V,4eS i't6) POLICY RESEARCH WORKING PAPER 1481 Corporate Control in A governance system based on bank ownership and Central Europe and Russia control of firms is not yet feasible, but a good case can Should Banks Own Shares? be made for allowing banks to gradually gain experience through limited equity Peter Dittus ownership especially for Stephen Prowse allowing the swapping of bad debt for equity. The World Bank Policy Research Department Transition Economics Division June 1995 I POLICY RESEARCH WORKING PAPER 1481 Summary findings Dittus and Prowse review corporate governance countries. This reflects limited understanding of the arrangements in the West and conclude that for a system financial sector's poor condition, the many institutional based on bank ownership and control of firms to and political obstacles to bank reform, and the initial succeed, the banking system must be free of perverse decision in many countries to focus first on the "real incentives and state interference, as well as subject to economy" (a decision that in hindsight seems adequate supervision by banking authorities and unfortunate). competition from market forces. Admirable progress * Dispel the belief (which still exists in some over the past few years notwithstanding, these conditions countries) that poor lending and investments will do not now exist in the countries of Central Europe and eventually be underwritten by the government, with few Russia, so a corporate governance system based on bank consequences for managers. ownership is not appropriate. That is not to say that such Greatly strengthen competition in the banking a system would not eventually be appropriate - but not system, in part by encouraging new private banks and the before much more effort is made to create a competitive, entry of foreign banks. (Some countries, such as Poland, private, well-supervised banking system (which is needed have taken the opposite tack, refusing to issue new in any case). licenses.) Changes in the banking system that are prerequisites - Provide effective bank supervision and an effective for any large-scale bank involvement in the ownership prudential and regulatory framework. This requires and governance of firms are simple to enunciate but less investing substantially in setting up institutions, easy to implement: accounting systems, and information networks, in hiring * Sever existing relationships between the state and and training qualified personnel, and in ensuring that the banks. Privatization is the strongest giuarantee that bank system is immune from political intervention. investment decisions will not be subject to state Developing such a system will surely be long and drawn influence, but bank privatization has been slow in most out, and may require foreign assistance. This paper - a product of the Transition Economics Division, Policy Research Department - is part of a larger effort in the department to explore issues of corporate governance in transition economies. The study was funded by the Bank's Research Support Budget under the research project "Corporate Governance in Central Europe and Russia" (RPO 678- 42). Copies of this paper are available free from theWorld Bank, 1818 H StreetNW, Washington, DC 20433. Please contact Grace Evans, room Nl1-041, extension 85783 (39 pages). June 1995. The Policy Researcb Working Paper Series disseminates the findings of ulork in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be used and cited accordingly. The findings, interpretations, and conclusions are the authors' own and should not be attributed to the World Bank, its Executive Board of Directors, or any of its member countries. Produced by the Policy Research Dissemination Center Corporate Control in Central Europe and Russia: Should Banks Own Shares? Peter Dittus and Stephen Prowse* Peter Dittus Stephen Prowse Bank for International Settlements Federal Reserve Bank of Dallas 4002 Basle 2200 N. Pearl Street Switzerland Dallas, TX 75201 Phone: 41 61 2809249 Phone: 214922 5230 Fax: 41 61 280 9100 Fax: 214 922 5194 Internet: 100420.146@compuserve.com Internet: stephen_prowse@frb-main- 5.ccmail.compuserve.com The views expressed in this paper are our own and not necessarily those of the Bank for International Settlements or of the Federal Reserve Bank of Dallas. We owe a great debt of gratitude to staff at the Czech, Hungarian and Polish National Banks for very useful discussions and for support in setting up meetings, and to managers of commercial banks for their willingness to take the time and discuss with us their banks' strategy. Herbert Baer, Stijn Claessens, Cheryl Gray, Mirsolav Hrncir and Ian Mladek provided valuable comments or ideas. We are grateful to participants at a BIS seminar for their suggestions on an earlier draft of this paper. 1. Introduction Five years after the start of economic reforms in central and eastern Europe, the issue of corporate governance remains largely unresolved. While privatisation, foreign direct investment, increasingly competitive markets and the introduction of hard budget constraints have enforced some discipline on managers and encouraged rationalisation and downsizing in firms, there is undeniably a great need for a strong mechanism of corporate governance in these countries which can effectively replace managers where appropriate and actively restructure firms. Some observers have argued that banks should fill the vacuum by swapping debt for equity, or purchasing equity outright, and taking an active role in the restructuring and governance of nonfinancial enterprises. They base their recommendation on the experience of countries like Japan and Germany where banks have played an important role in corporate governance.1 Do these countries provide an appropriate role model? Should banks in central Europe and Russia own shares and become involved in the governance of nonfinancial firms? These are the questions we attempt to answer in this paper. We do so by analysing the corporate governance systems in different western countries and attempting to derive lessons for the countries in central Europe, recognising the existing conditions in these transition countries with regard to their capital markets and banking systems. Doubtless banks are already playing a major role in corporate governance in central Europe and Russia, and will continue to do so. The reason is simply that they hold the lion's share of enterprise debt, most of it being of short maturity. The possibility to refuse to roll over credit gives bank a potentially powerful control instrument over enterprises. A companion paper by Baer and Gray (1994) deals with this aspect of corporate control by banks. Our focus is narrower. We ask whether banks can and should play a role in corporate governance based on ownership rights. To anticipate the conclusion, in our view a corporate governance system based primarily on ownership and governance by banks is not appropriate at this stage for central Europe and Russia. This is not to say that such a system may not eventually be appropriate. But for this banks must be free from state interference and subject to strong competition from market forces and adequate supervision. Beliefs that the state will continue to bail out banks in difficulties must be credibly dispelled. Clearly, some progress has been made on all these fronts in the last few years. Nonetheless we are convinced that much more needs to be done before an active role of banks in governance based on ownership can be envisaged. Interviews with bank managers in the region have reinforced our view and lead us to believe that banks' involvement in corporate governance based on ownership interests will remain marginal in the foreseeable future. There may however be benefits from allowing a limited role for some banks to own equity and restructure firms that have defaulted on loans to the bank. This would enable them to obtain experience in the ownership and governance of firms that may bear fruit later when the banking system is in a position to play a more active role in this area. The arguments we advance to make our case against a major role of banks in the ownership and control of non-financial firms might be understood as suggesting that existing banks in some countries should not lend either in current circumstances. This question has been subject to heated debate before, but is not addressed here. The reason is that we feel in all the countries under review the decision to build a functioning banking system on the basis of existing banks has been taken. The question now is not what to do with the existing banks, but how to develop them into efficient intermediaries as quickly as possible. The contribution of this paper to this debate is to point out that tight restrictions on the ownership of non-financial firms should form part of the strategy to develop an efficient financial industry. I See for example, Hoshi, Kashyap and Loveman (1994). Others advocating a primary role for banks in the governance of firms include van Wijnbergen (1992), Corbett and Mayer(1991), Scharfstein (1992) and Sarcinelli (1992). 2 Section 11 analyses corporate governance systems in the West. We compare the market- based system in the United States to the more bank-based systems in Japan and Germany. No attempt is made to evaluate which system is the best. The focus is on the conditions necessary for different models of corporate governance to flourish. Section III analyses the current conditions existing in central and eastern Europe. We concentrate on the degree of liquidity of corporate securities markets, the restructuring requirements of firms, and conditions in the banking sector including the degree of state influence, the bad debt burden, competition, supervision, and the incentives banks face with regard to their asset portfolio. Section IV draws the threads of the two previous sections together and compares the conditions necessary for successful operation of the various governance models in the west with the existing conditions in the transition countries. Implications are drawn for the appropriate role of banks in the ownership and governance of firms. I. The Role of Banks in Corporate Governance in the West As a field of research, comparative corporate governance is in its infancy. The major industrialised countries in the west appear to present us with dramatically different institutional arrangements for corporate governance. Yet we lack a consensus as to why these different arrangements came about, what the advantages and disadvantages of each system are, and whether one system is clearly better at the resolution of agency problems between the various stakeholders of the firm. This poses a problem for scholars attempting to derive lessons from the west for the countries in central Europe and Russia undergoing the transition from socialism. In putting forward recommendations for these countries on the role of banks in corporate governance, this paper will consequently be somewhat circumspect. It will try to identify common points of agreement among researchers on westem models of corporate governance, and derive implications for the countries in transition. We start with a description of the corporate governance mechanisms in the U.S., Japan and Germany. Particular attention is paid to the role of banks and their ownership of firms in each country. Then we describe the main costs and benefits of having banks play a primary role in the governance mechanism through the ownership of firms. We do not draw any conclusions about the relative superiority of one western model over another. Instead we identify the underlying conditions of capital markets, the restructuring requirements of firms, and bank supervision, governance and competition that appear necessary for the successful operation of each system. By comparing these requirements with the existing conditions in central Europe and Russia, we draw implications for the feasibility of the different western models in the transition countries. A. Western Models of Corporate Governance We describe the role of banks in the corporate governance mechanism in the U.S., Japan and Germany, with a particular focus on three important characteristics of the mechanism in each country: the structure of corporate ownership, bank-enterprise relations, and the market for corporate control. The structure of corporate ownership Differences in the ownership of firms across countries are illustrated to some degree by simple inspection of the aggregate statistics on the ownership of listed companies in table 1. Table I reveals the far heavier weight of banks in Japan compared to the U.S. Germany appears to lie somewhere between Japan and the U.S. in terms of bank ownership of corporate equity. In Germany, 3 however, banks have wide powers to exercise the voting rights owned by individuals but held in trust.2 Table 1 Ownership of common stock in 1990 Percentage of outstanding shares owned All corporations ................... 44.5 72.9 64.0 Financial institutions ........... 30.4 48.0 22.0 Banks ................... 0 18.9 10.0 Insurance Companies ...... 4.6 19.6 Pension funds; .................... 20.1 19.5 1 l2.0 Other ................... 5.7 J Non-financial corporations 14.1 24.9 42.0 Individuals ............. ...... 50.2 22.4 17.0 Foreign ................... 5.4 4.0 14.0 Government ................... 0 0.7 5.0 Sources: US Federal Reserve Flow of Funds, Japanese Flow of Funds, Deutsche Bundesbank Monthly Report. These aggregate figures, however, reveal nothing about the concentration of ownership nor who the large shareholders are in a typical firm, which is important from a corporate control perspective. Table 2 gives some data on ownership concentration in a sample of large, listed U.S., Japanese and German nonfinancial firms. The table highlights a number of differences: first, ownership concentration varies quite widely across countries. In the U.S. the largest 5 shareholders hold on average about a quarter of the outstanding shares of the firm. Ownership concentration is significantly higher in Japan, but is by far the highest in German companies, where the holdings of the largest 5 shareholders average over 40 percent. In addition, voting power in German firms may be more concentrated than even these data suggest. As mentioned above, banks in Germany have traditionally been given a wide latitude to vote the shares they hold in trust for smaller shareholders. Proxy votes exercised by the banks on behalf of beneficial shareholders are very important in the large German corporations, and concentrate voting rights in the hands of the banks.3 This may be particularly important in those few widely held companies in Germany where small shareholders typically deposit their shares with the banks.4 Finally, many large German firms exhibit a pyramid holding company structure, where ownership is concentrated in successive layers of holding 2 The law on proxy voting has been tightened over the years but it still appears to confer a wide power to banks to vote stock according to their wishes. Purrucker (1983) has estimated that 95 percent of private shareholders do not make use of their rights to instruct the bank on voting matters, leaving the bank with substantial discretionary voting power. 3 For example, of the 100 largest AGs in Germany in 1978, banks had a combined voting power (from their direct holdings and proxies) of greater than 25% in 41 of them. In the 56 AGs in which banks had a combined voting power of greater than 5% their average share of the vote was almost 57%. In the remaining 44 AGs, there existed a dominant nonbank shareholder owning greater than a 25% stake, but even among these firms, banks often held large combined voting stakes. 4 Even those companies that are widely held in terms of beneficial ownership may have their voting rights concentrated in the hands of a few banks. For example, Pfeiffer (1989) reports that throughout most of the 1980's, the big 3 universal banks controlled over half the voting rights in BASF, a widely held company. 4 companies, many of which are ultimately controlled by either a wealthy family or a bank (see Prowse (1994)). This structure serves to increase the control of banks over firms through their equity ownership. Table 2 Summary statistics of ownership concentration of large non-financial corporations Percentage of outstanding shares owned by the largest five shareholders Mean ................... 25.4 33.1 41.5 Median ......... .......... 20.9 29.7 37.0 Standard deviation .................. 16.0 13.8 14.5 Minimum ............ ........ 1.3 10.9 15.0 Maximum ............ ....... 87.1 85.0 89.6 Mean firm sizel (millions of US$, 1980) ......... 3,505 1,835 3,483 Mean firm size2 (millions of US$, 1980) .........1 1,287 811 1,497 I Measured by total assets. 2 Measured by market value of equity. Samples: United States: 457 non-financial corporations in 1980. Japan: 143 mining and manufacturing corporations in 1984. Germany: 41 non-financial corporations in 1990. Sources: For the United States and Japan, Prowse (1992); for Germany, Prowse (1993). Size data converted to US$, using 1980 average exchange rates and deflated by US consumer prices. One major reason for the differences in the ownership of enterprises by banks are differences in the legal and regulatory environment in the three countries. Banks in the U.S. are simply prohibited from owning any stock on their own account, while bank holding companies cannot own more than 5 percent of the firm and their holdings must be passive. Japanese commercial banks are not prohibited from owning corporate stock, although they are subject to anti-monopoly regulations that until 1987 limited a single bank's holdings of a single firm's shares to 10 percent (the limit has since been lowered to 5 percent). Banks in Germany can hold whatever share of equity they like in any nonfinancial firm, limited only by a number of prudential rules which do not appear to be particularly binding. Bank-enterprise relationships There are likely similar differences between countries in the ownership concentration of debt claims on firms, although the available data is much more sparse. The greater reliance on securities markets than on intermediated markets for debt finance in the U.S. (illustrated in table 3) may be taken as a rough indicator that fragmentation of debt claims is higher there than in Japan or Germany. A second indicator may be differences in the regulatory limits on banks' exposures to individual customers--these limits are generally less binding in Germany and Japan than in the U.S.. For example, in the U.S., banks are limited to lending 15% of their capital to an individual borrower, whereas in Japan and Germany the limits are 30% and 50% respectively.5 There are also legal and regulatory differences in the degree to which investors are permitted to take debt and equity stakes in 5 See Borio (1990). 5 the same firm. The legal doctrine of equitable subordination discourages all creditors in the U.S. from taking an equity position in the borrower, since the creditors are potentially liable to subordination should they be perceived to exert control over the firm. Table 3 Composition of companies' credit market debt, 1985 In percentages .. ---..Unit tites ..n...... Intermediated debt ................. 45 9 1 94 of which, from banks ..................... 36 n.a. 88 Securities ..................... 55 9 6 Note: Credit market debt excludes trade debt. Intermediated debt refers to loans from financial intermediaries. Securities includes commercial paper and other short-term bills and long-term bonds. Sources: Borio (1990) and national data. Differences in the strength of bank-firm relationships are consequently significant. Bank- enterprise relations in Japan have traditionally been very strong. Not only have banks typically held equity in firms, they have also been an extremely important source of financing. From 1970 to 1985, for example, Japanese firms relied on loans from financial institutions for more than 40% of their total financing needs (table 4). This may have given the main bank of a Japanese firm effective control over the firm's access to working capital and thus a much enhanced power to influence the firm's activities. In Germany, the story is somewhat different. German firms have never relied heavily on banks to finance their investment needs, because they have obtained most of their finance from internal funds.6 From 1970 to 1985, for example, German firms raised only about 20% of total funds from financial institutions (primarily banks). What governance power the banks have does not appear to stem from any control over the firm's access to funds.7 In the U.S., banks don't own equity nor are they an important source of funds to the typical large firm. Bank-firm relationships in the U.S. are typically arms length. The strength of bank-firm ties in Japan is reflected in the frequency with which creditors are simultaneously equity holders of firms. For example, Prowse (1990) reports that in over 40% of a sample of large Japanese manufacturing firms the largest debtholder was also the largest shareholder; 6 This is partly a function of the differing institutional arrangements for pension provision in Germany. About two-thirds of the funds earmarked for the payment of private pensions is retained by the company as an unfunded liability. Only the remainder is invested outside the company via private pension funds. The funds retained by the company are used for general corporate purposes. The result is that there is less capital available for the capital market and less demand for outside financing than in Anglo-Saxon countries where the bulk of private pensions are channelled through private pension funds (see Edwards and Fisher (1994)). The Japanese private pension system is relatively small compared to the public pension system and is funnelled largely through the banking system via trust banks. 7 It is true that as universal banks, German banks typically underwrite equity and bond offerings by the firm, giving them more influence in this sense than Japanese banks, which are precluded from acting as investment banks. However, the low demand by German firms for external finance means that this power may not amount to much. 6 on average the largest debtholder held almost one quarter of the firm's outstanding debt and over 5% of the firms outstanding equity, while the largest 5 debtholders of the firm held over half of the firm's debt and almost 20% of its equity. In contrast, Clyde (1989) finds in a sample of U.S. Fortune 500 companies that in only a few cases do any of the largest 5 shareholders hold any of the firm's debt and that in these few cases the debtholdings of the largest shareholders are minuscule. Table 4 Gross funding of non-financial corporations, 1970-85 As a percentage of total gross financing Retentions ...... ................ 52 76 External finance .48 24 of which: Intermediated debt 41 21 Securities .7 3 Note: Total gross financing excludes trade credit and some overseas financing. Intermediated debt refers to loans from financial institutions. Securities includes public equity and short and long-term bills and bonds. Sources: OECD Financial Statistics, Part Ill and national data. Marketfor corporate control Another dramatic difference between the U.S. and Germany or Japan is the frequency of mergers and acquisitions. The market for corporate control is much less active in Japan and Germany. Data on the volume of completed domestic merger and acquisition transactions for the second half of the 1980's is displayed in table 5, which reveals large differences between countries. Part of the reason for the greater merger activity in the U.S. is of course the larger number of companies listed on the U.S. stock market. However, even normalising the dollar value of mergers and acquisitions by stock market capitalisation fails to alter the impression that the merger market is much more active in the U.S.--15 to 20 times more so according to table 5.8 The differences across countries in the frequency of hostile take-overs are even more striking.9 Since WWII, for example, there have only been 4 successful hostile take-overs in Germany.10They appear almost as rare in Japan. Kester (1991) claims that the use of take-overs as a device for replacing inefficient management in large Japanese firms is very infrequent. Conversely, in the U.S. almost 10% of the Fortune 500 in 1980 has since been acquired in a transaction that was hostile or started off as hostile. 8 Other empirical evidence supports the claim that mergers and acquisitions in Japan are far lower than in the U.S.. Kaplan (1993a) finds that just over 2% of his sample of large Japanese firms are taken over or merged between 1980-89 in contrast to over 22% of his sample of large U.S. firms. 9 Hostile take-overs are thought to be motivated by corporate control considerations to a greater extent than friendly mergers. I10 See Franks and Mayer (1993). 7 Table 5 Average annual volume of completed domestic mergers and corporate transactions with disclosed values, 1985-89 Volume (in billions of USS).. 1,070 61.3 4.2 As a percentage of total market capitalisation ............ 41.1 3.1 2.3 Dollar values calculated at current exchange rates for each of the five years covered. Market capitalisation figures are for 1987, converted to dollars at prevailing exchange rates. Sources: For the United States and Germany, Securities Data Corporation, Mergers and Corporate Transactions database; for Japan, Yamaichi Securities Corporation, as reported in Beiter (1991). B. The Role of Banks In Corporate Governance in the West In the U.S., the banks' primary role in the corporate governance mechanism is through their role as lenders. Even here the equitable subordination fear may make banks reluctant to play a big role in the governance of the firm. They typically own no equity in the firm, are infrequently represented on the board, and maintain only arms-length relationships with their borrowers. Their main governance tool appears to be covenants they insert in the loan contract and the maturity they set for the loan. This does not mean however that banks do not exert control over firms in the U.S. when the law allows them to do so. For example, in firms that file for bankruptcy or restructure their debt privately, bank lenders and other financial institutions with debt outstanding to the firm frequently receive significant blocks of voting stock, appoint their representatives to the board of directors, and insert restrictive covenants in the companies' restructured lending agreements to give them more say in the firm's investment and financing policies (see Gilson (1990)).! 1 U.S. firms that undergo financial distress thus appear to take on some of the characteristics that the typical Japanese and German firms exhibit--notably, high ownership concentration, large equity and debt stakes held by banks and bank representation on the board of directors. In Japan, the corporate control framework is related to the "keiretsu" form of corporate organisation, where a group of firms based in different industries are centred around a bank. The nonfinancial firms in the group tend to have product market links among them, and take small equity stakes in each other. They all tend to have strong but not exclusive borrowing links with the banks and other financial institutions in the keiretsu, who also take large equity stakes in them. While individual banks will take stakes of less than 10% in firms, collectively banks will own about 20% of the firm, and banks and life insurance companies together may own around 35-40%. Ownership concentration is relatively high, as is the concentration of debt claims. Collectively, banks are the most important large shareholders of firms and until recently have also been their only major source of external finance. Consequently, they have had a potentially very powerful position as active monitors of management either through the board or more informally through the Presidents Club 11 U.S. law allows banks a significantly more active role in the governance of the firm once it defaults on loans owed to the bank. 8 meetings, and by controlling the firm's access to external funds. Banks monitor firms on a continuous basis in the sense of having regular and substantive discussions with management on policy, and actively intervene in the case of financial distress. The market for corporate control among large firms is inactive. Hostile take-overs and other transactions between firms involving corporate control changes are rare. This does not mean that management turnover in response to poor performance is infrequent, merely that it occurs by other means, namely pressure from banks. In Germany there appears to be more reliance on nonbank direct shareholder monitoring than in Japan. Ownership concentration is very high among large firms, high enough to give the large shareholders large incentives to monitor management.12 Banks generally take small stakes directly in firms (with some important exceptions such as Daimler-Benz in which Deutsche Bank has a 25% stake) but monitor and exert control through proxy votes, a pyramid structure of holding companies and representation on the supervisory board (often in the chairman position). Banks appear to have the potential to engage in monitoring and influencing management particularly in the diffusely held firms where their control of voting rights may be important. They appear to have much less control (compared to Japanese banks) over firms through their control of external sources of finance, since German firms rely more on internally generated funds. Hostile take-overs are almost non-existent. The success of the Gernan corporate control mechanism appears, in brief, to stand on the ability of large shareholders, often banks, to monitor management. C. Costs and Benefits of Banks Having a Primary Role in the Ownership and Governance of Firms The academic literature suggests a number of potential advantages, from a corporate governance perspective, of a system that allows large equity and debtholders of the firm to be the same agents, that encourages the concentrated holding of debt and equity claims, and that restricts firm's sources of external finance to one investor.13 The legal and regulatory environment of the German and Japanese economies has encouraged this concentration of firm's claims in the hands of banks to a much greater extent than that of the U.S.. It has also given the banks--in Japan in particular--a potentially very effective monitoring role by allowing them to control much of the firm's external financing. A priori then, one might expect that the German and Japanese model would encourage a more efficient form of corporate governance than that in the U.S.. Recent research provides some empirical evidence on this issue. Prowse (1990) finds that the tendency for financial institutions to take large debt and equity positions in the same firm may mean that the agency problems of debt finance are lower for firms in Japan than in the U.S., where institutional investors are prohibited from being large debt and equity holders in the same firm: leverage ratios are higher in Japanese firms than U.S. firms partly because of their lower agency costs of debt. Hoshi et al (1990a) find that Japanese firms that are members of keiretsu experience fewer liquidity effects on investment than those firms that are not members. Keiretsu members typically have strong ties to a main bank, which along with other group firms, takes a significant equity position in the firm and holds a large fraction of the firm's outstanding debt. They conclude that membership of a group and close ties to a group main bank are important in mitigating the information problems that are typically associated with external finance and governance. Lichtenburg and Pushner (1993) provide evidence that suggests that financial institutions are the major monitors of 12 This of course assumes that the monitors are themselves value-maximisers which may not necessarily be the case. The issue of "who monitors the monitor?" is addressed in the following section. 13 See, for example, Stiglitz (1985). 9 firm behaviour in Japan. Japanese firms with high financial institution ownership show higher levels of productivity and profitability than other Japanese firms. 14 There is less empirical evidence available on Germany. Cable (1985) provided until recently the only comprehensive analysis of German firm performance and its relation to the tightness of its bank ties. He provided evidence that banks do play an important corporate control function: profitability among the large German firms was positively related to the proportion of equity voting rights controlled by the 3 big universal banks, and by bank representation on the board of directors. 15 He concluded that banks did use their control of voting rights and their presence on the board to monitor and influence management towards profit maximisation. Elston (1993), in a study modelled on that of Hoshi et al (1990), finds evidence that firms with tighter bank ties in Germany exhibit investment functions which are less sensitive to liquidity constraints than firms with weaker ties, suggesting that bank ties are important in mitigating the information and agency problems associated with extemal finance and governance.16 Overall, the general consensus of the existing empirical literature is that strong bank-firm relationships may significantly reduce agency and information costs and improve efficiency. However, there are clearly some potential disadvantages of allowing banks to own equity and to take a primary role in the governance mechanism of nonfinancial firms. One potential disadvantage is the conflicts of interest such arrangements may generate within banks. Saunders (1994) identifies a number of concerns regarding such conflicts of interest. The most important is that the bank may have incentives to make either subsidised or unsound loans 14 They also analyse how financial institutions exert their control over firms in which they have large stakes. They distinguish between two methods of monitoring: continuous direct monitoring of the firm versus significant intervention in the firm's business only after the firm has encountered financial difficulties. They find that high financial institution ownership reduces both the severity and the frequency of lapses from efficiency and profitability. This would suggest that, in some contrast to the conventional wisdom on how Japanese financial institutions behave towards the firms which they own, financial institutions conduct both types of intervention. The conventional wisdom has been that banks use their monitoring and disciplining role only in times of financial distress. See Sheard (1989). 15 The 3 big universal banks are Deutsche Bank, Dresdner Bank and Commerzbank. 16 Edwards and Fisher (1994) take issue with much of the conventional wisdom on the role of banks in the corporate governance system in Germany. They present a variety of evidence suggesting banks may not have the degree of influence as lenders, shareholders, voters of proxies, or representatives on the board as has been widely thought. They find that bank control of voting rights is only weakly related to the number of bank representatives on the supervisory board, a result somewhat at odds with the belief that banks use their control of voting rights to further their control over management. (Though not completely inconsistent if banks use their votes to elect members to the supervisory board who are sympathetic to bank interests even though they are not themselves bankers--a distinct possibility in Germnany where there has been a tradition of criticism of the banks for exerting undue influence over firms, and where there may be a strong incentive for banks to conceal their influence.) They also find that bank representation on the supervisory board does not mean that the firm borrows more from banks, again something that might be expected if banks used their board membership to reduce the costs of providing external finance to the firm. As Edwards and Fisher admit, the evidence provided both for and against the importance of banks in the corporate control function of German firms is very incomplete. Probably the most that can be said about the issue at the moment is that while it is likely that banks in Germany probably play some role in the corporate control of firms, it is unclear that they play the primary role. The structure of corporate ownership in Germany presented earlier suggests that nonfinancial firms and individuals that are large shareholders may play an equally important role as banks. Indeed, Edwards and Fisher's results may be a product of the fact that banks play an important corporate governance role only in those firms that are widely held and that do not enjoy the benefits of large shareholder monitoring. The Monopolkommission's (1978) findings that banks' voting power from direct holdings and from proxies is greatest in those firms which are widely held is consistent with this view. 10 to enterprises in which it has an equity stake. Such practices might ultimately threaten the safety of the bank itself. 17 Another potential disadvantage is that allowing banks to invest in equity as well as debt might lead to an increase in the risk of bank failure. By their very nature equity investments involve more risk than do loans. Allowing banks to take equity stakes in firms could therefore increase the riskiness of the banking system and the potential cost to the deposit insurance system. Of course it is not clear whether expanding the permissible asset class for bank investments will result in an overall increase or decrease in bank risk. Allowing banks to invest in equity could conceivably lower overall bank risk by allowing the bank more avenues for diversification and opportunities to hedge their loan risk more effectively. Although this issue has not been addressed directly by researchers, it has been addressed in the context of empirical investigations into the possible effect on overall bank risk of the introduction of universal banking in the U.S. (see Saunders and Walter (1994)). The bulk of the research on this issue suggests that in fact there are potential risk reduction gains from allowing commercial banks to expand into universal banking activities. However, these empirical results are very hard to translate into implications for banks in the transition countries with their drastically different conditions and environment. 18 A final possible disadvantage in allowing banks to own equity in firms stems from the power a large equity holding confers on a bank to influence the firm to take out loans from the bank (or purchase other bank services) at premium prices. The bank may thus be able to extract rents from the firm by virtue of its large equity stake in it. Here the equity stake held by the bank may frustrate the ability of the firm to take advantage of a competitive loan market. This problem may be all the more important in banking systems where there is not much competition in the first place. Empirical evidence about the importance of these potential disadvantages of allowing banks to own equity in firms is scarce. The Gessler Commission (1979), in its study of universal banking arrangements in Germany, concluded that potential conflicts of interest in universal banking probably were only responsible for minor restraints of competition that could be remedied by marginal changes in the law. It appears that most of the potential disadvantages mentioned above can be substantially mitigated by a combination of a high degree of competition in the banking and financial sector and a sophisticated and intensive prudential supervision and regulation of banks. Absent such an environment, of course, these arguments are likely to take on a much greater importance. In particular, they are likely to be far more relevant for the transition countries D. Requirements for the Efficient Operation of the Different Systems In analysing the relevant lessons for eastern Europe and Russia, a clear understanding of the conditions necessary for the successful operation of each system observed in the west is needed. What are the requirements of the different systems in the west in terms of liquid capital markets, the severity of the information asymmetries between insiders and outsiders, the independence and governance of the banking system, the degree of competition within the banking system, and the type of monitoring 17 Other potential conflicts of interest involve i) the bank restricting the supply of credit to an enterprise that is a competitor with an enterprise in which the bank has an equity stake (an "affiliate"), ii) the bank passing on privileged information about a borrower to competing affiliates, and iii) the bank using its lending powers to tie customers in to the products of affiliated enterprises. 18 This is partly because the introduction of universal banking involves a very wide expansion of bank powers-- into insurance and corporate securities underwriting--in addition to the holding (and trading) of corporate equity securities. Thus the results cited here probably do not have much to say on the narrower issue of the effect on bank risk of the holding of equity securities. In addition, and perhaps more important for the focus of this paper, banks in the U.S. are completely different entities from their namesakes in the transition countries in terms of their sophistication, the reputations they have at stake, and the supervisory and competitive environment. 11 and restructuring that firms require? The requirements of each system can then be compared with the existing conditions in eastem Europe and Russia and some implications for the operation of each system under these conditions can then be drawn. Capital market liquidity The Anglo-Saxon system of corporate govemance relies crucially on liquid securities markets which can transmit new information rapidly and efficiently and which come close to being efficient markets in the sense that there is a large amount of publicly available information about the financial and operating characteristics of firms which is incorporated into prices. Without such conditions, hostile take-overs become virtually impossible. In contrast, the German or Japanese system of large shareholder monitoring by banks and others does not rely at all on liquid corporate securities markets and may even be hindered by their presence. The Anglo-Saxon requirement for capital market liquidity means a requirement for a large amount of publicly available information about firms so that the information asymmetries between insiders and outsiders are not severe. If they were, outside investors would never risk their money by buying corporate securities, and a corporate securities market would consequently be very illiquid. It is not a coincidence that corporate disclosure requirements in the US are much more comprehensive and demanding than those in Japan and Germany (see Prowse (1994)). Capital market liquidity not only requires large amounts of information be available to outside investors, but also that there be sufficient expertise among outsiders (investment banks, ratings agencies, buyout firms, and take-over specialists) to analyse, interpret and understand the financial and operating information disclosed by firms, so as to evaluate whether a firm is being run badly or not. In the German and Japanese systems, this expertise is also required but need only be concentrated amongst the large shareholders who, as insiders, have access to the appropriate information and do most of the monitoring. There is thus very likely to be a lower requirement for such expertise overall in the Japanese or German system compared to the Anglo-Saxon system. Some authors claim that the role of the large, active shareholder in Germany and Japan is buttressed by or may even require the existence of illiquid securities markets. One reason proposed for this is that illiquid equity markets preclude the possibility of large equity blockholders simply selling their stakes in firms in which they become dissatisfied ("exit"), thereby increasing the incentives for large shareholders to monitor and influence management ("voice") (see Coffee (1991)). Another reason may be that in undeveloped capital markets firms often have only one source of extemal capital - bank loans. This sole reliance on banks for extemal finance may strengthen banks' corporate governance role by giving them control over the cash raised by the firm to finance activities (see Scharfstein (1992) and Prowse (1994)). There is some evidence - particularly from Japan - that the existence of illiquid securities markets contributes importantly to the power of large shareholders to monitor the firm and indeed that they may be a necessary condition for the large shareholder system to operate efficiently. For example, in Japan corporate securities markets until the early 1980's were very illiquid. In addition, banks have traditionally been very important providers of extemal finance to firms through their role as lenders. As capital markets have been deregulated in Japan and firms have had access to a wide variety of sources of finance apart from bank loans, their reliance on bank loans has fallen sharply. If the banks' role as corporate monitors depended crucially on their role as providers of extemal finance to firms we might expect that their effectiveness as monitors may have declined recently. Some evidence suggests that this is indeed the case. Hoshi et al (1990b, 1993) provide evidence of agency problems among Japanese firms, the source of which appear to be the increasing accessibility to nonbank finance. They find that despite the advantages they document for Japanese firms from tight bank relationships in terms of less information problems and liquidity restraints, many Japanese firms were actively reducing the strength of their bank ties in the 1980's as deregulation led to many 12 opportunities for external finance apart from bank finance. They analyse the firms that have increased their use of the public debt markets and reduced their reliance on bank loans. They find that more profitable and successful firms, as well as less successful firms in which managers have significant share ownership, have been the most aggressive in reducing their reliance on banks.19 Their findings indicate that the corporate control mechanism in Japan may have ceased to operate as effectively as it had done over the period when the legal and regulatory environment precluded firms from tapping nonbank sources of extemal finance. In Germany, the story is somewhat different. The German financial system has typically been characterised by illiquid and shallow securities markets. This may have induced large shareholders to exert voice rather than exit. However, as mentioned earlier, despite the absence of many alternatives to bank loans German firms have never relied heavily on banks to finance their investment needs. What governance power the banks have does not appear to stem from any control over the firm's access to funds. In sum, while deep and liquid corporate securities markets are clearly a necessary condition for the operation of the Anglo-Saxon system of corporate governance, there is some evidence that they may be a hindrance to the operation of the Japanese model. This is clearly of importance to eastern European countries which currently have very underdeveloped capital markets and are considering whether to devote resources to making them more liquid. Governance of banks The German/Japanese system requires that the institutions that are the primary monitors perform their monitoring function properly. How is this ensured? In other words, who monitors the monitor? In most German and Japanese banks, there appears to be no large shareholder to perform such "monitoring of the monitor". Banks in both Germany and Japan are typically widely held (see Prowse (1994)), suggesting that this may be an important issue in both these countries. Diamond (1984) has demonstrated that under certain conditions, banks can alleviate problems of monitoring the performance of enterprises and financial institutions. By investing in a large number of firms, banks smooth their earnings and are able to offer investors fixed return debt (deposit) contracts. By issuing debt, managers of banks are provided with incentives to maximise the performance of the firm. However, there are a number of real world complications that may cloud the applicability of Diamond's result to the German and Japanese banks. The most important is the existence of deposit insurance, which removes depositors' incentives to penalise banks, through higher interest rates, if the bank does not monitor efficiently. The monitoring function may also be performed in part by the banking supervision authorities. Japan and Germany have fairly intensive prudential supervision and regulation frameworks with which to monitor and control banks. Admittedly, the primary focus is to protect the government from failure of the bank rather than the maximisation of bank equity value, which is what would ensure optimal monitoring behaviour by the bank. However, adequate supervision would presumably mitigate one of the conflict of interest problems that Saunders (1994) identifies--namely 19 They interpret these findings in the following way: First, because intermediated finance involves a significant amount of monitoring by the bank, it has costs that successful firms with good investment opportunities would prefer not to bear. Thus when the opportunity to tap public debt markets arises, these firms react in a profit maximising way by tapping these cheaper forms of finance. However, managers of all firms have incentives to tap such finance, because they will then be subject to less scrutiny and monitoring from stakeholders and will have more opportunity to shirk. Thus firms in which managers are entrenched may also be expected to reduce their reliance on banks and finance themselves through the public markets. In this case their response is not the profit maximising one, but the one that maximises the managers' preferences. Hoshi et al's evidence point to there being some degree of management entrenchment in Japanese firms. 13 that banks would not easily be able to make unsound or subsidised loans to affiliates in financial distress. Ultimately, whether or not the "who monitors the monitor?" issue is a problem for Germany and Japan is a matter of empirical evidence, of which we have little to date. However, it is clear that the German and Japanese model work more efficiently the greater are the incentives for the banks to perform efficient monitoring of the firms they own and lend to. Regardless of how well banks monitor in Germany and Japan, this issue clearly acquires great significance for the countries of eastern Europe, where the issue is likely to be important since banks are still state owned, suffer from incentive problems, and are subject to fairly weak prudential supervision and regulation by the banking authorities. Competition among banks Given the potential for conflicts of interest in a system that allows banks to own equity in firms, the greater degree of competition in the banking system, the less this potential is likely to be a reality. For example, many of the conflicts of interest that a bank might be susceptible to would be mitigated to the extent that competitive market conditions imposed external discipline--through the loss of reputation--on the bank. Competitive conditions in the banking system would for example, make product market tie ins impossible for a bank to insist upon. Similarly it would make it impossible for a bank to threaten to disrupt credit flows to the competitor of an affiliate. And the incentive to pass on private information to affiliates about their competitors who borrow from the bank would be mitigated by the threatened loss of reputation. One reason these problems have not appeared to be serious in Japan and Germany may well be because their banking systems are reasonably competitive (see Corbett and Mayer (1991)). There are large numbers of banks and concentration levels tend to be low. For example, the 13 largest city banks in Japan held just under half of all banking assets in 1986, and in Germany the 6 largest banks accounted for less than 10% of all banking assets. However, this issue is clearly of importance for eastern Europe, where the concentration of the banking system is much higher. Restructuring problems of enterprises The most effective system of corporate governance for any particular firm is likely to depend on its line of business, its stage in the life cycle, and the technological and market conditions it faces. For example, old established firms in basic industries suffering from large over capacity problems that need to downsize by selling off assets, cutting capacity, laying off workers and cutting wages may be subject to a particular set of agency and information problems that are quite different from a young firm in an expanding industry that needs to invest large arpnunts of capital in research and development in order to stay competitive. These different agency and information problems can plausibly be addressed more efficiently by particular corporate governance mechanisms rather than others. What can we say about how corporate control arrangements are suited to different problems of adjustment and restructuring? First, there is evidence that hostile take-overs are a particularly effective mechanism of achieving the painful restructuring required of a firm in an industry with large over capacity problems. Shleifer and Vishny (1989) found that U.S. firms suffering from such problems were more likely to undergo a corporate control change through a hostile take-over than through internally precipitated removal of management by the board of directors. This may be because hostile take-overs involve the intervention of an outsider who has no previous relationship with the managers and employees of the firm, and who thus has more freedom to enact the painful measures necessary for maximisation of equity value (see Shleifer and Summers (1990)). Conversely, firms which are governed by a large institutional shareholder (bank or otherwise) may plausibly not be as 14 responsive to over capacity conditions by virtue of the long-term relationships that they have built up with other stakeholders (such as management and workers) over time.20 Miszei (1994a) claims that the reliance on the impersonal take-over mechanism has meant that firms in the U.S. in industries characterised by over capacity have restructured and downsized themselves much more quickly than firms facing similar requirements in Japan and Europe. Second, there is evidence that the corporate governance arrangements we observe in risky start up firms may be a response to their particular needs. These needs are typically for an active investor that brings with him not only capital but also an expertise in the marketing, accounting and financial aspects of the firm's operations. Sahlman (1991) and Fenn et al (1994) describe the corporate governance arrangements of venture capital firms in the U.S. as being dramatically different from those of large firms. Venture capital firms typically have an active investor (the venture capitalist) who takes a majority equity stake in the firm, sits on the board, finds other sources of capital for the firm if necessary, and provides marketing, financial and technical expertise to the firm. This large shareholder role seems particularly suited to mitigating the problems facing risky start up firms where the original management team may have plenty of technical expertise but little or no business expertise. Third, Franks and Mayer (1992) speculate that there are certain lines of business that are likely to benefit more from the Japanese or German style of corporate governance and others that benefit from the U.S. style. They suggest that the U.S. system is best suited to situations in which corporate activities involve subjective assessments of future prospects and evaluation of the quality of employees and managers is of little relevance, on the assumption that the market for corporate control is really a market for corporate strategy with managers competing for the rights to implement their preferred strategies. They assert that this function is quite different from the one performed by the German or Japanese system which relies on banks to be active investors. The primary role that the main bank or Hausbank plays in corporate governance along with other interested stakeholders (such as suppliers and customers) ensures that the influential parties have a lot of information about the quality of management and employees. Thus the German or Japanese system is particularly well suited to evaluating managerial performance in industries where there are standard operating procedures and known skills that can be evaluated. Franks and Mayer associate the former type of activities where the U.S. system is preferred with more speculative risky industries such as biotechnology and pharmaceuticals. The German or Japanese system may be superior in lines of business associated with standard methods of production where there are known skills that can be transferred from one firm to another, such as basic manufacturing. It may be no accident that the comparative performance of different countries in different industries appears to be related to the advantages of alternative corporate governance systems.21 What implications does this have for central Europe and Russia? Ultimately it depends upon the view one takes of the nature of the restructuring activity facing the firms in these countries. The restructuring requirements of firms in eastern Europe and Russia are likely to be quite diverse, with a large number of firms requiring fairly radical downsizing or even liquidation, but also a significant number requiring expansion and investment in new lines methods of production and new lines of business. In almost all cases there will be a need for the injection of expertise in marketing, accounting and financial operations. In relatively few cases will there be firms that already have invested in the most up-to-date operating procedures and require little or no radical restructuring. 20 Although, as Shleifer and Summers point out, these long-term relationships may be of some significant value in many firms in market economies. 21 Which way the causality runs is of course unclear: the German system may have evolved as a response to its comparative strength in manufacturing, not the converse. 15 Ill. Exlsting conditions In central Europe and Russia The U.S., Japan and Germany present us with three different models of the governance role of banks. These models have unique requirements in terms of the conditions under which they can operate efficiently. These conditions comprise the degree of liquidity in the capital markets, the governance and supervision of banks, the degree of competition within the banking system and the particular restructuring needs of firms. These conditions are now examined for the countries of eastern Europe and Russia. A. Business Skills, Capital Market Liquidity, and Firm Restructuring Requirements Lack of business skills andfinancial expertise Perhaps the most serious legacy of central planning is the acute lack of business skills and expertise throughout all firms, banks included. Among nonfinancial firms there is an urgent need for marketing, accounting and financial skills. Many managers need to acquire these skills or be replaced. Among banks, there is a need for loan evaluation skills, which require accounting and other financial expertise. Under central planning these skills were not needed. As far as banks were concerned, the rudimentary banking system was in effect part of the budget. Its role was to distribute funds according to the plan and to monitor consistency of expenditures with the plan. Much of the activity of the monobank and the sector-specific specialised banks revolved around keeping accounts and extending working capital credit in accordance with plan targets. In short, banking then was equivalent to book keeping by a monopolist.22 It is not surprising that banking skills like evaluating creditworthiness, pricing risk, loan-vetting procedures and offering products tailored to customers' preferences in order to attract funds did not develop in such circumstances. Information asymmetries between firm insiders and outsiders Firms in the transition countries suffer from an acute asymmetric information problem between insiders and outsiders, in particular as the systemic transformation radically altered the economic environment in which enterprises were operating. Changes in relative prices, the collapse of CMEA export markets, sudden import competition, reduced subsidies and the sudden sharp increase in real interest rates made a large number of existing enterprises unprofitable while others found their profit situation improved. This debased the value of accumulated information on enterprises, including that held by the banks. In addition the lack of reliable enterprise accounting and the absence of auditing reduced the amount of information outsiders could gather on companies. Uncertainty surrounding the ownership situation and the legal framework has made the situation more difficult still. The level of uncertainty is illustrated by the fact that in cases where independent auditors from the West were called in to evaluate the value of a firm in order to set a sale prices, their estimates often differed by several hundred percent. Capital market liquidity Severe information asymmetries and the lack of expertise in the transition economies in evaluating information about firms has meant that securities markets have been very slow to develop. Stock markets (with the exception of the Czech and Slovak Republics) are small, consisting of perhaps the largest 20 to 30 firms in the country. Corporate bond markets are almost non-existent. 22 Sometimes enterprises did not have their own accounting section and relied mostly on the account-keeping services of banks. 16 Further, the institutional and regulatory infrastructure that can provide for the growth of corporate securities markets is currently lacking in these countries. Bond rating agencies, investment advisory firms and securities regulatory bodies are in the very early stages of development if they exist at all. This suggests that it will be a long time before corporate securities markets in the transition countries approach the breadth and liquidity of even the Japanese and German securities markets, let alone those of the Anglo-Saxon countries. Restructuring requirements offirms The restructuring requirements of firms in eastern Europe and Russia are very diverse. A large number of firms will have to undergo radical downsizing and even liquidation. Others must be relieved of their debt burden and restructured. In almost all cases there is an urgent need for a complete reorientation of the management of the business. This will require the replacement of management in many cases. As Phelps et al (1993) point out, the need to replace old management with new is hard to underestimate in the transition countries, where the old managers have a skill set appropriate for success in a command economy but in many cases completely inappropriate for success in a market economy. In other firms, the infusion of new funds will be necessary to invest in new technologies and new lines of business. Finally and perhaps most important, there will almost always be the need for additional expertise in marketing, accounting and finance.23 Obviously all these elements are linked. Firms that are well managed will have better access to external finance and will more quickly attain the necessary business skills. In practice, however, marketing and financial skills may often be better imported through foreign trade partners or investors. In sum, there appears to be a premium in the transition economies on a corporate governance system which can effectively fire unsuitable managers and that involves the active investor contributing to the firm's expertise in financial and accounting skills. B. Conditions in the banking sector Non-performing loans The appearance of a large portfolio of non-performing loans was to be expected once the reform of eastern European economies was under way24. As a result of market-oriented reform and the shocks stemming from the collapse of CMEA export markets, a large number of existing enterprises became unprofitable. Not surprisingly, a substantial debt overhang developed: many firms could cover operating costs but were not able to service their debt. Banks were saddled with a large stock of non-performing loans and often found themselves with negative net worth. 23 The Czech economy provides a vivid example of this point. Arguably, the Czech's have advanced the furthest in addressing problems of governance and access to external finance by enterprises. Yet, some success stories notwithstanding, a large number of firms still languish, suffering from an acute lack of marketing and financial skills. 24 This and the following paragraphs draw on Dittus (1995), "Bank reform and behaviour in central Europe", Joumal of Comparative Economics, forthcoming. 17 Chart 1 Non-performing loans (1992) 30 25 20 .~15 10l 5 Czech Republic Hungary Poland Estimates for non-performing loans at the end of 1992 are shown in chart 1.25 It suggests that about a third of loans were non-performing, although variations between banks are considerable and average figures have to be interpreted with caution. Many banks were weighed down with large amounts of non-performing loans. The undesirable incentive effects resulting from a low net worth of banks are well known: in many cases they lead to opportunistic behaviour by bank managers and owners. Banks have an incentive to "gamble for resurrection" by making risky loans or may engage in activities which benefit the management and owners at the expense of depositors or, where these are insured, the taxpayer. The disincentive effects of a large debt overhang on firms are also well known. Enterprises are preoccupied by the debt burden instead of focusing on survival in the new market environment.26 The aim of financial restructuring is to eliminate the disincentives stemming from past lending. Bygones should be treated as bygones and should not influence current and future lending decisions. Countries have addressed problem loans and the resulting weak capital bases in various ways and have sometimes changed their approach over time.27 Broadly speaking, two approaches can be distinguished. The central European countries have recapitalised their existing banks to such a degree that by mid-1994 the portfolio problem of most banks could be considered as being under control (Dittus (1995)). In the Czech Republic, most of the non-performing loans were carved out of the commercial banks in 1991. This, together with an additional injection of fresh capital, brought the 25 Non-performing loans in banks' portfolios are not easily identified because of inadequate accounting standards, the only partial availability of portfolio reviews, and tax rules which discouraged provisioning. Moreover, once many enterprises are illiquid or insolvent they "contaminate" the economic environment if they are allowed to remain in operation because bankruptcy rules are either weak or not enforced. As they do not pay suppliers, these in turn cannot pay their suppliers and as payment arrears spread so does the difficulty of distinguishing good from bad enterprises. 26 In addition, a debt overhang makes it very difficult to borrow new funds, even for clearly profitable projects. Another risk is that borrowers may not behave in a prudent manner as long as they expect a bailout in the future (moral hazard). 27 This paper does not summarise or discuss the respective merits and drawbacks of the different proposals which have been advanced on how to deal with problem loans; there are numerous suggestions in the literature. Early papers were by Hinds (1990) and Brainard (1991). Convenient summaries do exist, for example in Schmieding and Buch (1993). Neither does the paper describe in detail the measures countries have taken to tackle the problem of non- performing loans; these are analysed in Dittus (1994). 18 capital-to-asset ratio of the major banks to 4.5 in 1991. In several rounds of recapitalisation, the capital-to-asset ratio of Hungarian banks was brought to 4% by mid-1994. Through the issue of sub- ordinated debt a further increase to 8% is planned by the end of the year. In Poland, the nine commercial banks which took over the former monobank's assets in 1989 were recapitalised to more than 8% in 1993.28 In contrast, in Russia the problem of non-performing loans has been reduced through hyperinflation (Claessens and Pohl (1994)). The broad money supply contracted from 80% of GDP in 1990 to 16% in 1993 and credit to the enterprise and household sector shrunk to a mere 12% of GDP (chart 2). Even if problem loans as a share of total loans were large, their size in relation to GDP would be modest. Moreover, the contraction of the overall balance sheet of the banking system was accompanied by a large change in its structure. The share in total assets of the formerly dominating specialised banks has declined to about a third. For example, Sberbank, which continues to hold most of the country's savings, now accounts for only 5% of the total29. Two thirds of total banking assets are held by the about 2000 newly-created banks, of which about 20 to 30 are relatively large (chart 3). In sum, both in central Europe and in Russia past problem loans have been provided for and therefore are no longer the most important issue. However, uncertainty with respect to the likelihood of future bail-outs remains and continues to provide many banks with bad incentives (see below). Chart 2 Financlal disintermedlatlon In Russia 80 70 60 X 50 . 40 Broad money 0 a 30 Credit to enterprises and households 20 10 0 1990 1991 1992 1993 State ownership and influence At the beginning of the reforms all banks were owned by the state. State-owned banks are generally bad at providing effective corporate control because of political interference in lending decisions and the direct dependency of management on government. Even in the absence of direct interference state banks are likely to give more weight to employment and political considerations than to narrow financial returns. For banks to play a role in corporate control, therefore, their links to government need to be severed through privatisation. 28 However, the situation of some other banks, in particular in the cooperative sector, and that of BGZ and the savings bank remains precarious. 29 Of the old state banks, only Vneshtorgbank (which holds most of the foreign exchange) and Rosselkhozbank (which channels subsidised credit to agriculture) remain of some importance. 19 Chart 3 Distribution of bank assets in Russia (1993) Vneshtorgbank Others 18% 28% Rosselkhozbank 10% Sberbank 5% Promstroibank 20 largest commercial banks 36% Generally, bank privatisation in eastern Europe has proceeded at a slower pace than hoped for. One reason has been that many foreign banks consider the retail markets in eastern Europe still very risky and are content to focus on a few profitable niches - trade finance, credit to joint ventures, and services for high net worth individuals. This can be done easily through the opening of a branch or subsidiary without taking a stake in the large state banks, thereby avoiding the daunting task of overhauling bureaucracies which have evolved under central planning. It has also taken much longer to restructure the banking sector than was initially anticipated. Although governments in central Europe have recognised the importance of putting the banking sector on a sound footing, in some cases the difficulties of doing so were significantly underestimated at first. In particular, correctly valuing bank portfolios has proved extremely difficult, and many more banks than initially anticipated appeared to have negative net worth. In other cases, it has proved difficult to implement the necessary measures. The result has been that financial sector reform has lagged, with the exception perhaps of the Czech Republic where balance sheets were cleaned early in the process. Privatisation has proceeded fastest in the Czech Republic. In the first wave of mass- privatisation in 1993, majority stakes in all state banks (with the exception of the Consolidation bank) were distributed to the public. In Poland, three state banks have been privatised through direct sale, and more are awaiting privatisation. At several others political influence has been reduced by insulating the management from ministries through a supervisory board. The laggard in privatisation has been Hungary which privatised the first state bank only this year. The other main banks are still directly owned by the government. In Russia, the old state banks have not been privatised. However, a large number of regional branches of these banks, often on oblast level, have established themselves as independent commercial banks (so-called "spin-offs"), contributing to the rapid growth of the total number of banks. Competition Competition in the banking sector has increased markedly in most countries. In Russia, market entry has been extremely rapid: the total number of banks has risen within a couple of years from a handful of state banks to about 2400 banks. Most of these are small and many are owned by 20 one or a few enterprises which use them mainly as cheap sources of credit.30 Nonetheless, about 20 to 30 banks appear to have developed into well-managed commercial banks. Within a few years these banks have come to dominate the commercial banking market while the role of the old state banks which are not involved in the distribution of directed credit has shrunk (see chart 3). Market entry has been rapid in the Czech Republic and also, though to a lesser degree, in Hungary. The number of banks with foreign participation (including joint-ventures, subsidiaries and branches) has increased substantially in these countries (chart 4). The new banks have contributed to heightened competition. In contrast, the issuance of new banking licences in Poland has been very restrictive.31 Chart 4 Number of banks with foreign participation 35 30 25 20 15 10 5 0 1990 1991 1992 1993 | Czech Republic 0 Hungary * Poland Despite the entry of new banks, the "old" banks, whether privatised or not, continue to dominate the lending and deposit business. In the Czech Republic and Poland, the old banks hold more than 80% of loans and deposits and in Hungary about two thirds. Only in Russia do new banks dominate the scene, accounting for two-thirds of the banking system's assets. Another element of competitive pressure, particularly in the Czech Republic and Hungary32, has been the access of enterprises to direct lending by foreign banks and of households to instruments denominated in foreign currency. The amounts borrowed by enterprises directly from abroad have risen substantially in the recent past and are likely to exceed $1 billion in the Czech Republic and Hungary. On the deposit side, foreign currency deposits have become available to households. They have used their newly-gained freedom to diversify their portfolios by increasing their holdings of foreign exchange accounts with domestic banks. As a result banks have to offer interest rates on domestic currency deposits which, after allowing for devaluation expectations, are competitive with those offered on foreign exchange accounts. 30 Many of these enterprise-owned banks appear to be plagued by bad loans. Reportedly, fraudulent and even criminal behaviour has been quite comrnmon. 31 After more than 50 new banks, mostly small and undercapitalised, had sprung up in the wake of reforms and presented the authorities with difficulties of supervision, the Polish authorities all but stopped issuing licenses. 32 Until recently, enterprises in Poland found it difficult to tap foreign sources of capital owing to the arrears on foreign debt. After a settlement was reached with commercial creditors in 1994 direct foreign lending and investment has tended to increase. 21 Overall, competition in all countries has increased substantially. In Russia, the seeds of a competitive banking market are apparent as a small number of new banks make up a large part of total lending; however, many of the 2400 banks have a captive market in the enterprises by which they are owned. In the Czech Republic, competitive forces in banking are quite strong, due both to the rapid entry of new banks and the access of many companies to foreign loans. Such access also plays a role in Hungary. In particular joint ventures between domestic and foreign companies have access to foreign finance, and competition to lend to these enterprises is fierce among the smaller, private banks. In contrast, the market for loans to large companies is dominated by a few banks and is much less competitive, partly due to the fact that many of these companies need restructuring and lending to them at present is not attractive. There is even less competition in Poland. New private banks have not played a major role, and until recently entry by foreign banks has remained limited. The market remains dominated by the major state banks. Despite heightened competition in these countries it would appear unwise, however, to rely solely on competitive forces as a disciplinary mechanism to ensure that banks use their control powers over enterprises in a profit-maximising way. Table 6 Elements of the legal and regulatory environment I~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~. ........ .... .. ... .... .. .. .... -....-.................... -hm - -- - l Re , e- ~ ~~~~~. i... . ........... ... ......... - - :-:- ~~~~~~~~.a: .S :1 :a bil -W-f. . ..^-_s A.J ......... Rules for share Ownership of shares is Ownership of shares is Ownership of purchased ownership restricted to a maximum of restricted to a total of 60% shares is restricted to 25% 10% of own capital and 25% of own capital and that of of bank capital but the of capital and reserves. one enterprise (direct and President of the National Limit does not apply to indirect) to 15% of capital Bank of Poland can raise this debt/equity swaps, where for universal banks and 40% limit to 50%. shareholdings have to be for investment banks. reduced to the legal limit Limits do not apply to within 2 years. separately recorded and handled shareholdings arising from debt/equity swaps for a maximum period of 18 months. Exposure rules As proportion of capital: As a proportion of capital: As proportion of capital: individual customers max. individual customers max. individual customers 25%; entities with special 25%; total of large loans (including guarantees) max. relationships (e.g. more than max. 600%. Restrictions on 1 5% and any one loan 10% of shares owned by insider lending. agreement max. 10%. bank): max. 20%; banks in President of the National Czech/Slovak Republic or in Bank can relax these rules OECD area: 80%; ten by allowing banks to count largest debtors: max. 230%. long-term loans towards To be attained by end- 1995, capital and can increase the with transitional provisions. limit to 50%. Bank representation No limitations No limitations No limitations on board of directors 22 Supervision In all countries, a regulatory framework has been put in place that is modelled closely on rules in western countries. For example, a capital-to-asset ratio of 8% or more is now required, and new accounting, loan classification schemes and mandatory provisioning have been introduced33. While bank regulations have been tightened and to a large degree brought into line with international practice, implementing them has not been easy. This is partly due to the difficulties of establishing reporting systems for banks and to the lack of experience of the supervisory agencies. All countries have established such agencies, either in the form of departments of the central bank (Czech Republic, Poland and Russia) or of a separate government body (Hungary). Initially, as was to be expected, the new supervisory bodies experienced difficulties, partly because of the inexperience of staff and the absence of established procedures for off-site and on-site inspection. These difficulties have been recognised, however, and have begun to be addressed in all countries, often with the help of foreign technical assistance. Despite substantial progress, two fundamental problems remain. The first is that despite some foreign assistance, there is still an acute lack of human capital and developed information systems that the supervisory authorities can draw on to perform their function. The second is the credibility of bank supervision. If banks do not or cannot comply with the new regulations, perhaps because the gap between the initial situation and the newly required standards is too wide, bank supervisors may be faced with a difficult choice between regulatory forbearance and the closure of banks. Given the banks' initial situation, many of the new rules could not be complied with immediately. To some extent the countries in question have attempted to deal with these circumstances through transitional provisions which phase in the new regulations over time. In some cases, however, even transitional provisions cannot be complied with. Under these circumstances it is difficult to enforce high regulatory standards. Strict application of the law might require the closure of a large segment of the banking industry or of major banks. Mergers with other banks may provide a partial solution but can work only if potential partners have a strong capital base, which in most cases appears unlikely. The threat to close large banks is not credible because of the large costs associated with it. There is an obvious danger of time inconsistency here. If the regulatory rules are such that banks can comply at reasonable cost to themselves and the economy, then the imposition of such rules can be expected to lead to changed bank behaviour and efforts to comply. The Government and the legislature are then in a position to induce changed behaviour through better regulations and tighter supervision.34 This is not the case when banks can argue that compliance with the regulations would be prohibitively costly or simply impossible. The Government can then be expected to show regulatory forbearance if many banks or a large bank do not comply with the rules, and the banks know this. In practice governments have opted for a mixed strategy to overcome the lack of credibility of the new regulatory regimes. To some extent they have showed regulatory leniency and have extended the period for complying with the new regulations. At the same time, they have helped banks meet the standards in question through measures to improve banks' balance sheets. They have also shown a willingness to merge and even close smaller banks. There is little doubt that supervision has become increasingly tight. But it is also clear that much room for improvement is left, and that bank supervision is not yet very effective in insuring that 33 An overview of bank regulation and supervision is in Group of Banking Supervisors from Central and Eastern European Countries (1992), Report on Banking Supervision in Central and Eastern European Countries, Basle. See also Dittus (1994), section 3. 34 In game-theory parlance they are able to occupy a Stackelberg leadership position. 23 regulations are respected. Currently, the supervisory capacity in all countries under review is not at the level required to provide effective monitoring and supervision for banks. The legal and regulatory framework In central Europe and Russia the current legal and regulatory framework does not show restrictions on bank involvement in corporate control to the extent found in the United States. Banks are allowed to hold shares in non-financial companies and to exercise control rights. However, prudential regulations apply, which are more restrictive than for example those in Germany (table 6). In the central European countries, ownership of shares is restricted to a certain percentage of own capital (10% in the Czech Republic, 25% in Poland and 60% in Hungary). Only in cases where shares are acquired through a debt for equity swap are these rules relaxed for a limited period of time in the Czech Republic and Hungary (see table 6). It is probably not the case that current regulations are a binding constraint on bank equity holdings in nonfinancial firms at present.35 However, they do clearly limit a large scale expansion of such activity. Bank incentives to make sound investments This section examines the behaviour of banks with respect to their investment policies. First it is argued that bank behaviour has changed substantially since the inception of reform, and that the days of large-scale, indiscriminate lending to old customers are over. Secondly, we analyse how strong the current incentives are for banks to make sound investments. Thirdly, we look at banks' strategy with respect to share ownership. This analysis is based to a large degree on interviews with bank managers in Hungary and Poland, but ancillary information on intentions and behaviour in the Czech Republic and Russia is used, too. In the past, bank lending was entirely passive and basically the accounting reflection of the plan. When a two-tier banking system was established, the new commercial banks maintained close links with their former clients, partly out of inertia and partly because of the lock-in effect created by large loans which many enterprise were no longer able to service. Initially, banks often rolled over loans or capitalised interest in order to avoid revealing their own insolvency. Table 7 Indicators of behavioural change Net lending to enterprises (X of GDP) Czech Republic 6.4 2.9 5.5 Hungary 1.6 -2.5 1.6 Poland 8.0 1.1 3.3 Interest rate spread (Q4) Czech Republic 5.9 6.7 7.2 Hungary 4 11.2 8.8 Poland 14.2 15.5 20.7 35 This is despite the fact that probably the lion's share of bank equity holdings in Poland and Hungary are in related financial companies rather than nonfinancial firms. 24 Since 1991, however, bank behaviour appears to have changed. An illustration of these changes is provided in table 7. Net lending by banks to enterprises declined dramatically in this year and even turned negative in Hungary. At the same time, the spread between lending and borrowing rates widened. Tighter lending policies and the increased spread are probably best understood as banks' reaction to the visible emergence of a large portfolio of bad loans, combined with a tightening of accounting standards, bank regulation and supervision. While other interpretations of developments in 1992 are also possible, the explanation that banks altered their behaviour as a reaction to changed circumstances seems to be the most convincing one.36 The National Bank of Hungary has come to this conclusion (Bod, 1993), as has Hrncir (1994) in the case of the Czech Republic. For Poland, the OECD (1995) notes that automatic extension of credit to troubled firms - which dominated in 1990/91 - has clearly come to an end. It would be premature, however, to conclude from this evidence that the incentive problems of banks have disappeared. A sound capital base is a necessary but not sufficient condition to ensure that banks have incentives to make good investments. The expectations of bank managers are critical. If they believe that poor performance will trigger a government bailout in the future with little or no consequences for themselves for themselves, then their incentive to take difficult decisions on existing borrowers and to screen carefully potential borrowers are not strong. If they cannot expect such government support, then the incentive for good performance is much stronger. For outsiders it is difficult to form a good understanding of the incentives banks are facing. Personal relationships and tacit understandings between bankers and government agencies are factors that are difficult to evaluate. Nonetheless, three indicators can be helpful in evaluating the chances of future bailouts. The first one is the degree of government influence and control over banks. The more control ihe government has, the higher is its responsibility for the banks' policies and the more it is likely to support banks that are experiencing difficulties as a result from decisions that were influenced, directly or indirectly, by the government. A second indication is provided by government behaviour in the past. Governments that have bailed out banks in the past without implementing tough measures are likely to be expected by bank managers to do so again in the future. A third indication may be provided by the degree of industry concentration. Often governments have been seen to allow smaller banks to fail, but to extend safety nets under the larger ones (too-big-to-fail). On all three counts the Russian banking system appears to be the least likely to get government support. Government influence in the new banks is almost non-existent. Banks have also not been bailed out in the past as there was no need to do so. Both debts and assets in the Russian economy were wiped out by hyperinflation in the past few years. The share and influence of the formerly large and powerful state banks has dwindled as a consequence. Finally, the degree of industry concentration is low. The failure of one of the larger private banks would hardly lead to contagion effects and therefore the government could afford to let it fail. In the Czech Republic majority stakes of all banks have been privatised37; the remaining share is held by the National Property Fund that provides an insulating layer between banks and the government. State influence is minimal. Government behaviour suggests that the likelihood of future bailouts is small. The recapitalisation of the banking system was done early and was clearly linked to past loans. Banks that got into difficulties later were either closed or taken over by other institutions and management paid for their mistakes by losing their positions. Since the inception of reforms, the banking industry has become less concentrated with the largest bank holding less than a third of total assets. Two or three banks may be (implicitly) covered by a too-big-to-fail insurance, but it is clear that existing management would be fired should such intervention be required. 36 The evidence is analysed in detail in Dittus (1994,1995). 37 Except for the Consolidation bank (the "loan hospital"). 25 Although three banks have been privatised in Poland, the major banks remain state-owned. They are shielded to some degree from government interference through independent supervisory boards and so far the owner of the banks--the Ministry of Finance--has acted to protect bank management from the requests of other ministries. The banks have been recapitalised in the context of a tough restructuring programme, involving major behavioural and organisation changes. Finally, industry concentration is low. However, the regional specialisation means that the failure of one of the major banks would probably be solved through mergers and take-over (as has already happened). The threat of no further bailouts is probably least credible in Hungary. The government continues to own the major banks directly and has in the .past exerted its influence through management reshuffles. Government support policies likewise suggest a certain likelihood of future bailouts should difficulties arise. The government has injected new capital into the banks in several rounds over the past few years, each one coming shortly after the previously one. Bank management was not changed in any of the rounds. The major banks remain too large for the government to consider letting them fail. In sum, in all countries the incentives to make good investments has improved. Differences between countries are pronounced, however. Powerful market incentives seem to be at work at the private larger banks in Russia38 and also in the Czech Republic. Incentives for making good investments are still strong in Poland. Market incentives are probably weakest in Hungary. Banks' attitude to share ownership and corporate control The purpose of this section is to understand how banks are using the leeway they have and to what extent they are interested, from a purely commercial point of view, to acquire share holdings and to exercise active shareholder control. Although there is a wide variation of opinions, depending on the individual bank's history, market position and ownership structure, three themes come out clearly from our interviews with bank managers and officials. i) New, private banks, whether in central Europe or in Russia, do not want to own companies and to play a role in corporate governance through the ownership channel. Four main reasons are given for this. First, private banks consider that an active shareholder role is extremely costly in terms of human capital and management time while the pay-off is low compared to concentrating on the core business of relationship banking. Several banks made it quite explicit that with a staff of a few hundred people and the given skill profile they could not take an active role in restructuring and supervising a company in more than a handful cases. Secondly, they want to avoid the "moral obligation" to extend financing to a company they own in part and perceive a very strong conflict of interest between ownership and independent credit appraisal. Most consider that a very tight Chinese wall would be necessary, so tight indeed that the investment banking part might as well be constituted as a separate legal unit. Thirdly, these banks do not see much synergy between the lending business and corporate control through partial ownership. Several indeed were quite explicit in that they consider commercial and investment banking to require completely different skills and would not want to use their loan officers for restructuring activities. Fourthly, these dynamic new banks consider that their lending and account relationship with customers give them enough information and leverage over companies already. They do not feel that ownership would give them much in addition. 38 To the degree that supervision is weak and legal enforcement lacking, these market incentives may, of course, in some cases very well lead to illegal behaviour. 26 The only exception these banks are willing to make - and this is equally true for the better Russian banks - concern small share holdings of up to about 5% in order to cement a business relationship with a client enterprise, provided the enterprise is requesting the bank to take such a stake. Indeed, in several cases have companies approached the banks and asked that they take up a small stake. In general the position of these banks can be summarised as a strong determination to focus on the core business of relationship banking and not to be distracted by other activities. ii) The large, and partly state-owned, banks see a role for themselves in restructuring and corporate control to the degree that it is necessary to recover their loans, but only to this degree. Both in Hungary and Poland, the recapitalisation programs force banks to take a more active role in the restructuring of insolvent clients. Thus banks have little choice. It is interesting to note, however, that even these banks do not usually actively seek taking equity stakes. First experiences with work- out units has taught them how skill-intensive and demanding an active role is, and the general conclusion seems to be to avoid being dragged too deeply into active shareholder roles. The behaviour of banks under the financial restructuring law in Poland provides perhaps the best illustration of these points. This law obliges recapitalised banks to come to agreements with their defaulting debtors. They have three options: they can swap debt for equity, write down and reschedule loans in a Chapter I I-type procedure which does not involve the courts, or sell non-performing loans at a discount on the secondary market. Banks resorted to debt-for-equity swaps in about 30% of all cases.39 For example, the Polish Development Bank engaged in two swaps and has equity investments in about 15 other enterprises. Its work-out department which handles equity investments has a staff of eight and often draws on foreign consultants to leverage its resources. Despite its limited involvement and staff this bank is one of most active large banks with regard to equity investment and restructuring activity. This only serves to highlight the general reluctance of banks to get involved in actions that involve an active role in corporate control. Clearly, the financial restructuring law has forced many banks to acquire expertise in this area and to build up expert staff, and may thus have contributed to overcome the reluctance of commercial banks to shoulder the set-up costs for equity ownership and the exercising of corresponding control rights. Nonetheless, the recently introduced strategy of the Polish Development Bank towards equity investments clearly indicates that the reluctance of entering this line of business remains high. PDB intends to limit equity holding to less 20%; it prefers investments where a strong partner takes care of corporate control, as it considers that its capacity for involvement in corporate governance and restructuring is limited to at most 3 to 4 firms40; there should be a possibility of exiting the investment after 3 years, a requirement that highlights the importance of liquid securities markets; and, finally, investments should in any case be exited after 5 years. Equity ownership and corporate control on a larger scale, so the management feels, is a distinct business from commercial banking and is best pursued in separate organisational units. Indeed, PDB is in the process of setting up separate investment funds and established a joint venture with Kleinwort Benson to manage one of the soon- to-be established National Investment Funds. It is also worth noting that equity ownership of Polish banks in general is already close to the legal limit, although most of this is in the form of equity stakes in other financial institutions. 39 See Baer and Gray (1994). 40 However, their capacity to search for attractive investment opportunities is much higher. 27 There appear to be two major exceptions to the general rule of banks' reluctance to own equity. First, in some cases the bank realises that enterprise management is consciously driving down the value of the firm in order to pick up the pieces at low prices. Then banks are sometimes prepared to buy stakes in these companies in order to replace management. The second exception concerns enterprises where banks perceive a clear under-exploited upside potential. In this case several of the larger banks are prepared to buy minority stakes of up to 20%, provided there is another good and active shareholder (e.g. a foreign investor) which can be assumed to provide effective governance. iii) Debt-for-equity swaps are not particularly popular with banks. In Poland, as mentioned above, swaps have occurred in only about 30% of all cases. In Hungary, debt-equity swaps appear even less popular. Debt-for-equity swaps are often avoided because of the implied loss of collateral. Banks are reluctant to "swap bad debt for bad equity". This may be particularly the case in Poland where bank loans--whether secured or not--are senior to all other types of loans. In addition, in the case of unlisted companies, Polish prudential regulations require that a zero value is assigned to such stakes in the balance sheet, forcing banks to fully provision against the swapped loan. C. Summary This section has analysed the existing conditions in central Europe and Russia. A number of conditions are of importance with regard to the issue of the most appropriate mechanism of corporate governance. In brief, they are: i) Very illiquid corporate securities markets, with little prospect that will develop quickly. ii) Diverse restructuring requirements of enterprises, with an emphasis the need for accounting, marketing and financial expertise. iii) In some countries, continued state influence over large parts of the system. iv) Relatively weak competitive forces in banking. v) The need for improved supervision of the banking system. vi) In some countries, continued expectations among bank managers the government will subsidise losses from bad investments. vii) Lack of bank expertise in the restructuring of firms. In some such as Poland, banks have been forced to attain such expertise. IV. Implications for the Role of Banks in Ownership and Governance The impracticality of relying on liquid securities markets and take-overs as a means of governance in eastern Europe and Russia leaves a primary role to some type of large shareholder monitoring. The combination of severe information asymmetries between insiders and outsiders, and a lack of expertise in evaluating any information that is publicly available means that securities markets in eastern Europe and Russia are likely to remain undeveloped for quite some period of time. This essentially rules out any role for the U.S. model of corporate governance which relies to a substantial extent on anonymous securities markets and the threat of take-over as a method of management discipline (Tirole (1991) and Phelps et al (1993) reach a similar conclusion. However, note that Miszei (1994a) argues the opposite point). Indeed there may be some dangers in trying to create more liquid securities markets while relying primarily on an active shareholder to monitor firms. First, a higher degree of liquidity may lessen the incentives of investors to monitor their investments rather than sell them. Second, as the example of Japan illustrates, if banks are chosen to play a primary role as active investors in firms, their ability to do so may be hampered if they cannot control firms' access 28 to external finance. The question we address here is whether banks are the appropriate institutions for this role. A. The Argument against Banks Bank capacity to monitor: limited managerial capacity and competing bank functions Our snapshot of existing conditions in eastern Europe shows that banks suffer form a severe lack of expertise and human capital in evaluating and monitoring firms. This would appear to provide a clear prima facie case against a large bank role in restructuring and corporate governance. But this lack of expertise characterises not just banks, but all domestic institutions in eastern Europe and Russia. Thus any arguments against banks based solely on their limited capacity in this regard are not particularly strong. However, there are reasons to suspect that state-owned banks might have less expertise compared to other possible domestic candidates for the role of active investor. Much of the current management and staff at the banks was hired and trained under the regime of central planning. The emphasis then was on book-keeping, not on commercial banking and certainly not on the skills required to manage a portfolio of large equity stakes in firms. Thus management and staff are unlikely to be able to develop quickly the entrepreneurial and business skills necessary to play a major role in corporate governance. Indeed, some observers have gone so far as to talk about the "negative human capital" embodied in the state-owned banks, suggesting that their break-up or scaling down would be beneficial. In countries such as Poland and Hungary where the state banks still comprise the vast majority of the banking sector, this may be a much more powerful argument against banks playing a primary role in corporate control than it is in Russia where the private banks are far more predominant. In addition, banks have other functions to perform that are just as important if not more important than the corporate control function. In the transition economies, banks provide the only payment services available; they are the almost exclusive providers of financial instruments; and they are the predominant sources of external finance for enterprises. The financial services they provide may be poor but currently they are the only game in town. Ensuring continued provision of these services and overall confidence in the financial system is clearly essential. Banks are also at the interface between the central bank and the economy through their large presence in the money markets. Given the limited experience of bank management, it will almost certainly be stretching their capacity to conduct normal commercial banking activities if the ownership and control of firms is an additional responsibility for them. In other words, banks may simply not have the capacity to play an important role in the ownership and control of firms given their limited ability even to conduct more traditional commercial banking operations.41 Bank incentives to monitor: governance and supervision issues Even if banks had the managerial capacity to be important players in the governance of firms, their incentives may be such that they fall prey to conflicts of interest in conducting their activities. Bank credit to enterprises in central Europe (though not in Russia) is highly concentrated among a few mostly state-owned banks that have yet to be fully weaned off of bail outs from the state, and that are inadequately supervised. Such a situation would appear to be very susceptible to banks abusing or misusing any powers of ownership and control over firms that they might have. 41 See Scott (1992) for an argument along these lines. 29 First, given that most large banks are still state-owned, they could be put under pressure by the government to subsidise loans to firms whose bankruptcy would create political problems. This of course is true whether or not the banks own and control firms. However, the situation is worsened if banks do have this power, since large stakes held by banks might preclude the ownership by a non- politicised agent that may have genuine interests to maximise the value of the firm. Second, the relationship between banks and the govemment is further complicated by the fact that in some countries the expectation persists that the govemment will bail out banks that get into trouble in the future by making unsound loans. Giving banks an ownership role in enterprises under conditions in which bank management is not held responsible for their investment decisions would mean that firms are unlikely to be monitored efficiently. Third, the current state of supervisory capacity in the transition countries means that even without the perverse incentives introduced by state ownership and expectations of future bail outs, there is little safeguard against banks making unwise or imprudent loans. Giving banks the opportunity to make equity investments which by their very nature are more risky than debt investments could be a recipe for disaster when bank supervision is very weak. Finally, the concentrated nature of the banking system means that the discipline imposed by market competition is weak in some market segments, notably the one of large loans to non-prime borrowers. In addition, the opportunities for banks to extract rents from firms in which they have large equity stakes is enhanced by a lack of competition. In sum, even if banks had the ability to engage in a govemance role without impeding their ability to conduct traditional commercial banking activities, several arguments suggest that they might not use their control powers in a constructive manner. State ownership, the persistence of expectations of future bail-outs, weak bank supervision and a lack of competition all mean that banks' incentives to use corporate control powers to maximise firm value are not strong. At worst, the incentives created by these conditions might lead not only to a failure to restructure firms properly, but to the re- emergence of bad asset problems in the future. Restructuring requirements offirms The description of the restructuring needs of eastem Europe and Russia suggests that banks may not be the most suitable institutions to play the primary role of active investor in the firm. First, as part of their inherited relationships with the large industrial firms the banks may have established links with current management or other stakeholders that make it harder for them to engage in radical downsizing and restructuring than other active investors with no such links.42 Second, banks are unlikely to be able to supply the expertise in financial management, accounting, marketing and inventory control that enterprises desperately need and that a successful active investor would be expected to provide. Once again, arguments against banks versus any other domestic active investor based solely on the ability to supply such expertise are not strong since there is no competing domestic candidate for an active investor role that has these skills in any abundance43--what it does point to is the desirability (as far as is politically possible) of a large foreign venture-capital-like element in the corporate governance mechanism. 42 Shleifer and Summers (1988) point out that long-term relationships between financiers, managers and other stakeholders in the western firm may be of considerable valuable. In the context of a firm in one of the transition economies, these long-term relationships may be a carry-over from the days of central bank management of the central plan and thus be of considerable negative value to the extent that they lead bankers to continue lending money to their friends in the state firms regardless of the economic prospects of the firm. 43 Although, as argued earlier, it may well be the case that the management of the state banks have substantially less such expertise than other domestic candidates. 30 B. The Argument for Banks Banks' current involvement and the lack of other viable institutions One argument for giving the current state banks a role in corporate governance stems from two political realities. The first is that it may not be politically feasible to supplant the state banks with completely new institutions, as proposed by a number of western economists.44 Given that these institutions are going to continue to exist, and given that they have a very large presence in the banking and financial sectors of many transition countries (Russia excepted), they are likely to play some role in the corporate control of large firms. The second reality is that the state banks in some transition countries have already been playing an important role in the restructuring of large firms for a number of years. Primarily motivated by their need to do something about the bad loans on their books, some state banks have swapped the debt for equity in a few of these firms and have made attempts at restructuring them.45 This has occurred to differing degrees in different countries and probably not to the extent anticipated a few years ago, but nevertheless it has occurred and not without some degree of success. Indeed, some state banks (for example, PDB in Poland) have arguably the greatest level of expertise in the country in the restructuring of large enterprises. Thus any policy on the development of a corporate control mechanism should recognise the state banks' current and likely continued role in this area. Related to the above point is the question of who could be an active investor if it is not to be the state banks? Private and foreign banks--who might appear to be the better agents to accomplish ownership and governance tasks because they are not burdened by bad debts or have any state involvement--do not appear to be doing so, but are content with sticking to basic corporate lending to the best credits. There are a limited number of other candidates that could be viable and important players in a governance role in all but the long term. For example, pension funds or insurance companies are not large enough to be important players in the financial markets or in corporate governance for some time to come. The only serious alternative to banks, in the medium term at least, appear to be investment funds. A detailed analysis of investment funds in the Czech Republic can be found in Shafik (1994), Coffee (1994) and Claessens (1994). How effective these bank-managed funds will be from a corporate governance perspective remains, however, to be seen.46 44 See Phelps et al (1993). Of course, in Russia this has been largely achieved as an unintended side effect of the hyperinflation. 45 The relatively small number of bankruptcies in these countries (Hungary excepted) is not a good indicator of bank's involvement in corporate restructuring. Bankruptcy proceedings in all countries are very slow. Banks have attempted to deal with defaults by threatening to shut down firm's access to trade finance to force a restructuring. And at times they have indeed taken equity stakes in companies in default to them to exert direct influence on management. 46 For the purposes of this paper, two features stand out. First, the largest bank-affiliated funds created very broad portfolios, comprising between 200 and 500 companies. Hence they deliberately refrained from gaining effective control in a smaller number of companies. This is in marked contrast to some of the larger and successful private investment funds and the ones established by foreign banks which deliberately went for controlling stakes in companies, sometimes in violation of the (apparently not too strongly enforced) rules. Why did domestic banks forego this unique opportunity? A number of explanations can be advanced, ranging from carelessness to fear of not being able to invest all voucher points. Perhaps the most convincing one is that bank-affiliated funds tries to obtain small stakes in a large number of firms in the hopes of attracting banking business through those relationships. Secondly, banks apparently do not enforce strict fire-walls exist between themselves and their investment funds. In some cases bank management or employees have been used to fill the supervisory board positions the banks' IPFs are entitled to. In others there is a regular exchange of information between the bank and its IPF. The close co-operation between banks and their IPFs has not however led to improved corporate govemance as some had hoped. On the contrary, providing effective corporate governance and being actively involved in restructuring is seen by banks as high cost activities with little return - given their generally low stakes in the companies. Instead, they are using their information actively to trade securities. These are early days, and any judgement must remain tentative. But it would appear as if bank involvement in company ownership has done more to promote the trading of securities than to improve corporate govemance directly. 31 While banks in the Czech Republic appear to have chosen not to become too involved in corporate governance and restructuring, at least through the IPF vehicle, it is noteworthy that other, private investment companies have chosen the corporate governance road. And judging from first, preliminary evidence, it seems is if concentrated ownership were indeed perceived by the market to be sign of better prospects: large, strategic ownership by domestic or foreign investors has had a strongly positive influence on share prices (Claessens, 1994). Investment funds based on the Czech model may therefore be a viable alternative to banks as primary agent of corporate governance. Banks'informational advantages Banks may be best placed to mitigate the severe information problems between managers and any potential owners. This still seems to apply in central Europe even though much of the banks' involvement with borrowers was under a completely different regime. This comparative advantage in gathering information about enterprises stems from their on-going relationship with the firm as a creditor and payment settler. To the degree that enterprises require new loans or the roll-over of existing ones, banks can and do request privileged access to internal information about the firm. They are also, in the case of large customers, likely to know the enterprise managers and their capacities fairly well. Banks' monitoring capabilities are strengthened by the fact that they require enterprises to hold their deposit accounts with the main lending bank. In Russia, regulations remain effective which forbid enterprises to open more than one current account, giving banks a full view of the cash flow of their customers. C. Are Banks Ready to Play a Major Role in Corporate Governance? We have argued that for some time to come effective corporate governance in central Europe and Russia will need to come from large active shareholders. Even if securities markets developed rather rapidly it would take years before a reliance on them and the take-overs associated with them as a means of governance could be envisaged. The question of whether banks, in their present condition, are the right institutions to play this role, has been the focus of this paper. The strongest argument for banks appears to be the lack of obvious alternatives and the (limited) experience they have gained already over the last few years. These arguments however currently appear to be outweighed by the potential disadvantages. The banking system has other tasks to fulfil that are clearly more along the lines of traditional commercial banking and arguably more important than that of the governance of firms. At present, many banks are struggling to master their more traditional businesses. In addition, the environment for banks needs to be improved substantially before a larger role in governance becomes a recommended option. Incentive problems have not been resolved completely and supervision is still weak. Some comparisons with the current situation in the transition countries and the situation in Japan in the immediate post World War II period may be instructive. As pointed out by some observers advocating the adoption of Japan-like bank-firm ties for the transition countries, in many ways there are considerable similarities between the two situations (see Hoshi, Kashyap and Loveman (1994)). In 1946, Japan was faced with a transition from extreme centralisation and govemment intervention in the economy to a market economy. Post-war Japan too had a severely limited productive capacity, a very high rate of inflation and extremely close ties between large banks and govemment administrators. The banking system was highly concentrated and burdened with an immense amount of bad debts as a result of previous govemment directed and unmonitored lending. Despite these problems, the Japanese were able to install a system of corporate governance of enterprises led by the banks that coincided with a rapid recovery from the immediate post-war chaos and 40 years of subsequent rapid economic growth. The conditions in Japan in 1946 mentioned above mirror those currently in the transition countries. However, there are two crucial differences between the current situation among the 32 transition countries and the Japanese situation in 1946. The first is that, although market based lending had been suspended in Japan for a number of years prior to and during the war, there was an inherited institutional and human capital base in the banking system embodying knowledge of how to evaluate investments and the soundness of loans. In contrast, market based lending and the evaluation of investment proposals in the transition countries has been suspended for over 40 years and there currently is no inherited expertise whatsoever in these areas in their banking systems. Perhaps more important, however, is the fact that the Japanese adopted a conscious policy of restructuring and placing the banking system on a sound financial footing before the restructuring of nonfinancial firms. Indeed, banks cleaned up their balance sheets relatively quickly in Japan, and by early 1948 they could be considered to be relatively sound and largely free of perverse incentives (see Hoshi (1994)). They had never of course been under the ownership of the state. That meant that the banks were very well placed to take an active role in the restructuring of the industrial sector without succumbing to the dangers involved in doing so while still burdened by substantial bad debts or being subject to undue state influence. In contrast, although the importance of putting the financial sector on a sound footing was recognised early on in central Europe, it has taken a long time to make substantial progress in this area. Although much has been achieved in the last few years, much also remains to be done. Indeed, perhaps a more relevant comparison to make is with the sad experiences in a number of countries with emerging financial markets--such as Argentina, Chile, Venezuela and Mexico.47 In these countries the existence of close bank-enterprise relationships (often including the holding of equity stakes in enterprises by banks) led to well-publicised asset problems and even bank failures.48 A major reason for these problems was a legal and supervisory structure in these countries that was unable to enforce "good" behaviour on the banking system. This may be the more relevant comparison because it is quite clear that the legal and supervisory structure in the transition countries is much closer to that in countries such as Argentina and Chile than it is to the current situation in Japan or Germany, or arguably even to the situation in Japan in 1946. D. Implications for regulation Our analysis suggests that, for the time being, banks should not be allowed to take large equity stakes and play a primary role in non-financial companies. The main reasons are that banks are still subject to perverse incentives arising from weak competition and supervision, and the weak credibility of government pledges that no further bank bailouts are in the offing. In addition, most banks currently appear not to have the managerial capacity or the requisite business skills to take on the governance role that many firms in the transition countries require. That is not to say that once banks are free from their current problems a system of corporate governance based primarily on banks might not be appropriate in the medium term, particularly since well-developed and liquid capital markets are unlikely to play a major governance role in these countries for the foreseeable future. However, if banks are to play such a role in the medium term, they would benefit from being given opportunities to experiment in a limited manner with equity participations and governance roles now in order to build up the experience necessary to play a more important role in the future. Thus, while banks should be prevented from becoming large shareholders in firms, there is a case for adopting a more lenient stance towards their equity participations and governance role in 47 We refer here to the banking system problems in Chile and Argentina in the early 1980's and in Venezuela in the last year. The problems in Mexico over the last year have been much less extensive than in the previously mentioned countries, being largely confined to one or two banks that have made large insider loans to affiliated companies. 48 In some cases, the nature of the bank-enterprise relationship was enterprises taking large equity stakes in banks, which in turn made large loans to the enterprise or enterprise affiliates. 33 firms which have defaulted on loans to them. Not only would banks be allowed to gain some experience in owning firms, but also it is in those firmns which are currently in default to the bank that the bank is likely to have some substantial informational advantages over other potential primary shareholders. Such opportunities should be strictly controlled however, both in terms of the length of time the banks is allowed to own equity obtained through a swap of debt, and also, perhaps, related to the perceived strength of the bank itself. The implications for regulation are the following: first banks should be restricted in the amount of equity they own. The current prudential limits in central European countries seem appropriate. But, secondly, banks should be given limited opportunities to swap bad debt into equity and become active in governance. The current waiver period in the Czech Republic and Hungary of 18 to 24 months for equity acquired through swaps of debt again appears to accord well with this principle. In addition, there may be a case for enabling the authorities to bar extremely weak banks from making use of this waiver. V. Conclusion Our review of corporate governance arrangements in the west suggests that for a system based on bank ownership and control of firms to be successful the banking system must be free of perverse incentives and state interference and subject to adequate supervision by the banking authorities and competition from market forces. Admirable progress over the past few years notwithstanding, these conditions do not currently exist in the countries of central Europe and Russia at present. Therefore, a corporate governance system based on bank ownership is not appropriate. That is not to say that such a system would not eventually be appropriate for these countries. But before that much more effort is needed to create a competitive, private and well-supervised banking system. It may be worth pointing out that this effort is needed in any case, regardless of any possible involvement of banks in corporate governance. The changes in the banking system that are a necessary precursor to any large scale bank involvement in the ownership and governance of firms are simple to enunciate but much less easy to implement. First, existing relationships between the state and banks must severed. Privatisation is the strongest guarantee that bank investment decisions will not be subject to state influence. However, bank privatisation in most countries has been slow. This reflects to some degree the difficulties, in the early years of reform, to fully understand the poor condition of the financial sector. It also reflects the many institutional and political obstacles to bank reform. In addition, the initial decision in some countries to focus first on the "real economy" may have played a role. With the benefit of hindsight, that decision now appears somewhat unfortunate, though understandable in the circumstances of the early reform years. Another important step is the dispelling of the belief - that still exists in some countries - that poor lending and investments will eventually be underwritten by the government with little consequences for managers. Here some countries have clearly made more progress than others. Political factors have weighed heavily in some cases. But this issue must be resolved before banks can be entrusted to make investment decisions and even traditional loans on a rational basis. Strong competitive forces in the banking system are another necessary condition. The foundation of new private banks and the entry of foreign banks should therefore be encouraged. Some countries, such as Poland, have been taking exactly the opposite tack here in refusing to issue new licenses. Effective bank supervision is essential for the soundness of the banking system in general and in particular for the success of any corporate governance system based on bank ownership and control. An effective prudential and regulatory framework requires a substantial investment in setting 34 up institutions, accounting systems and information networks, hiring and training qualified personnel, and ensuring the system is immune from political intervention. Given the size of this task this will inevitably be a drawn-out process. However, if the establishment of a privatised and competitive banking system is viewed as critical to the development of a successful financial system and a successful corporate governance system, then priority must be given to establishing a sound prudential and regulatory framework. Foreign assistance will be crucial to the development of such a system. While a governance system based on bank ownership and control of firms is not yet feasible, there is a good case to be made for allowing banks to gradually gain experience through limited equity ownership, in particular in the case of swaps of bad debt for equity. The regulations currently in place seem conducive to such experimentation while sufficiently tight to prevent the emergence of conflicts of interest on a large scale. 35 References Baer, H. and C. Gray (1994), Debt as a control device in transitional economies: The experience of Hungary and Poland, mimeo Bod, Peter (1993), Monetary policy and exchange rate policy in Hungary during the years of transition, paper prepared for the conference on monetary and exchange rate policies in small, open economies, Vienna, April. Borio, Claudio (1990), Leverage and financing of non-financial companies: an international perspective, BIS economic paper no.27. Brainard, L.J. (1991): Strategies for economic transformation in central and eastern Europe: role of financial market reform, pp. 95-108, in: Blommestein, H. and M. Marrese (eds.), Transformation of planned economies: property rights reform and macroeconomic stability, OECD: Paris. Claessens, Stijn (1994), Corporate governance and equity prices: evidence from the Czech and Slovak Republic, IBRD mimeo (October). Claessens, Stijn and Gerhard Pohl (1994), "Banks, capital markets, and corporate governance", IBRD Policy Research Working Paper No. 1326. Cable, J. (1985): "Capital market information and industrial performance: the role of West German Banks", The Economic Journal, No. 95. Clyde, P. (1989): "The institutional investor and the theory of the firm", unpublished manuscript (UCLA, Los Angeles, CA.) Coffee, John C. Jr. (1994), Investment privatisation funds: the Czech experience, mimeo (September). Corbett, J. and C. Mayer (1991), Financial reform in eastern Europe: Progress with the wrong model, Oxford Review of Economic Policy, 4, 57-75. Dittus, Peter (1994a), Bank reform and behaviour in central Europe, Journal of Comparative Economics, December. Dittus, Peter (1994b), Corporate governance in central Europe: the role of banks, BIS economic paper - no.42. Edwards, J. and K. Fischer (1994): Banks, Finance and Investment in Germany, Cambridge University Press (forthcoming). Elston, J.A. (1993): "Finn ownership structure and investment: theory and evidence from German panel data", unpublished manuscript. Fenn, G., N. Liang and S. D. Prowse, (1994), The economics of the private equity market, mimeo. Franks, J. and C. Mayer (1992): "Corporate control: A synthesis of the international evidence", unpublished working paper (London Business School, London, England) Franks, J. and C. Mayer (1993): "German capital markets, corporate control and the obstacles to hostile takeovers: Lessons from 3 case studies", London Business School Working Paper. 36 Gessler Commission (1979): Grundsatzfragen der Kreditwirtschaft, Scriftenreihe des Bundesministeriums der Finanzen, 28. Gilson, S.C. (1990): "Bankruptcy, boards, banks and blockholders: Evidence on changes in corporate ownership and control when firms default", Journal of Financial Economics, Vol. 27, No. 1, pp. 355-388. Group of Banking Supervisors from Central and Eastern European Countries (1992), Report on Banking Supervision in Central and Eastern European Countries, Basle. Hart, Oliver and John Moore (1994), Debt and seniority: an analysis of the role of hard claims in constraining management, NBER working paper no. 4886. Hinds, M. (1990): Issues in the introduction of market forces in eastem European socialist economies, IBRD: Washington, D.C., mimeo. Hoshi, T. (1993), Evolution of the main bank system in Japan, Center for Japanese Economic Studies working paper 93-4, Macquarie University, Sydney Hoshi, T., A. Kashyap and G. Loveman, (1994), Lessons from the Japanese main bank system for financial system reform in Poland, Economic Development Institute working paper 94-18 Hoshi, T., Kashyap, A. and D. Scharfstein (1990b): "The role of banks in reducing the costs of financial distress in Japan", Journal of Financial Economics, Vol. 27, No. 1, pp. 67-88. Hoshi, T., Kashyap, A. and D. Scharfstein (1993): "The choice between public and private debt: An analysis of post-deregulation corporate financing in Japan", unpublished manuscript Hmcir, M. (1994), Reform of the banking sector in the Czech Republic, in Bonon, J.P and I.P. Szekely (eds.), The development and reform of financial systems in central and eastern Europe, Edward Elgar. Jensen, M. (1986): Agency costs of free cash flow, corporate finance, and takeovers, American Economic Review, papers and proceedings: Nashville TN, vol. 76, p.323-329. Jensen, M. and W.H. Meckling (1976), Theory of the firm: managerial behaviour, agency costs and ownership structure, Journal of Financial Economics, 3, 305-360. Kaplan, S. (1993a): "Top executive rewards and firm performance. A comparison of Japan and the U.S.", University of Chicago Working Paper. Kester, W. Carl (1991): Japanese Takeovers: The Global Quest for Corporate Control, Boston, Harvard Business School Press. Lane, T. and D. Folkerts-Landau (1992): Financial sector reforms in formerly centrally-planned economies: banking, securities, and payments, paper presented at a conference on the role of democratic governments during and after the transition to a market economy at the University of Twente, Enschede, August 13-15. Monopolkommission (1978): Hauptgutachen II. Fortschreitende Konzentration bei Grossunternehmen. Miszei, K. (1994a), Financial intermediaries and corporate governance: Prospects and pitfalls, mimeo 37 Miszei, K (1994b), Bankruptcy and the banking system, in Bonin, J.P. and I.P. Szekely (eds.), The development and reform of financial systems in central and eastern Europe, Edward Elgar. Myers, S.C. (1977), Determinants of corporate borrowing, Journal of Financial Economics, 5, 147-175. OECD (1995), Country survey of Poland, Paris. Phelps, E.S., Frydman, R., Rapaczynski, A. and A. Shleifer (1993): Needed mechanisms of corporate governance and finance in eastern Europe, EBRD discussion paper no. 1. Pfeiffer, H. (1989): "BASF und die Deutsche Bank--Eine Fallstudie zum Bankeneinfluss in der Wirtschaft", WSI-Mitteilungen 42, pp. 9-16. Prowse, S.D. (1990): "Institutional investment patterns and corporate financial behavior in the United States and Japan", Journal of Financial Economics, Vol. 27, No. 1, pp. 43-66. Prowse, S.D. (1992): "The structure of corporate ownership in Japan", Journal of Finance, Vol. 47, No. 3, pp. 1121-1140. Prowse, S.D. (1993): "The structure of corporate ownership in Germany", unpublished working paper. Prowse, S.D. (1994): Corporate governance in an international perspective: A survey of corporate control mechanisms among large firms in the United States, United Kingdom, Japan and Germany, BIS Economic Paper, 41, July, Basle. Purrucker, M. (1983): Bank in der kartellrechtlichen Fusionskontrolle, Berlin. Sahlman, W., (1990), The structure and governance of venture capital organizations, Journal of Financial Economics, 27, 473-524. Sheard, P. (1989): "The main bank system and corporate monitoring and control in Japan", Journal of Economic Behavior and Organization, Vol. 11, pp. 399-422. Sarcinelli, M. (1992), Eastern Europe and the financial sector: Where are they going?, Banco Nacionale del Lavoro Quarterly Review, 183, 462-492. Saunders, A. (1994), The separation of banking and commerce, Journal of Banking and Finance Sanders, A. and I. Walter, (1994), Universal banking in the United States: What could we gain? What could we lose?, Oxford University Press, NY Scharfstein, D., (1992), Japanese corporate finance and governance: Implications for the privitisation of eastern European enterprises, mimeo. Schmieding, H. and Buch, C. (1992): Better banks for eastern Europe, Kieler Diskussionsbeitrage No. 192. Scott, D., (1992), Revising financial sector policy in transitional socialist economies: Will universal banks prove viable?, World Bank Policy Research Working Paper 1034. Shleifer A. and L. Summers, (1990), Breach of trust in hostile takeovers, Brookings Papers on Economic Acivity 38 Shleifer, A. and R. Vishny, (1989), Alternative mechanisms of corporate control, American Economic Review Stiglitz, J.E. (1985): "Credit markets and the control of capital", Journal of Money, Credit and Banking, Vol. 17, No. 2, pp. 133-152. Shafik, Nemat (1994), Marking a market: mass privatisation in the Czech and Slovak Republics, IBRD PR Working paper no. 1231. Tirole, J. (1991): Privatisation in eastern Europe: incentives and the economics of transition, NBER Macroeconomics Annual, pp.221-259. van Wijnbergen, S., (1992), On the role of banks in enterprise reform: The Polish example, CEPR Discussion Paper 898. White, M. J. (1989), The corporate bankruptcy decision, Journal of Economic Perspectives, 129-151. 39 Policy Research Working Paper Series Contact Title Author Date for paper WPS1464 How Does the North American Free Edward E. Leamer May 1995 S. Vallimont Trade Agreement Affect Central Alfonso Guerra 37791 America? Martin Kaufman Boris Segura WPS1465 Post Trade Liberalization Policy Sarath Rajapatirana May 1995 J. Troncoso and Institutional Challenges in 37826 Latin America and the Caribbean WPS1466 Ownership and Financing of Charles D. Jacobson June 1995 WDR Infrastructure: Historical Joel A. Tarr 31393 Perspectives WPS1467 Beyond the Uruguay Round: The Jeffrey D. Lewis June 1995 B. Kim Implications of an Asian Free Trade Sherman Robinson 82477 Area Zhi Wang WPS1468 Government's Role in Pakistan Rashid Faruqee June 1995 C. Anbiah Agriculture: Major Reforms are Needed 81275 WPS1469 The Role of Labor Unions in Fostering John Pencavel June 1995 WDR Economic Development 31393 WPS1470 Pension Systems and Reforms: Patricio Arrau June 1995 E. Khine Country Experiences and Research Klaus Schmidt-Hebbel 37471 Issues WPS1471 Pension Reform and Growth Giancarlo Corsetti June 1995 E. Khine Klaus Schmidt-Hebbel 37471 WPS1472 Fiscal and Monetary Contraction in Klaus Schmidt-Hebbel June 1995 E. Khine Chile: A Rational-Expectations Luis Serven 37471 Approach WPS1473 The Surge in Capital Inflows to Eduardo Fernandez-Arias June 1995 Developing Countries: Prospects and Peter J. Montiel Policy Response WPS1474 Are Stable Agreements for Sharing D. Marc Kilgour June 1995 C. Spooner International River Waters Now Ariel Dinar 32116 Possible? WPS1475 Decentralization: The Way Forward Andrew N. Parker June 1995 D. Housden for Rural Development? 36637 WPS1476 Public Spending and the Poor: What Dominique van de Walle June 1995 C. Bernardo We Know, What We Need to Know 37699 WPS1477 Cities Without Land Markets: Alain Bertaud June 1995 L. Lewis Location and Land Use in the Bertrand Renaud 30539 Socialist City Policy Research Working Paper Series Contact Title Author Date for paper WPS1478 Promoting Growth in Sri Lanka: Sadiq Ahmed June 1995 A. Bhalla Lessons from East Asia Priya Ranjan 82168 WPS1479 Is There a Commercial Case for Panayotis N. Varangis June 1995 J. Jacobson Tropical Timber Certification? Rachel Crossley 33710 Carlos A. Primo Braga WPS1480 Debt as a Control Device in Herbert L. Baer June 1995 G. Evans Transitional Economies: The Cheryl W. Gray 85783 Experiences of Hungary and Poland WPS1481 Corporate Control in Central Europe Peter Dittus June 1995 G. Evans and Russia: Should Banks Own Stephen Prowse 85783 Shares?