Economic subjects | Investments, Stock exchange » C. Justin Robinson - Dividend Policy Among Publicly Listed Firms In Barbados

Source: http://www.doksinet Dividend Policy Among Publicly Listed Firms In Barbados C. Justin Robinson ABSTRACT This paper adds to the sparse literature on corporate management practices in the Caribbean by undertaking an in-depth study of dividend policy among publicly listed firms in Barbados. The paper utilizes an analysis of historical data on dividend payments, a questionnaire survey and structured interviews with financial managers in Barbados. The paper finds that unlike their developed country counterparts, financial managers in Barbados retain a strong commitment to paying dividends as opposed to share repurchases as a means of providing cash for investors. The historical data and survey responses suggest that firms in Barbados are reluctant to omit or cut dividends and appear to pursue stable dividend policies. In particular, financial managers in Barbados seem to take a “Bird in the hand” view of dividends. Also, financial managers in Barbados appear to target the

dividend per share as the variable to be managed as opposed to the dividend payout ratio, suggested in the academic literature. The author contends that the attachment to dividends as a means of returning cash to shareholders and the high level of dividend stability are largely due to the closely held nature of firms in Barbados, which in turn leads to thin trading and an almost exclusive reliance on dividends as a means of generating cash from equity investments. While this may be beneficial to large shareholders, it has potentially negative consequences for minority shareholders and the stock market microstructure in Barbados. INTRODUCTION This paper seeks to add to the sparse literature on corporate management practices in the Caribbean, by undertaking a pioneering study of dividend policy among publicly listed firms in Barbados. Dividend policy is defined as the payout policy that management follows in determining the size and pattern of cash distributions to shareholders over

time. The issues surrounding dividend policy have generated much debate among financial economists over the years. Specifically, the research has generally sought to shed light on four issues: (1) the manner in which firms determine the amount and pattern of dividend payments, (2) the relationship between dividend policy and the value of the firm (share price), (3) inter-country differences in corporate dividend policy and (4) the emerging issue of “disappearing dividends”. Over forty years of research and debate has generated much insight into these issues, but as is typically the case in economic research, the issues are far from settled and the debates continue. Whatever ones view on the academic debate there can be no disagreement that to date the research has focused almost exclusively on firms in developed countries. This study carries the debate into the area of developing countries or so called emerging markets. From a broader perspective, compared to the U.S, other

developed capital markets and the larger emerging markets, where most of the research has been concentrated, Barbados is likely to constitute a different institutional context and the study will generate insights into whether or not the apparent differences in institutional context translate into significantly different dividend policies among corporations. The study also adds to the growing body of survey evidence on corporate financial management practices. 1 Source: http://www.doksinet In attempting to address the issues surrounding corporate dividend policy in Barbados, the paper utilizes a questionnaire survey, interviews of financial managers and historical data on dividend payments in Barbados as part of the research methodology. It is believed that this mix of methodologies provides an interface between theory, the beliefs of corporate managers and evidence from actions to date, which may facilitate a clearer the understanding of the issues surrounding corporate dividend

policy in Barbados. The rest of the paper is structured as follows. Section two presents a literature review, section three discusses the methodology, section four discuses features of publicly traded firms and the stock market microstructure in Barbados, section five presents the results and discussion and section six provides a conclusion. LITERATURE REVIEW There is a wide and varied literature on corporate dividend policy, and the author does not attempt to provide an exhaustive review of this literature here 1. Instead, the author tries to outline the main arguments, issues and findings in the literature to date. For convenience, the literature review is organized around what the author perceives to be the four main issues addressed in the literature to date, namely, the manner of determining dividend payments, dividend relevance, inter-country differences in corporate dividend policy and the emerging issue of “disappearing dividends”. In terms of the manner in which corporate

managers go about setting dividend payments, Lintner’s (1956) paper remains the most authoritative study to date. Lintner’s study essentially drew three conclusions about dividend policy among US firms. Firstly, firms have long-run target dividend payout ratios expressed as a percentage of earnings, secondly, managers believe that investors prefer corporations that follow stable dividend policies and accordingly smooth dividend payments over time (managers believe it is better to pay two moderate level dividends than to pay a high one today followed by a low one next period) and thirdly, managers focus more on changes in dividends than on the level of dividends (managers believe that announcing changes in the level of dividend payments provides important information to investors and must be carefully considered). Lintner’s behavioral model suggests that the change in dividends is a function of the target dividend payout less the last period’s dividend payout multiplied by the

speed of an adjustment factor. Lintner (1956) subsequently tests his propositions and finds that the partial adjustment model predicts dividends payments more accurately than naïve models. A number of other studies have confirmed the dividend policy beliefs of managers in the U.S, as described by Lintner (1956) Brittain (1964,1966) and Fama and Babiak (1968) reformulate the Lintner model by undertaking a more comprehensive empirical approach and confirm the findings of Lintner that corporations follow stable dividend policies. Specifically, Fama and Fabiak (1968) contribute to the Lintner model by suppressing the constant term and adding more levels of lagged earnings. Fama (1974) repeats the same study by using a larger sample and once again, he reaches the same conclusion about dividend policy stability. The Lintner model has also been tested in a number of other developed countries. Ryan (1974) tests the Lintner model on firms in Britain, McDonald, Jacquillat and 1 The interested

reader is referred to an excellent review by Lease et al. (2000) 2 Source: http://www.doksinet Nussenbaum (1975) test the model on the French market, Chateau (1979) tests the model on large Canadian corporations, Shevlin, (1982) and Partington (1984) test the Lintner model on a sample of firms in Australia, Leither and Zimmerman (1993) test the model on four major European markets, namely West Germany, U.K, France and Switzerland, Kato and Lowenstein (1995) test the model on Japanese corporations, Lasfer (1996) tests the Lintner model for a panel data of commercial and industrial corporations in the U.K In all of these studies the common result is that corporations follow stable dividend policies along the lines suggested by the Lintner model. In more recent studies, Dewenter and Warther (1998) use the Lintner model and apply it to a sample of U.S and Japanese corporations For the time period 1982-1993, they find that U.S managers smooth out the dividends even more compared to the

period of 1946-1964 covered in Fama and Fabiak’s (1968) study. However, the authors find that Japanese corporations are more willing to omit dividends and follow a relatively less stable dividend polic y compared to their U.S counterpart Based on an extensive analysis of changes in dividends, Bernatzi, Michaely and Thaler (1997) conclude that Lintner’s model remains the best description of the dividend setting process available. Brav et al (2003) report that in line with Lintner’s findings, managers express a strong desire to avoid dividend cuts, except in extreme circumstances. However, Brav et al (2003) report that in contrast to the Lintner era, managers are more reluctant to increase dividends in tandem with earnings increases and they no longer view the dividend payout ratio as the primary decision variable. The issue of “Dividend Relevance” is raised most poignantly in Modigliani and Miller (1961). In a seminal article, the authors argue that under conditions of perfect

capital markets the dividend payout rate is irrelevant to the value of the firm. Perfect capital markets can be summarized as markets where the following conditions exist: • • • • • • Information is costless and available to everyone equally; No distorting taxes exist; Flotation and transaction costs are non-existent; No contracting or agency costs exist; No investor or firm individually exerts enough power in the market to influence the price of a security; Investors are not systematically irrational. Much of the controversy surrounding Modigliani and Miller’s (1961) paper lies in the fact that it flies in the face of conventional wisdom, which suggests that dividend policy is critical to the value of a firm. Over the years researchers have generally offered four common explanations for dividend relevance, namely, the bird-in-the-hand, signaling, tax preference, and agency explanations. The “Bird-in-the-Hand” explanation argues that a relationship exists between

firm value and dividend payout because dividends represent a sure thing for shareholders as compared to capital gains. Due to the fact that investors value dividends, which are supposedly less risky than uncertain capital gains, firms should set a high dividend payout ratio and offer a high dividend yield to maximize stock price. Bhattacharya (1979) argues convincingly that the reasoning underlying the bird-in-the-hand explanation for dividend relevance is fallacious. He argues that the risk of a firm is 3 Source: http://www.doksinet determined by the riskiness of its cash flows, which is not altered by the firm’s dividend policy. Bhattacharya (1979) argues that an increase in dividend payout today will simply result in an equivalent drop in the stock’s ex-dividend price. In general, the majority of the financial economics literature is rather dismissive of the bird-in-the-hand explanation for dividend relevance. The “Signaling Explanation” argues that if the assumption of

relevant information being freely available to all parties interested in a firm is violated in practice then dividend policy may be relevant to the value of the firm. In particular, information asymmetry may exist between outside investors and corporate managers, in that managers may have information relevant to the value of the firm that outside investors do not have. Managers may therefore use a change of dividend as a way to signal this private information and thus reduce information asymmetry. Ross (1977), Bhattacharya (1979,1980), John and Williams (1985), Miller and Rock (1985), and Ofer and Thakor (1987) all provide support for a signaling perspective on dividend policy. According to the signaling explanation, cash dividends announcements convey valuable information about management’s assessment of a firm’s future value that other means cannot fully communicate. In turn, investors may use dividend announcements as information to assess a firm’s stock price. A major

drawback to this explanation, however, is that no one has been able to determine why dividends are better signals than other, less costly signals. Also, Black and Sholes (1974) and Ambarish et al (1987) provide evidence that suggests that it is unclear as to whether or not dividends send unambiguous signals send about a firm’s expected returns. Brav et al (2003) survey 384 CFOs and Treasurers in the U.S and find only moderate support for the signaling explanation of dividend relevance. In fact, the survey results suggest that while managers believe that dividends convey information to investors, the information conveyance does not appear to be related to signaling in the academic sense. However, on balance, in the financial economics literature, there is much theoretical and empirical support for the view of dividends as a signaling device, and it remains an influential theoretical framework for analyzing corporate dividend policy. The “Tax Preference” explanation focuses on one

of the most glaring violations of perfect capital markets, that is, the existence of distorting taxes. In many countries dividends and capital gains are taxed differently. In particular, dividends are generally taxed more heavily than capital gains. Given this fact, rational investors may prefer a low dividend payout to a high payout. The conventional view has been that taxes affect shareholders’ optimal dividend payout ratios because dividends result in an immediate tax liability for shareholders, while if the funds are retained and reinvested in the firm, they result in capital gains that gain preferential tax treatment. This view suggests that firms should keep dividend payouts low if they want to maximize shareholder value. From the tax preference viewpoint arises the idea of clienteles (Gordon and Bradford (1980)). Some important groups of shareholders may prefer dividends to capital gains because dividends provide cash flow and, for these shareholders, there is little or no tax

advantage to capital gains. The most important group is non-taxable institutions, but individuals with low marginal tax rates and other corporate shareholders are also in the low tax clientele. These shareholders will own stock in firms with high dividend payout ratios while other shareholders (the high tax clientele) will invest in firms with low dividend payout. Empirical evidence regarding the clientele hypothesis 4 Source: http://www.doksinet has been mixed. Studies that find a clientele effect include Petit (1977), Gordon and Bradford (1980), and Scholz (1989). Studies finding contradictory evidence include Long (1978), Litzenberger and Ramaswamy (1979, 1982), Hess (1982), Auerbach (1983), Poterba and Summers (1984), Poterba (1987) and Blume and Friend (1987). King (1977), Auerbach (1979) and Bradford (1981) present a newer view of dividend taxation, which holds that taxes are irrelevant because dividend taxes get capitalized in the value of the firm. The major drawback to

this theory is that it fails to deal adequately with the possibility of periodic share repurchases. Also, Poterba and Summers(1984) provide empirical evidence contradicting this view. Despite a plethora of empirical studies, the evidence on the tax-preference explanation of dividends is rather inconclusive. This may in part reflect the complexity of tax systems and the varying marginal tax rates among investors. In a comprehensive survey, Brav et al (2003) find that the role played by taxes in determining payout policy is only of second-order importance. It is rather striking that more than two-thirds of the executives surveyed say that the elimination of dividend taxation, as proposed by the Bush administration, would definitely not affect their dividend decisions. The “Agency” explanation for dividends focuses on another possible violation of perfect capital markets, that being the possible existence of agency and contracting costs. This view is most closely associated with the

work of Jensen and Meckling (1975), and the extensions by Rozeff (1982) and Easterbrook (1984). This theory derives from potential conflict of interests between corporate managers (agents) and outside shareholders (principals). For example, management may consume excessive perquisites out of undistributed corporate earnings and invest the retained earnings sub-optimally. This conflict leads to agency costs. Crockett and Friend (1988) argue that shareholders are not sufficiently well informed to know whether or not management is acting in their best interests. Agency theory posits that the dividend mechanism provides an incentive for managers to reduce the costs related to the principal/agent relationship. One way to reduce agency costs is to increase dividends. Paying larger dividends reduces the internal cash flow subject to management discretion and forces the firm to seek more external financing. Raising costly capital outside subjects the firm to the scrutiny of the capital market

for new funds and reduces the possibility of sub-optimal investment. This monitoring by outside suppliers of funds of capital also helps to ensure that managers act in the best interest of shareholders. Thus dividend payments may serve as a means of monitoring or bonding management performance. Several empirical studies provide support for the agency explanation for dividends. Rozeff (1982) finds a negative relationship between dividend payout and the percentage of insiders. Crutchley and Hansen (1989 and Moh’d, Perry and Rimbey (1995) conclude that managers make financial policy tradeoffs such as paying dividends to control agency costs. However, Brav et al (2003) in their survey of executives find only moderate support for agency costs explanations of dividend policy, in contrast to the extensive influence the agency explanation enjoys in academic circles. As suggested in the introduction to this paper, the literature on dividend policy has largely focused on developed countries.

The relatively few studies focusing on dividend policy in developing country firms have produced somewhat mixed results. In a survey study, Isa (1992) finds that firms in Malaysia follow stable dividend policies along the lines suggested by the Lintner model. Kester and Isa (1996), Annuar and Shamser 5 Source: http://www.doksinet (1993) and Gupta and Lok (1995) also provide evidence supporting the validity of the Lintner model in explaining the dividend policy of Malaysian firms. However, Pandey (2001) tests the Lintner model for firms listed on the Kuala Lumpur Stock Exchange and reports that Malaysian firms follow less stable dividend policies than developed country firms. A study by Pandey and Bhat (1994) in India supports the Lintner findings and reveals that Indian managers maintain an uninterrupted record of dividend payments and also try to avoid abrupt changes in their dividend policies. Ariff and Johnson (1994) confirm Lintner’s model for firms in Singapore. Adaoglu

(2000) tests the Lintner model on a sample firms listed on the Turkish stock exchange. Adaoglu reports that the firms follow unstable cash dividend policies and the main factor that determines the amount of cash dividends is the earnings of the corporation in that year. The suggestion is that when one moves away from developed countries the robustness of the Lintner model in describing corporate dividend policy is somewhat reduced. Glen et al. (1995) conducted a pioneering study of dividend policy in seven developing countries, namely, Chile, India, Jamaica, Mexico, the Philippines, Thailand and Turkey. The study finds that emerging market firms retain, on average, substantially more of their earnings than do their developed country counterparts. The dividend payout ratio, for a composite of all emerging markets followed by the IFC’s Emerging Market Database, ranged from 30 to 40 percent over the period 1986 to 1994. By comparison, a global composite index of developed countries had

a payout ratio of 66% in 1993, substantially higher than the level for any year tracked by the emerging markets index. Among the major developed markets there were no substantial differences, the U.S and Japan had similar ratios of 61 and 62 percent respectively whereas the UK had a ratio of 72 percent. The authors are, however, unable to explain this phenomenon In contrast to financial managers in developed countries, the study also suggests that financial managers in developing countries are less inclined to smooth earnings than their developing country counterparts (the empirical tests of the Lintner (1956) model in developing countries reported earlier in this section, provides some support for this notion). The authors concluded that developing country firms tend to adopt a target dividend payout ratio and attempt to maintain dividend payments in line with that target ratio regardless of short-term volatility in earnings. However, while the firms have a target dividend payout

ratio they are generally less concerned with volatility in dividends over time, and consequently dividend smoothing over time is less important. As a result dividend payments tend to be more volatile in emerging markets than in developed countries. This preliminary and largely exploratory literature suggests that there may be major differences between developed and developing country firms in terms of dividend policy, which may be a phenomenon well worth exploring in more detail. A more recent issue in the literature on dividend policy is the notion that dividends are becoming less important as a means for firms to return cash to investors. In an influential paper, Fama and French (2000) noted the phenomenon of “disappearing dividends.” In this paper the authors note that the percentage of firms in the US paying cash dividends fell from 66.5% in 1978 to 208% in 1999 Fama and French (2000) argued that this was not only due to a tilt in the population of publicly listed firms

increasingly towards small firms with low profitability and strong growth opportunities, characteristics typical of firms that have never paid dividends, but also to a declining propensity to pay dividends by all types of firms. Brav et al (2003) in their survey of 6 Source: http://www.doksinet corporate managers in the U.S report that the large majority of managers view share repurchases as a more flexible means of returning cash to investors than cash dividends and share repurchases are now used extensively as an alternative or supplement to cash dividends. The authors also report that among firms in their sample that don’t pay dividends, 70 percent say they never plan to initiate dividends, and among firms that currently pay dividends, the overwhelming majority state that they if they had not started paying dividends they would currently use repurchases instead of dividends as a means of returning cash to investors. In conclusion, while dividends appear to be

“disappearing” in developed countries, the “dividend puzzle” noted by Black (1976) remains a major unresolved issue in the corporate finance literature. While the dominant academic model suggests that under conditions of perfect capital markets dividend policy is irrelevant to the value of the firm, the majority of financial managers and investors appear to believe that dividend policy is relevant and stable dividends are preferred. The dominant academic explanations for dividend relevance, namely, information asymmetry, agency costs and tax preferences of explanations have at best a mixed record. The Lintner (1956) model appears to be a robust explanation of the manner in which managers in developed countries determine their dividend payments. However, the limited research from developing countries suggests that dividend policy is less stable than in developed countries and the Lintner (1956) model may not be as widely applicable. DATA AND METHODOLOGY This paper combines an

analysis of historical data on dividend payments of publicly listed companies in Barbados, a test of the Linter (1956) model in the Barbados context and a survey of CFOs in Barbados using a questionnaire and structured interviews. It is hoped that this combination of methodologies will result in a clearer picture of corporate dividend policy in Barbados. Historical Data on Dividend Payments Reliable information on the dividends per share and earnings per share for publicly traded firms in Barbados is available for the period 1985-2001 in the form of the annual Ernst and Young performance report. The study therefore limits itself to this time period. Over this time period, there were 25 Barbadian companies listed on the Securities Exchange of Barbados (SEB). Of the 25 companies traded over this period only 17 companies had at least a five-year history of dividend payments, which is generally viewed as a minimum for assessing dividend policy. The reason for this exclusion is to have

enough years of nonzero cash dividends for empirical analysis. The quantitative analysis in the paper therefore confines itself to these 17 companies. This information serves to provide factual information describing corporate dividend policy in Barbados and allows one to test the Lintner (1956) model in the Barbadian context. Specification of the Lintner Model Lintner (1956) built the following behavioural model in light of his survey findings: * D i,t = T i P i,t 7 (1) Source: http://www.doksinet D i,t – D i,(t-1) = a i + c i (*D i,t - D i,(t-1) ) + u i (2) The change in cash dividends (eq.2) depends on the difference between the targeted dividend payments (*D i,t ) and the actual dividend payments last period (D i,(t-1) ). The positive “a i “ intercept shows the reluctance of corporations in decreasing the dividend and their preference for a gradual growth in dividends. Coefficient “c i ” indicates the stability in dividend changes and is also the adjustment factor

towards the target payout ratio (T i,t ). The adjustment factor shows the level of management response in dividends to changes in the level of earnings (P i,t ). The higher the value of the adjustment factor, the lower the level of dividend smoothing. For the adjustment factor, a value of 1 indicates that the firm does not smooth dividends at all and a value of 0 indicates that the firm follows a maximum dividend smoothing policy. By combining equations (1) and (2) without affecting the error term, Lintner tests the following empirical model: D i,t = a i + b P i,t + d D i,(t-1) + u i (3) Where b = cT and d = 1-c The preceding model in equation (3) is modified to test for stability in the dividend policy of the companies listed on the SEB. As is the standard practice in the financial economics literature, the Lintner model is modified as per Fama and Fabiak (1968), and estimated as: DPS i,t = α 1 + β 1 EPS i,t + β 2 DPS i,(t-1) + µ I,t (4) In equation (4) DPS stands for dividends

per share and EPS stands for earnings per share. The per share values replace the variables in the original Lintner formulation to account for frequent capital increases and bonus dividend issues by firms. Estimation Methodology As has become common in recent years, panel data regressions are utilized to test the Lintner model for companies listed on the SEB. In panel data (pooled) regression, time-series and cross-sectional observations are combined and estimated. Gujarati (1995) states that in the case of a properly specified model, pooled regression can provide more efficient estimation, inference and even forecasts. Greene (1997) states “the fundamental advantage of panel data set over a cross section is that it will allow the researcher far greater flexibility in modeling differences in behavior across individuals.” Pindyck and Rubinfield (1998) also add that “incorporating relating to both cross-section and timeseries variables can substantially diminish the problems that

arise when there is an omitted variables problem. If there is no missing data for the cross sectional units over the sample period, the sample is a called a balanced sample in a panel data setting. However, the sample can also be unbalanced where there are some missing data for some of the cross-sectional units over the whole sample period. Due to the fact that the 17 companies with at least a 8 Source: http://www.doksinet five-year history of dividend payments were listed on the SEB at different times and some were subsequently de-listed, merged or acquired, the sample is unbalanced. For the 1985 to 2001 period the sample is unbalanced and the unbalanced number of pooled observations is 260. Questionnaire Survey and Interviews The cover letter requesting participation in this study, along with a stamped return envelope and the survey instrument, was first mailed to the Chief Financial Officer of the eighteen Barbadian firms listed on the Securities Exchange of Barbados in August

of 2003. A second mailing was sent in January of 2004 and by the end of March 2004, all 18 firms had responded to the survey. The survey instrument also elicited a broad range of demographic data about publicly listed firms in Barbados. As part of the study, fifteen one-on-one interviews with Chief Financial Officers were conducted. The interviews complement the survey information along several dimensions Interviews allow the use of open-ended questions, so the respondent’s answers can dictate the direction of the interview (versus pre-chosen questions in the survey). Interviews also allow for give- and- take and clarifications, which are not possible with a traditional survey. A major limitation of the study stems from the small number of listed companies in Barbados. The fact that there are only eighteen firms effectively eliminates the value of statistical analysis of responses to the survey questions in terms of various categories, due the small sizes of the groups providing

particular responses. However, due to the fact that fifteen of the eighteen publicly listed firms in Barbados participated in the study, the fact that additional clarifications were generated through the interviews and the use of supporting historical data on dividend payouts, the study is likely to represent a thorough analysis of corporate dividend policy in Barbados, despite the lack of statistical sophistication. FIRM STRUCTURE AND MARKET MICROSTRUCTURE IN BARBADOS In this section some likely differences between publicly traded firms and their business environment in Barbados and the firms typically studied in the finance literature are presented. The author contends that the major differences relate to firm size, sectoral distribution of firms, the degree of shareholder concentration, the level of stock market microstructure development and taxation policy. These features of equity trading and publicly listed companies in Barbados may conceivably condition managers and

shareholders attitudes to dividends thereby influencing the dividend policies of firms. These features are therefore elaborated upon in this section. Size of Publicly listed Firms in Barbados Of the publicly listed firms in Barbados, three had sales of less than $12.5 million, five had sales of between $12.5 and $49 million, six had sales of between $50-$249 million and four had sales of between $250 and $450 million. Using standard international classifications 2, all of the public listed firms in Barbados would be classified 2 Firms with sales of USA $ 1billion or more are usually classified as large, those with sales between $500 million and $ 1billion as medium sized and firms with sales of less than $500 million are classified as small. 9 Source: http://www.doksinet as small firms. Brav et al (2003) report that 51% of the firms in their sample of US firms had sales of over U.S one billion dollars, while Brounen et al (2004) report that 25% of European firms have sales over

U.S one billion dollars Publicly listed firms in Barbados are therefore typically smaller than their counterparts in the US and Europe and this difference in relative size may well create some differences in dividend policy. Sectoral Distribution of Firms and Foreign Sales The majority of publicly listed firms in Barbados are in the retail and wholesale sectors, unlike the case of the U.S and Europe where the majority of publicly listed firms are in the manufacturing sector. Seven of the eighteen publicly listed firms in Barbadian were in the Retail/Wholesale sector, four were in Banking/Finance/Insurance, three were in manufacturing, two Tourism, and one each in Transport/Energy and Communication. Again, this difference in sectoral distribution may well create some differences in dividend policy. Concentrated Share Ownership and Close Investor/Company Relationships In the case of publicly listed companies in Barbados, share ownership tends to be relatively highly concentrated. As the

first column of table 1 suggests, the large majority of firms listed on the SEB are rather closely-held. The information in table 1 suggests that more than 40% of the shares being held by the five largest investors for 12 of the 17 listed companies (if anything table 1 is likely to understate the extent of ownership concentration). The high shareholder concentration may simply be reflective of the relatively small size of the firms in Barbados, but it may also reflect the fact that many of the firms started life as family owned enterprises and the founding family has maintained a significant share ownership. As these firms have evolved, large institutional investors and other firms as opposed to the public at large have developed significant shareholdings. The second column of table 1 provides an example of the role of a large institutional investor, the Barbados Mutual Life Assurance Company of Barbados (now Sagicor Corporation). At the end of fiscal year 2000, Sagicor Corporation

held 5% or more of the shares of 11 of the 17 companies listed on the SEB. Another reason for the concentrated shareholdings may lie in the fact that the conglomerate structure is popular in Barbados and this is borne out by the fact that a number of the firms listed on the SEB are associated through share ownership. For example, at the end of fiscal year 2001, the largest conglomerate Barbados Shipping and Trading owns shares in the following listed companies, Banks Holdings Ltd. (24%) and Almond Resorts (37%). In turn, Banks Holdings Ltd owned 84% of the shares of another listed company, Barbados Dairy Industries Ltd, while Goddard Enterprises Ltd. owned at least 56% of West Indies Rum Distillery Ltd. The extent of cross shareholding is reflected in the fact that nine of the fifteen firms that responded to this question on the survey had their largest shareholder listed on a stock exchange. The available evidence also suggests that these large investors are very well represented on

the board of directors of the listed firms in which they have a shareholding (Beckles (1987) and Barrow (1974)). In addition the firms and investors often enjoy close relationships. The firms use each other as subcontractors or suppliers of parts or products Human ties and interlocking directorates further cement these relationships. 10 Source: http://www.doksinet These two factors suggest that publicly listed firms in Barbados are generally closelyheld by a few dominant investors who are typically present on the board of directors. Given the dominance of large investors who are typically represented on the board of directors and/or involved in management, agency costs would appear to be less of an issue in the Barbadian context than suggested in the corporate finance literature to date. Also, there is likely to be a prevalence of direct channels of communication between firms and major shareholders. This suggests that the information asymmetry problem noted in the corporate

finance literature is not likely to be especially severe for the large majority of investors in the Barbadian context. The fact that a significant number of shareholders are typically present on the board of directors and/or management team or are closely related to such raises the possibility that shareholders may have other sources of cash flow from the firm other than dividends or capital gains in the form of salaries and other perks. If this is so, it may also impact on the dividend policy desired by shareholders. Ballon and Tomita’s (1988) characterization of the Japanese investor as being neither an outsider, nor an individual, nor particularly concerned by the short-term may be a reasonably apt description of the dominant investors in Barbados. Following Ballon and Tomita (1988), the Barbadian firm listed on the SEB may be construed as having four types of shareholders: (1) Stable, family and related company shareholders who are generally well informed about the firms’

prospects, have a prominent monitoring role and are unlikely to sell their shares for capital gains in the regular course of affairs and may thus prefer a stable dividend; (2) Institutional investors who are generally well informed about the firms’ prospects, have a prominent monitoring role and are unlikely to sell their shares for capital gains in the regular course of affairs and may thus prefer a stable dividend; (3) Employee shareholders (largely through bonus shares) who are likely to be relatively well informed about the firms’ prospects, not inclined to sell their shares and would like steadily increasing dividends; (4) A minority of “external investors” who are unlikely to be well informed about the firms’ prospects (may suffer from asymmetric information), do not have a prominent monitoring role (may suffer agency costs), are flexible in terms of selling their shares and are likely to be happy with steadily increasing dividends. Thin Trading and Absence of Share

Repurchases Despite the efforts of the SEB and successive governments in Barbados, trading on the SEB is extremely thin, and in the absence of major corporate restructurings such as mergers and acquisitions, only a very small percentage of shares change hands each year (see table 2). This may in part be due to the relatively underdeveloped market microstructure. Transactions costs are relatively high, there a limited number of trading days on the exchange (for most of its life the exchange only traded two days a week, a third day has recently been added) and market makers are conspicuously absent. 11 Source: http://www.doksinet However, the low trading levels may also be influenced by the pattern of shareholding described in the previous sub-section. Also, to date share repurchases by publicly listed firms have not occurred in Barbados. This is in sharp contrast to a number of other countries where share repurchases have become a major alternative to dividends as a means of

returning cash to shareholders. In fact, this phenomenon has become so pervasive that the notion of “disappearing dividends” has emerged as a major issue in the finance literature. The combination of thin trading and the absence of share repurchases suggests that while equity investors in developed countries (and a number of developing countries as well) rely on a mix of capital gains, dividends and share repurchases as sources of cash from their equity investments, equity investors in Barbados are heavily reliant on dividend payments as a source of cash from their equity investments, which may have implications for the dividend policy desired by investors. Taxation Policy In Barbados There is no capital gains tax in Barbados and dividends are taxed at a rate of 12.5%, while normal income above $25,000 is taxed at 40%. This indicates the presence of distorting taxes Barbados, which may lead to an investor preference for capital gains over dividends and dividends over cash flows in

the form of salaries. Historical Analysis of Dividend Payout in Barbados Dividend Payout Ratios Over the sample period, the average annual dividend payout ratio for Barbadian companies listed on the SEB ranged between 45% and 321% 3 (see tables 3 and 4). The payout ratios for publicly listed companies in Barbados appear to be higher than the 3040% reported by Glen et al (1995) in their survey of ten developing country stock markets and closer to the 66% average reported for developed countries. The variations in the payout ratios among Barbadian firms appear to be driven primarily by changes in the EPS. A casual observation of the data suggests that dividends per share (DPS) is rather more stable than earnings per share (EPS). Over the sample period the average DPS ranges between 6.64 cents and 97 cents, while the average EPS ranges between 13.5 cents and 221 cents This raises questions as to whether or not management in fact targets a particular payout ratio as suggested in the

theoretical literature, or a particular DPS. Dividend Changes Table 5 classifies dividend payments in terms of whether or not the DPS for the year represented a constant DPS, an increase in the DPS, a decrease in the DPS, an omission of the dividend payment or an initiation of dividend payments. It is quite noticeable that dividend omissions are extremely rare among publicly listed firms in Barbados. Dividend omissions only occurred in 1993 and 2001 In 1993 two of the sixteen listed firms omitted dividends, while in 2001 one of the 18 listed firms omitted 3 The very high payout ratios in some years, arises from companies paying dividends even when earnings were negative (in which case the payout ratio was capped at 100%) and paying stable dividends per share in the face of significant declines in earnings per share. 12 Source: http://www.doksinet dividends. The issue of the disappearing dividend raised in the financial economics literature does not yet appear to apply in the

Barbadian context. The data also suggests that firms are relatively reluctant to cut dividends, even when faced with a decline in earnings. For example, in 1990 while 10 firms recorded a decrease in EPS, only 1 firms reduced DPS, again in 1999, 10 of the 18 firms recorded a decline in EPS but only five firms reduced DPS. In a number of cases the DPS was unchanged even though the EPS had changed. A casual analysis of the data therefore suggests that dividends are alive and well in Barbados, dividend policy is relatively sticky and managers attempt to smooth dividends. THE LINTNER MODEL AND DIVIDEND POLICY IN BARBADOS The Lintner model stands out as a rigorous model for testing the stability and regularity of dividend policy. Table 6 presents the results of the estimation of the Lintner model for publicly listed firms in Barbados. The intercept term is positive and statistically significant indicating the reluctance of Barbadian firms to avoid payment of dividends. The regression

coefficients of current earnings (EPS i,t ) and past dividends (DPS i,(t-1) ) are highly significant. But the larger coefficient and the associated t-statistic of DPS i,(t-1) imply the greater importance of past dividends in determining the dividend payment, as compared to the current level of earning The computed target payout ratio is 33%, which is somewhat lower than the sample dividend payouts over the sample period. The speed of adjustment is 0.48 indicating that is significant level of dividend smoothing The data analysis therefore suggests that dividends are in no immediate danger of disappearing in Barbados, and financial managers in Barbados like their developed country counterparts engage in dividend smoothing and follow stable dividend policies along the lines suggested by Linter (1956). The results appear to provide further support for the robust nature of the Lintner model in explaining the manner of setting dividend payments across a variety of countries and firms.

However, the results raise questions as to whether or not management in fact targets a particular payout ratio as suggested by Lintner or a particular DPS. These findings are subjected to further analysis and probing with the aid of a questionnaire survey and interviews of financial managers in Barbados, the results of which are presented in the following section. SURVEY RESULTS AND DISCUSSION The Relationship Between Dividend Policy and Firm Value A major issue in the finance literature is the relevance of dividend policy to the value of the firm. According to Modigliani and Miller (1961) under conditions of perfect capital markets dividend policy is irrelevant to the value of the firm in that it affects neither the firm’s stock price nor its cost of capital. Four statements on the survey instrument seek to elicit the views of corporate managers in Barbados on this issue. The survey responses suggest a strong belief among financial managers in Barbados that dividend policy is

relevant to the value of the firm in that they believe it affects both a firm’s stock price and its cost of capital 4. Thirteen of the responding firms (76%) expressed agreement with the notion that “a change in dividend policy affects the value of the firm,” and 82% agree with the statement, “a firm’s dividend policy affects its cost 4 This view is also strongly expressed in the interviews. 13 Source: http://www.doksinet of capital.” In line with these views, sixteen (94%) of the respondents agree that “an optimal dividend policy strikes a balance between current dividends and future growth that maximizes stock price,” and 71% of respondents agree that “ a firm should formulate its dividend policy to produce maximum value for shareholders.” These views are consistent with the findings of Baker, Farrelly and Edelman (1985), Baker and Powell (1999) and Brav et al (2003) in their surveys on management views about dividend policy in the USA. Interpreting these

views in terms of modern finance theory would lead one to suggest that corporate managers in Barbados believe that one or more of the assumptions of perfect capital markets are consistently violated in the Barbadian context. In the finance literature a number of explanations have been advanced for dividend relevance in light of the Modigliani and Miller (1961) position. These explanations are namely, the bird-in-the-hand, signaling, tax preference, and agency explanations. The Bird-in-the-hand Explanation Two statements on the survey instrument can be seen as reflecting the Bird-in-thehand explanation of dividend relevance, and the responses to these statements suggest strong support for the Bird-in-the-hand explanation for dividend relevance among corporate managers in Barbados. In their responses, 77% of respondents agree that “investors prefer a certain dividend stream to uncertain price appreciation” and 88% agree with the statement, “investors prefer certain, current

dividends to possibly higher but riskier future dividends.” This is in contrast to Baker and Powell (1999) who find mixed responses to these statements among corporate managers in the USA. In fact, during the interviews conducted, financial managers in Barbados were rather quizzical of the notion of capital gains and dividends being substitutes for each other and had a strong sense of dividends being a reward for investing, quite separate and distinct from capital gains. The finance literature has tended to be rather dismissive of the Bird-in-the-hand explanation for dividend relevance, and this view appears to be somewhat mirrored by the survey responses in the USA, which makes the responses among corporate managers in Barbados rather interesting. This may well be due to differences in institutional context relating primarily to the nature of shareholding in Barbados. As discussed in section 4, for a variety of reasons, the majority of shareholders (dominant family shareholders,

institutional shareholders and employee shareholders) in Barbados rarely sell their shares in the absence of a major corporate restructuring such as a merger or takeover, and firms have never engaged in share buyback programs. Therefore, in such an environment dividends remain the major source of cash from equity investments and investors are unlikely to view capital gains and dividends as substitutes, which may well foster the view of dividends as the proverbial Bird-in-hand, and explain the reluctance to omit or cut dividends in Barbados. The Signaling Explanation Six of the statements on the survey instrument explore information content or signaling effects of a firms’ dividend policy. Corporate managers in Barbados appear to believe that investors use dividends as signals about a firm’s future prospects and also use dividend announcements as information in valuing a firm’s shares. In their responses, 14 Source: http://www.doksinet 94% of respondents agree that

“investors regard dividend changes as signals about a firm’s future prospects” and of 88% respondents express agreement with the statement “investors use dividend announcements as information to assess a firm’s stock value.” Sixty percent of the respondents also agree with the statement “the market rewards unexpected increases in dividends,” while only 55.5% believe that the market punishes an unexpected decline in dividends. In line with these views 83% of respondents express agreement with the notion that “a firm should adequately disclose to investors its reasons for changing dividend policy.” However, 72% agree “dividend increases are ambiguous because they can suggest future growth or lack of investment opportunities.” These responses are consistent with the findings of Brav et al (2003) and Baker and Powell (1999) in their survey if corporate managers in the USA. It would appear obvious that the large majority of shareholders in Barbados have alternative

and likely more accurate sources of information about the prospects of the firms they invest in, than the potentially expensive and ambiguous signals sent by changes in dividends. Thus the notion that dividends matter in the valuation of firms in Barbados, because of their information content, is in the author’s view a rather contentious one. The discussions in the interviews suggest that while financial managers believe that dividends provide investors with information about the financial well-being of the firm, they do not see dividend payout policy as an expensive signal that can be used to correct information asymmetry. Therefore, while financial managers in Barbados appear to believe that dividend policy are an important source of information for investors they do not appear to subscribe to dividend policy being a signaling device in the academic sense 5. This finding is similar to Brav et al (2003) who find little support for both the assumptions and predictions of signaling

theories that are designed to explain dividend policy, at least not in terms of the conscious decisions executives make about dividend policy. The Tax-Preference Explanation Four statements on the survey address the tax-preference explanation for dividend relevance. The responses to the statements can be classified as ambiguous or mixed While, 94% of the respondents agree that “a firm should be responsive to the dividend preferences of its shareholders,” and 72% agree that “investors are attracted to firms that have dividend policies appropriate to that investors’ particular tax circumstances,” 83% disagree that “investors prefer that a firm retain funds over paying dividends because of the way capital gains are taxed as compared with dividends” and 72% disagree that “stocks that pay high (low) dividends attract investors in low (high) tax brackets.” These responses are similar to the findings of Baker and Powell (1999) and Brav et al (2003) in their surveys of

corporate managers in the USA. In Barbados, capital gains are not taxed while dividends are taxed at 12.5%, which gives a potential tax advantage to capital gains relative to cash dividends. Along the lines of Brav et al (2003) one can interpret the responses of corporate managers in Barbados to be suggesting that while they are aware of the tax disadvantages of dividends relative to capital gains, they believe that tax concerns are second order concerns for investors, relative to their desire for cash dividends. 5 The author tried to spell out the academic view of signaling to the managers and they expressed strong disagreement with that view. 15 Source: http://www.doksinet The Agency Explanation The survey contains two statements that provide an agency explanation for paying cash dividends. The majority of respondents provide little support for an agency explanation. This is, in the view of the author, not surprising given the ownership structure of publicly listed firms in

Barbados. Major shareholders in publicly listed firms in Barbados often enjoy potentially powerful monitoring rules due to their presence on the Board of Directors which creates little incentive for the use of alternative monitoring devices. Again, one should note that the responses provide little support for both the assumptions and predictions of one of the leading academic theories of dividend policy. Setting Dividend Payments In terms of the manner in which corporate managers go about setting dividend payments, Lintner’s (1956) paper remains the most authoritative study to date. There are two key results from Lintner’s (1956) study. Firstly, the starting point for most payout decisions was the payout ratio (dividends as a percentage of earnings) and secondly corporate dividend decisions were made very conservatively, especially reflected in the reluctance of management to cut dividends. The Linter (1956) model has been tested and re-tested over the years in a range of countries

and while the model has proven to be a robust one in describing dividend policy among developed country firms, the results in terms of dividend policy among developing country firms have been somewhat mixed. Table 6 presents the results of the estimation of the Lintner (1956) model for publicly listed firms in Barbados. The results therefore suggest that publicly traded firms in Barbados engage in a dividend smoothing and follow stable dividend policies along the lines suggested by Linter (1956). The responses to the survey questions and subsequent interview, however, appear to provide a richer view of the dynamics of setting dividend payments in Barbados. The responses from corporate managers in Barbados suggest that unlike the postulations of the Lintner (1956) model the payout ratio might not be the starting point for payout decisions. While 76.4% of respondents agree that a firm should have a target dividend payout and gradually work towards it, 66% of the firms had no target or a

flexible target range while only one firm reported a strict target range and 94% agree that the market places greater value on stable dividends than stable dividend payouts. In the subsequent interviews, the large majority of financial managers generally agree that they in fact target the DPS rather than a payout ratio. These responses and views in Barbados, contrast with those of Baker and Powell (1999) but are consistent with those of Brav et al (2003) in their surveys of corporate managers in the USA. Brav et al (2003) report that corporate managers in the USA appear to have changed this aspect of setting dividend payments compared to the time of Lintner’s (1956), in that the dividend per share instead of the payout ratio appears to be the variable now targeted by corporate managers in the USA. However, the survey responses do appear to confirm a belief in dividend conservatism along the lines of Lintner (1956). Ninety-four percent of respondents agree that “ the market places a

greater value on stable dividends than stable payout ratios,” 88% of respondents agree that “a firm should avoid changing its regular dividend if that change might have to be reversed in a year or so,” and 94% agree that “a firm should 16 Source: http://www.doksinet strive to maintain an uninterrupted record of dividend payments.” These responses are consistent with the findings of Baker and Powell (1999) and Brav et al (2003). It thus appears that Lintner’s notion of a sticky dividend policy is relevant in Barbados, but unlike Lintner’s postulation the variable targeted appears to be the DPS rather than the dividend payout. SUMMARY AND CONCLUSION This paper seeks to add to the nascent literature on corporate financial management in developing countries by undertaking an in-depth study of dividend policy among publicly traded firms in Barbados. The paper utilizes historical financial data on dividend payments in Barbados over the period 1985 to 2001 along with a

questionnaire survey and interviews of financial managers in Barbados. The paper finds that all publicly listed firms in Barbados pay dividends and are committed to continue such, and none of the firms have utilized share repurchases as a means of returning cash to shareholders. Thus the phenomenon of “disappearing dividends” and the increased popularity of share repurchases noted in developed markets does not appear to have yet spread to publicly traded firms in Barbados. The paper also finds that like their counterparts in developed markets, corporate managers in Barbados believe that dividend policy affects the value of the firm. However, corporate managers in Barbados tend to favour the Bird-in-the-had explanation for dividend relevance while corporate managers in the developed markets tend to favor signaling and agency costs explanations for dividend relevance. Finally, the paper finds that while corporate managers in Barbados exhibit conservatism in their dividend policies,

especially in the form of a strong reluctance to cut or omit dividends along the lines suggested by Lintner (1956), they depart from Lintner (1956) in that the dividend payout ratio does not appear to be the variable targeted when setting dividend policy. This is similar to the views expressed by corporate managers in the developed markets. The research findings therefore suggest that financial managers in Barbados depart from their developed country counterparts in terms of their strong commitment to paying dividends and a tendency to favour the Bird-in-the-hand explanation for dividend relevance. It is the author’s contention that these differences stem primarily from differences in the ownership and the corporate governance structure of Barbadian firms, and to a lesser extent the stock market microstructure in Barbados as compared to the U.S The perpetuation of closely held firms tends to lead to thin trading (and possibly retards the development of the market microstructure) and

a consequent difficulty in realizing capital gains in the absence of major corporate restructurings, making dividends the main means of generating cash from an investment in stocks in Barbados thereby fostering a Bird-in-the-hand view of dividends. Also, the fact that major shareholders are often present as members of the board of the companies they invest in and enjoy close personal and business relationships with these key personnel of these firms, serves to reduce information and agency costs and a need for alternative monitoring devices. 17 Source: http://www.doksinet Table 1: Share Ownership Concentration on the SEB Company Almond Resorts Inc. A.S Bryden’s & Sons (B’dos) Ltd. Barbados Farms Limited Barbados Shipping and Trading Co. Ltd. Barbados Dairy Industries Ltd. Banks Holdings Ltd. Bico Ltd. Cave Shepherd & Co, Ltd, Cable & Wireless Bartel Cable & Wireless BET CIBC Courts (B’dos) Ltd. Goddard Enterprises Ltd. Life of Barbados Ltd. Light and Power

Holdings Ltd. The West India Biscuit Company Ltd. % of shares held by the five largest shareholders Sagicor % Ownership 76 15 46 18 27 7 45 32 84 NA 37 54 .61 NA 53 6 74 NA 74 91 NA NA 87 NA 33 6 43 23 NA 6 NA NA 7 NA 6 The annual report does not list any investors as owning five percent or more of the firm’s shares. The annual report lists five organizations as owning 5% or more of the firm’s share capital but the amounts are not disclosed. 7 18 Source: http://www.doksinet Table 2: Level of Trading On The SEB 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Total Volume of Shares Outstanding (Mls) 155 178 181 198 195 210 234 239 254 228 256 303 % Traded 0.76% 069% 060% 105% 085% 135% 091% 070% 100% 143% 195% 079% 0 Table 3: Dividend Payout Year Number Average Median Minimum Maximum Range Standard Of Firms Payout Payout Payout Payout Payout Deviation 1985 13 59.87% 6452% 7.94% 13193% 12399% 32.83% 1986 13 44.63% 4444% 7.91% 75.00%

67.09% 19.22% 1987 13 44.52% 4054% 15.37% 93.66% 78.29% 20.27% 1988 14 44.85% 3582% 20.66% 78.95% 58.29% 20.34% 1989 14 45.66% 3880% 22.39% 75.52% 53.13% 17.97% 1990 14 52.70% 5149% 29.85% 10000% 70.15% 19.60% 1991 13 52.16% 4500% 29.59% 10257% 72.99% 23.11% 1992 15 65.11% 6145% 31.25% 12128% 90.03% 26.81% 1993 17 45.71% 4584% 0.00% 11770% 11770% 27.82% 1994 18 58.27% 4435% 0.00% 18750% 18750% 43.81% 1995 19 101.23% 5012% 0.00% 60714% 60714% 14761% 1996 18 52.21% 4448% 32.57% 10061% 68.04% 17.75% 1997 18 53.03% 4377% 27.79% 10000% 72.21% 22.46% 1998 18 321.87% 4287% 18.47% 500000% 498153% 116768% 1999 18 53.82% 4809% 24.51% 10000% 75.49% 21.79% 2000 18 77.56% 4562% 18.59% 33333% 31474% 88.72% 2001 18 35.65% 3274% 0.00% 10044% 10044% 29.01% 19 Source: http://www.doksinet Table 4. Dividends per Share Year Number Average Median Minimum Maximum Range Standard Of Firms DPS DPS DPS DPS of DPS Deviation 1985 13 $0.0803 $00500 $00125 $02500 $02375 $00698 1986 13 $0.0814 $00555 $00185

$01920 $01735 $00516 1987 13 $0.0906 $00700 $00375 $01920 $01545 $00495 1988 14 $0.0872 $00713 $00375 $02250 $01875 $00504 1989 14 $0.0938 $00750 $00375 $02500 $02125 $00572 1990 14 $0.0960 $00825 $00375 $02750 $02375 $00617 1991 13 $0.0850 $00800 $00500 $01795 $01295 $00371 1992 15 $0.0683 $00600 $00250 $01795 $01545 $00364 1993 17 $0.0664 $00625 $00000 $01795 $01795 $00416 1994 18 $0.0759 $00700 $00000 $01800 $01800 $00450 1995 19 $0.0782 $00738 $00000 $01800 $01800 $00454 1996 18 $0.0787 $00775 $00150 $02000 $01850 $00477 1997 18 $0.0849 $00838 $00200 $02110 $01910 $00498 1998 18 $0.0877 $00750 $00200 $02260 $02060 $00518 1999 18 $0.0977 $00850 $00300 $02560 $02260 $00566 2000 18 $0.0913 $00710 $00200 $02810 $02610 $00616 2001 18 $0.0787 $00675 $00000 $02760 $02760 $00714 Table 5 Dividend Analysis Year Number Increases Increases Decreases Decreases DPS DPS Of Firms EPS EPS 1986 13 12 9 1 1 1987 13 10 9 3 0 1988 14 7 8 6 2 1989 14 8 7 6 2 1990 14 4 7 10 1 1991 13 5 3 8 4 1992 15 0 1

13 9 1993 16 10 8 5 3 1994 17 7 8 8 2 1995 18 13 9 5 2 1996 18 10 8 6 1 1997 18 12 13 6 1 1998 18 12 10 6 2 1999 18 8 7 10 5 2000 18 12 6 6 5 2001 18 8 5 10 8 20 Source: http://www.doksinet Table 6: Lintner Model Dependent Variable: DPS Method: Least Squares Date: 10/14/03 Time: 18:48 Sample(adjusted): 1 259 Included observations: 259 after adjusting endpoints Variable Coefficient Std. Error t-Statistic Prob. C 0.011115 0.003657 3.039038 0.0026 EPS 0.159684 0.017013 9.385907 0.0000 PREDPS 0.517763 0.043387 11.93365 0.0000 R-squared 0.670304 Mean dependent var 0.082721 Adjusted R-squared 0.667728 SD dependent var 0.051905 S.E of regression 0.029920 Akaike info criterion -4.169093 Sum squared resid 0.229166 Schwarz criterion -4.127894 Log likelihood 542.8975 F-statistic 260.2360 Durbin-Watson stat 1.833962 Prob(F-statistic) 0.000000 21 Source: http://www.doksinet Bibliography Adaoglu, Chait, 2000, ‘Instability in the Dividend Policy of the Instanbul Stock Exchange Corporations:

Evidence from an Emerging Market,” Emerging Market Review, 1 (3), 252-270. Allen, F., and R Michaely, 1995, “Dividend Policy,” in: R A Jarrow , V Maksimovic and W. T Ziemba, eds, Handbooks in Operation and Management Science: Finance, Vol.9, (North-Holland, Amsterdam) Allen, Franklin, A.Berbardo and I Welch, 2000, “A Theory of Dividends Based on Tax Clienteles,” Journal of Finance, 55 (6), 2499-2536. Amihud, Yakov and Maurizio Murgia, 1997, “Dividends, Taxes, and Signaling: Evidence from Germany,” Journal of Finance 52 (1), 397-408. Ariff, M. and L W Johnson, 1994, “Securities Markets and Stock Pricing: Evidence from a Developing Market in Asia, Singapore, Sydney and London,” Journal of Social Science and Humanities, 1 (2), 171-177. Bahttacharya, Sudipto, 1979, “Imperfect Information, Dividend Policy, and the ‘Bird in the Hand Fallacy,’” Bell Journal of Economics, 10(1) 259-270. Black, Fischer, 1976, “The Dividend Puzzle,” Journal of Portfolio Management,

2, 5-8. Brav, A. J Graham, C Harvey and R Michaely, 2003, “Payout Policy in the 21st Century,” Working Paper, Duke University. Brennan, Michael J. and Anjan V Thakor,1990, “Shareholder Preferences and Dividend Policy,” Journal of Finance 45 (4) 993-1019. DeAngelo, Harry, L. DeAngelo, and D Skinner,1996, “Reversal of fortune, Dividend Signaling and the Disappearance of Sustained Earnings Growth,” Journal of Financial Economics, 40, 341-371. DeAngelo, Harry, L. DeAngelo, and D Skinner, 2000, “Special Dividends and the Evolution of Dividend Signaling,” Journal of Financial Economics, 57, 309-354. DeAngelo, Harry, L. DeAngelo, and DSkinner, 2000, “Are Dividends Disappearing? Dividend Concentration and the Consolidation of Earnings, Working Paper, University of California.” Easterbrook, Frank H., 1984, “Two Agency-Cost Explanations of Dividends,” American Economic Review, 74 (4) 650-659. 22 Source: http://www.doksinet Fama, Eugene F. and H Babiak, 1968,

“Dividend Policy: An Empirical Analysis, Journal of the American Statistical Association 63(324), 1132-1161. Fama, Eugene, and K. French, 2001, “Disappearing dividends: Changing firm characteristics or lower propensity to pay?” Journal of Financial Economics 60, 3-43. Fama, Eugene, and K. French, 2002, “Testing Trade-off and Pecking Order Predictions about Dividends and Debt,” The Review of Financial Studies 15, 1-35. Glen, J.D, Y Harmiklias, RMiller, and S Shah, “Dividend Policy and Behaviour in Emerging Markets,” Discussion Paper No. 26, International Finance Corporation, 1995 Greene, W. H, 1997, “Econometric Analysis,” 3rd ed, (Prentice Hall, New Jersey) Graham, John R., and C Harvey, 2001, “The Theory and Practice of Corporate Finance: Evidence from the Field, Journal of Financial Economics 60, 187-243. Gupta, G.S and KS Lok,1995, “Dividend Behaviour in Malaysia,” Capital Market Review, 73-83. Jensen, Michael C.,1986, “Agency Costs of Free Cash Flow,

Corporate Finance, and Takeovers,” American Economic Review 76 (2), 323-329. Jensen, Michael C and W. H Meckling, 1976, “Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure,” Journal of Financial Economics, 3 (4) 305-360. Kester, George W. and Isa Mansor, 1996, “Dividend Policy in Malaysia: A Comparative Analysis,” Malaysian Journal of Economic Studies, 33(1), 33-48. Lintner, John,1956, “Distribution of Incomes of Corporations Among Dividends, Retained Earnings, and Taxes,” American Economic Review, 46 (2), 97-113. McDonald, J., B Jacquillant and M Nussenbaum, 1975, “Dividend, Investment and Financing Decisions: Empirical Evidence on French Firms,” Journal of Financial and Quantitative Analysis 10, 741-755. Michaely, Roni, Richard Thaler and K. Womack, 1995, “Price Reactions to Dividend Initiations and Omissions: Overreaction or Drift?,” Journal of Finance 50 (2), 573-608. Miller, Merton., 1987, “The Information Content of

Dividends,” J Bossons, R Dornbush and S. Fischer, (eds) Macroeconomics: Essays in Honor of Franco Modigliani (MIT Press: Cambridge MA) 37-61. Miller, Merton and Franco Modigliani,1961, “Dividend Policy, Growth and the Valuation of Shares,” Journal of Business, 34, 411-433. 23 Source: http://www.doksinet Miller, Merton and Kevin Rock,1985, “Dividend Policy Under Asymmetric Information,” Journal of Finance 40 (4), 1031-1051. Myers, Stewart C., and Nicholas S Majluf,1984, “Corporate Financing and Investment Decisions when Firms Have Information that Investors Do Not Have,” Journal of Financial Economics, 13 (2), 187-221. Pindyck, R.S and D L Rubinfield, 1998, “Econometric Models and Economic Forecasts,” 4th ed., (Irwin/McGraw-Hill, Boston) Shefrin, Hersh M. and Meir Statman,1984, “Explaining Investor Preference for Cash Dividends,” Journal of Financial Economics, 13 (2) 253-282. Shevlin, T., 1982, “Australian Corporate Dividend Policy: Empirical Analysis,”

Accounting and Finance, 1-22. Shleifer Andrei and Robert Vishny,1986, “Large Shareholders and Corporate control,” Journal of Political Economy 94 (3), 461-488. 24 Source: http://www.doksinet 25