斯特林大学英国金融学博士毕业dissertation辅导-seasoned equity offering and dividend policy in UK
1 Simultaneously
不错的引言 (inf asy, pecking, stock overvalue)
Prior studies suggest that firms should not pay dividends and issue equity at the same time because the cost of using retained earnings is lower than the cost of issuing new equity to finance new investment projects (e.g. Fazzari et al., 1998). In the presence of information asymmetry, equity issuance is costly because market participants who believe that firms only issue new equity when their stocks are overvalued would buy new stocks at a discount price.
Pecking order theory (e.g. Myers and Majluf, 1984) suggests that firms should use retained earnings, then debt before utilizing funds from new equity issues to finance new projects. If new projects are expected to be worth investing, then firms are more likely to save cash out of cash flow for their upcoming investments by ceasing to pay dividends to their shareholders. But why do firms pay dividends and issue equity simultaneously?
此前的研究表明,公司不应支付股息、发行股票的同时,因为使用留存收益成本低于发行新股票的金融投资的新项目的成本(例如Fazzari等人。,1998)。在信息不对称的存在,股票发行是昂贵的因为市场参与者认为,公司只发行新股票时,股票估值过高会以折扣价购买新股票。
啄食顺序理论(如梅尔斯和Majluf,1984)认为,企业应该利用留存收益,然后利用基金从新股发行债务融资的新项目之前。如果新项目预计将是值得投资的,那么公司更容易保存现金的现金流为他们即将到来的投资停止向股东支付股息。但为什么公司支付的股息和发行股票?
(Ngo, 2011)
Lit 的引言(dp)
There is vast literature on corporate payout policy over the last four decades. Scholars have developed various theories to help explain why firms pay dividends, including signaling, tax preference, clienteles, agency cost, and catering theories. According to the literature, firms may use dividends for several reasons: (i) managers pay out cash to shareholders to signal the prospect of future earnings to uninformed outside investors; (ii) managers may adopt a dividend policy that appeals to investors whose preferences are not met by other firms in order to enhance the share prices via clientele effect (O’ Connon et al., 2003); (iii) managers are forced to pay dividends to reduce agency cost or free cash flow problem; (iv) managers use dividends to cater to irrational investors who prefer the categories of growth stocks and dividend paying to boost share prices above their fundamental values (Baker et al., 2005).
#p#分页标题#e#
(Ngo, 2011)
Pecking order
Firms’ investment projects are apparently not financed first with retained earnings to the greatest extent possible. Based on the concept of residual dividend policy, non-zero dividends would be distributed if and only if a firm’s internally generated cash flow exceeds the desired capital expenditure. Firms are expected to minimize the use of external equity financing because flotation costs to issue new shares can be substantial.
(Mori, 2012)
同时的理由 (从控股股东利益
Based on capital budgeting theory, value-maximizing firms should finance all attractive investment opportunities. If the specially tailored dividend payment for the controlling corporate shareholder outweighs the level of free cash flow, then the firm must depend on external financing to fund the shortfall to undertake positive NPV projects. Under these circumstances, if the benefit from self-interested tax saving outweighs pro-rata flotation costs, then it is reasonable for the controlling corporate shareholder to receive non-zero custom-made dividends, and to compel the firm to issue equity. That is to say, the combination of non-zero dividend payment and simultaneous equity issue is tailored for the controlling corporate shareholder at the expense of non-controlling shareholders in a concentrated ownership structure. Such a perquisite can motivate large-block ownership. It is worth noting that non-controlling shareholders usually cannot enjoy tax-saving benefits, although they must incur pro-rata flotation costs. Within our framework, managers are assumed to be perfect agents for the controlling corporate shareholder, but not for non-controlling shareholders. (具体细节需要看他的模型)
(Mori, 2012)
对Mori 2010的解释 (why 同时)
Mori (2010) develops a time preference fitted dividend model demonstrating that the controlling power over payout policy confers a valuable perquisite to the dominant shareholder in the firm.
His model suggests that controlling shareholders who prefer dividends to reduce their tax burden may compel firm’s managers to adopt a custom made dividend policy. According to the model, firms would issue new equity to finance the short fall of undertaking all positive NPV investment projects if the benefits of specially tailored dividend payment for controlling shareholders outweigh the level of free cash flows.
Obviously, this combination strategy will make controlling shareholders happy at the expense of minority shareholders. Large shareholders may reduce agency problem between shareholders and managers, but large shareholders also may increase agency problems. If large shareholders actively monitor managers, they may reduce agency problem between managers and shareholders. However, large shareholders may also expropriate small shareholders, thereby increasing the agency problem. If controlling shareholders successfully compels the entrenched managers to pay custom made dividends while issuing new equity at the same time, controlling shareholders actually gain benefits at the expense of small shareholders. Therefore, corporate governance clearly has significant effects on the dividends decisions. #p#分页标题#e#
(Ngo, 2011)
Why 同时 (agency cost, signal)
According to free cash flow hypothesis (Jensen, 1986), firms can employ dividend policies as a substitute for other corporate governance characteristics to convince shareholders that they will not be expropriated. To establish reputations, the managers of poorly governed firms may signal their good financial positions to the market participants by paying dividends if they intend to raise external financing in the market in the future. Thus, poorly governed firms will pay dividends. However, we argue, in the event of paying dividends and issuing equity simultaneously, that the managers use dividends to pacify the controlling shareholders to avoid dismissals rather than to establish firm’s reputation on capital markets.
(Ngo, 2011)
Past explanation for the simultaneously
“Sleeping dogs” theory (Warther, 1993) provides some explanations to such situations. The theory helps explain that managers set high payout levels to avoid management intervention by shareholders who prefer high dividends. For examples, Zwiebel (1996) and Myers (2000) have developed a model in which managers must pay minimum dividend or face retaliation by stockholders. Signaling theory (e.g., Ross (1977), Bhattacharya (1979), John and William (1977)) explains that dividends are used as signaling device of high quality firms even though dividends are costly for individuals because of high personal income tax. De Angelo and De Angelo (1990) find that managers of firms with long histories of paying continuous dividends are especially reluctant to omit dividends when their firms encounter protracted financial difficulties. In this case, the managers may foresee benefits of paying continuous dividends obviously outweighing the costs of issuing new equity. Signaling theory also predicts that dividends can establish a reputation for managers, and therefore firms have incentives to pay dividends to establish reputation. More recently, Mori (2010) presents a new theoretical perspective on dividend policy based on controlling shareholders’ view. He has developed a model based on inter-temporal consumption choice to explain tax clientele effect of dividends. The model shows that corporate investors are expected to prefer “time preference fitted dividends” if tax rates remain constant over time. And tax saving problems should be linked with inter-temporal consumption choices. From this model, Mori (2010) argues that if the benefits from self-interested tax saving are greater than pro-rata flotation costs, a group of firm shareholders who would be better off receiving dividends can compel the firm to issue equity to finance new positive NPV projects. The Mori’s (2010) model implies that a group of controlling shareholders who have a substantial control over dividend policy can avoid tax burdens at other shareholders’ expenses. #p#分页标题#e#
(Ngo, 2011)
Tax and institutional shareholders
In contrast to individual investors, marginal tax rates on dividends are usually lower for corporate investors, but those on realized capital gains are higher. Corporate investors do not necessarily prefer high-dividend paying stocks because the tax rate on dividend income is simply lower for them. Assuming intertemporal consumption allocation, corporate investors find it optimal to receive time-preference-fitted dividends if the tax rates are constant over time. If dividend shortfalls exist, they must realize capital gains and thereby suffer unfavorable tax treatment, although excessive payments cause intertemporal double taxation of reinvested dividends.
Our model demonstrates that the controlling power over dividend policy confers a valuable perquisite to the largest shareholder in a concentrated ownership structure. To minimize its tax burden, the controlling shareholder has a strong incentive to compel the firm’s managers to adopt a custom-made dividend policy. Concretely speaking, the firm would pay a non-zero dividend if the largest shareholder is a corporate investor who prefers dividend income to realized capital gain for tax saving.
(Mori, 2012)
发行成本 税 DP overvalue
Financial literature shows that the payout versus retention decision is affected by flotation cost of raising external equity and by personal tax on payouts. Theoretical models of imperfections in capital markets imply that external financing is more costly than internal financing (Myer and Majluf, 1984; Hubbard, 1998; Fazzari et al., 2000). Myer and Majluf (1984) develop a model in which they show that investors will discount firms’ stock prices if firms engage in new equity issuances. That is, if a firm decides to issue new shares through SEOs, then investors might interpret that the firm is overvalued because the managers are assumed to have no incentives to sell undervalued shares. Thus, firms are more likely to use retain earnings rather than to use proceeds from SEOs to finance their new investment projects. Consistent with Myer and Majluf’ (1984) predictions, Asquith and Mullins (1986) find that SEOs announcement has a negative impact on the share prices on the announcement date, and that price reduction is positively related to size of flotation. This evidence, which is attributed to Myers (1984) pecking order of financing preference, is consistent with adverse selection and transaction costs. Pecking order theory suggests that firms should utilize internal financing sources (retained earnings) rather than external financing sources (debt, new issues) to finance new projects in order to avoid costly financing due to information asymmetry. Therefore, it is costly to issue new equity and pay dividends simultaneously. This combination may incur additional tax (personal tax on dividends) and wasteful flotation cost for individual shareholders. And it seems that it is not optimal for firms to pay dividends and issue equity simultaneously (Fazzari et al, 1998). But why have many firms been paying dividends and raising new equity at the same time?#p#分页标题#e#
(Ngo, 2011)
Governance, managers, on DP and equity issue
Literature on corporate governance shows that controlling stockholders with their large equity stake have strong incentives to monitor operating management (La Porta et al, 2000, Shleifer and Vishny, 1986). In some cases, controlling shareholders can interfere the management if they displeasure with top management. Given that managers may be replaced or fired in some cases by controlling stockholders, managers have incentives to pay sufficient dividends to keep the controlling stockholders happy while raising equity to fund projects that they find attractive that might not pass muster by controlling stockholders or even might harm the minority stockholders. In addition, controlling stockholders who prefer a custom made dividend policy have strong incentives to compel the firms’ managers to pursue a dividend policy that minimize their tax burden. For almost all shareholders, the strategy of paying dividends and issuing equity simultaneously itself is not beneficial because of flotation costs of new offerings and additional tax burden. However, controlling stockholders can force managers to adopt such a combination strategy, making themselves better off at the expenses of other non-controlling shareholders.
(Ngo, 2011)
Sum
To our current understanding, there is no prevailed theory that helps explain why firms engage in such a strategy. Given above literature review and discussions, one can expect that firms engage in such a suboptimal dividend policy because of managerial entrenchment or benefits of high reputation on capital markets (firms keep signaling by paying dividends even such actions are costly).
HY development 里有一些解释可以看
(Ngo, 2011)
Empirical
Existing empirical findings suggest that some firms pay non-zero dividends and issue equity simultaneously. According to Loderer (1989), 43% of the traded firms in Switzerland raise funds through rights offerings. Moreover, 98% of these firms distribute dividends simultaneously. Fama and French (2005) report that, on average, 58% of dividend-paying firms simultaneously issued net equity during 1973–2002 in the United States. A survey by Brav et al. (2005) revealed that 65% of the U.S. dividend-paying firms in their sample tended to raise external funds rather than conside omitting dividends. Less than half of financial executives reported that desirable investment opportunities should affect dividend decisions.
(Mori, 2012)
2. SEO DP day return
引言 signal inf asy
Although the issue of whether corporate payout decisions convey valuable signals remains controversial in the literature, an examination of the impact of payout declarations on equity offerings may provide an alternative means of testing the reliability of such signaling. If such payout announcements correctly signal the future prospects of a firm, then this should mitigate the magnitude of any information asymmetry on the firm’s valuation. Thus, if there is a smaller price decline on the equity offering date, the existence of the signaling effect should be discernible close to the payout announcement period; alternatively, if the differences in returns are insignificant, even with a lower announcement return, then the signaling hypothesis can be rejected.#p#分页标题#e#
(wang et al, 2011)
引言 Inf Asy 和empirical
Over the past two decades a substantial empirical literature has developed, documenting a significant decline in equity prices around the announcement of a Seasoned equity offering (SEO). These studies average the announcement-day returns across all SEOs and find that the typical decline in the stock price of an American firm is around 2%–3%. A variety of hypotheses have developed to explain this reaction but the most prevalent is based on information asymmetry.
These tests generally find that when there is less information asymmetry, the SEO announcement return is less negative. For example, Korajzyk, Lucas, and McDon-ald (1991) show that disagreement in interpreting earnings announcements affects announcement-day returns, and Bayless and Chaplinsky (1996) show that the return is affected by whether it is a “hot” issue market.
(Booth et al, 2011)
为什么Paying div 降低不了Inf asy
The reason for the cash dividend not mitigating the information asymmetry in the SEO procedure is rather perplexing, particularly when share repurchases do effectively mitigate such asymmetry. Generally speaking, a cash dividend is a commitment payment which has a stronger signaling effect; in contrast, a share repurchase is a more flexible and temporary payment (Jagannathan et al. 2000). If a repurchase could be regarded as a signal of the future prospects of a firm, and thereby increase the SEO announcement return, then the dividend should at least have the same market reaction effect on the equity offering, and this would then be an accurate assessment of the cash dividend increase. However, the empirical results of Loderer and Mauer (1992) do not support this notion. The purpose of this study is therefore to re-examine the relationship between dividend declarations and equity offerings using updated data.对于现金股利不缓解信息不对称在SEO的过程的原因是相当复杂的,特别是当股票回购可以有效地减轻这种不对称。一般来说,现金股利是一种承诺支付具有较强的信号效应;相反,股票回购是一种更灵活的和暂时的付款(贾甘纳坦等人。2000。如果回购可以看作是一个企业的未来前景的信号,从而提高增发公告回报,那么股利至少应该对股票提供相同的市场反应的影响,这将是对现金股利增加一个准确的评估。然而,罗德里和莫尔的实证结果(1992)不支持这个概念。本研究的目的是以关系的重新审视和公平使用更新的数据分配方案产品。
(wang et al, 2011)
Inf asy 和市场反应的关系
Given that such under-pricing is positively related to information asymmetry, then the greater the asymmetric information, the more significant the market reaction would be to the signals provided by the firms. Applying this inference to the dividend declaration and the SEO announcement returns, we find that the sub-sample covering the 1990s (and subsequent years), a period which is likely to contain greater information asymmetry, shows a positive relationship between dividend payments and SEO announcements. This is referred to as the ‘information asymmetry hypothesis’.#p#分页标题#e#
HY development 里有一些解释可以看
(wang et al, 2011)
Inf asy 和div payer 的关系 ,Loderer and Mauer 1992’s empirical
It seems well accepted that dividend payers face less information asymmetry than nonpayers (e.g., Howe and Lin 1992; Khang and King 2006; Li and Zhao 2008). If this result is true, asymmetric information theory implies that the market should react less negatively to dividend payer SEO announcements than to nonpayer announcements. However, the only study that has empirically addressed this issue is that by Loderer and Mauer (1992), who use U.S. data from 1973 to 1984 and do not support this argument. This discrepancy between theory and empirical evidence is a puzzle. One possible explanation is that Loderer and Mauer’s data cover the years 1973 and 1974, a time when dividend payments were frozen by law under the Nixon administration. However, because this two-year period covers only approximately 4% of their total sample, it cannot adequately solve the puzzle.
Another possible explanation is the change in firm characteristics documented by Fama and French (2001). They identify a disappearing dividend puzzle: the percentage of dividend-paying firms declines significantly from 1978 to 1999. They also find that the total population of firms shifts in composition toward smaller firms with lower profitability and stronger growth opportunities. This result would indicate that structural changes may have occurred in the way the market reacts to the declining proportion of dividend-paying firms.
(Booth et al, 2011)
Iny asy 理论和 announcement return 的关系
The empirical literature finds that stock prices typically drop by 2%–3% on the SEO announcement day. Among the many attempts to explain this price drop, the effect of information asymmetries is the most prevalent. Asymmetric information theory suggests that managers have superior information about the true value of both the firm’s assets-in-place and its future growth opportunities. Myers and Majluf (1984), for example, contend that managers issue equities only when a firm’s equity is overpriced, resulting in a wealth transfer from new to current shareholders.
Consequently, a rational market anticipates this opportunistic behavior by managers and discounts the price on the announcement day. Several empirical studies support the implications of asymmetric information on SEO announcement-day returns by using a variety of proxies for the asymmetry. Korajczyk, Lucas, and McDonald (1991), for example, argue that the degree of information asymmetry increases with the lapse in time since the firm’s last information release. Consistent with this idea, they find that the magnitude of the price drop increases with the length of time since the firm’s last earnings release. Dierkens (1991) uses four proxies for information asymmetry: earnings announcement-day returns, idiosyncratic volatility, number of public announcements per period, and trading intensity. She finds that firms with larger information asymmetries experience larger price drops on the SEO announcement day. Bayless and Chaplinsky (1996) show that the size of the price drop is also affected by whether the announcement is made during a hot new issue market. D’Mello and Ferris (2000) find that SEO announcement returns are more negative for firms with fewer analysts following them and whose earnings forecasts exhibit less consensus.#p#分页标题#e#
Finally, D’Mello, Tawatnuntachai, and Yaman (2003) find that the SEO announcement return is affected by the frequency of SEOs, on the basis that this affects the degree of information asymmetry and whether the behavior is opportunistic. The implications of asymmetric information are also widely applied in both the theoretical and empirical work on dividend policy. Miller and Rock (1985) and John and Williams (1985) pioneered the theoretical work to support the smoothing behavior first noted by Lintner (1956). Given the assumption of an information asymmetry between insider managers and outside investors, they both derive a signaling equilibrium based on differential costs to providing a false signal. Empirical work confirms that dividend payers have less information asymmetry than nonpayers. Howe and Lin (1992) find an inverse relation between dividend yield and bid–ask spread, and support the information content associated with dividend policy. Khang and King (2006) show that the amount of dividends is negatively related to returns to insider trades across firms, where the return from insider trades is a proxy for the information asymmetry. Finally, Li and Zhao (2008) find that firms with less earnings forecast error and dispersion are more likely to pay dividends. Myers and Majluf (1984, p. 220) analyze the joint effect of information asymmetry and dividend status, and suggest one way to solve the SEO announcement-day price drop is for the firm to issue equity only when there is no information asymmetry and to maintain a dividend policy where dividend changes are closely c or related with value changes. Ambarish, John, and Williams (1987) generalize miller and Rock (1985), John and Williams (1985), and Myers and Majluf (1984). They construct an efficient signaling equilibrium with dividends and investments or, equivalently, dividends and net issues of common shares. In their equilibrium, dividends and new issues collectively communicate a ll private information a t a lower cost than new issues alone. Despite this extensive theoretical literature on the effect of asymmetric information on dividend policy and the extensive empirical literature on its effect on SEO announcement-day returns, there is only one empirical study that jointly investigates dividend status and SEO announcement-day returns. Loderer and Mauer (1992) use U.S. data from 1973 to 1984 to assess whether dividend status reduces valuation uncertainty and whether announcement-day returns are lower for non-dividend-paying firms. Surprisingly, they do not find any evidence supporting either argument and recognize that “the results are not easy to explain” (p. 219). The discrepancy between the implications of asymmetric information and Loderer and Mauer’s findings regarding the effect of dividend status are paradoxical. One possible explanation for the paradox lies in the structural changes that occur i n payout policy and the c hanging composition of firms sampled to test the theory. Of note is that the theoretical models depend on an underlying information structure, so that a s this changes over time we can expect changes in the value of a firm’s dividend-paying status. Fama and French (2001), for example, show that the propensity to pay dividends has declined significantly from 1978 to 1999, a result that has come to be known as the “ disappearing dividends” puzzle. Similarly, Aivazian, Booth, and Cleary (2006) show that it is firms with public debt outstanding, that is, f irms dealing i n arms-length financial markets, that primarily both pay and smooth their dividends. It is possible therefore that the information content of dividends and its effect on asymmetric information may have evolve d over time, consistent with Fama and French’s disappearing dividends result. Revisiting the paradoxical results of Loderer and Mauer (1992) and extending them to this later period thus allows an investigation of both the effect of dividend status on SEO announcement-day returns and the implications of the disappearing dividends phenomenon.#p#分页标题#e#
(Booth et al, 2011)
Seo announcement effect on stock return and inf asy (quality of the firm and using the DP)
Several papers predict that CARs following the announcement of an SEO will be negative. Myers and Majluf (1984) argue that information asymmetry between firms and the market leads to this outcome. Other papers build on the same argument and predict a similar outcome under different assumptions. Several empirical findings have confirmed that market reactions are along predicted lines (see Eckbo, Masulis, & Norli, 2007, and Kale et al., 2011, for a comprehensive review of this literature). In a world characterized by information asymmetry, higher-quality firms resort to signaling mechanisms to convey their quality. Previous studies have argued that dividends are a costly signal of firm quality (Bhattacharya, 1979; John & Williams, 1985; Miller & Rock, 1985). Consistent with this hypothesis, a recent survey by Brav et al. (2005) found that managers attach importance to the dividend initiation decision because they consider dividends to be more inflexible than share repurchases are. If paying dividends conveys information, dividend -paying firms should have a less negative abnormal return than non-dividend paying firms do when SEOs are announced. However, evidence supporting this hypothesis was missing from the literature until recently. In a recent study, Booth and Chang (2011) reported that the information asymmetry between dividend-paying and non-dividend-paying firms has increased sharply since themid-1980s. Prior to that time, the market did not differentiate strongly between these types of firms, but, subsequently, the market has reacted less negatively to SEO announcements by dividend-paying firms. We note that significant differences, depending on a firm’s dividend status, were observed in SEO announcement-day CARs from 1985 to 2002, and these differences coincided with fully taxable dividends (see Appendix A). These observed differences were followed by a structural shift in the economy when President George W. Bus h signed JGTRRA on May 28, 2003.
(Prakash et al, 2011)
Div Tax rate and capital gain tax rate, structural change
Several papers have studied JGTRRA’s differential impact on dividend-paying and no n-dividend-paying firms. The sharp reduction in dividend taxes, which resulted in the equalization of dividend and capital gains tax rates, has attracted considerable attention from academics in public economics, corporate finance, and accounting. Recent studies have covered a wide range of topics that include investment incentives (Carroll et al., 2003), firm payout policies (Blouin et al ., 2004; Chetty & Saez, 2005; Brown et al., 2007), ex-dividend returns (Chetty et al., 2007), taxes and valuation (Amromin et al., 2008), and capital cost (Dhaliwal et al., 2007). Chetty and Saez (2005) and Brown et al. (2007) analyzed JGTRR A’s effects on corporate behavior. Both studies found that dividend taxes affect agency issues. Changes in dividend taxes affect the incentive structure in the principal -a gent relationship, resulting in changes in payout policies. Furthermore, Chetty and Saez reported a surge in dividend payments and special dividends and increases in total payout. Brown et al. (2007) also found that executives with higher ownership were more likely to increase dividends after JGTRRA. Moreover, they reported that firms with higher individual ownership and high executive ownership actually declined in value. Specifically, for a firm with an average level of individual ownership, a one-standard-deviation increase in executive ownership led to a 1.7% lower return. The stud y’s findings are consistent with agency conflicts within firms. The studies cited above examine the link between JGTRRA and firms’ agency issues. Several complications arise, however, when a firm’s dividend status is also taken into consideration. As Brown et al. ( 2007) reported, high individual dividend tax rates may provide an incentive for some firms not to pa y dividends (perhaps due to double taxation effect on clientele ) , but shareholders ma y simultaneously bid down the fir m ‘s share price, fearing that managers will misappropriate funds. Rozeff ( 1982) and Easterbrook (1984) argue that the act of pa yin g dividends reduces agency costs for the firm and prevents one group of investors from gaining relative to another group after securities have been issued Paying dividends therefore can lower agency costs. Because taxes are exogenous to the firm, changes in dividend taxes present an opportunity to stud y the role of dividends in reducing agency costs. Easterbrook (1984) argues that dividend payments make it more likely that projects will not be financed out of retained earnings, which prevents a transfer of wealth from shareholders to bondholders. Just as bondholders do not want to increase dividends after interest rates have been set, shareholders want to increase dividends to avoid being taken advantage of by bondholders. Because the risk aversion of undiversified managers aligns their interests with those of bondholders, dividend payouts disincline them from doing so, thus lowering the agency costs of shareholder -manager conflict. Consequently, we use JGTRRA as a natural experiment to study whether dividend status makes a significant difference in SEO announcement-day CARs. We include both pre- and post-JGTRRA announcements, and we employ the agency theory of dividends to ex plain our findings. #p#分页标题#e#
(Prakash et al, 2011)
Loderer and Mauer (1992)
Loderer and Mauer (1992) examine the announcement returns from seasoned equity offerings (SEOs) in their study of dividend declarations and equity offering announcements in the US, and find that the SEO announcement abnormal returns are lower if the SEO firms pay a cash dividend to the shareholders around the offering date. Specifically, this indicates that the dividend declaration does not convey a good signal to the stock market, a result which is inconsistent with the signaling hypothesis.
(wang et al, 2011)