At a discussion meeting hosted by Commerzbank in Frankfurt-am-Main on 22 February, Jeremy Edwards and Klaus Fischer presented the main results of a new study which cast doubt on the merits of Germany's distinctive, bank-based system of investment finance. Edwards is University Lecturer in the Faculty of Economics and Politics at Cambridge University, a Fellow of St John's College and a Research Fellow in CEPR's Financial Economics programme. Fischer wrote his PhD thesis on house-bank relationships in Germany at the Universität Bonn and now works in banking in Germany. The research reported in this volume was financed by the Anglo-German Foundation for the Study of Industrial Society and formed part of CEPR's research project, `An International Study of the Financing of Industry', which was also funded by the Bank of England, the European Commission, the UK Economic and Social Research Council, the Esmée Fairbairn Charitable Trust, the Japan Foundation and the Nuffield Foundation. Financial support for the meeting from Cambridge University Press is gratefully acknowldeged. The views expressed by Edwards and Fischer were their own, however, not those of the above organizations nor of CEPR, which takes no institutional policy positions.

Economic theory suggests that German banks might enjoy both economies of scale in information-gathering and economies of scope between information-gathering and the exercise of control rights. Representation on supervisory boards might allow a relatively small number of banks to provide loan finance at low cost by reducing informational asymmetries between borrowers and lenders; it might also prevent returns from being dissipated by inefficient management and reduce the costs of financial distress. Universal banks might also monitor managerial performance for firms with widely-dispersed share ownership, while their close involvement as lenders might enable them to ensure that only high-quality firms approach the stock market for equity finance, so shareholders would be more willing to take up new issues.

Edwards and Fischer reported, however, that this compelling theoretical case for the `German-style' system bears little relation to the operation of German banks in practice. During 1970-89, for example, non-financial enterprises in the UK financed a higher proportion of gross capital formation by loans from financial institutions than their German counterparts. Nor did a relatively small number of banks dominate the market for loans to all German firms: while the `big three' undertook most of the duties associated with representation on supervisory boards, control of proxy votes and underwriting new equity issues, much bank lending was undertaken by banks with no such influence. Indeed, most German firms do not possess supervisory boards, and those larger companies that are legally required to do so financed a smaller proportion of investment by loans than others in Germany or indeed comparable public limited companies in the UK.

Edwards and Fischer then considered the influence of German banks on equity finance, noting that they supply only some 3% of the value of their loans to non-bank firms as equity, while their role as underwriters of new share issues does not lead to significant flows of equity finance for investment. Nor do they effectively limit the costs of financial distress or ensure the smooth reorganization of troubled firms; since their loans are typically secured by collateral, their incentives to reorganize rather than liquidate are rather limited.

They then compared the German system of corporate governance with that achieved by hostile take-over elsewhere. Most large listed companies in Germany have at least one `large' shareholder with an incentive to monitor management effectively, while most of their counterparts in the UK do not. This suggests that differences in the incidence of managerial failure reflect the different structures of share ownership rather than any role of banks as such in monitoring or control. German banks are often held to act on shareholders' behalf by using proxy voting rights to install their representatives on supervisory boards, but this relationship is empirically very weak; even if all bank representatives acted together, they could not determine the boards' decisions. Overall, there is little conclusive evidence to indicate whether German banks exercise effective control over firms. Cable's 1985 study of data on individual companies investigated whether banks act on shareholders' behalf under the assumption that profits should then be higher in firms over which banks have stronger control. A careful analysis of his results reveals, however, that the effect of bank control on profitability was statistically insignificant in most of the estimated equations. Edwards and Fischer noted in conclusion that the institutional structure that entails extensive representation of large firms on each others' supervisory boards of which the `big three' banks are the most prominent may benefit the German economy, but the available evidence on the supply of external investment to firms and the exercise of corporate control provides no support for widely-held beliefs concerning the merits of the German system of corporate finance.