Are dividends and share repurchases substitutes?
Traditionally, most companies used to make payments to their shareholders in the most direct form – dividends. In the last decades however, an alternative option to transfer wealth to shareholders has gained increasing importance. Firms repurchase their own shares, thereby diminishing the number of shares outstanding which in turn increases the stock price. However, there are widely differing views as to whether both payout methods can be as substitutes or whether there are significant differences between both.
Even though dividend policy is the second big topic of research in the field of finance besides capital budgeting, there is still no commonly accepted framework and empirical evidence seems to leave open more questions than it answers. This paper aims at reviewing some of the more important contributions to the question: Are dividends and share repurchases substitutes? Furthermore, it will try to reconcile widely differing views and come to a conclusion in how far both methods are indeed substitutes or not. In order to examine this main question, it is broken down into some aspects.
A first field of interest is the sources of the funds out of which either dividends or share repurchases are paid. If these sources were equal, that would mean
Need essay sample on "Are dividends and share repurchases substitutes?"? We will write a custom essay sample specifically for you for only $ 13.90/page
Finally, the paper will study the perspective of investors and managers respectively. It will examine their perspectives of dividends or share repurchases. The paper will thus break down the main question into the following four sub questions:
1. Are repurchases financed by funds that would otherwise be used for dividends?
2. Are the motives for paying dividends and repurchasing stock the same?
3. How do stock markets react to both ways of paying shareholders?
4. What is the relevance of the choice of payout method for managers? These sub-aspects will then allow a more structured approach to the main problem statement.
The article will be organized as follows. It will first clarify and contrast the theoretical base of both payment methods by analyzing in the simplified Miller-Modigliani (MM) world both before introducing real-world elements like taxation or transaction costs and after. This serves as a basis for understanding the following discussions of the differing views and theories concerning the four sub questions in the part three. Part four compares and evaluates the opposing views in an effort to reach syntheses for each of the sub questions. Subsequently, this will form a basis for the answering of the main problem in part five.
Finally, the paper ends with a conclusion that summarizes the findings and gives recommendations for further research.
2. THEORETICAL BASE (MM) In order to ease the understanding of the following discussion of the different viewpoints, this paragraph will introduce the problem of dividend policy by means of the Miller-Modigliani (MM) (Ross, Westerfield & Jaffe, 2002, p. 407, 416) model. This model is generally accepted among financial researchers and practitioners. It starts by stating a number of assumptions that make the MM world a simplified, easy-to-analyze one.
These assumptions are total market efficiency, rational, profit-maximizing behaviour and perfect certainty. (Megginson, ???? ). With the help of this model, it can be shown that in the absence of taxes, any investor is indifferent as to whether a corporation pays out profits or retains them and invests them at the current interest rate. Furthermore, investors are also indifferent of the form of payment if one is made. While the MM model has often been criticized for its lack of realism due to its extreme assumptions it can be seen as a starting point for introducing and examining one by one elements of realism.
Probably the most important market imperfection is taxation. Most current taxation regimes tax dividends higher than capital gains which are reached through share repurchases, and as an effect, rational investors should prefer share repurchases as payout method. (For an example and illustration of the effects of taxation, see Graham & King, 2000). The fact that despite these disadvantages companies still dominantly use dividends is generally called the dividend policy puzzle. Numerous theories exist to explain it, some of which are explained in the remainder of the paper.
The situation is not as clear when it comes to transaction costs. While (reference) see dividends as a way to pay out shareholders in relatively underdeveloped markets where investors would bear high costs when selling shares to make their own payout, others (e. g. University of Strathclyde, 2003) stress the cost of share repurchases.
3. ANALYSIS OF SUB QUESTIONS 3. 1 Source of funds Maybe the most important aspect when it comes to evaluating different policies in general is the source of funds. This part will review different thoughts as to how dividends and share repurchases are financed.
Examination of existing literature reveals strong evidence for both instruments being funded from residual cash flows. Ross et al. (2002) introduces both share repurchases and dividends as ways to transfer residual cash flows to investors after all investments with a positive NPV have been taken. This theoretical reasoning that stems from an MM type analysis is confirmed by a study of Brav, Graham, Harvey and Michaely (2003) who highlight that this is exactly the perception managers have of the two tools. De Jong, van Dijk and Veld (2001) come to a similar conclusion by studying the decision making process concerning payouts.
He opposes three different models which he calls unrelated, related and sequential decision-making (see Appendix A). The study confirms the sequential decision making process which means that firms decide on whether to payout first, and only then on the method of payout. This indicates that the decision to pay dividends and the decision to repurchase shares can only be made when funds have been defined and the decision on payout level made. The above-mentioned results are largely confirmed by many writers. (ganz viele Referenzen) However, there are also writers that point to different results.
Lilljeblom and Pasternack (2002) for example suggest based on Jagannathan, Stephens and Weissbach (2000) that dividends are used to “distribute relatively permanent cash-flow shocks and repurchases to distribute more transient shocks”. This hypothesis is supported by Brav et al. (2003) who found through interviews that “managers are very reluctant to cut dividends, [and] that dividends are smoothened through time”. A theoretical explanation for this the Lintner model which states that companies generally have a long-run target payout ratio and adjust dividends only slowly (Lintner, 1956).
Grullon and Michaely (2002) use differences between predicted and actual dividend payments according to the Lintner model as support for their hypothesis that repurchases and dividends are substitutes by reasoning that these differences are filled by share repurchase programs. 3. 2 Motives for payment Another important dimension for comparing different policy measures is the motive behind them. This part will consider a number of theories tat explain why payouts are made in general, but also how to choose which payout method and why dividends are still used at all despite the so-called dividend policy puzzle.
It starts with the two leading hypotheses, namely the signaling hypothesis and the free cash flow hypothesis. Then, a few special purposes for share repurchases will be stressed. 3. 2. 1 Free cash flow theory The free cash flow theory was mainly developed by Jenson (1986). Its main idea is that dividends and share repurchases are paid to get free cash flows out of the company. This becomes essential because of the principal-agent conflict. As Howe, He and Kao(1992) explain, managers “have pecuniary and nonpecuniary incentives to overinvest” as their status often depends on firm size.
If any excess cash after investing into positive NPV projects is given back to investors, the amount of this agency conflict is reduced. Notably, this approach makes no difference between the methods of payout. However, while seeming appealing at first sight, empirical evidence for this theory is only mixed. While Lang and Litzenberger (1989) as well as Ikenberry, Lakonishow and Vermaelen (2000) find no support for it, Fenn and Liang (1997) confirm it in their large sample empirical research. The overall picture that emerges is that of weak but nevertheless present prediction power. 3. 2.
2 Signaling theory The signaling theory starts with the idea of asymmetric information and is therefore also called asymmetric information model (Persons, 1995). Managers are believed to have more information about the state of the firm than investors. In order to signal the state of the firm, managers can use among others dividends and share repurchases. According to Grullon et al. (2002), the signaling theory shows that both are not interchangeable. The reasons for this distinction are manifold. They start from the above-mentioned idea that managers are very unwilling to cut dividends (Brav et al., 2003).
Raaballe and Bechmann (2002) reason that, on the margin, dividends are more costly for low quality firms than for high quality firms. Therefore, they represent an excellent, though not cost-free way to signal their higher quality. As these findings are not transferable to share repurchases, they are consistent with empirical evidence and thus present an explanation for the dividend policy puzzle. This is also in line with the study of Sarig (1999), who also finds the information content of dividends to be higher than the one of share repurchases.
While most researchers agree on these findings (e. g. John, Kose and Williams, 1985 or Bernheim, 1991 in Grullon et al. , 2002) and empirical evidence seems strong, it must also be noted that there are some critical voices present. John Persons (1995) shows the exact opposite, namely that share repurchases are the more efficient signaling device. 3. 2. 3 Other theories In addition to these two leading theories, there are a number of other explanations to explain the complex issue of dividend policy.
This part will mention the most important ones and show how they point to differences or similarities between share repurchases and dividends. One theory suggests that managers tend to use share repurchases when they believe that their stocks are undervalued. This allows them to buy back stock at a lower price, and if necessary reissue them again when market prices reflect the full value. As this theory also assumes asymmetric information, it is a close relative of the signaling theory. The difference is that it con only be used for share repurchases.
In addition, this hypothesis, which is also called the “good investment hypothesis” (Rau and Vermaelen, 2002), assumes non-efficient markets because otherwise stock prices would rise to their true value immediately after the announcement of the intended repurchase, thereby making any advantages for the repurchasing company obsolete (Rau et al. ). Even though this theory does not offer the elegance of the more theoretical theories and can only explain a limited range of transactions, its empirical support makes it seem rather convincing. (Ikenberry, Lakonishok & Vermaelen, 1995 and 2000; Stephen & Weii??bach, 1998 )
Even though it often is a starting point of discussions concerning payout policy, the tax advantage of share buybacks is often neglected when reasons for choosing payout methods are mentioned. As this part of the paper aims at giving a complete overview over reasons for payout in general as well as for choosing one method, it must stress the tax advantage of share repurchases once again at this point. While many writers see this issue as the most fundamental one, its strength differs over different countries (University of Strathcycle, 2003, Rau et al. , 2000)
Finally, it must be mentioned that a number of special purposes for share repurchases have been proposed. Among these are strategic considerations as the fighting of takeovers or issues of efficiency such as paying out very small shareholders that are too expensive to serve (Rau et al. 2000). One issue that is not mentioned in this part is the incentive of managers to prefer share repurchases over dividends. This is explained in detail in part 3. 4. 3. 3 Investors’ reactions Given the large amount of hypotheses concerning the justification and source of funds of payout programs, a crucial issue is the actual outcome of decisions.
Therefore, it is essential to monitor the reaction of the investors since they are the ones that define the success of any measure in business and even the success of companies as a whole. Rau et al. (2000) found positive abnormal returns in the long run for companies that announce share repurchases. While they interpret this as raising questions about efficiency, it could also be seen as supporting the good investment hypothesis which believes shares to be undervalued at the point of the repurchase announcement.
Another interesting investor reaction is the one discovered by Howe et al. (1992), who finds that increases in share prices following a repurchase announcement are not significantly different for low-quality and high-quality firms. Howe et al. Interpret this as giving evidence for the free cash flow hypothesis but contradicting the signaling hypothesis. While many studies use a measure of stock market reactions to support or defend their theses, there emerges no clear picture as to which payout method investors prefer. 3.
4 Managers’ perspective One point that is often forgotten in the analysis of dividend policy is the perspective of the managers. Most study aim at finding elegant theoretical explanations for empirical patterns that are consistent with generally agreed upon principles such as market efficiency. However, the firms’ managers are the ones that make the final decision, and therefore, their own interests must be taken into account, too. Managers care about dividend policy because it has the power to change the value of the firm’s stock.
As managers are often compensated, at least partly, with stock options, they have a genuine interest in maximizing share value. When a company pays a regular cash dividend, that company’s stock price decreases on the ex-dividend date by up to as much as the dividend itself is. Since most managers do not hold stock but stock options, they bear the full loss of that price drop. Fenn et al. (1997) use this relationship to explain that “managers can increase the value of its stock options by substituting share repurchases for dividend growth”.
This is a clear example of a conflict of interest, where managers can increase their own wealth at the cost of company wealth. Both Fenn et al. and Liljeblom and Pasternack (2002) found evidence that this is actually happening. They call this theory the managerial-wealth hypothesis. As a response to that, some companies pay their managers with dividend protected options, which means options that contain a mechanism that takes this loss away from the option holders. Liljeblom et al.
have also found that firms with dividend protected option programs have different payout policies and do rather tend to pay dividends than those without dividend protected options.
4. COMPARISON OF LITERATURE / EVALUATION OF SUB QUESTIONS After the presentation of the most important existing theories in part three, it becomes evident that there are many factors and theories that advocate a substitutional relationship between cash dividends and share repurchases just as well as there are many who seem to contradict such a relationship.
Few state their point of view as clear as Grullon et al. (2002) (“firms have gradually substituted repurchases for dividends”) or the University of Strathclyde (2003) (“[… ] these two methods are not necessarily [… ] substitutes”), but most display a strong predisposition for one of the two positions in their reasoning. This part tries to summarize, compare and evaluate the evidence in an effort to achieve a preliminary answer to the sub questions asked in the introduction. 4. 1 Source of funds
The first point that was discussed is the one of the source of funds. Here, one must say that the most important finding is that both methods of payouts are financed out of residual cash flows after all possible positive NPV investments are taken.
Also, evidence is strong that the rising importance of share repurchases has not resulted in a higher total payout ratio but rather was countered by an equivalent decrease in dividends. Therefore, one could answer the question: Are repurchases financed by funds that would otherwise be used for dividends?with a clear yes if it was not for some writers who point at the fact that dividends are used to pay out rather stable excess cash flows while repurchases are used for more temporary ones.
Nevertheless, this seems to be originating rather in historical and traditional reluctance to cut dividends than in economically sound reasoning. Therefore, this distinction will probably diminish as time goes on and the importance of share repurchases will further increase. 4. 2 Motives for payment The second sub question that was examined was: Are the motives for paying dividends and repurchasing stock the same?
Here, evidence is not as clearly pointing in one direction as with the source of funds. While the free cash flow hypothesis strongly advocates the substitution hypothesis, the signaling theory sees differences between both methods. However, even among proponents of the signaling hypothesis there exists no agreement as to which payout tool signals more effectively. Further, there are some theories that are not applicable to both dividends and share repurchases, such as the good investment hypothesis or strategic consideration that try to explain share repurchase behavior.
These special purpose theories are of course a hint that there is more to share repurchases than just a more efficient way to pay dividends. In conclusion, one must say that there is evidence in both directions. The emerging picture is that mostly, the reasons for paying out are the equal, regardless of the payout method, but there are also a number of cases where one method could not serve as a substitute for the other in achieving its purpose. 4. 3 Investors’ reactions After the motives for payment, the paper analyzed the reaction of investors.
To measure investors’ reaction, movements in stock prices were scrutinized. While many studies use these market reactions to prove their theses, the question How do stock markets react to both ways of paying shareholders? cannot be answered since there emerges no clear picture that might give a hint as to which method of payment investors prefer. 4. 4 Managers’ perspective The last question that was investigated is What is the relevance of the choice of payout method for managers?. Research in this area pointed towards two points.
First of all, managers who hold stock options should prefer share repurchases since they result in a capital gain that managers profit from instead of a drop in share prices as is the case with dividends. Second, this bias can be nullified if a firm includes in their option payments a mechanism to protect the options from this price drop. Therefore, one can say that if companies pay dividend protected stock options, managers see no significant difference between both methods of payout and regard both as substitutes.
5. SYNTHESIS Up to this point, the paper has broken down the research question Are dividends and share repurchases substitutes? into four sub questions. Then paper analyzed arguments concerning all of these sub questions and finally compared and evaluated these arguments. This part will try to synthesize the partial answers in part four and thereby to approach the main problem statement. 5. 1 Definition of subsitutes First of all, one must notice that the whole discussion whether both methods are substitutes suffers from one weakness that does not lie within but rather even before all research.
The problem is that no writer states exactly how he/she defines the word “substitute”. Some that come to the conclusion that both are no substitutes do so they use a definition of substitute that is so tight that it allows virtually no differences. Others also find differences in details but still think that the similarities are enough to talk of substitutes. Even dictionaries seem not very helpfull for this problem. The “Advanced Learner’s Dictionary of Current English” (Hornby, Gatenby and Wakefield, 1963) defines the word as follows: “person or thing acting on behalf of another”.
Clearly this gives no hint concerning the degree of difference bein allowed among two things in order to still be substitutes. In the further discussion in this paper, the word substitutes will be used if two things show enough similarities to be used interchangeably in most but not all of the cases and if the people and institutions concerned see them as closely related alternatives. 5. 2 Answering of problem statement The analysis of the four sub questions revealed that share repurchases and dividends comply with the above mentioned definition of substitutes. The source of funds is basically equal.
Only some historical factors that are already loosing force force ake up some differences in this field. Furthermore, motives for payment are the same in most but not all cases. These two findings are enough to say that the first part of the definition is complied with. While the analysis of the view of investors remains rather inconclusive, research of quesiton four, concerning the perspective of managers shows that mangers see both as alternatives with very few differences. This can be considered enough to comply with the second part of the definition of substitutes.
Therefore, after close research of the involved factors and a thorough definition of hte word substitute, the paper has founf that the question Are dividends and share repurchases substitutes? must be answered with a clear yes.
6. CONCLUSION The paper tried to answer the question whether dividends and share repurchases are substitutes by analyzing four sub questions. These were concerned with the source of funds used, the motives why one method was used, and with the views of investors and managers, respectively.
The thorough analysis of these sub questions was then put together to enable an answer to the initial problem. However, in the course of the paper, it became clear that the problem lied at least partly in a missing definition of the word substitute. Therefore, a sufficiently precise definition was given and the paper could show that the two payout methods comply with this definitions. The paper thus comes to the conclusion that despite some differences dividends and share repurchases are indeed substitutes. With this main question answered, some concluding remarks are still necessary.
First of all, more research is still needed, especially in the field of motives of payment, where there are still a number of theories existing that sometimes contradict. These should be integrated into one framework. Given the power that the existing theories already have by themselves, it can be expected that such a framework could cope with many empirical puzzles that existing theories fall short to explain. Also, research could still be undertaken to answer the question whether investors really prefer share repurchases as tax considerations might suggest.
Finally, one must emphasize that dividend policy still is a field that changes at a breathtaking pace. Therefore, retesting in most fields related to it is necessary. Especially, the tendency to use dividends for permanent and share repurchases for more temporary increases in earnings is expected to diminish as well as the extreme reluctance to reduce dividends. Also, it remains to be seen what the impact of globalization of markets and market habits and better access to information will be.
Bernheim, B.D. (1991). Tax policy and the dividend puzzle, RAND Journal of Economics 22, 455-476. Retrieved June 22, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Brav, A., Graham, J.R., Harvey, C.R. & Michaely, R. (2003). Payout Policy in the 21st Century, Retrieved June 20, from the World Wide Web: http://papers.ssn.com
Fenn,G.W. & Liang, N. (1997). Good News and Bad News about Share Repurchases. Retrieved June 20, from the World Wide Web: http://papers.ssn.com
Graham, R.C. & King, R.D. (2000). Do Share Repurchases Harm Uninformed Shareholders?. Financial Practice and Education, v. 10, iss. 1, pp. 11-16. Retrieved June 20, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Grullon, G. & Michaely, R. (2002). Dividends, share repurchases and the substitution hypothesis. Journal of Finance 57(4) 2002 pp1649-1684. Retrieved June 22, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Hornby, A.S., Gatenby, E.V., Wakefield H. (1963). The Advanced Learners Dictionary of Current English (2nd ed.). Great Britain: Oxford University Press
Howe, K.M., He, J. & Kao, G.W. (1992). One-Time Cash Flow Announcements and Free Cash-Flow Theory: Share Repurchases and Special Dividends; Journal of Finance, v. 47, iss. 5, pp. 1963-75. Retrieved June 22, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Ikenberry, D., Lakonishok, J. & Vermaelen, T. (1995). Market underreaction to open market, Journal of financial economics 39, 181-208. Retrieved June 20, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Ikenberry, D., Lakonishok, J. & Vermaelen T. (2000). Stock repurchases in Canada: Performance and strategic trading. Journal of finance 55 p.2373-239. Retrieved June 22, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Jagannathan, M., Stephens, C., weisbach, M. (2000). Financial flexibility and the choice between dividends and stock repurchases. Journal of financial Economics 57, 355-384.Retrieved June 22, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Jensen, M. (1986). Agency cost of free cash flow, corporate finance and takeovers. American Economic Review 76, 323-329.
John, K. & Williams, J. (1985). Dividends, dilution and taxes: A signalling equilibrium. Journal of Finance 40, 1053-1070. Retrieved June 20, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Jong, A. de, Dijk, R. van & Veld, C. (2000). The Dividend and Share Repurchase Policies of Canadian Firms: Empirical Evidence Based on a New Research Design, Retrieved June 20, from the World Wide Web: http://papers.ssn.com
Lang, L. & Litzenberger, R. (1989). Dividend announcements: cash flow signalling vs. free cash flow hypothesis?, Journal of Financial Economics 24, 181-191. Retrieved June 20, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Liljeblom, E. & Pasternack, D. (2002). Share Repurchases, Dividends, and Executive Options: Empirical Evidence from Finland, Retrieved June 20, from the World Wide Web: http://papers.ssn.com
Lintner, John (1956). Distribution of incomes of corporations among dividends, retained earnings and taxes. American Economic Review 46, 97-113.
Megginson, Corporate Finance Theory, Addison-Wesley
Persons, J.C. (1995). Heterogenous Shareholders and signaling with Share Repurchases, Retrieved June 20, from the World Wide Web: http://papers.ssn.com
Raaballe, J. & Bechmann, K.L. (2002). Taxable Cash Dividends – A Useful Waste of Money, Retrieved June 20, from the World Wide Web: http://papers.ssn.com
Rau, P.R. & Vermaelen T. (2002). Regulation, Taxes, and Share Repurchases in the United Kingdom. Journal of Business, v. 75, iss. 2, pp. 245-82. Retrieved June 20, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
Ross, S.A., Westerfield,R.W. & Jaffe, J.F. (2002). Corporate Finance (6th edition). New York: McGraw Hill,
Sarig, O. (1999). A Longitudinal Analysis of Corporate Payout Policies. Retrieved June 20, from the World Wide Web: http://papers.ssn.com
Stephen, C.P. & Weiï¿½bach, M. (1998). Actual share reacquisitions in open market repurchase programs, Journal of Finance 53, 313-333. Retrieved June 22, from the World Wide Web: http://www.ub.unimaas.nl/fdewb
University of Strathclyde (2003). Topics in corporate finance: Share repurchases. Retrieved June 25, from the World Wide Web:http://accfinweb.account.strath.ac.uk/tom/sharerep.doc