Agricultural Development Bank Reform:
The Case of Unit Banking System of
Bank Rakyat Indonesia (BRI)
Hans Dieter Seibel
Rural Finance Adviser, IFAD
IFAD – INTERNATIONAL FUND FOR AGRICULTURAL DEVELOPMENT
IFAD's thematic working papers disseminate the findings of work in progress to
encourage the exchange of ideas. The authors are IFAD staff, consultants, and
members of IFAD projects or partner organizations. The papers are unofficial
documents which carry the names of the authors and should be cited accordingly.
The findings, interpretations, and conclusions expressed in these papers are
entirely those of the authors. They do not necessarily represent the view of IFAD,
its Executive Directors, or member countries .
Table of contents
1 INTRODUCTION ...............................................................................................................................1
2 THE REFORM OF THE BRI UNITS ..............................................................................................2
2.1 The era of supply-leading finance (1969 to 1983): BRI as agent of agricultural development .....2
2.2 The „big bang reform“ in 1983.......................................................................................................2
2.3 The evolution of the new BRI Unit Banking System (1984 to 2000) ............................................2
3 LESSONS LEARNED FROM BRI’S REFORM EXPERIENCE................................................10
3.1 Prerequisites for the big-bang reform in 1983 ..............................................................................10
3.2 Safe-guarding the Units’ operational autonomy ...........................................................................10
3.3 Success factors and lessons learned from BRI’s own perspective ...............................................12
4 FUTURE CHALLENGES AND REFORM AGENDA AHEAD ..................................................13
4.1 Expansion of lending ....................................................................................................................13
4.2 The Unit Banking System as „bank“ within the bank BRI...........................................................14
4.3 Lack of Competition .....................................................................................................................15
The case of BRI is evidence that, in a deregulated policy environment, a government-owned
agricultural development bank can (a) be transformed into a highly profitable, self-reliant financial
intermediary, and (b) turn into a major microfinance provider, offering carefully crafted
microsavings and microcredit products to low-income people at market rates of interest. Making
good use of government seed money and a World Bank loan during an initial phase, it has now fully
substituted savings deposits for external loans as its source of funds. With an outreach to 25 million
saving accounts and 2.7 million borrowers through a network of over 3,700 banking units operating
as profit centers, BRI covers its costs from the interest rate margin and finances its expansion from its
profits. With non-targeted loans from $5 to $5000 at rural market rates of interest and unrestricted
deposit services, the BRI Microbanking Division has weathered the Asian financial crisis well,
making BRI the only profitable government bank in Indonesia.
Several lessons can be drawn from BRI's experience:
Financial sector policies work and are conducive to financial innovations
With attractive savings and credit products, appropriate staff incentives, and an effective system
of internal regulation and supervision, rural microfinance can be profitable
The poor can save; rural financial institutions can mobilize their savings cost-effectively
If financial services are offered without a credit bias, demand for savings deposit services
exceeds the demand for credit by a wide margin
Incentives for timely repayment work
Outreach of a financial institution to vast numbers of low-income people and profitability are
After a big-bang-type of reform in 1983/84, the BRI Units soon became a model case in Asia of the
transformation of an ailing government-owned agricultural development bank into a viable and self-
sufficient financial intermediary with ever-increasing financial resources and numbers of customers,
competing successfully with an array of other local financial institutions. Further strength was added
to BRI’s microfinance operations during the Asian financial crisis: When the Indonesian banking
system collapsed, BRI’s Unit Banking System remained profitable, the loan repayment high, and the
deposit volume more than doubled. The BRI Units emerged from the crisis stronger than before.
BRI Unit Banking System
Performance Indicators 2000
Number of active borrowers 2.7 million
% of households %
Loan amount outstanding US$ 816 million
Average loan size (disbursement) US$ 425
Number of deposit accounts 25.8 million
Deposit volume US$ 1,992 million
Average deposit size US$ 77
Market share rural deposits (1998) 86 %
Portfolio quality (Past dues/Loans outstanding) 2.5 %
Operational self-sustainability (Income/Operational cost) 413 %
Financial self-sustainability (Income/Operational cost and Cost of funds) 135 %
Profitability Return on Assets 5.4 %
Self-sufficiency in funds (Deposit to loans ratio) 244 %
Efficiency (Administrative cost/Average loans outstanding) 8.4 %
Productivity (Number of active borrowers/Credit officer) 534
AGRICULTURAL DEVELOPMENT BANK REFORM:
The Case of Unit Banking System of Bank Rakyat Indonesia (BRI)
Bank Rakyat Indonesia (BRI) is a state-owned commercial bank in Indonesia. For the past thirty
years, BRI’s mandate has been the provision of banking services to the rural and agricultural sector.
BRI's physical network is the most extensive of all banks in Indonesia and effectively covers the
entire nation. BRI has 318 branches that are located in the district capitals and municipalities.
Furthermore, the bank has an even more extensive network of 3,724 retail outlets at the sub-district
level, known as the BRI Units1. Although both branches and Units are part of the same bank, the Unit
Banking System can be clearly distinguished from the branch network in terms of target groups,
services and mode of operation.
It is the BRI Unit Banking System that won the international reputation as a successful rural financial
institution and more recently as the developing world’s largest microfinance institution. The Unit
Banking System underwent a substantial reform in 1983. In subsequent years, it was transformed from
a chronic loss maker to a highly profitable rural banking network. Most surprisingly, the BRI Unit
Banking System weathered the severe financial crisis in the late 1990s better than any other financial
The experience of the BRI Unit Banking System, the success factors and lessons learned have been
extensively documented elsewhere.2 This paper focuses on the reform process and reform steps, and
the lessons which can be learned by other banks in Asia and elsewhere. Section two presents a review
of the various phases in the evolution of the new BRI Unit System and an analysis of the impact of the
financial crisis on Unit operations, deposit mobilization and lending. Section three draws some of the
lessons learned from BRI’s experience. Finally, the fourth section identifies and outlines a few critical
areas for future reform.
Formerly the Units were called Unit Desa (desa = village). However, the Units are actually not located
at the village level but mostly in sub-district capital towns. More recently, many units have been opened in urban
areas as well.
Among others: Patten and Rosengard (1991), Robinson (1992, 1994), Yaron (1992), GTZ (1997), BRI
2. THE REFORM OF THE BRI UNITS
2.1 The era of supply-leading finance (1969 to 1983): BRI as agent of agricultural
Established more than 100 years ago during the Dutch colonial period in 1895 as De Poerwokertosche
Hulp-en Spaarbank der Inlandsche Hoofden, the bank underwent several changes in its subsequent
evolution. Following the independence of Indonesia, the bank was named Bank Rakyat Indonesia -
Indonesian People’s Bank - and became the first government bank covering the entire country. Under
the New Order Government of former president Suharto and the first Five-Year Development Plan
which commenced in 1969, BRI was assigned an active role as agent of development with primary
responsibility for the agricultural sector. In the spirit of the Green Revolution, a large credit program
with subsidized interest rates for rice farmers - called BIMAS - was designed to promote national rice
self-sufficiency. As BIMAS was to be implemented by BRI, the bank prepared itself by establishing
more than 3,600 field units - then called Unit Desa - all over Indonesia in the early 1970s. The
primary function of the Unit Desa was to channel cheap credit to farmers.
The operating costs of the program were high, and as the interest rates were fixed by the government
at low levels, BRI was saddled with annual operational losses. Hence, the implementation of BIMAS
could only be sustained through administrative subsidies from the Ministry of Finance. Furthermore,
the loan recovery gradually deteriorated and the capital stock eroded which required frequent
injections of fresh funds. In the early 1980s, the BIMAS lending volume declined, loan defaults
climbed above 50% and the program was rapidly approaching its collapse.
2.2 The big bang reform in 1983
BIMAS’ deterioration coincided with important events in the overall economic and political fields in
early 1983. On the macroeconomic front, oil prices collapsed and the sharp decline in oil revenues put
heavy strains on the government budget. On the political front, a new cabinet and most important a
new finance minister took office. Under this new constellation, the government launched some major
economic reforms. A major financial sector reform package was announced in June 1983. After
decades of financial repression, the measures were a decisive step towards deregulation of Indonesia’s
financial system. The most important element of the package was the liberalization of interest rates on
both deposits and commercial loans.
In the light of the new policies it also became clear that the massive support to the BIMAS program
cannot and will not be sustained by the government. When BIMAS was phased out at the end of 1983,
and with it the subsidies to unit operations, BRI was faced with a difficult choice: either introduce
drastic measures to increase the viability of the units, or close them down. With the encouragement of
the Ministry of Finance, BRI decided to convert the Unit Desa into a rural banking network that
would meet a wide range of financial needs in a flexible and financially viable way.
2.3 The evolution of the new BRI Unit Banking System (1984 to 2000)
Phase 1 (1984 to 1986): Establishing the new system
The task was enormous: 3,626 BRI Units had to be converted from loss-making BIMAS conduits to
profitable, full-service rural banking units. A massive restructuring of the entire Units was necessary.
Almost one third of the Units did not have the potential of reaching viability within a foreseeable
period of time. However, they were not closed down, but were designated village service posts (PPD),
kept open for one to six days a week depending on the volume of customers and transactions. Today,
many of the PPDs have been upgraded to full units and the number of units is roughly the same as it
was at the outset. Furthermore, many Units were physically relocated from out-of town field sites to
town centers and close to the markets.
One of the first steps taken was the reorganization of the Units into profit centers and the creation of
separate balance sheets and profit and loss statements for each Unit. A new accounting system,
focusing on transparency, allowed each Unit’s performance to be evaluated. A standardized
management information system centered around few but significant performance indicators was to
provide timely information to managers and supervisors at all levels. It was on this foundation that a
staff incentive system were built.
In February 1984, a single loan product for general rural credit (KUPEDES) was introduced.
KUPEDES is non-targeted and non-subsidized, and is available to any creditworthy customer for any
kind of productive enterprise. Loan terms are up to 24 months for working capital and up to 36
months for investment capital. KUPEDES interest rates were set at 1.5% flat per month for working
capital loans and 1% for investment loans. The flat rates are calculated on the borrower's original
balance and work out to annual effective rates of 33% and 22% respectively, 3 if loans are paid on
time. KUPEDES borrowers must provide sufficient collateral to cover the value of the loan, usually in
the form of land titles, but also by the pledging of buildings, motorcycles or other property. The
maximum loan size has gradually been raised over the years from US$ 1,000 at the start to around
US$ 5,000 (Rp 50 million) today.
Savings mobilization has been an integral part of the Unit banking philosophy and strategy since the
beginning. The objective of creating a viable and financially self-sufficient network of financial in-
termediaries was based on the belief that the large demand for loans can be financed by local-
ly-mobilized savings. In November 1984, BRI introduced SIMPEDES, a rural savings product, on a
pilot basis. The pilot project quickly showed evidence of massive demand for a liquid, convenient and
safe deposit facility. After some modifications and refinements, the facility was successively expand-
ed to all Units by September 1986. Since then, SIMPEDES has developed into the main product ac-
counting for about two thirds of total deposits. In addition, the BRI Units offer demand and time de-
posits, as well as TABANAS, another passbook savings product originating from the pre-deregulation
era, which is mainly used by school children and students.
Based on a twelve month loan with monthly installments of equal principal amounts according to the
following formula: [i x 2 x t/(n+1)]x 12, where i=monthly flat rate, t=loan term in months, n=number of monthly
Table 1: Deposits and loans in BRI Units, 1984-2000 (Amount in Rp billion)
Deposits Loans outstanding Deposit to loan
Year Accounts Amount Number Amount ratio
1984 2,655 42 640,746 111 38%
1985 36,563 85 1,034,532 229 37%
1986 418,945 176 1,231,723 334 53%
1987 4,183,983 288 1,314,780 430 67%
1988 4,998,038 493 1,386,035 542 91%
1989 6,261,988 959 1,643,980 847 113%
1990 7,262,509 1,695 1,893,138 1,382 122%
1991 8,587,872 2,541 1,837,549 1,456 174%
1992 9,953,294 3,399 1,831,732 1,649 206%
1993 11,431,078 4,325 1,895,965 1,957 220%
1994 13,066,854 5,232 2,053,919 2,458 213%
1995 14,482,763 6,016 2,263,767 3,191 189%
1996 16,147,260 7,092 2,488,135 4,076 174%
1997 18,143,316 8,837 2,615,679 4,685 188%
1998 21,698,594 16,146 2,457,652 4,697 344%
1999 24,235,889 17,061 2,473,923 5,957 286%
2000 25,823,228 19,115 2,715,609 7,827 244%
Both KUPEDES and SIMPEDES had a successful start, and after three years of operation it became
clear that the BRI unit system could be financially viable. In 1984, only 14% of the Units were
profitable, two years later this share had increased to 72%.
Phase 2 (1987 to 1991):
Big surge in savings and manifestation as the player in rural finance
The next five years saw a major expansion of outreach and a big surge in savings mobilization as
illustrated in Table 1. During this period, the number of savings accounts grew rapidly to almost 8.6
million and - at a slower pace - the number of borrowers to almost 1.5 million. This dimension of
outreach clearly established the BRI Unit system as the major player in rural finance in Indonesia.
The development of the BRI Units during the second phase received support from a conducive policy
environment. In October 1988, the government announced a set of reform measures known under the
acronym PAKTO which opened entry in the financial markets to new participants and aimed at
deliberate promotion of competition among banks. New financial services such as factoring, consumer
credit, venture capital, and securities dealing and underwriting were opened up to commercial banks.
Banks were allowed to introduce their own savings products, and the minimum reserve requirement
was lowered to a uniform 2% on all deposits. This caused a fierce competition among banks and
resulted in an outright explosion of bank deposits in the following years.4
The BRI Units particularly benefited during the boom years. The Units added one million new
depositors every year and the deposit volume increased annually by 80% on average. In the sixth year
of operation in 1989, the BRI Unit system achieved self-sufficiency in funding when the volume of
deposits was equal to the loan amount outstanding. As deposits continued to rise and lending
Total bank deposits in Indonesia increased from US$16 billion in 1987 (before the reform) to US$118 billion
operations did not expand at the same pace, the BRI Units soon experienced a situation of ever
increasing excess liquidity - a situation which has prevailed for the past ten years in varying degrees
until today. The excess liquidity was transferred from the Units to the BRI branch system
Another reform package came in January 1990, when the Central Bank announced the immediate
termination of 34 government-sponsored credit programs, and with it, the phasing out of refinance
facilities which put increased pressure on banks, especially state banks, to search for their own funds.
The situation was further aggravated in February 1991 when the finance minister withdrew an amount
of US$ 4.2 billion in deposits from the banking system (so-called Sumarlin-shock) to curb the growth
in money supply and counter the danger of overheating of the economy.
BRI as a whole suddenly faced a liquidity problem. The management reacted swiftly and imposed a
general credit stop. This policy decision had considerable consequences for the BRI Units. For two
years, the Units did not accept any new borrowers, in fact, the number of borrowers with loans
outstanding dropped in 1991 and 1992 (see Table 1). Furthermore, the credit crunch led to a
deterioration of the loan repayment. The repayment rate dropped to 95% in 1991, the arrears climbed
over 9% in 1992 (see Table ), a level not seen in the entire history. The BRI management learned an
important lesson from this unexpected reaction. The reliability of financial services, in particular the
availability of future loans, is an important incentive for borrowers to repay their loans. Any
restriction and disruption may have an immediate effect on the borrowers’ motivation and discipline.
Table 2: KUPEDES repayment performance
Year 12-month Arrears ratio
1984 99.0% 5.4%
1985 98.2% 2.1%
1986 97.3% 4.5%
1987 97.0% 5.8%
1988 95.4% 7.4%
1989 97.7% 5.4%
1990 98.0% 4.1%
1991 95.1% 8.6%
1992 96.6% 9.1%
1993 97.8% 6.5%
1994 99.3% 4.5%
1995 98.9% 3.5%
1996 98.4% 3.7%
1997 97.8% 4.7%
1998 98.1% 5.7%
1999 98.3% 3.1%
2000 98.9% 2.5%
Phase 3 (1992 to 1997):
Stable growth in a conducive environment, improving systems and increasing efficiency
Growth continued throughout the 1990s, albeit at a slower pace than in the first decade. Every year,
the Units added an average of 100,000 borrowers and 1.5 million deposits. Deposits continued to
outstrip loans both in terms of number of accounts and volume. At the end of the third phase, the Unit
Banking System recorded more than 18 million deposit accounts against 2.6 million borrowers, a ratio
of 7:1. BRI’s experience has shown that most rural households are net savers most of the time.
In terms of volume, the deposit-to-loan ratio has been about 2:1. In other words, only half of the funds
mobilized by the Units were recycled as KUPEDES loans, the other half was transferred to the branch
system. For the Units, the fund transfer was attractive as they were paid an interest rate - the so-called
transfer price - which was normally set slightly above the interest rate paid on time deposits. The
branches easily absorbed the Units’ excess liquidity and channeled the funds into large corporate
loans which were rapidly expanded during this period.
The third phase also saw a considerable progress in productivity and efficiency. Staff productivity
was raised significantly. Nowadays, a credit officer is responsible for more than 500 borrowers and a
teller handles about 6,000 deposit accounts on average. As a consequence, administrative efficiency
increased. Administrative expenses as percent of average loans outstanding were brought down from
10% to 8% (see Table below). The gains in productivity and efficiency were partly due to the
computerization of the Units which was completed during this period. Another reason may have been
the staffing standards which were consistently applied and ensured maximum performance: one credit
officer for every 400 borrowers (excluding fixed income loans), one teller for every 200 daily cash
transactions, and one bookkeeper for 150 daily transactions.
Table 3: Staff productivity and efficiency
1993 1994 1995 1996 1997 1998 1999 2000
# active borrowers per credit officer 440 466 503 541 527 478 490 534
# deposit accounts per teller 3,008 3,320 3,598 3,887 4,220 4,962 5,614 6,181
Administrative expenses/average loan n.a. 10.1% 9.7% 9.4% 9.4% 8.4% 8.1% 8.4%
Source: Own calculations on the basis of data from BRI Monthly Report February 2001
Phase 4 (1998 to 2000):
The financial and economic crisis: challenges and opportunities
What started as a currency crisis in Thailand in July 1997, unfolded as the worst financial and
economic crisis for Southeast Asian countries in decades. Indonesia has probably been the country hit
most hardest. The major onslaught of the crisis came in late 1997 and early 1998. Within a few
months, the country's currency - the Rupiah - experienced a dramatic 80% plunge in its value against
the US dollar, followed by sharp increases in inflation (77% in 1998) and interest rates. The ensuing
political crisis mounted in May 1998 with the downfall of president Suharto. The banking system was
on the brink of collapse when the government stepped in and provided a blanket guarantee cover for
all bank deposits. Many private banks were closed down, major state banks merged and non-
performing assets transferred to the newly-created Indonesian Bank Restructuring Agency (IBRA).
It is hard to believe but the BRI Unit Banking System remained largely unaffected by the severe
crisis! Rather the opposite was the case as BRI benefited from its status as government-owned bank
and safe haven for depositors. From early 1998 onwards, the Units experienced a tremendous influx
of deposits as illustrated in Chart 1 below. Within little over a year, time deposits shot up from Rp 1
trillion to Rp 7 trillion. Savings - mainly SIMEPDES/SIMASKOT - continued to grow steadily and
their volume more than doubled in the course of two years. A slight friction was felt in the second half
of 1998 when some depositors shifted from savings to time deposits due to comparatively higher time
Loans stagnated for about two years. During 1998, the number of loans even declined by 6% from
2.61 million to 2.46 million accounts. However, most surprisingly loan repayment was affected only
marginally. The 12-months repayment rate remained at about 98% throughout the crisis years (see
Table 2 above). The arrears increased to a high of 6% for a few months in late 1998 but has come
down to 2.5% at the end of 2000, an extremely low level for the BRI Units in a historical perspective.
However, BRI had to pay its tribute in form of higher loan loss provisions.
Apparently, the micro business sector was not hit as hard by the crisis as medium and large
enterprises. Micro and small enterprises utilize family labor and capital to large extent, and can thus
more flexibly and easily adjust to changes. Furthermore, the agricultural sector, especially tree crops,
performed well during and after the crisis. And finally, the borrowers’ repayment discipline remained
strong despite the hardships of the crisis.
Chart 1: BRI Units during and after the crisis
Savings Time deposits Loans
In mid-1999, the crisis was effectively over for the BRI Units. Time deposits returned to normal levels
as ‘fugitive’ deposits were gradually withdrawn with the stabilization of the financial sector.
KUPEDES lending picked up again and the loan portfolio almost doubled from mid-1999 until end-
2000. However, the crisis has left its scars. The surge of inflation to a rate of 77.6% in 1998 has
dampened the performance in real terms, i.e. after taking into account inflation. Chart 2 presents the
development of deposits and loans in BRI Units in real terms, i.e. in constant prices of 1984. Deposits
exhibit a steady growth path, albeit at a slightly lower rate since the outbreak of the crisis. Loans
marked a sharp decline in real terms in the crisis year. If the current growth continues, the loan
portfolio will likely reach the pre-crisis level at the end of 2001.
Chart 2: Deposits and loans in real terms, 1984 to 2000
(Rp billion in constant prices of 1984)
The impact of the crisis on the viability of BRI Unit operations was only marginal. The profitability
suffered slightly in 1998 and 1999 due to (a) increased loan loss provisions and (b) higher cost of
funds which reduced the financial spread. The income and cost structure of the Unit Banking System
is presented in Table 4 below as percent of average loans outstanding. While the cost of funds move
with the market, the interest rates on KUPEDES loans have remained unchanged since inception. The
only exception was during the crisis when inflation and interest rates shot up and KUPEDES rates
were increased for a period of nine months. BRI Units operate with a comfortable spread of around
20% of average loans outstanding. Operating costs have moved at around 11% on average, leaving a
Table 4: Income and cost structure of the BRI Unit Banking System, 1994 to 2000
1994 1995 1996 1997 1998 1999 2000
(% of average loans outstanding)
Interest income from lending 32.4% 31.6% 30.3% 30.3% 30.9% 35.4% 33.2%
less: Cost of funds 8.5% 9.6% 10.7% 11.6% 12.7% 16.0% 8.0%
= Financial spread 23.8% 22.0% 19.7% 18.7% 18.2% 19.4% 25.2%
Administrative cost 10.1% 9.7% 9.4% 9.4% 8.4% 8.1% 8.4%
add: Loan loss provisions 1.8% 1.3% 1.3% 1.4% 2.9% 5.0% 2.8%
= Operating cost 12.0% 10.9% 10.6% 10.7% 11.3% 13.1% 11.1%
Profit 11.9% 11.1% 9.1% 8.0% 6.9% 6.3% 14.1%
Source: Own calculations on the basis of data from BRI Monthly Report February 2001.
Note: The calculation assumes KUPEDES lending as profit center and uses weighted cost of funds
and administrative cost.
The Unit Banking System as a whole has been profitable since 1986. During the 1990s, the
profitability increased steadily and the Return on Assets (RoA) reached more than 6% as illustrated in
Table 5. Return on Assets includes interest earned on excess funds held with BRI branches which
explains the slightly different profitability pattern compared to Table 4.
Table 5: Profitability of BRI Units
1993 1994 1995 1996 1997 1998 1999 2000
Net income (Rp billion) 140 265 403 423 417 861 1,190 1,160
Average annual assets (Rp billion) 3,344 5,401 6,482 7,680 9,224 14,068 19,487 21,676
Return on Assets (RoA) 3.1% 4.9% 6.2% 5.5% 4.5% 6.1% 6.1% 5.4%
Source: Own calculations on the basis of data from BRI Monthly Report February 2001.
A number of rural financial institutions have reported adequate financial returns in terms of Return on
Assets (ROA) and Return on Equity (RoE), but they depended heavily on subsidies. The favorable
financial ratios of the BRI units are not a result of subsidies but of efficient rural financial
intermediation. Since 1987, Yaron’s subsidy dependency index (SDI) has been negative, reaching –
48% in 1995 (Charitonenko et al.1998)5.
However, the rest of BRI fared much worse during the crisis. Due to heavy loan losses in the
corporate business unit, BRI almost collapsed (see section IV.3 below).
Positive values indicate subsidy dependency: the higher the value, the higher the dependency. Some comparative SDI
figures from donor-driven programs in Nepal, 1991/92: ADBN‘s Small Farmers Development Programme 135%; Intensive
Banking Programme 180%; Production Credit for Rural Women 366%. (Seibel, Pant and Dhungel 1998:4)
3. LESSONS LEARNED FROM BRI’S REFORM EXPERIENCE
A little more than a decade of KUPEDES and SIMPEDES has made the BRI unit system one of the
most successful rural financial institutions in the developing world. Some of the features are certainly
unique and relate to the specific context of rural Indonesia and/or to BRI as an institution, but there
are many others that can be generalized and applied in other countries. The BRI experience has
become a learning ground for policy makers and practitioners from institutions and countries around
3.1 Prerequisites for the big-bang reform in 1983
The timing of the reform was favorable. The timing of the introduction of a savings-based rural
banking system in early 1984 was appropriate as several events coincided. This included among
others: the financial sector deregulation in June 1983 allowing banks to set their own interest rates,
the choice forced by the collapse of the BIMAS credit program, and the declining acceptability of
subsidized programs in general, as well as key personnel changes in the government (new finance
minister) and in BRI (new president director) in 1983. Even after the big-bang reform in 1983, stable
macroeconomic conditions and a further reform of the financial sector provided a favorable and
conducive environment in which the new Unit Banking System could develop and prosper.
The political will to reform the BRI Units was strong. The political will to reform the BRI Units
originated primarily from economic pressures. The oil price collapse in 1983 and the decline in oil
revenues forced the government to impose austerity on budgetary expenses. The government simply
could not afford to continue funding the BIMAS credit program and the continuous losses it created
for BRI. Moreover, the political goal of achieving self-sufficiency in rice production - the rationale of
BIMAS - had been achieved, thus adding pressure to terminate the program. Furthermore, the
blueprint for reform was perfectly in line with the overall financial reforms announced in June 1983,
thus winning the support of the architects of the reform in the Ministry of Finance and the Central
Bank. The following period of transformation and development of the Unit Banking System, too, saw
strong support from the government, there was little or no interference in the Unit operations. Section
2 below explores how the Units managed to maintain operational autonomy.
A farsighted blueprint for a new Unit Banking System was on the table. The big-bang type of
reform which followed required a clear vision of the restructuring and transformation of 3,600 Units
and its 14,000 staff into a fundamentally different system. There was not much time for trial and error.
The blueprint for a sustainable rural banking system had been put on the table by a team of local
bankers and international rural finance experts in 1983. The essential components which today are
regarded as the success factors of the Unit Banking System were already included in the initial
blueprint. The concept was partly based on the positive experience of BRI with two non-agricultural
loan programs (Kredit Mini and Kredit Midi) and to a large extent on the inputs from the team of
international consultants who supported the design and implementation of the new system. Given the
success of the BRI Unit Banking System and given the fact that little was changed in the design for
almost two decades, the original blueprint was farsighted and the choice a right one. An overview of
the main components is presented in Section 3 below.
3.2 Safe-guarding the Units’ operational autonomy
With BRI being a government-owned bank, the issue of operational autonomy gains particular
relevance. As mentioned above, the government has maintained a hands-off position since the
re-design of the Unit Banking System in 1983. How was it possible to keep the Units free from typical
BRI’s International Visitors Program receives study groups from all over the world.
interventions such as credit targeting, interest rate restrictions, provision of cheap funds, etc. or from
interference in lending decisions? Several reasons may account for this:
BRI’s self-reliance in terms of fund mobilization and profitability has created the material base for
its autonomy and freedom from political interference which has so severely afflicted the rest of the
banking system. The Unit Banking System was profitable from the third year (1986) and self-
sufficient in funding from the sixth year (1989) onwards. Except for the initial injection of seed
capital at the tune of US$ 20 million in 1984, there were no further government funds needed. A
World Bank loan of US$ 97 million for onlending was provided at a time when the BRI Unit
Banking System no longer needed outside funds.
Strong leadership has proven a crucial factor. BRI's president director from 1983 to 1992 took
personal initiative and responsibility for the development of the Unit Banking System. Jointly with
the other members of the Board he protected the Units from interference and led the development
of what they called BRI's new culture and Indonesia's new rural banking system.7
The only targeting device applied in the Unit Banking System was a ceiling for KUPEDES loans
in order to focus the Unit operations on the micro business segment. Initially the maximum loan
size was fixed at about US$ 1,000 and was successively raised to US$ 5,000. The relatively small
loan sizes have effectively kept large and politically influential borrowers out of the system,
thereby reducing the probability of interference.
However, it would be naive to assume that BRI as a government-owned bank were free of
government influence and intervention. As agent of development, BRI has always played a role in
the implementation of rural credit programs designed by the government for priority sectors and
target groups until today. However, all these programs are administered by the BRI branches and
are overseen by a ‘political’ division for program credit in the head office. This way, BRI has
always accommodated specific requests and special programs from the government (and from
donors) without disturbing the development of the Unit Banking System. For example, BRI
branches continue to provide subsidized loans to farmers - successor programs to BIMAS - and to
politically-favored village cooperatives, however, strictly separated from the mainstream of Unit
operations. The lesson to be learned by government-owned banks is clear: As government
intervention cannot be avoided the bank could (and should) accommodate such programs in a
special unit, but keep the mainstream free from intervention.
Robinson 1992, pp.48-49.
3.3 Success factors and lessons learned from BRI’s own perspective
Key principles of the success of the Unit Banking System and Lessons Learned
During the past decade and a half, BRI has learned many valuable lessons in moving away from a
targeted, subsidized rural finance program into a full-service, fully commercial microfinance
institution. Some of the most important of these lessons are:
In order to minimize overhead costs, credit should be made available to all creditworthy
microcredit customers. If a bank can serve only one particular segment, such as agriculture, staff
and other overhead costs are too high. The bank should be able to lend for any productive
Credit decisions should be based solely on a borrower’s willingness and ability to repay - and
this decision should be left to the bank. The amount of the loan should be directly related to
need and the assessed repayment capacity of the enterprise and borrower. If the decision is made
by agricultural extension workers of other government officials, the creditworthiness criterion is
diluted. The borrower should know that loan decisions are based solely on the bank officer’s
judgement of creditworthiness, and that additional credits will be available if earlier loans are
repaid according to agreed terms.
Access to credit on a long term basis is more important than cheap credit . Interest rates
should be set to cover the costs of the bank so that the bank is financially viable, and can
therefore provide credit over the long term as needed. Consistency and sustainability are far
more important to borrowers than cheap credit programs which fade away when the government’s
budget is tight or donor fashions change.
Almost all households want to and are able to save, and serving their need to save is just as
important as serving their demand for credit. There is a large potential for savings among
relatively low-income people. However, savings instruments must be designed to meet the needs of
savers. People need a safe place to keep their savings. They need liquid savings instruments for
their routine enterprise and household financial requirements. They need to have funds available to
meet unforeseen emergencies as well as to meet regular, relatively large expenses such as school
fees and uniforms.
Operating a large network of small banks efficiently requires a combination of simplicity
and transparency, standardization, and delegation of authority and responsibility. The
cornerstone of the Unit Banking System’s transparency leis in the simplicity of the credit and
savings instruments offered and the Units’ financial statements. Savings and loan terms, interest
rates, reporting, training, and procedures are all standardized. With this simplicity and
standardization, it’s possible to create an efficient well-supervised system which delegates
authority to the local level. Thus, both the responsibility and the authority for amking good loans
and earning a profit lie at the local level.
Adopted from: Bank Rakyat Indonesia (1997): Introduction to BRI’s Unit Banking System, page 19
4. FUTURE CHALLENGES AND REFORM AGENDA AHEAD
Despite the overall success with the reform of the BRI Unit banking System over the past one and a
half decades, some challenges remain to be tackled in the near future. Three critical issues remain to
be addressed in the following.
4.1 Expansion of lending
A critical issue is the financial drain occurring within the BRI unit system. Since 1989, savings
mobilization has exceeded KUPEDES lending, which has led to a considerable transfer of funds to the
urban BRI branch network. Throughout the 1990s, more than half of the savings mobilized by the
units were deposited in the branches. The financial crisis has widened the gap even further in 1998
when 70 % of the deposits were channeled off to the BRI branches.
Despite BRI’s achievements in lending outreach, there is still a large unmet demand for loans in rural
Indonesia. With 2.7 million borrowers out of a population of more than 200 million - with the
majority living and working in rural areas, BRI so far has likely reached the smaller part of the
potential customers in all sectors of the economy. The distribution of KUPEDES loans outstanding by
economic sector show a clear concentration of loans in small trade (42%) and for fixed salary
employees (30%). Only 22% of the loans were provided for agriculture and just 2% for small
Table 6: Distribution of KUPEDES loans by economic sector (December 2000)
Economic sector Number Loan amount Share
of borrowers (US$ million)
Agriculture 589,834 177.0 22 %
Industry 66,329 19.2 2%
Trade 1,154,838 338.6 42 %
Fixed salary employees 817,041 242.9 30 %
Other 87,566 38.1 5%
Total 2,715,608 815.8 100 %
Source: BRI Microbanking Division
Given the potential of Indonesia’s rural economy and the likely credit demand of micro and small
businesses, the number of borrowers could probably double or triple, especially in agriculture and
agriculture-related activities. BRI Units may also consider to enter new segments such as housing and
consumer finance. Also, the time will come when a single loan product - KUPEDES - may no longer
be adequate to serve an increasingly diversified and heterogeneous clientele, when new products and
services are needed. The planned introduction of online banking services is an important step towards
the integration of the Units and their customers into the wider payments system.
However, on the supply side the expansion of lending is primarily a function of number of staff,
especially loan officers. The number of loan officers has not risen substantially over the past decade
and has remained almost unchanged since 1997. At the end of 2000, there were 5,082 loan officers
each serving an average of 534 borrowers (374 microenterprises and 160 fixed salary employee
loans). Hence, they have reached a productivity ceiling and there is not much scope to increase the
number of borrowers any further without sacrificing the service quality and loan quality. An
expansion of borrower outreach can therefore only be achieved with additional staff. A significant
increase in the number of loan officers would be an important step to reduce the financial drain, to
narrow the gap between deposits and loans and to strengthen local financial intermediation.
4.2 The Unit Banking System as „bank“ within the bank BRI
Another critical point relates to the role of the Unit Banking System within the overall organization of
BRI as a commercial bank. The transformation of the Units from a chronic loss-maker to a highly
profitable entity has also changed its role within BRI. In the 1970s and 1980s, the Units were totally
dependent on the mother institution in terms funding and operational subsidies. Within little more
than a decade, this has changed to the opposite. In the 1990s, the Units and the microbanking business
were generating the profits to sustain the rest of BRI! The rest of BRI, i.e. the branch network, has
been running losses since several years - a fact well hidden in consolidated figures.8 It was solely due
to the Units’ performance that, on a consolidated basis, BRI posted a positive net income until 1997
(see Table 7). Needless to mention that the bill was paid by the Unit customers, the micro borrowers
Table 7: Profitability of BRI Units versus BRI branches (Net income in US$ million)
1993 1994 1995 1996 1997 1998 1999 2000
BRI (consolidated before tax) 57 67 111 145 30 -3,308 -243 n.a.
a. Branch network -10 -54 -64 -33 -60 -3,397 -411 n.a.
b. Unit Banking System 66 121 174 178 90 89 168 121
The bomb exploded in 1998 when BRI was hit by the financial crisis. 56% of the branches’ loan
portfolio - large corporate loans for the major part - turned sour and had to be written off. This
resulted in a loss of US$ 3.3 billion against a capital base of only US$ 215 million. In 1998, the
consolidated BRI - including the Unit banking System! - was technically insolvent and effectively
bankrupt, like many other banks in Indonesia at the time. Being an integral part of BRI suddenly
turned into a major risk for the BRI Unit Banking System, the risk of being dragged into bankruptcy
by the ailing mother organization. However, the government stepped in to recapitalize BRI; the non-
performing loans were transferred to the Indonesian Bank Restructuring Agency (IBRA).
Table 8: Microbanking „market shares“ within BRI
1997 1998 1999
Savings Deposits 69 % 73 % 70 %
Total Deposits 37 % 40 % 43 %
Net Loans 14 % 21 % 25 %
Assets 25 % 53 % 67 %
Source: Own calculations on the basis of data from BRI Monthly Report
February 2001and BRI Annual Reports 1998 and 1999
Until 1997, the microbanking section had grown into a substantial business unit of BRI already but
the crisis and the faltering corporate business gave it major boost. At the end of 1999, the Unit
Banking System accounted for the major share in savings deposit volume (70%), almost half of all
deposits (43%) and micro loans (KUPEDES) made up one quarter of BRI’s net loans. The Units
brought in two thirds of BRI’s assets, although this ratio declined in 2000 when the recapitalization
bonds were issued and fresh money came in.
See also GTZ (1997) for critical comments on this issue.
Adopted as a sick baby in 1984, the Unit System recovered quickly with the right medicine and
developed into a healthy child. Today, the child has grown up and matured. The microbanking
business is not only able to stand on its own feet but has nurtured the mother organization for most of
the past decade. And when BRI was severely affected by the financial crisis, the very existence and
success of the Unit Banking System has convinced policy makers to save and recapitalize BRI. 9 Now
the responsibility lies with the rest of BRI, the branch network, to carry out fundamental reforms. In
this regard, much can be learned from the child, i.e. the Unit Banking System.
4.3 Lack of Competition
A third critical point relates to BRI’s market position in rural finance. The BRI Unit Banking System
has operated under little or no competition and has thus manifested a monopolist position in the rural
areas of Indonesia. The concern is that this may result in distortions of the market with negative
consequences for the customers and other market participants.
The financial reforms in Indonesia over the past two decades have resulted in a tremendous expansion
of financial institutions and services. Following the financial sector reform package of 1988, more
than 2,000 small, private rural banks (BPR) were established at the sub-district level, often close to
the BRI Unit. However, the private BPRs had to struggle hard to compete with the government-owned
BRI Unit in deposit mobilization as the latter offered government-guaranteed safety of deposits.10
Hence, BPRs are forced to offer higher returns on deposits 11 and better service, e.g., doorstep
collection of savings, in order to attract depositors. Furthermore, most BPR are independent unit
banks with no recourse to vertical risk diversification and with no access to a liquidity pool like the
BRI, exposing them to much higher risks and costs. This situation has placed the BPRs at a
comparative disadvantage vis-a-vis the BRI Units.
Despite growing competition in some areas, the BRI Unit Banking System has retained its dominant
position in the rural financial system in Indonesia. In terms of rural deposits, BRI is clearly a
monopolist accounting for almost 80% of the accounts and 86% of the deposit volume (see Table 9).
In the rural credit market, the BPRs were able to gain ground achieving a market share of more than
30% while BRI’s position with more than half of all borrowers has remained strong. In terms of
number of borrowers, the village credit institutions (BKD) emerge as a strong third player. The fact
that they account for just 2% of the loan amount is an indication of the very small loan sizes of BKDs.
It should be mentioned, however, that the BKDs are supervised and therefore effectively controlled by
BRI which puts BRI in an even stronger position.
Table 9: Market shares of rural financial institutions (June 1998)
Rural financial institutions Savings Deposits Loans outstanding
Type of institution Number Accounts Amount Number Amount
BRI Units1 4,085 78 % 86 % 52 % 69 %
Rural banks (BPR) 2,227 19 % 13 % 32 % 30 %
Village credit institutions (BKD) 5,345 3% 1% 16 % 2%
Total 11,657 100 % 100 % 100 % 100 %
Sources: BRI: Monthly monitoring data; BPR and BKD: Bank Indonesia 10/1998
Data for the number of BRI units include village posts
Deposit amount includes savings and time deposits; for BPR and BKD includes equity.
Other state commercial banks were merged into one, i.e. Bank Mandiri.
The BRI Units clearly demonstrated their competitive advantage as government-owned bank by posting the
national flag in front of every Unit.
The deposit rates of BPR are frequently 3-5 percentage points above SIMPEDES rates.
The market shares presented in Table 9 are based on mid-1998 data and, hence, do not yet reflect the
impact of the crisis. While the crisis has strengthened the BRI Unit Banking System even further as
outlined earlier, it has badly affected the private rural banks. Many BPRs have already been closed
and many more will have to be liquidated in the near future. The sad consequence from a market
policy viewpoint is that the crisis has stifled the slowly emerging competition in the rural financial
markets of Indonesia.
From the BRI Units’ viewpoint, this dominant position is undoubtedly beneficial - the profit margins
are a clear indication. However, in the medium and long run the exposure to competition is a
necessary prerequisite to maintain and continuously improve the quality of the financial services and
the efficiency of the Unit operations. Competition ensures that efficiency gains are passed on to the
customers and are not used for cross-subsidizing bad performance and inefficiency in other parts of
the organization as it happens in BRI. In this regard, policy makers and regulators have a
responsibility for creating the enabling environment.
Finally, competition is the driving force for innovations - an aspect which may need to be addressed
by the BRI Unit Banking System. The initial blueprint which has been the centerpiece of the success
story of the BRI Unit Banking System may require innovative modifications and adaptations to the
rapidly changing environment in Indonesia, in Asia and in the global economy.
Bank Rakyat Indonesia (BRI), Annual Reports, Jakarta 1994, 1995, 1998, 1999
BRI (1997): Introduction to BRI’s Unit Banking System. Jakarta
BRI Head Office, Laporan Bulanan BRI Unit, February 2001. Jakarta
Charitonenko, Stephanie, Richard H Patten, and Jacob Yaron (1998): Indonesia: Bank Rakyat
Indonesia – Unit Desa 1970-1996. Sustainable Banking with the Poor - Case Studies in
Microfinance. Washington DC, The World Bank
Deutsche Gesellschaft fuer technische Zusammenarbeit (GTZ). (1997). Comparative Analysis of Sav-
ings Mobilization Strategies - Case Study of Bank Rakyat Indonesia.
Robinson, M., Rural Financial Intermediation: Lessons from Indonesia, Part One: The Bank Rakyat
Indonesia: Rural Banking 1970-91, Development Discussion Paper 434, Cambridge, MA:
Harvard Institute for International Development 1992.
Robinson, M., Savings Mobilization and Microenterprise Finance: The Indonesian Experience, in: M.
Otero and E. Rhyne (eds), The New World of Microenterprise Finance, West Hartford:
Patten, R.H. and J.K. Rosengard, Progress with Profits - The Development of Rural Banking in
Indonesia, San Francisco, International Center for Economic Growth 1991.
Seibel, H.D. (1989): Finance with the Poor, by the Poor, for the Poor. Financial Technologies for the
Informal Sector, with Case Studies from Indonesia. Social Strategies vol. 3 no. 2. Basel
Seibel, H.D., & Uben Parhusip (1998):Microfinance in Indonesia. An Assessment of Microfinance
Institutions Banking with the Poor. Economics & Sociology Occasional Paper No. 2365.
Rural Finance Program, Dept. of Agricultural Economics, The Ohio State University,
World Bank, KUPEDES: Indonesia's Model Small Credit Program, Operations Evaluation
Department, OED Précis, Number 104, February 1996.
Yaron, J., Successful Rural Financial Institutions. World Bank Discussion Paper 150. Washington,