The Impact of Ownership Structure on the Dividend Policy
We find evidence in support of the hypothesis that a positive relation exists between dividends and free cash flow and it’s greater for low-growth firms than for the high-growth firms. The results also show that the impact of managerial ownership and bank ownership on dividend yield is positive particularly for the low growth firms. This is inconsistent with the view that the managerial ownership and institutional ownership reduce the need for the dividend mechanism. Finally, there is evidence that the Keiretsu classification affects relations between ownership structure and dividend payouts.
Overall, the dividend policy appears to be used by Japanese low-growth firms to control the overinvestment problem. Free cash flow hypothesis is to some degree supported. JEL classification codes: G32 G34 G35 Keywords: Ownership Structure, Dividend Policy, Free Cash Flow -2- 1. Introduction Why does a firm pay dividends? This question has been the subject of debate for many years, In the pre-Miller and Modigliani era, it was believed that increasing dividends would always increase market value.
Miller and Modigliani (1961) establish that in a perfect capital market, given an investment policy, dividend is irrelevant in determining share value. Empirically, however, we have observed that a change in dividend policy does have a significant impact on the share price. Different researchers have concentrated on different types of imperfections in the market in order to rationalize why dividends matter. Of these, a plausible idea is that corporate dividend policy addresses agency problems between shareholders and managers (Rozeff, 1982; Easterbrook, 1984; Jensen, 1986).
According to these agency theories, unless profits are paid out to shareholders as dividends, they may be committed to unprofitable projects that provide private benefits for the managers. Rozeff (1982) and Easterbrook (1984) argues that the payment of dividends expose companies to the possible need to raise external funds, and hence subjects them to greater monitoring by capital markets. Jensen (1986) argues that paying dividends reduces the discretionary resources under managerial control and so helps to mitigate the overinvestment problem.
In this study, we examine the implication of the free cash flow hypothesis in corporate dividend policy, and focus specifically on cross-sectional relations between dividend payout policy and ownership structure and free cash flow. Given the severity of the overinvestment problem, relations between dividend payouts and ownership structure, free cash flow may be conditioned on the existence of growth opportunities. This research examines how the sensitivity of relations between dividend payouts and ownership structure, free cash flow varies cross-sectionally with growth opportunities.
Previous studies have shown that in countries like the US, firm ownership is relatively dispersed, leading to a limited ability of owners to monitor or control management’s use of free cash flow. Thus the dividend payout is one of the primary control mechanisms whereby shareholders can reduce management access to or abuse of discretionary funds. In countries with 1) a higher concentration of ownership 2) extensive cross-shareholding and 3) strong banking relationship, like Japan, dominant shareholders are believed to have both the incentives and the ability to keep management in check.
Tests using a sample of 986 observations for 350 firms from 1992 to 2000 period indicate that the sensitivity of managerial ownership and bank ownership to dividend payouts varies directly with the relative abundance of growth opportunities. we find that dividend payouts for low-growth firms are significantly related to managerial ownership and bank ownership. In -3- contrast, there are no significant relations between dividend payouts and managerial ownership, bank ownership for high-growth firms. We also investigate associations between free cash flow and dividend payouts.
Consistent with the prediction by Jensen (1986), there is a strong positive relation between the level of free cash flow and dividend payouts. Furthermore, association between free cash flow and dividend payouts is stronger for low-growth firms. The rest of the paper is organized as follows. Section 2 reviews the previous theoretical and empirical research. Section 3 explains the Japanese institutional background. Section 4 describes the empirical framework.
The empirical results are presented in Section 5-6 and Section 7 concludes. . Dividend payouts, Ownership Structure and Agency Cost Theory 2. 1 Dividends and Agency Costs Corporate dividend policy has been viewed as a control mechanism that mitigates agency conflicts between shareholders and managers. Jensen and Meckling (1976) suggest that one way to reduce agency costs of equity is to pay a larger proportion of its earnings as dividends to its stockholders. A high dividend payout ratio will result in lower “discretionary” cash flows available to be squandered away by managers.
Rozeff (1982) argues that dividend payments are part of the firm’s optimal monitoring/bonding package and serve to reduce agency costs. Easterbrook (1984) lists some of the mechanisms by which dividends and the consequent raising of capital can control agency costs. Agency costs “are less serious if the firm is constantly in the market for new capital. When it issues new securities, the firm’s affairs will be reviewed by an investment banker or some similar intermediary acting as a monitor for the collective interest of shareholders, and by the purchasers of the new instruments”.
Free cash flow hypothesis The free cash flow hypothesis is a variant of the agency argument based on the Principal-Agent framework. According to this framework, dividends are used by shareholders as a device to reduce overinvestment by managers. Jensen(1986) argues that managers with substantial free cash flow tend to invest it in wasteful projects rather than pay it out to shareholders, because managerial compensation and perquisites increase even with poor investments. These unnecessary investments lead to poor performance, creating conflicts between shareholders and managers.
Jensen emphasizes the disciplinary role of dividends that restrain managerial unprofitable expansionary tendencies by limiting financial resources available to managers. Dividend payments represent an ongoing commitment to maintain higher payments in future periods, because firms are reluctant to cut dividends and have been greeted by a significant -4- negative stock market reaction when they do. Jensen suggests that dividends should be paid out in ways that instigate managers to gorge the cash beyond the optimal amount.
This implies that free cash flow positively determines dividend payments. . 3 Ownership structure and dividend policy One criticism of the agency cost theory is that if managers want to overinvest or spend more on jets, what is the mechanism that will force them self-commit to an action that will prevent them from doing so? Several authors address this issue in the context of ownership structure 2. 3. 1 Institutional Ownership There are several important ways in which institutions differ from individual investors. In general, institutions manage large pools of funds and therefore invest larger amounts in each stock.
Because they have larger amounts at stake, they should have incentives to devote resources to monitoring (Grossman and Hart, 1980; Shleifer and Vishny,1986). Institutions are also likely to be better informed than are individual investors. Not only do institutions devote resources to gathering information, but they are also sometimes privy to corporate information that individual investors do not have ( Michaely and Shaw,1994). However, the prediction on the relationship between dividend policy and institutional ownership are mixed.
The first line of research suggests a positive relation. Zeckhauser and pound (1990) suggest the arm’s length view of investment held by many institutional investors, coupled with the incentives to free ride with respect to monitoring activities, implies that institutional shareholders are unlikely to provide direct monitoring themselves. The institutions, rather than providing monitoring themselves, forces firms to increase their dividends in order that they are subsequently forced to go to the external capital market for future funds.
Eckbo and Verma (1994) argue that institutional shareholders will prefer free cash flow to be distributed in the form of dividends in order to reduce the agency costs of free cash flow. From this perspective, it may be argued that institutional shareholders may counter a tendency for managers to prefer the excessive retention of cash flow and, by virtue of their voting power, force managers to pay out dividends. Moh’d, Perry and Rimbey (1994) and Short, Zhang, and Keasey (2002) also provide additional support.
The second line of research suggests a negative relation. Jensen and Meckling (1976) argue that external monitoring activity is an important controlling element when agency conflict exists. If large institutional investors act as monitoring agents, and if dividends are paid to reduce agency cost, then according to this theory, there should be a substitute relation between dividend policy and institutional ownership.
This implies a negative relationship between the percentage of -5- hares held by institutions and the dividend payout. D’Souz, and Saxena (1999) provide the empirical evidence. 2. 3. 2 Managerial Ownership There are several lines of argument on the role of managerial ownership. The first line of argument suggest that managerial ownership may better aligning the interest of management and shareholders and helps mitigate free cash flow problems. Therefore it results in a higher level of total payouts when managers hold more shares. ( White,1996; Fenn and Liang,2001).
The second line of argument suggest that insider stock ownership provides direct incentive alignment between managers and shareholders while dividends serve as a bonding mechanism reducing management’s scope for making unprofitable investment out of internal funds. Thus, insider stock ownership and dividend policy are viewed as substitute means of addressing potential agency problem.
Empirical articles have shown that managerial ownership is not a linear function of agency costs. Morck, Shleifer and Vishny (1988) and McConnell and Servaes (1990) found insider ownership is related to performance in a nonlinear fashion. Schooley and Barney (1994) report a nonmonotonic relation between CEO stock ownership and dividend yield. Farinha (2003) documented the U-shaped relationship between insider ownership and dividend payout in the UK. He argues that it stems from the effects of managerial entrenchment.
Japanese Institutional Background Japan offers us a valuable opportunity to examine issues related to dividend policy under an institutional setup quite different from that in U. S. , such as the main bank, the cross-shareholding among corporations. This distinctive Japanese institutional background may result in important difference between Japanese and U. S. firms in terms of corporate monitoring and information sharing. Since, most theoretical explanations of dividends rely on agency and information issues, they would suggest different choices of dividend policy in Japan. 3. 1 Main bank -6-
Aoki, Patrick and Sheard (1994) highlight a significant governance role played by the main bank for Japanese firms. The main bank effectively monitors the client firms by becoming well-informed about the firm (Diamond, 1984)). The main bank’s equity stake in the client firm mitigates agency costs between creditors and shareholders (Prowse (1990)). The main bank sometimes intervenes the management of the client firm that performs poorly by appointing bank employees to the board of directors in the client firm (Kaplan and Minton,1994; Kang and Shivdasani, 1995; Morck and Nakamura, 1999).
In case of financial distress, the main bank acts as a guarantor for other creditors, reducing the cost related to the restructuring of the client firm (Hoshi, Kashyap and Sharfstein,1990).. In contrast, several authors suggest that there is a cost in having a main bank. Firms relying on the main bank for financing are likely to be constrained in raising the additional capital when the banking sector as a whole has a financial difficulty (Kang and Stulz, 2000). The main bank can extract surplus from the client firms due to its monopolistic power of information production (Rajan, 1992).
In a similar context, the main bank has an incentive to force the client firms to undertake low-risk, negative NPV projects (Weinstein and Yafeh, 1998). Firms that do not depend on bank borrowing exhibit higher profitability than the matched sample of firms that have a main bank ( Kang and Shivdasani, 1999). 3. 2 Keiretsu Group There exist differences between keiretsu or industrial groups centered around affiliated banks and financial institutions and unaffiliated independent firms with weaker banking ties.
Japanese industrial organization is characterized by groups of enterprises (keiretsu) composed of firms based in different industries but bound by ties of fractional ownership and reliant on a large commercial bank as the major but not sole lender. The large shareholders of keiretsu firms often are also large creditors of the firm as well as important long-term commercial business partners. The keiretsu and non-keiretsu firms are facing different liquidity constraints in their investment spending. Investment spending is very sensitive to liquidity constrains for non-keiretsu firms, but not so for keiretsu firms.
Since keiretsu firms are likely to have better access to financing sources, keiretsu firms seemingly face less liquidity constraints in making investment decision. The differences in institutional arrangements between keiretsu and non-keiretsu firms may influence the behavior of shareholders as monitors. Kester (1990) describes the corporate governance system of keiretsu firms in terms of a complex interaction between shareholdings, credit holding and long-term business relationship that exist between the firm and its stake holders.
Aoki, Patric, and Sheard (1994), and Berglof and Perotti (1994) suggest a two-tier monitor system. In the first stage, corporate cross-shareholders serve as the monitors under -7- normal circumstances because they have specific industry knowledge and observe each others’ performance through their business relations. In the second stage, the financial institutions take an active intervention role when member firms get into financial distress, replacing incumbent managers and requiring restructuring and liquidation of assets.
Managerial equity ownership Because the well-known keiretsu structure and influential bank shareholders, the agency problems between Japanese managers and shareholders are considered to be minimal (e. g. Nakatani, 1984; Hoshi, Kashyap and Scharfstein, 1990, 1991; and Prowse, 1990). The manager ownership, as a way of aligning interests between managers and shareholders, has been viewed as an unnecessary corporate governance mechanism. However Kang and Stulz (1998), Mock and Nakamura (1999), and Weinstein and Yafeh (1998) questioned the effectiveness of bank oversight in Japan.
Morck and Nakamura (1999) argue that for independent firms, bank equity holders pursue their interests as creditors at the expense of their equity claims. Gibson(1995) and Kang and Stulz (2000) argue that poor bank health may adversely affect their dependent firms’ investment prospects, which, in turn, would affect their ability to monitor effectively. This particular contention is especially relevant to the late 1980’s and early 1990’s as it is well known that Japanese banks have been experiencing significant financial difficulties during this time period.
In light of these findings, Morck and Nakamura (1999) contend that some independent firms may require corporate control mechanisms other than bank oversight. Due to the decline in power of Japanese banks, the rarity of incentive-based compensation contracts for Japanese managers, and the fact that many Japanese firms are not affiliated with a keiretsu group, the managerial-ownership may represents an alternative mechanism to ensure that firms operate efficiently.
Thus, the unique Japanese institutional arrangements provide an interesting backdrop to investigate whether cash flow theory explanation for dividend policy still apply given the differences. 4. Empirical framework 4. 1 Hypotheses If one assumes, as suggested by Jensen (1986), that managers receive utility from increasing the size of the firm, the control function of dividend payouts on the overinvestment problem varies with the firm’s growth opportunities. Management may have an incentive to pay out as few dividends as possible at shareholders’ expense.
The overinvestment problem is less important and may be trivial for firms with many growth opportunities, because the objectives of managers and shareholders are more likely to coincide. On the other hand, when good projects are not -8- available, managers with substantial free cash flow must find ways to spend it and hence choose poor projects. Thus, the overinvestment problem is higher for low-growth firms than for high-growth firms, and divergence of interests between shareholders and managers over the firm’s payout policy are more severe in firms with few growth opportunities.
These firms can limit management’s temptation to overinvest by paying out a larger percentage of their earnings. Their high-growth counterparts with lots of investment opportunities are likely to pay low dividends because they have profitable uses for the capital. For this reason, we expect stronger relations between free cash flow and dividend payouts for low-growth firms. Hypothesis I: relations between the level of free cash flow and dividend payouts are positive and are stronger for firms with low growth opportunities.
Most of the existing agency explanations of payout rely on the implicit assumption that firms can get refinanced on the capital markets when they need funds to undertake new investment projects. Consequently, the strategy that minimizes agency costs is to maintain a high payout (to reduce the amount of free cash flow and to avoid overinvestment problems) and to raise new outside capital whenever and attractive investment opportunity emerges.
Outside shareholders are harmed by a potential overinvestment and therefore they have preferences for high payout, which curbs the amount of corporate resources that can be spent by management on value reducing projects. Managerial ownership helps to align interests of management and shareholders that may yield the reduction of agency costs stemming from payout smaller. Consequently, payout ratios in a firm with managerial block holdings may be low because the severity of manager-shareholder agency conflict is low.
This traditional agency view generates a set of hypothesis that the payout is negatively related with the managerial ownership. Institutional investors are more effective at monitoring management than retail investors. Due to the size of their investments and the resources at their disposal, institutional investors have greater incentive and ability to gather and analyze information pertaining to their investments, as well as a greater ability to discipline management and push for changes when management performs poorly.