TESTS OF THE EFFICIENCY PERFORMANCE

OF CONGLOMERATE FIRMS

J. FREDWESTONAND SURENDA
K. MANSINGHKA*

MANYASPECTS of conglomerate firms have now been studied. The literature

on portfolio theory is relevant for appraising diversification aspects of
conglomerate firm performance? Other aspects of the economic theory of
conglomerates have been treated in various papers? A comprehensive analysis
of legal aspects has been made? Business motivations for conglomerate diver-
sifications have been described: The conglomerate merger movement has also
received considerable scrutiny with regard to its public policy effects? Most
empirical work tests aspects of the effects of mergers generally? Only three
statistical studies of conglomerate performance have been published.’
Professor Samuel R. Reid’s studies include data evaluating a sample of
conglomerate firms for the decade ending in 1961. He utilized three measures
he characterized as reflecting the interests of managers and three reflecting
the interests of stockholders. Professor Reid concluded that more actively
merging firms and firms that diversified to a greater extent in their merging
activity scored higher on the criteria related to managers’ interests and .lower
on criteria related to stockholders’ interests.*
Lorie and Halpern studied the performance of 117 mergers taken from the
Federal Trade Commission listing for 1954-1967 of all mergers in manufactur-
* Professor of Business Economics and Finance, University of California at Los Angeles, and
Assistant Professor of Finance, University of California a t Riverside, respectively. Research sup-
port was provided by the Research Program in Competition and Business Policy, UCLA. Helpful
comments were received from A. A. Alchian, M. Goudzwaard, N. H. Jacoby, R. H. Mason and
K. Smith. Research assistants participating in data compilation included: Kent Alves, Scott Ditt-
rich, Jorge Ramos, Whitt Jones and Nanci Grottke. Ronald Shrieves carried a major responsibility
in the computer calculations.
1. See Fama (12), Higgins (19a), Lewellen (23a), Lintner (24), Markowitz (26), Sharpe (36),
and Tobin (39).
2. A useful sample would include Gort (14, 15), Mueller (27), Narver (29, 30), Reid (34),
Stone (38), Udell (41), Weston (43, 4 9 , Weston & Peltzman (42) and 32 papers in the Spring
1970 Edition of the St. John’s Law Review (35).
3. Some of the relevant papers are Asch (4), Bicks (6), Blair (7), Bock ( 8 , 9 ) , Day ( l l ) , Pat-
terson & Patterson (31), Turner (40) and 18 items in (35).
4. Various aspects are treated in Ansoff (1, 2), Berg ( 5 ) , Weston (44) and in a number of
management journal articles.
5. See especially the discussion in Studies by the Staff of the Cabinet Committee on Price
Stability (lo), pp. 69-81; see also (13) with the introductory statements by Senator Philip A.
Hart of Michigan and by members of the Federal Trade Commission.
6. Gort and Hogarty (16), Heiden (19), Hogarty (213, Kelly (22).
7. The 1171-page special edition of the St. John’s Law Review on conglomerate mergers con-
tains no article with empirical data on the comparative performance of conglomerate firms (35).
T w o empirical articles in the volume deal with other aspects of performance. The S. E. Boyle paper
analyzes the pre-merger growth and profitability characteristics of acquired companies. The paper
by T. F. Hogarty reviews earlier historical studies of the success of mergers generally.
8. Reid (34).

919
920 The Journal of Finance
ing and mining in which the acquired firm had $10 million or more of assets?
I n their study the focus is particularly on the possibility of “deception of
investors.” The mergers which they studied were therefore those in which
the shareholders of the acquired company received relatively complicated in-
struments such as convertibles or warrants rather than cash, common stock,
ordinary preferred stock or bonds.1°
The investment return to stockholders of the acquired firms was analyzed
on various bases comprehended in the period six months prior to the merger
to two years after the merger. I n general, the investment return performance
to stockholders of the acquired firms was superior to the market performance
of broad market indexes for comparable periods of time. For example, the
mean rates of return for the periods 12 and 14 months subsequent to the
mergers were 9.34 per cent and 9.52 per cent, respectively, while correspond-
ing rates for the market index were 7.7 per cent and 7.38 per cent.
Smith and Schreiner compared the efficiency of portfolio selection by con-
glomerates with that of investment companies.11 Their findings indicated
superior portfolio performance for investment companies. Their data on in-
vestment companies was based on actual industry distributions of investments.
However, portfolio results of the conglomerates was a simulation study based
upon the industries in which the conglomerate firms appeared to be operating.
Further, the distinctive economic function of conglomerates is not portfolio
diversification. Indeed, it has been demonstrated that “homemade diversifica-
tion” is an effective substitute for corporate or other institutional diver-
sification under the assumptions of perfect capital markets.12
Since the data of the previous empirical studies ended predominantly in
the early or mid-l960’s, an updating is clearly required.13 Also, most of the
studies dealt with mergers generally. Of the three empirical studies dealing
with conglomerate mergers, the Lorie and Halpern study is specialized to the
point of view of stockholders in acquired companies. The Smith and Schreiner
study focuses on aspects of conglomerate diversification under a form of
simulation for conglomerates. Only the Reid study is directly comparable to
this paper.
To evaluate the findings of the previous studies and related issues, the
present study begins with a comprehensive empirical analysis of the per-
formance of conglomerate firms. We employ a number of measures parallel-
ing the performance measures presuming to reflect the interests of managers
and individual stockholders as well as institutional investors. However, our

9. Lorie and Halpern (25).

10. Since this kind of “funny money” (Lorie and Halpern’s term) was alleged to be characteris-
tic of conglomerate mergers, their sample of firms is presumed to be of “conglomerates.” However,
no formal criteria for selection were employed.
11. Smith and Schreiner (37).
12. Myers (28).
13. The Heiden and Reid studies ended in 1961, the Kelly and Hogarty studies ended in 1963
and 1964, respectively. The conglomerate merger movement did not really get underway until the
mid-l960’s, and peaked in 1967 and 1968. In fact, a fair test of the conglomerate merger movement
is not possible even today since a number of additional years would be required for implementa-
tion of the strategies associated with these corporate development programs.
 Performance of Conglomerate Firms 92 1
greatest emphasis is on criteria related to economic efficiency. After our
results are presented, an evaluation is made of Professor Reid’s study. Next,
some supplementary data sources for evaluating conglomerate performance
are presented. Finally, some comments on the significance of our findings are
made.
I. IDENTIFICATION AND SELECTION OF CONGLOMERATE FIRMS
The term “conglomerate” heretofore has been ambiguous and its implica-
tions have undergone a substantial tran~formation.’~The confusion is
aggravated by the classification system employed by the Federal Trade Com-
mission in reporting on merger activities in three categories. Horizontal
mergers involve firms in the same industry. Vertical mergers represent firms
at different levels of the production process in the same industry. Conglom-
erate mergers are “all other” mergers. Conglomerate mergers are further
divided into “product extension and geographic extension” and “other” merg-
ers.
Much of the ambiguity stems from the emphasis in the Census Standard
Industrial Classification (SIC) system on characteristics of products for
defining “industries.” The more relevant consideration from the standpoint
of the economic effects of mergers is the extent to which research, manufactur-
ing, marketing and other managerial capabilities are extended or comple-
mented in a merger.16 From this standpoint it would be more informative to
label product and market extension mergers as “concentric” or “congeneric.”
These terms suggest that the merging firms possess related characteristics.
Since the word “conglomerate” carries some ambiguity, we attempt to give
some explicit content to the term.la We are focusing on firms that have entered
into a broad program of diversification achieved to a substantial degree by
external mergers and acquisitions rather than by internal development. I n
selecting the companies to be included in the study, two criteria are therefore
relevant. The first is the extent to which growth was achieved from external
sources and the second is the degree of diversification represented by the ex-
ternal mergers and acquisitions.
To implement these criteria, we first compiled a list of all firms involved
in three or more mergers during the period 1960 through 1968.1? In making
our compilation we sought to adjust for underreporting. If a firm’s assets
grew at a rate greater than 10 per cent per year, the possibility of growth
by merger was further investigated. I n addition, if acquisitions, adjusted for
post-merger growth, aggregated more than 500 millions, or if the firm ap-
peared to have been involved in at least two major acquisitions or a number of
smaller acquisitions as indicated by the increase in the number of subsidiaries

14. For a more complete discussion, see Weston (43), (45).

15. For further development of these criteria see Weston (43), pp. 52-54.
16. Previous studies have not employed analytical criteria for identification of conglomerate
firms. Hence this paper sets forth the selection criteria explicitly.
17. We relied on a number of sources t o provide candidates for potential inclusion in our study
including the Federal Trade Commission reports and a number of lists compiled by financial
magazines and brokerage firms.
 Performance of Conglomerate Firms 925
No great importance is attached to these measurements of the differential
growth rates. For any of the measures calculated the difference in growth rate
is primarily a function of the degree of merger activity of the conglomerate
firms. This applies to the growth rates of the earnings per share as well as
the growth rates in market price per share. The more critical measurements
reflect the effectiveness of the economic utilization of resources by the three
groups.

Comparative Performance Measures

Earnings performance measures for the-years 1958 and 1968 for 63 con-
glomerate firms were compared with those of the two other samples of firms.
Multiple measures are provided, representing alternative methods of measur-
ing earnings performance. To measure earning power before the influence of
financial leverage, the earnings are related to total assets. Since this rep-
resents all forms of assets financed by debt or equity, the costs of sources of
funds utilized to finance total assets must be included in the numerator?'
Our first measure is earnings before interest and preferred stock dividends to
total assets. Whether the numerator should be before or after taxes is a
matter of dispute in the financial literature, so we made the calculations both
ways.28 For 1968, we calculated the returns on net worth by two methods,
including and then excluding preferred stock.
I n Table 3, statistical tests of the significance between the means of the
earnings ratios are set forth. For the year 1958, the earnings rates of the
sample groups were significantly higher than the earnings rates of the con-
glomerates f i r m ~ By
. ~ ~1968, there were no significant differences observed in
earnings performance among the groups?o
Two possible explanatory influences on earning power performance are
examined in Table 4. Price/earnings ratios are calculated, along with their
significance measures, on two bases. The first set of price/earning ratios in-
cluded all firms. I n the second set of calculations, unusual price/earnings
ratios were excluded.3l For 1958, in the early stages of the conglomerate

27. For a precise measure, we would deduct non-interest-bearing liabilities from total assets and
then calculate the ratio of earnings before interest and taxes or earnings before interest alone to
total assets minus non-interest-bearing liabilities. However, since the non-interest-bearing liabilities
do not ordinarily represent a large percentage of total assets, for comparing groups of companies
it is more convenient to use total assets.
28. See, for example, the discussion in M. H. Miller and F. Modigliani, American Economic
Review LVI (June, 1968), pp. 333-391.
29. The sample of industrial companies (Rl) exhibited earnings performance in 1958 higher
than the conglomerate companies by a difference with significance a t the 1 per cent level for all
three comparisons.
30. In analyzing the data on preferred stock, we noted a substantial disparity between the
stated balance sheet values of preferred stock for the conglomerate firms and estimates of more
realistic values of the preferred stocks based on prevailing dividend yields, call prices or liquidating
values. For 1968 we calculated adjusted earnings rates on total assets, adding to the total assets of
each of the three sets of companies, the difference between our estimate of an appropriate balance
sheet value of preferred stock and the stated value. As shown in Table 3, the differences in the
earning power among the three groups of companies remained nonsignificant.
31. If the price/earnings ratio was greater than 100 times, or a negative multiple of SO or more,
we excluded it in the second set of calculations.
926 The Journal of Finance
TABLE 3
SIGNIFICANCE TESTS FOR EARNINGS
PERFORMANCE MEASURES, 1958, 1968
1958
Sample Means F-Statistics
C R1 R2 CJR1 C,R2 Rl,R2 C,Rl,R2
1. EBIAT/TotalAssets 5.8 9.2 6.8 9.83s' 1.10 6.12S6 6.21s'
2. EBIT/Total Assets 8.7 16.7 12.6 17.13s' 5.47S5 4.52S10 9.54s'
3. Net Inc/Net Worth 7.6 12.6 10.7 10.52S1 4.30Slo 2.08 6.05s'
1968
Sample Means F-Statistics
C R1 R2 C,R1 CJR2 R l , R 2 C,RI,R2
1. EBIAT/TotalAssets 10.4 8.5 7.6 0.44 0.93 1.37 0.70
la. EBIAT/TotalAssets* 9.4 8.2 7.5 0.21 0.50 0.80 0.37
2. EBIT/TotalAssets 15.1 15.6 13.3 0.02 0.38 2.33 0.44
3. Net Inc/Net Worth 13.3 12.4 12.0 0.81 1.98 0.12 0.85
4. Net Profit/Equity 14.2 12.4 12.2 0.99 1.44 0.06 0.98
Abbreviations have the same meaning as in Table 2. New measures introduced: EBIAT refers to
earnings before interest and preferred stock dividends, but after taxes. EBIT refers t o earnings
before interest, preferred stock dividends and taxes. Net Inc refers to net income plus preferred
stock dividends. Net worth includes preferred stock. Net profit to equity represents net income less
preferred dividends to net worth excluding preferred stock.
Total Assets* represents total assets plus the difference between the balance sheet values of pre-
ferred stock and estimated book value of preferred stock. The additions to total assets were re-
spectively $5.7, $0.5, and $1.7 billions for the conglomerates, sample 1 and sample 2.
Source: Same as Table 2.

movement, there were no significant differences between either of the two

sets of price/earnings In 1968, at the height of the popularity of
conglomerate firms, the difference between the means of their price/earnings
ratios and the means for Random Sample #1 remained non-significant, but
became significant compared with Random Sample #2.
These data are contrary to the general impression that one of the reasons
for the high growth in earnings per share of the conglomerate firms was that
they were able to play the differential price/earnings ratio game. On the
contrary, their price/earnings ratios as a group average were not significantly
different from the price/earnings ratios of other groups of industrial firms.
A dilemma confronted the conglomerate firms with depressed price/earn-
ings ratios. If they merged with firms with higher price earnings, the conglo-
merates would suffer dilution in earnings per share which would at least
initially depress growth rates in earnings per share. The disadvantage of a
differentially lower price/earnings ratio in merger activity can be offset by
offering non-equity forms of securities in exchange or utilizing delayed equity
forms such as convertibles and warrants.
Table 4 provides information on this point also. Debt to net worth ratios
32. For 1958, the second set of price/earnings ratios, excluding the large negative ratios for some
of the conglomerate firms, raised the average for the remaining conglomerates somewhat above the
average price/earnings ratios of the other two samples. However, the differences in the means re-
mained nonsignificant.
 Performance of Conglomerate Firms 927
TABLE 4
TESTOF FACTORS
SIGNIFICANCE EARNINGS
INFLUENCING PERFORMANCE,
1958, 1968
1958
F-Stati~ti~~
Sample means
C R1 R2 C,R1 C,R2 R1,R2 C,Rl,
R2
1. Price/Earnings
Ratios 15.2X 14.6X 18.3X 0.02 0.37 0.96 0.36
2. Price/Earnings,
Adjusted* 17.2X 14.9X 15.8X 0.83 0.34 0.22 0.51
3. Debt/Net Worth
Incl. P/S** 95% 56% 155% 8.19S1 2.13 6.24S5 4.33S5
1968
F-StatiSti~~
Sample means
C R1 R2 C,R1 C,R2 Rl,R2 C,R1,
R2
1. Price/Earnings
Ratios 19.4X 18.7X 16.7X 0.10 4.60S10 0.74 0.97
2. Pricefiarnings,
Adjusted* 19.4X 20.2X 16.7X 0.26 4.61Sl0 4.22S10 2.82
3. Debtmet Worth
Incl. P/S** 169% 87% 201% 10.25S1 0.33 5.45S5 3.46S10
+
4. (Debt P/S)/
Equity 233% 94% 210% 5.95S6 0.09 5.6736 3.00S10
* The adjustment for the price earnings ratios was to exclude ratios exceeding 100 times and
those with negative values exceeding 50 times.
** P/s refers to preferred stock. Other abbreviations have the same meanings as in Tables 2
and 3.
Sowce: Same as Table 2.

were measured on two bases in 1968. On the first basis, preferred stock was
included in net worth, but not in debt. With net worth including preferred
stock, the predominant portion of which for conglomerates was convertible,
the ratio of debt to net worth was 169 percent. This compared with 87 percent
for Random Sample #1 and 201 percent for Random Sample #2.Sa The lever-
age ratios of the conglomerate firms were significantly higher than those of
Random Sample #1 firms in both 1958 and 1968. In 1958 the debt ratios of
the conglomerate firms were significantly lower than those of Random Sample
#2, including non-industrial firms, but they were not significantly different
in 1968. As compared with industrial firms, therefore, the conglomerate firms
employed higher debt ratios, both in 1958 and in 1968. In addition, as com-
pared with both samples, the debt ratios of the conglomerate firms grew more
rapidly during the decade.

33. Random Sample 12 included banks and utilities, accounting for leverage ratios higher than
those for industrial lirms.
928 The Journal of Finance
111. IMPLICATIONS OF DEFENSIVE DIVERSIFICATION
Some preliminary implications of the data in the foregoing tables may be
indicated. The conglomerate firms had lower profitability rates in the late
1950’s. On the average, their price/earnings ratios were not significantly dif-
ferent from the price/earnings ratios of other firms. Utilizing leverage, how-
ever, they were able to raise the return on their net worth. But measurement
of returns on total assets, which is free of the effects of financial leverage, is a
more meaningful figure. The returns on total assets for the conglomerates in
1958 were generally significantly lower than for the other samples. However,
by 1968 the returns on total assets were not significantly different.
This improvement in earnings performance is evidence consistent with the
successful achievement of defensive diversification by the conglomerate firms.-
Defensive diversification may be defined as diversification to avoid adverse
effects on profitability from developments taking place in the firm’s traditional
product market areas.8a Hence the foregoing data are consistent with the pro-
position that the conglomerate firms perform the economic function of preserv-
ing the values of ongoing organizations as well as restoring the earning power
of the entitiesFBIn addition, the conglomerate firms avoid the costs of bank-
ruptcy.8’
Analysis of the backgrounds and acquisition histories of the conglomerate
firms suggests that they were diversifying defensively to avoid (1) sales and
profit instability, (2) adverse growth developments, (3) adverse competitive
shifts, (4) technological obsolescence, and ( 5 ) increased uncertainties as-
sociated with their industries. These potentially unfavorable developments
occurred in various degrees and combinations in a number of industrial cate-
gories.
The largest single category of firms came from the aerospace industry. This
industry is subject to wide fluctuations in total market demand as well as
abrupt and major shifts in its product mix. Excess capacity developed, ag-
gravated by entry of firms from other industries. The industrial machinery
and auto parts companies were also subject to considerable instability in sales.
Low growth prospects were associated with a major secular decline in some
markets. This was characteristic of the railway equipment industry, textiles,
movie distribution (at an earlier period), and the tobacco industry. A related
problem was slow growth in sales or declining profit margins due to persistent-

34. Other aspects of a theory of the timing of the conglomerate mergers movement are discussed
in Weston (43), pp. 50-52.
35. This could also be expressed as changing the product market mix of the firm’s portfolio of
capital budgets. Again, from the standpoint of the economy as a whole, it would make no dif-
ference whether investors provided their own diversification or whether the individual firms
achieved diversification, under the assumptions of no bankruptcy costs and no costs of rebuilding
an effective organization.
36. An elaboration of how the conglomerate firms achieve these functions is set forth in Weston
(43, 44).
37. It is probable that some individual conglomerate firms w ill go through reorganization or
bankruptcy. This would not be inconsistent with the above proposition. The point is that more of
these entities would have been forced into reorganization and bankruptcy if they had not achieved
successful defensive diversification.
 Performance of Conglomerate Firms 929
ly unfavorable excess capacity conditions in these (and other) industries. A
number of firms sought to diversify in the attempt to acquire access to the new
technologies that had developed after World War 11. Illustrative was the need
of firms in the electrical industry to develop an electronics capability.
Another large category of conglomerates are firms originally based in the
natural resources industries. These industries are characterized by instability
in demand and supply relationships resulting in substantial fluctuations in
product prices and in profit margins. I n addition, rising corporate tax rates
stimulated them to seek diversification to utilize more fully the tax shelter
afforded natural resource operations.
The classification of these diverse firms into a small number of forms of
defensive diversification involves some overgeneralization. In addition to fac-
tors such as tax considerations, some of the conglomerates pursued positive
programs such as applying advanced technology in industries and firms where
technology had lagged (e.g., Litton Industries). Others were attempting to
utilize effectively special capabilities in financial planning and control (e.g.,
IT&T and Transamerica) .
But defensive diversification was a strong motivation among the conglom-
erate firms. Consistent with the important role of defensive diversification are
our findings that profitability of the conglomerate firms was below average in
the late 1950’s and had improved to average levels by 1968. In this connec-
tion, Professor Reid found that on measurements similar to those employed
in the present study the conglomerate firms had less favorable performance.
Professor Reid’s analysis distinguished between tests that represented the
interests of the managers of the firms and those that represented the interests of
the owners of the The three measures Professor Reid used to reflect
managers’ interests are growth in sales, (2) growth in assets, and (3) growth
in employment. The three measures he uses to reflect stockolders’ interests are
(1) growth in the market value of shares, ( 2 ) growth in the ratio of net in-
come to total assets, and (3) growth in the ratio of net income to sales. Pro-
fessor Reid found that growth rates in the three measures reflecting managers’
interests were favorable for conglomerates compared with other companies,
while the conglomerates performed less effectively on measures reflecting the
stockholders’ interests.
But Professor Reid’s conclusions are at variance with his own data. In his
detailed industry analysis, the results for his measures of performance were
either not significant or significant in the wrong direction for ten or more of
the 14 industries for each of his six measures.88aWhen Reid does not group
his firms by industry, his data on the conglomerates compared to non-merging
38. Reid (33, 34). There are a number of justifications for examining the Reid study in some-
what more detail. Most of the other studies dealt with the success of mergers or merging firms
generally. These studies are at least indirectly relevant since if merging generally is unsuccessful
for firms, by implication the probability is that conglomerate merging would also be unsuccessful.
But the possibility also exists that the diversification and other characteristics of conglomerate
mergers could also increase the probability of their success. These possibilities provide a rationale
for studies focussed on conglomerate merger activity.
38a. Reid (34), Tables 9.3-9.8, pp. 186-188. “Significant in the wrong direction” indicates that
the performance of the conglomerates was superior to the control groups.
930 The Journal of Finance
firms is consistent with his generalization. But the performance of the con-
glomerate firms, by his own data, was more favorable for all six ratios than for
firms engaging in other forms of merger Furthermore, the fact
that the data grouped by industry shows no significant difference between
the conglomerate firms and all other firms suggests that it is the greater
weighting of a small number of industries that produced the results for his
total sample.
Conceptual difficulties mar two of his three measures purporting to reflect
stockholder interests. Professor Reid measured profitability by net income
available to common stockholders divided by total assets. If profits are to
be related to total assets, financial charges should be added back to reflect
other sources of funds utilized to finance total Otherwise the measure
of net income divided by total assets penalizes firms with greater leverage and
in particular firms that are increasing leverage over the time period measured.
Since the conglomerate firms were utilizing more debt and they were increas-
ing their debt at a faster rate over time than other firms, dividing net income
available to common stockholders by total assets biases the performance of
the conglomerate firms downward.
Professor Reid’s other measure of profitability was the time trend in the
ratio of profits to sales. This ratio has little economic significance in the con-
text of the present analysis. The ratio of profit to sales varies greatly among
industries, depending primarily upon the characteristic sales to asset turnover.
On the average, industries with high ratios of profits to sales are capital
intensive with a low turnover of assets and conversely for industries with high
Trends in the ratio of profits to sales, therefore, reflect primarily
the industry mix of the firms rather than any measure of inherent profitability
or managerial performance. Since the conglomerate firms were by definition
changing their product mix, Reid’s measure of the ratio of profits to sales
simply reflects the fact that conglomerate firms were moving into product
areas with higher asset turnover, characteris tically associated with lower
profit margins on sales than industries with lower total asset turnovers.
Thus, two of the three measures purporting to reflect stockholder interests
contained conceptual defects. The third test Reid employed was growth in
market price of common shares-a measure which suffers from the defect of
the early terminal year (1961) for the data. Professor Reid’s conclusion,
therefore, that the conglomerates operated effectively for the managers, but
not for the stockholders has no valid empirical basis. Nor is his implication
that conglomerate firms were not efficient supported by the data. The data
from our own studies support the conclusion that the performance of the
conglomerate firms, on the average to date, satisfies the criterion of efficient
economic performance. However, the question may be raised whether our
terminal year of 1968 may be too early to reflect the critical impacts on
conglomerates during 1969. Data were not available for 1969 while our study

38b. See Reid (34), Table 9.1, 0. 185.

38c. See any standard textbook on business finance.
38d. For details see Gupta (IS).
 Performance of Conglomerate Firms 931
was in progress. However, from a number of other sources, data through 1969
for segments of the analysis performed in our broader study are available
as a check on our results.
IV. OTHEREMPIRICAL EVIDENCE ON CONGLOMERATE PERFORMANCE
Forbes, Fortune and News Front magazines compile annual performance
results for up to 1,000 firms. These data provide a basis for further analysis
of conglomerate performance?’
The Forbes Data40
Forbes compiles data which permit calculation of six performance measures
for 500 or more firms as well as for two categories termed by Forbes “con-
glomerates” and agglomerate^."'^ The Forbes measures summarized in Table
5 permit comparisons with each of the major categories of analysis employed
in our study. Forbes calculates a five-year annual increase in The
median values of the growth rates in sales of the multi-industry companies
were at least double the growth rate in sales for the broader sample of firms.
The growth rate in earnings per share for the multi-industry companies ranged
from 12.2 per cent to 12.7 per cent for the two five-year periods, ending in
1968 and 1969, respectively, about 50 per cent better than the performance
by the total of the Forbes firms. Forbes measured the gain in market price in
terms of a total percentage increase for a five-year period. The multi-industry
companies performed more than twice as well as the broad sample; growth for
both declined when 1969 is taken as the terminal year.
Forbes calculates three profitability measures. The first profitability mea-
sure is a five-year average combined return on debt and equity. This is
comparable to our measure of earnings before interest and preferred stock
dividends but after taxes, to total assets, except that the Forbes denominator
does not include current liabilities. The results confirm the findings of our
analysis. The five-year average combined return on debt and equity for the
multi-industry companies for the period ending in 1968 was slightly higher
than the Forbes broader group. For the period ending 1969 there was no

39. These sources provide rankings which permit easy determination of median values. Our
worksheet analysis of alternative measures of central tendency indicated that the median values
are relatively free of the distortions produced by weighted or “unweighted” averages. Therefore
the medians were used for the Forbes and Fortune data test. Use of the median values precludes
utilization of the usual statistical tests of significance. However, the observed differences between
the medians were either so large as to be clearly significant or so small as to be clearly nonsignifi-
cant.
40. Forbes, Vol. 103 (January 1, 1969), Vol. 104 (January 1, 1970).
41. Forbes defines conglomerate companies as those which have found a central focus for their
operations, while the agglomerate companies have diversified without finding a central thread. The
two categories are combined for analysis of “multi-industry” firms.
42. The five-year growth percentages are measured by Forbes by averaging the three initial
years and calculating the growth to the values for the terminal years. For example, in the January
I, 1970 issue of Forbes, the three initial years averaged were 1963, 1964 and 1965; the terminal
year was 1969. The five-year growth rates for sales and earnings per share are medians of the
compound annual growth rates calculated for individual companies. For market prices, the total
percentage changes for the five-year periods are calculated.
932 The Journal of Finance
TABLE 5
PERFORMANCE
COMPARISONS
BETWEEN MULTI-INDUSTRY COMPANIES
LARGEFIRMS, 1963-1969*
TOTAL
AND FORBES’

Multi-Industry
Forbes’ Firms Companies
1962-64 1963-65 1962-64 1963-65
to 1968 to 1969 to 1968 to 1969
Number of Companies 500 578 38 47
1. Five-year annual increase in sales 9.0% 9.9% 20.0% 19.05%
2. Five-year annual increase in per-
share earnings 8.6 7.05 12.7 12.2
3. Price gain for five-year periods 70.5 17.15 140.0 60.2
4. Five-year average combined
return on debt and equity 9.7 9.6 10.0 9.6
5. Five-year average return on
equity 12.4 12.5 12.6 13.4

-1968
-1969 1968
- 1969
-
6. Last year’s return on equity 12.0 11.65 12.2 11.85
* Median values compared on all items.
Sources: Forbes, Vol. 103 (January 1, 1969),pp. 37-41,53-66,78, 80, 91.
Forbes, Vol. 104 (January 1, 1970), pp. 44-81,96-99,227-231.

difference in the median values of the five-year average returns for the two
groups.
The results of the Forbes data on the other two measures of earning per-
formance were also the same as for our study. Both the five-year average
return on equity and last year’s return on equity for the multi-industry com-
panies was slightly higher than the earnings performance of the Forbes total
number of large companies. The Forbes data, therefore, provide confirmation
of the results of our study.
The News Front Data
News Front compiles data on the 1,000 largest manufacturing companies
in the United States. 43 The firms are grouped by SIC categories. In its 1969
presentation, New Front presented a category, SIC 3999, “diversified/con-
glomerate companies,” in which are listed 47 firms. Presenting data for 1968,
the diversified/conglomerates averaged a return on stockholders’ equity of
12.1 per cent, compared to 12.2 per cent for “average man~facturing.”~~ The
difference between the performance of the multi-industry companies and
“average manufacturing” described by News Front is negligible.
The Fortune Data
Fortune magazine each year caIculates the ten-year compound annual
growth rates of earnings per share for its sample of 500 industrial firms and
43. News Front (June-July, 1969), p. 70.
44. News Front (June-July, 1969), p. 56. The averages computed by News Front are the
weighted arithmetic means.
 Performance of Conglomerate Firms 933
the ratio of net income to net worth. These results are summarized for our
sample of conglomerate firms compared with the total Fortune 500 firms
for the most recent five years through 1969. The results are tabulated in
Table 6. For the ten-year periods ending in 1965 and 1966, the compound
TABLE 6
COMPARISONS
OF TEE PERFORMANCEOF CONGLOMERATES
WITH THE
FORTUNE 500 FIRMS,1965-1969*
Ten-Year Compound Annual Net Profit to
Growth Rates of EPS Stockholders’Equity
Gonglom- Conglom-
erates less erates less
Years Fortune Conglom- Fortune Fortune Conglom- Fortune
Ended 500** erates 500 500 erates 500

1969 7.10 9.91 +2.81 11.3 11.8 +0.5

1968 8.71 11.98 +3.27 11.7 12.9 + 1.2
1967 6.59 8.57 + 1.98 11.3 12.4 +1.1
1966 6.09 5.99 -0.10 12.7 13.2 +0.5
1965 6.04 5.74 -0.30 11.8 12.4 +0.6
* Medians of the growth rates or percentage returns for each group of companies are reported.
** Comparisons with our two random samples does not change the direction of the differences,
but the magnitudes are sometimes larger, sometimes smaller.
Source: The Fortune Directory, 1966-1970.

annual growth rates of the Fortune 500 companies were higher than for the
conglomerate firms. Beginning in 1967, the conglomerate firms began to out-
perform the Fortune 500 firms by relatively large differentials. For the ratios
of net income to net worth, the conglomerates were higher than the Fortune
500 in every year but the differences are not significant.
V. CONCLUSIONS
Most of the earlier studies of mergers covered periods ending in the early
or mid-1960s. The present study, supplemented by compilations from Forbes,
Fortune and News Front magazines, extends the analysis through 1969. Our
empirical findings suggest three generalizations.
1. The conglomerate firms outperformed samples of other firms or broader
groups of firms on all of the growth measures. Contrary findings of some
other studies resulted from their early terminal dates, before the mid-1960s.
However, we do not attach great significance to these measures, since they
reflect primarily the extent of past merger activity by the conglomerate firms.
More important will be the growth in market price per share over a period of
years in the future, after the ability of the conglomerate fir- to sustain a
growth rate in earnings per share over an extended period has been tested.
2. The earnings performance measured by the ratio of net income to net
worth is somewhat higher for conglomerate firms, but the difference is not
statistically significant. This somewhat higher return on net worth did not
result from the differentially higher price/earnings ratios of conglomerate
firms in mergers because on the average the price/earnings ratios of conglom-
erate firms were not significantly different from the price/earnings ratios of
934 The Journal of Finance
other samples of firms. The higher returns on net worth of the conglomerate
firms resulted from the larger and increasing percentage of leverage employed
by them during the decade of the ’60s. The effects of the higher leverage of
conglomerate firms on their longer-run success has not been fully tested.
3. We attached greatest significance as a measure of the economic per-
formance of conglomerate firms to the ratios of profitability free of the in-
fluence of leverage. The earnings rates of the conglomerate firms in the late
1950s or the early 1960s were significantly lower than the earnings on total
assets or net worth-plus-long term debt for other groups of firms. However,
by 1968 no significant differences in earnings performance were observed.
I t appears, therefore, that an important economic function of conglomerate
firms has been raising the profitability of firms with depressed earnings to the
average for industry generally?6 This function was particularly important in
connection with defensive diversification. A large proportion of the conglom-
erate firms developed out of industries in which adverse developments in their
economic environments had either taken place or were in prospect. Therefore,
the most appropriate test of the earnings performance of conglomerate firms
is not superior earnings performance, but their ability to achieve at least
average earnings
I n many respects the conglomerate movement may be regarded as the
development of a new industry. Economic principles predict, therefore, that
the earning power of this group of firms would tend to equality with the earn-
ing power of firms on the average. Theory also predicts great variations in
earnings performance among individual conglomerate firms.
The superior performance of individual firms in new industries has charac-
teristically attracted an excess number of firms and firms without the requi-
site managerial or product characteristics for survival. Hence, as a new in-
dustry, one would expect the failure rate among conglomerates in the near
future to be somewhat higher than in industry generally. But there will be
individual conglomerate firms that will achieve higher-than-average profitabil-
ity over periods of time. The success of individual conglomerates, as is true of
firms generally, will depend upon the relative abilities of their managements
for achieving operating efficiency and for making the shifts in their product-
markets required by the changing economic environment.
REFERENCES
1. H. Igor Ansoff. “Strategies for Diversification,” Harvard Business Review (September-October,
1957), pp. 113-124.
2. .“A Model for Diversification,” Management Science 4 (July, 1958), pp. 392-414.

4 5 We have also studied a number of aspects of the dispersion of returns, comparing measures
of dispersion among the conglomerate firms with the two samples of firms. In addition, we are
analyzing the risk/return characteristics in the framework of a capital asset market model. The
results will be presented in a separate paper.
46. On the one hand, if the earnings performance of the conglomerate firms had been markedly
superior to industry generally, they would be subject to charges that they possessed monopoly
power, as have other large firms which have achieved high profitability rates over a period of
time. On the other hand, because their earnings as a group have not been significantly higher than
industry generally, the conglomerate firms have been charged by their critics as not contributing
to improvements in economic efficiency. But to improve earnings up to the average, as well as to
avoid falling below the average, represents a contribution to economic efficiency.
 Performance of Conglomerate Firms 935
3. and J. Fred Weston. “Merger Objectives and Organization Structure,” Quarterly
Review of Economics and Business 2 (August, 1962), pp. 49-58.
4. Peter Asch. “Conglomerate Mergers and Public Policy,” MSU Business Topics 15 (Winter,
1967), pp. 61-67.
5. Norman Berg. ‘Strategic Planning in Conglomerate Companies,” Haward Business Review 43
(May-June, 1965), pp. 79-92.
6. R. A. Bids. ‘LConglomeratesand Diversification Under Section 7 of the Clayton Act,” Anti-
trust Bulletin 2 (November-December, 1956), pp. 175-186.
7 . J. M. Blair. “The Conglomerate Merger in Economics and Law,” Georgetown Law Journal 46
(Summer, 1958), p. 672.
8. Betty Bock. Antitrust Issues in Conglomerate Acquisitions, Studies in Business Economics,
Number 110. New York: National Industrial Conference Board, 1969.
9. . L‘ConglomerateMergers, Joint Ventures, and Potential Competition,” The Con-
ference Board Record 5 (February, 1968), pp. 2-6.
10. Cabinet Committee on Price Stability. Studies by the Staf of the Cabinet Committee on
Phce Stability. Washington, D.C.: US. Govt. Printing Office, 1969.
11. Richard E. Day. L‘ConglomerateMergers and the Curse of Bigness,” North Carolina Law
Review 42 (1964), p. 539.
12. Eugene F. Fama. <‘Risk, Return and Equilibrium: Some Clarifying Comments,” Journal of
Finance 23 (January, 1968), pp. 29-40.
13. Federal Trade Commission Staff. Economic Report on Corporate Mergers. Washington, D.C.:
U.S. Govt. Printing Office, 1969.
14. Michael Gort. Diversification and Integration in American Industry. Princeton, N.H.: Nation-
al Bureau of Economic Research, 1962.
15. . “An Economic Disturbance Theory of Mergers,” Quarterly Journal of Business
83 (November, 1969), pp. 624-642.
16. and Thomas F. Hogarty. “New Evidence on Mergers,” Journal of Law and
Economics XI11 (April, 1970), pp. 167-184.
17. Maurice B . Goudzwaard. “Conglomerate Mergers, Convertibles and Cash Dividends,” Quar-
terly Review of Economics and Business (Spring, 1969), pp. 53-62.
18. Manak C. Gupta. “The Effect of Size, Growth, and Industry on the Financial Structure of
Manufacturing Companies,” Journal of Finance XXIV (June, 1969), pp. 517-529.
19. Edward J. Heiden. “Mergers and Profitability.” Unpublished manuscript, University of
Wisconsin.
19a. R. H. Higgins, “Discussion,” Journal of Finance XXVI (May 1971), pp. 543-545.
20. Thomas F. Hogarty. “The Profitability of Corporate Mergers,” Journal of Business 43 (July,
1970), pp. 317-327.
21. Neil H. Jacoby. “The Conglomerate Corporation,” The Center Magazine 2 (July, 1969), Re-
print.
22. Eamon M. Kelly. The Profitability of Growth Through Mergers. The Pennsylvania State
University, 1967.
23. Henry A. Latanb. “Portfolio Balance-The Demand for Money, Bonds, and Stocks,” Southern
Economic Journal XXIX (October, 1962), pp. 71-76.
23a. W. G. Lewellen, “A Pure Financial Rationale for the Conglomerate Merger,” Journal of
Finance XXVI (May 1971), pp. 521-537.
24. John Lintner. “Security Prices, Risk, and Maximal Gains from Diversification,” Journal of
Finance XX (December, 1965), pp. 586-591.
25. James H. Lorie and Paul Halpern. “Conglomerates: The Rhetoric and the Evidence,” Journal
of Law and Economics XI11 (April, 1970), pp. 149-166.
26. Harry Markowitz. Portfolio Selection. New York: Wiley, 1959.
27. Dennis C. Mueller. “A Theory of Conglomerate Mergers,” Quarterly Journal of Economics
83 (November, 1969), pp. 643-659.
28. Stewart C. Myers. “Procedures for Capital Budgeting Under Uncertainty,” Zndustrial Manage-
ment Review 9 (Spring, 1968), pp. 1-19.
29. John C. Narver. Conglomerate Mergers and Market Competition. Berkeley and Los Angeles:
University of California Press, 1967.
30. . “Some Observations on the Impact of Antitrust Merger Policy on Marketing,”
Journal of Marketing 33 (January, 1969), pp. 24-31.
31. James M. Patterson and Jeanne Patterson. “Conglomerates: The Legal Issues,” Bwiness
Horizotw 11 (February, 1968), pp. 39-48.
32. E. F. Renshaw. “The Theory of Financial Leverage and Conglomerate Mergea,” California
Management Review 11 (Fall, 1968), pp. 79-84.
936 The Journal of Finance
33. Samuel R. Reid. “Mergers for Whom: Managers or Stockholders?,” Hearings before the Sub-
committee on Antitrut and Monopoly of the Committee on the Judiciury. United States
Senate, 89th Congress, Second Session, pp. 1914-1931.
34. . Mergers, Managers and the Economy. New York: McGraw-Hill, 1968.
35. St. John’s Law Review. Conglomerate Mergers and Acquisitions 44 (Spring, 1970), Special
hue.
36. William F. Sharpe. Portfolio Theory and Capital Markets. New York: McGraw-Hill, 1970.
37. K. V. Smith and J. C. Schreiner. “A Portfolio Analysis of Conglomerate Diversification,”
Journal of Finance 24 (June, 1969), pp. 413-428.
38. James M. Stone. Conglomerate Mergers: Their Zm$lications for the Eficiency of Cagital and
the Theory of the Firm. Unpublished thesis, Harvard University (March 27, 1969).
39. James Tobin. “Liquidity Preference as Behavior Towards Risk,” Review of Economic Studies
67 (February, 1958), pp. 65-86.
40. D. Turner. ‘LConglomerateMergers and Section 7 of the Clayton Act,” Harvard Business
Rewiew 78 (May, 1965), pp. 1313-1395.
41. Jon G. Udell. “Social and Economic Consequences of the Merger Movement in Wisconsin,”
Wisconsin Economy Studies Number 3 (May, 1969).
42. J. Fred Weston and S. Peltzman (Eds.). Public Policy Toward Mergers. P a f i c Palisades,
Calif.: Goodyear Publishing Company, 1969. Part V, on Conglomerate Mergers.
43. J. Fred Weston. “Diversification and Merger Trends,” Business Economics 5 (January, 1970),
pp. 50-57.
44. . “Mergers and Acquisitions in Business Planning,” Rivtta Znternuzionule di
Scienze Economiche e Commerciali XVII (April, 1970), pp. 309-320.
45. . “The Nature and Significance of Conglomerate Firms,” St. John’s Law Review
44 (Spring, 1970), pp. 66-80.