behavior
(Calem and Carlino, 1991). Further, the research claimed that the concentrated market is inefficient
because prices are higher than marginal cost. It is also not equitable because concentrated
market produces higher profits at the expense of consumers (Berger and Hannan, 1989).
The structural approach has become the foundation of antitrust policy in many countries.
The US Department of Justice has long adhered to this view by maintaining an explicit policy of
challenging mergers between rival firms that result in concentration levels above certain threshold
(Shaffer, 1994b). The Indonesia Legislation number 5, 1999 employs the structural approach to
determine the unlawful conduct of firms based on its impact to competition. The Law applies
to all industries. Further, according to the Government Regulation number 28/ 1999 on article
8 (2), we can find that the merged banks could not have the asset (after merger) of more than
20 percent of total assets in the banking industry. Thus, 20 percent is perceived as the threshold
where the merger would bring a negative effect into competition.
2.2. Non Structural Approach
The first argument against the traditional structure performance hypotheses came from
Demzets (1973) and Pelzman (1977). According to them, the source of concentration is efficiency
instead of market power. Their finding is labeled as Efficiency-Structure hypotheses. They
explained that the difference in firm-specific efficiencies within markets can create unequal
market shares and high level of concentration. The difference in efficiency might be derived
from superior management and production technology (Neuberger, 1997).
Manson implicitly assumes that the concentrated market facilitates firms to boost price.
Therefore market, where the price is higher than the marginal cost, perceives as less efficient.
Shaffer (1994b) argued that lower price is not a good indication to measure the market efficiency.
He defended his idea with the explanation that any extra efficiency among leading firms in a
competitive market would tend to show up as lower, not higher, prices.
SCP assumes that there is a single-way relationship between structure, conduct and
performance. Hence, market structure affects the firms» behavior further the behavior influences
149Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
the market performance. The causal relationship becomes unclear because firms» decision to
enter the market might be affected by the expectations of the degree of competition in the
market stage (Vesalla, 1995, p. 18). The non-structural approach argued that both the market
structure and firms» conduct are endogenous, since there is a feedback effects from conduct
back to market structure (Vesalla, 1995, p.18). In addition, market performance may also
influence the firms» decision to enter the market.
The non-structural approach claimed that the firms» profit is a poor measure of market
power. Vesalla (1995) argued that market power and profits are not necessarily positively
correlated (p. 19). The monopolists tend to be less efficient and it will affect the firms» profitability.
Hence inefficient behavior will reduce the profit rather than increase the firms» profit.
After that some researchers tried to use the Lerner Index to measure the market
performance. Shaffer evaluated the usage of the Learner Index as a measurement of market
performance. Shaffer (1994b) proved that the Learner index is a poor proxy of social welfare
because the linkage between the index and total welfare is not monotonic. The decreasing
number of firms reduces consumers surplus, but it increases producer surplus. Thus, mergers
may increase total surplus if the cost efficiency derived from mergers is higher than the reduction
of consumer surplus. Further, Shaffer (1994b) suggested measuring the cost structure as an
impact of consolidation policy, whether the cost structure becomes more efficient or not.
The non-structural approach argues that the link between concentration and competition
is not linear because it depends on various factors. The contestable market developed by Baumol
et al (1982) was referred by some researchers to explain competitive pricing in a concentrated
market. The main characteristic of contestable markets is free entry and exit. In such markets
an entering firm can attract customers by charging a low price and can recover any cost of
entry while abandoning the market if older firms retaliate by under-pricing in turn. As entry and
exit are free, the market might produce competitive pricing because less efficient firms can be
excluded from the market and replaced by the more efficient entrant.
Next, the non-structural approach focused on the information derived from competitive
behavior of firms. They believed that in learning about the market, we need to concentrate on
firms conduct instead of market concentration. In assessing competition, Panzar and Rose
(1987) developed a method based on the firm cost structure3. The method tries to distinguish
the competition level in markets by finding the relation of firms» revenue to factor price changes.
3 Panzer, J.C., Rose, J.N., (1987) developed a measurement to assess the competition in the market by observing the banksbehaviour. The H-statistic reflects the elasticity of changes on input price to banks revenue.
150 Bulletin of Monetary Economics and Banking, October 2011
They derived a generally applicable testable implication of monopoly profit maximizing behavior.
Their model is focusing on the sum of factor elasticity of the reduced form of the revenue
equation. The factor elasticity capture the effect of a proportional shift in average, total, and
marginal cost curve even when cost data themselves are unavailable.
They concluded that in monopoly markets, the value of elasticity is negative because an
increase in input prices will increase marginal costs, reduce equilibrium output and subsequently
reduce revenues; hence elasticity will be zero or negative (Bikker and Haaf, 2002). In perfect
competition, the value of elasticity is unity, because the average cost will shift proportionally
with the changes in factor price.
Many studies employed the Panzar-Rose (PR) method to measure the impact of
consolidation policies on competition in the banking industry. In the early 1980s, the deregulation
of deposit interest rates in some developed countries had heightened competition and led to a
wave of mergers and acquisitions. Shaffer (1982), Molyneux et al (1994), Bikker and Haaf
(2002) and Bandt and Davis (2000) examined the impact of the banking consolidations in the
United States and European countries on competition.
The consolidation in the US banking market was mostly driven by markets as «merger
was perceived as one way to improve their diversification, efficiency or possibly market power»
(Shaffer, 1994, p. 3). By using samples from banks in some states in the United States between
1979 and 1980, Shaffer (1982) proved that the banking market was competitive. The study
found that in most of the market, the significance test of the elasticity rejected the existence of
market power.
In addition to that, the European banks also faced structural changes after the
implementation of the European Single Market, included the financial services market. Under
the policy, there was limitless access to enter markets of the other member countries (Molyneux
et al, 1994). Molyneux et al (1994) tried to assess the impact of consolidation on competition
in the early stage of implementation of the Single Market policy. They focused on assessing the
banks behavior between the period 1986 and 1989. They had quite similar findings to Shaffer
(1982). The competitive environment in the banking industry was well sustained. Monopolistic
competition was established in countries such as Germany, United Kingdom, France and Spain.
However, the banking market in Italy was performed under monopoly or conjectural variation
short run oligopoly.
More recent data was employed by Bandt and Davis (2000). Quite similar findings appeared
with Molynuex et al (1994). Bandt and Davis (2000) aimed to measure the level of competition
across different groups of banks, which are large, medium and small during the period of 1992
151Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
to 1996. According to them, the large banks in German and France were working under
monopolistic competition, while the smaller banks were acting as a monopoly. Regarding Italy,
both small and large categories were working under monopoly market.
Extending the coverage of countries, Bikker and Haaf (2002) tried to investigate the
competitive behavior in 23 developed countries all over the world during 1989 to 1998. The
banking market in 23 countries was indicated to be working under monopolistic competition.
The findings supported De Band and Davis (2000) where competition among larger banks was
stronger than other groups.
Quite a fewer studies have been conducted in emerging countries. Gelos and Roldos
(2002), Claessens and Laeven (2004) and Yeyati and Micco (2007) assessed the competitive
behavior of banks in some emerging countries. It is important to conduct the study for emerging
markets as we might find quite different conditions compared to mature markets. The Latin
American countries also experienced consolidation with a substantial number of mergers and
acquisitions. The consolidation was firstly initiated by government through the Central Bank.
Further, it was also driven by market forces. Gelos and Roldos (2002) explained that the three
largest private banks in Brazil initiated the consolidation to sustain their competitiveness. In
addition, the five largest private banks in Argentina gained substantial market share from the
combination of organic growth and acquisitions.
Consolidation did increase concentration in emerging market of Latin American countries.
However, the concentrated market did not lead to a less competitive situation. This conclusion
was generated from three different studies of Gelos and Roldon (2002), Claessens and Laeven
(2004) and Yeyati and Micco (2007). The Contestable market principle might be helpful to
explain this situation. The removal of the restriction of entry to the market effectively preserved
the competitive environment (Classens and Laeven 2004).
On the other side, the consolidation which was mostly driven by government, as happened
in most Asian countries, showed different outcomes. Instead of increasing the concentration,
the wave of mergers and acquisitions in Asian countries reduced concentration (Gelos and
Roldos, 2002). In relation to competition, the less concentrated market in the Asian banking
market was perceived as competitive.
This illustration of findings under the non-structural approach clearly supports the notion
that the relationship among market concentration, firms conduct and market performance
may not linear. The consolidation itself, which was taken from broad varieties of policies, might
create concentrated or less concentrated markets. Further, competition in concentrated market
might be sustained by lessening the entry barrier into the market (Classens and Laeven 2004).
152 Bulletin of Monetary Economics and Banking, October 2011
(1)
(2)
2.3. Panzar and Rose Approach
As illustrated in the above section, we could not rely on market structure information to
determine the competition level in the banking market. It might be misleading to depend on
the information about market structure as some studies proved a non-linear relation between
market structure and competition. Moreover, Bikker and Haaf (2002) proved that concentration
indices such as the Concentration Ratio (CR) and the Herfindahl-Hirschman Index (HHI) appear
to be inversely correlated with the number of firm. The countries with smaller number of banks
tend to have a higher concentration level.
In order to measure competition in the market, this study will employ a method used
by Panzar and Rosse (1987). The method is based on properties of reduced form revenue
equations at the firm level, and the data of revenues and factor prices (Panzar and Rose,
1987). Therefore the method will directly assess the competitive behavior of banks to define
the market structure.
The Panzar-Rose method will calculate the sum of elasticity of the reduced form revenues
with respect to factor prices. This sum of elasticity is given the symbol H (Vesalla, 1995). The
value of elasticity will provide information about banks conduct, and furthermore it will
determine the structure of the market. The assumption underlying this method is that market
power of banks is measured by the extent to which changes in factor prices (unit costs) are
reflected in revenue earned (Vesalla, 1995). If the industry is competitive the elasticity will be
high, otherwise the elasticity will be low or even negative in the case of monopoly and
collusive oligopoly.
The properties of H allow us to distinguish empirically between common imperfect
competition theories of price formation as characterizations of the competitive behavior of
Indonesia»s banks, whether monopoly or perfect collusion, monopolistic competition or perfect
competition (Vesalla, 1995).
The Panzar-Rose empirical model assumes that banks have a log-linear marginal cost and
revenue function (Bikker and Haaf, 2002).
153Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
Where OUT is output, n is the number of banks, FIP denotes factor input prices, and EXirevenue
and EXicost
are other variables affect banks» revenue and cost functions, respectively. The empirical
application of Panzar and Rose approach assumes a log-linear marginal cost function, where
dropping subscripts referring to bank»i (Bikker and Haaf, 2002).
Further, the Rose Panzar model assumes a profit maximizing for individual banks. The
profit maximizing banks will produce at the level where marginal cost is equal to the marginal
revenue, yielding the equilibrium value for output:
In the empirical analysis, the following operationalization of the reduced-form revenue
equation is used (Bikker and Haaf, 2002, p.2196):
Where, TIR is the ratio of interest income to the total balance sheet; AFR is the price of funding
(funding rate); PPE is the price of personnel expenses (wage rate); PCE is price of capital
expenditure (capital price); OI is ratio of other income to the total balance sheet; and BSF is
bank-specific exogenous factors, such as the risk component, differences in the deposit mix
and size (banks» real assets) (Yeyati and Micco, 2007, p.1637).
Under this condition, H shows the sum of the elasticity of the reduced-form revenue
function with respect to factor prices.
The level of competition determines the value of H, whether monopoly or perfect collusion,
monopolistic competition or perfect competition. Below is a formula to calculate H, where H is
the sum of elasticity, which consists of the elasticity of revenue with regard to funding cost
changes ( β ), elasticity of revenue with regard to changes on human resource expenditure ( γ )
and elasticity of revenue with regard to capital price changes ( δ ).
(3)
(4)
(5)
154 Bulletin of Monetary Economics and Banking, October 2011
Panzar and Rose proved that under monopoly, an increase in input prices will increase
marginal costs, reduce equilibrium output and subsequently reduce revenues; hence H will be
zero or negative (Bikker and Haaf, 2002). In other words, markets where monopoly power
exists will yield a negative relation between those two variables as gross revenue responds in
the opposite direction to a change in a unit cost (Vesala, 1995). The same outcome is also
found in monopolistic competition without the threat of entry, i.e. with the fixed number of
banks. The market consists of some banks; however there are barrier to enter the market,
therefore the number of bank is unchanged. Vesalla (1995) proved that in such market H is
zero or negative, where similar with Panzar-Rose findings on monopoly market.
In analyzing monopolistic competition, the Panzar-Rose approach is based on the
comparative statics properties of the monopolistic competition Chamberlinian equilibrium model
(Bikker and Haaf, 2002). In the limit case of the monopolistic competition model, where banks»
products are regarded as perfect substitutes of one another; the Chamberlinian model produces
the perfectly competitive solution, as demand elasticity approaches infinity (Bikker & Haff: 2002:
pp. 2195). In a perfectly competitive market, increasing input prices will increase average cost
Figure 2.H in perfect competition
P1AC
P1
P0
y* y
AC1 (W1)
AC0 (W0)
(6)H = β + γ + δ
155Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
P MC
AC
B
MR
Pc
P
y y c y D
proportionally. Exit of some banks increases the demand faced by each of the remaining banks,
leading to increases in prices and revenues equivalent to the rise in costs (Bikker and Haaf,
2002). Finally, the value of H in perfectly competitive markets is equal to one.
In the case where monopolistic competition model recognizes product differentiation
the value of H is positive but less than one. As shown in figure 3, monopolistic competition is
an intermediate case between monopoly and perfect competition. Monopolistic competition
produces «excess capacity» where yi* is produced at average cost (point A) higher than minimum
average cost (point B). In addition, price pi* exceeds marginal cost (MC), compared to competitive
solution pc (Vesalla, 1995).
Figure 3.Monopolistic Free Entry (Chamberlinian) equilibrium
Banks in Chamberlinian equilibrium are not free from market power but they cannot
receive supernormal profits, since banks price at average cost (Vesalla, 1995). Therefore, the
value of H in this condition is positive but less than one. It means that the changes to input
prices positively affect revenue but are less than one.
This model is more consistent with the observation that banks tend to differentiate
themselves by various product quality and advertising, although the core services provided by
them are fairly homogenous (Vesalla, 1995, p. 50). The product differentiation is important in
156 Bulletin of Monetary Economics and Banking, October 2011
order to create less elastic demand (Tirole, 1988). As product is less differentiated, less market
power is exercised and the value of H will be higher. High switching cost is also the source of
market power. The account holders of the banks bear high switching cost to move from one
bank to others. If customers plan to switch the supplier then they must create a new account
number. It implies a lot of paperwork. In addition, the customer must inform about the changes
to their business relation (Canoy et.al, 2001).
At last, in an oligopoly market, the value of H may also be positive, where there are
strategic interactions among fixed number of banks (Bikker and Haaf, 2002). However, in the
case of perfect collusion in an oligopoly, the Rose and Panzar method produces a negative
value for H, similar to the monopoly model.
Table 4.Summary of the Discriminatory Power of H
Competitive environment
Monopoly equilibrium: each bank operates independently as under monopoly profit
maximization conditions (H is decreasing function of the perceived demand elasticity) or
perfect cartel.
Monopolistic competition with free entry equilibrium (H is an increasing function of the
perceived demand elasticity).
Perfect competition. Free entry equilibrium with full efficient capacity utilization.
Value of H
H < 0
0 < H < 1
H = 1
Source: Bikker and Haaf (2002, p. 2195)
There are five assumptions need to apply the Panzar-Rose method. First, banks are treated
as single product firms that act as financial intermediaries. Therefore banks produce interest
revenues by employing intermediate funds, labor and capital as inputs (De Bandt and Davis,
2000). Second, we have to assume that higher input prices are not associated with higher
quality services that generate higher revenues. Gelos and Roldos (2001) explained that if the
correlation exists, there might be bias in interpreting H. The third assumption is that the market
is in equilibrium in the long run. The fourth and fifth assumptions are considering banks as
profit maximization institutions and they have normally shaped revenue and cost functions
(Gelos and Roldos, 2002, p. 13-14).
157Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
III. METHODOLOGY
3.1. Empirical Model
In assessing the competition in the Indonesia banking market between 2001 and 2009,
we will run a regression model to estimate the following reduced form revenue equation. In
following equation, the four explanatory variables outside the bracket represent bank specific
factors.
Table five provides detail information about the variable specification. It explains the
definition and proxy used to measure each variable.
The study will employ a pooled regression, more specifically the Fixed Effect Model. There
are some reasons underlying the application of the pooled regression model. First, since the
application of a cross-section regression implicitly assumes that all banks have access to the
same factor markets therefore they only differ in terms of the scale of operation (De Bandt and
Davis, 2000, p. 1050) the cross-sectional regression is unable to capture the time-series dimension
inside the model. Vesalla (1995) concluded that it is difficult to infer whether the changes on
competition over time are statistically significant if we rely on the estimation of H using the
(7)
Table 5.Variable Spesification
Ratio of annual interest income to total assets
Ratio of annual interest expenses to total deposit
Ratio of annual wage and salary expenses to total assets
Ratio of other expenses to fixed assets
Ratio of others income (operational and non-operational incomes) to total assets
Equity divided by total assets
Loans divided by total assets
Ratio of interbank deposits to total deposit
Ratio of demand deposits from customers to total deposit and short term
funding
TIR (revenue)
AFR (funding rate)
PPE (wage rate)
PCE (capital rate)
OI (other income)
EQ (capital risk)
LO (loan risk)
BDEP (deposit mix)
DDC (deposit mix)
VariableVariableVariableVariableVariable Variable SpesificationVariable SpesificationVariable SpesificationVariable SpesificationVariable Spesification
158 Bulletin of Monetary Economics and Banking, October 2011
cross-section regression for each year. Some studies such as Vesalla (1995) and Yeyati and
Micco (2003) suggested consolidating the individual cross sections into a single pool data set.
Polled estimation will produce more reliable estimates of H as it examines the behavior of
banks over time.
The second consideration is that the study will also explore the relation between
concentration and competition across time. Therefore we emphasize the dynamic dimension,
which cannot be incorporated in the cross-section method (Yeyati and Micco, 2003a).
The third benefit of the implementation of pooled regression method relies on the ability
to capture the non-time-varying determinants of banks» revenues. The application of the Fixed
Effect Model allows the inclusion of bank fixed effects that can be used to control the
heterogeneity between banks that are not captured in the model. The Fixed Effect Model will
treat the heterogeneity of the non-time-varying determinants of revenues by entering the cross-
section dummies (for each bank). Therefore the model will introduce the different intercepts
capturing the banks-specific variables that are not explicitly addressed in the regression
specification (De Bandt and Davis, 2000).
In order to capture the impact of the consolidation policy on the banks» behavior, this
study tested the changes on the input prices coefficients. Referring to Gelos and Roldos (2002)
the test can be implemented by dividing the observation period into two sub-periods and
interacting the input price variables (ln(AFR), ln(PPE) and ln(PCE)) with a dummy variable that
takes the value of one in the second sub-period» (p. 15). The interaction variables will show
whether the consolidation policies significantly changed the banks» behavior; if the interaction
variables generate significant values, they indicate a structural break in the statistical relationship
between revenues and input prices (Gelos and Roldos, 2002, p. 15). In addition, the value of
interaction variables will determine the direction of changes on competition. If they are positive,
we can conclude that the consolidations increase competition or otherwise. Further if the H-
statistics is positive between 0 and 1 and the cumulative value of the interaction variables is
positive, it implies a stronger competition (Vesala, 1995, p. 56)4.
The first period is from 2001 to 2003 and the second period starts from 2004. This study
chooses 2004 as a year marking of structural break because starting from 2004; the Central
Bank of Indonesia formally enforced the consolidation process by introducing the Indonesia
Banking Architecture policy.
4 It refers to Bikker and Haaf (2002, p. 2203). They explained that the result of Vesala (1995) implies that the interpretation of H-statistics between 0 and 1 is a continuous measure of the level of competition. Further Bikker and Haaf (2002) described thehigher value of H can be used as an indication of the stronger level of competition (p. 2203).
159Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
One of the assumptions under the Panzar and Rose method is that the market is in
equilibrium (Claessens and Laeven, 2003). We need to check that the assumption is fulfilled by
running the test on market equilibrium. Referring to some studies, the market equilibrium test
must be able to validate that the Panzar-Rose statistics can deliver reliable results (De Bandt and
Davis, 2000).
In order to have an equilibrium test, some studies suggest testing whether the input
prices are related to the industry return. Here, we will modify the reduced form revenue equation
by replacing the dependent variable with the ratio of net income to total assets as an endogenous
variable (De Bandt and Davis, 2000).
The equilibrium is defined as a condition where the value of equilibrium E-statistics is
zero. The E-statistic is defined as the summation of β, γ, dan δ. Further, we can employ a F-test
to verify the statistical significance of the test whether the E = 0 (Claessens and Laeven, 2003).
3.2. Data
Data were obtained from the Central Bank of Indonesia. It consists of the unconsolidated
annual balance sheets and income statements of the commercial banks. The sample contains
128 banks for each year approximately; however the exact number in the sample for each year
varies. The yearly variation on the number of banks is caused by mergers, acquisitions, bank
liquidations and banks entry during the observation period. In the case of mergers and
acquisitions, the database only keeps the data of the new institution which is usually the bigger
bank.
Table 6 provides detail information about the number of bank for each category. According
to the database, the Central Bank divides commercial banks into six categories. Currently
Indonesia has 4 State Owned Banks, 68 local private banks which are comprised of banks
which engaged with foreign exchange services and those which are not. Even though the
Central Bank has grouped the 68 banks into local banks, actually some of them were bought
by overseas investors during the privatization policy in the late 1990s. There are 20 joint ventures
banks, where the owners are foreign and local investors. There are 11 foreign banks which are
usually branch offices of foreign banks. Finally, we have 25 regional banks that usually operate
(8)
160 Bulletin of Monetary Economics and Banking, October 2011
in their province. The shareholders of the regional banks are the government of the province
and the municipalities.
Compared to other developing countries in South East Asia, the number of banks in
Indonesia is large. However, the market is concentrated as a few large banks dominate it. The
banks» balance sheet shows that on average 3 from 113 banks control more than forty percent
of the industry between 2001 and 2009. Descriptive statistics information about the distribution
of»the assets and capital support the notion of the highly concentrated market. Table 7 illustrates
the distribution of the assets and equity from all banks in 2009. In terms of assets, a quarter of
market is dominated by single bank which is Bank Mandiri with assets 370 trillion rupiah and
Table 6.Sample Size
4
39
29
20
11
25
128
State Owned Banks
Private Banks - Foreign Exchange
Private Banks - Not engage in foreign exchange
Joint Venture Banks
Foreign Banks
Regional Banks
Total
Type of Banks (based on the Central Bank Categorization)Type of Banks (based on the Central Bank Categorization)Type of Banks (based on the Central Bank Categorization)Type of Banks (based on the Central Bank Categorization)Type of Banks (based on the Central Bank Categorization) Number of BanksNumber of BanksNumber of BanksNumber of BanksNumber of Banks
Source: The Indonesian Banking Statistics, 2009, the Central Bank of Indonesia
Table 7.The Descriptive Statistics of Asset and Equity All Banks in 2009 (in Million Rupiah)
StatisticsStatisticsStatisticsStatisticsStatistics AssetAssetAssetAssetAsset EquityEquityEquityEquityEquity
Mean
Median
Q4 (100% data)
Q3 (75% data)
Q2 (50% data)
Q1 (25% data)
Skewness
Kurtosis
Jarque-Bera
Probability of JB
22,158,195
3,978,396
370,310,994
13,356,445
3,923,234
1,523,057
4,44
23,26
2,242,80
0,00
2.267.144
547.938
35.108.769
1.370.931
539.862
164.954
4,16
21,00
1.803,03
0,00
161Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
equity more than 35 trillion rupiah. The asset of Bank Mandiri is 15 times larger than the
average and the equity is 17 times larger than the average.
Since the market is positively skewed, the median is a preferred measure of central
tendency. Table 7 shows that all statistics to assess the normality distribution of data are violated.
The values of skewness are positive both for assets and equity, implying the positively skewed
distribution. The JB values are positive and significantly different from one, indicating that the
data rejects the normality assumption. Referring to the data, half of the bank (56 banks) had
equity less than 550 billion rupiah. In addition, more than 66 percent (85 banks) had equity less
than 1,371 billion rupiah. On the other hand, a very small portion of banks - less than 1 percent
or only 7 banks - has equity more than 10 billion.
The above figures will be used to calculate break point to divide the sample based on the
size. The sample division into large, medium and small banks is important to capture the real
behavior of banks in competing with other banks in their categories. Large banks are assumed
to work at the national or even international level, while small banks focus on local or specific
industries and often do not have customers outside their defined market.
5 The categorization is based on the data of Indonesia banking in 2009.
Table 8.Three Categories of Banks based on the Value of Equity
CategoryCategoryCategoryCategoryCategory Equity (IDR)Equity (IDR)Equity (IDR)Equity (IDR)Equity (IDR) Sample Size Sample Size Sample Size Sample Size Sample Size 44444
> 10 trillion
1 - 10 trillion
< 1 trillion
7
36
70
113113113113113
Large Banks
Medium-sized Banks
Small Banks
TotalTotalTotalTotalTotal
The value of equity in 2009 will be employed as a basis to break the sample into large,
medium, and small banks. Large banks are those with equity more than 10 trillion rupiah. The
medium-sized banks are working with the capital from 1 to 10 trillion rupiah. Finally, the small
banks are those with equity less than 1 trillion rupiah. In terms of the number of banks, the
small bank category dominates the market; however the market is controlled by the larger
group.
162 Bulletin of Monetary Economics and Banking, October 2011
IV. RESULT AND ANALYSIS
4.1. Competition in Banking Industry
The finding is robust using the two different measurements of revenue with respect to
the level of competition in the Indonesia banking industry. Some studies employed the ratio of
interest income to total asset to measure the banks» revenue. Bikker and Haaf (2002) explained
that this proxy is more consistent with the assumption underlying the Panzar and Rose method,
where banks are assumed to be a single product firms that acts as financial intermediaries.
Therefore the main business of banks is financial intermediation.
On the other hand, De Bandt and Davis (2000) argued that as a response of tighter
competition, banks increase their interest to non-interest income activities √ including asset
management income, mutual funds and insurance (p. 1047). They proposed to employ both
measurements on the ratio of interest income to asset and the ratio of total revenue to asset as
a validation method to test if the model is robust.
Overall both specifications produce similar outcomes. However the interest income
specification provides some larger coefficients. We choose to employ the first proxy because
the Indonesian banking industry still relies heavily on interest-based-income. The data of the
annual financial report of all banks between 2001 and 2009 shows 88 percent of banks» income
was originated from interest income. Therefore the first proxy will capture the real portrait of
competition in the Indonesian banking industry. In addition, the study includes other income
which means income originated from non-interest activities as one of the explanatory variables
in order to capture the growing trend of non-interest based income in banks revenue.
One critical assumption under the Panzar-Rose method is whether the observations are
in long-run equilibrium. As described in the previous part, we have conducted an equilibrium
test by running a regression of the revenue reduced-form by replacing the dependent variable
by return on equity. We employ the method introduced by Claessens and Laeven (2003) in
deriving the Return on Asset (ROA). As ROA can take on a small (negative) value, they suggested
computing the dependent variable ROA» by using the formula of Ln (1+ROA) (p. 11). The
estimation showed that the joint coefficient of β, γ and δ is not significantly different from
zero (please refer to the appendix for the complete estimation result). Therefore, the Panzar-
Rose method is suitable to assess competition in the Indonesia banking market between 2001
and 2009.
The Panzar-Rose model has been applied to around 128 banks between 2001 and 2009.
The redundant test was employed to examine whether the introduction of an intercept for
different cross section units in order to account the heterogeneity across banks are significant.
163Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
The restricted F-test and x2 test of fixed effect are significant. They confirm that fixed effect
model is more reliable capturing the information about what contributes to the difference
across banks. Thus the model can capture the information of what contributes to the differences
across banks. Please refer to appendix C for detail statistical results on redundant fixed effects
tests.
Table 9 provides complete information about the estimation results of the reduced-form
revenue equation. The H-statistics for all banks is .69. The F test shows that the value is
significantly different from zero thus it rejects the monopoly hypotheses. It is also significantly
different from one, thus it also rejects the perfect competition hypotheses. We can conclude
Table 9.Empirical Results using Fixed Effect Model Interest Income as Dependent Variable,
All Banks between 2001 and 2009
Funding rate
Wage rate
Capital price
Risk of loans
Risk of capital
Deposit Mix √
Demand Deposit
Deposit Mix √
Interbank Deposit
Other Income
Number of observation
R2
H (competition) MarketH (competition) MarketH (competition) MarketH (competition) MarketH (competition) Market
structurestructurestructurestructurestructure
All Banks Large Banks Medium Banks Small Banks
a) the value of F test shows that H is not significantly different from 0 and 1 (level confidence 99%)The value in the bracket is the t-statistic;*** Significant at 99% confidence level; ** significant at 95% confidence level; * significant at 90% confidence level
0.46***
(34.74)
0.21***
(11.52)
0.02
(1.50)
0.45***
(14.04)
0.002
(0.19)
0.03***
(2.06)
-0.001
(-0.60)
-0.008
(-0.83)
892
0.89
0.690.690.690.690.69aaaaa
MonopolisticMonopolisticMonopolisticMonopolisticMonopolistic
CompetitionCompetitionCompetitionCompetitionCompetition
0.49***
(18.56)
0.08***
(2.53)
0.01
(0.40)
0.17
(1.20)
0.18***
(2.68)
0.025
(0.81)
-0.021***
(-2.17)
0.04
(1.54)
63
0.94
0.590.590.590.590.59aaaaa
MonopolisticMonopolisticMonopolisticMonopolisticMonopolistic
CompetitionCompetitionCompetitionCompetitionCompetition
0.51***
(28.52)
0.26***
(10.10)
0.03***
(2.12)
0.52***
(11.51)
0.10***
(4.02)
0.007
(0.33)
-0.003
(-0.97)
-0.022
(-1.64)
309
0.94
0.800.800.800.800.80aaaaa
MonopolisticMonopolisticMonopolisticMonopolisticMonopolistic
CompetitionCompetitionCompetitionCompetitionCompetition
0.46***
(24.50)
0.19***
(7.50)
0.018
(0.88)
0.457***
(9.41)
-0.028
(-1.83)
0.02
(1.29)
0.001
(0.40)
-0.010
(-0.77)
520
0.87
0.660.660.660.660.66aaaaa
MonopolisticMonopolisticMonopolisticMonopolisticMonopolistic
CompetitionCompetitionCompetitionCompetitionCompetition
164 Bulletin of Monetary Economics and Banking, October 2011
that the banking sector is working under monopolistic competition. The changes in factor price
will increase the revenue less than proportionally because the products are differentiated and a
switching cost is high. Further, competition has risen over time. This conclusion was based on
a comparison of our findings with the study conducted by Classens and Laeven (2003). According
to them, from 1994 to 2001, the H-statistic for Indonesia banking was .62.
The across groups analysis provides broader understanding of the market. The H-statistics
vary from .59 to .80 suggesting that the banking market across all groups was working under
monopolistic competition. The large banks groups was the least competitive market, with the
H-statistics .59. The medium banking group was the most competitive market. The H-statistic
for the medium bank group is quite high (.80), and it is comparable with the banking market in
developed countries and other emerging market such as Latin American bank (Yeyati and
Micco, 2007). It is quite surprising as in other countries including Europe; the banking market
of larger banks is more competitive than that of the smaller banks which usually service the
local market (Bikker and Haaf, 2002).
As expected, deposits or funds are the most important input. The coefficient for the
funding rate is the main contributor of the elasticity. This figure is consistent across different
groups. Human resources are the second main input in the banking industry. The sign of the
coefficients are also positive indicating that the increase in the wage rate is transmitted into the
revenue. The least important input is capital. The coefficients on the capital price are small, and
in some estimates, they are not significant.
Risk of loans is an important explanation of the pattern of revenue. Positive coefficients
for risk of loans imply that banks with a higher proportion of loans on their balance sheets
generate higher interest revenue per rupiah of asset. A similar conclusion is also applied to the
risk of capital variable. The coefficients are positive for large and medium banks. Thus in the
larger banks, it can be concluded that the higher portion of equity per asset earns higher
revenue. The deposit mix variables are not really significant in explaining the movement of
revenue, except the proportion of interbank deposits to total deposits in the large banks market.
The sign of the coefficient is negative which implies that the bigger the portion of interbank
deposit to total deposits, the less revenue that might be earned.
4.2. Market Structure and Competition
This section will describe the relationship between concentration and competition. The
study will employ two-frequently applied types of such indices as a proxy of market concentration.
The first index is called k-banks concentration ratio (CRk) which takes the market shares of the
165Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
k biggest banks in the market, and ignoring the remaining banks. The second index is Herfindahl-
Hirschman index (HHI), which takes market shares as weights, and stresses the importance of
larger banks by assigning them a greater weight than smaller banks (Bikker and Haaf, 2002)6.
These concentration measurements can be considered reliable for assessing market
concentration. As all proxies produces similar outcomes. The data shows that the banking
market in Indonesia has become less concentrated. This finding is surprising. Unlike the developed
countries where consolidation increased the market concentration7, policy changes reduce the
concentration in the Indonesia banking market.
Our findings are aligned with a study conducted by Gelos and Roldos (2002). They argued
that consolidation did not increase market concentration in Asian countries because the structural
changes were mostly driven by government. While in the developed countries and more mature
markets of developing countries in Latin America, mergers and acquisitions were driven by the
market.
In those markets, tight competition forced large banks to merge to enhance their
competitiveness. On the contrary, mergers and acquisitions in Indonesia banking were mostly
6 The following are formulas to calculate the concentration ratio (CR)and Herfindahl-Hirschmanindexes, where s is market share ofbanks.
7 Please refer to De Bandt and Davis (2000) and Bikker and Haaf (2002).
Figure 4.The Banking Concentration by Using Four Different Measurements
All Banks, 2001 - 2009
0.7
0.6
0.5
0.4
0.3
0.2
0.1
02001 2002 2003 2004 2005 2006 2007 2008 2009
CR3CR4CR5HHI
CRk = Σk sii = 1 HHI = Σm s
i2
i = 1
166 Bulletin of Monetary Economics and Banking, October 2011
conducted by medium-sized and smaller banks in order to comply with the Single Presence
Policy and Minimum Capital Requirement. These consolidations increased the scale of economies
of the merging banks. However it did not increase industry concentration because the mergers
were dominated by small and medium-sized banks. It implies that consolidation in the Indonesian
banking market does not create more concentrated markets but it improves the market share
distribution.
In the case of large banks, the Single Presence Policy has effectively reduced concentration
by the consolidation of PT Bank Niaga Tbk. and Bank Lippo in 2008. The market share of the
merged bank, PT. CIMB Niaga, increased about 3 percent. This merger significantly reduced the
market concentration because the new merge bank was not in the top three large banks.
Therefore as the share of merge bank increased, the market share distribution was changed
where the share of biggest three banks decreased from 74 percent in 2001 to 66 percent in
2009.
The mergers and acquisitions in the medium-sized and small banks market aimed to
improve the banks» performance after the 1997 economic crisis or to comply with the Single
Presence policy or Minimum Capital Requirement policy. There were seven mergers in the
medium-size banks category between 2001 and 2009. The number of banks in the small
group also reduced. The consolidation in small banks market was dominated by the
liquidations of banks with poor performance and consolidation to comply with the Minimum
Capital Requirement policy. Please refer to table 3 for a detail list of merger between 2001
and 2009.
Table 10.The Banking Concentration across Groups 2001 - 2009
All Banks Large Banks Medium-sized Banks Small Banks
20012001200120012001 0.47 0.10 0.74 0.24 0.30 0.06 0.35 0.05
20022002200220022002 0.46 0.10 0.74 0.23 0.30 0.06 0.30 0.04
20032003200320032003 0.45 0.09 0.73 0.22 0.27 0.05 0.30 0.04
20042004200420042004 0.42 0.08 0.71 0.21 0.26 0.05 0.29 0.05
20052005200520052005 0.38 0.07 0.67 0.20 0.25 0.05 0.33 0.05
20062006200620062006 0.36 0.06 0.65 0.17 0.23 0.05 0.29 0.05
20072007200720072007 0.37 0.07 0.64 0.18 0.23 0.04 0.30 0.05
20082008200820082008 0.37 0.06 0.64 0.18 0.23 0.04 0.32 0.06
20092009200920092009 0.39 0.07 0.66 0.18 0.24 0.05 0.28 0.05
Average 0.0.0.0.0.4141414141 0.0.0.0.0.0808080808 0.0.0.0.0.6969696969 0.0.0.0.0.2020202020 0.0.0.0.0.2626262626 0.0.0.0.0.0505050505 0.0.0.0.0.3131313131 0.0.0.0.0.0505050505
CR3CR3CR3CR3CR3 HHIHHIHHIHHIHHI CR3CR3CR3CR3CR3 HHIHHIHHIHHIHHI CR3CR3CR3CR3CR3 HHIHHIHHIHHIHHI CR3CR3CR3CR3CR3 HHIHHIHHIHHIHHIY e a r
167Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
Compared to large banks, the medium-sized and small banks have a much lower degree
of concentration. They have a better market share distribution, as the biggest three banks in
the group only controlled less than 35 percent of the market share. Analogous with the large
banks, the concentration level in medium-sized banks and small banks has also reduced. But if
we compare the value of the concentration changes, the changes of the concentration level in
medium-sized banks and small banks were relatively smaller, especially if we refer to Herfindahl-
Hirschman index. The information of the market concentration and competition in the medium-
sized banks and small banks helps in understanding this phenomenon. The medium-sized and
small banks markets were highly competitive and less concentrated, therefore the mergers and
acquisitions just slightly reduced the market concentration.
There is a consistency between market concentration and competition in the three different
groups. Larger banks have the lowest degree of competition because the market is more
concentrated (CR3average
=0.69; HHI average
=0.20). On the other hand, the market of medium-sized
banks is the most competitive because it is less concentrated (CR3average
=0.26; HHI average
=0.05).
Small banks also work in a quite competitive environment because each bank has a small share
of the total market (CR3average
=0.31; HHI average
=0.05).
Table 11. Empirical Results using Fixed Effect Method with Interaction Variables(All Banks between 2001 and 2009)
Rincian
H (Competition Level)
∆H (After 2004 Consolidation Policy)
Significance test on ∆H , F test¡
Test of change in H
Market Structure
N (Number of Observation)
0.68
0.027
3.5074240.0615*
Can not reject the increasing of H-statistics
Monopolistic Competition
892
Note: *** significant at 99% confidence level; ** significant at 95% confidence level; * Significant at 90% confidence level
The outcome of fixed effect model using the interaction variables for all banks implies
that the second wave of consolidation through the introduction of Indonesia Banking Architecture
increased the value of H-statistics. It increases about 0.027 from 0.68 in the first period to 0.70
in the second period. The significance test by using F test shows that the increase on the level
of competition is significant at 90 percent of confidence level. The result confirms that the
consolidation policies enhanced the competition in the Indonesia banking industry. Consolidation
168 Bulletin of Monetary Economics and Banking, October 2011
policies were effective to encourage medium-sized and small banks to merge in order to comply
with the Minimum Capital Requirement or Single Presence Policies. It significantly increased
the scale of economies of the merging banks and equipped them with bigger capacity to
compete with other banks. Further the mergers of smaller banks improved the distribution of
the market share. The improvement of the market share distribution reduced the market
concentration and enhanced the competition.
V. CONCLUSION
There are some interesting findings of this study. First, the data implies that the structure
of Indonesia banking market is vulnerable. Comparing to other East Asian countries, the number
of bank in Indonesia is larger. However, the market is concentrated in a few banks. Large banks
control a substantial share of the market. On the other hand, more than a half of banks are
small with the equity less than 1 trillion rupiah. In addition, the market concentration in the
large banks group is much higher than in smaller banks.
Secondly, banking market has become less concentrated during the implementation of
consolidation policy. This finding showed that the consolidation policy in Indonesia has similar
pattern with those implemented in other Asian countries. On the contrary, consolidation policy
in Latin American countries and other developed countries increased the market concentration.
As described by Gelos and Roldos (2002), the consolidation policy that was driven by market
will increase the concentration. On the other hand, the consolidation policy in Indonesia was
driven by the authority through the banks liquidation, mergers of the State Owned Bank and
the implementation of the Indonesia Banking Architecture. They have effectively encouraged
medium-sized and small banks to consolidate. This improved the distribution of market share
and reduced the market concentration.
Thirdly, during the implementation of consolidation policy the banking industry was
working under monopolistic competition. The H-statistic was .69. If we compare to the value
of H-statistic within 1994-2001 as calculated by Claessen and Laven (2003), the competition
has risen over time. The competition analysis by sub-groups shows that large banks were working
in the least competitive market while the medium-sized banks are working under the most
competitive market. The nature of competition in Indonesia banking was different with developed
countries where large banks were more competitive (Bikker and Haaf, 2002).
Next finding is discussing the impact of consolidation policy on competition. The estimation
shows that banking industry was more competitive in the second period of the implementation
169Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
of the consolidation policy. The increasing of the economies scale of the merging banks and
the improvement of market share distribution enhanced the competition.
Finally the study demonstrates that a concentrated market contributes to a less competitive
environment. It is possibly the reason why the large banks in Indonesia are working in a less
competitive market rather than the smaller banks. Large banks possess monopoly power to
enable them to behave as monopolist or oligopolies.
170 Bulletin of Monetary Economics and Banking, October 2011
REFERENCES
Bain, Joe E. (1951). Relation of Profit Rate to Industri Competition: American Manufacturing,
1936 √ 1940. The Quarterly Journal of Economics 65.hal. 293-324
Bain, Joe E. (1956). Barriers to New Competition. Cambridge. Mass: Harvard University Press.
Berger, Allen N., and Timothy H. Hannan. (1989). The Price-Concentration Relationship in
Banking. Review of Economics and Statistics 71.Hal.291-99.
Bikker, Jacob A., and Katharina Haaf. (2002). Competition, concentration, and their relationship:
An empirical analysis of the banking industri. Journal of Banking and Finance 26.hal. 2191-
2214.
Boumol, William, John C. Panzar, and Robert D. Willig.(1982). Contestable Markets and the
Theory of Industri Structure. San Diego: Harcourt Brace Jovanovich.
Buletin Bisnis. (2007). Daftar Akuisis dan Merger Bank di Indonesia. Retrieved from website:
http://buletinbisnis.wordpress.com/2007/12/11/daftar-akuisisi-dan-merger-bank-di-
indonesia/.
Calem, Paul, and Gerald Carlino. (1991). The Concentration/ Conduct Relationship in Bank
Deposit Market. Review of Economics and Statistics 73.hal. 268-76.
Canoy, Marcel, Machiel van Dijk, and Jan Lemmen. (2001).
Competition and Stability in Banking.CPB Netherland Bureau for Economic Analysis.Netherland.
Bank Sentral Indonesia.(2010). Arsitektur Perbankan Indonesia.dikutip dari website:
www.bi.go.id.
Aturan Bank Sentral IndonesiaRepublik IndonesiaNomor 10/15/PBI/2008 TentangKebutuhan
Modal Minimum
Aturan Bank Sentral Indonesia Republik Indonesia Nomor 8/16/PBI/2006 TentangKebijakan
Single Presence.
Central Bank of Indonesia.(2009). Indonesia Banking Statistics 2009.
Chan, Donald, Christopher Schumacher, and David Tripe.(2007). Bank Competition in New
Zealand and Australia.Finsia_MCFS.The Melbourne Center for Financial Studies. Retrieved
from website: http://www.melbournecentre.com.au/
Chua, H.B. (2003). FDI in financial sector: the experience of ASEAN countries over the last
decade, in CGFS (2004), the Central Bank paper submitted by Working Group members.
Retrieved from website: www.bis.org/publ/cgfs22mas.pdf
171Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
Claessens, Stijn, Asli Demirhuc-Kunt, and Harry Huizinga.(1998). How Does Foreign Entry
Affect the Domestic Banking Market.Policy Research Working Paper.The World Bank.
Claessens, Stijn, and Luc, Laeven. (2004). What Drives Bank Competition? Some International
Evidence.World Bank Policy Research Paper.No.3113, August.
De Bandt, O., and Davis, E.P., (2000). Competition, contestability and market structure in
European banking sectors on the eve of EMU. Journal of Banking and Finance 24, 1045√
1066.
Demsetz, Harold. 1973. Industry Structure, Market Rivalry, and Public Policy. Journal of Law
and Economics 16 (April).hal. 1-10.
Gelos, Gaston., R and Jorge Roldos. (2002). Consolidation and Market Structure in Emerging
Market Banking Systems.IMF Working Paper WP/02/186.
Legislation number 5/ 1999 Regarding the Indonesian Competition Regulation
Legislation of Republic of Indonesia Number 7/1998 Regarding Banking Sector.
McLeod, Ross. (1999). Control and competition: Banking deregulation and re-regulation in
Indonesia. Journal of the Asia Pacific Economy, No 4(2), p. 258-297.
Molyneux, Phillip, John Thornton, and D. Michael Llyod-Williams.(1991). Competition and
Market Contestability in Japanese Commercial Banking.Mimeo. Washington, D,C.
International Monetary Fund.
Molyneux, Phillip, John Thornton, and D. Michael Llyod-Williams.(1994). Competition condition
in European Banking.Journal of Banking and Finance 18.hal.445-459.
Nathan, Ali, and Edwin H. Neave. (1989). Competition and Contestability in Canada»s Financial
System: Empirical Results. Canadian Journal of Economics 22 (June).Hal.576-94.
Nueberger, Doris. (1997). Structure, Conduct and Performance in Banking Markets.University
Rostock, Working Paper No. 12.
Panzer, John C., and James N. Ross. (1987). Testing for ≈Monopoly Equilibrium∆. The Journal
of Industrial Economics.hal.443-456.
Pelzman, Samuel. (1977). The Gains and Losses from Industrial Concentration. Journal of Law
and Economics 20.hal. 229-63.
Shaffer, Sherrill. (1982). A Nonstructural Test for Competition in Financial Markets. Proceedings
of a Conference on Bank Structure and Competition (Federal Reserve Banks of Chicago)
on May. hal.225-43.
Shaffer, Sherrill. (1994a). Bank Competition in Concentrated Market.Business Review.March/
April.
Shaffer, Sherrill. (1994b). Structure, Conduct, Performance, and Welfare.Review of Industrial
Organization 9.hal. 435-450.
172 Bulletin of Monetary Economics and Banking, October 2011
Tirole, J. (1987). The Theory of Industrial Organization.MIT Press, Cambridge, MA.
Vesala. (1995). Testing Competition in Banking: Behavioral Evidence from Finland. Bank of
Finland Studies Working Paper No. E1.
Yeyati, Eduardo, Levy and Micco, Alejandro. (2003a). Concentration and Foreign Penetration
in Latin American Banking Sectors: Impact on Competition and Risk. Inter-American
Development Bank.Retrieved from website: http://www.oecd.org/dataoecd.
Yeyati, Eduardo Levy and Micco Alejandro.(2003b). Banking Competition in Latin America.
Latin-American Competition Forum. Retrived from website: http://www.oecd.org/dataoecd.
Yeyati, Eduardo, Levy and Micco, Alejandro. (2007). Concentration and Foreign Penetration in
Latin American Banking Sectors: Impact on Competition and Risk. Journal of Banking and
Finance 31.hal. 1633-1647.
173Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
Appendix A.Alternative Specification, Empirical Results Polled Regression Model of Fixed Effect Method,
Total Revenue as Dependent Variable (All Banks between 2001 and 2009)
Funding rate
Wage rate
Capital price
Risk of loans
Risk of capital
Deposit Mix -
Demand Deposit
Deposit Mix -
Interbank Deposit
Other Income
Number of Observation
R2
H (Competition)H (Competition)H (Competition)H (Competition)H (Competition)
All Banks Large Banks Medium Banks Small Banks
a) the value of F test shows that H is not significantly different from 0 and 1 (level confidence 99%) The value in the bracket is the t-statistic.*** significant at 99% confidence level;** significant at 95% confidence level;* significant at 90% confidence level
0.43***
(35.05)
0.17***
(10.30)
0.067***
(5.32)
0.39***
(13.34)
0.037***
(3.10)
0.03***
(2.61)
-0.005***
(-2.52)
0.12***
(13.19)
892
0.87
0.660.660.660.660.66aaaaa
0.43***
(19.02)
0.065***
(2.13)
0.01
(0.58)
0.08
(0.66)
0.16***
(2.69)
0.04
(1.22)
-0.022***
(-2.66)
0.16***
(7.85)
63
0.94
0.510.510.510.510.51aaaaa
0.40***
(23.65)
0.21***
(8.91)
0.04***
(2.60)
0.41***
(10.50)
0.09***
(4.00)
0.001
(0.08)
-0.008***
(-2.55)
0.15***
(11.90)
309
0.93
0.650.650.650.650.65aaaaa
0.46***
(29.85)
0.14***
(6.93)
0.05***
(3.01)
0.41***
(10.21)
0.02
(1.52)
0.02
(1.63)
-0.003
(-1.26)
0.08***
(7.48)
520
0.86
0.660.660.660.660.66aaaaa
174 Bulletin of Monetary Economics and Banking, October 2011
Appendix B.The Equilibrium Test √ Return on Asset
as dependent variable
All Banks
H = 0 cannot be rejected (level confidence 99.9%)The value in the bracket is the t-statistic.
-0.067
(-1.55)
0.16
(2.63)
-0.074
(-1.52)
-0.70
(-6.16)
0.004
(0.095)
0.093
(2.06)
-0.028
(-3.26)
0.19
(5.61)
862
0.75
0.0240.0240.0240.0240.024aaaaa
0.0720.0720.0720.0720.072
0.7890.7890.7890.7890.789
Funding rate
Wage rate
Capital price
Risk of loans
Risk of capital
Deposit Mix √ Demand Deposit
Deposit Mix √ Interbank Deposit
Other Income
Number of observation
R2
H (competition)H (competition)H (competition)H (competition)H (competition)
Equilibrium testEquilibrium testEquilibrium testEquilibrium testEquilibrium test
F testF testF testF testF test
ρρρρρ (1,742) (1,742) (1,742) (1,742) (1,742)
175Competitive Conditions in Banking Industry: An empirical Analysis of the Consolidation, Competition andConcentration in the Indonesia Banking Industry between 2001 and 2009
Appendix D.Empirical Results of Polled Regression Model of Fixed Effect with Time Dummies
All Banks between 2001 and 2009
All Banks Large Banks Medium Banks Small Banks
a) the value of F test shows that H is not significantly different from 0 and 1 (level confidence 99%)The value in the bracket is the t-statistic.*** significant at 99% confidence level;** significant at 95% confidence level;* significant at 90% confidence level
0.43***
(24.73)
0.23***
(12.62)
0.028***
(2.1)
0.42***
(12.70)
0.016
(1.25)
0.0009
(0.069)
0.00045
(0.178)
-0.0004
(0.177)
892
0.90
0.690.690.690.690.69aaaaa
0.53***
(9.62)
0.033
(0.61)
-0.0025
(-0.063)
0.0008
(0.0061)
0.21***
(2.30)
0.03
(0.77)
-0.028***
(-2.75)
0.064
(1.66)
63
0.95
0.550.550.550.550.55aaaaa
0.41***
(17.87)
0.28***
(9.96)
0.06***
(3.34)
0.40***
(9.51)
0.19***
(6.73)
-0.046
(-1.94)
0.0005
(0.124)
-0.042***
(-2.82)
309
0.94
0.760.760.760.760.76aaaaa
0.48***
(17.61)
0.18***
(7.18)
0.015***
(0.76)
0.46***
(9.15)
-0.024
(-1.56)
0.011
(0.62)
0.002
(0.70)
-0.015
(0.70)
520
0.87
0.680.680.680.680.68aaaaa
Appendix C.Redundant Fixed Effects Tests
F-test
df (degree of freedom)
x2 statistics
df (degree of freedom)
Number of Observation
All Banks Large Banks Medium Banks Small Banks
H = 0 is (level confidence 99.9%)
13.73
(112;771)
978.28
(112)
892
9.42
(6;48)
49.04
(6)
63
20.27
(35;265)
NA
(NA)
309
9.42
(69;442)
470.23
(69)
520
Funding rate
Wage rate
Capital price
Risk of loans
Risk of capital
Deposit Mix √
Demand Deposit
Deposit Mix √
Interbank Deposit
Other Income
Number of observation
R2
H (competition)H (competition)H (competition)H (competition)H (competition)
176 Bulletin of Monetary Economics and Banking, October 2011
This page is intentionally left blank
177Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
RELATIVE EFFECTIVENESS OF INDONESIAN POLICY CHOICESDURING FINANCIAL CRISIS1
Tumpak Silalahi2 and Tevy Chawwa3
The objective of this paper is to review the impact of crisis and policy measures taken during the
crisis, to evaluate the effectiveness of those measures and to analyze the exit strategy in Indonesia. The
econometric model was used to evaluate the impact of monetary and fiscal policy to economic output
using quarterly data from 1990 - 2010. The result shows that monetary and fiscal policies have significant
impact to economic output. In the short run the changes in real GDP is significantly affected by changes
in real monetary supply in the previous three quarter and real fiscal expenditures. The lesson learned from
this research among other are that cooperation and coordination among the policy makers and the timely
responses are very important in tackling the crisis; an effective conventional monetary policy in normal
times may become less effective in a crisis thus unconventional monetary policy indeed necessary as
timely policy response and the improvement for more timely disbursement of government expenditure is
important to increase the effectiveness of this policy to stimulate economic output. Moreover, several
Indonesian exit strategy and policies to face future challenges are very important to reach the ultimate
objective of sustainable economic growth while maintaining macroeconomic stability.
1 This is a collaborative research project with several central banks coordinated by SEACEN Research and Training Centre.2 Associate Senior Economist at Directorate of Economic Research and Monetary Policy - Bank Indonesia, email : [email protected] Junior Economist at Directorate of Economic Research and Monetary Policy - Bank Indonesia, email : [email protected] . The views
expressed in this paper are those of the authors and do not reflect the views of Bank Indonesia
Abstract
JEL Classification: E52, E62, E63
Keywords: monetary policy, fiscal policy, financial sector policy, global financial crisis.
178 Bulletin of Monetary Economics and Banking, October 2011
I. INTRODUCTION
Prior to the global crisis, particularly during 2007, optimistic expectation about Indonesian
economy was a general mindset of the economic forecasters. This expectation was supported
by various macroeconomic indicators which showed a remarkable achievement of the Indonesian
economy in 2007 after the Asian crisis in 1997. Indonesian GDP growth trend was increasing
continuously since 2005 and for the first time after crisis, the GDP growth reaches more than
6% in 2007. Basically, the growth in 2007 was driven by robust domestic consumption and
external demand which led to a surplus in the current account. In addition, positive sentiment
accompanied by attractive yield on rupiah portfolio investment helped encouraged capital
inflows. Shifting funds into emerging market assets contributed to a positive appreciation of
currency in the region. The surplus in current account and the rise of portfolio capital inflows in
2007 had increased Indonesian foreign currency reserves to 13% of GDP or sufficient to cover
import of goods and services for an average of seven months.
Indonesia financial market and institutions were also in a stronger condition based on
financial indicators. Lessons learned from the 1997 crisis resulted in a fairly strict implementation
of prudential regulations in the corporate and banking sectors as a result of which led Indonesian
banking industry became much sounder with a more robust foundation to absorb various
shocks in the economy. Demand pressure in 2007 was relatively high indicated by positive
output gap, although it was still far below output gap in 1996. Moreover, the government had
tried to reduce its dependency on foreign debts, both short-term and long-term. All of these
improvements resulted in Indonesia been assessed as a low risk country and it achieved the
highest ICRG (International Country Risk Guides) scores since 1997.
In 2007, fiscal policy was targeted at maintaining price stability for energy and staple
needs, while also delivering an economic stimulus. Escalating world oil prices in combination
with below-target lifting of domestic oil led to a considerable pressure in the government
budget deficit. In the monetary policy, BI»s stance could be divided into 2 periods, the period of
decline in the BI Rate (January-July 2007) and the period of flat movement in the policy rate
(August-November 2007). Bank Indonesia also applied a flexible exchange rate policy, allowing
the rupiah to move in line with economic fundamentals. To manage the volatility in the rupiah,
Bank Indonesia conducted foreign exchange market interventions on a limited scale. Although
many improvements such as better monetary and fiscal coordination were introduced to
strengthen the effectiveness of policy choices in Indonesia, the disturbances from external
shocks such as the rise in commodity/oil price and internal shocks such as crop failures or
seasonal events were important factors that influenced the Indonesian macroeconomic
conditions.
179Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
The global financial crisis in 2008 had reversed the previous optimistic mindset of economic
forecasters. The pressures in the global liquidity had caused a massive short term portfolio
capital outflow followed by a decline of Indonesia»s financial market performances. In the real
sectors, reflecting the input of global slowdown, exports declined and it had an indirect impact
on household and private sector»s income, leading to a decline in Indonesia»s consumption and
investment. As a result, Indonesia»s GDP growth»in 2009 declined to 4.5% (yoy).
Some policy measures were implemented in monetary, fiscal and financial sectors to deal
with the global financial crisis. Bank Indonesia had implemented an accommodative monetary
policy in order to keep a moderate growth achieve at least by maintaining financial markets
liquid which was facilitated by relatively low inflation. The policy rate was brought down in
December 2008 with the intention to decrease banks» lending rates. Some unconventional
monetary policy measures such as narrowing the interest rate corridor for standing deposit and
lending facility had also been taken to address liquidity issues. On the fiscal side, the government
provides policy response to keep domestic demand by several fiscal stimulus and trade policies.
There were also coordination between Ministry of Finance, Central bank and other institutions
in order to maintain financial and macroeconomic stability.
With this background, this paper is aimed at reviewing the policy measures taken during
the crisis, evaluating their effectiveness and analyzing the exit strategy to reach the ultimate
objective of sustainable economic growth while maintaining macroeconomic stability in
Indonesia. In turn this is expected to make a contribution to a comprehensive evaluation of the
effectiveness of the policy measures in SEACEN Economies. To handle the broad issues in the
paper, two methodologies are adopted: firstly, by descriptive analysis using simple statistics and
graphics; and secondly, by econometric model to analyze the relative effectiveness of policy
choices.Second session of this paper discusses the theory and the empirics of the crisis, session
three discuss methodology while session four present result and analysis. Session five will give
conclusion and close the presentation.
II. THEORY
As a small open economy, Indonesia could not be immune from external shock impact.
The integration in financial sector has left many countries particularly for open economy to
contagion risk. An empirical study by Santoso et al (2009) shows that Indonesia has a contagion
relationship with several countries in Asia, such as Japan, Taiwan, Korea, Hong Kong and India.
The domestic financial market moves closely with the movements in global financial markets.
The research also showed that there is no direct contagion between Indonesia stock exchange
180 Bulletin of Monetary Economics and Banking, October 2011
and the Dow Jones index and NASDAQ index. Thus, if Indonesia is affected by the global crisis,
it is not a direct effect from the US market but rather indirect effects from capital markets in
Asia that share a direct relationship with the US capital market. Furthermore, the research
indicated that Indonesia is a shock absorber rather than shock transmitter, particularly with
regard to developed countries (Japan, Australia, Germany, United Kingdom and the US).
In the real sector, Indonesian export is also affected by external condition. Research by
Kurniati»s et al (2008) study show that Indonesia»s exports are most sensitive to the economic
growth of Singapore (1.19), followed by the US (0.84), Japan (0.81) and China (0.3).
Related to the current global financial crisis, Kurniati and Permata (2009) found that a
shock in global risk aversion have immediate negative impact on capital inflows to Indonesia,
particularly from portfolio investments that leading to rupiah depreciation. The impacts through
financial channel on financial variables are temporary and relatively faster to recover (self-market
correction). The second round effects of global crisis occur through trade channel. Negative
shock of US GDP growth leads to contemporaneous decline in Indonesia»s exports which
subsequently result in decreasing domestic real GDP growth, capital outflow and rupiah
depreciation. The impacts on exports seem to persist and need policy responses from the
authorities.
The effectiveness of monetary policy depends on the environment of domestic economy
and the disturbance from external shock. Study done by Arifin (1998) which analyze effectiveness
of interest rate policy for rupiah stabilization during crisis 1997/1998 in Indonesia conclude that
interest rate policy is effective for rupiah stabilization only if there are no disturbance from
other non-economic factor, such as negative rumors, mass mobilization and riots. Thus, interest
rate policy becomes less effective for reducing inflation level because inflation is also affected
by supply factor pressures such as production and distribution.
The current research done by Simorangkir and Adamanti (2010) examines the impacts of
fiscal stimulus and interest rate cut on Indonesian economy during global financial crisis using
Financial Computable General Equilibrium (FCGE) approach. The simulation results showed
that the combination of fiscal expansion and monetary expansion boosts economic growth of
Indonesia effectively. Relative to the effectiveness of fiscal expansion without monetary policy
expansion or monetary expansion without fiscal expansion, the combination of those two policies
is more effective. Another result of this paper showed that looking into the components of
GDP, the combination of fiscal and monetary expansion has a large multiplier effect, boosting
aggregate demand through increasing consumption, investment, government expenditure,
exports and imports. Meanwhile, from production side, the combination of fiscal and monetary
181Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
expansion has positive effects on increasing production of all economic sectors. This effect
comes from fiscal incentive (lower tax, lower import duties, etc) in increasing investment.
Moreover, the increase in aggregate demand also encourages enterprises to increase their
production. This paper also found that institutionally fiscal stimulus and monetary easing has
increased income and purchasing power of the poor and rich households in rural and urban
area. This increase in turn results in higher all household consumption.
Several other research related to the effectiveness of fiscal policy during crisis noted that
the important thing to concern when government gives fiscal stimulus are: (i) the coordination
between central and local government and (ii) the ability of local and central government to
disbursing the money quickly. In order to reach this, the complicated procurement processes
and administration practices will need to be simplified and made more transparent.
Survey done by The SMERU Research Institute in 2009 investigating the roles of Social
Safety Net programs and Fiscal Stimulus Program (especially infrastructure and labor intensive
program) in mitigating the impact of global financial crisis (GFC) found that:
- There are several problem related with the implementation of social safety net programs
that disturbing its effectiveness: (i) most local government officials lack of specific information
on the impact of GFC to the community, (ii) no information system on the socioeconomic
condition of community available and provided in hierarchical, periodic & systematic way,
(iii) lack of responsiveness on prices changes or other signs of crisis.
- Related with fiscal stimulus programs (FSP), several problems happened are: (i) Central
government does not formally socialize FSP to the local governments, (ii) There is no linkage
between the level of impact of GFC and the fund allocation to the regions, (iii) There is no
linkage between the impacted sector and the funded sector projects, (iv) There is important
role of Central MP to inform local governments on the availability of projects to be funded,
(v) Fund allocated to the regions also depend on the active role of local governments efforts
to lobby Central Government.
These problems also pointed in Yudo et al (2009) analysis about social safety net program,
known as JPS (Jaring Pengaman Sosial) that Indonesian government introduced to response to
the 1997/1998 crisis. An evaluation of this anti-poverty program which encompassed food
security, employment and income maintenance, and preservation of access to education and
health showed that in many cases the target groups had largely been missed and that the
effectiveness of the efforts varied across programs and regions. This suggested that some districts
were better than others in implementing common national program, further highlighting the
need for large improvements in program implementation, in particular in targeting the
182 Bulletin of Monetary Economics and Banking, October 2011
beneficiaries of a particular program and raising coverage within the target groups. Further
research reaffirmed the earlier conclusion on the heterogeneity of performance both across
programs and regions. This could be due to programmed design, budget allocation across
programs and regions, and regional capacity to implement the program. One main lesson to be
drawn from these assessments is that targeting requires detailed administrative guidance as
well as community involvement if it is to be both effective and socially and politically acceptable.
Furthermore, static administrative targeting was unable to capture the newly poor or shocked
households.
As a part of emerging market economies, Indonesia was impacted by the global financial
crisis which resulted in a sudden stop in capital inflows into emerging market countries and a
decline in global economic growth. The impact on macroeconomic indicators can be grouped
into the first and second round effects as follows:
2.1. First Round Effects of Global Financial Crisis
2.1.1 Impact on BOP and Exchange Rate Movements
During quarter III 2008, global economic developments placed pressure on the Indonesia»s
balance of payments. The pessimistic outlook for the global economy in 2008 signaled by
international institutions reinforced pessimism among market actors. Investors saw gloomier
prospects and higher risk in fund placements in emerging market, including Indonesia. The
high risk perceptions of Indonesian market could be seen in indicators such as Credit Default
Swap (CDS) and Government Bond Yield that increased significantly (Figure 1 and 2). Then, in
order to avert risk, the investors moved their funds to the safe haven of US Treasuries.
Figure 1.CDS (Credit Default Swap)
Figure 2.Government Bond Yield
0
200
400
600
800
1000
1200
1400
2008 2009 2010
Jan Feb Mar Apr MayJun Jul Aug Sep Okt Nov Des Jan Feb Mar Apr MayJun Jul Aug Sep Okt Nov Des Jan Feb Mar Apr MayJun
Source : Bloomberg
5
10
15
20
25
1 YR
5 YR
10 YR
2008 2009 2010
Jan Feb Mar Apr MayJun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr MayJun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr MayJun
Source : Bloomberg
183Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
A negative sentiment caused by the turbulence in global financial markets prompted a
wave of capital outflows as seen in Figure 3. At that time, foreign investors cut back their
holding of Indonesian government securities by Rp 4.4 trillion or approximately US$ 387 million.
The foreign investors» behavior to terminate their portfolio investment then followed by domestic
investors drawing their assets and these actions brought the portfolio investment in quarter
IV-2008 to a recorded net outflow. Further, the domestic agents moved their accounts from
domestic banks to overseas banks and some of them had failed to get new foreign financing
as indicated by the other investment component that recorded a deficit. The increased deficit
of other investment also explained by higher drawing on corporate lines of credit spurred by
heavy corporate foreign exchange demand to pay for imports in 2008. In contrast, direct
investment still recorded as net surplus due to acquisition of activities of domestic banks by
foreign investors.
Global financial crisis also weakened the performance of Indonesian current account in
the quarter II until quarter IV 2008 (Figure 4). This escalation in the current account deficit
resulted primarily from falling exports as global economy contracted and the falling of export
commodity prices. Current transfers buoyed by worker remittances were also decreasing although
remain positive. In 2008, incoming transfers from worker remittances generated a surplus of
US$ 5.2 billion then decreased to US$ 4.8 billion in 2009.
The downturn in the BOP in turn triggered a strong exchange rate depreciation
accompanied by high volatility. Demand pressure for foreign currency which was derived from
foreign portfolio capital outflow and the drop in foreign currency supply because of the collapse
Figure 3.Financial /Capital Account
Figure 4.Current Account
bn US$
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun2008 2009 2010
Portfolio Investment Direct Investment
Other Investment Capital Account Balance (Quarterly)
Source : Bank Indonesia
bn US$
2009 :10.75
2008 :0.13
-2.0
0.0
2.0
4.0
6.0
8.0
10.0
12.0CA Balance (Quarterly)
CA Balance (Annual)
Mar Jun Sep2008
Dec Mar Jun Sep2009
Dec Mar Jun2010
Source : Bank Indonesia
184 Bulletin of Monetary Economics and Banking, October 2011
of export led to heavy depreciating pressure on the exchange rate. Downward pressure was
also sustained by regional currencies that weakened from the spillover effects of external
turbulence. Additionally, the side effect of Bank Indonesia»s policy in lowering reserve requirement
ratio may also gave effects to the excess rupiah liquidity in the market and led to the depreciate
rupiah values decreased. These developments placed to put pressure on the rupiah, fell to its
lowest level of Rp 12,400 per US$ in November 2008 (Figure 5).
Bank Indonesia responded by issuing a series of policies to ease the pressure and prevent
excessive volatility in the rupiah such as exchange rate stabilization, policy governing the purchase
of foreign currency through banks, policy governing the transaction of foreign currency against
rupiah and prohibition of structured product transaction. These policies will be discussed in
detail in the next section.
2.1.2 Impact on Stock Market
The investor»s behavior to withdraw their funds from emerging countries during global
crisis led to fall in the stock market index of emerging market including Indonesia. The falling of
mining and agricultural commodity prices on the world market also affected the stock market
index adversely. The Indonesian Composite Index (IDX) continued to decrease sharply and closed
at 1.355 at the end of the period 2008, a drop of 50.64% compared to the quarter II 2008
(Figure 6). With this worst performance, Indonesian Stock Exchange was placed at level 5 in
Asia and the Pacific, after Vietnam, Shanghai, Shenzhen and Mumbai.
Figure 5.Exchange Rate Movements
November :Policy governing thepurchase of foreign
currency through bank
8000
8500
9000
9500
10000
10500
11000
11500
12000
12500
13000October :
- Several exchange ratestabilization policy : RR
Ratio, Daily BalancePosition, FX Swap tenor,
Foreign currency provision-Policy governing
lowering rupiah RR ratio
December :Policy governing the
transaction of foreigncurrency against
rupiah & prohibition ofstructured product
transaction
2008
Source : Bank Indonesia
Jan Feb MarApr May Jun Jul AugSep Oct Nov Dec
2009Jan Feb MarApr May Jun Jul AugSep Oct Nov Dec Jan Feb MarApr May Jun
2010
185Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
The disappointing performance of the IDX has led to a fall in both the volume and value
of transactions in the stock market especially in the fourth quarter of 2008 (Figure 7). Number
of firms who already held principle permits to issue shares also decided to postpone their share
issuances.
To avoid further fall on the stock market performances, government decided to suspended
IDX trade on 9th and 10th October 2008, issued regulations regarding buyback, banned short
selling and limited margin trade. These policies were intended to provide time for investors to
think rationally amidst the financial market turbulence caused by the crisis.
2.1.3 Impact on Market Liquidity
The global financial pressures also led to a liquidity shortage in the domestic money
market, which was reflected in a slower pace of growth in narrow money (M1) and broad
money (M2) (Figure 8). In quarter IV-2008 average M1 and M2 growth slower at 1.5% (yoy)
and 14.9% (yoy), decline from 19.9% (yoy) and 17.2% (yoy) in the preceding quarter.
Strong perception of tight bank liquidity and spillover from global conditions was also
reflected in the rising liquidity premium, which widened progressively for longer tenors. With
the tightened conditions on the money market, some banks who customarily supply liquidity,
reviewed their credit lines and credit limits to individual counterparties. This resulted in more
uneven distribution of liquidity on the market tending towards greater segmentation, due to
loss of confidence in transaction. The movement in the overnight interbank rate remained
above the policy rate (BI Rate) during July-October 2008, alongside drastically reduced transaction
Figure 6.Jakarta Stock Index
Figure 7.Trading Volume & Value
1000
1500
2000
2500
3000
3500
Source : CEIC
2008 2009 2010Jan FebMarAprMayJun Jul AugSepOct NovDecJan FebMarAprMayJun Jul AugSepOct NovDecJan FebMarAprMayJun
0
50,000
100,000
150,000
200,000
250,000
300,000Trading Volume (mn)
Trading Value (IDR bn)
2008
Jan FebMarApr MayJun Jul AugSepOct NovDec
2009
Jan FebMarApr MayJun Jul AugSepOct NovDec
2010
Jan FebMarApr MayJun
Source : CEIC
186 Bulletin of Monetary Economics and Banking, October 2011
volume and a growing spread between the highest and the lowest overnight rate (Figure 9).
Following this, banks with long positions on the money market chose to shift their liquidity to
Bank Indonesia short term investment.
2.1.4 Impact on Financial Institutions
The impact of financial crisis on Indonesian financial institutions was not as severe as in
other countries because Indonesian banks and national financial institutions» exposure to
subprime mortgages was minimal. One of the contributory factors in this regard was the
characteristics of Indonesian banks and financial institutions, which still leaned towards
conventional instruments of investment. Another factor was the quality of surveillance in the
banking sector and non-bank financial institutions as well as the capital market which has been
improved. The lessons from Asian crisis 1997 had caused Bank Indonesia to strengthen Indonesian
banking structure under Indonesian Banking Architecture as a part of financial landscape. Beside,
the banks were well disciplined in following prudent regulations which led to limit their exposure
to the larger problems associated with derivative products.
However, the tight liquidity in the money market had made it difficult for banks to manage
their fund. On October 2008, three state banks proposed a liquidity support from the government
in amount of Rp 15 trillion in total (approximately US$ 1.36 billion). Medium and smaller banks
also had more severe problems as depositors moved their fund to bigger banks because they
worried of possible bank liquidations as experienced in 1997. Under the circumstances, banks
competed to attract deposits by offering high deposit interest rates. In turn, lending rates were
Figure 8.M1 and M2 Growth
Figure 9.Daily Interbank Market Volume
%
0.0
5.0
10.0
15.0
20.0
25.0
2008 2009 2010Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun
Source : IFS-IMF
M2 Growth (yoy)
M1 Growth (yoy)
Billion of Rp
0
5000
10000
15000
20000
25000
30000
35000
40000
Jan FebMarAprMayJun Jul Ags SepOct NovDecJan FebMarAprMayJun Jul AugSepOct NovDecJan FebMarAprMayJun
2008 2009 2010
Source : Bank Indonesia
187Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
also increased. These conditions led to a decrease of the banks» performance as indicated by
Capital Adequacy Ratio and Non Performing Loans (Figure 10).
2.2 Second Round Effects of Global Financial Crisis
2.2.1 Impact on Exports
The weakening global demand and collapsing world commodity prices had deteriorated
Indonesia»s export earnings significantly, especially in quarter I 2009 (Figure 11). The growth of
exports dropped drastically to -18.73% yoy from 13.64% yoy in the same quarter 2008 (Figure
..
Figure 10.Commercial Bank Performances
%%
2.5
2.7
2.9
3.1
3.3
3.5
3.7
3.9
4.1
Source : CEIC
14
15
16
17
18
19
20
21CAR (RHS)
NPL Ratio
2008 2009 2010Mar Jun Sep Des Mar Jun Sep Des Mar Jun
Figure 11.Export and Import
Figure 12.Export Growth (yoy)
bn US$
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0ImportExport
2008 2009 2010Mar Jun Sep Dec Mar Jun Sep Dec Mar Jun
Source : IFS - IMF
-25
-20
-15
-10
-5
0
5
10
15
20
25
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q22008 2009 2010
13.64 12.36 10.63
1.99
-18.73-15.52
-7.79
3.67
19.99
14.60
Source : IFS - IMF
188 Bulletin of Monetary Economics and Banking, October 2011
12). Indonesia»s exports were concentrated in a select group of countries and relatively lacked
of diversification of export commodities and there was a higher degree of the vulnerability of
Indonesia»s export to external shocks.
Even though there was a shift in the main destination of Indonesia»s exports to China
since 5 years ago, but the Japan and the US markets still become the most destination of
Indonesian export. Prior to crisis in 2007, based on destination country, the proportion of
Indonesian export to Japan was 21.71% and to USA was 10.67% from total exports (Figure
13). The next main destinations were Singapore (9.65%), China (8.89%) and South Korea
(6.97%). The implication of this concentration is that any slowing of economic growth in the
major export destinations as experienced in the global crisis will have an adverse impact on
Indonesian exports.
Unlike other developing countries in Asia in which exports are dominated by electronic
appliances and office machinery, the structure of Indonesia»s export is dominated by oil and gas
as well as low technology industrial product. Accordingly, the primary sector especially oil,
natural gas, mining and agricultural commodities accounts for a substantial portion in Indonesia»s
export. This dependency on primary sectors made Indonesian export more vulnerable to external
shocks especially fluctuations in international commodity prices.
The growth in export volumes in all main export»s sectors decreased (Figure 14). Annual
growth of mining sector»s export decreased sharply from 47.83% in 2006 to 7.98% in 2007
and become -0.43% in 2008. The growth of agriculture and manufacturing sectors continued
to remain negative in 2009, but export in mining sector turned positive and recorded a growth
of 13.79%.
Figure 13.Indonesian Export Destinations in 2007
Figure 14.Growth of Export Volumes by Sectors
Japan, 21.71%
China,8,89%South Korea,
6,97%
Malaysia,4,68%
India,4,54%
Australia,3,12%
Thailand,2,81%
Netherlands,2,53%
Taiwan,2.39%
Germany,2,13%
Others, 19,92%
USA,10,67%
Singapore,9,65%
Agriculture, Hunting And Fishing
Mining And Quarrying
Manufacturing
%
-40
-30
-20
-10
0
10
20
30
40
50
60
2001 2002 2003 2004 2005 2006 2007 2008 2009
Source : Bank Indonesia
189Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
2.2.2 Impact on the Industrial Sector, Unemployment and Poverty
Tradable sectors were affected worst by the fall in exports during the crisis. During
quarter I 2008 until quarter III 2009, the growth of manufacturing sector declined from above
4 % to 1.5% (yoy) on average, in line with the deterioration in manufacturing products export.
Subsectors that impacted deeply in 2008 were other non-metallic products, chemicals products
and products of wood, which their export volumedeclined by 20 √ 30% from previous year. In
2009, volume of exports in machinery and motor vehicles decreased by 44% and 36% compared
to export in 2008 (Figure 15 and 16).
Figure 15. Volume of Export Decrease inManufacturing Subsector 2008
Figure 16. Volume of Export Decrease inManufacturing Subsector 2009
Other Non-Metallic Mineral Products
Chemicals and Chemical Products
Wood And of Products Of Wood And
Office, Accounting And Computing
Coke, Refined Petroleum Products
Furniture
Textiles
Basic Metals
Rubber And Plastics Products
%-35 -30 -25 -20 -15 -10 -5 0
-31%-22%
-21%
-14%
-13%
-9%
-8%
-8%
-6%
Source : Bank Indonesia
Machinery And Equipment N.E.C.Motor Vehicles, TrailersChemicals And Chemical ProductsOther Transport EquipmentWearing Apparel; Dressing And DyeingFabricated Metal ProductsBasic MetalsElectrical Machinery And ApparatusMedical, Precision And OpticalFurnitureTanning And Dressing Of LeatherPaper And Paper ProductsOther Non-Metallic Mineral ProductsTobacco Products
%-50 -40 -30 -20 -10 0
-44.25%-36.20%
-25.90%-25.07%
-21.66%-19.85%-19.40%
-11.50%-9.54%-9.03%
-6.30%-3.99%-2.77%-1.83%
Source : Bank Indonesia
The slowdown of tradable sector»s performance in turn led to a reduction in employment.
Pressure from the global crisis compelled several companies to make changes in their operations
and upgrade business efficiency. Consequently some factories were closed or began laying off
workers, driving down purchasing power even further in 2008. According to Ministry of
Table 1.Impact of Export Deterioration on Labor Absorption
Scenario
Source: Indonesia Economic Outlook, January 2008
Labor Absorption (%)
All Sector Industrial Sector
Export total decreased by 1%
Agricultural export decreased by 1 %
Mining export decreased by 1 %
Manufacture export decreased by 1 %
-0.166
-0.009
-0.005
-0.091
-0.416
-0.001
-0.002
-0.400
190 Bulletin of Monetary Economics and Banking, October 2011
Manpower and Transmigration, total number of workers that had been temporarily laid off
was of 10,306 until December 2008. Bank Indonesia»s analysis using Indonesia Input Output
Table shows that for each 1% decline in Indonesian exports resulted in 0.42% reduction in
industrial employment (Table 1).
The fallout in the export dependent sectors from the global financial crisis seems to have
put considerable pressure on prosperity levels. The impact of reductions in working hoursand
dismissals in some industries caused many households to lose their income. Additionally, farmer»s
income in the estate sector began to suffer in October 2008, following the collapse in commodity
prices. The high inflationary pressure in 2008 has also produced a reduction in real wage levels
for workers. Fortunately, the relatively buoyant economic growth until QIII-2008 helped bring
improvement in various indicators of welfare such as poverty and unemployment. Government
program designed to combat unemployment, such as the National Community Empowerment
Program (PNPM) for block grants, disbursement of Grassroots Business Credit (KUR), the
Unemployment Reduction Movement and distribution of Direct Cash Transfers, also seems to
have some positive influence in the improvement of welfare indicators.
2.2.3 Impact on GDP Growth
The decline in exports, deterioration in production and lower income simultaneously
reflected in the decline of Indonesia»s economic growth from 6.3% in 2007 to 6.0 in 2008 and
4.55% in 2009. Compared to other countries in the world, Indonesian GDP performance during
crisis were relatively remarkable. The GDP growth in 2009 was the third highest in the world
Agriculture 6.44 4.81 3.25 5.12 4.83 5.91 2.95 3.29 4.61 4.13 3.00 3.08
Mining and Quarrying (1.62) -0.37 2.32 2.43 0.68 2.61 3.37 6.20 5.22 4.37 3.08 3.77
Manufacturing 4.28 4.23 4.31 1.85 3.66 1.50 1.53 1.28 4.16 2.11 3.70 4.29
Electricity, Gas and Water Supply 12.34 11.77 10.41 9.34 10.92 11.25 15.29 14.47 13.99 13.78 8.18 4.76
Construction 8.20 8.31 7.76 5.88 7.51 6.25 6.09 7.73 8.03 7.05 7.05 7.18
Trade, Hotels and Restaurant 6.75 7.68 7.59 5.47 6.87 0.63 -0.02 -0.23 4.17 1.14 9.36 9.63
Transportation and Communication 18.12 16.57 15.64 16.12 16.57 16.78 17.03 16.45 12.22 15.53 11.92 12.91
Financial, Rental and Business Services 8.34 8.66 8.60 7.42 8.24 6.26 5.33 4.90 3.77 5.05 5.34 6.10
Services 5.52 6.51 6.95 5.93 6.23 6.70 7.19 6.04 5.69 6.40 4.62 5.25
GDP 6.21 6.30 6.25 5.27 6.01 4.53 4.08 4.16 5.43 4.55 5.69 6.17
20102008 2009
Q1 Total Q1 Q2Q2 Q3 Q3 Q4 Total Q1 Q2Q4
Table 2.GDP Growth by Sector (% yoy)
Source: Bank Indonesia
191Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
after China and India. This achievement was supported by the performance of some sectors
not related to external developments such as electricity, gas and water utilities, construction,
transport and communications, and the services sector. Growth in the electricity, gas and water
utilities sector reached 13.78% and growth in transport and communications sector reached
15.53%, respectively (Table 2). On the demand side, economic expansion in 2009 was driven
by strong domestic demand, especially consumption of both households and government that
grew by 6.21% (Table 3).
2.3 Comparison between 1997/1998 Crisis and 2007/2008 Crisis
During the last ten years, Indonesia and most other Asian countries had experienced two
financial crises. The first, Asian crisis occurred in 1997 and the second crisis known as global
financial crisis occurred ten years later in 2008. In magnitude and breadth, there are similarities
between those crises. Therefore, computation of the duration, amplitude, slope and cumulative
losses of each of the two crises will show the big picture of those events.4
Compared to 1997/1998 crisis, the impact of the current crisis on the real sector was
relatively smaller. In 1997/1998 crisis, the GDP growth continued to decline for 8 quarters and
the amplitude of the decrease was -28.53%. In fact, the GDP growth was negative in 5 quarters.
Under the GDP growth declined by 2008/2009 crisis for 3 quarters and the amplitude was only
2.16%.
As indicated by the cumulative losses, the impact of the recent financial crisis on domestic
credit was smaller than that of the 1997/1998 crisis. The figure shows the cumulative losses
Table 3.GDP Growth by Demand Side
20102008 2009
Q1 Total Q1 Q2Q2 Q3 Q3 Q4 Total Q1 Q2Q4
Consumption 5.47 5.49 6.34 6.42 5.94 7.28 6.27 5.44 5.91 6.21 2.52 3.12
Gross Fixed Capital Formation 13.88 12.16 12.30 9.40 11.86 3.46 2.37 3.24 4.18 3.32 12.45 10.13
Export 13.64 12.36 10.63 1.99 9.53 -18.73 -15.52 -7.79 3.67 -9.70 19.99 14.60
Import 17.99 16.11 11.10 -3.73 10.00 -24.42 -21.04 -14.67 1.62 -14.97 22.60 17.74
GDP 6.21 6.30 6.25 5.27 6.01 4.53 4.08 4.16 5.43 4.55 5.69 6.17
Source: Bank Indonesia
4 The more explanation about similarities and differences of 1997 crisis and 2008 crisis could be seen in Appendix
192 Bulletin of Monetary Economics and Banking, October 2011
Table 4. Duration, Period, Amplitude, Slopeand Cumulative Losses for Recent Crisis And 1997/98 Crisis
Source : CEIC, estimated
1997/1998 Crisis 2008/2009 Crisis
Economic Growth Recession (% yoy)Duration 8 quarters 3 quarters
Period Q1 1997 √ Q4 1998 Q4 2008 √ Q2 2009
Amplitude 28,53% 2,2%
Slope 4% 0,72%
Cummulative Losses 115,16% 4,86%
Domestic Credit Growth (% yoy)Duration 4 quarters 4 quarters
Period Q3 1998 √ Q2 1999 Q4 2008 √ Q3 2009
Amplitude 150,4% 25,2%
Slope 37,6% 6,3%
Cummulative Losses 370,8% 56,8%Stock Prices (Index)
Duration 5 quarters 4 quartersPeriod Q3 1997 √ Q3 1998 Q1 2008 √ Q4 2008
Amplitude 448 1390
Slope 90 348
Cummulative Losses 1411 2999
%
Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul NovMar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
-80.0
-60.0
-40.0
-20.0
0.0
20.0
40.0
60.0
80.0
100.0Jun-98, 90.5%
Jun-99, -59.9%
Sep-08, 34.8% Sep-099.6%
GDP Growth Indonesia (yoy)%
-20
-15
-10
-5
0
5
10
15
Dec-96, 10.28%
Dec-98, -18.26%
Dec-99, 5.36%
Jun-09, 4.08%
Jun-10, 6.17%Sep-08, 6.25%
Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul NovMar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Figure 17b.Peak and Trough of Credit Growth
Figure 17a.Peak and Trough of GDP Growth
Source : CEIC, estimated Source : CEIC, estimated
193Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
consecutively by 370.8% and 56.8%. Even the duration of the credit decline was almost equal
at 4 quarters, the amplitude of 2008/2009 crisis was lower (25.2%) than that of 1997/1998
crisis (150.4%).
In the financial market, although it seems that the impact of recent crisis was bigger
because the stock market index decreased by 1390 points, in relative terms (percentage change)
it was lower (50,6%) than that of 1997 crisis (61,9%). The duration was also shorter.
III. METHODOLOGY
In examining the relative effectiveness of both monetary and fiscal policies on economic
growth, we use the Engle and Granger two step estimating procedure which allows an explicit
testing of co-integration and specification of the Error Correction Model (ECM).
Empirical model in this paper is aimed at testing the relationship of economic growth
with monetary policy, fiscal policy and other control variables.
Figure 17c.Peak and Trough of Stock Market Index
Source : CEIC, estimated
Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul NovMar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
500
1000
1500
2000
2500
3000
3500
Jun-97, 725
Sep-98, 276
Dec-07, 2,746
Dec-08, 1,355
(1)
where Y is a measure of economic activity, MP, a measure of monetary policy, FP measure of
fiscal policy and Z is other control variables that may affect economic activity. Finally, the general
form of the Error Correction Model (ECM) specification in this paper is:
194 Bulletin of Monetary Economics and Banking, October 2011
(2)
Where:
Y = dependent variables (economic output)
X = independent variables, consists of monetary variables, fiscal variables, and other
control variables
ECM = residuals from long run relationship between variable
n = number of explanatory variable in the model
p = number of lags used to represent the short run dynamics in the model
There are several variables which could be used as proxies for of economic activity,
fiscal policy, monetary policy and control variables as outlined in the table below. We use
quarterly data from 1990 Q1 to 2010 Q2. To capture the effect of financial crisis period to the
effectiveness of policy choices, we add interaction variables of dummy recession and policy
variables. Some of variables were adjusted for seasonality using Census X12 method.
Table 5.List of Variables
Growth
Fiscal
Monetary
Inflationary
Effect
External Sector
Dummy
Recession
IFS, staff estimated
IFS
BI
BI
BI
1.1.4.1.1.1
BI
IFS
IFS
BI
IFS, staff estimated
BI
IFS
IFS
Estimated
Estimated
Government Expenditure √
Interest Payment
Prior to Q3 2005, we use SBI
1 month as proxy for policy
rate
Q1 1997 √ Q4 1998
Q4 2008 √ Q2 2009
Indicator Variable Sources Notes
GDP Real
GDP Nominal
Fiscal Balance
Government Revenue
Government Expenditure
Primary Expenditure
Primary Balance
M1
M2
Policy Rate
GDP Deflator
CPI
Exchange Rate
Current Account Balance
1997/1998 Recession
2008/2009 Recession
Notes: all variable are in logarithm, except Fiscal balance, primary balance, current account balance (because they contains negative values) and dummyrecession (binary 1/0)
195Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
IV. RESULT AND ANALYSIS
4.1. Empirical Result
To characterize the time series property of the variables, we utilize the Augmented Dickey-
Fuller (ADF) and the Phillips Perron (PP) methods. This PP approach is more appropriate than
ADF since the data shows a structural break as effect of 1997/1998 crisis. Both of the ADF and
PP test indicate that most of the series are non-stationary when the variables are defined in
levels, except Fiscal Balance, Primary Balance, Policy Rate and Current Account Balance. But
first-differencing the series removes the non-stationary components in all cases and the null
hypothesis of non stationary is clearly rejected at the 5% significance level suggesting that all
variables are integrated of I(1). Thus, the next step of testing for possible cointegration relationship
will be done only with the I(1) variables.
t-stat p-values t-stat p-values t-stat p-values t-stat p-values
Real GDP RGDP -2.631 0.268 -2.139 0.032 I(1) -2.327 0.415 -8.320 0.000 I(1)
Nominal GDP NGDP -2.038 0.572 -5.929 0.000 I(1) -2.117 0.529 -5.491 0.000 I(1)
Real GDP_Adjusted RGDP_SA -2.214 0.475 -5.216 0.000 I(1) -1.921 0.634 -5.159 0.000 I(1)
Nominal GDP_Adjusted NGDP_SA -1.913 0.639 -1.957 0.049 I(1) -1.869 0.661 -2.939 0.004 I(1)
Fiscal Balance FB -4.231 0.006 -6.526 0.000 I(0) -9.025 0.000 -29.488 0.000 I(0)
Government Revenue GR 5.943 1.000 -17.544 0.000 I(1) 2.972 0.999 -42.299 0.000 I(1)
Government Expenditure GE 5.327 1.000 -20.666 0.000 I(1) 2.100 0.991 -36.833 0.000 I(1)
Primary Expenditure PRIM_GE 5.137 1.000 -20.657 0.000 I(1) 1.976 0.988 -42.757 0.000 I(1)
Primary Balance PB -5.436 0.000 -6.289 0.000 I(0) -7.132 0.000 -26.702 0.000 I(0)
Fiscal Balance_Adjusted FB_SA -3.074 0.003 -12.525 0.000 I(0) -8.552 0.000 -20.894 0.000 I(0)
Government Revenue_Adjusted GR_SA 4.040 1.000 -15.234 0.000 I(1) 3.211 1.000 -15.292 0.000 I(1)
Government Expenditure_Adjusted GE_SA 3.562 1.000 -10.192 0.000 I(1) 4.154 1.000 -18.761 0.000 I(1)
Primary Expenditure_Adjusted PRIM_GE_SA 4.400 1.000 -11.337 0.000 I(1) 3.046 0.999 -18.449 0.000 I(1)
Primary Balance_Adjusted PB_SA -1.843 0.063 -12.502 0.000 I(0) -7.297 0.000 -23.681 0.000 I(0)
M1 M1 -2.348 0.404 -1.880 0.058 I(1) -2.369 0.393 -7.610 0.000 I(1)
M2 M2 -0.933 0.947 -8.250 0.000 I(1) -0.933 0.947 -8.248 0.000 I(1)
M1_Adjusted M1_SA -1.496 0.823 -3.371 0.001 I(1) -1.641 0.768 -5.250 0.000 I(1)
M2_Adjusted M2_SA -0.864 0.955 -6.885 0.000 I(1) -0.941 0.946 -3.738 0.000 I(1)
Policy Rate PR -3.265 0.020 -7.633 0.000 I(0) -3.000 0.039 -7.646 0.000 I(0)
GDP Deflator GDPDEFL -2.125 0.524 -4.443 0.000 I(1) -1.871 0.661 -4.293 0.000 I(1)
GDP Deflator_Adjusted GDPDEFL_SA -2.271 0.444 -3.798 0.000 I(1) -1.849 0.672 -3.798 0.000 I(1)
CPI CPI -2.402 0.376 -3.049 0.003 I(1) -2.208 0.478 -4.278 0.000 I(1)
Exchange Rate ER -1.845 0.673 -5.637 0.000 I(1) -1.221 0.899 -6.479 0.000 I(1)
Current Account Balance CAB -3.021 0.003 -11.764 0.000 I(0) -3.021 0.003 -16.024 0.000 I(0)
GDP US USGDP -0.215 0.992 -2.199 0.028 I(1) -0.249 0.991 -3.144 0.002 I(1)
GDP Japan JPGDP 1.014 0.917 -3.321 0.001 I(1) 1.351 0.955 -16.482 0.000 I(1)
Variables Abbreviation
ADF Test Result Phillips Perron Test Result
Level 1st difference Level of
integration
Level 1st difference Level ofintegration
Table 6.Unit Root Test for Variables
196 Bulletin of Monetary Economics and Banking, October 2011
Following the Engle and Granger two step-method, in the next step we estimate the
long run equilibrium relationship among variables by OLS and test for stationary of the residuals,
using critical values for the Engle - Granger Cointegration Test provided in Enders (2004). After
estimating several alternatives model based on the variables, we found the best long run
cointegration equations as follow:5
5 RM1_SA defined as real seasonally adjusted M1 which equal to nominal M1/GDP Deflator. RGE_SA defined as real seasonallyadjusted government expenditure which equal to nominal government expenditure/GDP Deflator. value in ( ) shows standard error.*** significant at a = 1%, ** significant at a = 5%, * significant at a = 10%. We realize that there might be endogenityrelationshipbetween RM1_SA and GDP_SA, but to be inline with the agreed methodology we assume the one-way relationship between themand use ECM. For robustness, we also use VECM and found the long run relationship between those variables.
(4)
The critical value of residual unit root test from this equation is -6.61, and given the
critical value of Engle - Granger Cointegration Test for 2 variables which is -4.123 for significance
at 1%, then variables real GDP, real M1 and real government expenditures are said to be
cointegrated.
We next switch to a short run model with an error correction mechanism in the form:
RGDP_SA = 5.13 + 0.88 RM1_SA + 0.20 RGE_SA
(0.05)*** (0.03) *** (0.25) ***
R2 = 0.94
(3)
Where X consists of real M1, real government expenditure and other control variables, meanwhile
ECM is residuals from equation (3). To address the impact of crisis, we also use dummy recession
variables CR97 and CR08 and interaction variables between recession in 2008 and policy
measures. With general to specific approach to several combinations of cointegrated variables
and lags, we found the best models as follows:
197Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
The empirical results show that Real GDP cointegrated with real M1 and real fiscal
expenditure. On the basis of this information, an error correction model was developed which
was shown to be well-specified relative to its own information set and capable of parsimoniously
representing the data set.
From the error correction models, we could conclude that in the short run the changes in
real GDP is significantly affected by changes in real fiscal expenditures and real monetary changes.
The previous real GDP changes are not significant to affect real GDP changes. The crisis 1997
decrease the real GDP significantly, meanwhile the crisis 2008/2009 effect is not significant.
The model shows that in the period of 2008 crisis, the impact of M1 changes and fiscal
expenditure to output relatively similar with the impact in the normal period. Thus, both of
variables are still significant in affecting output although in crisis period.
The result also shows a well-defined error correction term, and indicates a feedback of
7% of the previous quarter»s disequilibrium from the long run money supply, and fiscal
expenditure to economic activity. To evaluate the goodness of the model, we did some in
sample forecasting and compared the result with the actual data. The result was quite good as
shown in Figure 20. The root mean square error (RMSE) of the forecast was only 0.01.
Table 7.Error Correction Models
Real M1(-3)
Real Gov_Exp
ECM(-1)
Exchange Rate (-1)
Inflation (-1)
Crisis 1997
Crisis 2008
Real M1(-3)*Crisis 2008
Real Gov_Exp * Crisis 2008
Constant
R2
DW-Stat
SIC
0.06
0.03
-0.07
-0.05
-0.29
-0.01
0.01
0.14
0.08
0.02
0.66
1.98
-5.40
1.83
3.80
-2.04
-3.00
-5.47
-2.04
0.75
0.51
0.42
11.36
6
8
10
Variable Dependent : Real GDP
*** significant at α = 1%, ** significant at α = 5%, * significant at α = 10%
Coefficient t-stat
*
***
**
***
***
**
3
4
5
***
7
9
11
198 Bulletin of Monetary Economics and Banking, October 2011
Figure 18.Evaluation of Model
The relative effectiveness of policy actions are determined by the size of the contribution
of policy instruments implemented to limit the severity of the downturn in order to achieve a
more positive outcome. While some economists believed that monetary policy should be the
first line of defense during the turbulence, others argued that fiscal policy has a more important
role particularly when conventional monetary policy measures were not sufficient in addressing
losses in output due to vulnerable in a weakening economy. Although it is not specific in the
crisis period, but the lag in the effect of monetary changes relative to fiscal policy as shown in
table 6 indicates that the impact of fiscal policy on GDP is relatively faster than monetary policy.
This result is in line with the Elmendorf and Furman (2008) which consider that a key
potential advantage of fiscal stimulus relative to monetary stimulus was that it could boost
economic activity more quickly, and true fiscal stimulus implemented promptly can provide a
larger near-term impetus to economic activity than monetary policy.
4.2. Analysis
In the empirical model above, we use M1 for indicator of monetary policy and real
government expenditure for indicator of fiscal policy. Meanwhile, after implementing Inflation
Targeting Framework (ITF) in July 2005 BI has shifted the focus of monetary operation toward
short term interest rate, replacing the former operating target of monetary base.
At operational level in the ITF, the monetary policy stance is reflected in the setting of the
policy rate (BI Rate) with the expectation of influencing money market rates and in turn the
12.2
12.4
12.6
12.8
13.0
13.2
13.4
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010
LRGDP_SAF LRGDP_SA
199Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
deposit rates and lending rates in the banking system. At early stage of implementation of BI
Rate, BI rate was intended to only be reflected in the discount rate of Bank Indonesia one-
month paper (known as 1-month SBI). Since late January 2008, some gradual steps have been
taken to focus more on managing short term market interest rates around the BI Rate level.
Effective from June 9, 2008, BI has officially changed the operating target from 1-month
SBI rate to overnight (O/N) interbank money market rate (PUAB O/N). Changes in these rates
will influence output and inflation.Ω
Beside those rates, there are also several monetary instruments used in controlling the
agregate demand of the economy. Regarding to that, in this part, the analysis of the monetary
policies and fiscal policies that have been done will be broaden, not only M1 and government
expenditure. Because exchange rate also significant in affecting output, there will be analysis
about foreign exchange policy.
4.2.1 Monetary Policy
Bank of Indonesia took a series of policy measures in response the global financial instability.
Generally, the monetary policy measures are decided by taking into considerations, economic
circumstances and macroeconomic characteristic.As shown in the figure below, during the
period from January 2008 to November 2008, Bank Indonesia increased the policy rate from
8% to 9.50% in order to constrain the pressure of hiking CPI. Although the pressure from
financial stability point of view was higher as indicated by the high overnight interest rate,
banking interest rates and yield rate of government bonds were increasing due to shortage of
global liquidity at the time, BI decided to increase the policy rate in view of concern on inflation.
In line with slow down economic growth, inflation was also declined during 2008 √ 2009.
Accordingly, the policy rate has been adjusted downward since December 2008.
Further, as a response to the global financial crisis Bank Indonesia took some measures to
address liquidity issues in the financial sector. Following the declared bankruptcy of Lehman
Brothers in September 2008, the interbank market rose sharply from 6.98% to 9.37% Q-1 to
Q-III 2008. Beside that the volume of interbank rupiah transaction declined by 41% during the
period due to falling of market confidence among the banking institutions. The main objective
of the unconventional policy as shown in the table 4 was to bring the interbank interest rate to
the policy rate and feasibility its convergence with the policy rate. In order to reduce excessive
volatility in the interbank money market the interest rate corridor was narrowed on 4 September
2008 and on 16 September 2008.
200 Bulletin of Monetary Economics and Banking, October 2011
3-A
pr-0
8
O/N
inte
rban
k ra
te re
plac
ed th
e 1
mon
thSB
I rat
e as
the
oper
atio
nal t
arge
t- F
asbi
Cor
ridor
: BI
Rat
e - 3
00 b
psRe
po C
orrid
or :
BI R
ate
+ 30
0 bp
s
16-S
ep-0
8
Fasb
i Cor
ridor
: BI
Rat
e - 1
00 b
psRe
po C
orrid
or :
BI R
ate
+ 10
0 bp
s
4-Se
p-08
Fasb
i Cor
ridor
: BI
Rat
e - 2
00 b
psRe
po C
orrid
or :
BI R
ate
+ 30
0 bp
s
23-S
ep-0
8
Expa
ndin
g tim
e pe
riod
ofFT
O fr
om 1
4 da
ys to
3 m
onth
4-D
es-0
8
Fasb
i Cor
ridor
: BI
Rat
e - 5
0 bp
sRe
po C
orrid
or :
BI R
ate
+ 50
bps
9-D
es-0
8
- Ope
n w
indo
w re
po o
f 2-1
4da
y te
nure
Jan-
08
Feb-
08M
ar-0
8A
pr-0
8M
ay-0
8Ju
n-08
Jul-0
8A
ug-0
8Se
p-08
Oct
-08
Nov
-08
Dec
-08
Dec
-08
Dec
-08
BI R
ate
9.25
%
Oct
-08
- Nov
-08
BI R
ate
9.50
%
Sep-
08BI
Rat
e9.
25%
Aug
-08
BI R
ate
9%
Jul-0
8BI
Rat
e8.
75%
Jun-
08BI
Rat
e8.
5%
May
-08
BI R
ate
8.25
%
Jan-
08 -
Apr
-08
BI R
ate
8%
14-O
ct-0
8
Low
erin
g Re
serv
e Re
quire
men
tRu
piah
Dep
osit
to 7
.5%
Fore
ign
Dep
osit
to 1
%
23-O
ct-0
8
Low
erin
g Re
serv
e Re
quire
men
tRu
piah
Dep
osit
to 5
%
Apr
-09
Ope
n w
indo
w re
po o
f 1 m
onth
tenu
re
Sep-
09
Ope
n w
indo
w re
po o
f 3 m
onth
tenu
re
Oct
-09
Impl
emen
tatio
n of
2.5
% s
econ
dary
rese
rve
requ
irem
ent
Jan-
09D
ec-0
8
Feb-
09M
ar-0
9A
pr-0
9M
ay-0
9Ju
n-09
Jul-0
9A
ug-0
9Se
p-09
Oct
-09
Nov
-09
Dec
-09
Jan-
09Fe
b-09
Mar
-09
Apr
-09
May
-09
Jun-
09Ju
l-09
Aug
-09
- Dec
-09
BI R
ate
8.75
%BI
Rat
e8.
25%
BI R
ate
7.75
%BI
Rat
e7.
50%
BI R
ate
7.25
%BI
Rat
e7.
00%
BI R
ate
6.75
%BI
Rat
e6.
50%
Pict
ure
1.
Mo
net
ary
Polic
y in
200
8 an
d 2
009
201Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
Like other emerging central banks, Bank Indonesia also responded to the liquidity
problem in the financial sector by reducing reserve requirement, followed by providing liquidity
facility in order to flow the fund into the financial market. Furthermore, there were coordination
between the central bank and the government taken to address confidence issues and also to
respond to the asset price burst. Several policies also had been taken to address confidence
issues and asset price burst. The description of policies taken by Bank Indonesia during the
global crisis could be seen in the appendix.
a. Evaluation of Interest Rate Policy
There were significant differences between the movements of the policy rate in 1997/
1998 crisis and the recent crisis. To contain the inflation pressure which was very high in the
1997/1998 crisis (almost 83% in Q-IV 1998), the central bank increase the SBI rate significantly
to 68% and then gradually brought down to 11.03% in Q-I 2000. Meanwhile, prior to the
recent crisis, the inflation pressure was not so high, and hence, the policy rate was raised only
up to 9.25%. At that moment, BI Policy to increase the policy rate in quarter II, III and IV 2008
1997/1998 Crisis 2008/2009 Crisis
Policy Rate (%)Duration 6 quarters more than 6 quarters
Period Q4 1998 √ Q1 2000 Q1 2009 √ Q2 2010 (still not increase yet)
Amplitude 57.73% 2.75%
Slope 10% 0.46%
Lending Rate (%)Duration 9 quarters more than 6 quarters
Period Q4 1998 √ Q4 2000 Q1 2009 √ Q2 2010
Amplitude 18.1% 2.0%
Slope 2.0% 0.3%
Inflation (% yoy)Duration 6 quarters 5 quarters
Period Q4 1998 - Q1 2000 Q4 2008 - Q4 2009
Amplitude 83.6% 9.36%
Slope 13.9% 1.9%
Table 8.Duration, Period, Amplitude and Slope of Interest Rate and Inflation
Source : CEIC, estimated
202 Bulletin of Monetary Economics and Banking, October 2011
was contrary to other central banks in the region and across the world as then policy was to
lower their interest rates to address liquidity issues and reduce economic activity. BI commenced
gradually reducing its policy rate in QI 2009 and now it l stays in 6.5%.
During both of crisis periods, the magnitude of the decline in average lending rate in
the banking sector was much smaller than the decline of BI Rate which could be seen in the
wider spread between lending rate and policy rate. From micro perspective of banks, some
factors contributing to the lending rate movements included the cost of fund and risk premium
which tended to rise during the crisis, and profit margin. A previous study has showed that
the decrease in the aggregate banking cost of fund throughout 2009 tended to be slower
than the decline in BI Rate. Furthermore risk premium in the economy was still perceived to
be high and there were indications that the banking industry preferred to maintain their
profit margin. All of these contributed to reduce the power of policy rate pass through to
lending rates.
Figure 19a.Peak and Trough of Interest Rate
Figure 19b.Peak and Trough of Inflation
Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul NovMar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
0
10
20
30
40
50
60
70
80Policy Rate
Working Capital Lending RateSep-98, 68.76
Mar-00, 11.03Dec-08, 9.25 Jun-10, 6.5
Sep-98, 35.7
Dec-00, 17.65Dec-08, 15.22 Jun-10, 13.26
Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul NovMar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar Jul Nov Mar
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
%
-10.0
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0Sep-98, 82.4%
Mar-00, -1.2%Sep-0812.1% Dec-09
2.8%
Source : CEIC, estimated Source : CEIC, estimated
Non - Crisis Crisis
Q2 1996 - Q1 1997
6.12 %6.12 %6.12 %6.12 %6.12 %
Q1 2006 - Q4 2006
4.29 %4.29 %4.29 %4.29 %4.29 %
Q2 1999 - Q1 2000
8.23 %8.23 %8.23 %8.23 %8.23 %
Q1 2009 - Q4 2009
7.41 %7.41 %7.41 %7.41 %7.41 %
Table 9.Average Spread between Lending Rate and Policy Rate During Expansion Period
203Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
b. Evaluation of Lowering Reserve Requirement
Reserve Requirement Policy was affected to provide more rupiah liquidity to the banking
system to achieve liquidity constrain and reduce volatility in the interbank market. Its impact
could be seen in the interest rate of interbank market which declined after the policy
announcement. The volatility of interbank interest rate also decreased. As measured by the
standard deviation, during the period of 1 month before the implementation of the policy, the
volatility of interbank interest rate was 0.22% and it was decreased to 0.08% during the 1
month period after the implementation of the policy.
The volume of interbank transactions also increased although the impact was not as
immediate as that on the interest rate. One of the reasons for lower increase in the interbank
market transaction could be expansion in the fine tuning operations of the Central bank at that
time, which enhanced the access to liquidity from the Central bank and hence reduce the need
for borrowing from the interbank market. This was also supported by the amendment of liquidity
facility for commercial banks from central bank which provide wider access for banks to receive
funding for a longer time horizon than the usual inter-day facility. Another reason was to
reduce the loan disbursement by commercial bank which was in line with the lower demand of
credit at that time due to global crisis and hence the lower need to borrow funds.
Figure 20.Interbank Market Interest Rate
Figure 21.Volume of Interbank Market Transaction
Lowering RRPolicy, 23 Oct '08
8.80
9.00
9.20
9.40
9.60
9.80
10.00
10.20
2 0 0 8
1Aug
11Aug
21Aug
31Aug
10Sep
20Sep
30Sep
10Oct
20Oct
30Oct
9Nov
19Nov
29Nov
9Dec
19Dec
29Dec
Source : Bank Indonesia
Billion of Rp
Lowering RRPolicy, 23 Oct '08
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
2008
1Aug
11Aug
21Aug
31Aug
10Sep
20Sep
30Sep
10Oct
20Oct
30Oct
9Nov
19Nov
29Nov
9Dec
19Dec
29Dec
However, there is a side-effect of this policy. The increasing supply of rupiah has led to
the rise of money in circulation in the market (M0), and this could cause a depreciation of
rupiah.
204 Bulletin of Monetary Economics and Banking, October 2011
Figure 22.Interest Rate Corridor (%)
Table 10.Spread of Policy Rate and Interbank Market Rate
Period Spread
1 week before the 1st adjustment
1 week after the 1st adjustment
1 week after the 2nd adjustment
1 week before the 3rd adjustment
1 week after the 3rd adjustment
9.08 bps
11.72 bps
27.42 bps
35.57 bps
17.59 bps
Notes
Higher
Higher
Ω
Lower
Note:1st : 4 September 2008; 2nd : 16 September 2008; 3rd : 4 December 2008Source: Bank Indonesia, estimated from daily data
Adjustment of Corridor ofOvernight Interest Rate
Adjustment ofCorridor of O/N
Interest Rate
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
2 0 0 8
1Aug
11Aug
21Aug
31Aug
10Sep
20Sep
30Sep
10Oct
20Oct
30Oct
9Nov
19Nov
29Nov
9Dec
19Dec
29Dec
Interbank O/N RateBI RateOperational Corridor
Source : Bank Indonesia
c. Evaluation of Narrowing the Interest rate corridor for Standing DepositFacility/ Lending Facility (Repo)
This policy was aimed to reduce excessive volatility in the interbank money market and
increase credibility of Bank Indonesia. The lower spread between interbank interest rate and
policy rate indicate as higher degree of credibility of central bank.
From the figure above, we could see that after the first and second adjustment of the
corridor, the spread between interbank market rate and policy rate was relatively higher and it
contradicted with the purpose of the policy. One of the reasons would be the increasing pressure
of liquidity because of higher risk perception in September following the bankruptcy of Lehman
Brother. Meanwhile, the third adjustment was successful to decrease spread between interbank
205Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
market and policy rate (Table 10). For comparison, best practices of this spread in several ITF
countries are about 20bps.
4.2.2 Fiscal policies
Like other countries, Indonesia launched fiscal policy measures as a countercyclical measure
to fight the slowdown direct effect on aggregate demand. In order to reduce the impact of
current crisis, the Indonesian government took ten steps for the purpose of economic stabilization
and securing the state budget. Additionally, the government offered a fiscal stimulus package
amounting to 73.3 trillion Rupiah or 7.56 billion US$ (2.6% of GDP) with following aims,
a. Maintaining household purchasing power to keep consumption growth above 4%.
b. Improving real sector resilience and competitiveness to prevent more worker layoffs.
c. The government issued some traded policies.
d. Creating job opportunities for unemployed/laid-off workers by launching labor √intensive
infrastructure projects
e. Social protection and poverty alleviation has decided to state in the budget for 2009.
In the period of crisis, the realization of fiscal balance at the end of 2008 was a significant
improvement with a marginal deficit, while in 2009, fiscal stimulus and the lower government
revenue due to economic slowdown leads to a higher fiscal deficit of 1,6% of GDP (Figure 23).
Based on its component, the largestgovern,entexpenditure in 2008 and 2009 were transfer to
region and subsidies (Figure 24).
Figure 23.Government Finance Operation
Figure 24.Total Expenditure Component
% of GDP% of GDP
-2.00
-1.50
-1.00
-0.50
0.00
0.50
1.00
1.50
2.00
2.50
2004 2005 2006 2007 2008 2009 2010*0.0
5.0
10.0
15.0
20.0
25.0
-1.0%
-0.5%
-0.9%
-1.3%
-0.1%
-1.6% -1.6%
1.7% 1.8% 1.5% 0.8% 1.7% 0.1% 0.3%
Fiscal Balance Primary BalanceExpenditure (RHS) Revenue (RHS)
Poly. (Fiscal Balance)
Source : Ministry of Finance
% of GDP
0.0
5.0
10.0
15.0
20.0
25.0
2004 2005 2006 2007 2008 2009
Transfer to Region Other Routines Social AssistanceCapital Expenditure Subsidy Interest PaymentMaterial Personnel Total Expenditure
5.7% 5.4% 6.8% 6.4% 5.9% 5.5%
2.7% 1.2%
1.6%1.6% 1.5% 1.3%
4.0% 4.4%3.2%
3.8% 5.6%2.9%
2.7% 2.4%2.4% 2.0%
1.8%
1.7%
2.3% 2.0%2.2%
2.3% 2.3%
2.3%
18.6% 18.4%20.0% 19.2% 19.9%
17.1%
Source : Ministry of Finance
206 Bulletin of Monetary Economics and Banking, October 2011
The success of the government to hold fiscal sustainability during the financial crisis in
2008 was attributed to (1) over the past ten years, fiscal policy actions have reduced the high
public debt ratio. (ii) The government has taken significant measures to reduce domestic fuel
subsidies enabling it to increase spending at both the central and local government levels. (iii)
The government has increased the total spending on education.
However, in the Indonesian case, the problem in the fiscal stimulus package was that
regional government has a limited ability to complete their budgets on time. Therefore, the
stimulus package hadn»t used optimally as can be interpreted from the data for October and
December 2009 which show that only 52.1% of fiscal stimulus plan was realized. Lack of
socialization, frugal spending and slow regulation implementation led to this low absorption of
the fiscal stimulus. Thus, to increase the effectiveness of fiscal policy, it is needed to configure
an effective and understandable standard operating procedure for fiscal policy implementation,
both in central and regional areas.
To analyze the effectiveness of fiscal policy to boost economic activity after the crisis is
not an easy task, thus we try to adopt the evaluation of fiscal stimulus (FS) principles based on
Elmendorf and Furman (2008) as follows:
Table 11.Fiscal Stimulus Plan and Realization
Source: Ministry of Finance and other sources
Notes : 1) Realization until October 2009, more updated data is not available2) Realization until December 2009
1 43.0 20.5 47.7%1)
Reductions in Income Tax Rates 32 18 56.3%1)
Lower Corporate Tax Rate 18.5 12.8 69.2%1)
Lower Personal Income Tax Rate 13.5 5.2 38.5%1)
Income tax-free band raised to IDR 15.8 million 11 2.5 22.7%1)
2 13.3 3.7 27.8%1)
VAT on oil/gas exploration, cooking oil 3.5 2.5 71.4%1)
Import duties on raw materials and capital goods 2.5 0.3 12.0%1)
Payroll tax 6.5 0.1 1.5%1)
Geothermal tax 0.8 0.8 100.0%1)
3 17 14.0 82.2%Reduced price for automotive diesel 2.8 2.8 100.0%1)
Discounted electricity billing rates for industrial users 1.4 1 71.4%1)
Additional infrastructure expenditures+subsidies+government equity injection 12.2 10.18 83.4%2)
Upscaling of Community Block Grants (PNPM) 0.6 n/a73.3 38.2 52.1%
Plan Realization
(IDR Trillion) (IDR Trillion)Tax Savings
Tax/Import Duty Subsidies for Business/Targeted Households
Pro-business/Jobs subsidies + budget expenditures
TOTAL
DescriptionNo% of
Realization
207Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
7 SILPA: Excess of government budget utilization in the previous year
Table 12.Evaluation of the Effective Fiscal Stimulus Principles
Principles Explanation
Timely
Targeted
Temporary
Although government immediately gave instruction for FS,there were problems that delayed the implementation ordisbursement of the fund.The government expenditure was mostly disbursed in Q-4(Figure 26). It would have been better if the disbursementin each quarter were quite balanced. Thus in terms of thetimely principles, the Indonesia»s fiscal stimulus was lesseffective.
The biggest proportion of the FS was tax reductions. Thiscould stimulate economic output from investment andindirectly would increase employment and wages. Then,there will be increases in consumption and economic output.Moreover, the big spending in infrastructure was goodbecause it would boost a sustainable growth in longer terminstead of only short term.Thus from the targeted principles, the Indonesia»s fiscalstimulus was effective.
The source of fund for fiscal stimulus came from the excessof budget utilization (SILPA)7 in 2008 and debt.Fund from excess of budget utilization won»t affect the nextgovernment budget but the usage of debt, in the long termcould impact the budget deficit.Additionally, the budget deficit plan in 2010 still relativelyhigh (1.6% GDP) (Table 15). Thus from the temporaryprinciples, the Indonesia»s fiscal stimulus was quite effective
Explanation Measure for Indonesia»s Fiscal Stimulus
FS should not be enactedprematurely, delayed too long, orconsist of tax cuts or spendingincreases that would take too longto be implemented or to boostoutput
Tax cuts and spending increasesshould be directed so that theyprovide the greatest benefit topeople who are affected mostadversely by an economic slowdown
The FS should not increase thebudget deficit in the long run
Note: The effectiveness of means is based on authors» opinion.
Figure 25.Government Expenditure by Quarter
Figure 26.Tax Revenue versus GDP Changes
0
100,000
200,000
300,000
400,000
90 92 94 96 98 00 02 04 06 08 10
Q1Q2Q3Q4
Source : CEIC
Trillion of Rp
-
200
400
600
800
1,000
1,200
1,400
2003 2004 2005 2005 2007 2008 2009 2010*
Change of Tax Revenue (yoy)
Change of GDP (yoy)
208 Bulletin of Monetary Economics and Banking, October 2011
Figure 27.Impact of Foreign Exchange Policy
Table 14.Volatility of Foreign Exchange after Policy
1 weekbefore
1 weekafter
15 October's policies 94.58 69.61 Lower
13 November's policies 349.90 244.21 Lower
16 December's policies 338.22 73.40 Lower
Std Deviation of Rp/US$FX Policies Result
- Abolishing the limit of dailybalance position of short term
foreign loan- Foreign Exchange Provision for
Domestic Corporation- The extension of FX Swap tenor
Regulation governingthe purchase ofForeign Exchangeby Banks
Prohibition ofstructured
producttransaction
8000
8500
9000
9500
10000
10500
11000
11500
12000
12500
13000
2 0 0 8
1Aug
11Aug
21Aug
31Aug
10Sep
20Sep
30Sep
10Oct
20Oct
30Oct
9Nov
19Nov
29Nov
9Dec
19Dec
29Dec
Source : CEIC, Estimated
Table 13.Impact of Fiscal Policy on Government Budget
PeriodFiscal Balanceas a% of GDP
1990-1996 Average
1997-1998 Average
1999
2001-2008 Average
2009
2010 Estimates
0.26%
-1.45%
-2.84%
-1.16%
-1.56%
-1.59%
Primary Balanceas a % of GDP
2.05%
1.13%
1.05%
1.92%
0.12%
0.28%
Source: Ministry of Finance
4.2.3 Foreign Exchange Policy
To address and avoid more Rupiah depreciation, Bank Indonesia also introduced following
foreign exchange policy measures such as extending FX swap tenor, issuing regulation about
the purchase of foreign exchange by banks, etc (details could be seen in Appendix).The foreign
exchange policy measures were effective to reduce the volatility of exchange rate during the
crisis. Although in term of level, rupiah currency continue to depreciate after several exchange
rate policy measures in October and November 2008 because of the massive capital outflow,
the volatility of the exchange rate became relatively lower after the introduction of policy
measures.
209Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
4.2.4 Condition of Indonesia Economic after the Crisis
There have been improvements in Indonesian economy during 2009 and 2010 which
boosted optimism over the sustainability of the ongoing economic recovery. The positive
economic performances include an improvement in the risk indicators, performance in stock
market, balance of payment, the strengthening of rupiah, and the fairly high economic growth.
Economic growth in quarter II 2010 has increased to 6.17%, which was 2.07% points
higher than economic growth in the same quarter on the previous year. From the demand side,
the highest contribution to economic growth came from domestic demand especially from
investment (gross fixed capital formation) 2.35% and consumption 2.06%. From sector
perspectives, the main contributors were transportation and communication (1.13%); trade,
hotel and restaurants (1.61%) and manufacturing (1.12%).
The indications that the global recovery is proceeding sooner than previously expected
have boosted optimism over Indonesian economic outlook. Such optimism is also supported by
domestic economic resilience that endured the effects of the global crisis. The increased optimism
over Indonesian economic outlook is confirmed by the upgrading of Indonesia»s rating by
international rating agencies in early 2010. These positive conditions support the empirical
result of this research that both monetary policies and fiscal policies give significant impact to
the economic output.
However, there are several challenges for Indonesian economic development. From the
external side, the challenge is primarily related to the impacts of likely strategies developed
countries to unwind the measures adopted by them in response to the global crisis, which
included monetary easing and fiscal expansion, the polarizing trend of global trade and the
Table 15.Sector Contribution to GDP Growth 2010
Tabel 16. Demand Side Contribution toGDP Growth 2010
Q1 Q2
Agriculture 0.42 0.43Mining and Quarrying 0.25 0.31
Manufacturing 0.97 1.12Electricity, Gas and Water Supply 0.06 0.04
Construction 0.45 0.45Trade, Hotels and Restaurant 1.55 1.61Transportation and Communication 1.02 1.13Financial, Rental and Business Services 0.52 0.58
Services 0.44 0.50GDP 5.69 6.17
S e c t o r 2 0 1 0Q1 Q2
2 0 1 0 Demand
Consumption 1.65 2.06
Gross Fixed Capital Formation 2.82 2.35
Net Export 1.18 0.42
Export 7.85 6.02
Import 6.67 5.60
Statistic Discrepancy 0.04 1.34
GDP 5.69 6.17
Sumber: Bank Indonesia
Sumber: Bank Indonesia
210 Bulletin of Monetary Economics and Banking, October 2011
large imbalances in global economic performance. From the domestic side, challenges are
related to several issues that could disrupt the effectiveness of monetary policy, such as excess
bank liquidity, the dominance of short-term inflows in the structure of capital inflows, the
potential asset price bubble, a shallow financial market and numerous structural problems in
the real sector.
V. CONCLUSION
Indonesia has been affected by sudden stop of capital inflow into emerging market
countries and declining on global economies growth following global financial crisis. Their first
and second round effects on macroeconomic indicators were identified. The first lesson from
the recent crisis is that Indonesia as an emerging country clearly demonstrated the effectiveness
of timely monetary, fiscal and financial sector policies which helped Indonesia to recover from
the economic crisis. Indonesia and most Asian countries had experienced the two financial
crises during the last ten years. The first Asian crisis episode occurred in 1997, led to introduce
significant reforms both policy reforms and institutional reforms in the financial sector, However,
in the second crisis ten years later known as global financial crisis occurred in 2008, the reforms
introduced can be categorized as soft reform compared to those in the first Asian crisis.
The second lesson is that the closer cooperation and coordination among the policy
makers is very important in identifying and dealing with challenges posed by a global crisis. As
the Central Bank authority, Bank Indonesia had implemented an accommodative monetary
policy in order support a moderate growth with a relatively low inflation. The policy rate
commenced sliding on December 2008 with the intention to decrease bank»s lending rates.
Some unconventional monetary policy had also been taken to address liquidity issues. On the
fiscal side, the government responded to keep domestic demand by several fiscal stimulus and
trade policies. There were also coordination between Ministry of Finance, Central bank and
other institutions in order to maintain financial market and macroeconomic stability.
The policy measures taken during the crisis had been formulated in a timely manner with
the ultimate objective of sustainable economic growth while maintaining macroeconomic stability
in Indonesia. Based on error correction models, we concluded that in the short run the changes
in real GDP is significantly affected by changes in real money supply in the previous three
quarter and real fiscal expenditures. This indicated that the impact of fiscal policy to GDP is
relatively faster than monetary policy.
As conclusion of this research, policy implications that should be concerned in the future
are as follows:
211Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
1. The cooperation and coordination among the policy makers and the timely responses are
very important in tackling the crisis. Thus, in addressing the crisis, monetary policy could not
stand alone but requires coordination with fiscal policy and other sectoral policies.
2. An effective conventional monetary policy in normal times may become less effective in a
crisis because of the high degree of uncertainty particularly with pressure from external
circumstances. Thus, unconventional monetary policy indeed is necessary as timely policy
response.
3. Regarding to fiscal policy, more timely disbursement of government expenditure is important
to increase the effectiveness of this policy to stimulate economic output.
212 Bulletin of Monetary Economics and Banking, October 2011
Arifin, Syamsul, (1998), ≈Effectiveness of Interest Rate Policy for Rupiah Stability during Crisis,
Bank Indonesia Buletin Ekonomi Moneter dan Perbankan, December 1998. (Available on
Bahasa).
Bank Indonesia, Indonesian Economic Outlook 2009-2014, ≈Global Financial Crisis and the
Impact on Indonesian Economy∆
Enders, Walter, (2004), ≈Applied Econometric Time Series∆, John Wiley∆& Sons, Inc, USA
Elmendorf, Douglas and Furman Jason, (2008),∆≈If, When, How: A Primer on Fiscal Stimulus∆,
Bookings Institution, The Hamilton Project Strategy Paper
Kurniati Y, et all, (2008), ≈Sensitivity of Mainstay Export Commodities to Slowing Trading Partner
Growth and Changing International Prices∆, Bank Indonesia Research Notes: October.
(Available on Bahasa)
Kurniati Y and Meily Ika Permata, (2009), ≈Transmission of External Shock to Indonesian
Economy∆.Bank Indonesia Working Paper.(Available on Bahasa).
Santoso et all, (2009), ≈Impact of Contagion Risk on the Indonesian Capital Market∆. Bank
Indonesia Financial Stability Report No 12: March.
Simorangkir, Iskandar and JustinaAdamanti, (2010), ≈The Role of Fiscal Stimulus and Monetary
Easing in Indonesian Economy during Global Financial Crisis: Financial Computable General
Equilibrium Approach∆, Presented at Call for Papers - EcoMod2010, Istanbul, July 7-10,
2010.
Yehoue, Etienne B et all, (2009), ≈ Unconventional Central Bank Measures for Emerging
Economies∆. IMF Working Paper No. 09/226
Yudo, Teguhet all, (2009),
≈The Impact Of The Global Financial Crisis On Indonesia»s Economy∆. Centre for Strategic and
International Studies (CSIS)
Bank Indonesia (2008) Economic Report On Indonesia. Bank Indonesia Annual Publication
Bank Indonesia (2009) Economic Report On Indonesia. Bank Indonesia Annual Publication
REFERENCES
213Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
APPENDIX
Note : * calculated by HP Filter Method using annual Real GDP data in Billion RpSource : IFS, CEIC, Bank Indonesia and staff estimates
69.22
4.64
19.30
70.50
2,745.83
6.3%
BB -
5.60%
10,852
Indicator
Bank Loan to Deposit Ratio (LDR)
Non Performing Loan %
Bank Capital Adequacy Ratio
ICRG
Stock Market Index (last position)
GDP Growth (yoy)
S & P Rating
Inflation
Output Gap*
1996 1997
109.26
11.82
70.00
637.43
7.8%
5.12%
149,293
Table A. 1 Key Indicators : Measuring Vulnerability of Economy to External Shocks √ 1997/98 and 2008/09 Recession
432.04
118.01
54.74
85.26
7.10
7.70
10.49
147.51
141.18
Ω
27.49
113.69
2.8
3.3
-5.5
-
27%
13%
20%
-
-
2.4%
34.1%
32.7%
Ω
6.4%
26.3%
0.7%
0.8%
-1.3%
Indicator
GDP (Current)
Exports of Goods and Services
Foreign Currency Reserves
Imports of Goods and Services
Average Monthly Imports
Months of Imports Covered
Balance on Current Account
Total Government Debt
Foreign Debt
Composition of External Liability
Short term
Long term
Debt Service Payment (foreign)
Primary Balance
Fiscal Deficit
1996 1997
US $ Billions % of GDP US $ Billions % of GDP
227.37
50.19
17.82
44.24
3.69
4.83
-7.80
Ω
110.17
Ω
22.03
88.14
2.82
4.55
1.73
Ω
22%
8%
19%
-
-
-3%
Ω
48%
Ω
9.7%
38.8%
1.24%
2.00%
0.76%
214 Bulletin of Monetary Economics and Banking, October 2011
Table A. 2 Similarity and Difference between 1997/1998Crisis and 2008 Global Crisis in Indonesia
Similiarity
Both crisis were the consequence of theglobal economy, because of the economicand financial interdependence amongcountries;The impact of the crisis led to falling valueof the rupiah against foreign currencies;The impact of the crisis will affect theeconomic sectors which resulted in lossesfor the community.
Difference
1998 crisis was multidimensional witheconomic crises, political, social, ideologi-cal, defense and security , meanwhileglobal crisis is tend to caused by financialand economic crisis;1998 crisis started from currency crisis inthe Bath-Thailand while the global crisisstarted from the breakdown of Sub-PrimeMortgage in the United States;1998 economic crisis led to on societyanarchism action while the global crisis didnot;1998 crisis led to the demanding forchange of leadership, while global crisis didnot;The focus of monetary policy in 1998 crisiswas tightening, meanwhile in global crisiswas loosening.
215Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
Table A. 3Monetary Policies Taken During Financial Crisis 2008 - 2009
Policies
Policy rate increased gradually to 9.25% onSeptember 2008
Policy Rate (BI Rate) increased to 9.5%(October andNovember 2008, decrease to 9.25% (December2008) and then decreased gradually to 6.75% inJuly 2009Lowering Reserve Requirement for Rupiah currencyfrom 9.1% to 7.5% consist of 5% primary reserve(cash reserve) and 2.5% secondary reserve (23October 2008)Lowering Reserve Requirement for foreign currencyfrom 3% to 1%. (23 October 2008)
Objectives
To contain prevent inflationary pressure such assecond round effect of the fuel price hike and foodprices on other goods.To sustain business momentum amid the globaleconomic slowdown while safeguarding macro-economic stability
To provide more rupiah liquidity in to the bankingsystem
To increase USD liquidity availability to be used bybanks in their transactions with customers.
Conventional Monetary Policy
Policies Objectives
Unconventional Monetary Policy
Narrowing the interest rate corridor for Standing DepositFacility (Fasbi) to BI Rate - 200 bps (from BI Rate - 300bps and maintain Lending Facility (Repo) at BI Rate +300 bps (4 September 2008)Narrowing the interest rate corridor for Standing DepositFacility (Fasbi) to BI Rate - 100 bps and Lending Facility(Repo) to BI Rate + 100 bps (16 September 2008)
Expanding time period of Fine Tuning Operation from14 days to 3 month (23 September 2008)Amendment of Regulation regarding the Liquidity Facilityfor Commercial Banks (18 November 2008).
Open Standing Facility(repo) of 2-14 day tenure (9December 2008)Regulation regarding a Liquidity Facility for Rural Banks(BPR) (10 December 2008)
Narrowing the interest rate corridor for Standing DepositFacility (Fasbi) to BI Rate - 50 bps and Lending Facility(Repo) to BI Rate + 50 bps (4 December 2008)Open 1 month window repo (FTE) (17 April 2009)
To reduce excessive volatility in the interbank moneymarket
To reduce excessive pressure in the interbank moneymarket and maintain sufficient liquidity in bankingindustry sufficiently.
To provide a wider flexibility for liquidity managementin interbank money marketTo smooth the operation of the payment systemsupported by high-value, liquid collateral
To provide wider access to banks by offering fundingwith a longer time horizon than the inter-day fundingfacility
To allow banks suffering from insufficient liquidity toremain solvent and avoid systemic impacts
To facilitate the longer-term bank liquidity requirement
To provide an equal opportunity for rural banks tomake use of this funding facility if a short-termliquidity shortfall is experiencedTo resolve the issue of segmentation in the interbankmoney market
To facil itate the longer-term bank liquidityrequirement.
216 Bulletin of Monetary Economics and Banking, October 2011
Table A. 4Policies Taken to Address Confidence Issues and Asset Price Burst
Joint Policies
Executing Government Bond buyback andpreparing a state-owned enterprise equitybuyback program.
Allowing alternative security evaluation techniquesuch as discounted cash flows beside marked tomarket value (Joint press release- Bank Indonesia,Bapepam,-LK and Accounting Association).Allowing commercial bank to switch bondportfolio from trading and available for salecategories to held to maturity category.
Maintaining a sufficient level of foreign reserves.
Putting some restrictions on short-selling on thecapital market. Limiting purchases of foreigncurrency without underlying transaction toUS$100.000 to curb speculation.
Banning trading on banks» structured product/derivative products that provide chances for bankcustomers to purchase foreign currencies includingdual currency deposits that are callable forward.
Objective
To reduce the high risk perception in Indonesianfinancial portfolio which can distort the monetarypolicy transmission mechanism, The Minister ofFinance has bought back IDR41 billion (US$3.89million) worth of Government Bond using thegovernment fund in the central bank account.
To provide market confidence for the governmentbond particularly when no market prices areavailable.To minimize the impact of Indonesian Financialturbulence by providing an opportunity to thecommercial bank to arrange portfolio categories.
To support the rupiah and focused more onpreventing too volatile movement of the rupiah.
To reduce high risk perception
To reduce speculation and exchange rate volatilityexpectation.
217Relative Effectiveness of Indonesian Policy Choices During Financial Crisis
Table A. 5Foreign Exchange Policy
Policies
Abolishing the limit of daily balance position ofshort term foreign loan (13 October 2008)
Foreign Exchange Provision for DomesticCorporation through Banks (15 October 2008)
The extension of FX Swap tenor from a maximumof 7 days to a maximum of 1 month (15 October2008)
Regulation governing the purchase of ForeignExchange by Banks (13 November 2008).
Amendment to Bank Indonesia Regulation onConcerning Derivative Transaction (prohibition ofstructured product transaction) (16 December2008)
Coordination with Other Central Banks, such as :- Signing of a Bilateral Currency Swap
Arrangement (BCSA) between Bank Indonesiaand People»s Bank of China (23rd March 2009)
- Signing of the agreement on an increase in themaximum amount of the Bilateral SwapArrangements between Japan and Indonesiaunder the Chiang Mai Initiative (6 April 2009)
Objectives
To decrease pressures in USD purchase due totransfer of rupiah account to foreign currencyaccount by foreign customers.
To enhance assurance in fulfilling foreign currencydemand by domestic companies
To fulfill the temporary demand for USD currencyand in order to provide sufficient adjustment timefor banks/market players before actually adjustingtheir portfolio composition
To support the balance of supply and demandcondition of foreign exchange in the domesticmarketTo moderate excessive pressure on rupiah exchangerateTo mitigate foreign currency purchase for speculativepurposesTo support banks» prudential actions through KnowYour Customer Principle (KYC).
To minimize speculative foreign currency transaction
To improve trade and direct investment betweenboth countriesTo assist in providing short-term liquidity for financialmarket stabilization and help Indonesia address tightinternational liquidity.
218 Bulletin of Monetary Economics and Banking, October 2011
This page is intentionally left blank
219
WRITING GUIDANCE
1. The paper should be original and should not violate any copyrights. The submitted paper
should have never been published or not being submitted to other publisher. The copyright
of the published paper is retained to the author.
2. The Bulletin of Monetary Economics and Banking provide a financial incentive between IDR
1,000,000 to IDR 3,000,000.
3. The paper should be sent in 2 formats (1) Microsoft Word (*.doc) and (b) portable digital file
(*.PDF). The 2 files should be sent to the following mail address:
[email protected] and Cc to: [email protected]
You may save your files in CD and send it to the following Editorial Address:
BULLETIN OF MONETARY ECONOMICS AND BANKING
Directorate of Economic Research and Monetary Policy-Bank Indonesia
Building Syafruddin Prawiranegara, 20th Floor
JI. M. H. Thamrin No.2 Central Jakarta, Indonesia
Ph. +62-21-3818202, Fax. +62-21- 3800394
4.4.4.4.4. To avoid missing fonts or other compatibility issue, any special characters or mathematicalTo avoid missing fonts or other compatibility issue, any special characters or mathematicalTo avoid missing fonts or other compatibility issue, any special characters or mathematicalTo avoid missing fonts or other compatibility issue, any special characters or mathematicalTo avoid missing fonts or other compatibility issue, any special characters or mathematical
expression (equations, symbol, matrix, etc.) must be written using Microsoft Equation.expression (equations, symbol, matrix, etc.) must be written using Microsoft Equation.expression (equations, symbol, matrix, etc.) must be written using Microsoft Equation.expression (equations, symbol, matrix, etc.) must be written using Microsoft Equation.expression (equations, symbol, matrix, etc.) must be written using Microsoft Equation.
5. The submitted paper should contain (i) an abstract of maximum one page A4, (ii) keywords
and (iii) JEL classification code. See the JEL code at http://www.aeaweb.org/journal/
jel_class_system.html.
6. The paper must contain the followings:
The background, the aim of the paper and its distinction to previous study
Theory and review of literatures
Methodology (quantitative methodology is preferred)
Result and analysis
Policy and further study implication
7. The citation should be in footnote and not in endnote.
8. The reference must obey the following rule:
220 Bulletin of Monetary Economics and Banking, October 2010
a. Book:
John E. HankeJohn E. HankeJohn E. HankeJohn E. HankeJohn E. Hanke and Arthur G. ReitschArthur G. ReitschArthur G. ReitschArthur G. ReitschArthur G. Reitsch, (1940), Business Forecasting, Prentice-Hall, New
Jersey.
b. Article in journal:
Rangazas, PeterRangazas, PeterRangazas, PeterRangazas, PeterRangazas, Peter. ≈Schooling and Economic Growth: A King-Rebelo Experiment with
Human Capital∆, Journal of Monetary Economics, October 2000, 46(2), page. 397-416.
c. Article in book edited by other people:
Frankel, Jeffrey AFrankel, Jeffrey AFrankel, Jeffrey AFrankel, Jeffrey AFrankel, Jeffrey A. and Rose, Andrew KRose, Andrew KRose, Andrew KRose, Andrew KRose, Andrew K. ≈Empirical Research on Nominal Exchange Rates∆,
in Gene Grossman and Kenneth Rogoff, eds.,∆Handbook of International Economics.
Amsterdam: North-Holland, 1995, page. 397-416.
d. Working papers:
Kremer, MichaelKremer, MichaelKremer, MichaelKremer, MichaelKremer, Michael and Chen, DanielChen, DanielChen, DanielChen, DanielChen, Daniel. ≈Income Distribution Dynamics with Endogenous
Fertility∆. National Bureau of Economic Research (Cambridge, MA) Working Paper
No.7530, 2000.
e. Mimeo or unpublished work:
Knowles, JohnKnowles, JohnKnowles, JohnKnowles, JohnKnowles, John. ≈Can Parental Decision Explain U.S. Income Inequality?∆, Mimeo,
University of Pennsylvania, 1999.
f. Article from web or other electronic form:
Summers, RobertSummers, RobertSummers, RobertSummers, RobertSummers, Robert and Heston, Alan WHeston, Alan WHeston, Alan WHeston, Alan WHeston, Alan W. ≈Penn World Table, Version 5.6∆
http://pwt.econ.unpenn.edu/, 1997.
g. Article in newspaper, magazine or equal periodicals:
Begley, SharonBegley, SharonBegley, SharonBegley, SharonBegley, Sharon. ≈Killed by Kindness∆, Newsweek, April 12, 1993, page. 50-56.
9. The paper should be submitted along with curriculum vitae complete with mail address and
phone number.