Corporate governance – bank versus market systems

1. The United Kingdom and the United States

In his study of financial systems, Berglof (1990) argues that there are essen­tially two models of corporate governance – market oriented and bank ori­ented. Using this taxonomy, both the system in the UK and in the USA fall into the category of market oriented. A market-oriented system is charac­terised by ‘outside’ shareholders who play little part in monitoring manage­ment behaviour, prefer to spread their risks widely, and trade actively in shares (selling them when dissatisfied with managerial performance). Falling stock prices and the market for corporate control provide an impor­tant source of management incentives. To use Albert Hirshman’s (1970) terminology, ‘exit’ (that is, the severing of a relationship and the search for new ones) is the preferred mechanism for exerting pressure rather than ‘voice’ (that is, investing time and effort in reforming the existing relation­ship).

The impersonal and distant relationship between the shareholder and the firm in the market-oriented system is reinforced by several other factors. Shareholders possess both control rights and residual rights. These are rarely separated, so that widely spread residual rights result in equally spread voting and control rights. The principle of equality of shareholder status is also defended by regulatory bodies who try to ensure equal access to information, prevent dealing by ‘insiders’ who may have privileged infor­mation, and insist on the disclosure of the build-up of large stakes in a company. Further, trade in shares on the UK and US stock markets is dom­inated by institutions. These are financial intermediaries, such as pension funds and insurance companies, owing a fiduciary duty to the people who have entrusted them with their money. This does not entirely prevent them from holding a stable portfolio of shares, but the prompt disposal of a failing stock may well be considered a cheaper and safer way of protecting the saver than embarking on a costly and possibly vain rescue attempt.

A central characteristic of market-oriented systems is therefore the low incidence of direct monitoring by ‘inside’ shareholders. Section 5 of Chapter 8 discussed the circumstances in which incentives to undertake monitoring would exist in a market system even where shareholdings were fairly dispersed. Accepting, however, that monitoring costs will limit the extent of such activity, Boards of Directors will find themselves relatively free of direct shareholder intervention. Shareholders are able to appoint outside ‘non-executive’ directors to the Board of a company, and auditors owe a duty to the body of shareholders to prepare suitably informative accounts.²⁸ If, however, the appointment of non-executive directors and auditors is under the effective control of executive directors, and if the flow of information reaching auditors and outside directors is controlled by the management group, the effectiveness of these control mechanisms is open to question.²⁹

Dimsdale (1991) and Sykes (1991) have discussed these problems in the context of the UK. The latter argues strongly for a greater commitment by institutional investors to monitoring and control. He suggests the use of managerial incentive contracts by which managers might hope to achieve ten to twelve times their annual salaries by way of stock options. These, however, would be tied to five to seven year performance targets. The stabil­ity of control required for the system to operate would be achieved by finan­cial institutions combining to take significant (twenty per cent or more) stakes in a company. Sykes wishes to encourage three to five institutions to build up expertise and knowledge about the firm in which they are to take a long-term interest. He suggests that investment institutions should each establish a secretariat which would support non-executive directors nomi­nated by the institution. His objective is clearly to move away from the market-oriented system towards an ‘investment-institution’-oriented (though non-bank) one. This would re-establish some of the conditions of classical capitalism by providing significant shareholders with an incentive to become knowledgeable ‘insiders’. He argues that agreements between financial intermediaries to build up expertise in specific areas, but outside these areas to spread risk widely, could limit the adverse consequences for risk pooling. Each financial institution would implicitly be conferring mon­itoring and control benefits on other institutions in its specialised field. The full benefits, therefore, would require an agreement by all institutions to participate and not simply free ride on the services of others. At present, there does not seem to be a working example of an ‘investment-institution’- oriented system of corporate governance.

2. Japan and Germany

In earlier chapters, some of the characteristics of Japanese transacting have already been discussed. The flexible, long-term, non-specific and obliga- tional aspects of transactional relations in Japan mentioned in the context of buyer-supplier relations (Chapter 7) and labour contracts (Chapter 6) can also be seen in the area of corporate governance. Shareholders in Japan are ‘insiders’. Although institutional shareholdings make up a similar pro­portion of the total in Japan as in the UK (about seventy-five per cent in the late 1980s), the institutions concerned are not independent pension funds or insurance companies. Instead, shares are held by institutions the commercial operations of which may be quite closely intertwined. Shareholders may include banks who also provide loan finance, firms that are customers or suppliers, and other companies in ‘Keiretsu’ groups linked by cross-shareholdings.

As a result of these reciprocal long-term arrangements, trade in shares is far less active. Implicit agreements to hold shares as friendly and passive ‘insiders’ are widespread and the most active trading of shares therefore takes place in the household sector. This is quite different from the market- based system where institutional shareholdings are the most frequently traded. Because markets in shares are thinner and holdings are stable, con­tinuity of control can be relied upon in Japan and the takeover is virtually unknown.

In Germany, there are no equivalent barriers to the transfer of shares that exist in Japan. In spite of this, continuity of control is greater than in a market-oriented system and changes in control are effected without fre­quent recourse to the takeover. A number of features of the German system produce these results. Public companies in Germany can issue non-voting stock up to an amount equal to all voting shares. Thus, it is possible, up to a point, to raise capital by means of an issue of non-voting shares without affecting the control of a company. Further, the banks in Germany are able to exercise by proxy the voting rights attached to shares deposited with them. This gives the banks an influence far in excess of that which would be expected solely on the basis of their own shareholdings.

As briefly mentioned in Chapter 8, section 5, monitoring incentives in the German and the Japanese systems seem to derive from a splitting of control rights from residual rights. Individual shareholders use other institutions not simply (as in the market-oriented system) to make trading decisions but to exercise a monitoring and control function on their behalf. The agency costs associated with this arrangement might be thought to be prohibitive, but the fact that the banks themselves hold a substantial minority stake in a firm³⁰, are providers of loan finance, and will often have seats on the Board means that they have the incentive and the influence to act as mon­itors of performance. In the Japanese context, Aoki (1989) sees the ‘main bank’ as implicitly providing monitoring and control services. He even argues that certain financial characteristics of the Japanese system, such as the tendency for firms to borrow more from the main bank than would be expected on the basis of normal commercial requirements, can be rational­ised as a payment for monitoring services. ‘The deviation from share price maximisation by the company at the sacrifice of individual stockholders and the phenomenon of overborrowing in the interests of the bank may in part be thought of as the “agency fee” paid by the individual stockholders to the bank for that service’ (p. 148).³¹

It would be a mistake to oversimplify the nature of bank-oriented systems by forming the impression that they recreate the conditions of clas­sical capitalism. Control rights may be more concentrated, and greater direct monitoring effort may occur through the ‘Hausbank’ in Germany or the ‘Main bank’ in Japan, but the conclusion that shareholder interests thereby dominate business decisions would be quite erroneous. In Japan, managers on Boards are mainly promoted internally through the ranking hierarchy described in Chapter 6. This mechanism is not designed to reward the singleminded pursuit of shareholder value. Indeed, Aoki (1984) sees managers in the Japanese system as mediators between the ‘quasi-perma­nent employees’ and the body of stockholders. The firm is a coalition of cooperating resources and a central managerial function is to make sure that the rents generated by the team are not dissipated in rent seeking. Instead, they are distributed according to a cooperative bargain mediated by the managers. This bargain is likely to involve higher growth rates and a greater concern for market-share objectives (in the interests of the promo­tion prospects of employees) than would be chosen by the stockholders alone.

The German system likewise gives voice to more than shareholder inter- ests.³² In particular, the two-tier Board structure is deliberately designed to provide a vehicle for worker representation. Membership of the Supervisory Board is divided equally between the representatives of share­holders and of employees. These Supervisory Boards appoint the Management Boards but they do not exercise very close control over day- to-day operations. They are there to oversee matters of broad policy and will not interfere unless business conditions deteriorate. However, clear institutional recognition of the claims of other ‘stakeholders’ in the firm is important. Vulnerable holders of firm-specific assets may feel more secure in a bank-oriented system with representation on supervisory boards than in a stock-market system where takeover by new owners is more likely.

3. Differences in Capital Structures

The protean nature of economic systems makes it advisable not to assume that observed differences in financial variables between countries are immutable. Nevertheless, bank systems and market systems do seem to be characterised by differing financial structures. Berglof (1990), for example, concludes that bank systems ‘generally have higher debt/equity ratios, higher shares of bank credits in their liabilities, more concentrated holdings of both debt and equity, and a lower turnover of these holdings’ (p. 257).

Studies of balance sheets indicate that companies in the UK and the USA have higher proportions of equity than do companies in Germany or Japan.³³ OECD data for 1975 to 1989 show the debt-equity ratio for non­financial enterprises in Japan falling steadily from 5.6 to 4.2. Over the same period, the German figures rose from 2.56 to 4.25. In contrast, the debt-equity ratio in the USA was much lower, although it increased during the 1980s. It rose from 0.52 to 0.82 (in 1989). The UK ratio was fairly stable at or just below 1.1 over the same period.

Care is required when interpreting this type of information, however. It certainly seems consistent with the contrast between ‘bank-oriented’ and ‘market-oriented’ systems. Greater stability, longer-term relations, better information flows and closer direct monitoring by banks would be condu­cive to a higher ratio of debt to equity. On the other hand, the nature of some of the ‘debt’ instruments is important. Much of the Japanese debt is made up of short-term trade credits between contracting firms. The close supply chain associations between Japanese companies and the lower degree of complete vertical integration was discussed in Chapter 7. These closer links and the high level of trust that is apparently engendered are reflected in the short-term credits that are advanced and which appear on both sides of Japanese balance sheets. The German case is complicated by the problem of pension provisions. Unlike ‘arms-length’ systems, where pension funds are governed by trustees who are independent of firms, in Germany pension contributions are retained by each company as a liabil­ity on their balance sheets. Financial ratios are very sensitive to the treat­ment of these provisions. Treating them as a form of debt produces a much higher ratio of debt to equity than treating the claims of pensioners as a form of equity stake.

If bank-oriented systems result in higher debt-equity ratios, market- oriented systems have certainly produced a greater incidence of takeover activity. The experience of the USA and the UK compared with Japan has already been mentioned in section 9 above. Franks and Mayer (1990) con­trast Germany with the UK and argue that the influence of the banks reduces the level of contested takeover activity.

In sections 5 and 6, the possibility that dividends would be important in market-oriented systems either as signals or as commitments to minimum levels of performance was discussed. The evidence does suggest that divi­dend payout is higher in both the USA and the UK compared with Germany and Japan. In the late 1980s, the ratio of dividends to gross income was around forty per cent in the UK, compared with 28 per cent in the USA and only 10 per cent in Japan. Mayer and Alexander (1990) esti­mate that dividend ratios in the UK were more than double those in Germany during the 1980s.

Source: Ricketts Martin (2002), The Economics of Business Enterprise: An Introduction to Economic Organisation and the Theory of the Firm, Edward Elgar Pub; 3rd edition.