Download Corporate Mergers and the Co-Insurance of Corporate Debt and more Study notes Corporate Finance in PDF only on Docsity!
THE JOURNAL OF FINANCE • VOL, XXXII. NO, 2 MAY 1977
CORPORATE MERGERS AND THE CO-INSURANCE OF
CORPORATE DEBT
E. HAN KIM AND JOHN J. MCCONNELL*
WHILE NUMEROUS STUDIES have been devoted to examining the returns to the
stockholders of merging firms, the same detailed analysis has not been extended to
another group of emminently interested security owners—namely the bondholders
of these same firms. Indeed, there exists a fundamental unresolved controversy
concerning the impact of corporate merger on the value of the merging firms'
bonds, and, by implication, the value of their common stock. The controversy
revolves around the notion of a "co-insurance" effect for corporate debt and the
wealth-transfers thereby engendered.
The idea of a co-insurance effect for corporate debt was first advanced by
Lewellen (11). He argued that the joining-together of two or more firms whose
earnings streams were less-than-perfectly correlated would reduce the risk of
default of the merged firms (i.e., the co-insurance effect) and thereby increase the
"debt capacity" or "borrowing ability" of the combined enterprise. He concluded
that the increased total borrowing capacity of the resulting firm, in combination
with the well-known effect of tax-deductible interest payments, provided an
economic incentive for shareholder-wealth-maximizing firms to engage in merger.
However, Lewellen's thesis was incomplete because he failed to examine carefully
the impact of the co-insurance effect on the value of the merging firm's already
outstanding debt.
Higgins and Schall (6) and Galai and Masulis (5) extended the analysis to show
that the co-insurance effect would lead to an increase in the market value of the
merging firms' debt and a concomitant decline in the market value of their equity.
Thus, the net financial result of non-synergistic mergers would be a wealth-transfer
from stockholders to debtholders. They concluded that unless firms can neutralize
this wealth-transfer they should not engage in merger. However, if firms are either
controlled by stockholders or if managers at least seek to maintain shareholders'
wealth—let us define these more generally as shareholder-wealth-/;ro/ec//«^ firms—
we would expect to observe that merging firms do take steps to neutralize this
wealth-transfer.
This paper is a theoretical and empirical examination of corporate merger and
the co-insurance of corporate debt. The theoretical section uses a cash-flow
analysis to re-examine the co-insurance effect. The comparative advantages of the
theoretical framework used here is that the valuation consequences of the co-
insurance effect are provided with no distributional assumptions and without
- Ohio State University and Purdue University, respectively. We wish to thank A. Chen, T, Langetieg,
G. Racette, L. Schall, and J. Warner for their helpful comments and C. Chang and G, Rhee for
assistance in data collection, A special thanks is due to S. Kon for his critical reviews of several earlier
versions of the paper. Financial support for this study was generously provided by The Ohio State
University Research Foundation,
349
350 The Journal of Finance
assuming the validity of any specific capital market pricing mechanism. The
empirical section provides a comprehensive examination of the returns to the
bondholders of merging firms and the subsequent actions taken by merging firms
to neutralize the potential wealth-transfer. The empirical results are important for
at least three reasons: First, they represent the first effort to determine the impact
of corporate merger on the market value of the merging firms' debt. Second, they provide some direct evidence toward resolving the controversy raised by the co-insurance effect. Third, they provide evidence on the effectiveness of stock- holders in protecting their own interests through the media of appointed managers and delegated authority. In this regard, the results provide some indirect evidence on the theory of agency.'
I. THE CO-IINSURANCE EFFECT AND SHAREHOLDER-WEALTH-PROTECTING FIRMS
This section re-examines the co-insurance effect within a perfect capital market framework in which the values of all securities depend only upon the distribution of cash flows which they confer upon their owners and in which bankruptcy is assumed to be costless. Bankruptcy is defined as occurring whenever the face value of the firm's debt obligation exceeds the terminal market value of its assets. When bankruptcy occurs, ownership of the firm is transferred to debtholders, who may then decide either to liquidate it or to continue its operations. In this setting the end-of-period gross dollar return to the debtholders of firm /, f,, may be expressed as
where y, is the gross dollar amount promised to debtholders, and A',, is the firm's end-of-period gross dollar return after paying all non-capital factors of production. Suppose that firms A and B combine in a nonsynergistic merger (i.e., A'^ = A"^ + Xg) to form firm C. The total pre-merger cash flow distribution owned by the debtholders of firms A and B may be written as
•r V" "^ V V "^ V 11 .\ A ^ I As ^ n ^ * ft .f X,>Y,,X,<Yg ^2)
if X^ < y^, Xg < Yg.
If the merged firm leaves unchanged the debt obligations of the combined pre- merger firms, then Y^^Y^-V Yg, and from (1), the cash flow distribution owned by the merged firm's creditors is
- \YA+Yg if X^--Xg>Y^-^Yg ^.. Yc=i / / / / /. (3) ^A^-t-Ag U y^A^ ^B^ ' A^ ' B-
1, For an extensive discussion of the economic theory of agency, see Jensen and Meckling (9).
352 The Journal of Finance
of such transactions effectively reduce shareholder wealth. A less expensive and
more realistic option open to these firms is to increase their use of financial
leverage to the point where the post-merger default risk of the previously outstand-
ing debt is increased sufficiently to negate the co-insurance effect and to cancel any
wealth-transfers from equity-holders to debt-holders.
Therefore: if (1) corporate mergers produce a co-insurance effect and (2)
merging firms are shareholder-wealth-protecting firms, then we would observe that
(1) bondholders of merging firms do not earn any abnormal returns around the
time of merger, and (2) merged firms increase their use of financial leverage
relative to the participating firms' combined pre-merger financial leverage. The
following sections provide empirical investigations of the returns to the bondhol-
ders of merging firms and the use of financial leverage by these firms.
II. EMPIRICAL TESTS OF BONDHOLDER RETURNS
This section describes two separate empirical methodologies and the data used to
examine the returns to the bondholders of merging firms. The first test employs a
paired-comparisons procedure, while the second uses a two index market model.
A. Paired-Comparisons Procedure
In the paired-comparisons test one bond issued by a non-merging company was
selected to match each of the bonds issued by a sample of merging firms on four
observable risk-return characteristics. These characteristics were: (1) bond rating as
given by Standard and Poor's Bond Guide; (2) term to maturity; (3) coupon interest
rate; and (4) industrial bonds were paired with industrial bonds and utility bonds
were paired with utility bonds. The theoretical and empirical justification for
matching the bonds on these characteristics is contained in Kim, McConnell, and
Greenwood (10). If the underlying risk-return characteristics of the two sets of
bonds are identical, except that the issuers of one set of bonds engaged in merger
while the issuers of the other did not, then any difference in returns between the
two groups may be attributable to the merger. The paired-comparisons tests
examines the differences in rates of return between the two groups of bonds over
the period from 24 months before to 23 months after the mergers in question.
Using the month of merger as month 0, the ki\i month following the merger as
month -I- k, and the A:th month before the merger as — k, monthly rates of return,
including both capital gains and accrued coupon interest payments, were computed
for each bond over the period from month —24 to month -1-23.^
The monthly rate of return on each matching bond /, /?,*, was subtracted from
the monthly return earned in the same calendar month on its corollary merger
bond /, /?,^, and the average difference (AD) in rates of return across pairs of bonds
was computed for each relative month k as:
M
/M, A:=-24,...,-t-23, (8)
where M is the number of merger bonds in the sample. The measure b,^ may be
3. For a description of the way in which the monthly rates of retum were calculated, see Bildersee (2, p. 509),
Corporate Mergers and the Co-insurance of Corporate Debt 353
thought of as the average "abnormal" return to the bondholders of the merging
firms. The average differences in rates of return between the two groups of bonds
were then summed over the period from month - 24 through month T to compute
the cumulative average differences (CAD) in month T as
bT= 1 K r = - 2 4 ,... , - h 2 3. (9) *** = - 2 4**
If the holders of both sets of bonds earned equivalent returns we would expect the AD's to be randomly distributed around a mean of zero. If, however, the
bondholders of merging firms earned windfall gains, we would expect the AD's to
be positive more often than negative (especially around the time of merger); and
we would expect to observe an upward or positive trend in the CAD.
B. Data The sample of merging firms was selected from the FTC's Statistical Report on
Mergers and Acquisitions (16). In order to be included in the sample, a firm had to
satisfy five criteria:
- The firm engaged in a major merger between January 1, 1960 and December
31, 1973.'' A merger was classified as major if the book value of the assets of the
smaller firm was greater than 10% of the book value of the assets of the larger firm.
- The merger was complete (i.e., partial mergers were excluded).^
- The merging company had long-term publicly-traded debenture bonds out-
standing for at least twenty-four months before the merger and the same bonds
were outstanding for at least twenty-four months after the merger.*
- The firm did not engage in any other non-minor^ merger during the forty-
eight month interval surrounding the month of the merger in question.
- The merger was classified as "conglomerate" by the FTC*
Out of a total of 2,286 firms involved in a merger or acquistion, we found only 39
firms which had 44 bonds outstanding that satisfied these criteria. A comprehen-
sive description of the companies and the characteristics of the sample and control
bonds is contained in the Appendix.
4. The major merger requirement was imposed to exclude small mergers wherein any co-insurance
effect might be overwhelmed by other phenomenon,
5. To be a complete merger the common stock of one company had to be purchased entirely by the
other company and the debts of both companies had to be assumed by the merged firm. We might note
in this regard that a substantial proportion of the mergers reported by the FTC were partial mergers,
wherein the acquiring company controlled 80% to 90% of the common stock of the acquired company.
One explanation for this phenomenon is that all earnings synergies could be achieved through partial
merger, but because both firms continued as separate entities each was liable for its own debts only,
thereby avoiding any possible co-insurance generated wealth-transfers.
6. We excluded convertible bonds which represent both a debt claim and an equity claim.
7. To distinguish from a major merger, a non-minor merger was defined as one wherein the book value
of the assets of the smaller firm was greater than 5% of the book value of the assets of the larger firm.
8. This requirement was imposed to isolate the "co-insurance" effect from the possible synergistic
effect of merger. Even in the absence of any "co-insurance", a favorable synergistic effect may increase
operating profitability and consequently the market value of the merging companies' debt. See footnote
Corporate Mergers and the Co-insurance of Corporate Debt 355
C. Two-Index Market Model
The two-index market model was used to account for a bond's systematic risk more explicitly. We included a bond index as well as a stock index as the independent variables (1) to hedge against the possibility of a segmented capital market and (2) to alleviate the omitted asset problem.''' The following version of the market model was used:
SrU,) O;E{B,U,) O;
0 if t¥=k
where /?„ is the return on security /; 5, is the return on a stock market index; B, is the return on a corporate bond index;" [/„ is the stochastic disturbance term for security i; and a,, y8,,, and ^Sj,, are coefficients to be estimated; and the subscript t represents the time period. The estimates of the coefficients in equation (10) were derived from a moving time series OLS regression. The coefficients were reestimated for each bond in each month using the previous 12 monthly returns (i.e., month / - 1 2 to month t-l). This procedure yielded 36 market equations for each bond. The reestimation procedure was used to account for instabilities in the relationship between the security return and the market indexes. Each estimated equation was then used to generate an estimate of the conditional expected return on the security in month t. The difference between the actual return, /?,,, on each bond and its estimated conditional mean return was then computed as:
where a,, ^,,, and /5,2 were estimated using data from periods t-l2 to t-. The difference, e,,, represents the residual or abnormal return to bond / in month t. These residuals were averaged across firms to compute the average residual (AR) in each relative month k as
M p AR - y —
and then summed to yield the cumulative average residual (CAR) from month - 12 to month K as
CAR^= 2 ,
k=-\l
The stock market rate of return used is a value-weighted index of NYSE stocks
- See Mayers (13, pp, 230-232),
- Since the stock and bond indices are correlated, the bond index was orthogonalized by regressing it against the stock market index. The residuals of the preliminary regression were then used as the second index in estimating the market equations for the individual securities.
356 The Journal of Finance
compiled by Myron Scholes, while the bond market rate of return is an equally-
weighted index of high-quality corporate bonds created by Ibbotson and Sin-
quefield (7).'*
III. RESULTS ON BOND RETURNS
A. Average Differences (AD'5) and Cumulative Average Differences (CAD'5)
The AD's and CAD's for the paired-comparison test are presented in Table 1.
The AD's (Column 2) were almost evenly split between positive and negative
observations with 25 and 23, respectively. Comparison of the before and after
merger period shows 13 positive AD's before the merger and 12 after. The largest
positive AD (1.5%) did occur one month before the merger, but its /-statistic
(Column 3) was only .35. Examination of the remainder of the /-values in Column
3 shows that there was no single relative month in which the bondholders of
merging firms reaped statistically significant abnormal gains.
Looking at consecutive AD's we see that the longest run of positive observations
was only 3 months in duration. However, it is interesting to note that in the
12-month interval around the month of merger, beginning in month —5, nine out
of twelve AD's were positive. This suggests that the bondholders of merging firms
may have achieved some positive abnormal returns around the time of merger.
Looking at the CAD (Column 4) we see that it was positive in month - 24 and
remained positive until month - 6 when it first became negative. The run of
positive AD's then shows up in the CAD which reached a level of -I- 2.0% in month
- 1 and remained at about the same level until month -1-6 after which it began to
decline until becoming negative in month +16. Column 5 shows that at no time
was the CAD statistically significantly different from zero.'^''^
B. Average Residuals (AR's) and Cumulative Average Residuals (CAR's)
Table 2 contains the AR's (Column 2) and the CAR's (Column 4) computed
from the estimated two index market model." There is a certain consistency
between these and the paired-comparisons results. Here, as with the AD's, the one
month before the merger exhibits a large positive abnormal return (in fact, it is the
second largest observed). Further, the AR's just before and just after the merger are
predominantly positive. As with the CAD, the CAR reaches its highest level right
after the merger month, remains positive for several months, and then becomes
16, Both indexes are available from the Center for Research in Security Prices at the University of
Chicago,
17, To compute this ^-statistic, the CAD at each month was divided by the cror^-sectional standard
deviation of the individual bonds' cumulative differences at that month. That is for each month T, bj- in
(9) was divided by the standard deviation of 2! - -24 (^i* ~ ^i*)-
18, As a second test on the paired comparisons procedure, we computed the average price of the
bonds in each group in each relative month. In no month was the difference between the average prices
of the two groups statistically significant,
19, The average i?-squared value for the 1584 estimated market equations (36 equations per bond for
44 bonds) was approximately ,20,
AVERAGE RESIDUALS AND CUMULATIVE AVERAGE RESIDUALS COMPUTED USING 12-MONTH MOVING MARKET EQUATIONS Relative
10 11 12 13 14 15 16 17 18 19 20 21 22 23 Average
r-Values
Average Cumulative Average r-Values Cumulative Average
negative again. In no case, however, are the AR's or the CAR's statistically
significantly different from zero.^° There is, however, one troubling aspect to these results. Over the 36-month interval the AR's were predominantly negative, exhibiting 21 negative observations out of a total of 36. In fact, 7 out of a total of 15 positive AR's occurred in the
20. The r-statistics for CARj^ were obtained by dividing CAR^^^ by the standard deviation of
2 ^ - -i2^i- Thus, although the /-values in Columns 3 and 5 may contain an upward bias [Black, Jensen,*
- TABLE - MONTH + AVERAGE DIFFERENCES AND CUMULATIVE AVERAGE DIFFERENCES FROM MONTH - 24 TO
- 0, Difference
- -0, - 0,
- -0,
- -0,
- -0, - 0, - 0, - 0,
- -0,
- -0, - 0, - 0.
- -0. - 0,
- -0. - 0.
- -0.
- -0. - 0. - 0. - 0,
- -0, - 0.
- -0, - 0, - 0,
- -0. - 0, - 0. - 0.
- -0.
- -0,
- -0.
- -0.
- -0.
- -0.
- -0.
- -0.
- -0.
- -0,
- 0, - 0, Difference - -0. - 0. - -0. - -0. - -0. - 0. - 0. - 0. - -0. - -0. - 0. - 0. - -0. - 0. - -0, - 0, - -0. - -0, - 0, - 0, - 0, - -0. - 0. - -0. - 0. - 0. - -0. - 0. - 0. - 0, - -0, - 0. - -0. - -0. - -0. - -0, - 0, - 0. - -0, - -0. - -0, - 0, - 0, - -0. - 0, - -0. - 0, - 0. Difference - 0. - 0. - 0. - 0. - 0. - 0. - 0. - 0. - 0. - 0. - 0. - 0. - 0, - 0. - 0, - 0. - 0, - -0. - 0, - 0, - 0, - 0, - 0, - 0. - 0, - 0. - 0. - 0. - 0. - 0. - 0, - 0. - 0. - 0, - 0, - 0. - 0. - 0. - 0, - -0. - -0. - 0. - 0. - -0, - 0. - 0. - 0. - 0, CAD - 0, - 0, - 0, - 0, - 0, - 0, - 0. - 0. - 0, - 0. - 0. - 0. - 0, - 0, - 0, - 0. - 0. - -0. - 0, - 0. - 0, - 0, - 0, - 0, - 0, - 0, - 0. - 0. - 0. - 0, - 0. - 0, - 0, - 0, - 0, - 0, - 0. - 0, - 0. - -0, - -0. - 0. - 0. - -0, - 0, - 0, - 0, - TABLE 358 The Journal of Finance
- -0, Residuals - -0, - 0. - -0. - 0, - -0, - -0, - 0, - -0, - 0. - 0. - 0, - -0. - 0, - 0. - -0. - -0, - -0. - 0. - -0. - 0, - -0, - -0, - -0, - -0, - -0, - -0. - -0, - -0. - 0. - -0, - 0.
- -0, - 0, - 0. - 0, - -0, Residuals - -0, - 0, - -0. - 0. - -0. - -0, - 0, - -0, - 0. - 0. - 0, - -0. - 0. - 0, - -0, - -0. - -0, - 0. - -0, - 0. - -0. - -0. - -0. - -0, - -0, - -0. - -0. - -0, - 0, - -0. - 0, - -0. - 0. - 0. - 0, - -0. Residuals - -0. - -0. - -0, - -0, - -0. - -0. - -0. - -0, - -0. - -0. - 0, - 0, - 0. - 0. - 0. - 0. - 0. - 0. - -0, - 0. - -0. - -0. - -0. - -0. - -0, - -0. - -0, - -0. - -0. - -0, - -0. - -0. - -0, - -0. - -0. - -0, Residuals - -0. - -0, - -0, - -0, - -0. - -0. - -0. - -0, - -0, - -0. - 0. - 0, - 0. - 0, - 0, - 0, - 0, - 0, - -0, - 0, - -0, - -0, - -0. - -0. - -0. - -0, - -0. - -0, - -0. - -0. - -0. - -0, - -0, - -0, - -0.
Corporate Mergers and the Co-insurance of Corporate Debt 359
12-month period beginning in month - 5 and ending in month +6. Furthermore,
the CAR was negative in all periods except the 10-month interval right around the
merger, beginning in month - 1, and ending in month +9.
One possible explanation for this phenomenon is that the estimated market
equations tend to provide upward biased estimates of the conditional mean return.
Since the residuals were obtained by subtracting the conditional mean return from
the actual return, an overestimate of the conditional mean will result in an
underestimate of the "true" value of the residual. Such an overestimate of the
conditional mean may occur if the relationship between the bond return and the
market indices (i.e., the slope of the market equation) is shifting systematically over
time as the bond approaches maturity. For example, while the maturity of the bond
index remains relatively constant over time as bonds are added and deleted and the
stock index has a constant infinite maturity, the maturities of the sample bonds
decrease systematically. For an index with a constant maturity, the effect of a given
shift in term structure of interest rate may be constant over time. However, for a
bond with a decreasing maturity, a given shift in term structure of interest rate will
cause a systematically declining reaction in the bond price as the term to maturity
TABLE 3
AVERAGE RESIDUALS AND CUMULATIVE AVERAGE RESIDUALS COMPUTED USING MARKET EQUATION ESTIMATED WITH MONTHS - 2 4 TO - 13 AND + 12 TO +
Relative Month
10 11
Average Residual -0, -0, 0, -0,
-0. -0, 0,
-0. -0. 0, -0, 0, 0, -0,
-0, 0, -0. 0, 0, -0,
Cumulative Average Residual -0. -0. -0, -0. -0. -0, -0. -0.
-0, -0, 0, 0, 0,
0, -0, 0, 0, -0. -0.
Corporate Mergers and the Co-insurance of Corporate Debt 361
companies, whereas the creditors of smaller firms reap substantial benefits when they join with larger companies. As it turns out, splitting the sample on these characteristics showed no statisti- cally significant results in either the AD's, the CAD's, the AR's, or the CAR's. In short, we found no evidence of systematic abnormal returns to the bondholders of any category of merger.
V. THE USE OF LEVERAGE BY MERGING FIRMS To examine the issue of whether or not merging firms tend to increase their financial leverage following the merger we computed three leverage ratios for the
yth merged firm at time t:
where BLDj, is the book value of long-term debt; 5 , is the average market value of common plus preferred stock; BTDj^ is the book value of total debt (long-term plus short-term); and BTAj, is the book value of total assets. The combined data of both the acquiring and the acquired firms (collected from Moody's Industrials) were used to calculate each ratio for each of the two years before the merger (/ = - 1, - 2) and each of the two years after the merger ( / = 1,2). The year of merger was omitted from the calculation. For each merger we computed the average of the before and the after merger leverage ratios, and the changes in financial leverage as follows:
^LJ = {Ll, + Ll:^/{Ll^ + Ll_:^,_ for i=LM,TM,B,
which measures the relative use of leverage after merger vs. the relative use of leverage before merger. A AL' greater than 1.0 shows an increase in the combined leverage ratio after the merger. Table 4 shows the average of AL^'s along with the number of mergers that showed an increase vs. a decrease on each measure.
TABLE 4
MEASURES OF CHANGE IN THE USE OF FINANCIAL LEVERAGE
Measure
AL/'^ tiL™ A L /
Mean
1, 1, 1,
Number Increase
20 18 26
Number Decrease
11 13
The mean of each of the ratios across mergers is greater than 1.0. The number of firms showing an increase in AL'''^ is 20 out of the 31 mergers for which complete data were available;^" for AL™ it is 18 out of 31, and for AL it is 26 out of 31.
24, The sample size drops from 39 to 31 companies for several reasons. Two companies were lost because we have two cases in which both the acquiring and acquired companies are in the sample. Since we used aggregate data, these were counted as one leverage observation. The remaining six firms were lost because of inadequate data for one or both firms.
362 The Journal of Finance
Assuming an equal probability of increases and decreases, and using a binomial
test [Conover (1971), p. 97], the number of increases in the first and the third
measures are significant at the 5% and .5% level, respectively. The overall results
support our hypothesis that merging firms tend to make greater use of leverage
after merger than the combination of individual firms did before the merger.^'
VI. COMMENTS AND CONCLUSIONS
As was discussed earlier, we are concerned with two issues. The first is whether or
not the bondholders of merging firms earn abnormal positive returns. The second
is the use of financial leverage by merging companies. Using two separate statisti-
cal procedures we could not reject the null hypothesis that the bondholders of
merging firms do not earn abnormal positive returns around the time of merger.
Thus, for this group of corporate mergers there was no statistically significant
transfer of wealth from stockholders to bondholders. Using three different mea-
sures of financial leverage, we found that merged firms do make greater use of
financial leverage after the merger than the combination of independent firms did
before the merger. In the absence of any co-insurance effect we would expect this
increase in financial leverage to generate windfall losses for the bondholders of the
merging firms.^* Since these bondholders did not suffer abnormal negative returns,
we have evidence that a co-insurance effect did take place. While it is not possible
to prove empirically a cause and effect relationship, these results are consistent
with the argument that (1) a co-insurance effect did exist, and (2) the wealth-
transfers to bondholders that would have been generated were negated by the
increased use of debt financing. The evidence also is consistent with the notion that
managers act in the best interests of stockholders when a conflict arises between
stockholders and bondholders.
There may, of course, be other reasons for our failure to discover any statistically
significant abnormal returns to the bondholders of merging companies. It may be
that a significant wealth-transfer existed, but our tests were not powerful enough to
detect it. However, given the assortment of tests employed that seems unlikely. It
may be that we are observing the wrong set of firms, in the sense that managers (or
stockholders) only engage in mergers with firms that will generate little co-
insurance or that they take other steps to circumvent the wealth-transfer. For
example, they may engage in partial mergers (see footnote 5) or they may call the
bonds." Having stated these caveats, however, we continue to feel confident in
concluding that stockholders need not be overly concerned with the potential
wealth-transfers that may be generated by corporate mergers. The evidence indi-
cates that whatever gains mergers generate, they are not reaped by bondholders.
- Melicher and Rush (14) provide further evidence on this report. They showed that firms engaged in conglomerate type mergers (i,e., mergers which are likely to generate a large co-insurance effect) make greater use of debt financing than firms involved in other types of mergers,
- See Kim, McConnell, and Greenwood (10) for further explication of this point,
- As an aside, two firms which qualified for the sample on all other measures were excluded because they called their bonds right around the time of merger.
(^364) The Journal of Finance
X -
z
I §
u
\
II
3 •3^ i f^?^ ^
a I
.5 O Z J <
(^00) n <, COaa us < to
S h
0 t
z a
I "
i l
< < s
< < a
oa ooa <
f n v - > o * o o
iS 2
•S 3 • Vi Vl Ui
S S t
S 3 H^ "2^ -ri^ <5^ -
•S c c c i: <3 a a a a
t t
P z
.
a
d S
5 <S S 0 •o s "^ S
? 8 _g "
I I
.3 s
I I
Co, ical £
rsha X^ a
Inc. If §• e .2duiG
.c U
a
d rs
u o
5 i
2
I
X
E 5
Corporate Mergers and the Co-insurance of Corporate Debt 365
£
2 I g E
m o fn o O «n o Tl (^ O
•s e- I a
11 0 .i
- o , 3 1 ^
n 03
a a 11 I M ^
ffi Z <
— • 5 B
E ^ « S I
s s s
1 2 5 » O. H Ou B s s s i= g
11 a si
o < ca ^ CO ( n < < < < < 0 9 < < 0 9 m o 9
•2 a Vi Vi
o o o o U O O O
a g
11, U -O I
t^ to cn 'O
£ 3 < X
Tf m m m
g a I I « I S
Co. rake D <
•atin
a 1 Dry (^1) c
Corp
c idated sol c a
E E E E < <
~ o.
D 1
. "
I- '3 ' t a a
a I
u c (if inoi ISc
Inc. Co.. 1 «
E ff Moi
o v O t
2
i I
£ a
•o \ r"^ r^s. r-
ON O 00 —**
a 00 1 eufacl n
eem
M
Co,
- H
Lorilla
Inc. s 1
chle-G< Mi.
Inc, ing I 1
ndard
'.
Sta
si
I I 1 I
•g i
Z^ o <nS
a
E E E E
d «
O I
s 2
I 1 s 1
III
caca 03 oa 03 oa CD < <
•2 •a • (/3 M M
IN IN (N
I
