Report No. 8403-th Thailand Financial Sector Study (In Two Volumes) Volume I The Main Report May 25, 1990 Industry and Energy Operations Division Country Department 11 Asia Region FOR OFFICIAL USE ONLY Docuet of te World Bnk' This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents nay not otherwise be disclosed without World Bank authlrization. CURRENCY EQUIVALENTS (December 31, 1989) US$1 - Baht 25.61 Baht 1 US$0.039 FISCAL YEAR Government I October 1 to September 30 Most financial institutions s January 1 to December 31 LIST OF ACRONYMS BAMC - The Bank for Agriculture and Agricultural Coooperatives BOT - The Bank of Thailand CCC - Civil and Commercial Code of Thailand CDs - Certificates of Deposits FIDF - Financial Institutions Development Fund F&S Companies - Finance and Securities Companies FRNs - Floating Rate Notes GDP - Gross Domestic Product IFC - International Finance Corporation IFCT - The Industrial Finance Corporation of Thailand IMF - International Monetary Fund MFC - The Mutual Fund Company PNs - Promissory Notes PGs - Peer Groups RGs - Risk Groups SET - Securities Exchange of Thailand FOR OMCIAL USE ONLY Preface This report has been prepared by a mission that visited Thailand in the second half of September to early October 1989. The mission members included Zamir Hasan (Chief), William Easterly, Patrick Honohan, Vincent Polizatto, Choeng Chung, Arie Chupak and Correia da Silva from the World Bank; Gayle McGuigan from International Finance Corporation; Robert Effros from International Monetary Fund; and Christopher Barltrop and Harry Sasson as consultants. Dr. Amara Sriphayak and Ms. Sirinart Nang-Anong of the Bank Resident Mission, Bangkok also helped the mission. Editorial assistance was provided by Ms. Pat Bretton-Miller. The report was discussed with the government in April 1990. During the field work, the mission benefitted from discussions with senior government officials. The mission would like to express its thanks in particular to Khun Chavalit Thanachanan, Governor, and the staff of the Department of Economic Research, Bank of Thailand, for their help and advice. Thanks are also due to the senior officials of a number of banks, finance and securities companies, Securities Exchange of Thailand, TDRI, IFCT and BAAC for their cooperation. The mission alone is, however, responsible for the conclusions and recommendations contained in the report. This documnent has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. THAILAND FINANCIAL SECTOR STUDY VOLUME I Table of Contents Page No. EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . 1 I.3 IITRODUCTION . . . . . . . . . . . . * . . .1 II. ECONOMIC PERSPECTIVE AND INTEREST RATE POLICY . . . . . . . 3 A. Macroeconomic Perspective . . . . . . . . . . . . * . 3 Overview of the Real Economy . . . . . . . . . . 3 Openness of the Financial System . . . . . . S financing of Domestic Investment . . . . . . . . . . . . 6 Bank of Thailand Credit Allocation Policies . . . . . . 8 D. Interest Rate Po olic................ 10 The Main Interest Rates and Recent Trends . . . . . . . 10 Interest Rate Ceilings . . . . . . . . . . . . . . . . . 12 Determinants of Short-Term Interest Rates . . . . . . . 16 Targets and Instruments of Monetary Policy . . . . . . . 18 Mechanisms of Intervention . . . . . . . . . . . . . . . 20 Long-Term Interest Rates ... . . . . . ..... . *... . 22 C. Resiliency of the Financial Sector and Susceptibility to Economic Slowdown . . . . . . . . . . . . . . * . * . 22 III. COMMERCIAL BANKING IN THAILAND . . . . . . . . . . . . . . . 25 A. Overview of the Financial System . . . . . . . . . . . . 25 Components of the Financial System . . . . . . . . . . . 25 Development and Depth of the Financial System . . . . . 25 B. The Present Banking Scene . . . . . . . . . . . . . . . 28 Historical Perspective . . . . . . . . . . . . . . . . . 28 Ranking of Banks and Market Shares . . . . . . . . . . . 28 -it - Page No. C. Performance of Domestic Banks . . . . . . . . . . . . . 30 Peer Grouping . . . . . . . . . . . . . . . . . 30 Operational Performance .o.a.n. ce. . . . . . . . . . . 32 Liquidity . . . . . . . . . . . . . . . 34 D. Portfolio Quality..... ...... 36 Assessing Portfolio Problems . . . . . . . . . . . . . 36 Portfolio Diversification . . . . . . . . . . . . . . . 39 E. Capital Adequacy . . . . . . . . . . . . . . . . . . . 40 Nominal and Actual Capital Adequacy Ratios . . . . . . 40 increasing the Core Capital . . . . . . . . . . . . . . 41 Form of Capital .... . . . . . . . . . . . . . . . . 44 F. Bank Branching . . . . . . . . . . . . . . . . . . . 46 G. Foreign Banks . . . . . . . . . . . . . . . . . . . . . 48 H. Staffing and Training .... . . ..... . . . . . . 50 Training .... . . . . . . . . . ...... . . . . 50 Staffing . . . . . . . . . . . . . . . . . . . . . . . 50 IV. THE FINANCE AND SECURITIES COMPANIES . . . . . . . . . . . 52 A. Origin and Evolution of Finance & Securities Companies 52 Origin and Legal Framework . . . . . . . . . . . . . . 52 Evolution of F&S Company Business . . . . . . . . . . . 53 B. Operational Performance and Financial Condition . . . . 54 Operational Performance . . . . . . . . . . . . . . . . 54 Troubled F&S Companies .... . . . . . . . . . . . . 57 C. The Role of F&S Companies . . . . . . . . . . . . . . . 59 Functional Demarcation .... . ...... . . . . . 59 Distinction Between Banks and Non-Banks . . . . . . . . 60 Separation of Stock Brokerage Activities . . . . . . . 61 Foreign Exchange Business . . . . . . . . . . . . . . . 63 Branching .... . . . . . . .... . . . . . . . . 63 Funding . . . . . . . . . . . . . . . . . . . 64 D. The Issue of Ownership of F&S Companies . . . . . . . . 64 Restrictions on Ownership of F&S CoLpanies . . . . . . 64 - iii - Page No. V. CONCENTRATION AND COMPETITION IN THE FINANCIAL SYSTEM . . . 70 Introduction .... . . . . . . . . . . ...... . 70 A. Structure and Concentration . . . . . . . . . . . . . . 70 B. Conduct: Evidence of Market Power . . . . . . . . . . . 74 C. Performance and Profitability . . . . . . . . . . . . . 75 D. Concentration in the Finance Companies . . . . . . . . 79 E. Industrial-Finance Groups . . . . . . . . . . . . . . . 80 F. Government Policy on Competition . . . . . . . . . . . 80 VI. REGULATION AND SUPERVISION OF FINANCIAL INSTITUTIONS . 83 A. Supervision and Regulations . . . . . . . . . . . . . . 83 Organizational of Prudential Supervision . . . . . . . 83 The Supervisory Process .... . . . . . . . . . . . . 84 The Legal Framework .... . . . . . ....... . . 86 Prudential Controls and Limits . . . . . . . . . . . . 87 Sanctions .... . . . . . .... . . . . . . . . . . 90 B. Resolzation of Banking Insolvencies and DeposLtor Protection .... . . ..... . . . . . . 91 The Finance Company Crisis . . . . . . . . . . . . . . 92 Financial Assistance Provided to Troubled Institutions . . . . . . . . . . . . . . . . . . . 93 Financial Support for Ailing Commercial Banks . . . . . 96 The Financial Instituttons Development Fund . . . . . . 97 Cost of Support Arrangements . . . . . . . . . . . . . 98 Weaknesses in the Support Arrangements . . . . . . . . 100 Recommendations for Strengthening the Failure Resolution Process . . . . . . . . . . . . . 101 Deposit Insurance .... . . . ...... . . . . . . 102 VII. INDUSTRIAL SECTOR FINANCING AND THE INDUSTRIAL FINANCE CORPORATION OF THAILAND . . . . . . . . . . . . . . . . . . 104 A. Industrial Investment and Financing . . . . . . . . . . 104 Industrial Growth and Investment . . . . . . . . . . . 104 Financing of Industrial Investment . . . . . . . . . . 105 - iv - Paae Mo. B. The Industrial Finance Corporation of Thailand . . . . 109 Legal Framework and Ownership . . . . . . . . . .. . 109 IFCT Position in Thai Financial System . . . . . . . . 110 The Special Status of IFCT . . . . . . . . .. . . . . 112 Operations, Resources, Financial Record . . . . . . . . 113 IFCT's Pioneering Activities . . . . . . . . . . . . . lS Foreign Exchange Risk Exposure and Losses . . . . . . . 116 Key Issues, Future Role and Outlook . . . . . . . . . . 118 VIII. AGRICULTURALIRURAL CREDIT AND THE BANK FOR AGRICULTURE AND AGRICULTURAL COOPERATIVES . . . . . . . . . . . . ... 123 A. Agricultural Czedit Sector . . . . . . . . . . . . . . 123 Agriculture Sector in the Thai Economy . . . . . . . . 123 Trends in Rurel Finance .... . .. ... . . . . . . 123 Distortionary and Allocative Effects of Government Credit Policies . . . . . . . . . . . . . 127 B. The Role and Performance of BAAC . . . . . . . . . . . 132 Functions, Organization and Management . . . . . . . . 132 Operations . . . . . . . . . . . . . . .. . . . 133 Financial Performance . .*.......... . 135 Resources . . . . . . . . . . . . . . . . . . . . . . . 137 Key Issues and Outlook .. . . . . . . .*. . . . . . 138 1X. THE CAPITAL AND MONEY MARKETS . . . . . . . . . . . . . . . 142 A. The Thai Money Market ... . . . . . . . ...... . 142 Primary Market Size .... . . . . .. ....... . 142 The Secondary Market .... . . . . . . . . .... . 145 B. The Equities Market . . . . . . . . . . . . . . . . . . 146 The Securities Exchange of Thailand . . . . . . . . . . 146 Market Size and Characteristics . . . . . . . . . . . . 147 C. Long Term Debt Securities Market . . . . . 150 Primary Market Size .y.a.e.tS.................. . 150 Page No. D. Problems and Constraints ............... 152 Developing a Market in Government Bonds . . . . . . 152 Statutory Restrictions on Who May Issue Debt Obligations . . . . . . . . . . . . . . . . . . . 153 Law on Trusts . . . . . . . . . . . . . . . . . .. . . . 154 Registration and Listing Related Constraints . . . 155 Tax Related Issues . .... , * # * .. ... l55 Constraints on Insurance Companies and Othar Institutional Investors . . . . . . . . . . 156 Absence of a Rating Agency . . . . . . . 156 Availability of Financial Services . . . . . . . . . . 157 Other Suggestions . . . . . . . . . . . . . . . . . . ................... 157 E. New Financing Methode.s .............. 158 Financial Leasing .a .................. . 158 Factoring . . . . . . . . . . . . . . . . . . . . . . 160 Trust Receipt Financing .... . . ............... . 161 TABLES IN TEXT Table 3.1: Ranking of Thai Conmercial Banks, December 31, 1988 . . 29 Table 3.2: Evolution of Market Share, 1986-88 . . . . . . . . . . 30 Table 3.3s Classification of Thai Banks into Peer Groups . . . . . 31 Table 3.4: Operating Expenses of the Thai Banks . . . . . . . . . 32 Table 3.5: Evolution of Interest Differential Margins by Peer Group, 1986-1988 . . . . . . . . . . . . . . . . 34 Table 3.6: Liquidity and Growth of the Thai Comercial Banks, 1987-1988 . . . . . . . . . . . . . . . . . . 35 Table 3.7: Shortfalls in Gross Receipts Tax Payments, 1980, 1985-1988 . . . . .... . . .... . .... . .... . ... . . . 37 Table 3.8: Impact of BOT Soft Loan Support, 1988 . . . . . . . . . 38 Table 3.9: Risk Grouping of Thai Banks, 1988 . . . . . . . . . . . 41 Table 3.10: Comparative Performance Ratios . . . . . . . . . . . . 43 Table 3.11: Comparison of Performance and Branch Distribution . . . 47 Table 4.1: Performance & Strength of F&S Companies, December 31, 1988 .8................. . 55 Table 4 2: Business Focus and Profitability of F&S Companies December 31, 1988 . .................. 56 Table 4.3: Comparative Performance of F&S Companies, 1988 . . . . 57 Table 4.4: Troubled F6S Companies, December 31, 1988 . . . . . . . 58 Table 4.5t S&F Company Affiliation with Thai Banks December 31, 1988 ........ ......... . 66 - vi - Page No. Table 5.1: Commercial Bank Concentration . . . . . . . . . . . . . 71 Table 5.2s Bank Concentration, Operating Expensee and Interest Margins . . . . . . . . . . . . . . . . . . 73 Table 5.3: Thailand Bankss Performance Data by Peer Group . . . 76 Table 6.1: Amounts Outstanding Under Various Support Schemes . . . 99 Table 7.1: Industrial Investment, 1981-89 ... . . . . .. . . 104 Table 7.2s Increase in Authorized Capital of Companies . . . . . . 105 Table 7.3s Return on Equity of Listed Nonfinancial Firms . . . 106 Table '.4s Gearing of SeleaLed Itndustrial Subsectors, 1980, 1985-88 . . . . .. 107 Table 7.5: Net Inflow of Foreign Capital for Manufacturing Sector, 1980-88 .109 Table 7.6: IFCT Share in Thai Financial System Assets and Credit, 1980-88 .. . . . 111 Table 7.7t Principal Sources of Credit to Manufacturing Sector, 1981-88 ..112 Table 7.8: IFCT Loan Approvals, 1980-88 ..114 Table 7.9: IFCT Interest Rate Margins, 1980-88 . . . . . . . 115 Table 8.1: Agricultural (Farm) Credit, by Institutional Source . . 124 Table 8.2: Share of Formal and Informal Credit for Agriculture . . 125 Table 8.3: Mandated Targets and Actual Lending of Commercial Banks for Agricultural Sector, 1975-88 . . . . . 127 Table 8.4: Estimation of Subsidies in Agricultural Lending, 1983-87 .... . . . . . . . 130 Table 8.5: BOT Rediscount Facilities for Agricultural/Rural Development . . . . . ... . . . . . . . 131 Table 8.6: Disbursement of BAAC Loans, 1976-88 . . . . . . . . 133 Table 8.7: BAAC's Interest, Cost and Profit Margins, 1983-88 . 136 Table 8.8: BAAC's Sources of Funds, 1983-88 . . . . . . . . . 137 Table 9.1: Primary Money Market Instruments, 1980-88. 142 Table 9.2: Holders of Treasury Bills, 1980-88 . . . . .143 Table 9.3: Secondary Market in Treasury Bills, 1980-88 . . . . . . 145 Table 9.4: Secondary Market in Bond Repurchase Agreements, 1980-88 .. . . . . . . . . 146 Table 9.5: Market Capitalization of Set Listed Equities, 1980-88 . 147 Table 9.6: Market Capitalization by Sectors, June 30, 1989 . . 148 Table 9.7: Foreign Portfolio Investment, 1982-89 . . 148 Table 9.8: Relative Size and Valuation of Thai Stock Market, September 30, 1989 ....150 Table 9.9: Primary Debt Securities Market, 1980-88 . .1.5.1. . lSl - vii - LIST OF FIGURES Pane No. Figure 2.1: Current Account & Public Sector Deficit . . . . . . 7 Figure 2.2: Change in Foreign and Domestic Assets . . . . . . . 7 Figure 2.3: Money Market Interest Rates . . . . . . . . . . . . 12 Figure 2.4: Baht and US Dollar Money Market Rates . . . . . . . 12 Figure 2.5: Cumulative Interest Gain Relative to US Dollar Investment . . . . . . . . . . . . . . . . . . . . 12 Figure 2.6: Baht Interest Yield in US$ Terms . . . . . . . . . . 12 Figure 2.7: Real Deposit Rates ..14 Figure 2.8: Ceiling and Actual Deposit Interest Rates . . . . . 14 Figure 2.9: Ceiling and Actual Prime Lending Interest Rates . . 14 Figure 2.10s Distribution of Bank Loans by Lending Rates . . . . 15 Figure 2.11: Thai Government Repurche and Eurodollar Rates 1984-1989 .... . . .. 18 Figure 3.1: Ratio of M2 to GDP. 1960-1988 ... ..... . . . 27 Figure 5.1: Deposit and Lending Rates, by Size of Banks . . 78 Expenses and Profits, by Size of Banks . . . . . . . 78 - viii - LIST OF ANNEXES (Volume II) Annex is GDP by Industrial Origin and Expenditure, 1970-88 Annex 2: Inflation Rates, 1982-88 Annex 3: Gross Investment and Savings, 1978-88 Annex 4s Consolidated Non-F_nancial Public Sector Fiscal Account Sumary, 1981-89 Annex 5: Results of Regression Analysis Annex 6: External Financing Ratios Annex 7: Financial Aggregates, Ratios to GDP, Investment and Savings Annex 8: Bills, Loans and Overdrafts of Commercial Banks as Percentage of GDP Value Added, Classified by Sectors Annex 9: Development Credits bv Bank of Thailand, 1982-88 Annex 10: Main Interest Rates, 1983-89 Annex 11: Determination of Short-term Interest Rates Annex 12: Assets of the Financial System, 1970-88 Annes 13: Analysis of Income Statements and Balance Sheets of all Commercial Banks Annex 14: Analysis of Peer Group 1 Commercial Banks Annex 15: Analysis of Peer Group 2 Commercial Banks Annex 16: Analysis of Peer Group 3 Commercial Banks Annex 17: Analysis of Peer Group 4 Commercial Banks Annex 18: Analysis of Bangkok Bank, Ltd. Annex 19: Analysis of Farmers Bank, Ltd. Annex 20: Analysis of Siam Commercial Bank, Ltd. Annex 21: Analysis of Bank of Ayudhya Annex 22: Analysis of Thai Military Bank Annex 23: Analysis of Krung Thai Bank Annex 24: Analysis of Sayam Bank, Ltd. Annex 25: Analysis of Siam City Bank, Ltd. Annex 26: Analysis of First Bangkok Bank, Ltd. Annex 27: Analysis of Bangkok Metropolitan Bank, Ltd. Annex 28: Analysis of Bangkok Bank of Commerce Annex 29s Analysis of Bank of Asia, Ltd. Annex 30: Analysis of Union Bank of Bangkok, Ltd. Annex 31: Analysis of Thai Danu Bank, Ltd. Annex 32: Analysis of Nakornthon Bank Annex 33: Analysis of Laem Thong Bank, Ltd. Annex 34: Interest Differential Margins of Thai Commercial Banks, 1988 Annex 35: Sectoral Distribution of Credit, 1975-88 Annex 36: Estimated Direct and Indirect Taxes on Thai Banks, 1988 Annex 37: Branch Distribution of Thai Commercial Banks Annex 38: Percent Branch Distributi',n Annex 39: Summarized Balance Sheets of All Finance and Securities Companies, 1988 Annex 40: Summarized Income Statements of All Finance Securities Companies, 1988 Annex 41: Ratio Analysis of All Finance and Securities Companies, 1988 -ix- List of Annexes (Volume ii) (C"nt.) Annex 42: Analysis of Securities Companies, 1988 Assets Annex 43s Sectoral Distribution of Finance and Securities Companies' Portfolic, December 31, 1980-88 Annex 44: Financial Institutions Development Fund Falance Sheet as of September 5, 1989 Annex 45s IFCT Summarized Balance Sheets as of December 31, 1984-88 Annex 46s IFCT Summarized Statements of Income and Unappropriated Retained Earnings, Years Ending December 31, 1984-88 Annex 47s Volume of Agricultural Credit by Institutional Source, 1978-88 Annex 48: Number and Share of Agricultural Clients/Families Served by Institutional Lenders Annex 49: Percentage Share of Farmers' Loans, 1980-1986 Annex 50: Commercial Banks: Actual and Mandated Targets for Agriculture and Agri-Business, 1975-88 Annex 51i Commercial Banks: Actual and Mandated Targets as a Percentate of Total Deposits Annex 52s Calculation of Implicit Subsidies from Credit Requirement (Mandated Lending) Policy Annex 53: Calculation of Implicit Subsidies from BAMC's Average-Cost Pricing Rule Annex 54: Calculation of Subsidy-Free Interest Rates Annex 55: BAAC Summarized Balance Sheets as of December 31, 1979-88 Annex 56: BAAC Summarized Income Statements and Ratio Analysis, 1978-85 Atnex 57: Alternative Net Income Scenarios for BAM.' Annex 58: Private Capital Inflow and Outflow, 1980-88 THAILAND FINANCIAL SECTOR STUDY Executive Summary Introduction 1. This Study of the Thai financial system has been undertaken with the endorsement and active cooperation of the Government of Thailand and the Bank of Thailand (BOT). Although Thailand's financial system showed symptoms of serious distress during the early to mid-19qOs, that situation has been ame- liorated (though not fully resolved) by the country's strong economic growth of recent years. But questions about the financial system still linger. Finance and secarities companies were particularly hard hit by the earlier problems, and their role in an increasingly liberalized financial system needs to be reviewed. Because of the rapid overall growth, questions have also been raised about the efficiency of the existing financial intermediation process to raise and channel needed savings into productive investments. In this respect, the role of the capital market, particularly the private sector fixed-income securities market, and possible new financing methods and instru- ments will be crucial. The interest rate policy followed by the authorities is also very important from this point of view. In addition, basic questions have been raised about the future role in the financial system of two major sectorally-specialized financial institutions: the Bank for Agriculture and Agricultural Cooperatives (BAAC); and the Industrial Finance Corporation of Thailand (IFCT). Given these concerns, the Bank and senior government offi- Lials agreed that it was appropriate to undertake a financial sector study at this time. The same concerns defined the focus and scope of the Study. The Study is issue-oriented and makes specific operationally feasible recommenda- tions. Most of the data contained in the report is as of, or cumulative to, December 31, 1988; later data have been added as necessary. Overview of the Real Economy 2. In the aftermath of the first oil crisis in the early- to mid-1970s, Thailand employed conservative macroeconomic policies, with the maintenance of financial prudence and low inflation rates as major policy objectives. Fiscal and monetary policies were well coordinated and were used to stimulate growth, curb inflation, reduce deficits in the ralance of payments, and restrain budgetary red ink. 3. For the last three years, growth has averaged close to 10? annually. It is broadly based, with particularly strong growth in construction. The financial sector has also benefitted from the boom: value added in financial services was up a strong 33.6Z in 1987 and 16.3? in 1988. Asset prices are soaring and financial markets are strong with the stock exchange index lp 36? in 1988 and 84? in the first nine months of 1989. Such a significant gain is itself likely to become a contributing factor to the boom in private invest- ment, which has been leading the growth of GDP on the demand side. Private investment surged at a rate of 23.4? in 1988 and is likely to grow by 18? in 1989. At the same time, private consumption and public sector investment have both lagged. Public sector investment has been particularly weak, and fiscal - ii - policy has successfully played a counter-cyclical role, with a budgetary surplus recorded in 1988 and expected again for 1989. The result has been an impressive mix of rapid growth and low inflation. 4. Thailand's private savings performance has been impressive, too. However, private sector investment demand has been so high in recent years that a large external resource gap became inevitable. A notable feature of investment financing is the relatively high proportion of external financing in public investments (peaking at 87Y in 1985); the share of external finan- cing in private investment has never exceeded 202. Openness of the Financial SXstem 5. The capital account in Thailand is not formally open. Capital move- ments in and out of the country are closely regulated. Thai nationals are not permitted to buy foreign exchange except for stated purposes and foreign currency accounts are generally not allowed. Before March 1990, individual banks could not have an open foreign exchange position in excess of US$5 mil- lion or 202 of total capital, whichever was smaller. This has since been 4 relaxed to US$6 million or 252 of a bank's capital whichever is larger. Despite these restrictions, however, Thailand's financial system exhibits many characteristics of a high degree of openness. This naturally has implications for the conduct and effectiveness of monetary policy by BOT (see para. 8). BOT Credit Allocation Policies 6. Apart from exempting priority sector lending from capital adequacy requirements, BOT credit allocation policies have taken three forms. The first and major policy is to require commercial banks to lend a certain percentage (currently 20Z) of their previous year's deposits to the rural sector. If there are shortfalls, they are to be deposited at BAC at the going deposit rate, which is generally lower than the interbank rate and thus involves a cost to the banks. Second, BOT engages in development lending for specific purposes through its rediscount windows. Finally, it uses differen- tial tax rates as an instrument of policy. As to the first, there is no economic justification for directing commercial banks to lend to agriculture. A better alternative would be to expand the scope of BAAC's activities and allow it to expand commercially (see also para. 72). As for the second, deve- lopment cr dit through discount windows has never been a major factor in total credit, amtunting to less than 10? of total commercial bank lending. In October 1988, the extent of rediscounting was reduced to half of the face value of loans and, therefore, such development lending will be, proportionately, even lower henceforth. As for the third, the tax related policy, it is feared that the exemption of tax on deposits of less than B 200,000, announced in August 1989, may result in the splitting of deposit accounts simply to evade taxes, and could result in a substantial and unnecessary loss in government revenues. Similarly, exemption from the withholding tax on foreign borrowings of more than three years' maturity is also considered an unnecessary give-away. - iii - Interest Rates and Monetary Policy Instruments 7. Because of the dominant role of the banking sector, bank interest rates are the most important indicators of the cost and price of capital. The more prominent of these are, of course, the lending and deposit rates. Both of these rates have historically been subject to ceilings imposed by BOT (although the deposit rate for more than one year maturity was freed in mid- 1989 and all other deposit rates except savings deposit rate were freed in March 1990). Given the low inflation rate, real lending as well as deposit rates have remained positive throughout the 1980s. But the ceilings at the levels at which they have been pegged in recent years have proved to be constraining. There is evidence suggesting that, without ceilings, the spread over prime charged to many borrowers would have been higher. It is also probable that higher risk borrowers were pushed out of the formal sector because they could not be accommodated within the ceiling rates. For deposit rates, too, there are indications that they would have gone higher than the ceiling permitted during some periods when liquidity was tight and the repurchase rate was higher than the deposit ceiling rate. This is confirmed by the fact that the deposit rate for maturities of more than one-year has actually risen since it was freed in mid-1989. The ceilings on interest rates distort the functioning of the banking system. To improve the efficiency of the system, it is recommended that the ceilings be removed. 8. Within the ceilings, lending rates are influenced by domestic mone- tary conditions and foreign interest rates. Econometric analysis undertaken as part of this Study shows both the importance of domestic monetary condi- tions in the short run and the longer term tendency for Thai interest rates to converge with foreign rates. 9. Monetary Policy Instruments. The BOT currently imposes a number of reserve and liquidity requirements. From the monetary policy point of view, a single ratio should suffice for the control of credit and liquidity: the basic reserve requirement. It is recommended that all other reserve require- ments should be eliminated. In fact, the bond portion of the basic reserve requirement should also be dropped. The secondary reserve requirements for opening of branches, etc. are not needed and, in fact, cause distortions in the free operation of the system. A not insignificant benefit of such a change would be the freeing of bonds for secondary trading purposes. 10. The BOT has used intervention in the repurchase market as its main instrument for influencing liquidity in the system. With the increasing sophistication of the market, BOT should consider additional instruments to enhance the feasibility with which it may, at its initiative, influence interest rate movements, particularly in an upward direction. It could, for example, consider the sale of short term bills or the occasional use of aggressive bidding for deposits from the banks as a supplementary, flexible means to influence the movement of rates. The Commercial Banking Sector 11. The commercial banking sector dominates the financial system in Thailand. Its relative share has gradually but steadily increased from 51? of total financial system assets in 1970 to 59X in 1988. The financial system - iv - (comprising banks, nonbanks, insurance companies and a few sectorally special- ized financial institutions, in addition to BOT) itself has grown at a compound annual rate of 19? since 1970. As of the end of 1988, systemwide assets amounted to B 1,934 billion (US$75.5 billion equivalent). Apart from the gain of commercial banks, the other change worth noting is the gradual decline in the share of BOT assets in total systemwide assets, from 36Z in 1970 to less than 162 in 1988, indicating a healthy structural development of the financial system. 12. There are at present 15 domestic banks with 2,061 branches as well as 14 foreign banks (or more accurately branches of foreign banks). The foreign banks have steadily lost market share and now account for only about 4? of total banking assets. For analytical purposes the domestic banks have been divided into four categories or peer groups: five private banks with a market share of at least 5 each; a single government-owned bank with its own charac- teristics and a 13? market share; five banks with a market share of between 2? to 52 each; and four small banks. 13. OPerational Performance. A detailed analysis of the domestic banks' interest differential income. expenses (divided into five categories) and returns on equity and assets reveals no consistent relationship between size and performance. The medium-size banks as a group (with returns on equity and assets of 12.42 and 0.76Z, respectively, for 1988) are more profitable than others (and include the most profitable bank); the small banks are as a group the least profitable (and include the only bank with a negative return in 1988). The biggest banks had the best spreads (3.62Z) but also had high expenses being a close second to the small banks in this respect. Compared to banks in some other countries for which comparable and recent data were avail- able (Indonesia, Morocco, Pakistan, the Philippines and Turkey), Thai banks are quite efficient and operate on narrower margins. Compared to banks in the five comparator countries taken together, interest and other income margins are also lower for the Thai banks. On the other hand, capital formation for Thai banks is significantly lower. 14. Liquidity of banks deteriorated in 1988 almost across the board, with several banks having loan to deposit ratios of 100? or higher. This was the result of very high growth in loans relative to deposit growth. For many banks the rate of growth in loans also exceeded that in equity growth; these banks are ignoring the constraints that their lack of capital formation should impose. 15. Portfolio Quality. Data on individual banks which is necessary for a proper analysis of the banks' portfolio quality, were not provided by BOT because of confidentiality considerations. But a careful analysis of the banks' published annual reports indicates that several banks are continuing to suffer from portfolio problems. A rough estimate is that BOT has provided about B 21 billion in soft loans to various banks at a cost of about B 3 billion annually. The condition of six or seven banks seems to require continuing close attention by BOT. The preferred approach would be to restructure these banks and encourage mergers between small/medium-size banks to form larger institutions to compete effectively. However, close family controls pose serious constraints. The alternatives of continued BOT support, - v - outright failures, or acquisition of weak institutions by the largest banks, leading to greater concentrations, are worse. 16. Another aspect of portfolio quality - sectoral diversification - raises no particular concerns. The manufacturing sector accounts for a fourth of total loans. Loans for real estate (which includes condominiums) did increase sharply during 1987-88, but their share in the total portfolio still remained small (6X). The BOT is keeping a close eye on the banks' exposure to this sector. 17. Capital Adequacy. There is a p-assing need to expand bank capital in the face of a rapidly growing financial system, itself the result of rapid overall economic growth. Inevitably, the banks have requested a lowering of the 8X capital adequacy ratio in effect since 1962. This should certainly not be done. In fact, the application of this requirement is already too lax in that certain priority-sector (and higher-risk) loans are currently excluded from the calculatio'n base defined by BOT. If such loans were counted and contingencies were also included (as recommended by the Basle Agreement on capital adequacy), the actual ratios could fall below 5.52 in each of the last four years. It is therefore a matter of some urgency to strengthen the banks' capital if they are to continue to support the expanding economy while main- taining sound balance sheet structures. 18. The banks' capital may be increased either by new issuances or by larger retained earnings. What is necessary is to increase the attractiveness of investment in bank capital. Two different reform approaches are necessary. Under the first, the initiative has to come from the banks themselves. Existing ownership groups must be prepared to accept dilution of their holdirgs and price new issues accordingly if that is what it would take to attract new investors; they have to give priority to obtaining new capital over retaining their proportionate shareholdings. Banks' have also to realize that cross-ownerships with commercial/industrial interests create a perception of less than arms-length dealings and consequently create doubts about the accuracy of banks' reported earnings. It is up to the banks to remove such perceptions. It will take time, but banks need to be seen as professionally managed and having transparent relationships with their related entities. 19. Under the second approach, the initiative has to come from the Government. What is needed is to review the direct and indirect fiscal impositions on financial intermediation in the context of competitiveness (domestic and international) of banking system. This Study estimates that the Goverrnent's total take from bank operations amounted to 622 of their pre-tax earnings in 1988. This is high in absolute terms as well as relative to the level in many other countries. The Study makes three recommendations with the primary objective of achieving rationalization of taxation on financial inter- mediation. First, it is desirable to fold the 3.3Z gross receipts tax into the regular corporate tax on banks, even if the net tax burden is not reduced. Secondly, the Government should considert (a) removing all taxes on trans- actions between financial intermediaries; (b) removing transaction-related taxes such as stamp duties; and (c) modifying the tax treatment of loan loss provisions to conform to BOT prudential guidelines. Finally, the Study is recommending separately, for other reasons, to; (a) remove the 122 reserve requirement related to branch openings, and (b) eliminate subsidization of - vi - agricultural lending through the banking system. These recommendations are desirable irrespective of their impact on the burden of taxation on banks; to the extent they help bank profitability, they will also increase the attrac- tiveness of investment in bank capital. It is also possible that, given the competitiveness of the system, the impact of any net reduction in taxes falls principally on interest rates rather than on profits. 20. The BOT may provide some impetus to capital formation by redefining "capital fundsa to allow banks greater flexibility of the forms in which they raise capital. It is recommended that BOT favorably consider any requests from the banking community for modification of the definition of "capital funds", as long as the conditions laid down by the Basle Committee in July 1988 are met. For example, noncumulative perpetual preferred stock should be an acceptable form of capital. In addition, as part of secondary capital, authorities may consider allowing disclosed general reserves that are over and above specific and identified risks, and hidden reserves to the extent that banks are willing to disclose them and after their valuation has been verified by external auditors to the satisfaction of BOT. 21. Bank Branching. The different elements of BOT policy in respect of branching are perhaps not clearly understood by the banks because they were surprised when BOT approved substantially all branch opening requests made in 1989. For the sake of clarity, it is recommended that BOT restate its policy. The new statement should drop the requirement that the bank must buy bonds (secondary reserve requirements), and there should be no quid pro quo between rural and urban branch openings. In effect, the branch opening decision should become a business decision of the banks; BOT's role should be limited to satisfying itself that the applicant banks have met capital adequacy and liquidity requirements, are in compliance with all regulatory requirements and have presented a satisfactory business plan in support of the branch opening decision. 22. Staff and Training. The banks' past personnel policies have general- ly been geared to in-house training and lifetime employment. However, the recent surge in the growth rate of the Thai financial system and the need for weaker institutions to upgrade expertise are creating internal and market demands for staff that exceed past hiring and training plans. As a result, some of the larger banks are experiencing increasing turnover rates in manage- ment staff, while weaker institutions are having difficulty attracting good quality staff. Possible licensing of nine new foreign banks is likely to further tighten the labor market. This shortage of human capital may be as serious as the shortage of financial capital, and represents a potentially critical constraint on the ability of the Thai banking system, and particular- ly of its weaker members, to keep up with market growth. To some extent this problem is being addressed, at the clerical level, through a high level of investment in automation. But this form of institutional change only increases the already high demands made on management level staff. While the major banks have their own training centers and the capacity to expand throughput over time, this will not solve the short-term problem. The govern- ment may wish to consider relaxing work permit rules for expatriates as an effective and quick solution to the problem. In the longer term, institution- al training facilities need to be substantially strengthened and enlarged. - vii - Finance and Securities Companies 23. Finance and securities companies (F&S companies) constitute the second largest segment of the financial system, accounting for 10? of total assets. This segment is characterized by a large number of companies, with a very wide range in size of individual institutions. Some F&S companies are larger than the small commercial banks; at the other end of the spectrum, half of all F&S companies in operation at the end of 1988 accounted for only 13Z of the assets of all F&S companies. Their operational performance also varies widely. As of end-1988, 22 F&S companies were in the red, accounting for 21? of total F&S assets. Generally, there is a direct correlation between capital adequacy and profitability: the higher the equity-to assets ratio, the higher the profitability. Although F&S companies engage in many kinds of businesses, lending is dominant and provided 58? of total gross income in 1988. 24. Twenty-two F&S companies should be regarded as in difficult financial condition because their equity amounted to less than 6Z of assets, as required by BOT. If a more conservative 82 capital adequacy criteria is used, the number of troubled F&S companies would increase to 34. Ten of these are in serious condition, with negative returns on assets and equity. The continued functioning of these ten companies during 1988 added B 706 million to the losses that must be borne by creditors (or by BOT should it decide to bail them out). The BOT should seriously consider arranging mergers/acquisitions of the 34 weak F&S companies, but close family controls make this a difficult task. Alternatively, these companies should be closed. 25. The Role of F&S Companies. The F&S companies generally originated either (a) as affiliates of commercial banks, in some cases with the par- ticipation of foreign financial institutions, established to provide services that the parent bank could not legally provide directly; or (b) as formalized small institutions engaged in quasi-banking activities, generally high-risk, high-margin consumer finance. As of the end of 1988. 38 F&S companies were affiliated with commercial banks, including 12 with Krung Thai Bank as a result of its role in rescue operations. The assets of the 26 F&S companies affiliated with banks other than !Krung Thai Bank accounted for 48X of the assets of all F&S companies; they were all profitable. 26. ConsideLable debate is now occurring in Thailand on whether the legally-mandated functional segregation between commercial banks and F&S com- panies is appropriate and, more generally, what should be the role of F&S companies in the financial system. As a general principle, there should be as little institutional fragmentation as possible. This Study has concluded that legal separation of functions or activities to be undertaken by different entities may be justified only on the basis of (a) the degree of vulnerability to runs on the principal funding source; (b) the degree of risk in the busi- ness undertaken; and (c) the potential for conflict of interest in the various activities undertaken. As for the first, demand deposits allow for immediate, without notice, withdrawal and represent the highest-risk funding source in the financial system. Borrowing on the basis of commercial paper represents a different level of risk. Differentiation on this basis is appropriate- Dif- ferentiation in terms of business (as opposed to funding) risk considerations may also be justified. There is merit in the legal channeling of high-risk business activities to a separate entity in order to protect the parent from - viii - the impact of losses incurred on the high-risk dealings. Finally, it may also be justified to require that activities that together may raise a serious conflict of interest are undertaken by separate legal entities. 27. If the above principles are applied, only a relatively few activities will necessarily need to be undertaken by legally separate entities. Most financial intermediation activities should, therefore, be allowed to be under- taken by banks as well as F&S companies and, in the case of bank-affiliated F&S companies, either by the bank or the affiliated F&S company as decided by their managements. Some activities which managements had earlier decided to undertake in the name of affiliated F&S companies, because the latter could charge higher interest rates, should also be thrown open after interest rate ceilings have been abolished. There is no justification for functional specialization of institutions if it is based on distorted macroeconomic policies; the correct course of action is simply to remove the distortion. Once artificial and unnecessary restrictions have been removed, the result should be to allow the top tier of the F&S companies to graduate to under- taking a broad range of financial intermediation functions similar to banks, with only a few specific limitations. 28. The foreign exchange business represents a special case. It is cur- rently restricted to banks, and is highly profitable. It is debatable whether a finance company could achieve an adequate market penetration in a cost-ef- fective manner. On the other hand, there is nothing inherently wrong in a finance company undertaking foreign exchange dealings, and there should be no doubts about the ability of many finance companies to acquire/develop the necessary skills, given the innovation they have already introduced in the financial system. It is recommended, on balance, that healthy finance companies should be allowed to undertake foreign exchange business after they have obtained BOT's authorization. 29. Two other critical issues relating to F&S companies are branching and funding. The current restrictions on branching by F&S companies should be relaxed. It is recommended that F&S companies that are in compliance with all regulatory requirements, that receive satisfactory inspection reports from BOT and that present a satisfactory business plan should be authorized to open branches as their managements decide. On funding, the Study considered F&S companies' sources as akin to savings deposits of banks. Consequently, for reasons of parity, they should be subject to the same reserve requirements, etc. as apply to bank savings deposits. The F&S companies should continue to be barred from accepting demand deposits and from offering checking account services (unless they seek and obtain a banking license). 30. The application of the principle of potential conflict of interest dictates that the securities dealing business of a typical F&S company should be segregated. In this particular case, the lack of synergies of securities business with other F&S activities also argues for a segregation. The argument often used to justify the retention of securities business with other F&S activities, that it was not big enough to stand alone, is no longer valid after the rapid growth of the stock market in recent years. It is recommended therefore that BOT not relent on its initiative to segregate the securities business of F&S companies. As for the F&S function of underwriting new issues it is recommended that this function be retained as part of the authorized - ix - merchant banking activities of a finance company, but with some additional Lules to prevent conflicts of interest. 31. The Ownership of F&S Companies. The current law limits the share- holding of a bank in its affiliated F&S company to 10% of total share capital. It is commonly acknowledged, however, that the banks have complete control over their affiliated companies. This is effected through a maze of stock holding relationships involving relatives, shareholders of the 'parent" bank, friends, officers of fully owned/controlled institutions, and cross ownerships between group companies. The 10? restriction has been reduced to a legal fiction, and it is generally acknowledged as such. 32. Dispersal of ownership of financial sector institutions is an under- standable sociopolitical objective. Experience has proven, however, that it cannot be achieved through legislation alone. The practical impact of the 1OX ownership restriction has been to create a lack of transparency in the system. Even worse, it limits the pass-through of the F&S company's earnings to the parent bank to 102 of total earnings; however, if the affiliate becomes troubled, the parent may be implicated through its management control and thus obligated to absorb 1002 of any losses incurred. 33. The Study suggests two alternative methods to deal with the problem. Under the first alternative, working under the present legal framework, banks would be allowed to segregate all financial intermediation activities that entail above-normal risks (as defined by BOT) compared to a bank's usual prudent activities into a separate legal entity. There would be no limit on the extent of ownership. The bank's equity investments in such separate entity or entities would then be considered as no longer available to support its own financial position, and thus would be excluded from the determination of its capital adequacy. The same treatment should be given to loans by the parent to the subsidiary if they are not collateralized by assets with a determinable market value satisfactory to BOT. The same principles should be applied when the parent is an F&S company and the question arises of the F&S company's investments in or loans to its subsidiary. 34. The second alternative would place the bank and its related F&S company (or the F&S company and its subsidiary) on an equal plane through the use of a common holding company, which is essentially a nonoperating legal and tax entity that owns the bank and each related F&S company. In this way the balance sheet of each entity is kept clean and uncluttered by investments in other related financial institutions (thus improving transparency for regula- tors, investors and clients). The issue of double counting of capital is also avoided. This alternative might require changes in the present legal frame- work, including perhaps in the Thai tax code. 35. Two other matters related to this issue should be noted. As a corol- lary to the risk containment principle, investments (beyond clearly defined strict limits) in companies that are not involved in financial intermediation should be prohibited. Similarly, the limits on lending to related interests, already in place, should be reiterated. Finally, the violators should be forcefully prosecuted (there may be a need to review and revise penalties). Concentration and Competition in the Financial System 36. The share of foreign banks in the overall banking sector has declined sharply, from about 172 of total banking assets in the early 1960s, to under 6Z in 1980 and to 42 by 1988. This loss particularly benefited Bangkok Bank, which increased its market share from about one fifth of total banking assets in the early 1960s to slightly more than one third by 1980. The sha . of the top three banks grew from 451 to 59? and that of the top five banks from 56t to 702. This trend has now been halted and slightly reversed, the share of Bangkok Bank has also declined to 282. There is no doubt, however, that Thai banking is an industry with a high degree of concentration. 37. But is the high degree of concentration to be equated with unfair competitive behavior? The general perception in Thailand is in the affirma- tive. For the evidence, the Study analyzed the banks' conduct (in terms of pricing and quality of services) and performance (in terms of profitability of the industry and its leading participants). There is little firm evidence to substantiate monopolistic abuses. However, most banks in Thailand are parts of (may be even controlling entities in) large groups with widespread interests, and it is possible that group-related dealings could involve hidden excesses of power. The explanation for the perception that the largest banks hold an unfairly strong grip on the market may lie in their sheer size, par- ticularly when viewed in the context of much bigger groups of which each is the most visible component, and the social and political power these groups command and do not hesitate to exercise. 38. Even if there is no clear evidence of flagrant anticompetitive behavior in the Thai financial system, it would still be desirable to make it publicly known that the government authorities will not tolerate anti-competi- tive practices, and to establish a 'banking ombudsman' to address grievances brought forward by the public. The ombudsman may also be appointed under the auspices of the Thai Bankers' Association. Such a complaints procedure need not be costly in terms of resources and might provide a useful channel of information to the authorities in their attempts to improve the competitive- ness and efficiency of the banking system. 39. Policy with regard to competition could require both positive actions and the removal of some existing interventions. Positive actions could include a more positive attitude to licensing new banks, domestic as well as foreign. The authorities are already considering the admission of several new foreign banks. This would be a step in the right direction. But the author- ities need to go further. The restrictions on foreign banks should also be eliminated; the domestic banks do not need protection. 40. In addition, the authorities should consider removing certain res- trictions and privileges which help create niches of market power. Examples include the restriction of foreign exchange dealin.3 to commercial banks. The securities market privileges of finance companies would also need to be looked at in this context. A relaxation of the restrictions on branching, for both foreign banks and finance companies would also help. Other examples of regu- lations which tend to enhance the market power of the larger banks are res- trictions on interest payable on demand deposits and the privileges of the Krung Thai Bank with respect to state enterprise deposits. - xi - 41. An entirely different route to promote more competition among banks would be to encourage the development of the capital market, particularly the market in private debt securities. In practice, the banking sector has a monopoly on providing credit to the real sectors. There is a need to develop an active market in notes and debentures to create an alternative source of funding for the users of capital. Supervision and Retulation 42. Banking SuDervision. Over the years, the emphasis in the Thai banking regulatory process has rightly shifted from a concern with the com- pliance with laws and procedures to the determination and monitoring of per- formance and solvency of the institutions. Today, the overall framework for supervision and examination is satisfactory and the skills of individual examiners rank high among their peers from other countries. The process of supervision, including on-site examination and off-site surveillance, is gene- rally thorough and satisfactory. An improvement would be to pay more atten- tion to the institutions' internal management systems and processes to control risks and anticipate problems. Such an approach would move supervisors beyond simply determining an institution's present condition to reasonably ensuring its health in the future; it would also make them positive agents for changes in the way financial institutions operate. Organizationally, the function of supervision is divided between two departments, one each for banks and F&S companies. Some concern was expressed by the institutions that the standards of the two departments might not be uniform. So long as there are two depart- ments, that risk cannot be totally eliminated. But given the size of total staff (600 persons), creation of one huge department is not a practicable solution. The fact that both departments are under the same Assistant Governor is a reasonable assurance of consistency in the application of standards. A deliberate policy of more frequent transfer of staff between the two departments would also be helpful. 43. There are at present two exposure limits: the single borrower limit (SBL) of 25 of a bank's capital funds (302 in the case of an F&S company) and S0Z for contingent liabilities. The additional limit for contingent liabil- ities exceeds a prudent level. It is recommended to include contingent liabi- lities also in the SBL and retain it at 252. 44. Banks and nonbank financial institutions are prohibited from making loans to directors and related interests. It is commonly known in Thailand that banking groups also have investments in real sectors. It is unrealistic to believe that banks would not make loans to group-related compaties. An absolute bar on loans to directors and related interests in all probability has led to circumvention of the law rather than strict adherence to it. It is suggested that (a) loans to directors and related interests should be allowed within well-defined limits, (b) sanctions should be provided for violators of the limits, (c) full disclosure in the banks' published annual reports should be required of loans to directors and related interests, and (d) directors/ officers should be required to abstain from the decision-making process related to such transactions. - xii - 45. Regulations relating to loss provisioning and interest accrual can be improved. Provisions are required for loans classified as 'doubtful' or 'loss.' A conservative approach would require that provisions should also be made for assets classified as 'substandard.' in addition, general reserves should be established for the balance of the portfolio based on historical loan loss experience, economic trends, concentrations of credit, etc. A related issue is tax deductibility of provisions. The present rules act as a disincentive to making adequate provisions since nearly all legal remedies (which may take years) must be exhausted before. a loan loss becomes tax deductible. Tax authorities should consider a more reasonable policy. As for interest accrual, the decision to continue it should not take into account the collateral aspect. Only in exceptional cases should interest accrual be allowed to continue based on collateral considerations (where it is highly liquid and already in the process of collection). Resolution of Banking Insolvencies and Depositor Protection 46. Starting in the Iate-1970s, Thailand faced a financial crisis which, by the mid-1980s, affected nearly a third of the country's domestic banks and F&S companies. At the onset of the crisis, an appropriate arrangement for dealing with failing and failed institutions did not exist, so these institu- tions were handled on an ad hoc basis. Many were simply supported with soft loans. In many cases problems were considered to be short-term and liquidity- related; in retrospect, they were clearly solvency-related. The amount of resources needed to deal with the problems was underestimated, and the legal framework also constrained the authorities. In 1985, the Bank of Thailand Act was amended to strengthen the legal framework, and the Financial Institutions Development Fund (FIDF) was created to deal with troubled institutions more systematically. 47. An assessment of the financial crisis since its onset, and even since 1985, must conclude that while the actions taken by the Thai authorities stabilized the problems, they did not resolve them. As of the end of 1989, 13 F&S companies and several banks were still being supported through various subsidies such as soft loans. The amount outstanding under the Life Boat Scheme is estimated at B 8.6 billion. In addition, support is also being provided under various other schemes. A rough, order of magnitude type, estimate of the cost of rescue operations is B 3 billion annually. This is not the cost incurred to date, but the recurring annual cost on the basis of the amounts outstanding of support already provided. 48. Experience shows that successful resolution of solvency-related problems of financial institutions requires three critical actions: the non- performing assets are carved out of the balance sheet and removed from the failed institution, shareholders are wiped out, and management culture is changed. In retrospect, the more prudent manner in dealing with the crisis would have been either to (a) close the institution, pay off depositors, and liquidate the institution's assets; or (b) restructure the institution by removing bad atsets, replacing management, and eliminating shareholders. The need at this point may be to change the law which handicaps a receiver or another institution from obtaining a clean title to assets purchased or acquired from an insolvent institution's portfolio, after legal proceedings have started. In addition, a more formal approach to dealing with failures is - xiii - required. This will have the advantages of being (a) efficient and consistent over time; (b) somewhat shielded from political influence since the rules of the game would already have been established; and (c) prefunded through perio- dic assessments against the institutions. 49. It is also recommended that FIDF should be made a stronger, more independent and interventionist organization so that it can rlaey a leading eole in dealing with failing and failed financial institutionls. This can be accomplished by: (a) providing FIDF with its own specialized and highly skilled staff, preferably with a private-sector orientation; (b) mandating the appointment of FIDF as receiver as soon as BOT has determined a financial institution to be insolvent; (c) providing FIDF the authority to examine financial institutions, although in practice it would normally defer to the authority of BOT to conduct examinations as the primary supervisor; (d) re- quiring BOT to routinely inform FIDF of institutions rated 4 or 5 on the CAMEL rating and to share examination reports of institutions so rated; (e) re- quiring FIDF to elaborate its policies for restructuring or liquidating finan- cial institutions, using the principles discussed above as guiding factors; and (f) amending the laws which may handicap the transfer of claims after legal proceedings have been instituted. so that FIDF can act more effectively in the process of restructuring financial institutions. 50. Deposit Insurance. The question of formal vs the present informal system of deposit protection is currently being debated in Thailand. The primary concern of Thai authorities should be with improving the institutional arrangements to deal with failing and failed institutions, as discussed above. If Thailand opts for a formal deposit protection system, it should be intro- duced after the financial system reaches a high degree of stability, a level the authorities may judge it has already reached, given the continuing econo- mic growth and rapid appreciation in the value of real as well as financial assets. Formal or informal, a deposit protection system should not seek to protect every depositor; large depositors are expected to exercise due care in choosing the depository institution. Before adopting a fontal system, it would be important to develop and reach a firm, explicit consensus on the limits on the size of depositors to be protected and the extent of protection. If there were fears that despite such consensus, the limits might be swept away by "exceptions" whenever an episode of distress occurs, it would be better not to introduce a formal system. An informal system is inherently more flexible and generally better able to handle "exceptions." Also, despite the theoretical feasibility of differentiating among institutions in terms of the fee to be charged, it should be uniformly and compulsorily applied to all financial institutions; to do otherwise would raise almost unsurmountable problems and could even prove counterproductive. Industrial Finance Corporation of Thailand 51. The Industrial Finance Corporation of Thailand (IFCT) was established in 1960 with a mandate to support industrialization by providing long-term capital to the private sector. For statutory and policy reasons, IFCT's oper- ations have excluded many sectors (e.g., power and transportation), and para- statal enterprises; until recently it had also not made medium term or working capital loans, confining its operations strictly to long-term loans and limited equity financing. This may have partly resulted from the domination - xiv - of commercial banks in IFCT's sharehol4ing, which amounted to 70a of the total until 1975. Commercial banks' shareholding has now declined to 332 of the total. The Government holding amounts to 24Z (152 directly and 9Z through Krung Thai Bank). 52. The Special Status of IFCT. IFCT has never been regarded as a fully private sector, commercially-oriented, financial institution. It has always been regarded as a quasi-state enterprise and is accorded special statust it pays no corporate income tax; it is exempted from payment of business taxes and revenue stamps; its debt securities are eligible as commercial banks' secondary reserve assets; no withholding tax is payable on interest from IFCT bonds; it is exempted from the limitation on shareholdings in other companies; and it is the only company outside the F&S company sector which may underwrite securities. The Government has guaranteed IFCT's foreign borrowings from the World Bank, the Asian Development Bank, and from capital markets; its deben- tures issued domectically are also guaranteed by the Government. The Govern- ment has an agreement to cover IFCT's foreign jxchange losses, for a fee, but interpretation of the agreement during the la - decade has left IFCT with significant losses, as is explained later. 53. The special status of IFCT carries with it some substantial costs and burdens. It lacks autonomy in making policy decisions without prior approval of the Government. The Government also appoints IFCT's Chairman, and must approve its President, the chief executive officer. IFCT's charter, together with other constraints, significantly limits its ability to adapt to changed economic and market conditions, and to respond to substantially increased competition from commercial banks and F&S companies. It is required to devote significant staff resources to financing small- and medium-scale enterprises; it is prohibited from taking deposits; and it has restricted itself in the spreads it may charge. Because it is supposed to focus on "developmental" investments and has special privileges, it is, by corollary, not supposed to be a profit maximizing institution. IFCT, therefore, absorbs considerable costs from activities which do not yleld a profit. 54. More positively, however, in a financial system dominated by a few large influential groups, IFCT is one of the very few nongroup, independent, professionally managed institutions. It has successfully undertaken many pioneering activities, including the mutual fund business which has raised large amounts of domestic and foreign capital. 55. Operations and Financial Record. Despite statutory and self imposed policy restrictions on the scope of its activities, IFCT's operations have grown at a brisk pace; its total assets have increased at an average annual rate of 27Z during 1960-88, and amounted to B 24 billion (about US$940 mil- lion) at end-1988. It has always been profitable (except for foreign exchange losses since 1985) and has paid a dividend. In recent years, its margins have declined significantly as it has relied substantially more on domestic borrow- ings than in the past, to reduce exposure to foreign exchange risks. The return on equity has declined as a consequence to about 112. It is conserva- tively capitalized, with a total-debt-to-equity ratio of about 6:1. 56. Foreign Exchanae Risk Exposure and Losses. Due to IFTC's early reluctance to assume an open ended exchange risk, the Government agreed to - xv - reimburse IFCT for realized losses (and to retain gains) over and above provi- sions made by IFCT. For its part, IFCT agreed to repay the amounts advanced by the Government in reimbursement of the realized losses. The agreement was thus, strictly speaking, to provide liquidity-support to IFCT to enable it to make the repayments on foreign d*'.t installments as they became due. IFCT was to pay a substantial fee for this service. 57. IFTC had no foreign exchange losses until 1985, but after 1986 when the US dollar declined, losses increased dramatically. In 1987 and 1988, the losses amounted to B 705 million and B 885 million respectively. IFCT expected the Government to reimburse it for these losses, but this did not occur, because of a technicality. In 1972, when the IFCT Act was amended for some other reasons, the article dealing with foreign exchange losses was also amended: the phrase sany change in the par value of the bahz in relation to any of the borrowed currencies" was substituted for the phrase "any change in the valuation of the baht in relation to any of the currencies." Since the baht has had no par value since 1978, when it was floated, Thai legal opinion has held that the Government is not obliged to reimburse losses incurred by IFCT after 1978. 58. Although the dispute is still being negotiated, it is unlikely that the Government will reimburse IFCT fully. However, because of the strengthen- ing of the US dollar and other steps IFCT has taken, the potential loss on its total outstanding foreign borrowings has declined sharply, from B 6.3 billion at end-1987 to B 3.4 billion in August 1989. Therefore, even assuming no reimbursement from the Government, IFCT is likely to avoid insolvency because its profitability and operational cash flow supplemented by small sales of assets and/or new borrowings are expected to be sufficient for it to meet foreign debt installments as they become due. 59. Future Role and Outlook. The key question concerning IFCT is its longer-term futur. in an increasingly liberalized and competitive financial system. IFCT is a relatively small institution, accounting for about 1Z of the total assets and outstanding credit of the financial system; its share in the financing of the manufacturing sector is higher but still small (about 3Z). Nevertheless, IFCT has a clear cut market niche, of providing long term investment capital. With the size of its market expanding, IFCT's future from this point of view appears assured. It should try to build on this strength. 60. To make its future role r-,d orientation clear, the Government and IFCT should explicitly decide on a number of measures. As a strong signal of allowing IFCT more autonomy, the government should decide not to require its approval for the appointment as President of IFCT of the person elected by its Board. Other needed changes relate to IFCT's activities, discussed below; these will have the effect of transforming IFCT essentially into a broad based finance company. 61. The *market' for IFCT may be divided into three segments: companies affiliated with family groups and controlled by large financial institutions, independent companies not affiliated with bank groups, and state or quasi- state enterprises. The second category of clients, the independents, is the natural IFCT constituency; to ensure that it does not lose its market share in this segment, IFCT should seek to offer a broad range of services, not just - xvi - long term loans. But IFCT should not voluntarily abandon the other two busi- ness segments. The state and quasi-state enterprises constitute a sizeable segment of the Thai economy and have huge investment requirements. Most are well run and lending to them would represent safe and profitable business. IFCT should therefore seek to remove the legal and policy restriction on its lending to state enterprises in order to sub.tantially enlarge its market. IFCT should also fight for business from companies affiliated with bank groups on the basis of its comparative advantage in long term lending. 62. IFCT should also seek to broaden the sectoral coverage of its activities and to remove existing legal and policy restrictions on its invol- vement in many kinds of business activities. It should accept large time deposits from the general public as well as from corporations (but not current deposits where it would be at a competitive disadvantage with commercial banks because of their branch networks). It should also issue large denomination certificates of deposit and promissory notes. On the lending side, it should offer a broad range of services, including working capital loans. 63. IFCT should offer to give up its tax-exempt status and at the same time discontinue activities which are clearly loss making. Should the Govern- ment wish IFCT to continue an activity on social equity grounds, IFCT should offer to do so on behalf of the government only on payment of a fee (including a reasonable profit margin) and on an off balance sheet basis. 64. It is suggested that IFCT reduce its shareholding by commercial banks but maintain a government shareholding participation of up to 25Z to safeguard IFCT status as an independent non-group dominated financial institution. The government could also offer to guarantee IFCT's borrowings, in domestic as well as in international markets, for an appropriate, mutually negotiated fee. 65. Once a settlement has been reached between IFCT and Government on the foreign exchange risk agreement in respect of past foreign currency borrow- ings, the agreement should be terminated. In retrospect, the agreement may have proved more harmful than helpful (by giving IFCT a false sense of security). In any case, better techniques of managing foreign exchange risk are now available and IFCT itself has proved quite adept in using them. 66. More research and a legal opinion will be needed to ascertain whether the new course suggested by the foregoing recommendations can be effected by amendments to the existing IFCT Act. If not, it may be necessary to abrogate the IFCT Act, and reincorporate IFCT as a finance company to engage in broad- based activities. It should be possible for IFCT to grow from within to become such an institution. This would be a phased, gradual approach. Alter- natively, IFCT may consider acquiring/merging with an existing finance company to hasten the transformation. The Bank for Agriculture and Agricultural Cooperatives 67. Agricultural and Rural Credit. During the last decade, lending by commercial banks to the rural/agricultural sector has increased by about 21Z per annum. Institutional credit, including that of BAAC now accounts for about 85? of total rural/agricultural credit. The agricultural credit clientele is divided among the principal sources as follows: commercial banks - xvii - lend to better off, commercially oriented, borrowers in larger amounts (loan size averages B 106,000 per client); BAMC clients are typically small to medium sized individual farmers (average loan size is B 14.000); and the informal sector which is losing market share to institutional sources, lends still smaller amounts (about B 7,000 per client). By 1988, commercial banks provided 572 of total credit to about 102 of all rural borrowers. BAMC about 262 to half of all rural borrowers, and the informal lenders accounted for the remaining share. 68. The increase in agricultural credit since 1976 has been the result of not only the dynamic growth of agriculture itself but also of the following policies: (a) commercial banks were subjected to increasing mandatory agri- cultural lending targets; (b) interest rates in commercial bank credit to agriculture were controlled at levels equivalent to the market rates for urban borrowers, and far below informal market rates; (c) BAAC took advantage of its mandate to substantially increase its lending, and its lending rates were pegged at slightly below commercial bank rates (with subsidies provided to ensure some profits); and (d) BOT expanded its rediscount facilities directed towards agriculture. 69. Although the above policies have proved successfu. in enlarging the amount of agricultural credit available, they should now be reviewed in the context of changing circumstances. First, since primary agricultural produc- tion will continue to decline in importance in its contribution to future economic growth, credit should be made available for other rural commercial ventures like marketing and agro-industries as well as for the non-agricul- tural needs of the rural populace. Second, the costs of direct and indirect subsidies will become substantially higher if nothing is done to contain them. Particularly critical will be the opportunity cost of mandated agricultural lending in a fast industrializing economy. Furthermore, as a higher propor- tion of the agricultural population becomes commercialized, there is a danger that conflicting credit policies (e.g. government subsidized credits to BAAC's regular borrowers) will undermine lending discipline. The trade-offs will be particularly critical since Government is embarking on an overambitious small- farmer outreach program. Finally, BAAC has matured into a well- managed com- mercial entity, with comparative advantage in operating in a specific credit market niche (rural areas and medium-small scale farmers) and thus has a good chance, if not forced to lend at subsidized rates, of becoming independent of government subsidies. 70. BMAC's Operations, Performance and Condition. Since its inception in 1966, BAMC has performed well its difficult task of being an agricultural development agency and a viable financial institution at the same time. It has concentrated, as mandated, on direct lending to small farmers; in 1988 disbursements on such loans accounted for 87Z of its total disbursements. Its disbursements in the last 15 years have grown at about 16% annually. Short- term loans account for 702 of BAAC's portfolio. In recent years, BAAC's long- term loans have grown rapidly, at the expense of medium-term loans. Most short-term loans are unsecured; BAAC relies mainly on the collective guarantee of a farmers' group, a system which has enabled it to expand its outreach without jeopardizing collection efficiency. - xviii - 71. MAC has always been profitable, albeit marginally, with a return on equity averaging 7.22 during the last five years. Operating costs amounted to 3.8Z of assets in 1988. Considering the nature of its operations and the fact that it provides fully for doubtful debts and foreign exchange losses, BAAC's costs are reasonable. At end-1988, its total assets amounted to 8 40 billion, with the loan portfolio accounting for 632; and the total debt to equity ratios was 11.5. BAAC's total assets were financed by commercial bank depo- sits (36X), deposits from the public (292), foreign borrowings (18Z), BOT loans (7Z), other borrowings (22) and its equity (82). 72. Key Issues and Outlook. BAMC has a natural comparative advantage in rural areas, which it should exploit to the fullest in the future. This would require removing the restrictions under which it now operates. It sh'..ld be allowed to cater to all the financial needs of farmers, including the provi- sion of credit for non farming activities. Such an expansion in the scope of BAAC's activities would benefit the farmers (who would be able to make BAAC their one stop banker) and BAAC itself by reducing its per unit costs; it could also increase competitiveness in the financial system. Once BAAC is allowed to act essentially as a commercial bank it should be subject to the same obligations and have the same rights in relation to BOT as do commercial banks. 73. The case for allowing BAMC to charge commercial rates of interest is also strong. Most of its lending is for profitable, commercial activities; its clients do not need to be subsidized. If BAAC were allowed to charge the same interest rates as charged by commercial banks (implying a 22 increase in its present short-term rate), it would be possible for BAAC to phase out all subsidies and still be profitable. 74. The present policy of agricultural lending targets for commercial banks has served only one unintended purpose: to provide resources to BAAC at a cost to the banks. Once BMAC is allowed to compete fairly with commercial banks it would be inequitable to require the banks to subsidize BAAC. For this reason and because such mandated lending hinders rational resource allocation, agricultural lending targets for the commercial banks should be phased out. 75. In its future operational planning, BMAC should be realistic. Unfor- tunately, its current five-year plan sets an unrealistic target of reaching 145,000 new clients annually for several years, compared to less than 100,000 reached annually in the last five years. Trying to achieve such a target would seriously affect the quality of its portfolio; furthermore, the target volumes of lending imply a resource gap of B 22.3 billion during 1989-94. Instead of emphasizing quantity and setting unrealistically high lending targets, BAAC should improve the quality of its service. It should continue to expand its outreach, but within prudent limits so that its institutional strengths are not compromised. The Capital and Money Markets 76. In relation to the overall financial system, only the stock market component of the capital market has reached a respectable size; the money market at the short end and the debt securities market (particularly the - xix - market in private sector debt instruments) at the long end are both very small. In the primary money and debt markets, the total amounts issued in 1988 were respectively, about B 20 billion (including B 16 billion in commer- cial paper and B 2 billion in treasury bills) and B 25 billion (including B 20 billion in government bonds and B 2 billion in state enterprise bonds). By contrast, outstanding credit extended by banks and F&S companies as of December 31, 1988 amounted to B 1,116 billion. 77. In recent years, treasury bills have been held principally by BOT and the Exchange Equalization Fund, one of the entities empowered to hold foreign reserves. There is no active secondary market. The principal holder of government bonds are the commercial banks, to comply with primary and secon- dary reserve requirements. There is also no active secondary market in government bonds; they are normally held by the commercial banks to maturity. The lack of active secondary markets in both short- and long-term government papers is because their supply is so small. In 1988, only B 2 billion of treasury bills and B 20 billion of government bonds were issued, all taken up by banks to meet reserve requirements. Given the government's surplus budget- ary position, the primary issues of treasury bills and bonds ire not expected to increase in coming years. 78. The equities component of the capital market has grown rapidly over the last four years. From the end of 1985 to mid-1989, the market capitaliza- tion of all stocks listed on the Securities Exchange of Thailand (SET) has increased by more than seven times to B 367 billion. Still, the Thai market ranks behind those in Taiwan, Korea, Inxia and Malaysia (not to mention Japan and Singapore) among seven Asian markets' and ranked fourth if measured as a percentage of respective GDP. Foreign holdings amounted to 122 of all listed shares and 23Z of total capitalization as of the end of 1988. 79. Constraints and Problems. An active market in government bonds normally precedes the development of a market in corporate bonds because individual corporate issues are too small to develop an active secondary market large enough to support the necessary institutional infrastructure. Two factors are responsible for the lack of an active secondary market in government bondss short supply relative to demand for meeting reserve require- ments (so that most bonds are bought and held to maturity by commercial banks), and lower than market coupon rates. The recommendation made in para.9 to eliminate the secondary reserve requirements and the bond portion of the primary requirement is reiterated. Furthermore the interest rate should be allowed to rise to bring about a positive yield curve. Once these recommenda- tions have been implemented, an adequate supply of .onds would be freed for trading and bonds would also become attractive for iiivestment purposes. A well functioning government bond market would also be helpful to the authori- ties to sterilize excess liquidity if, in the medium to long term, Thailand becomes a balance of payment surplus country, and avoid the kind of problem faced by Korea. After the government bond market has developed, the stage would be set to make a serious attempt to develop a market in private sector fixed-income securities. 80. That attempt should start with a basic change in the law regulating who may issue debentures. The Law on Limited Public Corporations restricts the privilege of issuing debentures to public companies. The Thai Commercial - xx - Code on the other hand prohibits private companies issuing debentures. The objective of both laws -was presumably to encourage companies 'to go public," and it was thought that the power of public companies to issue debentures would provide sufficient incentive. This has not been effective, and, in any case, there is no reason in principle to restrict the issuance of debentures to public corporations. Incentives for companies to go public should take other forms than restrictions on issuing debt instruments, which have had the unintended effect of perpetuating the strong hold of banks on credit business and of denying most firms legitimate and perhaps less expensive ways of raising capital. It is recommended that the Law on Limited Public Corporations and the Commercial Code be changed to allow all companies (whether public or private) to issue debentures. 81. Related to issuance of debt instruments is the subject of trusts. It appears that Thai law makes trusts invalid. But since debt instruments are commonly issued under the provisions of a trust agreement between the issuing corporation and a trustee representing the holders of the obligation, doubts on the validity of trusts would have a negative impact on debt issues. While there are legal techniques which make it possible to issue debt securities even in the absence of corporate trustees, these would probably be less prac- ticable in the Thai context. It is recommended, therefore, that the Thai authorities consider amending the law to make the validity of trust agreements clear beyond doubt. 82. In many countries an incentive for companies to go public and be listed on the stock exchange is the provision of a lower tax rate for the listed companies. In Thailand, too, the tax rate for listed companies is 30X against 352 for non-listed companies. In reality, many small owner-managed companies do not pay taxes at the higher 352 rate; for them, a tax rate of 30X on the presumably better audited results they would have to show as listed companies may be a disincentive. It is suggested that the Government consider imposing an even lower tax rate, say 25?, for newly listed companies for a limited period after listing, say five years. Such an inducement may pr se to be effective but without adversely affecting total tax receipts. 83. The authorities might also consider a number of other measures to encourage the development of the capital market: (a) the listing process should be modified to shorten the time involved and to centralize decision making functions performed by the Ministries of Finance and Commerce and SET; (b) SET's registration requirements should be revised to take into account the development of new instruments and increased market ac ivity; (c) the yield of the 3.3? business tax and 1? stamp duty should be examined in relation to the possible stimulating effects of eliminating them or at least merging them into the overall corporate tax structure; (d) limitations formulated on insurance companies and other institutional investors should be reviewed to remove un- necessary constraints (e.g., the limit of 2? on holdings in each security, making investment at par only, etc.); (e) a rating agency should be set up, preferably as a joint venture with, or at least initially with technical assistance from, a reputable international firm; (f) trading in rights and warrants should be allowed; (g) more companies should be allowed to enter the mutual fund business; (h) the number of authorized brokers (32 at present) should be substantially increased; (i) trading methods on the floor of the exchange are in need of modernization; (j) stock splits should be allowed; - xxi - (k) preferred shareholders should not be given the same voting rights as com- mon shareholders: (1) the over-the-counter market should be brought under the supervision of SET, perhaps as a Third Board or Third Tier market; and (m) principles and detailed safeguards should be developed to govern take over and acquisition activities so that these may evolve in an orderly fashion. 84. New Financing Methods. Thailand has reached the stage in its econo- mic development where the needs of business would be better served by more varied and sophisticated means of financing. The Study recommends that an adequate legal framework for financial leasing, factoring accounts recei- vables, and trust receipt financing should be developed. Concluding Remarks: An Agenda 85. Preparing a Plan of Action. This Study has addressed a number of specific topics--the overall macroeconomic policy as it affects the financial sector, commercial banking, finance and securities companies, two large sector-specific financial institutions, and issues related to money and capItal markets. Although the recommendations made in this Study are wide ranging and relate to diverse subjects, there are some common threads. The underlying objectives of the recommendations are to create more competitive financial markets, which are integrated rather than fragmented and func- tionally specialized by law, offer more freedom of choice to providers as well as users of capital not only in terms of institutions but also in terms of instruments, and allow as much flexibility and freedom of action (including the ability to compete) to the institutions comprising the system as is prudent from the point of view of the system's overall stability. Collective- ly, these recommendations, when implemented, should result in the provision of better, more diversified financial intermediation services, offered at lower costs, thus furthering the overall economie development of the country. The financial system is also expected to become more resilient to possible future shocks as a result of improved supervision and regulation, strengthening of banks' capital, greater transparency, and strengthened arrangements to deal with troubled financial institutions. 86. The detailed recommendations made in this Study cannot be implemented all at once, and it is suggested that the authorities assign priority to the following four areas. First, it is an anachronism that interest rates, de- posit as well as lending, are still controlled in an environment as competi- tive and open as Thailand's. Interest rate ceilings have led to a number of distortions; they should be removed as soon as practicable. Second, in the banking sector, there is an urgent need to augment banks' capital, bring about transparency in ownership of subsidiaries and in transactions, and reduce obstacles to entry. Third, the arrangements to deal with distress in the system and with troubled financial institutions need considerable strengthening. The treatment in the past was ad hoc and the cost has been high without final resolution of the problems. Finally, there is a need to develop a well functioning secondary market in government bonds, for the fillip this would give to the private securities market, as well as the additional instrument it would provide for monetary control. Some of the recommendations can be implemented through administrative actions; others would require changes in the laws. THAILAND FINANCIAL SECTOR STUDY I. INTRODUCTION 1.1 This Study of the Thai financial system has been undertaken with the endorsement and active cooperation of the Government of Thailand and par- ticularly of the central bank, the Bank of Thailand (BOT). Thailand's finan- cial system showed symptoms of serious distress during early to mid-1980s which have been ameliorated (though not fully resolved yet) by the strong economic growth of recent years. The finance and securities company segment was particularly hard hit, and their role in an increasingly liberalized financial system is being debated. Because of the rapid overall growth, ques- tions have also been raised about the efficiency of the financial intermedia- tion process to raise and channel the needed savings into productive invest- ments. In this respect, the role of the capital market, particularly private sector fixed income securities market and possible new financing methods and instruments, becomes crucial. The interest rate policy followed by the authorities is also very important from this point of view. In addition, basic questions of the future role in the financial system have also been raised about two major sectorally specialized financial institutions: the Bank for Agriculture and Agricultural Cooperatives (BAAC) and the Industrial Finance Corporation of Thailand (IFCT). Given these concerns, the Bank and senior government officials agreed that it was appropriate to undertake a financial sector study at this time. 1.2 The report is organized as follows. The chapter following this introduction provides the macroeconomic perspective for the financial system. It is divided into two sections. The first section briefly discusses policies towards the financial sector in Thailand from a macroeconomic perspective. Given the importance of interest rates in the development of financial sectors, the second section contains a detailed discussion of the interest rate policy and how BOT may improve its toolkit for the conduct of monetary policy. It is followed by two chapters dealing with the two main segments of the banking system: commercial banks and securities and finance companies. The focus is on the major problems or issues and on how improvements may be made; clear cut recommendations have been made. The next two chapters deal with two specific aspects which have generated considerable discussion: com- petition/concentration in the financial system, and regulation and supervision of banks and securities and finance companies (particularly focusing on how to deal with failing and failed financial institutions). The specific problems of industrial and agricultural credit and more specifically of BMAC and IFCT are the subject of the following two chapters. These are followed by the chapter on capital market which analyzes the present state of the money, equi- ties and fixed income securities markets. The emphasis of the chapter is on identifying constraints and problems; it also proposes how three specific methods of financing (financial leasing, factoring accounts receivables and trust receipt financing) may be made more feasible in Thailand. An erecutive summary is provided up front. 1.3 Most of the statistical data contained in this report and Volume II (Statistical Annexes) was provided by BOT. However, BOT was understandably constrained from providing data on individual banks and finance and securities companies because of the confidentiality consideration. Data on these institutions conform to their published annual reports. Some of these annual reports raised doubts about accuracy of data and full disclosure; analysis of such data is therefore qualified to that extent. Data was also provided by IFCT and BAAC. Most of the data are as of, or cumulative to, December 31, 1988; later data have been added at places where considered necessary. -3- II. ECONOMIC PERSPECTIVE AND INTEREST RATE POLICY A. Macroeconomic PersDective Overview of the Real Economy 2.1 Thailand, like all oil-importing countries, suffered from various external shocks in the decades following the first oil crisis in 1973. How- ever, Thailand has gone through those periods by employing conservative macro- economic policies. Maintenance of financial prudence and keeping low domestic inflation rates have been major policy objectives. There is coordination between monetary and fiscal policies, which have been used to stimulate growth, curb inflation, and reduce deficits in the balance of payments. The trading syscem in Thailand is relatively open, as is the capital account despite the existence of formal controls on capital movements. Nevertheless domestic liquidity conditions do have a short-run impact on interest rates so that there is a continuing need for monetary policy action to be consistent with domestic and external stability. The government has also followed a responsible fiscal policy so that budgetary deficits were restrained and in fact the budget has been in surplus for 1988 and 1989. Thus, Thailand has avoided the prolonged crises that many other developing countries have suffered. 2.2 Thailand has had one of the most rapidly growing economies in the world for the last three years. After growth of 8.42 in 1987 and 11? in 1988 (Annex 1), growth of about 112 is estimated for 1989. The financial sector has been in the forefront of this growth, as value added in financial services shot up by 33.6Z in 1987 and 16.32 in 1988. This is partly due to the rapid growth of sectors that use financial services intensively, such as construc- tion--13.72 growth in 1988--and real estate transactions--26.8% growth in 1988. This reflects the real estate boom that the economy has experienced, especially in construction of office and residential condominiums. Although no hard figures are available, informal reports speak of a doubling of real estate values in prime areas in 1989. The surging Thai stock market also indicates that the valuation of physical capital has risen sharply. The Securities Exchange of Thailand (SET) index rose by 84? in the first nine months of 1989, after rising by 36S in 1988. The implied capital gain on the stock market alone for the first nine months of 1989 amounts to about 15Z of GDP. Such a significant gain seems likely to be contributing to the boom in private investment, increasing both the expected rate of return on investment and the liquidity of corporations and private households. 2.3 The private investment boom has been leading GDP growth on the demand side. Private investment growth of 18.5? is estimated for 1989, after previous year's surge of 23.4?. Private consumption has been lagging behind the overall growth of the economy, while both government investment and consumption have been growing more slowly or even falling, as explainti in mcre detail below. 2.4 A noteworthy feature of the economic expansion is that it has taken place with little acceleration of inflation. Even the signs of acceleration - 4 . in the second half of 1989 seem to be reflecting mainly a relative price shift. The average inflation for 1989 is expected to be 5.5?, up from 3.8S in 1988 (Annex 2). Items that are rapidly rising in price are either those whose international price has risen rapidly--rice and other food commodities--or those that are nontraded and in booming sectors--construction materials. As standard analysis would predict, the demand expansion is putting some upward pressure on the ratio of nontraded to traded goods' prices. 2.5 The main force of the expansion has been felt in the widening external deficit, as shown in Annex 3. Although private saving has risen, it has lagged behind the extraordinary rise of private investment, so that a current account deficit (saving-investment) of 3S of GDP appeared in 1988, with a like figure expected for 1989. The public resource balance has played the desired countercyclical role, with a resource surplus achieved in 1988 for the first time since 1974. 2.6 Although external deficits were significant in 1988-89, this was following a successful external adjustment where previously much higher external deficits were eliminated. A rough calculation suggests that a cur- rent account deficit of no more than 2.3Z of GDP would be required in 1990 to keep the ratio of net external liabilities to GDP stable, assuming that growth remains at 102 in 1990. In the medium term, current account deficit of 1.62 of GDP would be required to keep the debt ratio stable, assuming the long run growth rate of 7X. As Annex 3 shows, the reduction in the public resource deficit has been the most important force in restoring external balance to the economy. The current public surplus is desirable to restrain private demand and to keep current account deficit within bounds. 2.7 Annex 4 shows the components underlying the behavior of public deficits. After several years of high deficits in 1981-85, a major fiscal adjustment was achieved in 1986-89. Although there was some decrease in capital expenditure, the fiscal adjustment was mainly achieved through an increase in revenue. The relative decline of public investment was partly unintentional, reflecting project delays as shortages of construction materials and rapidly rising materials prices wreaked havoc on budget execu- tion procedures. This problem is now being addressed by introducing an adjustment for construction costs into project execution procedures. 2.8 The revenue increase was also partially endogenous to the expansion, reflecting the rapid rise in the taxes most sensitive to economic activity-- tariffs on imports (242 increase in 1988-89), corporate income tax receipts (32Z increase), and business tax receipts (35? increase). This suggests that the achievement of the fiscal surplus not merely be seen as the completion of the adjustment undertaken since the early 1980s, but also as reflecting bene- ficial feedback from the economic expansion itself. 2.9 Although the fiscal adjustment has been very impressive, it has not been sufficient to keep up with the rapid rise in private investment, so that the external resource gap has been large in 1988-89. It is expected that when the fixed investment comes on stream, it will result in a higher value of exports, thereby helping to reduce the external resource gap. It remains unclear at this point, however, whether the gap will be fully bridged. This leads to the issue of the adequacy of financing for domestic investment over the long tenm. This is discussed below after bringing out the extent of open- ness of the Thai financial system. Openness of the Financial System 2.10 In analyzing the financing of public and private investment by the financial system, it is important to analyze the degree of openness of the Thai economy to external capital flows. Theory predicts that an economy perfectly open to capital flows and integrated with the world capital market would display several important characteristics. The domestic interest rates would not be affected by private sector excess demand or fiscal deficits but would be determined solely by international interest rates (adjusted for expected changes in currency values). Changes in the fiscal deficit or auto- nomous private demand would pass through into the current account deficit of the balance of payments rather than increasing domestic interest rates. 2.11 Public borrowing in the domestic financial market would not crowd out private investment; instead, there would be offsetting private borrowing in external markets. Similarly, credit expansion by the central bank would be entirely offset by a loss of foreign exchange reserves, other things being equal. Credit expansion by the entire financial system would be offset by a decline in net foreign assets to the extent that it exceeded the expansion in money demand. In such an economy, the level and composition of the financing of the public deficit would have no effect on private investment.l/ 2.12 With perfect capital mobility, fiscal policy would be effective in restraining aggregate demand to improve the current account.21 Monetary policy would be ineffective either to affect the current account deficit or private investment, but would have a strong effect on the level of foreign exchange reserves. 2.13 The Thai economy is relatively open to trade flows, as shown by the large increase in the share of exports in GDP over the last two decades. An analysis of trade policy across countries in the World Development Report 1987 classified Thailand as *outward-oriented." The capital account in Thailand, though, is not formally open. Movements of foreign exchange and Thai currency out of Thailand are regulated closely by BOT. Residents are not permitted to buy foreign exchange except for documented purposes such as imports, travel, etc. Foreign currency accounts are not permitted in the banking system except 1/ One caveat to this result must be mentioned. If the private sector anticipates that excessive public sector deficits will lead to a balance of payments crisis, devaluation, and/or the rationing of external credits sometime in the future, then deficits will affect expectations and thus private investment. 21 There is the theoretical possibility that private saving could offset changes in fiscal policy ("Ricardian equivalence"), but most empirical tests have shown that this does not hold in either industrial or deve- loping countries. - 6 - for narrowly circumscribed purposes. Commercial banks can move capital both in and out, but individual banks cannot have a net foreign exchange position of more than US$ million or 20Z of its capital, whichever is smaller.3/ While banks can use the forward market to cover their positions, the market is limited to commercial banks and importers/exporters, and then, too, esseatial- ly for short terms. 2.14 Though this degree of capital controls makes the Thai capital market less than perfectly open, the economy nevertheless exhibits many of the characteristics of a high degree of openness. Figure 2.1 shows the behavior of the public and current account deficits. In many periods, the two deficits have not moved in tandem, indicating that there are many other short run influences on the current account deficits (e.g., terms of trade movements, changes in ta:iffs, private sector demand shocks. etc.). However, as a long term trend, there is a significant correlation, indicating a high degree of pass-through of fiscal into external deficits.4/ Given the high degree of de- facto openness of the capital account, the authorities may wish in the future to consider a gradual dismantling of the formal restrictions mentioned in the previous paragraph. This would be consistent with the increasing linkage of Bangkok with international money and capital markets. 2.15 Although these results suggest that the current account deficit is probably affected by fiscal deficits, it does not mean that a fiscal deficit is the only cause of external deficits. Excess private demand can also lead to current account deficits, as is the case for 1988-89. 2.16 Figure 2.2 shows the behavior of the changes in BOT domestic credit and net foreign assets over 1971-89. Again a negative correlation is displayed, which is suggestive of a high degree of offset of reserve changes to domestic credit expansion. This would suggest that monetary policy is relatively ineffective in influencing output or money demand. Furthermore, at least since 1986, the cost of forward cover has equalled the difference between foreign and domestic interest rates, providing some evidence of capital market integration. However, it is not clear that the gap between foreign and domestic rates has always mirrored exchange rate expectations. There have been periods when baht interest rates were higher than could be explained by exchange rate expectations, notably in 1985. In the medium term, however, the forces of capital mobility push domestic interest rates towards foreign levels. This has been true even for the administratively controlled bank deposit rates as well as for money market rates. 3/ This was changed to US$6 million or 25Z of a bank's capital, whichever is larger, in March 1990. 4/ A simple regression of the current account deficit on the fiscal deficit shows an R2 of .21 and a coefficient of .50 on the fiscal deficit (significant at the 5Z level). While more analysis is needed to resolve potential problems of simultaneity and omitted variables, this result is suggestive of the high degree of openness of the economy. The regression results are given in Annex 5. - 7 - Figure 2.1: Current Account & Public Sector Deficit 2 2 - _ a a . I I I I I I a a I 7{2 73 74 75 76 77 78 7Q 80 81 82 83 84 85 86 87 88 aear Figure 2.2: Change in Foreign and Domestic Assets so 40 -~~~~a 20 16 0 a * 4 -20 _ .4 - Net Forelgn Assets\ 40 ~~Net Domesfic Assets\ -60 I, , , , I I, I, I... 71 72 73 74 75 76 77 78 79 80 81 82 8 84 8 86 87 88 89 Years _446Wfta a a 44 a~~~~~~~~~~{7 l 4 - 8- Financing of Domestic Investment 2.17 With a high degree of openness, the relevant concept for analyzing the role of external financing is the ratio of total external borrowing to total investment. The ratio was fairly modest--less than lOX--over the 1970s, which helps to explain how Thailand avoided the debt crisis by which many other middle-income countries were beset. Annex 6 shows that during the first half of the 19809, the ratio increased significantly, but then decreased again in 1986-88 as the fiscal adjustment improved the external situation. Another factor is the large share of direct foreign equity investment in total external financing in 1988-89. The external debt ratio peaked in 1985 at 34.2? of GDP, then declined to 26.52 by 1988. A long-run tendency toward excessive dependence of investment on external financing is not apparent in these numbers. 2.18 Annex 6 also shows the financing breakdown of public and private investment. As discussed earlier, this does not have much macroeconomic significance in a highly open economy, but it is still of interest on institu- tional grounds. Public investment consistently has a higher ratio of external financing than private investment, going as high as 87? in 1985. The role of external financing in private investment has never exceeded 20?. 2.19 The achievement of a surplus by the government allowed a reduction in its external financing needs. Since there continued to be a small positive amount of foreign borrowing, however, the surplus led to a sharp reduction in domestic government debt. The domestic debt reduced most sharply was that owed to the banking system, as was shown in Annex 4. 2.20 The sharp reduction in public borrowing from the banking system allowed a rapid expansion of private credit, even though total domestic credit stayed roughly constant as a ratio to GDP (Annex 7). The ratio of public domestic credit to GDP peaked at 17Z in 1984, then declined to 9.5? in 1988. Private credit increased to 57.3? of GDP in 1988 from 48? in 1984. Over 60? of private investment was financed by domestic credit during 1987-88. Since total credit stayed roughly constant while there was an increase in demand for financial assets, as reflected in a high ratio of H2 to GDP, there was an increase in net foreign assets of the financial system. Bank of Thailand Credit Allocation Policies 2.21 A concern that the agricultural sector might be discriminated against by the financial system led BOT to initiate policies to direct credit towards this sector. As shown in Annex 8, the ratio of bank credit received to sectoral value added in 1988 was lower in agriculture than in other sectors except utilities. The BOT currently requires commercial banks to lend 20? of their previous year's deposits to the agricultural sector. Of this a maximu-i of 6? of deposits can be lent to agribusiness concerns. Any shortfall of the lending from the target must be deposited by banks at BAAC at below-market interest rates. As discussed in detail in Chapter VIII, the target for lending has been revised several times and the commercial banks have not reached the target although the subtarget for agribusiness lending has been met or exceeded every year. - 9 - 2.22 The agricultural lending requirement does not appear to have a strong economic justification. The lower level of credit utilization by the agricul- tural sector probably reflects technical and institutional characteristics of production in that sector, and not necessarily a market failure on the part of lenders. To the extent that it is enforced, the requirement to lend to agriculture may result In credit being directed away from uses where it has the highest rate of return. There are also other regulations that affect the agricultural sector. In particular, the ceiling on lending rates may itself be causing discrimination against the agricultural sector. A move toward greater efficiency of credit allocation across sectors would require the elimination of both the agricultural lending requirement and the ceiling on loan rates. These agriculture related issues are discussed in detail in Chapter VIII. 2.23 The other major effort of BOT to influence credit allocation takes the form of refinancing of certain kinds of bank lending at subsidized rates, such as promotion of exports, small-scale industry, and agricultural produc- tion. The BOT will rediscount bills for these purposes presented to it by commercial banks. The rate of interest on these rediscounts varies between 32 and 52 depending on the purpose of the loan. Annex 9 shows the evolution of sluch development credits since 1982 in relation to total commercial bank credit by sector. Only in the export sector have these been significant relative to total sectoral credit. Total development credits have not been more than 92 of total com ercial bank lending since 1982; they would be even lower as a share in total system wide lending. 2.24 Since the last quarter of 1988, there has been a sharp proportionate decline in development credits. This is the result of the policy announced by 3OT in October 1988 that it would henceforth only rediscount 50 of the face value of loans to preferential borrowers, instead of 802 as previously. The impact of this measure was probably felt most strongly in the export sector, where the share of development credits in the total was estimated to have declined from 49S to 20Z. For the agriculture sector, however, despite the reduction in the percentage of refinancing, the amount of loans increased in 1989. 2.25 The reduction in directed" development credits is desirable, since it is not necessary for activities that have a high enough rate of return to be economically viable and not normally desirable for nonviable activities. Development loans have a fiscal cost that is financed implicitly through taxes on other users of financial services. Although these distortions do not appear to have been large. their reduction is a wise policy move during a time of good performance in the sectors being favored. 2.26 The 3OT has a more informal policy instrument in the *moral suasion" that it exercises to restrain excessive credit increases to particular sec- tors. This has been used recently to try to dissuade banks from excessive lending for real estate speculation. It is difficult to evaluate the success of this policy since it cannot be estimated vhat bank lending would have been in the abseni of "moral suasion*. The rapid growth despite official dis- couragement of credit to real estate business (482 in 1988) might suggest that - 10 - the effectiveness of the policy has been limited. However, the ratio of credit to value added in this sector in 1988 was only moderately higher than in 1984 after an intervening drop (see Annex 12), so it is possible BOT did play a role. 2.27 There are two other policy instruments that should be mentioned: the withholding tax on interest paid to non-residents, and the tax on interest earned by depositors. Although they do not directly influence sectoral credit allocation, they affect portfolio decisions by the financial system and the private sector and thus credit behavior in general. 2.28 The tax of 152 on interest income is part of the overall income tax, and thus serves the purpose of preventing large disparities in tax rates between different types of income. The government announced in August 1989 the exemption of the tax on depositor accounts of less than B 200,000 as a measure to promote saving of 'small savers". While the objective of encou- raging smail savers to save is understandable, it seems likely that all deposit interest income earners will now be able to evade taxation through splitting of accounts, a potential loss to the treasury of as much as B 7 bil- lion.5/ While the public sector is currently in surplus, it is not desirable to provide a tax evasion measure at a time when passive public sector sur- pluses serve to restrain aggregate demand. If all interest income becomes de facto exempt from taxes, it would worsen the regressive impact of the tax system. In addition, attempts to enforce the measure will entail needless manpower costs for banks and the government. Even if the measure did not lead to tax evasion, the creation of unequal rates for different types of income is a violation of sound and efficient tax policy. 2.29 The 102 withholding tax on foreign borrowing is a measure to reduce the disparity between the tax rates on interest income received by foreigners and by nationals. Given the de facto high degree of capital mobility, changes in the tax are one of the few means available for the authorities to affect the domestic interest rate. However, an exemption was granted for foreign borrowing of longer than three years maturity during May-August, 1989 (later extended). The avowed purpose--to promote long-term finance for the current account--again is more worthy than the means chosen. The temporary change in policy led to revenue loss with probably little or no permanent change in private behavior, since the private sector may have rescheduled previously planned borrowings to fall within the period of exemption. Private capital inflows grew spectacularly during 1989, but it seems likely they would have done as well in the absence of an exemption. The loss in revenue again was inadvisable on stabilization grounds, since the overheating of the economy called for the only effective countercyclical instrument--passive fiscal surpluses. Also, if the private sector does interpret the exemption as a more lasting measure, it might worsen the overheating of the economy by lowering the effective domestic interest rate and increasing investment further. S/ The estimated revenue from income tax on deposit income in FY1988-89. - 11 - B. Interest Rate Policy The Main Interest Rates and Recent Trends 2.30 What Interest Rates are Important. Despite the recent surge in the importance of the equity markets in Thailand, the banking system still dominates financial intermediation. Neither quoted equity nor marketable debt securities approaches the value of the assets of the banking system. Flow of funds analysis also shows that most domestic loans borrowed by the private sector are intermediated through the banking system. The banking system is funded in large part by deposits from the public and most of its assets are held in the form of private credit. This domination by banking of the structure of flows define the main interest rates to be analyzed. They are the interest rates on bank deposits and loans, on government bonds, on the interbank market and in the repurchase market. These interest rates are determined, at least in part, through market forces. 2.31 Trends. Interest rates have moved quite considerably over the years in response to market pressures. Following a period of high nominal and real interest rates during the late 1970s and early 1980s, monetary stability was restored during the first half of 1986 with money market rates falling as low as 5.5Z. Since late 1987 there has been a general upward trend, with a surge of rates during late 1988 and early 1989. Annex 10 shows the main interest rates in effect in Thailand in recent years; some of the rates are also plotted in Figures 2.3-2.6. Figure 2.3 plots the inter-bank and repo rates, while Figure 2.4 displays the movement of baht and US dollar money market rates over twelve years. Though there is evidence of correlation in Figure 2.4, it is clear that, up to 1986. baht rates have normally been higher than US dollar rates. Monthly US dollar returns (annualized) on baht money market investments are shown in Figure 2.5. The stability of the exchange rate before 1984 is evident, as is the importance of exchange rate fluctua- tions thereafter. As shown in Figure 2.6, despite the depreciations of 1981 and 1984, baht investments showed a cumulative gain to mid-1989. 2.32 Looking more closely at the past five years, the two interest rate peaks in 1985 and 1986 are clearly evident in Figure 2.3, which also shows how closely interbank and repurchase market rates move together. The two peaks were not driven by international interest rate movements, nor can they be explained by subsequent exchange rate movements. They correspond to a period of tight monetary and credit policy. Interest Rate Ceilings 2.33 Both lending and deposit rates of the commercial banks and the finance companies have been subject to ceilings, though the ceiling on rates for deposits of more than 12 months was removed in mid-1989. (Some more deposit rates were freed in March 1990.) Considering the low inflation rate in Thailand--well under 52 p.a. until the second half of 1989--the interest ceilings have always permitted positive real rates throughout the 1980s both on th-te deposit and on the lending side (see Figure 2.7). -12- 20 - g ~~~~~~Figure 2A3: ataWU olrMony Mlr Ro i~~~~~~~~~~~~hnn U$d ~~~~~5 g~~~~1 W O- , .i .S.w>.;S. o.s "85129 .~~~~1 g Fu" : 3blWM_ S3]o .*-Ji ; <.9 > 5- ;;/ - v Thd U86 5 ~ ~ ~ -. >gaA _................................................................... Svg T 8 " S " " g " 73 W~~~~~~~~~7 so el 02 e -e s e g -° J0 g ~~~20- | S 0 }~-2D- | g ~~~40- | 0 77 78 s 9 80 " 82 8~~~~~~~S es 85 8 7 se se _ - 13 - 2.34 On the lending side, the ceiling for banks was lowered from 19Z to 17.5? in early 1983 only to be increased again for non-priority sectors to 19? one year later (the lower ceiling of 17.5? being maintained for loans to priority sectors). In early 1986 the ceiling was lowered again in two steps and unified at 152 where it stood until March 1990. A higher rate has applied to finance companies (18.5Z until March 1990). In March 1990, the ceiling rates were increased by 1.5Z. 2.35 Before 1986, deposit rate ceilings varied by term. Looking at the rates for deposits between 12 months and two years, the ceiling on rates remained at 132 until early 1986 when it was lowered, in two steps, to 9.5Z. As mentioned, the ceiling was removed in mid-1989 for deposits of over one- year's maturity. The ceiling on savings deposit rates is currently 7.5?. Finance company promissory notes are subject to an interest rate ceiling cur- rently at 13.5?. 2.36 The degree to which these ceilings have actually bitten has varied over the years. As for deposit rates, Figure 2.8 illustrates that they have tended to fluctuate with money market rates: when the repurchase rate has exceeded the ceiling deposit rate, for example, the one-year deposit rate has been at or close to the ceiling. Deposit rates show less volatility than money market rates: when money market rates fell to about 5? in 1987, the one- year time deposit rate remained at 7.25Z. When money market rates once more approached the ceiling on deposit rates in late 1988, the ceiling again began to bite. The authorities' response was to remove the ceiling on long-term deposit rates in mid-1989: this has resulted in some banks increasing their deposit rates by between one-half and one percentage point over the old ceiling. Shorter term time deposits remained subject to the ceiling until March 1990 when they, too, were freed; only savings deposit rates now remain subject to ceiling. 2.37 On the lending side, despite the fact that prime rates have been well below lending rate ceilings (Figure 2.9), lending rate ceilings do seem to bite continuously to a greater or lesser extent. This is illustrated by Figure 2.10 which shows the distribution by interest rate of bank loans in 1985-88. For each year the distribution has two peaks: one at or about the prime lending rate, the other at the ceiling. (It is noteworthy that little lending occurred above the separate lending ceiling for priority sectors in 1985). Clearly, there would not have been such .. bunching of loans made at 17.5? in 1985 or 152 in the other years had it not been for the fact that these were precisely the ceiling rates. Without ceilings the spread over prime charged to many borrowers would have been higher. 2.38 A further striking point from Figure 2.10 is the widening of the gap between the rates at which the two peaks in the distribution occur after 1986. This is reflected in a widening intermediation margin and increased profitabi- lity--at least of the larger banks--in those years. It is likely that lending rates to prime borrowers are not much influenced by the interest rate ceilings because of their greater bargaining power and alternative borrowing possibili- ties. The data suggests that a lowering of the ceilings will have its impact in the short-run on bank profits, and on some middle level borrowers who will have to pay a higher spread over prime. Theoretical considerations suggest - 14 - .).; (Detared 2.7: Real Deposit Rates (Do M.d by subsequent year CPI) 12- / _2 1 0 . 4 - 83 84 865 86 T 88e O Figure 2.8: Cellng and Actual Deposit Interest Rates 14 - _ 10 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~0 .. 's' :'s>~~ ,o Actulal 0 -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~- * ~~~~83 84 85 se 87 8 8-av ;,. .;'.' :Fiure 2.9: Clillng and Actual PrCme Lending Interest RatAs 20 - '''''''"";' 8 14 8686 8788 t;9>;*;X.4 838 868 8 4.~~~~~~~~~~~~ss t Figure 2.10: Distribution of Bank Loans by Lending Rates (in Percentages) 1987 1988 030-so -.---- os --- U U t t 25 8 40 t t 20n d d so I IsI n n o g 20 to C C r r * d t 0 t 0iIi*m ~ mummmIumUUiIIloom 8.6 9 9.6 10 V061 11.6 ISU U 31.6 14 14.6 16 1.6 10616.t?17.SISto6195 '0.6 99.6 1010.61 MGS1212MG 13 .6 14 4.6 1M 1.61S16MI17t7.6 a S.6 9 Interest Rate Charged Interest Rate Charged 1987 1988 t * 20,20 t t a a n 5n I d d 1 n n g to g 10 d d~ItretRt hrgdItrs Rt hre - 16 - that the higher risk borrower could also be expected to suffer from reduced availability, and that in the long-run interest rates for small depositors would probably fall. 2.39 The conclusion must be that lending rate ceilings, at the levels at which they have been pitched in recent years, do distort the functioning of the banking system. To improve the efficiency of the system, it is recom- mended that the ceilings on lending rates are also removed as the constraints on deposit rates are eliminated (a process now well advanced). Determinants of Short-Term Interest Rates 2.40 The Influence of the Bank of Thailand. Determination of bank interest rates within the ceilings is influenced by demand and supply condi- tions as expressed in the effective interest rate for wholesale bank liqui- dity. Banks adjust their liquidity in the interbank market, in the repurchase market for government bonds or through borrowing from foreign correspondents in foreign exchange, which may in principle be covered. The BOT intervenes only in the repurchase market and in the spot foreign exchange market (where it fixes a buying and selling rate each morning) but not in the forward foreign exchange market. 2.41 The BOT operates tne repurchase market in Thai government bonds as a means of influencing liquidity in the banking system. The BOT acts as a market maker in that all bids and offers are made to it each morning. It can choose to satisfy the market in whole or in part, or to leave unsatisfied bids or offers. The sale and repurchase agreements are for term of 1,3, 15 or 30 days, with most activity concentrated at the shorter end. Any unsatisfied or emergent liquidity needs or surpluses can be addressed by individual banks in the interbank market in the afternoon. The BOT does not intervene in the interbank market. 2.42 In many countries the Treasury bill is the key money market instru- ment and its yield is taken as the pivot on which other rates depend. For several years the quantity of Thai Treasury bills was substantial. Further- more there was a regular auction of the bills. However, the improved fiscal situation has resulted in the supply of Treasury bills effectively drying up. There were no auctions between February and September of 1989. Furthermore, especially in recent years, the chief bidders for bills were public entities, and the yield determined at auction tended to be well below money market rates. Treasury bill auctions and rates are no longer influential in setting market rates. 2.43 Role of Foreign Interest Rates. Some banks make quite heavy use of borrowing from correspondents abroad. Such borrowing is limited to a ceiling per bank of B S billion or 202 of the bank's capital, whichever is smaller. The exchange risk of borrowing in foreign exchange could be covered by individual banks in the forward market, which has become quite active at the short end, though contracts beyond six months maturity would be rare. Even though there is a limit on the banks' foreign borrowings, such borrowings represent an important source of funds. The foreign interest rate adjusted for the cost of forward cover, therefore, constitutes an important influence - 17 - on the domestic rates. In the last few years, the government repurchase rate has been closely correlated with the foreign interest rate inclusive of the forward premium (see Figure 2.11); interest parity has normally prevailed during this period. However, in the mid-1980s, periods of stress were associated with higher domestic interest rates than were reflected in the forward discount. 2.44 Domestic Monetary Conditions. Although the evidence is not absolute- ly conclusive, the interest rates for bank liquidity appear not to have been insensitive to domestic monetary conditions as they would be if capital mobil- ity was perfect. Indeed, there has been a close correlation between the repurchase rate in Bangkok and the amount of banks" net borrowing from abroad and the amount of their borrowing from BOT. This suggests that capital does not move with perfect smoothness from foreign centers to meet any local requirements. This would suggest in turn that monetary policy can influence domestic lending interest rates and, through these, demand conditions. On the other hand, it was not possible to identify any statistically significant domestic factors affecting deposit rates after account was taken of the effect of foreign interest rates and the forward premium. The technical note given as Annex 11 provides some econometric evidence of the influence of domestic conditions on Thai (repurchase) interest rates, and the lack of such influence on the deposit rates. This area deserves further study in order to help the authorities plan their interventions on short-term interest rates and monitor their effectiveness. 2.45 However, because of the de facto high degree of openness of the Thai capital market, the impact of domestic conditions on Thai interest rates is essentially short-run in character. The econometric analysis shows that the tendency after several months is for Thai rates to converge to foreign rates (as adjusted for the effects of anticipated exchange rate movements). This implies in particular that any attempt to push domestic interest rates below foreign ones on a sustained basis would result in capital outflows and become ultimately unsustainable. The authorities have wisely refrained from any such attempt. Targets and Instruments of Monetary Policy 2.46 Money-Market Interest Rates as Indicators. Although the monetary policy is generally formulated in terms of monetary and credit aggregates, it is implemented by intervening to influence the availability and cost of bank liquidity. The attempt to achieve targets for the aggregates while monitoring interest rates poses difficult questions of judgment on an operational basis. In particular, deviations during the progr&, year from the originally expected path of external variables can result in higher interest rates than expected. The authorities recognize this by reviewing their targets for the monetary base quarterly, and their interest rate policy monthly. It is important for the authorities to keep their short-term operating procedures under review so that they maintain an accepted framework within which they can respond with confidence to unexpected developments within the course of the program. They need both a prepared methodology for analyzing market trends--so that they can quickly decide whether or not to lean against the wind--and a flexible mechanism for effecting whatever interventions they decide upon. Figure 2.11: Thai Government Repurchase and Eurodollar Rates, 1984-1989 19% - 18% 17%- 16% - 16% 14% 13%- 12% - 11% 10%- 1v \} 9% 8% y 7%- 6% - 5% Jan1984 Jan1985 Jan1986 Jan 1987 Jan1988 Jan1989 3 EURODOLLAR+1-MO. FORWARD PREMIUM + 3-DAY GOVERNMENT REPURCHASE RATE - 19 - 2.47 For example, the experience of the last few years is that BOT has not fully accommodated within year shifts in the demand for bank liquidity. Towards the end of 1988 especially, though BOT was providing considerable liquidity support, it was not enough to prevent interest rates from rising quite sharply. Since BOT took action to increase bank liquidity, this episode could be (and has been) considered one of expansionary monetary policy. But since interest rate differentials with abroad were allowed to rise, it could be argued that the episode was one of monetary tightness. The authorities need to be able to form their judgements in such situations even as they unfold. The increasing exposure of the financial markets to fluctuations in external flows will probably increase the difficulty of making these assess- ments. 2.48 By paying greater attention to interest rates, the authorities might also be in a position to smooth out seasonal fluctuations in liquidity condi- tions. A fully effective interest rate policy would have succeeded in eliminating any systematic seasonal fluctuations or predictable transitory movements in interest rates. However there is some evidence from analysis of the past five years that some systematic seasonal patterns may remain. It is recommended that the authorities should intensify research efforts to enable them to improve their forecasting of the factors that influence bank liquidity conditions so that transitory and seasonal influences can be smoothed out, and so that unplanned conditions of monetary ease or tightness can be recognized for what they are and dealt with accordingly. 2.49 Reserve Reauirements. A number of reserve or liquidity requirements, some of them overlapping, apply to banks and finance companies. The basic reserve requirement is to hold an amount equivalent to 72 of total deposits in the form of deposits with BOT (at least 2Z), holdings of eligible (government) securities (from 2.52 to 52), and vault cash (up to 2.52). Finance companies are also subject to a 7X requirement, though the composition is different: 0.52 in deposits with BOT, 12 in deposits or call loans at commercial banks and 5.52 in the form of government securities or securities guaranteed by the government. 2.50 To be eligible for the award of new branch licenses, banks must satisfy a separate ratio, namely 162 of deposits to be invested in eligible securities (including government and state enterprise and IFCT bonds). Securities counted towards the basic reserve requirement may also be counted here. This branching requirement has recently been relaxed to allow banks to sell up to 42 of the total 162 on the repurchase market. 2.51 In addition, according to the "local lending requirement*, each bank branch established outside Bangkok since 1975 must lend at least three-fifths of its deposit resources locally, i.e. in its own or adjacent provinces. Failure to make such loans gives rise to a compulsory interest-free deposit at BOT in the amount of the shortfall, or alternatively to the holding of govern- mant bonds in the amount of 4.5 times the shortfall. In mid-1989 the short- fall (apart from waivers granted) was B 1.6 billion or about 0.172 of total bank deposits, of which B 1.5 billion was deposited with BOT. - 20 - 2.52 There is also a BO guideline that banks should hold at least the equivalent of 20a of deposits in liquid assets. This would automatically be satisfied by the combination of the basic reserve and the full 162 branching reserve. 2.53 Since cash reserves perform no useful role as a safety net in a modern banking system, these requirements--apart from the local lending requirement, which is comparatively sAall--should be seen as primarily the fulcrum on which monetary policy is levered and secondly as a form of assured and fairly inexpensive finance for the government. This fiscal role of the requirements is discussed in para. 3.34 in the context of the impact of fiscal impositions on intermediation costs. From the monetary policy point of view, a single ratio should be sufficient to control the evolution of domestic credit. The basic reserve requirement is the instrument most attuned to such a purpose. The other requirements should, therefore, be phased out. As a further simplification, the government bonds portion of the primary require- ment should also be eliminated. Indeed, so long as government bonds are held in an amount far greater than the requirement, that portion of the basic reserve requirement is redundant. By phasing out the requirement to hold government bonds as part of various reserve requirements, the authorities would also help develop a transparent and active secondary market In govern- ment bonds. 2.54 The fact that finance companies are permitted to satisfy almost all of their reserve requirement through holdings of government bonds and bank deposits could prove to be a serious loophole in a period of mo.etary restraint, as banks could refer depositors to associated finance companies where the required reserves were less onerous for the group to satisfy. Since finance company promissory notes are essentially equivalent to savings deposits at banks, the requirements should be made equivalent. The "level playing field, argument used in the context of the discussion of roles of banks and finance companies (see para. 4.35, would also require that finance companies' promissory notes should be subject to the same reserve requirements as are imposed on banks' savings deposits. 2.55 The branching requirement should be removed in a phased manner with attention to the transitional steps, to avoid any inflationary impact of the change. The basic principle should be to increase the yield on government bonds to the point where they are held voluntarily. If at that stage the banks are still the main holders, then the removal of the requirement will clearly have no immediate impact on the volume of bank lending. Even if the process of bringing the rate up to market leaves a greater volume in the hands of the non-bank public still, as a 'irst approximation, this should have little or no effect on internal balance. The tax status of interest on the government bonds in the hands of the banks, together with the banks' own tax situation will clearly influence the rate at which the bonds would be held voluntarily by the banks. There will be no substitute for a gradual process of feeling the water in order to find and converge on the necessary rate. - 21 , Mechanisms of Intervention 2.56 Considering the steadily increasing sophistication of the Thai banking system, the instruments used by BOT for influencing liquidity condi- tions seem somewhat inflexible. The main instrument is intervention in the repurchase market. In 1987-88, BOT also issued its own bonds to the banks in order to mop up liquidity. The banks may also access the loan window of BO3 for last resort facilities, but this use is limited in duration and generally at a penal rate. 2.S7 The repurchase market is conducted as a daily auction where, by con- vention, the participants generally make bids or offers only at a single interest rate. By netting the bids and offers, BOT arrives at the net deficiency or surplus in the market, which it can choose to satisfy or leave uncovered. The BOT has no mechanism for taking the initiative in signalling its intentions for the direction of the interest rate except by failing to satisfy the net demand. Thus if interest rates are, in the view of BOT, too low, then it will leave an unsatisfied demand for liquidity at the end of the morning repurchase market fixing, leaving it to the deficit banks to make up their deficit in the interbank market or eventually at the loan window. Considering that it will not be easy for the banking system as a whole to generate additional liquidity within the day, the likelihood is that substan- tial deficits in the repurchase market will result in the loan window being used that day. In the case where BOT wishes to see interest rates lower, it will arrange that the repurchase market is left with unsatisfied offers of liquidity. Once again this will push the banks into the interbank market. That market should experience a falling of rates, but it may well be that banks will end the day with excess holdings of cash. 2.58 Under the existing arrangements, it seems that 3OT may not have an easy way to take the initiative in influencing the trend of interest rates, especially if it wishes to see interest rates increase. For example, the recent strength of inward capital movements has created the need for morb vigorous action to sterilize the impact on domestic liquidity. The issue of BOT bonds has been employed for this purpose, but this has been seen as an exceptional measure, and Ithe bonds had relatively short maturities. The BOT should consider se.e of shorter-term bills or the occasional use of aggressive bidding for depos$:s--either on an overnight basis or at term--from the banks as supplementary flexible means of allowing it to influence rates in an upward direction when that is needed. Any such deposit taking arrangement should not, however, become an open ended and permanent repository for surplus bank funds. The BOT bills would have the added advantage of being negotiable in the secondary market at a time when the diminished supply of government paper is probably impeding the development of the money market. 2.59 The BO3 might also consider intervening in the forward foreign exchange market; this could perform several useful purposes. First, at times when the covered interest parity condition fails to hold, BOT could step in to correct the market imperfections by buying forward foreign exchange. On the whole, outright sale of forward foreign exchange involves more risk than benefits especially at times of lack of confidence in local currency. If the currency has in fact depreclated by the time the forward contract matures, the - 22 - effect will be an undesired expansion in bank liquidity. A second use cf forward foreign exchange intervention could be in the form of a swap trans- action whereby BOT could acquire foreign exchange under a swap or repurchase agreement and thereby influence domestic liquidity conditions--and lower interest rates--without requiring the banks to have holdings of government bonds as at present. However, the adverse experience of other central banks in supporting the forward foreign exchange market at unrealistic rates is a reason for exercising great caution in this area. Lon2-Term Interest Rates 2.60 For reasons discussed in Chapter IX, the long-term bond market is not well developed in Thailand. Promotion of this market will clearly be an objective of the authorities. For the time being the main instruments are likely to continue to be government bonds with some state enterprise bonds. Therefore the authorities' decisions as to issue price and buy-back procedures will continue to be of some importance in establishing interest rate condi- tions at long maturities. 2.61 The government issues bonds at two maturities. The so-called 'loan bonds', are issued with a five-year maturity and are aimed especially at the household saver. The "investment bonds", used by institutional investors including the banking system, carry a slightly (25 basis points) higher coupon and have a maturity of ten years. At present the policy has been to set the interest rate on government bonds a fraction above the ceiling rate on term deposits.6/ Pricing of long-term bonds by reference to current short-term rates is not an altogether satisfactory approach. Ideally the long-term rates would be determined in a competitive market. Pending the emergence of an active market among professional fund managers in medium and long-term paper, the authorities should seek to ensure that rates approximate to those which would emerge in such a market. In particular this implies that they take a view on the equilibrium slope of the yield curve in Thailand in establishing the interest rate on new issues. Considering the interest risk involved, a reasonable rule of thumb might be that long-term paper should yield lz-2Z more than money market instruments in normal times. In times of monetary restraint, it would be reasonable to tolerate an inverted yield curve. Actually the policy adopted over the last several years has not deviated too far from what would be implied by this prescription, except that the long yields were still pitched a little low on average. 2.62 While the ten year bonds can be used by financial institutions in the repurchase market, the five-year bonds effectively carry a buy-back clause, whereby the authorities stand ready to buy the bond back at par. This results in massive redemption of bonds at the time of new issues when coupons are rising, and passes all of the interest rate risk to government. It seems that there should be a larger gap between the coupon on such a bond with buy-back at par and that on bonds that can only be sold at market value. 6/ After the removal of the ceiling in mid-1989, the rates fell below the highest deposit rates. - 23 - C. Resiliency of the Financial Sector and Susceptibility to Economic Slowdown 2.63 Although Thailand has experienced uninterrupted growth for over two decades, there have been two intervals of relatively slower progress in the past decade. GDP growth averaged only 5Z during 1978-82, and only 42 in 1984- 86. No single cause can be adduced for these slower periods; some of the causes were external, including the adverse terms of trade which hit the value of agricultural exports in some years, and the loss of competitiveness resulting from the strength of the US dollar to which the baht was linked. An important contributing factor was the restrictive fiscal policy which suc- ceeded, after 1985, in reducing and eventually eliminating the budget deficit. Monetary policy was also restrictive in late 1984. The growth recession of 1985-86 particularly affected the usual pro-cyclical sectors such as housing, while the volume of imports actually fell in 1985. With exports rebounding from their 1983 trough, helped by the local currency depreciation of late 1984, the balance of payments turned around, moving from a deficit of over 7? of GNP in 1983 to a small surplus in 1986. 2.64 Although the banking sector was not uninfluenced by the slower economic growth of the early and mid-1980s, it would be a mistake to see the difficulties experienced by the financial sector in those periods as being primarily caused by cyclical economic fluctuations. As already foreshadowed in a speech by the Governor of BOT in 1981, the main problems of the Thai banking sector had more to do with management failures allowing a concen- tration of lending in interrelated enterprises and excessive growth of expenses. 2.65 Even though a recession is considered less llkely than sustained growth over the next few years, it is worth considering how well the financial sector might cope, and in particular whether the kind of banking crisis of the early 1980s might be avoided. The experience of the last decade has taught Thai bankers some lessons which will help reduce their future vulnerability. They are probably less prone than in the past to make credit decisions based heavily on collateral. After the difficulties of the earlier recession, all banks are now seeing the value of collateral, which seemed so depreciated only a few years ago, jump to levels that make the prospects of loan recovery much better than they were. It is true that almost one-quarter of commercial bank loans go to real estate, construction, or financial services companies and that these sectors are considerably more volatile than the economy as a whole. Thus banking is vulnerable to downturns in these sectors. Recognizing this, BOT has acted to discourage excessive lending to property development, lest the banks should become too exposed to another possible fall in collateral values. 2.66 Bank profitability over the past few years has improved sharply, through the rapid expansion in the balance sheet size has meant that reported capital adequacy ratios have not strengthened by much between 1985 and 1988. An audit of the condition of the banks' portfolio did not form part of this Study, and therefore no view can be taken of the robustness of reported capital. However, this much can be said with confidence: the average - 24 - portfolio quality (after provisions) has improved. This would imply that the banks are better placed than before to weather a downturn. 2.67 If the banking sector is already better placed than before to weather a downturn, the recommendations made in this Study should further increase its resiliency. Strengthened capital base of the banks (Chapter III), rational- ized, integrated institutional structure, and transparency in transactions and ownership of subsidiaries (Chapter IV), policies to strengthen competition in the system (Chapter V), improved supervision and regulation, and strengthened arrangement to deal with financial distress in the system if and when there is a shock (Chapter VI), should all help to increase the resiliency of the system in dealing with future shocks. - 25 - ITI. COMMERCTAL BANKING IN THAILAND A. Overview of the Financial System Components of the Financial System 3.1 The formal financial intermediation system in Thailand consists of the central bank (Bank of Thailand, BOT); the banking system comprising com- mercial banks, finance and securities companies, credit foncier companies and a government savings bank; private and government insurance companies; and a number of sectorally and functionally specialized financial institutions. The total assets of the financial system amounted to B 1,934 billion (US$77.5 billion equivalent) as of the end of 1988. There is also a stock exchange which has been very vibrant in the last couple of years; market capitalization of listed equities amounted to B 367.3 billion as of the end of 1988, increasing to B 659.5 billion by the end of 1989. 3.2 The comercial banking sector dominates the financial system. Its relative share has gradually but steadily increased, from 512 of the total financial system assets in 1970 to 59? in 1988. There has been a corres- ponding gradual and steady decline in BOT's share, from 362 of the total system wide assets in 1970 to less than 162 in 1988. The relative share of the central bank in the country's financial system assets is significantly lower (and the banking sector's share higher) in Thailand than is typical in developing countries, and represents the success of commercial banks in mobil- izing and intermediating funds. Within the banking sector, domestic private commercial banks dominate. The non-bank financial intermediaries have gained a little in terms of relative share but still remain, as a group, a small part of the financial system - as of the end of 1988, they accounted for 8.4? of total system wide assets. Within this group, the 12 insurance companies toge- ther accounted for 1.62 of total system wide assets, and no single specialized financial institution accounted for more than 22 share as of the end of 1988. The components of the Thai financial system are given in Annex 12. Develonment and Depth of the Financial System 3.3 Since 1970, the assets of the financial system have grown at a com- pounded annual rate of 192. The lowest growth rate was recorded during the 1980-85 period when it was 162 annually, and actually became negative in 1986 (-5Z). But the growth has rebounded strongly since then, with rates of 30S and 252 in 1987 and 1988 respectively; there are indications that the growth rate remained high in 1989. 3.4 Thailand's financial system is also very deep. As Annex 7 shows, the ratio of broad money to GDP in 1989 was 65.42, a figure exceeded only by a - 26 - handful of developing countries.7/ Much of this monetization has taken place relatively recently as shown in Figure 3.1. As recently as 1981, the ratio of money to GDP vas only 38.5?. The rapid monetization took place because an increasingly large share of private saving (net of depreciation) was channeled into accumulation of financial assets. Annex 7 shows how in 1984-88, 802 of saving was in the form of broad money. This contrasts with an average of 40t in the 19709 and 602 in 1980-83. This increased "financialization" of saving seems to be taking place both in response to high real interest rates (see para. 2.33) and increased confidence in the economy as a result of the suc- cessful macroeconomic adjustment. The rapid monetization itself must have contributed to the adjustment (as well as preventing a more serious crisis earlier in the 1980s) by lessening the adverse impact of any given fiscal deficit on international reserves. The high monetization ratio in 1987-89 also suggests that the financial sector is keeping up with the booming economy. 3.5 The monetization of the economy has led to a complementary rise in credit. As shown in Annex 8 and discussed in detail later in this Chapter, this increase in financial intermediation has been well distributed across sectors. All of the sectors identified except for utilities show an appre- ciable increase in the ratio of total credit outstanding to value added over 1981-88. The distribution of credit across sectors continues to differ from their sectoral ahares in GDP. In particular, the agricultural sector received far less credit relative to value added than the other sectors. This is normal, and at least partly reflects structural or technical characteristics of the sector. 3.6 Although the financial system has grown rapidly and is also deep, it is not very well developed in certain other respects. It does not offer a wide choice of instruments to the users and providers of funds; the private debt securities market is particularly underdeveloped. The money market is also inactive. Certain methods of financing commonly used in other countries are not widely available to Thai users of capital, e.g., financial leasing, accounts receivable factoring and trust receipt financing for non-import sectors. As discussed in paras. 2.33-2.39, lending and deposit rates are still controlled. Freedom of entry and exit in the banking sector also does not exist. Most banks and major finance and security companies are still family controlled, and truly independent and professionally managed financial institutions are very few. 7/ The World Development Report 1989 lists only eight low and middle income countries (out of 73) with a higher ratio of broad money to GDP in 1987-- China, Yemen PDR, Yemen Arab Republic, Egypt, Jordan, Malaysia, Portugal, and Greece. Even some of these countries have price controls and goods rationing that artificially inflates the money to GDP ratio (the "money overhang'). Thailand's monetization ratio even compares favorably with that of high income countries--their median ratio of broad money to GDP in 1987 was 64.8?, compared to Thailand's 64.9?. - 27 - Figure 3.1: Ratio of M2 to GDP, 1960-1988 70% 66% - 60% - Ss% - ,, 50% l z w 0 m 45% uiU a. = 40%9 35% - 30% - 26% - 20% - 1960 1966 1970 1976 1980 1985 Q M2/GDP - 28 - B. The Present Banking Scene Historical Perspective 3.7 Commercial banking in Thailand has a long history. The Hongkong & Shanghai Banking Corp. opened its branch in 1888, followed by Standard Char- tered Bank in 1894 and Banque Indosuez in 1897. These international banks followed their clients along trade routes and were originally sub-branches of Hong Rong or Saigon (Ho Chi Minh City) offices. At the turn of the century, banking in Thailand was controlled by international institutions. The first domestic bank, The Siam Comercial Bank Ltd., was formed in 1906 by the Crown. As has been the general practice in most countries, special interest groups led the establishment of domestic banks: The Thai Military Bank was established by the armed forces in 1957, Krung Thai Bank by the government in 1966, and the rest by individual family groups between 1933 and 1960. Each bank focussed on providing financial support to the owning group's commercial and industrial activities. While in other countries the ties between the group owned bank and other group enterprises have loosened over time as the bank's ownership has been diluted by new stock issuance, in Thailand the original ownership groups continue to maintain tight control over their institutions. Regulatory restrictions that provide the domestic banks with competitive advantages over foreign banks (discussed in paras. 3.50-3.51) and the preference of the leading industrial and trade groups to use their own institutions have assured the dominance of domestic banks over the foreign banks. 3.*8 The legal and regulatory framework within which the banks operate has evolved in response to changes in the financial system. Two sets of changes are noteworthy. First, in the 19709, several laws and regulations were changed to discourage concentrations of economic power by requiring dispersion of ownership and markets through restrictions on shareholdings, allowable business activities and lending concentrations. Unfortunately, the unintended result of these changes has been the creation of a complex and opaque struc- ture of cross ownership which has essentially defeated the attainment of the intended objectives. The second and more successful change has been a shift in the primary focus of supervision and regulation from concern with com- pliance with laws to the assessment of financial condition and solvency. Both aspects are dealt with in detail in the relevant parts of the report (see paras. 4.36 to 4.46 and 6.7). Ranking of Banks and Market Shares 3.9 Commercial banks represent the single largest component of the Thai financial sector. Measured in terms of assets, commercial banks as a group now account for 592 of the total assets of the entire financial sr 'em as of the end of 1988; this relative share has increased from 51Z in 19; Within the banking segment of the financial system, commercial banks now account for about 78Z of the total segment assets. The B 1,147 billion in coamercial - 29 - banking assets are held by 15 domestic banks 8J with 2.061 domestic and 30 overseas offices, and 16 branches of 14 foreign banks. These banks are ranked in Table 3.1. Table 3.1: RANKING OF THAI COMMERCIAL BANKS. DECEMBER 31, 1988 (Billion baht) Total Total Domestic Banks Assets Foreign Banks Assets Bangkok Bank Ltd. 350.8 The Bank of Tokyo 10.8 Krung Thai Bank Ltd. 176.4 The Mitsui Bank Ltd. 10.5 Thai Farmers Bank Ltd. 156.4 Citibank NA 6.1 The Siam Comercial Bank Ltd. 103.3 Deutsche Bank (Asia) 5.0 Bank of Ayudhya Ltd. 70.2 The Chase Manhattan Bank NA 4.5 The Thai Military Bank Ltd. 69.6 Banque Indosuez 3.6 First Bangkok City Bank Ltd. 47.6 Hongkong & Shanghai Banking Corp. 3.2 The Siam City Bank Ltd. 46.6 Bank of America NT&SA 3.0 Bangkok Metropolitan Bank Ltd. 44.7 Standard Chartered Bank 2.8 The Bangkok Bank of Commerce Ltd. 38.2 United Malayan Banking Corp. Ltd. 1.7 The Bank of Asia Ltd. 32.4 Bharat Overseas Bank Ltd. 1.1 The Union Bank of Bangkok Ltd. 19.4 Security Pacific Asian Bank 0.9 The Thai Danu Bank Ltd. 14.2 Internat'l Commercial Bank of China 0.7 Nakornthon Bank Ltd. 11.0 Four Seas Communications Bank Ltd. 0.5 The Laem Thong Bank Ltd. 5.5 Total assets 1,186.0 54.5 Source: Bangkok Bank Ltd. 3.10 Evolution of Market Share. Relative market shares remain generally quite stable. The changes in market share occuring in Thailand between 1986 and 1988 are however, significant, with the three largest banks losing market share to the smaller banks (Table 3.2). Such movements can be indicative of a highly competitive market, uneven management strength, the relative growth rates of related industrial and trading groups, or significant corrections to the book value of loans to troubled clients. 81 The Sayam Bank has effectively merged into the Krung Thai Bank and is not considered to be a separate entity although it continues to exist as a corporate shell. - 30 - Table 3.2: EVOLUTION OF MARKET SHARE, 1986-88 (As percentage of Total Loans and Advances) Bank 1986 1e87 1988 Change Bangkok Bank Ltd. 33.3 31.9 30.1 -3.2 Krung Thai Bank Ltd. 10.9 12.5 12.2 -0.4 * Thai Farmers Bank Ltd. 13.0 12.7 12.4 -0.6 Siam Commercial Bank Ltd. 8.2 8.0 11.1 +2.9 Bank of Ayudhya Ltd. 4.7 5.6 6.1 +1.4 The Thai Military Bank Ltd. 4.9 5.2 5.4 +0.5 First Bangkok City Bank Ltd. 2.8 3.6 3.7 +0.9 The Siam City Bank Ltd. 3.3 3.2 3.7 +0.4 Bangkok Metropolitan Bank 4.0 3.7 3.6 -0.4 The Bangkok Bank of Commerce 3.6 3.3 3.0 -0.6 The Bank of Asia Ltd. 1.9 2.3 2.6 +0.7 The Union Bank of Bangkok Ltd. 1.6 1.6 1.5 -0.1 The Thai Danu Bank Ltd. 1.1 1.1 1.2 +0.1 Nakornthon Bank Ltd. 0.7 0.7 0.9 +0.2 The Laem Thong Bank Ltd. 0.5 0.5 0.4 -0.1 Sayam Bank Ltd. 1.7 - - - Foreign Banks 3.7 4.1 4.9 +1.2 * Including impact of absorption of Sayam Bank Ltd. Sources Based on Bangkok Bank Ltd. data. In Thailand, all four factors are present. However, regardless of the cause of these changes, the fact that smaller institutions can increase their market share at the expense of the larger institutions is indicative of a competitive market. The subject of concentration/competition in the financial system is discussed in detail in Chapter V. C. Performance of Domestic Banks Peer GrouRini 3.11 For analytical purposes, domestic commercial banks may be divided into four peer groups by market share, but separating the government-owned Krung Thai Bank. This is shown in Table 3.3. - 31 - Table 3.3: CLASSIFICATION OF THAI BANKS INTO PEER GROUPS December 31, 1988 Return on Loan Market Equity Assets Share 2 Peer Group 1 Bangkok Bank Ltd. 11.19 0.57 31.3 Thai Farmers Bank Ltd. 10.75 0.71 13.0 The Siam Commercial Bank Ltd. 14.64 0.78 8.5 Bank of Ayudhya Ltd. 17.49 1.21 6.4 The Thai Military Bank Ltd. 10.34 0.60 5.8 12.15 0.69 65.0 Peer Group 2 Krung Thai Bank Ltd. 2.20 0.08 13.0 Peer Groun 3 The Siam City Bank Ltd. 4.37 0.26 4.0 First Bangkok City Bank Ltd. 7.86 0.65 3.9 Bangkok Metropolitan Bank Ltd. 5.94 0.32 3.9 The Bangkok Bank of Commerce Ltd. 4.73 0.22 3.2 The Bank of Asia Ltd. 16.33 1.00 2.8 7.58 0.46 17.8 Peer Groun 4 The Union Bank of Bangkok Ltd. 8.54 0.52 1.6 The Thai Danu Bank Ltd. 11.55 0.83 1.3 Nakornthon Bank Ltd. 14.12 0.67 0.9 The Laem Thong Bank Ltd. -48.28 -8.92 0.4 -5.62 -0.42 4.2 All Banks 9.25 0.51 100.0 Source: Published Financial Statements. 3.12 Peer Group 1 (PGI) comprises the five non-government banks with market shares of more than 52 each. PG2 consists of a single bank, the Krung Thai Bank, which as a government owned institution is a class by itself--it has the special privilege of being the depository institution for government agencies but also carries the special burden of undertaking financial restruc- turing projects at the government's invitation. The PG3 includes the five medium sized institutions that each have market shares between 2.52 and 52, while PG4 includes the smaller banks with market share of less than 2.52 each. Annexes 13-33 provide detailed financial analyses of all commercial banks combined, each peer group, and each individual domestic bank, based on pub- lished financial data. _ 32 - Ogerational Performance 3.13 Oneratina Efficiencv. On a consolidated basis, the overall efficien- cy of Thai baoks has been steady to slightly mproving for the last three years. Table 3.4 analyses the relative operating efficiency for 1988 using the ratio of each major expense category to total assets. Table 3.4s OPERATING EXPENSES OF THE ThAI BANKS As Percentages of Average Total Assets, 1988 Taxes & Other Staff Occupancy Equipment License Operating Total Peer Group 1 Bangkok Bank 1.04 0.12 0.20 0.27 0.33 1.97 Thai Parmers 1.00 0.18 0.38 0.29 0.56 2.40 Siam Comercial 0.97 0.17 0.31 0.25 0.44 2.14 Ayudhya 0.85 0.12 0.28 0.33 0.37 1.95 Thai Military 1.07 0.04 0.24 O.SO 0.44 2.19 1.01 0.14 0.26 0.28 0.41 2.10 Peer Group 2 Krung Thai 0.93 0.13 0.26 0.24 0.43 1.99 Peer Group 3 Siam City 0.91 0.08 0.18 0.23 0.30 1.71 First Bangkok City 0.54 0.09 0.12 0.26 0.23 1.24 Bangkok Metro. 0.96 0.23 0.18 0.27 0.35 1.98 Bank of Commerce 1.13 0.18 0.21 0.25 0.56 2.33 Bank of Asia 0.88 0.37 0.20 0.26 0.40 2.11 0.88 0.18 0.17 0.25 0.36 1.84 Peer Groun 4 Union Bank 1.38 0.29 0.32 0.30 0.39 2.69 Thai Danu 1.08 0.19 0.34 0.30 0.47 2.37 Nakornthon 0.97 0.36 0.26 0.3S 0.32 2.25 Laem Thong 0.89 0.1 .22 0.34 Q.45 2.02 1.15 0.26 0.30 0.31 0.40 2.43 All Banks 0.98 0.15 0.25 0.27 0.40 2.05 Sources Published Financial Statements. The ratio of total operating expenses (excluding loan loss provisions) to average total assets ranged from a low of 1.241 to a high of 2.69Z. Since this ratio excludes loan loss provision, it represents the basic measure of - 33 - efficiency. Too low a ratio can indicate an expansion of assets beyond the level normally supportable by the resources allocated, which could be expected to lead to a loss of quality control over the risk management process and a subsequent increase in troubled loans. Too high a number indicates a lack of cost control (and possibly weak accounting and management information) in the case of a large bank, or specialization on fee based service and loans to smaller borrowers at higher margins, as is likely to be the case with the PG4 banks. Clearly, while low operating expenses are a major contributor to the Bank of Ayudhya's high return on assets, this relationship does not hold for most of the banks, since interest differential income and loan loss provisions also play an important part. 3.14 Interest Differential Income. Annex 34 takes a somewhat simplistic look at the interest differential income of the banks, by developing the spread between the average interest rate paid on funding liabilities and the average interest rate earned on all interest earning assets.91 Interestingly, PG2 and PG3 banks, which had the most erratic and generally lowest operating expense per total assets, also have the most erratic and generally lowest interest differential margin.101 Particularly P03 banks' spread is distorted by BOT support program (see para. 3.24), indicating that the underlying earnings capacity of these middle level institutions may in fact be lower by some 27 basis points than the stated figure for the group. 3.15 The differences by peer group shown in Annex 34 are not new. Table 3.5 shows the evolution from 1986-1988 of the vpread by peer group. 'hile each group other than PG3 shows steady recovery from the depressed levels of 1986, PG3 actually showed a decline in 1988, indicative of sustained weakness in the core earnings capacity of this group. 91 This ignores the impact of non-interest sources (such as equity) and uses of funds. 101 The spread for Laem Thong Bank is distorted by the existence at the beginning of the year of significant non-performing loans in the port- folio. While these were provisioned for in 1988, and thus taken out of net earning assets, their presence at the beginning of the year raises the average loans outstanding, thus lowering the net interest earned. - 34 - Table 3.5: EVOLUTION OF INTEREST DIFFERENTIAL MARGINS BY PEER GROUP 1986-1988 1986 1987 1988 2 X X Peer Group 1 2.86 3.30 3.62 Peer Group 2 2.27 2.41 2.72 Peer Group 3 0.46 /a 2.24 2.21 Peer Group 4 2.41 2.79 3.25 All Banks 2.36 2.97 3.23 La Due to the Siam City Bank, which had a margin of -3.96 in 1986. Source: Published financial statements. Liquidity 3.16 Table 3.6 shows the banks' liquidity as measured by loan to deposit ratios at year-end 1987 and 1988. The table also shows growth in the banks' loans and equity in 1988 over 1987. A comparison of changes in liquidity on one hand and growth in loans and equity on the other shows that it is those banks which had a higher rate of growth in loans and equity that also had a higher loan to deposit ratios in 1988 than in 1987 (the Thai Danu Bank is an exception). Balance sheets of the affected banks show that the increase in loan to deposit ratios was caused by a much higher rate of increase in loans than in deposits. 3.17 In a soundly managed institution the loan to deposit ratio tends to be fairly stable, with significant movements more likely to be attributable to changes in the general market. Significant changes in this ratio that are not consistent with movements within the entire industry are thus likely to signal changes in management philosophy, extraordinary circumstances or lack of ef- fective asset and liability management. 3.18 Of the PG1 banks, Bangkok Bank's loan volume decreased relative to deposits, some of which was accounted for by new provisions of 1.62 of end- 1987 net loans, which reduced new loans correspondingly. Bank of Ayudhya shows a 5.1X growth in the loans to deposit ratio during 1988 to a relatively high 104.8Z, although this was not significantly out of line with this bank's historical levels. The remaining PG1 banks show a healthy level under 902, but loan growth rates that exceed equity growth, indicate weakening financial resilience. - 35 - Table 3.6: LIQUIDITY AND GROWTH OF THE THAI COMMERCIAL BANXS 1987-1988 Loans To Deposits Growth during 1988 In 1987 1988 Loans Equity I I I I Peer Group 1 Bangkok Bank Ltd. 105.7 100.9 16.3 20.8 Thai Farmers Bank Ltd. 84.0 88.8 20.9 3.6 The Siam Commercial Bank Ltd. 79.4 88.3 24.2 8.3 Bank of Ayudhya Ltd. 99.7 104.8 32.8 43.0 The Thai Military Bank Ltd. 86.9 88.9 28.3 2.7 94.6 95.7 20.7 15.0 Peer Group 2 Krung Thai Bank Ltd. 79.0 77.4 20.1 26.7 Peer Groun 3 The Siam City Bank Ltd. 91.0 105.3 40.6 66.0 First Bangkok City Bank Ltd. 106.2 115.4 27.6 30.0 Bangkok Metropolitan Bank Ltd. 102.3 105.6 18.1 16.9 The Bangkok Bank of Commerce Ltd. 83.1 87.4 16.2 3.6 The Bank of Asia Ltd. 100.9 114.6 38.0 44.2 96.0 104.8 27.2 31.6 Peer Group 4 The Union Bank of Bangkok Ltd. 90.0 92.9 20.0 17.0 The Thai Danu Bank Ltd. 88.8 99.9 43.0 3.4 Nakornthon Bank Ltd. 81.1 93.1 48.2 57.2 The Laem Thong Bank Ltd. 108.4 97.6 3.0 -38.9 89.7 95.4 29.7 -1.4 All Banks 92.2 94.2 22.1 18.2 Source: Published financial statements. 3.19 PG2--Krung Thai Bank--continues to show the effects of its role as depository institution for government institutions. As shown in Annex 34, the Krung Thai Bank paid its depositors a lower average rate than that paid by other banks, and achieved a return on risk assets better than only two of its competitors, in part because of its extensive funding of the money market. Use of the Krung Thai Bank as depository institution may not be in the best financial interests of parastatal enterprises. 3.20 Within PG3, Siam City Bank, First Bangkok City Bank, Bank of Asia and to a lesser extent Bangkok Metropolitan Bank all show substantial increases in - 36 - the loan to deposit ratio. Loan to deposit ratios in excess of lOO1 are troubling, particularly for institutions that have significant earnings problems. The Bangkok Bank of Commerce has shown much sounder loan to deposit ratios, modest growth, and inadequate capital formation more characteristic of some of PGl banks. 3.21 The PG4 banks show substantial variation. The Union Bank of Bangkok shows controlled growth. The Thai Danu Bank shows excessive loan growth and declining equity in real terms, since the equity growth rate was below the inflation rate. The Nakornthon Bank also shows excessive loan growth but has been able to expand its equity base. The Laem Thong Bank remains well capitalized despite the substantial loan loss provisions taken during 1988 that prevent rational further analysis. 3.22 While two years do not form a trend (1986 data shows a number of distortions due to the then prevailing economic difficulties), and it is rea- sonable to assume that sound managements will respond to changes in their institution's performance, repetition of the movements seen during 1988 would result in improvement only in Bangkok Bank. The "trend" in all remaining banks was negative for 1988, leading to impaired liquidity, weakened capital adequacy or both. D. Portfolio Quality Assessina Portfolio Problems 3.23 While the financial statements provided by the Thai banks are remark- ably informative compared to those in other developing countries, little data is included on the condition of the banks' loan portfolios, other than rela- tively frequenc audit qualifications to the effect that agreement had been reached with BOT to defer provisions that would normally be required. There is, however, an indirect way to gauge the presence and extent of non-perform- ing loans. The 3.32 gross receipts tax is based on interest (and fees) actually received in cash rather than on accruals. Therefore, if the gross receipts tax expense is lower than what would be indicated by the interest (and fees) revenue shown, it would indicate that some of that revenue was not actually received in cash, thus indicating the presence of non performing loans.ll/ Table 3.7 shows the development of these tax payments over time. r'egative amounts would indicate recoveries of past due interest from earlier periods. As would be expected from the evolution of the economic environment, 1986 shows a general upsurge in 'missing' tax payments, reflective of an increase in non-performing assets. This deterioration continues in 1987, but is reversed in 1988--consistent with the improved financial condition of Il/ This analysis is indicative only, since the effect of receipts from loans booked in foreign branches that are not subject to the tax is ignored, other taxes may be included, and gross receipts tax may be paid on over- draft interest that is debited to the borrower's account, even though the client has lost the capacity to pay. However, the trend lines should reflect changes in portfolio quality. - 3? - borrowers In an improving economy. If the basic argument is correct, the data for 1989 should show a substantial further improvement. Table 3.7s SUORTPALLS L_ IN GROSS URCZIPTS TAX PAYMENTS, 1980, 1985-1988 (in percentages) 1980 1985 1986 1987 1988 Peer Group 1 Bangkok Bank Ltd. 3 14 27 25 22 Thai Farmers Bank Ltd. 24 19 20 20 14 The Siam Commercial Bank Ltd. 29 31 26 38 31 Bank of Ayudhya Ltd. 1S -9 8 1S 3 The Thai Military Bank Ltd. 9 6 9 12 10 15 22 23 19 Peer Group 2 Krung Thai Bank Ltd. 21 -6 6 22 15 Peer Grouo 3 The Siam City Bank Ltd. 16 -5 11 -3 19 First Bangkok City Bank Ltd. 17 -22 -17 -1 12 Bangkok Metropolitan Bank Ltd. 20 11 14 25 11 The Bangkok Bank of Comerce Ltd. 24 9 11 12 11 The Bank of Asia Ltd. 32 22 12 23 13 21 2 7 10 13 Peer Groun 4 The Union Bank of Bangkok Ltd. 5 8 10 9 The Thai Danu Bank Ltd. 22 10 13 15 14 NIakornthon Bank Ltd. 24 5 1 7 5 The Laem Thong Bank Ltd. 39 23 -13 -16 -29 27 8 1 9 6 All Banks 14 7 11 20 17 la From the levels indicated by interest (and fees) figures shown in income statements. Sources Calculated from published financial statements. 3.24 The BOT Sunort. The SOT has provided a significant amount of finan- cial assistance to Thai banks in the form of long term soft loans that are secured by treasury instruments. These soft loans are reflected in the banks' published financial statements as long term loans, and typically bear very low interest rates. While some of these long term loans may be independent of support programs (particularly Siam Commercial Bank, where the amount seems - 38 - too small to be worth the effort), it is estimated that OT had provided some B 17 billion in such support programs.121 If account is taken of this support, the return on assets (ROA) figures need to be adjusted; the adjusted figures are shown in Table 3.8. These figures should be interpreted as indicative only because, as stated earlier in the paragraph, some long-term loans may not represent soft loans. Table 3.8. IMPACT OF BOT SOFT LOAN SUPPORT, 1988 (million bahtj Peer Group Average 1988 Impact of Revised Term Loans 5.62 Spread la ROA* Peer Group 1 Bangkok Bank Ltd. 5,962 334 0.46z The Siam Commercial Bank Ltd. 323 18 0.762 The Thai Military Bank Ltd. 1.366 76 0.48Z 7,651 428 Peer Group 2 Krung Thai Bank Ltd. 1,482 83 0.03? Peer Groun 3 The Siam City Bank Ltd. 3,500 196 lb -0.22? First Bangkok City Bank Ltd. 3,300 185 lb 1.612 The Bank of Asia Ltd. 1.500 84 0.702 8,300 465 Seven Bank Total 17,433 976 la Assuming the soft loans carry interest at an average rate of 22 and the treasury bill rate averaged 7.6? during 1988. lb Are not paying income taxes, so no tax offset. Source: Staff calculat4ons. 3.25 The adjusted return on assets figures for Bangkok Bank Ltd., Siam City Bank Ltd., First Bangkok City Bank Ltd. and Bank of Asia Ltd. still remain distorted by delayed recognition of loan loss provisions under agree- ments with BOT. Although the published financial statements do not reveal any 12/ This figure differs from the estimated amount of BOT soft loans shown in Table 6.1 on page 99. The probable explanation is that banks' published annual reports do not accurately identify all BOT soft loans. Also any BOT soft loans to Sayam Bank are not included in the B 17 billion figure stated in para. 3.24. - 39 - soft loans to Bangkok Metropolitan Bank Ltd. or Bangkok Bank of Commerce (PG3), financial statements of both these banks are also qualified with respect to the adequacy of loan loss provisions. Thus the two largest banks and at least six other banks are in sufficient financial difficulty to warrant support arrangements. These eight institutions represented 68Z of the assets of the domestic commercial banking system as of December 31, 1988. The Laem Thong Bank is left out of this adjustment process since it is now well capitalized and may have already taken adequate provisions to accurately state its financial condition. 3.26 Size clearly plays a rolet institutions that exceed a total asset threshold of some B 25 billion and that continue to rely heavily on family management have difficulty competing with professionally managed institutions above that threshold or with limited service (and thus more manageable) small- er institutions. Some of the medium sized institutions are unlikely to achieve the adequate economies of scale and professional management needed to compete with the more efficient banks, leading to pressure for further con- solidation within the banking system. Ideally, this should lead to a pooling of resources through mergers between medium sized institutions to form larger institutions that can effectively compete. Unfortunately, such mergers are difficult because of the tight family controls over prospective merger can- didates where neither side wishes to lose the control and confidentiality, as would be necessary in a merger. If successful, however, the results of such mergers of necessity, in which groups lose their controlling interests, could lead to the emergence of truly independent private banks that would take the Thai financial system to the next level of development. The BOT may need to assume the role of marriage broker to achieve this objective, and should have the tools available to apply both the stick and the carrot to achieve such mergers. The alternative of outright failure of small and medium sized banks or takeover of failing institutions by the largest banks would increase con- centration in the industry, and will not be desirable. However, as has been shown by experience, bail-outs that preserve the identity of the failed institution and attempt to resuscitate it are expensive and rarely successful, since the institution that is burdened with non-performing assets or incom- petent management cannot compete effectively in the market. Portfolio Diversification 3.27 Annex 35 shows the evolution of the Thai banking system's loans by economic sector,131 in absolute amounts as well as in percentages, over time. It shows a remarkable degree of stability, with some clearly defined under- lying trends. Particularly significant is the steady increase in the finan- cing of manufacturing from a low of 18.42 in 1980 to 25.9? in 1988, while wholesale and retail financing has returned to historical levels after an increase of some 42 between 1975 and 1986. This reflects focus on the finan- cing of the most economically productive activities. Both import and export finance have declined substantially in relative terms, indicating greater 13/ While this data is for all banks, the international barks forn a suffi- ciently small segment of the bankin- system that thcLr inclusion has no material impact on the applicability of this data to t1 e Thai banks. - 40 - independence of the domestic economy from international trade. A possible cause for concern is the growth in real estate finance which increased by B 31.9 billion or 153S in just two years. However, as a percentage of total portfolio, the share of real estate increased from 3.8? in 1986 to 6.1S in 1988. The BOT is well aware of possible speculative excesses in this area and is closely monitoring the banks' real estate financing activities. E. CaRital Adeauacv Nominal and Actual Canital Adequacy Ratios 3.28 The shortage of equity capital has been voiced as a serious problem by Thai bankers. In this regard, Thailand is not unique. The adequacy of capital is a problem for banks in many other countries. In the U.S., the problem for many banks became serious in recent years because the huge write downs of developing countries' debt reduced the banks' capital base. In Thailand, the reason the banks face capital inadequacy problems is to some extent far more positives rather than the massive write-offs seen in other countries, in Thailand the recent explosive growth of the economy and its need for banking services has outpaced the growth of banking equity needed to maintain the capital adequacy ratio. Inevitably the banks are asking for a lowering of the 82 capital adequacy ratio in effect since April BE 2505 (1962). However, any lowering of the 8! capital adequacy ratio would run counter to the generally accepted international trend towards increasing capi- tal adequacy requirements, and is not recommended. 3.29 The bankers complain that the capital adequacy requirement is hampering the growth of banking. This should more accurately be interpreted to mean that the requirement to maintain a minimum capital adequacy ratio is serving one of its principal objectives, namely to act as a check on unre- strained growth, obviating the need for direct credit controls. Unfortunate- ly, this restraint is not operating as effectively as it should. While the statutory capital adequacy requirement is nominally 8S of risk assets, certain categories of (high) risk assets, such as loans to preferential sectors, are excluded from the calculation base defined by BOT. As a result, actual capital adequacy is significantly lower than the nominal requirement as shown in Table 3.9. - 41 - Table 3.9s COMMERCIAL BANK CAPITAL ADEQUACY RATIOS, 1980-88 1980 1985 1986 1987 1988 All banks - equity to normal risk assets 7.4 7.0 6.8 7.8 7.4 All risk assets 6.6 6.2 5.8 6.7 6.4 Total assets 5.4 5.2 4.8 5.8 5.6 All RA & Contingent 5.4 5.2 4.8 5.4 5.2 liabilities * * Incomplete data. Sources Banks' published financial statements. 3.30 In Table 3.9, the ratios shown on the last line make a further refinement in that the denominator includes not only all risk assets (defined as all loans and advances plus investments other than in treasury instruments) but also (incomplete) contingent liabilities somewhat akin to the level that might be achieved through a risk weighting process. Based on information con- tained in published financial statements, this is more indicative of the figure that would be derived from a rigorous application of the Basle Agree- ment on capital adequacy. These ratios indicate a shortfall of some 2.8Z against the required level of capital.14/ Increasing the Core Calital 3.31 Given the above analysis, it is a matter of some urgency to strengthen the banks, capital if banks are to continue to support the expanding economy while maintaining sound balance sheet structures. Measures need to be taken to increase the attractiveness of (or reduce the disincen- tives to) investing in banks' capital stock and to encourage banks to retain a larger portion of their profits in the business. Two entirely different approaches are being suggested. Under the first approach, the initiative has to come from the banks themselves. It should be reflected in the pricing of new issues of capital. Understandably, the controlling owner-groups would wish to peg the price of new issues at a level which would not result in any dilution of their holdings. Experience has shown, however, that most new issues have to be priced lower to be successful. The controlling ownex-groups must assign priority to strengthening of capital base, and accept dilution of 14/ This calculation assumes that equity as reported is not overstated due to insufficient valustion (specifically loan loss) reserves, particularly in the wesker institutions. In many countries, insufficient provisions for loan losses are the rule, rather than the exception. - 42 - their holdings if new issues of capital can be sold only at a price which would have that result. 3.32 Impact of Groun Related Activities. The controlling owner-groups should also consider that concentrated group holdings of banks' stock and cross ownership between banks and related commercial/industrial groups may also be factors in the relative unattractiveness of bank stocks as investment vehicles. These factors raise doubts as to the accuracy of reported bank profits in that income could be diverted through below market pricing of loans, above market pricing of deposits, and absorption by banks of losses/ expenses which should legitimately be borne by the group related borrowers. Such doubts and mistrust (which may be legitimate or wrongly perceived) can be overcome only over a period of time and only at the initiative of the banks themselves with the support of rigorously imposed accounting and disclosure standards. The banks need to be perceived as professionally managed, with arms length and transparent relationships with their related group entities. 3.33 Impact of Fiscal Impositions on Capital Formation. Under the second approach, the initiative will have to come from the Government. It is sug- gested that the Government review the fiscal impositions on financial inter- mediation. Annex 36 estimates the size of the direct and indirect tax over- head borne by the banking system. It indicates 15/ that the government re- ceived, in total some 62? of the benefit of the domestic banks' operations, thus equivalent to a 62? tax rate. The underlying earnings capacity of the banking system, if this external overhead is totally removed, would be some B 16.1 billion, representing a 2.3Z return on average assets in 1988 and a 402 return on equity. This is the theoretical maximum in that it assumes no taxes at all. 3.34 Since bank capital formation through retained earnings and through new issues of stock is directly related to after tax bank profitability, the authorities may wish to rethink the issue of what constitutes an acceptable level of bank profits. The government should review the taxation on financial intermediation primarily from the point of view of rationalizing it and in the context of domestic and international competitiveness. It is recommended that the government should consider to eliminates (a) all taxes on transactions between financial intermediaries,16/ (b) transactions-related taxes such as stamp duties, and (c) modifying the tax treatment of loan loss provisions to conform to BOT prudential guidelines (see also para. 6.32), These changes 15/ These estimates are somewhat simplistic, for example they do not take into account overseas activities that may be exempt from some of the direct and indirect taxes shown in this table, and assume that the for- eign exchange tax is calculated on the net industry profit, rather than the gross profit in each institution. However, some of the imprecisions appear to cancel each other, and the net result should be fairly indica- tive. 16/ This means applying taxes only at the edges of the financial sector where it interacts with the real sector. This will favor the most efficient reallocation of resources within the financial sector. - 43 - will strengthen financial intermediation and are desirable per se. The report has separately recommended, for other reasons, to (a) remove the 12-162 reserve requirement related to branch openings, and (b) eliminate bank sub- sidization of agricultural lending through the banking system. These changes would also improve bank profitability and make investment in bank capital more attractive. Later, the Government may also consider phasing out the 32 national and 0.3? local turnover tax, possibly in conjunction with some modi- fication to the basic income tax rates to achieve a fair, uniform tax standard for all financial intermediaries. 3.35 Comparison With Other Countries. Table 3.10 compares key performance ratios of the Thai commercial banks against similar ratios for five other countries. Table 3.10t COMPARATIVE PERFORMANCE RATIOS (Income and Expense items to Average Total Assets, in percentages) Interest Other Staff Profit Income Capital /a Country Year Margin Income Expense Before Tax Tax Formation Thailand 1988 2.72 0.59 0.98 0.91 0.34 0.50 Indonesia 1987 2.13 3.37 1.21 1.32 0.10 1.22 Morocco 1988 3.23 1.55 1.42 1.68 0.78 0.90 Pakistan 1988 2.35 1.02 2.05 0.58 0.23 0.28 Philippines 1986 2.39 2.76 1.37 1.40 0.15 0.92 Turkey 1986 2.81 4.81 1.87 1.79 0.22 1.56 Average 2.58 2.70 1.58 1.35 0.30 0.98 /a Net income after tax less dividend. Source: Staff analysis. The Thai interest margin of 2.72Z is somewhat above the comparator group aver- age of 2.58Z, while other income (generally fee income) is far lower at 0.59? compared to the group average of 2.7?. Because of possible differences in classification of income into interest and fees (other income), it is better to compare total (interest and other income) margins. This total margin for Thai banks was 3.312 of average total assets compared to 5.28Z for the com- parator group, indicating lower core earnings capacity of Thai banks, amounting to only 63? of the comparator group average. Staff expenses, at 62Z of the comparator group average, indicate a relatively high degree of opera- ting efficiency; this obviously helped profit before income tax but they still - 44 - amounted to only 672 of the comparator group.l71 Income taxes (without any transfer of gross receipts tax) were slightly higher than the comparator group average (and substantially higher if Morocco is excluded from the group). As a re: t, net nominal capital formation for Thai banks was about half that of the comparator group average. 3.36 It is not certain to what extent any reductions in direct and indirect taxes would result in lower intermediation margins, i.e. the benefits would be passed on to depositors and borrowers (as a result of competition in the system) and how much of the *savings* would be retained by the banks to bolster their margins and profitability. in any case, a lowering of fiscal impositions would be a positive development for financial intermediation per se that would also improve the attractiveness of equity investments in banks and improve the formal sector's ability to compete with the informal sector. Form of Capital 3.37 Additional impetus to capital formation could come by giving the banks greater flexibility in choosing the form in which to raise capital. The current definition of acceptable capital may be overly restrictive. The Cor- mercial Banking Act of 2505 (1962) defined Ocapital funds, as the sum of paid up capital, statutory reserves, general reserves (excluding valuation re- serves) and net retained earnings. Some discussion is already occurring on expanding this definition to provide the banks with greater flexibility in raising new capital, and it would be appropriate for BOT to respond construc- tively. 3.38 While the definition of capital should legitimately evolve with changes in the financial instruments available within the market, the key objective of capital should be maintained, namely its 'ability to absorb losses' --it is the source of funding that should remain at greatest risk in case of financial difficulty within the institution. It should thus, by defi- nition, not include any provision for withdrawal or compensation that is in- dependent of the performance of the institution, and should be subordinated to all other claims on the institution except those of other shareholders. It should thus be *permanent' unless the financial condition of the institution is so exceptional as to allow a partial return of invested equity to share- holders without jeopardizing capital adequacy requirements and the interests or confidence of creditors. 3.39 Within these definitional parameters, there is scope for relaxing and expanding the Thai definition of 'capital funds' . It is recommended that BOT favorably consider any requests from the banking community for modification of the definition of capital, including the introduction of the concept of primary and secondary capital, provided that the conditions suggested in July 71 The Thai banks' profit before tax figure is after the gross receipts tax which equaled 0.272 of average total assets. - 45 - 1988 by the Basle Committee 181 are met. This would allow inclusion in primary capital of any category of resources that are permanent in nature, defined as (a) having no provision for repayment of the original funds to the source unless the financial condition of the institution is such that the repayment would not prejudice the interests of any of the institutions credi- tors or cause the institution to violate statutory capital adequacy require- ments, or (b) providing no specific compensation to the source that is not directly tied to the institution's financial performance during the specific time period. As an example, non-cumulative perpetual preferred stock should be acceptable to the authorities (if also to existing common stockholders). 3.40 In addition to primary capital, there is general recognition that some other sources of financial strength may represent secondary capital that should be acceptable, up to a point, in the calculation of capital adequacy. These include: o Revaluation reserves on assets that can be readily liquidated in case of financial difficulty,191 provided that (a) the revaluation re- flects a partial write-up to market where the higher market value is reasonably certain to be permanent according to rigorous tests estab- lished by regulatory guidelines, and not the result of a temporary Obubbleu,201 and (b) the write-up is net of any related capital gains or income tax liability. O Disclosed reserves that are not allocated against specific, iden- tified business risks. For example, general loan loss reserves based on many generally accepted considerations that a certain percentage of the loan portfolio will eventually prove uncollectible may be included up to prescribed limits. Specific loan loss provisions that directly adjust the value of specifically identified troubled credits to market should not be eligible. 181 International Convergence of Capital Measurements and Capital Standards by the Basle Committee on Banking Supervision which meets under the aus- pices of the Bank for International Settlements in Basle, Switzerland. 191 Fixed assets owned by the institution and that are used in the normal course of business are usually essential to the functioning of the insti- tution as a going concern, and thus cannot be liquidated without impair- ing the viability of the institution. Normally revaluation of such assets is not recommended other than to correct significant distortions caused by a major depreciation of the value of the nation's currency in a highly inflationary environment, which has not been the case in Thailand. It should be noted, however, that this is an issue on which there is no universal agreement. 20/ An argument against even such restricted write-ups is that the institu- tion already has the option of selling those assets, thus recognizing the capital gain. - 46 - 3.41 It is also recommended that BOT ask the banks to disclose all hidden or undisclosed reserves and allow their inclusion in capital to the extent that their availability and valuation is verified by independent auditors to the satisfaction of BOT. Along the lines of the Basle Agreement, BOT may also prescribe that the sum of paid in capital, free reserves, and retained earnings (the primary, core or tier one capital) may not be less than a certain percentage of capital brought about by other means. F. Bank Branching 3.42 Annexes 37-38 show the distribution of the 2,061 domestic branches as of December 31, 1988 among the Thai commercial banks by region, in absolute numbers and in percentages. Peer grouping indicates relative size, with the exception of Krung Thai Bank, which is the second largest bank and would fit into PG1 based on size, but has been separately classified because of govern- ment ownership. It shows a greater emphasis on branching outside of the main Region 1, which includes Bangkok, while the smaller banks in PG4 that do not have equivalent economies of scale are also more heavily concentrated in Region 1. This appears entirely reasonable. Unfortunately, this analysis does not indicate the relative value of individual branch locations--this would require a breakdown by branch profitability, which was not available. 3.43 The percentage distribution of each bank's total domestic branch network (Annex 38) provides a more meaningful picture, and shows that the larger, older banks in PG1 have a significantly higher concentration of branches in the more desirable Region 1 than the smaller institutions in PG3 (except for Bangkok Metropolitan Bank). This lends credence to many bankers' complaints that the larger banks operate at a competitive advantage. 3.44 Table 3.11 shows that there is a generally good correlation between PG1 through PG3 banks, performance and the concentration of its branch network in Region 1. PG4 shows inconsistencies, indicating that other factors over- whelm the trend established by the sounder banks. The Bangkok Metropolitan Bank has poor performance despite a very high concentration of branches in Region 1, while First Bangkok City Bank and The Bank of Asia results, as previously mentioned, are overstated by the amount of under-provisioning authorized by the BOT. 3.45 In the absence of management problems, the geographic distribution of branch networks thus has a direct impact on a bank's operating efficiency and on its ability to compete in the market. Until recently banks were required to open branches in uneconomic rural locations before receiving authorization for more desirable city locations. Perhaps not surprisingly, this effort to use the banking system to lead rural economic development, rather than the more business oriented approach of allowing the banks to make their own deci- sions on following business opportunities, may have been a factor contributing to the substandard performance of Bank of Asia, Bangkok Bank of Commerce, Krung Thai Bank and Siam City Bank. This analysis also appears to indicate that the smaller banks do not need help through any policy of favoring the smaller banks on branch approvals. - 47 - Table 3.11s COMPARISON OF PERFORMANCE AND BRANCH DISTRIBUTION Banks by Peer Group Adjusted Z of Branches Return on Assets in Region 1 Peer Groun 1 Bangkok Bank Ltd. 0.46 33 Thai Farmers Bank Ltd. 0.71 39 The Siam Commercial Bank Ltd. 0.78 42 Bank of Ayudhya Ltd. 1.21 49 The Thai Military Bank Ltd. 0.48 32 Peer Group 2 Krung Thai Bank Ltd. 0.03 29 Peer Group 3 The Siam City Bank Ltd. -0.22 30 First Bangkok City Bank Ltd. 1.62 32 Bangkok Metropolitan Bank Ltd. 0.32 52 The Bangkok Bank of Commerce Ltd. 0.12 21 The Bank of Asia Ltd. 1.00 28 Peer Group 4 The Union Bank of Bangkok Ltd. 0.52 46 The Thai Danu Bank Ltd. 0.83 67 Nakornthon Bank Ltd. 0.67 72 The Laem Thong Bank Ltd. -8.92 70 3.46 In 1989, BOT approved a high proportion of new branch applications, much to the surprise of the banking system, which had routinely been inflating its branch licence requests in order to obtain the actual number of approvals wanted. This willingness to allow more branch openings represents a positive step towards allowing the banks to make their own branching decisions--BOT could aid this process by requiring that the banks submit brief branch level loan, deposit and profitability data, possibly on an annual basis, so that consolidated demographic data can be provided to assist the banks in making rational branch opening (or closing) decisions. In any case, the general economic boom is dragging up the non-Bangkok economy, so *rural" branches are becoming more attractive as sources of deposits and loans. In addition, several banks have introduced mobile branches,21/ some even equipped with PC accounting systems, that may be significantly more cost effective in pene- trating rural financial markets than expensive fixed branches. Since banks 211 An example would be mobile branches aimed at tourists that provide for- eign currency conversion, but the concept is equally valid for deposit and lending activity. - 48 - will normally follow business opportunities, the need to force rural branching to ensure general access to the banking system is diminilshing. 3.47 It is recommended, therefore, that BOT authorize banks that meet capital adequacy and liquidity requirements applicable to banks, that are in compliance with all regulatory requirements, that receive satisfactory inspec- tion reports from BOT and that present a business plan including market analysis in form and substance acceptable to BOT, to open branches at their managements' discretion. The BOT should confine its role to ensuring that (a) the bank is making a conscious management decision based on adequate back- ground information on costing and business opportunity (but without second guessing that decision), and (b) the bank is financially able to support the expenditure of resources involved in market entry. The BOT should be informed of the banks' branching decisions in advance so that it can modify its inspec- tion process accordingly. 3.48 An additional source of interference with the branch decision making process takes the form of a requirement that banks wishing to open new branches hold treasury instruments equivalent to 162 of their total deposits. Since August 1989, this requirement has been relaxed so that effectively the banks are now required to hold treasury instruments equal to 122 of their deposits. This requirement appears to have more to do with ensuring that the banks would finance government deficits than as a rational financial perfor- mance related market entry hurdle, and the substantial growth of the banks' deposit base and recent government surpluses have ensured that available treasury debt instruments are inadequate to meet this requirement. The net effect has been an unavoidable erosion of enforcement of this 162 requirement and an undesirable concentration of government domestic debt in the banks. The related aspect of the use of reserve requirements as monetary policy tool was discussed earlier (para. 2.50). The recommendation made there to eliminate this 162 branching reserve recuirement is reiterated. G. Foreign Banks 3.49 As listed in Table 3.1, there are presently fourteen foreign banks operating sixteen branches in Thailand. Represented are major banks from the U.S., Japan, Germany, Hong Kong, Britain and France; the two largest foreign banks are both from Japan. Although the number of foreign banks is almost equal to the number of domestic banks and they were the first to be estab- lished, the market share of foreign banks is small - the fourteen foreign banks taken together account for only 32 of the total assets of the financial system and less than 42 of banking assets compared with domestic banks' 742, as of December 31, 1988. 3.50 The foreign banks' small market share is the result of tight govern- ment restrictions on their activities to maximize the benefits for, and mini- mize the impact on, the domestic banks. The treatment of foreign banks vis-a- vis domestic banks is unequal in three important respectss (a) foreign banks are restricted from generating deposits by a prohibition on branching (only two grandfathered sub-branches exist), (b) their tax rates are higher than for domestic banks, since the domestic banks' stocks are quoted on the SET, giving eligibility for the 30t income tax rate, whereas the foreign banks are subject - 49 - to the 3SSZ income tax on non-quoted companies, and (c) they pay a 16.5Z witn- holding tax on dividends transferred overseas. 3.51 The foreign banks focus mainly on financing international trade transactions where their branch networks in Thailand's trading partner countries allow them to compete on fees while obtaining fee income from both ends of the transaction. While they are active in the ioreign exchange market, most of their business relates to trade transactions of their own clients, and the need to cover the steady foreign exchange losses on their unhedged foreign borrowings obliges them not to compete too aggressively with the domestic banks on spreads. 3.52 The unequal treatment of foreign banks mentioned in para. 3.50 above restricts the foreign banks' ability to compete with the domestic banks for business from the major economic groups. Such competition could (a) lessen the interdependence between these groups and their related bank, thus reducing loan concentrations and forcing the banks to diversify their borrowing client base, and (b) help strengthen accounting standards through the imposition of more rigorous requirements on large corporate borrowers. At this stage in the evolution of the Thai financial system, both should be considered positive steps. 3.53 The BOT should expect the foreign banks to behave like good neigh- bors. In many countries, the high level of professionalism and the wish to avoid adverse publicity ensures that the major foreign banks often behave more responsibly than domestic institutions. Since they are not beholden to any domestic industrial or trade group, they may also be more willing than domestic banks to support capable but unconnected entrepreneurs (without equity kickers), thus expanding the country's economic diversity. They take a leading role in the development of rigorous accounting and lending standards, public disclosure of banking terms and conditions, and introduction of new products and services. Given these factors, the Thai authorities are to be applauded for considering authorizing new foreign banks to operate in Thailand. However, benefits to the country of the presence of foreign banks are not being maximized because of the authorities' continued reluctance to allow foreign banks to competitively challenge domestic institutions on a level playing field. The domestic banks should not need the *protection* of restrictions on foreign banks' activities and unequal tax treatment, since such protection perpetuates complacency on the part of the protected. While a foreign bank will wish to operate its Thai operations at a profit, and that it will attempt to transfer such profit back to its parent organization, the current restrictions on their activities constitute an unnecessary hindrance to a free and competitive market. It is recommended, therefore, that the authorities place the foreign banks on a level playing field with domestic institutions bys (a) recognizing the quotation of their shares on foreign stock exchanges as achieving the same objective of public access to stock ownership as quotation on SET of domestic shares - thus allowing a 30X tax rate; (b) allowing them to open branches under the same terms and conditions as applied to domestic banks (see also paras. 3.48), and (c) removing the double taxation of dividends. These measures when implemented can be expected to lead to system wide improvements in cost-efficiencies of banking services - 50 _ in the country, although it is recognized that this may involve some painful adjustments by some of the weaker institutions. H. Staffing and Trainina Staffing 3.54 Personnel policies have generally followed in-house training and life-time employment. However, the recent surge in the growth rate of the Thai financial system and the need for weaker institutions to upgrade exper- tise is creating internal and market demands for staff that exceed past hiring and training plans. As a result, some of the larger banks 221 are experien- cing increasing turnover rates in management staff, while weaker institutions are having difficulty attracting good quality staff. This is driving up middle management salary levels. In addition, possible licensing of nine new foreign banks is likely to further tighten the labor market. This shortage of 'human capital may be as serious as the shortage of financial capital (see paras. 3.29-3.42), and represents a potentially critical constraint on the ability of the Thai banking system, and particularly of its weaker members, to keep up with market growth. To some extent this problem is being addressed, at the clerical level, through a high level of investment in automation. But managing this form of institutional change if anything increases the demands on the management level staff. The government may wish to consider relaxing work permit rules for expatriates as an effective and quick solution to the staff shortage problem. In the longer term, staff training facilities need to be substantially strengthened and enlarged. Training 3.55 Internal Training. This problem is aggravated by the traditional approach of a relatively long learning curve for managers, in which expertise is passed down through lengthy rotational assignments, individual training and osmosis rather than through the application of standaraized, codified policies and procedures. As a result, many of the banks have only limited ability to accelerate formation of management staff by increasing reliance on written procedures to complement basic training. 'While the major banks have training centers and the capacity to expand throughput over time, this will not solve the short term problem and particularly the smaller institutions should be expected to have increasing difficulty in this area, increasing the risk of mismanagement among the weaker institutions and a concentration of resources among the larger institutions. 3.56 External Training. The Thai Bankers Association has been providing a two year evening course (90 minutes every week-night) with some 50 students per class and some 1,000 graduates. This course is at the undergraduate level and is relatively basic. It was opened to foreign banks since demand from the domestic banks was inadequate--Thai employees of the foreign banks now account for some 502 of students, and these are generally better motivated than their 221 Apparently particularly the Bangkok Bank and the Thai Farmers Bank. - 51 - counterparts from the Thai banks. There is an effort to develop a Thai Institute of Banking examination process modeled on the British system, but this ia still in the formative stage. Clearly, an opportunity exists for strengthening of the role of outside training targeted at those institutions that have less internal training capacity, particularly by expanding into middle and senior management training. Some assistance, possibly through foreign sources, may be appropriate in developing a program that targets the brightest and most promising candidates. Such a program should provide enough value to justify paid leaves of absence from the staff's normal bank duties for off-site total immersion programs where rapid progress can be achieved and measured. - 52 - IV. THE FINANCE AND SECURITIES COMPANIES 4.1 The non-commercial bank financial institutions in the banking sector include Finance and Securities Companies authorized to conduct both finance company and securities business, Securities Companies engaged solely in secu- rities trading, Credit Foncier Companies which are predominantly involved in builder finance of real estate developments and the Government Savings Bank. This Study deals with the Finance and Securities Companies (F&S companies). Apart from analyzing their operational performance and financial condition, the major issues discussed relate to their role in the financial system and the present restrictions on their ownership. A. Oriain and Evolution of Finance and Securities Companies Orizin and Lesal Framework 4.2 The F&S companies generally originated in one of two ways: o As affiliates of commercial banks, in some cases with the participa- tion of foreign financial institutions, established to provide ser- vices that the parent bank could not generally provide directly; and • As formalized small institutions engaged in quasi banking activities, generally high margin, high risk consumer finance. 4.3 As of December 31, 1988, 93 F&S companies were authorized to under- take finance company business, almost three quarters of which were also autho- rized to engage in securities company business;231 11 Securities Companies were authorized to undertake securities company business only; and 19 Credit Foncier companies engaged predominantly in the finance of real estate develop- ment. The operations of these non-bank financial institutions are governed by the Act on the Undertakina of Finance Business. Securities Business and Credit Foncier Business, dated May 8, B.E. 2522 (1979), as amended by the EmerQencv Decree of B.E. 2526 effective December 16, B.E. 2526 (1983) and the Second Emergency Decree of B.E. 2528 effective November 27, B.E. 2528 (1985). 4.4 The F&S companies may engage in the activities set out in the Act on the undertaking of Finance Business, Securities Business and Credit Foncier Business ("F,S & CFO) Section 4: 'Finance business' means the business of procuring funds and using such funds for any kind of business operations, which may be categorized as follows: 231 There is some confusion whether there were 93 or 94 F&S companies as of December 31, 1988. The analysis in this chapter is based on the data on 93 companies whose names and on which detailed data are given in Annexes 39-42. Conclusions and recommendations would remain the same even if one firm (which must be very small) has been missed. - 53 - (1) Business of finance for comuerce; (2) Business of finance for development; (3) Business of finance for disposition and consumption; (4) Business of finance for housing; (5) Other types of finance business as prescribed in ministerial regulation. Securities companies may operate as separate legal entities or as departments of finance companies, and may perform the following activities at set out in F, S & CP Section 4s "Securities business" means securities business of the following typess (1) Business of brokerage for buying and selling of securities; (2) Business of trading in securities; (3) Business of providing investment advice; (4) Business of managing sale of securities; (5) Business of investment management; (6) Other types of business concerning securities, as prescribed in ministerial regulations." Thus the non-credit services generally provided by stock brokerage companies and in part by merchant and investment banks in other countries are provided by F&S companies in Thailand. 4.5 The F&S companies come under the supervisory authority of BOT, as do banks, but are subject to a separate legal framework and are prohibited from engaging in foreign exchange transactions, from offering checking accounts and from opening branches. Funding is obtained through the issuance of promissory notes under the Thai Civil and Commercial Code. The F&S companies account for 13.3Z of the banking system assets. 4.6 An F&S company is required to have at least 100 shareholders, with no single shareholder holding more than lOZ,241 and at least 75S of shares and board seats must be held by Thai nationals. The company is required to verify its lists of registered shareholders and notify BOT if these limits exceed, notify the shareholders involved that they must divest, and refrain from paying dividends or allowing voting rights on any shareholding in excess of these limits. Evolution of F&S Company Business 4.7 The specialized characteristics of S&F companies have evolved over a period of time and reflect the historical development of the Thai financial system. Hire purchase--principally automobile finance--has become significant in recent years as a result of general economic prosperity. This form of finance requires a statistical approach, high margins to reflect the 24! Unless specifically exempted by BOT (which is usually the result of a financial support operation). - 54 - relatively high default rate, and an effective collateral repossession process. Companies engaged in hire purchase require highly specialized staff and procedures, which may not easily fit within the general staffing and organizational framework of a bank. In the absence of a reliable individual credit history and rating service in Thailand, this business is profitable only if effective annual interest rates are high. In Thailand, interest on hire-purchase transactions is calculated as an add on to the original amount, s0 12? p.a. nominal translates into some 21.52 p.a. This does not violate the 152 limitation on interest rates contained in the commercial code, since the code does not specify the method of calculation. 4.8 Corporate finance services appear to have been developed to bypass the cost of bark intermediation, very much like the inter-company financial markets development in other countries when large prime companies felt they could bypass the banking system. This money brokering activity by definition focussed on prime risk borrowers, but has now been displaced by improvements in the banks' treasury management services. Next, a window of opportunity arose in loan syndication, since the banks were not inclined to cooperate directly with each other in syndicated credits. The leading finance companies undertook analysis, advisory and packaging functions and brought the debt issues to market, often taking ur a small share for their own account and syn- dicating the rest to banks. As the banks responded to this competition by themselves engaging in loan syndication, the nature of F&S companies' corporate finance business progressed from debt into equity involving merchant banking and venture capital like functions. 4.9 Stock brokerage activities evolved in some F&S companies as a logical extension of the corporate finance activities. Given the limited size and variable activity of the stock market, independent stock brokerage operations were difficult to cost justify until the recent boom. The ability to switch staff between different business lines was quoted as the major reason for re- taining this activity within the existing F&S companies. 4.10 The more dynamic F&S companies have thus represented the leading edge of product development in the Thai financial system, and since they are typi- cally smaller and more efficient than the banks, they could attract some of the best financial talent available. However, given the number of players in- volved, the thin market, and the propensity of the banks to introduce competing services once these have been proven successful at the S&F company level, the service mix provided by these S&F companies has shifted, with ex- isting practices and margins constantly under pressure. B. ODerational Performance and Financial Condition Operational Performance 4.11 Annexes 39-42 provide financial data for the F&S companies for 1988 plus ratio analyses grouped by size of institution, capital adequacy and return on assets. This data shows a wide disparity between institutions within the market, and is summarized in Table 4.1. On average, there is a di- rect correlation between capital adequacy and return on assets. - 55 - Table 4.1s PERFORMANCE & STRENGTH OF F&S COMPANIES, DECEMBER 31, 1988 Group Number Avg Total Year End Staffing To Return on Avg Assets Equity Tot Assets Total Assets -----(billion baht)---- (Z) (?) By Equity to Total Average Assets 8? + 57 86.7 12.58 0.8 1.7 7-7.9? $ 23.1 1.98 0.7 1.4 6-6.9Z 12 28.0 2.01 0.7 0.9 0-5.9? 6 13.0 0.44 0.6 -1.6 Negative 10 16.3 -1.83 0.4 -4.3 By Return on Equity 2? + 24 45.0 6.33 1.1 3.2 1-1.9? 24 51.6 4.95 0.7 1.5 0-0.9? 23 31.0 2.92 0.6 0.4 Loss 22 39.6 0.97 0.4 -3.1 Total 93 167.2 15.18 0.7 0.7 4.12 As shown in Table 4.1 by the 1.7Z return on assets of the 57 F&S com- panies that have capital adequacy in excess of 8?, it would appear that a well structured and well managed F&S company can substantially out-perform a com- mercial bank. Unfortunately, the blending of finance company and securities brokerage activities in many of these companies prevents meaningful analysis beyond identifying institutions that are clearly failing. For example, the most successful F&S companies have, on average, the highest staffing expense to total average assets - yet this is more indicative of a greater concentra- tion in the securities business than on any direct relationship between resources applied and earnings capacity. 4.13 Im2act of Business Focus. Analysis of the business focus of F&S com- panies shows that 58? of gross earnings came from lending business, while 13? came from hire purchase business, 12? from securities trading, 10? from divi- dends on investments and 7? from other sources. Table 4.2 shows that the institutions that have a dominant focus on income derived from lending have, on average, lower staffing costs to to'.al average assets than the banks (0.7? versus 1.02), and have higher returns on assets (0.8? versus 0.6?), and thus appear somewhat more efficient than the banks. Table 4.2 also shows the hard- ly surprising fact that hire-purchase business is staff intensive but provides relatively high returns on assets. While all F&S companies have at least some income from lending, 65 (or 70? of institutions) showed dependence for over - 56 - 5O0 of gross income on lending business, 12 (or 13? of institutions) showed dependence on hire-purchase business, while 4 showed dependence on returns from investments which are inherently volatile (all 4 showed substantial losses in 1988), and 12 institutions showed no clear business focus, although two had dependence on other income for half of their total income. There is also no discernible correlation between business focus and profitability as measured by return on equity. The quality of management seems to be the most important factor, with the lack of a clearly defined focus on one of the three core businesses indicating lack of direction and related poor results. Table 4.2: BUSINESS FOCUS AND PROFITABILITY OF F&S COMPANIES, DECEMBER 31, 1988 Number Avg Total Staffing Return on Avg Assets To Tot Assets Total Assets Business Focus /a (billion baht) (1) (S) Lending 65 125.8 0.7 0.8 Hire-Purchase 12 15.9 1.3 2.1 investment in Securities 4 4.4 0.4 -6.1 No Focus 12 21.1 0.7 0.3 /a Business focus means more than half gross earnings were derived from the stated activity. 4.14 Impact of 'Blendinx" Institutions. The balance sheet and ratio analysis of a finance company should in many respects be similar to that of a bank. A stock brokerage company, on the other hand, should not be receiving client funds beyond those directly related to current stock purchases or sales, and should not be engaging in general lending business in any signifi- cant fashion. Its balance sheet should thus look more like that of a service company in any other industry, with limited gearing ratios and performance measured more in terms of return on sales than on assets. Clearly, an F&S company that is predominantly involved in the securities business yet that displays a balance sheet and ratios more appropriate for a finance company is a cause for concern. 4.15 Imoact of Bank Affiliation. Twenty-six F&S companies are affiliated with private Thai commercial banks, and a further 12 with the government owned Krung Thai Bank as part of the April 4, 1986 rescue effort. Table 4.3 examines the comparative performance of the three groups by ownership. Clear- ly, affiliation with a strong private bank (as shown in Annex 41, page 6) gen- erally translates into solid performance, although 17 unaffiliated F&S compa- nies with average total assets of B 18.7 billion had returns on assets of 22 or more in 1988, indicating that affiliation with a bank is not essential for good financial performance. The negative performance of F&S companies af- filiated with Irung Thai Bank is no surprise; their bad financial condition - 57 - and performance in the first instance was the reason they had to be rescued and became affiliated with Krung Thai Bank. 4.16 Sectoral Diversification of Portfolio. The portfolio of F&S com- panies remains well diversified sectorally. The top three sectors at the end of 1988 were the same as at the end of 1980, although the ranking between these three sectnrs did change. At the end of 1980, manufacturing accounted for 262 of the total portfolio, followed by personal consumption and construc- tion and real estate with share of 17? and 142 respectively; at the end of 1988, loans for personal consumption accounted for the highest share (25ZY, followed by manufacturing (21X) and construction and real estate (18?). Sec- toral diversification of F&S companies' portfolio is given in Annex 43. Table 4.3s COMPARATIVE PERFORHANCE OF F&S COMPANIES, 1988 Equity to Return on Number Avg. Total Assets -----Total Assets----- Group No. ? Amount X Z Z (bil.baht) Bank affiliated 26 28 80.9 48 9.7 1.6 Krung Thai affiliated 12 13 23.3 14 -5.0 -3.3 Unaffiliated 55 59 63.0 38 10.8 0.9 Total 93 100 167.2 100 8.2 0.7 4.17 Performance of Securities Companies. Of the eleven pure securities companies, seven had healthy returns on equity in 1988 while two had minimal returns and two losses, of which one lost 1.25 times its average equity, moving it closer to insolvency. The conclusion is that high earnings are pos- sible for a well managed securities company in a relatively favorable environ- ment; the risks are also high, underlining the need for segregation of the securities business from that of banks and finance companies (see also para. 4.29). Troubled F&S Companies 4.18 If BOT capital adequacy requirement for F&S companies (62) is used as the cut off point, there would be 22 companies which would be classified as troubled. However, if a more conservative 8? capital adequacy criteria is used, and assuming that loan loss provisions have been appropriate and that equity is fairly stated, then 34 F&S companies would be judged as troubled, - 58 - representing 39S by number but accounting for 482 of total average assets.25/ These F&S companies can be classified into three groups, as follows: Group 1 Equity less than 8? of total average assets but return on equity above the inflation rate, thus a positive growth in equity base. These institutions may be able to earn their way out of difficulty or attract new equity. This group should not be considered in serious trouble. In fact, it includes some of the better known F&S companies in Thailand. One may even argue that if an 8X capital requirement had been mandatory, many F&S companies in this group would be able to manage it over time. Group 2 Equity less than 8? of total average assets, still positive but eroding. Continued operation with unchanged trend lines .ikely to lead to failure. Group 3 Negative equity, thus insolvent, and for the most part with negative earnings that will cause further losses to creditors, since with loss of equity the shareholders' funds have already been exhausted. 4.19 Table 4.4 summarizes F&S companies into these three categories. As shown by the details in Annex 41 (page 5), the last three F&S companies in group 2 had losses in 1988 significantly higher than their total remaining equity at year end--if these losses are repeated in 1989, then they would move from group 2 to group 3. Table 4.4: TROUBLED F&S COMPANIES, DECEMBER 31, 1988 Group Number Avg. Total Equity to Net Income Assets Total Assets After Tax (billion baht) z (billion baht) 1 16 44.9 7.0 0.63 2 8 19.2 4.2 -0.27 3 10 16.3 -10.9 -0.71 Total 34 80.4 3.0 -0.35 4.20 The continued functioning of the group 3 institutions in 1988 added B 706 million to the losses that must ultimately be borne by creditors, (or by BOT should it decide to bail them out). In addition, the continued existence of so many institutions whose balance sheets clearly show their precarious 25/ Since loan loss provisions are unlikely to be adequate in all cases, given BOT's past flexibility on requiring immediate recognition of asset values, the actual number of troubled F&S companies is probably higher. - 59 - financial condition could be destabilizing to the Thai financial system and could lead to a *flight to quality' (often equated with size), thus impairing the ability of even those smaller institutions which are soundly managed, and encouraging further concentration of economic resources. 4.21 Management Problem of Family Controlled F&S Companies. While most oz the 45 'affiliated' F&S companies have access to experienced and capable man- agement, possibly even among the best available to their related banks, the 48 independent institutions face increasingly serious difficulties in obtaining capable management. This is due in part to the close family control that blocks the career path of potential management candidates in small companies. and because of the increasing need of the major institutions to attract the best available candidates which is already causing increases in middle and junior management salary levels. While some of these institutions (and some of the banks) seek to grow their own management through extensive training of family scions, institutions that limit themselves to the resources of the family may have difficulty competing effectively over the long term with institutions that draw their human resources from the larger general labor pool. The BOT should expect continuing failures among these small, indepen- dent institutions, if only because of inadequate management resources. 4.22 Treatment of Failina Institutions. Consolidation among the large number of F&S companies may be advisable, particularly in view of the already apparent shorta,64 in management staff. However, any decisions to consolidate should be a management prerogative based on a business analysis of the costs and benefits involved, although BOT should exercise its responsibility towards the financial system in general by closing down or merging failing institu- tions that also fail to develop a rational action plan to deal with such failure. The BOT should ensure that no avoidable administrative or legal impediments exist to this process. As indicated above, however, there are 34 companies which are in weak financial condition, including ten which have negative returns on assets and equity. In respect of these companies, BOT should seriously consider arranging mergersiacquisitions; alternatively these companies should be closed. C. The Role of F&S Comnanies Functional Demarcation 4.23 The functions that F&S companies can and cannot perform were stated in para. 4.4. The Thai banks are also barred from engaging in certain activities (e.g., banks cannot undertake hire-purchase business although they may lend to companies that engage in hire-purchase business). As implied in para. 4.2, this functional demarcation began when banks started to establish F&S companies for activities that they could not undertake themselves as banks. Since many such services form part of normal banking services in other countries, the F&S companies, to a significant extent, owe their origin to artificial constraints on the Thai commercial banks' activities. Later, the continued functional segregation was justified in terms of preserving certain profitable businesses for F&S companies to ensure a base for overall opera- tional profitability, and to ensure diversity within the Thai fianmcial market. These arguments have lost much of their validity, if they ever were - 60 - valid, because of the tremendous growth in the volume of banking and near banking business which has also enabled the growth in the size of financial institutions; some F&S companies are now larger than the small commercial banks. 4.24 Considerable debate is currently occurring in Thailand on whether the legally mandated functional segregation between commercial banks and F&S companies is appropriate or, more generally, what should be the role of F&S companies in the financial system and particularly in relation to the commer- cial banks. As is generally known and discussed in detail in para. 4.38, many F&S companies are *affiliates (more appropriately recognized as subsidiaries) cf banks. As a practical matter, therefore, a captive F&S company can conduct any financial activity that a bank may not engage in. In that sense, the issue of functional demarcation is moot. It is examined here from the point of view of rationalization of the institutional structure of the Thai finan- cial system to achieve operational efficiencies and enhance competition, rather than in terms of functionaliy specialized institutional structure ver- sus universal banking. Distinction Between Banks and Non-Banks 4.25 The elements that differentiate a non-bank financial institution from a bank should be the key issue in this debate. A common distinction is based on the degree of vulnerability to runs on deposits. Demand deposits that allow for immediate, without notice, withdrawal represent the highest risk funding source for a financial intermediary, particularly if economic shocks occur within the financial system. Institutions that accept such demand deposits are rightly subject to more rigorous prudential supervision measures, including a requirement to maintain liquidity reserves, than institutions that take only the more stable term and savings deposits. Another important aspect of this distinction is the potential impact on unrelated, non-client holders in due course of checks (or other instruments) that have been used by an account holder as a medium of payment, should the institution (as opposed to the drawer) be unable to cover checks presented for payment. Dishonor of pay- ment media spreads the impact of an insolvency beyond the institution's direct client base, and thus increases the impact on the market and economy at large. The rationale underlying this distinction based on funding risk is reflected in the existing restrictions on deposit taking by the F&S companies, which seek to ensure that the funds raised by them represent savings (even if short term) rather than transitory funds circulating in the payment system. Differentiation on this basis appears entirely appropriate. This is consis- tent with practices in other countries, where nonbanks generally fund them- selves with stable savings and time deposits. 4.26 A secondary distinction may also be made in terms of product special- ization: a finance company may focus on a narrowly defined service range using higher priced staff and lower levels of internal controls than are feasible within the general purpose structure of a bank. This kind of dif- ferentiation is based on economics and rational business objectives (rather than on risk considerations) and should be left to be made by the managements of the financial institutions comprising the financial system; it should not be made for them by regulation. - 61 - 4.27 Another distinction may be made in terms of business (as opposed to funding) risk considerations. There is merit in a legal separation of certain high risk business activities into a separate legal entity (a subsidiary) in order to protect the parent from the impact of losses. In practice, such a risk control separation may have limited effect. An F&S company must be financially sound to compete for funding from the public. If its business focus involves high risk, its providers of funds may be required to accept a level of risk that the parent bank is unwilling to take itself (thus the justification for the separation); they will require risk premia, which will raise the institution's cost of funds and may make its business uneconomic. If the parent provides the resources itself through direct funding, or is implicated through issuance of formal guarantees or indirectly through manage- ment control, then the need to protect its reputation may increase the actual risk far beyond its equity investment, thus negating the value of the separa- tion. 4.28 In countries where interest rates are administratively controlled, non-banks have been used to circumvent regulatory restrictions to provide higher risk financing than the parent bank can economically justify within its interest rate, liquidity and capital adequacy constraints. Non-banks that are used primarily to circumvent regulatory restrictions on the commercial banks provide uncertain economic benefit to the country, since their business focus may be more the result of distortions in the environment, rather than of any underlying business and management efficiency considerations. To the extent that F&S companies were created in Thailand because the commercial banks had to operate under mandated interest rate ceilings, the result has been artifi- cial institutional fragmentation. As recommended elsewhere (para. 2.39), interest rate ceilings should be eliminated. There is no justification for functional specialization of institutions if it is based on factors such as mandated interest rate ceilings. Separation of Securities Com_anies Business 4.29 Certain activities may Justifiably be segregated into separate legal entities based on transparency and ease of supervision, the potential for con- flicts of interest were those activities handled within the parent bank, or where no synergies exist with the other normal business of the institution. These considerations argue for a separation of securities related business from the other business of a typical F&S company: securities related business should normally come under the supervision of a stock exch-nge regulatory agency rather than under bank supervisors, no synergy seems evident between stock brokerage and hire-purchase activities, for example, and a risk of conflicts of interest exists between securities related business and credit risk management functions. Additionally, as indicated in para. 4.14, securities related activity requires a different balance sheet structure than that of a banking type institution, and deficiencies in the management of the securities business that would become evident in a separate organization may be masked by consolidation with a finance company. 4.30 The booming stock market has enabled the securities business activities to grow and to become very profitable. This has increased depen- - 62 - deuce on brokerage business of some 34 F&S companies; they now derive more than 101 of their total income from stock trading. But it has also increased the risk that 'deposits' from the public might be invested in speculative stock purchases.261 In effect, purchasers of promissory notes ("depositors*) could unwittingly invest in a form of 'mutual fund" managed by the issuing F&S company without the upside potential of successful investments, but with the downside risk in case the institution should suffer losses on its portfolio and fail. While a soundly managed institution will generally avoid betting its depositors' funds, the potential gain could be tempting. 4.31 A securities business company cannot easily avoid dealing in any given issue of shares where it may have a conflict of interest, since it must respond to externally generated client buy and sell orders. The initiative taken by the Thai authorities to separate the debt and equity functions of the existing F&S companies correctly responds to the issues involving supervision, non-synergy and conflict of interest indicated above. Stock trading is deci- dedly distinct from the debt or credit business that forms the core of banking and finance business. The size of stock brokerage business has also grown very rapidly in the last few years to a level where it should now be con- sidered a viable stand alone business without the support of finance company business. It is recommended, therefore, that BOT not relent on its initiative to segregate the businesses of finance and securities companies. The existing F&S companies should be given a choice of opting for either the finance compa- ny business or the securities company business. A reasonable period should be allowed to phase out one line of business or, alternatively, to reorganize the combined business into two separate legal entities, possibly using the device of a holding company, as recommended in para. 4.44. If a company is currently engaged in one kind of business but v1shes also to undertake the other business, the same rules should apply. In other words, market entry and exit rules should be uniform and applied consistently. 4.32 Merchant Bankint Business. A more difficult issue relates to the in- vestment/merchant banking activities of F&S companies, such as underwriting new issues. There is obvious potential for conflict of interest if an under- writer has an existing creditor relationship with the firm whose issue is being underwritten. On the other hand, it can be argued that the existing creditor relationship provides comfort to the underwriter. On balance, this Study concludes that the objective of avoiding conflict of interest should override other considerations. It is recommended, therefore, that an under- writer should avoid involvement (by disqualifying itself) in any issue in which it has pre-existing credit exposure to the issuer, and thus avoid a conflict of interest, while at the same time the new issue underwriting function should be retained as part of the authorized merchant banking activities of a finance company. It is further recommended that BOT should institute new regulations to prevent a finance company (a) from underwriting an issue for a client where a prior debt, equity or management relationship exists, (b) from establishing a debt relationship with a client where an 261 Although there is certainly a general market understanding that this is not prudent, institutions that are in financial difficulty have been known to gamble on imprudent activities that promise a high return. - 63 - underwriting operation was undertaken within a prescribed period of timel and (c) to introduce adequate capital adequacy requirements based on the degree of risk involved in this type of transaction. If the recomendation to bar a finance company from underwriting a new issue where a creditor relationship pre-exists is considered too drastic, BOT should, as a minimum, require full disclosure of the creditor relationship in the documentation used for the underwriting operation. The BOT may also consider requiring that all stock temporarily acquired during a stock issuance must be sold on the open market or through private placement within a specific, limited period of time so that no speculative positions can be taken. Foreign Exchange Business 4.33 The F&S companies are not permitted to engage in foreign exchange business. This reflects the original role of F&S companies as small, special purpose, limited service institutions that did not have the economies of scale, specialized staff resources, and internal sophistication in policies and procedures required to manage and control the risks involved in trade documentation and foreign exchange exposure. At this stage in the development of the Thai financial system, some of the professionally managed finance companies appear at least as capable as many of the banks, and significantly more capable than some of the weaker banks. However, the Thai foreign exchange market is thin, focussing on the settlement of trade transactions with very little trading or hedging of currencies. Recent experience of some banks thst have tried to become more aggressive in this area has not been en- couraging (Sayam Bank, for example), yet each bank is trying to improve its market share of this highly lucrative business in direct competition with the foreign banks, with the major emphasis being on securing export clients that will assure a steady stream of foreign currency inflows. It is debatable whether a finance company could achieve an adequate market penetration without a full trade documentation capability, and whether such a capability can be cost effectively introduced in many of the present F&S companies. On balance, it is recommended that healthy finance companies that meet BOT's criteria should be authorized to engage in foreign exchange activity, subject to with- drawal of such authority if BOT later determined that the appropriate skills, policies, procedures and controls were not being developed in a timely fashion. Two factors underlie this recommendation (a) participation in for- eign exchange business in an open market should be an unregulated internal management decision of the financial sector institutions within the framework of a uniform regulatory environment that ensures that no segment of the finan- cial market receives an unfair competitive advantage, and (b) finance companies have been the innovators in the financial system, have the demons- trated capability to attract the most competent staff and, therefore, can be expected to provide effective competition to domestic and foreign banks in this segment of banking business. Branching 4.34 The F&S companies are restricted from establishing branches except for those that were established before the introduction of this restriction. In practice, however, marketing companies and commissioned sales agents have sprung up in remote sites to provide access to F&S company services, effec- - 64 - tively circumventing the restriction on branching. While the restriction on branching may make sense for small, unprofessionally managed institutions, many of the F&S companies have now evolved into competent institutions that are providing valuable services, but most effectively to those who have ready access to the institution's offices. Since these institutions are located in Bangkok, the restriction on branching acts to discriminate against those living outside of Bangkok. This does not seem in the national interest; in fact it is against stated policies and priorities. It is recommended, therefore, that the F&S companies that are in compliance with all regulatory requirements, that receive satisfactory inspection reports from BOT and that present a business plan including market analysis in form and substance acceptable to BOT, should be authurized to open branches as their managements decide. 4.35 In Thailand, nonbanks fund themselves through the issuance of prom- issory notes. Where these are payable at call in small denominations, they have the liquidity characteristics of demand deposit accounts but not the risk to holders in due course.27/ While a check may be presented through the in- termediary of a holdir in due course, a promissory note must be physically presented to the finance company itself before payment is received. These promissory notes thus have the character of bank savings depos'its and applica- tion of the principle of a level playing field would require that such promis- sory notes be subject to the same liquidity and reserve requirements as apply to bank savings deposits. The issuance of promissory notes under the commer- cial code represents an anomaly resulting from the blending of Napoleonic code concepts into a predominantly common law framework. In effect, the use of prom.ssory notes achieves the same general objectives as would be served by the use of bank-type savings accounts, except that dishonor subjects manage- ment to criminal prosecution.28/ D. The Issue of Ownership of F&S Companies Restrictions on Ownershil of F&S Comoanies 4.36 Increased dispersion of economic power represents a central socio- political objective in many developing countries. One of the principal chosen vehicles is normally a restriction on the extent of ownership of wealth or a requirement for wide distribution of share ownership of joint stock companies. In Thailand, in so far as the financial system is concerned, these objectives are supposed to be achieved, inter-alia, by the application of the law which prohibits financial institutions from owning more than lOZ of the stock of 27/ This assumes that these promissory notee are not negotiable, or at least not negotiated. 28/ It should be noted that one important benefit to the use of promissory notes is that such use alerts a partially informed depositor that he is not dealing with a full bank. - 65 - other fianmcial institutions.291 This legal restriction on the extent of own- ership by one financial institution in other financial institutions has resulted in a complex network of multiple and cross ownerships between com- mercial banks and S&F companies. The question is whether this restriction is now proving to be counter productive. 4.37 As stated in para. 4.2, the origin of F&S companies principally lay in the restrictions placed on commercial banks. The commercial banks were in a way forced to create separate S&F companies to be able to undertake func- tions that they could not undertake as banks. Full control of S&F companies by the commercial banks establishing them was, therefore, understandable and .eely acknowledged. Ownership was another matter. Because of the 102 legal restriction, the S&F companies were established as "affiliates' but, to retain vll control, it became necessary to create stock holding relationships t rough relatives, shareholders of the "parent" bank, friends, officers of tJ), owned/controlled institutions, and cross ownerships between group tr.t%tutions. These cross relationships have been so effectively created and maintained that the 1O0 restriction has been reduced to legal fiction and it is generally acknowledged as such. The extent of commercial banks' control/ownership of S&F companies is shoun in Table 4.5. It shows that domestic banks effectively control 45 of tLe 93 F&S companies.30/ These affiliated institutions account for 65Z of total assets and 702 of total borrowings of all S&F companies. The table also shows that the larger banks have multiple affiliated F&S companies. These bank affiliated S&F companies engage in specialized businesses that either do not fit cleanly into the related bank's structure, or that the parent bank is barred from undertaking by prevailing regulations. 29/ Although this restriction is waived by BOT in special situations, for example to facilitate a bail out of a failing S&F companies. 30/ The 13 affiliates of Krung Thai Bank represent the failed institutions that were taken over on behalf of the Bank of Thailand during the crisis in 1986, and are referred to as the April 4 group (see para. 6.48). - 66 - Table 4.5: S&F COMPANY AFFILIATION WITH THAI BANKS, December 31, 1988 Bank Affiliated S&F Companies Krung Thai Bank Ltd 13 la Thai Farmers Bank Ltd 7 Bangkok Bank Ltd. 5 Bangkok Metropolitan Bank Ltd. 4 Siam Commercial Bank Ltd. 4 Thai Military Bank Ltd. 4 Bank of Asia Ltd. 2 Siam City Bank Ltd. 2 Bank of Ayudhya Ltd. 1 Nokornthon "ink Ltd. 1 First Bangkok City Bank Ltd. 1 Thai Danu Bank Ltd. 1 Total 45 Independent 78 .la Include twelve that were acquired as a result of rescue operations. Sources BOT. 4.39 The Thai authorities implicitly recognize the problem of continued group control over the ownership of F&S companies' stock by holding the latter's managements responsible for ensuring that no single group owns more than the allowable 102 of their companies' stock. However, where the stock is listed on the stock exchange, managements have no direct control over trades in their institution's shares, and thus on stockholder concentrations, and the requirement becomes impossible to observe. 4.40 The restrictions on direct investment by the parent bank limit the pass through of a captive F&S company's earnings to the parent bank to 102 (except where the 102 rule has been waived by BOT). However, if an F&S company becomes troubled, the parent may be implicated through its management control and thus obligated to absorb 1002 of any losses incurred. The restriction on ownership in F&S companies, therefore, acts to reduce the benefit without a commensurate reduction in the risks to the parent bank, and thus may act to weaken the parent bank. 4.41 The practical impact of the legislative restrictions on ownerships has been to hide cross-ownership and connected lending beyond what would have been necessary without such legislation. Expressed in another way, the result of this legal restriction has been a lack of transparency. On balance, such legislation is doing more harm than good. Increased dispersion of economic power and isolation of any problems within the financial system are understan- - 67 - dable and desirable objectives, but achieving these in Thailand is unlikely through legislative restrictions, since these are too easily circumvented. 4.42 This Study suggests two alternative methods to oval 'With the problem. One would work within the present legal framework while the other is based on the holding company concept and might require some changes in the basic legal framework. Both alternatives are based on the fundamental prin- ciple that higher or uncertain risk activities should be limited and segregated in such a way that any significant related losses do not impair the viability of a bank. This and other considerations were discussed earlier in para6. 4.25-4.28. 4.43 Under the first alternatives, banks would be required to segregate all significant (as defined by BOT) financial intermediation activities that entail above normal risks compared to a bank's usual, prudent lending activity into separate legal entities. The bank's investment(s) in such separate legal entities would then be governed by the following: o The equity invested by a bank in any financial subsidiary or affiliates should be considered as no longer available to the bank to support its own financial position, and thus should be deducted from its own equity for capital adequacy purposes.31/ o The lOS limit on bank ownership of nonbanks should be repealed and replaced with ownership disclosure requirements that would allow BOT to better monitor ownership interlocks and potential sources of financial strain. Strict and full disclosure would also enable investors and clients to obtain an accurate picture of the real situation. O Loans made to financial subsidiaries or affiliates 32! should be deducted from the parent's equity for capital adequacy purposes to the extent that such loans are not fully collateralized by assets with a determinable market value satisfactory to BOT. O Parent bank financial support for an insolvent subsidiary or affiliate should not be permitted by BOT unless the cause of the insolvency is clearly identified as extraordinary and is judged correctable by the application of sound management techniques, based on presentation to BOT of a detailed business plan in form and substance acceptable to BOT. 31/ This eliminates ;±s need for a specific limit on the percentage of a bank's capital that may be invested in financial subsidiaries, although many countries do set such a limit. 32/ These tend to be unsecured short-term loans or money market placements that are unlikely to be repaid on demand, should the parent bank or the subsidiarylaffiliate experience financial difficulty. - 68 - The same principles stated above should also apply when the parent io an F&S company and questions arise of the F&S company's investments in or loans to its subsidiary or affiliate. The legal insulation only works if problems of a failed institution are not allowed to be transferred to the parent beyond the extent of that parent's equity investment. 4.44 The Holding ComDanv Alternative. The above recommendations assume that the existing structure of a direct parent-subsidiary relationship between banks and F&S companies would be maintained. However, this structure can impair the public's (and management's) perception of the extent of the segre- gation between a bank and a failing F&S company, since the owners may be legally or morally presumed to be Oresponsible" for their related F&S company, and any failure to protect creditors may be improperly perceived as a sign of financial weakness in the parent bank itself. The risk containment objectives laid out in para. 4.27 can be better achieved through placing the bank and its related F&S company on an equal plane through the use of a common holding company, which is essentially a non-operating legal and tax entity that owns the bank and each related F&S company. In this way, the balance sheet of each individual financial entity is kept clean and uncluttered by other related financial institutions that have differing balance sheet structures (thus improving transparency for regulators, investors and clients). The issue of double counting of capital is also avoided, since each institution is a separate entity requiring its own capital. Consolidation for tax purposes at the holding company level is essential to ensure that the group is properly motivated to innovate--this may require some modernization of the Thai tax code (it is not examined here any further because it is outside of the scope of the Study). 4.45 ?rohibition on ownership in Non-Financial Companies. As a corol- lary to the risk containment principle, investments in subsidiaries or affiliates that are not involved in financial intermediation business should be prohibited (except maybe venture capital companies and where the risks are clearly isolated). The BOT should have the responsibility for defining those activities that form part of the financial intermediation business. Allowable activities should be uniformly defined as published by BOT from time to time in the Gazette. 4.46 Lending to Related Companies. Insulation of !nstitutions from difficulties in related entities is a key concern of bank regulators. In country after country, bank and nonbank failures have been linked to lending to related entities, since ownership interests can easily subvert prudent credit policies and procedures, particularly credit risk evaluation and approval standards, loan concentration limits, documentation standards, and recovery action should the borrower experience difficulties. If decisions on lending to related entities are controlled by the bank's owners on nonbanking criteria, then such loans in effect represent a removal from the financial institution of all or part of the equity cushion needed to protect depositors - 69 - against unforseen impairment of assets.33/ The Thai regulatory environment already recognizes these dangers through strict limits on lending to related entities. Unfortunately, these limits are largely unenforceable--leaving the Thai banking systet% vulnerable to abuse in comparison to the systems in countries that have moved from closely held organizations to general stock ownership and professional management answerable to diverse, balanced economic interests. The BOT has been actively seeking to identify lending to related entities, but this is a difficult task, and identification of the problem does not solve it. It is recommended that BOT actively pursue all available legal measure to identify related lending and to force compliance with the existing restrictions--it is essen'ial that individual managements feel that they run a serious risk of criminal penalties for failure to disclose such loans and for any unsound lending practices to related entities that put depositor and BOT funds at risk. 33/ In a somewhat extreme example in another country, loans to shareholders totaled seven time paid up capital, and accumulated losses totaled fif- teen time paid up capital. - 70 - V. CONCENTRATION AND COMPETITION IN THE FINANCIAL SYSTEM Introduction 5.1 Thailand's top four or five banks are widely perceived in the country as holding a strong grip on the market and even, by some, to have an unfair competitive advantage in the system. Although this perception is not univer- sally held, it is, significantly, shared by government authorities. An earlier Bank report 341 did, in fact, identify excessive concertration in the banking system as a problem. Noncompetitive behavior in the banking sector is generally perceived to result in the provision of services of lower quality and at higher price than warranted, to the detriment of both the consumer and efficient use of economic resources. There is also the sociopolitical concern that concentration of power in a few financial institutions can lead to un- desirable concentrations of wealth. Given these concerns, this chapter examines the issues related to concentration/competition in the financial sys- tem in some detail. 5.2 The analysis is organized under three h adingst structure, conduct and performance. Under "structure" are examined the openness of the market both to entry and to exit, the degree of concentration and the regulatory environment. 'Conduct' refers to the behavior of the market participants in terms of pricing and quality of services. Under "performance' is considered the profitability of the industry and its leading participants. To be set against the aforementioned concerns is the possibility that considerable economies of scale may exist in banking. If so, and this remains a very con- troversial issue worldwide, it may be desirable to tolerate high concentration in the banking system, with a few dominant banks, ever if their pricing and provision of services display some elements of oligopolistic behavior. A. Structure and Concentration 5.3 The institutional structure of the Thai financial system was des- cribed in Chapter III of the report. With some 15 domestic banks, 14 foreign banks 351 and scores of finance companies, there is no absolute shortage of banks and near-banks in Thailand. There is no doubt, however, that banking in Thailand is a highly concentrated industry (see Table 5.1). 341 Thailand: Perspectives for Financial Reform, World Bank, Report No. 4085-TH, July 31, 1983. 351 The Government is expected soon to allow several more foreign banks to operate in Thailand. - 71 - Table 5.1s COMMERCIL BANK CONCENWTATION (in percentage shares La) ----- Total Assets ------ Deposits Branches Bank 1962 1972 1980 1988 1988 1988 Bangkok Bank 21.5 29.5 34.5 28.3 28.7 16.4 Krung Thai Bank 18.7 14.6 13.4 14.2 15.5 13.8 Thai Farmers Bank 4.5 6.9 11.4 12.6 13.5 14.5 Siam Commercial Bank 5.6 4.9 5.5 8.3 8.9 9.8 Bank of Ayudhya 5.3 6.3 4.7 5.7 5.6 8.1 Thai Military Bank 2.5 3.3 2.8 5.6 6.0 7.3 First Bangkok City Bank 2.4 3.4 3.3 3.8 3.1 2.5 Siam City Bank 3.6 4.0 2.9 3.8 3.5 4.8 Bangkok Metropolitan Bank 3.5 4.9 4.5 3.6 3.4 5.4 Bangkok Bank of Commerce 4.3 3.6 3.7 3.1 3.4 6.8 Bank of Asia 3.2 2.4 1.8 2.6 2.3 2.3 Union Bank of Bangkok 3.0 2.3 1.5 1.6 1.6 4.2 Thai Danu Bank 1.6 1.3 0.9 1.1 1.2 1.4 Nokornthon Bank 0.9 0.9 1.2 Laem Thong Bank 2.3 0.8 0.6 0.4 0.4 0.5 Foreign banks 16.7 9.7 5.9 4.4 2.0 0.9 3-bank concentration 44.7 51.0 59.3 55.1 57.7 44.8 4-bank concentration 50.3 55.9 64.8 63.4 66.6 54.5 5-bank concentration 55.6 62.2 69.5 69.1 72.2 62.6 Herfindahl Index 0.113 0.138 0.165 0.138 0.145 0.102 ia Columns may not add to 100 because of rounding. Between 1962 and 1980, the share of tobal assets held by the top three banks grew from 452 to 592 and that of the top five banks grew from 562 to almost 702. This trend has now been halted and slightly reversed, with the top three banks' holding declining to 55Z of total assets at end-1988 and the top five banks holding 69S of the total. Using the numbers-equivalent inverse of the Herfindahl indes,36/ the 1962 structure was approximately equivalent to 361 The Herfindahl index is a commonly-used measure of concentration in an industry. It uses information about the market share of each bank in the system to give a single summary measure. The reciprocal of t a Herfin- dahl index is used as a numbers-equivalent measure of concentration: it is the number of equally-sized banks which would result in the same level of concentration as is actually observed. A refinement used in this report is to adjust the Herfindahl index for country size (para. 5.6). This is computed by multiplying the Herfindahl index by the square root of GDP. - 72 - 1u8t under nine equally-sized banks, which fell to six equally-sized banks by 1980 but recovered by 1988 to better than seven equally-sized banks. The *hares of deposits held by the banks follow a similar pattern. 5.4 The increasing share of the top banks in the 19608 and 19709 re- flected the declining role of the foreign banks, whose share of total assets shrank from 172 to nder 6S in 1980 and about 42 by 1988. This loss par- ticularly benefited Bangkok Bank, which increased its market share from about one fifth of total assets in 1962 to slightly more than one third by 1980, then fell to 282 by 1988. Bangkok Bank's lose of market share during the 1980s was to the gain of what are now the fourth- and sixth-ranked banks, Siam Commercial and Thai Military. From being tenth-ranked in 1962, the latter is now poised to replace the Bank of Ayudhya as the fifth largest holder of total assets and, as has already occurred, of deposits. 5.5 The high degree of concentration in the banking industry in Thailand is better assessed, however, in the context of the industry structure in other developing countries. As indicated in the following table, in only one of 15 countries (other than Thailand) for which data have been collected is the market share of the top four banks less than one half.371 371 In making international or historical comparisons of concentration, it is never clear whether to look at the three- four- or five-bank concentra- tion ratio. Different choices among these alternatives give widely dif- fering results. The four-bank concentration ratio has been used here simply because the data for other countries were available in terms of four-bank concentration. _ 73 - Table 5.2: BANK CONCENTRATION, OPERATING EXPENSES AND INTEREST MARGINS Selected Developing Countries, 1986-88 Various Years 4-Bank Adjusted Concentration Herfiudahl Herfindahl Operating Interest Country Ratio Index Index la Expenses lb Margin lb Thailand 65 0.13 0.9 2.28 2.50 Algeria 100 0.29 2.2 1.17 3.56 Bangladesh 74 0.20 0.8 2.35 2.50 Botswana 100 0.45 0.4 0.69 3.38 Cameroon 82 0.18 0.6 4.25 3.39 Chile 54 0.09 0.4 1.68 1.78 Cote d'Ivoire 67 0.14 0.4 4.10 Indonesia 66 0.10 0.9 2.40 2.24 Malaysia 64 2.80 3.60 Morocco 70 0.18 0.6 2.51 3.29 Nigeria 52 0.09 0.7 3.48 5.80 Pakistan 0.26 1.5 2.70 2.15 Philippines 40 0.06 0.3 3.90 2.70 Senegal 78 0.16 0.6 Tunisia 80 0.19 0.6 1.60 2.20 Turkey 52 0.04 0.3 5.44 2.82 /a Adjusted Herfindahl ig Herfindahl multiplied by square root of GDP to correct for market size. lb These are expressed as percentages of total assets. 5.6 Whether the Thai banking industry is highly concentrated by inter- national standards is subject to debate. Though the Herfindahl index suggests that Thai banking is more concentrated than in the Philippines or Indonesia, for example, of the 15 developing countries for vhich data are readily avail- able, Thailand ranked only tenth in banking concentration in 1986, with a Herfindahl index of 0.13. This suggests that the degree of banking industry concentration is not unusual in Thailand. But if an adjustment is made for market size, considering that there is more scope for numerous banks in large economies, Thailand rises sharply in the ratskings. in particular, the adjusted concentration measure, which takes account of market size, pushes Thailand up in the rankings to third place out of the fifteen, along with Indonesia. Thai banking therefore seems to be rather highly concentrated when account is taken of market size. 5.7 A particular aspect of possible unfair competitive behavior relates to branching activity. Some observers have alleged that the large banks have limited the smaller banks' access to the most lucrative branch locations, though this is denied by local authorities. Although a high proportion of branch requests were granted in 1989, the small banks complain that it was those branches which they mcst wanted that were denied. Taking a longer view, - 74 - comparison can be made between the distribution of branches in 1978 and 1988, respectively. Over this period, the total number of branches increased by some 70X. The largest number of new branches went to barngkok Bank, and this number represented a somewhat higher percentage increase than the average. The second-largest bank, Krung Thai, received almost as many new branches, representing an even higher proportionate increase, althcagh the absorption by Krung Thai of some failed banks complicates the picture somewhat. Large increases in the number of branches by some banks was accompanied by an increase in market share. The Thai Military Bank ar. the Siam Commercial Bank almost trebled and doubled, respectively, their branch network and, as has already been noted, gained sharply in market share. On the other hand, the Bangkok Metropolitan Bank also more than doubled its branch network, but with a decline in market share. The distribution of branches is somewhat less unequal than the distribution of market share (Table 5.1). For example, the top four banks have only 552 of the branches, but about 632 of total assets. However, there is no evidence that the authorities have systematically favored smaller banks in the allocation of new branches over the decade, or succeeded if such a policy were followed: the Herfindahl index of branches is the same now as it was ten years ago. As discussed more fully and recommended in paras. 3.42-3.48, BOT should restate its policy on branching so that ambiguities and misunderstandings are removed and the objectives of its policy with respect to branching are clearly understood. 5.8 One other aspect of the policy affecting the share of the largest banks is the requirement that many state enterprises must give banking busi- ness to the government-owned Krung Thai Bank. Continuation of this policy may be due to the special considerations surrounding the Krung Thai's management of failed banks, but it clearly contributes to the market power of the second largest bank and helps perpetuate concentration of the industry. B. Conducts Evidence of Market Power 5.9 The aspect of conduct is examined in terms of market power exer- cised by large banks individually vis-a-vis small banks, possible collusive behavior on the part of the largest banks as a group, and market power of the banks vis-a-vis the non-bank financial intermediaries. An important point to recognize here is that monopoly power in banking, if at all present, is not generally observed across the whole range of banking services, but lies chiefly in some market segments. Some segments, such as wholesale deposit gathering and lending to medium-scale corporate clients, tend to have the characteristics of a contestable market: even though only a few players are active, the potential entry of others deters the incumbents from monopolistic behavior. In other segments, monopoly profits can be sustained because the potential entrant would have to incur high sunk costs. 5.10 The Thai situation seems to confirm these general observations. Large banks do appear to have size advantages in certain types of business such as money transmission, large loan syndications and retail deposit gathering. At the same time, they also experience some competitive disadvantages of scale: for example, because of greater managerial flexibility, the small banks may be able to operate more effectively in the newly deregulated deposit market. The large banks appear to use the - 75 - advantages of sise when they can, though concrete illustrations of significant abuses are not easily documenteds even the small banks are unable to point to notable examples. The small banks acknowledge that the large banks are better placed to effect money transmission services, but see this as having little effect on their own ability to provide the services for which they have a comparative advantage. Significantly. although large banks may attempt to tie in other services to those for which they have a comparative advantage, there does not appear to be collusion among the large banks in shutting out smaller banks from business. Furthermore, competition among the large banks helps limit margins on most lines of business, 5.11 One exception to this general practice of fairly competitive pricing is foreign exchange business, which appears to be excessively profit- able and which the banks may be using to cross-subsidise other lines of busi- ness in order to attract customers away from the finance copanies, which are not allowed to handle foreign exchange dealings. This is one example where banks as a group may have an undue compet' .ve advantage over non banks. Tie ia sales may be most important in this area.381 C. Performance and Profitability 5.1? The published accounts of Thai banks over the past few years. indicate some apparent differences between the cost and performance of large and small banks. However, the differences are small, and it cannot be ruled out that they reflect a chance coincidence of random factors rather than a systematic pattern. 5.13 For this analysis, the same *peer grouping" has been used as dis- cussed in Chapter nII. Peer group 1 (PG1) comprises the Bangkok Bank, Thai Farmers Bank, Siam Commercial Bank, Bank of Ayudhya and Thai Military Bank, being five of the six largest. Krung Thai Bank, also one of the six largest, has been classified separately as PG2 because of its special characteristics as a government-owned bank including its assumption of the business of some failed banks. The data for P02 are not considered in this analysis. PG3 represents the next tiers five medium size banks. The four smallest banks, Union, Thai Danu, Nakornthon and Laem Thong, form PG4. The performance data are given in Table 5.3. 381 A recent Bank of Thailand study also showed that large banks charged more for letter of credit services than did small banks. - 76 - Table S.Ss THAILAND BANKS: PERFOKMANCE DATA BY PEER GROUP PG1 PG3 P04 FinCos Pe1 P03 PG4 Funding Cost (7) Lending Rate (2) 1985 9.93 10.06 10.89 - 13.25 12.83 13.45 1986 8.54 9.65 9.55 - 11.40 10.10 11.96 1987 6.32 6.68 7.01 7.68 9.62 8.91 9 79 1988 6.32 6.61 6.74 7.14 9.94 8.82 10.00 Operating Expenses * Intermediation Costs * 1985 2.11 1.97 2.02 3.26 2.33 2.89 1986 2.15 1.90 2.22 3.14 1.17 2.89 1987 2.16 1.98 2.37 3.74 2.68 3.46 1988 2.10 1.84 2.43 4.04 2.89 3.62 Profits* Interest Margin* 1985 0.54 0.16 0.40 3.32 2.78 2.56 1986 0.41 -0.94 0.24 2.86 0.46 2.41 1987 0.61 0.40 0.62 3.30 2.24 2.79 1988 0.69 0.76 -0.42 3.62 2.21 3.25 Loan-Loss* 1985 0.32 0.09 0.30 1986 0.39 0.17 0.26 1987 0.63 0.20 0.17 1988 0.82 0.18 1.32 * As percentage of total assets. Notes Intermediation costs - operating expenses + loan losses + taxes + profits. Interest margin - lending rate - funding cost. 5.14 On the average cost of funds, an unambiguous pattern seems to emerge from the accounts for 1985-88. P¢1 banks have the lowest funding costs in each year and P64 banks the highest. On average, the funding cost for PG1 banks was 7.782, compared with 8.25Z for PG3 banks and 8.552 for PG4 banks. One reason for the lower cost of funds for PG1 banks, especially in 1985 and 1986 whes baht interest rates were higher than US dollar rates, is probably this group's greater recourse (and presumably easier access) to foreign funding. Foreign deposits formed about 71 of the total liabilities of PG1 banks, but did not figure in the balance sheets of other banks. Another reason could be the larger branch network of PG1 banks allowing them access to - 77 - lower-cost deposits. However, the PG1 banks also had a smaller than average share of interest-free demand deposits. In terms of operating expenses and intermediation costs, too, there does not appear to be any consistent pattern in favor of P61 banks; if at all, the P03 banks seem to have an edge. 5.15 The performance v7ith respect to lendins rates was somewhat different. The lowest lending rates were charged by PG3 banks--on average almost a full percentage point less than the rate charged by the PGI banks. The ranking of the groups was the same in each year (Figure 5.1). These lending rates are not risk-adjusted, though some indication of the relative degree of risk might be adduced from the provisions made against loans. In fact, P01 banks made much larger provisions than PG3 banks though not enough to offset the interest differential completely. Taking this adjustment at face value would indicate that loan charges by PG1 banks and PG3 banks were less different on a risk- adjusted basis. (PG4 banks also made high provisions, but they were concen- trated in 1988 and are attributable to the exceptional level of provision made by a single bank.) At February 1989, the rates in effect on the loan port- folio of the different groups of banks no longer showed a clear ranking. The general conclusion should be that, while individual banks have had different lending policies, no competitive advantage of size is evident from the interest yield of loans. 5.16 The lower funding costs of the PG1 banks is the main contributor to that group's slightly higher recorded profit margin, which everaged 0.562 on total assets during 1985-88, compared with 0.102 and 0.212, respectively, for PG3 and PG4 banks. It should be noted tnat PG1 banks account for just under two thirds of total bank assets and PG3 banks for about one fifth. An indica- tion of the scale of the profit differential between large and small banks is that if all banks (including both PGl and PG3 banks) had earned the lowest profit margin, i.e., only 0.102 on total assets, total bank profits would have been on average about B 3.5 billion less a year. The eeneral conclusion from this analysis by peer groups is that the largest banks may have a slight com- petitive advantage in raising deposits, but cannot command higher yields on their loans. The advantage on the deposit side has translated into somewhat higher profit margins for the large banks. 5.17 Both intermediation margins and bank profitability have increased in the past two years since the lowering of depositor interest rates between 1986 and 1987 was sharper than and preceded that in lending rates. The difference and the time lag allowed an improvement in profits despite tho. need to make higher loan loss provisions. For instance, intermediation margins of PG1 banks rose from 3.32 of total assets in 1986 to 4.02 in 1988. Over one half of this increase was absorbed in higher loan loss provisions, the remainder being equally divided between additional profits and additional income tax paid. 5.18 Onerating expenses have been more stable, at just over 22 of total assets per annum, except for PG4 banks which experienced an increase under all subheadings. In an international comparison of developing countries (Table 5.2), operating expenses in Thailand appear to be on the low side: Thailand ranks fifth out of the sample of 15 countries. The same is true of gross interest margins (fifth out of the same 15). How2ver, such direct 78 - Flgure 6.1: Deposit and Lending Rates, by Size of Banks 14 Small 12 Med[ium E Bg-5 10 4- 2- 1985 1986 1987 1988 1985 1986 1987 1988 Deposit Lerding Expenses and Proftts, by Size of Banks 2.5 2 - Medium 1.5 - Big-5 I 0E.5- 0 -0.5 1985 1986 1987 1988 1985 1986 1987 1988 Expenses Proflls OW46473 - 79 - comparisons are not always reliable because of the structural, accounting and regulatory differences among banking systems. Furthermore, Thailand should now aspire to even greater efficiency than is suggested by a position in the second quartile of developing countries. D. Concentration in the Finance Comoanies 5.19 Finance companies represent a heterogeneous category of institutions. Concentration is relatively low: the three largest firms account for about 18S of total assets, the five largest for less than 27?. Neverthelces, each of the two largest finance companies has more assets than the two smallest com- mercial banks. 5.20 Net interest income as a percentage of year-end total assets varied between 1.2? and 1.6? for the four largest companies in 1988, with much higher figures having been recorded in the previous year. This fall in the share of net interest income reflects the rapid asset growth of the two largest compa- niess assets of TISCO, the market leade: in 1988, grew by 125? in a year. For the finance companies as a whole, net interest represented 0.8? of assets, up from 0.42 the previous year.39/ This low net interest percentage reflects in part the higher funding costs experienced by finance companies compared with banks: in both 1987 and 1988, average funding costs for finance companies, at 7.7? and 7.1?, respectively, exceeded those for the small (PG4) banks (see Table 5.3). 5.21 Net income for finance companies depends heavily on income from secu- rities trading. This is true both for large individual companies-- TISCO's gain from securities trading exceeded its net income for 1988--and for the sector as a whole: trading gain amounted to 98? of net income for the total of all finance companies in 1988, and far exceeded net income in 1287. 5.22 The finance companies thus operate intermediation on very fine margins; securities trading is the mainstay of their profitability. Their survival alongside the banks may depend on their privileged position with regard to securities trading,40/ and, to a lesser extent, on the hire-purchase business. A relaxation of the restrictions which inhibit finance companies from doing foreign exchange business and from branching (as recommended earlier in paras. 4.33 and 4.34) might help some of the larger companies not only to improve tbeir margins but also to provide greater competition for the banks. 39, Spreads for finance companies were a good deal higher in the early 1980s. 401 However, as so many finance companies are associated witb commercial banks, it is possible that a comparison of their profits and funding costs may be marred by the existence of transactions between finance companies and their associated banks which may not always be at a fair market price. - 80 - E. Industrial-Financial Grouns 5.23 Each of the major banks is associated, through overlapping ownership and control, with a variety of non financial companies as well as with at least one, and usually more than one, finance company. Therefore, the poten- tial anticompetitive role of the overlapping ownership and control of banks, finance companies and nonfinancial enterprises must also be considered in the context of a discussion of competition in the banking industry. This over- lapping and common ownership of financial and non-financial businesses by individual groups is not uncommon in developing countries with predominantly private-owned banking systems. 5.24 The first issue in this respect is that a bank's relationship with enterprises which are part of its industrial-financial group may not be con- ducted at arm's length and fair market prices. If so, loans may be made which would not be unambiguously in the interest of the minority shareholders or the depositors of the bank. This issue is primarily a matter for prudential regu- lations it is also discussed earlier in para. 4.46. 5.25 Second, there is the risk that, due to conflicts of interest, non- group borrowers may be unfairly treated. Thus, for example, a client who is the competitor of a nonfinancial company in the group may be denied access to credit by the group's bank and finance companies. In such a case the client would have to turn to another group's bank. No evidence is available to indicate that the different groups act in concert to shut out nongroup cli- ents, but that is a theoretical possibility. The existence in the system of at least some nongroup financial institutions (IFCT and a few finance companies, for example) also mitigates this risk. 5.26 Third. banks may use their nonfinancial enterprises to capture bank- ing business. Reportedly, this does happen (for example, a joint venture partner of a group concern is obliged to use the group's bank). 5.27 Fourth, reported bank profits may be misleading to the extent that transfer pricing is practiced between banks and their nonfinancial associates. Banks might be more profitable than shown in the accounts if they were passing some of these profits (resulting from below market pricing of loans and absorption of costs) to nonfinancial associates. 5.28 If there are significant monopolistic abuses in the Thai banking sys- tem ,they may lie chiefly in these kinds of within-group dealings. Though little quantitative evidence of such practices is available, and they are de- nied by the banks, it is generally believed that relations between banks and their associated companies are not completely transparent. It would require a separate study to measure quantitatively the extent of uncompetitive behavior in these respects and express a judgement authoritatively. F. Government Policy on Competition 5.29 On balance, the foregoing discussion belies the view that the top banks have established a stable and powerful cartel. The top five banks have been successfully challenged by a second-tier bank. The market leader's share - 81 - has been eroded (though it remains about twice as large as its nearest competitor). Powerful though they are, the leading banks are not immune to competition. Nevertheless, the absence, for a quarter of a century, of new domestic entrants and, perhaps more importantly, the restrictions on branching which continue to apply to foreign banks mean that the system is not fully open to competitive forces. In summary, though concentration indicators for Thailand remain on the high side for such a large country, and although entry is extremely limited, the reversal in the fortunes of some of the larger banks suggests that their power is by no means absolute. 5.30 At the same time, the top five banks are widely (though not univer- sally) perceived as holding a strong grip on the market. How can these two facts be reconciled? There are three possible explanations. First, the concern may be with size oer se rather that with competition. The fact that Bangkok Bank alone accounts for one third of the entire commercial banking system stands out and almost unavoidably attracts fears and allegation of unfair competition. Second, the concern may be with the size and power of 'the groups' rather than with the banks, which are only one albeit the most visible component in each group. No systematic study has been undertaken of the concentration of wealth in the hands of the various groups but it is generally believed in the country to be substantial. Finally, "unfair competition' may be a euphemism for the resentment against the social and political power that the groups command because of their size and which they do not hesitate to exercise. 5.31 Even if there is no clear evidence of flagrant unfair competitive behavior in the Thai financial system, it would still be desirable to make it publicly known that the government authorities will not tolerate anticompeti- tive practices and to establish a 'banking ombudsman' to address grievances brought forward by the public. Such a complaints procedure need not be costly in terms of resources and might provide a useful channel of information to the authorities in their attempts to improve the competitiveness and efficiency of the banking system. 5.32 Policy with regard to competition would require both positive actions and the removal of some existing interventions. Positive actions could include a more positive attitude to licensing new banks, domestic as well as foreign. The authorities are already considering to allow several new foreign banks to operate in Thailand. This would be a step in the right direction (see para. 5.2). While this might give foreign enterprises a larger share in the banking industry, the benefits to both the consumer and the national economy of more competitive banking services should easily outweigh such a concern. The arrival of foreign banks is also likely to weaken the power of existing industrial-financial groups more effectively than, say, anti-trust legislation. 5.33 In addition, the authorities should consider removing certain res- trictions and privileges which help create niches of market power. Examples include the restriction of foreign exchange dealing to commercial banks. The securities market privileges of finance companies would also need to be looked at in this context. A relaxation of the restrictions on branching by both - 82 - foreign banks and finance companies would also help.41/ These aspects are discussed in more detail in relevant sections of the report. If, as proposed elsewhere in the report, the regulatory environment facing the larger and more viable finance companies were brought much closer to that now facing banks, there might be a considerable shake-down and rationalization of the industry, with the viable part of many of the finance companies merged into larger entities. 5.34 Other examples of regulations which tend to enhance the market power of the larger banks are the restriction on interest payable on demand deposits and the privileges of the Krung Thai Bank witb respect to state enterprise deposits. 5.35 An entirely different route to promote more competition for banks would be to encourage the development of the capital market, particularly the market in private debt securities. The banking sector has a practical mono- poly on providing credit to the real sectors. There is a need to developing an active market in notes and debentures. This is discussed in detail in Chapter IX. 5.36 In conclusion, although little firm evidence is available substan- tiating monopolistic abuses in the banking system, and although the magnitude of any associated inefficiency does not seem high compared to the situation in some other developing countries, government authorities should take every opportunity to ensure the maximum degree of competition in the system. Banking is potentially an important export industry, but only those banks which have achieved a high degree of efficiency can expect to compete interna- tionally. Alr-eady, market participants complain that the more sophisticated banking services are not being introduced as rapidly as they might be because the market is not open to newcomers. A more open and competitive banking system would benefit the consumer of bank services, in addition to providing a long-term strengthening of the industry itself. 5.37 Although the analysis of this chapter enabled the conclusion to be drawn as stated above, there is also a need to look at competition from the perspective of different categories or classes of borrowers. This analysis would require data on who borrows from where at what terms and conditions, and the objective would be to disclose the presence of systemic biases, if any. Authorities may wish to note this as an item on future work agenda. 411 It is not recommended, however, tnat the authorities go beyond this by restricting branching by large banks while encouraging branching by small banks. - 83 - VI. REGULATION AND SUPERVISION OF FINANCIAL INSTITUTIONS A. Sunervision and Regulations Organization of Prudential Suoervision 6.1 Thailand's Commercial Banking Act empowers the Minister of Finance to appoint inspectors to examine and report on the affairs of banks; however, in practice, supervision of financial institutions in Thailand is delegated to BOT. The actual performance of the supervisory function is carried out by the Departments of Bank Supervision and Examination and Financial Institution Supervision and Examination. These departments report to the same Assistant Governor, together with the Legal Department and the Financial Institutions Develonment Fund, thus assuring a certain amount of consistency in the formu- lation, development, and application of supervisory policies and practices. 6.2 Approximately 600 persons are employed within the two supervisory units where they are assigned to the tasks of supervision, examination, analysis, licensing,421 and administration. Supervisory coverage is extended to commercial banks, finance companies, securities companies, finance and securities companies, and credit foncier companies. A small number of specialized, government-owned institutions are under the direct control of the Ministry of Finance. These include the Government Savings Bank, the Bank for Agriculture and Agricultural Cooperatives, the Industrial Finance Corporation of Thailand, and the Government Housing Bank. 6.3 The Department of Bank Supervision and Examination employs approxima- tely 350 persons in the supervision of commercial banks. Of this number, slightly more than 200 are involved in the actual performance of on-site bank examinations. At the end of September 1989, there were 15 Thai banks opera- ting a nationwide branch system of 2,103 branches in Thailand and 17 branches and four agencies abroad. In addition, there are 14 foreign banks operating 16 offices in Thailand. To supervise these institutions, the Department is organized into several divisions. Three divisions are devoted to the task of on-site examinations. In addition, there are three regional units in the north, south, and east responsible for the examination of larger branch offices. Another division, the Bank Supervision and Analysis Division, is responsible for off-site surveillance of banks and the banking industry. Lastly, the Special Assignments Division is responsible for researching new supervisory methods, performing computer audits, and implementing special projects. 6.4 The Department of Financial Institution Supervision and Examination employs approximately 250 persons in the supervision of the 123 nonbank finan- cial institutions, i.e., the Finance and Securities Companies (F&S companies) and credit fonciers. This department is organized like the Department of Bank 3upervision and Examination, with three divisions containing approximately 150 421 The Bank of Thailand's role in licensing is that of providing a recommenda- tion to the Minister of Finance who actually grants the license. - 84 - persons, devoted to on-site examinations of non-bank financial institutions. In addition to their responsibilities for on-site examination, these divisions are also responsible for off-site analysis of the nonbanking financial insti- tutions. Two other divisions complete the organizational framework. The Financial Institution Supervision and Development Division is responsible for proposing and issuing new regulations, promoting and developing the business of nonbank financial institutions, managing the Fourth of April Scheme,431 and commenting on applications for new licenses. It also serves as a 'think tank' on issues related to the nonbanks. The Special Assignments Division is responsible for investigating alleged illegal activities, developing and updating supervisory methodologies, assisting in debt recovery and liquidation efforts under the Lifeboat Scheme (see para. 6.48), and performing specJal assignments as required. 6.5 There is a perception in the institutions supervised (banks as well as F&S companies) that the supervision standards of the two separate super- visory departments might not be uniform. It is hard to express a definitive judgement on this point. Absolute uniform standards can be assured only if there were only one department under a single director. But this is not a practical proposition since the total staff of 600 persons is too large for effective supervision. If the total staff and work is to be divided into two departments, organizing them on the basis of the type of institutions super- vised is the most logical basis. The only suggestion offered is for BOT to adopt a policy of frequent staff transfers between the two Departments. The BOT should also always keep the two Departments under the same Assistant Governor, as 8s the case at present. -The Sunervisory Process 6.6 Bank supervision in Thailand is correctly viewed as being very important in ensuring the stable development of the banking industry, with its positive benefits for economic growth and tne protection of depositors. Thus, the objectives of bank supervision today focus on monitoring the changes occurring within the banking industry and assessing their possible impact on the health of the system and the monetary policies being pursued by the authorities. 6.7 To carry ovt this monitoring function, supervisory activities for banks as well as F&S companies are performed both on-site and off-site. During the past decade, the emphasis of the on-site examination process has evolved from one of determining compliance with laws and regulations to one designed to assess the financial condition and solvency of the institutions supervised. The supervisory authorities had already begun to change their examination focus by the time of the finance company crisis in 1983. However, the process was not yet mature and the legal powers necessary to enforce corrective actions on the financial institutions were lacking. The Emergency Decree Amending the Commercial Banking Act (1985) substantially strengthened 431 The Fourth of April Scheme refers to the take-over of insolvent finance companies on April 4, 1984. - 85 - the powers of supervisors to act. Today, the overall framework for bank supervision and examination (as distinct from the failure resolution process) is highly satisfactory, and the skills of individual examiners rank high among their peers from other countries. 6.8 On-Site Examinations. On-site examinations are performed with the principal objective of determining a bank's condition and performance. To accomplish this, targeted on-site examinations of all head offices are gene- rally conducted once a year. Branch offices are selected for examination based on the bank's internal reporting processes and the importance of the branch to the overall health of the institution. In addition, other branches are selected for on-site visits on a sample basis. In planning examinations, known or suspected problem banks are given first priority in the scheduling and allocation of resources. 6.9 During the conduct of on-site examinations, the examiners attempt to form opinions on the bank's overall condition by evaluating the five CAMEL components (capital adequacy, asset quality, management, earnings, and liquidity) through a process of verification, sampling, and inquiry. To make the best use of its scarce resources, the areas to be examined are targeted to higher-risk activities, e.g., loans and advances, foreign exchange, speculative activities (for example, the recent condominium toom has been the subject of special interest), and insider transactions. 6.10 'While the staff's on-site examination work is generally thorough and satisfactory, there is some concern that they may not yet be focusing enough attention and rigor on certain aspects of the institutions' internal manage- ment systems and processes, such as strategic planning and formally adopted written policies. An emphasis on the prudent development of these management systems is necessary to ensure that financial institutions control risks within reasonable tolerances and anticipate and prevent future problems. The further development of such an approach moves supervisors away from simply determining an institution's condition today to reasonably ensuring an institution's health in the future. In the process, the supervisors become positive agents or catalysts for change in the way financial institutions operate. 6.11 Off-Site Surveillance. The off-site surveillance function is under- taken based on reports submitted by the institutions. For banks, these includet (a) a quarterly summary statement of assets and liabilities; (b) a monthly financial condition report with attachments; (c) a monthly status report listing the major debtors and contingent liabilities for large custo- mers; (d) a semi-annual income and expense statement with attachments; (e) an annual report of fixed assets; (f) a monthly report on inter-bank transac- tions; (g) a semi-annual report on deposits; (h) a quarterly report on clas- sified assets; (i) an Annual balance sheet and profit and loss statement; and (j) semi-annual reports on overdue debts. In addition, a variety of statis- tical reports are required at various intervals. These provide information regarding the amount of credit granted to different economic sectors, types of debtors or businesses, the public sector, etc. - 86 - 6.12 The appropriate reports are compiled and analyzed versus trends, peers, etc. Where appropriate, microcomputer modeling through simulations and forecasting is performed. Through this process, actual and potential problem areas are identified and communicated through the appropriate channels for follow-up. This process, as described, is relatively well developed and distinguishes the supervisory authorities in Thailand from their colleagues in many other developing countries. The Legal Framework 6.13 The legal basis for bank supervision and prudential regulation of the financial system is contained in a variety of laws and decrees. The most important of these are the Bank of Thailand Act, 1942, the Commercial Banking Acts of 1962 and 1979, the Emergency Decree Amending the Commercial Banking Act, 1985, and the Act on the Undertaking of Finance Business, Securities Business and Credit Foncier Business, 1979. The main features of these laws, as they relate to prudential aspects, are discussed below. 6.14 Licensing. Commercial banks may be established only in the form of a limited public company with a license granted by the Minister of Finance. The application for a license is required to contain information on: the number and value of shares; the names and biographical notes of the organizers; the proposed name of the commercial bank; the expected deposits at inception and the sources of those deposits; the main business activities to be conducted; the scope and powers of executive officers; and the names, work records, and qualifications of the persons to be appointed as executive officers. The Minister may also require such other information or conditions as he determines necessary. After fulfilling all conditions, the bank obtains the registration as a limited public company. It then submits copies of the certificate of registration, the Memorandum of Association, the minutes of the statutory meet,ing of the company, the Articles of Association, and the register of shareholders to the Minister for the issuance of the banking license. The information requirements and the process of approval are thorough and rigorous. However, in practice, all this seems irrelevant in that no new bank has been licensed in the last quarter of a century. A similar, but somewhat less complicated procedure, is required for the establishment of branches. As discussed earlier in Chapters III and IV, branch opening decisions should essentially be left to the judgements of the banks and F&S companies concerned. 6.15 Foreign banks wishing to establish a branch must maintain assets (branch capital) in Thailand of not less than B 5 million. This can take the form of: (a) deposits with BOT (separate and distinct from required reserves); (b) unencumbered government securities; tc) bonds issued by BOT; (d) shares, debentures, or debt instruments issued by BAAC or IFCT, and debentures or bonds guaranteed by the Minister of Finance; and/or (e) fixed assets in Thailand used for bank premises or employee housing. In addition, the Minister of Finance may impose such other conditions as deemed necessary before granting a branch license. 6.16 Finance companies, securities companies, and credit foncier companies must follow licensing procedures which are similar to banks. The finance - 87 - business and credit foncier business may only be undertaken by limited public companies after a license has been obtained from the Minister of Finance. The securities business may be undertaken only by a limited company 441 after the grant of license by the Minister. 6.17 The capital requirements for licensing of nonbank financial institu- tions differ based on the nature of the license granted and the business activity to be undertaken. For finance companies, a minimum capital of B 60 million is required. Credit foncier companies require capital of B 30 million. Securities companies which manage the sale of securities require capital of B 10 million. Securities companies which only engage in brokerage, trading, investment advice, and investment management require minimum capital of B 5 million. In every instance, the Minister of Finance muky prescribe a higher level of capital. Prudential Controls and Limits 6.18 Capital Adequacy. The Bank of Thailand is granted the power to prescribe capital adequacy ratios with the approval of the Minister. The ratios may be established with regard to: total assets, subject to maximum and minimum levels of 152 and 52, respectively; each type of asset; and contingent liabilities. The following assets are excluded from the computations: cash, deposits with BOT, deposits with other banks, Thai government securities, and such other assets as may be prescribed by BOT with the approval of the Minister of Finance. 6.19 Using the powers granted under the Commercial Bankizig Act, BOT has established a minimum capital adequacy ratio of 82 for banks. This minimum level for capital is based on a narrow definition of capital which includes only core capital items;45/ however, it has permitted 31 exemptions (in addition to those included in the Act and discussed above) for different classes of assets, including certain categories of risky, priority sector loans. The total exemptions from the capital adequacy computations account for approximately 402 of total assets. Approximately 252 of the exempted amount comprises items waived for policy reasons. By permitting these exemp- tions, BOT has used capital adequacy as a tool of economic regulation to encourage directed credit, rather than as a buffer to absorb unusual losses. This has had the effect of substantially weakening the capital adequacy requirements. (This aspect of capital adequacy was discussed in detail in paras. 3.28-3.41.) 6.20 Additionally, because of the manner in which the Conmmercial Banking Act is drafted, two capital adequacy standards exist: one for balance sheet risks, as described above, and one for off-balance sheet contingencies. The 441 Note that the definition of a limited company includes limited public companies. Finance companies and credit foncier companies must be limited public companies. Securities companies may be limited public companies. 45/ The core capital items included in the definition of capital are: paid-up capital, share premium, retained earnings, and statutory reserves. - 88 - latter requires capital of 202 in relation to the amount of avals, acceptance bills, and loan guarantees outstanding. Contingencies in the form of letters of credit, foreign exchange transactions, and other types of guarantees are exempted. 6.21 Similar standards have been prescribed for nonbank financial institu- tions, where a minimum capital adequacy ratio has been established at 62 of risk assets for balance sheet risks and 252 for off-balance sheet contingent liabilities. 6.22 Given the weaknesses described above and Thailand's increasing im- portance as a financial center, it is recommended that the authorities consider amendments to the Commercial Banking Act that will move Thailand closer to convergence with international practices by adopting the capital adequacy standard formulated by the Basle Committee on Banking Supervision. This approach establishes risk weights for both balance sheet and off-balance sheet items following a single standard based solely on prudential considera- tions and broadens the definition of capital into core and secondary com- ponents. Other actions, such as permitting the tax deductibility of loan loss provisions (see para. 6.32), should also be taken to ensure that banks are adequately capitalized. 5.23 Concentration of Ownership. The Commercial Banking Act, as amended, limits shareholding by one person to not more than 52 of the total shares outstanding. Further, a commercial bank should not have fewer than 250 persons as shareholders, each of whom holds not more than 1/2 of 1? of the total shares sold, but who together hold shares not less than 5O0 of the total shares outstanding. Three quarters of the shareholders must be Thai nationals. 6.24 For finance companies and credit fonciers, limits on sha-reholding by one person are lOS of the shares outstanding. No fewer than 100 persons, each holding not more than 6/10 of 12 of the total shares sold, must hold not less than 502 of the total shares outstanding. As with banks, three quarters of the shareholders must be Thai nationals. Although limits or. the ownership of finance companies and credit fonciers have been established, it is commonly assumed that the ownership limits have been exceeded through nominees. 6.25 Exposure Limits. The total amount of credit that can be extended to a single borrower is limited to 252 of a bank's capital funds and 302 of the capital of a finance company or credit foncier. In addition, a higher limit of 40? of capital has been established for finance companies if contingent liabilities are also included. For purposes of applying these limits, all related borrowers in a group are combined as if they are one person. 6.26 While the bank lending limit for a single borrower is within reanon- able constraints, the additional limit for contingent liabilities exceeds a prudent level in that the bank may be forced to disburse credit under those contingent liabilities at an inopportune time, with the result that the bank may become overexposed to a borrower in distress. Given such risks, many countries have combined such limits into one lending limit with a maximum exposure of 252 of capital or less. Therefore, it is recommended that the - 89 - authorities seek changes in the law which will establish such an exposure limit not exceeding 252 of capital. 6.27 Eauity ITnvestments. Banks, finance companies and credit fonciers may invest in the shares of a limited company in an amount up to 102 of the company's shares outstanding or, in the case of banks and credit fonciers, 202 of their capital funds. Securities companies need only to notify BOT if they acquire 20? or more of the shares of a limited company. Banks and finance companies may not hold shares of other banks and finance companies respective- ly except where the shares are acquired for debts previously contracted. In such cases, they have six months to dispose of the shares. 6.28 Loans to Insiders. Banks and nonbank financial institutions are prohibited from extending or guaranteeing credit to or for their own direc- tors. This prohibition extends to the spouse or minor child of a director, an ordinary partnership in which the director is a partner, and limited partner- ships or companies in which the director has unlimited liability and/or holds more than a 302 beneficial interest. 6.29 It is a known fact in Thailand that banking groups also have invest- ments in real sectors. It is unrealistic to believe that banks would not be making loans in group-related companies. An absolute bar on loans to related companies (insiders) in all probability has led to circumvention of the law instead of strict adherence to it (like the 102 limit on ownership of F&S companies, discussed in paras. 4.36-4.46). It is suggested that BOT consider allowing loans to directors and other related interests within well-defined limits. At the same time, BOT should also provide for (a) strict full disclosure of all such transactions in the banks' published annual reports, (b) abstention of the concerned directors/officers from the decision making process related to such transactions, and (c) sanctions for the violators of the limits. 6.30 Asset Classification. Provisioning, and Interest Suspension. Banks and nonbank financial institutions are required to write off their worthless or irrecoverable assets and make provisions for doubtful accounts at the end of each semi-annual accounting period. The BOT has prescribed guidelines to assist banks in determining their worthless, irrecoverable, or doubtful assets. 6.31 During their on-site examinations, the examiners classify credits using a grading system similar to that used in the United States and many other countries, although the definitions have been modified for the Thai context. Classified asset categories are substandard, doubtful, and loss. Assets classified in these categories exhibit well-defined credit weaknesses. Another category, especially mentioned, includes assets which possess poten- tial weaknesses. While this system for asset classification is ba6ed on sound principles, loan loss provisions are now required only for assets classified as doubtful or loss. In most countries, a more conservative view is taken that loss potential exists within the aggregate of substandard assets and this loss potential should be provided against. In addition, general reserves should be established for the balance of the portfolio based on historical loan loss experience, economic trends, loan growth, changes in loan policies - 90 - or lending management, concentrations of credit, and the inherent riskiness in different types of loans. It is recommended that BOT adopt this more conser- vative approach. 6.32 Related to this topic is the tax deductibility of provisions. At present, the tax treatment of provisions acts as a disincentive to adequate ;vovisioning since an institution must have exhausted nearly all legal reme- dies before the tax authorities will consider the loss a deductible expense. As is known, resolution of civil suits may take years. It is suggested that the tax authorities consider a more reasonable policy for permitting the tax deductibility of provisions so that banks are encouraged to set aside adequate reserves for losses. 6.33 Rules have been established for the suspension of interest on non- performing assets. Based on these rules, the accrua' of interest should cease on loans which are 180 days in arrears. However, financial institutions are permitted to continue accrual if adequate collateral is pledged against the indebtedness. Real estate, up to its assessed value, is considered suitable collateral by the authorities for this purpose. There is a concern that this policy is overly generous in permitting financial institutions to record pro- fits by capitalizing interest even though its borrower is not performing. Collateral, in all cases, should only be viewed as the ultimate source of repayment in case the borrower fails as a going concern, and not as a sub- stitute for the borrower's ability to repay through cash flow or the normal business asset conversion cycle. Further, real estate is illiquid, and liqui- dation and realization of proceeds after legal civil suit proceedings may take an indeterminably long time. Therefore, it is recommended that the decision on suspension of accrual of interest should not take into account the aspect of collateral. A possible exception may be where the collateral is liquid and already in the process of collection. 6.34 Annual Accounts. Both banks and nonbank financial institutions are required to publish their annual balance sheet and profit and loss statement within four months of the end of the accounting year. The statements must be in the form prescribed by BOT and must be audited by an auditor apnroved by BOT. Sanctions 6.35 The Emergency Banking Decree Amending the Bank of Thailand Act, 1985, significantly strengthened the sanctions and other actions which could be imposed by the Thai authorities. In addition to the traditional sanctions of levying monetary fines, imposing penalties, or withdrawing an institution's license, the authorities can now exercise a full range of intermediate powers. These include the powers to: (a) order a financial institution to suspend all or part of its operations for a temporary period of time, in which case the Minister of Finance may also prescribe rules, procedures, or conditions as - 91 - deemed necessary;"j/ (b) order the closure of a branch; (c) place the finan- cial institution under control; (d) prohibit the payment of dividends or other distributions of profits; (e) prohibit or restrict the lending of money or making of investments; (f) require a financial institution to reduce or increase its capital; (g) remove and/or appoint directors or other persons; (h) require a financial institur ion to take or omit certain actions within a specified time frame; and (i) impose any other conditions necessary to rectify the condition or operation of a financial institution. 6.36 Despite the vast array of powers granted to the authorities under the Emergency Decree. the use of sanctions, in practice, has been reasonable and measured. In one respect, however, it is debatable whether BOT has been too lenient. It is known that BOT has permitted those troubled financial institu- tions with reasonable prospects for recovery to achieve an adequate level of loan loss reserves over a phased period subject to conditions it has pres- cribed. The result could be that at a given point in time, loan loss provi- sions may be underprovided, and this fact is not d closed to the depositors or Investors in the concerned banks' stock. On ti-- other hand, if banks were required to make full provisions at the time the provisions were first de- termined to be necessary, they would show reduced profits or even operating losses, which might affect their ability to attract and retain deposits or to access the money and capital markets. The BOT has been willing to pursue a remedial program with such institutions because it fears that the impact of making full provisions immediately could prove destabilizing. If any finan- cial institution does not adhere to the remedial program agreed to with BOT, it would be subject to further sanctions and, depending on its condition, even intervention. The balanced approach used thus far by BOT satisfies both the financial institutions involved and BOT in trying to resolve a problematic situation. 6.37 If the Minister of Finance determines it is necessary to intervene and take control of a financial institution, the Minister is granted the power f *-point a control committee consisting of a chairman and two others. This committee has the power and duty to manage the affairs of the institution in every respect, including the appointment of officers. If it is determined that the institution cannot be returned to health, the Minister may order the institution dissolved and appoint a receiver to oversee its liquidation. B. Resolution of Banking Insolvencies and Denositor Protection 6.38 The experiences of most countries suggest that the restructuring of insolvent financial institutions is best handled when three critical actions are taken: 461 In general, if a financial institution incurs losses which reduce its registered capital by one quarter, it may no longer accept deposits. If it loses one half of its registered capital, it is required to stop doing business and to submit a plan which addresses the cause of the problems and propose corrective actions to be undertaken. - 92 - (a) The management culture is changed through the replacement of the bank's chief executive officer and other key decision-makers. Frequently, this involves the replacement of several layers of management. It is important that managers who remain do not become obstacles to a successful restructuring of the institution; (b) Shareholders of the institution are first to suffer the consequences of the diminution of capital through provisions, write-offs and other losses. When an institution becomes technically insolvent, super- visors order its closure and the institution ceases to exist as a corporate body. Because time is of the essence in restoring the public's access to their monies on deposit, banks, unlike other companies, should not go through normal bankruptcy proceedings. When a bank is declared insolvent, shareholders should lose all rights and claims to wlhich they were previously entitled; and (c) the non-performing assets are carved out of the balance sheet and removed from the failed institution. This is facilitated by the purchase of the bad assets by a restructuring mechanism, such as a deposit protection fund, or the government through cash or government securities at market rates of interest. The use of securities may be preferable since the initial outlays of cash are limited to interest payments, thus the monetary effects are minimized. The appointed receiver is given the responsibility of collecting the non performing assets so that the restructured institution will not be unduly burdened by having to continue collection activities. This process of carving out bad assets creates a shell of a good institution which can then be recapitalized by new investors, an acquiring institution, or the government (in which case it would, ideally, be reprivatized within a specific time period). Importantly, the institution will be able to operate profitably since performing assets have replaced non- performing ones. Subsequent actions should involve operational restructuring to rationalize redundant staff, branch networks, management information systems, etc. 6.39 As will be discussed below, all these actions were not taken in Thailand when the need arose to deal with technically insolvent institutions. Instead, insolvent institutions were assisted through a variety of support mechanisms, including soft loans used to purchase government securities. The result is that costs continue to mount, and problems remain unsolved. The following provides a brief background as to the crises of the early 1980s in the financial system and the support arrangements devised to deal with it. The Finance CompanY Crisis 6.40 Starting in the late-1970s, Thailand faced a financial crisis which, by the mid-1980s, affected nearly a third of Thailand's domestic banks and - 93 - nonbank financial institutions.47/ Twenty-four financial institutions were closed during a three-year period between late-1983 and November 1986. Another 30 financial institutions were intervened. As a consequence of the crisis, the legal framework was substantially strengthened through the Emergency Decree Amending the Bank of Thailand Act, 1985; the supervisory process was also strengthened by accelerating its change in focus from compliance to assessment of risk. The Financial Institutions Development Fund (FIDF) was created at the same time to deal with failing and failed financial institutions, and as a means to spread the financial burden across the financial system. While the actions taken by the Thai authorities stabilized the problem, they did not resolve it. Today, many of the institutions continue to be supported through various subsidies such as soft loans at substantial cost with no end in sight. 6.41 At the onset of the crisis, an appropriate arrangement for handling seriously distressed and/or insolvent institutions did not exist and so these institutions were handled on an ad hoc basis. For many institutions, rather than to close and liquidate them or restructure them using a purchase and assumption method in which non-performing assets are carved out of the balance sheet, the choice was simply to support them with soft loans under the Fourth of April lifeboats scheme. In retrospect, the authorities acted as they did for a combination of reasons: (a) they did not ask for or provide sufficient resources to deal with the crisis, thus limiting the actions which could be taken; (b) in the minds of some, the problems were initially thought to be ones of temporary illiquidity, not insolvency, thus support was provided in the form of liquidity through loans and other means, rather than substantive restructuring and removal of the bad assets; (c) the actions taken were expedient under the existing legal framework in which B0T powers were limited; and (d) the legal framework handicapped, and still does, the transfer of claims once legal action is commenced. The consequence of this was to block a purchase and assumption transaction. Therefore, the troubled financial institution had to remain in existence in order to foreclose on debtors or to collect on bad assets. Financial Assistance Provided to Troubled Institutions 6.42 To support the distressed and insolvent institutions, BOT and other financial institutions provided financial assistance through a variety of mechanisms. These are described below. 6.43 Security Purchase Fund. Beginning in late-1979, short-term financial assistance was provided to finance and securities companies through a Security Purchase Fund. The state-owned Krung Thai Bank, acting in response to government policy, began purchasing listed and authorized securities of finance and securities companies using funds borrowed from BOT. On July 16, 47/ For a discussion of the causes of the financial system crisis, see R. Barry Johnstons 'Distressed Financial Institutions in Thailand: Structural Weaknesses, Support Operations and Economic Consequences," IMF Working Paper WP/89/4, January 13, 1989, pp. 12-25. - 94 - 1987 Krung Thai Rank sold securities for B 1,655 million and repaid BOT loans in the amount of B 1,067 million. The Fund ceased operations on December 31, 1988, by transferring accumulated net profits of B 514 million to the Ministry of Finance. for its services, Krung Thai Bank was compensated with a 1U management fee equal to B 65 million.48/ 6.44 The P odmissory Note Exchante Proaram. This program was used to assist the holders of promissory notes of failed F&S companies and credit fonciers. During the initial phase of the program, four major F&S companies 49/ were chosen by the authorities to exchange a series of their own promissory notes for those of the failed companies. Although the face amounts of the notes given in exchange were the same, the maturities were spread out over a period of up to ten years and no interest was paid, thus the note holders suffered losses in present value terms. 6.45 For their participation in the exchange program, the four selected F&S companies were compensated with soft loans from BOT collateralized by (invested in) goverrment bonds. At the height of the program, the four F&S companies had outstanding soft loans of B 7,690 million Invested in government securities. As of mid-July 1989, the amount outstanding had teen reduced to 5 4,330 million. Altogether, the promissory notes of 19 failed F&S companies and credit fonciers were exchanged for the notes of the four selected FIS companies. As of June 30, 1989, the promissory note obligations of these companies totalled B 5,633 million. 6.46 During the second phase of the program, FIDF which had been set up in 1985 took over the function of the four F&S companies. The current procedure is the same, except that the cost of exchanging promissory notes is now fully borne by FIDF. At the beginning of September 1989, the PTDF had B 108 million of promissory note obligations outstanding. 6.47 Liguiditv Fund. A liquidity fund was established in late 1983 to provide liquidity support to troubled F&S companies. An amount of B 5 billion was contributed by Thai commercial banks under the auspices of the Thai Rankers' Association. Of this amount, B 1 billion w-as provided by Krung Thai Rank acting on behalf of the Government. The Fund was jointly managed by SOT, hrung Thai Bank representing the Thai Bankers' Association, and the Ministry of Finance. Rates charged to financial institutions were adjustable at the prevailing market rate. The rate paid by the Fund was set at the discount rate of BOT. Loans were initially granted for periods of up to three years, but have since been rolled over. As of December 31, 1988, Krung Thai Bank had extended overdrafts and other credit to the Fund in the amount of B 1,921 million. An additional B 678 million had been disbursed to support the Fund. 48/ Krung Thai Bank Limited, Annual Report 1988, pp. 28-29. 491 Thai Financial Syndicate Co., Ltd., Thai Financial Trust Co., Ltd., Thananant Finance and Securities Co., Ltd., and Thanasap Finance and Securities Co., Ltd. - 95 - 6.48 The Lifeboat Scheme. Under the Fourth of Aprll 1984 "Lifeboat Scheme, the Ministry of Finance initiated a program to provide additional financial support to troubled financial institutions. However, unlike the earlier programs, this scheme enabled the authorities to intervene in the troubled financial institutions. Twenty-five finance and securities companies agreed to participate in the scheme. As a condition of participation, each of the participating finance and securities companies had to agree to write off its losses and provide reserves for doubtful accounts, strengthen credit and collection procedures, suspend dividends, and submit a financial plan for returning the institution to health. In addition, depending upon the condition of the company, 25Z of the outstanding sharee plus 502 of the voting rights, to 902 of the outstanding shares were to be transferred to the Ministry of Finance. The 50X voting rights were to be transferred back to the original owners within five years at a price to be determined by the Ministry of Finance. Thus, the existing shareholders would see a dilution in their share ownership. but would retain the hope of recovery in the value of their shares due to the government's intervention. 6.49 Several forms of support were provided through the Lifeboat Scheme. (a) Account 1. Account 1 was a program set up to provide liquidity support through the Krung Thai Bank to companies in the Lifeboat Scheme. Krung Thai Bank, acting at the behest of the government, established overdraft facilities for the financial institutions participating in the scheme. In turn, BOT extended a loan, with a letter of comfort or aval from the Ministry of Finance, to the Krung Thai Bank in the amount of the overdraft facilities granted. In general, the overdraft facility extended to each of the financial institutions in the Lifeboat Scheme was established in an amount equal to one half of the promissory notes outstanding to the public. The overdraft facilities were for the sole purpose of permitting distressed F&S companies to meet their promissory note obligations as they came due. A rate of 14.52 was initially charged by the Krung Thai Bank, but this rate was gradually reduced and has been at 7.5? since June 1988. As of June 30, 1989, B 2,420 million was out- standing under the overdraft facilities in a total credit line of B 3,155 million. (b) The Baht 4 Billion Credit Line. In an attempt to resolve the problems of F&S companies still under the Lifeboat Scheme, the authorities asked Krung Thai Bank to take over the management of these companies in January 1987. The Bank appointed its own staff to the management teams of the companies and took actions to restructure their activities, in some cases merging weaker institutions into new companies. In addition, Krung Thai Bank made B 4 billion in demand loans available to the participating companies. These loans carried a subsidized rate, initially 8? (currently at the interbank rate), and were used by the financial institutions to reduce their funding costs, to pay back loans from othc.-r financial institutions, and to expand their business activities in the normal course of business. The BOT issued a letter of comfort to Krung Thai Bank to assure it _ 96 - that it would not suffer any losses from loans extended to the financial institutions under the scheme. As of June 30, 1989, the distressed F&S companies had B 3,467 million outstanding under the B 4 billion line. (c) Capital Iniections. The BOT provided a sum of B 2.4 billion in January 1986 in the form of soft loans to Krung Thai Bank (B 1.4 billion) and FIDF (B 1 billion) to be used for capital injections in 17 troubled companies. The loan to Krung Thai Bank carries a letter of comfort or aval from the Ministry of Finance. By the close of 1988, Krung Thai Bank held more than 602 of the shares in six F&S companies, more than 301 of the shares in two additional companies, and less than 20? in the remainder. As of June 30, 1989, Krung Thai Bank had invested a total of B 1,381 million (B 71 million has since been repaid from the sale cf shares) in these F&S companies. At the same time, FIDP invested B 670 million in seven institutions. (d) Soft Loans. Soft loans have been used as a support mechanism in Thailand for many years. They were widely used in dealing with the institutions in the Lifeboat Scheme. Soft loans, at interest rates ranging from one tenth of 12 to 8X, were extended by BOT to distressed financial institutions for periods ranging from six months to five-years and eight months (maximum maturity is five years but a loan may be extended). The proceeds of these loans were used to purchase government securities required as collateral. The effect of this arrangement was to provide the institutions a subsidy in the form of the interest rate differential between the interest earned on tne securities and the interest paid on the borrowings. This form of assistance was the most common used by BOT in dealing with the finance company crisis. As of June 30, 1989, soft loans of B 5,270 million were outstanding to 28 finance and securities companies. (e) Aval Credit Line. In addition to the other financial support provided by it, the Krung Thai Bank made available a B 3.5 billion credit line for promissory notes aval at an aval fee of 0.52. As of June 30, 1989, B 2,754 million was in use under the line. Financial Support for Ailing Commercial Banks 6.50 Support arrangements for commercial banks in distress include soft loans and fresh capital. Since 1984 through 1987. eight commercial banks received financial assistance. One of these, Sayam Bank TLimited (formerly Asia Trust Bank) was taken over by the authorities in Aug.ist 1984, although its insolvency had been recognized much earlier. In 1986, the authorities injected B 1.5 billion into Sayam Bank, of which B 1 billion was used to absorb loan losses. An additional B 2 billion was provided as a soft loan. The Sayam Bank was subsequently merged with Krung Thai Bank on August 17, 1987 at the government's behest. By the time of the merger, losses had mounted to the point that Sayam Bank had a negative capital of B 7,372 million. Its - 97 - liabilities of B 14,445 million were nearly twice the value of its assets.50/ The difference was recorded on the books of Krung Thai Bank in "Other Assets, as a noninterest-bearing account called 'Sayam Bank Limited debtor account." The BOT provides a soft loan subsidy to Rrung Thai Bank in order to neutralize the negative effect of its non-performing assets. 6.51 Seven other banks have also received financial support from BOT. Bangkok Bank is the largest recipient with nearly B 6 billion in soft loans. The First Bangkok City Bank, plagued by bad loans and foreign exchange losses, received soft loans of B 3.8 billion and fresh capital. The loans, which were granted in May 1986, carry a ten-year maturity. The Bank of Asia received a soft loan of B 1.7 billion in 1987. The Siam City Bank received soft loans of B 3.5 billion. The weight of managing the finance companies in the Lifeboat Scheme and the merged Sayam Bank took its toll on Krung Thai Bank and it, too, needed assistance from the authorities. It currently has soft loans of B 1.5 billion from BOT; in addition, the FIDF increased its stake in the bank to 38.51S as of December 31, 1988, compared to 26.93X one year earlier. Thai Military Bank and Siam Commercial Bank have also received soft loans of B 1.4 billion and B 0.3 billion respectively. The Financial Institutions Development Fund 6.52 One outgrowth of the financial system crisis was the establishment of the Financial Institutions Development Fund (FIDF). This Fund was established as a separate and distinct legal entity in 1985 by the emergency Banking Decree Amending the Bank of Thailand Act. The Fund is managed by the Fund Management Committee comprised of the Governor of BOT as Chairman, the Permanent Secretary of Finance as Deputy Chairman, and five to nine other committee members appointed by the Minister of Finance. Staffing is provided by BOT and currently numbers about 30. Capitalization and operating monies are provided through assessments against banks and other financial institu- tions, resources provided by BOT, other monies and properties devolving on the Fund, and earnings from investment of the Fund's assets. In addition to B 1.2 billion in capital, BOT has provided soft loans to the Fund, aggregating B 8.5 billion as of September 5, 1989. A balance sheet of the Fund is given in Annex 44. 6.53 The FIDF's powers are wide-ranging and include the powers to lend or borrow money, purchase assets, invest in or hold shares, hold ownership of or possessory rights to property, provide financial assistance to depositors or creditors of financial institutions who sustain losses, and guarantee bills. These powers are sufficient to enable the FIDF to carry out its objectives which are, loosely stated, to reconstruct and develop the financial institu- tion system. Although FIDF collects premiums from financial institutions, it does not provide explicit deposit insurance. However, it is widely assumed that FIDF, and the Government, would not permit depositors to sustain a loss of their monies on deposit. 50/ Krung Thai Bank Limited, Annual Report 1988, p. 24. - 98 - 6.54 In practice, FIDF acts principally as an arm of BOT to provide assistance to distressed financial institutions through the provision of soft loans, capital iljections, and management expertise; it is run as a department of BOT. It has no authority to conduct bank examinations and, therefore, it must rely solely on BOT for its information; however, it has no legal right to receive the examination and supervisory information prepared by BOT. It has no authority to intervene or close banks. It is not expressly authorized to act as receiver; in any case it lacks specialized staff skilled in workout situations. The FIDF does not have the full-range of powers or functions necessary to be considered an effective mechanism for restructuring financial institutions. Nonetheless, it is thought that FIDF can play a greater role in resolving the incidence of financial distress which now exists if certain improvements were to be made in its functioning. These are discussed in para. 6.62 below. Cost of SupDort Arrangements 6.55 Quantification of the costs incurred under the several programs pro- viding support to distressed financial institutions is difficult because of the pa.U4ty of data. But since even a rough, order of magnitude estimate has merit, it has been attempted. As a first step, the following table has been constructed to show the amounts outstanding under the various schemes. - 99 - Table 6.1. AMOUNTS OUTSTANDING UNDER VARIOUS SUPPORT SCUUSS 1 Institution Outstanding Providing Type of Balance Support Beneficiary Support (Million B) Bank of Thailand Pour exchangers Soft loans 4,330 FIDF Soft loans 8,460 FIDF Capital 1,200 Finance companies Soft loans 6,283 Commercial banks Soft loans - est. 18,651 Krung Thai Bank Soft loans 2,449 41,373 Financial Institutions Development Fund Finance companies Exchange of P/Ns 108 Finance companies Capital 1,181 Couumercial Banks Capital 5,294 Commercial Banks Soft loans 2.965 9,548 Xrung Thai Bank Finance companies Account 1 2,420 Finance companies Credit line 3,467 Finance companies Aval line 2,754 Finance companies Capital 1,222 Liquidity Fund Soft loans 735 Liquidity Fund Expenses 678 Raja Finance Pay-off PIN holders 54 11,330 Sub-Total 62,251 Less: Duplications 15,074 TOTAL 47,177 la The BOT amounts are as of December 1988 or June 1989; FIDF amounts are as of September 1989; and Krung Thai amounts are as of December 1988. 6.56 The cost is then estimated by using the differential between the interest rate actually charged and the comparable market rate. On this basis, the cost (in the sense of revenues lost) is roughly estimated to be about B 3 billion annually. This is not the total cost incurred to date but is an estimate of the recurring annual cost on the assumption that the support provided remains about B 48 billion. The cost in the future would be dif- ferent if the outstanding amount increases (or declines) and if the market rates change. The current boom in real estate has had a beneficial effect in redeeming some non-performing assets that were secured by real estate, and DO? has successfully found interested parties to invest in many of the instLtu- - 100 - tions in the scheme. If the process of rehabilitation and disposition continues, the costs in the future could be lower. 6.57 Although details of the current status of financial assistance provided to the financial sector are sketchy, it is clear that about 13 511 finance and securities companies were still receiving financial assistance from Krung Thai Bank totalling B 8.6 billion as of the end of 1989 under the Lifeboat Scheme. Seventeen of these companies have also received capital injections from FIDF or Krung Thai Bank in the aggregate amount of B 2.4 bil- lion. Soft loans provided to these F&S companies and several others by BOT total an additional B 6.3 billion. In addition, soft loans outstanding to the four companies which exchanged promissory notes for nineteen defunct F&S companies total B 4.3 billion. An additional amount of B 735 million in soft loans is outstanding to finance and securities companies other than under the schemes mentioned above. The direct exchange or payoff of F&S company promissory notes totals B 162 million. Unremunerated expenses absorbed by Krung Thai Bank to support the Liquidity Fund total B 678 million. Of the fifteen Thai commercial banks, eight are receiving financial assistance in the form of soft loans amounting to B 18.6 billion and/or have received capital injections totalling B 5.3 billion. Weaknesses in the Support Arrangements 6.58 The support arrangements discussed above have shown several serious weaknesses. Perhaps the most serious, the non-performing assets were not removed from the institutions' balance sheets. Therefore, the problems of insolvency were not resolved, only stabilized. Since the institutions were not closed, new chief executives were faced with the dual responsibilities of attempting rescue operations while servicing existing or new clients. Such a task is extraordinarily difficult for even the most talented banker. It is only now, through the economic boom and surge in the value of real estate held as collateral for non-performing assets, that some recoveries are being made. While this occurrence is fortuitous, it should not be viewed as a failure- resolution methodology to be followed in the future. 6.59 Shareholders invest in equities expecting a return on their invest- ment. As a necessary corollary, shareholders also bear the risk of invest- ment. Under the support arrangements followed in Thailand so far, the existinR shareholders of insolvent financial institutions did see a serious dilution of their investments, but they were not eliminated completely as shareholders. To the extent that an institution is rehabilitated, they will benefit. The support arrangements have, therefore, negated an essential 51/ Initially there were 25 F&S companies joining the Scheme. Since then six institutions have merged into Thananant Finance and Security Company and three others have merged into Capital Development Finance Co., Ltd. Five other companies exited the Lifeboat Scheme in 1989. Since the report was prepared, one company has been sold to original investors and one to new investors. As of March 1990, there were 11 companies left under the Lifeboat Scheme. - 101 - element of discipline by preserving shareholders' interests at the public's expense. 6.60 While, in most cases, the chief executive officers of the troubled financial institutions were removed, other managers at the senior level were retained. These managers often proved to be obstacles to rehabilitation efforts and, in some instances, deliberately sabotaged rehabilitation efforts. The failure to remove managers resvonsible for oroblems often permitted the existing weak banking culture to remain intact, making rehabilitation even more difficult. It is clear that in some instances, changes in management must reach deeper down into the ranks if substantive change in the banking culture is to occur. 6.61 As these weaknesses and the foregoing discussion demonstrate, the support arrangements only stabilized problems, but did not resolve failures. Host of the troubled institutions remain troubled even today. As a general observation, the three most critical components of a successful financial restructuring were violated: (a) the bad assets remained; (b) the shareholders were not eliminated; and (c) the management culture was not sufficiently changed by eliminating the responsible layers of management. Recommendations for Strengthening the Failure Resolution Process 6.62 As discussed above. the methodologies chosen in dealing with the financial crises were ones of expediency. Sufficient resources were not devoted to solving the problems upfront, and the approach was flawed. As a result, the losses continue to mount at a rate estimated at B 3 billion per year. In retrospect, the more prudent manner in dealing with the crisis would have been either to: (a) close the institution, pay off deposit holders, and liquidate the institution's assets; or, (b) restructure the institution by removing bad assets, replacing management, and eliminating shareholders. In the latter instance, the removal of bad assets would occur by using government securities to purchase them. A shell of a bank would be created with good assets matching liabilities. This shell would then be sold to new investors and recapitalized or merged with a healthy institution. The decision between liquidation or restructuring would depend largely on the estimate of the relative ultimate costs to be borne by the government. There may also be a need to review the law to make sure that a receiver or another institution can obtain a clean title to assets purchased or acquired from an insolvent insti- tution's portfolio. 6.63 With the changes in the legal framework if necessary, the lessons of experience, and improved supervisory capabilities, there is scope for improving the manner in which bank insolvencies are handled. Clearly, a more formal approach to dealing with insolvency is required to resolve problems than the ad hoc approaches of the past. A formal approach would have the advantages of dealing with failures in a way that is: (a) efficient and consistent over time; (b) somewhat shielded from political influence since the rules of the game would already have been established; and (c) prefunded through periodic assessments against the institutions. It is recommended that FIDF should be made a stronger, more independent and interventionist organiza- - 102 - tion so that it can play a leading role in dealing vith failing and failed financial institutions. This can be accomplished bys (a) providing FIDF its own specialized and highly skilled staff, prefer- ably with a private-sector orientation: (b) mandating the appointment of FIDF as receiver as soon as B0T has determined a financial institution insolvent.; (c) providing FIDF the authority to examine financial institutions, although in practice it would normally defer to the authority of B0S to conduct examinations as the primary supervisor; (d) requiring BOT to routinely inform FIDF of institutions rated 4 or 5 on the CAMEL rating and to share examination reports of institutions so rated; (e) requiring FIDF to elaborate its policies for restructuring or liqui- dating financial institutions, using the principles discussed above as guiding factors; and (f) amending the laws which handicap the transfer of claims after legal proceedings have been instituted, so that FIDF could obtain a clean title to assets acquired in the process of restructuring financial institutions. Deposit Insurance 6.64 Whether a financial system is better off with or without a formal deposit insurance arrangement is controversial, and much literature is available to support either viewpoint. Even the actual experience of countries that have deposit insurance arrangements can be used to argue the case either way, depending on one's viewpoint. Given the controversial nature of the subject, no clear-cut recommendation is being made regarding formal arrangements for depositor protection in Thailand. Instead, the following general observations are being made in the hope that they would be helpful to the authorities. 6.65 Deposit protection arrangements should not be formalized and institu- tionalized unless the financial system as a whole has reached a high level of stability. In a financial system where episodes of institutional distress have occurred frequently in the recent past and where symptoms of distress are continuing, it would be undesirable to introduce, or convert an informal arrangement into, a formal deposit protection system. The cost of operating a formal deposit protection arrangement would under such conditions be prohibi- tively high. In the case of Thailand, the necessary level of financial system stability will be reached soon, if it does not already exist, given the continuing overall economic growth and the rapid appreciation in the value of real as well as financial assets, and assuming that most of the recommenda- tions in this report are accepted and implemented. - 103 - 6.66 Any system of deposit insurance (formal or informal) ought not to seek to protect every depositor. Financial sector discipline brought about by discriminating players is the best kind of descipline. Large depositors are expected to exercise due care in choosing the depository institution and be able to obtain the required information to form their judgements. The consequence of a failure on their part to choose a sound depository institu- tion should be a loss on their account; it should not become a loss for the system or economy as a whole. A formal system of deposit insurance could be justified only if it is intended to protect small depositors only. As an indication of what might be considered 'small" in case the authorities in Thailand opt to introduce a formal system of deposit protection, it is suggested to consider B 100,000 or such smaller amount as would result in covering about IOZ of the total amount on deposit in the system. A variation of the proposal would be to provide for insurance of larger deposits also, but at a steeply declining rate so that thp point of no protection is reached relatively quickly. Before a formal system were introduced, it would be important to develop and reach a firm, explicit consensus that the limits would be strictly adhered to. If there were fears that, despite any such consensus, the limits would be swept away by "exceptions' made whenever an episode of distress occurred, it would be better not to introduce a formal system. An informal system is inherently more flexible and generally better able to handle "exceptions." 6.67 Although a theoretical argument can be made for differentiating among institutions in terms of the fee to be charged (based on riskiness of assets/ portfolio), in practice it would be very difficult to implement a system which reflected the true risks and it might even be counterproductive to try. Therefore, if a formal system were to be introduced in Thailand, it should be applied uniformly to all institutions covered, and the coverage should be compulsory. 6.68 Finally, it should be noted that the financial sector institutions in Thailand seem to favor a formal system of deposit insurance and more disclosure of how the present informal system works if it is to be retained. The Thai authorities' primary concern should be to improve the framework for dealing with failing and failed institutions, as discussed earlier in this chapter. In terms of relative importance, the debate about formal vs the present informal arrangement for depositor protection is secondary. - 104 VII. INDUSTRIAL SECTOR FINANCING AND THE INDUSTRIAL FINANCE CORPORATION OF THAILAND 7.1 Like typical development finance institutions in many developing countries, the Industrial Finance Corporation of Thailand (IFCT) specialized in long-term lending, funded its operations principally by borrowing in foreign currencies, and has now run into problems and faces an uncertain future. The nature of and reasons for the problems are however very dif- ferent; its portfolio remains sound. Given its topicality. the scope of the Study specifically included a critical examination of IFCT's future and role in the financial system. This is the subject of this chapter. Since IFCT's operations relate principally to the industrial sector, the first part is devoted to a brief discussion of the aspects of investment and financing of this sector. A. Industrial Investment and Financing Industrial Growth and Investment 7.2 Industry, and manufacturing in particular, has gradually become more important in the Thai economy. Manufacturing now accounts for about a quarter of GDP (current prices) and is its single largest component. If construction is included, the share would go up to 30Z. For 1987 and 1988, manufacturing has grown at about 13? annually, the second highest sectoral growth rate (after banking, insurance and real estate) in the economy. 7.3 The high growth rate of the manufacturing sector has been sustained by a corresponding high rate of investment. Industrial investment has increased rapidly as the following table shows. Table 7.1s INDUSTRIAL INVESTMENT, 1981-89 (Billion baht) Industrial Machinery Non-Residential & Equipment Construction Total 1981 30.2 38.9 69.1 1982 36.9 38.1 75.0 1983 34.3 30.1 64.4 1984 41.9 28.5 70.4 1985 47.1 32.7 79.8 1986 42.2 32.3 74.5 1987 39.7 26.8 66.5 1988 57.0 34.7 91.7 1989 la 90.6 42.8 133.4 /a Estimate. Sources NESDB. - 105 - A breakdown of industrial investment into public and private sectors is not available, but there is no doubt that the private sector has provided the main thrust. This is clear from a public/pri4ate breakdown of total investment. For example, in 1988 the overall investment growth was 17.7S, with private investment expanding by 23.4Z but public investment by only 0.72. Financing of Industrial Investment 7.4 The sources of financing industrial investment can be broadly clas- sified into two categoriess equity and debt. Equity is derived from retained earnings and new capital; debt (or loan) sources are more varied. Both equity and debt sources of capital are discussed in the following paragraphs. 7.5 Equity Capital. Although detailed and accurate data on equity finan- cing of industrial investment are not available, there is sufficient indirect evidence and aggregate data to indicate clearly that equity financing has been quite adequate. Corporate enterprises have increased their capital at a rapid rate as the following table shows: Table 7.2: INCREASE IN AUTHORIZED CAPITAL OF COMPANIES (Billion baht) New Firms Expansion Firms Total 1980 10.6 4.9 15.5 1981 9.6 5.6 15.2 1982 17.7 6.0 23.7 1983 24.6 14.3 38.9 1984 19.4 22.0 41.4 1985 19.4 29.0 48.4 1986 17.3 23.2 40.5 1987 31.9 44.6 76.5 1988 59.5 54.0 113.5 Source: SET. It can be safely assumed that with such an increase in authorized capital, actual paid-in capital must have also increased at a significant rate. The increase in equity through retained earnings can be estimated by comparing the level of profitability of nonfinancial firms listed on the stock exchange (data on nonlisted firms is not available but may be assumed to be generally similar) and dividend payout ratios. The figures on return on equity of listed nonfinancial firms are given in the following table: - 106 - Table 7.3s RETURN ON EQUITY OF LISTED NONFINANCIAL FIRMS (in percentages) 1983 120.8 1984 12.9 1985 -15.1 1986 11.2 1987 34.S 1988 28.6 Sourcet CMRI. A comparison of the profitability level Implied by the above figures with the pay out ratio of less than 60S in 1987 and 1988 521 indicates that the equity of nonfinancial firms has increased substantially through retention of profits. Overall, the equity of Thai industrial firms has increased more rapidly than their debt, as indicated by the following data on the major industrial subsectors. 7.6 Availability of Medium- and Long-Term Finance. The choice of industrial sector companies in Thailand in need of borrowed capital is essen- tially limited to the banking segment. The suitability and availability of a given type of loan in any individual case depends, in practice, on the amount required, the size and creditworthiness of the borrower, the purpose for which the funds are required, and whether the borrower is a public sector agency or is in the private secLor. As a generalization, there is no shortage of funds and no constraint on lending or borrowing, other than that which is dictated by prudential considerations. The scale of medium- and long-term finance has thus kept pace with the recent substantial growth of the Thai economy. As discussed in detail in the chapter on capital market, it should be noted, however, that a great majority of Thai firms are not allowed to issue deben- tures, with possible implications for competition and costs. 52/ Sources SET. - 107 - Table 7.4: GEARING OF SELECTED MIDUSTRIAL SUBSECTORS, 1980, 1985-88 1980 1985 1986 1987 1988 Food & Beverage Long-tenm debt/Eqvi'ty 0.69 1.28 0.52 0.30 0.19 Total debt/Equity 3.82 5.05 4.03 3.37 2.52 Automotive Long-term debt/Equity 0.26 0.14 0.11 0.12 0.06 Total debt/Equity 1.21 1.14 1.52 1.24 1.59 Construction Materials Long-term debt/Equity 1.32 0.95 0.79 0.51 0.53 Total debt/Equity 2.23 1.93 1.54 1.35 0.92 Electrical Eauipment Long-term debt/Equity 0.20 0.09 0.21 0.20 0.12 Total debt/Equity 1.40 0.86 0.96 1.08 1.04 Packatina Long-term debt/Equity 0.19 0.30 0.19 0.11 0.35 Total debt/Equity 1.56 1.37 1.23 1.26 1.51 Textiles Long-term debt/Equity 0.81 1.08 0.64 0.19 0.09 Total debt/Equity 2.10 4.01 2.64 1.43 1.01 Source: CMRI. 7.7 The principal types of medium- and long-term finance available includes (a) bills discounted; (b) medium- and long-term loans extended by commercial banks, finance companies, insurance companies, and IFCT; (c) bank overdraft facilities extended by commercial banks and routinely renewed; (d) revolving issuance facilities for underwritten notes or bills of exchange; (e) floating rate note issues; (f) leasing finance; (g) bond issues; and (h) issues of debentures. (a) Bills Discounted. There is substantial activity in discounted bills: business -nd industrial firms in need of funds may sell post-dated checks, promissory notes, or bills of exchange to commercial banks or finance companies at a discount, and with recourse. As a category, - 108 - these instruments are known as bills discounted. They represent in total about one third of total commercial bank credits. Their maturity, however, is usually under one year. Post-dated checks in particular are popular with borrowers (because they offer the fastest way to obtain funds without collateral) and with lenders (because defaulting on checks is a violation of both the criminal and civil codes. They represent about 102 of total commercial bank credits. Apart from the amounts sold in the organized financial sector (to commercial banks and finance companies), post-dated checks are also sold in the informal financial vector. (b) Kedium- and Long-term Loans. These are made directly by various financial institutions, mainly commercial banks, to borrowers. The pattern of lending generally follows the close business relationship that exists between banks and groups of companies, but large faci- lities are syndicated between groups of lenders. Depending on the lender and the term of the facility, the interest rate may be fixed or floating. Floating rate facilities are based on short-term interest rates, and the interest rate is reset at contractually agreed intervals during the life of the loan. At end-1987, short- term and long-term loans by commercial banks accounted for about 35.4? of total commercial bank credits. (c) Overdraft Facilities. These are revolving credit commitments backed by some collateral and extended by commercial banks to their cus- tomers. The interest rate applied is floating. On May 1, 1985, BOT limited overdrafts to B 50 million per customer. As a result, the percentage share of bank lending represented by overdraft commitments fell from 45.42 at end-April 1985 to 33.4? at end-1987. td) Revolving Issuance Facilities. These are a relatively new instrument in Thailand, and are becoming more prominent. A recent example is a B 975 million bill of exchange issuance standby credit facility to three members of the Charoen Pokphand group, which was arranged in July 1989 by one bank and underwritten by two others. (e) Floating Rate Notes. These are also a relatively new instrument in Thailand. The first was issued by Thai Acrylic Fiber Co. in 1987. In that year a total of B 1,825 billion was raised this way by five business and industrial firms. The longest maturity was 4.5 years. The interest rate was in each case tied to other market rates, e.g., Citinote, time deposit, or repurchase rates. Despite a margin of 0.25? to 1.5?, such facilities were nevertheless less costly to blue chip companies at the time than borrowing at bank prime rates. (f) Leasing Finance. Leasing has been available for many years, general- ly for periods of up to seven years. The market has yet to gather momentum, and has not done so because of the present lack of clarity in respect of specific legislation governing leasing transactions-- the companies that may engage in leasing business, the rights and liabilities of the parties to a financial lease, transfer of title, etc. (see also paras. 9.49-9.52). The tax treatment of leasing - 109 - business and lease payments also needs clarification by the authori- ties. (g) Bond Issues. Official statistics describe all medium- and long-term bonds, i.e., those with a maturity in excess of one year, as long- term bonds. Long-term bonds have been issued only by the Government, by BOT and by some government enterprises, e.g., the Electricity Generating Authority of Thailand, the National Housing Authority, the Telephone Organization of Thailand, etc. (h) Debentures. Debentures, in Thai usage, are similar to bonds but are not government-guaranteed or backed by real assets or security. They may be issued only by public companies and companies already regis- tered with SET. Debentures may be straight, guaranteed, convertible, or guaranteed convertible, with maturities ranging from five years to eleven years. They are registered with SET and trade on the exchange. 7.8 Despite the many sources of debt financing and instruments used, the principal providers of credit to the industrial sector are the commercial banks. finance companies and IFCT, and it is in the form of straight loans. The amounts of outstanding credit provided by these sources (1981-88) are shown in Table 7.7. 7.9 Foreign Capital. Foreign capital (loans as well as equity invest- ments) has also been an important source of financing for the industrial sector, as the following figures show: Table 7.5: NET INFLOW OF FOREIGN CAPITAL FOR MANUFACTURING SECTOR, 1980-88 (Million baht) 1980 1981 1982 1983 1984 1985 1986 1987 1988 1,014.8 2,526.0 1,231.0 2,567.5 3,167.0 1,358.1 2,123.8 4,749.2 16,345.0 Source: BOT. A breakdown of the above data into equity and loans is not available. How- ever, total equity and loan inflow and outflow (without sectoral breakdown) are available, indicating that foreign capital inflow in the form of equity far exceeds that in the form of loans. - 110 _ B. The Industrial Finance Corporation of Thailand Legal Framework and OwnershiD 7.10 The Industrial Finance Corporation of Thailand (IFCT) was established in 1960 under the IFCT Act of 1959 to take over the role of its failed prede- cessor, the Industrial Bank, which had been set up under the Industrial Bank Act 1952 and which was liquidated in 1959. The Act states that IFCT's main objectives are tog (a) assist in the establishment, expansion and moderniza- tion of industrial enterprises in the private sector; and (b) encourage and bring about the participation of private capital, both internal and external in such private enterprises, mobilize and pool funds and assist in development of the capital market. 7.11 Unlike the Industrial Bank, which was government-owned, IFCT was set up as a privately-managed private sector corporation because the Government wanted the private sector to play a bigger role in the development of private industrial enterprises. Over the years, the structure of IFCT's shareholding has changed substantially. In 1975, commercial banks as a group held 7.32 of its shares. That percentage declined to less than 502 by 1980, and now com- mercial banks hold only 332 of IFCT's shares. The Government, which became a shareholder for the first time in 1982 when it bought 20X of IFCT's shares, now directly holds only 152 but holds a further 9Z through the Krung Thai Bank. The finance and securities companies as a group now hold 13S, private companies 242 and individuals 122 of IFCT's share capital. Insurance com- panies have been static shareholders at 2?. Foreign shareholders, now with 41?, are limited to a maximum of 49Z under the Articles of Association of IFCT. IFCT became a listed company with SET in 1975. 7.12 The ten largest shareholders in IFCT, and their percentage holdings at end 1988, were as follows: Thai State through Ministry of Finance 15.48S Thai State through Krung Thai Bank Ltd. 9.221 Mitsui Bank Ltd. 5.282 Deutsche Finanzierungs Gesellschaft 3.94? Bangkok Bank Ltd. 2.57? Chase Nominees Ltd. 2.53? Bank of Tokyo Ltd. 2.01? Investment Siam Pte Ltd. 1.782 Banque Internationale a Luxembourg a/c Asian Development Equity Fund 1.32? IFCT Position in Thai Financial System 7.13 Although an active financial institution, IFCT is not governed by the country's Banking Act, and does not fall under the supervisory jurisdiction of BOT; it reports directly to the Ministry of Finance. It is Thailand's only specialized financial institution providing project-based long-term capital, and has an excellent reputation for its professionalism in the financial com- munity. - III - 7.14 Despite rapid growth in IPCT's operations, IFCT remains a relatively small institution in Thailand's financial system, largely because the finan- cial system itself has grown so rapidly. As the following table shows, IFCT's share in system-wide assets has hovered slightly above 12, and IFCT's loans outstanding have been about 12 of total outstanding credit of the financial system as a whole. Table 7.6: IFCT SHARE IN THAI FINANCIAL SYSTEM ASSETS AND CREDIT, 1980-88 (In percentages) Total Assets Total Credit Outstanding 1980 0.7 1.0 1981 N.A 1.1 1982 N.A 1.1 1983 N.A 0.9 1984 N.A 1.0 1985 1.2 1.1 1986 1.3 1.1 1987 1.4 1.0 1988 1.2 1.0 N.A. Data not available. Source: Calculated from basic data supplied by BOT. 7.15 Because of IFCT's emphasis on the manufacturing sector, its share of system-wide financing for this sector is considerably higher than its overall share as indicated above. Still, IFCT accounts on average only for about 42 (and declining in recent years) of the total credit to the manufacturing sector provided by commercial banks, finance companies and IFCT, as the fol- lowing table shows. - 112 - Table 7.7: PRINCIPAL SOURCES OF CREDIT TO MANUFACTURING SECTOR, 1981-88 (Billion baht) Commercial Banks Finance Companies IFCT Year Amount x Amount S Amount 2 1981 59.0 74.6 16.7 21.1 3.4 4.3 1982 64.9 73.1 20.1 22.6 3.8 4.3 1983 88.7 77.5 21.9 19.1 3.9 3.4 1984 106.0 79.6 22.1 16.6 5.1 3.8 1985 122.6 80.1 23.5 15.4 6.9 4.5 1986 124.9 80.4 23.4 15.1 7.1 4.6 1987 162.2 83.9 24.1 12.5 7.0 3.6 1988 223.9 84.6 33.6 12.7 7.1 2.7 Source: Calculated from basic data supplied by BOT. 7.16 IFCT's share in total medium- and long-term credit to the manufac- turing sector is probably somewhat higher than shown in the table, but this is difficult to quantify since a substantial part of commercial banks, term financing is technically on a short-term basis and recorded as such. The Special Status of IFCT 7.17 Even before the Government bought a minority shareholding in IFCT in 1982, IFCT has never been regarded as a fully private sector, commercially- oriented, financial institution. Legally, it is a private sector institution and its charter limits its lending to private sector enterprises only. How- ever, it is regarded as a quasi-state enterprise and is accorded special statuss it pays no corporate income tax; it is exempted from payment of business taxes and revenue stamps; its debt securities are eligible as commer- cial banks' secondary reserve assets; no withholding tax is payable on interest from IFCT bonds; it is exempted from the limitation on shareholdings in other companies; and it is the only company outside the F&S company sector which may underwrite securities. 7.18 The Government has guaranteed IFCT's foreign borrowings from multi- lateral institutions (the World Bank and Asian Development Bank) as well as from the capital markets; its debentures issued domestically are also guaran- teed by the Government. The Government also has an agreement to cover IFCT's foreign exchange losses, for a fee. But interpretation of the agreement during the last decade has left IFCT with significant losses (see paras. 2.32- 7.40). 7.19 The special status of IFCT carries with it some substantial costs and burdens. Its charter, together with other constraints, significantly limits its ability to adapt to changed economic and market conditions, and to respond to substantially increased competition for profitable business from commercial - 113 - banks and F&S companies. For example, it is required to devote significant staff resources to financing small- and medium-scale enterprises; it is prohi- bited from taking deposits; it has restricted itself in the spreads it may charge; it may not offer a wider range of services to clients than term loans and working capital to its term loan clients; and it may not lend to para- statal agencies. Because it is supposed to focus on "developmental" aspects and has special privileges (e.g., exemption from taxes), it is, by corollary, not supposed to be a profit maximizing institution. IFCT, therefore, absorbs considerable costs from activities such as undertaking specific studies, con- ducting training programs and seminars, and establishing and running programs like the Capital Market Development Fund, Industrial Development Fund and Small Industries Credit Guarantee Fund. 7.20 Perhaps IFTC's most important special feature is its lack of autonomy in making policy decisions without prior approval of the Government. The Government also appoints IFTC's Chairman,531 and must approve its President, the chief executive officer. The Chairman has for more than a decade been either the current or a former Finance Minister, and has always exercised considerable influence on major decisions. 7.21 Finally, in a financial system dominated by a few large influential groups, IFCT is one of the very few nongroup, ind.3pendent, professionally managed, institutions (subject to the limitation noted above). In fact, until the commercial banks' shareholding in IFTC was abstantially reduced, IFCT was in the anomalous situation where its sharehold A.s at times might not have acted in its best interests. Operations. Resources. Financial Record 7.22 Operations. IFCT operations consisted almost exclusively of long- term loans until 1986, when it started also to make medium-term loans (three to five years); in 1988, medium-term loans exceeded long-term loans for the first time. Working capital loans, being limited to long-term loan clients only, remain relatively small. IFCT's loan approvals have grown steadily (except in 1986) and peaked in 1988, as shown in the following table: 53/ The Government has the right to nominate one director who is always by tradition elected the Chairman. - 114 - Table 7.8s IFCT LOAN APPROVALS, 1980-88 (Million baht) Long-term Medium Term Working Capital Total 1980 1,027.5 - 1,027.5 1981 1,208.6 - 1,208.6 1982 1,037.5 - - 1,037.5 1983 1,574.6 - 177.3 1,751.9 1984 2,211.4 - 215.8 2,427.2 1985 3,007.2 - 101.2 3,108.4 1986 1,449.8 129.0 323.9 1,902.7 1987 2,264.5 1,755.2 554.0 4,573.7 1988 2,276.7 2,374.3 530.0 5,181.0 Sources IFCT. 7.23 The portfolio of IFCT is well diversified within the broadly defined industrial sector. The greatest proportion of its loans have gone to agroin- dustries (including food processing, and non-metallic mineral products, which accounted for about 192 and 182, respectively, of the total outstanding port- folio at end-1988; chemicals and chemical products accounted for the third largest share at 92. Geographically, Bangkok and its surrounding areas account for about 352 of all loans. 7.24 Resources. As of end-1988, IFCT's total long-term resources amounted to B 21.9 billion. These were derived from shareholders' funds (B 3.2 bil- lion), foreign borrowings (B 9.2 billion), domestic borrowings (B 5.2 bil- lion), and issuance of debentures (B 4.3 billion). The share of domestic borrowings has increased rapidly in the last two years as a result of a deli- berate shift in policy (because of the foreign exchange risk problem, see para. 7.31-7.39). Most domestic borrowings are in the form of debentures. 7.25 Financial Record and Condition. The most noteworthy feature of IFCT'r cperational record is the decline in its interest margin.541 As the share of low-cost foreign borrowings has declined (ignoring the exchange losses) and emphasis has shifted to domestic borrowings, the interest margins have been squeezed, as the following table shows: 54/ Defined as interest income minus interest cost, as a percentage of earning assets. 115 - Table 7.9s IFCT INTEREST RATE MARGINS, 1980-88 1980 1981 1982 1983 1984 1985 1986 1987 19ea 3.22 3.24 2.35 2.54 1.25 0.43 0.43 0.64 0.47 Sources IFCT. 7.26 The IPCS has always been profitLble although the level of profitabil- ity has declined appreciably since 1984, reflecting the interest margin squeeze.551 The return on equity has declined from the historic peak of 192 in 1984 to a range of 102 - 112 in the last three years. The return on earning assets has averaged 32 during the last three years. For a comparative perspective, all domestic banks as a group had, in 1988, an interest margin of 3.232, return on equity of 10.22, and return on risk assets of 0.66?. IFCT has paid a dividend regularly; in 1988, the dividend rate was 6.8Z having been reduced from 13.52 in the previous two years as a result of actual and poten- tial foreign exchange losses. Summarized balance sheets and income statements of IFCT are given in Annexes 45 and 46 respectively. 7.27 The total assets of IFCT have grown at an average annual rate of aboat 272 during 1960-88 and amounted to B 24.1 billion (about US$950 million equivalent) at the end of 1988. These have been financed by long-ternm borrow- ings (B 14.8 billion), issuance of debentures (B 4.3 billion) and equity (B 3.2 billion). IFCT is conservatively capitalized with a total debt to equity ratio of 6X1 as of the end of 1988. These ratios would, of course, change significantly if IFCT is itself to meet all its actual and potential foreign exchange losses. IFCT's Pioneering Activities 7.28 By far the most successful new activity undertaken by IFCT was the establishment of the Mutual Fund Comnany (MFC) in 1975. The Bank Group's International Finance Corporation (IFC) has been a participant in this company since its beginning. The MFC is still the only mutual fund company in Thailand and is responsible for attracting large amounts of foreign capital into Thailand. Up to the end of 1988, MFC had established and managed a total of 15 mutual funds, both open- and closed-end type, with total funds of almost B 16 billion and nearly 60,000 subscribers. The MF- is also licensed to manage provident funds and as of end-1988, it was managing 85 provident funds amounting to B 427 million for more than 18,000 members. 55! Profitability has been calculated, and certified by IFCT's auditors, after due provisions were made for foreign exchange losses as required by an agreement with the Government but without offsetting total losses, see also paras. 7.32-7.40. - 116 - 7.29 The Thai Factory Development Company (TFD) was established by IFCT in 1977 in a joint venture with the United Kingdom's Commonwealth Development Corporation (DCD) and the Federal Republic of Germany's State Development Agency (DEG). It builds modern factories equipped with basic utilities for sale to investors on a deferred payment basis or long-term lease. The TFD operations have been successful and the company is profitable. 7.30 The IFCT was also the first to introduce equipment leasing in Thailand when in 1978 it set up Thai Orient Leasing (TOL) as a joint venture with the Orient Leasing Company of Japan and IFC. By end-1988, TOL had 550 clients leasing machinery and equipment worth B 410 million, about 40Z of which was in the industrial sector. The TOL is also a profitable operation. 7.31 The IFCT also set up an industrial research and consultant services company, the Industrial ManaRement Company (IMC), in 1977. The IMC's services include the conducting of project feasibility studies, marketing research, system planning, research to identify problems or to improve production sys- tems and the organization of staff training and recruitment services for owners of industrial projects. ForeiRn Exchange Risk Exposure and Losses 7.32 In earlier years, IFCT was used by the Government as a conduit for channeling low-cost funds from abroad into the private sector, because domes- tic funds were insufficient to finance many industrial projects. Much of the foreign debt of IFCT is in the form of concessional loans made to Thailand by foreign governments or their agencies. Half of it is formally guaranteed by the Government of Thailand. The Government was always well aware of the pos- sible harmful consequences to IFCT of its foreign currency exposure, and under the ItCT Act of 1959, the Ministry of Finance entered into an exchange risk agreement with IFCT, whereby the Ministry of Finance would shoulder any realized foreign exchange losses and also take over any realized foreign exchange gains. The way the agreement was constructed, the Government obliga- tion was, strictly speaking, to provide liquidity support to IFCT to enable it to meet the losses and make the repayments on foreign debt installments as they became due. IFCT was to pay a substantial fee for this service. 7.33 At the end of 1988, ItCT's borrowings stood at B 17.6 billion (approximately US$700 million). The distribution by currency was as follows: Baht 48Z Yen l9Z Swiss Franc 6Z Deutsche Mark 7Z Other (negligible US$) 201 Most of the foreign currencies in which IFCT borrowed during the 1970s and 1980s have appreciated relative to the baht, which is valued against a basket of currencies in which the US dollar represents a weighting of about 8O?. As IFCT's assets and earnings are denominated in baht, its foreign currency exposure has resulted in heavy actual and unrealized losses. - 117 - 7.34 At the end of 1988, unrealized foreign exchange losses amounted to B 4.8 billion (approximately US$200 million). Foreign exchange losses (net of provisions) which have been realized, i.e., in which the underlying liability has crystallized or matured, amounted to B 1.2 billion. Until 1984, there were actual foreign exchange gains; these turned into losses beginning in 1985 and increased dramatically after 1986 when the US dollar declined. In 1987 and 1988 the losses amounted to B 705 million and B 885 million, respectively. Since the beginning of 1989, the strengthening of the US dollar has had the effect of reducing the extent of the unrealized foreign exchange loss (by well over B 1 billion per 102 appreciation of the dollar), and has given IFCT the opportunity to swap out of some of its foreign debt on relatively attractive terms. 7.35 Despite the provisions and the intentions of the exchange risk agree- ment entered into by the Ministry of Finance to protect IFCT from the poten- tially damaging consequences of its foreign exchange exposure, the Ministry has accepted liability to compensate IFCT for only the B 240 million of losses to 1978, and not for the full amount of B 763 million which includes the losses incurred after 1978, because of a technicality. 7.36 In 1972 when IFCT Act wras being amended for other reasons, the phrase 'any change in the par value of the baht in relation to any of the borrowed currencies' was substituted for the phrase many change in the valuation of the baht in relation to any of the currencies' which was the expression used both in the original Article 25 of the IFCT Act of 1959 and in the amendment to Article 25 by the IFCT (No. 3) Act of 1963. This has meant that the original intent and agreement could not be followed for losses and gains after 1978, when the fixing of the par value of the baht (a legal term relating to the currency) was canceled via a royal decree in accordance with the law on cur- rency. 7.37 This was the ruling of the Thai Juridical Council in 1985, and again in November 1988. Although the problem has not yet been finally resolved, it seems unlikely that the Government will fully compensate IFCT for the losses actually incurred, although the Ministry of Finance has reiterated its support for IFCT. 7.38 However, even if IFCT gets no further help from Government, it should be able to survive as a result of a number of operational and strategic measures it has taken to manage its outstanding foreign currency debt: a currency swap in 1983 from yen to Swiss francs, reducing the unrealized exchange loss by approximately B 200 million; a maturity swap in 1986 out of a six-year sterling obligation, resulting in a gain of approximately B 53 mil- lion; prepayment in 1987 of the counterpart funds of the World Bank Structural Adjustment Loan; and defeasance of part of its yen obligations by the purchase of yen at a favorable rate. In addition, it is looking at forward purchases, refinancing, and the restructuring of repayments. In 1986, the reserve for foreign exchange risk was also increased from 1.12 to 1.62 of borrowings, raisin; the reserve by approximately B 20 million per year. During 1985 to 1987 ths special reserve was increased by setting aside 5 of each year's profit, amounting to approximately B 49 million. In 1988, IFCT increased the - 118 - reserve for foreign exchange risk from 1.6Z of all borrowings to 2.5? of out- standing loans, thus raising the reserve by approximately B 80 million per year. 7.39 A strengthening US dollar would help IFCT, as nearly all of its US$400 million equivalent of long-term foreign debt is denominated in curren- cies other than the US dollart the Japanese yen, the Deutsche mark and the Swiss franc together account for 561 of the outstanding foreign debt. The combined effect of the appreciation of the 'IS dollar and all the measures taken by IFCT has been a steady reduction in unrealized foreign exchange losses from B 6.32 billion in 1987 to B 4.83 billion in 1988, B 3.43 billion in June 1989, and B 3.39 billion on August 7, 1989. 7.40 With regard to new debt, IFCT's strategy since 1987 has been to switch the emphasis to domestic funds, raising approximately B 3.73 billion, while only B 1.3 billion equivalent was raised from foreign sources. Also, in 1988, IFCT borrowed pound sterling 15 million which was immediately swapped with local commercial banks into baht, thus effectively mobilizing approximately B 692 million from foreign sources. Key Issues. Future Role and Outlook 7.41 The outlook for the financial system as a whole is for increasing liberalization, competitiveness and efficiency. The major institutions, com- mercial banks as well as finance companies, are already multifunctional and, in an increasingly rational and liberalized system, are expected to fight fiercely for all business opportunities. Can IFCT survive in such an open competitive financial system? The answer must consider IFCT's strengths and weaknesses, and take into account its short-term as well as long-term problems. In the process of answering the question, the justification of the support the Government provides to it must also be questioned, and account taken of the costs it incurs as a quid pro quo for such support. 7.42 The question of IFCT's survival in a narrow sense of assets and liabilities must be settled at the outset because without an assured short- term future, the question of long-term survival and future does not arise. The short-term aspect is particularly important because of the overhang of the foreign exchange risk and loss problem. As stated earlier (para. 7.34), actual foreign exchange losses are now running in the range of B 700-800 mil- lion annually. The Government did not dispute IFCT's claims for reimbursement amounting to about B 240 million in 1989. This means that if the Government and IFCT are unable to settle their dispute amicably, IPCT would have foreign exchange losses of about B 450-550 million annually. IFCT's net profit in 1988 amounted to B 352 million after a piovision of B 225 million for exchange losses. On a rapidly increasing portfolio, IFCT's profitability in the future should also improve over 1988. On a pre-provision basis, therefore, IFCT's level of profits in the future can be expected to be sufficient to meet the entire burden of foreign exchange losses if the Government is unable to help IPCT. 7.43 The operational cash flow aspect should also be considered, however. In 1987 and 1988, IFCT's operational cash flow amounted to more than B 460 - 119 - million annually. Therefore, from an operating cash flow point of view, too. IFCT should be able to survive with small borrowings or liquidation of assets for the purpose of meeting the foreign exchange losses. Sacrifices will be needed (e.g., dividends will need to be cut to conserve cash) but IFCT would be able to survive. It should be noted that the foregoing, effectively, sketches the worst possible scenario. If IFCT can survive this, in actual practice it would in all probability fare better. 7.44 Still considering the narrow aspects of assets and liabilities but shifting the focus to the longer term, the situation is as follows. As of January 1, 1989, total potential (inrealized) foreign exchange losses were estimated at B 4,830 million, having declined from B 6,320 million a year earlier. A strengthening US dollar had the effect of reducing the potential loss to less than B 3,400 million by August 1989. Against these potential losses, the book value of IFCT's total equity was B 3,257 million as of December 31, 1988. But it is understood that many of IWCT's assets--real estate (i.e., its own office building) as well as financial assets (i.e., many of its equity investments, including its affiliated companies)--are substan- tially understated. While the potential foreign exchange losses weigh heavily on IFCT's future, in actual practice, its long-term solvency is not really in doubt because of its actions during 1989 to substantially strengthen its capital resources. In any case, the foreign exchange losses wi'll materialize only gradually as loan repayment installments become due. Therefore, both from a profitability and operating cash flow point of view, IFCT is likely to be able to meet its obligations. 7.45 Perhaps the real question, then, is not about IFCT's solvency or viability in the narrow assets/liabilities sense, but should be posed in terms of its strengths and weaknesses in a competitive finncial system. 7.46 Its staff is its greatest asset. It is highly qualified and experienced in the techniques of project-based medium- to long-term lending. IFCT's reputation in the financial community as a professional institution is also an asset; in fact its reputation is now international. It is a well respected institution in Thailand and abroad. 7.47 IFCT's record if almost 30 years is another major asset. It has operated prudently, pro,; itably and without scandal. The quality of its assets is considered good (in fact there are substantial hidden reserves). A long- unblemished record for a financial institution must be considered a great strength. 7.48 IFCT is among the very few large financial institutions in Thailand which are not owned/controlled by a family group. This is both an advantage and a disadvantage. It affords independence, autonomy and enables IFCT to have truly professional management and staff. On the other hand, IFCT does not possess the political clout that large family groups have; in fact IFCT becomes vulnerable when its interests clash with those of family group-related financial institutions which have considerable political influence. 7.49 IFCT*s special relationship with the Government has also proven to be a double-edged sword. Tax free status, government guarantee of its foreign - 120 - borrowings, use of its bonds as commercial banks' reserve assets are by no means small privileges. These have to be balanced by the constraints and costs discussed earlier in para. 7.18. It should also be noted that Govern- ment help has not been readily forthcoming in meeting the very substantial foreign exchange losses being sustained by IFCT. Quite a few foreign exchange borrowings (in hard currencies) were government-to-government credits for which IFCT became the Thai Government's designated conduit. It is a moot point now to speculate whether IFCT would have contracted these loans on its own. But the most serious negative aspect of IFCT's special relationship with the Government is the risk of its politization. This risk is unfortunately likely to increase unless it is consciously recognized and steps are taken to avoid it. If not, IFCT will have lost its major assets, i.e., independence, autonomy and professionalism. 7.50 It is feared that if status quo is maintained, IFCT could drift without a clear mandate or conception of its role and destiny, with a tendency to gradually come under increasing government control with resultant politiza- tion. This would inevitably result in a gradual decline in the morale of management and staff and ultimately in its professionalism and autonomy. 7.51 Thailand is currently characterized as an emerging newly industrial- ized country. Given the broader regional picture of East and Southeast Asia, the Thai economy is expected to do well in the future which means the recent high levels of investments will also be maintained. This in turn would result in continuing high demand for capital, particularly long-term funds for fixed investment. IFCT has the clear market niche of providing long-term investment capital. With the size of its market expanding, IFCT's future from this point of view appears assured. It should try to build on this strength. 7.52 To make its future role and orientation clear, the Government and IFCT should explicitly decide on a number of measures. These relate to its future activities and functions, and would have the effect of transforming IFCT into a broad based finance company. These measures would also enhance its autonomy. They are discussed in the following paragraphs. 7.53 The "'arket' for IFCT may be divided into three segments: companies that are affiliated with family groups controlled by large financial institu- tions, independent companies not affiliated with bank groups, and state or quasi-state enterprises. The second category of clients, the independents, is the natural IFCT constituency. With the overall growth of the economy. this group will also grow, and it may be that its size would be sufficiently large to support IFCT's own growth sufficiently. However, IFCT may already be losing some clients from this category because IFCT does not offer a full range of financial services. So as not to lose its market share in this seg- ment, IFCT should seek to offer a broad range of services, and not just long- term loans. This is discussed further in para. 7.56. 7.54 However, IFCT should not voluntarily abandon the other two business segments. While companies affiliated with bank groups will not easily become IFCT clients, IFCT should fight for business on a comparative advantage basis (in long-term lending). As for state and quasi-state enterprises, these constitute a sizeable segment of the Thai economy and their investment re- - 121 - quirements are huge. Most of them are well run, and lending to them would represent a safe and profitable business. IFCT should seek to remove the legal and policy restriction on its lending to state enterprises; the result would be a substantial enlargement of its market. 7.55 IFCT should also seek the same objective of market enlargement by broadening the sectoral coverage of its activities. In its history so far, IFCT has concentrated on the broadly defined (manufacturing) industrial sector. In the future, IFCT should not neglect the industrial sector, but there is no reason why it should ignore other sectors, e.g., power, transpor- tation, etc; they are also "developmental* and offer good lending oppor- tunities. 7.56 IFCT is also under legal and policy restrictions to undertake many kinds of business activities. For example, it is not allowed to accept deposits and does not make working capital loans (except to its long-term loan clients). IFCT should seek to remove all legal restrictions on its business activities contained in its Act. It may decide, for policy reasons, to refrain from undertaking certain activities. It is suggested that it should accept large time deposits (but not current deposits where it would be at a competitive disadvantage with commercial banks because of their branch net- works) from the general public as well as from corporations; it should also issue large denomination certificates of deposit and promissory notes. On the lending side, it should offer a broad range of services, certainly including working capital loans. 7.57 IFCT should offer to give up its tax exempt status. It should at the same time also discontinue those activities which are clearly loss-making. Should the Government consider any such activity to be of very high priority from a social point of view and would wish IFCT to continue it, IFCT should offer to do so on behalf of the Government on payment of a fee (including a reasonable profit margin) and on an off-balance sheet basis. 7.58 It is suggested that the Government maintain its shareholding in IFCT. However, it would be better to consolidate its holdings into all direct investment by the Ministry of Finance. This would imply that the Ministry would buy out the Krung Thai Bank. This is suggested to avoid potential conflict of interest between IFCT and the Krung Thai Bank, as IFCT expands its activities to include many new banking activities not now undertaken. IFCT should at the same time seek to reduce the shareholding by commercial banks as a group. As stated in para. 7.6, commercial banks' holdings have already declined from 70Z of total share capital in 1975 to 33? at the end of 1988; this decline is welcome and should continue. This implies that when new capital is issued in the future, IFCT should seek to pre-empt a more than proportionate share for new and existing nonbank shareholders. A shareholding by the Government of up to 25? is expected to prove a stabilizing influence and safeguard IFCT's status as an independent non-group-dominated financial institution. The government investment would continue to be justified on a financial yield basis as IFCT is a profitable and dividend-paying institution. 7.59 While the government may continue to appoint one director and the tradition of electing the government nominated director as the Chairman of the - 122 - Board of Directors may also be continued, the election of the President by the Board should not be subject to subsequent approval and confirmation of the government. Such a change would be a strong signal of the government's inten- tion to let IFCT operate with more autonomy. 7.60 Since the Government will continue to hold 25? of ZFCT share capital, it could offer to guarantee IFCT's borrowings, in domestic as vell as in international markets. However, this should not be for free; the Government should charge an appropriate, mutually negotiated fee for assuming a contin- gent liability. 7.61 Once a settlement has been reached between IFCT and Government on the foreign exchange risk agreement in respect of past foreign currency borrow- ings, the agreement should be terminated. This arrangement has not worked and may have proved more harmful (by giving IFCT a sense of protection which it now finds does not exist) than helpful. Better techniques to manage foreign exchange risk are now available, and IFCT itself has recently demonstrated its adeptness in using them. 7.62 More research and legal opinions will be needed to ascertain whether the new course suggested by the foregoing recommendations can be effected within IFCT's present corporate set up, by amendments to the IFCT Act. If not, it may be necessary to abrogate the IFCT Act, and reincorporate IFCT as a finance company with license(s) to engage in broad-based activities. 7.63 Once basic decisions have been made along lines suggested above, IFCT would need to prepare a two to three year phased plan to 'Implement changes. The most challenging part would be to change the corpo.ate culture. Fortui- tously, both the Chairman and President have recently resigned so that the new management would have the opportunity to make a fresh start.56/ It may be necessary to reduce the total staff substantially (when such activities as small-scale lending are given up). At the same time, it may also be necessary to recruit some new persons at senior levels to oversee the expansion into new activities. 7.64 It should be possible for IFCT to grow from within to become a broader-based financial services institution. This would be a phased, gradual approach. Alternatively, IFCT may consider acquiring/merging with an existing finance company to hasten the transformation. 56/ A new President of IFCT was appointed in March 1990. - 123 - VIII. AGRICULTURAL/RURAL CREDIT AND THE BANK FOR AGRICULTURE AND AGRICULTURAL COOPERATIVES A. Airicultural Credit Sector ARriculture Sector in the Thai Economy 571 8.1 About 44 million of Thailand's 55 million population lives in the rural areas, of which nearly 35 million are involved in agriculture. Agricul- tural GDP (in constant prices) averaged an impressive 6.32 p.a. during the period of the first three Five-Year Development Plans (1961-1976) compared with 7.92 for the overall economy. The main factors responsible for the growth of agriculture were the presence of surplus land, extensive road construction, and a dynamic private sector which was responsive to oppor- tunities for production of, and world trade in, primary commodities. The construction of multi-purpose dams and irrigation facilities in the 1950s and 1960a also helped. However, by the late 1970s, as expansion neared the limits of the land frontier, agricultural growth slowed to an average of 3.52 p.a. and further to 2.62 p.a. in the early 1980s. Nevertheless, as agriculture began to switch from extensive to intensive agriculture, the 19809 saw fertilizer use expand by 132 p.a., sprayers by 23Z p.a. and tractors by 122 p.a., reflecting the fast rate agriculture was being commercialized. 8.2 In addition to increased commercialization, the growth of Thai agri- culture and the relatively even distribution of its holdings also reduced the incidence of poverty and provided the basis for rural industry and employment growth. Absolute rural poverty declined from 611 in 1962/63 to 362 in 1975/76 and then to 272 in 1980/81. In 1985/86, however, poverty incidence increased to 35Z. Although, agriculture now contributes only 172 to GDP, it accounts for nearly 552 of total export earnings and 612 of total employment. For the agricultural household family, income from non-farm sources is now growing at a faster rate than agricultural income; non-farm income now accounts for about 602 of total cash income of farm families compared with about 552 in 1975. Trends in Rural Finance 8.3 The formal sector providing credit to the agricultural and rural economy consists essentially of the commercial banks, the Bank for Agriculture and Agricultural Cooperatives (BAAC) and some finance companies. Most of the agricultural cooperatives which provide credit to their members obtain their loans from BAAC. Agricultural lending increased rapidly after the mid-1970s when the new government took over and wanted, for socio-political reasons, to 57/ For more details on Thai agriculture, see (a) World Bank Report No. 3705- TH, Program and Policy Priorities for an Agricultural Sector in Transition, Volume I, December 3, 1982; (b) World Bank Report No. 3847-TH, Thailand Irrigation Subsector Review, April 25, 1986; and (c) Kajonwan Itharattana, Review of Aaricultural Economic Situation in Thailand; ASEAN Agricultural Development Planning Center, 1988. - 124 - substantially increase resource transfer to rural areas. To achieve its objectives, it adopted the following policies which are essentially still in effects (a) commercial banks were subjected to various mandatory agricultural lending targets (see paras. 8.11-8.14 below), and rural branching was 'encouraged'; (b) interest rates on commercial bank credit to agriculture were controlled at levels equivalent to the market rates for urban borrowers, and far below informal market rates in rural areas (see paras. 8.15-8.17 below); (c) The B0T expanded its rediscount facilities directed toward agricul- ture (see paras. 8.18-8.19 below); and (d) BAAC was given the mandate to substantially expand its lending (see paras. 8.23-8.29). As a result of the above-mentioned policies, agricultural credit from institu- tional sources increased rapidly, as shown in Annex 47 and summarized in Table 8.1 belows Table 8.1: AGRICULTURAL (FARM) CREDIT, BY INSTITUTIONAL SOURCE Volume (Million Baht) Lending Share (Z) Z Av. Annual Growth 1978 1980 1988 1978 1980 1988 1978-80 1980-88 Commercial Banks 8,657 12,588 57,184 46.2 49.0 67.4 20.6 20.8 BAAC 9,703 12,464 26,400 51.7 48.5 31.1 13.3 9.8 Other 394 638 1,306 2.1 2.5 1.5 27.2 9.4 TOTAL 18,754 25,689 84,890 100.0 100.0 100.0 17.0 16.1 Sources BOT. 8.4 The growth in commercial banks' agricultural lending between 1980-88 (21? p.a.) was even higher than the rate of growth in their overall lending (18? p.a.) and nearly twice the 10? p.a. rate for BAAC. These trends reflect the greater participation of the large farmers (the commercial bank clientele) in the commercialization process compared with BAAC's clients who were (and still are) the medium and small farmers. Although quite dynamic in the finan- cial system, other institutional sources (essentially the Finance Companies) remain rather insignificant as agricultural credit lenders. - 125 - 8.5 A remarkable feature of the Thai rural credit scene is the demarca- tion of total clientele in terms of who meets their credit requirements.58/ There is hardly any overlap between the borrowers from formal (BAAC, commer- cial banks) and informal credit sources. Even between BAAC and the commercial banks, there is a clear demarcation of clientele; only about 52 of the 500,000 clients of commercial banks (with loans averaging B 106,000 per client) also borrow from BAAC (whose clients borrow an average of about B 14,000 each). The informal credit sector (largely comprising of individual traders, millers, other farmers and relatives) typically lends smaller amounts per client (about half that of BAMC) and does not typically compete with the formal sector. Various estimates also exist in terms of the outreach of the various credit sources (see Table 8.2 below). Table 8.2: SHARE OF FORMAL AND INFORMAL CREDIT FOR AGRICULTURE (in percent) Credit Source Loan Value No. of Clients No. of Loans 1988 1985 1988 1980/81 1986/87 Informal 20.0 la 49.0 /a 41.4 la 42.1 29.6 Formal 80.0 51.0 58.6 57.9 70.4 Comprising: Commercial Banks 53.9 7.5 9.8 11.5 10.5 BAAC 24.0 43.5 48.8 46.0 59.2 - Individuals (n.a.) (26.1) (31.7) (26.7) (45.5) - Coops/Assocs. (n.a.) (17.4) (17.1) (19.3) (13.7) Other 1.2 n.a. n.a. 0.4 0.7 /a Imputed from total number of farmers and assuming no leakage in the volume and coverage of the loans to commercial banks. Source: Annexes 48 and 49. 8.6 Among the institutional (formal) c.ddit sources, commercial banks account for two-thirds of the total agricultural credit volume (1988). But in terms of farm families reached by the lending institutions, commercial banks account for only 17? of the 3.1 million families who borrow from institutional 58/ This is brought out by studies undertaken in six regions of Thailand with emphasis on the Northeast, and are contained in Ammar Siamwalla et.al, The Thai Rural Credit System: A Descrivtion and Elements of a Theory, a paper presented at the World Bank Conference on Agricultural Development Policies and the Theory of Rural Organization, June 14-16, 1989 (Annapolis, Maryland). Field work (in September 1989) by the mission preparing this report confirmed these findings. - 126 - sources. The BAAC, on the other hand, accounts for less than one-third of institutional credit by volume but reaches more than 802 of institutional clients in agriculture (about half of all 5.3 million agricultural borrowers). The informal sector is estimated to account for about 402 of all agricultural clients but only 15Z of total loan volume. Both the share of clients and especially the share of total number of loans accounted for by the formal sector, have increased substantially over the years. Compared to an outreach of only about 152-202 for institutional credit prior to 1975, the formal sector now lends to nearly 602 of all agricultural families. In effect, since the late 19708, the informal sector has lost ground to the formal sector, essentially because of the increasing competitiveness of BAAC. 8.7 Even with rapid commercialization of agriculture in Thailand, the cost of providing rural credit remains higher than lending for other sectors. Average loan amounts are generally much smaller because farm holdings are small and the clients are also more geographically scattered than those for the industrial or service sectors. Furthermore, given the typical farm enter- prise (rice farming), agricultural sector incomelcapita is only about one- tenth that for non-agriculture in 1986. Lending risks are also higher because of agriculture's dependance on the vagaries of nature (drought, floods) and the greater volatility associated with its prices. 8.8 But the strongest contributor to the high cost of lending to agricul- ture in Thailand is probably the lack of legal rights to land attachable by the lender in the event of loan default. Generally all long-term lenders use land as collateral. However, land titles and land documents accepted by credit institutions (NS-3 or NS-30) have so far been issued to only about 571 of the area under agricultural holdings. Loan recovery risks, therefore, have to be minimized either through the control of borrowers' output or by lending through groups where joint liability can be arranged. Given this collateral related problem, the informal sector normally opts for controlling output as the main alternative (especially since they are often traders who also supply the farmers' inputs). Commercial banks, on the other hand, opt for lending to large farmers who can offer clean collateral, or to farmers whose primary source of income is non-agricultural.591 The BMAC has focused the bulk of its lending efforts on joint liability groups of medium to small farmers (whose primary source of income is agricultural) rather than lending to single individuals or through cooperatives or farmers associations. 8.9 Partly as a reflection of the costs and risks, the purpose for borrowing and, in the far outlying areas, credit monopoly, the interest rates charged by informal sector lenders tend to be the highest (typically 52 per month). Interest rates charged by BMAC range between 11.52 (long term loans) and 12.5? p.a. (short term) compared with the 132-14.5Z rate for commercial banks. Surveys of the informal sector also indicate a partial market failure in the system: working-capital for crop production is not available fror 'he informal sector during low-rainfall or potentially bad harvest years. T. is 59/ See: Siamwalla, Ammar, et al, OD cit; and Satsanguan Ploenpit, The Role of the Commercial Banks in the Provision of Agricultural Credit, TDRI Discus- sion Paper, ARD 4, 1988 (In Thai). - 127 - partly why BAAC, which also lends without land collateral through joint liabi- lity groups and does not withhold credit during adverse weather years, is preferred by its clients who try to maintain a good repayment performance with it. Distortionarv and Allocative Effects of Government Credit Policies 8.10 The sources of distortions or policy issues in agriculturallrural credit revolve around three broad areas: (a) mandated lending for commercial banks; (b) interest rate structure, subsidies and target group programs; and (c) rediscount facilities of BOT. 8.11 Mandated Lendina for Commercial Banks. The most prominent vehicle to achieve government's goal of expanding credit for the agrtcultural sector is to require commercial banks to lend a percentage of their deposits to this sector. It began in 1975, when banks were mandated to allocate five percent of their previous year's lending; shortfalls were to be deposited with BAAC. Changed to 71 of the previous year's total deposit in 1976, the rate gradually increased to llZ by 1978 when an additional 21 lending for agribusiness/agro- industries was also mandated. These mandated rates remained unchanged until 1987, when they were increased to 201 (14Z for direct lending to the rural sector and 61 for agribusiness). The new definition of directed lending" to the rural sector allowed the commercial banks to include in the quota loans to small scale rural industries (SSRI's) and to finance the cost of rural inhabitants' job search (and installation) in the Middle East. Table 8.3: MANDATED TARGETS AND ACTUAL LENDING OF COMMERCTAL BANKS FOR AGRICULTURAL SECTOR, 1975-88 (Amounts in million iht) Mandated Lending Deposit Actual Lending Tarset Shortfall with BAAC Amt. Z of Dep. Amt. Z of Dep. Amt. S of Dep. Amt. I of Dep. 1975 2,234 3.2 4,333 6.2 a/ 2,099 3.0 1,682 2.4 1978 8,100 6.2 11,771 9.0 3,671 2.8 5,511 4.2 Av.1979-81 12,029 6.6 20,060 11.0 8,031 4.4 7,045 3.9 Av.1982-84 28,233 8.5 35,988 11.0 7,755 2.5 8,915 2.8 Av.1985-86 38,907 7.5 57,084 11.0 18,177 3.5 10,899 2.1 1987 55,864 9.0 86,932 14.0 31,068 5.0 12,340 2.0 1988 73,211 9.8 104,111 14.0 30,900 4.2 14,119 1.9 1989 bl 74,752 n.a. 123,809 14.0 49,057 n.a. 14,397 n.a. la Target was 5Z of previous year's loans rather than of previous year's deposit. /b As of August 31, 1989. Source: BOT. - 128 - 8.12 Since the inception of mandatory lending, commercial banks' "directed lending' has always fallen short of government's targets (see Table 8.3); they have always achieved, on the other hand, the target for lending to agri- business (see Annexes 50 and 51). Despite the measures taken to broaden the definition of direct lending in 1987, the shortfall (as a proportion of pre- vious year's total deposits) has continued to increase because the contribu- tion of SSRI lending and financial support for Middle East employment have not been able to keep up with the increased deposits and consequently the ever increasing level of mandated targets. 8.13 The basic reason for the shortfalls is the fact that many of the smaller commercial banks have very few branches in rural areas. However, even where banks have such branches, other mandatory requirements of BOT have com- pounded the problem. First, 602 of the deposits mobilized at each branch has to be lent in the local area. Second, a third within this 60? (i.e. 20X) has to be for direct lending to agriculture. If these conditions are not met, all shortfalls have to be deposited at BOT without interest. Because the establishment of full service branches in the more outlying provinces is a costly business, many branches have been primarily set up as deposit-taking offices even if it meant, in effect, that they were mobilizing funds for BOT. The commercial banks see this as the price they have to pay to be permitted to set up other branches in the more lucrative urban areas. 8.14 The impact of the mandatory lending policy is hard to quantify (see also para. 8.17). While such lending may have resulted in a number of agri- cultural or rural investments which might not have otherwise materialized, it is questionable whether such high mandated rates are presently needed. First, agribusiness targets mandated for commercial banks, having been always exceeded, should be eliminated. Second, permi'ting SSRIs to be counted as part of the 'direct lendingw quota may not necessarily result in substantial expansion of such lending because reclassification of existing loans into this category is permitted. In any case, the availability of good SSRI projects has been declining thus limiting the scope for such loans. The risk is that commercial banks may be pushed into funding marginal SSRIs. Third, with relative scarcity of credit for urban investments, the opportunity cost of forced lending in agriculturelrural sector will be even higher. A better policy stance would be to encourage the commercial banks to expand their rural lending in as cost efficient a manner as possible but without setting mandated targets. In any case, the 202 mandated use of locally mobilized deposits for agriculture (para. 8.13) should be eliminated or at least defined more flexi- bly. Such a move would, for instance, enable the branches in the Bangkok Metropolitan Region (where agriculture constitutes a very small proportion of value added) to make more efiective use of their loanable funds. 8.15 Interest Rate Structure and Subsidies. Since 1975, there are no restrictions on the rate commercial banks may charge on their agricultural loans except that the rate has to be within the overall ceiling (presently 15Z p.a.). Interest rate ceilings for BAMC loans are established by its Board and the current ceiling is 12.5? p.a. for short term and 11.52 for long term loans. It charges lower interest rates on loans for special projects (usually sponsored by the government). In practice, commercial banks presently lend at below-ceiling rates of between 13Z - 14.52 p.a. (differentiating the rates by - 129 - size of customer or loan rather than by the sector). Given the high risks and costs of agricultural lending (para. 8.7), maintaining uniform rates for all borrowers/sectors has meant, in effect, that commercial banks have cross-sub- sidized their agricultural loans. BAAC, on the other hand, receives about 182 of its sources (1988) at subsidized rates and can thus afford financially to charge lower interest rates than commercial banks and still remain at least marginally profitable. Economically, it is hard to justify BMAC's lower rates. 8.16 Interest rate ceilings for the commercial banks have served to restrict the coverage of their clients in the sector. They have essentially penalized potential clients who may be willing and able to pay higher rates (although not at the 52 per month rate generally charged by the informal sector) for riskier, but potentially viable, investments. In the non-agricul- tural sectors, higher risk investments have been able to obtain loans from Finance Companies which can lend at an interest rate of up to 181, but such options are practically non-existent for agricultural borrowers. 8.17 The quantification of explicit and implicit subsidies accompanying the substantial expansion in institutional credit to agriculture over the past decade, has to be indicative at best. It is still useful, however, to do so. If the commercial bank rate (14.5Z) were to be taken as the unsubsidized rate, the subsidy value (based on a 2.64Z difference with its weighted average interest rate charged) would have been B 875 million in 1988. Another way of looking at the subsidies for BAAC's lending would be to separately examine the cost of subsidies posed byt (a) the mandatory requirement--which implies an opportunity cost for higher earnings from alternate lending (paras. 8.11 and 8.14); and (b) the use by BAAC of an average-cost pricing rule based on the average rate of all its sources of funds (including subsidies) in setting its lending rate instead of a marginal cost pricing rule based on an incremental source of funds at the market rate (e.g., 12 month deposit rate). The results are given in Annexes 52, 53 and 54, and summarized in Table 8.4. - 130 - Table 8.4s ESTIMATION OF SUBSIDIES IN AGRICUALTURAL LENDING, 1983-87 (million baht) Type of Subsidy 1983 1984 1985 1986 1987 Av.1983-87 Mandated lending /a 61 75 80 67 88 74 Implicit subsidy from average pricing rule of BAAC lb 1.155 1.157 668 390 507 775 Total 1,216 1,232 748 457 595 849 Ls Assuming 502 leakage (i.e., only half actually reach the agricultural sector). Lb Assuming the administrative cost of deposit-taking to be 0.75Z. The average annual subsidy for 1983-87 (B 849 million) is little different from the initial method of subsidy computation, but it illustrates the rela- tively minor interest-rate distortion impact of mandated lending (average B 74 million) compared with that arising from BAAC's average pricing policy (B 775 million, including the implied subsidies from commercial banks' deposits with BAAC). Expressed in terms of subsidy amounts per client (B 298 or US$12) the interest rate subsidies may be considered small. 8.18 Rediscount Facilities of BOT. Since 1967, BOT has been authorized to rediscount promissory notes arising out of agricultural transactions. Although other rediscount facilities also affect agriculture (eg. export promotion), from the total allocation of about B 42 billion in 1989, about 25S has directly benefitted agriculture. The rediscounted loans were for (a) agricultural production/husbandry, (b) agricultural product trading, and (c) Rural Development credit refinancing. The first two rediscount credit facilities are usually short term and the third is medium to long term in nature. For agricultural production, the banking system (including BAAC) may have 50? of their loans rediscounted at 5? p.a. and the maximum rate the banking system may charge the subborrowers is 102 p.a. For loans for agricul- tural product trading, the rediscount rate is 4? but the maximum lending rate remains 10?. For these two facilities, a maximum limit per banking institu- tion is reviewed every six months. The rural development facility, started in FY 1988189, is essentially an investment project financing facility with 50? of the promissory notes (up to five years) drawn by the project owner to be rediscounted at a maximum rate of 32 p.a. and relent at a maximum rate of 10? p.a. Details of BOT rediscount facility are given in Table 8.5. - 131 - Table 8.5 : OT RDISCOUNT FACILITIES FOR AGRICULTUREIURAL DEVELOPMENT (Amounts in billion baht) Annual Disbursement 1989 Pacility Allocation 1980 1985 1986 1988 1989 Total Amt. Amount per Type of Facility Client la Ag. Production 0.88 0.51 0.68 1.24 1.72 1.50 1.61 - Animal Husbandry (0.88) (0.51) (0.58) (0.63) (0.82) (0.60) (4.0) - Sugar cane l_) C-) (0.10) (0.61) (0.90) (0.90) (0.012) Ag. Product Trading 6.62 (-) 1.00 6.09 4.22 n.a. 11.0 - Paddy (n.a.) (-) (n.a.) (6.09) (4.22) (5.00) (11.0) - Others (n.a.) C-) (n-a) C-) C-) (n.a.) 1-) Rural Development - - - - - 2.00 n.a. la In million baht. Source: BOT. 8.19 The objectives of these facilities vary from improving agricultural production (sugar cane), providing cheap sources of animal protein for the general population (animal husbandry), product price and income stabilization to rural development. The need for the facility to provide about B 4.0 mil- lion per client for animal husbandry support, now that low cost meat (prima- rily broiler chicken) is widely available, is questionable. Similarly, short- term price stabilization of fresh oil palm (under agricultural product trading support) runs the risk of being a long-term price support program because most investments assumed they would receive the high aomestic prices rather than the lower world market price which would prevail once domestic demand has been catered for. The need for a new rediscount facility for rural development operated by BOT is also questionable. The BOT appears, more and more, to be assuming the functions of a direct lender or that of the Ministry of Agricul- ture and CooperativeJNESDB in its mobilization of staff to undertake the fea- sibility assessment of the projects submitted for financing under this facil- ity. The projects being considered (eg. production of dairy production, chicken, tiger prawns, rice, baby corn and cashew nuts) are the usual projects which can be handled by the commercial banks and BMAC independently and should, in any case, not be singled for additional subsidy support. Given these concerns, BOT should re-examine the objectives and performance of agri- cultural rediscount facilities and its appropriate role in their operation. - 132 - B. The Role and Performance of MAAC Functions, Organization and Management 8.20 BAAC was established in 1966 under the Bank for Agriculture and Agricultural Cooperatives Act (1966) as a government-owned bank to provide financial assistance to the agricultural producer, either directly or through agricultural cooperatives and farmers' associations. Loans are based on regu- lations which specify inter alia, the type of borrowers, the purposes of the loans, repayment periods, loan ceilings and interest rates. Authorized share capital of BAAC is presently B 4.0 billion of which the amount paid in is B 2.1 billion (about US$82.2 million). Under the BAAC Act, individual farmers, agricultural cooperatives, farmers' association, and other financial institutions are allowed to hold up to 102 of the capital; in practice, the Ministry of Finance holds 1002. The policies and activities of BMAC are con- trolled largely by the Minister of Finance through a Board of Directors con- sisting of eleven members appointed by the Council of Ministers. The Chairman is the Minister of Finance, who is in charge of implementing the BAAC Act. The Vice Chairman is usually a representative of the Ministry of Agriculture and Cooperatives and the BMAC President is the Secretary of the Board. The main responsibility of the Board is to formulate the operating policies of BAAC and to control and supervise its activities. 8.21 In addition to its head office in Bangkok (with nearly 171 of its 6,900 staff), MAC has 71 branches, 23 subbranches and 584 field units all over the 73 provinces in the country--the most widespread among banking institutions. The staff are generally well qualified, and of good caliber. The BAAC is free to establish its own staffing policies including recruitment, salary structure, staff promotion, and discharge. The salaries and allowances of B MC staff are somewhat above those of the regular civil service but below the commercial banks' levels. Promotions are dependent upon yearly perfor- mance evaluation of branch personnel on the basis of their contribution to profit, deposit mobilization, loan disbursements, loan repayments and opera- ting costs. On the whole BMAC staff are generally dedicated and hardworking. 8.22 Overall, BAAC is organizationally sound, and well managed. However, as BAAC has expanded and is further expanding its activities in terms of number of branches and field units and the volume and breadth of its banking activities, it should consider substantial decentralization of its organiza- tional and decision-making structure in order to provide better service to its clients. More specifically, BAAC should consider tos (a) upgrade the field offices, (particularly those which are remote from their supervisory branches) by permitting them, within certain limits, to handle certain cash transaction activities; and (b) establish as a first step, four to six regional management offices with the responsibility to manage the branches in their respective areas. These steps and a reorganization of the Head Office are already under BAAC management consideration. - 133 - ODerations 8.23 The BAAC lends to: (a) farmers directly (about 872 of total disbur- sements in FY88), and (b) cooperatives and farmers' associations mostly for onlending to their members (132). Farmers who are members of the coope:atives or farmers' associations are not entitled to borrow directly from BMC. The share of farmers' associations in BAAC's disbursements has now declined to insignificant amounts (see Table 8.6). Table 8.6: DISBURSEMENT OF BAAC LOANS, 1976-88 (Amounts in million baht) Fiscal Direct Lending. By Term Onlendina Throuah Total Years Short Medium Long Subtotal Cooperatives Farmers Assoc. Disbursement (Amt) (Amt) (Amt) (Amt) 2 (Amt) (2) (Amt) (Z) (Amt) (Z) 1976 1,951 1,160 123 3,234 77 691 16 282 7 4,207 100 1980 4,136 1,254 560 5,950 77 1,700 22 88 1 7,738 100 1985 10,919 1,222 1,984 14,125 82 3,113 18 25 17,263 100 1986 13,988 83 1,746 15,817 82 3,565 18 23 - 19,405 100 1987 15,010 4 2,085 17,099 82 3,808 18 21 - 20,928 100 1988 15,4951a 1.204 3,364 20,063/b 83 4,054 17 21 - 24,138 100 Av. Annual Growth 1976-80 20.72 2.OZ 46.1Z 16.52 25.2Z -25.32 16.52 1980-88 18.0S -0.5S 25.12 16.4Z 11.52 -16.4Z 15.32 Source: BAAC. 8.24 Short Term Loans. Short-term loans account for more than 702 of BAAC's portfolio. They fall under three main categoriest (a) production loans, which now account for about 952 of BMAC short-term lending (particu- larly for rice, 60Z); (b) storage loans for assisting farmers with their cash- flow in the event that they are unwilling to sell their products in a depressed market; aad (c) cash credit lines. Individual borrowers who are unable to provide collateral are required to join the joint liability groups, to ensure favorable repayments through socio-economic pressure from its members. The minimum group size is five and the maximum is 30. The maximum loan amount per group member is B 60,000 (US$2,350). Most of BAAC's short- term loans are unsecured relying mainly on the collective guarantee of the farmers' groups, a system which enabled it to expand its outreach without significantly jeopardizing its repayment record. 8.25 Term Loans. Medium- and long-term loans are mainly for purchase of farm machinery and equipment, draught animal, planting materials for tree crops, and land development. Some refinancing loans are also made (less than 12 in 1987-88) to enable farmers to repurchase or redeem previously owned land mortgaged to (or forfeited by) private money lenders. Term loans represent - 134 - about 302 of BMAC's portfolio. The share of long-term credit (over five years) has been growing over the years with disbursements growing Lt 252 p.a. in the 1980s (Table 8.6). Medium-term lending has declined sharply and now accounts for less than 52 of BAAC total lending, primarily as a result of the availability of BAAC's credit line. 8.26 "Credit-in-Kind" Program. Essentially an input delivery credit scheme. this program was initiated in 1980 with the objective of providing quality inputs directly from wholesalers to farmers at reasonable prices and, in the process, provide competition to the local private sector suppliers. The principal input financed is fertilizer. Started to counter adulteration by certain private dealers, lending for fertilizer increased at 442 annually, from 33,000 tons in 1980-81 to 290,000 tons in 1986-88. The program now also covers other inputs, machinery and equipment, breeding stock, and agricultural services. Fertilizer and agro-chemicals now account for about 502 of a total lending value of nearly B 2.9 billion in 1987-88 with equipment machinery and tools accounting for another 402. 8.27 A recent independent evaluation of the credit-in-kind program 601 found that 932 of the farmer beneficiaries wanted the program to continue with the reasons given validating the original objectives of fair price, good quality, timeliness/convenience and keeping private trader explcitation under control. Recipient farmers in less accessible areas have benefitted most. From a banking point of view, the program ensures that loans are used for the purposes for which they were granted and thus improve loan repayment perfor- mance. However, care has to be taken to ensure that the program does not stifle private sector competition. The BAMC should consider phasing out the credit in kind program from areas where competition for certain inputs is already at a high level. There is also scope for reducing the administrative costs by transferring some of the logistical burdens to the clients after giving them proper training. 8.28 Special Development Proiects. The special projects activity was started in 1977 in response to government's requests to provide credit on concessional terms to farmers participating in various government sponsored projects. With total loan outstanding at the end of FY88, amounting to B 2.4 billion for about 148 projects, they may be divided into three broad catego- ries: (a) projects in response to natural calamities (e.g. flood disaster in the South, Paddy Pledging) or widespread distress (e.g. depressed prices) among farm producers--54Z of special projects outstanding loans; (b) agricultural development projects primarily involving technology transfer to diversify the farm enterprise and/or improve yields (382); and 601 See Khamphon Puapanich, "Evaluation of the Need for the Credit-In-Kind provided by BAAC (in Thai), September 1989, Kasetsart University. - 135 - (c) integrated projects formulated and implemented jointly with relevant goverment agencies and private companies (82). Except possibly for the integrated projects (at regular BAAC terms) and the Paddy Pledging Scheme (subsidized), both of which have loan repayment rates in excess of 901, repayment for the rest were rather poor (averaging about 402). Many of these initiatives, which are primarily medium-to-long-term investment loans, were poorly planned and received inadequate technical and marketing support. Even for the high repayment Paddy Pledaina Scheme (begun in PY86), Government's objective of raising prices early in the harvesting season does not appear to have been fulfilled. Greater storage capacity on the farm was not encouraged and whatever price increases which resulted later in the harvesting season were essentially due to world market trends. Those able to avail themselves of the 3Z p.s. Paddy Pledging loans normally do so to benefit from the subsidy rather than to explicitly take advantage of higher intra- cropping year prices. In any case, the favored farmers stand to gain by using Paddy Pledging loan proceeds to prepay their short-term production loans (borrowed at 12.51 p.a.). The integrated projects have been the most success- ful and should be the primary focus of future special project activities. 8.29 Other Activities. BAAC, in trying to meet what it perceives to be major gaps in the agricultural sector's needs, has developed new programs which may be considered to be departures from traditional banking activities and possibly marginal to 8MC's omn mandate. The New Generation Farmers' Settlement Proiect is an example. This project was established as a pilot (benefiting 80 farm families in FY87) to demonstrate that, with proper planning, effective commercial cooperatives can be established using "new generation* farmers (former graduates) to produce a range of export-oriented fruit and vegetable products. Because of its newness, these projects cannot be evaluated. However, even if the projects were viable and worthwhile, the involvement of BMAC in a primarilv non-banking activity like land settlement is questionable. The task is particularly staff and skill intensive and heavy involvement of BMAC would be difficult to justify, especially since it does not have the required knowledge and staff to undertake more than a few pilot projects. Another new program is the Central Paddy/Alricultural Produce Marketing Scheme. Based on the apparently successful Paddy Warehouse Scheme (implemented in FY85 and FY86), BMC initiated this marketing support scheme in FY87 to complement its Government-sponsored Paddy Pledging Scheme by providing farmers with the opportunity to sell their paddy directly to buyers in an open and more competitive trading environment and to .-educe the control of local traders. The BAAC has opened two trading centers and plans to construct more. The arguments made by BMC for supporting this non-banking activity are the fairer prices accorded to farmers and the improvement of its loan collections. However, these types of activities, like the settlements, require knowledgeable and experienced staff which BMC might have only for a few pilot projects. Financial Performance 8.30 Key financial indicators are given in Table 8.6 and more details are provided in Annexes 55 and 56. In the last ten years, BAAC's gross income increased at an average annual rate of 11.51, from about B 1.5 billion in FY79 - 136 - to about B 4.9 billion in FY88. The cost of funds during the same period increased at about the same rate from B 865 million (582 of gross income) to about B 2,345 million (592 of gross income). Operating costs as a percentage of total lons outstanding appears to be relatively stable (4.52 to 5.02). Interest margins, about 52 on average, appear adequate. Net income has grown at an average rate of 5.72 p.a. during the last six years, from B 167.5 mil- lion in 1983 to B 221 million in 1988. It has never incurred a loss, and it does not pay a dividend. In addition, BAAC is not subject to institutional taxes and reserve requirements (imposed on commercial banks). Table 8.7: BAAC'S INTEREST, COST AND PROFIT MARGINS, 1983-88 (Percentages to average total assets) 1983 1984 1985 1986 1987 1988 Interest income 13.8 13.8 14.3 13.8 11.9 10.0 other income 1.0 1.0 0.9 0.5 0.6 1.7 Total Income 14.8 14.8 15.2 14.3 12.5 11.7 Interest expenses 8.5 8.3 9.2 8.6 6.8 6.5 Interest margin 5.3 5.5 5.1 5.2 5.1 3.5 Operating cost 3.9 3.4 3.9 3.6 3.4 3.4 Bad debt allowances 1.4 2.0 1.1 1.4 1.2 0.8 Exchange losses 0.1 0.1 0.4 0.2 0.5 0.4 Total ExPenses 13.9 13.8 14.6 13.8 11.9 11.1 Net profit before taxes 0.9 1.0 0.6 0.5 0.6 0.6 Source: BAAC. 8.31 Compared with commercial banks in Thailand, BAAC's operating costs (at 3.8? of total assets in 1988, including provisions of 0.72 and 0.42 for doubtful debts and foreign exchange loss respectively) are undoubtedly high. However, this comparison should not be interpreted to mean that BMC is an inefficient organization. By and large the high operating cost of BAAC is the result of the nature of its operation. BAAC essentially finances farming activities of small and medium size farmers all over the country. It has to maintain a large number of branches and field offices (particularly in remote areas), its average loan size is quite small (only B 14,000/client), and it is involved in relatively high credit risk activities innate to agriculture. In fact, compared to similar institutions in other countries, BAAC's operating costs, adjusted for recovery performance (822 overall in 1987-88), are on the low side. 8.32 A feature of BAAC's financial performance is that most of its special projects are handled on a cost plus basis, and includes margin to cover poor - 137 - repayment. This arrangement partly explains why even with the low repayment rates of special projects (averaging 47-502), BAAC has been, albeit marginal- ly, profitable. Relatively, the collection rate for short-term credit (85- 882) is better than that for term loans (69Z). The repayment performance of cooperatives and farmers' association (taken together) is among the poorest (about 402), but such a poor performance has also not been reflected in adverse profit rates for BAAC because deferred payments are rolled-over by reducing future loans to individual borrowers of the cooperatives (i.e., the repayment losses are not yet tin the books"). Thus, a major crisis could result when members' individual loan limits are exhausted and repayments cease. While BAAC has prudently reduced the share of farm institutions' loans in its total lending portfolio from 402 in FY79 to 132 in FY88, increased government pressure to lend to small farmers via these institutions require a thorough reassessment of the problem. 8.33 The profitability of BMC is, not surprisingly, quite low. Its return on equity (before tax because it does not pay any taxes) has ranged between 72-92 since 1983, compared with commercial banks' (after tax) return of about 102. Its total debt to equity-ratio reached its highest level in 1986 (17.4 times) but has since declined to 11.5 times (1988). Its loan to deposit ratio was a low 712 (1988) indicating excess liquidity. Resources 8.34 The sources of funds for BAAC's operations are summarized in Table 8.8 below; Table 8.8: 8AAC'S SOURCES OF FUNDS, 1983-88 Am-ounts in mil ion baht) 1983 1984 1985 1988 1987 1988 Average Source Amount X K*ount lrun- Amount 7 Eunt -Iount X Growth Rate (X) 1983-88 Shareholders' Equity 2,428 10.58 2,381 2,665 1,849 1,978 3,218 0.04 5.80 Deposits from the genoral public 4,785 20.75 4,498 5,548 7,118 8,812 11,459 20.62 19.08 Deposito from Comercial banks 9,018 89.11 9,847 10,685 11,113 18,640 14,895 35.95 9.80 Borrowings 8,832 14.45 5,382 6,869 7,808 7,183 7,048 17.69 16.15 Bank of Thailand 8,000 13.01 8,000 8,500 6,848 8,000 8,000 7.49 0.00 Others 497 2.16 601 600 687 826 927 2.82 18.28 Total 28.060 100.03 26,429 29.35t 84.923 36.089 40.042 100.00 11.87 - 138 - 8.35 BAAC's funding comes from four main sources: (a) Denosits from Commercial Banks. A benefit (for BAMC) of mandated lending (see para. 11), such deposits provided 36? of BAAC's source of funds. Although these deposits with BAAC pays the prevailing market interest rate for these deposits, there is an element of subsidy in that BMAC does not have to incur any costs in raising these deposits. (b) Deposits from the General Public. One of the highlights of BA_^' development is that public deposits have grown at 192 p.a. since PY83, faster than any other source of funds. From 212 of BMAC total resources in FY83, to about 292 in FY88, this impressive development is the outcome of a rural savings mobilization campaign launched by BMAC back in 1981 complemented later by a salary incentive system which assigned it the highest points (weight) to branch merit increases. (c) Borrowings (Mainly from Foreisn Sources). This is another source of fund which has grown quite rapidly, at an annual average rate of about 16?. At the end of FY88, borrowing stood at about B 7.0 billion, of which OECF provided 45?; IBRD, 42?; IFAD, 6?; ADB, USAID and KfW together, 32; and domes- tic sources (excluding commercial banks, SOT and public deposits), 4?. (d) BAMC equity and BOT soft loans. These represent subsidized sources of funding for BMAC. Its equity is a subsidized source because it is not required to pay any dividends and includes grants of B 957 million and B 34 million made by EEC and the government respectively. The BOT preferential interest rate loans to BMC, made to finance its special programs, have declined from about B 4.5 billion to about B 3.0 billion or 24.52 of BAAC's total resources in FY80 to about 7.5S in PY88. Kev Issues and Outlook 8.36 To date, BAAC has tried to undertake the very difficult task of being both a government development agency and a commercially viable specialized institution for medium to small scale farmers. Overall it has handled this delicately balanced act quite well. The overriding issue for BAAC is how it should face the trade-offs between expanding its business to still smaller farmers (the priority target group of government) and maintaining its integrity and effectiveness as a commercial lending institution. It is recom- mended that BAAC should not only be permitted but encouraged to increase its effectiveness as a specialized commercial lending institution in a broader capacity than at present. Restrictions on its operations should be reduced if not eliminated. The long term goal should, therefore, be to afford it the opportunity to expand its comparative advantage to its fullest as an effective rural banking institution; at the same time BAAC should be expected to compete on a more equal footing with commercial banks. Such a correction of course in BAAC's future, requires a reevaluation of the following issues: (a) expansion of BAAC's scope of activities; (b) interest rate and financial policies; - 139 - (c) the framework and ground rules for BMAC's special project operations; (d) options for reaching the poor farmers; and (e) implications for BAAC's corporate plan. 8.37 Exiransion of BAAC's Scove of Activities. With the decline in relative importance of agricultural production activities (associated with the gradual maturing of the Thai economy) and the comparative advantage of BAAC in the rural sector, the case for allowing BMAC to lend for all rural development activities is compelling. First, since total farm family income should be the main concern, supporting feasible options for improving their non-farm income (now accounting for 422 of their total family income) would make eminent sense. Second, because the clientele of BAAC and commercial banks are essen- tially different, the non-production credit needs of BAAC's clients are not being met by commercial banks--a large opportunity unnecessarily foregone for BMAC. Third, because its creditworthiness assessments already take into account the overall family status of its clients, permitting BAAC to lend for the farmers' non production needs would be the most effective means to reduce its unit operating costs and, concurrent with the upgrading of its field offices, also permit more efficient financial resource mobilization. Fourth, credit risk would be reduced for BMAC because farmers would be able to pledge all their available collateral to one financial institution; and, for the farmer, his overall source of income would be less dependent on natural condi- tions. Fifth, by permitting BAAC to extend its rural mandate, it can more successfully cater to the two extremes of its spectrum of clients. At the high-income farmer end, for example, it will prevent BAAC from losing such clients to commercial banks and provide the needed competition with them. At the small farmer end, it will permit BAAC to better reach the near-landless farmer whose income generating options in agricultural production are likely to be limited, but whose earning potential would be substantially brighter in the service and manufacturing sectors at the provincial level. Thus, expan- sion of BMAC's scope of activities would result int (i) greater overlap of market segments with commercial banks, leading to more competition and efficiencies; azLd, more importantly, (ii) the small and medium size farmers' total credit needs would be met more comprehensively, more convenientely and at a lower cost. 8.38 Interest Rate and Financial Policies. BAAC's commercial viability is constrained by limitations imposed by the Government on its lending rates, although these rates are officially set by its own Board. The BMAC would not have operated profitably under its prevailing interest rates (12.5? on short and medium term loans, 11.5? on long term loans and between 3? to 11.5? on special projects) without the following indirect subsidies: (a) soft loans from BOT and specially constituted development lending funds provided at sub- sidized interest rates; (b) exemption from corporate taxes and other imposi- tions on commercial banks; (c) no mobilization cost on the substantial deposits that it receives from commercial banks; and (d) the non-payment of dividends from its profits. Interest rate ceilings for BAAC have not only restricted the coverage of its clients but, because they are also below com- mercial bank rates, have created distortions detrimental to the commercial credit sector. While not large, they are not needed. If BAAC were allowed to _ 140 - charge the rates commercial banks do, it would not need the tax exemption and most of the other subsidies that are currently provided to it (see Annex 57 for details). At the current, below commercial bank lending rates, net income would have been negative had commercial banking taxes and charges been applied to BAAC's gross income. If BMC had been permitted to charge 14.52 on its direct lending to farmers in FY86-88, its net income would have been positive (ranging from B 203 million to B 470 million) and its return on equity would have increased (ranging from 10.3Z to 15.62). These results also illustrate the leeway available to incur the necessary costs for self-mobilization of domestic savings and to substantially reduce the subsidies on its sources of funds (e.g. BOT soft loans). 8.39 A Framework for BAAC's Special Proiects. Given the fact that there will inevitably be some trade-offs in the use of BMC's scarce staff (and financial, resources, certain conditions should be set to permit the screening out of non-viable socio-politically imposed programs or projects. First, it should clearly be understood that BMAC would act strictly as an agent without assuming the credit risk and procedures should be established accordingly for compensating/reimbursing BAC. Second, account should be taken of the adverse impact ons (a) BAAC's farmer client income and the undermining of financial discipline through their poor repayment habit; and (b) BAAC staff morale and attitude, particularly those in the branches and field units (as the experience with the Paddy Pledging Scheme has shown). Conditions should, therefore, be set which stipulate the expected minimum viability of the special projects to be undertaken. The criteria to be used, which could include internal rates of return, expected repayment performance, employment and income effects relative to cost, should be tailored to whether the project would be of a pilot nature or would be undertaken on a large scale. BAAC's progressiveness in this area should not be stifled by over-stringent condi- tions; but, at the same time, it should not be encouraged to start pilot initiatives which cannot be replicated. Or even if it can be replicable, explicit account should be taken to phase out BMC involvement to other, more appropriate, developmental agencies or the private sector (e.g. lending-in- kind program). 8.40 Reaching the Small/Poorer Farmers. In 1989, BMAC initiated a five- year Small Farmer Lending Program (FY89-FY94) with financial support from a special Small Farmer Fund (contributed by BOT, Government budget, Government Savings Bank and commercial banks at a blended cost of 6.1Z). About one bil- lion baht will be available annually from this fund for lending to new small farmer clients (defined as those with a net farm income of less than B 10,000/ year) at an interest rate of 12.52. This implies that (assuming an average loan size of B 7,000) nearly 145,000 new clients annually, or a total of 3.3 million farm families by 1994, would have to be reached by BMC. Given a total client expansion rate of 96,000/year between 1983 to 1987, such a target seems unrealistic unless client creditworthiness and project viability crite- ria are to be seriously compromised. Instead of being target oriented, BAAC should concentrate on improving the quality of its financial support services to all its clients and simultaneously expand its outreach. Three promising and connected avenues should be exploited with the requisite staff support: (a) expansion of new as well as existing joint liability groups with emphasis on: (i) identifying new income generating opportunities, (ii) complementary - 141 - line agency support, and (iii) maintaining good repayment incentives; (b) branch field office expansion, especially in small-farmer concentrated areast and (c) expansion of the scope of activities to cover non-agricultural investment or working capital needs. 8.41 Iamlications for BAAC. The push for BAAC to expand the extent of its special projects and to reach the smaller farmers (partly through coopera- tives) is reflected in its corporate plan for 1989-94 (October, 1989). For example, the share of special projects in total loan outstanding (year-end) would increase from 4.3? in 1988 to 12.22 in 1994 and cooperatives share would increase slightly (from 12.22 to 13.6Z). Such a trend which would be counter to the orientation suggested above, also reflects an overambitious increase in lending volume (16.7? p.a. versus 10.7Z historically). From the financing side, BAAC would be relying more on direct government funding (2.3? p.a. in the past six years to 10.4Z p.a. by 1994) rather than comercial banks' depo- sits (9.9? to 2.22) and embark on no new external borrowing. Despite increased effort to mobilize public deposits, BAAC expects a slower rate of growth in this activity than in the past, which means that it would be faced with a resource gap of B 22.3 billion or US$871 million from 1989 to 1994. It is unrealistic to expect that BAAC could borrow this amount. It needs to reexamine its corporate plan after taking into account the suggestions made in the foregoing. - 142 - IX. THE CAPITAL AND MONEY AURT5 9.1 The cap.tri, and money markets are both networks of providers and users of fundes tlhdy are distinguished in temss of maturities and therefore the underlying lnstruments of debt. The money market provides funds for up to one year; the capital market for maturities of more than one year. The capital market includes the market in equity securLties (i.e. common and other kinds of stocks); the money market comprises debt Instruments only. both markets deal in public sector (i.e. government and quasi government) as well as private sector securities. The instruments traded In the Thai money market include government treasury bills, notes issued by commercial banks and IFCT, transferable certificate of deposits (CDs), and comercial paper or billst the capital market instruments currently traded are government, state enterprise and BOT bondst debentures issued by private sector firms floating rate notess common and preferred shares, and convertible debentures. 9.2 This Chapter is divided into five parts. The first three provide details about development, size and characteristics of the money market, the market in equities and the market in long term debt securities. The fourth part discusses the problems and constraints to development of the money and capital markets. The final section discusses how threo now methods of finan- cing may be introduced in Thailand; these are financial leasing, factoring and trust receipt financing. A. The Thai :Monev Market Primary Market Size 9.3 The size of the primary money market in terms of Instruments, and its development since 1980 is shown in Table 1 belows Table 9.1: PRIMARY MONEY MARKET INSTRUMSNTS, 1980-88 (million baht) 1980 1981 1982 1983 1984 1985 1986 1987 1988 Treasury bills 8,400 8,400 9,900 11,000 12,000 12,000 12,000 11,500 2,000 Commercial paper - - - 1,284 - 2,382 8,423 11,196 16,685 Transferable CDs - - - - 19,484 18,858 1,917 950 301 Notess Citi Notes - - - . . - 325 650 650 Chase Notes - - - - - . - 200 200 IPCT Notes - - - - - - - - 400 Sources BOT. - 143 - The relative size of the Thai money market in the financial system as a whole is quite small. Compared with the amounts shown above for money market instruments, total credit extended by financial institutions and outstanding as of the end of 1988, for example, amounted to B 1,116 billion. 9.4 Treasury Bills. As Table 9.1 indicates, treasury bills are no longer an important trading instrument. Not only have the amounts issued declined, but all the amounts outstanding are held by BOT and Exchange Equalization Fund. In the past, treasury bill auctions would be held on a weekly basis and commercial banks were also active players. But since 1987, commercial banks do not hold any treasury bills, as the following table showss Table 9.2: HOLDERS OF TREASURY BILLS, 1980-88 (million baht) 1980 1981 1982 1983 1984 1985 1986 1987 1988 Bank of Thailand 6,258 3.385 5,756 8,154 4,850 9,709 4,486 9,460 - Exchange Equalization Fund La 1,409 1,240 1,568 1,570 1,000 1,800 2,067 2,000 1,960 Commercial Banks 330 3,335 2,197 935 5,260 35 1,850 - - Others 403 440 379 341 890 446 3,597 40 40 /a This is a government-owned account which holds a portion of the country's foreign exchange reserves. Source: BOT. The last treasury bill auction was held in February 1989; there has been no need to issue any bills since then. 9.5 Repurchase Agreements. The government bond repurchase market was established in 1979 as part of the Money Market Development Project initiated by BOT. Although the primary market and the secondary market in government bonds are both capital market activities, the bond repurchase market is a money market activity. Following its introduction, the repurchase market rapidly became the principal instrument by which BOT intervened in the money market, to provide or to mop up liquidity. One to three day transactions are the most usual, but up to 180 days is possible. 9.6 Following the relaxation in August 1989 in commercial banks' reserve requirement for branch openings from 162 to 122, the difference of 4Z has been used in the repurchase market, and has boosted the market's liquidity by B 8 billion to B 18.5 billion. The relaxation itself reflects the inadequate supply of bonds for banks to hold to satisfy the primary and secondary reserve requirements, which in absolute amounts have grown rapidly in recent years in - 144 - line with the growth in their deposits. The new issues of bonds have also been declining because of the government's budget surplus and cash rese.ves. In 1988 the goverAment issued bonds equivalent to 14.92 of banks' total depo- sits, 1.12 lower than the regulated level at the time. In the first half of 1989, the proportion fell to 132 as total outstanding deposits of banks rose 11.42 to B 950.3 billion. It has been recommended earlier (in part B of Chap- ter II) to eliminate completely the secondary reserve requirements for branching and also to drop the bond portion of the primary reserve; this would free the supply of bonds and help develop the secondary market. 9.7 Interbank Loans. The interbank loan market involves lending between banks at call and for fixed periods ranging from overnight to six months or longer. The market has tended to be one way, with the largest Thai banks as the lenders and the smaller Thai banks, foreign bank branches, and finance companies as the borrowers. In 1985 commercial banks introduced BIBOR, the Bangkok Interbank Offered Rate, which is the average of the rate at which several large banks offer deposits to each other. Interbank rates tend to be more responsive to money market conditions than other market rates. At present, the seven day, one month, and three month BIBOR are quoted daily, as are the bid rates at which the banks bid for deposits from each other. Besides direct lending. interbank deposits are sometimes used as a means to adjust liquidity positions between banks. 9.8 Transferable Certificates of Deposit. Transferable certificates of deposit were first introduced in Thailand in 1984, but only three banks have so far issued them, with maturities ranging from three months to a year. Their attractiveness to investors has declined with the disappearance of the interest premium they originally offered to large depositors over other types of fixed deposit. With the recent lifting of ceilings on long term deposits, the opportunity may return for their pricing to become more attractive. 9.9 Commercial Bills. Commercial bills are the category of short term financial instruments which includes promissory notes, bills of exchange, and bankers' acceptances. They have been issued, either directly or through underwriters, by business and industrial companies of high creditworthiness; by certain foreign banks' branches e.g. the Citinotes and Chase notes which are transferable bills of exchange issued by Citibank and the Chase Manhattan Bank; and by certain other financial institutions e.g. the IFCT notes which are transferable bills of exchange issued by IFCT. Maturities of commercial bills are mostly between three months and six months. The market for com- mercial bills, though still small, has been growing rapidly. 9.10 The Intercompany Market. There is also an intercompany market for liquid funds, but this is mainly between affiliated companies. Intercompany trading in short term promissory notes is not goverred by any regulations. There is an informal commercial paper market for issuers with first class names. _ 145 - The Secondarv Market 9.11 Treasury Bills. In addition to its role as the sole underwriter in the issuing market for treasury bills, BOT also makes a secondary market by selling treasury bills from its own portfolio and purchasing treasury bills before maturity. This secondary market has been limited because returns on treasury bills are low in comparison to the returns on other securities, and because the Initial buyers of treasury bills tend to hold them to maturity. In recent years it has also declined sharply in line vith the decline in total primary issues, and as stated in para. 9.4, all outstau.ding amounts of treasury bills are held by BOT and the Exchange Equalization Fund. The amounts traded in the secondary market are shown in the following table. Table 9.3. SECONDARY MARRET IN TREASURY BILLS, 1980-88 (Million baht) 1980 1981 1982 1983 1984 1985 1986 1987 1988 Repurchases 22,408 25,759 52,967 62,956 58,403 5,998 9,348 5 1,004 Sales 39,327 45,257 84,262 91,875 89,469 38,147 42,271 1,998 3,673 Source: BOT. 9.12 Transferable Certificates of Degosit. The decline in the outstanding amount of CDs (from B 19.5 billion at end 1984 to B 950 million at and 1987) has caused a similar decline In secondary market activity. At present, only three finance and securities companies deal in the secondary market for trans- ferable CDs. 9.13 Renurchase Anreements. The activity in the secondary market, assumed to be mainly in repurchase agreements, in government bonds and the trends since 1980 have been as shown in Table 9.4. B. The Equities Market The Securities Exchanae of Thailand 9.14 The Securities Exchange of Thailand (SET) commenced operations in April 1975, following the enactment in May 1974 of enabling legislation (the SET Act BE 2517) which established the exchange and provided a regulatory framework for the securities market. An earlier stock exchange, the Bangkok Stock Exchange, had been formed in 1962 but becam moribund after a troubled early life and as a result of a lack of both supply and demand for shares. _ 146 - Table 9 4; SECONDARY MARKET IN BOND REPURCRASE AKERENTS, 1980-88 (Million bait) 1980 1981 1982 1983 1984 1985 1986 1987 1988 Government bond 77.2 88.8 109.5 134.5 151.4 179.0 200.9 204.6 184.6 State Enter- prises bonds 0.9 0.9 0.9 1.4 1.4 1.4 1.4 2.3 2.4/a /a Up to September 1988. Source: BOT. 9.15 Membership of SET is limited to socurities companLes licensed by the Ministry of Finance to engage in the securities business as stock brokers. Towards the end of 1989, SET had 35 members; all members are Thai nationals as foreign brokers are not yet allowed to be members. 9.14 Listings of companies on SEST are either on the Main Board (listad companies) or the Second Board (authorized companies). Securities registered in the name of, and eligible to be traded by, foreign investors can be traded only on the Special Section (Alien) Board when foreign holdings in such securities reach the maximum permitted by law or by the companies' Articles of Association. The same rule applies to the offshore funds. The scale of demand and limLted supply of shares have pushed prices on the Special Section way above those on the Main Board. This is reflected In the high premiums for offshore funds above net asset values. 9.17 Overall control of SET and the securities industry rests with the Ministry of Finance, which has delegated certain powers and duties to BOT. The BOT is, for example, responsible for the supervision of the securities companies. In 1989, the Ministry of Finance established a new Division in the Fiscal Policy Office, to exercise the functions of a securities exchange com- mission. The Division is still in its early developmental stage, and had a five member staff only at the end of September 1989. Since the supervision and regulation of securities industry is a highly technical job, the Ministry of Finance would be well advised to staff the new Division with skilled tech- nical persons. If the new Division is staffed principally with civil servants transferred from other parts of the government, it may simply become a bureau- cratic bottleneck and not be helpful to the healthy development of the capital market. In the long term, the government should consider establishing an autonomous Securities and Exchange Commission, and move the securities industry towards self regulation. - 147 - Market Size and Characteristics 9.18 The market capitalization of SET listed equities has increased steadily since 19811 the rate.of increase jumped after 1986, as the following table showst Table 9.S MARRET CAPITALIZATION OF SET LISTED EQJITIES, 1980-89 (Billion baht) 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Covoon shares 28.8 23.0 28.9 34.2 46.7 48.8 74.2 135.0 218.1 659.50 Preferred shares 0.06 0.05 0.05 0.16 0.18 0.17 0.21 0.37 0.55 NI/A Unit Trusts 0.43 0.33 0.44 0.40 0.54 0.51 0.70 2.04 3.34 NIA Source: BOT. In 1989, the values of the listed equities further increased substantially. As of mid-1989, the market capitalization had increased to B 367.3 billion. 9.19 As of June 30, 1989, the sectoral breakdown of the total market capitalization was as shown in Table 9.6. - 148 - Table 9.6: MARKET CAPITALIZATION BY SECTORS, DECEMBER 31, 1989 Sector Companies Market Capitalization I (billion baht) Banking 15 88.32 13.4 Finance & Securities 23 56.65 8.6 Insurance 10 28.94 4.4 Commerce 16 37.78 5.7 Services 7 9.90 1.5 Warehouse & Silo 3 2.36 0.4 Hotel 6 30.05 4.6 Packaging 9 15.15 2.3 Construction 12 210.05 31.8 Automotive 3 8.04 1.2 Textiles 21 30.55 4.6 mining 4 17.62 2.7 Food & Beverage 14 21.54 3.7 Electrical 5 17.92 2.7 Others 70 84.63 12.8 218 659.50 100.0 Source: Phatra Research Institute, Stock Market Mid Year Review, 1989. 9.20 Foreign Participation. The SET has been open to foreign investment since its inception but the Thai market has become popular with foreign inves- tors only since 1985. Since then, however, foreign investment in Thai secur- ities has been increasing rapidly. Details of foreign capital inflow and outflow are given in Annex 58; portfolio investments are summarized in the following table. Table 9.7: FOREIGN PORTFOLIO INVESTMENT, 1982-89 (Billion baht) 1980 1981 1982 1983 1984 1985 1986 1987 1988 Inflow 1.1 0.3 0.6 0.4 2.4 4.1 3.0 17.1 27.8 Outflow 0.1 0.2 - 0.1 2.5 0.2 0.5 4.3 16.6 Source: BOT. - 149 - Of the total capitalization of the Thai stock market, it is estimated that roughly 601 is closely held by the founding families or groups. Foreign par- ticipation is in general legally limited to 492; in fact direct foreign owner- ship is virtually at the standard limit established in a company's articles of incorporation of 25Z for almost every issue. Such direct ownership accounts for roughly 602 of the estimated float. In addition, foreigners also own Thai shares indirectly through funds (almost all of which are closed end in charac- ter). Inclusive of such indirect holdings, overall foreign ownership is estimated to amount to roughly three-quarters of the floating supply. Domes- tic households are believed to be holding only 102 of the market's total capi- talization. The ratio of the market value of shares held by Thai households (excluding controlling groups) and institutions to GDP is estimated to be very low, around 2.0Z, and it would be even lower if foreign demand had not raised valuations in recent years. It has been further estimated that less than 12 of all Thai households directly own any listet Thai shares. 9.21 Much of the foreign portfolio investment is institutional money looking to diversify globally; it is probably concentrated in a fairly limited number of institutions. Foreign institutional portfolio funds are quite mobile, and there tends to be strong consensus behavior in the international institutional commmuity. Therefore, if for whatever reason, foreign portfolio investors were to decide to reduce their exposure at the same time, wholesale liquidations by foreigners (closed end fund managers could in principle switch to monev =r;,ret and debt instruments) who dominate the float could have an extremely destabilizing effect on market valuations. 9.22 Comparison with Other Asian Markets. In a seven-country comparison (excluding Japan and Singapore) the Thai market as of the end of September 1989 had the third lowest market capitalization. The relative size of the Thai market and its valuation is shown in Table 9.8. - 150 - Table 9.8s RELATIVE SIZE AND VALUATION OF THAI STOCK MARIET, September 30, 1989 _Valuation Market Capitalization Price Price/ Dividendl Amount 2 of Earnings Book Yield US$ million GDP la (Times) (Times) (2) Taiwan, China 243,768 118 49.6 7.4 0.6 Korea 139.676 71 39.7 2.7 1.2 Malaysia 33,851 69 31.6 3.3 2.2 India 24,908 9 16.8 3.1 1.9 Thailand 17,787 17 18.2 5.7 2.5 Phillippines 11,960 11 19.6 4.7 1.1 Pakistan 2,425 7 8.9 1.8 7.0 la Based on data as of December 31, 1988. Sources IFC: Zmerging Stock Markets Factbook. 9.23 Preferred Stock and Warrants. As is clear from Table 9.5, the great majority of stocks listed and traded on SET is common stock, but there are also four issues of preferred stock and five of unit trusts. Warrants have been issued, attached to common shares and/or to debentures (warrants cannot be issued on their own). New rules apply from September 1989s a listing of not less than three years, not over 302 of the outstanding, and warrants not exceeding five years. Warrants are interestingly seen as delayed subscrip- tions instead of as options to buy shares. Local practice is unfamiliar with covered warrants, and SET has not made its position clear on them. C. Lons Term Debt Securities Market Primary Market Size 9.24 The size of the primary market in long term debt securities, and its develoument since 1980 is shown in Table 9.9. - 151 - Table 9.9: PRIMARY DEBT SECURITIES MARKET, 1980-88 (Billion baht) 1980 1981 1982 1983 1984 1985 1986 1987 1988 Government bonds 14.0 23.2 24.5 18.7 35.0 26.3 40.8 29.2 20.5 State enterprise bonds 1.5 2.6 0.5 - 0.1 3.4 3.9 1.7 2.2 BOT bonds - - - - - - - - 0.1 Debentures (by private firms) - - 0.05 - - - 0.8 2.5 0.5 Floating rate notes - - - - - - - - 1.2 Source: BOT. 9.25 Bonds ("Pantabat'). Long term bonds with maturities of over one year have been issued only by the Thai Government and by certain government enter- prises e.g. the Electricity Generating Authority of Thailand, the National Housing Authority, and the Telephone Organization of Thailasd. Their effective interest rate is fixed slightly higher than commercial banks' one year fixed time deposit rate. Tax on interest on government bonds is levied at the rate of 15? on the amount of interest that is in excess of 7.252 p.a.; the full amount of interest on state enterprise bonds is taxed at 15Z. 9.26 The Thai Government has issued government bonds since 1933 as a means of financing budget deficits. The progressive reduction in fiscal deficits, culminating in the surplus recorded in 1988 for the first time since 1974, had reduced the need for bond issues and has led to a shortage in the supply of bonds for reserve requirements. Because of this shortage, banks typically hold bonds to maturity. Earlier (Chapter II) in discussing the monetary policy control instruments, it was recommended to eliminate secondary reserve requirements and the bond portion of primary reserve requirement as well. This should improve the availability of bonds for trading purposes. Otherwise tta present shortage would continue, because of the expectation that the budget surplus in 1989 will be larger than in 1988, and that budget surpluses may continue for several years. 9.27 Government enterprise bonds have been issued since 1972, with maturities ranging from three to ten years. They have been underwritten by certain finance companies. Most but not all government enterprise bonds are registered at SET. 9.28 Debentures (AHoonkoo"). Debentures have most of the features of long term bonds, but are not guaranteed by the government, and are not backed by any real assets or security. They may be straight debentures, guaranteed debentures, convertible debentures, or guaranteed convertible debentures. They are issued by companies of high reputation and high creditworthiness, which must either be public companies or companies that are registered with SET. Maturities range from five to eleven years. Debentures placed with - 152 - insurance companies must be guaranteed by commercial banks or secured by land collateral. 9.29 Issuance of debentures by private firms is the weakest feature of the Thai capital market. Table 9.9 shows that debentures have been issued only in four of the nine year 1980-88 period and then, too, in very small amounts. In fact there are only eight issues of debentures registered at SET. Why the market in private long term debt securities is so limited is examined in detail in Section D below. D. Problems and Constraints 9.30 The regulatory framework contains the legal basis under which con- straints are imposed on financial intermediaries and market participants in order that prudential behavior is ensured, that monetary policy may be exer- cised, that the investors are assured of due protection, and that an optimum level of liquidity is maintained in individual firms and in the system as a whole. Ideally, it should do all these without impeding the evolution of the financial intermediaries' role of fulfilling the needs of society and the economy, by developing the necessary instruments and performing the necessary functions of an efficient market. In pr A:ice some seemingly harsh constraint may be needed at times in the interests of maintaining stability in the system as a whole. However, apart from such exceptions, distortions such as those caused by detailed regulations that have become out of date, and constraints imposed by legal restrictions and too-stringent requirements which have side- effects that may not have been intended, cause uncertainty and impede the development of an efficient, thriving and growing market in financial services and instruments. Discussed below are a number of specific constraints that have been identified as impeding the development of the capital market in general and a market in private sector fixed income securities in particular. Developing a Market in Government Bonds 9.31 An active market in government bonds normally precedes the develop- ment of a market in corporate issues. It is because individual corporate issues are inherently too small around which to develop active secondary markets large enough to support the institutional infrastructure needed for active markets. The development of an active secondary market in government bonds has been inhibited principally by two factors: short supply relative to demand arising out of reserve requirements imposed on commercial banks (resulting in most bonds being held by them and to maturity), and lower than market coupon rate. As discussed earlier in the context of interest rate and reserves policies and recommended there, secondary reserve requirements should be eliminated and the bond portion of primary reserve requirement should also be dropped. Also, as already recommended there, the interest rate should be raised to bring about a positive yield curve until an active market develops at which point the market itself can be depended upon to determine the rates. Once these two major recommendations have been implemented, adequate supply of bonds would be freed for trading and they would also become attractive for investors to buy rather than for banks to hold them simply to meet reserve requirements. An actively functioning bond market would also prove very use- ful to the authorities to sterilize excess liquidity, when and if in the - 153 - medium to long term, Thailand achieves balance of payment surplus, thus avoiding the kind of problem faced by Korea. After the market in government bonds has been reasonably well developed, the stage will be set to make a serious attempt to develop a market in private sector fixed Income securities market. That attempt should take into account the constraints discussed in the following paragraphs. Statutory Restrictions on Who May Issue Debt Obli$ations 9.32 Most Thai firms rely for their expansion and even for current opera- tions on (i) funds generated by their business, or (ii) bank loans; more recently new equity issues are also being made. They do not normally issue debt to the public. This is the result of the enactment of a statute that sought to encourage the conversion of private companies Into public companies. Public companies were defined as companies whose ownership was broadly held. The privilege of issuing debentures was limited to such companies so as to create an incentive for the transformation of private companies into public companies. In practice, however, most pr$vate companies have chosen to remain private, finding the burden of going public, dispersing ownership and meeting disclosure and other requirements of the Law on Limited Public Corporations, enacted in 1979,611 onerous. To reinforce the objectives of this law, the Thai Civil and Commercial Code (CCC) was also amended to proviee categorically that private companies cannot issue debentures. In 1984, the amended SET Act permitted the issuance of debentures by corporations registered at or approved by SET or whose registration/approval was being considered. 9.33 Companies which have chosen to remain private and have not otherwise complied with the requirements of SET Act are prohibited from issuing deben- tures, and have considered issuing other kinds of debt $nstruments. They have hesitated to issue to the general public short term commercial DaDer In the form of vromissory notes for fear that DOT would charge them with doing finance company business without a license. Accordingly, while some companies have issued promissory notes, they have avoided issuing them to the general public and have limited their issuance to private placements. 9.34 Another method that a few companies have explored is the issuance of bills of exchange. These are thought to be sufficiently different from deben- tures and promissory notes so as to escape their vices in the eyes of the supervisory authorities. Two rather cumbersome approaches have been tried. In the first mechanism, a company draws a bill of exchange on itself, accepts it and then sells it to the public through a bank. Some investors Insist on a bank guarantee or an aval of the accepted bill before purchasing it. The second mechanism that has been used is for a company to draw a bill of exchange on a bank which then accepts the bill. Thereafter the drawer company sells the bill to the public through the bank. It is clear that these cumber- some mechanisms involving the use of a bill of exchange are resorted to In 61/ It is understood that only about 28 companies went public under this law and of these the majority were breweries or distilleries which complied only in order to obtain a liquor license. - 154 - the absence of more tr!ditional methods for a corporation to borrow from the public. 9.35 What is the explanation for this unusual state of affairs where most firms are in effect prevented from borrowing from the public? It is difficult to provide a definitive answer. It is possible that the framers of the Law on Limited Public Corporations simply misjudged the power of the positive and negative incentive the law was providing for companies to go public or to get listed i.e. that only public companies could issue debentures and non-public companies could not. Experience has proven the incentive to be ineffective. In fact whether a firm is a public or a non-public company should be irrelevant in the consideration of whether it may borrow from the public. Incentives for firms to go public or for them to be listed should take dif- ferent forms from the prohibitions against issuing debentures or commercial paper in the form of promissory notes. The present law also has the unin- tended effect of perpetuating the strong hold of the commercial banks on credit business, and of denying most firms of legitimate and may be less expensive ways of raising capital. It is recommended that the Law on Limited Public Corporations and CCC should be changed so as to allow all corporations (whether public or private) to issue debentures. 9.36 There is nothing intrinsically wrong with a corporation issuing debentures to the public. From the corporation's standpoint, it allows the corporation to supplement its capital funds with borrowed funds. It permits corporations to go to the market if bank funds are unavailable or too expen- sive. From the investing public's standpoint, debentures provide another instrument of investment and one which they may prefer since, unlike the case of dividends on stock, interest payments must be made on debentures whether or not the corporation earns a profit. Opening up another avenue for the users and providers of capital would also result in more competition for the commer- cial banks in the business of credit. 9.37 It would be instructive for Thai authorities to review the Japanese experience in this regard. In Japan, too, corporations were not allowed to issue debentures until 1979. Commercial banks opposed it for fear of losing business. But gradually, responding to international liberalization and integration of financial systems, commercial banks' opposition was overcome and ultimately they joined in; Japanese corporations are now permitted to issue both debentures and promissory notes. Law on Trusts 9.38 Related to issuance of corporate debenture and other debt instruments is the subject of trusts. Although not absolutely clear, it appears that the Thai Commercial Code makes trusts invalid. One interpretation is that this prohibition applies to non-commercial situations only. Nevertheless, the prohibition appears to have been given a broader construction so that it is believed to apply to commercial situations as well. It is in these latter situations that the prohibition against trusts may impinge negatively on cor- porate debt issues because debt instruments are commonly issued under the provisions of a trust agreement between the issuing corporation aud a trustee who represents the holders of the obligations. A trust agreement normally - 155 - sets out a form of the obligations, provides for the authentication and issuance of the obligations, contains various covenants, provides for the redemption of the obligations before maturity pursuant to a sinking fund, and for the acceleration of principal repayment upon the happening of certain events. The agreement may also authorize the trustee to take other actions on behalf of the holders of the obligations in the event of default by the corporation. If the Thai law does make trusts invalid, it would amount to denying debt instrument holders useful protection, thus making debt obliga- tions less attractive. 9.39 The existence of a corporate trustee is not indispensable for the corporate issuance of debt obligations. Thus, various countries have dealt with the matters incident to such issuance through other devices. It is pos- sible, for example, to envisage the holders of the obligations pursuing their remedies individually against the corporate issuer or even together as a class. But such techniques would probably be less practicable in the Thai context. The authorities should, therefore, consider to amend the law to make clear beyond any doubt the validity of trust agreements. Reaistration and Listina Related Constraints 9.40 The process of listing involves not only SET but also the Ministries of Commerce and Finance. It normally takes four months and may take longer. The government contends that delays occur because the applicants take a long time to comply with the requirements and provide the necessary information. But the effect of delays, irrespective of the reasons thereof, is that an underwriter is left totally exposed unless he resorts to grey market opera- tions. Either way the result is undesirable. It would be of great help if all the overseer functions can be centralized at one place. Could it be pos- sible for the Ministries of Finance and Commerce to delegate their roles to SET? Alternatively the new Division recently set up in the Ministry of Finance, with some delegated staff from SET, may be made the focal point for reviewing and approving all listing applications. 9.41 The registration and listing procedures at SET are in themselves slow and cumbersome, and have not kept pace with the development of new financial instruments and with increased market activity. The conversion to equity of a convertible debenture, for example, requires it first to be exchanged for cash which is then used to purchase equity which is then registered. A somewhat related negative is the requirement in underwriting of new issues to provide five year forecasts, which is impossible with any degree of accuracy, and meaningless if done with large assumptions or merely as an extrapolation. Divergence of actuals from such forecasts may bring unwarranted criticism of the underwriter and the issuer. Tax Related Issues 9.42 As an inducement to go public, the tax rate for listed companies is 30? against 35? for unlisted companies. In reality, this concession acts as a disincentive for many owner managed private companies to seek a listing at SET. It is in the nature of many small owner-managed companies that the ef- fective tax :ate at which they actually pay taxes is likely to be lower than - 156 - the concessional rate of 30a on the rather better audited results they would have to show as listed companies. It is not being implied that to make listing more attractive, the tax rate for all listed companies be reduced further; that would have serious fiscal impact. It is however, suggested for authorities' consideration that the tax rate for newly listed companies be reduced to say 25Z for a limited period of time after listing, say for five years. Such an inducement for listing may prove to be an effective attrac- tion, but may not adversely affect total tax receipts. 9.43 The 3Z business tax (increased to 3.3Z because of 102 municipal tax) is a disincentive to active trading. When VAT is introduced, it is proposed to retain the special business tax as a tax on gross interest received. Similarly the 1? stamp duty on bond sale transactions may be inhibiting trading activity. Given the small size of secondary market in bonds, the yield of this tax is probably small; on balance it may be desirable to eliminate this tax. Also tax concessions that require yearly renewal should either be permanent or cancelled altogether, to avoid uncertainty. Finally, the point at which a tax is levied and the basis on which the liability is calculated are also important considerations. It is preferable for tax (if it cannot be eliminated altogether) to be levied at the "bottom line', leaving actual transactions in the secondary markets to develop more freely. Constraints on Insurance Companies and Other Institutional Investors 9.44 The rules governing investments by insurance companies and other similar institutional investors should be reviewed so that undue restrictions on their investments that act as constraints to trading could be removed. For example, there is a limit of 22 maximum holding in each security, or that stock be valued at book value. The restriction on government bodies to invest at par only is another example. While investment rules should always be prudent, they should not be so conservative as to become unduly restrictive. Given the significant developments in the financial market since those invest- ment rules were formulated, it is now desirable to have a second look. Absence of a Rating Agency 9.45 At present there is no rating agency in Thailand. It is not prac- ticable for investors, particularly small investors, to evaluate individual debt securities. A rating agency to provide credit rating on new issues as well as current rating on existing issues would be extremely helpful both to issuers and to investors, and should result in efficient pricing. it is necessary, however, that the rating agency itself be highly credible. To achieve this, it would be highly desirable that, initially, the rating agency is set up as a joint venture with, or at least with technical assistance from, an internationally reputable firm such as Standard and Poor or Moody's. It is recommended that the establishment of an inuependent rating agency be con- sidered seriously and expedited. - 157 - Availability of Financial Services 9.46 The development of financial services sector is still at an early stage in Thailand. A truly discriminating financial market must provide savers with information of quality and quantity necessary to evaluate the prospective risk adjusted returns of many competing investments; it must mobilize funds from a broad array of savers, and it must allocate them within a decentralized market domain of competing investments. This requires an extensive service sector in the economy: one that generates and processes quality information on numerous issues and makes the financial market trans- parent; one that solicits funds from a wide array of prospective savers, util- izing a varied menu of instruments that cover the entire risk/reward spectrum; one that conducts competitive auctions that price the financial issues of prospective users appropriately. Such market transparency calls for vendors of securities with analytical skills and buyers that are able to discriminate on the basis of such information, and an infrastructure of communications and information dissemination that brings them together. The distribution of securities requires well staffed institutions that sell securities to savers and institutions that act as portfolio intermediaries. Additionally, competi- tive auctions need well informed rules of fair play and a respect for them by market participants. It should be noted that this type of elaboration takes time. Unlike the basic industry where development can occur in quantum leaps with the construction oS each plant or facility, financial services develop- ment is incremental, gradual, and evolutionary. Other Suggestions 9.47 The following other suggestions are being offered for consideration of Thai authorities: o Traded rights issues would provide flexibility in raising additional equity. Trading in rights should therefore be allowed. Similarly, trading freely in warrants should also be allowed. O Only ten companies are allowed to manage provident funds. The SET has requested the Ministry of Finance to amend the governing regula- tions. The rules should be reviewed and the necessary changes made expeditiously. O The Mutual Fund Company is currently the only holder of a license to manage mutual funds. The government should consider issuance of additional such licenses thereby promoting competition and efficiency. O The OTC market operates without effective supervision. It would be better to bring it under SET as a Third Board or Third Tier market. The responsibility placed on the introducing stock broker as the sponsor should counterbalance the lack of detailed disclosure and track record, and ensure smooth progress towards eventual authoriza- tion or listing. - 158 - o Present regulations require that share certificates be signed individually. This should be reviewed. O The law does not prohibit preferred shareholders from having voting rights although in practice they are not usually given this right. It would be preferable to amend the law to make clear the distinction between the two kinds of shares in terms of risk taking and assurance of a financial return. o There are currently no rules governing take over activities. 'While mergers and acquisitions are not important yet, the authorities should develop principles and detailed safeguards and rules/proce- dures now so that such activities may evolve in an orderly fashion. O The number of brokers is very small in relation to the size of market and trading activity. The SET should plan to increase the number of brokers substantially. o Stock analysts are not allowed to make buy/sell/hold recommendations. This restriction should be reconsidered. o The trading methods on the floor of the exchange are badly in need of modernization. The SET should aim at fully electronic trading methods in a phased manner within a reasonable time frame. E. New Financing Methods 9.48 Thailand has reached the stage in its economic development where the needs of business would be better served by more varied and sophisticated means of financing than are available at the present time. The basic commer- cial codes must be adapted or supplemented by additional legislation to enable the introduction of new financing methods and to establish clear rules to govern their application. The following paragraphs discuss three specific financing techniques where the elaboration of clear rules would assist the needs of Thai business at this time and where by the same token, the current inadequacy of the law makes for uncertainty and acts as a disincentive for financial institutions to engage in these methods of financing. These are: (a) financial leasing, (b) factoring accounts receivable, and (c) trust receipt financing. Financial Leasing 9.49 Financial leasing, which began in the United States, is now a multi- billion dollar worldwide phenomenon. Offered by banks and other institutions, it is the functional equivalent of lending. It may give rise to advantages for both the lessor and the lessee over lending. Thus, the lessor (creditor) may enjoy certain tax benefits related to depreciation and investment tax credits, if provided by local law. From the point of view of the lessee (borrower), leasing may eliminate the need for a downpayment on the initial purchase of goods and offer better balance sheet treatment than a loan. While financial leasing may be applied to such items as automobiles, it is of par- - 159 - ticular interest in the conte3:t of economic development when applied to so- called Obig ticket" items (ships, locomotives, airplanes, etc.) and equipment leasing. In this context, the following features of the financial lease are typical: (i) There are three parties to the transaction: the supplier, the lessor (creditor) and the lessee (debtor); (ii) The lessor does not acquire the equipment that is to be the subject of the lease with a view to offering it to a class of undetermined customers, but rather purchases the equipment at the express request of a specific customer. It is the latter (who will become the lessee) who chooses the property he wishes to lease, selects the supplier and negotiates with the supplier the terms of purchase. If the bank (or other creditor) finds the customeL an acceptable credit risk, it then purchases the property specified by the customer and leases it to him. Delivery by the supplier is made directly to the customer (lessee) who then makes lease payments to the creditor (lessor); (iii) The lease is irrevocable during the period of the lease. The period of the lease is calculated to approximate the economic life of the property. It follows that it is of importance to the lessor that it has the assurance, after it has paid the supplier at the lessee's request, that the lessee will have no right to eject, terminate or otherwise claim against the lessor on account of the equipment or other property which has been purchased; and (iv) The burdens of operating costs and risks are borne by the lessee who agrees to maintain insurance, pay for maintenance and assume the risk of loss or damage. 9.50 These characteristics of a financial lease differentiate it from the more familiar operating lease in which all or most of these burdens are the responsibility of the lessor. In a financial lease the lessor remains a creditor and does not hold himself out as expert in the maintenance of equip- ment or other property. Such expertise falls in the purview of trade, from which banks and others are ordinarily excluded by law. 9.51 There are provisions in Thai CCC that arguably might apply to a financial lease. These are the provisions relating to the hire of property set out at sections 537-571. However, they are inadequate to cover financial leases. Thus, section 546 provides that the lessor is responsible for delivering the property in a good state of care, while Section 548 provides that if the property delivered is not in condition or suitable for the purpose for which it is let, the lessee may terminate the lease. According to Section 550, the lessor is responsible for any defects which arise during the lease and must make necessary repairs. If the lessor fails to make such repairs within a reasonable time. Section 551 permits the lessee to' terminate the contract. None of these provisions would be consistent with the features of a financial lease, as set out above. - 160 - 9.52 As a consequence of the inadequacy of these laws, financial leasing has not developed in Thailand. A number of countries have adopted specific financial leasing laws to govern the rights and liabilities contemplated by that activity. It is recommended that Thai authorities also do so. In addition, UNIDROIT, an international organization based in Rome, Italy, has produced a Convention on International Financial Leasing which was signed by 55 countries at Ottawa in 1988. While the convention is intended to cover international leases, it may readily be adapted for domestic use. According- ly, the Thai authorities may wish to consider such adaptation in order to produce a set of rules that, by clarifying the rights and responsibilities in this area, would encourage its development and use by banks and other finan- cial institutions to meet the increasing needs of Thai business. Factoring 9.53 Factoring, as a concept and an activity, has evolved over the years. Originally a factor was a selling agent and often he came to guarantee to his principal, the seller, the accounts of the purchasers to whom he made sales. Gradually, the financing function of the factor gained ascendance over his selling function. In order to safeguard their interests, factors often sought and obtained clarification of the law of security where they were doing busi- ness. It was important for the law to allow a factor to obtain a nonposses- sory lien on merchandise and its proceeds and for that lien to cover after- acquired merchandise (that is to say, merchandise acquired in the future as well as that which was already the subject of the lien) so that business could flow smoothly without the necessity of repetitive formalities with respect to each new batch of goods and associated accounts. The nature of facto. ing continued to evolve and its practitioner2, once individuals, came to include banks and finance companies. Today the activity usually encompasses purchasing the accounts receivable of a supplier for cash by a financial institution. 9.54 In the typical accounts receivable situation, a supplier will sell merchandise to this customers on time and will acquire an inventory of accounts receivable from them evidencing the sales. Instead of tying up his own funds while awaiting maturity of the accounts, the supplier will sell the accounts receivable at a discount from their face value to a factor. The advantage that occurs to the supplier through the factoring of his accounts receivable is an augmentation of the supply of operating cash through the increased turnover of working capital. At this point alternative procedures may be used. 9.55 The factor or the supplier may notify the supplier's customers of the assignment of their debts and the customers may then make payment directly to .he factor as the accounts become due. Alternatively, there may be no notice of the assignment and the customer may continue to make payments to the sup- plier who will accept such payments as agent of the factor and forward them to him. Another set of alternatives involves whether or not the arrangement between the supplier and the factor contemplates that, in the event of nonpay- ment by a customer, the factor may have recourse against the supplier. If the answer is in the affirmative, the supplier would in such case guarantee pay- ment of the accounts. _ 161 - 9.56 In Thailand, factoring of accounts receivable is done by a few finance companies. It is done on a notification to the debtors basis in which the supplier stands guarantor of the accounts. In effect the accounts receivable serve as collateral to a loan made to the supplier. The legal situation is, however, unclear and the activity is consequently limited. There is some thought that factoring might be accommodated under the provi- sions of the Thai CCC (Section 229 and Section 306). It is unclear, however, whether these provisions would be adequate to cover factoring. 9.57 UNIDROIT has also produced a Convention on International Factoring which was signed in Ottawa in 1988. As with the Convention on International Financial Leasing, the provisions of the factoring convention may be considered as a basis for a domestic legal framework which could remove uncer- tainties and encourage expansion of financing techniques for Thai business. Trust Receipt FinancinR 9.58 The Thai CCC contemplates two bazic forms of sscurity for a creditor. It provides for a pledge and a mortgage. A pledge is a possessory lien, while a mortgage does not require the debtor to transfer the subject property to the possession of another. However, the mortgage must be registered with a competent official. Thai law, however, only provides for the registration of mortgages on real estate and certain specified items such as ships, floating houses, beasts of burden and machinery fixed in place. Ordinarily, the regis- tration requirements of a mortgage law stipulate that the property must be described in specific detail. This requirement, together with other formal- ities, makes the mortgage procedure cumbersome with respect to some types of collateral, such as inventory, which are vital for the financing of business. Moreover, the express limitations on the types of items subject to registra- tion described above would appear to exclude inventory.62/ 9.59 In order to deal with the problems of financing inventory, banks in the United States developed a new security device in the nineteenth century which was subsequently given statutory expression. This is the trust receipt. The statute which was passed to validate trust receipt financing provided for the bank to file notices in order to secure its interests against other creditors of the buyer. The notices, however, did not require the formalities involved with chattel mortgages. Moreover, it was not necessary to describe the subject of them with great particularity. A simple description of the kind of goods covered was sufficient such as 'coffeel, *silk*, or 'auto- mobiles'. Thus the arrangement could contemplate a moving series of trans- actions based on a changing inventory. As a result of the development of trust receipts, banks could remain assured of a valid security interest and business could receive needed financing of inventory. 9.60 Trust receipt financing is not unknown in Thailand. However, it is practiced to finance import trade only, by including the outstanding amount in 62/ Other securities devices used in Thailand include the use of warehouse receipts, hire purchase and assignment. - 162 - the reported bills discounted item. There is no express recognition of trust receipt financing in Thai statutes. In order to provide a generally accepted means of financing inventory for Thai business, it is recommended that con- sideration be given to drafting new legislation that would expressly validate this or a similar security interest.