【正文】
id back the lost equity capital on the open Consequently, a high dividend payout helps in minimizing agency costs. The implication of both these arguments is that dividend payments will be higher where there are dispersed outsider investors, as long as the firm is in continuous need of equity capital and thus forced to interact with the capital role of the institutional structure through which the firms raises capital is thus important for dividend policy. Rajan (1992) showed the difference between bonds and bank debt is in the information acquisition process and the potential for renegotiating the contract. The key is that bank debt is a contract between an informed provider of debt capital who has access to current corporate information, much of it confidential. The banking relationship usually requires the filing of quarterly financial information in a standardized form, as well as regular site visits by the lending officer, so that the lending officer is familiar with the pany. Moreover, much of the bank debt is either short term or involves ‘‘material adverse conditions’’ clauses that effectively give the bank an almost continuous call option on its loan. Consequently, there is a significant reduction in the extent of the moral hazard problem, as well as of agency costs. This risk reduction is accentuated by the practice of recovering the loan principal through monthly mortgage type payments. In essence this practice serves the same premitment function as a dividend. It is not surprising therefore, that James (1987) found that the announcement of a credit facility was acpanied by a % 2day abnormal equity return. In many ways initiating a banking relationship is equivalent to the initiation of a dividend: both signal higher quality firms and a premitment to cash outlays. In contrast to bank debt (‘‘inside’’ debt), Diamond (1991) characterizes public debt or bonds as involving less monitoring. As a result there is less emphasis (and cost) on information gathering and on the renegotiation process discussed by Rajan. The key features instead are the existence of a track record and a public reputation. In stark contrast to bank debt, bond investors tend to rely heavily on public information reflected in bond ratings, a history of dividend payments, and simple financial ratios. In such a context the premitment to pay dividends may be more important than for a firm relying solely on bank debt, particularly when many institutions are restricted to investing in the public debt of dividend paying firms only. Consequently, dividend policy may be a more useful premitment and signaling mechanism in financial systems that are more heavily reliant on arms length transactions. We term this role for dividend payments the substitute hypothesis, in the sense that the dividends substitute for direct munication with the external investors in the firm, both debt and equity. This substitute hypothesis implies that there should be significant differences in dividend policy across countries with different institutional structures. In particular, dividend policy should be more important for firms operating in arms length capital markets, rather than in ‘‘internal’’ bank centric markets. Dewenter and Warther (1998) made a similar point when they argued that stable dividend payments may not be as important for firms in Japan that are part of a ‘‘Keiretsu’’ work, due to the close ties between managers and investors. In contrast to the substitute hypothesis, one could argue that dividends reinforce rather than substitute for other mechanisms in controlling agency and information problems. We term this the plement hypothesis. La Porta et al(1998) argue that the lack of transparency, inadequate legal infrastructure, and weak investment protection in emerging markets all enhance the role of dividends as a reputation mec