bulletin of monetary economics and banking · peter abdullah, pakasa bary, rio khasananda, rahmat...

BULLETIN OF MONETARY ECONOMICS AND BANKINGCentral Banking Research Department

Bank Indonesia

PatronBoard of Governors

Board of Editor

Prof. Dr. Anwar NasutionProf. Dr. Miranda S. Goeltom

Prof. Dr. InsukindroProf. Dr. Iwan Jaya Azis

Prof. Iftekhar HasanProf. Dr. Masaaki Komatsu

Dr. M. SyamsuddinDr. Perry Warjiyo

Dr. Iskandar Simorangkir Dr. Solikin M. JuhroDr. Haris Munandar

Dr. M. Edhie PurnawanDr. Burhanuddin Abdullah

Dr. Andi M. Alfian Parewangi

Editorial ChairmanDr. Perry Warjiyo

Managing EditorDr. Darsono

Dr. Siti AstiyahDr. Andi M. Alfian Parewangi

SecretariatIr. Triatmo Doriyanto, M.S

Nurhemi, S.E., M.ATri Subandoro, S.E

This bulletin is published by Bank Indonesia, Central Banking Research Department. Contents and results research in the writings in this bulletin entirely the responsibility of the authors and not an official view of Bank Indonesia.

We invite all parties to write in this bulletin paper delivered in the form files to Central Banking Research Department, Bank Indonesia, Tower Sjafruddin Prawiranegara Floor 21; Jl. M.H. Thamrin No. 2, Central Jakarta, email: [email protected].

The Bulletin is published quarterly in April, July, October and January.

Quarterly Outlook on Monetary, Banking, and Payment System In Indonesia:

Quarter II, 2016

TM. Arief Machmud, Syachman Perdymer, Muslimin Anwar,

Nurkholisoh Ibnu Aman, Tri Kurnia Ayu K,

Anggita Cinditya Mutiara K, Illinia Ayudhia Riyadi

Competition and Leader-Follower Interactions: Panel Estimates on Indonesian Banking

Peter Abdullah, Pakasa Bary, Rio Khasananda, Rahmat Eldhie Sya’banni

Underground Economy In Indonesia

Sri Juli Asdiyanti Samuda

Determinant of Microcredit Repayment

Farida Hermanto Siregar, Nunung Nuryartono, Eka Intan KP

The Islamic Banking and The Economic Integration In ASEAN

Solihin, Noer Azam Achsani, Imam T. Saptono

BULLETIN of moNETary EcoNomIcsaNd BaNkINg

Volume 19, Number 1, July 2016

21

55

1

39

77

This page intentionally left blank

1Quarterly Outlook On Monetary, Banking, and Payment System In Indonesia: Quarter II, 2016

QUARTERLY OUTLOOK ON MONETARY,BANKING, AND PAYMENT SYSTEM IN INDONESIA:

QUARTER II, 2016

TM. Arief Machmud, Syachman Perdymer, Muslimin Anwar, Nurkholisoh Ibnu Aman, Tri Kurnia Ayu K,

Anggita Cinditya Mutiara K, Illinia Ayudhia Riyadi1

1 Authors are researcher on Monetary and Economic Policy Department (DKEM). TM_Arief Machmud ([email protected]); Syachman Perdymer ([email protected]); Muslimin AAnwar ([email protected]); Nurkholisoh Ibnu Aman ([email protected]); Tri Kurnia Ayu K ([email protected]); Anggita Cinditya Mutiara K ([email protected]); Illinia Ayudhia Riyadi ([email protected]).

The growth of Indonesian economy on Quarter II, 2016 increased with a well-maintained financial

system and macroeconomic stability. Though the growth was not uniform across sectors, the aggregate

growth has increased during this quarter, supported by domestic demand, fiscal stimulus, along with

monetary policy ease. On the other hand, the macroeconomic stability was well preserved as reflected on

inflation within the band target, a better current account deficit, and relatively stable Rupiah’s rate. This

stable macroeconomic condition enabled the monetary authority to ease their policy. In the future, the

policy coordination between the fiscal and the monetary authority is required, particularly on accelerating

the implementation of structural reform, to support a sustainable economic growth.

Abstract

Keywords: Macroeconomy, Monetary, Economic Outlook

JEL Classification: C53, E66, F01, F41

2 Bulletin of Monetary Economics and banking, Volume 19, Number 1, July 2016

I. GLOBAL DEVELOPMENT

Global economic growth is predicted to remain strong. Despite improvements due to increased consumption and improvement in the labor sector, the US economy in Q2/2016 grew well below expectations as investment slowed. US economic developments are still overshadowed by uncertainty, so the Fed Fund Rate increase (FFR) in 2016 is expected to be conducted only once in 2016. Meanwhile, the European economy is expected to grow moderately, overshadowed by post-Brexit uncertainty. Similarly, China’s economy is expected to grow limited because public investment has not been able to provide a boost to the private sector that still faces overcapacity and high corporate debt. On the other hand, in commodity markets, world oil prices are rising, although still low. Prices of some Indonesian export commodities have also improved, such as CPO, coal, and tin.

The US economy in the second quarter of 2016 grew below expectations as investment slowed. US Gross Domestic Product (GDP) in the second quarter of 2016 grew by 1.2%, below nowcasts (1.8-2.9%) and growth last year (Graph 1). US economic growth that is below the forecast is influenced by slowing investment, both residential and nonresidential investment (Graph 2). This condition occurs, among others, due to increased uncertainty and contraction of inventory. Nevertheless, the US economy improved from the previous quarter due to increased consumption. US consumption in the second quarter of 2016 rose significantly, driven mainly by consumer goods. The increase in US consumption is reflected in the highest growth in PCE consumption in 2016, which is the highest since 2014, at 4.2% (SAAR). In addition, the labor sector also improved, reflected in higher nonfarm payrolls than the historical average of the past 5 years accompanied by an increase in nominal wages.

US economic developments are still overshadowed by uncertainty, so the Fed Fund Rate (FFR) rise in 2016 is expected to be made only once in 2016. High uncertainty, among other things, is driven by financial market volatility post Brexit.

Graph 1.Contribution to US GDP Growth

Graph 2.US Investment Growth

��

��

��

��

��

�

�

�

�

�

�

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

�

��

��

��

��

��


Meanwhile, the European economy is expected to grow moderately, overshadowed by post-Brexit uncertainty. Consumption tends to be weak in line with declining retail sales and reduced consumption credit growth (Graph 3). From the sectoral point of view, the manufacturing, manufacturing and construction sectors are indicated to be slowing, despite strengthening in Q1 / 2016. On the other hand, CPI in June 2016 announced at 0.1% (yoy) or 0.2% (MoM) By the decline in energy prices that were not as deep as before. Meanwhile, post-Brexit uncertainty continues and affects the confidence of market participants. This is reflected in the decline of Sentix Investor Confidence to 1.72 (previously 9.9) and Zew Survey Expectation which dropped to -14.7 (previously 20.2), the lowest since 2012.

China’s economy is expected to grow limited. This is because public investment has not been able to provide a boost to the private sector that still faces overcapacity and high corporate debt. Deseleration of private investment continues even though interest rates are kept low. The weakness in private expenditures is compensated by public sector expenditures (Graph 4). Meanwhile, the rebalancing of the economy into consumption-led driven is slow, along with retail sales that are still around the lowest level.

Graph 3. European Retail Sales and Market Retail PMI

Graph 4.Public and Private Investement of China

In commodities, world oil prices are rising although still low (Graph 5). The supply disruptions of some countries and the decline in US production are still driving up prices. Oil production disruptions are mostly from Libya, Nigeria and Canada. The upsurge has accelerated the rise in world oil prices over the past few months. The decline in US oil production is due to low investment levels by 2015, while US drilling depletion rates are relatively high (2-3 years). Meanwhile, prices of some Indonesian export commodities also improved, such as CPO, coal, and tin. CPO prices rose due to lower production, reflected in Malaysian CPO inventories that continued to fall below historical levels. China’s coal production cuts are bigger than the decline in world demand, helping to boost coal prices. In addition, tin prices rose due to lower inventories.

��

��

��

��

�

�

�

�

�

��

��

�

�

��

��

��

��

��

��

��

�

�

�

��

��

��

��


Graph 5.Balance of Supply/Demand and Brent Oil Price

II. MACRO ECONOMIC DYNAMICS OF INDONESIA

2.1. Economic Growth

Indonesia’s economic growth increased in the second quarter of 2016, although not evenly spatially or sectorally. Economic growth in the second quarter of 2016 reached 5.18% (yoy), higher than the previous quarter of 4.91% (yoy). Rising economic performance in Q2 / 2016 was driven by increased domestic demand, particularly government consumption and investment and household consumption. Fiscal stimulus and loose monetary policy began to give impetus to government consumption and private consumption (Table 1).

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� �

��

��

��

��

��

��

��

��

� �

��

��

��

��

��

��

��

��

� �

��


Government consumption is increasing, in line with the ongoing fiscal stimulus. Government consumption rose from 2.94% (yoy) in the first quarter of 2016 to 6.28% (yoy). The increase is influenced by the acceleration of government spending, with significant increases in personnel expenditures and goods expenditures. In addition to the continued acceleration, the increase in government consumption growth also occurred due to the base effect due to the change in the new nomenclature handled towards the end of second quarter 2015.

In addition to government consumption, household consumption has boosted economic growth in Q2/2016. This was reflected in the growth in household consumption which rose from 4.94% (yoy) in the first quarter of 2016 to 5.04% (yoy). Increased household consumption occurs in food and non-food groups. The fiscal multiplier effects and accommodative monetary policy began to provide impetus to household consumption. Strong household consumption is supported by a number of consumption indicators that show positive developments. Retail sales rose from the improvement of sales of all commodity groups, with the highest growth in the information and communications equipment group (Graph 6). In line with the positive retail sales, car sales continued in Q2/2016. This corresponds to a seasonal pattern ahead of Idul Fitri celebrations. In addition, the Consumer Confidence Index (IKK) in the second quarter of 2016 also showed improvement.

Investment growth slowed, amid accelerated Government capital expenditures. Investment grew 5.06% (yoy) in the second quarter of 2016, lower than the previous quarter which grew 5.57% (yoy). The slowdown was driven mainly by slowing construction investment, due to weak private investment interest. Meanwhile, government capital expenditures related to infrastructure projects recorded significant increases. The decline in building investment is reflected in the declining return of cement (Graph 7). Meanwhile, although not strong, non-construction investment has grown positively (2.02%, yoy) compared to the contraction in

Graph 6.Retail Sales

Graph 7.Cement Sales

� ��

��

��

��

�

��

��

��

�

��

��

��

��

��

�

��

��

�

�

��

��

��

��

��

��

�

�

�

�

�

��

��

��

��

��


previous quarter (-0.28%, yoy). Improvements in non-construction investments were driven mainly by high growth in Cultivated Bioligical Resources (CBR) and improved investment in machinery and equipment and vehicles, although still not solid.

Externally, export performance showed improvement, although still contracted, among others supported by exports of some commodities that began to improve. Exports in the second quarter of 2016 recorded a contraction of 2.73% (yoy), better than the previous quarter’s contraction of 3.53% (yoy). The improvement of non-oil and gas exports is underpinned by improved exports of agricultural products and manufactured products. Agricultural exports in June 2016 grew better than the previous month, buoyed by improved exports of shrimp and fish, spices and tea. In June 2016, manufacturing exports also improved driven by exports of TPT, processed wood, and chemical products. The performance improvement of manufacturing exports is in line with the positive growth of Indonesian exports to the United States in June 2016, most of which are exports of manufactured products.

In line with rising domestic demand, import contraction improved in Q2 / 2016. The contraction in imports improved in Q2 / 2016 to 3.01% (yoy) from 5.08% (yoy) in Q1 2016. The retention of import contraction was mainly supported by import growth Raw materials and consumer goods. Raw material import growth continues to increase driven by the growth of raw materials for industrial mamin. Meanwhile, in line with weak private investment until Q2 / 2016, imports of capital goods continued to contract in June 2016, despite improving from the previous month.

From the sectoral perspective, the economic recovery is sustained by the financial and agricultural services sectors. Financial services increased due to the widening Net Interest Margin (NIM) due to the spread of lending rates and increased deposit rates. With the relatively stable operating operating cost (BOPO) trend, the increase in NIM has boosted the performance of the financial services sector. In addition to the financial services sector, the agricultural sector becomes a driving force of the domestic economy. The improvement in the agricultural sector was mainly driven by the growth in the food crops subsector due to the shifting of high crops to the second quarter of 2016 since 2015.

Spatially, economic growth in Q2/2016 was driven by increased growth in Java and Sumatra, while economic growth in Kalimantan and KTI was weak (Picture 1). Acceleration in Sumatra is driven by improved performance of agriculture sector, trade sector, hotel and restaurant (PHR), and building sector. Meanwhile, the acceleration of economic growth in the Java region stems from the increasing performance of financial services and buildings. On the other hand, economic growth in Kalimantan and KTI slowed with considerable contraction in East Kalimantan and Papua. The economic slowdown in the KTI region is influenced by the still contraction of mining, while the economic slowdown in Kalimantan is affected by the slowing down of all economic sectors, except for financial services.


2.2. Indonesia’s Balance of Payments

The current account deficit in Q2/2016 declined, driven by a surplus in non-oil and gas trade balance. The current account deficit decreased from 4.8 billion US dollars (2.2% of GDP) in the first quarter of 2016 to 4.7 billion US dollars (2.0% GDP) in the second quarter of 2016 (Graph 8). The decline was supported by the increase in non-oil/gas trade surplus due to higher non-oil/gas exports from higher non-oil and gas imports. The performance of non-oil and gas exports was mainly supported by an increase in exports of manufactured products, such as textiles and textile products, vehicles and parts, as well as mechanical machinery and equipment. Meanwhile, the increase in non-oil and gas imports was mainly supported by rising raw material imports. On the other hand, oil and gas trade balance deficits are widening, in line with rising world oil prices. In addition, the service account deficit also increased following a seasonally adjusted surplus in the balance of travel services during the quarter under review.

PDRB ≥ 7.0% 5.0% ≤ PDRB < 6.0% 4.0% ≤ PDRB < 5.0% PDRB < 0%6.0% ≤ PDRB < 7.0% 0% ≤ PDRB < 4.0%

Aceh3.5

SUMUT5.7

RIAU2.4

KEP. BABEL3.7

DKI JAKARTA5.9 JATENG

5.7

SULTENG15.5

KALTIM-1.3

KALBAR4.2

SULUT6.1

MALUT5.6

PAPBAR3.4

PAPUA-5.91

BALI6.5 NTT

5.3

KEP. RIAU5.4

LAMPUNG5.2

BENGKULU5.4

BANTEN5.2

SULSEL8.1

SULBAR4.6

JABAR5.9 JATIM

5.6NTB9.9

KALTENG5.7

KALSEL4

GORONTALO5.4

MALUKU6.5

SULTRA6.8

SUMSEL5.1

JAMBI3.6

KALIMANTAN

National : 5.18%Q1 : 4.91%

I III III IV2015 2016

2.01.4

0.4

1.5 1.4

II

1.1

SUMATERA


3.5 3.0 3.14.6 4.2

II

4.5JAWA

I I IIII III IV2015 2016

5.3 5.25.5

5.9

5.35.7

KTI


6.59.4 8.9 8.6

6.3

II

5.9

DIY5.6

SUMBAR5.8

Picture 1. Map of Regional Economic Growth, Quarter II 2016


Graph 8.Current Account

Meanwhile, capital and financial account surpluses increased, buoyed by positive investor perceptions of the domestic economic outlook and easing of uncertainty in global financial markets. The capital and financial account surplus in Q2/2016 reached 7.4 billion US dollars, higher than the previous quarter’s surplus of US $ 4.6 billion, mainly supported by portfolio investment capital inflows. Net capital portfolio investment inflows increased significantly to US $ 8.4 billion in Q2 2016, largely supported by the government’s global bond issuance and net inflows from foreign investors buying in the stock market and the rupiah SBN market. In addition, the direct investment surplus also recorded increased to 3.0 billion US dollars from 2.7 billion US dollars in the first quarter 2016, in line with the positive outlook for the domestic economy.

Overall, Indonesia’s balance of payments in Q2/2016 recorded a surplus, buoyed by a decrease in current account deficit and an increase in capital and financial account surpluses. Indonesia’s balance of payments surplus was 2.2 billion US dollars, after a deficit of 0.3 billion US dollars in the previous quarter (Graph 9). This development shows an improved external balance of the economy and contributes to the maintenance of macroeconomic stability.

The development of Indonesia’s balance of payments in turn strengthens foreign exchange reserves. Foreign exchange reserves increased from 107.5 billion dollars at the end of the first quarter of 2016 to 109.8 billion US dollars at the end of second quarter 2016 (Graph 10). The amount of foreign exchange reserves is sufficient to finance the need for payment of imports and government foreign debt for 8.0 months and is above the international standard of adequacy.

��

��

��

��

��

��

��

�

�

��

��

��

��

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��


Graph 9.Indonesia’s Balance of Payments

Graph 10.The Development of Foreign Exchange Reserves

2.3. Rupiah Exchange Rate

The stability of the rupiah was maintained. During Q1/2016, the rupiah exchange rate, point to point (ptp), rose 3.96% and reached the level of Rp13,260 per US dollar (Graph 11). The strengthening of the rupiah in Q1/2016 was driven by continued foreign capital inflows in line with optimism over the outlook for the domestic economy and the maintenance of external risk factors. The strengthening of the rupiah in the first quarter of 2016 was driven by domestic and external factors. From the domestic side, the continued strengthening of the rupiah is supported by positive perceptions of the domestic economy due to the maintenance of macroeconomic stability and optimism for future economic growth. This is in line with the decline in the BI Rate and government policy packages to improve the investment climate, and accelerate the implementation of infrastructure projects. In addition, the strengthening of the rupiah was also supported by domestic export-oriented corporate foreign exchange supply. From the external side, the strengthening of the rupiah is driven by the easing of risks in global financial markets related to the rise in FFR and the continued easing of monetary policy in some developed countries. The movement of the rupiah is accompanied by maintained volatility. In the first quarter of 2016, rupiah exchange rate volatility recorded a decline and was relatively lower than that of some peers. This is in line with the strengthening of the rupiah exchange rate that occurred gradually since February 2016 (Graph 12).

��

��

�

��

��

�

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

�

��

��

��

��


Graph 11.Rupiah Exchange Rate

Graph 12.Regional Exchange Rate

2.4. Inflation

Inflation remained under control within the 2016 inflation target range of 4 + 1%. In the second quarter of 2016, the Consumer Price Index (CPI) recorded inflation of 0.44% (qtq) or 3.45% (yoy) lower than the previous quarter of 0.62% (qtq) or 4.45% Yoy). Lower CPI inflation in Q2/ 2016 stems from volatile foods (VF) and core group (Graph 13).

Core inflation was quite low. Quarterly, core inflation in Q2/2016 was recorded at 0.72% (qtq), lower than core inflation in the preceding quarter of 0.80% (qtq). The low core inflation is in line with the limited domestic demand, the strengthening of the rupiah exchange rate, and controlled expectations of inflation. In addition, global commodity prices tend to be lower than the previous quarter, especially CPO and maize. Sources of core inflationary pressure in the second quarter were sugar and gold jewelry.

Inflation expectations at the retail and consumer level are still showing a downward trend. In the next 3 months, inflation expectations at consumer and retailer levels show decline as demand for post-Eid al-Fitr falls. The decline also occurred on inflation expectations 6 months to come at the retailer level. Nevertheless, inflation expectations for the next 6 months have increased in the consumer level, along with seasonal factors such as Christmas and year-end holidays (Graphs 14 and 15).

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� �

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��


Graph 13.Inflationary Progress

Graph 14.Inflation Expectation Retailer

Graph 15.Consumer Inflation Expectation

Inflation of volatile foods is maintained. Quarterly, the volatile foods category recorded inflation of 0.98% (qtq) or 8.12% (yoy), lower than the volatile foods inflation in Q1 / 2016 at 2.47% (qtq) or 9.15% (yoy). The lower volatile foods inflation in Q2 / 2016 was driven by the harvest of rice and horticultural crops during the second quarter of 2016. The inflation was mainly due to increased commodities of chicken, carrot, chicken eggs, cooking oil, garlic and beef. Increased volatile foods inflation can further be detained by the deflation of rice commodities, along with the high harvest and deflation of horticultural commodities (red pepper, cayenne pepper and onion), along with the commodity harvest.

Quarterly, administered prices (AP) in Q2 / 2016 recorded lower deflation. The administered prices group in Q2 / 2016 recorded a deflation of 0.73% (qtq), lower than the previous quarter’s deflation of 1.64% (qtq). The lower deflation in administered prices was driven mainly by

� ��

��

�

�

�

�

�

��

��

��

��

��

�

�

�

�

��

��

��

��

��

��

��

� � � � ��

��

��

��

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

�

�

��

��

� � � � ��

��


increases in air transport rates and inter-city transportation tariffs, in line with the high demand ahead of Idul Fitri. The inflation of AP groups was held back by the downward trend in gasoline prices, electricity rates and inner city transport. The decline was driven by government policies that reduced the price of general fuel Pertamax, Pertamax Plus, Pertamina Dex and Pertalite in mid-May 2016 with an average decrease of Rp200 per liter.

III. MONETARY DEVELOPMENT, BANKING AND PAYMENT SYSTEM

3.1. Monetary

The loose monetary policy stance continues, followed by a reduction in interest rates on the money market. The stance of monetary easing continued into Q2/2016, reflected in the BI Rate cut by 25 bps to 6.50% followed by the decline in the Deposit Facility (DF) rate to 4.50% and the Lending Facility (LF) to 7.00%. The decline was followed by interest rates on both the O / N tenor and longer tenors.

The condition of liquidity in the money market remains intact. The overnight interbank rates fell from 5.26% in the first quarter of 2016 to 4.88% in the second quarter. In addition to the O/N tenor, declines also occur in longer tenors, in line with the declining demand for liquidity. In the second quarter of 2016, liquidity conditions in the money market remain intact. This is reflected in the average volume of overnight interbank money market rose to Rp8.06 trillion from Rp7.11 trillion. On the other hand, the average spread on overnight inter-market P/A interest rates increased from 15 bps in the first quarter of 2016 to 23 bps in Q2 2016. This increase was due to the temporary shock at the end of June 2016 in line with the seasonal pattern of Ramadhan. However, the condition returned to normal after Lebaran holiday, along with the return of currency to banking.

Bank deposit interest rates fell, responding to the loosening of monetary policy. Compared to the first quarter of 2016, the weighted average deposit rates in the second quarter of 2016 fell by 43 bps to 7.14%. Thus, on a year-to-date basis (ytd), the weighted average of deposit interest rates in Q2 / 2016 had fallen by 80 bps. Decrease in deposit interest rate occurs in the tenor of 1 to 12 months. The biggest decrease occurred in the short 3-month tenor which decreased by 75 bps (qtq) to 7.00% followed by 6-month tenor which decreased by 56 bps (qtq) to 7.75%. Meanwhile, the 24-month long tenor recorded an increase of 4 bps to 9.16%. Movements in long-term deposit rates tend to be more rigid due to their relatively longer tenure maturities, resulting in a response to slower rate cuts.

Bank lending rates in the second quarter of 2016 recorded decline compared to the previous quarter. Compared to the first quarter of 2016, the weighted average lending rate in the second quarter of 2016 fell by 32 bps to 12.38% in line with the decline in the BI Rate and deposit rates. Year-to-date (ytd), the weighted average loan interest rate in Q2/2016 fell by 45 bps, slower than the decline in the weighted average of deposit rates. The decline in


loan interest rates occurred in all types of loans, with the largest decrease in interest rates on the type of working capital credit (KMK) which fell 46 bps to 11.82% (Graph 16). Late decline in the weighted average loan interest rates amidst monetary policy easing stance is influenced by rising credit risk factors (NPLs). The spread between deposit rates and lending rates in Q2/ 2016 widened to 524 bps from the previous quarter 513 bps (Graph 17).

Economic liquidity growth (M2) increased. In Q2 II 2016, M2 was recorded at 8.7% (yoy), up from 7.4% (yoy) in the previous quarter. The increase in M2 growth was driven by both quasi money and M1. The increase in quasi money growth was mainly driven by rupiah deposits and Rupiah savings. M1 growth in second quarter 2016 was recorded at 13.94% (yoy), up from the first quarter of 2016 at 11.18% (yoy). The increase in M1 growth in Q2/2016 was driven by the increase in currency currency (currency outside bank), which was influenced by the seasonal factor of Lebaran.

Graph 16.Loan Interest Rate: KMK, KI and KK

Graph 17.Spread of Interest of Banking

��

��

��

��

� ��

� ��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� �

��

��

��

��

��

��

��

��

��

� � � �

��

��

��

��

3.2. Bank Industry

Financial system stability (SSK) remains stable and is supported by the resilience of the banking system. The stability condition of the financial system is supported by adequate banking liquidity and strong banking capital. Conditions The stability of the financial system in the future is still maintained to support the intermediation process which is expected to grow higher.

Loan growth in Q2/2016 is limited. Credit growth stood at 8.9% (yoy), up from 8.7% (yoy) in the preceding quarter. Credit growth in Q2/2016 was driven by an increase in consumption credit (KK). In the meantime, investment credit (KI) and working capital credit (KMK) remained slow, although at the end of second quarter 2016 KMK started to show positive growth (Graph


18). By sector, credit in the second quarter of 2016 in all sectors grew positively except the mining sector in line with weak demand for credit in the sector.

Third party funds growth (DPK) in Q2/2016 was 5.9% (yoy), down from the previous quarter’s growth of 6.5% (yoy) (Graph 19). The slowing growth in third party funds in Q2/ 2016 primarily stems from slowing growth in deposits and demand deposits. Slowing growth in deposits, among others related to the transfer to other financial instruments, while the decline in demand growth associated with the government’s financial behavior and payment of THR. Meanwhile, savings increased significantly to 16.3%, increasing the current account, saving account (CASA) ratio to 54.5%.

Graph 18.Credit Growth by Use

Graph 19.Growth of DPK

��

��

��

��

��

��

��

�

�

�

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

�

��

��

��

� �

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

The condition of the banking sector is still well maintained amidst limited credit growth. At the end of Q2/2016, capital adequacy was still adequate with the Capital Adequacy Ratio (CAR) at 22.3%, well above the minimum requirement of 8% (Table 2). In line with the credit slowdown, credit risk (NPL) in Q2/2016 was at 3.1% (gross) or 1.5% (net). In terms of liquidity, banking liquidity in Q2/2016 was sufficient, as reflected in the ratio of Liquid Equipment to Third Party Funds at 20.3%.


3.3. Stock Markets and State Securities Markets

The domestic stock market during Q2/2016 showed steady improvement, driven by various domestic and global positive factors. The performance of the JCI in Q2/2016 reached 5,016.65 (30 June 2016), up by 3.5% (qtq) (Graph 20). From the domestic side, the improvement of the Composite Index was driven by positive perceptions related to the maintenance of macro stability and legalization of Tax Forgiveness Law. From a global perspective, JCI improvements are underpinned by positive sentiment related to the limited impact of Brexit and the expected delay in FFR increases. The performance of JCI movement is still better than the regional stock exchange (Malaysia, Singapore and Thailand). JCI growth was above Malaysia (-3.7%), Singapore (0.0%), and Thailand (2.6%).

In line with the stock market, the SBN market showed a positive performance. Improved SBN market conditions are characterized by declining yield on government securities across all tenors. Overall yield fell 27bps to 7.46% in the second quarter of 2016 from 7.73% in the first quarter of 2016. Short, medium and long term yields decreased by 25 bps, 24 bps and 39 bps to 7, 12%, 7.47% and 7.89% respectively. Meanwhile, the benchmark 10-year yield fell by 22 bps to 7.45% from 7.67. The improvement is driven by global and domestic positive factors, which are relatively similar to the positive factors driving the improvement of the JCI. Amid declining yield on government securities, nonresident investors posted a net buy of Rp37.90 trillion in Q2 / 2016, down from the preceding quarter of Rp47.53 trillion (Graph 21).

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

�

�

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��


3.4. Non-Bank Financing

Non-bank economic financing increased significantly. Total financing during Q2 / 2016 through initial issue, rights issue, corporate bonds, medium term notes, promissory notes and other financial institutions increased to Rp83.1 trillion from Rp24.2 trillion in Q1 2016 (Table 3). The increase was mainly driven by rising stock issuance and bond issuance. The increase in non-bank financing is a financing alternative for the private sector amid still limited bank loan financing.

Graph 20. JCI and Global Stock Indeks,Quarter II 2016 (qtq)

Graph 21. Yield SBN and Net Selling/Buying Foreign, Quarterly

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

�

�

�

�

�

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��


3.5. Payment System Development

The development of rupiah money management in general is in line with the development of the domestic economy, especially from the household consumption sector. The position of UYD at the end of the second quarter of 2016 was recorded at Rp642.0 trillion, growing 26.7% (yoy), or 26.2% (qtq) (Graph 22). The increase in UYD shows the still high role of cash as a means of payment in national economic activities, especially household consumption. Household consumption is growing supported by the impact of accommodative monetary policy. In addition, the increase in government consumption as a result of the government’s fiscal expansion during Q2/2016 also became one of the factors driving the growth of UYD in Q2/2016.

Graph 22.UYD Developments (yoy)

Bank Indonesia is committed to providing publicly circulated money, which is the original Rupiah currency that meets the requirements to be circulated based on quality standards stipulated by Bank Indonesia. During the second quarter of 2016, Bank Indonesia destroyed Unscrupulous Money (UTLE) in various denominations, especially Rp 1.4 billion of Rupiah paper, or Rp49.9 trillion. The number of UTLE destruction is higher than the same period in the previous year which stood at 1.2 billion shares or Rp33.4 trillion. Increasing the number of pieces and nominal destruction of UTLE is a consequence of setting a higher standard of money quality.

The implementation of the payment system during Q2 / 2016 is safe, smooth and well maintained. The condition is in line with the Bank Indonesia System - Real Time Gross Settlement (BI-RTGS System), Bank Indonesia - Scripless Securities Settlement System (BI-SSSS) Generation II and the National Banking Clearing System of Bank Indonesia (SKNBI) Generasi II. The volume of payment system transactions organized by BI was recorded at 33,875.40 thousand transactions, up 9.71% compared to the previous quarter of 30,877.25 thousand transactions. The increase in transaction volume occurred in all payment systems held by BI covering BI-RTGS, SI-SSSS and

��

� � � � � � � � � � � � � � � � � ��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��


SKNBI, which increased by 6.10% (qtq), 16.76% (qtq), and 9.87 % (Qtq) (Table 4). Meanwhile, transaction value decreased by 1.84% (qtq) from Rp40,844.77 trillion to Rp40,094.25 trillion (Table 5). The decline in transaction value was driven by a decline in BI-SSSS transaction value of 9.37% (qtq) or down Rp1,217.76 trillion.

During Q2 / 2016, transactions in the BI-RTGS system increased in both volume and value compared to the previous quarter. The volume of payment system transactions completed through the BI-RTGS system increased by 6.10% (qtq) from 1,436.25 thousand transactions to 1,523.86 thousand transactions. The increase in volume was in line with the increase in transaction value by 1.41% (qtq) to Rp27,117.76 trillion in Q2 2016. In general, the increase in the volume and value of BI-RTGS transactions mainly resulted from increases in community funds transfers (among customers) Both in terms of volume and transaction value, respectively by 7.26% (qtq) and 13.67% (qtq).

The increase in the value of SKNBI transactions is driven by increased credit clearing/fund transfer transactions. This is the result of the implementation of the nominal upper limit of SKNBI fund transfer policy and the nominal lower limit of funds transfer through the BI-RTGS2 System. Through this policy, the volume of transactions through SKNBI increased by 9.87% (qtq) to 32,271.09 thousand transactions in Q2/2016. In line with the increase in volume, transaction value through SKNBI also increased by 8.02% (qtq) than the previous quarter.

Meanwhile, Card-Based Payment Instrument (APMK) transactions in Q2/2016 recorded positive growth. The volume of APMK transactions increased by 7.31% to 1,388.41 million transactions. It shows the increasing frequency of APMK usage in the community, especially ATM Card and/or Debit Card. The value of transactions in the reporting period also recorded an increase of 10.21%. ATM/Debit cards still dominate the volume and value of APMK transactions compared to credit cards with the proportion of respectively 94.58% and 5.42% (volume share) as well as 95.37% and 4.63% (nominal share). The increase in volume and value of APMK transactions during the quarterly report is quite high is suspected to be a cyclical increase considering that during the reporting period there is an Idul Fitri holiday.

The value of electronic money transactions has increased. In the second quarter of 2016, electronic money transactions grew by 26.91% compared to the previous quarter to Rp1,78 trillion. In addition, the transaction volume increased by 22.32% to 169.51 million transactions. The average value of the use of electronic money in a single transaction during the reporting period was Rp10,473.

2 The nominal limit of transactions through SKNBI, which was originally a maximum of Rp500 million to be unlimited, while the nominal limit of funds transfer through the BI-RTGS System which was originally at least Rp100 million increased to Rp500 million.


��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

� ��

��

��

� ��

� ��

��

��

� ��

��

��

� ��

��

��

��

� ��

� ��

��

��

� ��

��

��

��

� ��

� ��

��

��

� ��

��

� ��

��

� ��

��

��

��

� ��

��

��

��

� ��

� ��

��

��

� ��

��

� ��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

� � ��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

� � ��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

IV. ECONOMIC PROSPECTS

Economic growth in 2016 is expected to remain well maintained, although the 2016 growth projection range is slightly lower than previous estimates. The maintenance of economic growth in 2016 is expected to be supported by the loosening of monetary and macropudential policies that have been taken and the acceleration of the implementation of the Government Policy Package. Nevertheless, the saving of government spending in the second half of 2016 has the potential to reduce growth this year. With these developments, economic growth for the whole of 2016 is expected to be in the range of 4.9-5.3% (yoy), slightly lower than the previous range of 5.0 - 5.4% (yoy).


Inflation in 2016 is predicted to be within the inflation target range of 4+1%. On the domestic front, inflationary pressure from the demand side is predicted to be relatively limited. Increased domestic demand is expected to be responded by production capacity. Inflation expectations are also expected to be maintained with policy support and coordination between Bank Indonesia and the Government. On the external side, inflationary pressure is predicted to be maintained. This is supported by the limited estimates of rising international commodity prices, in line with the gradual pace of world economic recovery. Pressure from the exchange rate is also expected to be under control in 2016. In an effort to achieve the inflation target, Bank Indonesia will continue to strengthen coordination with the Central Government and Local Government through TPI and TPID forums to control inflationary pressures, particularly volatile food inflation.

Bank Indonesia will keep a close watch on some external and domestic economic risks. From the global side, the global economic recovery is still weak. Although improving, the US economy in the second quarter of 2016 grew below expectations as investment slowed. The development of the US economy is still overshadowed by uncertainty, so that the Fed Fund Rate (FFR) increase in 2016 is expected to be conducted only once in 2016. From the domestic side, the risk that needs to be noticed is the limited fiscal space due to potential slowing of tax revenues. The proceeds from Tax Amnesty are still running and are expected to contribute in overcoming the limitations of fiscal space. In addition, it is necessary to watch out for lower-than-expected economic growth risks due to a reduction in government spending to keep the state budget deficit at a healthy level.

21Competition and Leader-Follower Interactions: Panel Estimates on Indonesian Banking

COMPETITION AND LEADER-FOLLOWER INTERACTIONS: PANEL ESTIMATES

ON INDONESIAN BANKING

Peter Abdullah1 Pakasa Bary

Rio Khasananda Rahmat Eldhie Sya’banni

This paper discusses banking competition and leader-follower relationship. Banking competition is

investigated using some specification from Monti-Klein model that allows leader-follower (i.e. Stackleberg)

relationship, the possibility of Cournot competition and other form of competition. We use monthly

observations across 119 banks listed in Indonesia using the standard panel fixed effect methodology

to absorb time-invariant unobserved heterogeneity and dynamic panel data to minimize the risks of

endogeneity. The estimation suggests the leader-follower relationship among banks exist both on loan and

deposit markets. The results are mostly consistent across different groups and on full sample estimates,

although are quite different in magnitudes. While leader-follower relationship is dominantly occur in

credit market, there are some evidence of simultaneous appearance of both leader-follower and Cournot

interactions on the deposit market.

Abstract

Keywords: Banking, Monetary Policy

JEL Classification: C70, E50, G21

1 Peter Abdullah is a Senior Economist at Bank Indonesia, while Pakasa Bary, Rio Khasananda and Rahmat Eldhie Sya’banni are Economists. The views expressed on this paper are those of the authors’ and not necessarily represents the views of Bank Indonesia.


I. INTRODUCTION

Our preliminary investigation indicates that the response of deposit interest rate and lending rate towards monetary policy in Indonesia has been asymmetric. The response of deposit rate has been relatively proportional and timely, whereas the response of lending rate has been lagging and relatively rigid. This could be an indication of uncompetitive market (Cottarelli dan Kourelis, 1994; Borio and Fritz, 1995). Moreover, responses towards monetary policy among group of banks with different assets are heterogenous. Therefore, it indicates that some behavior related to individual market power and interaction among banks affect the industry response.

Those problems, which related to competition behavior in banking industry, are likely to affect the monetary policy transmission, particularly through interest rate channel and lending channel. Further, competition is also a relevant factor to increase efficiency (Hafidz dan Astuti, 2013) and to determine interest rates (Muljawan et. al. 2014).

Previous literatures conduct empirical estimates on this issue by applying widely-used competition indicators such as Lerner Index (Amidu and Wolfe, 2013), Hirchman-Herfindahl Index (Adams and Amel, 2011), Panzar and Rosse H-Statistic (Gunji et. al, 2007; Oliviero et. al., 2011) and Boone Indicator (van Leuvensteijn, 2013). This method, particularly by using Lerner, Boone Indicator or H-Statistic can indicate “conduct and performance” effect, whereas HHI only capture market structure effect. However, those approaches are generally explains competition on the whole industry, and does not capture asymmetrical interactions among individual banks.

Ariefianto (2009) suggest estimating specification that derived from Monti-Klein model that allows possible indication of leader-follower or Cournot interactions. This model is originally based on Cournot interactions (see Klein, 1971; Frexias and Rochet, 2008). In addition, Toolsema-Veldman and Schoonbeek (1999) had derived a Stackleberg version of this model. However, Ariefianto (2009) estimates are based on arbitrary choice of samples.

This research will examine competition in banking industry using industrial organization approach, also by improving methods to determine sample selection. Further, this research will model interest rate setting on a bank towards monetary policy using game theory and analyze the implications on monetary policy transmissions. Particularly, this research tries to answer three questions; first, how is the competition behavior on Indonesian banking industry? Second, if the leader(s) exist, how the followers will respond to leader’s decisions? Third, how does the bank competition indirectly affect monetary policy transmissions?

This research is aim to contribute a more interactive indication about competition behavior on banking industry. In addition, this research potentially indicates a recommendation to increase the effectiveness of monetary policy transmission. We limit the analysis on the case of Indonesia.


II. THEORY

Bank competition is essential to be discussed. Competition between banks tends to raise efficiency (Hafidz and Astuti, 2013). Empirically, the degree of competition is one of determining factors of interest rate (Muljawan et. al. 2014). Moreover, a more concentrated banking industry has more rigid interest rate movement (Hannan and Berger, 1991; Neumark and Sharpe, 1992). However, Adams and Amel, (2011) said that the relationship between banking competition and monetary policy response are ambiguous.

Previous empirical studies conduct estimates on this issue by applying a widely-used competition indicators such as Lerner Index (Amidu and Wolfe, 2013), Hirchman-Herfindahl Index (Adams and Amel, 2011) and Panzar and Rosse H-Statistic (Gunji et. al, 2007; Oliviero et. al., 2011). This method, particularly by using Lerner or H-Statistic can indicate “conduct and performance” effect, whereas HHI only capture market structure effect. However, those approaches generally provide insights on the industry as a whole, and cannot capture asymmetrical interactions among individual banks.

Ariefianto (2009) suggest estimating specification that derived from Monti-Klein model that allows possible indication of leader-follower relationship or Cournot interactions. This model is originally based on Cournot interactions (see Klein, 1971; Frexias and Rochet, 2008). In addition, Toolsema-Veldman and Schoonbeek (1999) had derived a Stackleberg version of this model.

We recall a form of Monti-Klein model, by noting the following assumptions:

1. Two bank products, deposit and credit, are homogenous. Bank 1 and bank 2 have linier function of deposit and credit demand:

rL = α - βL ; L= L1 + L2 (1)

(2)rD = a + bD ; D = D1 + D2

2. Banks using deposit and credit quantities as strategic instrument

3. Linier cost function:

(3)

(4)

C1 (L1, D1) = γL,1L1 + γD,1D1

C2 (L2, D2) = γL,2L2 + γD,2D2

4. Interbank money market rate (r) is exogenous variable as it affected by monetary policy of Bank Indonesia.

5. Profit function of bank:

�i = rLLi - rDDi - r(Li - Di) - Ci(Li, Di) (5)


(7)

(8)

Li = LP-α - r - γL,i

2β12

L-i-12

Di = DP-r - a + γD,i

2b12

D-i-12

First partial differentiation of (6) to lending and credit variable derives equation (7) and (8) as follows:

(6)

Combining equations (1) to (5) above, obtained maximization utility function of bank:

Equation (7) and (8) show that the quantity of credit (deposit) of a bank is affected inversely by that the quantity of credit (deposit) of the leader and those of other competitor.

III. METHODOLOGY

3.1. Empirical Specification

For the first analysis, we use a general specification that allows leader-follower (i.e. Stackleberg) relationship, the possibility of Cournot competition and other form of Competition, as in Ariefianto (2009).

From (7) and (8), we have basic understanding that one bank’s lending (deposit) depends on its leader and other bank’s lending (deposit). Combining (7) and (8) with macroeconomics and banking variables, where , the general specification can be represent as follows:

(9)Xit = α + β1Xpt + β2X-i,t + Σk

γkYkt + Σm

μmZmit + uit

is the amount of loan (deposit) of a particular bank at , is the amount of loan (deposit) supplied by the leader, is the amount of loan (deposit) supplied by the rest of followers, is the kth panel-invariant factor, is the mth panel-variant factor, and is the stochastic error. is a constant.

The possible key hypotheses on Equation (9) are as follows: 1. If and , it indicates that the leader and other competitor are significant to affect bank’s quantity of credit/deposit; 2. If and , then the market is indicated to be consistent to leaderfollower relationship (i.e. Stackleberg competition). 3. If , , then the market is indicated


to be consistent with Cournot model, or it can be inferred that there is no leader exist. 4. If and/or , it may indicate other form of competition that is not usually predicted.

Control variables, the “panel variant” or “panel invariant” factors consists of macroeconomic variables (GDP, inflation, exchange rates, and benchmark interest rates), specific internal bank variables (non-performing loans, capital adequacy ratio, etc), and industry variables (HHI). This list of control variables are partly based on Claessens and Laeven (2004) and Angellini dan Certorelli (2003). Further details of these variables are reported on the appendix.

We use two approaches to estimate Equation (9), namely: 1. the standard panel fixed effect methodology to absorb time-invariant unobserved heterogeneity (by replacing with

); 2. dynamic panel data as in Arellano-Bond (1991) to minimize the risks of endogeneity (by replacing with ).

3.2. Grouping of Observations

Observations consist of monthly data of 119 banks listed in Indonesia. The main data source are Bank Indonesia and CEIC. Observations are grouped based on an identification of whether banks compete on a relevant market, where the products across banks have a high degree of interchangeability. This step is crucial as competition is the central issue in this paper. To facilitate this matter, on estimations on credit market we use degree of similarity on credit across economic sectors between each bank and the leader candidate, using the following formula:

(10)Xi = -xik

Σk xik

K

kΣ xIk

Σk xIk

is bank ’s lending on sector , and is bank leader’s lending on the corresponding sector. This formula was modified and inversed from trade complementary index (Michaely, 1996), which is used generally in international trade analysis. The leader candidates are Bank A, Bank B, Bank C, Bank D, and Bank E.

Similarly, for estimations on deposits, we use degree of similarity on deposit spatially (i.e. individual bank’s deposit distribution across provinces) between each bank and its leader candidate, using following representation:

Xi = -xim

Σk xim

M

mΣ xIm

Σk xIm(11)

is bank ’s deposits on province , and is the leader’s quantity of deposits on the corresponding province.


Each group estimates applies to a group of observations that consists of 30 banks with the lowest value of . As we have 5 suspected leaders (Bank A, B, C, D, E), then we have 5 groups (Group A, B, C, D, E, respectively) to estimate. In addition, we conduct estimation using all observations as a robustness test for omitted variable bias regarding omitted competitors across groups. To note, for full-sample estimations, we define the quantity of leaders’ deposit/credit is the sum of those of all leader candidates.

IV. RESULTS AND ANALYSIS

4.1. Credit Market

The results of fixed effect panel regression (Table IV.1-1) shows that four groups of banks, each with Bank A, Bank B, Bank D, and Bank E as the leader respectively, follow Stackleberg competition, without any indication of Cournot competition. The highest follower response to the leader’s choice of credit quantity is indicated on the group of banks with Bank B as the leader (i.e. Group B). Moreover, GDP have positive impact to dependent variables with elasticity close to unity. Non-performing loan (NPL) has negative impact to bank lending.

Table 1.Fixed Effect Panel Estimation for Credit

Group A Group B Group C Group D Group E

VARIABLES Leader Credit Follower Credit CONTROL VARIABLES 1. Macroeconomics 2. Structural 3. Internal bank Constant Observations R-squared Number of bank Hausman Prob

GDP, inflation rate, interbank rate, exchange rate HHI, credit diversification, credit to PDB ratio

NP L, CAR, BOPO

-0.0189**(0.00799)

0.0961(0.204)

-4.483***(1.107)

4200.902

120.008

-0.0557**(0.0232)

0.0209(0.168)

-3.426*(1.601)

5250.862

150.099

-0.0107(0.0307)

-0.110(0.253)

-4.104***(1.236)

5250.914

150.093

-0.0175*(0.00923)

0.215(0.285)

-3.940***(1.196)

5250.900

15

-0.0292*(0.0148)

0.375(0.345)

-3.360(3.170)

5250.720

150.000

Dependent variable: Log(Credit) (…) = Robust standard errors Significance level: *** p<0.01, ** p<0.05, * p<0.


Table 2.Dynamic Panel Estimation for Credit

Group A Group B Group C Group D Group E Full Sample

VARIABLES Leader Credit Follower Credit CONTROL VARIABLES 1. Macroeconomics 2. Structural 3. Internal bank Constant Observations Number of bank Sargan test prob Arellano-Bond AR(2) prob

GDP, inflation rate, inter-bank rate, exchange rate HHI, credit diversification , credit to PDB ratio

NPL, CAR, BOPO

-0.0638**(0.0263)-0.0084

(0.0414)

0.983***(0.00495)

91327

0.331

0.967

-0.0585***(0.0184)-0.122**(0.0521)

0.997***(0.00595)

94527

0.278

0.762

-0.0447**(0.0197)

-0.167***(0.0419)

0.992***(0.0102)

94527

0.212

0.393

-0.134**(0.0680)-0.0834

(0.0942)

0.983***(0.00611)

83226

0.267

0.105

-0.0247**(0.0108)-0.0422

(0.0541)

0.970***(0.0105)

84024

0.514

0.413

-0.208*(0.121)

0.163(0.138)

0.965***(0.00647)

3,521105

0.898

0.110Dependent variable: Log(Credit) (…) = Robust standard errors Significance level: *** p<0.01, ** p<0.05, * p<0.1

Dynamic panel data for credit indicates a leader-follower relationship for all groups estimated, and also consistent for full-sample estimates, that contain all banks in the industry. Bank C and Bank B groups follows Stackleberg and Cournot competition model simultaneously. The highest response to the leader’s decision is indicated on the group of banks with Bank D as the leader. GDP have positive impact with short term elasticity 0.14 – 0.30. Interbank money market rate and NPL variables are negative, tends to be inelastic. The lag dependent parameters are estimated below unity, thus indicate dynamic stability.

4.2. Deposit Market

The fixed effect panel data regression shows that a leader-follower relationship occurs in all groups of observations, except for the groups of observations with Bank C and Bank A as the leader, respectively. Cournot competition model applies for all groups, except for the group of banks with Bank A as the leader. The highest follower response to the leader’s decision occurs on the group of banks with Bank D as the leader. GDP have positive impact to quantity of deposit with elasticity that close to unity, especially in groups of banks with Bank C, Bank A, and Bank D the leader, respectively. Non-performing loan (NPL) has a negative impact to bank’s deposit amount.


In general, dynamic panel regression indicates that Stackleberg and Cournot competition models apply simultaneously for all group estimates and full-sample estimates.

The highest response occurs on the group of banks with Bank D as the leader (i.e. Group D), where 10% increase on Bank D’s deposit will be responded by, on average, 3.6% decrease on follower’s deposit. GDP have positive impact to deposit variables. Herfindahl–Hirschman Index (HHI) has negative impact with small effect. Lag dependent coefficients below 1 indicates dynamic stability of the model.

Table 3Fixed Effect Panel Estimation for Deposit


VARIABLES Leader deposits Follower deposits CONTROL VARIABLES 1. Macroeconomics 2. Structural 3. Internal bank Constant Observations R-squared Number of bank Hausman Prob.

GDP, inflation rate, interbank rate, exchange rate HHI, credit diversification, credit to PDB ratio

NPL, CAR, BOPO

-0.106(0.138)-0.607

(0.426)

17.76**(6.755)

3910.430

120.000

-0.122**(0.0421)-0.887**(0.316)

14.50***(3.571)

4080.558

12

-0.0968(0.103)-0.831*(0.390)

8.653***(2.630)

4080.493

12

-0.359**(0.127)

-1.893***(0.562)

23.26***(3.152)

3800.526

12

-0.166**(0.0639)-0.974*(0.461)

4.441(7.023)

4030.271

140.029

Dependent variable: Log(Deposits) (…) = Robust standard errors Significance level: *** p<0.01, ** p<0.05, * p<0.1

Table 4.Dynamic Panel Estimation for Bank’s Deposit

VARIABLES Leader deposits Follower deposits CONTROL VARIABLES 1. Macroeconomics 2. Structural 3. Internal bank Lag dependent Observations Number of bank Sargan Test ProbArellano Bond AR(2) prob

GDP, inflation rate, interbank k rate, exchange rate HHI, credit diversification, credit to PDB ratio

NPL, CA R, BOPO

-0.466***(0.0541)-0.231***(0.0860)

0.989***(0.00834)

79825

0.228

0.512

-0.269***(0.0419)-0.165*

(0.0851)

0.991***(0.00797)

82225

0.202

0.632

-0.185***(0.0455)

-0.370***(0.116)

0.992***(0.00968)

77425

0.363

0.733

-0.298***(0.0711)

-0.256***(0.0900)

0.999***(0.0122)

84625

0.181

0.515

-0.209***(0.0402)

-0.777***(0.170)

0.937***(0.0250)

75426

0.569

0.394

-0.471***(0.128)

-0.339***(0.0582)

0.988***(0.0171)

3,212106

0.147

0.429Dependent variable: Log (deposits) (…) = Robust standard errors Significance level: *** p<0.01, ** p<0.05, * p<0.



V. CONCLUSIONS

The estimation results suggest a leader and follower relationship among banks on most of the grouped observations, although with some variations in magnitude. Generally, competition between followers is insignificant on credit market, but is significant on deposit market. Leader and follower competition result can be viewed on Table 1 below. Control variables, such as: GDP, inflation rate, interbank rate, exchange rate, HHI, credit diversification, credit to PDB ratio, and operational bank ratios are generally show consistent parameters as expected. All-sample estimates also suggesting similar results, and hence confirming the robustness of selected-sample regressions. For estimates using dynamic panel regressions, all estimations fulfill dynamic stability and well represent the data variations. Moreover, the estimations met the exogeneity assumptions for instrumental variables.

Table 5.Group Summary

Fixed Effect

Arellano-Bond

Arellano-Bond

Competition

Highest Response

Competition

Highest Response

Competition

Stackleberg,Except Group CGroup BStackleberg with Cournotfor Group C and Group BGroup D and Group B

Stackleberg

Stackleberg and Cournot,Except Group C and Group AGroup D

Stackleberg and Cournot

Group DStackleberg and Cournotsimultaneously

Panel Data Credit Deposit

Although our results indicate that leader-follower relationship is generally hold on both credit and deposit market, either in separate groups or full sample, this paper still put restrictive assumptions on the leaders’ behavior. The interactive responses between leaders have yet to be analyzed without any prior restrictions. The suggestion for further research is to improve the empirical specification. For instance, to use credit/deposit similarities as a weight matrix that attached to the leader-follower coefficients to allow multiple leaders at once as well as to allow heterogeneous response towards each leader’s decisions.


REFERENCES

Adams, Robert M. and Dean F. Amel (2011), Market structure and the pass-through of the federal funds rate. Journal of Banking & Finance, 35: 1087–1096.

Amidu, Mohammed and Simon Wolfe (2013), The effect of banking market structure on the lending channel: Evidence from emerging markets. Review of Financial Economics, 22: 146–157.

Angelini, Paolo dan Nicola Cetorelli (2003), The Effects of Regulatory Reform on Competition in the Banking Industry, Journal of Money, Credit and Banking, Vol. 35, No. 5: 663-684.

Arellano, M., and S. Bond. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies, 58: 277–97.

Ariefianto, M.D. (2009). Perilaku Persaingan Industri Perbankan Di Indonesia Pasca Krisis, Analisa Dengan Pendekatan Teori Oligopoli Dan Ekonometrika Panel Data 2002-2008, PhD Dissertation, University of Indonesia.

Borio, C. and Fritz, W. (1995) The response of short-term bank lending rates to policy rates: a cross-country perspective, BIS Working Paper No. 27. 5.

Claessens, Stijn and Luc Laeven (2004), What Drives Bank Competition? Some International Evidence, Journal of Money, Credit and Banking, 36 (3): 563-583.

Cottarelli, C. and Kourelis, A. (1994) Financial structure, bank lending rates and the transmission of monetary policy, IMF Staff Paper, 42: 670–700.

Freixas X. and J.-C. Rochet (2008), The microeconomics of banking, Boston: The MIT Press, 2nd edition.

Gunji , Hiroshi, Kazuki Miura and Yuan Yuan (2009), Bank competition and monetary policy, Japan and the World Economy, 21: 105–115.

Klein, Michael A. (1971), A Theory of the Banking Firm. Journal of Money, Credit and Banking, 3(2): 205-218.

Olivero, María Pía, Yuan Li, and Bang Nam Jeon (2011), Competition in banking and the lending channel: Evidence from bank-level data in Asia and Latin America, Journal of Banking &

Finance 35, 560–571.

Roodman, D. (2009), How to do xtabond2: An introduction to difference and system GMM in Stata. The Stata Journal, 9 (1): 86–136.


Toolsema-Veldman, L. & Schoonbeek, L. 1999, Bank behavior and the interbank rate in an oligopolistic market. Working paper, University of Groningen, SOM research school.

Van Leuvensteijn, Michiel (2013), Impact of bank competition on the interest rate passthrough in the Euro area, Applied Economics, 45: 1359–1380.


APPENDIx

Table 1.List and Notes of Dataset

Main Variables

Credit

Deposit

Credit/deposit of the leader

Credit/deposit of the followers

Macroeconomics

GDP

Inflasi

Interest rate

Exchange rates

Industry

HHI

Credit to GDP ratio

Internal Bank

NPL

CAR

BOPO

Credit to asset ratio

Bank Indonesia

Bank Indonesia

Bank Indonesia

Bank Indonesia, authors’

calcula tion

CEIC

CEIC

CEIC, Bank Indonesia

CEIC

Bank Indonesia, authors’

calculation.

Authors’ calculation

Bank Indonesia

Bank Indonesia

Bank Indonesia

Bank Indonesia,

authors’ calculation

Total credit of individual bank data

Total deposit of individual bank data

5 banks as candidate (Bank A, B, C, D, E)

Industry data – data on a particular bank observed

Nominal, interpolated to monthly using quadratic

match sum

Year-on-year terms

Bank Interbank call money

Using credit/deposit approach

As a proxy to indicate the industry

significance on the economy

Share of credit with quality 3 to 5

Operational cost / revenues

As a proxy to indicate business

diversification

Variables Source Notes


Table 2.Fixed Effect Estimates on Credit Market

VARIABLES Kredit leader Kredit follower CONTROL VARIABLES 1. Macroeconomics PDB Inflation Suku bunga PUAB Real Exchange Rate 2. Struktural HHI Diversifikasi Rasio kredit/PDB 3. Internal bank NPL CAR BOPO Constant Observations R-squared Number of bank

-0.0189**(0.00799)

0.0961(0.204)

1.041***(0.236)

-0.000251(0.00135)-0.00530

(0.00556)-0.182*

(0.0836)

-0.000474(0.000411)

-0.199***(0.0459)

0.0694***(0.0180)

-0.0197**(0.00648)

-9.92e-05***(1.31e-05)

1.91e-05(0.000271)

-4.483***(1.107)

4200.902

12

-0.0557**(0.0232)

0.0209(0.168)

1.124***(0.208)

-0.00365(0.00298)-0.00556(0.0113)

-0.355***(0.0730)

0.000980(0.000663)

-0.238***(0.0688)

0.0685***(0.0209)

-0.0288***(0.00416)0.000721(0.00162)0.000229

(0.000165)-3.435*(1.795)

5250.862

15

-0.0107(0.0307)

-0.110(0.253)

1.185***(0.230)

-0.000910(0.00200)-0.0127*

(0.00652)-0.214***(0.0575)

0.00102**(0.000437)

-0.125***(0.0324)0.107***(0.0190)

-0.00764(0.00826)-0.00197

(0.00362)0.000250**(9.83e-05)

-4.104***(1.236)

5250.914

15

-0.0175*(0.00923)

0.215(0.285)

0.908***(0.280)

0.000500(0.00164)

-0.0116(0.00935)

-0.211(0.131)

-8.79e-05(0.000573)

-0.148***(0.0454)

0.0661***(0.0150)

-0.0280***(0.00151)-0.00249

(0.00376)0.000244

(0.000200)-3.940***

(1.196)525

0.90015

-0.0292*(0.0148)

0.375(0.345)

0.933***(0.199)

0.000488(0.00417)-0.00582(0.0231)-0.250**(0.105)

-0.00108(0.000646)

-0.0771(0.0813)

0.144*(0.0695)

-0.0248**(0.00956)-0.00345

(0.00669)-0.000663

(0.000514)-5.745

(4.673)525

0.72015



Table 3.Dynamic Panel Estimates on Credit Market

VARIABLES Kredit leader Kredit follower CONTROL VARIABLES 1. Makroekonomi PDB Inflasi Suku bunga PUAB Exchange Rate 2. Structural HHI Rasio Credit/PDB Diversifikasi 3. Internal bank BOPO NPL CAR Lag dependen Observations Number of bank

-0.0638**(0.0263)0.00843(0.0414)

0.131**(0.0545)

0.00279*(0.00148)-0.0157**(0.00643)

-0.111(0.0698)

-0.000484*(0.000280)

0.00409*(0.00241)

-0.0131(0.0111)

-0.000100(0.000197)

-0.00170(0.00128)-0.0007***(0.000243)

0.983***(0.00495)

91327

-0.0585***(0.0184)-0.122**(0.0521)

0.205***(0.0632)

0.000804(0.000889)-0.0155***(0.00449)-0.142***(0.0492)

0.00055***(0.000202)

0.000956(0.00156)

0.00464(0.00459)

5.70e-05(0.000125)-0.00558***

(0.00157)0.00113**

(0.000526)0.997***

(0.00595)94527

-0.0447**(0.0197)

-0.167***(0.0419)

0.305***(0.0545)

0.00250***(0.000825)-0.0236***(0.00513)-0.251***(0.0541)

-5.84e-06(0.000188)

0.00460(0.00402)-8.21e-05(0.00584)

-0.000195(0.000152)-0.0064***(0.00208)0.000980

(0.000717)0.992***(0.0102)

94527

-0.134**(0.0680)-0.0834

(0.0942)

0.189***(0.0487)

-0.000652(0.000915)-0.0176***(0.00474)

-0.0198(0.0486)

0.00075***(0.000211)0.00719***(0.00222)0.0120***(0.00380)

-0.000186(0.000155)-0.0070***(0.00139)-6.58e-05

(8.72e-05)0.983***

(0.00611)83226

-0.0247**(0.0108)-0.0422

(0.0541)

0.144**(0.0731)0.00178

(0.00163)-0.0178**(0.00759)

-0.115(0.0812)

-0.000501(0.000321)

0.00609(0.00457)0.0174***(0.00496)

5.03e-05(0.000154)

-0.00219(0.00379)-0.000149

(0.000722)0.970***(0.0105)

84024

-0.609*(0.314)0.642*(0.365)

0.0970(0.0847)

-0.000783(0.00165)-0.0109*

(0.00640)0.0374

(0.0473)

0.00104**(0.000522)0.00516**(0.00248)-0.00155

(0.00350)

-2.89e-05(0.000102)

0.00305(0.00643)

-0.00086***(0.000310)

0.986***(0.00939)

3,319105



Table 4.Fixed Effect Estimates on Deposit Market

VARIABLES DPK leader DPK follower CONTROL VARIABLES 1. Macroeconomics PDB Inflation Suku bunga PUAB Exchange Rate 2. Structural HHI Diversifikasi Rasio kredit/PDB 3. Internal bank NPL CAR BOPO Constant Observations R-squared Number of bank

-0.106(0.138)-0.607

(0.426)

0.222(0.336)

-0.0365***(0.0107)0.179**

(0.0788)0.0575(0.235)

-0.00513**(0.00166)

0.124***(0.0307)0.164***(0.0395)

-0.000784(0.0246)

0.00509***(0.000583)

0.000440(0.000889)

17.76**(6.755)

3910.430

12

-0.122**(0.0421)-0.887**(0.316)

1.078***(0.301)

-0.00629**(0.00277)0.0574***(0.0140)-0.489**(0.158)

-0.00624***(0.00118)0.215***(0.0443)0.121***(0.0273)

0.00186(0.0141)

-0.0159***(0.00454)4.76e-05

(0.000387)14.50***(3.571)

4080.558

12

-0.0968(0.103)-0.831*(0.390)

1.575***(0.429)-0.0103

(0.00690)0.0312

(0.0207)-0.749***

(0.228)

-0.00669***(0.00170)

0.0191(0.0610)0.105***(0.0331)

-0.0185(0.0172)

-0.0187***(0.00414)0.000648

(0.000605)8.653***(2.630)

4080.493

12

-0.359**(0.127)

-1.893***(0.562)

1.580**(0.628)

-0.0170***(0.00314)

0.128***(0.0154)-0.491**(0.204)

-0.00620***(0.00127)0.0762***(0.0216)0.165***(0.0367)

0.0156(0.0154)

-0.000991(0.000869)

0.000629(0.000538)

23.26***(3.152)

3800.526

12

-0.166**(0.0639)-0.974*(0.461)

2.199***(0.523)

-0.0121*(0.00569)-0.00557(0.0331)-0.855*(0.443)

-0.0150***(0.00437)0.0228***(0.00370)

0.418(0.270)

0.00430(0.0314)

-0.0111**(0.00386)-4.96e-05

(0.000194)4.441

(7.023)403

0.27114



Table 5.Dynamic Panel Estimates on Deposit Market

VARIABLES DPK leader DPK follower CONTROL VARIABLES 1. Macroeconomics PDB Inflation Suku bunga PUAB Exchange Rate 2. Structural HHI Rasio kredit/PDB Diversifikasi 3. Internal bank BOPO NPL CAR Lag dependen Observations Number of bank

-0.466***(0.0541)-0.231***(0.0860)

0.857***(0.0963)0.00259

(0.00269)0.0161**

(0.00799)-0.0597

(0.0856)

-0.00352***(0.000461)

0.00301(0.00209)-0.0137***(0.00519)

0.000596(0.000379)

-0.00265(0.00261)0.000204

(0.000298)0.989***

(0.00834)79825

-0.269***(0.0419)-0.165*

(0.0851)

0.394***(0.114)

0.00409*(0.00240)0.0328***

(0.0110)0.419***(0.123)

-0.00266***(0.000462)

0.00318(0.00253)-0.00451

(0.00768)

0.000765**(0.000366)-0.00428***

(0.00158)0.000230

(0.000292)0.991***

(0.00797)82225

-0.185***(0.0455)

-0.370***(0.116)

0.643***(0.153)

0.00252(0.00231)

0.0149*(0.00837)

-0.0256(0.0966)

-0.00206***(0.000495)

0.00245(0.00234)0.00949*(0.00516)

0.000229(0.000344)

-0.00225(0.00150)-0.000343

(0.000355)0.992***

(0.00968)77425

-0.298***(0.0711)

-0.256***(0.0900)

0.652***(0.156)

0.00539**(0.00262)

-0.0178(0.0109)-0.205*(0.119)

-0.000923(0.000591)

0.00505(0.00532)

0.0133(0.0124)

0.000398(0.000506)

-0.00583(0.00532)-0.000152

(0.000396)0.999***(0.0122)

84625

-0.209***(0.0402)

-0.777***(0.170)

0.903***(0.188)

-0.00336(0.00609)0.0602**(0.0236)

0.351*(0.196)

-0.00369***(0.00118)0.0763**(0.0344)-0.00125

(0.00194)

0.000542*(0.000327)

-0.0161(0.0129)

-0.000823(0.00117)0.937***(0.0250)

75426

-0.471***(0.128)

-0.339***(0.0582)

0.776***(0.218)

-0.00392(0.00315)0.0380**(0.0190)

0.240(0.169)

-0.000771(0.00114)

0.00844(0.00541)

-0.00322**(0.00154)

0.00101*(0.000590)

0.0109(0.00853)

0.00778***(0.000680)

0.988***(0.0171)

3,212106



Graph 1 Credit Market: Dynamic Panel Actual Vs Fitted Values

Graph 2 Deposit Market: Dynamic Panel Actual Vs Fitted Values

39Underground Economy in Indonesia

UNDERGROUND ECONOMY IN INDONESIA

Sri Juli Asdiyanti Samuda1

This paper estimates the size of underground economic activity in Indonesia. Underground economy

covers market production of goods and services, legal and illegal, which are sold or purchased illegally.

Using monetary approach, this paper concludes the average size of the underground economy in Indonesia

during 2001-2013 was 8.33 percent of GDP. Consequently, the average size of potential tax loss was

Rp11,172.86 billion or about one percent of GDP.

Abstract

Keywords: Underground Economy, Tax Evasion

JEL Classification: E26, H26, K42

1 Sri Juli is working at BPS (Badan Pusat Statistik) Kabupaten Kepulauan Sula, Indonesia; [email protected]


I. INTRODUCTION

Underground economy is the unseparated element of economic activities in almost all countries. Underground economy is both legal and illegal economic activities that have missed from the calculation of Gross Domestic Product (GDP). Underground economy is also known as unofficial

economy or black economy. It has been an issue worldwide (Scheineider & Enste, 2000). It should be also take into account as the intensifying underground economy leads to bias or underestimated output of Gross Domestic Product (GDP). In the other hand, the intensifying underground economy also creates massive losses through missing potential tax revenue. In general underground economic activities have missed from the surveillance of fiscal authority thus it obliterates the obligation of tax payment from those underground economic actors, leading to massive losses suffered from the country. In the case of Indonesia, some of the underground economic activities that have missed from publications are smuggling commodities abroad such as timber, oil fuel, and even endangered animals protected by the government as well as other cases such illegal goods imported from China without customs and clearance. These phenomenon lead to massive losses for the country due to unpaid import customs and the unstoppable flows of imported goods in Indonesia.

The measurement of underground economy is urgent due to several factors: first, huge amount of tax imposed to the economic agents. The increase in underground economic activities can be considered as individual reactions who perceive burden from the government policy and decide to take the “exit option” rather than “voice option”. In this regard, the increase in underground economic activities is indicating that tax liabilities the agents have to pay is relatively high. The second, the intensifying underground economy leads to inefficiency in decision making process by stakeholders as some of economic measurement indicators such as poverty, labor force, income, and consumption become inaccurate. The third, other considerable effects such as shifting from legal to illegal economic activities by domestic and foreign workers due to underground economy which finally creates competition between them.

During the process, it is still found some obstacles in measuring underground economy as it is conceptually various, moreover those underground economic actors prefer to remain underground. Some to the previous research revealed that accurate data could be utilized as a proxy to measure underground economy such as observing the elasticity of demand on fiat money to tax expense proposed by Vito Tanzi (1980). This approach is also utilized in this research. It actually assumes that underground economic activities occurred as the underground

economic actors preferred to avoid to pay the tax liabilities. This research proposes to measure underground economic activities that has occurred in Indonesia since 2001 – 2013.

The next part of this paper is literature review and other relevant empirical studies. The third part provides data and methodology to measure the size of underground economy while the fourth session provides result and analysis. Conclusion and recommendation are provided in the fifth session of this paper.


II. THEORY

2.1. Definition of Underground Economy

According to Smith (1994) in Faal (2003), underground economy is a production of goods and services, both legally and illegally through the measurement of Gross Domestic Product. (GDP). Illegal activity is actually illegal market where goods and services are produced, traded, and consumed illegally. It is classified as illegal activities as the process are not justified by the law (for instance illegal drugs and prostitution). Moreover legal activities from such underground

activities is production of goods and services that are not publicly traded on purpose due to the following reasons: (i) to avoid payment of tax; (ii) to avoid payment of social protection contribution; (iii) to avoid standard regulation such as required minimum wages, required maximum working hours, required safety standard, etc; (iv) to avoid approval on the required administration procedure.

Moreover according to Smith (1994) in Faal the classification of underground economy

is exhibited as follow:

Table 1.Classification of Underground Economic Activities

ActivitiesType of Transactions

Monetary Transaction Non-Monetary Transaction

Illegal • Trade of stolen goods

• Illegal drugs

• Gambling

• Prostitution

• Money Laundering

• Smuggling

• Fraud

• Barter of illegal drugs

• Money laundering for personal purpose

• Production of illegal drugs for personal purposes

Legal • Unreported Income

• Wages, income, and asset from any activities

that are not reported as legal goods and services

• Non-invoice payment

• Allowances for employee

• Other allowances

• Disclosed Income

• Wages, income, and asset from any activities that

are not disclosed as legal goods and services

• Non-invoice payment

• Allowances for employee

• Other allowances

Moreover Feige (1990) divided underground economic activities into four categories explained as follow:

1. The Illegal economy is the illegal economic activities which contain revenue earned from any economic activities that break the constitution or are not in accordance with the


regulation. Such activities include trade of stolen goods, piracy, and smuggling. Moreover gambling, illegal drugs trade, and narcotics are economic activities that are fully ban by the law.

2. The Unreported Economy is income that are not reported to tax authority in order to avoid any payment of tax.

3. The Unrecorded Economy is income that should have been recorded in the government statistics. As the consequences, gap exists between the recorded revenue or expenditures and the real revenue or expenditures.

4. The Informal Economy is income generated by economic agents informally. Economic agents of these sectors may not have any formal permissions from the authority, employment agreement, or financial loan.

2.2. Measurement Method of Underground Economy

There are several alternative approaches to measure underground economy. From the overall approaches, there are three approaches frequently used to measure underground economy such as direct approach, indirect approach, and modelling approach.

Direct Approach

Direct approach is a micro-based approach through survey on a particular group of sampling with certain sampling method. The survey is designed to identify number of activities classified into underground economic activities. The estimation of underground economy using survey-based direct approach is commonly held in developed countries. There are several strengths of direct approach such as detail information through questions within the survey. Meanwhile the weakness of this approach is that the accuracy of the data highly depends on tendency of respondents to give objective answers. As mentioned earlier, the underground economic actors tend to remain underground.

Indirect Approach

Indirect approach is frequently known as indicator approach as it uses macroeconomic variables to measure underground economy. The most frequent indicators to measure the number and the growth of underground economy are:


2 Smith, J.D (1985): Market motives in the informal economy.

1. Discrepancy between GDP Expenditure and GDP Revenue

This approach is based on statistical discrepancy between GDP measured using expenditure approach and revenue approach. Theoretically, GDP measured using expenditure and income approaches would generate the same value. The gap between those GDP measurements indicates an underground economic activities within the country2.

2. Discrepancy between legal and illegal employment participation rate

This approach is used to measure the gap between legal and illegal employment rate. If the participation rate remain unchanged, but the participation rate in legal sectors diminishes, then it indicates underground economic activities.

3. Monetary Approach (Demand on Fiat Money)

The estimation using monetary approach is one of the most frequent methods to measure underground economic activities. This method is developed by Vito Tanzi (1980) that he proposed it to measure underground economy in the United States. Tanzi defined underground economy as economic activities that are unreported or unrecorded by fiscal authority in order to avoid tax payment. According to Tanzi (1980) tax burden is part of motives in underground economic activities.

In the proposed model, Tanzi estimated underground economic activities using elasticity of demand on fiat money to tax expenditure. This assumption is driven by a perspective that economic agents in underground economy prefer to use cash for their transaction to avoid any control from the government mainly from fiscal authority. The model measures sensitivity of public desire of holding cash to tariff changes or tax burden. The incentive of tax avoidance by using cash in any transactions leads to higher intention of those underground economic

actors to hold higher amount of cash.

Modelling Approach

In this approach, the estimation of underground economic activities is obtained from modelling technique that represent indicators of underground economic activities. The explicit model consists of several factors leading to the increase in underground economic activities.

There are number of prior researches that estimate underground economy in several countries. Asaminew (2010) conducted an estimation on underground economy in Ethiopia and revealed that underground economic activities in Ethiopia had increased during 1977 – 1991 when the country fell into long run civil war. During the periods, the under economy accounted for 41.5% to total GDP and decreased to 30% in 1993 after the war ended. Haque (2013)


also estimated underground economy in Bangladesh during the period of 1973 – 2010. The estimation result shows that during 1973 the underground economy accounted for only about 7% to nominal GDP of Bangladesh, and drastically increased to 62.75% to nominal GDP in 2010.

III. METHODOLOGY

This research uses quarterly secondary data that include period range of 2001 – 2013. The data is accessed from Central Bureau of Statistics (Badan Pusat Statisik/BPS), Ministry of Finance, and Bank Indonesia (BI).

This research run the estimation of underground economy using monetary approach, similar to the one proposed by Tanzi (1980). Some of economic variables to indicate underground

economy is explained as follow:

a. Demand on Fiat Money (C)

It is the number of fiat money, both in the form of paper and metal (coin), circulating in the public. In order to reflect its real value, this research uses real fiat money which is nominal fiat money adjusted to general price level.

b. Inflation (I)

It reflects changes of price in particular group of goods and services consumed by the public, moreover it can also serve as public purchasing power. Theoretically, inflation has positive impact on the demand of fiat money.

c. Tax Burden

The proxy of tax burden is the ratio of tax revenue to nominal GDP. It is in accordance with the definition proposed by Organization for Economic Co-Operation and Development (OECD). It is expected that the tax burden would have positive relationship with money supply.

d. 1-month Deposit Rate

1-month Deposit Rate is assumed as the opportunity cost of holding cash. Theoretically, 1-month Deposit Rate has negative relationship with demand for fiat money as the higher the interest rate the higher the intention of public to save or invest their money.

e. Gross Domestic Product (GDP)

The research uses quarterly current GDP data from 2001 – 2013. GDP is expected to have positive impact on demand for fiat money.

Moreover the equation of demand for fiat money can be mathematically expressed as follow:


where ln C is fiat money used for transaction in an official economy.

Underground economy is measured based on the extent to which economic agents produce and trade goods and services beyond surveillance from fiscal authority so that those economic agents could avoid their obligation to pay tax. This research constructed the following systematical stages to measure underground economy.

(1)

Picture 1.Measurement Stages of the Size of Underground Economic Activities

Estimation of Demand forFiat Money (C)

Calculation of Demand for Fiat MoneyUnderground economy

CUE = C - COE

Calculation of the Velocity of Money Suppy inUnderground economy

VUE = PDB /(M1- CUE)

Calculation of the size of Underground economy

UE = CUE x VUE

IV. RESULT AND ANALYSIS

4.1. Estimation Result and Model Validation

Before estimating the model, the stationary test is required in the first place. The test is applied to all variables included. Stationary test has the same value of means, variance, and auto-variance (on lag variance) at any range of periods the data is utilized. It means stationary data of time-series model is relatively stable. Theoretically, spurious regression is indicated by high


R2, nevertheless it does not mean significant relationship between independent and dependent variables exist.

The research utilizes Augmented Dickey-Fuller Test to estimate the data as exhibited in Table 1. According to the Table 1 it is seen that there some variables that are not stationary at level, thus differencing process is required to identify on which level those remaining variables are stationary.

As explained in the above paragraph the estimated model is actually the elasticity of demand for fiat money to tax expenditure as fiat money is the most preferable tool of transaction for underground economic actor to avoid any detection from fiscal authority. The model will also measure the influence of tax burden on demand for fiat money.

The estimation result exhibited in Table 2 shows that tax burden, 1-month deposit rate, and GDP have significant impact on demand for fiat money, while inflation performs no significant impact.

Table 2.Result of Regression Model

Variables Coefficient Std. Error t-Statistic Prob.

INF

Tax

R

LNPDB

C

R-squared

Adjusted R-squared

F-statistic

Prob(F-statistic)

Durbin-Watson stat

0,015139

0,015137

-0,006631

0,169989

2,751379

0,915849

0,906498

97,95012

0,000000

1,912247

0,016199

0,003341

0,003819

0,036330

0,573883

0,934605

4,530057

-1,736099

4,679079

4,794318

0,3550

0,0000

0,0894

0,0000

0,0000

Source: data processing

The above model needs to be further tested using cointegration test to identify whether or not all variables have long run equilibrium. Technically the test is done to identify the existence linear combination between two or more variables in the model. The research uses Engle Granger method using Augmented Dicky Fuller Test.

From the overall statistical tests, it is seen that residual of the equation is stationary which means there is linear combination between all variables included or it can be said that they are cointegrated. It also explains that the estimation result of the equation would not form a spurious regression.


Model robustness test refers to determination coefficient (R2), t-statistics, and F-statistics. Based on the regression result it is found that determination coefficient (R2) is 0.92. It indicates that 92% of demand for fiat money can be explained by inflation, tax burden, 1-month deposit rate, and Gross Domestic Product.

Meanwhile the probability of F statistics is 0.000 (1% level of significance) indicating that the model consisting of inflation, 1-month deposit rate, tax burden, and GDP are simultaneously significant towards dependent variable. Meanwhile among those four independent variables, it is only inflation that performs insignificant impact on demand for fiat money, while 1-month deposit rate, tax burden, and GDP perform significant impact on demand for money.

Model validation also includes classical assumption tests to ensure that the model is robust. The research conducts multicollinearity test. Through VIF score, it is seen that none of those independent variables that contain higher-than-10 value if VIF indicating that all variables are free from multicollinearity assumption.

Table 3.Tolerance and VIF Scores

VariableCollinearity Statistics

Tolerance VIF

R

PDB

Inf

Tax

0,505

0,377

0,91

0,645

1,982

2,653

1,099

1,551

Source: Data processing

The autocorrelation and heteroscedasticity tests are also included in model robustness test. The statistical value of Durbin-Watson of the initial model is 1.05 indicating that autocorrelation exists in the model. By adding lag period, the value of Durbin-Watson increased to 1.91 indicating that the model is free from autocorrelation problem. In terms of heteroscedasticity, the result of the test, using White method, shows that the value of F-statistics is 1.38493 with the probability of 0.2050 (level of significance by 5%) indicating that there is no heteroscedasticity problem within the model.

4.2. Analysis of Model Estimation Result

The Relationship Between Inflation and Demand for Fiat Money

Demand for real money is basically demand for fiat money in regards with price changes of goods and services which in general perform significant impact on purchasing power of money. When inflation occurs with the same nominal value of the money, then the consequences is


3 Peirre Lemieux, (2007) The Underground economy Cause, Extent, Approach, Montreal Economic Institute Research Papers

that the money can only be used for less-transaction, or in other words there is diminishing purchasing power of money. Therefore, the theory states that the relationship between demand for fiat money and inflation is positive. It mainly due to the increase in inflation would require public with higher amount of money for the same transaction.

It is in accordance with the empirical findings of the research where the coefficient of inflation is positive by 0.0151. Eventhough inflation is not statistically significant, nevertheless it does not mean that inflation does not perform any impact on demand for fiat money. It is mainly due to low impact of inflation indicating that inflation does not directly drive changes in demand for money, still, it requires time to change.

The Relationship Between Tax and Demand for Fiat Money

Variable of tax is the important one to measure size of underground economic activities where it must have statistically significant impact on demand for fiat money. The use of fiat money as transaction tool would ease the underground economic actor for tax evasión. The result shows that variable of tax has positive impact on demand for fiat money. Coefficient of tax is 0.0151 indicating that the increase in tax by 1% (others remained unchanged/ceteris paribus) then demand for fiat money would increase by 1.51%. It can be also interpreted that tax has significant influence on the increase in underground economic activities in which the higher the tax the higher the intensity of underground economic activities.3

The Relationship Between 1-Month-Deposit Rate on Demand for Fiat Money

Interest rate is actually an opportunity cost of holding money where it has negative correlation with demand for money theoretically. It is in accordance with the empirical result where the coefficient of 1-month-deposit rate is negative by 0.007. Therefore 1% increase in 1-month-deposit-rate then demand for money will diminish by 0.7%. According to the theory, the higher the deposit rate the lower the intention to hold cash as public tend to save money in the bank would be higher as well due to higher potential return.

The Relationship Between Gross Domestic Product and Demand for Fiat Money

According to Keynes, income positive correlation with demand for money. In this research, income is proxied by Gross Domestic Product (GDP). This approach is also used by Ebrima Faal (2003) who measured the size of underground economy in Guyana. The use of GDP as proxy of income is mainly due to value added generated by the economic agents in producing goods and services.


According to the empirical result, it is found that income has positive impact on demand for fiat money. Coefficient of GDP by 0.17 indicates that every 1% increase in GDP would lead to the increase in demand for fiat money by 17%.

4.3. Measurement of Underground Economic Size

The estimation result of the model would drive certain level of fiat money, both used in official

economy and underground economy. The estimation result of demand for fiat money with tax minus estimation on fiat money without tax is used to estimate level of fiat money in underground economy. The result is then multiplied by velocity of money to obtain precise size of underground economy in Indonesia during the period of 2001 – 2013 as exhibited by the following Table 4.

Table 4.Estimation of Underground Economy in Indonesia, 2001-2013

Underground economy(in Billion Rupiah)


2001

2002

2003

2004

2005

2006

2007

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

31899.25

29573.92

56849.40

40314.81

33758.58

36216.79

45594.91

43631.31

28158.74

37544.56

57594.81

36615.73

38516.53

49338.56

75939.52

70395.54

44235.53

51890.20

69986.94

63599.30

51828.99

61261.79

91011.78

76500.37

54863.54

81395.83

114195.66

72423.59

104035.46

101441.25

129354.40

102143.43

81833.64

93224.54

118759.42

34622.35

157109.82

80328.31

119214.58

81957.28

134793.25

138138.11

151378.10

124896.16

137148.55

196591.67

212969.48

106301.74

188571.63

140849.84

327739.00

286806.74

94141.25

473.51

431.50

524.86

411.09

328.55

364.88

457.41

130.84

583.93

290.56

424.07

285.51

459.06

464.03

496.61

406.18

440.20

625.09

676.01

336.33

590.78

438.49

986.12

847.73

396.53

293.80

256.26

489.50

229.73

284.80

301.96

374.62

234.75

219.45

297.49

301.76

283.88

192.65

240.06

360.45

327.34

298.44

337.53

440.34

368.51

296.96

344.95

496.62

272.17

283.58

412.46

561.33

345.07

2008

2009

2010

2011

2012

2013


Period PeriodTW TW

Nominal Real (2000) Nominal Real (2000)

Average


Moreover development of underground economy in Indonesia during 2001 – 2013 is exhibited by Graph 1. It can be seen that underground economic activities performed positive trend. It was only the period 2009 where underground economic activities diminished drastically. The highest underground economic activities was in the quarter III 2013.

Table 4.Estimation of Underground Economy in Indonesia, 2001-2013



2001

2002

2003

2004

2005

2006

2007

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

31899.25

29573.92

56849.40

40314.81

33758.58

36216.79

45594.91

43631.31

28158.74

37544.56

57594.81

36615.73

38516.53

49338.56

75939.52

70395.54

44235.53

51890.20

69986.94

63599.30

51828.99

61261.79

91011.78

76500.37

54863.54

81395.83

114195.66

72423.59

104035.46

101441.25

129354.40

102143.43

81833.64

93224.54

118759.42

34622.35

157109.82

80328.31

119214.58

81957.28

134793.25

138138.11

151378.10

124896.16

137148.55

196591.67

212969.48

106301.74

188571.63

140849.84

327739.00

286806.74

94141.25

473.51

431.50

524.86

411.09

328.55

364.88

457.41

130.84

583.93

290.56

424.07

285.51

459.06

464.03

496.61

406.18

440.20

625.09

676.01

336.33

590.78

438.49

986.12

847.73

396.53

293.80

256.26

489.50

229.73

284.80

301.96

374.62

234.75

219.45

297.49

301.76

283.88

192.65

240.06

360.45

327.34

298.44

337.53

440.34

368.51

296.96

344.95

496.62

272.17

283.58

412.46

561.33

345.07

2008

2009

2010

2011

2012

2013


Period PeriodTW TW

Nominal Real (2000) Nominal Real (2000)

Average

Table 4.Estimation of Underground Economy in Indonesia, 2001-2013 Continued

Graph 1. Development of Quarterly Underground Economic Activities in Indonesia 2001-2013

0

50

100

150

200

250

300

350

I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV I II IIIIV

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013


Based on the estimation result of underground economy in Indonesia, it can be predicted amount of potential revenue loss due to underground economic activities. Value of potential tax is obtained from the multiplication of the size of underground economy with the average tariff of tax. The estimation of the average tariff of tax is obtained from total tax revenue divided by tax base, which is GDP in this case. Table 5 exhibits the result of potential tax revenue


calculation. During the period of 2001 – 2013, the average potential tax revenue of Indonesia had lost due to underground economic activities by 11,172.864 billion quarterly. Table 5 also provides ratio of potential tax revenue from underground economic activities to nominal GDP. The highest ratio of potential tax revenue losses to nominal GDP occurred in the quarter IV 2013 by quarterly 2.33% or 55,233.76 billion.

Table 5.Potential Tax Revenue of Underground economy in Indonesia, 2001-2013

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

I

II

III

IV

4309.59

3318.19

6566.11

7220.38

3693.19

3835.36

4814.82

6104.02

3238.25

3900.88

5989.86

5770.64

4317.70

5136.14

8315.38

11368.88

5193.25

5879.16

8258.46

9438.14

5867.04

7124.75

9983.99

11429.16

5629.00

9303.54

13829.09

11348.78

11672.78

12842.46

16803.14

14269.44

8363.40

10534.37

11769.06

4878.29

6259.95

14032.77

11891.72

10679.04

10777.22

18777.20

19081.11

15830.77

11473.43

21365.66

29811.47

14999.95

13898.73

16129.42

28430.04

55233.76

11172.864

1.05

1.05

1.27

1.11

0.64

0.76

0.81

0.34

0.42

0.88

0.71

0.64

0.62

1.03

0.99

0.83

0.58

1.04

1.41

0.72

0.65

0.73

1.20

2.33

0.9992

1.11

0.80

1.54

1.73

0.85

0.85

1.02

1.32

0.65

0.78

1.16

1.15

0.80

0.91

1.40

1.90

0.82

0.88

1.16

1.24

0.75

0.88

1.15

1.31

0.61

0.97

1.34

1.10

Source: Author’s tabulation

Period PeriodQuarter QuarterNominal Potential

Tax Revenue(Billion)

Ratio toGDP

Ratio toGDP

Nominal PotentialTax Revenue

(Billion)

Average


The growth of underground economic activities over 2001 – 2013 can also be calculated by measuring the increasing size of real underground economy. By ignoring price changes over periods, then development of underground economy over 2001 – 2013 is exhibited as follow.

Graph 2. Growth of Underground economyin Indonesia, 2001-2013

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Growth -5.76 -7.82 1.62 28.95 -2.36 13.59 14.88 -30.3 23.59 15.26 13.79 37.81

-40.00

-30.00

-20.00

-10.00

0.00

10.00

20.00

30.00

40.00

50.00

According to the above Graph 2, it can be seen that the growth of underground

economic activities of Indonesia is quite fluctuating over 2001 – 2013. The highest growth of underground economic activities occurred in 2013 by 37.81%. Meanwhile the underground

economy drastically decreased by 30.3% in 2009.

V. CONCLUSION

This paper is an empirical research that measures and analyzes underground economy in Indonesia. Based on the empirical in the previous section, it can be concluded that, first, monetary approach can be utilized to measure underground economy in Indonesia as indicated by significant impact from tax expenditure on changes of demand for fiat money. The second is that the use of monetary approach leads to the conclusion that the average underground

economy is Rp94,141.25 billion or 8.33% to Gross Domestic Product quarterly over 2001 – 2013 in Indonesia. As the consequences, the research delivers the third conclusion which is potential tax losses due to underground economic activities by averagely Rp11,172.86 or 1% to Gross Domestic Product (GDP) quarterly over 2001 – 2013.

Three conclusions above have strong implications that the use of cash for transactions widely enlarge the opportunity of underground economic activities. In the near future, electronic money should be more frequently used as the ultimate solution to reduce underground economic

activities.


REFERENCES

Asaminew, Emerta. “The Underground economy and Tax Evasion in Ethiopia: Implications for Tax”. 2010.

Faal, Ebrima. “Currency Demand, the Underground economy, and Tax Evasion: the Case of Guyana. International Monetary Fund Working Paper. 2003.

Feige, Edgard. “Defining and Estimating Underground and Informal Economies: The New Institutional Economic Approach, World Development, 18, no. 7, pp. 989 – 1002. 1990.

Gujarati, Damodar. “Ekonometrika Dasar”, third edition, Jakarta; Erlangga, 1993.

Haque, Sheikh Touhidul. “Underground economy of Bangladesh: An Econometric Analysis”. Research Study Series No – FDRS 01/2013. 2013.

Lemmiuex, Pierre. “The Underground Economy Causes, Extent, Approach”. Montreal Economic Institute Research Papers. 2007.

Schneider, Friedrich and Enste, D. H., “Shadow Economies: Size, Causes, and Consequences”, Journal of Economic Literature, Vol.38, pp. 77-114. 2000.

Smith, J. D. “Market motive in the Informal Economy”. The Economics of the Shadow Economy. 1985.

Tanzi, Vito. “The Underground Economy in the United States: Annual Estimates, 1930-80, “IMF Staff Papers, Vol. 30 (June), pp. 28.

55Determinant of Microcredit Repayment

DETERMINANT OF MICROCREDIT REPAYMENT

This paper investigate the determinants of microcredit repayment by employing the logistic

regression on micro-business households in Pati, Central Java. The result of this study reveals that loan

repayment is affected significantly by the business lines, spending on food consumption, side job, other

loan sources, collateral, and credit constrained. Interestingly, the result concludes that the loan repayment

are no longer influenced by moral hazard, since the characteristics such as gender, education level, age,

experience do not significantly encourage borrowers to repay. This paper also conforms the important

role of peer-screening process on hindering the credit default.

Abstract

Keywords: Micro-Credit, Repayment, Credit Constrained, Moral Hazard

JEL Classification: C13, G21

1 Farida (corresponding author; [email protected]) is a lecturer at Economic Department, UPI YAI Jakarta. Hermanto Siregar ([email protected])is a professor at Department of Economic, Institut Pertanian Bogor; Nunung Nuryartono ([email protected]) and Eka Intan KP ([email protected]) are lecturers at Department of Economic, Institut Pertanian Bogor.

Farida1

Hermanto SiregarNunung Nuryartono

Eka Intan KP


I. INTRODUCTION

The principal problem for any credit scheme in Indonesia is bad debt. Several programs such as Mass Guidance Program (Bimbingan Masal or Bimas), the Mass Intensification Program (or Inmas), Special Intensification (or Insus), and Loan for Farming Business (Kredit Usaha Tani or KUT) ended with arrears and consequently these programs were forced to shut down. Another scheme, Loan for Food Sustainability Program (Kredit Ketahanan Pangan or KKP) is having difficulty to find an optimum distribution. The difference between KUT and KKP is the source of funding and the responsibility for the risk (Supadi and Sumedi, 2004). KUT is funded by the Liquidity Credit of Bank Indonesia (or KLBI) and the risk is borne by the government meanwhile KKP’s financing and 50% of the risk is borne by the executive bank. The arrears on KUT has accumulated to Rp 5.7 trillion, or 81.4 percent (Voice of Indonesia, 2012). The failure of KUT was basically due to 1) its distribution to the wrong target recipients, 2) the distributing institutions simply served its function to distribute the find, without any tracing follow-up which made the fund tracing difficult to do. This failure raises the question: how was the initial screening process? A rigorous peer-screening would reduce the risk of default. But there is no doubt that the initial screening is plagued by asymmetric information, meaning that not all borrowers’ information might be observed throughoutly (Ofonyelu, et al., 2013).

Learning from previous experience and putting in place improved credit requirement, affordable interest rate and safer credit risk, the government then rolled out the Credit for the Poor (Kredit Usaha Rakyat or KUR) to replace the previous programs. The distribution of KUR has seen a significant increase, indicating an early success. KUR serves as working capital loan or investment credit for feasible but not bankable businesses. The funds are disbursed by banks appointed by the government and the government acts as the insurer. This synergy between government subsidies with commercial banks is expected to increase the volume of available loan and to keep the risk of default in check. The amount of KUR disbursed to micro-enterprises in Indonesia since its deployment in 2007 until October 2014 has shown an extraordinary increase from Rp 17.19 trillion to Rp 171.67 trillion. The average level of non-performance loan (NPL) in 2014 is around 3.2 percent which is smaller than the requirement of Bank Indonesia. Formal credit is more organized and has more motivated, high quality staff compared to the informal one (Onyeagocha, 2012). Although the NPL rate is low, similar to other credit market, certain inherent risks remain such as delayed or overdue repayment. Therefore the aspect of micro-financing risk management should not be overlooked (Setargie, 2013).

KUR was set up in order to reduce the level poverty by improving women empowerment as one of its approaches. However, only 21.93% of women in this study have access to KUR, although women possess lower risk of moral hazard than men (D’Espallier, et al., 2011) thus should have been prioritized to get the loan. Research of (Armendariz and Morduch 2005; Chakravarty, et al., 2013) found that women show better repayment in both patrilineal and matrilineal communities. However this kind of empirical studies is usually for informal loans,


where the dominant customer is female. According to the microcredit summit campaign, more than 70 percent of the informal financial customers are women. Okojie, et al. (2010) estimated that 84.2 percent of the poorest microfinance clients are women. In fact there are many informal loans where the female customers also failed to repay. Research of Godquin (2004) also could not prove that the female borrower have a better repayment, where the coefficient result is positive but not significant. Microfinance program (e.g. KUR microcredit) is addressed to micro-business from various sectors, where the data in Figure 1 shows that the majority is in trading or retail business. The agricultural sector seems to have received little attention based on the small portion of the sector that makes up the credit quota. Agriculture is by many still considered as a business field with a high level of default risk. In his study, Gebeyehu (2013) found that farmers have failed to repay their credit due to social, economic and institutional factors.

KUR is distributed to finance working capital and/or investment in the intention to develop micro-business. However, many banks are not able to control at what extent the loan is used for productive efforts as working capital or for other consumptive purposes, such as family’s health expenditure or children’s school fee. When the credit is deviated to other use, a micro-business would end up having some difficulties to repay the loan. Credit diversion is an example of moral hazard that some clients take to change the purpose of the microcredit from productive intent to non-productive purpose.

This study will also observe which economic sector performs better (and worse) in terms of KUR repayment, and what factors determine the repayment level of KUR. The role of gender in the formal loan refund will be examined as well to see if the risk of moral hazard is still relevant in this study.

The remainder of this paper will review the theoretical and empirical studies related to the microcredit repayment rate. The third section shall review the data and logistic regression used in this paper. The fourth section will review the estimation result and analysis, while the conclusions will be presented in the fifth part to complete the paper.

II. THEORY

The previous loan programs such as KUT, KKP, National Community Empowerment Program (Program Nasional Pemberdayaan Masyarakat or PNPM) are group-based which use the groups to provide peer-screening, peer-monitoring and peer-enforcement all at the same time, meanwhile KUR is an individual-based loan model provided by commercial banks which requires a more objective and applicable requirements. Business prospects, feasibility and collateral give an added value to determine the loan approval. KUR is intended for individual borrowers and is not a group-based loan model.

Lukman et al (2008) in a study on strengthening effort of the microbanking’s role through a group approach in West Sumatra, concluded that the success of this group-based financing


approach is largely determined by socio-cultural characteristics of the community where microfinance institutions are located, so that the success rate varies in relation to the locals’ attitude and behavior in managing a loan. Some ethnic groups show a higher commitment where the level of compliance of the group is high, following the group leader as the referrer. The responsibilities borne by this group customers are linked from one member to another, so that the social control function largely determine the motivation and commitment of the members to repay the credit. Usually the loan repayment is guaranteed by the group leader by a form of joint liability or social collateral.

The already developed communication network at this era seems to reduce the high transaction cost of formal banks to collect the loan repayment. The communication cost is no longer expensive and is deemed more effective than a direct visit to the borrowers. Another factor that is expected to affect the success rate of credit repayment is the home-business owner characteristics and behavior such as gender, level of education, diverted use of credit, experience in using loan and household consumption. These behavioral characteristics can be included as internal factors of the micro-business or simply referred as borrower’s behavior. Setargie (2013) incorporated poor business performance as a factor that influence the delay in credit repayment. Diversion in loan use for non-productive purpose, number of dependents and issues in business ownership are some other factors that may lead to failure in loan repayment. On the other hand education, income, monitoring, sufficient repayment period and availability of other microcredit sources are important factors that may significantly improve the repayment rate of the loan.

Jalaluddin (2002) identified several factors that affect the success rate of loan repayment from small entrepreneurs in economic and non-economic category. The economic factors include: 1) The net income earned by small entrepreneurs both from farming and non-agriculture activities. 2) The number of dependents. The higher the number of family members means more household responsibilities thus lesser capacity to repay the loan. 3) The business scale, which is measured by the amount of capital required to run the business. 4) The frequency and amount of the loan installment. The non-economic factors include the level of education, training frequency, and business fields. In addition, Roslan and Karim (2009) have identified borrower characteristics, company characteristics, and loan characteristics as influencing factors, while Nawai and Shariff (2013) added a financial institution characteristics in the list.

III. METHODOLOGY

Field surveys are conducted to obtain specific information related to micro-enterprises, income and welfare indicators that will be used to supplement the secondary data. Primary data are collected through questionnaires and interviews using structured questions addressed to micro-businesses. The survey location is determination by a multistage purposive sampling. The first phase takes place on provincial level where Central Java is chosen as the largest recipient of


KUR, or about 15.12 percent of the total plafond of KUR in 2012. The second phase is on the regency level and the regency of Pati, as the biggest distributor of KUR, is nominated. The next stage is to select the districts based on their distance to the municipal market where the closest and the farthest districts from the municipal district are chosen. Distance or location are deemed influential in terms of access to market, banking, and information that will determine the success of a business. The two chosen districts are Margorejo as the nearest district with a distance of 4 kilometers and Dukuhseti with a distance of 36 kilometers.

The number of samples taken using the Slovin2 formula is n = N / (1 + N(e)2) where N is the population or the total customer KUR in Pati regency (which is 25,080 people). With a desired/tolerated level of sampling error at 8 percent then 155 samples of micro-business households who received KUR are needed. Based on the KUR distribution proportion, 86 units of microenterprise samples shall be taken from the district of Dukuhseti and 69 more from Margorejo.

The research in this paper will use logistic regression to determine the factors that influence the KUR repayment from micro-enterprises. Another study by Tundui and Tundui (2013), and another one from Mokhtar, et al., (2012) have used a similar approach. A model variant which is still classified in discrete choice models such as the probit model has also been done by Godquin (2004), Vitor (2012), Setargie (2013), Wongnaa and Vitor (2013), and also Gebeyehu, et al. (2013). The dependent variable in logistic regression is a response variable in two or more category, while the explanatory variables can be in categorical or continuous form.

The specification of a logistics model for a chance of occurrence P (Y=1|x) with a series of explanatory variables Xi are as the following:

2 See Umar (2004)

and its multiple logistic regression model:

(1)

(2)

Generally if a variable with a nominal or ordinal scale has k possible values, then a number of k-1 dummy variable will be required, so the logit transformation model shall be:

(3)

where:u : 1,2,3…,kj-1

Dju : kj-1 dummy variablebju : coefficient of dummy variableXj : the j-th independent variable at the level of kj


Estimation of parameters in logistic regression is using the maximum likelihood approach. If the probability of an event is assumed to be independent from with other events, then the probability function shall be:

The principle of MLE procedure is to determine b which value will maximize the joint probability with n times of observations. To facilitate the calculation, we performed natural logarithm transformation and obtained the following log likelihood function:

(4)

(5)

Where ModelUR is an unrestricted models or a complete model with bj ≠ 0. On the other hand, the Wald statistic can be used to test the effect of each individual explanatory variables with W statistics as follows:

(6)

(7)

In line with the research of Roslan and Karim (2009), a series of explanatory variables are used including borrower characteristics, business characteristics, and loan product characteristics as shown in Table 2.

in the condition of the first derivative on α and b, we can gain the estimator b that maximizes the l(b).

The overall model testing can be performed with likelihood ratio test, P, which is scattered following the shape of Chi-square distribution with a degree of independence (k-1) as below:

This statistic of W is spread following the shape of the normal distribution.

The interpretation of coefficients for a binary logistic regression model can be performed using odds ratio. Odds ratio can be defined as the probability of Option-1 among individuals with x = 1 compared to individuals with x = 0 (see, among others Juanda (2009)). The odds ratio is formulated as follows:

(8)


IV. RESULT AND ANALYSIS

KUR is a formal loan with distinctive difference from other types of loans. KUR was first launched in 2007. Initially KUR was distributed through seven executive bank executive: Bank Nasional Indonesia (BNI), Bank Mandiri, Bank Rakyat Indonesia (BRI), Bank Tabungan Nasional (BTN), Bukopin, Syariah Mandiri, and BNI Syariah. To accelerate distribution, the executive banks were assisted by 26 Regional Development Bank (Bank Pembangunan Daerah or BPD) in each region. According to data from the KUR Committee, there are 11.92 million of user in total until October 2014. On average the loan approved for each customer is Rp13.1 million. The

Table 1.Variable description in logit model

Name of Variable Type Description

p(xi)

Borrower characteristics;

District (X1)

Gender (X2)

Age (X3)

Level of education (X4)

Education_1

Education _2

Education _3

Spending on food consumption (X5)

Side job (X6)

Working spouse (X6)

Number of dependents (X7)

Business characteristics;

Distance to bank (X8)

Type of business (X9)

Type of business_1

Type of business _2

Type of business _3

Lifespan of business (X10)

Business obstacle (X11)

Sales (X12)

Working capital (X13)

Credit diversion (X14)

Other loan resource (X15)

Loan characteristics;

Collateral (X16)

Collateral_1

Collateral_2

Unrestricted credit (X17)

Installment period (X18)

Binary

Binary

Binary

Continuous

Binary

Binary

Binary

Continuous

Binary

Binary

Continuous

Continuous

Binary

Binary

Binary

Continuous

Binary

Continuous

Continuous

Binary

Binary

Binary

Binary

Binary

Continuous

KUR repayment (1= on time, 0= late)

Research location (1= Dukuhseti, 0= Margorejo)

Gender of microenterpreneur (1 = male, 0 = female)

Age of microenterpreneur

1 = SD/Elementary school, 0 others

2 = SMP/Middle school, 0 others

3 = SMA/High school, 0 others

Amount of expense on food consumption per month

1 = having side job(s), 0 no side job

1 = working spouse, 0 no working spouse

Number of dependent children in the family

Distance from business location to bank (in km)

1 = trade and retail, 0 others

2 = service, 0 others

3 = fabrication, 0 others

How long have the business been running (in years)

Obstacle faced by the micro-entreprise

(1= capital, 0 marketing)

Sales per week (Rp)

Working capital per week (Rp)

Deviated use of loan, 1 = yes, 0 no

Other loan resource, 1 = yes, 0 limited to KUR

1 = no collateral, 0 others

2 = BPKB (ownership certificate) of motorcycle, 0 others

Is the loan approved as requested (credit constrained),

1= all approved, 0 others

Period of installment (in months)


trade sector dominates the KUR with a portion of Rp 97.2 trillion or 56.62 percent. The second sector is agriculture which absorbs Rp30.1 trillion or 17.53 percent. The rest is distributed in other sectors as shown in Picture 1 below.

Picture 1.Distribution of KUR by Business Sector

17%

3%1%

57%

1%5%

3% 13%

Distribution of KUR, 2014

Agriculture Industry Manufacture Trade

Transportation Rents Public Services Others

Source: Primary data (processed)

Overall, the value of Non-Performance Loan (NPL) of KUR by the executive banks is at 3.2% which is still below 5% (November 2014). It is expected that in the subsequent periods all the bank with an NPL higher than 5% can cut down this value by have a better KUR distribution target. The general characteristics of KUR scheme are shown in Table 2 below;

Table 2.General characteristics of KUR scheme

No. Description Explanation

1.

2.

3.

4.

5.

6.

7.

8.

Purpose

Loan plafond

Interest rate

Duration

Guarantor

Insurance company

Loan resource

Micro business

Providing working capital and investment capital to feasible yet not-bankable micro

business

Rp 20,000,000

Max. 22% per year

3 years for working capital, 5 years for investment

Government; 80% for agriculture, fishery, micro-business, and 70% for the other sectors.

PT Jamkrindo, PT Askrindo, PT Jamkrida Jatim, PT Jamkrida Bali Mandara

100% by executive banks

Max of net asset <= Rp 50 million, excluding land and building, or max sales of

Rp300 million per year, or employee less than 5 person.

Source: KUR-Committee


4.1. Borrower Characteristics

In this study, one dependent variable is a binary between micro-business households with an on-time KUR repayments, and the others with difficulties and delays in repayments. An on-time repayment would mean the borrower has never been late more than twice during the loan repayment period, otherwise the borrower will be considered delayed (more than two times delay). This refers to the dependent variable in the study of (Mokhtar et al., 2012) which classifies a bad repayment if the delay is more than four times. From 155 micro-business respondents, 134 units or 86.45 percent show an on-time repayment, and the remaining 21 units or 13.55 percent are in difficulties for repayment. Women have a 14.29 percent rate of delay. The characteristics of microenterprise borrowers are shown in Table 3.

Table 3.Demographic Characteristics

Description

Delayed repayment (N1=21)

On-time repayment (N2=134 )

Number of respondents (N3=155)

% N3Sub-Total(N4=n1+n2)

% of N2% of N1

Total(n2)

Total(n1)

Demography:

Gender

Male

Female

Age

23 – 37 th

38 – 52 th

53 – 67 th

Level of education

SD

SMP

SMA

D3 or S1

18

3

4

15

2

8

8

4

1

85.71

14.29

19.05

71.43

9.52

38.10

38.10

19.04

4.76

103

31

64

60

10

38

38

54

4

76.87

23.13

47.76

44.78

7.46

28.36

28.36

40.30

2.98

121

34

68

75

12

46

46

58

5

78.07

21.93

43.87

48.39

7.74

29.68

29.68

37.42

3.22

Source: Primary data (processed)

Age is categorized into 3 group where the highest proportion is age between 38-52 years old by 48.39 percent and between 23-37 years old at 43.87 percent. Borrowers from 38-52 years old bracket experience more delays than the others at a rate of 71.43 percent. The highest proportion in education level grouping is the graduates of elementary (SD) and middle schools. Only 5 KUR borrowers or 3.22 percent have a higher education graduates. Among these 5 borrowers, one experienced difficulty in repayment, which makes this group has the highest delay rate percentage of all other demographic groups. The borrower characteristics related to economy situation is shown in Table 4 below.


Around 25 percent of KUR users have a second job and 50% have a working spouse. The number of dependent is defined as the number of children who are still dependent on their parents. In this study, 75 percent of the respondents have less than 3 dependent children. This means that many microentrepreneurs are following Planned Family (Keluarga Berencana) program.

The business characteristics are presented in Table 5. Distance shows how far the business site from a bank. The average distance of all respondents is 4 km, but for the groups experiencing difficulties in repayment, the average distance is at 6 km. The average business lifespan with repayment delay is 6 years, a little longer than those with good record in repayment at an average of 5 years.

Table 4.Economic Characteristics of Borrowers

Description



Total Respondent (N3=155)

% of N3Sub-Total(N4=n1+n2)

% of N2% of N1Jumlah

(n2)Total(n1)

Side job

Yes

No

Working spouse

Ye

No

No. of dependent children

< 3

>= 3

Spending on food consumption

<= Rp 1,5 million

>= Rp 1,5 million

Spending on non-food consumption

<= Rp 2,5 million

>= Rp 2,5 million

2

19

11

10

16

5

19

2

20

1

9.52

90.48

52.8

47.62

76.19

23.81

90.48

9.52

95.24

4.76

38

96

65

69

101

33

122

12

129

5

28.36

71.64

48.51

51.49

75.37

24.63

91.04

8.96

96.27

3.73

40

115

76

79

117

38

141

14

149

6

25.81

74.19

49.03

50.97

75.48

24.52

90.97

9.03

96.13

3.87Source: Primary Data (processed)


Overall, retail and trade is the dominating business sector that profits from KUR, which usually do commerce in daily necessities, food (e.g. noodles and meatballs), and other commerce stalls or kiosks both in traditional market or at home. The service business group includes workshop, salon, food catering and other services. Other business group category covers agriculture, livestock farming and fisheries sectors. This business sector suffers the most from difficulty in loan repayment.

The problem of capital makes more than 65 percent of business obstacles faced by microenterprises. This includes frequently damaged engine, difficulty of raw materials and crop failure. On the other hand, the common marketing problems encountered by microenterprises are lack of economies of scale, or disability to expand or increase production and lack of customers.

Table 5.Business Characteristics

Description



All respondent (N3=155)

% to N3Sub-Total(N4=n1+n2)

% of N2% of N1Total(n2)

Total(n1)

Average distance (km)

Average lifespan (th)

Business type:

Retail/trade

Service

Manufacturing

Others

Business obstacle:

Capital

Marketing

Average sales per week (Rp)

Average working capital

per week (Rp)

Frequency of KUR

Once

More than once

Credit diversion

Yes

No

Other loan resource

Ye

No

Intent for more loan from bank

Yes

No

6

6

4

3

6

8

15

6

5.32 million

4.57 million

9

12

8

13

12

9

17

4

19.05

14.29

28.57

38.09

71.43

28.57

42.86

57.14

38.10

61.90

57.14

42.86

80.95

19.05

4

5

66

27

20

21

88

46

4.41 million

3.61 million

88

46

19

115

27

107

75

59

49.25

20.15

14.93

15.67

65.67

34.33

65.67

34.33

14.18

85.82

20.15

79.85

55.97

44.03

4

5

70

30

26

29

103

52

4.54 million

3.74 million

97

58

27

128

39

116

92

63

45.16

19.36

16.77

18.71

66.45

33.55

62.58

37.42

17.42

82.58

25.16

74.84

59.35

40.65

Source: Primary Data (processed)


The average sales turnover of microbusinesses each week micro-businesses ranges from Rp 4.54 million and the average working capital at Rp 3.74 million. 62.58 percent from the entire microenterprise beneficiaries of KUR is a first-time KUR recipient, while the remaining (37.42 percent) has received KUR more than once or has done suppletion. Suppletion is allowed if the rest of the loan and the new loan do not exceed USD 20 million. KUR is intended for working capital or investment capital, but some borrowers may actually use it for other purposes. Approximately 17.42 percent of the KUR borrowers use the loan for non-profit making purposes, such as school fee payment, daily necessities purchasing or motor-cycle installment. Most household micro-businesses do not rely solely on KUR, but also on other loan sources such as cooperatives, National Community Empowerment Program (Program Nasional Pemberdayaan

Masyarakat or PNPM), neighbors, relatives, moneylenders and others, where around 25.16 percent of micro-business customer has other loan sources. On the question of whether the microenterpreneurs are planning to get another loan from the bank to support the business, about 59.35 percent of the respondent said yes.

Loan characteristics in this study could also mean the loan’s institutional characteristics, as the inherent characteristics of the following factors represent the institutional characteristics and the loan characteristics at the same time. These factors are observed based on the types of collateral submitted to the bank, the time needed for approval completion, the approved portion of the loan and the installment period until the loan is completely repaid. These characteristics can be seen in Table 6.

Table 6.Loan Characteristics

Description



Total respondent (N3=155)

% of N3Sub-Total(N4=n1+n2)

% of N2% of N1Total(n2)

Total(n1)

Collaterals;

1 = N/A

2 = BPKB

3 = others

Duration (days)

Approved amount;

All

Partial

Avg installment period

(months)

3

3

15

13.57

13

8

20.76

14.29

14.29

71.43

61.90

38.10

4

64

66

10.85

121

13

20.60

2.98

47.76

49.25

90.30

9.70

7

67

81

11.22

134

21

20.62

4.5

43.23

52.29

86.45

13.55

Source: Primary Data (processed)

Due to the asymmetric information about prospective borrowers, the bank are imposing the collateral requirement from the customer. The amount of the collateral shall determine the amount of the loan to be provided. Micro KUR is a loan without collateral, however in the reality


the banks still require certain collateral. Despite that, there are about 4.5% of the entire micro KUR borrower who do not provide any collateral. Around 43.23 percent use their motorcycle as collateral. The remaining 52.29 percent provide BPKP of their car or other certificate. It takes around 11.22 days (including the weekend) to process the loan application before the loan can be disbursed to the borrower. The norm is around 3-5 business days if the all the requirements are met. The borrower is normally required to open an account in the respective executive bank to put and withdraw the loan. Not all the loan application is approved: approximately 86.45 percent of KUR application has the entire loan approved, while the remaining 13.55 percent only received part of the required loan. The approval is part of the peer-screening by the financial institutions to its prospective borrowers by restricting the loan plafond. The average length of the installment period is 20.6 months, which is less than two years.

4.2. Model Estimation Result

The estimated factors that determine the loan repayment are given in Table 7 with the likelihood ratio (LR) at 60.35 and degree of freedom at 24, this logistic regression model as a whole can explain or predict household micro entrepreneur’s decisions in terms of KUR repayment.

Table 7.Logit estimation result of factors that influence loan repayment

On-time/delayed repayment Odd ratio Z P>│z│

DistrictGenderAgeEducation_SDEducation_SMPEducation_SMADistance to bankType of business_tradingType of business_serviceType of business_manufactureBusiness lifespanBusiness obstaclesSalesWorking capitalSpending for foodSide jobWorking spouseOther loan resourceNumber of dependent childrenNo collateralMotorcycle BPKB as collateralUnlimited loan requestCredit diversionInstallment periodConstant

2.190.95

0.91 1.840.401.770.897.880.880.601.070.691.000.990.9949.84.570.161.700.035.898.360.410.9449.1

0.78 -0.03 -1.69 0.31-0.43 0.29-0.94 1.23-0.10-1.99 0.72-0.29 2.39-2.38-2.23 2.60 1.60-2.09 1.10-1.99 1.67 2.03-1.02-0.84 0.81

0.4360.976

0.092*0.7550.6670.7740.3450.2170.922

0.047**0.4690.7740.0170.017

0.026**0.009***

0.10*0.03**

0.270.04**0.09*

0.04**0.03**

0.400.41

Log likelihood =-31,3, No of obs = 155, LR chi2(24) = 60,35, Prob>chi2=0,000Pseudo R2 = 0,4907, *** = significance 1%, ** = significance 5%, * = significance 10%


By looking at the p-values (α = 1%) of the logit estimated result above, we can determine the variable that affects positively to the repayment of the KUR credit with a confidence level of 99% is a side job. For the p-values (α = 5%), factors that affect negatively to the KUR repayment is the type of manufacturing business, spending on food, other non-KUR credit source availability, collateral unavailability and credit diversion. The factors that positively affect the repayment of loans with 95% confidence level (α = 5%) is the loan application that gets approved from the beginning. all from scratch. The factors that bring a significant positive loan repayment with p-values (α = 10%) are working spouse and collateral in form of BPKP, meanwhile the factor with negative effect is age.

Type of Business Field

This study categorizes business field into four type: retail/trade, services, manufacturing (production), and agriculture which includes livestock farming and fisheries. The manufacturing business type significantly brings negative effect on the repayment of the loan.

The probability of a manufacturing business type to be not able to repay the loan on time is 0.6 times higher compared with other types of businesses, or in other words, it has a probability of 1.67 times for delayed repayment compared to other types of businesses.

Manufacturing business is vulnerable to changes in raw material prices. Increased fuel price, for example, will directly increase the cost of production, meanwhile the microenterprises cannot immediately raise its product prices, and so reducing its profit margin becomes the only option. This directly results in difficulties for manufacturing microenterprises to repay their loan. This study is in line with Roslan and Karim (2009), which revealed that the odds of default for the economic activities in service/support drop down compared to production activities. Manufacturing business also counts working capital as an influential factor. The higher the cost of production, the higher the required working capital for micro-enterprises. As a result, the business will suffer profit loss if sales volume cannot be improved.

Spending on Food Consumption

Household microenterprises usually spends more on the food consumption than on the non-food consumption. At the significance level of 5% then the spending on food consumption variable with a p-value of 0.026 will have a very significant negative effect on the loan repayment.

This estimation result indicates that the higher expenditure on food, the difficulty for a micro-enterprise to repay the loan is 1.01 times greater the one who spends less on food.


Loan Source

The estimation result shows that the availability of other loan sources significantly gives negative effect on the loan repayment, meaning that the more the installments to be paid to the other loan sources, the more problematic the KUR repayment will become.

In quantitative terms, the value of likelihood ratio for this variable is 0.16. This means that the probability of a delayed repayment of a KUR customer who borrows from other sources is 6.25 times higher compared with a customer who only borrows from KUR.

Collateral

In this study it was found that micro-business KUR borrowers encounter a greatly varied requirements on collateral. Some banks require no collateral, the other demands BPKB of motorcycle or certificate. The results from model estimation shows that the customer with no collateral requirement will face a significant problem in loan repayment.

The probability of a borrower who does not provide collateral to repay the debt is 0.033 times of customers with collateral. In other words, the probability of a delayed repayment is 30.3 times higher compared to customers who provide collateral. Meanwhile, customers who put their motorbike BPKB as collateral would have a better probability to repay the loan on time 5.8 times higher compared to no-collateral case. Moral hazard is reduced due to the absence of any detrimental aspect to the borrower or due to any advantage gained by not paying the loan on-time. But with the presence of requirement for collateral, the micro businesses would think twice about not repaying the loan.

Credit Diversion

Credit diversion to non-productive and non-profitable activities affects negatively the rate of loan repayment. Micro-enterprises who use the loans for non-productive outcomes will have 2.5 times higher probability of loan repayment default compared to micro-business keep the loan for business purposes. This result is consistent with research (Setargie, 2013), which revealed that the credit diversion adversely affect the loan repayment as it was not used to gain any benefit or advantage.

Instant Loan Approval

The loan approval decision is related to the initial screening performed by the bank or lender. This screening procedure with constrained credit is intended to prevent the possibility of moral hazard and to reduce credit risk. Bank, based on its analysis can predict the level of inherent


risk in the loans they manage and can avoid risky borrowers. By providing the entire loan as required by the borrower, the bank shows a high level of trust that the borrower will repay back the loan.

KUR borrower with a full loan approval or without credit constrain has an 8.3 times higher probability to repay the loan compared to KUR borrower who gets credit constrain. This means that information about the customer can be obtained by the bank, if the bank is not usually use the collateral deemed acceptable to prevent wan-achievement of customer.

The presence of asymmetric information can also be overcome by imposing a high interest rate which is not applicable for the KUR program as its interest rate has been set by the government. Raising the interest rate might not be the best way anyway for it will increase the complicate the ability of microenterprises to repay of the loan. For informal credit markets, the measure to raise interest rates is often taken to cope up with asymmetric information. The accuracy of customer screening is very important that the predicted event can be approximately equal to the risk of ex-post default. The efficiency of this screening approach can be observed from the capacity to estimate and to secure risky borrower in order to minimize the risk of default (Ofonyelu and Alimi, 2013).

Side Job

Business owners normally have more flexible working time based on how they manage their business. By having a second job, microenterpreneurs are expected to greater income so they have more room for breathing in repaying the loan. Micro business owners with a side job would have a 49.8 times better chance in paying their KUR credit on time than the ones who do not.

Wongnaa and Vitor (2013), in their research revealed that off-farm side jobs have more positive impact on loan payments. (Tundui and Tundui, 2012) also found that owning different business at one time will reduce the problem of loan repayment where the profit from other businesses could help for the repayment.

Working Spouse

Household microenterpreneur with a working husband or wife has a better probability to repay the loan on time by 4.5 times than the one who does not have a working spouse. Couples with a working couple will have a greater income, thus better ability in loan repayment.

Age

Older business owner will have less probability to repay a loan by 0.9 time compared to a younger microenterpreneur, or the chance of the latter to repay the loan on time is 1.1 time better than the former.


Factors that are related to the borrower’s characteristics but do not affect significantly to the loan repayment are gender and education. This study is not in line with the results from Wongnaa and Vitor (2013) and Setargie (2013) who noticed that education is a very important factor affecting the repayment of the loan. According to their research, higher education background gives higher efficiency and productivity. While in relation to gender factor, Roslan and Karim (2009) stated that gender has a positive effect on loan repayment, or in other words the probability for repayment failure is higher for among the man than among the women (Wongnaa and Vitor, 2013). Meanwhile according to research of Setargie (2013), gender is not significant but the negative coefficient indicates that women repay their loan in better rate than men.

Business Location

The location of micro-enterprises does not significantly affect the loan repayment.in paying the loan, in this case KUR, due to the homogeneous characteristics of micro-enterprises and the uniformity of distribution scheme and requirements of KUR in every district in Indonesia. District of Margorejo is located very close to urban areas, while District Dukuhseti is more rural and coastal and very far from urban areas. However, the characteristics of KUR repayment is not significantly affected among the people in both rural and urban area.

Rural communities are considered to have more social and culture connection and cohesion compared to urban communities. Nevertheless, it does not play a significant factor that change the likelihood of KUR repayment. However, although not significant, Dukuhseti shows a coefficient with a positive direction.

The model specification in this paper is different from the research of Srinivas (2015) that incorporated socio-cultural factors such as borrower’s place of origin, local language and main religion in the area. In his research Srinivas find these factors play an important role in the informal loan repayment. However, formal and informal credit loans cannot be compared face to face because informal loan is very dependent on the trust level between the two parties, meanwhile the formal loan emphasis on the economic value of the collateral.

Number of Dependents

A bigger family means higher expenditure for food. The estimation result indicates that the number of dependents does not significantly affect the probability of a micro businesses not to repay the loan. This is in contrast with the research of Setargie (2013), where number of dependents significantly brings negative effect to the loan repayment.

In this study, more than 75 percent of borrowers have less than three children. Dependents in this study is defined as the number of children still economically dependent to their parents.


The estimation may yield different results if the measured dependent variable includes all dependents of a family, including parents, non-working spouse, nephew(s) and niece(s), and others.

Business Characteristics and Repayment Period

For business characteristics, the factor that does not significantly affect the loan repayment is the business lifespan. A long-time business can be associated with experience. This is not in line with the research of Wongnaa and Vitor (2013) and Tundui (2013) where experience have a significant positive connection to the repayment of loans.

The repayment period does not bring significant effect on repayment, although the direction shows a negative trend. This result is the opposite of research by Vitor (2012) who found that repayment period have a significant negative effect to the loan repayment.

V. CONCLUSION

Empirical studies in this paper have analyzed several factors that significantly affect the KUR credit repayment by household micro-enterprises, including age, type of business, expenditure on food consumption, side job, working spouse, availability of other non-KUR credit, collateral, initial screening (unrestricted credit) and credit diversion.

In general, some borrower characteristic factors such as gender (female), longer age, higher education and rural location are considered as a representative of moral hazard factor of a better the loan repayments, however the fact show they do not significantly affect the loan repayment. It is safe to say that when it comes to loan, moral hazard factor cannot be used as reliable indicator to predict one’s capability to repay the loan on time. Objective factors such as the type of business and collateral is shown to influence the loan repayment more significantly.

It should be underlined that the empirical research outlined in this paper has a limited number of samples, and is limited only from the same district. Therefore this paper can be developed by expanding the scope of data in national level.


REFERENCES

Armendariz, B. and Morduch, J. 2005. The Economics of Microfinance, MIT Press, Cambridge, MA.

Chakravarty, S., Iqbal, S.M.Z., Shahriar, A.Z.M. 2013. Are Women “Naturally” Better Credit Risks in Microcredit? Evidence from Field Experiments in Patriarchal and Matrilineal Societies in Bangladesh. Annual Meeting of the American Economic Association in Philadelphia in 2014.

D’espallier, B. Guerin, I. Mersland, R. 2011. Woman and Repayment in Microfinance: A Global Analysis. World Development Vol. 39, No. 5. Pp 758-772. 2011.

Gebeyehu, Z. Beshire, H. And Haji, J. 2013. Determinants of Loan Repayment Performance of Smallholder Farmers: The Case of Kalu District, South Wollo Zone, Amhara National Regional State, Ethiopia. International Journal of Economics, Business and Finance Vol. 1, No. 11, December 2013, PP: 431- 446, ISSN: 2327-8188 (Online) Available online at http://ijebf.com/.

Godquin, M. 2004. Micro Finance Repayment Performance in Bangladesh: How to Improve the Allocation of Loan by MFIs. World Development, Vol. 32, No. 11, pp. 1909-1926.

Ibeleme, Sylvester N.O., Godwin C. O., and Odionye, J. C. 2013. Determinants of loan size and repayment performance of small oil producers in Nigeria: The case study of Abia State. International Journal of Business Management and Administration Vol. 2(3), pp. 043-054.

Jalaludin. 2002. Studi Komparasi Kinerja Bank Perkreditan Rakyat (BPR) yang Berdasarkan Syariah dengan BPR Konvensional dalam Pemberian kredit untuk Pengusaha Kecil Perdesaan di Nusa Tenggara Barat (NTB). Not Published. Postgraduate School of IPB. Bogor.

Juanda, B. 2009. Ekonometrika Pemodelan dan Pendugaan. IPB Press.

Lukman, S. et al. 2008. Kajian Upaya Peran Microbanking dan Pendekatan Pembiayaan Kelompok dalam Rangka Pengembangan UMK di Sumatra Barat. Bank Indonesia and Centre for Banking Reasearch Universitas Andalas.

Mokhtar, S.H., Nartea, G., and Gan, C. 2012. Determinants of microcredit loans repayment problem among microfinance borrowers in Malaysia. International Journal of Business and Social Research (IJBSR), Vol. 2, No. 7, December 2012.

Nawai, N., Shariff, M.N. 2013. Determinants of Repayment Performance in Microfinance Programs in Malaysia. Labuan Bulletin of International Business & Finance, Vol. 11, 2013, 14 – 29.


Ofonyelu, C., and Alimi, R.S. 2013. Perceived Loan Risk and Ex Post Default Outcome: Are the Bank’s Loan Screening Criteria Efficient?. Asian Economic and Financial Review, 2013, 3(8):991-1002.

Ojiako, I. A, and Ogbukwa, B. C. 2012. Economic Analysis of Loan Repayment Capacity of Smallholder Cooperative Farmers in Yewa North Local Government Area of Ogun State, Nigeria. African Journal of Agricultural Research Vol. 7(13), pp. 2051-2062.

Oke, J.T.O., Adeyemo, R., and Agbonlahor, M.U. 2007. An Empirical Analysis of Microcredit Repayment in Southwestern Nigeria. Humanity & Social Sciences Journal 2 (1): pp. 63-74, ISSN 1818-4960.

Okojie, C., Monye-Emina, A., Eghafona, K., Osaghae, G., and Ehiakhamen, J.O. 2010. Institutional Environment and Access to Microfinance by Self-employed Women in the Rural Areas of Edo State. International Food Policy Research Institute. IFPRI. Brief No.14.

Onyeagocha, S.U.O, Chidebelu, S. A. N. D, Okorji, E.C.O, Ukoha, A.H. Osuji, M.N and Korie, O.C. 2012. Determinants of Loan Repayment of Microfinance Institutions in Southeast States of Nigeria. International Journal of Social Science and Humanities Vol.1 No.1 April 2012 ISSN 2166-7721.

Roslan, A.H. and Karim, M.Z.A. 2009. Determinants of Microcredit Repayment in Malaysia: The Case of Agrobank. Humanity & Social Science Journal 4(1): 45-52, 2009.

Srinivas, H. 2015. Borrower Evaluation in the Informal Credit Markets. GDRC Reseaarch Output E-110. Kobe, Japan: Global Development Research Center. Retrieved from http://www.gdrc.org/icm/b-eval.html on Thursday, 31 December 2015.

Supadi and Sumedi. 2004. Tinjauan Umum Kebijakan Kredit Pertanian. ICASERD working paper No. 25. Center of Research and Development on Agiculture Social Economy, Department of Agriculture.

Voice of Indonesia. 2012. Kredit Usaha Tani Macet. . http://www.id.voi.co.id/voi-komentar/1853-kredit-usaha-tani-macet.

Setargie, S. 2013. Credit Default Risk and its Determinants of Microfinance Industry in Ethiopia. The Journal of Young Economists. http://joyeconomists.com/2014/07/19/setargie-2013-credit-default-risk-and-microfinance-in-ethiopia/.

Tundui, C., and Tundui, H. 2012. Microcredit, Micro Enterprising and Repayment Myth: The Case of Micro and Small Women Business Entrepreneurs in Tanzania. American Journal of Business and Management Vol. 2, No. 1, 2013, 20-30.


Vitor, D.A. 2012. Determinants of Loan Repayment Default among Farmers in Ghana. Journal of Development and Agricultural Economics Vol. 4(13), pp. 339-345, November 2012. Available online at http://www.academicjournals.org/JDAE.

Wongnaa, C.A., and Vitor, D.A. 2013. Factors Affecting Loan Repayment Performance Among Yam Farmers in the Sene District, Ghana. Agris on-line Papers in Economics and Informatics. Volume V Number 2, 2013.

77The Islamic Banking and The Economic Integration in ASEAN

¬THE ISLAMIC BANKING ANDTHE ECONOMIC INTEGRATION IN ASEAN

The efficiency level of the banking industry is the most important indicator to identify the soundness

of the banking system. This paper uses a non-parametric frontier approach, DEA, to analyze the efficiency

the Islamic bank in ASEAN. The control variables use the price of deposits from customers, deposits and

placements of banks, labor, and others operational expenditures; and the output variables use financing,

deposits and placements in other institutions, securities, and investment others. The joint-venture bank

was found to be the most efficient group within the observation period. Furthermore, the average

efficiency of Islamic banking in Indonesia, using the intermediation approach, is lower than the average

of ASEAN, unless it can reduce the cost of labor and other operational expenses. Also this paper examines

the determinants of efficiency of Islamic Banking in ASEAN. Internal factors are Total Assets, ROA, BOPO,

and ETA, and the external factors are Market Power and Inflation. Using Tobit regression, the results show

the factors that most influence Islamic banking efficiency in Indonesia is the total size of the bank or its

assets, OPEX / OR, and Market Power.

Abstract

Keywords: ASEAN, Bank Efficiency, DEA, Islamic Banking

JEL Classification: C14, F65, G21

1 Postgraduate Student of Institut Pertanian Bogor (IPB) Email: [email protected] Director of Academic and Student Affair of MB-IPB Program Email: [email protected] Director of Bank BNI Syariah. Email: [email protected].

Solihin1 Noer Azam Achsani2

Imam T. Saptono3


I. INTRODUCTION

In 2015, the Association of Southeast Asian Nations (ASEAN) countries entered the era of the collective agreement of the ASEAN Economic Community (AEC). ASEAN is an association of countries in Southeast Asia consisting of 10 countries, namely Indonesia, Malaysia, Thailand, Philippines, Singapore, Brunei Darussalam, Vietnam, Laos, Myanmar, and Cambodia.

In the MEA agreement, one of the most influential sector is the banking industry. One type of bank is based on the type of payment services a bank conducts business based on the principle of profit sharing, such as the Sharia bank. In connection with the implementation of the MEA, Indonesia will become a Sharia banking market that is very attractive for investors in ASEAN banking. Some of the factors that are the main attraction for Sharia banking managers of ASEAN countries to enter and operate in Indonesia are the largest Muslim population in the world, the potential for Sharia economic development, and market share in Indonesia is still very large. Secondly, Indonesia only ranks fourth as a country with potential and conducive in developing sharia financial industry (Alamsyah, 2012). Thirdly, the scale of the Sharia Banks operating in Indonesia is still very small compared to the scale of the Malaysian Sharia Bank. Fourthly, the majority of Sharia banks in Indonesia are still in the expansion phase which requires significant infrastructure investment costs. These conditions give Malaysian Sharia banks already existing in Indonesia with greater assets, more opportunities for expansion within the country. And lastly, although the percentage of operating costs compared to the operating income (ROA) of Sharia banks in Indonesia is much higher than Islamic banks in Malaysia (and other ASEAN countries), but the Indonesian NOM, ROA and ROE indicators shows higher value. No wonder many foreign investors are interested to establish or buy Islamic banks in Indonesia. High profitability will certainly accelerate the asset growth of Sharia banks in Indonesia so as to achieve economies of scale efficiently.

Apart from giving a big opportunity for investors, including owners of Islamic banks already operating in Indonesia, these conditions also are a tough challenge for Islamic banks in Indonesia. With increasing Islamic banks that will operate in Indonesia, including Islamic banks either already operating or those newly established from other ASEAN countries, the Islamic banks already operating in Indonesia are required to manage its operations efficiently. An efficient Bank will be better able to defend itself in the competitive banking industry that will be encountered.

To get an idea of the competitiveness of the Islamic banking industry in each ASEAN country and Sharia banks operating in each the ASEAN country, it is necessary to analyze the level of efficiency of ASEAN Islamic banks that have been running in recent years.

In connection with the management of ASEAN Islamic banking in the future, problems may arise when competition is increased in the banking market, and where Islamic Banks are not able to operate efficiently so as not to compete and they eventually fail to survive. This


led the importance of measurement of the level of efficiency of each Islamic bank in ASEAN. Departing from the data level of efficiency and its position in the competition, there is a need evaluate other inputs used these banks, as well as the output generated by these Islamic banking.

To help managers and regulators of Islamic banking, there is also a need to perform analysis of the determinants of bank efficiency levels, both internal factors and external factors. Accurate and sharp mapping on this would help bank managers create a appropriate strategies to increase efficiency. For the government or policy-makers and regulators of Islamic banking in each country, these results may help in formulating policies that encourage the growth and development of the Islamic banking industry.

Explicitly, the purpose of this paper first, is to analyze and evaluate the efficiency of the Sharia banking industry in ASEAN countries to illustrate the power of Islamic banks against competition from other ASEAN countries; second, to analyze and identify competition and market structure in the ASEAN Islamic banking industry, and Islamic banking management policies of countries that have efficient banking; third, to provide information and suggestions to Indonesian banking regulators and practitioners of Islamic banking in Indonesia in an effort to improve the performance and financial efficiency so as to face the competition in the free market ASEAN countries.

Research conducted includes 32 banks Islamic banks in the following ASEAN countries, 11 in Indonesia, 18 in Malaysia, the first Islamic banks in Singapore, Brunei Darussalam, and the Philippines. Other Sharia Banks in ASEAN, if any, and UUS from conventional banks are not subject to this research due to the limited available data. The research years is from 2008- 2013, or the year the Islamic banks began operating if after 2008.

II. THEORY

2.1. Efficiency Theory

Efficiency is a relative term, which must always be associated with certain criteria. An activity is said to be efficient if the company can produce more output compared to another company (with similar resources), or from an established output standard using available inputs, or use fewer inputs than others to produce the same output as another company. Therefore, the concept of efficiency actually begins with the concept of the production theory that explains the technical relationship between input and output factors. The production function describes the process of transforming inputs into outputs at a given period of time.

According Pass and Lowes (1997), efficiency is the relationship between input factors that are scarce with the output of goods and services. This relationship can be measured physically (technical efficiency) or through its cost (economic efficiency). The concept of efficiency is used as a criterion in the assessment of how well the market allocates resources.


In the banking sector, the measurement of efficiency (performance measurement) is also one thing that is needed to determine the performance of the banking system. A company is said to be efficient if the company can minimize the cost of producing a particular output or can maximize profits by using a combination of existing inputs (Srivastava, 1999).

According to Hadad, et. al (2003), efficiency is one of the performance parameters that theoretically underlies all of the performance of an organization. The ability to generate maximum output with existing input is a measure of expected performance. At the time efficiency measurement is done, the bank faces with the challenges of how to get the optimum output level with the existing input level, or get a minimum input level with the level of specific output.

Generally, technical efficiency measurement can be performed based on the input (input-

oriented) or output (output- oriented). In other words, the efficiency can be measured with minimal input to achieve a certain output or maximize output with the use of certain inputs, in order to obtain alternative calculation:

The calculations in this study focuses on the second alternative, known as efficiency based on input, so that by using multiple outputs and multiple inputs, will obtain the following formula:

where: u1 = weighting on output i, y1j = 1 unit of output j, v1 = weighting on input i, dan x1j = 1 unit of input j.

2.2. Data Envelopment Analysis (DEA)

DEA is a technique of mathematical programming that is used to evaluate the relative efficiency of a set of units in managing resources (inputs) to the same type that into results (output) with the same type as well, where relationships form the function of input to output is not unknown (Coelli, et al., 1998). In other words, the DEA is a linear programming model that can include many fractional outputs and inputs without the need to determine the weight of each variable in advance, and without the need for explicit explanation about the functional relationship between input and output (unlike regression). DEA calculates the scalar measure of efficiency and determines whether the level of input and output are efficient for the unit being evaluated (Cooper, et al, 2007). DEA is used to measure the relative efficiency level, based mainly on technical efficiency.

......

2211

2211

++

++=

jj

jj

xvxvyuyu

Efficiency


Frontier approaches such as DEA actually started from the ratio analysis of output to input (output / input), the difference is DEA combines all inputs and outputs integrated. The first measurement of the efficiency in modern times was developed by Farrell in 1957. Farrell proposed that the efficiency of a company consists of two components, namely technical

efficiency, which reflects the company’s ability to obtain the maximum output of the set of available inputs, and allocative efficiency reflecting the company’s ability to use the inputs in optimal proportions with due regard to the price of each input (Ahmad, et al. 2012). Both types of efficiencies are then combined to give a total size of economic efficiency (Coelli, 1996 in Ahmad et al. 2012).

From the theory proposed by Farrell and developed by Coelli mentioned above, the efficiency can be grouped into three types of efficiency, namely:

1. Pure Technical Efficiency (PTE).

The size of the PTE is obtained by estimating the efficiency boundary with the assumption of the return-to-scale variable (VRS). PTE is a measure of technical efficiency without scale efficiency, its size reflects managerial performance in managing inputs for production processes in order to maximize output. Thus, the PTE size has been used as an index to capture managerial performance. This efficiency is what Farrel referred to as Technical Efficiency.

2. Scale Efficiency (SE)

Scale or allocative efficiency scale indicates the proportion of the reduced use of inputs used by the bank to produce output in optimal scale of constant returns to scale (CRS). SE demonstrates the ability of banks to operate on an optimum scale (Hauner, 2005). Farrel calls this efficiency, Allocative Efficiency. SE is obtained by calculating the ratio of OTE divided by PTE.

3. Overall Technical Efficiency (OTE)

Efficiency is related to the productivity of inputs. Technical efficiency of a company is a measure of how well the ratio of input to achieve its output, compared with the maximum potential to achieve it, which is represented by the production possibility frontier (Barros and Mascarenhas, 2005). Thus, the technical efficiency of the bank is its ability to convert some resources into financial services (Bhattacharyya et al., 1997). This efficiency is what Coelli calls Total Economic Efficiency.

In the DEA model, the efficiency level can be measured using either the input orientation or the output orientation. The measurement of Input-oriented efficiency shows that some inputs can be reduced proportionately without changing the amount of output produced. While, with the measurement of output-oriented efficiency, the number of outputs can be increased proportionally without changing the number of inputs used.


2.3. Input-Output Relations

The efficiency level is determined by the selection of the variables for input and output. Variables may represent different information, even though they carry the same label. Selection of input and output variables are affected by the limitations in the selection of variables due to the reliability of the data obtained.

There are three approaches to explain the relationship of input and output used in the measurement of the efficiency of financial institutions (see, among others Hadad et al. (2003)), i.e:

1. Production Approach is where the activity of Islamic banks is seen as a process of production of services for the owners of the funds (Shohibul mal) and recipients of financing or fund manager (mudharib).

2. Intermediation Approach is where the bank is seen to function as an institution that receives or collects funds from parties who have excess funds, and then distributes those funds to parties in need. In the Islamic bank, the intermediary approach describes banking activities as transforming owned funds derived from wadiah deposits, savings and time deposits of mudaraba, and the placement of funds by a 3rd party into a fund used for financing by mudharib.

3. Approach Asset is where the primary function of a financial institution is seen as a creator of credit for conventional bank loans or funds that can be collaborated for Islamic banks.

2.4. Research Accomplished

Research on the effectiveness of banking in general, and Islamic banking in particular, has been done, both with parametric approach and nonparametric approach. Berger and Humphrey (1997) mentioned that there are 130 studies analyzing efficiency boundaries for financial institutions in 21 countries, highlighting the importance and increasing frequency of this research in recent years. Several research and research conclusions related to banking efficiency with ratio analysis methods, SFA, and DEA are as follows

a. Indonesian Banking Efficiency Research

Hadad concluded that with the nonparametric approach (DEA) in measuring the efficiency of banks in Indonesia, it was found that foreign banks were efficient in 1997, while foreign national private banks were most efficient in 1998-1999, and the non-foreign national private banks were the most efficient in 2001-2003 (Hadad et al, 2003). Other DEA research concluded that non-foreign national private banks were most efficient during three years (2001-2003) in an eight-year analysis period (1996-2003) compared to other banks (Hadad et al 2003). By using the SFA, the efficiency of Islamic banking during the 2003-2006 experienced an annual average efficiency of 94.37 percent and Islamic


banking income was strongly influenced by the financing facilities and placement in Bank Indonesia (Suswadi 2007). Ascarya (2008) research results which analyzed and compared the efficiency of Islamic banks with conventional banks in Indonesia 2002-2006 using DEA shows that Islamic banks were relatively more efficient than conventional banks. The performance of Islamic banks consistently increased every year (during the study period), except in 2004 due to the Islamic banks conducting expansive phase. In addition, the average value of BUS efficiency is relatively better than UUS or SRB. And the results of the Bank Indonesia research department of Islamic Banking concluded that the overall level of efficiency of the intermediation of Islamic banks, both OTE, PTE and SE was relatively a good, particularly the intermediation of Islamic banks in channeling third-party funds, and the lower value of SE than the value of PTE shows inefficiency was caused BUS’ operational scale, rather than on pure technical efficiency. Inefficiency on a scale of SE shows that Islamic banks have not been producing at a rate of production capacity. And several other studies that add to the knowledge of Indonesian banking efficiency was conducted by Ramli (2005), Pratama (2011), Magrianti (2011), and Herlambang (2008).

b. Banking Efficiency Research Malaysia

DEA with non-parametric methods analyzing the performance of the Malaysian Islamic banking sector and foreign banks in Malaysia during the period 2001-2005, found that the scale inefficiencies (scale inefficiency) dominate the inefficiencies; and foreign banks showed a level of technical efficiency higher than domestic banks was incomparable (Sufian 2006).

c. ASEAN Banking Efficiency Research

Research by Yudhisthira (2003) analyzed the efficiency of 18 Islamic banks in 12 countries with DEA non-parametric methods showed that, overall, Islamic banks had little inefficiency during the global crisis of 1998-1999. A study of the efficiency of the banking sector in ASEAN countries in the years 1989-2000, using the SFA approach showed that in general banks in ASEAN were in a situation of increasing returns to scale in the production phase. Large size banks had an efficiency of greater costs compared with other banks, although the scale of the banking economy was inversely proportional to the size of the bank. The interesting thing to observe is that the level of banking efficiency had negative effects on GDP growth in the short term, but in the long term increase / decrease in the efficiency affected the growth of GDP (Abdul Karim and Mohd Zaini, 2001). Thangavelu and Findlay (2010) conducted a study to determine the efficiency of approximately 600 banks located in countries East Asia (Indonesia, Malaysia, Philippines, Singapore, Thailand, and Vietnam) in 1994 to 2008 with the conclusion that the characteristics of the banks have an important impact to the efficiency of the bank. This study indicates that the regulation and supervision of banks will be crucial to improve bank efficiency and stability of financial markets in Southeast Asia. In general, regulation and supervision of banks is


very important to improve the efficiency of banks in the Southeast Asia region, compared with the monitoring of the private sector in banking activities. In particular, restrictions on bank risk activities tends to produce a more efficient bank.

d. Other Banking Efficiency Research

Other research done by Barr, RS, et al. (1999) on the production efficiency and performance in the Commercial Bank of America in 1984-1989. Lang and Welzel (1996) analyzed the efficiency level of 757 banks in Germany in the period 1989-1992 using panel data with the conclusion that the average bank in Germany suffered a deviation from best performance limit. In addition, there is research by Berger and Mester (1997) regarding the calculation of the efficiency of financial institutions using DEA and DFA-SFA methods. Berger and Mester tried to test some of the possibilities that were the source of differences in results obtained from each study, including the different concepts of efficiency, the methods used, the number of samples and other sources that could lead to differences from the calculations. Framework

2.5. The framework of this research can be seen in the following Picture 1.

Graph 1 Framework

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��


III. METHODOLOGY

3.1. Data

The data used in this research is data published annual financial statements of Islamic banks from 2008 to 2013. This data includes 32 Islamic banks, which consists of 1 sharia bank in Brunei Darussalam, 11 sharia banks in Indonesia, 18 sharia banks in Malaysia, 1 sharia bank in Philippines, and 1 sharia bank in Singapore.

In data processing, this study used a two-steps; first, measure the efficiency of each Islamic bank by using the DEA method, by specification of the constant return to scale (CRS) and variable returns to scale (VRS). DEA calculation results of each Islamic bank is then grouped by country to be compared. In the second stage, measurements of certain factors that allegedly have an influence on the level of efficiency were used in Tobit regression analysis.

3.2. Selection of Input and Output

The approach used in this study is the intermediation approach. Selection of the appropriate variables as inputs and outputs become very crucial as it determines the level of results of efficiency assessment in research. Input and output variables are selected using the intermediation approach as follows:

• Input variables:Deposit (covering all third-party funds in current accounts, savings,deposits, borrowings, securities issued, and temporary funds syrkah), obligation to other banks, and costs Opex which includes cost of HR and other operating expenses.

• Outputvariables:FinancingProductive(coveringtheentirefinancing, investments, and receivables), placement in other banks and securities held.

3.3. Construction of the Banking Efficiency Determinant Model

Factors that affect the value of efficiency using Tobit regression includes internal and external factors. Internal factors include the size of Islamic banks (reflected by a logarithm of total assets), profitability (reflected by the ratio of return on total assets, ROA), operations (reflected by the ratio of operating expenses to operating income, ROA) and equity (reflected by the ratio of equity to total assets, ETA). While external factors selected were market power of 1 which is reflected by the ratio of the amount of deposits in each bank to total deposits in the banking industry of the country concerned; the market power of 2 which is reflected by the ratio of the total amount of assets in each bank to total gross domestic product (GDP) within the banking sector in the country concerned; and the inflation rate.


The general equation in the panel regression is as follows:

Yit = C + X1 LNTAit + X2 ROAit + X3BOPOit + X4 ETAit + X5 MP1it + X6 MP2it + X7 INFit

where Y indicates the level of efficiency of the bank; LNTA is the size of the bank; MP1 is market power 1 proxied by the ratio of the amount of bank deposits to the total amount of deposits of the banking industry of the country concerned; MP2 is the market power of 2 which is proxied by the ratio of total assets to GDP of the country concerned in the sector; while INF indicates the level of inflation.

This paper does three types of measurements, i.e. the Overall Technical Efficiency (OTE), Pure technical Efficiency (PTE), and Scale Efficiency (SE). Thus, there are three (3) models estimated by the Tobit panel, as well as an attempt at robust testing the of the model estimates.

IV. RESULTS AND ANALYSIS

The following can be put forward the research and discussion related to the measurement of the efficiency of Islamic banks in ASEAN Countries.

4.1. Efficiency Results Scores

DEA calculation results with the intermediation approach shows that the efficiency of both input oriented and output oriented, are not relatively different. This means that efforts to optimize the input to produce the same output, is not unlike the effort to optimize the output with the same input level. Based on the magnitude, the average value of the efficiency of Islamic bank in ASEAN in the intermediation approach indicates that banks are classified nearly efficient. This applies both when using the Overall Technical Efficiency (OTE) with a value of 0.78324, Pure Technical Efficiency (PTE) with a value of 0.88067, and Scale Efficiency (SE) with a value of 0.89172.

Of the 174 Decision Maker Unit (DMU) measured, the number of DMU that is relatively efficient in general (OTE) is as much as 48 DMU with a distribution of 9 DMU for Indonesia, 29 DMU for Malaysia, 6 DMU for Philippines, and 4 DMU for Singapore. If measured by Pure

Technical Efficiency (PTE), then 78 DMU is relatively efficient and the distribution is 4 DMU for Brunei, 17 DMU for Indonesia, 47 DMU for Malaysia, 6 DMU for Philippines, and 4 DMU for Singapore. Lastly, if you use the Scale Efficiency (SE), then 48 DMU is efficient with a distribution of 9 DMU for Indonesia, 29 DMU for Malaysia, 6 DMU for Philippines, and 4 DMU for Singapore. Results of efficiency of ASEAN and the deployment of the three types of efficiency are shown in the table and Graph below:


The most efficient DMUs and benchmarks for other DMUs are the Alliance Islamic Bank Berhad Malaysia 2008, Maybank Indonesia 2010, Islamic Bank of Asia (IB Asia) Singapore, Bank Panin Syariah Indonesia 2011, Bank Victoria Syariah Indonesia 2010, Standard Chartered Saadiq Berhad Malaysia 2010, and Hong Leong Islamic Bank Malaysia 2011.

Brunei, the Philippines, and Singapore only has one Islamic bank or banks that run Islamic transactions. Bank Islam Brunei Darussalam Berhad, technically is able to manage all the resources owned by the intermediary function to score efficiently from 2008 to 2011, and was slightly down in 2012 and 2013 in connection with the increase of the DMU technical capability in other countries. However, the size of the total assets managed, throughout the

Table 1Total DMU Efficiency in Indonesia and Malaysia

EFF

OTE 174 54 9 16.67 104 29 27.88 PTE 174 54 17 31.48 104 47 45.19 SE 174 54 9 16.67 104 29 27.88

ƩSDMU ASEANƩSDMU ƩSDMUƩSDMU EFF ƩSDMU EFF% %

Indonesia Malaysia

Table 2Total DMU Efficiency in Brunei, Philippines, and Singapore

EFF

OTE 174 54 9 16.67 104 29 27.88 PTE 174 54 17 31.48 104 47 45.19 SE 174 54 9 16.67 104 29 27.88

ƩSDMU ASEANSDMU SDMUSDMU EFF SDMU EFF% %

Indonesia Malaysia

Graph 1. Sharia Bank Efficiency Distributionin Five Asean Countries

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��


years of research, the DMU in Brunei showed to be at an inefficient level (below 0.500) and slightly increased to 0.503 in 2012 and 0526 in 2013. This is due to the total assets managed by Islamic banks in Brunei Darussalam is very small compared to the total assets managed by Islamic banks in other ASEAN countries. This condition caused the value of OTE of the Bank Islam Brunei Darussalam Berhad to be very efficient (below 0.500).

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

The Islamic bank in the Philippines, Al-Amanah Islamic Investment Bank, showed the value of the maximum efficiency all year round for all kinds of efficiency studies. It showed that the bank is able to manage input and output optimally. This means that the number of 3rd party funds placement in the bank, which is an input variable, is smaller than the funds that are assigned to 3rd parties and financing, which is a variable output. The amount is shown with a high ratio of Equity to Total Assets (ETA). Most sources of funds placement to 3rd-parties come from the funds of banks owners, not the successful of collection public funds. High efficiency value in the intermediation approach does not show that the bank managers were able to get a higher profit. The performance of Al-Amanah Islamic Investment Bank always been at a loss. In 2008, the accumulated losses exceeded US $8 million, so that the bank’s owners then added a capital injection of US $ 17 million in 2009. Unfortunately, the capital increase was not accompanied by ability to place funds on financing products that provides great returns. Most of the funds were placed on Bangko Sentral Ng Philippines, Bank Central with a small returns. Coupled with substantial operational costs, the return does not bring the performance of the bank into profit. In addition, the profit was only from the transactions Al-Bai Bithaman

Ajil financing which is not permitted in Islamic transactions in Indonesia.

Similar to Islamic bank in the Philippines, the Islamic Bank of Asia (IB Asia) in Singapore is also an investment bank that does not perform the intermediary function of sharia banks, which is shown by the amount of third party funds which is very small compared to the amount of fund placement to party to third parties. The total assets of Islamic Bank of Asia (IB Asia) also continued to decline from USD 735 million at the end of 2008 to USD 366 million at the end of 2011, and no longer publishes its report since 2012.


The graphs of the efficiency level of Sharia Banks in Brunei Darussalam, the Philippines and Singapore compared to the average efficiency levels of ASEAN from 2008 to 2013 are as shown in Graph 2-4.

Graph 2.Level of Efficiency of the Sharia Bank in Brunei

Graph 4.Efficiency Level of the Sharia Bank in Singapore

Graph 3.Efficiency Level of Sharia Banks in the Philippines

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

While Islamic banks in these three countries are efficient in almost all types of efficiencies, they do not have the potential to take advantage of the MEA open market. In addition to the relatively small scale of banks compared to Sharia banks in Indonesia and Malaysia, the company’s performance is also not good and has not been supported by banking policies and regulations in the countries concerned. Therefore, efficiency analysis with other approaches and or with different input and output variables is required.

Although the average efficiency level of DMU in Indonesia is almost efficient (above 0.75), the value is slightly below the average of ASEAN. The addition of Islamic banks in 2010 and the addition of a number of placements of third party fund from collected for pre-existing banks brought the OTE in 2010 and 2011 above the ASEAN average, but fell back in 2012.


Efficiency scores for Islamic banks in Malaysia showed slightly better efficiency levels compared with the average Islamic ASEAN bank. This means that when compared to the level of efficiency of DMU Indonesia, DMU Malaysia is more efficient. Similarly, the number of Malaysian DMU is efficient and became more of a benchmark than the DMU Indonesia.

A comparison of the average efficiency level of Islamic banks in Indonesia with the average Islamic bank in Malaysia is on the graph in Graphs 5 and 6.

Graph 5.Efficiency Level of Islamic Banking in Indonesia

Graph 6.Efficiency Level of Islamic Banking in Malaysia

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

When the level of efficiency in the intermediation approach measures only the input of the number of third party funds, meaning that the variable cost of human resources and other operational costs are incurred from the input variables, the results obtained differ. The average efficiency score for ASEAN is lower than the previous result of OTE 0.61810 (inefficient); PTE 0.82457 (less efficient); And SE 0.27401 (less efficient), with the trend only rising in 2009 to 2010, and relatively flat from 2010 to 2013 for OTE, PTE, and SE. A PTE value higher than the value of SE indicates that the source of inefficiency occurs because of problems at the operational scale of Islamic banks rather than on pure technical efficiency of Islamic banks. This means that inefficiency in SE shows that Islamic banks have not been producing at a rate of production capacity. From these efficiency values, it can be concluded that without taking into account the cost, the existing Islamic banks in the five ASEAN countries have not performed their intermediary function well, in contrast to the previous intermediation approach. In this intermediary approach, the efficiency score of Islamic bank Brunei Darussalam Berhad is at an inefficient level, below the ASEAN average for OTE and SE. While at PTE, originally efficient, it declined and was below the ASEAN average in 2012 and 2013. The condition was similar to the previous conditions of the intermediation approach. It shows that the total cost of human resources and other operating expenses in the bank does not relatively affect the value of efficiency. Similarly, the efficiency of Al-Amanah Islamic Investment Bank in the Philippines and the Islamic Bank of Asia (IB Asia) in Singapore remain above the ASEAN average.


The graphs of the efficiency levels of Sharia Banks in Brunei Darussalam, the Philippines and Singapore compared to the average efficiency levels of ASEAN from 2008 to 2013 with a pure intermediation approach are shown in Graphs 7-9.

Graph 7. Efficiency Level of the Islamic Bank in Brunei with Pure Intermediation Approach

Graph 9. Efficiency Level of the Islamic Bank inSingapore with Pure Intermediation Approach

Graph 8. Efficiency Level of the Islamic Bank in the Philippines with Pure Intermediation

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

In contrast to the efficiency scores for Islamic banks in Indonesia in the previous intermediary approach, where the average Islamic bank of Indonesia is below the ASEAN average efficiency score, in this intermediation approach the value of all efficiency is above ASEAN, and rises significantly from 2008 to 2009 and 2010. But with the increase of new Islamic banks in 2010, the efficiency scores declined in 2011 and 2012, and rose again in 2013 for PTE. It shows that the cost of human resources and other operational costs in Islamic banks in Indonesia is higher than the ASEAN average.

The efficiency level of Islamic Banks in Indonesia with a pure intermediation approach is shown in Graph 10.


While efficiency scores for Islamic banks in Malaysia show a level of efficiency below the average Islamic bank of ASEAN, this is different from the resulting measurement with the previous intermediation and shows that the cost of HR and other operational costs are input variables with small values.

Graph 10. Efficiency Level of Islamic Banks inIndonesia with Pure Intermediation Approach

Graph 11. Efficiency Level of the Islamic Bank in Malaysia with Pure Intermediation Approach

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

4.2. Potential Improvement

DEA calculations also show potential improvements that can be made by banks that are not yet operating efficiently. Based on the input-oriented intermediation approach, it can be concluded that the majority of Islamic banks in ASEAN countries must reduce their total inputs, while increasing their output to produce ideal output by DMU in those years. Improvements that must be done include a reduction in the amount of deposits amounting to 14.96%, HR costs reduced by 27.34%, other operational expenses reduced by 31.87%, and liabilities reduction


in other banks by 36.34%. While the addition of the amount of the variable output consists of additional financing at 0.0002%, placements in other banks can be increased by 50.59%, along with an increase in securities held by 3.77% (see Table 4).

��

��

��

��

��

��

��

� ��

��

� ��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

� ��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

�

�

��

� ��

��

��

�

�

��

��

��

��

�

�

��

� ��

��

��

�

�

��

�

�

�

�

�

�

��

� �

��

�

�

��

��

��

��

�

�

��


When the DMU in Indonesia and Malaysia are compared, the amount of the reduction of the variable input liabilities with other banks amounted to 36.64% at DMU in Malaysia and 35.36% at DMU in Indonesia. For Operating Costs for the input variables, the DMU in Indonesia should reduce by 40.04% of the actual cost of current operation, while the DMU in Malaysia only reduces to 29.82% from the actual. As for the HR Cost variables, the DMU in Indonesia should reduce by 31.89% of the actual cost of the current human resources, while the DMU in Malaysia only reduces by 24.89% from the actual. For Deposit input variables, the DMU in Indonesia should reduce by 24.91% of the actual current deposits, while the DMU in Malaysia only reduces by 13.77%. The output is to be expanded significantly is the Placement in the Other Banks, which is 75.55% for DMU in Indonesia and 37.67% for DMU in Malaysia. Securities must be increased by 41.96% for DMU in Indonesia and 1.93% for DMU in Malaysia. In input orientation, the absolute value of output addition is much smaller than its input reduction value.

4.3. Determinants of Banking Efficiency

Islamic banking competition is only available in Islamic banks in Indonesia and Malaysia. From the data processing, it is known that the determinants of the efficiency of Islamic banks in Indonesia are the total assets that negatively affect the Efficiency Scale of 0.05960, which means that each increase of one unit of asset amount will cause the decrease of the efficiency scale level

��

��

��

��

��

��

��

� ��

��

� ��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

� ��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

�

�

��

� ��

��

��

�

�

��

��

��

��

�

�

��

� ��

��

��

�

�

��

�

�

�

�

�

�

��

� �

��

�

�

��

��

��

��

�

�

��

��

��

��

��

��


of 0.05960, but does not affect the technical efficiency and overall efficiency. Therefore, an increase in the total amount of assets should be coupled with improved management capabilities to manage the company for a total efficiency level does not decrease. Another factor, BOPO has a positive effect on OTE at 0.27920, which means that every increase of 1 unit of BOPO will cause a total increase of banking efficiency by 0.27920. Thus, Islamic banking in Indonesia can increase its cost to earnings in order to increase its efficiency value. The largest cost of the bank is the profit-sharing cost. Therefore, Indonesian Islamic banking can provide a higher profit-sharing rate to the owner of the funds paid so far, when compared to the profit-sharing rate received by the bank.

��

��

�

��

��

��

��

��

��

��

��

��

� ��

� ��

� ��

��

� ��

��

��

� ��

� ��

� ��

� ��

��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

��

� ��

� ��

� ��

��

��

� ��

� ��

��

� ��

� ��

� ��

� ��

� ��

� ��

��

��

While in Malaysia, the Determining factors that influence the total assets negatively affect the PTE, ROA positively influences the OTE and PTE, and the MP1 has a positive influence on OTE. The explanation is that an increase of one unit of total assets will lower the technical efficiency level of the bank by 0.05643. The amount of assets that have been maximized will cause the level of technical efficiency to decrease, if not followed by the development of bank management techniques already in place.

An increased ratio factor for return on total assets will result in an increase in technical efficiency level at 4.765507 and an increase in total efficiency level at 4.80701. Thus, Islamic banking in Malaysia must increase of the level of ROA ratio over the years. While other factors also that affect the total efficiency rate is a factor of Market Power 1, which means that the increased ratio of deposit of sharia banks to total banking deposits in Malaysia will give a positive influence of 27.62184 level of total efficiency.


Islamic banks in Malaysia is more efficient compared with the average of Islamic banks in Indonesia. Individually, several Islamic banks in Malaysia have reached a large enough scale, and strong enough to be able to control the market of sharia banks in ASEAN. With total assets of more than 90% of the total assets of Islamic banks ASEAN, Malaysia has more capital to expand its operational areas. The development of Islamic banks in Malaysia is larger and faster than with Islamic banks in other countries in ASEAN, which is supported by several factors as follows (Nadratuzzaman. 2013):

• IslamicbanksinMalaysiaaremoreexperiencedbecausetheyareeightyearsolderthanSharia banks in Indonesia, founded on government initiatives, and 30% of the shares are owned by the Government affecting the policy of Sharia banking development in Malaysia.

• MalaysiahasanIslamicbankingregulatoryframeworkthatiswell-organizedwiththeestablishment of the Shariah Advisory Council of Bank Negara Malaysia (BNM SAC), General Practices Sharia 1 is a guide for Sharia institutions, and Islamic Financial Services Board (IFSB).

• Despitedisagreementsaboutthehalalness of some types of products, the products of Sharia banks in Malaysia are more numerous and varied than those in Indonesia. Similarly, Sharia products in the capital market.

V. CONCLUSION

This paper provides some conclusions, firstly, the input and output orientation yields a relatively similar level of efficiency. When a DMU is declared efficient on the input orientation, then the output orientation will give relatively the same result. Second, Brunei Darussalam, the Philippines and Singapore have only one Islamic bank. The DEA data result shows that each bank in the

��

��

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� �

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��


three countries is efficient except PTE and SE in the Islamic bank in Brunei Darussalam because the scale is too small and not optimal. However, they are not a threat to Sharia banking in other ASEAN countries. Thirdly, of the five ASEAN countries that have registered Islamic banks or banks that run Sharia transactions, Sharia banking competition of Islamic banks is only in Indonesia and Malaysia. The other three countries, Brunei Darussalam, Singapore and the Philippines do not have Sharia banks that are strong enough to compete, although they are relatively efficient. Fourth, Islamic banks in Indonesia will be more efficient if they can reduce their operational costs. The average level of DMU efficiency in Indonesia in the intermediation approach is below the ASEAN average. In contrast, the average level of DMU efficiency in Malaysia is above the ASEAN average. This condition shows that Sharia banks in Malaysia are more efficient than Sharia banks in Indonesia. However, if the operational costs that include human resource and other operational costs are excluded from input variables, then the efficiency level of Sharia banks in Indonesia is above the ASEAN average and Malaysia. To be able to see the potential, it should be seen and reviewed through further analysis of each bank so that it can be compared with other banks, and not the average value per country

Results of measurement of efficiency with DEA method can be a complementary measurement tool in assessing the overall performance of Islamic banking. With more detailed data and measurement of the efficiency with other approaches (asset approach and the approach of production), further research can be developed to illustrate the performance assessment of each bank more comprehensively.

To better compete with Islamic banks in other countries, Islamic banking in Indonesia should be able to suppress the HR Costs and Other Operating Expenses. The efficiency level of the Islamic bank in Indonesia is lower than the ASEAN average and Malaysia for HR Costs, and Other Operating Costs in Islamic banking in Indonesia is larger.

Banking in Indonesia should be able to increase the lending capacity to the community (real sector) to create new products, without losing the halal aspect. Business development activity needs to be accompanied by increased efforts to encourage research and development / innovation of more attractive Islamic banking products. Islamic banks should be encouraged to innovate new products that meet the challenges of the growing field of business / industry.

On the other hand, total assets and the ratio of ROA is an important element to improve the efficiency of Islamic banks in Indonesia, therefore it is necessary to formulate measures for the total assets of Islamic banks and increase of the revenue share paid against the results received by banks, in order to compete and develop / expand business activity.

Islamic banks in Malaysia have greater potential to be able to enter the market of Islamic banking in other ASEAN countries. Besides controlling 90% of the total assets of the existing Sharia ASEAN banking industry, the Malaysian Islamic banking institutions better off with more innovative and varied products for modern business transactions. In conclusion, this paper suggests the need for more in-depth study of the Indonesian Islamic bank opportunities to expand into other ASEAN countries, especially ASEAN countries that do not have Islamic banks.


REFERENCES

Abdul Karim, Mohd Zaini, 2001. Comparative Bank Efficiency Across Select ASEAN Countries. Asean Economic Bulletin. Vol. IX.(3): pp. 289-304.

Ahmad M. AA, Hans-Peter B, and Walayet AK. (2012), Islamic Commercial Banking In Europe: A Cross-Country And Inter-Bank Analysis Of Efficiency Performance. International Business

& Economics Research Journal. June 2012 Volume 11(6): pp. 647-676.

Alamsyah H. 2012. Perkembangan dan prospek perbankan syariah Indonesia: tantangan dalam menyongsong MEA 2015. Scientific Forum of Islamic Economy Experts. [Accessed in 2015 September 30]. Available on http://www.bi.go.id/id/ruang-media/pidato-dewan-gubernur/Documents/ 6bf00812e40b4d0cb140ea80239c4966 Perkembangan Prospek Perbankan Syariah Indonesia MEA 201.pdf.

Ascarya. 2008. Efficiency analysis of islamic and conventional banks in Indonesia using parametric SFA and DFA methods. Journal of Islamic Business and Economics.2(2).

Barr RS, Killgo KA, Siems TF, Zimmel S. 1999. Evaluating the productive efficiency and performance of US. Commercial Bank. Managerial Finance. 28 (8): pp. 3 – 25.

BarrosCP,MascarenhasMJ.2005.Technicalandallocativeefficiencyinachainofsmallhotels.International Journal of Hospitality Management. 24: pp. 415–436.

Berger AN, Humphrey DB. 1997. Efficiency of financial institutions: international survey and directions for future research. European Journal of Operational Research 98 (2): pp. 175-212.

Berger, A. N. and L. J. Mester. 1997. “Inside The Black Box: What Explains Differences in The Efficiencies of Financial Institutions?”. Journal of Banking and Finance, pp. 21: 895-947.

Bhattacharyya A, Lovell C, Sahay P. 1997. The impact of liberalization on the productive efficiency of Indian commercial banks. European Journal of Operational Researc. 98: pp. 332-345.

Coelli TJ, Rao DSP, Battese, O’Donnell CJ. 1998. An Introduction To Efficiency And Productivity

Analysis. USA (US): Kluwer Academic Publisher.

Cooper WW, Seiford LM, Tone K. 2007. Data Envelopment Analysis: A comprehensive Text

with Models, Applications, References and DEA-Solver Software. New York(US): Springer.

Department of Syari’ah Banking Bank Indonesia. 2014. Analisis Efisiensi Perbankan Syariah. Jakarta (ID): Bank Indonesia.

Hadad M, Santoso W, Mardanugraha E, Ilyas D. 2003. Pendekatan Parametrik Untuk Efisiensi

Perbankan Indonesia. Jakarta (ID): Bank Indonesia.


Hadad M, Santoso W, Mardanugraha E, Ilyas D. 2003. Analisis Efisiensi Industri Perbankan

Indonesia: Penggunaan Metode Non Parametrik Data Envelopment Analysis (DEA). Bank Indonesia, Jakarta.

Hauner David, 2005, Aging: Some Pleasant Fiscal Arithmetic, IMF Working Papers 05/71, International Monetary Fund.

Herlambang, Sigit. 2011. Analisis tingkat efisiensi faktor-faktor operasional BRI dan BTN [tesis]. Bogor (ID): Institut Pertanian Bogor.

Lang, G. and P. Welzel. 1996. “Efficiency and Technical Progress in Banking: Empirical Results For A Panel of German Cooperative Banks”. Journal of Banking and Finance, 20: 1003-1023.

Magrianti, Tessa. 2011. Analisis perbaningan efisiensi bank umum syariah dengan bank umum konvensional di Indonesia. Bogor (ID): Institut Pertanian Bogor.

Nadratuzzaman, Muhamad. 2013, Produk Keuangan Islam di Indonesia dan Malaysia. Jakarta (ID): PT Gramedia Pustaka Utama.

Pass C, Lowes B. 1997. Collins Kamus Lengkap Ekonomi. Ed ke-2. Jakarta (ID): Erlangga.

Pratama, Fajar P. 2011. Analisis efisiensi bank umum persero di Indonesia. Bogor (ID): Institut Pertanian Bogor.

Ramli, Mahyuddin, 2005. Studi tingkat rfisiensi bank komersial di Indonesia dan beberapa faktor penentu [desertasi]. Jakarta (ID): Universitas Indonesia.

Srivastava P. 1999. Size, Efficiency, and financial reforms in Indian Banking. [Working Paper]. New Delhi, India (IN): Indian Council For Research On International Economic Relations. No. 49.

Sufian, Fadzlan. 2006. The efficiency of islamic banking industry: a non-parametric analysis with non-discretionary input variable. Islamic Economic Studies. 14 (2): 147-175.

Suswadi. 2007. Analisis efisiensi perbankan syariah di Indonesia (metode stochastic frontier approach). Jakarta (ID): Universitas Indonesia.

Thangavelu, Mugan S, Findlay, Charles C. 2010. Bank efficiency, regulation, and response to crisis of financial institutions in selected Asian Countries. Economic Research Institute for

ASEAN and East Asia (ERIA) Research Project Report 2009: 288-315.

Yudistira, Donsyah. 2003. Efficiency in Islamic Banking: An Empirical Analysis of 18 Banks. [Paper]. United Kingdom (GB): Loughborough University.

MANUSCRIPT TERMS OF DELIVERY

1. The manuscript must be an original work of the author (individuals, groups or institutions) that do not infringe copyright. The submitted manuscript has never been published and is not being sent to other publishers at the same time. The copyright of the received manuscript is the author’s.

2. Each approved manuscript will receive Rp 5.000.000,- financial compensation.

3. The manuscript template can be downloaded at www.journalbankindonesia.org/index.php/BEMP

4. Manuscripts can be submitted in softcopy (file). We strongly suggest to send your softcopy to:

[email protected] (Cc. to: [email protected].)

5. Manuscripts can also be sent via www.journalbankindonesia.org

6. Mathematical equations and symbols are written using Microsoft Equation.

7. Each manuscript should include abstract, a maximum of one (1) page size A4. For Indonesian manuscript, the abstract should be in English and vice versa.

8. The manuscript must include keywords and two-digit classification numbers of the Journal of Economic Literature (JEL). Please refer to JEL classification, http://www.aeaweb.org/journal/jel_class_system.html.

9. The manuscript must include author’s bio (address, phone number, bank account and e-mail). It is suggested to write the biographical data in the form of a complete CV (curriculum vitae).

102 Buletin Ekonomi Moneter dan Perbankan, Volume 19, Number 1, July 2016
