THE CASE OF CHINA
SIJMEN H. PEREBOOM THESIS (MAIN TEXT)
INTERNATIONAL BUSINESS & MANAGEMENT University of Groningen
Faculty of International Business & Economics Landleven 5
ABSTRACT
Leading indicators are generally used either to explain past financial crises or to examine if they could have predicted past financial crises. This paper uses leading indicators to predict the likelihood of a future financial crisis in China. The results obtained with the signals approach indicate that there is a below average likelihood of a future financial crisis in China, using the most detailed data available for the period 1997-2004.
Keywords: financial crisis, China, leading indicators, signals approach.
INTRODUCTION
The 1990s reflected several financial crises: the ERM crisis in 1992-1993, the Mexican crisis in 1994-95, the Asian crisis in 1997-98 and the Russian crisis in 1998. These financial crises disrupted financial markets in many countries, which resulted in large costs to recover from the experienced financial crises. The large costs were a reason for policy makers to examine the possibility to predict financial crises. Several authors examined different financial crises and tried to find indicators that could have predicted these financial crises. Early warning systems (EWS) were developed, which is a tool to monitor the condition of financial systems. Typically, the backbone of an early warning system is an empirical model, which includes a precise definition of a crisis and a number of selected economic indicators. The values of these indicators are usually different before, during and after a crisis, so their movements can be used to forecast the probability of the onset of a crisis (Krznar, 2004: 2).
There are only very few studies that focus their attention on predicting a future financial crisis for countries. To study the possibility of a future financial crisis for countries, the results of the EWS studies may be of use. The leading indicators from the existing studies can be used to examine if these leading indicators play an important role in the financial system of the examined countries. However, different authors have mentioned that specialists should continuously search for other leading indicators, because crises seem to be substantially different from each other. For example, Kaminsky et. al. (1998: 5) already mentioned in their paper that leading indicators might also include political variables. Therefore, studies that will focus on predicting a future financial crisis for countries should keep in mind that other factors might be important besides the existing leading indicators.
The aim of this study is to examine the likelihood of a financial crisis in China, focusing on existing indicators from studies on other countries, which may be important in China’s financial system1. As noted above, other factors may be important in China's financial
system, for the prediction of the likelihood of a financial crisis. Therefore, a search for other factors other than already known leading indicators will be included as well. The main research question used in this study is as follows:
“What is the likelihood of a financial crisis in China?”
For the answering of the main question the literature written on financial crises needs to be reviewed. Reviewing the financial crises literature will provide an overview of leading indicators that may predict financial crises. Moreover, China’s financial system needs to be examined to see which factors are important. The following investigative questions are used to gather the needed information:
− Which leading indicators are given in the existing literature for predicting financial crises?
− What are the main characteristics of the financial system in China?
− Which leading indicators from the existing literature are important in China’s financial system and thus relevant for the prediction of the likelihood of a financial crisis in China?
− Besides the leading indicators from the existing literature, are there other factors important for China’s financial system?
The main research question and the investigative questions are the input for the conceptual framework. The conceptual framework presents a visualisation of the steps that need to be taken before a prediction about the likelihood of a financial crisis in China can be made.
Insert figure A1 about here
A substantial body of literature exists on financial crises, including many leading indicators that may explain and/or predict financial crises. Leading indicators are fundamental determinants of financial crises. In other words, leading indicators are factors that are the primary causes of financial crises. The indicators that may be relevant for predicting the likelihood of a financial crisis in China may include indicators from this body of literature. However, before a review of this body of literature will be given, the theory behind financial crises will be explained first. Moreover, common methodologies used in financial crises studies will be explained as well.
I. FINANCIAL CRISIS THEORY
The efficiency of an economy is dependent on the efficiency of its financial system. The most important function of a financial system is to facilitate the allocation of capital. Furthermore, a financial system provides investors the opportunity to share risk, access to capital and access to information (Garretsen et. al., 1999: 20). Although financial markets provide information, the division of information is asymmetric. According to Mishkin (1996: 2), asymmetric information occurs when two parties are involved in a contract, while one party has more accurate information than the other party does. Moreover, the presence of asymmetric information is a hindrance to the efficiency of financial markets. Asymmetric information may lead to moral hazard and adverse selection when (potential) borrowers of capital are aware of the presence of asymmetric information (Garretsen et. al., 1999: 23). Adverse selection and moral hazard are two basic problems that may occur in financial systems.
Adverse selection
Adverse selection occurs when the most likely chosen parties in a trade or credit relationship are parties who most likely produce an adverse (undesirable) outcome (Mishkin, 1996: 2). For example, the lender of capital selects those parties who are the least risk averse. Moreover, due to adverse selection the parties who are the most eager to take out a loan are the parties who are the least risk averse. Adverse selection thus increases the probability of bad credit risks for lenders, which may result in a situation where lenders decide not to make any loan at all, because this probability is too high.
Moral hazard
Moral hazard occurs after the occurrence of a transaction when the borrower shows undesirable behaviour. A borrower shows undesirable behaviour when the borrower’s behaviour deviates from the preferred behaviour from a lender’s point of view. In other words, the borrower engages in activities that make the repayment of the loan less likely (Mishkin, 1996: 3). For example, moral hazard occurs when the borrower has the incentive to invest in high-risk projects, because the borrower does not bear most of the costs in case of failure. Moral hazard may result in a situation where lenders decide to make fewer loans. Therefore, both adverse selection and moral hazard may thus result in situations where the level of lending decreases. A lower level of lending results in a lower level of investments, which in turn results in a decline in aggregate economic activity. Furthermore, adverse selection and moral hazard decrease the efficiency of financial markets.
Financial systems, in general, are thus characterised by imperfect (efficiency is suboptimal) financial markets. However, a financial system that is characterised by imperfect markets not necessarily means an unstable system. A financial system that is unable to fulfil its most important function (to facilitate the allocation of capital) properly after a sudden disturbance of the system is considered an unstable system (Garretsen et. al., 1999: 31). When a sudden disturbance occurs in an unstable financial system, a financial crisis may occur. A financial crisis occurs when a sudden disturbance leads to a sharp contraction in economic activity, eventually leading to stagnation in economic activity due to the inability of financial markets to function efficiently (Mishkin, 1999: 18). The occurrence of a financial crisis thus depends on the stability of a financial system.
Definition of a financial crisis
“A financial crisis is a nonlinear disruption to financial markets in which adverse selection and moral hazard problems become much worse, so that financial markets are unable to efficiently channel funds to those who have the most productive investment opportunities” (Mishkin, 1996: 17).
Increases in uncertainty
A sharp increase in uncertainty in financial markets, for example due to a failure of a prominent (non-) financial institution or political instability, makes it harder for lenders to measure bad credit risks (Mishkin, 1996: 19). Moreover, a sharp increase in uncertainty makes information in financial markets even more asymmetric. Therefore, problems of adverse selection increase, because screening of credit is more difficult. Lenders are thus less able to solve problems of adverse selection, making them less willing to make loans. There will be less lending activity, leading to a lower level of investment, because investments are often financed by loans. The lower level of investment results in diminished aggregate economic activity. A sharp contraction in economic activity may lead to a financial crisis, especially when it leads to stagnation.
Increases in domestic interest rates
According to Mishkin (1996: 18), higher domestic interest rates attract more high-risk projects, because borrowers who are willing to pay the higher interest rate offer these high-risk projects. Furthermore, fewer low-high-risk projects are offered, because borrowers who offer these are unwilling to pay the higher interest rates. Higher domestic interest rates thus increase the probability that a borrower is a bad credit risk, thereby increasing adverse selection. Lenders will have more difficulties to discriminate between good and bad credits, making them less likely to make a loan. The amount of loans made decreases resulting in a lower level of investment and aggregate economic activity. Increases in domestic interest rates thus may help precipitate a financial crisis.
Hence, both factors may lead to a contraction in economic activity. Leading indicators are direct or indirect related to these two factors, which explains why leading indicators are used as fundamental determinants of financial crises.
Leading indicators of financial crises
Different leading indicators may explain why a disturbance in financial markets leads to a sharp contraction in economic activity, which in turn leads to a financial crisis. Furthermore, different leading indicators may explain different types of financial crises. Many studies divide leading indicators in five categories (sectors): external sector/current account, external sector/capital account, financial sector, domestic real/public sector, and global economy sector.
External sector/current account
The real exchange rate and the ratio current account to GDP are widely used indicators of the external sector/current account. The real exchange rate is a measure of the change in a country’s international competitiveness and a proxy for the valuation of the domestic currency. An overvalued domestic currency inhibits the risk of speculative attacks, especially when investors believe that the currency will depreciate or will be devaluated in the future. Monetary authorities often increase the domestic interest rate to counter speculative attacks. A higher domestic interest rate makes it more attractive for foreigners to invest. The resulting increased demand for the domestic currency prevents depreciation. However, higher interest rates increases problems of adverse selection (see the previous subparagraph). Therefore, using domestic interest rates as a tool to defend the domestic currency may lead to less lending activity. An overvalued domestic currency thus increases the probability of a currency crisis, because a sharp rise in interest rates may lead to a decrease in economic activity.
The ratio current account (balance) to GDP is generally associated with capital flows. A positive ratio means that a country’s exports exceed its imports, thus the country has net capital inflows. Deterioration of this ratio is associated with lower capital inflows, when the ratio remains positive. However, a negative ratio is associated with capital outflows, deteriorating a country’s reserve position. Deterioration of the reserve position may make investors uncertain about the solvability of the country. The previous subparagraph outlined that increased uncertainty makes investors less willing to make loans. Furthermore, due to increased uncertainty, foreign investors may withdraw their capital massively. A massive withdrawal of capital leads to liquidity problems when borrowers are unable to meet the demands of the foreign investors that withdraw their capital.
withdraw their deposits from solvent as well insolvent banks, because they do not know the loan portfolios of banks and thus cannot distinguish between them (Mishkin, 1996: 23). The activities of banks and between banks diminish, leading to a decline in investment and aggregate economic activity (Garretsen et. al., 1999: 43). Therefore, a negative ratio current account to GDP in combination with uncertainty among both foreign and domestic investors increases the probability of a financial crisis.
External sector/capital account
Leading indicators often associated with the external sector/capital account are foreign debt growth and the ratio M2 to (foreign exchange) reserves. Foreign debt is debt denominated in foreign currency. However, the value of the domestic currency determines the burden of foreign debt. Furthermore, firm’s assets are typically denominated in the domestic currency. A depreciation or devaluation of the domestic currency increases the foreign debt burden in terms of the domestic currency, while the firm’s assets’ net value (net firm value) decreases. A firm’s balance sheets thus deteriorate due to a depreciation or devaluation of the domestic currency.
The decrease in the firm’s net value increase adverse selection and moral hazard due to a decline in the quality of the collateral. Collateral can solve the asymmetric problem by reducing the lender’s losses in case of default (Mishkin, 1996: 19). Collateral reduces the lender’s losses, because it can be sold to make up for the incurred losses due to default. In other words, high-quality collateral reduces uncertainty among lenders about the repayment of their loans. According to Mishkin (1996: 19), a firm’s net value can perform a similar role as collateral. Therefore, a firm with a high net value is less likely to default, i.e. to go bankrupt, than a firm with a low net value due to the higher value of assets to prevent a default. A higher net value means that there is more at stake for borrowers in the case of default, decreasing the incentives for moral hazard. As a result, lenders are more willing to make a loan, because problems of adverse selection and moral hazard are less likely to occur (Mishkin, 1996: 20).
The ratio M2 to (foreign exchange) reserves captures the amount of bank liabilities that are backed by (foreign exchange) reserves. In other words, it captures the ability of the central bank to meet the demands of investors that withdraw their bank deposits or convert their currency deposits into other currencies in the event of a financial crisis. A higher ratio M2 to (foreign exchange) reserves decreases the ability of the central bank to meet the demands of investors when they withdraw their deposits. As a result, investors may be uncertain about the country’s solvability, leading to a withdrawal of deposits. The previous subparagraph external sector/current account described that uncertainty about the solvability may lead to liquidity problems and bank runs, thereby increasing the probability of a financial crisis. Financial sector
Leading indicators related to the financial sector often include money growth (M1 and M2 growth) and domestic credit growth to explain/predict financial crises. An increase in money growth may be the result of a rise of the amount of bank liabilities. Large increases in money growth may result in inflationary pressures. When the growth of money is the result of a rise of the amount of banking liabilities, there may be an increase in uncertainty. For example, money growth resulting in a higher ratio M2 to (foreign exchange) reserves may make investors uncertain about the solvability of the country. Hence, there is a greater probability of a financial crisis when money growth results in increased uncertainty due to a higher ratio M2 to (foreign exchange) reserves.
However, money growth resulting in higher inflation rates means a decrease in the value of money. Domestic firms and investors can invest less with the same amount of money, which may lead to a lower level of investment and aggregate economic activity. Furthermore, a high inflation rate is often associated with a high nominal interest rate. High interest rates increase the probability of bad credit risks due to increases of adverse selection. Increased probability of bad credit risks may lead to lower levels of investment and aggregate economic activity, increasing the probability of a financial crisis (see for more details the external sector/current account subparagraph).
low given the weak legal structure. Furthermore, the lender’s ability to screen out good from bad credit risks make it more attractive for borrowers to offer high-risk projects.
Large amounts of domestic credit due to a weak legal structure thus mean a large amount of low-quality credit due to adverse selection and an increased probability of default due to moral hazard. The probability of default increases, because large amounts of low-quality credit increase the probability of non-performing loans (NPL’s). Therefore, an increase in domestic credit because of a weak legal structure may increase the probability of a financial crisis.
However, there is another explanation for the increased probability of a financial crisis due to an increase in domestic credit. An increase in domestic credit may make a country more vulnerable to a reversal in capital inflows. Investors may be more uncertain about the solvability of a country when large increases in domestic credit occur. Uncertainty may lead to bank runs, which in turn may lead to liquidity problems increasing the probability of a financial crisis.
Domestic real/public sector
Variables often used in the domestic real/public sector are changes in stock market prices and GDP growth. Changes in stock market prices directly influence a firm’s capacity to invest (Garretsen et. al., 1999: 42). A fall in stock market prices decreases the total firm value, which decreases the investment capacity. Moreover, lower stock market prices, i.e. a lower total firm value, increases the risk of adverse selection and moral hazard. Lenders may be less willing to lend, because the incentive to involve in more risky investments (moral hazard) increases for borrowing firms due to lower costs in case of default (Mishkin, 1996: 20). Therefore, lower stock market prices may result in a decline in investments and aggregate economic activity, increasing the probability of a financial crisis.
Global economy sector
GDP growth of OECD countries2 and growth of world oil prices are examples of indicators used in the global economy sector. Increases in GDP are often associated with increases in imports. GDP growth of OECD countries may thus increase the imports of these countries, which may result in increases in the exports of a country. On the other hand, a decline in GDP of OECD countries may result in decreases in the exports of a country. Decreases in exports (ceteris paribus) may lead to a negative current account balance, which increases the level of uncertainty. Increased uncertainty makes information more asymmetric, increasing problems of adverse selection. Lenders will have more difficulties to discriminate between good and bad credits, making them less willing to make loans. The resulting decrease in lending, investment and economic activity may lead to a financial crisis. Therefore, a negative growth in GDP of OECD countries increases the probability of a financial crisis.
A rise in world oil prices may worsen the current account balance of a country when it is dependent on the import of oil to meet its fuel need. As mentioned above, a negative current account balance may increase the level of uncertainty. A rise in the world oil price may thus increase the level of uncertainty, which in turn increases the probability of a financial crisis.
Methods
The most common approaches used in studies on financial crises are the signals approach, the logit approach and the probit approach. The aim of all three approaches is to identify indicators as main determinants of financial crises to explain and/or predict them. However, the used method to identify these indicators differs between the approaches.
Signals approach
The signals approach compares the behaviour of indicators in periods preceding a crisis with their behaviour during tranquil times. The behaviour of an indicator during tranquil times is seen as its normal behaviour. When one of these variables deviates from its normal level beyond a certain threshold, this is taken as a warning signal about a possible crisis within a specified period of time (Kaminsky et. al., 1998: 15).
2 The OECD, which is the Organisation for Economic Cooperation and Development, has 30 member
An index is constructed to examine if the behaviour of a variable exceeds a certain threshold. For example, Kaminsky et. al. (1998: 16) identified currency crises by the behaviour of the index of exchange market pressure and used a threshold of three standard deviations. Periods in which the index is above this threshold are identified as crises.
According to Kaminsky et. al. (1998: 17), threshold levels are chosen to balance between the risk of having many false signals (which would happen, if a signal were issued at the slightest possibility of a crisis) and the risk of missing many crises (which would happen if the signal were issued only when the evidence is overwhelming).
Furthermore, a so-called signalling horizon is chosen, which the period is within is expected that indicators have the ability to anticipate financial crises. The performance of each indicator can be explained using the following matrix3.
Crisis No crisis
Signal was issued A B
No signal was issued C D
When an indicator issues a signal, it is either a good or a bad signal. A is the number of months in which a good signal is issued and B the number of months in which a bad signal (noise) is issued. Moreover, C is the number of months in which no signal is issued, which would have been a good signal and D is the number of months in which no signal is issued, which would have been a bad signal.
The indicators that can best predict crises are indicators with a low signal-to-noise ratio, which is the ratio of false signals to good signals. The noise-to-signal ratio is calculated as follows: [(B/B+D)]/[(A/A+C)].
Logit and probit approach
The method used in the logit and probit approaches is basically the same as the method used in the signals approach. The approaches also compare the behaviour of indicators in the periods preceding a crisis with their behaviour during tranquil times. However, the logit and probit approaches do not chose a threshold and an index for the identification of crises.
The logit approach defines crises as a dependent variable that can have two values: 1 or 0. A dependent variable takes value 1 when the method predicts a crisis and takes value 0 in any other case, which means that the method does not predict a crisis.
The probit approach, on the other hand, defines crises as a dependent variable that can have any value other than 0 or value 0. A value other than zero means that the method predicts a crisis. When the dependent variable takes value 0, no crisis is predicted.
In general, the probability that a crisis occurs at a particular time is hypothesised to be a function of a vector of explanatory variables X (i, t). Letting P (i, t) denote the crisis dummy variable; β denote a vector of unknown coefficients and ε denote random errors.
P (i, t) = β * X (i, t) + ε
Hence, the greater the amount of explanatory variables signalling a financial crisis is, the higher the probability that a financial crisis will actually occur.
II. LITERATURE REVIEW
The authors of the studies included in this literature review share the opinion that a number of indicators can be identified as main determinants to explain/predict financial crises. However, there is no general agreement among the authors. Different authors find different indicators for different types of financial crises. Moreover, they find different indicators for the same type of financial crisis. Table B1 presents an overview of the theory behind the indicators found as main determinants for financial crises in the included studies.
Insert table B1 about here
A review of 21 studies is by far not complete. However, the limited time and scope given for this study made it not realistic to aim at a review of all the literature on financial crises. The studies that will be reviewed focus on either one of the three types of financial crises, a combination or on all three types. The starting point for the literature review is the study of Lestano et. al. (2003). They give an overview of indicators found significant in eight important papers on currency, banking, and debt crises as causes for these financial crises. Their overview is extended with other studies on financial crises that seem to be relevant for this study.
The studies included in the literature review are certainly not a random choice. Studies were chosen with the use of three, not mutually exclusive, selection criteria. The first selection criterion is that the paper has been written during or after 1999. The boundary was set for 1999, giving the author the possibility to include only (relatively) recent written papers. The second selection criterion is that the study examines leading indicators for financial crises, because the aim of this study is to find leading indicators for China to predict the likelihood of a financial crisis. The third and last selection criterion is that Asian countries are included in the study, because China belongs to the Asian continent. Therefore, findings from studies that examine Asian financial crises may be relevant for the prediction of a financial crisis in China. The idea is that studies on Asian financial crises found indicators that are especially relevant in the Asian context.
Using these selection criteria resulted in zero studies found only on debt crises. Therefore, debt crises are only included in studies that review all three types of financial crises.
The 13 studies added to the eight studies of Lestano et. al. (2003) will be reviewed in more detail. See appendix tables B3 and B4 for a complete overview of the findings and the indicators found in all the included studies. Table B3 presents the findings and explanatory variables, while table B4 gives an overview of significant and insignificant variables for each study.
Studies on all three types of financial crises
They found indicators that can predict all three types of financial crises. The indicators that correlate with all three types of financial crises are growth of money (M1 and M2), a fall in bank deposits, negative growth in the GDP per capita, and falling national savings. Deterioration of the ratio M2 to foreign reserves (=a higher ratio M2 to foreign reserves), increases in the domestic real interest rate and the inflation rate play an additional role in banking crises and some varieties of currency crises (Lestano et. al., 2003: 3).
These results confirm for a great part the results of an earlier study on the Asian currency and financial crisis by Corsetti et. al. (1999). They examined the causes of the Asian currency and financial crisis, using annual data over the period 1990-1997. The following countries were reviewed: Malaysia, Indonesia, Philippines, Singapore, South Korea, Thailand, Hong Kong, China and Taiwan.
In line with Lestano et. al. (2003), Corsetti et. al. (1999) found that a higher ratio M2 to foreign reserves, high domestic interest rates and high inflation rates caused the financial crises. Furthermore, Corsetti et. al. (1999) found a fall in stock market prices and a high ratio short-term debt to foreign exchange reserves as other causes for the financial crises. However, in contrast with Lestano et. al. (2003) findings, national saving, i.e. falling national saving was not found as a cause.
Sussangkarn and Tinakorn (2002), who examined the financial crises experienced by Indonesia, Philippines, Korea (South), and Thailand also found a fall in stock market prices and a high ratio short-term debt as causes for the financial crises. Moreover, according to Sussangkarn and Tinakorn (2002: 11), a real exchange rate misalignment4, declines in exports, and deterioration of both the terms of trade and the fiscal balance to GDP caused the financial crises.
Bustelo (2000), who compared the ERM crisis of 1992-93, the Mexican crisis of 1994-95 and the East Asian crisis of 1997-98 with each other, also found that large short-term foreign debt as a proportion of foreign reserves played a role in the Asian financial crises (Indonesia, Malaysia, Philippines, South Korea, and Thailand). However, Bustelo (2000: 28) argues that
4 Exchange rate misalignment occurs when a currency has a fixed exchange rate at a level that too high and
the financial crises were not caused by high inflation rates, but by a lending boom5 and a high ratio short-term debt to foreign exchange reserves.
Studies on currency crises
In their study on leading indicators of currency crises for 15 emerging countries, Burkart and Coudert (2002) found that there are some fundamental determinants (leading indicators) of crises. Using Fisher’s linear discriminant analysis and quarterly data over the period 1980-1998, the following causes were found for currency crises: overvaluation of the domestic currency (in general, speculative attacks starts with overvaluation of the domestic currency), large short-term debt, high inflation and losses in foreign exchange reserves. Furthermore, they found differences between the regions considered: Latin America and Asia. Deviation of the real exchange rate from its long-term value, a rise in real domestic credit and a highly open economy (high ratio import plus export to GDP) are indicators that were especially found for Asia.
Tambunan (2002) developed an EWS for Indonesia with the signals approach using data from January 1990 to December 2001 and found relatively similar indicators. He found that a large depreciation of the rupee, import growth, and a rise in domestic interest rates caused the currency crisis in Indonesia.
Tinakorn (2002), using both the signals and the probit model, found leading indicators for Thailand using monthly data for the period 1992-2000. A wide range of leading indicators for the currency crisis in Thailand were found: a misalignment of the real exchange rate, high inflation, a deterioration of the terms of trade, a high ratio short-term to foreign exchange reserves and a sharp rise of domestic credit as a proportion of GDP.
Krznar (2004) examined the currency crises in Croatia, using both the signals approach and the probit approach for the sample period: January 1996-March 2003. He found that a high inflation rate and an appreciation of the exchange rate marked both the currency crisis of 1998 and the currency crisis of 2001. Indicators that marked either the first or the second currency crisis are a sharp rise in domestic credit, a fall in freely available bank reserves, and a high level of the multiplier M2.
In contrast to Tambunan (2002) who found a depreciation of the rupee as a cause for the financial crisis, Krznar (2004) found an appreciation of the exchange rate as explanation for the currency crises in Croatia. A possible reason for this difference is that Indonesia had an overvalued domestic currency, while Croatia had an undervalued domestic currency.
Burkart and Coudert (2002) findings are largely confirmed by the above studies. However, a study done by Kibritcioglu et. al. (1999) does not confirm their findings. Using the leading economic indicators approach, Kibritcioglu et. al. (1999) review four financial crises that occurred in the 1990s: the European ERM-crises of 1992-93, the Mexican Peso crisis of 1994-95, the Turkish financial crisis in 1994, and the Asian crises in 1997-98. Furthermore, they used monthly data for the period of 1986-1998.
The indicators that explain the financial crisis of Turkey in 1994 are deterioration in the terms of trade, a misalignment of the exchange rate, negative export expectations, and negative export growth (Kibritcioglu et. al., 1999: 22).
Studies on banking crises
Demirgüç-Kunt and Detragiache (2005) examined 94 countries that have experienced a banking crisis, during the period 1980-2002. They used the multivariate logit approach to update their analysis of 1998. The indicators that were found significant are: low GDP growth, high real interest rates, high inflation, large broad money (M2) as a proportion of foreign reserves, large credit to the private sector, high lagged credit growth (may capture a credit boom), and low GDP per capita. Their findings are largely consistent with those of the earlier paper, indicating fairly robust relationships (Demirgüç-Kunt & Detragiache, 2005: 72). Demirgüç-Kunt and Detragiache (2005) searched for leading indicators for banking crises in general, while Adalet (2005) examined the effect of financial structures on crises for 13 countries from interwar Europe. Adalet (2005) runs logit regressions and found that universal banking, a high number of bankruptcies, and low foreign reserves are significant correlated with banking crises using monthly data for the period 1920-36. An indicator found with annual data for the period 1924-38 is a low level of branching.
economic shocks that influence a particular industry in a negative way. For example, when a recession hit an industry hard, universal banks lose money due to declines in the stock market prices and loan defaults. The capital-to-asset ratio declines as a result, which may increase the level of uncertainty due to lower liquidity.
A high number of bankruptcies increase the probability of a banking crisis, because it indicates the inability of investors to repay their loans. As a result, bank balance sheets worsen, making the banks unable to extend further loans. The resulting decline in lending activity leads to a lower level of investment and aggregate economic activity. On the other hand, the probability of bank failure decreases when it geographically diversifies, because geographical diversification makes banks less vulnerable to geographic shocks. A high level of branching means higher geographic diversification (Adalet, 2005: 12). Therefore, a high level of branching decreases the probability of contagion. In this case, contagion occurs when a bank failure in one region infects banks in another region. In other words, a bank failure in one region leads to a bank failure in another region due to contagion. Lower probability of contagion decreases the probability of financial crises.
Study on currency and banking crises
Kaminsky (2000) examined 76 currency crises and 26 banking crises in 20 countries. The indicators found significant for predicting currency crises are a misalignment of the real exchange rate, large gross foreign debt, large short-term foreign debt as a proportion of total foreign debt, hikes in world interest rates and large inverse capital flows (capital flight). The indicators misalignment of the real exchange rate and large gross foreign debt were also found in the studies on banking crises that have been reviewed.
Study on China
From his last chapter one can conclude that it is not very likely that China will experience a financial crisis in the (near) future. He (2002) draws his conclusions on data and on characteristics of the Chinese financial system at the time available. However, a lot changed since 2002, for example, the exchange rate regime. These changes may result in other conclusions about the possibility of a financial crisis in China. Furthermore, a more structured study may also result in different or even additional conclusions.
This literature review shows that leading indicators are very useful for the explanation and/or prediction of financial crises. Although the authors disagree on which leading indicators have the best explanatory/predictive power, there seem to be a few leading indicators that are important for a variety of regions. These leading indicators are a real exchange rate misalignment, a fall in foreign exchange reserves, high inflation rates, high real interest rates, deterioration in the terms of trade, an increase in the ratio M2 to (foreign exchange) reserves and domestic credit growth.
III. CHINA’S FINANCIAL SYSTEM
The chapter on financial crises theory outlined the importance of a financial system for an economy. In general, financial systems are characterised by imperfect markets due to the presence of asymmetric information. The basic function of financial systems is to facilitate the allocation of capital by providing investors the opportunity to share risk and give them access to both capital and information. Financial systems thus reduce adverse selection and moral hazard by making information less asymmetric. Financial systems that are able to minimise the effects of adverse selection and moral hazard are efficient financial markets. Efficient financial markets promote economic growth. On the other hand, low efficiency of financial systems may be a hindrance to economic growth (Allen & Oura, 2004:1). In situations where a financial system is unable to reduce asymmetric information, financial crises may occur.
Furthermore, the issuance of more securities of low-quality firms increases borrowers’ incentives to commit moral hazard. The incentives to commit moral hazard increases, because a lower net value of firms means that borrowers have less to lose when they default on their loans. Hence, both problems of adverse selection and moral hazard may occur due to excessive risk taking to obtain the high returns needed for economic growth. Risk taking increases the probability of a financial crisis when it leads to adverse selection and moral hazard.
A financial system and thus in particular the efficiency of a financial system may thus be helpful in explaining/predicting a financial crisis. Therefore, it may relevant for the prediction of the likelihood of a financial crisis in China to examine its financial system. The examination of China’s financial system largely follows the recent examination by Allen et. al. (2005). They included four sectors in their examination of China’s financial system. The examination in this study includes three of these four6 sectors, which are the banking sector, the capital
market sector, and the insurance market sector. Furthermore, the inflation rate, not mentioned in the study of Allen et. al. (2005) that may be relevant for the prediction of the likelihood of a financial crisis in China is included in the examination as well.
Insert figure A2 about here
Banking sector
According to Hansakul (2004: 3), China’s banking sector still exhibits the legacy of a centrally planned economy in which the government, both at central and local levels, continues to play an instrumental role in credit allocation and pricing of capital. The domestic banking system is still the primary channel of financial intermediation between savings and investment, which implies that the capital market in China is less important. The importance of China’s banking system over its capital market7 may indicate a bank-based financial system. Allen et. al. found the dominance of the banking system in China’s financial system in their 2002 paper and again in their 2005 paper. They compared the relative importance of capital markets vs. financial intermediaries or banks, using the relative size measures: structure activity (log of
6 Allen et. al. (2005) describe the foreign sector according to international financial capital flows, for example
FDI flows. The subparagraph foreign exchange reserves will deal with the international financial capital flows. Therefore, China’s foreign sector is not included in the examination, the way Allen et. al. (2005) do in their study.
7 China’s capital market includes stock markets, bond markets and equity capital market (including venture
the ratio of value traded to total bank credit) and structure size (log of the ratio of market capitalisation to total bank credit). China has the lowest scores for both categories compared to the LLSV8 sample countries, indicating that China’s capital markets are smaller than its
banking system. The bank dominance over capital markets is stronger than the average of all LLSV sample countries (Allen et. al., 2002: 9, and 2005: 11). Moreover, it is even stronger compared to the most bank-based country in the LLSV sample.
Furthermore, many researchers share the opinion that China may have the weakest banking sector of the world. Allen et. al. (2002, 2005) examined the efficiency of the banking sector. They found the lowest measure of finance efficiency (log of (total value traded/GDP)/overhead cost) for China compared to all the LLSV countries and concluded that China has a financial system, dominated by a large but inefficient banking sector (2002: 9, 2005: 11). This inefficient banking sector consists of four types of banks: wholly state-owned banks, commercial banks, credit co-operatives, and foreign banks.
The “big four” state-owned banks dominate the banking sector: the agricultural bank of China (ABC), the people’s bank of China (PBOC)9, the China construction bank (CCB) and the industrial and commercial bank of China (ICBC). The “big four” account for around 60 per cent of the banking sector’s total assets and a similar share of loans (Hansakul, 2004: 1). The commercial banks, credit-cooperatives and the non-bank institutions account for a great part for the remaining 40 per cent of the banking sector’s total assets and loans share.
Although the presence of foreign banks10 has increased since China’s WTO accession, foreign banks still do not play a significant role in China’s current banking sector. Despite the increase of the asset share of foreign banks in absolute numbers, the asset share in percentages is still very small. According to He and Fan (2004: 10), the asset share of foreign banks was 1.36 per cent at the end of 2003. The insignificant role played by foreign banks is due to the restrictions on entry of foreign banks that still exist. For example, it is not allowed that the equity investment proportion of a single foreign financial institution in a Chinese financial institution exceeds 20 per cent. Moreover, the activities must be granted the approval of the China Banking Regulatory Commission (CBRC) and fall under its regulation and supervision (He & Fan, 2004: 6).
8 LLSV stands for Rafael La Porta, Florencio Lopez-de-Silanes, Andrei Shleifer and Robert W. Vishny, who
conducted a study comparing the financial systems of 49 countries.
9 The people’s bank of China is the central bank of China.
10 New openings of foreign bank branches since 2002 are mainly from foreign banks based in regions such
Bank loans
China’s total bank lending is growing, currently equal to around 140 percent of GDP, which is more than twice the levels observed in most industrial countries. A large fraction of the total bank lending is short-term. In 2002, short-term loans accounted for 56.6 per cent of total bank lending (see table C1). Lending in China is thus excessive, which is partly due to official lending policies and the lack of accountability and regulation in the financial system.
China lacks good corporate governance and accounting standards. Moreover, China has weak shareholder and creditor protection (Allen et. al., 2002: 5 & 2005: 55). They argue that the most glaring problem in China’s accounting system is the lack of independent, professional auditors, similar to the situation of legal professionals. There is an estimate of 150,000 lawyers in China and only 4 per cent of the business enterprises have regular legal advisers (Tadesse, 2005: 7).
Furthermore, there is a complete lack of transparency and an effective judicial system to protect companies’ assets from embezzlement and other frauds. Allen et. al. (2002: 30) compared China to the countries in the LLSV and Levine11 sample on credit and shareholder rights, law enforcement, and legal and accounting professionals. They find a significantly underdeveloped legal system for China compared to the LLSV and Levine sample countries. Although formal corporate governance mechanisms are poor, there exist effective, informal mechanisms to support the financing and growth of the economy. Cultural beliefs, like relationship and reputation are fundamentals on which informal mechanisms are based (Allen et. al., 2002: 31, 2005: 57).
A significant factor in credit expansion is the deficiency in the knowledge and training of credit risk. Moreover, the low carefulness of state-owned banks’ loan officers is also a significant factor in credit expansion. The credit expansion may thus be the result of an increase in bad credit due to this lack of knowledge, training and low carefulness. Therefore, a slowdown in both the economy and personal income may lead to high default rates and large amounts of new NPL’s (Allen et. al., 2005: 19).
Non-performing loans
The total assets of the “big four” are high, but the quality of the assets is low. State-owned banks were in charge of public finance and supporting state-owned enterprises (SOEs) through the provision of funds without considering operating performance. As a result, the SOEs accumulated a huge amount of NPL’s and bad debt. In 1997, the government set up asset-management companies (AMC’s) to absorb the most problematic assets of the “big four”.
Despite these asset-management companies, the “big four” still accumulate new bad debt, i.e. new NPL’s. Table C1 shows that loans made to the Hybrid12 sector is much less than the loans made to manufacturing industries with many SOEs (Allen et. al., 2005: 16). The accumulation of new bad debt is thus the result of large share of total loans still going to state sectors. Another source for new NPL’s is the lack of a national consumer-credit database to spot overstretching debtors. Therefore, overstretched debtors can get new loans without being refused, thereby increasing the probability that they will default on their loans.
Insert table C1 about here
The exact amount of NPL’s is hard to estimate, because data on the amount of NPL’s is scarce. Furthermore, the available data may even be deficient. Despite these data problems, many authors share the opinion that the amount of NPL’s is large in China. For example, Allen, et. al. (2005: 82) estimate for 2002 an amount of NPL’s that accounts for 12.6 per cent of total loans and even for 15.2 per cent of GDP. The estimated amount of NPL’s of the “big four” is even larger: 26.1 as a percentage of total loans in 2002 (Hansakul, 2004: 10). When comparing China’s estimated amount of NPL’s with estimates of NPL’s for other Asian countries, China’s estimated amount is the largest (see table C2).
Insert table C2 about here
Allen, et. al. (2005) base their estimate on government and official records. However, the accuracy and reliability of government and official records of a (former) central planned
economy is questionable. According to Roubini and Setser (2005: 13), in 2002, the estimated amount of NPL’s accounts for 37.0 per cent of total loans and even for 60.9 per cent of GDP. The estimates of Roubini and Setser (2005: 13), based on records of the four AMC’s and the “big four”, are significantly larger than the estimates of Allen et. al. (2005: 82). The above example shows that the actual amount of NPL’s may be higher than the estimates based on government and official records.
Foreign exchange reserves
China is accumulating foreign exchange reserves on a rapid and huge scale. China’s current account balance surplus and its large scale of foreign direct investment (FDI) inflows are the driving forces behind the growing foreign exchange reserves (see table C3 and C5). The current account balance surplus is largely due to the large trade balance surplus with the United States. Table C4 shows a growing trade balance surplus with the United States, which grew from 49,695.5 (millions of U.S. dollars) in 1997 to 161,938.0 (millions of U.S. dollars) in 2004.
China’s large scale of FDI inflows is a result of the globalisation of its financial sector. The globalisation of China’s financial sector attracted foreign capital, especially FDI. According to Allen et. al. (2005: 13), China’s inflows dominate its outflows, which only account for about 10 per cent of total FDI.
Insert table C3 to C5 about here
Interest rates
According to Roubini and Setser (2005: 16), credit is cheap due to low real interest rates and they argue that cheap credit will fuel an investment boom13. An investment boom may lead to an investment bubble. A reason for the low real interest rates is the way interest rates are determined. The Chinese government determines the interest rates without considering the cost of bank capital. China thus lacks a market-determined structure of interest rates. As a result, investors have difficulties measuring credit risks.
Although the Chinese government relaxed the regulation on interest rates, financial authorities remain to determine the deposit and lending rates (He & Fan, 2004: 13). For example, bank loans going to SOEs are low interest rate loans, resulting in a trivial deposit-lending rate difference (Yang, 2004: 10).
Insert table C6 about here
Capital markets
As already mentioned in the examination of the banking sector, China’s capital markets are less important in its financial system. China’s capital markets are less important, for example due to weak investor protection and poor and ineffective regulation (see bank loans for more details). China’s capital markets consist of stock markets, bond markets, and private equity markets14, including a venture capital sector15.
Stock markets
The primary reason for creating stock markets in China was to allow SOEs to raise capital from Chinese households and from foreign entities as a substitute for continued central government funding of such capital investment (Abacus, 2005). From the beginning, state policies strongly influence the stock markets. The government protects Chinese companies from foreign take-overs through the division and fragmentation of the stock markets (Hansakul, 2004: 6).
13 An investment boom describes a period in which investments rapidly increase (expand).
14 Private equity deals with equity capital not quoted on a stock market and made available to companies or
investors (Abacus, 2005).
15 Venture capital is a general term to describe financing for start-up and early stage businesses. Venture
An example of the division and fragmentation is the types of shares traded on the stock markets: A-shares, B-shares, and H-shares. The A-shares are quoted in renminbi and only Chinese citizens and selected foreign institutional investors are allowed to trade in A-shares. The B-shares are quoted in foreign currencies. B-shares are reserved for foreign investors and mainlanders that have access to foreign currencies. The so-called H-shares are shares traded on the Hong Kong stock market and other foreign stock markets.
The stock markets are dominated by SOEs (under government control) due to the current listing process. The current process of listing companies strongly favours (former) SOEs with connections to government officials (Allen et. al., 2005: 26). Moreover, in many cases the government owns a great part (up to two third) of the shares of the listed enterprises, thus strongly influencing the stock markets.
Due to the predominant policy-driven nature of the stock markets, share prices do not always truly reflect the corporate fundamentals. For example, the government consistently pushed stock market prices up with price-to-earnings (P/E) ratios rising to above 60 at the beginning of 2001. This P/E ratio fell to around 35 per cent in 2002 through an attempt to sell state-shares (Drysdale & Huang, 2003: 15). A P/E ratio of 35 per cent is high when compared to the P/E ratio of 10 per cent for the H-share companies listed in Hong Kong. The P/E ratio of 10 per cent for the H-share companies listed in Hong Kong (a regional and international integrated stock market) may be considered as normal, because it represents corporate fundamentals (Drysdale & Huang, 2003: 14). Hence, Chinese stock market prices may be overvalued, which inhibit the risk of future declines in stock market prices.
Bond markets
China’s bond markets consist of three types of bonds: government bonds, policy-financial bonds16 and corporate (private) bonds. In general, China’s bond markets are underdeveloped compared to foreign bond markets (see figure C1). The underdevelopment of China’s bond markets is due to the inability of investors to estimate the probability of default and recovery rate in the case of default. This inability is the result of a lack of sound accounting/auditing and high-quality bond-rating agencies (Allen et. al., 2005: 19). Moreover, China lacks a market-determined structure of interest rates that accurately reflects the opportunity cost of funds at each maturity. Allen et. al. (2005: 30) argue that it is hard for
16 Policy-financial bonds are bonds issued by policy banks, which belong to the Treasury Department, and
firms and investors to manage risk due to the lack of such a market-determined structure of interest rates.
Insert figure C1 about here
Venture capital markets/Private equity sector
In general, the private equity sector and venture capital markets provide financing for the funding of new industries. China’s venture capital markets are underdeveloped. Table C7 shows that China has a low number of private equities funds and a low total capitalisation of the funds compared to other countries and regions. According to Allen et. al. (2005: 39), this indicates that supporting the growth of young firms is very limited in China.
China’s venture capital has increased in recent years, but its venture capital industry remains insignificant. In 2003, the government accounted for 48 per cent of total venture capital investment (Sood, 2004). The government thus has a great influence in the venture capital industry. However, the influence of the government will probably diminish due to increases in the inflow of venture capital investment from foreign funds. China has become one of the most attractive venues among emerging economies for private equity17 and venture capital.
Insert table C7 about here
Insurance market sector
According to Allen et. al. (2005: 23), China’s insurance market sector is underdeveloped in terms of importance in the financial system. They compare China’s insurance markets with the insurance markets of South Korea, Taiwan and Singapore. China’s insurance market is significantly smaller than the insurance markets of the other economies in terms of the ratio of total assets managed by insurance companies over GDP and premium income over GDP. Evidence for the underdevelopment of China’s insurance markets is, for example, the large domestic savings. According to Rajan (2005), the high saving rate is partly due to high households’ savings. The lack of a social security network forces many people to save for
17 Private equity investment includes venture capital, leveraged buyouts, mezzanine investment (late-stage
their own pension. Furthermore, credit given to households is limited, forcing households to save for the purchasing of durables that increased in their availability. Therefore, the dominant form of China’s savings is bank deposits, which is also due to the high risks associated with anything invested outside the banks18 (Morrison, 2005: 9).
Domestic savings and investments
China have a small saving surplus for years (see table C8), meaning that gross domestic saving19 is in excess of gross domestic investment (both as a percentage of GDP). Furthermore, China has a high saving rate as well as a high investment rate and both grow on an annual basis. For example, gross domestic saving grew from 38.0 per cent of GDP in 1997 to 49.1 per cent of GDP, in 2004. On the other hand, gross domestic investment grew from 33.6 per cent of GDP in 1997 to 45.7 per cent of GDP in 2004.
Insert table C8 about here
Inflation rate
China experienced a small internal deflation in the first year after its WTO accession. However, since 2003, China is experiencing modest, but rising inflation rates. The inflation rate rose from 1.2 percent in 2003 to 4.0 percent in 2004 on an annual basis (see table C9). When using quarterly data, the increase is slightly bigger: from 0.5 percent in the first quarter of 2003 to 3.6 percent in the fourth quarter of 2004. Table C10 shows that the inflation rate in 2004 is higher than foreign inflation rates, for example, the inflation in 2004 was 2.1 percent in the European Union and in the United States 2.7 percent (annual rates).
Insert table C9 and C10 about here
IV. CHINA’S ECONOMIC POLICY
A number of theoretical papers have mentioned the importance of other non-economic factors that may influence the occurrence of financial crises. Several authors mention
18 Investment in the stock and bond markets is associated as high-risk investment due to the weak legal
structure and investor protection.
19 According to standard national accounting, gross domestic saving is calculated as the difference between
political factors as explanations and/or as causes of financial crises. For example, many authors suggest that financial liberalisation may have caused the Asian financial crises. Financial liberalisation may result in financial instability, especially when it occurs before financial regulation and supervision are sufficiently effective to prevent moral hazard among banks. Ineffective financial regulation and supervision leads to a weak legal structure, in which screening and monitoring is difficult. Without effective financial regulation, it is hard for lenders to discriminate between good and bad credit, while ineffective supervision makes it hard to monitor borrowers’ behaviour. The resulting increased probability of bad credit and default make lenders less willing to make loans. Lending, investment, and aggregate economic activity will decline, increasing the probability of a financial crisis. Hence, financial liberalisation resulting in financial instability may lead to a financial crisis.
Moreover, financial liberalisation may be associated with falling savings. According to Jappelli and Pagano (1994), financial liberalisation may increase the access to consumer credit, resulting in falling savings. Falling savings may lead to higher consumption or even a consumption boom, which in turn may lead to an investment bubble. The occurrence of one or both the examples increases the possibility of a financial crisis. Financial liberalisation is a result of economic policy; therefore, political factors have a direct impact on the speed and dept of financial liberalisation and indirect on the occurrence of financial crises.
Furthermore, political factors influence the level of country risk, in this case, the level of country risk for China. The level of country risk is related to the (in) stability of a country’s political system. The more unstable a country’s political system is, the more vulnerable a country is to capital flight and negative changes in investor’s confidence. Capital flight and a loss of investor’s confidence may lead to a financial crisis.
Therefore, financial liberalisation and country risk, i.e. political risk may be important for the prediction of a possible financial crisis in China. However, the examination of China’s economic policy will only focus on financial liberalisation due to problems of data collection on the measurement of political risk.
Financial liberalisation
reforms of the government. The financial reforms are presented below and for each financial reform; the impact on the financial system is explained.
Exchange rate regime
China has no longer a fixed exchange rate regime. It changed it exchange rate regime to a strictly managed floated exchange rate regime, improving the convertibility of the Yuan. In July, there was a revaluation of the Yuan by 2.1% to 8.11 Yuan to 1 U.S. dollar from 8.28. Since its revaluation, the Yuan has appreciated by 0.33 percent with a trading band for the Yuan of 0.3 percent for a basket20 of currencies. In the future, market supply and demand determine the renminbi exchange rate and there will not be another one-off adjustment, according to Xiang (Deputy Governor of the PBOC). However, the outside world expects a further revaluation and/or appreciation in the future. When a revaluation and/or appreciation actually happens China’s current account balance may deteriorate, because exports become more expensive and imports become cheaper.
Banking sector
China prepares to open up its banking sector by 2007 in accordance with its WTO accession agreement concessions. Foreign banks are being led in only one-step at a time (The Economist, 2003). The increased openness of the banking sector increased competition from foreign financial institutions and among domestic banks. The competition will further increase in the future, especially from 2007 on when there are no more restrictions at all.
According to Drysdale and Huang (2003: 11), the increased competition revived the anxieties about the sustainability of the Chinese banking system. The government developed a three-step reform program for the banking sector. The first three-step involves continuation of reforming the internal management systems of the banks, including implementation of the international standard five-category loan classification system and improvement of the credit culture. State-owned commercial banks and joint shareholding commercial banks adopted a five-category loan classification system in 2004. The five-five-category loan classification system classifies bank loans according to their inherent risks as pass, special-mention, substandard, doubtful and loss (People’s Daily, 2003). See table D1 for a detailed explanation of these five categories.
20 The basket of currencies consist of the following currencies: the US dollar, the euro, the Japanese yen,
Insert table D1 about here
Moreover, the PBOC removed the ceilings on lending rates at the end of 2004 (People’s Daily, 2004). The adoption of the five-category loan classification system and the liberalisation of interest rates improved the banks’ ability to measure credit risks, improving the transparency of bank credit in general. However, the adoption of the five-category loan classification system may result in an actual amount of NPL’s that is higher than the currently estimated amount.
The second step of the three-step reform program includes the transformation of the “big four” into share-holding entities, possibly introducing strategic and foreign investors. The third and last step involves that all banks should aim at public listing, domestic or overseas (Drysdale & Huang, 2003: 12-13).
Another planned reform in the banking sector is the implementation of a deposit insurance scheme. The implementation of a deposit insurance scheme makes it easier for banks besides the “big four” to take in deposits, because currently the government only offers insurance on the deposits of the “big four” (China Daily, 2004).
The aim of a deposit insurance scheme is to enhance stability in the banking sector. In general, a deposit insurer guarantees a financial institution, for example a bank to cover potential losses to give the bank the ability to pay their depositors and other lenders. In exchange of such a guarantee, the insurer demands openness of information of the financial institution. A deposit insurer can monitor the health of the financial institution and force the financial institution to operate legally and according the rules. A deposit insurance scheme thus enhances stability in the banking sector through regulation - and transparency improvement.
The aim of reforming the banking sector is to improve the economic performance and market viability of the domestic banks. According to Morrison (2005: 10), three out of the “big four” may be insolvent. Moreover, there is the threat of corruption, because loans are often political loans. The “big four” need to improve their economic performance and market viability before being capable face foreign competition, successfully. Therefore, the occurrence of bankruptcies among the “big four” may indicate failure of the Chinese government in reforming the banking sector.
The banking sector’s fate is thus determined by the government’s ability to successfully reform the banking sector. However, the dual role of the government as regulator and as majority owner of the state banks may hinder the reform process. Hence, the government’s dual role may diminish the effectiveness of the reform.
Capital markets
China’s capital markets are gradually liberalised in the same pace as the banking sector is liberalised. Foreign investors are encouraged to invest in China, especially in the stock markets. For example, in November 2002, the China Securities Regulatory Commission (CSRC) and the PBOC announced the new qualified foreign institutional investors (QFΙΙ) system. The QFΙΙ system allows foreign banks, insurance companies or fund management firms, with more than US$ 10 billion assets under management, to invest in bonds or stocks listed in the domestic markets.
According to Drysdale and Huang (2003: 8), the QFII system is a very important step toward liberalisation of portfolio investment. Liberalisation of portfolio investment may lead to massive inflows of “hot” money21.
Although financial liberalisation has started, the conditions of the domestic stock markets are a major constraint for further liberalisation. Stock markets are inefficient and need substantial consolidation, in terms of company structure and stock market prices. In addition, another major constraint on further liberalisation is China’s weak legal structure. For example, there is still widespread corruption, financial speculation and misallocation of investment funds. However, China is improving its legal structure. For example, an anti-monopoly law is under review by the National People’s Congress. An antitrust law like an anti-monopoly law will
improve the legal structure of China’s economy (China Economic Review, 2006). Moreover, China drafted a new bankruptcy law. The new bankruptcy law better protect the creditor’s interests by setting up simple and streamlined bankruptcy procedures, improvement of the court’s standing in bankruptcy case hearings, and the establishment of a trustee system and a special procedure for SOEs bankruptcies (Harvard Asia Quarterly, 2006).
These reforms improve the health of China’s capital markets, which decreases the probability of bad credit risks. Furthermore, these reforms may decrease the level of uncertainty. However, liberalisation of portfolio investment makes China more vulnerable to the risks generally associated with short-term debt22.
Insurance market sector
China’s WTO accession agreement concessions included a gradual opening up of the insurance market sector as well. For example, in 2005 all geographic restrictions will be eliminated. This means that foreign investors are free in their decision where to invest in China. Moreover, as from 2007 on, liberalisation of a full range of group, pension and annuity products takes place (Drysdale & Huang, 2003: 7).
Opening up usually intensifies competition among firms through the entrance of foreign competition. Intensified competition may speed up the reform process when domestic firms need further reform/develop to be able to face the foreign competition. Therefore, opening up the insurance market sector may speed up the development of China’s insurance market sector. However, it may also result in bankruptcies among insurance companies that are unable to face foreign competition. Liberalisation of the insurance market sector thus inhibits the risk of increasing the probability of a financial crisis.
Free trade
The government gradually abolish trade restrictions, encouraging free trade. The foreign trade policy gives provincial government much autonomy in foreign trade and allowing private enterprises to engage in foreign trade (Chow, 2005: 1). As a result, China’s foreign trade grew significantly and is still growing. In 2002, China’s foreign trade (imports and exports) was 620.8 billion U.S. dollars and accounted for 65 per cent of GDP, growing at a rate of 35 per cent per year. The growth of the trade volume continued and reached 1.1
22 Large amounts of short-term debt make a country more vulnerable to large reversals in capital inflows.
trillion U.S. dollars in 2004 with a growth rate of 30 per cent. China surpassed Japan and became the third largest trading country in the world, next to the United States and Germany (Chow, 2005: 1). China’s economy seems to integrate more and more in the global economy, being more and more open to trade (increased vulnerability to contagion).
V. ANALYSIS
The literature review resulted in many potential leading indicators. Among the potential leading indicators are explanatory variables found significant, insignificant or significant as well insignificant (see appendix table B4 for an overview of these potential leading indicators). Including indicators insignificantly found in other studies on financial crises seems illogical. Therefore, the idea to select leading indicators follows the selection idea used in the paper of Kaminsky et. al. (1998: 10). The idea is to select leading indicators found significant in the original papers as main determinants of financial crises. Furthermore, only leading indicators found significant in at least two original papers are selected. The resulting set of selected leading indicators only includes leading indicators found significant in studies from the literature review. However, this set of selected leading indicators may differ from the set of important factors found for China in the examination of both its financial system and economic policy. These two sets will be matched, which may result in a set important factors that are the same as the selected leading indicators, but also in a set important factors that differ from the selected leading indicators. The important factors that are different from the selected leading indicators are included in the analysis as well.
Nevertheless, the selected leading indicators/important factors from the literature review and both examinations (financial system and economic policy) may be irrelevant for the prediction of the likelihood of a financial crisis in China. Therefore, the selected leading indicators/important factors will be examined on their relevance for the prediction.
