Information about Asia and the Pacific Asia y el Pacífico
Chapter

9 Capital Market Development and the Monetary Transmission Mechanism in Malaysia and Thailand

Editor(s):
John Hicklin, David Robinson, and Anoop Singh
Published Date:
July 1997
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

Capital market development and financial deregulation in Malaysia and Thailand are opening up new and increased financing opportunities for businesses and households. This paper looks at how capital markets have developed in these countries and assesses how the development of these markets, together with the deregulation of financial intermediaries, has affected the financing of economic activity and, thereby, the transmission of monetary policy to the economy.

In both Malaysia and Thailand, the use of external finance (debt and equity) by the corporate sector has increased significantly in recent years. In Thailand, this has occurred mainly through a rapid increase in borrowing from financial intermediaries, while in Malaysia there has been greater use of the capital markets. This difference reflects both the more developed nature of the capital markets in Malaysia and the less stringent controls imposed on financial intermediaries in Thailand. The increased reliance on external financing has important implications for the transmission of monetary policy to the real economy. While there may be some loss in the potency of direct controls on the banking system, the increased use of external financing may make the economy more sensitive to changes in interest rates. Evidence from vector autoregressions provides some support for the view that the potency of interest rate policy has increased in recent years.

Financial Systems in Malaysia and Thailand

The relative size of financial intermediaries and the capital markets in Malaysia and Thailand is shown in Figure 1.1 Two measures of the size of the financial intermediary sector are given: the ratio of commercial bank assets to GDP and the ratio of total financial sector assets to GDP.2 In both countries, these measures show steady financial deepening throughout the sample period.

Figure 1.Relative Size of Financial Intermediaries and the Capital Markets

(In percent of GDP)

Sources: IMF, International Financial Statistics (various issues); Bank of Thailand, Quarterly Bulletin (various issues); Stock Exchange of Thailand (1995); and Bank Negara Malaysia.

Various indicators of the structure and size of the Malaysian and Thai equity markets are shown in Figure 2. Given that the concept of equity is firmly entrenched in Malaysia (securities have been traded there in some form for more than a century), market capitalization of the Kuala Lumpur Stock Exchange already exceeded 80 percent of GDP in 1980. By the end of 1995, market capitalization had reached 265 percent of GDP, making the Malaysian market about three times larger than the Thai market and larger than the markets in many other Asian and industrial countries (Figure 3). Even though the Kuala Lumpur Stock Exchange has always been much larger (relative to GDP) than the Stock Exchange of Thailand, the general pattern of equity market development has been similar across the two countries since 1980. The size of the stock markets remained broadly unchanged between 1980 and 1985, began to accelerate in 1986, and rose particularly rapidly with the inflow of foreign capital during 1992-93. Although market capitalizations subsequently fell, the markets in both countries were still 60 percent larger relative to GDP in 1995 than in 1992. In Malaysia, privatization has been an important spur: 24 public enterprises have been listed on the stock market, accounting for 22 percent of total market capitalization at the end of 1995.

Figure 2.Indicators of Stock Market Development

Sources: Data provided by Bank Negara Malaysia and the Kuala Lumpur Stock Exchange; Stock Exchange of Thailand, Monthly Review (various issues), and Stock Exchange of Thailand (1995).

1 In percent of nominal value.

2In percent of turnover.

Figure 3.Stock Market Capitalization

(In percent of GDP)

Sources: International Finance Corporation, Emerging Stock Markets: Factbook, 1995; IMF, World Economic Outlook Database; Bank Negara Malaysia; and Stock Exchange of Thailand (1995).

The bond market in Thailand is comparatively small, while that in Malaysia is the second largest in the region after Singapore; both markets have declined in size since 1988 (Figure 4). In Thailand, out standing issues fell to 10 percent of GDP during the period and, while the stock and bond markets were the same size in 1988, by 1995 the stock market was nearly nine times as large as the bond market. Similarly, in Malaysia, outstanding issues declined from 65 percent to 48 percent of GDP, and the relative size of the stock market increased from a factor of less than two to a factor of more than five.

Figure 4.Outstanding Bond Issues

(In percent of GDP)

Sources: World Bank (1995); Bank of Thailand, Quarterly Bulletin (various issues); and Bank Negara Malaysia.

1Includes bonds issue by Cagamas, the national mortgage corporation, and nonfinancial public enterprises.

2Refers to 1993.

The declines in the relative sizes of the bond markets between 1988 and 1995 are due to the fiscal surpluses run by both the Malaysian and Thai governments and the resultant reduction in government debt out-standing, which has been only partly offset by increased issuance from other borrowers (Figure 5). The state enterprise sector is now the largest issuer in the domestic bond market in Thailand. The corporate bond markets3 in both countries have grown rapidly in recent years, albeit from a low base. In Thailand, restrictions on corporate bond issues were eased in 1992; this, together with the establishment of the Thai Rating and In-formation Service, a credit rating agency, in March 1993 and the Bond Dealers Club in September 1994, led to a large increase in the number of companies tapping the debt market. The outstanding stock of corporate bonds was B 134 billion at the end of 1995, with over B 100 billion issued in 1994-95.

Figure 5.Outstanding Domestic Bond Issues

Sources: Bank of Thailand, Quarterly Bulletin (various issues). Thai Bond Dealers Club; and Bank Negara Malaysia.

1Includes bonds issued by nonfinancial public enterprise.

In Malaysia, outstanding corporate issues were RM 22 billion in 1995 (4 percent of stock market capitalization), compared with RM 0.4 billion (½ of 1 percent of stock market capitalization) in 1987. As in Thailand, the establishment of a credit rating agency (Rating Agency Malaysia Berhad) in 1990 has been an important spur to the growth of the private bond market.4 Two of the greatest impediments to further development of the private bond market—which these countries are currently addressing—are the absence of a benchmark interest rate (owing to short supply and the sluggish turnover in the government bond market) and the lack of an efficient trading and settlement system.

Figures 6 and 7 illustrate that in recent years Malaysian and Thai entities have raised similar amounts of finance, as a proportion of GDP, through the international capital markets and syndicated bank loans. In Thailand, however, banks and other financial institutions have accounted for a larger proportion of funds raised. The operation of the Bangkok International Banking Facilities (BIBF)5 since 1993, the stable exchange rate, and the large interest rate differential vis-à-vis the United States have all made the issue of debt in overseas markets attractive to domestic issuers. In Malaysia—where potential borrowers or issuers of equity in the international markets must obtain approval from Bank Negara Malaysia—the nonfinancial public enterprises have been the primary participants. Commercial banks and other financial institutions have not had a major presence.

Figure 6.Funds Raised from International Sources by Instrument

(In percent of GDP)

Source: IMF, Developing Country Bonds, Equities, and Loans Database.

Figure 7.Funds Raised from International Sources by Borrower

(In percent of GDP)

Source: IMF, Developing Country Bonds, Equities, and Loans Database.

Capital Market Development, Financial Deregulation, and the Financing of Economic Activity

Financial deregulation and the development of capital markets have significant implications for the role of financial intermediaries, the financing of investment and consumption, and the structure of private saving. Financial intermediaries face pressures on both sides of their balance sheets as the nature and range of the assets and liabilities available to them change. Firms have a larger choice of external finance options, and households a wider range of saving options, as capital markets develop and overseas markets become accessible. In this section, the banks, business enterprises, and households are in turn examined to see how their balance sheets have developed with financial deregulation and the growth of capital markets.

Behavior of Domestic Banking Systems

Before 1990, banks in Thailand were subject to controls on deposit and lending rates and had limited access to debt funds through the capital markets. This led to excess demand for loans by domestic borrowers and to credit rationing. Being limited in their ability to raise funds, banks acted as asset managers—accepting deposits that came their way and allocating them among various assets. Interest rate liberalization and greater access to capital markets—particularly international capital markets—increased financing opportunities and allowed greater competition for deposits. Hence, banks have become better able to act as liability managers by tailoring their funding more effectively to match the domestic demand for loans, subject to the credit plans agreed on with the government.6

These developments have had two main implications for the balance sheet of the Thai banking sector (Table 1). First, before 1990, bank deposits and lending moved broadly in line; over the past five years, lending has grown considerably more rapidly than deposits, and the loan-to-deposit ratio has increased to well over 100 percent. Second, the foreign component of banks’ balance sheets has expanded more rapidly than the domestic component, with the increased borrowing through the BIBF and other international markets being matched on the asset side by higher foreign currency lending through the BIBF.

Table 1.Thailand: Selected Developments in the Consolidated Balance Sheet of Commercial Banks
1980198119821983198419851986198719881989199019911992199319941995
(In billions of baht)
Assets
Loans and advances to:2032392843814434905086197971,0451,4071,6892,0322,5223,2604,008
Government0110111112122313
Nonfinancial public enterprises99121112111010101011910171213
Businesses and households1942292713694314784976097861,0331,3951,6792,0202,5033,2474,004
Of which:
Consumption credit182125374246486589122159202269340437523
Holdings of securities334359588684114127140143142145147159203254
Foreign assets16252625313442394570567378157169240
Liabilities
Deposits2132563204054905486217448841,1191,4241,7312,0112,3862,7023,152
Foreign liabilities283429435146323762851101241683527801,164
Capital accounts19212528364244576883111143170221306395
(In percent)
Memorandum items:
Loan-to-deposit ratio95.493.288.694.090.489.381.883.290.193.498.897.6101.1105.7120.7127.2
Ratio of foreign liabilities to total liabilities9.59.56.98.18.06.54.14.05.56.16.15.86.611.119.423.2
Interest rate spread16.06.56.04.65.86.07.25.55.55.54.35.38.77.05.93.5
Ratio of lending to businesses and households to GDP29.230.132.240.143.645.243.846.850.455.663.766.671.579.190.396.2
Ratio of consumption credit to GDP2.72.83.04.04.24.34.35.05.76.67.28.09.510.712.212.6
Sources: Bank of Thailand, Quarterly Bulletin (various issues); and IMF, International Financial Statistics (various issues).

Defined as lending rate minus deposit rate.

Sources: Bank of Thailand, Quarterly Bulletin (various issues); and IMF, International Financial Statistics (various issues).

Defined as lending rate minus deposit rate.

Financial institutions’ greater use of the international capital markets is simply a form of international financial intermediation and a primary mechanism through which international capital flows reach the domestic economy. This form of business has become attractive to banks for several reasons. First, the longer loan maturity generally available in international markets gives banks better scope for funding their own long-term loans. Second, loans extended through the BIBF were not included in the credit plans until 1995, and a reserve requirement on short-term foreign borrowing (less than one year) was not introduced until 1996. Most important, however, the stability of the currency has allowed banks to take full advantage of the spread between domestic and U.S. dollar interest rates.7

Malaysian financial institutions have not had as large a role in international intermediation as their Thai counterparts. Most of their international borrowing has taken place through head offices and branches or bilaterally with foreign financial institutions, and so Malaysian banks have accounted for only a small portion of total international bond issues and syndicated borrowing (Figure 7). Although foreign liabilities of commercial banks expanded rapidly between 1989 and 1993, they fell subsequently (Table 2), and the growth rates of foreign and domestic components of bank balance sheets have been comparable. Also, in contrast to Thailand, the loan-to-deposit ratio has fallen in Malaysia in recent years—although it is still over 100.

Table 2.Malaysia: Selected Developments in the Consolidated Balance Sheet of Commercial Banks
1980198119821983198419851986198719881989199019911992199319941995
(In billions of ringgit)
Assets
Loans and advances to:202529364348525156668096105116133174
Government0001111111111000
Households2345678899111415181924
Of which:
Housing2334567788101212151417
Consumption credit0000011111223457
Businesses11721243035394241455465788694106142
Other20001111112232388
Statutory reserves with
Bank Negara Malaysia112222222357781419
Holdings of securities and negotiable certificates of deposit4677889131715182224283037
Foreign assets2235234578865111110
Liabilities
Deposits23283337434648505360627693119131162
Of which:
Held by households10121417182023252631354152666983
Negotiable certificates of deposit and banker’s acceptances outstanding13356789910151818142940
Foreign liabilities344766655681219311716
Capital accounts122344456678891619
Memorandum items(In percent)
Loan-to-deposit ratio390.290.889.899.1102.2106.8107.9105.1107.9112.1129.7127.5114.198.4102.2108.0
Ratio of foreign liabilities to total liabilities9.08.88.011.58.98.67.85.95.25.66.37.710.714.17.15.4
Interest rate spread44.95.87.45.73.72.01.72.43.22.92.6
Ratio of lending to businesses to GDP32.836.839.142.544.250.858.351.950.052.756.060.258.557.657.366.5
Ratio of lending to households to GDP4.55.36.16.57.18.910.79.99.49.19.910.810.111.310.211.2
Source: Bank Negara Malaysia, Monthly Statistical Bulletin (various issues).

Private and public enterprises.

Purchase of stocks and shares.

Including loans to foreigners.

End-of-year average lending rate less the end-of year three-month fixed deposit rate.

Source: Bank Negara Malaysia, Monthly Statistical Bulletin (various issues).

Private and public enterprises.

Purchase of stocks and shares.

Including loans to foreigners.

End-of-year average lending rate less the end-of year three-month fixed deposit rate.

In part, these developments were due to changes in the regulatory environment. In Malaysia, the statutory reserve requirement has been raised steadily from 4½ percent in 1989 to 13½ percent currently, and was expanded in 1994 to cover all externally sourced liabilities. Thus, loans and advances to domestic borrowers have fallen as a share of total assets since the early 1990s, and negotiable certificates of deposit and banker’s acceptances have replaced foreign borrowing as a source of funding.8 Other factors included the need to obtain approval for foreign borrowing from Bank Negara Malaysia9 and lending guidelines designed to ensure that priority sectors obtain adequate funding.

Developments in the Balance Sheets of the Private Corporate Sector

While there is a large literature on the capital structure of companies in industrial countries, much less is known about the capital structure of companies in developing countries. In this section, the theoretical determinants of corporate capital structure are briefly discussed, and evidence on the financial structure of companies in a selection of industrial countries is provided. Evidence on the corporate capital structure in Thailand and Malaysia is then presented.

In simple accounting terms, the corporate balance sheet can be represented as

where K represents the nonfinancial assets of the company, F the financial assets, N internal capital, and X eternal liabilities. Taking first differences of (1) yields the flow of funds identity

so that any increase in the company’s assets is financed from either internal funds or an increase in external liabilities. This can be rewritten as

where I is gross investment (ΔK), R is retained earnings (corporate earnings after interest and taxes, less dividends paid), S is new equity issues, and L is debt.

Internal and external funds are unlikely to be perfect substitutes, because of both outright restrictions (e.g., the minimum size required for a stock market listing) and a number of direct and indirect costs that are likely to make external funds more expensive than funds generated internally by the firm. In particular, the existence of informational asymmetries between borrowers and lenders will result in a higher cost of external finance, because the lender—lacking full information—is unable to specify and verify the behavior of the borrower in all situations (contracts are “incomplete”). In such a situation, lenders will have to screen and monitor borrowers, seeking to identify high-quality borrowers so as to minimize the risk of default. However, the absence of full information raises the problem of adverse selection, where it is not possible for a “good” company to completely distinguish itself from a “bad” company. As a consequence, the good company may face a premium on its cost of borrowing. Once the loan has been made, the lender also needs to monitor the actions of the borrower to ensure that the borrower is not acting contrary to the lender’s interests (e.g., by undertaking higher-risk projects that may increase the potential payoff to the borrower, but that also raise the risk of default). As screening and monitoring are costly and imperfect, the price of credit will be higher than if there were full information and complete contracts.10

The presence of these external finance costs leads to a financing “hierarchy,” with internally generated funds being the cheapest source of finance, followed by debt, and then equity (Myers, 1984).11 Within debt finance, banks may have a competitive advantage over the capital markets for certain types of borrowers because their closer relationship with the borrower allows them to screen and monitor behavior more effectively.

A number of studies have found that capital structures in industrial countries generally conform to the predictions of the financing hierarchy (see, e.g., Mayer, 1988, 1989). Details of the corporate capital structure in a number of industrial countries are reported in Table 3. While there are important differences across countries and between subperiods, the results are generally as reported in previous studies, with retained earnings being the single most important source of financing in all but one country. Debt financing is the primary source of external funding, with bank loans being the most important component of debt financing in France, Italy, and Japan, and the bond market in Canada and the United States. New equity issues are much smaller. Furthermore, there has been no trend toward greater use of external finance in any of these countries over the sample period.

Table 3.Corporate Capital Structure in Selected Industrial Countries(As a percent of investment in financial and nonfinancial assets)
External FinanceRatio of Investment to GDP
Debt financeEquity financeTotalInternal Finance
BanksBondsOther1Total
United States
1980-946.011.27.124.4-2.122.377.718.4
1980-857.711.77.526.0-1.725.274.819.8
1986-9010.012.17.729.8-5.524.375.718.4
1991-94-1.79.35.913.52.215.884.216.5
Canada2
1980-935.47.920.033.311.745.054.020.8
1980-858.17.017.732.812.545.354.721.1
1986-904.713.017.635.310.345.554.521.7
1991-932.9-0.427.930.413.343.756.318.5
Japan
1985-9327.54.720.552.66.759.340.730.4
1985-8926.85.725.648.67.565.734.329.4
1990-9328.43.112.744.25.549.750.331.6
Italy3
1982-939.71.424.135.29.744.953.320.9
1982-8412.84.825.643.317.260.537.722.4
1985-896.61.626.735.09.044.054.121.5
1990-9311.6-0.220.932.37.539.858.719.4
France3
1981-925.80.718.625.214.940.147.520.7
1981-8521.91.430.253.511.765.227.420.3
1986-89-4.01.015.512.418.931.455.920.5
1990-924.8-0.112.817.512.429.954.021.3
Sources: OECD, Non-Financial Enterprises Financial Statements, 1994; and IMF, International Financial Statistics (various issues).

Trade credit and accounts payable.

“Other” also includes foreign bank loans and borrowing from affiliates.

Debt plus equity plus retentions does not equal total finance owing to the existence of capital transfers.

Sources: OECD, Non-Financial Enterprises Financial Statements, 1994; and IMF, International Financial Statistics (various issues).

Trade credit and accounts payable.

“Other” also includes foreign bank loans and borrowing from affiliates.

Debt plus equity plus retentions does not equal total finance owing to the existence of capital transfers.

Work on corporate financial structure in developing countries has been more limited (see International Finance Corporation, 1991; Singh and Hamid, 1992; and Singh, 1995). These studies find that, while retained earnings are an important source of finance for developing country firms, there is greater reliance on external finance than in industrial countries (Table 4). These firms are more reliant on equity financing in particular, but debt finance also plays a key role.12

Table 4.Evidence on Corporate Financing in Developing Countries(In percent of growth in net assets)
Whole SampleMalaysia1Thailand1
Internal finance38.835.627.7
Equity finance39,346.6
Debt finance20.817.8
Memorandum items:
Ratio of debt to net assets25.315.055.0
Ratio of debt to equity51.927.3154.4
Source: Singh (1995).Note: Sample is the largest 100 manufacturing companies in each country for 1980-90 (where data are available). Countries in the sample are Brazil, India, Jordan, the Republic of Korea, Malaysia, Mexico, Pakistan, Thailand, Turkey, and Zimbabwe.

Sample is 1983-90.

Source: Singh (1995).Note: Sample is the largest 100 manufacturing companies in each country for 1980-90 (where data are available). Countries in the sample are Brazil, India, Jordan, the Republic of Korea, Malaysia, Mexico, Pakistan, Thailand, Turkey, and Zimbabwe.

Sample is 1983-90.

The results for Malaysia and Thailand are broadly consistent with those for the sample as a whole.13 For the average company in the sample, retentions account for 28 percent of finance in Thailand and for 36 percent in Malaysia. In Malaysia, equity finance accounted for a more significant proportion of external finance than debt (47 percent against 18 percent). Equivalent data for Thailand are not available. However, other statistics indicate that debt is a more important source of external funding in Thailand than in Malaysia. The average ratio of debt to net assets is 55 percent in Thailand, compared with 15 percent in Malaysia, while the average ratio of debt to equity is 154 percent in Thailand, compared with only 27 percent in Malaysia.14

An alternative to using balance sheet data is to take a more macro approach through an analysis of financial flows to the private corporate and public enterprise sectors.15 Lending from banks and other financial intermediaries (domestic and international) and capital raised in the domestic and international capital markets reveal how the capital structure of companies has changed in recent years. In Thailand, funding from domestic sources accounted for almost all finance raised before 1991 (Table 5). Since then, the open capital account has allowed funds raised overseas to become an increasingly important source. Within the domestic component, financial intermediaries account for most of the finance provided. Although equity and bond markets have developed, these still provide a relatively small proportion of the funds raised. Foreign bond issues have increased since 1993, but international equity issues have been insignificant. This may be because firms have access to reasonably cheap equity funds at home, but cannot issue debt or borrow from banks as cheaply or for as long. Also, there are limits on the proportion of domestic equity that foreigners can hold. State enterprises have not been big borrowers in the international markets, although they have accounted for a little over one-half of domestic bond issues.

Table 5.External Financing of Thai Enterprises
1980198119821983198419851986198719881989199019911992199319941995
(In percent of total external funds)
Domestic97.796.697.694.791.871.076.096.797.796.695.689.990.877.885.089.6
Debt95.594.093.793.282.564.968.384.992.187.986.179.879.974.180.681.2
Financial intermediaries95.594.093.793.282.564.968.384.992.187.984.671.974.165.572.273.8
Of which:
Commercial banks83.477.577.684.577.862.754.477.778.873.269.056.554.949.454.959.8
Bonds0.00.00.00.00.00.00.00.00.00.01.47.95.88.68.47.4
Equity12.32.63.91.69.26.17.711.95.68.79.510.010.93.64.38.4
International2.33.42.45.38.229.024.03.32.33.44.410.19.222.215.010.4
Debt2.33.42.45.38.229.024.03.32.13.24.19.79.220.913.49.3
Syndicated loans22.33.42.43.27.514.524.03.32.13.04.19.69.018.210.67.9
Bonds0.00.00.02.00.714.50.00.00.00.30.00.10.22.72.81.4
Equity0.00.00.00.00.00.00.00.00.20.20.30.50.01.41.61.1
(In percent of GDP)
Memorandum items:
Total external finance4.75.56.610.07.16.42.59.412.515.721.616.817.726.832.526.9
Borrowing by nonfinancial public enterprises in international markets0.00.10.20.40.51.90.60.30.20.20.50.70.90.50.90.3
Sources: Bank of Thailand, Quarterly Bulletin (various issues); Stock Exchange of Thailand (1995); data provided by the Thai Bond Dealers Club; and IMF, Developing Country Bonds, Equities, and Loans Database.

Includes issues by the financial sector.

Excludes short-term (less than one year) loans.

Sources: Bank of Thailand, Quarterly Bulletin (various issues); Stock Exchange of Thailand (1995); data provided by the Thai Bond Dealers Club; and IMF, Developing Country Bonds, Equities, and Loans Database.

Includes issues by the financial sector.

Excludes short-term (less than one year) loans.

While domestic funds are also the main source of finance for companies in Malaysia, the proportion of overseas finance is significantly higher than in Thailand (Table 6), mainly because of borrowing by non-financial public enterprises, particularly through syndicated bank loans. Within the domestic funding component, loans from financial intermediaries have fallen significantly, while funds raised through domestic bond and equity markets have become more important and, in some years, have come close to the amounts borrowed from financial institutions. Because the domestic equity market has consistently been an important source of financing during the sample period, this increased funding from domestic financial markets has mainly reflected the rising importance of the bond market, which now accounts for over 15 percent of funds raised.

Table 6.External Financing of Malaysian Enterprises
1980198119821983198419851986198719881989199019911992199319941995
(In percent of total external funds)
Domestic75.196.989.496.296.097.377.273.892.496.693.192.678.474.171.683.0
Debt73.379.977.180.275.287.572.15.284.979.060.074.948.161.652.670.1
Financial intermediaries173.379.977.180.275.287.572.1-5.079.672.350.863.735.941.535.055.5
Of which:
Commercial banks73.369.764.866.757.064.763.7-20.633.360.441.353.327.729.228.446.0
Bonds20.00.00.00.00.00.00.011.25.36.69.211.112.220.117.614.6
Equity1.816.912.416.020.89.85.168.77.517.733.117.830.312.519.013.0
International24.93.110.63.84.02.722.826.27.63.46.97.421.625.928.416.7
Debt24.93.110.63.84.02.722.826.27.63.46.97.418.325.928.412.6
Syndicated loans324.93.110.63.84.02.722.826.27.63.46.97.418.316.314.75.0
Bonds0.00.00.00.00.00.00.00.00.00.00.00.00.09.513.77.5
Equity0.00.00.00.00.00.00.00.00.00.00.00.03.20.00.04.2
(In percent of GDP)
Memorandum items:
Total external finance13.99.28.111.311.98.55.22.58.613.822.619.120.515.923.136.4
Borrowing by nonfinancial public enterprises in international markets2.20.30.90.40.50.11.20.71.00.50.81.33.44.04.62.5
Sources: Data provided by Bank Negara Malaysia; and IMF, Developing Country Bonds, Equities, and Loans Database.

Figures for 1981-88 include finance company lending only; those for 1989-95 include finance company and merchant bank lending.

Excludes issues from Cagamas, the national mortgage corporation, but includes commercial paper.

Excludes short-term (less than one year) loans.

Sources: Data provided by Bank Negara Malaysia; and IMF, Developing Country Bonds, Equities, and Loans Database.

Figures for 1981-88 include finance company lending only; those for 1989-95 include finance company and merchant bank lending.

Excludes issues from Cagamas, the national mortgage corporation, but includes commercial paper.

Excludes short-term (less than one year) loans.

Given data on investment and estimates of external finance, estimates of internal finance can be derived as the residual from equation (3).Table 7 reports the results of this exercise. Two sets of results are reported for each country. For Thailand, the first set is for the private corporate sector only, while the second is for the private corporate and public enterprise sectors combined. Both sets of results assume that companies are financing gross domestic physical investment.16 For Malaysia, both sets of results are for the private corporate and public enterprise sectors. The first set of results assumes that companies are financing gross physical investment, while the second assumes they are financing both physical and financial investments (approximated by the increase in corporate deposits with financial institutions).

Table 7.Estimates of Corporate Capital Structure in Malaysia and Thailand(In percent of gross investment)
1980-841985-901991-95
Malaysia
Physical investment1
Debt financing49.735.958.6
Financial intermediaries49.733.143.1
Bonds0.02.915.5
New equity issues7.110.014.8
Internal finance143.154.026.6
Investment to GDP19.420.730.9
Physical and financial investment2
Debt financing43.333.350.1
Financial intermediaries43.330.736.9
Bonds0.02.713.2
New equity issues7.110.014.8
Internal finance250.557.337.1
Investment to GDP22.122.236.3
Thailand
Private enterprises
Debt financing27.935.160.3
Financial intermediaries27.935.157.7
Bonds0.00.02.5
New equity issues1.23.65.7
Internal finance70.861.334.0
Investment to GDP22.225.233.4
Private and public enterprises
Debt financing24.933.660.6
Financial intermediaries24.732.854.3
Bonds0.20.86.2
New equity issues1.03.25.2
Internal finance74.163.234.3
Investment to GDP26.229.036.9
Source: IMF staff estimates.

Investment is defined as total private and public sector investment less government development expenditure.

Investment is defined to also include corporate deposits at financial institutions.

Source: IMF staff estimates.

Investment is defined as total private and public sector investment less government development expenditure.

Investment is defined to also include corporate deposits at financial institutions.

The results all indicate that there has been a significant shift toward external finance in both countries, particularly during the 1990s. In Thailand, between 1980 and 1990, about two-thirds of investment is estimated to have been financed from internal funds; this fell to one-third between 1991 and 1995. Within external finance, debt has remained by far the most significant financing source and, within this, financial intermediaries are the most important. Their importance increased markedly during 1991-95, with financial intermediaries accounting for more than one-half of funding. Bond and equity financing has increased, but remains small (totaling about 10 percent of funding in 1991-95). Similarly, in Malaysia, the role of external finance has grown, although this is due to increased use of the bond and equity markets rather than to increased borrowing from financial intermediaries. Business surveys conducted by the Bank of Thailand and Bank Negara Malaysia broadly confirm these results (Table 8).

Table 8.Survey Evidence on Sources of Financing for Gross Fixed Capital Expenditure
Malaysia1Thailand2
SurveyTable 73SurveyTable 73
Borrowing from financial institutions33.533.158.358.0
Equity1.96.8
Internal40.437.829.329.0
Other26.229.110.56.2
Sources: Bank Negara Malaysia; and Bank of Thailand.

Results for Malaysia are for 1993 and 1994. Survey covers approximately 1,000 companies in the agricultural, mining, manufacturing, services, transportation, and construction sectors.

Results for Thailand are for 1995. Survey covers 775 companies in the manufacturing sector.

Results for years corresponding to the survey years (not explicitly displayed in Table 7).

Sources: Bank Negara Malaysia; and Bank of Thailand.

Results for Malaysia are for 1993 and 1994. Survey covers approximately 1,000 companies in the agricultural, mining, manufacturing, services, transportation, and construction sectors.

Results for Thailand are for 1995. Survey covers 775 companies in the manufacturing sector.

Results for years corresponding to the survey years (not explicitly displayed in Table 7).

Compared with Singh’s (1995) findings, the results presented in this paper suggest a more important role for debt in Malaysia and for internal finance in both countries. This result may stem from differences in the data sets: Singh looked at only the largest quoted companies, and his sample ended in 1990. Large companies, given their greater access to the capital markets, are likely to rely less on internal finance than the corporate sector as a whole. For small companies, as deregulation has allowed risk to be priced more appropriately and has opened up new financing sources for the banks, credit should have become more easily available. Also, banks have increasingly had to look at the smaller company market as large companies have taken advantage of new financing options.

A number of potential explanations exist for the apparent increased reliance on external financing sources. For instance, one could argue that, as capital markets and financial intermediaries have developed, the informational asymmetries that exist between borrowers and lenders have declined (e.g., owing to better research and greater disclosure requirements), reducing the costs of external finance. But, it is also possible—and perhaps more likely—that the underlying high rates of economic growth and the sharp increase in the growth of investment in both Malaysia and Thailand have necessitated greater reliance on external finance, facilitated by the development of domestic capital markets and capital account liberalization (see Sussman, 1994, for a similar argument for differences in the importance of internal finance between Japan and the United States).

Developments in the Balance Sheet of the Personal Sector

Under the rational-expectations-permanent-income hypothesis of consumption behavior (Hall, 1978), individuals consume out of their expected lifetime “permanent” incomes. In broad terms, if current incomes are below those expected in the future, individuals will borrow so as to smooth their lifetime consumption. However, if capital markets are imperfect, individuals may be liquidity constrained and unable to smooth consumption optimally. A number of studies on industrial countries (e.g., Blundell-Wignall, Browne, and Cavaglia, 1991) have found that financial deregulation, by increasing the borrowing opportunities available to individuals, has reduced such liquidity constraints and allowed greater scope for consumption smoothing. As individuals have increased their borrowing, the stock of outstanding personal sector debt has risen.

Personal sector debt in Malaysia and Thailand has risen quite rapidly in recent years (Figure 8). In Malaysia, the outstanding debt stock had reached 30 percent of GDP by the end of 1995. Borrowing for consumption purposes has accounted for a large proportion of this rise, increasing from 6 percent of GDP in 1990 to 14 percent of GDP in 1995, while borrowing for housing purposes has remained broadly unchanged since the mid-1980s. In Thailand, borrowing for consumption (the only data available) has increased to 22 percent of GDP, compared with 11 percent in 1990. Consequently, the leverage of the personal sector, like that of the corporate sector, has significantly increased in recent years.

Figure 8.Personal Sector Debt

(In percent of GDP)

Sources: Bank of Thailand, Quarterly Bulletin (various issues); and Bank Negara Malaysia.

Banks, Capital Markets, and the Monetary Transmission Mechanism

This section looks at the implications for the conduct of monetary policy of the changes in the financial structure of the Malaysian and Thai economies described in the previous section.

The monetary transmission mechanism is defined by the impact of a change in the monetary policy instrument (usually the short-term interest rate or base money) on intermediate variables (such as broad money or domestic credit) and final objectives (output and inflation). In a standard, closed-economy model of this process, the demand for and the supply of money determines the short-term interest rate. The central bank, by operating on the supply of money, can bring about changes in this rate. For example, by selling bonds to the public, the central bank can reduce the supply of money. Interest rates must then rise in order to encourage increased bond holdings and restore equilibrium. If prices are slow to adjust, a change in the nominal interest rate results in a change in the real rate that, in turn, affects interest-sensitive consumption, investment, and output decisions. As prices adjust in the longer run, the real effects dissipate. However, this standard model does not explain how relatively small changes in short-term interest rates can generate large observed changes in economic activity—in particular, long-term investment and durable consumption spending—which often occur with substantial time lags.

The credit view of the monetary transmission mechanism enriches the standard model by considering the cost differential between internal and external funds (the external finance premium) that may arise under the existence of adverse selection, moral hazard, or both.17 Policy-driven changes in short-term interest rates may affect this premium through two possible channels: the bank-lending channel and the balance-sheet channel (see Bernanke and Gertler, 1995, for an overview of this literature).

Models of the bank-lending channel typically rely on two assumptions (see, e.g., Bernanke and Blinder, 1988). First, there is imperfect substitutability between bank lending and other sources of capital for some borrowers. As discussed in the previous section, this may be true—particularly for households and small companies—when large monitoring costs exist. Second, it is assumed that the central bank can affect the volume of bank credit through its open market operations. For example, a sale of government bonds enacted through the banking system reduces bank reserves, which, in turn, causes a reduction in bank lending if the reserves cannot easily be replaced, say, by issuing certificates of deposit. It follows that the central bank’s ability to affect the supply of credit provides monetary policy with a distinctive way to influence economic activity through its ability to directly alter the availability of funds to some sectors of the economy.

The balance-sheet channel focuses on the transmission of monetary policy through its impact on private balance sheets (see Bernanke, Gertler, and Gilchrist, 1996, for a broad overview). For companies, a tightening of monetary policy that raises interest rates will weaken their balance sheets by reducing cash flow. This may happen directly through higher interest payments on existing debt or by reducing the value of assets used for collateral (as higher rates of discount are applied to expected future asset earnings), or both. The deterioration in balance sheets may also occur indirectly, as reduced spending by firms that are directly affected feeds through to other firms in the economy. Given that changes in corporate net worth affect the premium on external finance, the decline in internally generated funds is likely to make external finance more expensive and thereby impinge on investment spending. Households may face a similar deterioration in their balance sheets and, in addition, may reduce consumption spending on the basis of reduced wealth from declining asset prices.

Studies of the structure of corporate balance sheets and informational asymmetries in capital and credit markets in industrial countries suggest that higher cash flows boost investment by providing more low-cost funding and collateral backing for external finance; higher leverage discourages investment by raising the cost of external financing; and cash flow is more important for small firms and highly leveraged firms because of their limited access to the capital markets (see Fazzari, Hubbard, and Petersen, 1988, for the United States; Devereux and Schiantarelli, 1989, for the United Kingdom; and Mills, Morling, and Tease, 1993, 1994, for Australia). These findings imply that changes in short-term interest rates can influence investment, not only through the discount rate used to assess investment projects, but also through the impact on corporate cash flow. Companies that depend on bank loans to finance investment will bear the brunt of any tightening in policy, although even firms that have access to other sources will face higher costs of external finance. In general, the sensitivity of companies to tighter monetary conditions will depend on their initial levels of debt, with the more highly leveraged firms being the most affected.

The impact of financial liberalization and capital market development on the bank-lending and balance-sheet channels of monetary policy is not clear. The bank-lending channel is likely to be weakened as alternative sources of finance develop. Companies have more scope to switch away from bank financing, and alternative sources of funding for financial intermediaries develop that allow them to become less reliant on deposits. However, the balance-sheet channel will be strengthened to the extent that corporate debt levels rise in response to deregulation and capital market development and make the corporate sector more sensitive to any given change in interest rates.

To investigate the monetary transmission mechanisms in Malaysia and Thailand, vector autoregressions (VARs)18 are estimated to characterize the dynamic relationships among the key variables in each economy. These can then be used to trace the predicted impact of a shock to one variable on the other variables in the system. In Malaysia, monthly data were available for January 1976—April 1996, whereas in Thailand data were available for January 1987–May 1996. Because of the short Thai sample, more emphasis is placed below on the results for Malaysia. Full details of the data and procedures used are provided in Appendix II.

For Malaysia, a five-variable VAR was estimated using the interbank interest rate, domestic credit, industrial production, consumer prices, and the exchange rate. Two exogenous variables—an index of oil and commodity prices and the London interbank offered rate (LIBOR)—were also included in order to control for external supply-side and monetary shocks. As Bank Negara Malaysia often targets the interbank interest rate, it is interpreted here as an indicator of the stance of monetary policy. A shock to the interest rate equation is thus taken to be a change in monetary policy, and the responses of the other variables are interpreted as the structural responses of economic activity to this change.

The top panel of Figure 9 shows the impulse response functions of three of the variables in the Malaysian system to a positive, onetime, 1 percentage point shock to the interest rate, using results from the system estimated on the full data sample.19 Domestic credit begins to decline about ten months after the shock. The decline in output begins about six months after the shock and then reverses after a year and a half. The decline in the price level is very small. However, it should be noted that most of these responses are not significantly different from zero (in a statistical sense).20

Figure 9.Malaysia: Responses to a 1 Percentage Point Increase in the Interest Rate

(In percent of initial value)

Source: IMF staff estimates.

To see whether the transmission of monetary policy has changed with the development of capital markets in Malaysia, the VAR was reestimated using only the 1987-96 observations. As shown in the bottom panel of Figure 9, the responses of all three variables to a 1 percentage point interest rate shock are significantly larger. Domestic credit begins a steady fall five months after interest rates rise. Output falls immediately, continues falling for about three years after the shock, and then returns to its initial value by the sixth year. The price level falls slowly and remains below its initial value.

At face value, these results confirm that interest rate policy has increased in potency in recent years. However, because both capital market development and financial liberalization have occurred during this time, it is difficult to say whether this change is due to a strengthened balance-sheet channel (increased interest rate sensitivity attributable to higher leverage ratios) or interest rate liberalization. The insignificant responses of all variables to interest rate changes predicted by the full-sample VAR suggest that the transmission of changes in interbank rates to other interest rates—such as lending rates—may have been weak during the earlier part of the sample.21

VARs for Thailand were estimated using data on the minimum loan rate,22 domestic credit, manufacturing production, consumer prices, and the exchange rate; again, an oil and commodity price index and LIBOR were included as exogenous variables. The top panel of Figure 10 plots the responses to a 1 percentage point shock to the minimum loan rate. Domestic credit is predicted to rise above its initial value for a few months and then to begin a sustained decline. Output contracts strongly for about a year and then slowly returns to its initial level. Prices fall by only a small amount. However, as in the full-sample Malaysian VAR, few of the responses are significantly different from zero. The bottom panel of Figure 10 displays results for a system estimated with a sample that begins three years later, in 1990.23 Although the magnitudes of the domestic credit and price responses appear to be smaller in the latter part of the sample, these responses are more precisely estimated than for the full sample and are more often statistically significant. While these results provide weak support for a stronger transmission mechanism during the 1990s, the lack of a more definitive result may be due to the short data sample that required an overly parsimonious model to be estimated and to a comparison across two samples that have more observations in common than not. It is also possible that changes in the minimum loan rate are simply not very informative about the stance of monetary policy in Thailand.24

Figure 10.Thailand: Responses to a 1 Percentage Point Increase in the Interest Rate

(In percent of initial value)

Source: IMF staff estimates.

The VAR analysis undertaken in this section provides some support for the hypothesis that interest rate policy has become more potent following financial deregulation and capital market development. In Malaysia, changes in the interest rate appear to have a larger impact on domestic credit, which feeds through to larger changes in output and prices. In Thailand, changes in the minimum lending rate appear to elicit more significant responses in domestic credit, output, and prices. While the results are consistent with a strengthening of the balance-sheet channel, which could occur as higher leverage ratios lead to greater interest rate sensitivity in these economies, they also reflect the substantial financial liberalization in both countries, which may have contributed to a stronger link between lending rates and policy instruments. While the latter should improve the potency of policy on its own, the impact will be magnified if firms and households are, in general, more sensitive to interest rate changes.

Conclusion and Policy Implications

Capital markets have developed rapidly in both Malaysia and Thailand in recent years, facilitating a noticeable shift from internal to external finance, as well as higher rates of investment and growth. In Thailand, the shift has occurred through increased borrowing from financial intermediaries, while in Malaysia, use of the capital markets has expanded. This pattern appears to reflect the greater development of debt and equity markets in Malaysia, and the more liberal regulation of the Thai banking sector and—associated with this—the rapid growth of the BIBF. This paper has provided some evidence that these developments, in combination with financial liberalization, have increased the efficiency of interest rate policy.

In the future, the implementation of monetary policy will be affected by three key trends. First, firms’ use of external finance—and thus their leverage ratios—will depend crucially on the outlook for growth and investment. In the short term, recourse to external finance is likely to remain high. This should decline as growth and investment rates slow over the medium term, although debt levels are still likely to be higher than they were before capital market development.

Second, there is likely to be a further shift from bank to bond financing and a resultant decline in the role of financial intermediaries relative to other sources of funding. Dalla and others (1995) predict that Asian bond markets will need to grow substantially over the next ten years because of the need for infrastructure spending, which requires especially long-term finance. If the current high rates of investment do not increase further, this would imply a reduction in the share of bank financing, although the role of financial intermediaries in the economy may not decline if more bank revenue is derived from fee-based activities.

Third, the composition of bank loan portfolios is likely to change. More firms will have greater access to nonbank sources of finance as disclosure requirements and the development of rating agencies reduce monitoring costs. Banks will increasingly look to smaller borrowers as large companies take advantage of alternative financing options, and the debt levels of small firms and consumers are likely to rise.

These trends have several implications for the conduct of monetary policy in the immediate future. The emergence of new financial instruments and nonbank sources of funding is likely to weaken the bank-lending channel and, also, to make direct controls on the banking sector less effective. However, as the debt levels of firms and individuals rise, economic activity should become more sensitive to interest rate changes. Given that financial liberalization is likely to continue—thereby increasing the speed with which changes in policy interest rates are transmitted through the system—there is scope for the potency of interest rate policy to increase over the medium term. Although the net impact on monetary policy effectiveness is theoretically ambiguous, the results in this paper indicate that, in the recent past, the improving strength of the balance-sheet channel may have outweighed the deterioration in the bank-lending channel. In the longer run, the effects of continued financial liberalization and capital market development may be offset by slowing investment and declining leverage ratios. Recent simulations conducted for the major industrial countries suggest that the response of GDP to changes in the policy interest rate has increased as financial deregulation and capital market development have continued in these countries (International Monetary Fund, 1996).

Another implication is that the sectoral impact of monetary policy is likely to change. As bank portfolios become more concentrated in loans to small companies and households, any policies that do act on the bank-lending channel will have the greatest impact on these small borrowers. Moreover, small firms and consumers are likely to be even more interest sensitive than large firms (and infrastructure projects) and, so, may also bear a greater burden of adjustment to policies acting through the balance-sheet channel.

Given an objective of maintaining (or increasing) the potency of interest rate policy within a sound financial system, this discussion highlights the importance of accompanying capital market development with continued financial liberalization and enhanced supervision, and of collecting detailed data—for example, on firm finance, intermediary deposits and loans by sector, and asset prices—to assess the impact of the trends described above for monetary management in the future.

Appendix I. Chronology of Financial Liberalization and Market Reforms
Thailand
April 1975Securities Exchange of Thailand begins trading (name changed to the Stock Exchange of Thailand (SET) in 1991).
March 1985The baht interbank offered rate (BIBOR) is introduced.
April 1987The requirement that at least 30 percent of initial public offering shares on the SET must be held by small shareholders is removed.
September 1987A “foreign board” is established on the SET.
June 1989Interest rate ceilings on time deposits with maturity exceeding one year are abolished.
Early 1990sBank of Thailand applies guidelines from the Bank for International Settlements on asset quality and capital adequacy to commercial banks and finance companies.
March 1990Interest rate ceilings on all time deposits are abolished.
May 1990Thailand accepts obligations under the IMF’s Article VIII, and exchange controls on all current account transactions are liberalized.
April 1991Individuals and companies are permitted to open limited foreign currency accounts for direct investment purposes; investment funds, dividends, and loan repayments may be freely repatriated; purchases of foreign property and securities by residents, and large foreign direct investments or loans by residents to foreign affiliates still require central bank approval.
The SET introduces a fully computerized trading system to replace its floor trading system.
June 1991The reserve requirement ratio is replaced by the liquidity ratio: banks must still maintain at least 7 percent of deposits in securities and cash, but the definition of “securities” is broadened to include nongovernment securities.
1992The preferential tax rate for listed (versus nonlisted) companies is removed.
The mutual fund industry is deregulated, and licenses are granted to seven new fund management companies.
March 1992Finance companies are authorized to act as selling agents for government bonds; to provide economic, financial, and investment information services; and to advise companies seeking listing on the SET
Commercial bank activities are expanded to include issuance, underwriting, and distribution of debt securities, trading of debt securities, and acting as supervisors and selling agents for mutual funds and securities registrars.
May 1992The Securities and Exchange Act (SEA) establishes the Securities and Exchange Commission (SEC) and defines rules, procedures, and supervision for each type of securities business, including private funds management. The SEA also allows limited and public companies to issue debt instruments with the approval of the SEC.
June 1992Ceilings on saving deposit rates and all lending rates are abolished.
September 1992A scripless clearing and settlement system is introduced on the SET.
March 1993The Bangkok International Banking Facilities are established. Participants may provide three types of services: banking to nonresidents in foreign currencies and baht (“out-out” transactions), banking to domestic residents in foreign currency only (“out-in” transactions), and international financial and investment banking services.
May 1993The requirement that banks must hold a proportion of their deposits in government and other eligible bonds before opening new branches is abolished.
July 1993The Thai Rating and Information Service is established.
October 1993Each bank is required to declare a minimum lending rate, which is the rate on term credits to prime customers. Banks are also required to declare rates for general and large depositors.
September 1994The Bond Dealers Club is founded as an officially sanctioned, but privately run, trading system operating under an established code of conduct and standardized dealing and settlement procedures.
August 1995Weekly auctions of Bank of Thailand bonds are introduced.
September 1995The Bangkok Stock Dealing Center is established as an over-the- counter market to provide capital to nonlisted companies.
June 1996Short-term offshore borrowing by financial institutions is subject to a 7 percent liquidity requirement.
Malaysia
January 1972Interest rate ceilings on commercial bank deposits with maturity exceeding one year are lifted.
1973With the termination of currency interchangeability between Malaysia and Singapore, the Stock Exchange of Malaysia is split into the Kuala Lumpur Stock Exchange and the Stock Exchange of Singapore, although many Malaysian companies continue to be listed on the Stock Exchange of Singapore, and vice versa.
August 1973Interest rate ceilings on all finance company deposits are lifted.
Discount rates for treasury bills are determined by open tender in the money market.
October 1978Commercial banks are allowed to freely determine all deposit and lending rates.
May 1979Negotiable certificates of deposit and banker’s acceptances are introduced.
March 1983The Islamic Banking Act of 1983 allows the establishment of Bank Islam Malaysia Berhad.
July 1983Non-interest-bearing government investment certificates are introduced.
November 1983Commercial banks and finance companies are required to declare a base lending rate based on their costs of funds; actual lending rates must be anchored to the base lending rate.
October 1985-January 1987Owing to tight liquidity conditions, the interest rates on commercial bank and finance company deposits with a maturity of up to one year must be pegged to the deposit rate of the two lead domestic banks.
September 1987Commercial bank base lending rates must not exceed that of the two lead banks by more than 0.5 percentage point; finance companies are subject to similar guidelines.
October 1987Cagamas, the national mortgage corporation, issues its first mortgage-backed bonds.
1988The Second Board on the Kuala Lumpur Stock Exchange is launched, enabling the listing of smaller companies.
January 1989Two-tier liquid asset ratio requirements are abolished for finance companies. Repurchase agreements are added to the list of eligible liabilities and thereby become subject to the statutory reserve requirement and the minimum liquidity ratio.
Bank Negara Malaysia issues guidelines on the operation of the corporate bond and promissory note markets.
A principal dealer system for government securities is introduced.
March 1989Non-trade-related swap transactions undertaken with offshore banks are subject to a daily limit.
May 1989The statutory reserve requirement is realigned to a uniform 4.5 percent for commercial banks and finance companies.
September 1989The weighted risk asset approach of the Bank for International Settlements is introduced as the uniform method of capital adequacy assessment.
Commercial and merchant banks are allowed limited participation in several initial public offerings and privatizations.
October 1989The Banking and Financial Institutions Act of 1989 places all banking institutions under the supervision of Bank Negara Malaysia.
January 1990Singapore-incorporated companies are delisted from the Kuala Lumpur Stock Exchange, and Malaysia-incorporated companies are delisted from the Singapore Stock Exchange.
March 1990Larger finance companies are allowed to issue negotiable certificates of deposit.
June 1990Two-tier liquid asset ratio requirements are abolished for commercial banks.
October 1990The Labuan International Offshore Financial Centre establishes a favorable tax environment, confidentiality rules, and no exchange controls for the conduct of international business activities in banking, insurance, corporate funding, investment and trust management, professional services, and other related activities. Financial institutions engage primarily in foreign currency business for residents and nonresidents.
February 1991Commercial banks and finance companies may declare their own base lending rates based on their own costs of funds; an institution may not lend at a rate below its base lending rate (except for loans with interest rates prescribed by Bank Negara Malaysia, such as low- cost housing loans), and the maximum spread between the declared base lending rate and actual lending rates is 4 percentage points.
October 1991All outstanding ringgit received through swap transactions with, and direct borrowings from, nonresidents (including offshore banks) are included in the eligible liability base.
Interbank borrowing limits on finance companies participating in the interbank money market are lifted. Deposit-taking activities of finance companies and merchant banks are expanded. Discount houses are permitted to invest, trade, underwrite, and manage issues of eligible private debt securities, as approved by Bank Negara Malaysia.
1992Rating Agency Malaysia Berhad is established to rate debt issues by corporations.
June 1992Non-trade-related swaps with all foreign customers are subject to a daily limit.
November 1992Trading on the Kuala Lumpur Stock Exchange becomes fully automated.
February 1993Bank Negara Malaysia bills are introduced.
March 1993The Securities Commission is established as the primary regulator and supervisor of securities markets.
The interest-free banking scheme is launched.
January 1994All funds sourced from abroad are included in the eligible liability base.
January 1994-August 1994All residents are prohibited from selling short-term monetary instruments to nonresidents; short-term private debt securities are also restricted as of February 1994.
January 1994-January 1995The outstanding net external liabilities position of each banking institution is subject to a ceiling specified by Bank Negara Malaysia.
February 1994Ringgit funds of foreign banking institutions held in non-interest-bearing vostro accounts at commercial banks are required to be placed with Bank Negara Malaysia in non-interest-bearing current accounts; the statutory reserve requirement is applied to vostro balances between February and May 1994.
February 1994-August 1994Commercial banks are not permitted to undertake non-trade-related swaps or outright forward transactions on the bid side with foreign customers.
December 1994The two-tier regulatory system allows larger commercial banks to engage in selected new activities (such as the operation of foreign currency accounts for exporters) and to conduct selected aspects of their operations under a more liberal regulatory environment.
1995Call warrants are listed and traded on the Kuala Lumpur Stock Exchange.
October 1995The formula used by commercial banks and finance companies to calculate their base lending rates is revised to better reflect the underlying costs of funds; an institution may not lend at a rate above its base lending rate.
December 1995The Kuala Lumpur Options and Financial Futures Exchange commences trading in stock index futures.
May 1996The Malaysia Monetary Exchange commences trading in interest rate-futures.
October 1996Short selling of selected stocks is permitted on the Kuala Lumpur Stock Exchange.
Sources: For Thailand, Bank of Thailand, Quarterly Bulletin (various issues); Beng (1994); Stock Exchange of Thailand (1995); Vichyanond (1994); and Wibulswasdi (1995). For Malaysia, bank Negara Malaysia (1994) and Annual Report (various issues); Kuala Lumpur Stock Exchange (1996); and Zamani (1995).
Sources: For Thailand, Bank of Thailand, Quarterly Bulletin (various issues); Beng (1994); Stock Exchange of Thailand (1995); Vichyanond (1994); and Wibulswasdi (1995). For Malaysia, bank Negara Malaysia (1994) and Annual Report (various issues); Kuala Lumpur Stock Exchange (1996); and Zamani (1995).
Appendix II. Statistical Results Reported in Main Text

Vector Autoregressions

A vector autoregression is a system of ordinary least squares regression equations that estimate how each variable is related to the lagged values of all the variables in the system. More formally,

xt = β1xt-1 + … + β2xt-2 + βjxt-j + γ0zt + γ1zt-1 + … + γkzt-k + et,

where x is an n × 1 vector of endogenous variables, z is an m × 1 vector of exogenous variables, β and γ are n × n and n × m coefficient matrices, respectively, and e is a vector of error terms. Based on the estimated coefficients and error terms, such a system can be used to trace the predicted impact of a shock to one variable on the other variables in the system.

Since the estimated errors, êt, are likely to be interrelated (e.g., interest rate shocks and domestic credit shocks are, in practice, contemporaneously correlated), some method needs to be adopted to disentangle them. This will enable a shock to the interest rate equation to be interpreted as a surprise movement in the interest rate (owing to a change in monetary policy), rather than as part interest rate surprise and part domestic credit surprise (perhaps caused by a change in the demand for credit). For the impulse response analysis, the error terms were orthogonalized by using the Cholesky decomposition of their estimated variance-covariance matrix. The ordering of the variables used before decomposition therefore dictates the recursive chain of causality among the shocks in any given period.

For example, in generating Figures 9 and 10, the ordering of the variables was the interest rate, domestic credit, output, prices, and exchange rate. This implies that a shock at time t to the interest rate can also affect all other variables in the system—domestic credit, output, prices, and the exchange rate—during period t. In contrast, a shock to domestic credit affects only domestic credit, output, prices, and the exchange rate, but not the interest rate (until the next period). Output shocks affect output, prices, and the exchange rate contemporaneously; price shocks affect prices and the exchange rate; and exchange rate shocks affect only the exchange rate in the same period that they occur. Relatedly, any contemporaneous correlation between êi,t (the estimated error term associated with the interest rate equation) and any of the other estimated error terms, say, êdc,t (domestic credit) is attributed solely to the orthogonalized interest rate shock. In contrast, the orthogonalized exchange rate shock is attributed to only that component of êer,t that was uncorrected with the other estimated errors. The economic interpretation of the chosen ordering is, therefore, that a shock to the interest rate equation represents a choice by the policymaker to peg the interest rate at a new level for the next month. While other variables in the system can respond immediately to the interest rate change, the interest rate itself will not begin responding to the changes in these other variables until the following month.

Data

The data set used for Malaysia included monthly observations on the seven-day interbank interest rate (monthly average), domestic credit from the monetary survey, the industrial production index, the consumer price index, and the ringgit-U.S. dollar exchange rate (monthly average). Bank Negara Malaysia’s Monthly Statistical Bulletin was the source of the interest rate data; all other series were taken from the IMF’s International Financial Statistics. Although Bank Negara Malaysia most often intervenes in the interbank market at the one-month maturity, data on one-month and three-month interbank rates were available only from July 1982. In the interests of a longer sample period, the seven-day rate was used instead. The correlation between monthly changes in the two series is 0.82 over their common sample period.

The data set used for Thailand included the minimum loan rate, domestic credit from the monetary survey, the manufacturing production index, the consumer price index, and the baht-U.S. dollar exchange rate (period average). The minimum lending rate series was obtained from the Bank of Thailand’s Quarterly Bulletin, and the manufacturing production index was obtained directly from the Bank of Thailand. All other series were taken from International Financial Statistics.

Two exogenous series were used in the vector autoregressions. The index of oil and commodity prices was constructed as the average of the oil price index and the commodity price index available in the World Economic Outlook Database. The seven-day London interbank offered rate (LIBOR) (used in the Malaysian vector autoregressions) and the three-month LIBOR (used in the Thai vector autoregressions) were from International Financial Statistics.

Model Specification

All series were tested for stationarity, and all were found to be stationary in first differences, but nonstationary in levels (integrated of order one, or I(1)). Therefore, the vector autoregressions were estimated with all variables in log levels, except the interest rates, which were in decimal form. This specification assumes that the variables are I(1) and allows for—but does not impose—cointegration. The Malaysian vector autoregression, as initially estimated over the full sample period, included 12 lags of each of the 5 endogenous variables, contemporaneous values and 12 lags of the 2 exogenous variables, a constant term, and 11 seasonal dummies.25 A sequence of likelihood ratio tests indicated that 8 lags of the endogenous and exogenous variables were preferable to 12. The analysis in the main text is based on this more parsimonious model.

For Thailand, the shortness of the data sample was a severe constraint, and the model specification was based as much on feasibility as it was on tests of model adequacy. The analysis in the main text is based on a model that includes 4 lags of the 5 endogenous variables, the contemporaneous and once-lagged values of the 2 exogenous variables, a constant term, and 11 seasonal dummies. Even with this comparatively terse specification, 36 parameter estimates were required for each equation.

Impulse Response Functions

Figures 11-14. show the impulse response functions for all the endogenous variables and all the vector autoregression estimations discussed in the main text. The instigating impulse is a positive one-standard-deviation shock to the interest rate equation. In contrast to the main text, the responses are not normalized to a 1 percentage point initial shock, and the response functions are plotted inside two-standard-error bands. In comparing the point estimate of the response function with any other point lying vertically inside the standard-error band, the interpretation is that the two points are statistically indistinct from one another at the 5 percent level of significance.

Figure 11.Malaysia: Responses to a Positive 1 Standard Deviation Interest Rate Innovation, Plus or Minus 2 Standard Errors

(Model estimated on 1976-96 data)

Figure 12.Malaysia: Responses to a Positive 1 Standard Deviation Interest Rate Innovation, Plus or Minus 2 Standard Errors

(Model estimated on 1987-96 data)

Figure 13.Thailand: Responses to a Positive 1 Standard Deviation Interest Rate Innovation, Plus or Minus 2 Standard Errors

(Model estimated on 1987-96 data)

Figure 14.Thailand: Responses to a Positive 1 Standard Deviation Interest Rate Innovation, Plus or Minus 2 Standard Errors

(Model estimated on 1990-96 data)
References

Note: Tim Callen and Patricia Reynolds are Economists in the IMF’s Asia and Pacific Department. The authors thank David Robinson, Mahmood Pradhan, and Coenraad Vrolijk for helpful comments.

The role of the short-term money markets is not discussed in this paper. Institutional details of financial deregulation and capital market development are given in Appendix I.

Noncommercial bank financial intermediaries in Thailand include finance companies, the Government Savings Bank, the Bank for Agriculture and Agricultural Cooperatives, the Government Housing Bank, and the Industrial Finance Corporation of Thailand. Nonbank financial intermediaries in Malaysia are defined here to include finance companies and merchant banks.

The corporate bond market in Malaysia includes bonds issued by nonfinancial public enterprises.

A second credit rating agency was introduced in September 1996.

The BIBF was established in 1993 as Thailand’s offshore banking system. BIBF transactions include taking deposits from foreign residents and lending these funds either in Thailand or abroad.

While commercial banks are free to determine their own lending practices, the Bank of Thailand monitors credit allocation through the submission of six-monthly credit plans by the banks. In general, the central bank encourages commercial banks to restrain lending to areas that might boost inflationary pressures and to focus on the “productive” sectors.

To the extent that banks and finance companies borrow in foreign currency and relend in baht, they are assuming exchange rate risk, whereas the BIBF transactions leave the exchange rate risk with the bank customer. There appears to be little hedging of this risk. Hedging is regarded as expensive, and the stability of the baht and the perceived low risk of devaluation encourage an open position. However, a bank’s net foreign currency position cannot exceed 25 percent of capital funds if assets exceed liabilities and 20 percent of capital funds if liabilities exceed assets. In April 1995, the Bank of Thailand increased the minimum loan disbursement for the “out-in” lending from $500,000 to $2 million. The objective was to ensure that BIBF loans were extended to large corporate borrowers who could better manage the exchange rate risk.

The narrowing Malaysia U.S. interest differential also played a role in this switch.

Approval is required for principal amounts exceeding the equivalent of RM 5 million. Permission is usually granted if the loan proceeds will be used for “productive” purposes.

Other costs of external finance include the direct costs of underwriting and administrative fees and indirect costs, such as potential financial distress costs as the probability of bankruptcy rises with leverage. There are also issues related to the dilution of ownership if equity finance is raised (this may be particularly important for a family-owned business thinking of listing), the need to present required information to the public (which may also be perceived as passing information to competitors), and the tax treatment of different types of financing instruments (debt-interest payments are tax deductible in many countries).

There are a number of reasons why the cost of equity finance may be higher than debt finance. Under a debt contract, it is easier to specify the actions of the borrower, for example, by attaching covenants to the loan. However, for equity finance, dividend payments are discretionary, and, hence, a higher level of monitoring of the manager’s performance may be required to assess the true position of the company given that the manager’s incentives may not coincide with those of the shareholders. Also, equity issues are often regarded as an adverse signal regarding a firm’s future prospects because the current owners are reducing their ownership of the company.

Singh (1995) lists a number of factors behind the importance of equity finance in many developing countries: governments have actively played a major role in the expansion and development of these markets; the relative cost of equity capital has fallen significantly as a result of large rises in share prices during the course of the last decade at the same time that the cost of debt finance has increased; and the supply of new equity appears fairly elastic in developing countries. Also, the nature and extent of shareholder monitoring is typically very different from that in industrial countries, as is evidenced by the relative rarity of takeovers.

Unfortunately, the data for Thailand are incomplete in many aspects.

A number of potential problems may hamper the interpretation of these results. The mean of the ratios that are reported in Table 4 may conceal significant differences across companies (however, the results for the median company are not that different); there may be questions about the quality of the accounting data used; the companies in the sample are unlikely to be representative of small and medium-sized companies that do not have access to the capital markets and may have more limited access to bank finance; and, given that the sample stops in 1990, the changes that have taken place since then may have altered the results.

While it is possible to derive estimates of private corporate financing for Thailand, the data on financial institution lending in Malaysia do not allow the separation of private corporate and public enterprise borrowing. Consequently, for the sake of consistency, the results for the combined private corporate and public enterprise sectors are discussed for both countries. It is also not possible to exclude mortgage lending from the Thai lending data, so some lending to the household sector is caught in these figures. The financing raised from the domestic bond market and the international financial markets by the banking and finance sector is not directly included in these figures, but is, instead, effectively intermediated through the domestic banking system and appears as domestic lending (in either foreign or domestic currency).

Two shortcomings are associated with this assumption. First, direct loans from parent companies abroad will not be picked up in these calculations; other forms of foreign direct investment, however, should be accounted for. Second, in the absence of data on outward foreign direct investment, any funds raised for this purpose will be erroneously attributed to the financing of domestic investment.

If a firm is not given access to some markets, its cost of external finance is effectively infinite.

A VAR model is a system of ordinary least squares regression equations that estimate how each variable is related to the lagged values of all the variables in the system, A more detailed discussion is given in Appendix II.

As the variables are graphed in log levels, the responses can be interpreted as percentages of initial value. Hence, a value for domestic credit of –1 in the period after the shock means that domestic credit is 1 percent lower than its value in the initial period.

“Statistically significant” means that the estimated response is statistically different from zero at the 5 percent level of significance. Appendix II displays the impulse response functions graphed within two-standard-error bands.

In fact, the change in the interbank rate explains less than 4 percent of the change in actual lending rates between 1976 and 1986, but explains 31 percent in the later period. (This result is based on a comparison of the R2s from regressions of the monthly change in the lending rate on the contemporaneous and six-lagged monthly changes in the one month interbank rate, over the two sample periods.)

VARs employing the average overnight interbank rate were also estimated, but indicated no significant responses in either sample period.

Given the number of parameters to be estimated, six years is the shortest subsample that could be reliably used for estimation.

We obtained somewhat stronger results by assuming domestic credit to be the policy instrument; however, to maintain comparability with the Malaysian VARs, results using the minimum loan rate were presented here.

Production, price, and domestic credit data are highly seasonal in both countries. Seasonally preadjusted series were not used, however, because procedures such as XII fit two-sided moving averages through the data, and the adjusted observation for any given month actually incorporates information about previous and subsequent months. This makes the interpretation of the error terms as unanticipated shocks suspect. The disadvantage of including the seasonal dummies in the vector autoregressions is that—owing to the underlying assumption of a log-linear pattern of seasonality—the seasonal pattern is not fully accounted for. Hence, some of the estimated impulse responses have a “jagged” shape in early periods.

    Other Resources Citing This Publication