Dividends, share repurchases and the substitution hypothesis


For decades, U.S. corporations have overwhelmingly preferred paying
out cash in the form of dividends over share repurchases, despite
the relative tax advantages of capital gains over ordinary income.
However, over the past twenty years share repurchase activity has
experienced an extraordinary growth.

• According to COMPUSTAT, expenditures on share repurchases
programs increased from 4.8 percent in 1980 to 48.1 percent in

• Share repurchases grew at an average annual rate of 26.1 percent
while dividends grew at an average annual rate of 6.8 percent.
• Share repurchases as a percentage of total dividends grew from
13.1 percent in 1980 to 113.1 percent in 1998.

• In 1999 and 2000, industrial firms spent more money on share
repurchases than on dividend payments.

The paper’s objectives are as follows:

• Analysis of the recent trend in share repurchases. Authors
show that despite the decrease in the dividend payout ratio in
the 80’s and 90’s, the total payout ratio did not decrease. As
well, the majority of firms initiate cash payout to shareholders
through repurchases rather than cash dividends.

• Substitution Hypothesis: Empirical evidence in this paper suggests
that the marked increase in share repurchase activity in
the U.S. has been financed with potential increases in the dividend.

– Number of firms repurchasing shares as a percentage of total
number of firms distributing cash to their equityholders
increased from 31.1 percent in 1972 to 80 percent in 2000.

– Number of firms initiating a buyback program as a percentage
of total number of firms initiating a cash distribution
to their shareholders increased from 26.6 percent to 81.3

As well, the authors find evidence that large-established firms
partially finance their repurchase programs with potential dividend
increases. Specifically, they find that dividend forecast
errors are negatively correlated with share repurchase activity.
Finally, evidence indicates that investor perceive that corporations
substitute share repurchases for dividends.

– Market reaction surrounding the announcement of dividend
decreases is significantly less negative for repurchasing
firms than for non-repurchasing firms.

– Market reaction to dividend decreases is not significantly
different from zero for repurchasing firms.

– Market reaction surrounding open market repurchases announcements
was significantly more positive before enactment
of the 1986 TRA when the benefit of substituting
was larger.

• The paper will show that after the SEC adopted rule 10b-18
(safe-harbor rule), the aggregate amount of cash spent on share
repurchases tripled.

Previous Research
Some of the previous research suggests that dividends and share
repurchases are perfect substitutes:

• Miller and Modigliani (1961) stipulate the dividend irrelevancy
theory that implies that share repurchases and dividends are
perfect substitutes.

• Agency theories of Easterbrook (1984) and Jensen (1986) imply
that one can control managers’ actions by taking excess cash
out of the firm.

• Most of the signaling models imply that dividends and repurchases
are perfect substitutes. Bhattacharya (1979) stipulates
that the signaling cost is the transaction cost associated with
raising new capital and in Miller and Rock (1985) its the cost
of shaving investment. However, John and Williams (1985)
model suggests that higher taxes on dividends are the costs of
the signal.

On the other hand, others believe that the relationship is not so
• Allen, Bernardo and Welch (2000) put forth a model in which
share repurchases and dividends are not substitutes because
the latter payout method attracts institutions.

• DeAngelo and Skinner examine the relationship between the
disappearance of special dividends and the appearance of repurchase
programs to investigate this issue. They do not find
that the latter has replaced the former.

• Jagannathan, Stephens, and Weisbach (2000) find that firms
that pay dividends have more stable earnings than firms that
use share repurchases. They also report that firms that only
pay dividends have higher earnings volatility than do firms that
payout cash using both dividends and repurchases.

From the Industrial COMPUSTAT files, authors create an initial
sample of all companies that appear on the files for at least one year
over the period 1972-1998. In order to remain in the sample, each
firm must have the following available information:

• Total earnings before extraordinary items (EARN)
• Market value of common stock at end of year (MV)
• Total dollar amount of dividends declared on the common stock
of the firm during the year (DIV)
• Total expenditure on the purchase of common and preferred
• Value of net number of preferred shares outstanding
For each observation in the final sample, the following variables were
• REPO: Equal to the expenditures on the purchase of common
and preferred stocks minus any reduction in the value of the
net number of preferred shares outstanding.
• MB: equal to the book value of the total assets plus any market
value of equity minus the book value of equity, scaled by the
book value of the total assets.
• CASH: equal to the book value of cash and short-term investments
scaled by the book value of the total assets.
• ROA: equal to the operating income before depreciation scaled
by the book value of the total assets.
• (ROA) : equal to the standard deviation of ROA over the
three years surrounding the year of the firm-year observation.
• NOPER: equal to the non-operating income before depreciation
scaled by the book value of the total assets.

The final sample contains 14.702 firms and overall 121,973 firm-year
observations over the period 1972-1998.

Trends in Corporate Payout Policy

Recent trends in corporate payout policy are examined. These are
tabulated in table 1 and depicted graphically in figure 1. Notable
items include:

• In the 70’s and early 80’s, share repurchases were a small fraction
of total earnings and total dividends. Between 1972 and
1983, repurchases amounted to an average of 10.9 percent of
dividend payments. Since the mid 80’s, share repurchases have
become a significant payout method. On average, between 1984
and 1998, the dollar amount distributed through repurchases
relative to dividends was 51.5 percent, and it reached a high of
104 percent in 1998.

• The mean dividend payout ratio has been declining since the
beginning of the 80’s from approximately 23.1 percent in 1980
to approximately 13.7 percent in 1998.

• The mean repurchase payout ratio has increased from 3.8 percent
in 1980 to 13.6 percent in 1998.

Table 2 reports characteristics of the firms in the sample by payout
policy. Notable items from here include:

• Dividend-paying firms (DIV > 0) are much larger and more
profitable than firms that do not pay dividends (DIV = 0).

• Firms that pay dividends have a lower variability of return on
assets than firms that do not pay dividends, irrespective of their
repurchase policy.

• It seems that firms that pay dividends but do not repurchase
shares (DIV > 0, REPO = 0) are very similar to those that
pay dividends and repurchase shares (DIV > 0, REPO > 0).

• Firms that do not pay dividends (DIV = 0, REPO > 0) appear
to have similar characteristics to firms that do not pay out any
cash (DIV = 0, REPO = 0).

• Repurchasing firms are younger than dividend paying firms.

• Conditioning on a firm paying dividends, there is no difference
between firms that repurchase or not repurchase shares.

Table 3 further analyzes the relationship between changes in variability
of earnings and propensity to repurchase. The table points

• Although the volatility of return on assets slightly declined from
3.35 percent in the period from 1972 to 1979 to 3.15 percent in
the period from 1992 to 2000, the average repurchase payout
ratio increased from 5.98 percent to 22.8 percent. There does
not seem to be a positive relation between share repurchase
activity and earnings volatility. As well the authors estimate a
cross-sectional regression of the change in the share repurchase
payout rate on the change in the standard deviation of ROA,
and find that the coefficient of standard deviation of ROA is

Figure 3 shows the cash distribution initiations. It shows that
the share repurchase programs have become the preferred method
of payout among firms initiating cash distributions to their equity
holders. Table 4 examines the dynamics of firms’ payout methods
during this time period and over three sub-periods. It reports the
transition probabilities of changing from payout policy i at time
T − 1 to payout policy j at time T. The four categories for firm
payout policy can be categorized as: 1) No cash distribution, 2)Only dividends, 3) Only repurchases, and 4) Both dividends and
repurchases. Results are as follows:

• Over entire time period, firms tended not to change their payout
policy. Comparing sub-samples, we see that:

– A higher portion of firms initiate a cash payment as dividends
in earlier periods.

– More firms initiate their cash payment in the form of repurchases
in the 1990’s then in the 1970’s.

– In recent periods, more firms that have been repurchasing
shares continue to do so.

– In the period from 1972-79, of the firms that only repurchased
in a given period, 15.1 percent switched to only
paying dividends in the following period and 29.06 percent
switched to using both methods of payments. Only 1.24
percent and 6.28 percent of firms, respectively, followed
this strategy in the later period.

– Firms that use both methods of payment are less likely to
switch to only dividends in later periods.

Share Repurchases, Dividends and the Substitution


Based on Lintner’s model:

ERRORt,i = [DIVt,i − (1,i + 2,iEARNt,i + 3,iDIVt−1,i)]/MVt−1,i (1)
Consistent with the substitution hypothesis, evidence indicates that
the dividend forecast becomes negatively correlated with the share
repurchase yield. Specifically:

• When there is no repurchase activity, dividend forecast error
is .044 percent when the preforecast period is 1973-1983 and
-.060 percent when the preforecast period is 1973-1990.

• When repurchase activity is high, the error is -.0144 percent
Cross sectional regression of dividend forecast on several factors
(Table Six): Here the authors regress the ERROR model against
firm characteristics like size, ROA, volatility of ROA, non operating
income scaled by total assets, and debt to total assets ratio. They
also use a Fama and Macbeth regression.

• Repurchase yield has a negative effect on dividend forecast error
when controlling for firm’s characteristics. RYIELD is equal to
-.0312 when preforecast period is 1973-1983 and -.01766 when
preforecast period is 1973-1990.

These results seems to suggest that dividend paying firms have been
substituting dividends with share repurchases.
Does the Market perceive dividends and repurchases
as substitutes?

The authors built a sample of 1,255 announcements of dividend
decreases between 1974 to 1996. Firms are classified as repurchasing
or non repurchasing firms. Results are in table 7.

• On average, the market reaction to dividend decreases is significantly
different from zero for non repurchasing firms. The
mean market reaction is -1.93 percent for nonpurchasing firms
and -.45 percent for repurchasing firms.

• The difference between the two means and medians are significantly
different from zero at the 1 percent level.

To control for other factors that may affect market reaction around
the announcement of the dividend decreases, the following crosssectional
regression is used:
CARi,t = 0+1DUMREPOi, t+2CHGDIVi,t+3SIZEi,t+4DY IELDi,t+5DROA0i,t+i,t
where CAR is the three-day cumulative abnormal return around the
announcement of the dividend. Results for this model are in table

• Coefficient of DUMREPO is positive and significantly different
from zero at the one percent level.

Thus, this suggests that investors perceive corporations substitute
share repurchases for dividends and penalize repurchasing firms less
then non repurchasing firms.

In order to check to see what happens if investors’ dividend tax
rates are higher than their capital gains tax, the authors test this by
examining the effect of the Tax Reform Act of 1986 on the market
reaction surrounding share repurchase announcements.

• The Tax Reform Act of 1986 drastically reduced the difference
between dividends and capital gains. Thus, the substitution
hypothesis predicts that a reduction in the market reaction
around share repurchases after this tax reform. The authors
form a sample of firms that announce open market share repurchase
program over the period 1980 to 1987.

• They divide this sample into two subsamples: Firms that announced
open market share repurchases prior to approval of the
tax reform and firms that announced open-market share repurchase
program after the approval of the tax reform. Results are
shown in table nine.

– Mean Market reaction around the announcement of share
repurchase programs declined after the Tax Reform Act of
1986 from 3.49 percent to 2.42 percent. The difference is
significant at the one percent level.

The decline in the average market reaction after 1986 could be because
the share repurchase programs became more predictable, so
the authors also test the market reaction to first time announcements
of share repurchases. They find that the average market
reaction declined after the TRA from 4.03 percent to 2.94 percent.
Controlling for other factors that might affect the market reaction
around share repurchase programs, the following cross sectional regression
was estimated:

CARi,t = 0+1TAXt+2log(PSOUGHTi,t)+3SIZEi,t+4DY IELDi,t+i,t (3)
Results are posted in table ten. Note that the differential tax variable
is positively related to the market reaction surrounding open
market share repurchase programs and similar results are noted for
first-time announcements.

The Effect of rule 10b-18 on share repurchase activity
If the evidence suggests that corporations substitute share repurchases
for dividends, why did it take so long?

• Prior to rule 10b-18, it seemed that firms were deterred from
repurchasing shares.

• Rule 10b-18 requires that firms repurchasing shares on the open
market only use one broker or dealer on any single day, avoid
trading on an up tick or during opening or the last half-hour
before the closing of the market, and limit the daily volume of
purchases to a specified amount.

• Does share repurchase activity increase significantly after adoption
of rule 10b-18 in 1982? Using the following model to test:

W = (μI0μII )2/(2
I + 2
II ) (4)
where μI(μII) is the mean value of the variable of interest over
the period from 1972 to 1983 (1984-2000) and 2
I (2
I ) is the
sample variance of μI(μII ). Results of this model are shown in
table eleven. In particular

– Before rule 10b-18, average annual expenditure on share
repurchase programs was 5.5 billions dollars. After, the
expenditure was 62 billion dollars.

– Rest of the table suggest that that share repurchase activity
drastically increased after adoption of rule 10b-18.

However, these differences can be due to some other factors such as
stock market activity, tax changes, or learning by the market. Using
the following model:

MV )t = 0+1REGt+2TAXt+3TIMEt+4MRETt+t (5)
with the assumption that the model seems to follow a MA(1) process,
t = μt + μt−1 (6)
Results are posted in table 12. Specifically:

• The coefficient, REG which captures the effect of rule 10b-18
is positive and significant.

• The coefficient for the tax differential, TAX is positive and


• Corporations have been substituting share repurchases for dividends

• The majority of firms that initiate cash payments do so through
share repurchases.

• Many firms that have been paying dividends have also started
to repurchase shares as well.

• Propensity of firms to initiate a dividend payment in the 90’s
is by order of magnitude smaller than it was in the 70[’s.
• Established corporations distribute more of their cash flows
through repurchases and less through dividends.

• Dividend forecast errors are negatively correlated with share
repurchase activity.

• Market reaction of dividend decreases is not significantly different
from zero for firms that are engaged in share repurchase

• Firms that do cut their dividend and are not repurchasing
shares do experience a significant negative price drop to the
announced dividend cut.

• Differential taxes between dividends and capital gains seem to
matter in that market reaction to repurchases is more positive
when the tax gains from repurchases relative to dividends are

• Share repurchase activity experienced an upward structural
shift after the adoption of rule 10b-18.

• A more appropriate tool of valuation is total payout rather than
dividend payout (due to combined trend of smaller dividend
payments and larger share repurchases).

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License