the cash- ow permanence and information content of...

qWe thank John Chalmers, John Core, Larry Dann, S.P. Kothari, Wayne Mikkelson, MichaelWeisbach, an anonymous referee, and seminar participants at the University of Alberta and theUniversity of Oregon for their comments.

*Corresponding author. Tel.: #1-541-3463242; fax: #1-541-3463341.

E-mail address: [email protected] (J. Harford).

Journal of Financial Economics 57 (2000) 385}415

The cash-#ow permanence and informationcontent of dividend increases versus

repurchasesq

Wayne Guay!, Jarrad Harford",*!The Wharton School, University of Pennsylvania, Philadelphia, PA 19103-6365, USA"Lundquist College of Business, University of Oregon, Eugene, OR 97403-1208, USA

Received 14 December 1998; received in revised form 16 September 1999

Abstract

We hypothesize that "rms choose dividend increases to distribute relatively permanentcash-#ow shocks and repurchases to distribute more transient shocks. As predicted, we"nd that post-shock cash #ows of dividend increasing "rms exhibit less reversion topre-shock levels compared with repurchasing "rms. We also examine whether the stockmarket uses the announcement of the payout method to update its beliefs about thepermanence of cash-#ow shocks. Controlling for payout size and the market's expecta-tion about the permanence of the cash-#ow shock, the stock price reaction to dividendincreases is more positive than the reaction to repurchases. ( 2000 Elsevier Science S.A.All rights reserved.

JEL classixcation: G35; G32

Keywords: Payout policy; Stock repurchase; Buy-back; Payout choice; Dividend signaling

FINEC=1086=KGM Shankar=Venkatachala=BG

0304-405X/00/$ - see front matter ( 2000 Elsevier Science S.A. All rights reserved.PII: S 0 3 0 4 - 4 0 5 X ( 0 0 ) 0 0 0 6 2 - 3

1. Introduction

Corporations use dividends and share repurchases as the two main methodsto distribute cash to shareholders. While each method has received considerableattention in the academic literature, fewer studies examine the choice betweenrepurchases and dividends. In particular, we do not know what factors drive thechoice between dividends and repurchases and what, if any, information inves-tors infer from this managerial choice.

We hypothesize that "rms choose between repurchases and dividends todistribute cash-#ow shocks based, in part, on the permanence of the shocks.Because dividend increases are implicitly permanent commitments, we predictthat repurchases disburse temporary cash-#ow shocks while dividend changesdisburse relatively more permanent shocks. Further, under the hypothesis thatmanagement's choice of payout method is driven by its expectations aboutcash-#ow permanence, announcements of dividend increases and repurchaseswill convey di!erent information to investors. Thus, we test a second, relatedhypothesis: When stock prices do not fully anticipate the permanence of thecash-#ow shock, the market uses the "rm's choice of payout method to updateits estimate of that permanence. All else equal, a dividend increase will conveymore favorable information about the permanence of the cash-#ow shock thanwill a repurchase. Thus, our second hypothesis is a prediction about theinformation conveyed by the method used to make the distribution, as opposedto the information contained in the amount or occurrence of the distribution.

We "nd, on average, that cash-#ow shocks preceding substantial dividendincreases are signi"cantly more permanent than cash-#ow shocks precedingrepurchases. We de"ne permanence as cash #ows in the period following thepayout decision relative to cash #ows preceding the decision. Compared withcontrol "rms that do not change their cash distributions, dividend-increasing"rms exhibit signi"cantly more permanent cash-#ow shocks. In contrast, repur-chasing "rms' cash-#ow shocks are no more permanent than those of nondis-tributing "rms. Further, we "nd that, on average, the market correctly assessesthat the cash #ows of "rms that subsequently increase their dividends are morepermanent than those of control "rms that do not increase their payout. Themarket-adjusted stock return over the two-year period preceding dividendincreases is signi"cantly greater than the return for control "rms matched oncash #ows and industry. In contrast, the returns of "rms that subsequentlyinitiate repurchases are no di!erent from the returns of control "rms.

We "nd that the stock market assesses the permanence of an individual "rm'scash #ows with error and uses the method of payout to update its assessment ofthe permanence of previous cash-#ow shocks. When the market has identi"eda cash-#ow shock as transient and management chooses to increase dividends,we expect the stock price reaction to the announcement to include an upwardreassessment of cash-#ow permanence. Similarly, when a repurchase is used to

Finec=1086=KGM=VVC

386 W. Guay, J. Harford / Journal of Financial Economics 57 (2000) 385}415

distribute a shock that the market believed to be relatively permanent, a down-ward reassessment of cash-#ow permanence is expected to be part of the stockprice reaction. Consistent with these predictions, we "nd that, controlling for thesize of the payout and the market's prior assessment of the permanence of thecash-#ow shock, stock price reactions to announcements of dividend increasesare greater than the reactions to repurchases and are increasing in our proxy forthe degree of transience the market assigned to the cash-#ow shocks.

In the next section, we review the relevant literature and develop our hypothe-ses. We describe the data and sample selection in Section 3. The cash-#owpermanence of dividend increasers, repurchasers, and control "rms is comparedin Section 4. In Section 5, we explore the information content of dividend-increase and repurchase announcements. We relate our results to prior literaturein Section 6 and conclude in Section 7.

2. Motivation and hypothesis development

In the "rst part of this section, we review the extant literature and place ourhypothesis in context. In the second part, we develop a simple expository modelto explain our hypothesis and frame the discussion for the remainder of thepaper.

2.1. Motivation

This study compares the two main methods of cash payouts: dividendincreases and share repurchases. Extensive literatures have focused on eachmethod in isolation. Researchers "nd clear evidence that the announcement ofeither method conveys positive information about the value of the "rm. Pettit(1972) and Aharony and Swary (1980) document positive stock price reactionsto dividend increase announcements. Dann (1981) concludes that the positivestock price reaction to repurchase announcements is the result of informationsignaling. Analysts react consistently with the market. Ofer and Siegel (1987),Denis et al. (1994), and Carroll (1995) provide evidence that dividend changescause analysts to change their earnings forecasts. Dann et al. (1991) show thatanalyst forecast revisions and earnings surprises follow repurchase tender o!erannouncements and that these surprises are related to the stock price reaction torepurchases. However, less success has been achieved in tying the informationcontent of payout announcements to an observable improvement in future cash#ows or earnings. An early study by Watts (1973) examines dividend announce-ments and "nds that the information content in dividends about future earningsis trivial. Subsequent studies use larger samples and di!erent techniques, butthey produce similar results. Little consistent evidence exists that dividends have

Finec=1086=KGM=VVC

W. Guay, J. Harford / Journal of Financial Economics 57 (2000) 385}415 387

incremental information relative to contemporaneous earnings in predictingfuture earnings (Leftwich and Zmijewski, 1994, DeAngelo et al., 1996; Benartziet al., 1997). Bartov (1991) "nds only what he terms `weaka evidence thatpositive earnings surprises occur in the same year as an open-market repur-chase.

Recent empirical evidence continues to question the roles of dividends andrepurchases as signals of higher future cash #ows. Benartzi et al. (1997) studya comprehensive sample of dividend changes and "nd that if dividend increasessignal anything about future earnings, it is that earnings are less likely to fallthan for similar "rms without dividend increases. They conclude that dividendsare a reaction to past and contemporaneous earnings changes instead of a signalof higher future earnings. Similarly, Lie and McConnell (1998) "nd little evid-ence that repurchase tender o!ers precede higher future performance. Rather,they show that the return on assets observed prior to the repurchase is less likelyto fall in the future. One explanation of these "ndings, tested here, is thatdi!erent payout choices signal varying degrees cash-#ow shock permanence, notgreater future performance.

The small body of research that jointly examines dividends and repurchasesfocuses either on their relative e$cacy as signals of future performance or ontheir relative e$ciency for distributing cash. Talmor and Titman (1990) andBagwell and Shoven (1988) contrast the two methods on tax e!ects whileBarclay and Smith (1988) highlight di!ering transaction costs. Ofer and Thakor(1987) demonstrate theoretically that repurchase tender o!ers should be used tocorrect large misvaluations while dividends are more e$cient for smaller mis-valuations. Choi and Chen (1997) "nd empirical support for the predictions ofOfer and Thakor (1987) by documenting that the stock price reaction is largerfor a repurchase tender o!er announcement than for a dividend change, evenafter controlling for the size of the distribution. A recent paper by Bartov et al.(1998) examines a sample of 260 "rms, half of which repurchase shares and halfof which pay dividends. The authors hypothesize that managers choose a repur-chase over a dividend if management views its shares as greatly undervalued,management has stock options that are not dividend-protected, or a largefraction of equity ownership is held by institutions (which seem to showa preference for repurchases). Their evidence generally supports their predic-tions. Fenn and Liang (2000) also recognize that dividends reduce executivestock option values while repurchases do not. Denis (1990) and Bagwell(1992) identify takeover defense as an alternative motivation for repurchases.While we agree that these factors in#uence the distribution decision, we view ourpaper as complementary in that we focus on a fundamental determinant of thedecision } the permanence of the underlying cash #ows. Contemporaneous andindependent work by Jagannathan et al. (2000) also focuses on cash-#owvariation as a determinant of payout method. We relate our results to theirs inSection 6.

Finec=1086=KGM=VVC


1 In a comprehensive study of actual share acquisitions in open-market repurchase programs inthe 1980s, Stephens and Weisbach (1998) "nd that only 10% of the sample "rms have more than onerepurchase in any three-year period.

2.2. Hypothesis development

We hypothesize that "rms use both dividend increases and share repurchasesto distribute past and contemporaneous cash-#ow shocks. This view is broadlyconsistent with the evidence in Bartov (1991) and Benartzi et al. (1997). We arguethat the choice between these two payout methods depends partly upon theexpected permanence of the realized cash-#ow shock. Lintner (1956), Fama andBabiak (1968), and Asquith and Mullins (1983), among others, provide evidencethat dividend increases are intended to be and are, on average, permanent. Sharerepurchases, by comparison, are generally stand-alone actions that are takeneach time shares are acquired (though they may be part of a pattern of plannedrepurchases).1 Further, Stephens and Weisbach (1998) document that not allannounced repurchase programs are completed and that the average comple-tion rate is approximately 80% of announced shares. This suggests that ifmanagers are unsure about the permanence or size of the contemporaneouscash-#ow shock, they may choose a repurchase over a dividend increase to allowthem #exibility in abandoning the planned distribution. Therefore, becausea dividend increase is implicitly more permanent, we hypothesize that dividendincreases follow cash-#ow shocks that are more permanent than those followedby repurchases.

When the market does not perfectly anticipate the permanence of a cash-#owshock, the hypothesized relation between the method of payout and the perma-nence of future cash #ows has implications for the information content ofdividend increases relative to repurchases. If, as we predict, "rms use dividendchanges to adjust cash payouts when cash-#ow changes are relatively morepermanent, an increase in the dividend provides the market with favorableinformation about the extent to which past and contemporaneous cash #ows arelikely to continue. However, when a "rm uses a share repurchase instead ofa dividend increase, the market, on average, receives less favorable informationabout the permanence of a cash-#ow shock. The magnitude of the price reactiondepends upon the extent to which a payout announcement causes the market tochange its expectations about the permanence of a "rm's cash #ows. As such, itis important to consider the information content of the payout choice condi-tional on the market's expectations about cash-#ow permanence.

A simple model aids the exposition of our predictions. Assume that a "rm'scash-#ow shocks either immediately dissipate or are completely permanent. The"rm receives a positive cash-#ow shock during period 1. Its cash #ows in period 1are its normal level of cash #ows (CF) plus the shock (Shock): CF#Shock. In

Finec=1086=KGM=VVC


2While our model is intended only to ease the exposition of the paper, the payout choice bymanagers can be made to be a separating equilibrium by continuing the life of the "rm past twoperiods and making it prohibitively costly for managers of "rms with temporary shocks to continuepaying higher dividends past period 2. For example, the separation can be achieved if we assumecostly external "nancing and a penalty for omitting a dividend in period 3.

the future, the "rm's cash #ows will be CF#p ) Shock, where p is the perma-nence parameter, taking the value of either zero or one. The "rm exists for twoperiods and makes a distribution announcement at the beginning of period 2.Because this model is purely expository, and because our predictions aredemonstrated most clearly in this simple setting, we will abstract from thepossibility that the "rm retains the cash-#ow shock instead of distributing it.Later, we will consider how this restriction a!ects the predictions.

Assuming a discount rate of zero without loss of generality, the price of the "rmin period 1 when the shock is observed will be P

1, where P

1"(CF#Shock)

#[CF#Pr(p"1 D c)Shock]. The managers observe the permanence para-meter, p, but the market does not. Therefore, the market must assess theprobability that the permanence parameter equals one based on its informationat the time of the shock, represented by c. The managers then make a distribu-tion announcement. If the shock is permanent, they choose a dividend; if theshock is temporary, they choose a repurchase.2 The market observes the choiceof distribution method and updates its belief about the permanence of the shock.Thus, the "rm's price after the distribution announcement is P

2, where

P2"2(CF#Shock) if a dividend is announced, and P

2"(CF#Shock)#CF

if a repurchase is announced.The change in price resulting from the announcement of the distribution is

decreasing in Pr(p"1 D c). If the managers announce a dividend increase, thenPr(p"1) goes to one, and the increase in price is a decreasing function of theprobability the market placed on observing a dividend increase. If the managersannounce a repurchase, then Pr(p"1) goes to zero. In this case, the price willfall, unless Pr(p"1 D c)"0, meaning the market completely anticipated therepurchase. As the market puts more weight on the possibility that the distribu-tion will be a dividend increase, the price change upon announcement ofa repurchase decreases. Thus, for both types of distribution announcements, theannouncement return, conditional on the chosen method, is decreasing in theprior weight put on p being equal to one.

One aspect of the above model that requires clari"cation is the prediction thatthe market responds negatively to share repurchases. This prediction, which isnot empirically descriptive, stems from the simplifying assumption that only twotypes of "rms exist: dividend increasers and repurchasers. When the model isexpanded to allow for "rms that make no cash payout, this counterintuitiveprediction is eliminated. To see this, reconsider the model with the followingchange: At the time of the payout decision, with some small probability, the

Finec=1086=KGM=VVC


3We consider self-tender repurchases in our sensitivity analysis at the end of Section 5. Weexclude self-tenders from the main sample because we believe that these repurchases re#ecta substantially di!erent event than open-market repurchases. Self-tenders normally involve at leasttwice the proportion of shares that open-market programs do and result in a signi"cant change inthe ownership concentration in the "rm. Further, other studies "nd that many self-tender o!ers areinitiated in anticipation of, or in competition with, an outside takeover o!er (Dann and DeAngelo,1988; Denis, 1990).

4An alternative approach is to start with a sample of "rms that experience large cash-#ow shocksand then track payout decisions. We feel that this alternative approach is better suited to addressinga di!erent question}why some "rms distribute cash-#ow shocks and others do not. Because weexplore how "rms choose between alternative payout methods, our approach o!ers two advantages:First, we do not force the sample "rms to have a cash-#ow shock of some arbitrary magnitude;second, our design identi"es all "rms that increase dividends or announce repurchases. This secondpoint is important because it allows us to begin with a complete sample of "rms making thesedistribution changes.

manager receives a negative signal about period 2 cash #ows (CF) and, insteadof distributing the shock, can choose to retain it to cover the expected cashshortfall. In this case, both dividend increases and repurchases will producepositive stock price reactions, though the size of the reactions continues to bedecreasing in the market's prior assessment of the permanence of the period 1cash-#ow shock.

To summarize, the two predictions of the permanence hypothesis are: (1) thecash-#ow shock preceding a dividend increase will have a larger permanentcomponent than a cash-#ow shock preceding a repurchase, and (2) the marketwill use management's choice of payout method to update its belief about thepermanent component of the cash-#ow shock.

3. Data

We identify a sample of "rms that declare either regular, quarterly dividendincreases or open-market repurchase authorizations.3 Working from the distri-bution decision date during a "scal year t, we compute baseline cash #ows overyears t!4 through t!2, the cash-#ow shock during years t!1 and t, andfuture cash #ows from years t#1 through t#3. Using these cash-#owmeasures, we compare the permanence of cash #ows between the two types of"rms.4

We begin with a sample of open-market repurchase authorization announce-ments from 1981 to 1993. These data are gathered from the Securities DataCompany database of repurchases. If a "rm announces two repurchases withinone "scal year, we eliminate the second one. These second repurchases could bereauthorizations of the previously announced programs. To this repurchasesample, we add all regular, quarterly dividend increases made by "rms in the

Finec=1086=KGM=VVC


5There are 111 "rms that announce both a dividend increase and a repurchase within the samemonth. Our hypotheses suggest that the cash-#ow shocks experienced by these "rms are likely to bepartially permanent and partially transitory. For parsimony, we do not treat these "rms asa separate sample. Because our tests examine di!erences in cash-#ow permanence between samples,the decision to include these "rms should bias against "nding the results we document. We also notethat excluding these "rms, or "rms that use one payout method in year t and a di!erent method inyear t#1, from the sample does not a!ect our inferences.

6Some "rms announce substantial dividend increases two years in a row. Because our researchdesign examines "rms' cash-#ow patterns over multiple years, these overlapping observations canbias our test statistics. To explore this possibility, we remove the second occurrence in each set ofoverlapping observations and repeat the tests. Our results are qualitatively unchanged.

Center for Research in Security Prices (CRSP) database over the same 13-yeartime period. A dividend increase is included only when other quarterly dividendchanges within the "scal year are either positive or zero; that is, when bothnegative and positive quarterly changes occur within the same "scal period, theobservation is excluded. The resulting data set contains 1,068 repurchases and5,007 dividend changes.5 Table 1 presents the distribution of the events acrosstime. While the use of repurchases increased over our sample period, theincidence of dividend increases has remained relatively stable. The robustness ofour results to subperiod analysis is discussed in Section 5.

We divide the dividend increase sample into two subgroups termed substantialdividend increases and small/routine dividend increases. The substantial dividendgroup contains increases that are not preceded by a dividend increase in theprevious year or are larger than the dividend increase in the previous year(dividend increases are measured as the dollar change in the quarterly dividendsover the entire year scaled by total assets at the beginning of the year). Thesmall/routine group has all remaining increases. We expect that small/routinedividend increases carry signi"cantly less information about cash-#ow perma-nence than a more substantial increase, because these changes are less likely tobe viewed as the primary payout decision in the two-year shock window. Thissubdivision results in 2,961 substantial dividend increases and 2,046small/routine increases.6

4. The cash-6ow permanence of dividend increasers versus repurchasers

In this section, we test our primary hypothesis that the distribution method isrelated to the permanence of previous and contemporaneous cash #ows. Our"ndings support the hypothesis that the cash-#ow shocks of "rms that substan-tially increase dividends are signi"cantly more permanent than the shocks ofrepurchasers or small/routine dividend increasers. The cash #ows of substantialdividend increasers are also more permanent than the cash #ows of control "rmsin the same industry with similar-sized shocks.

Finec=1086=KGM=VVC


Table 1The distribution over the sample period (1981}93) of the sample "rms increasing dividends orinitiating an open-market repurchase. The repurchases come from all open-market repurchaseannouncements recorded in the Securities Data Company database. We eliminate all repurchasesthat are preceded by another one in the prior four quarters. These repurchases could be reauthoriz-ations of the previously announced program. We use the Center for Research in Securities Pricesdatabase to identify all "rms increasing their dividends over the sample period. A dividend increaseis de"ned to occur when the current year's total quarterly dividend payout is greater than theprevious year's quarterly dividend payout. Annual dividend increases are included only when eachquarterly dividend change within a "scal year is either positive or zero; that is, when both negativeand positive quarterly changes occur within the same "scal period, the observation is excluded.

Year Repurchases Dividend increases

1981 4 4461982 6 2961983 30 2911984 126 4061985 41 3891986 62 3671987 158 3981988 53 4781989 156 4691990 160 4161991 65 3571992 106 3781993 101 316

Total 1,068 5,007

4.1. Summary statistics

Table 2 presents summary statistics for the samples of repurchasers anddividend increasers. While the repurchasers and substantial dividend increasersare similar in size, the small/routine dividend increasers are somewhat larger.The median market value of equity is $470 million for repurchasers, $440 millionfor substantial dividend increasers, and $656 million for small/routine dividendincreasers. The relatively bigger size of the routine dividend increasers is consis-tent with the notion that, on average, larger, more mature "rms use dividendsmore regularly. Although the market-to-book ratios di!er statistically betweensubstantial dividend increasers and repurchasers, the mean and median valuesfor the repurchasers are within 10% of the respective mean and median valuesfor both samples of dividend increasers. The leverage ratios indicate thatdividend increasers tend to be more levered than their repurchasing counter-parts.

Table 2 also contains information on the size of the cash distributions. Tocompare the cash payouts across the dividend increase and repurchase samples,

Finec=1086=KGM=VVC


Tab

le2

Sum

mar

yst

atistics

onge

nera

lcha

ract

eristics

ofth

esa

mpl

e"rm

s,al

ong

with

test

sfo

rdi!

eren

cesbet

wee

nth

esu

bsam

ple

s.R

epur

chas

ersar

eco

mpa

red

with

the

two

groups

ofdiv

iden

din

crea

sers

:sm

all/ro

utin

ein

crea

ses

and

subst

antial

incr

ease

s.A"rm

that

has

nothad

adi

viden

din

crea

sein

the

prior

year

orw

hose

incr

ease

isla

rger

than

that

from

thepre

vious

year

isin

clud

edin

thesu

bst

antial

incr

ease

sgr

oup.A

ll"nan

cial

char

acte

rist

icsar

eco

mput

edat

theen

doft

he"sc

alye

arprior

toth

epa

youtch

ange

.The

mar

ket

valu

eofeq

uity

(MV

)is

com

put

edas

the

endi

ng

shar

epr

ice

mul

tiplie

dby

the

num

berofs

hare

sou

tsta

ndin

g.M

arke

t-to

-book

(ass

ets)

isco

mput

edas

(boo

kas

sets!

book

equi

ty#

mar

ket

valu

eof

equity)

/book

asse

ts.

Lev

erag

era

tio

isboo

klia

bili

ties

/mar

ketva

lueofe

quity.

The

expe

cted

incr

ease

indi

vide

ndpay

outis

thean

nual

ized

chan

gein

qua

rter

lydi

viden

dsdi

vide

dby

a10

%di

scount

rate

and

scal

edby

the

stock

pric

e"ve

day

sprior

toth

edi

vide

ndch

ange

annou

nce

men

t.P

erce

nta

geofsh

ares

soug

htis

base

don

stat

emen

tsm

ade

byth

eco

mpa

nyas

reco

rded

by

Sec

urities

Dat

aC

om

pany.

Cum

ulat

ive

abno

rmal

retu

rns

(CA

R)fo

rth

ean

nou

nce

men

tper

iod

are

com

pute

dba

sed

on

mar

ket

model

regr

essions

for

day

s!

252

to!

20re

lative

toth

ean

noun

cem

ent.

Tes

tsfo

rdi!er

ence

for

the

mea

nsar

et-te

sts

and

the

t-st

atistics

are

pres

ente

d.

Tes

tsfo

rdi!er

ence

for

the

med

ians

are

sign

ed-r

ank

test

san

dth

eZ

-sta

tist

ics

are

pre

sente

d.

Cha

ract

eristic

ofsa

mpl

e"rm

sM

ean

Rep

urch

aser

sSub

stan

tial

Tes

tfo

rSm

all/

Tes

tfo

ran

ddiv

iden

ddi!

eren

cero

utine

di!

eren

cem

edia

nin

crea

sers

with

div

iden

dw

ith

repurc

hase

rsin

crea

sers

repurc

hase

rs

Mar

ket

valu

eofeq

uity

Mea

n2,

111.

541,

924.

070.

972,

168.

60!

0.78

Med

ian

469.

5443

9.90

0.52

655.

51!

4.57

!

Book

asse

tsM

ean

2,18

4.19

2,27

6.51

!0.

432,

235.

01!

0.24

Med

ian

447.

6049

5.52

!2.

09!

622.

49!

5.23

!

Mar

ket

-to-b

ook

(ass

ets)

Mea

n1.

671.

543.

94!

1.66

!0.

93M

edia

n1.

371.

284.

47!

1.39

!1.

57

Lev

erag

era

tio

Mea

n0.

730.

94!

6.62

!0.

80!

2.40

"

Med

ian

0.49

0.64

!7.

05!

0.55

!3.

31!

Exp

ecte

din

crea

sein

divi

den

dpay

out

(%of

MV

)M

ean

4.56

2.77

Med

ian

3.16

2.21

%sh

ares

sough

t#M

ean

7.69

Med

ian

5.85

Ann

oun

cem

ent

CA

R(%

)M

ean

2.06

1.19

2.94

!0.

266.

04!

Med

ian

1.81

0.88

3.47

!0.

167.

37!

Num

ber

ofob

serv

atio

ns1,

062

2,94

32,

045

!Sig

ni"

cantat

the

1%le

vel.

"Si

gni"

cant

atth

e5%

leve

l.#T

he

num

ber

ofobs

erva

tions

with

dat

aav

aila

ble

for

shar

esso

ugh

tis

676.

Finec=1086=KGM=VVC


7Alternatively, we could estimate a separate required return on equity for each "rm. However, themeasurement error introduced by estimating a beta for each "rm and a risk-free rate and expectedreturn on the market for each date is potentially as large, if not larger, than the error introduced byusing a constant discount rate. Following a similar logic, Jensen (1993) uniformly applies a discountrate of 10% to all "rms in computing performance measures. The perpetuity assumption is based onmanagers' reluctance to reduce dividends. To the extent that this assumption is incorrect, weoverestimate the size of the payout for dividend increasers, which biases against our later "ndingthat dividend increases receive a larger stock price reaction than repurchases, controlling for payoutsize.

we produce a payout measure that characterizes dividend increases as perpetu-ities. We estimate the present value of the cash paid out via a dividend increaseas the annualized increase in dividends divided by a discount rate of 10%.7 Wethen scale this measure by the "rm's share price "ve days prior to the dividendincrease announcement. The mean present value of the expected dividendpayout is 4.6% of market value for substantial dividend increasers and 2.8% forsmall/routine dividend increasers. As a percentage increase from the previousdividend level, the median dividend increase is 13.2% for substantial increasersand 8.6% for small/routine increasers (not tabulated). For the 676 repurchasing"rms with available data, the mean percentage of outstanding shares sought is7.7%. Thus, repurchases are larger than dividend increases, on average, even ifthe dividend increase is assumed to be permanent.

4.2. The size and permanence of the cash-flow shock

Consistent with previous studies (e.g., Dechow, 1994), cash #ow from opera-tions is computed as:

CFOt"Operating income before depreciation

t!Interest

t

!Taxest!DWorking capital

t.

We scale cash #ow from operations by beginning-of-period assets to reduceheteroskedasticity and spurious correlations stemming from "rm size.

Table 3 indicates that the size and permanence of the cash-#ow shocksexperienced by the sample "rms are broadly consistent with our predictions.The timing and de"nition of these measures are illustrated in Fig. 1. Year 0 is the"scal year in which the dividend is increased or the share repurchase is authoriz-ed. We measure the cash-#ow shock by comparing the average cash #ow inyears !1 and 0 with the average cash #ow in years !4, !3 and !2. The rawcash-#ow shock is reported in Panel A and is de"ned as

[Avg (Cash #ow/Total Assets) in years!1 and 0]

![Avg (Cash #ow/Total Assets) in years!4 to!2].

Finec=1086=KGM=VVC


Table 3Medians of the cash-#ow shock, reversion of the shock, and permanence of the shock for 1,068repurchasers, 2,961 substantial dividend increasers, and 2,046 small/routine dividend increasers.Cash-#ow shock, reversion, and permanence are measured as raw di!erences in the ratio of cash#ow-to-assets in Panel A and percentage changes in the ratio of cash #ow-to-assets in Panel B.! Thetable also contains test statistics for the hypotheses that the medians are not di!erent between therepurchasing sample and the dividend increasing samples. Dividend increases are subdivided intotwo groups: small/routine increases and substantial increases. A "rm that has not had a dividendincrease in the prior year or whose increase is larger than that from the previous year is included inthe substantial increases group.

Firms Cash-#ow shock Reversion Permanence(median) (median) (median)

Panel A: Raw diwerence in cash yow-to-assets

Repurchasers 0.005" !0.009" !0.009"

Substantial dividend increases 0.007" !0.006" 0.002Z-test for substantial dividendincreasers versus repurchasers

!1.587 !2.160# !4.425"

Small/routine dividend increases 0.000 !0.005" !0.008"

Z-test for small dividend increasersversus repurchasers

3.086" !2.303# !0.262

Z-test for small versus substantialdividend increasers.

!6.878" 0.329 !6.565"

Panel B: Percentage change in cash yow-to-assets

Repurchasers 4.218%" !7.068%" !8.578%#

Substantial dividend increases 6.427%" !4.879%" 1.709%"

Z-test for substantial dividendincreasers versus repurchasers

!1.785$ !2.292# !4.660"

Small/routine dividend increases !0.204% !4.844%" !7.089%"

Z-test for small dividend increasersversus repurchasers

3.279" !2.540# !0.405

Z-test for small versus substantialdividend increasers

7.413" !0.315 6.714"

!Cash-#ow shock is: [Average (Cash Flow/TA) in years !1 and 0]![Average (Cash #ow/TA) inyears !4 to !2] (TA is Total Assets). Reversion is: [Average (Cash #ow/TA) in years #1 to #3]![Average (Cash #ow/TA) in years !1 and 0]. Permanence is: [Average (Cash #ow/TA) in years#1 to #3]![Average (Cash #ow/TA) in years !4 to !2]. Percentage changes are computedby scaling the above raw di!erence by the second term in each expression."Signi"cantly di!erent from zero at the 1% level.#Signi"cantly di!erent from zero at the 5% level.$Signi"cantly di!erent from zero at the 10% level.

Because "rms di!er in their normal cash #ow-to-asset ratios, we also report thepercentage change in the cash #ow-to-assets ratios in Panel B as an alternativemeasure of the cash-#ow shock. We compute the percentage change measure byscaling the raw change in cash #ows de"ned above by the average cash

Finec=1086=KGM=VVC


8Approximately three hundred observations are removed because of negative average cash #owsin the denominator of the percentage change measures. The median cash-#ow shock for the removed"rms is negative compared with a signi"cantly positive median cash-#ow shock for the remainingsample "rms. Further, these contrasting median shocks persist when the negative cash-#ow "rms arepartitioned based on payout type. These di!erences suggest that negative cash-#ow observationsre#ect a substantially di!erent economic event than our treatment "rms.

9The analysis is robust to using the means of the raw changes. For the percentage changes, thedistributions are substantially more skewed, and while the di!erences have the same sign, thesigni"cance levels are reduced.

Fig. 1. Cash #ow time series. The Cash-#ow shock is (Shock Cash-#ows!Pre-shock Cash-#ows).Reversion is (Post-shock Cash-#ows!Shock Cash-#ows). Permanence is (Post-shock Cash-#ows!Pre-shock Cash-#ows). All cash #ows are scaled by contemporaneous total assets.

#ow-to-assets in years !4 to !2. We remove from the sample the mostextreme 1% of the observations for each of our shock and permanencemeasures, as well as observations with negative denominators.8 We report onlymedian values of our cash-#ow measures for parsimony and because thedistributions tend to be skewed.9

The median raw cash-#ow shock experienced by repurchasers is 0.005 com-pared with a somewhat larger shock of 0.007 for substantial dividend increasers.The median shock as a percentage of average pre-shock cash #ows is 4.2% forrepurchasers and 6.4% for substantial dividend increasers. Small/routine divi-dend increasers have a median raw cash-#ow shock of 0.000, considerablysmaller than for either of the other two samples. Thus, the two dividendsubsamples appear to make their payout choices in response to substantiallydi!erent cash-#ow patterns.

Table 3 also includes two measures of the permanence of the cash-#ow shock.The "rst is termed reversion and is de"ned as

[Avg (Cash #ow/Total Assets) in years#1 to#3]

![Avg (Cash #ow/Total Assets) in years!1 and 0].

Finec=1086=KGM=VVC


10The fact that the average reversion of the entire sample is negative is not surprising. Dechowet al. (1998) model the time-series properties of operating cash #ows and the accounting process thatincorporates expected future cash #ows in earnings. They show that when demand for a "rm'sproducts varies over time, di!erences in the timing of cash outlays and in#ows leads to negativeserial correlation in cash-#ow changes. For example, a "rm that experiences a positive shock inproduct demand is likely to make cash outlays in the current period for inventories and accountspayable but not fully receive the cash in#ows until some future period when sales are made andaccounts receivable are collected. They also provide empirical support for their predictions by"nding a signi"cant negative serial correlation of !0.28 between cash-#ow changes.

Reversion captures the extent to which future cash #ows remain at the level ofcash #ows during the shock period. We argue that managers who react toa cash-#ow shock by increasing dividends expect some portion of that shock tobe permanent. Further, the proportion of the shock that is permanent forsubstantial dividend-increasing "rms should be greater than the proportion foreither repurchasing "rms or small/routine dividend increasers. Consistent withour measurement of the cash-#ow shock, we also provide results for a percent-age change measure of reversion that scales the raw reversion measure above byaverage cash #ows in years !1 and 0.

Our second measure of the permanence of the shock, termed permanence,di!ers from the "rst by its comparison period

[Avg (Cash #ow/Total Assets) in years#1 to#3]

![Avg (Cash #ow/Total Assets) in years!4 to!2].

This measure is designed to capture the degree to which future cash #ows settleabove or below their pre-shock level. We predict that "rms with substantialdividend increases expect their cash #ows to settle above prior levels to a greaterextent than repurchasing "rms' cash #ows. We also report results for a percent-age change measure of permanence, which scales the raw permanence measureby average cash #ows in years !4 to !2.

The results in Panels A and B of Table 3 are consistent with our hypotheses.The reversion measures indicate that shocks followed by substantial dividendchanges are more permanent than those followed by repurchases. The medianraw reversion measures indicate that cash #ows of substantial dividend in-creasers drop by 0.6 percent of assets after the change in payout, while those ofrepurchasers drop by 0.9 percent of assets.10 The permanence measures, whichre#ect persistence of cash #ows above pre-shock levels, show that future cash#ows of the median substantial dividend increaser are slightly above the pre-shock level. In contrast, repurchasing "rms show a signi"cant reduction incash #ows. The percentage change permanence measure in Panel B revealsthat repurchaser cash #ows are approximately 91% of pre-shock levels

Finec=1086=KGM=VVC


11Though we do not explicitly explore the cash-#ow permanence of dividend-initiating "rms, weidentify a sample of 97 "rms that begin paying dividends between 1981 and 1993 and meet our datarequirements for computing the reversion and permanence statistics. Consistent with "rms initiatingdividends only when cash-#ow shocks are expected to be highly persistent, the median permanencemeasure for dividend-initiating "rms is very large, with future cash #ows remaining at 106% of pre-shock cash #ows.

while dividend-increasing "rms have cash-#ow levels approximately 102% ofpre-shock levels. Both percentages are signi"cantly di!erent from 100% of pre-shock levels.

While the reversion measures are similar for the substantial and small/routine dividend increasers, the permanence measures are markedly di!erentbetween these samples because of the considerably larger cash-#ow shocksexperienced by the substantial increasers. In contrast to the marginally positivepermanence measures of substantial dividend increasers, small/routine dividendincreasers, like the repurchasers, experience a signi"cant drop in future cash#ows to levels that are approximately 93% of pre-shock cash #ows. Thisdi!erence between the dividend groups is statistically signi"cant at the 1% level.These results highlight the importance of the earlier supposition that not alldividend increases convey similar information about the permanence of the"rm's cash #ows.

As a robustness check, we constrain the repurchase sample to contain only the697 "rms that also pay dividends. One could argue that these "rms legitimatelyhad both methods of cash distribution available to them: an increase in theirdividend or a repurchase. Firms without an established dividend would have toinitiate a dividend as an alternative to a repurchase}an event that is likely todi!er substantially from an increase in the regular dividend.11 Using thisrestricted sample does not change the inferences from Table 3. We discuss therobustness of our results using the restricted sample throughout the remainderof the analysis and, where appropriate, present tabulated results based on thatsample.

The evidence in this section is consistent with the hypothesis that, relative tosubstantial dividend increases, repurchases are used to distribute more transientshocks. However, these "ndings could be in#uenced by "rm-speci"c or industry-speci"c characteristics of the sample "rms. To explore this issue further, wereport reversion and permanence measures for a sample of control "rms that didnot increase dividends or initiate repurchases. The initial pool of "rms for thecontrol sample consists of all "rms with su$cient data available on CRSP andCompustat to compute our measures of cash-#ow shock, reversion, and perma-nence. From this group, we remove all "rm-year observations that overlap withour initial sample of repurchasers and dividend increasers. We create 20 equal-sized control portfolios based on the size of the cash-#ow shock and match each

Finec=1086=KGM=VVC


12We have repeated the matching procedure using size of cash-#ow shock and "rm size as thematching criteria. The results are qualitatively unchanged.

of our repurchase/dividend change sample "rms to a portfolio based on theranking of their cash-#ow shocks. To control for industry-speci"c di!erences inreversion and permanence, we also restrict each control portfolio to containonly "rms with the same two-digit Standard Industrial Classi"cation (SIC) codeas the sample "rm.12

Table 4 reports the median reversion and permanence measures for both thesample "rms and the matching "rms. It also presents the median di!erencebetween the sample and matching "rms for these measures and tests for thesigni"cance of the di!erences. The matching "rms exhibit considerably morenegative reversion and less permanence than the substantial dividend-increasing"rms. This "nding holds for both the raw di!erences and percentage changes incash #ow-to-assets. In contrast, the repurchasers and small/routine dividendincreasers have reversion and permanence measures that are similar to those ofthe matching "rms.

Finally, Table 4 includes two-year market-adjusted buy-and-hold stock re-turns measured over the cash-#ow shock period. Given that returns anticipatefuture cash #ows, we expect that, controlling for industry and the size of thecash-#ow shock, the stock returns contemporaneous with the cash-#ow shockare higher for "rms that subsequently increase their dividends than for control"rms or "rms that repurchase shares. The median return for the repurchasing"rms is marginally positive, at 2%, but not signi"cantly di!erent from thereturns of their matching "rms. This is consistent with our "nding that repur-chasers' cash-#ow shocks revert quickly and their cash #ows are no morepermanent than those of the control "rms. In contrast, the median market-adjusted return for "rms in the substantial dividend-increase sample is a signi"-cantly positive 12% and substantially greater than the median return for thecontrol "rms. The returns of the small/routine dividend-increase sample aresimilar to those of the substantial dividend increasers, but less pronounced.Thus, our results indicate that for dividend-increasing "rms, but not for repur-chasers, the market anticipates future cash #ows that are higher than those oftheir control sample counterparts.

To summarize, we establish that cash-#ow shocks experienced by "rms thatmake substantial dividend increases are more permanent and revert less thanthose experienced by "rms making repurchases or small/routine dividend in-creases and an industry-and-cash-#ow-shock-matched control sample. Further,consistent with the market accurately anticipating cash-#ow permanence onaverage, dividend increasers' stock returns over the cash-#ow shock period aregreater than those of "rms from the same industry with similar cash-#ow shocks,but that do not increase their payouts.

Finec=1086=KGM=VVC


Tab

le4

Sam

ple

repurc

has

ing

and

div

iden

d-in

crea

sing"rm

sm

atch

edw

ith"rm

sth

athav

eno

tch

ange

dth

eir

payo

utpo

licy.

The

cont

rolsa

mpl

eis

gene

rate

dby

crea

ting

20po

rtfo

lios

ofca

sh-#

owsh

ocks

and

mat

chin

gea

chsa

mple"rm

toa

port

folio

with

asim

ilar-

size

dca

sh-#

ow

shoc

k.E

ach

cont

rolpor

tfolio

isre

strict

edto

cont

ain

onl

y"rm

sw

ith

the

sam

etw

o-dig

itSta

ndar

dIn

dust

rial

Cla

ssi"

cation

(SIC

)code

asth

esa

mple"rm

.The

raw

and

perc

enta

ge-c

hange

reve

rsio

nan

dpe

rman

ence

mea

sure

s(a

sde"ned

abov

ein

the

footn

ote

sto

Tab

le3)

are

com

pute

dfo

rth

ere

purc

has

ing

subsa

mple

,the

divi

den

d-incr

easing

subs

ample

s,an

dth

em

atch

ing"rm

sfo

rea

chsa

mple

.The

med

ian

di!er

ence

betw

een

the

sam

ple"rm

san

dth

em

atch

ing"rm

sar

eal

sore

port

edfo

rea

chca

se.M

arket

-adju

sted

buy-

and-h

old

retu

rns

for

the

cash

-#ow

shoc

kpe

riod

are

also

com

pute

dfo

rsa

mple

and

cont

rol"rm

s.The

med

ians

for

each

subsa

mple

along

with

the

med

ian

paired

di!

eren

cebet

wee

nth

esa

mpl

e"rm

san

dth

eir

contr

ol"rm

sar

epr

esen

ted.

The

sam

ple

size

sar

eslig

htly

smal

ler

than

inTab

le3

bec

ause

adeq

uat

em

atch

ing"rm

sco

uld

not

beid

enti"ed

.Rep

urch

aser

snum

bere

d1,

051"rm

s;su

bsta

ntia

ldiv

iden

din

crea

sers

,2,9

00"rm

s;an

dsm

all/ro

utine

divi

dend

incr

ease

rs,2

,010"rm

s.Bec

ause

obse

rvat

ions

acco

unting

for

the

extr

eme

1%of

shock

,rev

ersion

,orpe

rman

ence

areex

clud

ed,t

hesa

mpl

e"rm

sdi!er

slig

htly

from

Pan

elA

toPan

elB

(acc

oun

ting

forth

edi!

eren

cein

sam

ple"rm

retu

rns

betw

een

the

two

panel

s).

Cont

empor

aneo

ustw

o-y

rre

turn

sR

ever

sion

(med

ian)

Per

man

ence

(med

ian)

(med

ian

%)

Sam

ple

Mat

chin

gD

i!er

ence

Sam

ple

Mat

chin

gD

i!er

ence

Sam

ple

Mat

chin

gD

i!er

ence

"rm

s"rm

s"rm

s"rm

s"rm

s"rm

s

Pan

elA

:Raw

diw

eren

cein

cashy

ow-t

o-as

sets

Rep

urc

has

ers

!0.

009!

!0.

012!

0.00

3!

0.00

9!!

0.00

80.

003

2.02

92.

225

!1.

734

Subst

antial

div

iden

din

crea

ses

!0.

006!

!0.

010!

0.00

7!0.

002

!0.

005

0.00

7!12

.033

!6.

201!

6.66

8!

Smal

l/ro

utin

ediv

iden

din

crea

ses

!0.

005!

!0.

006!

0.00

2!!

0.00

8!!

0.00

7!0.

002!

8.62

4!5.

692!

4.38

0!

Pan

elB

:Per

cent

age

chan

gein

cashy

ow-t

o-as

sets

Rep

urc

has

ers

(%)

!6.

900!

!10

.791

!!

1.88

9!

8.68

0"!

7.10

2#!

1.34

12.

386#

2.68

1"!

3.10

3

Subst

antial

div

iden

din

crea

sers

(%)

!4.

807!

!8.

528!

2.68

3!1.

728!

!5.

426

2.73

8!12

.007

!8.

563!

5.11

2!

Smal

l/ro

utin

ediv

iden

din

crea

sers

(%)

!4.

844!

!4.

969!

!0.

200

!7.

047!

!8.

204!

!0.

339

8.73

7!6.

100!

4.12

9!

!Sig

ni"

cantly

di!

eren

tfrom

zero

atth

e1%

leve

l."Si

gni"

cantly

di!er

ent

from

zero

atth

e5%

leve

l.#S

igni"

cantly

di!

eren

tfrom

zero

atth

e10

%le

vel.

Finec=1086=KGM=VVC


Fig. 2. Regression tests timeline.

5. The Information content of the method of payout

In this section we examine the stock price reaction to announcements ofdividend increases and repurchases to test our secondary hypothesis that themarket uses the payout announcement to update its estimate of the permanenceof the cash-#ow shock. If management's choice of payout method is driven by itsexpectations about cash-#ow permanence, announcements of dividend increasesand repurchases are predicted to convey di!erent information to investors.Fig. 2 illustrates an information timeline useful in understanding our regressiontests.

5.1. Development of the tests

To investigate whether the market updates its assessment of cash-#ow perma-nence based on the type of payout announcement, we require an estimate, foreach "rm, of the market's expectation of the permanence of the cash-#ow shock(analogous to c in our stylized model of Section 2). To produce this estimate, weregress each "rm's market-adjusted buy-and-hold return over the eight quartersleading-up to the payout announcement on its cash #ows over the same period:

8quarterReturni"a#b(8quarterCashFlows

i)#e. (1)

The residuals from this regression represent adjusted returns. If a "rm's residualis positive, then its return is higher than average after controlling for the level ofcash #ows, and we view the market as expecting the "rm's cash-#ow shock to berelatively permanent. Similarly, if the adjusted return is negative, we view themarket as expecting the cash-#ow shock to be relatively transient.

Finec=1086=KGM=VVC


We can now specify, in terms of the adjusted returns, our empirical predic-tions about the relation between the payout method, market expectations aboutcash-#ow permanence, and payout announcement returns. The market forecastscash-#ow permanence with error. Because our evidence indicates that the choiceof payout method is associated with cash-#ow permanence, we expect that themarket will update its expectations about permanence based on the payoutmethod that the "rm announces. If the adjusted stock return in the periodpreceding the payout announcement is high, and the "rm chooses a repurchaseto distribute cash #ows, we predict that the stock price reaction to the an-nouncement will be smaller than average. This is because part of the market'sreaction will re#ect a downward adjustment of its expectation of cash-#owpermanence. In terms of the model from Section 2, this is a case wheremarket participants assigned too high a probability to the greater permanence;that is, Pr(p"1 D c) is too high. Similarly, if the adjusted return is low,and the "rm chooses a dividend increase to distribute cash #ows, we predict thatthe market will react more positively than average as it adjusts upward itsexpectation of cash-#ow permanence. This is a case where market participantsassigned too low a probability to the greater permanence; that is, Pr(p"1 D c) istoo low.

Thus, the permanence hypothesis predicts a negative relation between theadjusted return and the stock price reaction to the payout decision. Speci"cally,conditional on a "rm choosing a dividend increase, thereby sending a strongsignal about cash-#ow permanence, the announcement return is predicted to bea decreasing function of the cash-#ow permanence expected by the market.Similarly, conditional on a "rm choosing a repurchase, thereby sending a weaksignal about cash-#ow permanence, the announcement return is predicted to bea decreasing function of the cash-#ow permanence expected by the market.

This argument implies that the expected negative relation between adjustedreturns and announcement returns is concentrated in the negative adjustedreturns for dividend-increasing "rms and in the positive adjusted returns for therepurchasing "rms. If the market expects a cash-#ow shock to be transitory(a negative adjusted return), and the "rm announces a dividend increase, pricesare predicted to react positively. However, if the market expects a cash-#owshock to be permanent (a positive adjusted return), and the "rm announcesa dividend increase, prices are not predicted to react as substantially because themarket correctly assessed the permanence of the cash-#ow shock prior to thedistribution announcement. Thus, we predict that, for substantial dividendincreasers, the coe$cient on negative adjusted returns in announcement returnregressions is negative and the coe$cient on positive adjusted returns is lessnegative or zero. Because small/routine dividend increases are not expected toconvey as much information as substantial increases, a less pronounced di!er-ence is expected between the coe$cients on their positive and negative adjustedreturns.

Finec=1086=KGM=VVC


Fig. 3. Interaction between stock price reaction to cash-#ow shock, the payout method choice, andthe stock price reaction to the announcement. In each cell, two mean returns are also presented. The"rst is for payouts below the median payout level, and the second is for payouts above the medianpayout level. For example, dividend increasers with positive return residuals and an above-medianpayout had an average announcement return of 1.19%, compared with 0.87% for the comparablegroup of repurchasers. For the dividend column, only substantial dividend increasers are included.

In the repurchasing sample, if the market expects a cash-#ow shock to betransitory (a negative adjusted return), and the "rm announces a repurchase,prices are not predicted to react substantially given that prices already re#ect thetransitory nature of the cash-#ow shock. However, if the market expects thecash-#ow shock to be permanent (a positive adjusted return), and the "rmannounces a repurchase, the announcement return, while still positive, shouldincorporate a negative component re#ecting the new negative informationabout the permanence of the cash-#ow shock. Thus, for repurchasing "rms, wepredict that the coe$cient on positive adjusted returns is negative and thecoe$cient on negative adjusted returns is less negative or zero.

Fig. 3 provides an illustration of our predictions and basic "ndings withrespect to payout announcement returns. For both substantial dividend in-creasers and repurchasers, we report the payout announcement returns for bothhigh and low expectations about future permanence (i.e., for adjusted returns' and (0). As a control for the cash distribution size, we partition theannouncement returns in each of these groups into above the median and belowthe median payout size subgroups using the measures described in Table 2.Consistent with the permanence hypothesis, the announcement return to divi-dend increases is greater than for share repurchases in each two-way compari-son. Further, the announcement response to a dividend increase is greater whenthe market's prior assessment of cash-#ow permanence is low, and, when thepayout is large, the announcement response to repurchases is lower when the

Finec=1086=KGM=VVC


market's prior assessment of cash-#ow permanence is high. However, thesedi!erences, based on such a coarse separation, are not all statistically signi"cant.

A more powerful test of the permanence hypothesis is presented in Table 5,where we report announcement return regressions. In addition to our estimateof the market's cash-#ow permanence forecast [the adjusted return from Eq.(1)], the regressions include indicator variables for repurchasers and substantialdividend increasers. We control for payout size using the measures de"ned inTable 2. We also control for other economic determinants of the stock pricereaction to dividend and repurchase announcements using variables suggestedby previous studies. Speci"cally, we include total assets, market value-to-bookvalue of assets, the leverage ratio, and the ratio of cash to total assets. Becauselarger "rms are expected to have less information asymmetry (e.g., Collins et al.,1987; Brennan and Hughes, 1991), the surprise component of their distributionannouncement should be lower, leading to a smaller stock price reaction. Weuse the market-to-book ratio to control for the "nding that the stock marketreacts favorably to distributions of cash by "rms with low growth opportunities(see Lang and Litzenberger, 1989; Nohel and Tarhan, 1998). Jensen (1986)argues that agency theory predicts a positive reaction to decisions that increasepreviously low leverage. Finally, we include "rms' cash reserves to control forthe agency theoretic prediction that the stock price reaction to distributions by"rms with large cash holdings will be positive, a "nding supported by Lie (2000).In all regressions, the control variables obtain coe$cients of the predicted signand, with the exception of leverage, are signi"cant in most of the speci"cations.

5.2. Multivariate results

Our "ndings are consistent with the hypothesis that part of the informationin the payout announcement is the payout method, repurchase or dividendincrease, and that investors use this information to update their beliefs about thepermanence of past and current cash-#ow shocks. Columns 1 and 2 presentresults for the full and restricted samples of dividend increasers andrepurchasers. The restricted sample contains all dividend increasers, but onlyrepurchasers that also pay dividends (see Section 4.2). In the full sample, thecoe$cient on the substantial dividend dummy is 1.04 and signi"cantly positive,compared with an insigni"cantly negative coe$cient of !0.75 for repurchasers.The univariate results reported in Table 2 con"rm the "nding in extant literaturethat the announcement return to repurchases is, on average, greater than that toa dividend increase. However, the multivariate results in Table 5 show that,controlling for the size of the payout and the market's expectation of thepermanence of the cash-#ow shock, the announcement return associated witha substantial dividend increase is greater than the response to a repurchase.

The coe$cients on the adjusted returns variables indicate that the announce-ment returns to both dividend increases and repurchases depend upon the

Finec=1086=KGM=VVC


Tab

le5

Ann

oun

cem

ent

retu

rnre

gres

sion

sfo

rdiv

iden

din

crea

ses

and

repu

rcha

ses.

The

anno

unc

emen

tper

iod

isda

y!

5to

day

#5

rela

tive

toth

edi

vide

ndde

clar

atio

nda

teorth

ere

purc

hase

pro

gram

announ

cem

entda

te.C

umul

ativ

eab

norm

alre

turn

sar

eco

mput

edba

sed

on

mar

ketm

odel

regr

ession

sfo

rda

ys!

252

to!

20.T

head

just

edre

turn

sar

eth

ere

sidual

sfrom

there

gres

sion

ofs

tock

retu

rnsove

rth

etw

o-y

earsh

ock

period

on

theca

sh#ow

sfrom

thesa

me

period

[see

Eq.

(1)]

.The

adju

sted

retu

rns

are

ourpro

xies

forth

ew

eigh

tth

em

arke

tpl

aces

on

the

per

man

ence

oft

heca

sh-#

owsh

ock.

Thi

sre

turn

isal

soin

tera

cted

with

indi

cato

rva

riab

les

for

each

ofth

ediv

iden

dsu

bsam

ples

.Fin

ally

,bec

ause

the

pre

dic

tion

softh

epe

rman

ence

hypo

thes

isar

est

ronge

stfo

rdi

viden

din

crea

sers

with

nega

tive

adju

sted

retu

rns

and

for

repurc

has

ers

with

pos

itiv

ead

just

edre

turn

s,th

ead

just

edre

turn

variab

les

are

split

into

thei

rpo

sitive

and

neg

ativ

eco

mpo

nen

ts.T

he

size

ofp

ayou

tva

riab

leis

de"

ned

asth

eex

pec

ted

incr

ease

indi

vide

nds

/mar

ket

valu

e(M

V)f

ordi

vide

nd-incr

easing

"rm

san

das

the

per

centa

geofs

hare

sso

ugh

tfo

rre

purc

has

ing"rm

s.In

Colu

mn

5,size

ofpay

outis

rede"ned

for

repur

chas

ersto

be

the

actu

alper

centof

shar

esre

purc

has

eddu

ring

the

2ye

ars

follo

win

gth

ean

nounc

emen

t.E

xpec

ted

Incr

ease

indi

viden

ds/

MV

,per

cent

of

shar

esso

ugh

t,to

tal

asse

ts,

mar

ket

-to-b

ook

ofa

sset

s,an

dle

vera

gera

tio

are

de"

ned

inT

able

2.C

ash/t

otal

asse

tsis

thera

tio

ofc

ash

and

short

-ter

min

vest

men

ts(C

ompu

stat

item

1)to

tota

lass

etsat

the

begi

nni

ng

oft

he"sc

alye

arco

nta

inin

gth

edistr

ibution

announc

emen

t.In

Colu

mns1,

2,an

d5,

there

gres

sionsin

clude

adum

my

variab

lese

teq

ual

toone

inth

eca

seof

subs

tantial

div

iden

din

crea

ses

and

zero

oth

erw

ise,

and

asim

ilar

dum

my

variab

lefo

rre

purc

has

es.S

mal

l/ro

utin

edi

vide

ndin

crea

ses

are

subsu

med

inth

ein

terc

ept.

The

spec

i"ca

tion

inC

olu

mn

2use

son

lyre

purc

has

ing"rm

sfrom

the

rest

rict

edsa

mple

of"rm

sth

atal

soha

da

div

iden

dpro

gram

inpla

ce.T

het-st

atistics

are

inpar

enth

eses

.

Reg

ress

or

Pre

dict

edFull

Res

tric

ted

Div

iden

dR

epurc

has

eU

sing

sign

sam

ple

sam

ple

only

only

actu

al(1

)(2

)(3

)(4

)re

pur

chas

eam

ount

s(5

)

Inte

rcep

t2.

28!

3.12

!1.

90!

3.20

#2.

93!

(3.2

4)(4

.19)

(3.9

0)(1

.66)

(4.2

2)

Subst

antial

div

iden

ddu

mm

yin

tera

cted

with

positive

adju

sted

retu

rns

!/0

!0.

56!

0.56

!0.

17!

0.55

(!1.

55)

(!1.

62)

(!0.

55)

(!1.

52)

with

neg

ativ

ead

just

edre

turn

s!

!1.

68"

!1.

41"

!2.

63!

!1.

71"

(!2.

38)

(!2.

08)

(!4.

45)

(!2.

42)

Smal

l/ro

utin

ediv

iden

ddum

my

inte

ract

edw

ith

positive

adju

sted

retu

rns

!/0

0.63

0.53

!0.

140.

65(1

.27)

(1.1

2)(!

0.33

)(1

.31)

with

neg

ativ

ead

just

edre

turn

s!

/0!

1.44

#1.

54"

!0.

07!

1.42

#

(!1.

83)

(!2.

06)

(!0.

12)

(!1.

82)

Finec=1086=KGM=VVC


Rep

urc

has

edum

my

inte

ract

edw

ith

positive

adju

sted

retu

rns

!!

1.36

#!

4.24

!!

2.12

"!

1.42

#

(!1.

67)

(!3.

37)

(!2.

01)

(!1.

75)

with

neg

ativ

ead

just

edre

turn

s!

/00.

663.

10#

!0.

440.

65(0

.51)

(1.9

5)(!

0.27

)(0

.50)

Rep

urc

has

edum

my

!!

0.75

!1.

66!

!1.

29"

(!1.

37)

(!2.

77)

(!2.

41)

Subst

antial

div

iden

ddu

mm

y#

1.04

!1.

08!

1.07

!

(3.4

2)(3

.75)

(3.5

0)C

ontr

ols

Size

ofpa

yout

#0.

10!

0.08

4"0.

32!

0.04

0.03

(3.1

7)(2

.55)

(4.5

0)(0

.79)

(0.9

9)

Cas

h-#

ow

shoc

k!

5.51

!4.

53"

4.26

"8.

695.

48!

(3.0

2)(2

.49)

(2.2

3)(1

.58)

(3.0

0)

Tota

las

sets

!!

0.17

!!

0.17

!!

0.19

!!

0.06

!0.

18!

(!3.

00)

(!3.

01)

(!3.

36)

(!0.

26)

(!3.

08)

Mar

ket

-to-

boo

kof

asse

ts!

!0.

64!

!0.

53!

!0.

45!

!1.

77!

!0.

67!

(!5.

09)

(!4.

26)

(!3.

53)

(!3.

78)

(!5.

28)

Lev

erag

era

tio

#0.

170.

160.

170.

160.

18(1

.40)

(1.3

0)(1

.36)

(0.3

4)(1

.44)

Cas

h/tota

las

sets

#2.

46!

0.92

0.85

10.6

9!2.

50!

(2.8

0)(1

.05)

(0.9

4)(3

.57)

(2.8

3)

Adj

ust

edR

20.

037

0.03

20.

031

0.05

90.

034

Num

ber

ofob

serv

atio

ns3,

612

3,42

93,

147

473

3,61

2

!Sig

ni"

cantly

di!

eren

tfrom

zero

atth

e1%

leve

l."Si

gni"

cantly

di!er

ent

from

zero

atth

e5%

leve

l.#S

igni"

cantly

di!

eren

tfrom

zero

atth

e10

%le

vel.

Finec=1086=KGM=VVC


market's prior assessment of cash-#ow shock permanence. In the full sample andrestricted sample regressions, the coe$cients on negative adjusted returns aresigni"cantly negative for substantial dividend increasers, indicating that thestock price reaction to a substantial dividend increase is larger if the market hadassigned low permanence to the cash-#ow shock. Similarly, the negative coe$c-ient on the positive adjusted returns for repurchasing "rms indicates that thestock price reaction to a repurchase is lower if the market assigned substantialpermanence to the cash-#ow shock. The generally smaller and insigni"cantcoe$cients on both positive adjusted returns for substantial dividend in-creasers and on negative adjusted returns for repurchasers support the predic-tion that the announcement return is less pronounced when the market moreaccurately assesses the permanence of cash-#ow shocks. The "ndings forsmall/routine dividend increases are similar to those for the substantial dividendincreasers.

In Columns 3 and 4 of Table 5, we present announcement return speci"ca-tions separately for dividend increasers and repurchasers. These speci"cationsallow the coe$cients on the control variables to vary across payout types. Theinferences in these regressions are very similar to those for the full sample andrestricted sample. Speci"cally, we "nd signi"cantly negative coe$cients only onnegative adjusted returns for dividend increasers and on positive adjustedreturns for repurchasers.

The announcement return regressions in Columns 1 to 4 of Table 5 arepotentially in#uenced by a systematic mismeasurement of the payout size forrepurchasers. Stephens and Weisbach (1998) show that "rms repurchase, onaverage, only about 80% of the announced amount. If the market incorporatesthis fact into its reaction to the repurchase announcement, then ceteris paribus,the return response to a repurchase will be lower than to a dividend increase,after controlling for the announced size of the payout. To address this issue, wereestimate the announcement return speci"cations after rede"ning the size of therepurchase to be: (1) a #at 80% of the announced size; and (2) the actualpercentage of shares repurchased in the two years following the announcement.To measure the actual shares repurchased, we follow the "rst method outlined inStephens and Weisbach (1998), which uses changes in CRSP shares outstanding.Our results are una!ected by these adjustments to the announced size of therepurchase. Column 5 of Table 5 reports the results of the estimation using theactual shares repurchased. For parsimony, we do not tabulate the results usingthe #at 80% estimate, though the inferences are similar.

Overall, our evidence is consistent with the permanence hypothesis'prediction that the market updates its assessment of cash-#ow permanencebased on the announcement of the payout method. Because dividend in-creases are associated with more permanent cash #ows than repurchases,dividends are a more favorable signal about future cash #ows than repurchases.Table 5 shows that, controlling for the market's ex-ante assessment of

Finec=1086=KGM=VVC


cash-#ow permanence, the size of the payout, and "rm characteristics, thestock price reaction to a repurchase is smaller than to a dividend increase.Further, the reaction to a dividend increase is stronger, the lower the market'sexpectation of permanence prior to the announcement, and the reaction toa repurchase is lower, the higher the market's expectation of permanence priorto the announcement.

5.3. Robustness checks

We have noted throughout that our results are robust to the choices we havemade in each test. We discuss further robustness checks here.

5.3.1. Characteristic diwerences between the payout samplesTo this point, our inferences implicitly assume that the partitioning of "rms

based on payout choice is the primary driver of the observed di!erences in thereversion and permanence of the sample "rms' cash #ows. However, the sum-mary statistics in Table 2 indicate that the repurchasing and dividend-increasingsamples di!er somewhat with respect to size, leverage, and the market-to-bookratio. While we have mitigated these di!erences when forming control samplesin Table 4 and in the announcement return regressions in Table 5, thesedi!erences are not incorporated into our estimates of the market's assessment ofcash-#ow permanence prior to the payout announcement (the adjusted returns).Further, several studies document variation in the return-earnings relation asa function of "rm size, growth, and risk (e.g., Collins and Kothari, 1989;Freeman, 1987).

To address this issue, we allow the coe$cient on cash #ows in the regressionsused to produce the adjusted returns [Eq. (1)] to vary with size, leverage, and themarket-to-book ratio. Although the coe$cients on each of these additionalinteractive variables are signi"cant, when the residuals from this augmentedmodel are used in the announcement return regressions, the results are un-changed.

5.3.2. Subperiod analysisGiven the clear secular increase in repurchase use over our sample period,

we conduct separate analyses for the subperiods 1981}87 and 1988}93. Whilethe results for both subperiods are similar to those documented in the fullsample, they are somewhat stronger in the 1988}93 subperiod. These "ndingsare consistent with the payout method choice conveying stronger informationonce both methods are "rmly established as accepted alternatives. We alsoexamine the peak merger years (1987}89) separately. Because repurchases havebeen argued by Denis (1990) and Bagwell (1992) to be e!ective takeoverdefenses, the information conveyed by their announcement may have been

Finec=1086=KGM=VVC


di!erent during these years. However, this possibility is mitigated somewhat byour concentration on open-market repurchases, which are expected to be a lesse!ective takeover defense than repurchase tender o!ers. Our "ndings are robustto the merger-wave analysis.

5.3.3. Other distribution methodsWhile open-market repurchases and regular quarterly dividend increases are

by far the most common methods used to distribute cash to shareholders,managers do have other options available. In particular, we separately examinerepurchase tender o!ers and specially designated dividends. Repurchase tendero!ers are similar to open-market repurchase programs in that they involverepurchasing shares and consist of a one-time commitment to distribute cash.However, repurchase tender o!ers are executed in the same way as inter"rmtender o!ers. They remain open for a speci"ed period of time (usually 20 tradingdays), o!er a premium for the "rm's stock, and handle oversubscription ona pro-rata basis. Further, they usually involve a much larger proportion of theshares and, at least for "xed-price o!ers, tend to be used by smaller, more closelyheld companies (see Comment and Jarrell, 1991).

In contrast to the implicit recurring commitment found in a regular quarterlydividend, specially designated dividends are similar to repurchase tender o!ersin that they involve a one-time commitment of cash. Brickley (1983) examinesspecially designated dividends and compares them with regular increases individends. His conclusions from the comparison of one-time and recurringdividends are consistent with our comparison of repurchases and dividendincreases. Speci"cally, while he "nds that both types of dividend changes aredeclared following good performance, earnings performance following regulardividend increases is signi"cantly stronger than that following specially desig-nated dividends.

We examine 102 repurchase tender o!ers and 90 specially designated divi-dends that match our sample period and meet our data requirements. We "ndthat both types of payout exhibit much larger cash-#ow shocks than our mainsample of "rms. However, the reversion of those shocks is very negative and thepermanence is insigni"cantly di!erent from zero. Thus, these "rms appear to becharacterized by large, dissipating shocks that have no discernible long-terme!ect on the magnitude of the underlying cash #ows. The use of a one-timedistribution instead of a recurring one is sensible for these "rms given that theirpermanence measure is roughly zero, on average. In comparison with our mainsamples, the permanence measure of zero puts them in between the !0.009 rawpermanence measure for open-market repurchasers and the #0.002 raw per-manence measure for substantial dividend increasers. We consider this to beconsistent not only with the permanence hypothesis, but also with the extantempirical evidence indicating that repurchase tender o!ers are stronger signalsthan open-market repurchases (see Comment and Jarrell, 1991).

Finec=1086=KGM=VVC


6. Discussion and relation to prior literature

It is useful to consider how our results complement previous research, becausemuch of the extant work on the stock price reaction and information content ofdividend announcements was performed before open-market repurchasesbecame an accepted viable alternative payout method. Further, much of thework on repurchases explores their characteristics and meaning without directlycomparing them with alternative payout methods such as dividends.

Previous studies of dividend announcements "nd that dividend in-creases produce upward revisions in analyst forecasts and yet do not precedeeither higher earnings or cash-#ows. We describe a setting where thesetwo empirical results can be observed simultaneously. We suggest that, prior tothe payout announcement, analysts incorporate an expected degree ofreversion and permanence into their forecasts of future cash-#ows. The signi"-cantly greater returns during the cash-#ow shock period for dividend in-creasing "rms relative to control "rms in Table 4 suggest that analysts'expectations of permanence are likely to be somewhat greater than that ob-served for the control "rms. With the announcement of the dividend increase,analysts are expected to revise their forecasts upward, to account for the newinformation conveyed by the announcement about lower reversion and greaterpermanence. In this setting, future cash #ows are not expected to be higher thancurrent cash #ows, but they are expected to be higher than the forecast prior tothe announcement.

The extant evidence on analyst revisions following open-market repurchasesis not as strong as the evidence for dividend increases and is consistent with our"nding that the cash #ows of repurchasing "rms behave similarly to the cash#ows of control "rms. Finally, our brief examination of repurchase tender o!ersin Section 5.3 provides a consistent interpretation of the evidence in both Dannet al. (1991) and Lie and McConnell (1998). Dann et al. "nd evidence of positiveanalyst revisions following repurchase tender o!ers, while Lie and McConnell"nd that slower mean reversion in return on assets is the only evidence ofperformance improvement following repurchase tender o!ers. Consistent withLie and McConnell, we "nd that, compared with open-market repurchases, thecash-#ow reversion for repurchase tender o!ers is less pronounced. Thus, ourexplanation for simultaneously observing positive analyst revisions and lower,but more slowly reverting, future performance following dividend increases alsolends itself to our evidence on repurchase tender o!ers.

By tying payout choice to the permanence of cash-#ow shocks, our study isrelated to, and consistent with, a growing stream of recent studies that suggesta relation between dividends and cash-#ow variance as opposed to cash-#owlevels. Decreases in earnings risk have been found in studies of dividendinitiations such as Dyl and Weigand (1998), and increases in earnings risk havebeen found for omissions by Sant and Cowan (1994). Further, Bradley et al.

Finec=1086=KGM=VVC


(1998) present evidence consistent with the hypothesis that managers pay lowerdividends when future cash #ows are uncertain. Finally, Jagannathan et al.(2000) show that repurchasing "rms have higher three-year cash-#ow volatilityprior to the distribution year. Their work provides further support for thepermanence hypothesis given that, based on a di!erent empirical approach, theyconclude, as we do, that cash-#ow variation and cash-#ow stability are determi-nants of payout policy.

In a recent literature review, Allen and Michaely (1995) emphasize that onemajor criticism of many signaling models of dividend policy, such as Bhat-tacharya (1979) and Miller and Rock (1985), is that repurchases would be a moree$cient signaling mechanism. This criticism is valid if the two methods areassumed to signal the same information, such as a high level of future earnings.In this study, we view the choice of dividend versus repurchase as primarily anoperational one made to match the characteristics of the expected future cash#ows. This relation between the permanence of the cash #ows and the method ofpayout is the basis for part of the information content of payout announce-ments. We suggest that repurchases do not dominate dividends as signalsbecause the two methods send fundamentally di!erent signals. That is, becauserepurchases are not a more e$cient method for signaling reversion or perma-nence of cash #ows, both payout methods are observed.

7. Concluding discussion

We hypothesize that the method used to distribute cash #ows re#ects thenature of the underlying cash-#ow process and shapes investors' expectationsabout the permanence of cash-#ow shocks. The hypothesis has two parts. First,"rms use repurchases to distribute cash-#ow shocks that are primarily transient,and they use dividends for cash-#ow shocks containing a larger permanentcomponent. Second, the market recognizes this association and uses theannouncement of a particular distribution method to update its belief about thepermanence of past and contemporary cash-#ow shocks.

We "nd that cash-#ow shocks followed by substantial dividend increases havea larger permanent component than those followed either by repurchases orsmall/routine dividend increases or by no payout at all. This means that the cash#ows of substantial dividend-increasing "rms are less likely to revert back tolevels prior to the cash-#ow shock and, hence, are less transitory. Thus, thepermanence of contemporaneous cash-#ow shocks is related to the type ofpayout method chosen. Our announcement return analysis demonstrates thatan important component to the information released by a distributionannouncement is the method used to distribute the cash #ows. We "nd that whenthe payout method does not match the market's expectations, the marketupdates its previous assessment of the permanence of the cash-#ow shock. These

Finec=1086=KGM=VVC


"ndings indicate that the distribution method used by managers signalsinformation about the permanence of the cash-#ow shock.

While many factors contribute to a manager's decision regarding the appro-priate method of payout to use for a given cash-#ow shock, we argue that theshocks' permanence plays an important role in this choice. By focusing on thepermanence of changes in cash #ows instead of on their levels, our explanationcan potentially reconcile evidence on favorable analyst reaction and littlechange in cash-#ow levels following a payout announcement. Further, bothpayout methods are observed because repurchases provide valuable #exibilityallowing managers to distribute transient cash #ows without committing to anincrease in a recurring dividend.

References

Aharony, J., Swary, I., 1980. Quarterly dividend and earnings announcements and stockholders'returns: an empirical analysis. Journal of Finance 35, 1}12.

Allen, F., Michaely, R., 1995. Dividend policy. In: Jarrow, R., Maksimovic, V., Ziemba, W. (Eds.),Handbooks in OR & MS, Vol. 9. North-Holland, Amsterdam, pp. 793}837.

Asquith, P., Mullins, D., 1983. The impact of initiating dividend payments on shareholders' wealth.Journal of Business 56, 77}96.

Bagwell, L., Shoven, J., 1988. Cash distributions to shareholders. Journal of Economic Perspectives3, 129}140.

Bagwell, L., 1992. Dutch auction repurchases: an analysis of shareholder heterogeneity. Journal ofFinance 47, 71}105.

Barclay, M., Smith, C., 1988. Corporate payout policy: cash dividends versus open-market repur-chases. Journal of Financial Economics 22, 61}82.

Bartov, E., 1991. Open-market stock repurchases as signals for earnings and risk changes. Journal ofAccounting and Economics 14, 275}294.

Bartov, E., Krinsky, I., Lee, J., 1998. Evidence on how companies choose between dividends andopen-market stock repurchases. Journal of Applied Corporate Finance 11, 89}96.

Benartzi, S., Michaely, R., Thaler, R., 1997. Do changes in dividends signal the future or the past.Journal of Finance 52, 1007}1034.

Bhattacharya, S., 1979. Imperfect information, dividend policy, and &the bird in the hand' fallacy. BellJournal of Economics 10, 259}270.

Bradley, M., Capozza, D., Seguin, P., 1998. Dividend policy and cash-#ow uncertainty. Real EstateEconomics 26, 555}580.

Brennan, M., Hughes, P., 1991. Stock prices and the supply of information. Journal of Finance 46,1665}1691.

Brickley, J., 1983. Shareholder wealth, information signaling, and the specially designated dividend:an empirical study. Journal of Financial Economics 12 (2), 187}209.

Carroll, T., 1995. The information content of quarterly dividend changes: professional adaptation.Journal of Accounting, Auditing and Finance 10, 293}319.

Choi, D., Chen, S., 1997. The di!erential information conveyed by share repurchase tender o!ers anddividend increases. Journal of Financial Research 20, 529}543.

Collins, D., Kothari, S.P., 1989. An analysis of intertemporal and cross-sectional determinants ofearnings response coe$cients. Journal of Accounting and Economics 11, 143}181.

Finec=1086=KGM=VVC


Collins, D., Kothari, S.P., Rayburn, J., 1987. Firm size and the information content of prices withrespect to earnings. Journal of Accounting and Economics 9, 111}138.

Comment, R., Jarrell, G., 1991. The relative signalling power of Dutch-auction and "xed-priceself-tender o!ers and open-market share repurchases. Journal of Finance 46, 1243}1271.

Dann, L., 1981. Common stock repurchases: an analysis of returns to bondholders and stockholders.Journal of Financial Economics 9, 113}138.

Dann, L., DeAngelo, H., 1988. Corporate "nancial policy and corporate control: a study of defensiveadjustments in asset and ownership structure. Journal of Financial Economics 20, 87}127.

Dann, L., Masulis, R., Mayers, D., 1991. Repurchase tender o!ers and earnings information. Journalof Accounting and Economics 14, 217}252.

DeAngelo, H., DeAngelo, L., Skinner, D., 1996. Reversal of fortune: dividend signaling and thedisappearance of sustained earnings growth. Journal of Financial Economics 40, 341}371.

Dechow, P.M., 1994. Accounting earnings and cash #ows as measures of "rm performance: the roleof accounting accruals. Journal of Accounting and Economics 18, 3}42.

Dechow, P.M., Kothari, S.P., Watts, R.L., 1998. The relation between earnings and cash #ows.Journal of Accounting and Economics 25, 133}168.

Denis, D., 1990. Defensive changes in corporate payout policy: share repurchases and specialdividends. Journal of Finance 45, 1433}1456.

Denis, D., Denis, D., Sarin, A., 1994. The information content of dividend changes: cash #ow,signaling, overinvestment and dividend clienteles. Journal of Financial and Quantitative Analy-sis 29, 567}587.

Dyl, E., Weigand, R., 1998. The information content of dividend initiations: additional evidence.Financial Management 27, 27}35.

Fama, E., Babiak, H., 1968. Dividend policy: an empirical analysis. The Journal of the AmericanStatistical Association 63, 1132}1161.

Fenn, G., Liang, N., 2000. Corporate payout policy and managerial stock incentives. Journal ofFinancial Economics, forthcoming.

Freeman, R., 1987. The association between accounting earnings and security returns for large andsmall "rms. Journal of Accounting and Economics 9, 195}228.

Jagannathan, M., Stephens, C., Weisbach, M., 2000. Financial #exibility and the choice betweendividends and stock repurchases. Journal of Financial Economics, this issue.

Jensen, M., 1986. Agency costs of free cash #ow, corporate "nance, and takeovers. AmericanEconomic Review 76, 323}329.

Jensen, M., 1993. The modern industrial revolution, exit, and the failure of internal control systems.Journal of Finance 48, 831}880.

Lang, L., Litzenberger, R., 1989. Dividend announcements: cash #ow signaling vs. free cash #owhypothesis. Journal of Financial Economics 24, 181}191.

Leftwich, R., Zmijewski, M., 1994. Contemporaneous announcements of dividends and earnings.Journal of Accounting, Auditing and Finance 9, 725}762.

Lie, E., 2000. Excess funds and agency problems: an empirical study of incremental cash disburse-ments. Review of Financial Studies 13, 219}248.

Lie, E., McConnell, J., 1998. Earnings signals in "xed-price and Dutch auction self-tender o!ers.Journal of Financial Economics 49, 161}186.

Lintner, J., 1956. Distribution of incomes of corporations among dividends, retained earnings, andtaxes. American Economic Review 46, 97}113.

Miller, M., Rock, K., 1985. Dividend policy under asymmetric information. Journal of Finance 40,1031}1051.

Nohel, T., Tarhan, V., 1998. Share repurchases and "rm performance: new evidence on the agencycosts of free cash #ow. Journal of Financial Economics 49, 187}221.

Ofer, A., Siegel, D., 1987. Corporate "nancial policy, information, and market expectations: anempirical investigation of dividends. Journal of Finance 42, 889}911.

Finec=1086=KGM=VVC


Ofer, A., Thakor, A., 1987. A theory of stock price responses to alternative corporate cashdisbursement methods: stock repurchases and dividends. Journal of Finance 42, 365}394.

Pettit, R.R., 1972. Dividend announcements, security performance, and capital market e$ciency.Journal of Finance 27, 993}1007.

Sant, R., Cowan, A., 1994. Do dividends signal earnings? The case of omitted dividends. Journal ofBanking & Finance 18, 1113}1133.

Stephens, C., Weisbach, M., 1998. Actual share reacquisitions in open-market repurchase programs.Journal of Finance 53, 313}333.

Talmor, E., Titman, S., 1990. Taxes and dividend policy. Financial Management 19, 32}35.Watts, R., 1973. The information content of dividends. Journal of Business 46, 191}211.

Finec=1086=KGM=VVC


the cash- ow permanence and information content of...

Documents