Dividend Policy (Ch. 16)

 

HOW WE VIEW DIVIDEND POLICY

·        The key question: “What dividend policy should the firm follow?”  Answer depends on how dividend policy affects s/h wealth (just dividend policy by itself).

·        Want to change just the dividend, holding everything else constant, and see what effect it has on s/h wealth

o       Investment decision does not change

o       Debt financing does not change

·        Think of the sources and uses of funds:

o       Sources: operating cashflows and new equity issues (since we don’t change debt)

o       Uses: investment and dividends

=> OCF + New Equity = Investment + Dividends

o       Holding the investment decision constant means both OCF and Investment are fixed (don’t change as you change the dividend)

o       If you increase dividends by $1, you will have to issue $1 worth of new shares to finance the dividend

·        So, to identify the impact of just dividend policy by itself on s/h wealth, we have to ask what happens to s/h wealth when you increase dividends and finance the increase by issuing new shares.

·        One thing to bear in mind:

o       if OCF > investment, the firm needs to pay out the difference to s/h

o       but dividends are not the only way to disburse cash to s/h

o       the firm can also repurchase shares

 

WHAT FIRMS ACTUALLY DO IN PRACTICE

·        They are reluctant to reduce dividends

o       Tend to increase dividends only when they feel the higher dividend can be maintained

o       This means dividends are increased when managers are confident future earnings will be higher

o       Typically, dividend increases follow earnings increases; and after dividend increases earnings stay constant or increase further

o       Earnings are volatile; dividends follow a step function (“dividend smoothing”)

·        Firms use target payout ratios, but only as a long range target

o       If you used a fixed payout ratio, dividends would increase every time earnings went up and decrease every time earnings went down.

o       With dividend smoothing, actual payout ratio fluctuates around target payout ratio

o       As earnings increase, actual ratio becomes consistently lower than target ratio for a while

o       If firm feels the earnings increase is permanent rather than temporary, they will step up dividends, and bring actual ratio close to target ratio again

 

INFORMATION CONTENT OF DIVIDENDS

·        Suppose two firms have the same increase in earnings today.  But firm P’s managers know the increase is permanent, while firm T’s manager’s know the increase is temporary.

·        Firm P will increase its dividend, firm T will not.

o       P and T announce the same increase in earnings 

o       T announces an unchanged dividend; P announces a dividend increase

·        The market understands what these dividend announcements imply about future earnings.  Both firms will have a positive reaction to their announcements, but P’s will be more positive.  The dividend increase announced by P conveys good news about future earnings.

·        Bottom line: looking just at earnings increases the market cannot tell which increases are permanent and which are temporary.  Dividend announcements allow the market to distinguish them apart.  So the market reacts to dividend announcements based on what information the dividends convey about future earnings.

·        Dividend increases are treated as good news; stock price goes up.  Dividend decreases are bad news; stock price falls.

 

THE DIRECT EFFECT OF DIVIDENDS

·        The information content of dividends is clearly relevant, and represents part of the effect that dividend policy has on s/h wealth. 

·        But it’s an indirect effect.  And ultimately all it achieves is that information that would have been reflected in the price later (as it became known through other channels) gets reflected in the price today.  Even if there’s no dividend increase today, s/h will get the same increase in their wealth later, as the market learns about the permanent increase in earnings.

·        We are more concerned about the direct effect of dividends.  Leaving information content to one side, when a firm pays out an extra dollar of dividends (financing it by issuing a dollar’s worth of new equity) is that good for s/h or bad?

·        Unfortunately, there is absolutely no consensus what the correct answer to that question is.  All we can do is understand what the issues are, and why the question remains unresolved.

 

PERFECT CAPITAL MARKETS

·        The best starting point is to consider what effect dividend policy would have in “Perfect Capital Markets (PCM)”

·        PCM is an artificial world, defined as follows:

o       No taxes

o       No transaction costs (in primary markets as well as secondary markets). So no issue costs, and no brokerage commissions or bid-ask spreads either.

o       Equal information; managers don’t know anything the market doesn’t

·        It’s a far cry from the real world but it sheds very useful light on how things should work in the real world

·        It’s the starting point for understanding both dividend policy and capital structure.

·        Turns out that dividend policy will be irrelevant in PCM: has no impact on s/h wealth

·        Why is that useful to know?  Tells us that there are only three reasons why dividend policy might be relevant in the real world: taxes, transaction costs or inside information

·        We’ve already discussed information effects.  That leaves us with two factors to consider: taxes and transaction costs.  Anything else is irrelevant.

·        Effect of dividend policy in PCM:

o       An all-equity firm has the following assets, and an investment opportunity today which requires an investment of $1,000 and has an NPV of $600:

§        Cash               1,000

§        Real assets     9,000

o       Suppose the firm pays no dividends.  Wealth of s/h = value of their equity = value of the firm = value of existing assets + PVGO = 10,000 + 600 = 10,600

o       Suppose the firm pays a $500 dividend.  It will now have to issue $500 worth of new equity to finance the investment.

§        Value of the firm once the new equity is issued and dividend is paid = value of existing assets + PVGO = 10,000 + 600 = 10,600

§        New s/h’s share of this value = $500  (they’ve been issued shares worth $500)

§        Old s/h’s share of this value = 10,600 – 500 = 10,100

§        Wealth of old s/h = dividend received + value of their shares = 500 + 10,100 = 10,600

=> no change in s/h wealth

o       Suppose initially there were 1,000 shares. 

§        Each share would originally be worth $10.60.  The dividend per share would be 50 cents.

§        From above, the ex-dividend value of these original 1,000 shares = 10,100

§        Ex-dividend value per share = $10.10

§        The new shares will be issued at this price.  To raise $500, the firm will have to issue 500/10.10 = 49.50 shares

§        Without the dividend, each old s/h has a share worth $10.60.

§        With the dividend, each old s/h gets 50c in cash and has a share worth $10.10.

§        Assume a s/h has bought his shares last year for $8 each, and sells today.  Without the dividend, his return consists of a capital gain of $2.60.  With the dividend, he gets a dividend of 50c and a capital gain of 2.10.  In PCM, s/h have no reason to care how their total return is divided up between capital gains and dividends.  So they are indifferent to whether there is a dividend or not.

o       The equal access argument:

§        Some stockholders may want cash from their portfolio today; others may not.

§        Neither type of s/h cares whether the firm pays a dividend or not.

§        If the firm doesn’t pay a dividend, a s/h who wants cash from her portfolio can generate cash by selling some of her shares. 

§        If the firm pays a dividend, a s/h who don’t want cash can simply reinvest the dividend by buying some of the shares the firm is issuing.

§        What the firm is doing by making a dividend payment, s/h can do or undo on their own.  So it doesn’t matter to them whether the firm pays a dividend or not.

o       Stock repurchase has the same effect as a dividend: no change in s/h wealth

§        Initially there are 1,000 shares outstanding, for a total value of $10,600

§        The company will use $500 to buy back 505/10.60 = 47.17 shares

§        There will now be 952.83 shares outstanding at $10.60 each, for the same total value as before, $10,100

§        S/h who sell their shares will now have $10.60 in cash; s/h who keep their shares will have $10.60 worth of shares

                §    The wealth of both types of s/h doesn’t change

EFFECT OF TAXES AND TRANSACTION COSTS

·        The effect of transaction costs is straightforward.  If you pay extra dividends financed by issuing new shares, the firm incurs issue costs.  (Issue costs can be substantial.  See figure 15.3 on p. 417.)  So dividends are costly, and higher dividends are more costly.

·        The effect of taxes is the source of all the controversy surrounding dividend policy.

·        A lot of empirical research has been done to try and figure out what the effect of taxes actually is.  However, there are serious problems with research design, and the studies are totally inconclusive.

·        So it becomes a theoretical issue.  And the reason for all the controversy is that there is no clear theoretical answer for what the effect of taxes should be.

·        The question we have to answer is very simple.  In PCM, investors have no preference between dividend income and capital gains income.  Do taxes make investors prefer either one over the other?

·        The problem is that there are three different groups of investors.  Individuals prefer capital gains income; tax-exempt institutions (like pension funds and mutual funds) are indifferent; corporations prefer dividend income.

·        Individuals:

o       Capital gains are taxed at 20%; dividends are taxed at an individual’s marginal tax rate.  Most people have marginal tax rates > 20%

o       Also taxes on dividends have to paid when the dividend is received. Taxes on capital gains are postponed till you actually sell the shares and realize the capital gain.  So the PV of taxes on capital gains is even lower.

o       Taxes on capital gains can be avoided completely if you bequeath shares to your heirs.

o       Bottom line: individuals pay higher taxes on dividend income than on capital gain income.  If a firm pays higher dividends, these investors have to pay higher taxes, so they are worse off.  Higher dividends reduce their wealth.

·        Tax-exempt institutions:

o       They pay no taxes on either capital gains income or dividend income.

o       So they are indifferent between dividend income and capital gains income.  Higher dividends have no effect on their wealth.

·        Corporations:

o       70% of their dividend income is exempted from tax, so they pay tax only on 30% of the dividends they receive.

o       The marginal tax rate is 35%.  On capital gains income they pay the full 35%.  On dividend income they pay 30% of 35% = 10.5%

o       Corporations prefer dividend income to capital gains income.

·        So we have three different groups with three different preferences.  The problem then becomes: which set of preferences is actually reflected in stock prices?  Do higher dividends increase s/h wealth or reduce it or leave it unchanged?

·        There is no consensus at all.  We end up with three schools of thought:

o       The rightists argue that higher dividends increase s/h wealth. 

o       Middle-of-the-roaders argue that dividends don’t affect s/h wealth.

o       The radical left argues that higher dividends reduce s/h wealth.

·        So it’s not clear what the best dividend policy is.  All we know for sure is:

o       The market treats today's dividend as conveying information about future earnings, and reacts to dividend announcements accordingly

o       Increasing dividends is costly because it will lead to higher issue costs.

o       Higher dividends will affect different stockholders differently.  They will increase the tax burden of individuals and decrease the tax burden of corporations.

 Return to Contents Page