Dividend Policy

1
Dividend Policy

• More Properly
• Payout Policy

2
Historical View

• Illustrated by the arguments of Gordon (1959) -
  more dividends more value.
• Follows from the discounted dividend approach to
  valuing a firm

3
Along Came MM

• Basic Point Firm value is determined by its
  investment policy, net dividends are simply the
  residual of earnings after investment.
• Dividend Irrelevance
• No Taxes or Transactions Costs
• Symmetric Information
• Complete Contracting Possibilities
• Complete Markets

4
Dividend Irrelevance

• Assume an all equity firm for simplicity.
• Firm value is the discounted value of the payouts
  to the equity holders.

5
Empirical Observations 1

• Corporations typically payout a significant
  percentage of their after-tax profits as
  dividends.
• Examination of dividend payouts over time shows
  that on average firms paid out between 40 and
  50 of their profits.
• Recently, a smaller percentage of all firms are
  paying dividends. Seems in part due to a lot of
  new firms and in part to the fact that fewer
  firms of all types are paying dividends. Some
  evidence suggests that firms are substituting
  repurchases for dividends.

6
Empirical Observations 2

• Historically, dividends have been the predominant
  form of payout. Share repurchases were
  relatively unimportant until the mid 1980s.
• Before 1984 repurchases amounted to between 2
  and 11 of corporate earnings. Since 1984 they
  have accounted for between 30 and 40 and have
  been on the rise.
• It is interesting to note that in the mid 80s
  the other major form of payout from the corporate
  sector, MA activity, also dramatically increased.

7
Empirical Observations 3

• Individuals in high tax brackets receive large
  amounts of dividends and pay large taxes on these
  dividends.
• That they choose to do so is referred to by Black
  as the Dividend Puzzle.
• Study by Peterson, Peterson, Ang (1985) showed
  that individuals received 33 B in dividends in
  1979 (2/3rds of total paid) and the marginal tax
  rate paid on the dividends was 40 (versus 20 on
  capital gains).

8
Empirical Observations 4

• Corporations smooth dividends.
• Lintner in a survey of companies noted that
• Firms are primarily concerned with the stability
  of their dividends.
• Changes in earnings are the most important
  determinant of changes in dividends.
• Dividend changes lag earnings changes.
• Dividend policy is set first then the investment
  and financing decisions made, taking dividends as
  given.
• Firms with many valuable investment projects are
  likely to set a low target payout ratio and those
  with few a higher target.

9
Lintner Model

• Two equations can be used to explain changes in
  dividends
• This is still one of the most useful ways to
  model dividend payouts by firms.

10
Empirical Observations 5

• There are positive stock price reactions to
  dividend increases and big negative reactions to
  dividend decreases.
• Pettit(1972), Charest(1978), Aharony
  Swary(1980).
• Consistent with asymmetric information models
  (dividends relay information) and with incomplete
  contracting models (dividends solve agency
  problems).
• Inconsistent with a large tax differential (or at
  least the tax effects are swamped by other
  effects).

11
Taxes

• A large part of the literature on dividends has
  focused on the impact of taxes on dividend policy
  and tried to reconcile the first three empirical
  observations.
• Basic aim of the tax literature is to determine
  if there is a tax effect do firms with high
  dividends have lower value than equivalent firms
  that pay low dividends?

12
Taxes

• Is there a tax effect?
• Fundamental question but not an easy one.
• Do tax clienteles exist?
• Simplest representation says pay no dividends.
• More clever ideas say firms target groups of
  investors.
• Is there dividend capture?
• Examine trading volume around dividend
  announcements.
• Managerial prescription?

13
Asymmetric Information

• Dividend signaling models.
• High (or higher) dividends signal good news.
• Good news about what?
• What is the signal cost?
• If the goal is to signal information to the
  market, why use dividends?
• Given the cost to the firm (transactions costs)
  and the cost to investors (taxes) there should be
  cheaper ways to communicate information about
  expectations credibly.

14
Asymmetric Information Predictions

• The information content of dividends.
• Signal of future cash flow.
• Signal of current earnings via sources and uses
  of funds identity.
• The predicted reactions.
• Dividend changes should be followed by changes in
  earnings in the same direction.
• Unexpected dividend changes should be followed by
  revised market expectations and/or stock prices.

15
Asymmetric Information Evidence

• The information content of dividends.
• In a statistical sense, dividends have relatively
  little information content beyond that contained
  in past and present earnings.
• Large changes seem to have some explanatory
  power for the firms next quarter earnings.
• Specially designated dividends and repurchases.
• Unexpected dividend changes positively related to
  changes in stock price.
• Not clear why the stock price reaction if there
  is no information conveyed. Unless it is not
  signaling that causes the reaction.

16
Agency Models

• Stockholder Bondholder conflicts.
• Bondholder wealth expropriation.
• Excessive dividends can expropriate wealth from
  bondholders and transfer it to stockholders.
• Bond prices do not drop when dividends are
  increased.
• Bond covenants restricting dividends.
• Covenants define a reserve out of which dividends
  may be paid.
• Firms tend to keep excess reserves.

17
Agency Models cont

• Stockholder Manager conflicts.
• Payouts as a disciplinary device.
• Control of free cash flow problems.
• More frequent scrutiny from the capital market.
• The empirical evidence.
• Evidence is inconsistent with the free cash flow
  story.
• Overall there is little convincing evidence that
  dividend payouts help control agency problems.

18
Other Stories

• Prudent Man rules.
• A stable dividend policy for firms that are
  invested in by a money manager may be a rule of
  thumb that helps constrain the agent.
• Transactions cost arguments.
• If investors are looking for current income
  dividends may be a low cost way of achieving that
  end.

19
Other cont

• Behavioral theories.
• People like to get dividends more than they like
  to participate in repurchase programs.
• Thaler and Shefrin (1981), Shefrin and Statman
  (1984)
• Irrational market stories.
• If managers have superior information so can time
  equity issues and the market doesnt fully adjust
  for this, dividends are a good policy.

20
The Prudent Prescription

• Firms should avoid having to cut back on positive
  NPV projects to pay a dividend.
• When personal taxes are a consideration, firms
  should avoid issuing stock to pay a dividend.
• Stock repurchases should be considered as an
  alternative way to get surplus cash out of the
  firm when there are few positive NPV projects and
  the firm has a surplus of unneeded cash.

21
Summary

• We have identified many of the things that should
  be important in establishing a firms payout
  policy.
• We have some evidence that some of them are
  important in shaping actual policies.
• We still dont know how they all fit together to
  establish an optimal policy nor do we know if
  there even is such a thing.