Endnotes

¹ Aoki and Dinc (1997) distinguish the rents accruing to a relationship bank by sources: information rents resulting from access to corporate inside information through monitoring; monopolistic rents coming from a financier’s informational monopoly over its clients due to the non-transferability of the information; reputational rents derived from 'good reputation' in assisting financially distressed firms often providing costly rescue financing; relation-specific rents accruing to the relational financier out of the economic value created from a specific relation; and policy-induced rents created by the government regulation or policies usually conditional on their compliance.

² The ten largest private banks held only 0.4% of the face value of corporate equity in 1994 compared with 1.3% in 1976 (Emmons and Schmid, 1998). A bank may also vote the shares held by an investment company if the bank has majority ownership. Furthermore, although many firms have restrictions on voting, they do not apply to banks’ proxy voting.

³ Sheard (1989) stresses that this role of main banks was particularly significant given that: (i) the standards of corporate accounting and disclosure have been poor in Japan by international standards; and (ii) the managerial labor market tends to be highly internalized with a high firm-specific skill component in the human capital of Japanese managers as well as consensus-based decision-making practices. This role of main banks may resolve the potential problems of the external takeover mechanism as an instrument of capital control arising from the difficulty in distinguishing between bona fide and opportunistic takeover agents engaging in the strategic exploitation of imperfect information.

⁴ However, this characterization of the main bank system as a "nexus of relations," and with main banks playing typical roles, seems to be less accurate or weak for small and medium-sized firms. Minato (1999) argues that financial institutions (with the largest share of) lending to a small and medium-sized firm typically do not provide any monitoring, bailing-out, or disciplining of incompetent managers. He also states there is no "nexus of relations" (as claimed by Aoki): bank loans are extended rather independently with little social obligation to bail out the firms in times of financial distress.

⁵ It is difficult to describe the MB system in specific terms, because the contracts involved are largely implicit. Boot, Greenbaum, and Thakor (1993) explain the use of legally unenforceable, discretionary financial contracts in circumstances where legally enforceable contracts are feasible. The discretionary contract fosters reputation enhancement, thereby increasing future fee income. The better the guarantor’s reputation, the greater is its incentive to write discretionary contracts, and a discretionary guarantee of a highly reputed guarantor can be more valuable than an enforceable guarantee of a less-reputable guarantor. Examples include holding company cross-guarantees ("comfort letters"); mutual fund contracts (discretionary guarantees); loan commitments; an investment bank’s “firm commitment” underwriting a contract; price-stabilization promises for new bonds and equity issues during the issuance period.

⁶ In this connection, interviews with Japanese bankers in 1999 reveal that corporate governance by Japanese banks is conducted through financing, rather than through shareholdings (Hirota, 1999).

⁷ Creditor banks that also hold equity shares in financially distressed firms face potentially serious conflicts of interest in the presence of other fixed claimants (Berlin, John, and Saunders, 1996). It was observed that German bankers seldom take equity stakes when firms enter financial distress (Edwards and Fischer, 1994). For US banks as well, it is mainly conflicts among fixed claimants rather than regulatory restrictions that restrict banks’ equity positions in distressed restructurings (James, 1993).

⁸ They include Fukuda and Hirota (1996) for Japan; Elsas and Krahnen (1998), Harhoff and Körting (1998) for Germany; Angelini, Di Salvo and Ferri (1998) for Italy; Petersen and Rajan (1994) and Cole (1998) for the US.

⁹ Kaplan and Zingales (1997), however, question the use of sensitivity of investment to cash flow as a reliable measure of "liquidity constraint" defined as "a wedge between the internal and external costs of funds a firm faces." While constrained firms should be sensitive to internal cash flow, it is not necessarily true that investment-cash flow sensitivities increase monotonically with the degree of liquidity constraints. There may be non-monotonicity in the cost function of raising external funds reflecting information and agency problems. Precautionary savings motives weaken the link between liquidity and investment, and irrational or overly risk-averse managers might choose to rely primarily on internal cash flows for investment despite the availability of low cost funds. Fohlin (1998a) notes that firms with the highest estimated liquidity sensitivity showed no real signs of liquidity constraints, indicating the potential for misleading conclusions when liquidity sensitivity is used to measure liquidity constraints.

¹⁰ Based on these findings, they suggest that informationally intensive firms (with many harder-to-value assets) are more likely to value a close (single) banking relationship (it is also more costly to establish multiple banking relationships or public debt facilities) and face costly information monopolies, and that banks are unable to completely resolve the resulting agency problems (adverse selection & moral hazard).

¹¹ There can be other modes of risk-sharing. For instance, lenders may subsidize borrowers in early periods (in order to discourage them from taking up excessively risky investments) and be reimbursed for this subsidy in later periods (Greenbaum, Kanatas, and Venezia, 1989; Sharpe, 1990). In arm’s length banking, banks are likely to charge higher rates and ask collateral early in the relationship when there are still large asymmetries of information. Only after these asymmetries are reduced through repeated transactions for some time will the banks lower interest rates and demand less collateral (Boot and Thakor, 1994).

¹² MB client firms are likely to adopt a profit smoothing accounting policy since they have strong incentives to show that they are good borrowers with low risk and stable profits. Consequently, their accounting information might be more useful as information about on current and future cash flow level than is the case for firms with a strong MB relationship. Uchida (1997B) finds a stronger association between unexpected profits (using the change in profit rate as a proxy) and cumulative abnormal return on equity for firms with a strong MB relationship.

¹³ Prowse (1992) shows evidence that financial institutions’ equity investments in independent firms are very sensitive to the benefits accruing from exerting control over firms with unstable environments (larger variation of profit rates measured by both stock and accounting returns). This indicates that financial institutions, including banks, can mitigate the high risk of their clients by better monitoring management as a major shareholder.

¹⁴ Collateral gives the lender greater incentive to liquidate failing firms, while an unsecured lender has little incentive to take actions that might lead to a run on the firm. This reduces the moral hazard problem on the part of borrowers. However, fully secured lenders have no incentives to monitor borrowers (Rajan and Winton, 1995). Also, collateral signals borrower quality, giving the lender additional information about the borrower in the process of inspecting the collateral (Picker, 1992).

¹⁵ Miwa and Ramseyer (2001) and Miwa (1985) presuppose that if MBs play their expected roles, the present MB of ‘a distressed firm’ is more likely to be the same as its MB ten years earlier, and that the MB loan share will increase in times of distress. They, however, find that the replacement rate of MB (the largest creditor) is higher for distressed firms, and that MB loan shares show no significant changes between 1973 and 1984.

¹⁶ Petersen and Rajan (1994) suggest that a single bank relationship might be a proxy for the lower quality or higher riskiness of firms. They provide other explanations as well: relationship increases information monopoly; cost of credit does not matter much because the marginal return from investment is usually much higher under credit rationing; and loan officers usually have larger discretionary power in deciding the availability rather than the pricing of the credit they supply.

¹⁷ Using the US Small Business Survey data, Scott (2000) shows that low accounting manager turnover and frequent social contact with the owner of the firm, which represent an aspect of strong relationship banking, have similar effects – significantly increasing credit availability and lowering loan rates.

¹⁸ Lichtenberg and Pushner (1994), however, find that ownership of other corporations had a negative effect on corporate performance. Their conjecture is that non-financial corporate ownership might insulate the firms from their own problems or market forces at the expense of profit/productivity; corporate owners might also encourage non-profit-maximizing behavior in their own interests. They also find a positive effect of director ownership indicating that ownership reduces the agency conflict between managers and shareholders.

¹⁹ However, they find q ratios rising monotonically with both ownership by management and corporate block holders. This indicates that large block holders are a way of overcoming the free-rider problems associated with dispersed ownership.

²⁰ The concern over conflicts of interest is not just a problem for commercial banks but for other institutional investors as well. Insurance companies and investment funds are rarely active in their governance role with various regulatory restrictions on their portfolio as well as conflict of interests as they belong to larger business groups. Company-sponsored pension funds run by corporate managers themselves are particularly vulnerable to conflicts of interest. Relatively active in corporate governance among institutional investors are union-sponsored and especially public-employee pension funds. Even public pension funds that are considered the best candidates for corporate-governance activities seem to be constrained by political pressures to accommodate local interests and refrain from such activities (Romano, 1995). The governance of these funds is often problematic due to the appointment of politically affiliated members to the boards of directors.

²¹ Having a banker on the corporate board, therefore, has two major functions. First, it is a means of reducing information asymmetry and disciplining corporate management. This may allow the firm to have better access to bank credit. Second, given the cost of lender liability, a banker joining the board of a firm is likely to have a certification role: signaling to the market that the firm is sound and unlikely to face financial distress. This signal lowers the firm’s costs of external financing.

²² If the number of banks is too small, however, banks may be able to extract rents from the borrowers without fulfilling the duties as relationship banks. Thus, relationship banking would not emerge (Aoki and Dinc, 1997).

²³ For the merits and demerits of universal banks, see Benston (1994), Diamond (1998), Fohlin (1999), and Guinnane (2001).

²⁴ In 1987, the Glass-Steagall provisions were relaxed: some banks were allowed to set up Section 20 subsidiaries that can underwrite corporate securities. They are subject to a substantial set of firewalls designed to limit information, resource and financial linkages between them and their parent HCs as well as with their commercial banking affiliates. Revenue generation from the securities activities of commercial bank holding companies is limited to 25% (adjusted upward from the initial ceiling of 10%).

²⁵ In Japan, the Financial System Reform Act of 1993 allows banks to set up securities subsidiaries (and securities firms to set up trust bank subsidiaries). On the basis of their experience as trustees for customers placing corporate bonds in the domestic market and as guarantors or co-underwriters (with Japanese securities houses) for bond issues abroad, Japanese commercial banks were able to penetrate the securities markets within a short period of time.

²⁶ Seoul Bank, one of the most distressed, was excluded from the study as it had been selected by the government for sale to reputable foreign banks. The planned sale was not materialized at the time of the study, but was under a management contract with Deutsche Bank following recapitalization by the government. The other bank, Korea First Bank, was sold to Newbridge Capital, which owns a majority share.

²⁷ The system of "principal transactions banks (PTBs)" started as a part of a credit control device on large business groups in the mid-1970s. As the capital structure of large business firms became increasingly fragile with the heavy burden of debt, the Korean government attempted to correct the situation by controlling bank loans to business groups while encouraging them to turn to the capital market for financing. The PTBs were given the task of implementing credit control and improving the capital structure of major borrowers by keeping an eye on the financial and business situations of the client firms and business groups. The system of PTBs has undergone some changes in its scope and in the roles of the banks. In later years, for instance, the PTBs were entrusted by the financial supervisory authorities with restricting the debt-financed acquisition of "non-productive" real estate or investment in other companies. The emphasis of the system is now on reducing credit concentration on large corporations and improving the capital structure of heavily-indebted business groups, thereby enhancing the soundness of the asset portfolio of financial institutions. In addition to managing credit and business information on large business groups and their affiliates and providing managerial guidance for the improvement of their capital structure, the "main creditor banks" (MCBs – the new name for the PTBs) organize a creditor bank consultation committee and work out and implement corporate restructuring measures with the help of other creditor banks when a client firm is in serious financial distress. Although the functions of PTBs or MCBs appear similar to those of main banks in Japan, there seems to be a distinct difference. The system was superimposed by the government with certain policy objectives and has failed to foster an autonomous relationship of mutual commitment. As such, the bank monitoring of client firms has tended to remain superficial with the major attention given to the compliance with government regulations (Nam, 1994).

²⁸ In the case of nationwide city banks in Korea, the maximum ownership share for any individual, corporation, or business group has been set at 4%, effectively preventing large business groups from controlling banks. The consequent diffuse ownership structure, which lacks large shareholders, resulted in a situation of governance vacuum, providing fertile soil for government interference in the management of private banks. The poor governance in these banks may have significantly contributed to the inefficient management of banks and the deterioration of their asset portfolios.

²⁹ The Korean financial supervisory authorities have required city banks to introduce a “bank president recommendation committee” to select their CEOs since 1993. The committee is composed of 3 previous presidents of the bank, 2 representatives each for large and small shareholders, and 1 representative each for corporate and individual customers. From 2000, the “candidates recommendation committee” composed of all outside directors selects the president and auditor of the bank. Outside directors are required to constitute a majority of bank board of directors; and 70% of outside directors are to be appointed by the shareholders, while the remaining 30% are selected by the board of directors. In reality, the financial supervisory authorities are believed to have substantial influence on the selection of presidents for government-owned banks.

³⁰ Yoshida (2001) provides a case study of Asahi Beer and its MB, Sumitomo Bank. In spite of its decreased reliance on bank loans, Asahi maintained a strong relationship with Sumitomo in the late 1980s: Sumitomo was its third largest shareholder, dispatched most top managers, underwrote corporate bonds, and guaranteed bonds issued by Asahi. Though Sumitomo was in a position to properly monitor Asahi, there was actually a vacuum in MB monitoring, leading to the deterioration of Asahi Beer’s financial status.

The views expressed in this paper are the views of the author/s and do not necessarily reflect the views or policies of the Asian Development Bank Institute nor the Asian Development Bank. Names of countries or economies mentioned are chosen by the author/s, in the exercise of his/her/their academic freedom, and the Institute is in no way responsible for such usage.

Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [0] comment(s) for this entry. Post a comment.