East Asian Financial Crisis:

Indonesia’s Weak Banking Sector

by

Ryan Pitylak

University of Texas-Austin

December 9^th, 2005

Abstract

Introduction

The banking sector consisted of many new small undercapitalized banks after deregulation in 1988.  Undercapitalization, and low reserve requirements, made banks particularly susceptible to liquidity crises.  The banking sector was not ready to handle the large levels of financial inflows, which led to overinvestment in the corporate and real estate markets.  Some banks were owned by local conglomerates, and these banks would lend money to the conglomerates at levels above legal requirements.  Poor governance existed because there was an information gap between what the bank managers knew about their loan portfolio and what the supervisory organization knew. 

The private debt levels increased dramatically during the 1990’s.  A lot of this debt was denominated in United States dollars, and when the rupiah (the Indonesian currency) depreciated against the dollar, the cost of this debt became burdensome.  Increasingly, loans were made for projects that were too risky.  Bailouts were required by the Bank of Indonesia to keep the banking and corporate sector from failing.  These bailouts led to moral hazard problems from borrowers, commercial banks, and government banks.  Borrowers did not believe that the government banks were aggressive about collecting loan payments, so some borrowers did not service their debts.  Commercial banks attracted depositors at interest rates above what they were able to receive from lending.  Government banks knew that they would be bailed out by the government, so they invested in very risky projects and had a very high non-performing loan rate.

            When the decision to float the rupiah was announced, the rupiah depreciated significantly against the dollar.  This led to a bank run so that deposits could be exchanged into dollars, which caused the rupiah to depreciate further.  A lack of confidence in the banking sector, due to poor transparency of the government, hurt the already weak banking sector.  Banks ran into liquidity crises when bank runs ensued because reserve requirements were too low.

Banking Sector’s Structure

The way in which financial liberalization was handled in Indonesia had a devastating impact on the economy.  Several factors lead to the financial turmoil during the East Asian Financial Crisis, but one main component was a weak banking sector.  Before financial liberalization, the banking community consisted of five large state owned banks (Giancarlo Corsettia, Paolo Pesentib, and Nouriel Roubini (1999)).  To get a picture of what happened to Indonesia, some critical statistics that are not included later in this paper are available in Table 1.

Financial liberalization followed several deregulation policies during 1988.  The impact of this deregulation was substantial because the number of banks increased from 111 in 1988 to 240 sometime between 1994 and 1996 (Charles Enoch, Olivier Frécaut and Arto Kovanen (2003)).  The new banks were undercapitalized, and were therefore exposed to liquidity crises. 

            One problem with the structure of this deregulation was that local banks were able to be setup by local conglomerates (Enoch, Frécaut and Kovanen (2003)).  Local conglomerates systematically over lent to their own corporations.  The corporations undertook projects that were too risky.  If the corporations were not tied to the banks, they probably would not have received the financial capital for these projects.  Taking depositors’ money and using it for risky projects was indicative of a banking system that would not last.  Furthermore, the projects undertaken by the corporations were in large scale projects that were given to them through exclusive contracts.  This exclusivity is another example of inefficiency in the Indonesian market.  These projects required the corporations to have large external debt levels because the money needed for these projects came from foreign banks.  This money was provided to the domestic Indonesian banks at high domestic interest rates.  The banks that were owned by conglomerates ignored legal lending requirements, and provided their conglomerates with levels of capital above and beyond what they should have received (Christian Chua (2005)).  The reserve levels were too low to be able to offset any potential problem that would come up, such as a depreciation that would make the external debt more expensive.  The dollarized external debt was made very expensive after the depreciation of the rupiah against the dollar because most of the debt was denominated in United States dollars.  The external private debt levels rose from $12.4 billion in 1989 to $80.1 billion in 1997 (Chua (2005)).  The high external public debt levels were a point of exposure for the financial market.

One of the main reasons for the collapse of the banking system in Indonesia was high levels of unhedged short-term financial liabilities.  After the currency depreciated for several reasons, big businesses dollarized some of their assets to decrease the effect of the deprecation.  This depreciated the rupiah further, which led to large scale capital flight.  The banks were not able to handle this scale of capital flight because capitalization was low.  “Lack of appropriate macroeconomic management, regulation, and governance ... induced [banks] to increase their external debts” (Chua (2005)).  Furthermore, banks did not have enough experience to handle market fluctuations (Pablo Bustelo (1998)).  These problems helped to create a weak banking sector.

A problem with the way the banking system was setup after financial liberalization is that there were no clear exit plans in place for failing banks (Enoch, Frécaut and Kovanen (2003)).  The typical modus operandi of the government was to have a failing bank merge with another bank, or for the government to provide other forms of support.  The government was hesitant to shut down banks because it did not want to show any weakness in the financial sector (Corsettia, Pesentib, and Roubini (1999)).

The deregulation of the financial market was coupled with an opening up of the capital account.  When the capital account was opened, foreign investors moved money to Indonesia, which created an excess supply of capital.  The savings and investment rates were already high in Indonesia, and capital account liberalization led to overinvestment (Bustelo (1998)).  Overinvestment leads to risky lending, which eventually creates financial instability. 

Confidence & Bank Runs

Insufficient governance controls, low capitalization levels, and high levels of non-performing loans contributed to a weak banking sector.  Many of these loans were funded in the form of foreign dollarized loans, which created risk if the rupiah depreciated.  Furthermore, the government should have also ensured people that when banks failed that deposits would not be lost; there are several ways to accomplish this, which will be discussed later. 

In November 1997, 16 insolvent banks were ordered closed by the government (Enoch, Frécaut and Kovanen (2003)).  There were no guarantees on deposits, so most depositors lost their money when the banks closed.  This created a crisis of confidence in the financial sector because many people did not want to lose their money if their bank was closed.  Money was shifted into state banks or foreign banks.  By mid-November, it became apparent that “a large number of banks were facing growing liquidity shortages, and were unable to obtain sufficient funds in the inter-bank market to cover this gap, even after paying interest rates ranging up to 75 percent”  (Steven Radelet and Jeffrey Sachs (1998) referencing Indonesia’s second agreement with the IMF).  By December 1997, bank runs had hit hard the monetary base of 154 banks; these banks represented half of the total assets of the banking system (Enoch, Frécaut and Kovanen (2003)).

            Banks in Indonesia were not required to keep high enough capital reserve levels.  The lower the capital reserve figure is for a developing country, the higher the risk that is associated with its banking sector.  Low reserve ratios are problematic because liquidity problems can arise when there is a small deterioration in loan quality.  Also, banks can run into liquidity problems when there are small runs on the bank.  With reserve levels set at 8%, the banks assets needed to fall by only 8% for the bank to become insolvent.  An increase in non-performing loans could easily put a bank in that position.  Reserve levels must be higher in developing countries because there are several factors that subject them to higher risks than developed countries.  These range from large terms of trade fluctuations, large GDP fluctuations, ineffective bankruptcy procedures, short supply of bank supervisors and manager, unenforceable bank sector regulations, and instable political systems.  (George Fane and Ross H. McLeod (2003))

Depreciation, an Expanded Monetary Base, and Interest Rates

                                                                                                                                                                                                                                                                                                                    Before August 1997, the Bank of Indonesia (BI) had a policy of fixing their exchange rate to the dollar.  BI decided to float the exchange rate to avoid regional currency market fluctuations, and to continue with its steady progress towards greater exchange rate flexibility (IMF (1997)).  In August 1997, BI floated the exchange rate, and a sharp depreciation followed; see Figure 1 for details.  The depreciation created fear for policymakers at the BI, and they decided to enact a monetary contraction policy.  This required some money to be shifted from commercial banks to BI.  Effectively, BI took over some of the deposits and reserves of the banks, without assuming any their loans.  This created a liquidity squeeze at the banks, and thereafter commercial banks’ deposits at BI fell by a total of 66% during the month of August alone.  In all, these policies created further instability in an already weak banking sector. (Fane and McLeod (2003))

Interest rates can be raised to combat against excessive lending, or to support the exchange rate.   In August 1997, interest rates increased significantly in an effort to stop financial outflows.  The Indonesian authorities also imposed capital controls to limit foreigners’ access to swap markets so that they could not use that currency for speculative attacks against the rupiah (Morris Goldstein and John Hawkins (1998)).  After this increase, interest rates fell back down to below 20% for the rest of 1997; see figure 2 for more details.  In April 1998, interest rates were bid up even further than during August 1997 because banks were trying to attract capital (Enoch, Frécaut and Kovanen (2003)).

The rupiah deprecated in real terms from 97.20 at 1996’s year-end to 62.40 at 1997’s year-end (Corsettia, Pesentib, and Roubini (1999)).  This is a depreciation of 56% during a one year time period; refer to table 2 for more details.  Table 3 shows the nominal exchange rate movement.  This created a large strain on corporations because they had large levels of unhedged external debt.  The depreciation of the rupiah, along with the rise in interest rates, made it difficult for corporations to service their debt payments.  This increased the level of non-performing loans, which in turn caused problems for the banks.  Liquidity support was necessary so that corporations would not go bankruptcy, which put further strain on the financial market (Warwick McKibbin and Will Martin (1998)). 

The high levels of dollarized external debt created the need for more financial support from BI.  By the end of October 1997, BI had provided 24 trillion rupiah in liquidity support to Indonesian banks (Enoch, Frécaut and Kovanen (2003)).  The support was equivalent to roughly 3.5% of GDP, and caused further depreciation of the rupiah.  In January 1998, the rupiah depreciated drastically and BI provided more emergency support.  Outstanding debt to BI for liquidity support reached 60 trillion rupiah by the end of January 1997 (Enoch, Frécaut and Kovanen (2003)).  This is a substantial figure, because 60 trillion rupiah is equivalent to 7% of the 1997 GDP. 

Monetary policy spiraled out of control when money supply and inflation rate targets were missed. Inter-bank interest rates became very high, moving from 16% in July 1997 to 65% in August 1997.  The money supply was supposed to grow by 7.7% during the nine months following September 1997, but the money supply actually grew by over 92% during that period.  Inflation was targeted at below 20% during 1998.  This was not a realistic target because the money supply grew 26.7% during January 1998 alone.  The Consumer Price Index, a measure of inflation, rose 20.5% during January and February of 1998.  Effective targeting, which was not accomplished, is an essential element of a successful floating exchange rate policy.  This created uncertainty in the financial stability of the country and made the problems during the crisis worse.  (Fane and McLeod (2003))

Government Guarantees, Moral Hazard, and Bad Loans

In January 1998 the Indonesian government provided protection to depositors against bank defaults (Enoch, Frécaut and Kovanen (2003)).  This protection was provided as an effort to protect against the capital outflows, and as an attempt to restore confidence.  The guarantee was provided to both creditors and depositors for claims denominated in rupiah or in foreign currency.  These claims were paid in the form of rupiah at the current nominal exchange rate.  The nominal exchange rate was depreciating rapidly against the dollar during this time period, but regardless, this policy helped to provide some additional confidence to money holders.  People would have preferred to see that the monetary system was under control, and that the banking sector’s problems were resolved, but because these resolutions were not immediately available, the protection was welcomed.

The downside to protection is that moral hazard ensues.  The poor governance associated with Indonesia’s weak banking system became an issue once again.  Moral hazard arose because bank owners and managers found it expedient to invest in risky projects.  They believed that their risk was limited because they believed that the government would provide them with support if the bank became insolvent.  Moral hazard also took the form of depositors who were willing to accept the risk of depositing their money into small banks at extremely high interest rates.  To limit this moral hazard issue, interest rate ceilings were imposed, but there was still enough opportunity to attract depositors to the weaker banks.  The official policy was that interest rates for deposits could not exceed 5% above the rates set by the Jakarta Interbank Offered Rate (JIBOR) (Enoch, Frécaut and Kovanen (2003)).  The JIBOR is a weighted average of interest rates from 18 bank members.  These banks attracted depositors by providing interest rates that were even higher than they were able to receive from their own investments.  It is certain that this created instability in the financial sector.  (Enoch, Frécaut and Kovanen (2003)) 

Another moral hazard problem was that government banks chose to provide loans for projects that were riskier than for projects that the commercial banks were willing to provide loans for.  The government banks knew that the government would bail them out if there were any problems (Fane and McLeod (2003)).  In 1994, the Finance Ministry went so far as to say that it would not permit a state bank to default on its obligations.  The willingness of the government to recapitalize government banks provided to be very costly for the entire financial sector.  To illustrate this problem, the level of non-performing loans held by government banks was 17%, compared to the 5% that were held by commercial banks (Corsettia, Pesentib, and Roubini (1999)).  The money provided to these banks had a cost, and this cost was borne to the financial sector as whole.  Furthermore, because government banks knew that they would be bailed out by the government, they were less aggressive about collecting loan payments (Enoch, Frécaut and Kovanen (2003)).  This information was available to some debtors, which created less incentive for borrowers to service their debts.  After mid-1998, loan performance dropped dramatically (Enoch, Frécaut and Kovanen (2003)).

The problem with the governance system in Indonesia was that “bank supervisors are reluctant, or unable, to act against politically well-connected banks” (Fane and McLeod (2003)).  Additionally, there was an informational gap between what supervisors knew, and what the banks knew about the loans that they provided.  Typically, the supervisors were bureaucrats who did not know enough about banking to be able to monitor the banks effectively.  Focusing on a system of better supervision would not have been a practical stop against moral hazard.  These supervision systems have often failed in developing countries and there is no reason to expect that Indonesia would have been any different. (Fane and McLeod (2003))

Financial markets will not perform well in the long-run unless depositors believe that their money is protected.  There are too many incentives to invest in foreign banks if the domestic banks do not provide adequate protection for deposits.  However, this does not suggest that blanket protection by the government is the appropriate method for this because it creates an incentive for moral hazard.  Fane and McLeod (2003) suggest that a better approach is to let insurance companies limit the risk instead of the government.  Government bailouts rely on bank mangers to limit moral hazard, but as we’ve already shown, this does not limit it.  An insurance company would be able to implement market-based limitations of moral hazard, and therefore remove the bank managers’ position of authority to limit moral hazard. 

Non-performing loans of local banks were 11% by the end of 1997 (Corsettia, Pesentib, and Roubini (1999)).  Loans were not only provided to corporations.  They were provided to the private sector as well for real estate.  After the market pressures caused interest rates to rise, and the rupiah to depreciate, the economy experienced a slow-down.  This slow-down was associated with falling property prices, which left borrower’s who had real estate loans exposed (Goldstein and Hawkins (1998)).  This exposure led to exposure for the banks because when property prices fall, the number of non-performing loans increases (Goldstein and Hawkins (1998)).  Table 4 shows the banking sector’s exposure to risk because of the real estate market.  As shown, the banks perceived value of collateral was overly optimistic.