The resurgence of banking institutions in post-crisis Korea: Risk and policy implications

Transcription

1 The resurgence of banking institutions in post-crisis Korea: Risk and policy implications Joon-Ho Hahm * Graduate School of International Studies Yonsei University June 2004 * Graduate School of International Studies, Yonsei University, 134 Shinchon-dong, Seodaemoon-ku, Seoul, , Korea. Tel: , Fax: , jhahm@yonsei.ac.kr The author gratefully acknowledges financial support from the Center for International Studies, Yonsei University.

2 The resurgence of banking institutions in post-crisis Korea: Risk and policy implications Abstract The Korean government pursued drastic financial reforms, and in accordance with these reforms, the financial structure also evolved dramatically. This article studies the nature of the financial transition by focusing on financial flows as an endogenous response to government restructuring policies. In the pre-crisis period, asymmetric and unbalanced financial liberalization policies resulted in the dominance of non-bank financial institutions and the deterioration in the maturity structure of corporate debts. After the crisis, the sequential restructuring approach to bank and non-bank financial institutions, coupled with delayed resolution of insolvent corporate firms, led to the concentration of financial savings in the banking sector. We identify various risk factors that arise from this unusual flow of funds and on-going bank consolidations, and draw policy implications to cope with them. Among others, strengthening financial supervisory roles and establishing market discipline are emphasized as effective ways to reduce systemic risk and to limit the negative consequences that could arise from the too-big-to-fail banking institutions that emerged after the crisis. Keywords Korea; bank; financial crisis; reform policy; consolidation.

3 Introduction While the foreign exchange liquidity problem and the severe terms of trade shock apparently caused Korea s economic crisis in 1997, the origins of the Korean crisis were deeply rooted in the nation s inefficient financial system. A prolonged period of interest rate controls and selective bank credit allocation, used as primary tools of development in the past, caused significant distortions in bank balance sheets. Extensive government regulations and influence over bank management also throttled and undermined the autonomy and accountability of banking institutions (Lee and Kim 2001). Notwithstanding the financial liberalization efforts of the 80s and 90s, the asset qualities of Korean banks remained poor, and the deterioration of bank balance sheets played a critical role in causing and propagating the currency crisis into a fullfledged financial crisis (Hahm and Mishkin 2000). 1 Among others, the legacy of the government s risk partnership with large conglomerates (chaebols) was a fundamental factor that brought about the deterioration of bank balance sheets (Cho and Kim 1997; Hahm 2003). In recognition of the financial sector s structural problems and in an effort to overcome an economic crisis of unprecedented severity, the Korean government pursued drastic financial sector reforms. As summarized in Table 1, under the government financial restructuring program, a total of 617 insolvent financial institutions (approximately 30% of all financial institutions that existed at the end of 1997) were either closed or merged by June In the process, the government 1

4 mobilized and spent trillion won or approximately 30% of Korea s 2002 nominal gross national income 2. As a result, these reforms substantially repaired the balance sheets of financial institutions. As Table 2 shows, even under a stringent forward-looking asset classification criterion 3, the amount of non-performing loans in the balance sheets of financial institutions fell sharply from 136 trillion won (21.8% of total assets) in June 1998 to 31.8 trillion won (3.9%) in December // Table 1 here // // Table 2 here // While the financial sector restructuring policies contributed much to the normalization of impaired financial intermediation and the rapid recovery of the Korean economy, structural transition of the financial system due to restructuring also brought about new challenges and risks. As we emphasize below, a prominent feature of the post-crisis financial transition in Korea is the rapid consolidation in the banking industry and the resurgence of commercial banks over non-bank financial institutions. The present paper focuses on this development and critically evaluates the nature and risks of the resurgence of banking institutions in post-crisis Korea. Throughout the paper we highlight the interaction between government policies and financial market responses. More specifically, this paper argues that, just as the unbalanced financial deregulation policies caused financial vulnerability in the precrisis period, unbalanced and unfinished financial restructuring brought about the asymmetry in financial flows and concentration of financial savings in the banking system in the post-crisis period. 2

5 The plan of the article is as follows: The first section describes the progress in bank restructuring and consolidation. Section 2 characterizes the key natures of the financial transition, based upon shifts in fund flows before and after the financial crisis. Section 2 also identifies the main factors behind the post-crisis financial transition. Section 3 presents and discusses emerging risk factors associated with the financial transition. Section 4 draws policy implications, and finally, section 5 offers brief concluding remarks. Progress in bank restructuring Since early 1998, the Korean government has attempted drastic financial structural reforms with a view to establishing an institutional foundation to clean up insolvent firms and financial institutions. As a first step, statutory infrastructure for financial restructuring was laid out. In December 1997, the Financial Industry Restructuring Act was amended to establish a relevant regulatory framework, and the prompt corrective action provision was formally instituted. In addition, according to the Act on the Establishment of Financial Supervisory Authorities newly enacted in December 1997, the Financial Supervisory Commission (FSC) was established in April 1998 as a supreme consolidated financial supervisory authority. Korea Asset Management Corporation (KAMCO) was also established in November 1997, and a non-performing loan resolution fund was created within KAMCO. In addition, the Korea Deposit Insurance Corporation (KDIC) announced a three-year temporary blanket guarantee on deposits of financial institutions in November

6 With the establishment of an institutional and statutory framework, the first comprehensive financial restructuring program was unveiled in April 1998, and bank restructuring was pursued according to the scheme. 5 The domestic banks financial condition turned out to be far worse than expected. Business transfers (purchase and assumption) were ordered to five insolvent banks in June 1998, which was the first case of bank closure in Korean history. All of the five closed banks posted negative BIS capital adequacy ratios, with non-performing loan ratios ranging from 21 to 49 per cent. Seven other banks self-rehabilitation plans were conditionally approved; however, five out of those seven banks were eventually merged in the recapitalization process. With the demise of the Daewoo group - the third largest chaebol in Korea in 1999, and as the strengthening of the asset classification criteria raised concerns over the potential insolvency of financial institutions, the government decided to raise additional public funds of 50 trillion won and announced the second round financial restructuring program in September Subsequently, four additional commercial banks that turned out to be insolvent were merged under the government owned financial holding company in March Unlike countries where bank consolidation occurred voluntarily as a natural way of reaping economies of scale and scope, bank consolidation in Korea responded to government restructuring initiatives. As shown in Table 3, the number of commercial banks declined by 42 per cent, from 26 in 1997 to 15 as of December 2002, and the number of branches and employees in the commercial 4

7 banking industry declined by 16 per cent and 41 per cent, respectively. Total bank assets, including bank trust accounts have increased by 22 per cent and average bank assets have sharply increased by 112 per cent from its 1997 base as the number of banks decreased as a result of consolidation. The consolidation has also brought about significant change in the competition structure in the banking industry. The degree of concentration has increased substantially, and as a result, a few large banks have emerged, dominating the banking industry. 6 // Table 3 here // Note that as a result of the restructuring program, both capital adequacy and profitability of commercial banks have improved substantially. The BIS capital ratio has increased to a level above 10 per cent, and the share of non-performing loans (NPL) classified as substandard or below in total bank assets fell sharply to 2.4 per cent by the end of With continuous NPL resolution efforts and improvements in the macroeconomic environment, the pre-provision profit of banks began to exceed loss provisions, and commercial banks finally recorded positive profits in 2001 and The productivity of commercial banks has also improved substantially, resulting in a significant reduction in the financial intermediation cost. The financial intermediation cost is the sum of operating expense, loan loss provision, tax and other expenses, and reflects operating efficiency as well as the efficiency in resource allocation. The commercial banks operating expenses as a percentage of total assets decreased to the 1 per cent level from the pre-crisis 2 per cent level, implying 5

8 productivity improvement in the commercial banking business. Indeed, in 2001, total financial intermediation cost to asset ratio dropped to 2.1 per cent from 2.7 per cent in the pre-crisis period as loan loss provisions fell sharply. Characterizing financial flows before and after the crisis Progress in financial sector restructuring as outlined above also caused an evolution in the financial structure. To highlight the fundamental nature of the transition, it is first necessary to understand the characteristics of the pre-crisis financial flow pattern. Financial liberalization and the rapid growth of NBFIs in the pre-crisis period Hahm (2003) studied the dynamic relationship between the government, chaebol and financial institutions in pre-crisis Korea, and argued that the weakening of the traditional risk partnership caused by financial liberalization brought about a shift in the financial structure in the late 80s and 90s. According to Hahm (2003), the corporate financing pattern before the crisis can be characterized by two notable features. First, the volume of credits intermediated by non-bank financial institutions (NBFIs) increased substantially - both in borrowings and in NBFI-related direct financing such as commercial paper. Second, the maturity structure of corporate debts - both domestic and foreign borrowing - became increasingly shorter. These features that characterized financial vulnerability of the Korean economy were significantly related to the nature of financial liberalization policies undertaken 6

9 throughout the 80s and 90s. The financial liberalization program in Korea revealed quite an unbalanced and asymmetric nature (Lee et al. 2000; Cho 2001; Choi 2002; Zhang 2002; Hahm 2003). The first asymmetry came from the unbalanced deregulation across commercial banking and non-bank financial industries. NBFIs were allowed much greater freedom in their management of assets and liabilities. Most crucially, the deregulation of interest rates on non-bank financial products, while effectively controlling bank deposit rates, contributed to the excessive and rapid expansion of NBFIs over commercial banks in the late 80s and early 90s. The share of NBFIs increased rapidly and accounted for more than 50 per cent of financial savings from the mid 80s, while the share of commercial banks decreased steadily from nearly 60 per cent in 1980 to 27 per cent in As a result of the shift, financial risks for the entire economy increased substantially because the NBFIs were not adequately supervised by financial authorities (Hahm and Mishkin 2000) and there were no effective market monitoring mechanisms to limit their excessive risk taking. Furthermore, given that funds intermediated by the NBFIs, such as commercial paper financing, were more shortterm in nature, 8 the resulting fund-flows increasingly shortened average maturity of corporate financing, thereby exposing the corporate sector to substantial liquidity and refinance risks. Furthermore, as a result of the differential ownership regulation toward NBFIs, the influence of chaebols over NBFIs began to steadily increase (Kim 1999). 9 While 7

10 privatizing the banking industry in early 80s, the government actually strengthened bank credit controls in an attempt to redirect credits to non-chaebol sectors such as small- and medium-sized enterprises. Faced with the tight control of bank credits, chaebols found the NBFIs to be an important alternative financing source. At the onset of the financial crisis, the 70 largest chaebols turned out to have ownership in 140 NBFIs. This dominance of chaebols in the non-bank financial industries, especially in the absence of effective supervision and monitoring schemes, brought about serious conflict of interest problems and resulted in various negative consequences during the crisis period (Kim 1999). Another asymmetry came from the unbalanced deregulation of capital accounts in the post-1993 period. In 1993, the Korean government expanded the list of usage for which financial institutions could provide foreign-currency denominated loans. Short-term foreign borrowing by financial institutions was allowed at the time, while the government maintained quantity restrictions on long-term foreign borrowing as a means of capital flow management (Cho 2001; Hahm and Mishkin 2000). Furthermore, gradual deregulation of trade related short-term external borrowing of corporate firms aggravated the maturity structure of external debt, exposing both corporate and financial sectors to foreign exchange liquidity risk. Resurgence of commercial banks in the post-crisis period As outlined above, the Korean government undertook a range of structural reforms, and these reforms in turn affected the financing and asset allocation 8

11 behavior of corporations and financial institutions. Hahm (2002) investigated the post-crisis financial transition and characterized it as a correction process of the pre-crisis financial intermediation pattern distorted under government implicit guarantees and unbalanced deregulations. Among the various characteristics of the post-crisis financial flows, the first notable feature is the dramatic reversal in the share of commercial banks relative to the NBFIs and direct financing. As can be seen in Figure 1, as depositors and investors began to perceive risks associated with financial products of NBFIs and with the massive failure of insolvent NBFIs, the share of deposits received at commercial bank accounts in total deposits at financial institutions increased sharply in the post-crisis period. Consequently, as shown in Figure 2, the share of commercial banks in corporate indirect financing, which was below 50 per cent during the 1990s, also increased sharply in the post-crisis period. // Figure 1 here // // Figure 2 here // With the contraction in non-bank financial industries, the maturity structure of corporate direct financing also improved significantly. The collapse of the merchant banking industry and commercial paper markets meant that corporations could no longer roll over short-term debt. As corporate borrowers redeemed commercial paper by issuing corporate bonds and stocks, the share of short-term debt in total direct financing fell sharply from over 22 per cent to 10 per cent (Hahm 2002). The flow of financial liquidity into the banking system accelerated from late 9

12 1999 after Daewoo s collapse. The average deposit balance of commercial bank accounts jumped up 2.9 times from 137 trillion won in 1997 to 400 trillion won in January This concentration of financial savings into the banking industry - the resurgence of commercial banks - reflected various factors. Above all, given that the unusual expansion of the NBFIs during the 80s and 90s reflected the existence of the implicit government guarantee over NBFIs that were under the influence of chaebols, this shift can be understood as a correction of the previously distorted financial flows. Indeed, the breakdown of the too-big-to-fail expectation for chaebols and the resulting depositors preference for safety helped commercial banks reclaim their share of the market, especially once they succeeded in restoring their capital adequacy ratios with the help of the government s bank recapitalization program (Hahm 2002). However, the recent overflow of financial saving into the banking system seems to be unusual in that it is also an outcome of the unbalanced and unfinished government restructuring programs. Note that the government adopted a bankfirst, NBFIs-later approach (KDI 1998) as a strategy in financial sector restructuring. In fact, the flow of funds during the restructuring episode showed much volatility and close correspondence with the actual restructuring policies. With the outbreak of the crisis, the government s ultimate priority was in normalizing the banking system as soon as possible, and the policy of allowing bond financing for chaebols through investment trust companies while restructuring the banking system was intended to prevent a severe credit crunch. 10

13 However, as a result of the sequential approach, a large volume of funds flowed through the capital market into insolvent chaebols like Daewoo, which not only postponed the resolution of insolvent corporate firms, but also magnified the scale of the problem (Oh and Rhee 2001). With heightened uncertainty in the aftermath of Daewoo s collapse in 1999, the flow of funds reversed, shifting back away from investment trust companies and toward commercial banks. Being politically constrained in procuring additional public funds for financial restructuring, the government failed to complete the corporate sector restructuring, and the restructuring of NBFIs also remain largely unfinished. With the high credit risks unresolved in the corporate sector and with the breakdown of the traditional implicit guarantee mechanism, financial savings could not flow into the capital market or shaky non-bank financial industries. 10 The second notable feature observed in post-crisis financial flows is the shift in the asset allocation behavior of commercial banks. The share of loans in commercial bank assets gradually recovered to the pre-crisis level, but there was also a shift in the underlying composition away from loans to enterprises and toward loans to households. As can be seen in Figure 3, the share of consumer loans - loans to households and loans for housing more than doubled from 20 per cent in 1996 to over 50 per cent in // Figure 3 here // Two factors underlie this shift. The first factor comes from the demand side. The development of capital markets and increasing availability of direct financing 11

14 reduce demand for bank credit, especially from large corporations with good credit standing. Note that in this situation, average credit risk of bank borrowers increases, and hence, bank asset quality may deteriorate due to adverse selection problems. Banks now recognize that, on average, risky corporate borrowers who cannot obtain credit from capital markets would remain and actively seek out bank loans. In recognition of this, commercial banks may reduce exposures to corporate loans. Indeed, with increasing emphasis on risk management, commercial banks themselves tried to avoid loan concentration by introducing exposure limits on corporate loans. This led them to increasingly emphasize more diversified consumer loans. Bank consolidations and new risk factors The bank consolidation process initially led by the government restructuring impetus was recently accelerated by enhanced competition among banks. Both mergers and acquisitions within the banking industry to exploit economies of scale and strategic alliances with other non-bank financial industries such as insurance and asset management to realize economies of scope have become increasingly popular. The consolidation process has brought about the shift in the competition structure of the banking industry, and the degree of concentration has increased as a few large banks have emerged. Figure 4 shows the asset shares of the top 3 and 5 commercial banks in total commercial bank assets, which clearly reveals the increased concentration in the banking industry in the post-crisis period. 12

15 // Figure 4 here // The combination of the financial savings concentration in the banking industry discussed above and the emergence of a few large banks dominating the banking industry as a result of the on-going bank consolidation presents potentially significant risks to financial stability. The first source of risk comes from the distortion of bank balance sheets as a result of the asymmetry between the demand for bank credit and the supply of funds to banking institutions. As emphasized above, relatively large firms with good credit standing would gradually increase the share of internal and direct financing as an alternative to indirect bank financing. In contrast, small- and medium-sized firms and firms with low credits are increasingly dependent on bank borrowing. The concentration of financial savings in the banking system and the increasingly less and lower quality demand for bank credits from the corporate sector may create significant distortions in bank balance sheets. As noted by Hoshi and Kashyap (1999), excessive supply of funds to the banking industry and resulting balance sheet distortions caused banking crisis in Japan. As a result of deregulations in the capital market from the late 70s, large firms bank dependency decreased substantially; however, regulations on savings and asset management of banks continued so that financial savers continued to supply funds to commercial banks. With this asymmetry, banks with increasingly abundant liquidity extended risky credits, such as property loans, which created real estate price bubbles and the subsequent burst and the banking crisis. Similar phenomena have been recently 13

16 observed in Korea. As shown above, with excessively abundant liquidity, commercial banks have increased consumer loans - especially housing loans. As a result, household debt as a ratio of GDP sharply increased from 50 per cent in 1997 to 65 per cent in 2001, and housing prices also increased substantially. Increased concentration in the banking industry also presents non-trivial risks to the financial system. It is controversial whether or not concentration in the banking industry undermines competition and the efficiency of financial intermediation. It is also controversial whether or not there exists a trade-off between financial stability and the efficiency of financial intermediation. While it is a general presumption that increased concentration (and thus higher market power) of banks leads to less agency costs and moral hazard problems and promotes financial stability (Keeley 1990), more recent literature on bank consolidation and financial market stability have found the opposite results. Note that bank consolidation leads to the emergence of a small number of large banks dominating the banking industry. A situation where dominant banks are toobig-and-few-to-fail creates incentives for financial supervisors to engage in regulatory forbearance, since the failure of large banks usually involves significant negative externalities. The liquidation of a large bank takes time with non-trivial costs as the process itself could be disorderly. The problem is that the recognition of the expectation on the bailout of large banks greatly reduces incentives for depositors and bank creditors to monitor banks and penalize them for taking excessive risk. As a result, excessive risk taking and moral hazard of large banks 14

17 may occur, leading to a higher degree of systemic risk (Edwards 1996). Recent empirical literature supports the hypothesis of increased risk taking of large complex banking organizations in advanced countries (Chong 1991; Demsetz and Strahan 1997; De Nicolo 2000). Furthermore, even without distortions in incentives, the financial market discipline and supervisory function could be weakened due to the increasingly complex asset and business structures of large banking institutions. Not only is it difficult to monitor the complex organizations on a timely basis, but belated discovery of the problem increases incentives for regulatory forbearance. In addition, although the extent of diversification may contribute to the financial stability of individual institutions, large banking institutions themselves will share increasingly similar characteristics in their business portfolios and asset structures. According to the G10 report on financial consolidation (2001), both direct exposure through various on- and off-balance sheet transactions (such as financial derivatives) and indirect exposure with increasingly similar asset and profit structures have substantially increased among large banking institutions in advanced countries. 12 Figure 5 shows the short-term inter-bank call loan exposure measured as a share of bank capital for the top 3 and top 5 banks in Korea. As can be seen, the inter-bank exposure seems to have increased in the post-crisis period, and this higher dependence among large banks may become a source of contagion and systemic risk. // Figure 5 here // 15

18 Policy lessons and implications As we have seen, the sequential restructuring approach to bank versus non-bank financial institutions, coupled with delayed resolution of insolvent corporate firms, has caused asymmetric fund flows and the over-concentration of financial savings in the banking sector in the post-crisis period. As the interaction of the unusual fund flows and the on-going bank consolidation may create new sources of systemic risk, policy-makers must establish a robust framework as a guard against potential risks. In this regard, following four policy tasks are critical. Resolution of remaining credit risk by accelerating corporate restructuring Note that the asymmetric fund flows and concentration of financial savings in the banking sector is an outcome of the low risk appetite of savers and investors as uncertainty and credit risk still remains high in the financial market. In this situation, financial flows cannot be directed to the capital market, as pricing functions may not work due to severe information asymmetry. Acceleration of corporate restructuring to sort out and resolve non-viable firms is critical to reduce market uncertainty over corporate viability. According to Kwon (2003), regardless of the visible improvement in corporate financial structure and debt service capacity of average listed firms, a significant number of firms, whose profitability and debt service capacity show no sign of improvement, are still operating in the market. 13 This indicates that banks are still refinancing the debts of these potentially insolvent firms, possibly due to the negative consequences of corporate resolutions 16

19 on bank profits and capital. In this regard, in order to accelerate corporate restructuring, further strengthening of the asset classification and loan loss provisioning requirements is important. 14 Also, the capital adequacy of commercial banks must be further improved for them to be able to effectively carry out corporate restructuring. Restructuring of non-bank financial institutions As emphasized above, the government restructuring effort for resolution of nonperforming assets and recapitalization was relatively more directed to the banking industry. As a result, the asset quality and capital adequacy of NBFIs remained poor. For instance, at the end of 2002, the non-performing loan ratios of merchant banking corporations and leasing companies were 10.5 and 46.8 per cent, respectively, and the NPL ratio of mutual savings banks remained at 14.8 per cent. 15 The resolution of insolvent investment trust companies was also delayed due to high-risk corporate bonds included in the portfolio of trust funds. Recent booms in consumer credit card loans and the subsequent hike in delinquency rates also substantially deteriorated the capital adequacy of credit card companies. In contrast to the banking industry, the overall profitability of NBFIs remained low as their market base was significantly eroded due to the collapse of market credibility during the financial crisis. Consequently, NBFIs failed to accumulate adequate loan loss reserves against the deterioration of asset quality. Note that delayed restructuring of insolvent NBFIs would deepen the distortion of fund flows. 17

20 Financial supervisory authorities must strictly apply prompt corrective action provisions to insolvent NBFIs, even if the resolution may require additional use of public funds. Early restoration of market credibility for NBFIs by accelerating the required restructuring would facilitate diversification of financial savings and encourage funds to flow into the capital market, which in turn would contribute to the balanced growth and stability of the financial system. Strengthening of supervision for large banks and limiting too-big-to-fail As discussed above, large banks may engage in moral hazard and aggressive risk taking given the possibility of regulatory forbearance and expectations of too-bigto-fail. An important way to ensure that bank supervisors do not engage in regulatory forbearance is through implementation of prompt corrective action provisions, which require bank supervisors to intervene earlier. Prompt corrective action is crucial to preventing bank failures because it creates incentives for banks not to take on too much risk in the first place, recognizing that if they do so, they are more likely to be subject to regulatory actions. In Korea, prompt corrective action provisions for commercial banks and merchant banking corporations were introduced in April 1998 and then subsequently extended to securities and insurance companies. According to the provision, banks are classified into five groups by the BIS capital ratio and the CAMELS-based evaluation results of bank management. 16 The supervisory authority could impose various corrective measures whenever banks BIS ratios and management evaluation 18

21 grades fall below predetermined criteria. Note that the key element in making prompt corrective action work is the mandatory nature of the scheme, which makes it a credible threat for financial institutions. Hence, discretionary application of the provision must be minimized. In the case of large financial institutions, systemic risk could be a concern when strictly applying the prompt corrective action. In reality, most countries at least implicitly have a too-big-to-fail policy in which all depositors are fully protected if a large bank fails. However, this systemic risk concern itself brings about moral hazard for large banks. Moreover, the expectation of future bailouts causes additional distortions in fund flows and increases market power of large banks, which in turn results in implicit government subsidies to large banks with taxpayers money as collateral. It is important to recognize that although large banking institutions may be too big to liquidate, they can be closed with losses imposed on uninsured creditors. Except under very unusual circumstances, the least-cost resolution procedure imposing loss to uninsured depositors and creditors must be strictly applied. In recognition of the criticism on the inefficient management of public funds, in December 2000, the Korean government enacted the Special Act on Public Fund Management. According to the Act, the Public Fund Oversight Committee (PFOC) was established under the Ministry of Finance and Economy. The principle of least-cost resolution was formally introduced in the act. However, the principle can be applied in a discretionary way by the judgment of the PFOC over systemic risk concerns. To prevent regulatory forbearance for large banks, the 19

22 conditionality for the systemic risk exception must be explicitly set out and strengthened further. Strengthening of disclosure requirements and market monitoring functions Note that the increasing complexity of the asset portfolio and business structure of large banking institutions substantially weakens both the financial authority s supervisory function and monitoring ability of concerned parties. An answer to these problems is to have the market discipline financial institutions by providing more transparent information on bank management and by establishing more marketbased supervisory schemes. Disclosure requirements are essential to the ability of the market to have information, which allows them to monitor financial institutions and keep them from taking on too much risk. A recent study by the U.S. Federal Reserve Board indicates that disclosure requirements for large complex banking organizations need to be strengthened in the areas such as securitizations and loan sales, internal asset risk rating and loan loss reserve calculations, credit concentrations by counterparty, industry, or geography, market risks, and risks by legal entity and business lines (Board of Governors of the Federal Reserve System 2000). In the same vein, public disclosure requirements need to be strengthened for large Korean banks too. With the effort to promote information transparency, supervisory authorities need to introduce more market-based measures, such as requiring banks to issue subordinated debt (Hahm and Mishkin 2000). Subordinated debt with a ceiling on 20

23 the spread between its interest rate and the interest rate on government bonds can be an effective disciplinary tool. If the bank is taking on too much risk, it is unlikely to be able to issue subordinated debt within the spread cap. Hence, compliance with the subordinated debt requirement would be a direct way for the market to force banks to limit their risk taking. Alternatively, deposit insurance premiums could be charged according to the interest rate on the subordinated debt. Information about whether banks can issue subordinated debt and the interest rate on the subordinated debt itself can help the public evaluate supervisors action, which in turn reduces the scope of regulatory forbearance. Concluding Remarks This article examined the nature of Korea s post-crisis evolution of the financial structure and sought to establish its relationship with government restructuring policies. We found that the sequential restructuring approach to bank and non-bank financial institutions and the delayed resolution of insolvent corporate firms caused excessive concentration of financial savings in the banking sector. Risk factors associated with the unusual fund flows were identified, and policy measures were suggested to cope with the risk factors. Given that the causes of the Korean financial crisis were deeply rooted in the structural deficiencies of the financial system, more fundamental structural reform is imperative to the sustainable growth of the economy. However, the Korean experience reveals that more prudent and balanced reform policies are critical in 21

24 order to have a smoother transition and to prevent the magnification of systemic risk during the reform process. Notes 1 As the recent asymmetric information view of financial crises (e.g. Mishkin 1997) indicates, the deterioration of bank balance sheets is an important factor in the causes of financial crises. Hahm and Mishkin (2000) diagnosed the degree of non-performing loan problems of Korean commercial banks at the onset of the financial crisis by applying more realistic asset classification criteria, and argued that Korean commercial banks had been significantly undercapitalized from the mid 90s much earlier than the actual financial crisis. 2 Among the trillion won, the Korea Deposit Insurance Corporation spent 99 trillion won to payoff deposits and recapitalize insolvent financial institutions. Korea Asset Management Corporation spent 38.7 trillion won to purchase non-performing assets from financial institutions. 3 From January 2000, the forward-looking criteria (FLC) asset classification system was adopted in Korea, according to which, banks are required to accumulate loss provisions against loans extended to firms whose debt service capacity is in doubt regardless of the firm s actual status of interest payment deferral. 4 From January 2000, the forward-looking criteria (FLC) asset classification system was adopted in Korea, according to which, banks are required to accumulate loss provisions against loans extended to firms whose debt service capacity is in doubt regardless of the firm s actual status of interest payment deferral. 22

25 5 Bank restructuring was carried out in accordance to the following strategies. First, the 12 commercial banks that failed to meet the BIS 8% capital adequacy ratio submitted restructuring plans in April Upon receiving the plans, 6 internationally recognized accounting firms conducted an assessment of their assets. The process was intended to provide a realistic assessment of the size of non-performing loans applying international loan classification criteria and thereby making the involved banks balance sheets more transparent. Second, the Bank Appraisal Committee evaluated recapitalization plans submitted with a view to the adequacy in capitalization, asset soundness, profitability, liquidity, management expertise and the prospects of future achievement of the target BIS ratios. With the input from appraisal results, the Financial Supervisory Commission (FSC) examined the feasibility and arrived at either an approval, conditional approval or disapproval of the plan for respective banks. See Shin and Hahm (1998) for a more detailed discussion on the financial restructuring framework. 6 Kim (2002) investigated the degree of concentration in the Korean banking industry and found that the Herfindahl-Hirschman index in the deposit market has sharply increased from 654 in December 1997 to 1,367 in December From November 1997 to June 2002, KAMCO spent 24.4 trillion won to purchase nonperforming loans from commercial banks. Banks themselves have also spent most of the operating income in accumulating loan loss provisions and writing-off bad debts. 8 For instance, merchant banking corporations almost virtually monopolized the commercial paper underwriting business. 9 The Banking Act amendment of December 1982 introduced an explicit ownership regulation. The ownership of commercial banks was limited up to 8 per cent of voting shares. The ownership limit was strengthened to 4 per cent from Different from the 23

26 banking industry, no outright ownership regulation was applied to the NBFIs. 10 For instance, the practice of guaranteeing corporate bond issues by commercial banks has been largely discontinued since Another feature that characterized the shift in commercial bank asset portfolio structure in the post-crisis period was an increase in the share of security holdings in total bank assets (Hahm 2002). 12 For instance, De Nicolo and Kwast (2002) investigated systemic risk presented in the U.S. banking industry over the period of , and found a positive trend in stock return correlations as well as a positive consolidation elasticity of correlations. They interpreted the evidence as suggesting that the systemic risk potential may have increased with on-going consolidations in the banking industry. 13 According to Kwon (2003), while the average debt service capacity of listed firms has improved substantially since 2001, the share of firms whose interest coverage ratio is less than one still remains at 14.9% as of June The interest coverage ratio is the ratio of EBITDA (earnings before interest and tax plus depreciation and amortization) to the firm s interest payments. If the ratio is less than one, the firm s operating cash flow is not sufficient to cover the firm s interest payments. 14 After the crisis, asset classification criteria have been substantially strengthened to conform to international best practices. From July 1998, the substandard or below category was strengthened to include assets with interest payments in arrears by 3 months or longer from the previous 6 months criteria, and precautionary asset category to include assets in arrears by 1 to 3 months from the previous 3 to 6 months criteria. In addition, from January 2000, forward-looking criteria (FLC) in asset classification were adopted, and banks are now required to accumulate loss provisions against loans extended to firms 24

27 whose debt service capacities are in doubt, regardless of the firm s actual status of interest payment deferral. However, since actual criteria for the FLC are to be established by banks themselves, there exist possibilities of discretionary and more lenient applications. 15 Monthly Financial Statistics Bulletin, Financial Supervisory Service, February CAMELS represents five criteria in evaluating bank management: capital adequacy, asset quality, management, earnings, and liquidity. 25

28 References Board of Governors of the Federal Reserve System (2000) Improving Public Disclosure in Banking, Staff Study 173. Cho, Y. J. and Kim, J. K. (1997) Credit Policies and the Industrialization of Korea, Seoul: Korea Development Institute Press. Choi, D. Y. (2002) Bidaechingjeok Kiupkeumyunggyujewa Oeywhanuiki [Asymmetric Regulations on Corporate Financing and the Currency Crisis], Seoul: Hankuk Kyongje Yeonguwon [Korea Economic Research Institute]. Cho, Y. J. (2001) The role of poorly phased liberalization in Korea s financial crisis, in G. Caprio, P. Honohan and J. E. Stiglitz (eds) Financial Liberalization: How Far? How Fast?, Cambridge: Cambridge University Press. Chong, B. S. (1991) The effects of interstate banking on commercial banks risk and profitability, Review of Economics and Statistics 73: Demsetz, R. S. and Strahan, P. E. (1997) Diversification, Size, and Risk at Bank Holding Companies, Journal of Money, Credit, and Banking 29(3): De Nicolo, G. (2000) Size, charter value and risk in banking: an international perspective, International Finance Discussion Paper, no 689, Board of Governors of the Federal Reserve System, December. Edwards, F. R. (1996) The New Finance: Regulation and Financial Stability, Washington D.C: AIE Press. Group of 10 (2001) Report on Consolidation in the Financial Sector, BIS, IMF, 26

29 OECD. Hahm, J. H. and Mishkin, F. S. (2000) Causes of the Korean financial crisis: lessons for policy, in I. S. Shin (ed.) The Korean Crisis: Before and After, Seoul: Korea Development Institute Press. Hahm, J. H. (2003) The government, the chaebol and financial institutions before the economic crisis, in S. Haggard, E. Kim and W. Lim (eds) Economic Crisis and Corporate Restructuring in Korea, Cambridge: Cambridge University Press. Hahm, J. H. (2002) Reconstruction and reform of the Korean financial system, mimeo presented at the RCIF International Conference, The Korean Economy: Beyond the Crisis, Seoul, October. Hoshi, T. and Kashyap, A. (1999) The Japanese banking crisis: where did it come from and how will it end?, in B. Bernanke and J. Rotemberg (eds) NBER Macroeconomics Annual 1999, Cambridge, MA: The MIT Press. Keeley, M. (1990) Deposit insurance, risk and market power in banking, American Economic Review, 80: Kim, J. K. (2000) Chaebols and financial sector restructuring in Korea, mimeo, Korea Development Institute. Kim, W. J. (2002) Woorinara Eunhangsanupe jipjungdo beonwha bunseok [An analysis on the concentration in the Korean banking industry], Kumyung System Review [Financial System Review], , Bank of Korea, pp Korea Development Institute (1998) Structural Reforms and Economic Prospects, 27

30 Seoul: Korea Development Institute Press. Kwon, J. J (2003) Kiupy Jaemukeonjunsunggua Eunhang Kiupkogaek Junryak [Financial soundness of corporate firms and bank strategy toward corporate borrowers], mimeo, Hankuk Keumyung Yeonguwon [Korea Institute of Finance]. Lee, C. H., Lee, K. and Lee, K. K. (2000), Chaebol, financial liberalization, and economic crisis: transformation of quasi-internal organization in Korea, mimeo, University of Hawaii. Lee, D. G. and Kim, D. S. (2001) Eunhangsanupe Gwageo Hyunjae Mirae [Past, present and future of banking industry in Korea], a chapter in Hankukkeumyungsanupe Gwageo Hyunjae Mirae [Past, Present and Future of Financial Industries in Korea], Hankuk Keumyung Yeonguwon [Korea Institute of Finance]. Mishkin, F. S. (1997) The causes and propagation of financial instability: lessons for policymakers, in C. Hakkio (ed.) Maintaining Financial Stability in a Global Economy, Kansas City: Federal Reserve Bank of Kansas City. Mishkin, F. S. (1999) Financial consolidation: dangers and opportunities, Journal of Banking and Finance 23(2-4): Oh, G. T. and Rhee, C. Y. (2001) Owewhan wuikiihu jakeumsoonwhane teukjing: hoesachaebusilwhaedaehan yeonku [Characteristics of the post-crisis flow of funds: a study on the corporate bond market], a paper presented in the 2001 Congress of Economic Associations in Korea. 28

31 Shin, I. S. and Hahm, J. H. (1998) The Korean crisis - causes and resolution, in L. Cho, Y. Kim and I. Shin (eds) Restructuring the Korean Financial Market in a Global Economy, Seoul: Korea Development Institute Press. Zhang, X. (2002) Domestic institutions, liberalization patterns, and uneven crises in Korea and Taiwan, Pacific Review 15 (3):

32 Table 1 Progresses in financial restructuring (number of financial institutions a ) Financial Institutions End Restructuring Newly 1997 Closed d Merged Total Entered Bank b Merchant banks Securities houses Insurance companies c Investment Non- Bank trust corp. Mutual Savings Credit unions 1, ,252 Lease Total 2, ,510 Source: Public Fund Oversight Committee, Republic of Korea. Notes: a Excluding branches of foreign financial institutions. b Including specialized banks. c Excluding postal insurance. d Transferred to bridge financial institutions, license revoked, liquidated, etc. Table 2 Non-performing loans of financial institutions (trillion Korean won, per cent) All financial Institutions Banks c a (21.8) b 88.0 (14.9) n.a (12.9) 64.6 (10.4) 42.1 (8.0) 39.1 (5.6) 18.8 (3.4) 31.8 (3.9) 15.1 (2.3) Source: Public Fund Oversight Committee, Republic of Korea. Notes: a Non-performing loans are assets classified as substandard or below under forward-looking criteria. b Numbers in parenthesis denote non-performing loan ratios as a percentage of total assets. n.a. = not available. c Including both commercial banks and specialized banks. 30

33 Table 3 Commercial bank restructuring and consolidations Number of commercial banks a (-42.3) e Number of branches a 4,557 5,987 4,780 4,709 4,776 5,021(-16.1) Number of employees a (1,000s) (-41.2) Total bank assets (trillion won) a,c (22.0) Average bank assets a (trillion won) (112.0) BIS capital ratio a (%) NPL ratio d (%) ROA b (%) ROE b (%) Source: Bank Management Statistics, Financial Supervisory Service, Republic of Korea. Notes: a End of period values b During the period, including trust accounts c Including trust accounts. d Ratio of assets classified as substandard or below. e Numbers in the parenthesis in the last column denote the rate of change from the 1997 base. 31

34 Figure 1 Share of commercial bank accounts in total deposits at financial institutions % Source: Complied from Money and Banking Statistics, Bank of Korea, various issues Figure 2 Share of commercial banks in corporate indirect financing % Source: Compiled from the Flow of Funds Accounts Statistics, Bank of Korea 32

35 Figure 3 Share of household and housing loans in commercial bank loan % Source: Compiled from Monthly Financial Statistics Bulletin, Financial Supervisory Service, various issues. Figure 4 Concentration of assets in commercial banking industry CR3 CR5 Source: Compiled from Bank Management Statistics, Financial Supervisory Service, various issues. 33

36 Figure 5 Commercial bank call loan to bank capital ratio Top 3 Banks Top 5 Banks Source: Compiled from Bank Management Statistics, Financial Supervisory Service, various issues. 34