An empirical examination of divestiture-related gains to buyers and sellers

MISSING IMAGE

Material Information

Title:
An empirical examination of divestiture-related gains to buyers and sellers
Physical Description:
vii, 101 leaves : ill. ; 28 cm.
Language:
English
Creator:
Sicherman, Neil Wayne, 1955-
Publication Date:

Subjects

Subjects / Keywords:
Corporate divestiture   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1986.
Bibliography:
Includes bibliographical references (leaves 98-100).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Neil Wayne Sicherman.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000893362
notis - AEK1883
oclc - 15293448
System ID:
AA00003394:00001

Full Text










AN EMPIRICAL EXAMINATION OF DIVESTITURE-RELATED
GAINS TO BUYERS AND SELLERS










By

Neil Wayne Sicherman


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1986














To my wife, to my mother, and in memory of my father.













ACKNOWLEDGMENTS


Special thanks are due Professor Richard H. Pettway who

chaired my supervisory committee. His guidance and support

ensured the successful completion of this dissertation. I

also thank Professor Roy L. Crum and Professor Richard

Elnicki for their assistance. Thanks are not sufficient for

the loving support of my wife, Terry, who typed this

dissertation and patiently endured the last four years.


iii















TABLE OF CONTENTS


Page


ACKNOWLEDGMENTS . . . . .

ABSTRACT . . . . .

CHAPTER


iii

vi


ONE INTRODUCTION . . .

Notes . . . .

TWO LITERATURE REVIEW . . .

Divestitures . . .
Mergers and Acquisitions . .
Summary . . . .
Notes . . . .

THREE RESEARCH DESIGN . . .

Data . . . .
Hypotheses . . . .
Methodology . . .
Tests of the Hypotheses . .
Notes . . .. .

FOUR EMPIRICAL RESULTS:
THE ACQUISITION OF DIVESTED ASSETS

Full Sample Results . .
First-Tier Stratification Results
Second-Tier Stratification Results .
Summary . . . .
Notes . . . .

FIVE EMPIRICAL RESULTS:
THE DIVESTITURE OF ASSETS . .

Full Sample Results . .
First-Tier Stratification Results
Second-Tier Stratification Results .
Summary . . . .


. 1

. 7

. 8

. 8
. 14
. 16
. 17


. .







SIX SUMMARY, CONCLUSIONS,
AND FUTURE RESEARCH . . 90

Summary and Conclusions . 90
Implications for Future Research .. 95

REFERENCES . . . . 98

BIOGRAPHICAL SKETCH. . . . 101













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



AN EMPIRICAL EXAMINATION OF DIVESTITURE-RELATED
GAINS TO BUYERS AND SELLERS

By

Neil Wayne Sicherman

August 1986


Chairman: Dr. Richard H. Pettway
Major Department: Finance, Insurance, and Real Estate



An increasing number of firms are divesting assets and

the economic consequences are significant. Sample

stratification to determine the divestiture-related wealth

effects to given classes of buyers and sellers is the

primary objective of this dissertation.

Both the divestiture and the purchase of divested

assets are analyzed. It is hypothesized that the divested

unit's relatedness with the seller and the buyer affect the

economic outcome. Further, it is hypothesized that the

seller's financial condition will impact divestiture-induced

shareholder wealth changes.

The results indicate that not all buyers of divested

assets gain. Positive significant abnormal returns are







earned by firms that acquire divested assets which are

related to the buying firm. Further, firms purchasing

related assets earn significantly more than firms purchasing

unrelated assets.

Many firms purchase unrelated assets despite possible

negative outcomes. Insider equity ownership is

significantly less for these firms than for firms purchasing

related assets. Managers may be diversifying employment

risk when they are not compensated by share performance.

Thus, a significant agency problem exists.

Divesting firms typically earn abnormal returns around

the announcement of a divestiture. The selling of unrelated

assets yields significantly higher cumulative abnormal

returns than the divestiture of related assets. Selling

firms in strong financial condition earn significantly

greater abnormal returns than sellers in weak financial

condition. Firms in weak financial condition gain at

announcement but accumulate no excess returns over the

period tested.

The findings of this dissertation provide evidence that

mergers and divestitures can be profitable to both the buyer

and the seller of divested assets. Wealth effects from

corporate combination should not be analyzed solely in terms

of buying-firm performance and selling-firm performance.

Rather, analyses should focus on the characteristics of the

buyer, seller, and divested unit of each specific

divestiture/acquisition.


vii














CHAPTER ONE
INTRODUCTION


Divestitures comprise a large subset of the population

of corporate mergers. Toy (1985) reports a 40 percent

increase in corporate divestitures over the previous four

years. Nearly 2,000 divisions and subsidiaries were

divested from parent firms in just 1984 and 1985. Beatrice

Companies has divested more than 50 units since 1983 and ITT

Corporation has sold off $1.7 billion in assets since

January 1985 (Guyon, 1986). Thirty percent of merger and

acquisition activity at Merrill Lynch is divestiture-related

(Toy, 1985). The decision to divest corporate assets is an

increasing phenomenon and has significant economic

consequences.

While divestiture activity is a significant proportion

of all acquisition activity, it must not be assumed that

the two actions yield identical responses from the buying

firm's shareholders. The acquisition market, in general, is

perceived as the active pursuance of targeted assets by

acquiring firms. This suggests some competition for ripe

targets. However, divestitures are usually initiated by the

selling firm. Usually only one buyer comes forth to bid for







the firm.1 These conditions suggest that the market for

divested assets is somewhat less than perfectly competitive.

In a real asset market that is not perfect, buyers and

sellers may disagree about the value that is assigned a

divested asset. Rosenfeld (1984, p. 1446) explains that

"both the acquiring firm and the acquired assets have unique

resources that earn positive economic rents when combined

with the assets of the other firm." The ingredients for the

successful acquisition of a divested asset depend upon the

characteristics of the parties to the transaction. These

characteristics also determine whether the divestiture

decision is beneficial to the seller.

An understanding of the economic consequences of mer-

gers and acquisitions cannot be obtained by studying mergers

and acquisitions as a homogeneous class. The underlying

implication that shareholder wealth effects are similar for

all corporate acquisitions is not accurate. Shareholder

response to an announced acquisition is a function of the

specific characteristics of the buying firm and the selling

firm (and the asset sold in the case of divestiture).

In any analysis of divestitures the characteristics

that impact the wealth of the firm's equityholders must be

discerned. The success of the acquisition, or divestiture,

is a function of these characteristics. Both the financial

press and financial theory offer insights about the nature

of the factors that affect shareholder wealth upon divesti-

ture and upon the purchase of a divested asset.








Recent articles in Business Week and the Wall Street

Journal highlight the key factor in predicting the success

of an acquisition: The acquired firm should be related to

the acquiring firm (Proketsch and Powell, 1985 and Toy,

1985). Conclusions from management consultants, as

reported, clearly indicate that related acquisitions are

most likely to succeed and that unrelated mergers are most

likely to fail. A common managerial error is the

"assumption by executives that the skills honed in one

business can be readily applied to another" (Business Week,

June 3, 1985, p. 89).

Several large firms have erred in acquiring unrelated

assets. Many conglomerates are now feeding a divestiture

wave by selling unrelated businesses and restructuring to

concentrate on managerial strengths.2 As one Business Week

report concludes:


The lessons are clear. Mergers can work under the
right circumstances. What is so astonishing is that so
few companies seem to have studied the historical evi-
dence before plunging ahead. (Business Week, June 3,
1985, p. 91)


It is not surprising that the popular financial media

are reporting the failures of unrelated acquisitions and

their subsequent divestitures. Financial theory and empiri-

cal evidence indicate that concentration of business assets,

not diversification, yields greater success. Business

diversification finds little support in financial theory.








Myers (1968) and others argue that the value-additivity

principle holds, that is, pure conglomeration does not

enhance shareholder wealth. This is due to the share-

holder's ability to diversify via security markets. Rumelt

(1974) finds that pure conglomerates do not perform as well

as non-conglomerates. He concludes that real asset port-

folios that are comprised of related businesses achieve the

best performance. Scanlon (1985) finds a statistically

significant correlation between merger-generated excess

returns and relatedness. Using SIC codes as a relatedness

measure, he finds greater shareholder wealth enhancement for

those firms that acquire related assets than for those firms

that purchase unrelated assets.

Copeland and Weston (1983) suggest that managerial

efficiency differences precipitate many mergers. Corporate

combinations are more likely to succeed if they are between

related firms, and thus can be managed more efficiently:


Hence another formulation of the differential effi-
ciency theory of mergers is that there are always many
firms with below average efficiency or that are not
operating up to their potentials, however defined. It
is further suggested that firms operating in similar
kinds of business activity would be most likely to be
the potential acquirers. They would have the back-
ground for detecting below-average or less-than-full-
potential performance and have the managerial know-how
for improving the performance of the acquired firm.
(Copeland and Weston, 1983, p. 562)


Thus, firms should expect to succeed in the acquisitions

market if acquiring related businesses. Firms that have not




5


done so will enhance value by divesting their unrelated

assets.

It is not suggested that any related acquisition will

result in positive net present values. However, acquiring

firm shareholders are more likely to gain from related

acquisitions than unrelated acquisitions. Managerial pref-

erences are not as obvious.

Managers may prefer to acquire unrelated units. A

substantial portion of the manager's wealth is represented

by human capital in the form of his or her employment

contract. Amihud and Lev (1981) have suggested that

conglomerate mergers are a result of management's desire to

reduce personal risk-bearing. Managers can diversify away

employment risk by acquiring unrelated businesses. Thus, a

possible conflict exists between shareholders and managers:

Managers may prefer unrelated acquisitions but shareholders

prefer related acquisitions.

Decisions to divest may be motivated by a change in the

parent firm's financial condition. As Jain (1985) suggests,

an unanticipated decrease in expected cash flows may imply a

higher probability of default. A firm in this position will

find difficulty in raising external capital. Therefore, the

company may sell off assets to raise needed cash. It is

expected that firms would sell unrelated assets initially

and thus relatedness is still the primary factor in the


decision.







An increase in a firm's probability of default may

induce the parent to sell assets at a lower price than might

be expected if the firm were in a stronger financial posi-

tion. The parent may find itself in a weaker negotiating

position and thus the acquiring firm may obtain a better

deal for its shareholders. Again, the acquiring firm's

primary consideration should be the relatedness of the

divested asset. However, additional gains may be achieved

by purchasing assets from sellers in a weaker negotiating

position.

In sum, the objective of undertaking this study is to

empirically examine the divestiture-related impact upon

shareholder wealth of both the acquiring firm and the

divesting firm. There are several issues that suggest that

all divestitures and all purchases of divested units do not

elicit identical shareholder response. Any complete

analysis must consider these factors. Specifically, the

issues of relatedness, shareholder-manager conflicts, and

selling-firm negotiating position are considered.

The following chapter presents a review of the academic

literature that is pertinent. Chapter 3 details the data

and presents the testable hypotheses and the methodology

utilized. Empirical results for the buyer and the seller

follow in Chapter 4 and Chapter 5 respectively. Summary,

conclusions, and future research implications follow in

Chapter 6.





7




Notes

1. Jain (1985) reports that typically only one buyer nego-
tiates for the divested asset. This is the case for
almost all of the transactions analyzed in this study.

2. For example, Gulf and Western, ITT, U.S. Industries,
and Beatrice.















CHAPTER TWO
LITERATURE REVIEW


This study builds upon existing literature in the areas

of divestitures and acquisitions. This chapter is dichoto-

mized to allow a review of the appropriate literature in

each area. First, a discussion of divestiture analyses is

presented, including the examination of divested asset

acquisitions. Second, a review of the voluminous mergers

and acquisitions literature is presented.


Divestitures

There have not been many academic studies of divesti-

tures and even fewer that analyze the acquisition of di-

vested assets. These studies have typically examined either

spin-offs or sell-offs.1 Hite and Owers (1983), Miles and

Rosenfeld (1983), Schipper and Smith (1983), and Rosenfeld

(1984) have all found positive abnormal returns around the

announcement date (AD) of an intended spin-off. Miles and

Rosenfeld (1983) found evidence that large spin-offs gener-

ate higher excess returns than small spin-offs. A spin-off

with equity market value of at least 10 percent of the

parent's equity market value is considered large. Rosenfeld

(1984) compared spin-offs and sell-offs and found spin-offs








generated greater excess returns than sell-offs, although

both events resulted in positive excess returns.

Alexander, Benson, and Kampmeyer (1984) suggest nega-

tive news usually precedes a sell-off announcement, thus

offsetting subsequent positive reaction. They find positive

abnormal returns for sell-offs around AD preceded by nega-

tive excess returns during the pre-event period. Jain

(1985) finds similar results around AD and for the pre-event

period. Zaima and Hearth (1985) also find positive abnormal

returns around AD for sell-offs. Thus, the general conclu-

sion can be made that the divestiture decision (both spin-

offs and sell-offs) generates positive excess returns for

the selling firm's shareholders. Yet, negative pre-event

abnormal returns suggest the sell-off announcement is pre-

ceded by bad news about the parent.

Prior studies analyzing the source of the parent firm's

abnormal returns lack agreement. Rosenfeld (1984),

Alexander, Benson, and Kampmeyer (1984), and Miles and

Rosenfeld (1983) suggest the spinning off of assets to

shareholders expands the opportunity set of investments

available to investors, thus enhancing wealth. This argu-

ment implies a clientele effect exists and security markets

are incomplete without divestitures. On the other hand, it

seems reasonable to assume that security markets are

complete, and that the seller's risk-return relationship can

be duplicated independently of the diversity of the firm's








asset portfolio. A stronger argument is presented by

Rosenfeld (1984). He cites the reduction in agency

monitoring costs, as discussed in Jensen and Meckling

(1976), associated with spun-off assets. The assets, once

spun off, can be monitored more closely and evaluated more

accurately than when being held as part of a large con-

glomerate.

Hite and Owers (1983), Schipper and Smith (1983),

Rosenfeld (1984), and Miles and Rosenfeld (1983) suggest the

possibility that excess returns to shareholders are the

result of a wealth transfer from bondholders to share-

holders. Their logic follows from Galai and Masulis (1976).

Wealth can be transferred since a spin-off is the expropria-

tion of collateral from bondholders to equityholders.

Schipper and Smith (1983) quite accurately suggest that

these wealth transfers are mitigated via debt covenants that

prevent such wealth redistribution.

Managerial decisions to both spin-off and/or sell-off

assets may be motivated by the desire to concentrate the

parent firm in one line of business. This most-compelling

argument is suggested by several authors. Schipper and

Smith (1983) found evidence to support this hypothesis.

They found high firm growth prior to divestiture, indicating

the need to concentrate, or scale down. They also found

that more than three-fourths of the spin-offs in their

sample were in a different line of business than that of the







parent firm. They did not, however, attempt to directly

test this.

Fama (1980), Jensen and Meckling (1976), Alchian and

Demsetz (1972) and others have described the corporate

setting as a "nexus" of contracts where managers act as

agents for the shareholders of the firm. An uninduced

manager will make decisions that always maximize the share-

holder's utility if and only if the utility functions of the

two parties are identical. If this is not the case, crucial

conflicts arise.

One such conflict occurs in the decision to acquire

another firm. Amihud and Lev (1981) suggest the manager's

objective diverges from the owner's objective in the merger

decision. The manager wishes to diversify personal risk-

bearing through the acquisition of unrelated assets. Jensen

and Smith (1985) call this "differential risk exposure:"


Managers typically have a nontrivial fraction of their
wealth in firm-specific human capital and thus are
concerned about the variability of total firm value,
including that portion of firm risk that can be
eliminated through diversification by the firm's stock-
holders. (p. 103)


Jensen and Smith (1985) also note that managers and share-

holders have different time horizons associated with the

firm. The manager's horizon is limited to his or her

transitory relationship with the firm while shareholders own

a perpetual claim on the corporation's cash flows.







Fama (1980) claims that natural market forces may re-

solve this conflict. That is, managers will settle up, ex

poste, for any decisions that do not maximize the owner's

utility. It is doubtful that the settling-up process is

complete since there are costs to both monitoring and penal-

izing the manager.

Zaima and Hearth (1985) suggest that the financial

condition of the parent firm affects the magnitude of gains

to shareholders. If the firm is in poor condition finan-

cially, it may be in a weak negotiating position. Yet, they

find no relationship between the abnormal returns in a sell-

off transaction and parent financial condition. Of course

any excess returns generated by the sale of poorly-per-

forming assets may be muted by a low selling price resulting

from a weak negotiating position. Thus, their evidence is

inconclusive. Rosenfeld (1984) does find evidence that

higher excess returns are generated when the parent is in

good financial condition. Again, there are two effects.

One, the parent's financial condition affects the outcome

via its negotiating position. Two, if the divested assets

were poor performers vis-a-vis the parent, the sell-off

should generate positive excess returns. Thus, if the

parent is in a good negotiating position (i.e. good

financial condition) and sells a poor performer, high posi-

tive excess returns will be found.

A final motive for sell-offs is offered by Jain (1985).

He claims assets will be sold off when they have a higher








value to outsiders than to the firm's shareholders. Posi-

tive excess returns can be generated for the parent's share-

holders if the cash received is greater than the present

value of expected cash flows from operating the divested

unit. This is simply an abandonment issue from the parent's

perspective. Thus where these valuation differentials

exist, both the divesting firm and the acquiring firm may

share in the excess returns generated by the decision to

sell off.

Jain (1985), Zaima and Hearth (1985), and Rosenfeld

(1984) have studied shareholder wealth effects of firms that

acquire divested units. All three studies analyze the ac-

quiring firms as a homogeneous class and conflicting results

arise.

Jain (1985) studies a sample of 304 acquiring firms.

He finds statistically significant positive abnormal returns

of 0.34 percent (t-statistic of 2.34) on the day prior to

the Wall Street Journal (WSJ) announcement day. However,

cumulative abnormal returns are generally negative and not

statistically different from zero. One exception is for the

5-day interval immediately following the initial announce-

ment when the cumulative abnormal return is 0.40 percent (t-

statistic of 1.34).

Zaima and Hearth (1985) find positive, but not statis-

tically significant, cumulative abnormal returns for all

intervals in their test of 75 firms. Rosenfeld (1984)







studied 30 acquiring firms and found statistically signifi-

cant positive cumulative abnormal returns for the three-day

period surrounding the WSJ announcement and for the entire

61-day test period. In fact, the cumulative excess return

for the interval beginning 30 days prior to announcement and

ending 30 days after announcement was 7.06 percent (t-

statistic of 1.98).

The examination of firms that acquire divestitures is

inconclusive. This result should surprise no one since

these studies treat these acquisitions homogeneously. Simi-

lar inconclusive results are found in the mergers and acqui-

sitions literature.


Mergers and Acquisitions

The research completed in this area has been extensive.

An overview is presented here as it applies to this study.

There are several good sources of thorough literature re-

views in the merger area (see Trifts, 1984, Jensen and

Ruback, 1983, Weston and Chung, 1983, and Copeland and

Weston, 1983).

Most studies have found positive abnormal returns for

target firms when mergers or tender offers are announced.

Eckbo (1983), Asquith (1983), and Dodd (1980) find highly

significant positive abnormal returns for targets around the

announcement day. Jensen and Ruback (1983) average cumula-

tive abnormal returns from several studies. They find







target firms accumulate abnormal returns of 20 percent in

mergers and 30 percent in tender offers.

Gains to acquiring firms are greater for tender offers,

on average, than for mergers. Jensen and Ruback (1983) find

average cumulative abnormal returns of zero and 4.0 percent

for mergers and tender offers, respectively. The acquisi-

tions of divested assets are facilitated through mergers,

which provide no gain to the buyer's shareholders, on

average.

Dodd (1980) finds statistically significant negative

abnormal returns around the two-day announcement period for

mergers.2 However, Asquith, Bruner and Mullins (1983) find

statistically significant positive abnormal returns for the

period beginning twenty days prior to merger announcement

through announcement day. They find similar results when

testing successive mergers in announced merger programs.

Wier (1983) and Eckbo (1983) find evidence that anti-

trust action reduces gains to target firms. They find that

the announcement of antitrust complaints leads to the elimi-

nation of abnormal returns associated with merger announce-

ments. This potential loss may lead divesting firms to seek

unrelated buyers to avoid antitrust action dependent upon

the enforcement objectives of the administration of the day.

Kummer and Hoffmeister (1978) find evidence that

target-firm managerial opposition increases abnormal returns

to the seller when the merger is consummated. However, they

also find that opposition increases the likelihood of the








merger's being canceled. Jensen and Ruback explain these

results:


The higher average return to targets with managerial
opposition to takeovers is also consistent with the
hypothesis that such opposition harms stockholders of
target firms by reducing the frequency of takeover
offers. For example, the higher returns could arise
because only the more highly profitable takeovers are
pursued when bidders believe managerial opposition will
lower the probability of success and raise the expected
costs. (Jensen and Ruback, 1983, p. 618)


Thus, the resulting average abnormal return is biased upward

due to the exclusion of lower-profit mergers.

This may help explain why divesting parents typically

do not seek multiple bidders when selling a unit. The

increased probability of failure and greater costs may dis-

courage potential buyers. In divestitures the seller ini-

tiates the action and does not want to drive away potential

bidders. In addition, a friendly merger will more likely

result in the acquired firm's management's remaining in

place.



Summary

Acquisitions, when treated as a like population,

typically add wealth to the selling firm. There is less

gain, if any, to the acquiring firm. For a subsample of

this population of acquisitions, the purchase of divested

assets, the evidence is inconclusive.







The lack of clarity in the results may be because both

the decision to divest and the acquisition decision have

firm-specific motivation. Divestitures may be motivated by

desires to restructure assets or by financial considera-

tions. Acquisitions may be pursued to benefit either the

shareholder or the manager. One can only conclude that

further stratification of the merger population is required

to obtain insight into shareholder response with respect to

acquisition decisions.


Notes

1. A spin-off is defined as the transfer of a portion of a
firm's assets to its existing shareholders. A sell-off
is the sale of a portion of the firm's assets to
another firm.

2. The two-day announcement period is the day before the
announcement and the Wall Street Journal announcement
day.















CHAPTER THREE
RESEARCH DESIGN


Data

An initial sample of 830 sell-offs was obtained from

Mergers and Acquisitions. These divestitures occurred

during the period beginning January 1, 1983 through

September 30, 1985. The distribution of these transactions

can be seen in Table 3-1. Of these 830 transactions, 636

parent (selling) firms and 464 acquiring firms were listed

on the daily stock return files of the Center for Research

in Security Prices (CRSP) at the time of divestiture. Both

seller and buyer were listed on CRSP for 269 of these trans-

actions.

The date of informational impact is defined as the Wall

Street Journal announcement day. An examination of the Wall

Street Journal Index yielded acquisition announcements for

149 transactions.2 3 Of these, two firms had missing data

on the CRSP tapes and were eliminated: therefore 147 trans-

actions were analyzed (see Table 3-2). After eliminating

firms for multiple divestitures, 130 parent firms remained.

Thus, 147 buying firms and 130 selling firms were studied.

It should be noted that several types of transactions

are not included in this sample. Leveraged buyouts or




19



acquisitions by the management of the divested asset are not


included.


Also, divested financial services are not in-


cluded in the sample.


Table 3-1
Time Distribution of Initial Sample of 830 Sell-Offs


Year


Quarter


1983

46
51
79
81
257


1984

61
79
76
95
311


1985

81
76
105

262


Total

188
206
260
176
830


Table 3-2
Time Distribution of Final Sample of 147 Sell-Offs


Year


Quarter


1983


1984


1985

12
11
19


Total

28
39
42
38
147








Hypotheses

A discussion of the methodology employed and the speci-

fic tests undertaken will follow this section. However, it

is useful to verbalize the hypotheses at this point. The

suggestions offered in this study are tested in three

stages for the buyer and the seller:

1. The full populations of divesting firms and firms

acquiring divested assets are studied.

2. A first-tier stratification based upon the factors of

relatedness and seller financial condition is employed.

3. A second-tier stratification is utilized to more

narrowly define the types of divestitures based upon

the above factors.


Buying Firm Hypotheses

As discussed, it is not clear that the acquisitions

market for divested assets behaves the same as the market

for any real asset. The lack of perfect competition for

divested assets is a function of the seller's being the

catalyst. Therefore shareholder wealth effects from ac-

quiring divested assets may differ from shareholder wealth

changes due to the acquisition of any real asset. The

subsample of acquisitions (divested assets) may lead to

greater shareholder gains than the population of all acqui-

sitions treated with homogeneity.







On the other hand, it is argued in this study that

homogeneous treatment of even this large subsample may lead

to false conclusions. To facilitate more robust analysis a

two-tier stratification process is employed. Theory and

empirical evidence clearly imply that the acquisition of

related assets will increase shareholder wealth more than

the purchase of unrelated assets. Additionally, a change in

the financial condition of the seller may result in a dis-

tress sale, placing the parent in a weaker negotiating

position. Acquiring firms may obtain greater gains if pur-

chasing from a distressed parent.

Finally, a second-tier stratification is undertaken.'

Buyers should enhance shareholder wealth most by acquiring

related assets that are sold off by parent firms in weak

negotiating positions. The acquisition of unrelated assets

from sellers in strong negotiating positions should not

enhance the wealth of the acquiring firm's shareholders.



Divesting Firm Hypotheses

For divestitures, sellers initiate the merger process

which may impact shareholder wealth differently than buyer-

initiated mergers. Further, all divestitures cannot be

studied homogeneously. The shareholder wealth effect of

divesting assets is dependent upon the characteristics of

the divestiture. A two-tier stratification process is again

employed.







The divesting of unrelated assets will enhance

shareholder wealth more than the selling of related assets.

A decline in the parent's financial condition may reduce the

gains obtained by the seller if the seller's negotiating

posture is weakened.

A second-tier stratification of the sample of selling

firms is undertaken. Sellers will increase wealth most if

they are in strong financial condition and are divesting

unrelated assets. Conversely, weak sellers that divest

related assets will not increase shareholder wealth.


Methodology

This analysis examines the informational impact upon

the shareholder wealth of firms divesting assets and of

firms acquiring divested assets. To best accomplish this,

an event-study methodology is used. Several variations upon

the event-study theme have appeared in the finance

literature. Brown and Warner (1985) have examined the use-

fulness of event studies and have compared various event

methodologies. They conclude that in most cases "the OLS

market model and using standard parametric tests are well-

specified under a variety of conditions" (Brown and Warner,

1985, p. 25). Further, Jain (1985) finds similar results

utilizing various methodologies.

Ordinary least squares parameter estimates are derived

from the market model during the estimation period:








Rit = a + PiRmt + eit


, t = -210 to -31


where eit is a random noise term.

Abnormal returns are then calculated as follows:



ut = Rit (ai + iRt)


where


uit = abnormal return for firm i on day t

Rit = actual return for firm i on day t


ai, Pi = OLS-estimated market model parameters

Rmt = actual CRSP equal-weighted market index return on

day t


Figure 3-1 depicts the time frame associated with this


methodology.


t=-210 t

[ Estimation Period


III

=-31 t=-30

] [


AD


AD
t=0

Test Period


where t (t = -210 to 30) is the number of trading days from
the announcement day, and AD (t = 0) is the Wall Street
Journal announcement of the transaction.


Figure 3-1
Event Time Frame


t=+30

I










Once the abnormal return for each firm is determined

for each day during the test period, uit is cross-

sectionally averaged to yield average abnormal returns for

each day:


AR = n1 E t t = -30 to 30
t



where ARt is the average abnormal return on day t and nt is

the sample population on day t.

To determine the cumulative effect, the average ab-

normal returns are accumulated across time to yield cumula-

tive abnormal returns:



CAKR = E ARt
t



where CART is the average abnormal return accumulated to day

T. The precise role of CARs is best described as follows:



If the initial announcement is unanticipated, and there
are no other information effects, this cumulative ab-
normal return includes the effects of all revisions in
expectations and offer prices and therefore is a com-
plete measure of the equity value changes for
successful bidders. (Jensen and Ruback, 1983, p. 597)










Tests of the Hypotheses


Buyers

The full sample hypothesis is that shareholder wealth

effects of firms acquiring divested assets are different

than shareholder wealth effects of homogeneously treated

acquiring firms. Specifically, there is more opportunity

for shareholders to gain in the case of divestiture acquisi-

tion versus acquisitions in general. The null and alterna-

tive hypotheses are as follows:

HO: CAR = CARA

The cumulative abnormal returns for firms acquiring

divested assets (CARD) are not significantly different

from the cumulative abnormal returns of acquiring firms

in general (CARA).

H1: CARD > CARA

The cumulative abnormal returns for firms acquiring

divested assets are greater than the cumulative ab-

normal returns of acquiring firms, in general.


To test these hypotheses, CARs are calculated for the

full sample of 147 firms that acquired divested assets.

These accumulated abnormal returns are tested to determine

significant difference from zero and compared with prior

studies to determine evidence of a difference from generic

acquisitions.








Shareholder wealth effects are more accurately deter-

mined if acquisitions of divested assets are classified by

correlated factors rather than tested homogeneously. A

first-tier stratification process based upon relatedness and

parent financial condition suggests hypotheses that lead to

stronger conclusions than nonstratified testing. The acqui-

sition of related divested assets should enhance shareholder

wealth more than the purchase of unrelated divested assets.

Also, the acquisition of divested assets from sellers in

weak negotiating positions should enhance value more than

the purchase of divested assets from sellers in strong

negotiating positions.

The first-tier relatedness hypothesis is as follows:

H0: CARr = CARu

The cumulative abnormal returns for firms acquiring

related divested assets (CAR r) are not significantly

different from the cumulative abnormal returns for

firms acquiring unrelated divested assets (CAR ).


H1: CARr > CARu

The cumulative abnormal returns for firms acquiring

related divested assets are greater than the cumulative

abnormal returns for firms acquiring unrelated divested

assets.







Cumulative abnormal returns are calculated separately

for related acquisitions and unrelated acquisitions. Re-

latedness is measured by two-digit SIC code. The CARs from

each subsample are tested to determine significant differ-

ence from zero and then compared to determine significant'

difference between substrata (see Figure 3-2).


sells to
P B> B > A

I I II L


If ASIC

If ASIC


H0:

HI:


P -- parent firm (seller)
B -- divested asset
A -- acquiring firm


Figure 3-2
First-Tier Stratification
The Acquisition of
Related Divested Assets versus Unrelated Divested Assets


= BSIC

# BSIC


CAR =
r
CARr >


=> CARr

=> CARu


CAR
u
CAR








The first-tier financial condition hypothesis is

as follows:

H0: CARw = CARs

The cumulative abnormal returns for firms acquiring

divested assets from parent firms that are in weakened

negotiating positions (CARw) are not significantly

different from the cumulative abnormal returns for

firms acquiring divested assets from parent firms that

are in strong negotiating positions (CAR ).


H1: CARw > CARs

The cumulative abnormal returns for firms acquiring

divested assets from parent firms that are in weakened

negotiating positions are greater than the cumulative

abnormal returns for firms that are in strong nego-

tiating positions.


Cumulative abnormal returns are calculated separately

for each class. Weakened negotiating position is inferred

from a downgrade in the seller's credit rating prior to an

announced divestiture.4 The accumulated abnormal returns

are tested for significant difference from zero and CARw and

CARs are compared for significant difference between them

(see Figure 3-3).

It follows from the above hypotheses that acquiring

firms should target, for acquisition, related assets being

sold by parent firms in a weakened negotiating position. A







second-tier stratification is thus required and will result

in four subsamples of cumulative abnormal returns:


1. The acquisition of related divested assets from sellers

in weakened negotiating positions (CARrw)

2. The acquisition of related divested assets from sellers

in strong negotiating positions (CAR rs)







sells to
P > B > A




If P downgraded* => CAR

If P not downgraded => CARs



Ho: CARw = CARs

H1: CARw > CARs


*A downgrade is a negative change in the seller's credit
rating prior to divestiture announcement.





Figure 3-3
First-Tier Stratification
The Acquisition of Divested Assets from
Weakened Parents versus Strong Parents







3. The acquisition of unrelated divested assets from

sellers in weakened negotiating positions (CARuw), and

4. The acquisition of unrelated divested assets from

sellers in strong negotiating positions (CAR us).


The resulting second-tier hypothesis is as follows:

HO: CARrw = CARuw = CARrs = CARus

The cumulative abnormal returns for each substratum are

not significantly different from the cumulative

abnormal returns for all other substrata.


H : CARrw > CARuw, CARrs > CARus

The cumulative abnormal returns for firms acquiring

related divested assets from sellers in weakened nego-

tiating positions will be greater than the cumulative

abnormal returns for firms in the remaining strata.

Further, the cumulative abnormal returns for firms

acquiring unrelated divested assets from sellers in

strong negotiating positions will be less than CARs

from all other strata.


Again, abnormal returns are accumulated for each class

and tested for significant differences from zero. The

stratum-specific CARs are compared and tested for signifi-

cant differences between them (see Figure 3-4).













P
I I


I I
I t
I I
I I I I


I I II
I I


I I I I
I I I I
II II
ASIC=BSIC ASICOBSIC ASIC=BSIC ASICOBSIC


A A A A



CARr CARuw CARrs CARus



HO: CARrw = CARuw = CARrs = CARus

H1: CAR > CARuw, CARrs > CARus






Figure 3-4
Second-Tier Stratification
The Acquisition of Divested Assets


Managerial motives may result in acquisition decisions

that diverge from shareholder preferences. Equityholders

prefer purchasing related divested assets from sellers in

weak negotiating positions. Managers may want to diversify

employment risk and acquire unrelated assets. The nego-

tiating position of the seller may not be a primary factor

in the decision. The following hypotheses will be tested:









H0: 10r = IOu, where 10 represents the percentage of

insider equity ownership

Insider equity ownership will not be significantly

different for firms acquiring related assets versus

firms acquiring unrelated assets.

H I Or > IOu

Insider equity ownership will be greater for firms

acquiring related assets than for firms acquiring un-

related assets.


The percentage of insider equity ownership is deter-

mined for each acquiring firm. The firms are then

segregated into two subsamples dependent upon the related-

ness of the buyer with the acquired assets. The Mann-

Whitney test is used to determine if there is significant

difference between the two strata.

The Mann-Whitney test is a nonparametric test used to

infer whether or not two samples come from the same

population. It can be utilized when it is not reasonable to

make the underlying assumptions of the t-test. The power of

the Mann-Whitney test is about 95.5 percent of the t-test

(see Ben-Horim and Levy, 1981). It can be shown that the

Mann-Whitney test yields the same result as the two-sample

Wilcoxon test.

The population is dichotomized on the basis of related-

ness and ranked per insider equity ownership proportions.







The insider ownership proportions are widely dispersed and

the sample contains many outliers (the range is from 1.0

percent to 55.0 percent). It is reasonable to test the

above hypotheses nonparametrically rather than rely on a t-

test. The U statistics are calculated for each class, as

follows:

nc(nc + 1)
U = R cc
c c 2


where Uc is the U-statistic for each class c, Rc is the sum

of the ranks for each class c, and nc is the number of

observations in class c.

Generally, for any nc > 10 the normal approximation can
2
be used. Thus, the expected valuAIumw and variance, a ,
mw mw'
of the Mann-Whitney U must be calculated:

nrnu
Lmw 2

2 nrnu(nr + nu + 1)
amw 12


Finally, the standard normal Z is calculated to test the

alternative hypothesis:


Ur mw
Z =
mw








Sellers

The full sample hypothesis is that shareholder wealth

effects of firms divesting assets are positive. The null

and alternative hypotheses are as follows:

H0: PCAR = 0

The cumulative abnormal returns for divesting firms

(PCAR) are not significantly different from zero.

H-: PCAR > 0

The cumulative abnormal returns for divesting firms are

greater than zero.


To test this hypothesis, CARs are calculated for the

full sample of 130 firms that divested assets. These

accumulated abnormal returns are tested to determine

significant difference from zero.

Shareholder wealth effects are more accurately deter-

mined if divestitures are classified by correlated factors

rather than tested homogeneously. A first-tier stratifica-

tion process based upon relatedness and parent financial

condition suggests hypotheses that lead to stronger

conclusions than nonstratified testing. The divestiture of

unrelated assets should enhance shareholder wealth more than

the divestiture of related assets. Also, sellers in strong

negotiating positions should enhance value more than sellers

in weak negotiating positions.








The first-tier relatedness hypotheses are as follows:

H0: PCARu = PCARr

The cumulative abnormal returns for firms divesting

unrelated assets (PCAR ) are not significantly

different from the cumulative abnormal returns for

firms divesting related assets (PCARr).

H1: PCARu > PCARr

The cumulative abnormal returns for firms divesting

unrelated assets are greater than the cumulative

abnormal returns for firms divesting related assets.


Cumulative abnormal returns are calculated separately

for related divestitures and unrelated divestitures. Re-

latedness is measured by two-digit SIC code. The CARs from

each subsample are tested to determine significant differ-

ence from zero and then compared to determine significant

difference between each substratum (see Figure 3-5).














sells to
P >- B > A




If PSIC = BSIC => PCAR
r
If PSIC # BSIC => PCARU


H0: PCAR = PCARr

H1: PCARu > PCARr


P -- parent firm (seller)
B -- divested asset
A -- acquiring firm





Figure 3-5
First-Tier Stratification
The Divestiture of
Unrelated Assets versus Related Assets


The first-tier financial condition hypotheses are as

follows:

H0: PCARs = PCARw

The cumulative abnormal returns for selling firms in

strong negotiating positions (PCARs) are not signifi-

cantly different from the cumulative abnormal returns

for selling firms in weak negotiating positions

(PCARw).








H1: PCAR > PCARw

The cumulative abnormal returns for selling firms in

strong negotiating positions are greater than the

cumulative abnormal returns for selling firms in weak

negotiating positions.


Cumulative abnormal returns are calculated separately

for each class. Weakened negotiating position is inferred

from a downgrade in the seller's credit rating prior to an

announced divestiture. The accumulated abnormal returns are

tested for significant difference from zero and PCARs and

PCARw are compared for significant difference between them

(see Figure 3-6).

A second-tier stratification will result in four

subsamples:


1. The divestiture of unrelated assets by sellers in

strong negotiating positions (PCARus)

2. The divestiture of unrelated assets by sellers in weak

negotiating positions (PCAR uw)

3. The divestiture of related assets by sellers in strong

negotiating positions (PCAR rs), and

4. The divestiture of related assets by sellers in weak

negotiating positions (PCARrw).













sells to
P > B > A




If P downgraded* => PCARw

If P not downgraded => PCARs


H0: PCARs = PCARw

Hi: PCARs > PCARw


*A downgrade is a negative change in the seller's credit
rating prior to divestiture announcement.



Figure 3-6
First-Tier Stratification
Divestitures by Strong Sellers versus Weak Sellers


The resulting second-tier hypotheses are as follows:
H0: PCARus = PCARuw = PCARrs = PCARrw

The cumulative abnormal returns for each substratum are

not significantly different from the cumulative

abnormal returns for all other substrata.

H1: PCARus > PCARuw, PCARrs > PCARrw

The cumulative abnormal returns for sellers in strong

negotiating positions that divest unrelated assets will

be greater than the cumulative abnormal returns for








firms in the remaining strata. Further, the cumulative

abnormal returns for sellers in weak negotiating

positions that divest related assets will be less than

CARs from all other strata.


Again, abnormal returns are accumulated for each class

and tested for significant differences from zero. The

stratum-specific CARs are compared and tested for signifi-

cant differences between them (see Figure 3-7).







P
II


No






PSIC=BSIC PSICOBSIC PSIC=BSIC PSICOBSIC

Sus > u P s >















Figure 3-7
I I i I
I BI
I I I I
I I












H1: PCAus > PCARuwPCCARrs > PCARrw










Notes

1. A sell-off occurs when a parent firm divests a portion
of its assets by selling them to another firm.

2. The announcement date for the seller often precedes the
announcement date for the buyer. In many cases, the
parent firm announces its intent to divest a unit(s)
prior to finding a buyer. Thus, for some divestitures
different event days are analyzed for the buyer and
seller. That is, the buyer's AD may differ from the
seller's AD. In all but one case only one bidder came
forth to negotiate, at least as reported in the Wall
Street Journal. Further, there appears to be no public
information released after the announcement and prior
to the effective date.

3. The 149 transactions that had public announcements
appear to be those whose relative size would indicate
new information to the market place.

4. The Wall Street Journal Index was reviewed for two
years prior to an announced divestiture for credit
rating changes by Standard & Poor's and Moody's. Any
credit rating downgrade during this period, that was
not reversed, is considered to signal a worsening in
the firm's financial condition and thus in its
negotiating position.















CHAPTER FOUR
EMPIRICAL RESULTS:
THE ACQUISITION OF DIVESTED ASSETS


The results of acquisition-related empirical tests

suggested in the previous chapter are presented in this

chapter in three parts. First, results from the full sample

tests are compared with results from previous studies. The

analysis based upon the first-tier stratification follows.

Finally the results of tests suggested by the second-tier

stratification are discussed.



Full Sample Results

Table 4-1 presents both the average abnormal returns

(AR) and cumulative abnormal returns (CAR) for the full

sample of 147 firms that acquired divested assets. These

firms represent a subsample of all firms that acquired any

real asset. There are both positive and negative ARs that

have significant t-values (Brown and Warner, 1985).1

Significantly positive average abnormal returns occur on day

two prior to announcement and on days seven, eight, and nine

after announcement. Significantly negative average abnormal

returns occur nine days prior to announcement and six days

after announcement.












Table 4-1
Abnormal Returns of Firms Acquiring Divested Assets:
Full Sample
n = 147


Abnormal Returns
Around
Announcement Day


Cumulative Abnormal
Returns Around
Announcement Day


t
-0.071
1.225
-0.019
0.091
0.363
-1.387*
0.998
1.186
0.071
-0.266
0.694
-0.480
2.009**
0.013
0.732
0.590
-0.519
1.206
-0.486
-0.227
-1.413*
1.296*
2.593***
1.342*
0.765
1.037
-0.784
-0.298
0.214


CAR
-0.011%
-0.297
-0.387
-0.562
-0.441
-0.655
-0.501
-0.318
-0.307
-0.348
-0.241
-0.315
-0.005
-0.003
0.110
0.201
0.121
0.307
0.232
0.197
-0.021
0.179
0.579
0.786
0.904
1.156
0.964
0.581
0.589


t
-0.071
-0.786
-0.756
-0.911
-0.624
-0.905
-0.677
-0.421
-0.398
-0.442
-0.301
-0.386
-0.006
-0.004
0.128
0.230
0.136
0.341
0.254
0.213
-0.022
0.188
0.601
0.806
0.915
1.105
0.875
0.503
0.489


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


Day
-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR
-0.011%
0.189
-0.003
0.014
0.056
-0.214
0.154
0.183
0.011
-0.041
0.107
-0.074
0.310
0.002
0.113
0.091
-0.080
0,186
-0.075
-0.035
-0.218
0.200
0.400
0.207
0.118
0.160
-0.121
-0.046
0.033










Average abnormal returns accumulated over the entire

61-day test period (day -30 to day +30) yield 0.589 percent

with an insignificant t-value (t-statistic is 0.4888).2

This result is consistent with results from studies of

acquiring firms in mergers. Jensen and Ruback (1983)

analyze several merger studies and find weighted average

abnormal returns over selected intervals. Table 4-2

compares their results with those of the present study.

While the two-day announcement effects appear to differ,

neither result is significantly different from zero. This

study is consistent with evidence that acquiring firms do

not gain from merger. However, this evidence results from

studies that treat all acquiring firms generically.

Table 4-3 facilitates comparison of results obtained in

this study with those obtained in other studies of firms

acquiring divested assets. The evidence is inconclusive.

Jain (1985) finds cumulative abnormal returns similar to

those found in this analysis. On the other hand, Rosenfeld

(1984) finds positive CARs for firms purchasing divested

assets. The results appear to be sample-specific and may be

biased by the small number of firms in Rosenfeld's sample.

When all mergers are studied as a homogeneous group, or

even as large subsamples such as divestitures, the evidence

is confounding. No conclusions or strong inferences are







The following sections suggest


clearer conclusions based upon sample stratification.




Table 4-2
Cumulative Abnormal Returns of Acquiring Firms:
Divested Assets versus Nonclassified Acquisitionsa


Cumulative Abnormal Returns

Nonclassifieg
Intervalsb Acquisitions This Studyc

Two-day announcement effect -0.0005 0.0021

One-month announcement effect 0.0137 0.0129


aNonclassified acquisition refers to the total population of
acquired firms, in merger, not classified by type.
bjensen and Ruback (1983) find weighted average abnormal
returns from existing merger studies for the intervals
presented.
cThe cumulative abnormal returns presented for this study
are based upon the intervals reported in Jensen and Ruback
(1985). These intervals are not identical for all studies,
thus perfect comparisons cannot be made.


produced in this approach.












Table 4-3
Cumulative Abnormal Returns of Firms Acquiring
Divested Assets--This Study versus Others


Intervals

-5 to -1
+1 to +5
+6 to +10
-30 to +30
-30 to -2
-1 to +1
+2 to +10
+2 to +30
+11 to +30


Cumulative Abnormal Returns

Jaina Rosenfeldb This Study

-.001 .003
.004 .001
-.005 -- .007
-- .0706 .0059
-- .0053 -.0001
-- .0210 .0021
-- .0004 .0070
-- .0422 .0039
-- .0418 -.0032


aFrom Jain (1985). The sample consists of 304 acquiring
firms.
bFrom Rosenfeld (1984). The sample consists of 30 acquiring
firms.


First-Tier Stratification Results


Relatedness

The population of firms acquiring divested assets is

stratified as a function of the acquiring firm's relatedness

with the acquired asset. Forty-nine firms acquired related

divested assets and 98 firms acquired unrelated divested

assets.








Table 4-4 lists the average abnormal returns for the

two strata over the 61-day test period. Several days yield

significantly positive abnormal returns for related acquisi-

tions, with the highest significance on days seven, eight,

and ten after announcement. Unrelated acquisitions yield

both positive and negative ARs that have significant t-

values.

Table 4-5 highlights the need for stratification based

upon relatedness. Firms acquiring related divested assets

earn statistically significant positive cumulative abnormal

returns of 3.2 percent (t-statistic of 1.51) over the test

period. Firms acquiring unrelated divested assets

accumulate negative (nonsignificant) abnormal returns over

the test period. Further, cumulative abnormal returns from

related acquisitions are greater than cumulative abnormal

returns from unrelated acquisitions for all days except day

-15. The CARs from related acquisitions are significantly

greater than the CARs from unrelated acquisitions on several

days. The accumulation of abnormal returns by firms buying

related assets exceeds the cumulative abnormal returns of

firms buying unrelated assets by almost 4.0 percent over the

61-day period (t-statistic of 1.52).3











Table 4-4
Average Abnormal Returns of Firms Acquiring Divested Assets:
Related versus Unrelated Acquisitions


Related Acquisitions
n = 49


Unrelated Acquisitions
n = 98


t

0.209
0.231
-0.736
-0.224
1.114
* 0.777
-0.059
1.539*
-0.451
1.422*
-0.081
0.799
1.381*
0.202
1.308*
0.454
0.656
0.528
-0.520
-0.638
-1.074
2.026**
2.202**
1.275
1.891**
-0.315
-0.344
0.103
-0.868


AR

-0.027%
0.236
0.088
0.076
-0.084
-0.402
0.194
0.092
0.061
-0.226
0.189
-0.211
0.256
-0.081
-0.016
0.097
-0.207
0.177
-0.091
0.032
-0.109
0.033
0.295
0.131
-0.083
0.296
-0.144
-0.064
0.146


t

-0.138
1.208
0.450
0.389
-0.430
-2.057**
0.993
0.471
0.312
-1.156
0.967
-1.080
1.310*
-0.414
-0.082
0.496
-1.059
0.906
-0.466
0.164
-0.558
0.169
1.509*
0.670
-0.425
1.515*
-0.737
-0.328
0.747


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


Day


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR

0.057%
0.063
-0.201
-0.061
0.304
0.212
-0.016
0.420
-0.123
0.388
-0.022
0.218
0.377
0.055
0.357
0.124
0.179
0.144
-0.142
-0.174
-0.293
0.553
0.601
0.348
0.516
-0.086
-0.094
0.028
-0.237












Table 4-5
Cumulative Abnormal Returns of Firms Acquiring Divested Assets:
Related versus Unrelated Acquisitions


Related
Acquisitions
n = 49


Unrelated
Acquisitions
n = 98


Related Minus
Unrelated


Day CAR


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


0.057%
-0.128
-0.255
-0.553
-0.270
-0.058
-0.074
0.346
0.223
0.611
0.589
0.807
1.184
1.239
1.596
1.720
1.899
2.043
1.901
1.727
1.434
1.987
2.588
2.936
3.452
3.278
3.044
3.387
3.227


t

0.209
-0.191
-0.282
-0.507
-0.216
-0.045
-0.057
0.259
0.163
0.439
0.415
0.559
0.806
0.829
1.050
1.114
1.211
1.284
1.177
1.055
0.864
1.181
1.518*
1.701**
1.975**
1.771**
1.562*
1.658**
1.514*


CAR

-0.027%
-0.387
-0.438
-0.515
-0.571
-0.973
-0.779
-0.687
-0.626
-0.852
-0.663
-0.874
-0.618
-0.699
-0.715
-0.618
-0.825
-0.648
-0.739
-0.707
-0.816
-0.783
-0.488
-0.357
-0.440
0.061
-0.120
-0.804
-0.748


t

-0.138
-0.808
-0.676
-0.659
-0.638
-1.061
-0.831
-0.718
-0.641
-0.855
-0.653
-0.845
-0.587
-0.653
-0.657
-0.559
-0.735
-0.569
-0.639
-0.603
-0.686
-0.650
-0.400
-0.289
-0.352
0.046
-0.086
-0.550
-0.490


CAR

0.084%
0.259
0.183
-0.038
0.301
0.915
0.705
1.033
0.849
1.463
1.252
1.681
1.802
1.938
2.311
2.338
2.724
2.691
2.640
2.434
2.250
2.770
3.076
3.293
3.892
3.217
3.164
4.191
3.975


t

0.250
0.315
0.164
-0.028
0.196
0.581
0.43.8
0.628
0.506
0.855
0.718
0.946
0.997
1.054
1.236
1.231
1.413*
1.375*
1.329*
1.208
1.102
1.339*
1.467*
1.551*
1.811**
1.413*
1.320*
1.668**
1.516*


*Significant
**Significant
***Significant


at 10% level using a one-tail test.
at 5% level using a one-tail test.
at 1% level using a one-tail test.








Comparisons of the stratum-specific CARs over selected


intervals are presented in Table 4-6.


Highly significant


cumulative abnormal returns result from related acquisitions

for both the 10-day period prior to announcement and the 10-


day period after the announcement.


The CARs resulting from


unrelated acquisitions are not significant and are negative

except for the 10-day period after announcement.




Table 4-6
Cumulative Abnormal Returns of Firms Acquiring Divested Assets:
Related versus Unrelated Acquisitions
Selected Intervals


Interval


-30
-30
-30
-10
+1
+1
+11


+30
-11
-1
-1
+10
+30
+30


Related Acquisitions
CAR t


3.227%
-0.574
1.239
1.813
1.856
1.631
-0.225


1.51*
-0.47
0.83
2.10**
2.15**
1.09
-0.18


Unrelated Acquisitions
CAR t


-0.748%
-0.487
-0.699
-0.212
0.275
-0.033
-0.308


-0.49
-0.56
-0.65
-0.34
0.44
-0.03
-0.35


Differences


Interval


-30
-30
-30
-10
+1
+1
+11


to +30
to -11
to -1
to -1
to +10
to +30
to +30


Related CAR Unrelated CAR Difference t


3.227%
-0.574
1.239
1.813
1.856
1.631
-0.225


-0.748%
-0.487
-0.699
-0.212
0.275
-0.033
-0.308


3.975%
-0.087
1.938
2.025
1.581
1.664
0.083


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


1.52*
-0.58
1.05
1.91**
1.49*
0.90
0.06









Significant differences between the related-acquisition

CARs and the unrelated-acquisition CARs occur for two

intervals other than the entire test period. A difference

of greater than 2.0 percent is found for the 10-day period

immediately prior to announcement (t-statistic of 1.91).

Related-acquisition CARs exceed unrelated-acquisition CARs

by 1.6 percent for the 10 days following announcement (t-

statistic of 1.49). The evidence supports the hypothesis

that firms acquiring related divested assets enhance

shareholder wealth more than firms acquiring unrelated

divested assets.

Firms still acquire unrelated assets despite the

evidence that the purchase of related assets enhances

shareholder wealth more than the purchase of unrelated

assets. For the sample used in this study, 98 firms

acquired unrelated assets and 49 firms acquired related

assets. It is hypothesized that managerial motivation to

reduce personal risk-bearing results in the acquisition of

unrelated assets.

The proportion of insider equity ownership to total

equity ownership is determined for each firm.4 The firms

are then classified as acquiring related divested assets or

as acquiring unrelated divested assets. To determine if the

two strata come from different populations, the Mann-Whitney

U test is applied. Table 4-7 highlights the results of this

procedure.












Table 4-7
Insider Equity Ownership and the Acquisition Decision:
Mann-Whitney U Test


Sample

Related acquisitions

Unrelated acquisitions

Total


4mw


Sample
Size

38

84

122


Group
Mean

11.38%

8.07%


Rank
Sum U-statistic

2704.5 1963.5**

4798.5 1228.5


= 1596.0


a2 = 32,718.0
mw


Ur ~mw
r -
r =
mw


1963.5 1596.0

180.88


= 2.03**


**Significant at 2.5% level using a one-tail test.


The hypothesis that firms acquiring related divested

assets are more likely to have greater insider equity

ownership than firms acquiring unrelated divested assets is

tested. The resulting Mann-Whitney test statistic is 2.03,

which is significant at the 2.5 percent level. The evidence

is strong that the two subsamples, firms purchasing related

acquisitions and firms purchasing unrelated acquisitions,







come from different populations. Firms that buy related

units tend to have higher insider ownership than firms that

buy unrelated units.


Financial Condition

Stratification as a function of the selling firm's

changes in financial condition is suggested by the

hypotheses offered in Chapter 3. Forty-two firms had credit

downgrades and are classified as weak. The remaining 105

firms are classified as strong.

Table 4-8 presents the average abnormal returns for

these two strata over the 61-day test period. Significant

positive ARs occur on days seven, eight, and fifteen after

announcement for firms acquiring assets from parents in weak

negotiating positions. The ARs are significantly positive

on days seven and two before announcement and days three and

eight after announcement for firms acquiring assets from

parents other than those in weak negotiating positions.











Table 4-8
Average Abnormal Returns of Firms Acquiring Divested Assets:
Weak versus Strong Parents


Weak Parents
n = 42


Strong Parents
n = 105


t

0.800
1.014
-0.388
1.145
-0.634
-0.831
-0.102
-0.018
-0.800
-0.053
1.036
-0.197
0.430
-0.782
0.395
0.416
-0.860
-0.204
0.070
0.564
0.148
1.638*
1.543*
0.993
0.310
1.874**
-0.916
-0.060
0.201


AR

-0.105%
0.149
0.039
-0.110
0.150
-0.205
0.227
0.258
0.106
-0.051
0.032
-0.082
0.385
0.091
0.113
0.079
-0.014
0.283
-0.113
-0.113
-0.322
0.094
0.386
0.177
0.130
0.012
-0.066
-0.058
0.023


t


AR

0.227%
0.288
-0.110
0.325
-0.180
-0.236
-0.029
-0.005
-0.227
-0.015
0.294
-0.056
0.122
-0.222
0.112
0.118
-0.244
-0.058
0.020
0.160
0.042
0.465
0.438
0.282
0.088
0.532
-0.260
-0.017
0.057


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


Day


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


-0.560
0.794
0.208
-0.586
0.800
-1.093
1.210
1.375*
0.565
-0.272
0.171
-0.437
2.052**
0.485
0.602
0.421
-0.075
1.508*
-0.602
-0.602
-1.716**
0.501
2.057**
0.943
0.693
0.064
-0.352
-0.309
0.123








The results presented in Table 4-9 again support

stratification. The accumulated abnormal returns resulting

from the acquisition of divested assets from weak parents

are 2.3 percent to day +20 and 1.5 percent for the entire

61-day test period. In contrast, the CARs resulting from

the acquisition of divested assets from strong parents are

less than 1.0 percent for the test period. The CARs

resulting from weak sellers exceed the CARs resulting from

strong sellers for accumulations to any given day. This

result is consistent with the hypothesis that acquiring

assets from sellers in weak negotiating positions enhances

shareholder wealth more than acquiring assets from sellers

in stronger negotiating positions (albeit statistical

significance is lacking).

The stratum-specific results are presented for selected

intervals in Table 4-10. For most intervals, the CARs

resulting from weak sellers exceed the CARs resulting from

stronger sellers. The lone significant exception is for the

10-day period prior to announcement. Acquisitions from

strong parents yield statistically significant positive CARs

and acquisitions from weak parents yield negative CARs

during this period.












Table 4-9
Cumulative Abnormal Returns of Firms Acquiring Divested Assets:
Weak versus Strong Parents


Weak Parents
n = 42


Strong Parents
n = 105


Weak Minus
Strong


Day

-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


CAR t CAR


0.227%
0.515
0.262
0.684
0.897
0.661
0.632
0.627
0.400
0.385
0.679
0.623
0.745
0.523
0.635
0.753
0.509
0.451
0.471
0.631
0.673
1.138
1.576
1.858
1.946
2.126
2.287
1.478
1.547


0.800
0.741
0.278
0.602
0.689
0.496
0.464
0.451
0.282
0.266
0.460
0.415
0.487
0.336
0.402
0.469
0.312
0.272
0.280
0.370
0.390
0.650
0.889
1.035
1.071
1.104
1.128
0.696
0.698


-0.105%
-0.620
-0.646
-1.060
-0.977
-1.182
-0.955
-0.697
-0.591
-0.642
-0.610
-0.692
-0.307
-0.216
-0.103
-0.024
-0.038
0.245
0.132
0.019
-0.303
-0.209
0.177
0.354
0.484
0.765
0.430
0.217
0.199


t

-0.560
-1.349*
-1.038
-1.413*
-1.136
-1.343*
-1.061
-0.758
-0.630
-0.671
-0.626
-0.697
-0.304
-0.210
-0.099
-0.023
-0.035
0.224
0.119
0.017
-0.266
-0.181
0.151
0.298
0.403
0.601
0.321
0.155
0.136


CAR

0.332%
1.135
0.908
1.744
1.874
1.843
1.587
1.324
0.991
1.027
1.289
1.315
1.052
0.739
0.738
0.777
0.547
0.206
0.339
0.612
0.976
1.347
1.399
1.504
1.462
1.361
1.857
1.261
1.348


t

0.976
1.362*
0.805
1.281
1.202
1.155
0.972
0.794
0.582
0.592
0.729
0.730
0.574
0.396
0.390
0.404
0.280
0.104
0.168
0.300
0.472
0.642
0.658
0.699
0.671
0.590
0.764
0.495
0.507


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.












Table 4-10
Cumulative Abnormal Returns of Firms Acquiring Divested Assets:
Weak versus Strong Parents
Selected Intervals


Interval

-30 to +30
-30 to -11
-30 to -1
-10 to -1
+1 to +10
+1 to +30
+11 to +30


Weak Parents
CAR t


1.547%
1.077
0.523
-0.554
1.311
0.912
-0.399


0.70
0.85
0.34
-0.62
1.46*
0.59
-0.31


Strong Parents
CAR t


0.199%
-1.127
-0.216
0.911
0.587
0.302
-0.285


0.14
-1.34*
-0.21
1.54*
0.99
0.29
-0.34


Differences


Interval

-30 to +30
-30 to -11
-30 to -1
-10 to -1
+1 to +10
+1 to +30
+11 to +30


*Significant
**Significant
***Significant


Weak CAR Strong CAR Difference


1.547%
1.077
0.523
-0.554
1.311
0.912
-0.399


0.199%
-1.127
-0.216
0.911
0.587
0.302
-0.285


1.348%
2.204
0.739
-1.465
0.724
0.610
-0.114


at 10% level using a one-tail test.
at 5% level using a one-tail test.
at 1% level using a one-tail test.


t

0.51
1.45*
0.40
-1.36*
0.67
0.33
-0.07








Second-Tier Stratification Results

Second-tier stratification results in four substrata:


1. Firms acquiring related divested assets from weak

parents,

2. Firms acquiring unrelated divested assets from weak

parents,

3. Firms acquiring related divested assets from strong

parents, and

4. Firms acquiring unrelated divested assets from strong

parents.



Related Acquisitions--Strong Sellers

Resulting average abnormal returns and cumulative

abnormal returns from the acquisition of related assets from

strong sellers are presented in Table 4-11. The CARs for

this stratum are significantly positive from the day prior

to announcement to thirty days beyond announcement.

Abnormal returns accumulate to 3.7 percent over the test

period (t-statistic is 1.58). Highly significant cumulative

abnormal returns of 4.5 percent occur through AD +10 (t-

statistic of 2.38).












Table 4-11
Second-Tier Stratification:
Related Acquisitions--Strong Sellers


CAR


Day


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


-0.008%
-0.183
-0.221
-0.544
0.247
0.417
0.477
0.975
0.938
1.416
1.396
1.688
1.998
2.193
2.488
2.622
2.857
3.301
3.305
3.248
2.619
3.112
3.545
3.811
4.545
4.354
3.799
3.803
3.684


-0.027
-0.251
-0.246
-0.456
0.181
0.298
0.334
0.668
0.630
0.932
0.902
1.071
1.245
1.344*
1.500*
1.555*
1.669**
1.900**
1.875**
1.817**
1.445*
1.694**
1.905**
2.022**
2.382***
2.154**
1.785**
1.705**
1.583*


*Significant at
**Significant at
***Significant at


10% level using a one-tail test
5% level using a one-tail test.
1% level using a one-tail test.


-0.027
0.148
-0.329
-0.577
1.792**
0.571
0.201
1.671**
-0.124
1.604*
-0.067
0.980
1.040
0.654
0.990
0.450
0.789
1.490*
0.013
-0.191
-2.111**
1.654**
1.453*
0.893
2.463***
-0.557
0.148
-0.406
-0.624


-0.008%
0.044
-0.098
-0.172
0.534
0.170
0.060
0.498
-0.037
0.478
-0.020
0.292
0.310
0.195
0.295
0.134
0.235
0.444
0.004
-0.057
-0.629
0.493
0.433
0.266
0.734
-0.166
0.044
-0.121
-0.186







Related Acquisitions--Weak Sellers

The ARs and CARs for related acquisitions from weak

sellers are listed in Table 4-12. Positive abnormal returns

are accumulated over the test period (1.8 percent with a t-

statistic of 0.42). However, for much of the test period

CARs are negative. Inferences derived from the results of

this stratification are limited due to the sample size of

this stratum.



Unrelated Acquisitions--Strong Sellers

Abnormal returns from the acquisition of unrelated

assets from strong sellers are presented in Table 4-13.

Negative, and sometimes significant, cumulative abnormal

returns occur throughout the test period. The CAR for the

entire 61-day interval is -1.8 percent (t-statistic of

-0.95).



Unrelated Acquisitions--Weak Sellers

Abnormal returns from the purchase of unrelated units

from weak sellers are listed in Table 4-14. The CARs are

positive but insignificantly different from zero. Abnormal

returns of 1.5 percent occur over the test period (t-

statistic is 0.55).











Table 4-12
Second-Tier Stratification:
Related Acquisitions--Weak Sellers


CAR


Day


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


0.125%
0.125
-0.505
-1.043
-1.665
-1.533
-1.414
-1.517
-1.783
-1.965
-2.156
-2.275
-1.487
-1.415
-0.786
-0.900
-0.938
-1.574
-1.778
-2.330
-2.099
-1.420
-0.193
0.495
0.278
-0.164
0.798
1.768
1.778


0.229
0.094
-0.279
-0.478
-0.666
-0.599
-0.541
-0.568
-0.654
-0.706
-0.761
-0.788
-0.506
-0.474
-0.259
-0.292
-0.299
-0.495
-0.551
-0.712
-0.633
-0.422
-0.057
0.144
0.080
-0.044
0.205
0.433
0.417


*Significant at
**Significant at
***Significant at


10% level using a one-tail test.
5% level using a one-tail test.
1% level using a one-tail test.


0.229
0.493
-0.898
0.207
-0.636
0.242
0.218
-0.189
-0.488
-0.334
-0.350
-0.218
1.445*
0.132
1.153
-0.209
-0.070
-1.166
-0.374
-1.012
0.423
1.245
2.250**
1.261
-0.398
0.143
-0.893
0.682
-0.405


0.125%
0.269
-0.490
0.113
-0.347
0.132
0.119
-0.103
-0.266
-0.182
-0.191
-0.119
0.788
0.072
0.629
-0.114
-0.038
-0.636
-0.204
-0.552
0.231
0.679
1.227
0.688
-0.217
0.078
-0.487
0.372
-0.221











Table 4-13
Second-Tier Stratification:
Unrelated Acquisitions--Strong Sellers


CAR


Day


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


-0.161%
-0.870
-0.888
-1.353
-1.670
-2.088
-1.766
-1.645
-1.457
-1.808
-1.746
-2.040
-1.613
-1.581
-1.570
-1.521
-1.677
-1.485
-1.664
-1.808
-1.956
-2.088
-1.729
-1.602
-1.814
-1.265
-1.473
-1.808
-1.768


-0.674
-1.486*
-1.121
-1.416*
-1.525*
-1.863**
-1.541*
-1.4Q5*
-1.220
-1.484*
-1.406*
-1.613*
-1.254
-1.208
-1.180
-1.125
-1.222
-1.066
-1.177
-1.261
-1.346*
-1.418*
-1.159
-1.060
-1.186
-0.781
-0.863
-1.011
-0.947


*Significant at 10% level using a one-tail test
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


-0.674
0.871
0.490
-0.310
-0.280
-1.749**
1.348*
0.506
0.787
-1.469*
0.260
-1.230
1.787**
0.134
0.046
0.205
-0.653
0.804
-0.749
-0.603
-0.619
-0.552
1.502*
0.532
-0.887
0.473
-0.536
-0.092
0.594


-0.161%
0.208
0.117
-0.074
-0.067
-0.418
0.322
0.121
0.188
-0.351
0.062
-0.294
0.427
0.032
0.011
0.049
-0.156
0.192
-0.179
-0.144
-0.148
-0.132
0.359
0.127
-0.212
0.113
-0.128
-0.022
0.142










Table 4-14
Second-Tier Stratification:
Unrelated Acquisitions--Weak Sellers


Day

-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR

0.262%
0.295
0.025
0.400
-0.121
-0.367
-0.082
0.030
-0.212
0.044
0.466
-0.033
-0.114
-0.326
-0.072
0.200
-0.317
0.146
0.100
0.413
-0.025
0.389
0.157
0.139
0.196
0.693
-0.180
-0.155
0.155


t

0.763
0.859
0.073
1.164
-0.352
-1.068
-0.239
0.087
-0.617
0.128
1.357*
-0.096
-0.332
-0.949
-0.210
0.582
-0.923
0.425
0.291
1.202
-0.073
1.132
0.457
0.405
0.571
2.017**
-0.524
-0.451
0.451


*Significant at 10% level using a one-tail test
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


CAR

0.262%
0.653
0.534
1.296
1.806
1.439
1.357
1.387
1.175
1.219
1.685
1.652
1.538
1.212
1.140
1.340
1.023
1.169
1.269
1.682
1.657
2.046
2.203
2.342
2.538
2.939
2.816
1.376
1.466


t

0.763
0.776
0.469
0.943
1.147
0.893
0.824
0.824
0.684
0.696
0.944
0.909
0.831
0.644
0.596
0.690
0.518
0.584
0.624
0.816
0.793
0.966
1.027
1.078
1.154
1.262
1.148
0.535
0.546










Summary


Table 4-15 summarizes the cumulative abnormal returns

over the 61-day test period for each stratum tested. The

acquisition of related assets yields significantly positive

CARs. The acquisition of unrelated divested assets yields

negative (insignificant) CARs. Further, the accumulation of

abnormal returns is significantly greater for related

acquisitions than for unrelated acquisitions. Purchases of

unrelated divested assets are more likely to be associated

with lower levels of insider equity ownership than purchases

of related divested assets.

When analyzed homogeneously, the acquisition of

divested assets results in slightly positive abnormal

returns to the buying firm's shareholders. The evidence is

clear, however, that some acquisitions enhance shareholder

wealth and others may reduce shareholder wealth. The

purchase of related assets increases shareholder wealth.

This result holds regardless of the seller's financial

condition.

Results from second-tier stratification are not as

clear. For related acquisitions, buyers enhance wealth more

by purchasing from strong sellers than by purchasing from

weak sellers. For unrelated acquisitions, buyers increase

shareholder wealth more by purchasing from weak sellers than

by purchasing from strong sellers.














Table 4-15
Acquisition of Divested Assets:
Stratum-Specific CARs over 61-Day Test Period


Stratum


Full
Related
Unrelated
Weak
Strong
Related/Strong Seller
Related/Weak Seller
Unrelated/Strong Seller
Unrelated/Weak Seller


Sample Size


147
49
98
42
105
38
11
67
31


CAR

0.589%
3.227
-0.748
1.547
0.199
3.684
1.778
-1.768
1.466


t

0.49
1.51*
-0.49
0.70
0.14
1.58*
0.42
-0.95
0.55


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.











Notes


1. See Brown and Warner (1985, p. 7). The t-test is:

ARt

S


where S is the estimated standard deviation of
abnormal returns over the estimation period.


average


The hypotheses tested throughout this analysis
suggest one-tail testing is appropriate.


2. Brown and Warner (1985, p. 29).
accumulated abnormal returns is:


E AR
t -


CARt


The t-test for


t 2 1/2 1/2
(S S ) (t S)
(t


3. See Miles and Rosenfeld (1983, p. 1603).
for a difference in CARs is:


The t-test


ARil ARi2


il i2

where AR is the stratum-specific average abnormal re-
turn over interval i, T. is the stratum-specific number
of days in interval i, and S is the pooled estimated
of each group's standard deviation. Note:










(1)
Sp =


2 2 1/2
(TBase(il) 1)S1 + (TBase(i2) 1)S2
TBase(il) + TBase(i2) 2


179(S1 2) + 179(S22) 1/2
358



since TBase(il) = TBase(i2) = 180


(2) AR = 1 AR( .) =
1


CART

i^


Thus the t-value can be rewritten as:


CART

T il


CARTi2
Ti2


1 1 31/2
Sp + I
peil 2J


CARi CAR i2

T i

S ( 2)12


since T. = T. for all tests undertaken.
Finally, to simplify and allow the reader to
clearly see the statistic as a test of the
difference between CARs:


CAR1 CAR2
T S -2) 1/2
i Ti





67





4. Insider equity ownership proportions are gathered from
Value Line. The data are not available for 25 firms,
leaving 122 firms to test for insider ownership-
relatedness correlation. Of these, 38 firms acquired
related assets and 84 firms acquired unrelated assets.


5. Weak and strong refer to the selling firm's negotiating
position rather than reference to the firm's overall
financial condition.















CHAPTER FIVE
EMPIRICAL RESULTS:
THE DIVESTITURE OF ASSETS


The results of empirical tests associated with the

purchase of divested assets were presented in Chapter 4.

The impact upon shareholder wealth of those firms that

divest assets are presented in this chapter.

First, results from the full sample tests are compared

with results from previous divestiture studies. An analysis

of the sample stratified on the basis of relatedness and

financial condition follows. Finally the results of

second-tier stratification tests of the joint effects of

relatedness and financial condition are discussed.



Full Sample Results

Table 5-1 presents both the average abnormal returns

(AR) and cumulative abnormal returns (CAR) for the full

sample of 130 firms that divested assets. There are both

positive and negative ARs that have significant t-values.












Table 5-1
Abnormal Returns of Firms Divesting Assets
Full Sample
n = 130


Abnormal Returns
Around
Announcement Day


Cumulative Abnormal
Returns Around
Announcement Day


t
-0.053
0.334
0.689
-2.056**
1.298*
0.620
0.716
1.839**
-0.249
1.876**
0.064
0.117
-0.551
5.576***
0.212
-0.975
-0.541
-0.318
-1.961**
1.108
0.339
0.064
1.511*
-1.081
0.986
1.150
0.869
0.811
1.537*


*Significant at
**Significant at
***Significant at


10% level using a one-tail test.
5% level using a one-tail test.
1% level using a one-tail test.


Day
-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR
-0.010%
0.063
0.130
-0.388
0.245
0.117
0.135
0.347
-0.047
0.354
0.012
0.022
-0.104
1.052
0.040
-0.184
-0.102
-0.060
-0.370
0.209
0.064
0.012
0.285
-0.204
0.186
0.217
0.164
0.153
0.290


CAR
-0.010%
0.100
-0.079
-0.713
-0.971
-0.854
-0.719
-0:372
-0.419
-0.065
-0.053
-0.031
-0.135
0.917
0.957
0.773
0.671
0.611
0.241
0.450
0.514
0.526
0.811
0.607
0.793
1.223
1.129
1.487
1.743


t
-0.053
0.216
-0.126
-0.945
-1.123
-0.965
-0.795
-0.403
-0.444
-0.068
-0.054
-0.031
-0.133
0.887
0.911
0.724
0.619
0.555
0.216
0.398
0.448
0.452
0.688
0.509
0.656
0.956
0.838
1.053
1.183








Significantly positive average abnormal returns occur on

days ten, seven, five, and one prior to announcement (AD

-10, -7, -5, -1) and days eight and thirty after announce-

ment (AD +8, +30). The largest positive AR occurs on AD -1,

the day prior to the Wall Street Journal announcement (1.05

percent with a t-statistic of 5.576). Significantly

negative average abnormal returns occur fifteen days prior

to announcement (AD -15) and four days after announcement

(AD +4).

Average abnormal returns accumulate to 1.74 percent (t-

statistic is 1.18) over the entire 61-day test period.

Table 5-2 facilitates comparison of these results with

results obtained in other divestiture studies. The full-

sample evidence is again inconclusive. Rosenfeld (1984)

finds post-announcement cumulative abnormal returns similar

to, yet larger than those found in this analysis. Jain

(1985) finds negative CARs for the pre-announcement period

which is consistent with this study.

Divestiture studies, like acquisition studies, yield no

clear inferences or conclusions when analyzed as a

homogeneous group. The following sections produce stronger

conclusions based upon sample stratification.







First-Tier Stratification Results

Relatedness

The population of firms divesting assets is stratified

as a function of the selling firm's relatedness with the

divested asset. Ninety-six firms divested unrelated assets

and 34 firms divested related assets.





Table 5-2
Cumulative Abnormal Returns of Firms Divesting Assets
This Study versus Others


Cumulative Abnormal Returns

Intervals Jaina Rosenfeldb This Study

-120 to -1 -3.25% -- % -- %
-30 to -2 -- 1.13 -1.35
-1 to 0 0.47 2.33 1.09
+1 to +30 -0.22 1.80 0.79
-30 to +30 -- 5.25 1.74
aFrom Jain (1985). The sample consists of 1062 divesting
firms.
bFrom Rosenfeld (1984). The sample consists of 62 divesting
firms.




Table 5-3 lists the average abnormal returns for the two

strata over the 61-day test period. Highly-significant

positive abnormal returns occur on the day prior to the

announcement of unrelated divestitures (1.36 percent with a

t-statistic of 6.14). Both significantly positive and







negative abnormal returns occur as a result of related

divestitures.

Table 5-4 suggests the need for stratification based

upon relatedness. Firms divesting unrelated assets earn

statistically significant positive cumulative abnormal

returns of 2.6 percent (t-statistic of 1.48) over the test

period. Firms divesting related assets accumulate negative

(nonsignificant) abnormal returns over the test period.

Further, the accumulation of abnormal returns from unrelated

divestitures exceeds the cumulative abnormal returns of

related divestitures by 3.1 percent over the 61-day test

period. The CARs from unrelated divestitures are signifi-

cantly greater than the CARs from related divestitures on

several days.

Comparisons of the stratum-specific CARs over selected

intervals are presented in Table 5-5. Highly-significant

positive cumulative abnormal returns result from unrelated

divestitures for the 10-day period prior to announcement

(2.5 percent with a t-statistic of 3.56). Significant nega-

tive CARs result from related divestitures for the 10-day

period after announcement (-1.6 percent with a t-statistic

of -1.31).












Table 5-3

Average Abnormal Returns of Firms Divesting Assets
Unrelated versus Related Divestitures


Unrelated Divestitures
n = 96


Related Divestitures
n = 34


t

-0.162
-0.491
0.072
-3.093***
1.371*
1.416*
0.171
1.078
-0.194
1.267
0.104
-0.334
0.225
6.145***
0.509
-0.460
0.203
-0.947
-1.407*
1.470*
1.456*
-0.117
1.267
-0.537
0.600
0.631
0.189
0.320
0.361


AR

0.065%
0.548
0.455
0.454
0.079
-0.440
0.409
0.650
-0.059
0.560
-0.018
0.291
-0.540
0.172
-0.167
-0.417
-0.518
0.361
-0.534
-0.119
-0.666
0.116
0.294
-0.441
0.336
0.435
0.507
0.384
0.884


t

0.170
1.435*
1.191
1.189
0.207
-1.152
1.071
1.702**
-0.155
1.466*
-0.047
0.762
-1.414*
0.450
-0.437
-1.092
-1.356*
0.945
-1.398*
-0.312
-1.744**
0.304
0.770
-1.155
0.880
1.139
1.327*
1.005
2.314**


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


Day

-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR

-0.036%
-0.109
0.016
-0.686
0.304
0.314
0.038
0.239
-0.043
0.281
0.023
-0.074
0.050
1.363
0.113
-0.102
0.045
-0.210
-0.312
0.326
0.323
-0.026
0.281
-0.119
0.133
0.140
0.042
0.071
0.080














Table 5-4
Cumulative Abnormal Returns of Firms Divesting Assets
Unrelated versus Related Divestitures


Unrelated
Divestitures
n = 96


Related
Divestitures
n = 34


Unrelated
Minus Related


CAR

-0.036%
-0.095
-0.195
-0.981
-0.826
-0.512
-0.474
-0.235
-0.278
0.003
0.026
-0.048
0.002
1.365
1.478
1.376
1.421
1.211
0.899
1.225
1.548
1.522
1.803
1.684
1.817
2.059
2.355
2.859
2.556


t


-0.162
-0.175
-0.265
-1.106
-0.813
-0.492
-0.446
-0.216
-0.251
0.003
0.023
-0.041
0.002
1.124
1.197
1.097
1.115
0.937
0.685
0.921
1.147
1.113
1.302
1.200
1.279
1.369*
1.487*
1.722**
1.475*


CAR

0.065%
0.656
0.256
0.053
-1.372
-1.812
-1.403
-0.753
-0.812
-0.252
-0.270
0.021
-0.519
-0.347
-0.514
-0.931
-1.449
-1.088
-1.622
-1.741
-2.407
-2.291
-1.997
-2.438
-2.102
-1.138
-2.340
-2.394
-0.561


CAR

-0.101%
-0.751
-0.451
-1.034
0.546
1.300
0.929
0.518
0.534
0.255
0.296
-0.069
0.521
1.712
1.992
2.307
2.870
2.299
2.521
2.966
3.955
3.813
3.800
4.122
3.919
3.197
4.695
5.253
3.117


t

-0.229
-0.694
-0.308
-0.585
0.270
0.628
0.439
0.239
0.242
0.113
0.129
-0.030
0.219
0.708
0.810
0.923
1.131
0.893
0.965
1.119
1.472*
1.400*
1.378*
1.476*
1.386*
1.067
1.488*
1.589*
0.904


*Significant
**Significant
***Significant


at 10% level using a one-tail test.
at 5% level using a one-tail test.
at 1% level using a one-tail test.


Day

-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


t

0.170
0.701
0.202
0.035
-0.784
-1.011
-0.766
-0.402
-0.425
-0.129
-0.136
0.010
-0.252
-0.166
-0.242
-0.431
-0.660
-0.489
-0.718
-0.760
-1.036
-0.973
-0.837
-1.009
-0.859
-0.439
-0.858
-0.838
-0.188
















Table 5-5
Cumulative Abnormal Returns of Firms Divesting Assets
Unrelated versus Related Divestitures
Selected Intervals


Unrelated Divestitures


Related Divestitures


Interval


-30
-30
-30
-10
+1
+1
+11


+30
-11
-1
-1
+10
+30
+30


CAR

2.556%
-1.130
1.365
2.495
0.339
1.078
0.739


t

1.48*
-1.14
1.12
3.56***
0.48
0.89
0.75


Differences


Interval


-30
-30
-30
-10
+1
+1
+11


+30
-11
-1
-1
+10
+30
+30


Unrelated CAR Related CAR


2.556%
-1.130
1.365
2.495
0.339
1.078
0.739


-0.561%
-1.451
-0.347
1.104
-1.588
-0.047
1.541


Difference t


3.117%
0.321
1.712
1.391
1.927
1.125
-0.802


0.90
0.16
0.71
1.00
1.38*
0.47
-0.41


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


Unrelated-divestiture CARs are significantly greater


than related-divestiture CARs for the


10-day period


following announcement (1.9 percent with a t-statistic of


CAR

-0.561%
-1.451
-0.347
1.104
-1.588
-0.047
1.541


t


-0.19
-0.85
-0.17
0.91
-1.31*
-0.02
0.90








1.38). Unrelated divestitures yield cumulative abnormal

returns in excess of CARs resulting from related

divestitures for all intervals except for the period

beginning at AD +11 and ending AD +30. This lone anomaly

results from significantly positive ARs on AD +29 and AD +30

for firms divesting related assets. The evidence supports

the hypothesis that firms divesting unrelated assets enhance

shareholder wealth more than firms divesting related assets.



Financial Condition

The hypotheses offered in Chapter 3 suggest

stratification as a function of the selling firm's changes

in financial condition. Thirty-five sellers had credit

downgrades and are classified as weak. The remaining 95

firms are classified as strong.

Table 5-6 presents the average abnormal returns for

these two strata over the 61-day test period. Both

significantly positive and negative ARs occur for firms in

strong negotiating positions. Significant negative ARs

occur on day fifteen prior to announcement (AD -15) and days

one and four (AD +1, +4) after announcement. Significant

positive ARs occur on days seven, five, and one prior to

announcement (AD -7, -5, -1) and day fifteen after announce-

ment (AD +15). Strong firms yield a highly-significant AR

of 1.3 percent (t-statistic of 5.97) on the day prior to

announcement.














Table 5-6
Average Abnormal Returns of Firms Divesting Assets
Strong versus Weak Parents


Strong Parents
n = 95


Weak Parents
n = 35


t


0.041
0.235
1.266
-2.145**
1.105
-0.216
0.925
1.722**
0.097
1.510*
-0.028
-0.239
0.612
5.970***
-0.925
-1.892**
0.134
0.460
-1.565*
0.672
0.783
-0.870
1.040
-0.327
0.525
1.478*
-0.060
0.783
1.275


AR

-0.059%
0.095
-0.261
-0.175
0.258
0.560
-0.043
0.273
-0.234
0.425
0.063
0.221
-0.749
0.387
0.694
0.433
-0.459
-0.495
-0.452
0.382
-0.222
0.556
0.443
-0.563
0.381
-0.064
0.644
0.105
0.326


t


-0.161
0.259
-0.712
-0.477
0.704
1.528*
-0.117
0.745
-0.638
1.159
0.172
0.603
-2.043**
1.056
1.893**
1.181
-1.252
-1.350*
-1.233
1.042
-0.606
1.517*
1.208
-1.536*
1.039
-0.175
1.757**
0.286
0.889


*Significant at
**Significant at
***Significant at


10% level using a one-tail test.
5% level using a one-tail test.
1% level using a one-tail test.


Day

-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR

0.009%
0.051
0.275
-0.466
0.240
-0.047
0.201
0.374
0.021
0.328
-0.006
-0.052
0.133
1.297
-0.201
-0.411
0.029
0.100
-0.340
0.146
0.170
-0.189
0.226
-0.071
0.114
0.321
-0.013
0.170
0.277










Firms in weak negotiating positions that divest assets

yield both significantly negative and positive ARs around

announcement. Average abnormal returns are significantly

negative at AD -2 (0.74 percent with a t-statistic of -2.04)

and significantly positive on announcement day (0.69 percent

with a t-statistic of 1.89).

The results presented in Table 5-7 again support

stratification and provide clearer inferences. The

cumulative abnormal returns resulting from strong sellers

are generally positive and accumulate to a significant 2.7

percent over the test period (t-statistic of 1.63). In

contrast, the CARs resulting from weak sellers are generally

negative and accumulate to -1.0 percent over the test period

(t-statistic of -0.36). Strong sellers yield significantly

positive CARs on day one prior to announcement (AD -1) and

on days fifteen, twenty, twenty-five, and thirty after

announcement (AD +15, +20, +25, +30). These results are

consistent with the hypothesis that firms divesting assets

from strong negotiating positions enhance shareholder wealth

more than firms divesting assets from weakened negotiating

positions.











Table 5-7
Cumulative Abnormal Returns of Firms Divesting Assets
Strong versus Weak Parents


Strong Parents
n = 95


t

0.041
0.925
0.459
-0.411
-0.684
-0.714
-0.506
-0.144
-0.122
0.177
0.168
0.120
0.232
1.318*
1.130
0.778
0.789
0.857
0.580
0.684
0.803
0.651
0.809
0.747
0.820
1.464*
1.538*
1.648**
1.631*


Weak Parents
n = 35


CAR

-0.059%
-0.962
-1.191
-1.676
-1.756
-1.196
-1.239
-0.966
-1.200
-0.775
-0.712
-0.491
-1.240
-0.853
-0.159
0.274
-0.185
-0.680
-1.132
-0.750
-0.972
-0.416
0.027
-0.536
-0.155
-1.307
-2.285
-1.748
-1.038


t

-0.161
-1.071
-0.979
-1.143
-1.045
-0.670
-0.705
-0.538
-0.655
-0.415
-0.374
-0.253
-0.628
-0.425
-0.078
0.132
-0.089
-0.318
-0.522
-0.341
-0.436
-0.184
0.012
-0.231
-0.066
-0.526
-0.873
-0.637
-0.363


Strong Minus
Weak

CAR t


Day

-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


CAR

0.009%
0.492
0.331
-0.357
-0.681
-0.728
-0.527
-0.153
-0.132
0.196
0.190
0.138
0.271
1.568
1.367
0.956
0.985
1.085
0.745
0.891
1.061
0.872
1.098
1.027
1.141
2.157
2.387
2.679
2.767


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


0.068%
1.454
1.522
1.319
1.075
0.468
0.712
0.813
1.068
0.971
0.902
0.629
1.511
2.421
1.526
0.682
1.171
1.765
1.877
1.641
2.033
1.288
1.071
1.563
1.296
3.464
4.672
4.427
3.805


0.160
1.393*
1.077
0.774
0.550
0.234
0.348
0.389
0.501
0.447
0.407
0.279
0.658
1.037
0.643
0.283
0.478
0.710
0.745
0.642
0.784
0.490
0.402
0.580
0.475
1.199
1.535*
1.388*
1.143







Further evidence in support of this hypothesis is

provided by examining the differences between CARs of strong

sellers and CARs of weak sellers (see Table 5-7). For every

day in the test period, the accumulation of abnormal returns

for strong sellers exceeds the cumulative abnormal returns

for weak sellers. The excess accumulation of abnormal

returns for strong sellers is significant for AD +20 and

AD +25. The accumulation of excess returns for strong

sellers is 3.8 percent greater than the cumulative abnormal

returns for weak sellers, over the entire 61-day test

period.

The stratum-specific results are presented for selected

intervals in Table 5-8. For most intervals, the CARs

resulting from strong sellers exceed the CARs resulting from

weak sellers. The only exception is the 10-day period

following announcement. The positive accumulation of

abnormal returns to strong sellers is highly significant for

the 10-day period prior to announcement (2.5 percent with a

t-statistic of 3.62). The strong-seller CARs are also

significant for the pre-announcement interval of AD -30 to

AD -1 and for the interval from AD +11 to AD +30. Weak

sellers produce negative CARs for all periods except the 10-

day intervals immediately prior to and after announcement.











Table 5-8
Cumulative Abnormal Returns of Firms Divesting Assets
Strong versus Weak Parents
Selected Intervals


Strong Parents


Weak Parents


Interval

-30 to +30
-30 to -11
-30 to -1
-10 to -1
+1 to +10
+1 to +30
+11 to +30


CAR

2.767%
-0.921
1.568
2.489
-0.226
1.400
1.626


t

1.63*
-0.95
1.32*
3.62***
-0.33
1.18
1.67**


Differences


Interval

-30 to +30
-30 to -11
-30 to -1
-10 to -1
+1 to +10
+1 to +30
+11 to +30


Strong CAR Weak CAR


2.767%
-0.921
1.568
2.489
-0.226
1.400
1.626


-1.038%
-2.014
-0.853
1.161
0.004
-0.879
-0.883


Difference


3.805%
1.093
2.421
1.328
-0.230
2.279
2.509


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.


The greater ARs around announcement for strong sellers


versus weak sellers


and the consistently


accumulation of abnormal returns for strong sellers versus


weak sellers is consistent with the hypothesis


CAR

-1.038%
-2.014
-0.853
1.161
0.004
-0.879
-0.883


t


-0.36
-1.22
-0.42
1.00
0.01
-0.44
-0.54


t


1.14
0.57
1.04
0.99
-0.17
0.98
1.32*


greater


that







divesting firms enhance shareholder wealth more if they are

in a strong financial position rather than a weakened

position. Further, the evidence presented here is

consistent with the hypothesis that unrelated divestitures

enhance shareholder wealth more than related divestitures.


Second-Tier Stratification Results

Second-tier stratification results in four substrata:

1. Strong firms divesting unrelated assets,

2. Weak firms divesting unrelated assets,

3. Strong firms divesting related assets, and

4. Weak firms divesting related assets.


Strong Firms Unrelated Divestitures

The ARs and CARs resulting from unrelated divestitures

by strong firms are presented in Table 5-9. There are both

significantly positive and significantly negative average

abnormal returns during the test period. The CARs for this

strata are generally negative (and insignificant) before

announcement and are positive (significant for AD +15, +20,

+25, +30) after announcement. The divestiture of unrelated

assets by strong firms yields an accumulation of abnormal

returns of 2.9 percent over the entire test period and 3.3

percent through AD +25. This result is consistent with the

hypothesis that strong sellers which divest unrelated units

enhance shareholder wealth and is consistent with evidence

presented in the previous section.












Table 5-9
Abnormal Returns and Cumulative Abnormal Returns
Strong Firms Unrelated Divestitures
n = 65


Abnormal Returns


Cumulative Abnormal Returns


t
-0.190
-0.463
0.235
-3.435***
0.917
0.608
0.477
0.981
0.194
0.507
0.455
-0.988
1.111
6.758***
-0.903
-1.294*
0.168
0.149
-1.160
0.794
1.033
-0.619
1.421*
-0.067
0.015
1.145
-0.045
0.559
0.313


CAR
-0.051%
-0.055
-0.182
-1.048
-1.002
-0.839
-0.711
-0.448
-0.396
-0.260
-0.138
-0.403
-0.105
1.707
1.465
1.118
1.163
1.203
0.892
1.105
1.382
1.216
1.597
1.579
1.583
2.373
2.746
3.317
2.856


t
-0.190
-0.084
-0.205
-0.977
-0.815
-0.667
-0.553
-0.341
-0.295
-0.190
-0.099
-0.284
-0.073
1.162
0.981
0.737
0.755
0.769
0.562
0.687
0.847
0.736
0.954
0.931
0.922
1.305*
1.434*
1.653**
1.364*


*Significant at
**Significant at
***Significant at


10% level using a one-tail test.
5% level using a one-tail test.
1% level using a one-tail test.


Day
-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR
-0.051%
-0.117
0.063
-0.921
0.246
0.163
0.128
0.263
0.052
0.136
0.122
-0.265
0.298
1.812
-0.242
-0.347
0.045
0.040
-0.311
0.213
0.277
-0.166
0.381
-0.018
0.004
0.307
-0.012
0.150
0.084








Weak Firms Unrelated Divestitures

The resulting abnormal returns and cumulative abnormal

returns from unrelated divestitures by weak sellers are

listed in Table 5-10. A highly-significant abnormal return

of 0.86 percent (t-statistic of 2.45) at announcement

indicates a positive shareholder response. However, there

are both significantly positive and significantly negative

ARs over the test period. Unrelated divestitures by weak

parents accumulate abnormal returns of 1.9 percent over the

test period. This indicates that sellers in weak

negotiating positions may still enhance shareholder wealth

by selling unrelated assets.


Strong Firms Related Divestitures.

Table 5-11 presents the ARs and CARs resulting from

related divestitures by strong sellers. The accumulation of

abnormal returns are generally positive and insignificant.

Abnormal returns accumulate to 2.6 percent over the 61-day

test period but are not statistically significant.












Table 5-10
Abnormal Returns and Cumulative Abnormal Returns
Weak Firms Unrelated Divestitures
n = 31


Abnormal Returns


Cumulative Abnormal Returns


AR
-0.004%
-0.093
-0.084
-0.192
0.426
0.629
-0.149
0.191
-0.243
0.585
-0.184
0.328
-0.469
0.424
0.857
0.414
0.046
-0.734
-0.314
0.561
0.420
0.268
0.073
-0.333
0.404
-0.210
0.155
-0.096
0.071


t
-0.011
-0.266
-0.240
-0.550
1.219
1.800**
-0.426
0.547
-0.695
1.674**
-0.527
0.939
-1.342*
1.213
2.453**
1.185
0.132
-2.101**
-0.899
1.606*
1.202
0.767
0.209
-0.953
1.156
-0.601
0.444
-0.275
0.203


CAR
-0.004%
-0.182
-0.224
-0.837
-0.457
0.172
0.023
0.214
-0.029
0.556
0.372
0.700
0.231
0.655
1.512
1.926
1.972
1.238
0.924
1.485
1.905
2.173
2.246
1.913
2.317
1.411
1.550
1.914
1.944


*Significant
**Significant
***Significant


at 10% level using a one-tail test.
at 5% level using a one-tail test.
at 1% level using a one-tail test.


Day
-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


t
-0.011
-0.213
-0.193
-0.599
-0.285
0.105
0.014
0.125
-0.017
0.312
0.205
0.379
0.123
0.342
0.777
0.974
0.982
0.608
0.447
0.708
0.896
1.009
1.029
0.866
1.036
0.595
0.621
0.732
0.712













Table 5-11
Abnormal Returns and Cumulative Abnormal Returns
Strong Firms Related Divestitures
n = 30


Abnormal Returns


Cumulative


Abnormal Returns


t

0.373
1.128
1.993**
1.414*
0.620
-1.365*
0.976
1.675**
-0.122
2.020**
-0.772
1.117
-0.606
0.495
-0.307
-1.493*
-0.011
0.623
-1.096
0.000
-0.169
-0.653
-0.294
-0.508
0.960
0.954
-0.044
0.582
1.890**


CAR

0.137%
1.676
1.445
1.146
0.017
-0.485
-0.126
0.490
0.445
1.188
0.904
1.315
1.092
1.274
1.161
0.612
0.608
0.837
0.434
0.434
0.372
0.132
0.024
-0.163
0.190
1.693
1.615
1.303
2.584


t

0.372
1.860**
1.184
0.779
0.010
-0.281
-0.071
0.272
0.242
0.633
0.473
0.676
0.551
0.632
0.567
0.294
0.288
0.390
0.199
0.197
0.166
0.058
0.010
-0.070
0.081
0.679
0.615
0.473
0.899


*Significant at
**Significant at
***Significant at


10% level using a one-tail test.
5% level using a one-tail test.
1% level using a one-tail test.


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR

0.137%
0.415
0.733
0.520
0.228
-0.502
0.359
0.616
-0.045
0.743
-0.284
0.411
-0.223
0.182
-0.113
-0.549
-0.004
0.229
-0.403
0.000
-0.062
-0.240
-0.108
-0.187
0.353
0.351
-0.016
0.214
0.695









Weak Firms Related Divestitures

The resulting ARs and CARs from this stratum are

presented in Table 5-12. There are both significantly

positive and significantly negative abnormal returns.

Cumulative abnormal returns are generally positive and

insignificant. Strong inferences cannot be made from these

results due to the small sample size.


Summary

Table 5-13 provides a summary of the stratum-specific

accumulation of abnormal returns over the 61-day test

period. Unrelated divestitures and divestitures from strong

parent firms yield significantly-positive abnormal returns

over the test period. Further, when both these factors are

present in a divestiture, significantly-positive abnormal

returns are earned. Negative (insignificant) abnormal

returns are accumulated for related divestitures and for

divestitures from weak parent firms.












Table 5-12
Abnormal Returns and Cumulative Abnormal Returns
Weak Firms Related Divestitures
n = 4


Abnormal Returns


Cumulative Abnormal Returns


t

0.645
1.473*
0.436
0.897
0.870
-1.381*
1.663**
0.618
-0.818
0.394
-1.782**
1.846**
1.016
-0.538
-0.883
-1.953**
-0.073
0.177
-1.476*
-0.378
-0.286
-0.052
1.110
-0.058
0.306
0.207
1.341*
0.917
2.871***


CAR

0.449%
3.594
2.369
1.606
0.818
-0.143
1.014
1.444
0.875
1.149
-0.091
1.193
1.900
1.526
0.912
-0.447
-0.498
-0.375
-1.402
-1.665
-1.864
-1.900
-1.128
-1.168
-0.955
2.220
3.740
3.156
6.983


t

0.645
2.109**
1.027
0.577
0.257
-0.044
0.304
0.424
0.252
0.324
-0.025
0.324
0.507
0.400
0.235
-0.114
-0.125
-0.092
-0.341
-0.399
-0.440
-0.443
-0.260
-0.265
-0.214
0.470
0.753
0.606
1.285*


*Significant at
**Significant at
***Significant at


10% level using a one-tail test.
5% level using a one-tail test.
1% level using a one-tail test.


-30
-25
-20
-15
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+15
+20
+25
+30


AR

0.449%
1.025
0.303
0.624
0.605
-0.961
1.157
0.430
-0.569
0.274
-1.240
1.284
0.707
-0.374
-0.614
-1.359
-0.051
0.123
-1.027
-0.263
-0.199
-0.036
0.772
-0.040
0.213
0.144
0.933
0.638
1.997












Table 5-13
Divestitures
Stratum-Specific CARs over 61-Day Test Period


Stratum

Full
Unrelated
Related
Strong
Weak
Unrelated/Strong
Unrelated/Weak
Related/Strong
Related/Weak


Sample Size CAR t

130 1.743% 1.18
96 2.556 1.48*
34 -0.561 -0.19
95 2.767 1.63*
35 -1.038 -0.36
65 2.856 1.36*
31 1.944 0.71
30 2.584 0.90
4 6.983 1.29*


*Significant at 10% level using a one-tail test.
**Significant at 5% level using a one-tail test.
***Significant at 1% level using a one-tail test.





When analyzed homogeneously, divestiture announcements

yield slightly positive abnormal returns to the selling

firm's shareholders. However, the evidence is clear that

some divestitures enhance shareholder wealth and others may

reduce shareholder wealth. The selling off of unrelated

units results in a positive shareholder response. Parent

firms in strong negotiating positions earn more in

divestiture than sellers in weak negotiating positions.

Regardless of the sellers negotiating position, the

divesting of unrelated assets enhances shareholder wealth

more than the selling of related assets.















CHAPTER SIX
SUMMARY, CONCLUSIONS, AND FUTURE RESEARCH


Summary and Conclusions

When studied as a single genus, the shareholder wealth

effects of firms acquiring divested assets are unclear.

Dependent upon the characteristics of the involved assets,

some firms earn positive economic rents by acquiring

divested units. Other firms decrease shareholder wealth by

acquiring divested assets. Thus, it is not surprising that

homogeneic treatment of firms acquiring divested assets

yields abnormal shareholder returns that are not signifi-

cantly different from zero.

Similarity of these results with merger study results

is less a function of the type of transaction than a

function of the type of empirical treatment. Studying a

diversity of events under the assumption that the economic

consequences of such events are identical will lead to false

conclusions. It has been concluded in merger studies that

acquiring firms do not gain. However, some acquiring firms

earn positive economic rents and other acquiring firms do

not enhance shareholder wealth.

Unambiguous conclusions cannot be reached by analyzing

the acquisition of divested assets as one economic group.








Comparisons of the full sample results of this study with

other divestiture-acquisition studies are inconclusive.

Sample stratification based upon economic characteristics is

required to obtain conclusive evidence.

Stratification based upon the relatedness of the

divested asset to the acquiring firm yields clear results.

The acquisition of related divested assets enhances the

shareholder wealth of acquiring firms. The acquisition of

unrelated divested assets affects shareholder wealth

negatively. The cumulative abnormal returns resulting from

related acquisitions are significantly greater than the

cumulative abnormal returns resulting from unrelated

acquisitions. Rational shareholders prefer the purchase of

related divested assets to the purchase of unrelated

divested assets.

The acquisition of unrelated assets may be motivated by

managerial preferences that differ from shareholder

preferences. .Firms that acquire unrelated assets tend to

have lower proportions of insider equity ownership than

firms that acquire related assets. Thus, there is an

economically significant agency conflict in decisions to

acquire divested assets.

The agency problem associated with divestiture

acquisitions are less likely to be resolved by natural

market forces than problems associated with other mergers.

The merger market, particularly when tender offers are made,







is more competitive than the divestiture market. Divesti-

tures tend to be friendly deals transacted directly between

seller management and buyer management. Tender offers are

made through capital markets, which may lead to a more

competitive bidding process. This allows the market to more

accurately evaluate the manager's decision.

The lack of external forces acting in the divestiture

decision gives management greater flexibility in the

acquisition decision. An owner-manager, or a manager whose

wealth is commensurate with share performance, is likely to

acquire related divested assets. Nonowner-managers prefer

to acquire unrelated divested assets.

The results are not as clear when stratification is

based on the inferred negotiating position of the seller.

Purchases from weak parents accumulate greater abnormal

returns than acquiring from strong parents. However, the

difference is not statistically significant and may be

interval-specific. There is some evidence that the seller's

negotiating posture has economic consequences but

measurement of this variable is not straightforward.

Measurement and small-sample problems limit confidence

in the results obtained from second-tier stratification.

The accumulated abnormal returns resulting from the

acquisition of related divested assets from weak parents

exceed the cumulative abnormal returns resulting from the

acquisition of unrelated divested assets from strong

parents. The difference accumulates to 3.5 percent but is









not statistically significant. The sample size (11) of

related assets acquired from weak parents reduces

significance.

The acquisition of related assets from strong parents

yields highly-significant positive cumulative abnormal

returns. It is clear that the relatedness variable is

dominant. The results from comparing related acquisitions

from weak parents with related acquisitions from strong

parents are confounding. Again, measurement problems and

small samples yield ambiguous results.

As with the study of acquisitions, the homogeneic

analysis of divesting firms results in ambiguous

conclusions. Some divestitures earn positive economic rents

and other divestitures may reduce shareholder wealth.

Again, it is not surprising that the study of divestitures

as a single class yields cumulative abnormal returns that

are not significantly different from zero. Sample

stratification based upon specific characteristics is

required to determine the economic impact of divestitures.

The divesting of assets unrelated to a firm's primary

line of business results in significantly positive gains to

shareholders. The divesting of related assets may have a

negative impact on shareholder wealth. A divesting firm

earns more for its shareholders if it is in strong financial

condition than if the firm's financial condition has

weakened as evidenced by a recent credit downgrade.