I

In contrast to the abundance of theoretical work on the company's capital structure1, practical research in this area is rather sparse. This is due to the fact that, firstly, in most cases, the indicators that are actually available for analysis can simultaneously reflect many theoretical hypotheses, and, secondly, there is a lack of reliable information on the capital structure of companies. As a result, most of the studies were based on data for the United States, and in some cases for Japan, Germany, and the United Kingdom. The exceptions are the works of K. Mayer (1990), R. Ryan and L. Zingales (1995), which used data for a number of developed countries, as well as L. Demirgyuk-Kunt and others.

V. Maksimovich (1999) and L. Bus et al. (2001)², who conducted a comparative analysis of the factors that determine the capital structure of companies in developed and developing countries.

There are two main sources of data that reflect how companies attract external financing: national capital flow tables and company financial statements. The first one provides information on capital flows between economic sectors. Until recently, it was more accessible, and its advantage is that it covers data for all companies in the country. The financial reporting database includes only companies listed on national exchanges, but the data itself is more detailed and allows you to track trends within individual companies, rather than by sector as a whole. In the 1990s, the quality of second-source information improved significantly. However, its disadvantage is that it mainly covers data on large companies and does not take into account a large layer of small and medium-sized enterprises. Therefore, conclusions should be drawn with some caution, as individual indicators may not be representative. Nevertheless, the whole set of data provides an overview of the processes that took place at the microeconomic level in the countries of the region.

Using data on capital flows of several developed countries, K. Mayer made the following main conclusions.

1. Retained earnings are the main source of financing in most countries.

2. In none of the countries studied do companies raise a significant part of their capital through the issuance of shares. Banks are the largest source of external financing

Polina DYACHKINA, PhD student at the Institute of Asian and African Studies at Lomonosov Moscow State University and Keio Gijuku University (Tokyo).

¹ The cornerstone of theoretical research on the structure of capital was the work of F. P. Blavatsky. Modigliani and M. Miller (Modigliani F., Miller M. The Cost of Capital, Corporation Finance and the Theory of Investment / / American Economic Review. 1958. N 48), in which they gave evidence that the capital structure, that is, the sources of financing of a company, does not have any influence on its value. The price of a company is determined only by what assets it has. According to the authors, in an economy with perfect financial markets, it does not matter whether companies attract funds through bank loans, through the issuance of shares, or simply by reinvesting retained earnings. The work sparked a flurry of research extending Modigliani and Miller's original model. Given the imperfection of the market, researchers noted the existence of such factors that significantly affect the choice of funding sources as transaction costs (Higgins R. How Much Growth Can the Firm Afford / / Financial Management. 1997), asymmetric information (Myers S., Majluf N. Corporate Finance and Investment Decisions when Firms Have Information that Investors Don't Have // Journal of Financial Economics. 1984. N13), проблема отношений заказчика и исполнителя (Jensen M., Meckling W. Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure // Journal of Financial Economics. 1976. N 3; Myers S. The Determinants of Corporate Borrowing // Journal of Financial Economics. 1977. N 5), comparative advantages of various tax shields (Miller M. Debt and Taxes / / Journal of Finance. 1977. N 32), the probability of bankruptcy (Jensen M. The Agency Costs of Free Cash Flow, Corporate Finance and Takeovers // American Economic Review. 1986. N 76).

² См.: Mayer C. Financial Systems, Corporate Finance and Economic Development, in Glenn Habbard. Chicago, 1990; Raghuram R., Zingales L. What do we Know about Capital Structure? Some Evidence from International Data // Journal of Finance. 1995. N50; Demirguc-Kunt A., Maksimovic V. Institutions, Financial Markets and Firm Debt Maturity // Journal of Financial. Economics. 1999. N 54; Booth L., Aivazian V., Demirguc-Kunt A., Maksimovic V. Capital Structures in Developing Countries // The Journal of Finance. 2001. N 56.

page 100

even in countries like the US, UK and Canada.

3. There is a clear inverse relationship between financing from retained earnings and bank loans. Small and medium-sized enterprises rely more on external financing than large ones. The latter use more equity than the smaller ones, and the smaller ones are more likely to resort to bank loans.

4. Countries with stock-market financial systems (USA, UK, Canada) are more dependent on retained earnings and less likely to resort to bank loans than "bank" countries (continental Europe and Japan).

Mayer's conclusions are based on data summarizing capital flows across individual sectors and do not reflect the processes occurring in individual companies, so they should be treated with some caution. The conclusion that the issue of shares is not a significant source of external financing in any of the countries studied (even in the United States, Great Britain, and Canada) is quite natural, since data on capital flows include all companies operating in the economy, and most of them do not resort to share issues. The role of stock markets can be traced only by the example of companies listed on the national exchanges of the studied countries.

R. Ryan and L. Zingales analyzed the factors that can determine the capital structure of the G7 companies. Contrary to the conclusions made by Mayer and others based on data on capital flows that the structure of the financial system affects the level of debt of companies, the analysis of balance sheets of large companies conducted by Ryan and Zingales did not support this hypothesis. It showed that the level of borrowed capital is approximately the same in the entire sample of countries (only in Germany and the UK it is slightly lower) and was determined by similar factors. These findings are at odds with the usual notion that Japanese and continental European companies have higher debt levels than Anglo-American ones.

Ryan and Zingales also examined the institutional factors that influence capital structure - the tax system, bankruptcy proceedings, corporate governance, and the traditional role of banks and stock markets in these economies. They concluded that the difference in countries with banking and stock market systems is mainly reflected in the comparative preference of loans to securities (not only stocks, but also bonds), and not in the level of debt as such.

Some researchers have tried to trace the difference in the capital structure of Japan and the United States-countries with banking and stock-market financial systems. K. Deventer and V. Barter found that information asymmetry ³ is less important in Japan - American investors are more closely monitoring changes in the dividend policy of companies than Japanese ⁴ . In Japan, information asymmetry is significantly lower in companies belonging to the Keiretsu industrial groups, which has led to a reduction in the cost of borrowing for these companies ^.5 When financial difficulties arise, US companies face an increase in the cost of borrowed capital, ⁶ whereas in Japan this does not happen - on the contrary, banks are more active in lending to distressed enterprises ^.7

II

Analysis of the capital structure of companies in developing countries has long been complicated by a lack of data. This gap was first filled in in the early 1990s by a group of researchers from the International Development Corporation (IDC). It collected financial information on companies in a number of developing countries (India, Pakistan, South Korea, Jordan, Thailand, Mexico, Turkey, and Malaysia). One might expect that in these countries, market imperfections (information asymmetry and transaction costs) are more pronounced than in developed countries, and the share of external sources of financing is less. However, the conclusions of the ICD researchers differ from this assumption: retained earnings are an important source of financing for companies in developing countries, but to a lesser extent than in developed countries; the share of external financing is low.

³ Borrowing companies have the most complete information about the prospects of projects for which funds are raised. Creditors or shareholders cannot be sure that the invested funds will make a profit and will be returned. This phenomenon is called information asymmetry.

⁴ См.: Dewenter K.L., Warther V.A. Dividends, Asymmetric Information and Agency Conflicts: Evidence from Comparison of the Dividend Policies of Japanese and US Firms // Journal of Finance. 1998. N 53.

⁵ См.: Hoshi Т., Kashyap A., Sharfstein D. Corporate Structure, Liquidity and Investment: Evidence form Japanese Panel Data // Quarterly Journal of Economics. 1991. N 106(1).

⁶ См.: Fazzari S.M., Habbard R.G., Petersen B.C. Financing Constraints and Corporate Investment // Brookings Papers of Economic Activities. 1988. N 1.

⁷ См.: Kaplan S.N., Minton В A. Appointments of Outsiders to Japanese Boards: Determinants and Implications for Managers // Journal of Financial. Economics. 1994. N 36.

page 101

more than in developed countries; the role of equity is gradually increasing in developing countries and decreasing in developed countries ^.8

It is quite possible that these conclusions are determined by the specifics of the data used and the sample of countries. The lack of alternative research makes it necessary to analyze this issue on the basis of broader material. In particular, it is interesting to study the capital structure of companies in the countries of East and South-East Asia that were most affected by the currency and financial crisis of 1997.

In a study of the financial statements of East Asian companies, the year 1997 represents a certain boundary. The currency and financial crisis was preceded by a deterioration in many microeconomic indicators, but debt restructuring and the recovery of companies improved their financial situation. When analyzing the external debt-to-equity ratio of companies for 1992-1996, we note a stable trend towards its increase as it approaches 1997 in all countries of the region, except Taiwan and the Philippines. So, the average debt of companies in Thailand, Indonesia and Malaysia has approximately doubled over the years, while in Hong Kong and Singapore it has increased by 39 and 57%, respectively. The average debt level of South Korean companies remained almost unchanged, and they continued to be the undisputed record holders in the region in this indicator.

Korean, Thai and Indonesian companies had the highest levels of debt in 1996 (the ratio of loans and borrowings to total assets) (1.94, 1.17 and 0.82), while Singapore and Hong Kong had fairly high levels (0.52 and 0.49). The most moderate levels were observed in Malaysia, Taiwan and the Philippines (0.45, 0.42 and O. S)⁹ .

For comparison, in 1991 this indicator was 0.38 in the United Kingdom, 0.59 in the United States, 0.61 in Germany, 0.63 in Canada, 0.88 in Italy, 0.92 in France, and 1.13 in Japan. These data show that in 1996, immediately before the monetary and financial crisis, the level of debt in the countries studied was somewhat lower. It is higher than in developed countries, but by 2001 these indicators were approximately equal. Note that the data are not fully comparable, since the analysis used different sources for different years.

The increase in corporate debt in 1993-1996 was largely due to the rapid increase in the fixed capital of Asian firms. During these years, the average annual growth of fixed capital was 33% in Indonesia, 29% in Thailand,25% in Singapore, 20% in Malaysia, 17% in South Korea and Hong Kong, 11% in the Philippines and 8% in Taiwan.

It is noteworthy that the share of short - term liabilities in the total debt of companies in the region in 1996 was extremely high-52% in South Korea, 80% in Singapore, 65% in Taiwan, 52% in Indonesia, 61% in Malaysia, 62% in Thailand, and 63% in Hong Kong. This debt structure is a feature of the East Asian method of lending. A significant proportion of loans were issued by banks for a period of less than a year, after which the contract was renewed. By insuring itself against "bad" debts in this way, the bank always reserved the right not to renew the contract for any reason. This did not create significant difficulties for companies during periods of stable growth, but in crisis years it could cause the borrower to go bankrupt. This is what happened in 1997-1998. All countries in the region had high levels of short-term debt, but the highest rates were in Indonesia, Malaysia, Thailand and Hong Kong.

Between 1997 and 2001, debt levels declined in all countries except Taiwan. The biggest declines from 1996 to 2001 were in South Korea (61%), Hong Kong (49%), and Thailand (43%). However, despite the decline, South Korea, Thailand and Indonesia continue to be the countries with the highest debt levels in the region (0.76, 0.67 and 0.76).

Given this circumstance, it is advisable to analyze the capital structure of Asian companies more closely. This will help you understand what factors influence companies ' choice of a source of financing, determine the share of external debt, whether these factors are similar to those in developed countries, and how various institutional characteristics (the level of development and structure of the financial system, the legislative framework, and the quality of law enforcement) affect the capital structure.

In the regression analysis ¹⁰ conducted by the author, accounting data from-

⁸ См.: Singh A., Hamid J., Salimi В., Nakano Y. Corporate Financial Structures in Developing Countries. Technical Paper #1. International Finance Corporation. Wash. DC, 1992.

⁹ Calculated from the financial statements of Asian companies from the Worldscope database of Thomson Corporation.

¹⁰ For more information on the methodology and results of regression analysis of factors affecting the capital structure of companies, see: Dyachkina P. Analysis of the capital structure of companies in Eastern and South-Eastern Asia / / Bulletin of Moscow University. Series N 13. Oriental studies. 2005. N 2.

page 102

parity of industrial companies collected in the Worldscope¹¹.

For the study, manufacturing companies were selected that correspond to the 20 - 39 lines of the US Standard Industry Classification (SOK USA)¹² . Screening by industry was carried out based on the following considerations: regardless of the quality of available financial services, companies may have different external financing needs that correspond to their type of activity. To minimize the impact of this factor, we used data for manufacturing companies only. As a result, we have data on 2,408 companies in Singapore, South Korea, Taiwan, Malaysia, Thailand, Indonesia, the Philippines, and Hong Kong.

In order to understand why the level of external debt of companies varies by country, it is necessary to analyze the factors that affect the structure of their capital. Among them, it is possible to name both specific to the companies themselves (size, share of fixed capital, profitability, etc.), as well as institutional, country-specific (structure of the financial system, taxation, legislative framework, strictness of law enforcement). A more in-depth analysis of all these indicators should show whether the same factors are at work in developing countries as in developed ones, and which of them are most important.

Our task is to confirm or refute the following assumptions.

1. The level of debt of companies depends on their size - larger companies have easier access to loans, therefore, the coefficient for a variable company size will be positive.

2. The share of fixed capital in the assets of companies in developed countries is positively related to the level of external debt, since the presence of collateral in the form of tangible assets facilitates the attraction of external financing. A similar dependence can be expected in companies in developing countries, although collateral is not as important for short-term lending, which prevails in this region, as it is for longer-term loans.

3. Dependent variables are influenced by market expectations in relation to the investment opportunities of the company, expressed in the market value of shares. Theoretically, there is a negative relationship between these indicators: a) lenders are more cautious than shareholders about new promising projects, so their financing can be carried out more through the issuance of new shares than by attracting new lending; b) companies tend to issue new shares at times when they think that the value of shares is too high (when shares are expensive). These considerations primarily apply to developed countries. In developing countries, the market is focused on growth, so an increase in stock prices can serve as a positive signal not only for portfolio investors, but also for lenders, so you can not expect an unambiguous negative relationship between these indicators.

4. As in developed countries, in developing countries we can expect a negative relationship between a company's profitability and the level of its external debt. Companies seek to reduce the cost of attracted investment funds. The cheapest source of financing is retained earnings, which means that more profitable companies will rely less on loans and the relationship between debt and profitability indicators will be negative.

5. More active stock markets contribute to the reduction of external debt of companies (negative dependence). In a number of the countries studied, a targeted policy was implemented to develop stock markets. Many companies issued a certain number of shares under state pressure, but they do not resort to this source of financing regularly, but continue to rely on a bank loan. Large but passive markets do not have a significant impact on the volume of attracted lending (there is no correlation between the size of the stock market and debt indicators).

6. There is a positive correlation between the size of the banking sector and the volume of loans and credits issued by companies.

7. Legal factors have a significant impact on the way companies are financed externally. Legislation protecting the rights of shareholders and creditors, in conjunction with the ef-

¹¹ In recent years, new databases containing financial statements of companies registered on national exchanges have appeared-Worldscope, Global Vantage, IFC Developing Countries Database, Moody's International Company Data, Financial Times Information's Extel Card Database. The author uses the Worldscope database available at Keio University.

¹² In the English-language literature, this classification is called SIC (Standard Industrial Classification). It is used in most modern databases on balance sheet indicators of companies.

page 103

the effectiveness of the application of these laws determines the amount of bank credit and equity in the financing of companies.

A regression analysis of the impact of intra-company factors (company size, profitability, share of fixed capital in assets, investment opportunities) on companies ' debt, conducted separately by country, showed the following.

1. In all countries of the region, fixed capital is positively associated with the level of debt. In other words, the more fixed assets that can serve as collateral in the event of a loan, the higher the share of borrowed capital in the company's liabilities. This indicator is statistically significant in Hong Kong, Singapore, Taiwan, Thailand, and South Korea.

2. The positive relationship between the size of the company (the natural logarithm of sales) and the level of debt is also confirmed by statistically significant coefficients in most of the sample countries. The exceptions are Malaysia and Hong Kong.

3. Profitability (calculated as the ratio of income before payment of bonds and taxes to the total assets of the company) negatively affects the decision of companies to raise additional funds, higher profits make it possible not to resort to external financing through loans and borrowings.

4. As for the company's prospects, expressed by the ratio of the market and book value of its assets ¹³, its impact on the dependent variable is ambiguous. The level of borrowed capital in market prices is negatively related to the assessment that the market gives to the company and its growth prospects. This corresponds to the theory that if there are promising projects that are too risky from the point of view of creditors, managers will prefer alternative sources of financing to debt - issuing new shares or reinvesting retained earnings. In contrast, when using a balance sheet debt indicator rather than a market indicator, the coefficients for this variable are either positive or statistically insignificant.

So, in general, in the studied countries, their characteristics affect the capital structure of companies in the same way as in developed countries. The capital structure is also affected by indicators that characterize the macroeconomic environment.

III

The countries studied vary greatly in their level of economic development, which may be one of the factors determining the capital structure of companies. Singapore, as well as Hong Kong, are among the developed countries ($24,505 and $ 27,118 per capita in 2001), while South Korea and Taiwan are about halfway there ($13,502 and $ 12,860). Malaysia and Thailand are three times behind South Korea and Taiwan (their per capita GDP was $ 4,708 and $ 2,853 in 2001), while the Philippines and Indonesia have the lowest rates in the region ($1,165 and $ 1,034).

The size of the stock market and banking sector relative to the GDP of these countries is not proportional to per capita GDP. The average figures for 1998-2001 were 323 and 162 in Hong Kong, 25.6 and 29 in Indonesia, 56.9 and 97.5 in South Korea, 146.3 and 149.4 in Malaysia, 164.2 and 120.3 in Singapore, 33.5 and 123.5 in Thailand, 131.5 and 136.6 in Taiwan, and 54 and 46.2 in Japan. Philippines ¹⁴ . The size of the stock market changed quite a lot from 1996 to 2001, which is associated with sharp fluctuations in securities prices. Some movement in the direction of reducing the size of the banking sector over the years was associated with the difficult situation in which banks, as well as businesses that are being credited, found themselves after the currency and financial crisis (a simultaneous decline in both supply and demand for loans). However, in general, this indicator was more stable than the size of the stock market.

One of the main parameters by which countries are classified as "banking" and "stock-market" is the relative size of the stock market and the banking sector. Despite fluctuations primarily related to the instability of the second half of the 1990s, it provides a clear picture of the ratio of these sectors. Hong Kong, Singapore and Malaysia belong to the stock market financial systems (1.85, 1.31 and 1.1015 ), Korea and Thailand belong to the banking systems (0.25 and 0.49). Taiwan (0.92), which has traditionally relied on the banking system to finance economic growth, occupies a borderline position due to the rapid growth of the stock market in recent years. late 90's. The Philippines and Indonesia belong to the market and banking groups (1.16 and 0.67), but

^{This indicator reflects the} difference between the book value and the market value of the company's assets, that is, it shows the mood of investors regarding its prospects. It is defined as the sum of the company's assets minus the book value of the share capital plus the market value of the shares divided by the total assets.

¹⁴ Calculated by: World Development Indicators on CD-ROM. World Bank, 2003.

¹⁵ Ratio of the average indicators of capitalization of the country's stock market and the total volume of loans of financial institutions to the private sector for 1996-2001.

page 104

not because of the large size of these sectors, but because of the underdeveloped financial system. This is confirmed by indicators of private sector lending and stock market capitalization. In 2001. They were only 20.5% and 15.8% for Indonesia and 40.1% and 29.9% for the Philippines. In the rest of the region, all indicators exceed their annual GDP.

The regression equations included parameters that reflect these macroeconomic characteristics, first in the form of dummy country variables, and then in the form of institutional and legal characteristics.

The coefficients for the dummy variables indicated that the debt of companies in all countries is lower than in Thailand, with the exception of Indonesia and South Korea. The statistical significance of the coefficients of these variables serves as evidence that the level of debt of companies depends on their country of origin.

In addition, the impact of a number of country characteristics on the capital structure of Asian companies was studied. These are per capita GDP-a parameter that controls the country's economic level of development; institutional characteristics - the size of the stock market (the ratio of its capitalization to GDP) ; its activity (the ratio of the size of transactions concluded on the stock markets per year to GDP) and efficiency (the ratio of the volume of transactions to the total market capitalization); the size of the banking system (the annual volume of private sector loans in relation to GDP); legal characteristics-composite indices that reflect the quality of laws that protect the rights of shareholders and creditors, as well as the effectiveness of their application.

A negative coefficient for the GDP per capita indicator indicates that as the economy grows, companies ' dependence on loans and borrowings will decrease. Of course, it is not possible to draw definitive conclusions about the impact of economic development on the financial system based solely on these data for eight countries in East and South-East Asia. However, it can be assumed that companies in developing countries are more dependent on external financing than in developed countries. Note that the same conclusion is reached by the research of a group of MKR economists.

The coefficients of institutional indicators behave as follows. As the capitalization of the stock market increases and its efficiency increases, the role of loans and credits in the capital structure of companies in the region decreases. On the contrary, the activity of the stock market has a positive effect on debt: the more active transactions with securities are made, the more accessible information about the quality of the company becomes not only for investors in the market itself, but also for credit institutions. This increases the attractiveness of companies as borrowers and leads to an increase in bank lending.

A negative coefficient for the variable "bank credit to the private sector" is a result that is at first glance unexpected, but coincides with the conclusions of other researchers. Increasing the volume and quality of services provided by credit and financial institutions does not lead to an increase in companies ' debt. In countries with a higher level of financial system development, companies try to attract funds from other sources. It can be assumed that this is due to the relatively large size of the companies studied (only those with access to stock markets), and the debt of small and medium-sized companies, on the contrary, will increase as the volume of lending to the private sector by banks increases.

Thus, the cumulative increase in the size and quality of financial services provided by both the stock market and credit and financial institutions leads to a gradual reduction in the share of loans and credits in the external financing of companies. This conclusion is confirmed by the results of a study conducted by the ICR group: in developed countries, retained earnings are the most significant source of financing, and developing countries are more dependent on external debt than developed countries ^.16 This can also explain the low level of debt of Taiwanese companies, despite the fact that Taiwan has a typically banking financial system.

The results of the regression analysis confirmed the significant role of the country's legal and regulatory framework in the question of the capital structure of companies. In particular, effective protection of creditors ' rights leads to an increase in the share of loans and borrowings in the financing of companies, and in countries where the rights of shareholders are better respected, the opposite effect is observed.

* * *

Thus, companies in East and South-East Asia (South Korea, Singapore, Hong Kong, Malaysia, Thailand, and Indonesia) had high rates of external debt before 1997, with short-term loans prevailing. This was one of the reasons for the Asian currency crisis.-

¹⁶ See: Singh A., Hamid J., Salimi V., Nakano Y. Op. cit.

page 105

By 2001, the debt level of companies in the region had significantly decreased in comparison with the pre-crisis period of 1995-1996 and was approximately equal to that of developed countries.

The factors that determine the capital structure of companies in developed countries and in the countries of East and South-East Asia are similar. However, country characteristics also play a role here, in particular the level of development of stock markets and credit and financial institutions. More developed stock markets contribute to a decrease in the share of loans and credits in external financing of companies. An increase in the volume of loans and the quality of services provided by banks also leads to a reduction in external debt, at least for large companies. Finally, the choice of a company's source of financing, as well as the actions of credit institutions and shareholders, is influenced by the protection of shareholders 'and creditors' rights and the effectiveness of law enforcement.

Permanent link to this publication: