A. Data on Indonesian Financial Structure

Indonesian finance has historically been dominated by banks. Flow of funds accounts presented show that approximately 40 percent of savings by the private sector flows into bank deposits (Table 2). Most of the rest of these savings remain within the private nonbanking sector, on which data exist only in aggregate form. It seems probable that much of the residual savings is retained by the original savers and a substantial part of capital formation is self-financed. The flow of funds accounts indicate that 30-45 percent of private capital formation is financed by bank credit and most of the rest is financed within the domestic private, nonbanking sector. According to these accounts, finance provided directly by the government and foreigners declined during 1991–94.

Table 2.

Indonesia: Role of Banks in Private Finance, 1991–94

Source: Bank Indonesia, Indonesia’s Flow of Funds Account Matrix, 1991–94.

Table 2.

Indonesia: Role of Banks in Private Finance, 1991–94

		1991	1992	1993	1994
		(In trillions of rupiah)
Domestic private nonbank sector
	Gross savings	39.1	49.0	58.8	67.5
	Gross capital formation	51.2	53.2	67.5	86.2
	Bank deposits and currency	14.5	20.2	26.3	29.5
	Credit from domestic banks	19.3	15.7	31.0	37.5
	Credit from Bank Indonesia and central government	7.9	4.9	2.7	1.3
	Credit from foreign sources	8.7	8.5	1.2	2.8
		(In percent)
Private nonbank savings intermediated by banks		37.1	41.2	44.8	43.7
Private nonbank investment
	Intermediated by banks	37.7	29.5	45.9	43.6
	Financed by government	15.4	9.2	4.0	1.5
	Financed abroad	16.9	16.0	1.8	3.2
	Financed within sector	30.0	45.3	48.3	51.7

Source: Bank Indonesia, Indonesia’s Flow of Funds Account Matrix, 1991–94.

View Table

Table 2.

Indonesia: Role of Banks in Private Finance, 1991–94

		1991	1992	1993	1994
		(In trillions of rupiah)
Domestic private nonbank sector
	Gross savings	39.1	49.0	58.8	67.5
	Gross capital formation	51.2	53.2	67.5	86.2
	Bank deposits and currency	14.5	20.2	26.3	29.5
	Credit from domestic banks	19.3	15.7	31.0	37.5
	Credit from Bank Indonesia and central government	7.9	4.9	2.7	1.3
	Credit from foreign sources	8.7	8.5	1.2	2.8
		(In percent)
Private nonbank savings intermediated by banks		37.1	41.2	44.8	43.7
Private nonbank investment
	Intermediated by banks	37.7	29.5	45.9	43.6
	Financed by government	15.4	9.2	4.0	1.5
	Financed abroad	16.9	16.0	1.8	3.2
	Financed within sector	30.0	45.3	48.3	51.7

Source: Bank Indonesia, Indonesia’s Flow of Funds Account Matrix, 1991–94.

Available data (Table 3) indicate that the bond market provides a very small part of commercial finance, while the stock market provides just over one-third and banks provide just under two-thirds of the total. However, this substantially overstates the role of the stock market, because capitalization includes shares which have never been sold on the stock market. A reported 70 percent of total shares are held by company founders, including those held by the government after partial privatizations. If a rough adjustment is made for these unissued shares, the stock market has provided about one–seventh of total business finance, and banks have provided a little under six–sevenths.

Table 3.

Indonesia: Role of Bank and Securities Finance, 1994–95¹

Source: Bank Indonesia, Report for the Financial Year 1994/95.

¹End-March.

Table 3.

Indonesia: Role of Bank and Securities Finance, 1994–95¹

		(In trillions of rupiah)		(In percent of total business finance)
		1994	1995	1994	1995
Bank credit (commercial credits only)		142.2	176.4	65.7	62.4
Total securities market capitalization		74.2	106.1	34.3	37.6
	Stock market	67.7	98.8	31.3	35.0
	Bond market	6.5	7.2	3.0	2.5

Source: Bank Indonesia, Report for the Financial Year 1994/95.

¹End-March.

View Table

Table 3.

Indonesia: Role of Bank and Securities Finance, 1994–95¹

		(In trillions of rupiah)		(In percent of total business finance)
		1994	1995	1994	1995
Bank credit (commercial credits only)		142.2	176.4	65.7	62.4
Total securities market capitalization		74.2	106.1	34.3	37.6
	Stock market	67.7	98.8	31.3	35.0
	Bond market	6.5	7.2	3.0	2.5

Source: Bank Indonesia, Report for the Financial Year 1994/95.

¹End-March.

The role of financial intermediaries including insurance companies and nonbank finance companies in Indonesian markets is small compared with that of banks. Banks own over 85 percent of the total assets of this group of intermediaries, which excludes pension funds because of a lack of consistent data (Table 4). The first domestic mutual fund in Indonesia began operations only in late 1995.

Table 4.

Indonesia: Size of Selected Financial Intermediaries, 1992–94¹

Source: Bank Indonesia, Report for the Financial Year 1993/94 and 1994/95.

¹Bank data as of end-March; insurance and finance company as of end-December of preceding year. Pension funds are excluded because of a lack of consistent data; Leechor (1996), indicates these controlled assets totaled Rp 18.6 trillion at end-1994.

Table 4.

Indonesia: Size of Selected Financial Intermediaries, 1992–94¹

		1992	1993	1994
		(In trillions of rupiah)
Commercial banks (total credit) Of which,		115.7	124.2	153.9
	State banks	61.8	69.1	73.7
	Private national banks	42.3	42.5	64.6
	Local government-owned banks	2.6	2.9	3.5
	Foreign and joint banks	9.1	9.7	12.1
Insurance companies (assets) Of which,		7.9	9.0	11.3
	Life insurance	1.6	1.9	5.6
	Social insurance	3.6	4.3	2.3
	Indemnity insurance and reinsurance	2.6	2.8	3.3
Finance companies (assets)		8.4	10.1	11.9
		(In percent of bank credit and assets of other companies)
Commercial banks		87.7	86.7	86.9
	State banks	46.8	48.2	41.6
	Private national banks	32.0	29.6	36.5
Insurance companies		6.0	6.3	6.4
Finance companies		6.3	7.1	6.7

Source: Bank Indonesia, Report for the Financial Year 1993/94 and 1994/95.

¹Bank data as of end-March; insurance and finance company as of end-December of preceding year. Pension funds are excluded because of a lack of consistent data; Leechor (1996), indicates these controlled assets totaled Rp 18.6 trillion at end-1994.

View Table

Table 4.

Indonesia: Size of Selected Financial Intermediaries, 1992–94¹

		1992	1993	1994
		(In trillions of rupiah)
Commercial banks (total credit) Of which,		115.7	124.2	153.9
	State banks	61.8	69.1	73.7
	Private national banks	42.3	42.5	64.6
	Local government-owned banks	2.6	2.9	3.5
	Foreign and joint banks	9.1	9.7	12.1
Insurance companies (assets) Of which,		7.9	9.0	11.3
	Life insurance	1.6	1.9	5.6
	Social insurance	3.6	4.3	2.3
	Indemnity insurance and reinsurance	2.6	2.8	3.3
Finance companies (assets)		8.4	10.1	11.9
		(In percent of bank credit and assets of other companies)
Commercial banks		87.7	86.7	86.9
	State banks	46.8	48.2	41.6
	Private national banks	32.0	29.6	36.5
Insurance companies		6.0	6.3	6.4
Finance companies		6.3	7.1	6.7

Source: Bank Indonesia, Report for the Financial Year 1993/94 and 1994/95.

¹Bank data as of end-March; insurance and finance company as of end-December of preceding year. Pension funds are excluded because of a lack of consistent data; Leechor (1996), indicates these controlled assets totaled Rp 18.6 trillion at end-1994.

B. Economic Implications of Indonesian Financial Structure

The key link between the financial system and economic growth is the mobilization of savings and the channeling of them into productive investment. Indonesia has high savings and investment rates relative to most other countries in the world, although not relative to some of its neighbors. Prima facie, this suggests that the financial system, by intermediating a large share of savings and investment, has successfully fostered growth. However, it is possible that with a different financial structure, Indonesian growth could be higher and the savings and investment rates could be larger.

At least two aspects of Indonesian financial structure have potential effects on macroeconomic stability. First, the dominance of bank debt over other forms of finance can facilitate the operation of monetary policy, by raising the importance of the banking system relative to other parts of the financial system. This makes the transmission from Bank Indonesia’s monetary policy to economic activity more direct and potentially more precise and predictable. Second, the dominance of banks also has negative implications, in that enterprises have fewer alternative sources of finance than in some other countries. Recent efforts to tighten regulation of the commercial paper market may exacerbate this effect, although the close links between that market and banks or bank affiliates reduce the degree to which it truly represents an alternative source of finance. The dependence of firms on bank finance increases the risk of a credit crunch that could magnify the effects of an economic downturn.

The financing of investment projects by internal corporate funds, or at least by funds internal to a conglomerate group, appears to play a major role in Indonesia. The literature suggests that this is an effective means to finance capital formation, because there tend to be fewer informational asymmetries in such arrangements than in more arms-length transactions. Nonetheless, the prevalence of self-finance may indicate that other channels of finance are not working well, which may inhibit growth. There is also the agency-theoretic concern that heavy use of self-finance indicates a high level of free cash-flow that is being inefficiently deployed by enterprise managers. More data would be needed to evaluate these questions. However, evidence on the effects of the 1983 liberalization of interest rates and bank credit on enterprise finance (Goeltom, 1995) shows that enterprise borrowing costs increased, but the reliance on internal finance by smaller firms and by firms not connected to a conglomerate decreased. Goeltom also finds that after the interest rate increase that accompanied liberalization, firms that were not entirely self-financed were more efficient, suggesting that the liberalization increased the efficiency of the allocation of investment.

The development of the stock market has positive implications for growth in Indonesia. A stock market is a necessary component of the development of sources of finance for risky investments in growing sectors of the economy. While venture capital so far plays a small role, the development of a larger venture capital sector could help channel initial finance to small firms, which could access the stock market later in their life cycle. Without that, self-finance augmented by bank loans is likely to be the primary source of finance to risky firms.

The financial structure has other impacts on economic efficiency. For example, Indonesian residents benefit as the set of available savings and financing vehicles becomes larger. The benefits result from financial markets’ becoming more complete, which standard microeconomic theory shows increases welfare. There has been much progress in recent years on expanding the range of financial products, with the development of the stock market and pension funds, as well as money market instruments and, most recently, stock warrants. Indonesia lacks an exchange for financial futures and options. Although such derivatives products could enhance efficiency, it will be argued later in this paper that the supervision of the financial system should be improved before such an innovation is contemplated.

While positive real interest rates have helped encourage short-term personal savings, there is a dearth of vehicles for long-term savings. Reform of the pension system, including efforts to make employer-sponsored plans more attractive (Leechor, 1996), could help attract more savings, and also channel the savings more to long-term investments such as stocks and bonds. Development of the life insurance industry could make a similar contribution.

Growth depends on the efficiency of investment, particularly the extent to which funds are allocated to high-return projects. One factor that influences efficiency is the enforceability of investment contracts and related agreements. The World Bank (1996) recently examined this issue and concluded that there are substantial legal impediments to financial contracting. Action to alleviate these impediments, which hinder both bank lending and bond finance, would ensure that the allocation of funds by intermediaries is done more according to the criteria of economic risk and return, and less according to the availability of other means to enforce contracts, such as social relations between counterparties.

A. An Overview of the Banking System

The number of commercial banks grew rapidly from 111 in 1989 to approximately 240 in 1994, when the authorities placed stricter limits on the issuance of new bank licenses. Large state-owned banks and regional government-owned development banks coexist with a rapidly growing sector of privately owned banks and partly or fully foreign-owned banks (Table 5). The large state-owned banks and the private national banks together accounted for 87 percent of total banking assets at the end of 1995. Although distinctions have diminished in recent years, the seven state-owned banks consist of five that began as sectoral lending banks, a former development bank, and a former savings bank. During 1991–95, assets of the private national sector grew at an annual rate of 26 percent, while the state bank sector grew at a 12 percent annual rate. By 1994, the assets of the private national sector had surpassed those of the state bank sector. Foreign-owned banks consist of joint ventures between foreign banks and domestic investors, and branches of foreign banks. Indonesia does not permit wholly owned subsidiaries of foreign banks in Indonesia, and only foreign banks with existing branches may open new branches. Foreign banks without a preexisting presence must enter through a joint venture.

Table 5.

Indonesia: Composition of the Banking Sector, 1991–95

Sources: Bank Indonesia, Indonesian Financial Statistics, January 1996; Report for the Financial Year 1993/94, 1994/95, and 1995/96.

Table 5.

Indonesia: Composition of the Banking Sector, 1991–95

			1991	1992	1993	1994	1995
Number of banks
	State commercial banks		7	7	7	7	7
	Local government-owned banks		27	27	27	27	27
	Private national banks		129	144	161	166	165
	Joint banks		19	20	29	30	31
	Foreign banks		10	10	10	10	10
		Total	192	208	234	240	240
Number of offices
	State commercial banks		1,395	1,434	1,455	1,490	1,635
	Local government-owned banks		580	613	639	645	705
	Private national banks		3,260	3,385	3,601	3,806	4,160
	Joint banks		24	31	45	50	52
	Foreign banks		99	94	98	100	103
		Total	5,358	5,557	5,838	6,090	6,655
Assets (in trillions of rupiah)
	State commercial banks		78.0	93.3	100.6	104.5	122.6
	Local government-owned banks		4.7	5.3	6.5	7.9	9.8
	Private national banks		58.5	66.3	88.2	113.8	147.5
	Joint banks		5.6	7.5	11.8	14.3	17.9
	Foreign banks		7.4	7.5	7.9	9.2	12.3
		Total	154.2	180.0	215.1	249.8	310.0

Sources: Bank Indonesia, Indonesian Financial Statistics, January 1996; Report for the Financial Year 1993/94, 1994/95, and 1995/96.

View Table

Table 5.

Indonesia: Composition of the Banking Sector, 1991–95

			1991	1992	1993	1994	1995
Number of banks
	State commercial banks		7	7	7	7	7
	Local government-owned banks		27	27	27	27	27
	Private national banks		129	144	161	166	165
	Joint banks		19	20	29	30	31
	Foreign banks		10	10	10	10	10
		Total	192	208	234	240	240
Number of offices
	State commercial banks		1,395	1,434	1,455	1,490	1,635
	Local government-owned banks		580	613	639	645	705
	Private national banks		3,260	3,385	3,601	3,806	4,160
	Joint banks		24	31	45	50	52
	Foreign banks		99	94	98	100	103
		Total	5,358	5,557	5,838	6,090	6,655
Assets (in trillions of rupiah)
	State commercial banks		78.0	93.3	100.6	104.5	122.6
	Local government-owned banks		4.7	5.3	6.5	7.9	9.8
	Private national banks		58.5	66.3	88.2	113.8	147.5
	Joint banks		5.6	7.5	11.8	14.3	17.9
	Foreign banks		7.4	7.5	7.9	9.2	12.3
		Total	154.2	180.0	215.1	249.8	310.0

Sources: Bank Indonesia, Indonesian Financial Statistics, January 1996; Report for the Financial Year 1993/94, 1994/95, and 1995/96.

The process of liberalization of the banking system began in 1983 (Binhadi, 1995, and IBCA, 1995) with a liberalization of interest rates, the elimination of credit ceilings, and the introduction of indirect monetary instruments. A package of deregulatory measures passed in 1988 included a reduction in reserve requirements from 15 percent to 2 percent, the reopening of licensing for new private banks and foreign joint-venture banks, and the granting of permission to state-owned firms to deposit 50 percent of their short-term funds with private banks, instead of only with state-owned banks. The number of privately owned banks exploded after this measure. Further deregulation the following year eliminated the need for Bank Indonesia (BI) approval for medium- and long-term loans and removed ceilings on offshore loans. Regulators also enacted a restriction on bank lending to related parties, a limit on net foreign exchange open positions, and limits on equity activities of banks.

The authorities have strengthened the regulation of the banking system in the 1990s. They introduced a risk-weighted capital adequacy ratio, effective end-1993, which mirrored that adopted by the Group of Ten countries through the Basle Committee on Banking Supervision, and enacted minimum loan-loss provisions. In the face of slowing activity in 1993, the authorities relaxed regulations somewhat, although they tightened the legal lending limit, which governs bank exposures to single borrowers and to parties affiliated with the bank.³ Banking Act No. 7 of 1992 converted state banks to limited liability companies and permitted them to lend to nonpriority sectors. Despite the imposition of limited liability, which in principle limits the amount of state support available, the Ministry of Finance announced in 1994 that it would not permit a state bank to default on its obligations (Chan, 1995a). In 1995, reserve requirements were raised from 2 percent to 3 percent, effective February 1996. In addition, the minimum capital required for banks with foreign exchange licenses was tripled, and the capital adequacy ratio for these banks was raised from the 8 percent that currently applies to all banks to 12 percent, with both of these measures to be phased in over a five year period ending in 2001. BI has developed a supervisory system patterned on the U.S. CAMEL system (Capital, Asset Quality, Management, Earnings, Liquidity), and undertakes annual on-site examinations of banks. BI remains responsible for bank supervision and regulation, while the Ministry of Finance has authority to grant and revoke bank licenses.

The substantial asset quality problems experienced by banks in the early 1990s were part of the impetus for the tightening of supervision and regulation. These problems appear to have been due to lax lending controls, exacerbated by the effects of a tightening in monetary policy in 1991. The problems of state banks culminated in the rescue in 1995 of Bapindo, which had built-up an overwhelming percentage of nonperforming loans. Earlier, a government program had been set up with the support of the World Bank to recapitalize and restructure the five state commercial banks. Problems among private banks led to the first bank failure in Indonesia in over twenty years, that of Bank Summa in 1992.

Some Indonesian banks continue to be adversely affected by problem loans and reported asset quality continues to be worse at state banks than at private banks. Official figures indicate that classified credits for the banking system as a whole declined from a peak of 14 percent of total loans at end-1993 to 10 percent at end-1995. However, the improving trend of this ratio appears largely due to the continued fast expansion of bank credit. Nonperforming loans comprised about 17 percent of total credits extended by state banks at end-1995, but only 5 percent of private bank credits (Table 6). Private foreign exchange banks (which tend to be the larger private banks) had substantially better asset quality than the smaller, private non-foreign exchange banks (Chan, 1995b; World Bank, 1996).

Table 6.

Indonesia: Commercial Bank Nonperforming Loans, 1993–95¹

Source: Bank Indonesia, in World Bank (1996b)

¹End–December.

²Bad loans are the lowest quality of three categories of nonperforming loans.

³Includes the five state commercial banks, Bapindo, and BTN.

Table 6.

Indonesia: Commercial Bank Nonperforming Loans, 1993–95¹

				1993	1994	1995
				(In percent of total credits)
All banks
	Nonperforming			14.2	12.1	10.4
		Bad2		3.3	4.0	3.3
State banks3
	Nonperforming			19.8	18.6	16.6
		Bad		4.2	5.9	5.3
Private banks
	Total credits			100.0	100.0	100.0
		Nonperforming		7.3	5.4	4.9
			Bad	2.1	1.9	1.4
				(In trillions of rupiah)
Memorandum items:
	Total credits			177.5	217.0	267.8
		State banks		99.1	104.1	120.9
		Private banks		79.8	108.5	141.3

Source: Bank Indonesia, in World Bank (1996b)

¹End–December.

²Bad loans are the lowest quality of three categories of nonperforming loans.

³Includes the five state commercial banks, Bapindo, and BTN.

View Table

Table 6.

Indonesia: Commercial Bank Nonperforming Loans, 1993–95¹

				1993	1994	1995
				(In percent of total credits)
All banks
	Nonperforming			14.2	12.1	10.4
		Bad2		3.3	4.0	3.3
State banks3
	Nonperforming			19.8	18.6	16.6
		Bad		4.2	5.9	5.3
Private banks
	Total credits			100.0	100.0	100.0
		Nonperforming		7.3	5.4	4.9
			Bad	2.1	1.9	1.4
				(In trillions of rupiah)
Memorandum items:
	Total credits			177.5	217.0	267.8
		State banks		99.1	104.1	120.9
		Private banks		79.8	108.5	141.3

Source: Bank Indonesia, in World Bank (1996b)

¹End–December.

²Bad loans are the lowest quality of three categories of nonperforming loans.

³Includes the five state commercial banks, Bapindo, and BTN.

There are indications that a significant number of banks are undercapitalized and have not yet complied with some important prudential rules, although compliance appears to be improving. According to BI, 15 banks did not meet the required 8 percent capital adequacy ratio in April 1996, down from 21 banks in December 1995, while 41 banks did not comply with the legal lending limit; this was an improvement from the 70 banks in December 1995. Twelve out of the 77 licensed foreign exchange banks did not meet the rules on net open foreign exchange exposure. Private banks have accounted for most violations (Table 7).

Table 7.

Indonesia: Number of Banks Not in Compliance with Prudential Rules, 1995

Source: Bank Indonesia, reported in “Banks Suffer US$4.5 Billion in Bad Loans,” Indonesian Observer, January 26, 1996; and Report of the Financial Year 1994/95.

¹End–March.

²End–October.

Table 7.

Indonesia: Number of Banks Not in Compliance with Prudential Rules, 1995

	Total number in category1	Capital adequacy ratio2	Legal lending limit2	Loan- deposit ratio2
State banks	7	0	2	1
Private banks	166	18	56	11
Local development banks	27	2	3	0
Foreign and joint-venture banks	40	1	9	6
Total	240	21	70	18

Source: Bank Indonesia, reported in “Banks Suffer US$4.5 Billion in Bad Loans,” Indonesian Observer, January 26, 1996; and Report of the Financial Year 1994/95.

¹End–March.

²End–October.

View Table

Table 7.

Indonesia: Number of Banks Not in Compliance with Prudential Rules, 1995

	Total number in category1	Capital adequacy ratio2	Legal lending limit2	Loan- deposit ratio2
State banks	7	0	2	1
Private banks	166	18	56	11
Local development banks	27	2	3	0
Foreign and joint-venture banks	40	1	9	6
Total	240	21	70	18

Source: Bank Indonesia, reported in “Banks Suffer US$4.5 Billion in Bad Loans,” Indonesian Observer, January 26, 1996; and Report of the Financial Year 1994/95.

¹End–March.

²End–October.

B. The Impact of the Structure of the Banking System on Growth and Efficiency

Poorly capitalized banks tend to make economically suboptimal lending decisions. In the theory of banking, a poorly capitalized bank has an incentive to make riskier loans, if depositors and shareholders expect to get bailed out if the bank fails. Bank shareholders in essence own a put option. If the loan portfolio does well, they gain the proceeds, but if the portfolio does poorly, their losses are limited by the willingness of the authorities to bail out the bank. Without the possibility of a bailout, creditors would demand a higher payout, and market pressure would operate on bank owners to reduce risks. On the other hand, if creditors expect to get bailed out, bank owners do not face this market pressure. Bank owners maximize the expected value in their put option by choosing a portfolio with high risk.

The consequence of a system with a number of banks that operate with low capital is that the economy will end up selecting production technologies that are riskier than optimal, given the expected returns of those projects, and taxpayers consequently bear more risk than optimal. The level of problem loans suggests that not only undercapitalized banks may have been making poor lending decisions. The ownership structure of Indonesian banks also influences the efficiency of asset allocation. State banks may not be required to make lending decisions on a commercial basis and, although direct information on this does not seem to be available, it is clear that problem loan ratios are much higher at state banks. In addition, many private banks are owned by affiliates of large corporate groups. There is a risk that they will make lending decisions in the interest of the owners of these groups, rather than those that maximize returns for the banks, and that the legal lending limit is not observed.

Problems in the legal system may also influence bank lending, and steer lending to areas with fewer legal risks rather than towards the highest economic return. The World Bank (1996) has cited difficulties in the use of collateral in Indonesia. Lenders have limited recourse if a borrower fails to make payments. The bankruptcy law is said to be inadequate, and there is also no adequate framework for restructuring corporations (IBCA, 1995).

A second, broader question regarding efficiency is whether banks are operating as competitive entities. Competition induces marginal cost pricing in both lending and deposit market, and also efficient use of resources to produce banking services. Direct evidence on this issue is absent in Indonesia. In general, however, there appears to be less focus by the authorities on ensuring a level of competition among banks adequate to produce competitive pricing on loans and deposits. The often expressed concern that Indonesia is overbanked presumes that the entire country can be considered a single and frictionless market, but this is not correct for many banking services. For small and medium-sized business lending and for some consumer finance, the local municipal market is the relevant market. These markets often have a small number of players, and relatively high informational barriers to entry on the lending side. For other products, the relevant geographic market is larger, perhaps national or even international, and the likely level of competition is higher. In addition, the current policy curtailing the issue of new banking licenses probably reduces efficiency by restraining entry by new banks that may have more efficient means of producing services. This is especially true of the restrictions on banks with foreign ties. These barriers to entry reduce incentives for existing banks to improve the efficiency of their operations.

C. The Impact of the Structure of the Banking System on Stability

There are two types of actual or potential events that may affect stability. First, destabilizing shocks can originate within the banking system, such as from a bank failure or a credit crunch. Second, the banking system can be part of the propagation mechanism for macroeconomic effects that originate elsewhere, such as aggregate demand or supply shocks, or changes in the exchange rate or international interest rates.

2. Problems originating outside the banking system

Macroeconomic shocks, including exchange rate or interest rate shocks, can increase the instability of the banking system, and initial problems can magnify their effects. In principle, BI regulations control bank sensitivity to exchange rate fluctuations by limiting bank net foreign exchange exposure to 25 percent of bank capital. This rule is applied to net on-and off-balance sheet exposures combined and separately to net off-balance sheet exposures. The exposure level specified by the regulation would appear to be low enough to prevent bank insolvency from occurring, but two concerns remain. First, it can be difficult to value bank exposures, especially those related to exchange rate options written by banks. There is little information available on how these derivatives are valued, so exposure may be larger than stated. Second, low net exposures may mask large gross exposures. Many banks have made substantial foreign currency loans, financed by foreign currency deposits. If the rupiah were to depreciate substantially, the rupiah-equivalent obligations of borrowers from the banks could soar, and if this were not offset by an increase in the foreign-currency resources available to these borrowers, defaults could occur.

Aggregate foreign exchange liabilities of Indonesian banks have grown rapidly in recent years and were 271 percent of total commercial bank equity capital by the end of 1994/95 (Table 8). Although total regulatory capital includes debt-based capital in addition to equity capital, the amount of equity capital alone determines a bank’s solvency level. Foreign exchange credits of banks were 169 percent of equity capital, so that net on-balance sheet foreign exchange liabilities were 102 percent of equity capital. Any large default on foreign currency loans could therefore quickly begin to erode equity capital. In addition, the separate net open exposure rule for off-balance sheet instruments prevents banks from hedging these exposures through forwards and options.

Table 8.

Indonesia: On-Balance-Sheet Bank Foreign Exchange Exposure, 1990/91–1994/95¹

Source: Calculations by author using data from Bank Indonesia, Indonesian Financial Statistics, 1996.

¹End-March.

²Current liabilities comprised 88 percent of total commercial bank liabilities at end-March 1995.

Table 8.

Indonesia: On-Balance-Sheet Bank Foreign Exchange Exposure, 1990/91–1994/95¹

	1990/91	1991/92	1992/93	1993/94	1994/95
	(In trillions of rupiah)
Current foreign exchange liabilities2	29.2	31.4	44.7	57.3	62.5
Outstanding foreign exchange credits	12.3	19.3	23.2	30.4	38.9
Equity capital	11.9	10.9	13.4	19.8	23.0
	(In percent)
Foreign exchange liabilities/equity capital	246	288	334	289	271
Foreign exchange credits/equity capital	104	177	173	153	169
Net foreign exchange liabilities/equity capital	143	111	161	136	102

Source: Calculations by author using data from Bank Indonesia, Indonesian Financial Statistics, 1996.

¹End-March.

²Current liabilities comprised 88 percent of total commercial bank liabilities at end-March 1995.

View Table

Table 8.

Indonesia: On-Balance-Sheet Bank Foreign Exchange Exposure, 1990/91–1994/95¹

	1990/91	1991/92	1992/93	1993/94	1994/95
	(In trillions of rupiah)
Current foreign exchange liabilities2	29.2	31.4	44.7	57.3	62.5
Outstanding foreign exchange credits	12.3	19.3	23.2	30.4	38.9
Equity capital	11.9	10.9	13.4	19.8	23.0
	(In percent)
Foreign exchange liabilities/equity capital	246	288	334	289	271
Foreign exchange credits/equity capital	104	177	173	153	169
Net foreign exchange liabilities/equity capital	143	111	161	136	102

Source: Calculations by author using data from Bank Indonesia, Indonesian Financial Statistics, 1996.

¹End-March.

²Current liabilities comprised 88 percent of total commercial bank liabilities at end-March 1995.

The degree to which interest rate shocks can harm banks depends on several factors. First, a typical bank borrows on a shorter-term basis than it lends, so that an increase in interest rates squeezes bank margins. BI statistics suggest that a substantial part of bank liabilities are extremely short-term. Second, the degree to which longer term loans are at variable rate affects bank exposure. It is not clear, however, to what extent interest rates on loans are variable in Indonesia. Finally, an increase in interest rates tends to increase nonperforming loans, which creates a negative bank exposure to interest rate increases. Borrowers at variable rates face higher interest payments, while interest rate increases also slow aggregate demand, reducing business profitability and the ability of borrowers—at both fixed and variable rates—to repay loans.

Banks’ sensitivity to exchange rate and interest rate shocks can be greater than their assets and liabilities indicate if they have significant off-balance sheet derivative exposures. (It can also be less, if derivatives are used to hedge balance sheet exposures.) BI recently recognized off-balance sheet risks and enacted a regulation that limits bank derivative exposures only to interest rate and exchange rate derivatives, except for case-by-case exceptions granted by BI for equity derivatives, and requires that potential losses from derivatives not exceed 10 percent of bank capital. Such a modest limit would appear appropriate for a banking system that is still in process of developing a reliable supervisory infrastructure. A potential weakness of the rule, however, its that it appears to offer little guidance on how these exposures should be computed. The valuation of derivatives, particularly options, can be difficult and this area requires careful supervision.

A. The Continued Development of the Stock Market

Continued progress in the development of market infrastructure and of supervision and regulation will probably contribute to the creation of a vigorous equity market in Indonesia. This is particularly likely if the supply of funds to the market also increases, through development of institutional investors and mutual funds, and if demand from growing enterprises for equity finance persists. The Jakarta Stock Exchange introduced a centralized settlement system for all listed securities in June 1994 (International Finance Corporation, 1995). It introduced the computerized Jakarta Automated Trading System in May 1995, which has already led to much larger trading volumes.

Recent years have seen great progress in stock market regulation, with a new capital markets law and a set of implementing regulations promulgated by Bapepam. The key area for further efforts appears to be in enforcement of these regulations, particularly in areas promoting transparency and fairness, such as insider trading. A transparent and fair market can make a far larger contribution to market liquidity than even large investments in infrastructure. If traders can be assured that their counterparties do not possess privileged information on the value of the security traded, these traders will more readily supply the counterparties with bids or offers at a narrow spread.

Better disclosure by corporations of their financial condition also promotes market transparency. In an important move, Bapepam (the capital market supervisory agency) and the Indonesian Accounting Institute implemented new accounting standards at the end of 1994, bringing financial accounting close to international standards, which helps international investors to evaluate their Indonesian exposures. Nonetheless, concerns remain among investors about the fairness of corporate disclosure, with poor auditing controls and with information leaking out to some parties before it reaches others (Montagnon, 1996). Continued monitoring of the production and release of financial information by enterprises is important to ensure that it is compiled promptly and accurately and disclosed publicly and fairly.

A deeper and more transparent stock market is, other things being equal, likely also to be a less volatile stock market. However, to the extent that a deeper market depends on international investors, volatility may actually increase as spillovers from other markets intensify. Volatility of stock prices on the Jakarta Stock Exchange (Table 9) was low until December 1988, when the market was opened to international investors, who soon dominated trading.⁵ The market was also less volatile relative to the U.S. market, chiefly because the crash in U.S. stock prices in October 1987 did not spillover into Indonesia. Since the opening of the Indonesian market in December 1988, it has been somewhat more volatile than the U.S. market. However, the growing liquidity of the Indonesian market in recent years helps explain why its absolute and relative volatility fell in 1993–96.

Table 9.

Indonesia and the United States: Daily Market Index Return Volatility, April 1986–July 1996

Source: Bloomberg.

¹Standard deviation of daily returns of Jakarta Composite Index (in rupiah).

²Absolute volatility divided by standard deviations of daily returns of Standard and Poor’s 500 Index (in dollars).

Table 9.

Indonesia and the United States: Daily Market Index Return Volatility, April 1986–July 1996

Period		Absolute Volatility1	Relative Volatility2
April 1986–December 1988		2.58	1.77
	April 1986–November 1988	0.72	0.49
	December 1988	13.11	23.41
January 1989–July 1996		1.30	1.78
	January 1989–December 1992	1.58	1.86
	January 1993–July 1996	0.84	1.47

Source: Bloomberg.

¹Standard deviation of daily returns of Jakarta Composite Index (in rupiah).

²Absolute volatility divided by standard deviations of daily returns of Standard and Poor’s 500 Index (in dollars).

View Table

Table 9.

Indonesia and the United States: Daily Market Index Return Volatility, April 1986–July 1996

Period		Absolute Volatility1	Relative Volatility2
April 1986–December 1988		2.58	1.77
	April 1986–November 1988	0.72	0.49
	December 1988	13.11	23.41
January 1989–July 1996		1.30	1.78
	January 1989–December 1992	1.58	1.86
	January 1993–July 1996	0.84	1.47

Source: Bloomberg.

¹Standard deviation of daily returns of Jakarta Composite Index (in rupiah).

²Absolute volatility divided by standard deviations of daily returns of Standard and Poor’s 500 Index (in dollars).

The flow of foreign investment in the stock market has expanded over time. For example, available data on net investment by U.S.-based investors in the Indonesian stock market show that there was almost no foreign investment before 1990. Investment then increased rapidly and reached a peak in 1994. Thus foreign portfolio investment was associated first with an increase in stock price volatility, in the early 1990s, and then a fall in volatility from 1993 onward. The finding that volatility fell from 1993 on calls into question the finding by Roll (1995), based on data running only through 1992, that foreign investment contributed to an increase in Indonesian stock market volatility.

In order to examine the linkages between international and Indonesian stock market volatility, following Folkerts-Landau et al. (1995), squared Indonesian daily stock returns were regressed on the previous day’s squared U.S. stock return (Table 10). The results of this analysis are striking. U.S. volatility has no effect on Indonesian volatility before 1993. For the period 1993–96, there is a strong linkage. Taken together, these results suggest that although internationalization has had mixed effects on the volatility of the stock market, foreign portfolio investment has increased the linkages in price movements between Indonesian and foreign stock markets.

Table 10.

Indonesia and the United States: Volatility Spillover Analysis, April 1986–July 1996

Source: Bloomberg.

¹Coefficient represents regression coefficient of squared daily return of Jakarta Composite Index (in rupiah) on squared daily return on the preceding day of the Standard and Poor’s 500 Index (in dollars). A constant was also included in the regression. * denotes significance at the 1 percent level. Other coefficients are insignificant at the 10 percent level. Data are from Bloomberg.

Table 10.

Indonesia and the United States: Volatility Spillover Analysis, April 1986–July 1996

Period		Spillover Coefficient1
April 1986-December 1988		–0.04
	April 1986-November 1988	0.00
January 1989-July 1996		0.17
	January 1989-December 1992	0.02
	January 1993-July 1996	0.71*

Source: Bloomberg.

¹Coefficient represents regression coefficient of squared daily return of Jakarta Composite Index (in rupiah) on squared daily return on the preceding day of the Standard and Poor’s 500 Index (in dollars). A constant was also included in the regression. * denotes significance at the 1 percent level. Other coefficients are insignificant at the 10 percent level. Data are from Bloomberg.

View Table

Table 10.

Indonesia and the United States: Volatility Spillover Analysis, April 1986–July 1996

Period		Spillover Coefficient1
April 1986-December 1988		–0.04
	April 1986-November 1988	0.00
January 1989-July 1996		0.17
	January 1989-December 1992	0.02
	January 1993-July 1996	0.71*

Source: Bloomberg.

¹Coefficient represents regression coefficient of squared daily return of Jakarta Composite Index (in rupiah) on squared daily return on the preceding day of the Standard and Poor’s 500 Index (in dollars). A constant was also included in the regression. * denotes significance at the 1 percent level. Other coefficients are insignificant at the 10 percent level. Data are from Bloomberg.

Two other issues will be important for the Indonesian stock market. First, with the rapid development of Southeast Asian economies, it will remain in competition with other stock markets in the region, including those in Malaysia, Singapore, and Thailand. As the technology and regulatory environment of these markets advance, the role for offshore trading will increase. On economic efficiency grounds, it is not desirable in the medium-term to force trading in Indonesian securities to remain solely in Jakarta. Instead, the creation of a market that is appealing to international investors will also benefit domestic investors, whom the authorities wish to attract to the market. Increased supervisory cooperation with financial regulators in neighboring countries will also be necessary.

Secondly, the development of stock market derivative products, such as futures and options, can help provide liquidity and aid in the process by which new information is incorporated into stock prices. However, such markets require assurances that brokerage firms and investors have the capability to deal with the risk inherent in these instruments. While derivatives merely repackage risk, they make it easier for investors to take leveraged and potentially explosive risk positions. The supervision and regulation of Indonesian financial markets do not appear to have reached the stage where many such products can be handled safely. This applies to the development of both exchange-traded and over-the-counter products, although with Bapepam’s approval of the first warrant issue in mid-1995 (Yu, 1995), one equity derivative market has been initiated.

B. The Development of a Domestic Bond Market

The small size of the bond market in Indonesia relative to other countries appears to be due to two factors. First, the legal environment for enforcing debt contracts is relatively weak. Secondly, there is a paucity of institutional investors, which reduces demand.

Difficulties with the legal system result in a bond market in which most issues are over-collateralized. Firms cannot issue bonds for general uses, but must earmark them for particular projects. The weakness of laws protecting collateral and governing corporate bankruptcy make even collateralized bonds risky and costly. To the extent that bonds are not collateralized, they are de facto required to have a guarantor, usually a state bank. Bonds therefore are indirectly obligations of a bank as well. Banks are more likely than a diffuse set of bondholders to succeed in collecting problem debts and to be in a position to evaluate the risk, although a new rule requiring bonds to be rated by the domestic credit rating agency Pefindo may reduce this advantage. For the bond market to grow, the legal system must reach the point where the abilities of creditors to protect their claims are strengthened.

Inflation also inhibits the issuance of long-term bonds. International evidence suggests that higher inflation tends to accompany more variable inflation, which in turn inhibits the issuance of long-term fixed rate bonds, because their real value is subject to great uncertainty. In fact, in recent years, almost all bond issues have carried floating interest rates, which reduce exposure to inflation down to a level similar to the exposure of short-term debt instruments. The popularity of fixed-rate bonds in some countries suggests that such bonds could be useful in Indonesian finance.

The Indonesian government currently issues no bonds in the domestic market because of its balanced budget rule. For the government to issue bonds of sufficiently large size to serve as liquid benchmarks, the amount of borrowing would have to be relatively large. Since this would raise domestic real interest rates and crowd out private borrowers, the development of a government bond market is therefore probably not desirable as a means to foster a corporate bond market. Instead, as suggested in IPAF (1995), state enterprise bonds, which currently dominate the bond market, could be used as benchmarks. Indonesia could borrow a technique used for government bond markets in other countries (see International Monetary Fund, 1994) and concentrate state enterprise bond issues into a smaller number of large issues, which could help to develop a secondary market.

Current developments are likely to augment the liquidity of the secondary market in bonds, which now suffers from low trading volume (PT Sigma Batara, 1995). The introduction of a centralized clearing and settlement agency for stocks, due to be extended to listed bonds, may reduce the cost of settling secondary market trades. Bond trades currently require the risky and costly physical delivery of bearer bonds. In addition, the Over-the-Counter Exchange has developed an on-line information system to disseminate bid and offer prices to all market participants. Plans exist to extend this to screen-based trading.

C. The Development of Nonbank Financial Intermediaries

A key contributor to the growth of securities markets is the development of institutional investors—mutual funds, insurance companies, and pension funds—to channel individual savings into them. These institutions can be efficient mechanisms for individuals to pool and repackage the risks of securities markets and to reduce transaction costs, thereby assisting in the provision of equity and long-term debt finance to the economy.

The role of domestic institutional investors remains small. Mutual funds have been introduced only very recently, with the first closed-end fund starting operation in October 1995.⁶ Other closed-end and open-ended funds have applied to Bapepam for approval. A new set of regulations drawn up by Bapepam in May 1996 permits funds to invest a maximum of 85 percent of net assets in the stock market, and also specifies procedures for portfolio valuation and financial reporting by funds. Regulations were also changed to eliminate the double taxation of mutual funds. For funds to operate properly and to provide income to investors that is equivalent to income received from direct securities holdings, funds must be able to pass through all income to shareholders on a before-tax basis.

Although Indonesia has made some progress in developing pension funds, existing funds now invest 85 percent of their assets in bank deposits (Yu, 1995). Pension funds can be separated into those sponsored by employers and those linked to the government, including a fund for civil-service pensions. A previous regulatory limit on the investment of employer-sponsored plans in securities markets has been lifted, although there continue to be ceilings on property investment, individual exposure limits, and limits on self-investment. A key issue for pension funds is the encouragement of investment in securities markets. The high real returns and relative security available in rupiah-denominated bank deposits have been cited as a reason for the low investment in securities.