UNIVERSITY OF
ILLINOIS LIBRARY
AT urbanacha;
BOOKSTACKS
b/4 t
CENTRAL CIRCULATION BOOKSTACKS
The person charging this material is re-
sponsible for its renewal or its return to
the library from which it was borrowed
on or before the Latest Date stamped
below. You may be charged a minimum
fee of $75.00 for each lost book.
Theft, mutilation, mid underlining of books are reasons
for disciplinary action and may result In dismissal from
the University.
TO RENEW CALL TELEPHONE CENTER, 333-8400
UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN
APR 291998
AUG 0 3 1998
When renewing by phone, write new due date below
previous due date. L162
Digitized by the Internet Archive
in 2011 with funding from
University of Illinois Urbana-Champaign
http://www.archive.org/details/onemoretimelooka1630duha
STX
COPY 2
BEBR
FACULTY WORKING
PAPER NO. 90-1630
One More Time: A Look at the Factors
Influencing Firm Performance
Irene M. Duhaime
J.L. Stimpert
mutu
m 2 4 m
16
College of Commerce and Business Administration
Bureau of Economic and Business Research
University of Illinois Urbana-Champaign
BEBR
FACULTY WORKING PAPER NO. 90-1630
College of Commerce and Business Administration
University of Illinois at Urbana- Champaign
February 1990
One More Time:
A Look at the Factors Influencing Firm Performance
Irene M. Duhaime*
J.L. Stimpert*
Department of Business Administration
"Department of Business Administration, University of Illinois, 350 Commerce
West, 1206 South Sixth Street, Champaign, IL 61820. Irene Duhaime: (217)
333-9344; J.L. Stimpert: (217) 333-4240.
ONE MORE TIME:
A LOOK AT THE FACTORS INFLUENCING FIRM PERFORMANCE
ABSTRACT
Several recent studies examine the factors influencing firm performance.
Schmalensee (1985) concludes that performance is largely the result of industry
effects — that some industries are more profitable than others, and so the choice
of industry in which to participate is a crucial management task. A study by
Wernerfelt and Montgomery (1988) concludes that performance is also influenced
by the extent of diversification, with narrowly diversified firms enjoying higher
performance than widely diversified firms. Hansen and Wernerfelt (1989) conclude
that both economic and organizational factors influence performance.
This study makes a further contribution to unraveling the performance
puzzle. We introduce an important new variable, comparative gross margin as a
proxy for management skill; use more recent data and alternative measures; and
reach conclusions that differ from previous studies. Based on our results, we
propose that relatively few new insights will come from additional studies
examining the relationship between diversification strategy and performance.
Rather, new strategic management research might focus on how management skill
moderates relationships among industry membership, diversification strategy, and
performance.
A large body of industrial organization research suggests that performance
is largely the result of industry effects — that some industries are more
profitable than others, and so the choice of industry in which to participate
is a crucial management task. Strategic management rejects this narrow view,
arguing that firm effects — strategies selected and implemented at the firm and
business levels — will also influence firm performance. Until fairly recently,
the term "strategies" often referred to the diversification strategies pursued
by firms, and strategic management researchers have given considerable attention
to the relationship between diversification strategy and performance. In fact,
a survey of the strategic management literature would suggest that a firm's
choice of diversification strategy is not only a key strategic decision, but also
a principal source of competitive advantage.
A newer stream of strategic management research, however, is challenging
both of these perspectives. These researchers are examining a variety of firm-
specific strategic factors, including firm-specific knowledge, business segment
strategies, and the structural relationship between corporate and business
segment strategies. This stream of literature points to management skill as a
key strategic variable, but work on management skill as a strategic variable has
not been well linked to other research predicting firm performance.
This paper provides evidence on the interactive relationships among
industry context, diversification strategy, management skill, and firm
performance, and suggests that management skill plays an important role in
influencing firm performance.
A REVIEW OF THE VARIOUS PERSPECTIVES ON PERFORMANCE
The Influence of Industry Membership on Performance
Long before strategic management researchers began exploring the topic of
diversification, economists maintained that the structural characteristics of
an industry would influence the profitability of firms in that industry. While
economists have been interested in the topic of diversification, their focus has
been on the possibility that firms would use the profits generated in one
industry to subsidize entry and expansion, and that continued expansion would
lead ultimately to increased concentration in another industry. By raising the
level of concentration in an industry, diversification might enable firms to
increase the profitability of that industry (Scherer, 1980). A study by Rhoades
(1973), for example, examined the relationship between diversification and
performance in 241 manufacturing industries and concluded that diversification
is associated with wider price-cost margins.
One of the key assumptions of industrial organization research, however,
is that firms don't matter much — that differences in profitability across firms
can be largely explained by industry membership and that industry performance
could be explained by barriers to entry and other structural characteristics.
Empirical work by economists appears to support this view (see for example Bain,
1951, 1956).
Perhaps the clearest expression of this view is in a recent article by
Schmalensee (1985). This research assesses the relative influence of industry,
firm, and market share effects on profitability. Using cross-sectional data from
the 1975 Federal Trade Commission Line of Business database, Schmalensee
concludes that 1) firm effects do not exist, 2) industry effects exist and are
important, 3) market share effects exist but have a negligible influence on
performance, and 4) industry and market share effects are negatively correlated
(1985:349).
The strategic implication of Schmalensee ' s research is straightforward —
firm performance is a function of the ability to acquire business units in
profitable industries. Schmalensee argues that Mueller's (1983) findings of
persistent firm-level profitability "are traceable to persistent differences at
the business unit or industry level, combined with relatively stable patterns
of activity at the firm level" (1985:349). Schmalensee also concedes, however,
that
it is important to recognize that 80 percent of the variance in
business unit profitability is unrelated to industry or share
effects. While industry differences matter, they are clearly not
all that matters (1985:350).
Diversification Strategy and Performance
A principal objective of strategic management research is to understand
how firms achieve competitive advantage over rivals in order to enjoy superior
economic performance. The field assumes that more is involved than simply
selecting individual industries or markets in which to compete. Instead,
strategic management researchers argue that the configuration of individual
businesses into a corporate portfolio — a firm's diversification strategy — can
be a source of competitive advantage. As a result, strategic management
researchers have shown a considerable interest in diversification, focusing
specifically on the relationship between diversification strategy and
performance.
Rumelt'a Strategy. Structure and Economic Performance (1974) provided a
basis and catalyst for this study of diversification. Building on the work of
Wrigley (1970), Rumelt developed a diversification taxonomy based on
relatedness, and nearly all subsequent research has defined diversification
strategy in terms of relatedness. Rumelt concluded that firms pursuing related
diversification strategies enjoyed higher levels of performance than firms
pursuing unrelated diversification strategies.
Rumelt 's original study was later replicated by Christensen and Montgomery
(1981) and Bettis (1981). Both of these studies concur with Rumelt' s conclusion
that firms pursuing related diversification strategies outperform firms pursuing
unrelated diversification strategies. In addition, Christensen and Montgomery
examine the intervening influence of market structure variables on the
relationship between diversification strategy and performance. Specifically,
Christensen and Montgomery suggest that firm size and market share and industry
concentration, growth rate, and profitability are also important influences on
performance. They also suggest that firms located in markets which constrain
growth or profitability are the most likely candidates to pursue unrelated
diversification strategies.
Bettis similarly examines the intervening influence of various strategic
decision variables including expenditures on advertising and research and
development, as well as capital intensity. He finds that these factors also
influence the higher levels of performance enjoyed by firms pursuing related
strategies.
The results of these studies are in general agreement with many of the
early studies in the finance literature which found few advantages for widely
unrelated diversification (Gahlon & Stover, 1979; Mason & Goudzwaard, 1976; and
Melicher & Rush, 1973). Many of these studies find that not only does unrelated
diversification fail to improve returns, but also that diversifying firms do not
achieve any significant risk reduction.
The literature, however, shows little consistency or consensus. While the
studies just cited suggest that firms pursuing related diversification
strategies may enjoy advantages over firms pursuing unrelated strategies, other
studies (Bettis & Hall, 1982; Lubatkin, 1987; Michel & Shaked, 1974; and Weston,
Smith, Si Shrieves, 1972) have reached the opposite conclusion — that unrelated
strategies can be more (or certainly no less) advantageous than related
strategies. In short, the diversification literature suggests that in spite of
extensive research, an understanding of diversification remains elusive; the
nearly two decades of research have produced few definitive conclusions, and the
findings of these studies are often contradictory.
One issue which has emerged from the diversification literature is the
question of the relative importance of diversification strategy versus the
choice of industry membership. The Christensen and Montgomery and the Bettis
and Hall studies noted above suggest that much of the high performance of firms
pursuing related diversification strategies could be attributed to industry
effects. Specifically, Bettis and Hall argue that the high returns of the
pharmaceutical firms in Rumelt's sample which were pursuing predominantly
related strategies may have been responsible for his findings. Rumelt (1982)
later replicated his original study and conceded that industry effects were
significant. He concluded, however, that performance differences across
categories persisted even after controlling for industry effects.
Similarly, Grant, Jammine, and Thomas also find a significant relationship
between diversification strategy and performance. They note, however, that the
importance of this relationship must be tempered by the fact that the
diversification variables in their study accounted for only "a small proportion
of interfirm differences in profitability. Industry membership accounted for
a larger proportion" (1988:795).
Management Skill As an Important Influence on Firm Performance
Prahalad and Bettis (1986) suggest that existing studies offer only
partial answers to understanding the relationship between diversification
strategy and performance. They suggest that we need to view management of the
large, diversified firm as a task which requires knowledge not only of each
individual business in which the firm operates, but also of the particular
requirements of managing a portfolio of businesses in a large, diversified firm.
More specifically, they propose that the relationship between diversification
strategy and performance will be influenced by the "dominant logic" of firms'
managers. This dominant logic is the shared understanding of the processes
needed to manage large diversified firms. According to this view, more complex
firms and unrelated diversification strategies require a broader dominant logic
to ensure high performance.
Kazanjian and Drazin (1987) similarly describe how successful
diversification requires a process of organizational learning. Through
diversification, firms enter new domains, and to be successful, managers must
acquire new knowledge. Other related research suggests that business unit
strategies require appropriate administrative relationships between the business
unit and the corporate central office if performance is to be enhanced (Gupta,
1987; Govindarajan, 1988). Taking a different perspective, Hill and Hoskisson
(1987) also examine the relationship between strategy and structure, and
8
Hoskisson (1987) concludes that firms pursuing market and product
diversification strategies enjoy higher performance when organized along a
multidivisional organizational structure.
Still other approaches have examined firm-specific knowledge and skills
which might be responsible for high performance. For example, Lippman and
Rumelt (1982) and Rumelt (1984) offer a theory of "uncertain imitability" in
which firms develop new production functions resulting in "firm heterogeneity
as an outcome rather than as a given" (Rumelt, 1984:562). According to this
model, unexpected events occur which are the source of potential rents. These
can include changes in technology, relative prices, consumer tastes, and laws
and regulations. The managers of successful firms are able to exploit these
changes, but do so in a way that leads to "causal ambiguity." As a result,
managers of competing firms in the same industries are uncertain as to how to
imitate the actions of these successful firms. Levels of performance can
therefore vary widely within the same industry.
Rumelt *s research supports this view. After analyzing the rates of return
on capital of 1,292 U.S. corporations over a 20 year period, he finds that "the
variance in long-run profitability within industries is three to five times
larger than the variance across industries" (1987:141).
Porter (1980, 1985) examines how firms and business segments can exploit
aspects of industry structure or the value chain to achieve competitive
advantage, and suggests that firms and business segments must pursue one of
three generic strategies — cost leadership, differentiation, or focus. Porter's
cost leadership strategy bears a remarkable similarity to the least cost
production techniques which characterized the so-called American System of
Manufactures. A more recent analysis suggests that problems of competitiveness
and the poor performance of many firms can be traced to an abandonment of these
least cost production techniques (Melman, 1983).
This underscores an important point — that we really know very little about
how high performing firms develop and implement strategies which lead to
competitive advantage. Yet, this would seem to be a very important influence
on the level of firm performance. This paper proposes that this newer, diverse
stream of strategic management literature points to management skill as a key
strategic variable, likely to be an important influence on firm performance.
We will argue that management skill is the reason why some firms consistently
enjoy levels of performance above industry averages. While this view is hardly
new as a theme in business policy and strategy, it has not been adequately
operationalized in quantitative research, and it has not been well linked to the
research stream explicitly concerned with performance.
TOWARD AN INTEGRATION OF THESE PERSPECTIVES
This review has sought to describe the major themes of three literatures.
Yet each seems to have its limitations. Schmalensee' s findings are remarkable,
but his analysis largely ignores the highly diversified nature of large firms.
Furthermore, we would disagree with his view of "relatively stable patterns of
A number of studies examine the importance of efficiency and productivity
in the growth of the United States' economy during the nineteenth and early
twentieth centuries. These studies emphasize how the impact of the American
System of Manufactures with an emphasis on least cost production techniques
resulted in the spectacular growth experienced by the United States economy.
See for example, North (1961), Rosenberg (1969), Layton (1973), David (1975),
Mayr and Post (1981), and Hounshell (1984).
10
activity at the firm level" (1985i349). Instead of the "relatively stable
patterns" described by Schmalensee, we see continuous acquisition and divestment
activity (Duhaime & Grant, 1984; Porter, 1987), suggesting that much more is
involved in managing a large diversified firm than one-time selection of the
right industries in which to participate.
On the other hand, not only has diversification research been plagued by
inconsistent findings, but diversification studies implicitly assume that firms
have equal abilities at developing and implementing strategies — an assumption
that does not seem realistic. This is a limitation we see in the recent study
by Werner felt and Montgomery (1988). Wernerfelt and Montgomery extend
Schmalensee ' s study and find that not only are industry effects important in
explaining differences in firm performance, but that firm focus (the extent of
diversification) is also important in explaining differences in performance —
that narrowly diversified firms enjoy higher performance than widely diversified
firms. •
Wernerfelt and Montgomery use a resource-based view of the firm
(Wernerfelt, 1984) to explain this finding, suggesting that narrowly diversified
firms are better able to transfer competencies and resources among business
segments. This resource-based view is appealing, but it ignores the
difficulties of implementing strategies which capture the benefits of
transferring resources among business segments. Duhaime and Grant (1984) and
Porter (1987) note the widespread acquisition and divestment activity of large
firms. These studies suggest that firms may find the transfer of competencies
and resources to be very difficult.
The managerial control literature offers an explanation for why synergies
are so elusive. Hamermesh (1977), for example, argues that information, and
11
especially "bad news" moves very slowly through organizations. Business segment
managers who detect unfavorable environmental circumstances have every incentive
to prevent this information from flowing to the central offices of large firms.
Furthermore, when information about or from a business segment does arrive at
the central office, senior managers there may face major challenges in
comprehending information and integrating this information with relevant facts
about the market conditions in which the various business segments operate. In
some large firms, these information lags are likely to constrain attempts to
share resources across business segments.
All of these considerations suggest that the level of management skill is
likely to be a key influence on firm performance. We agree with Grant, Jammine,
and Thomas when they conclude that the "total impact of diversification on
performance depends on complex interactions between diversification strategy,
corporate capabilities and resources, and external environment" (1988:795). A
key element of these corporate capabilities, in our view, is the level of
management skill. The managers of high performing firms are likely to have a
much better developed understanding of the cause-effect relationships which lead
to success in a particular industry or market; they also understand how to
coordinate and integrate the activities of large, multibusiness firms
successfully.
A similar conclusion is reached by Hansen and Wernerfelt who examine
economic and organizational influences on performance, and find both to be
significant. Using data gathered from questionnaires, they examine two
variables, emphasis on human resources and emphasis on goal accomplishment, to
assess organizational influences. They conclude "that the critical issue in
firm success and development is not primarily the selection of growth industries
12
or product niches, but it is the building of an effective, directed human
organization in the selected industries" (1989:409).
The dominant logic described by Prahalad and Bettis and the organizational
learning described by Kazan jian and Orazin have so far remained conceptual.
Furthermore, while Hitt and Ireland (1986) and Snow and Hrebiniak (1980) assess
the relationship between corporate level distinctive competencies and firm
performance using questionnaires to assess distinctive competencies, we know of
little additional research which has sought to analyze explicitly the
relationship between management skill and performance.
We believe that firms with higher levels of management skill will enjoy
either lower costs or higher prices than rivals in the same industries.
Specifically, a high level of management skill will enable a firm to operate
more efficiently than its rivals either because the firm is better able to
transfer competencies and resources among its business activities or because the
firm is better able to manage information requirements. Alternatively, a high
level of management skill might also enable a firm to better exploit
environmental and technological changes. This would permit the firm to
implement new strategies, and offer new products or services at a premium price,
thereby enjoying a higher gross margin than rivals.
Therefore, we believe that management skill can be represented by the
difference between a firm's gross margin and the average gross margins of the
markets in which that firm operates. Our interest is in how a firm's gross
margin compares with the gross margins of other firms and business units
operating in the same industries or markets. We believe that management skill
is a necessary condition to achieve a high gross margin relative to rivals, and
13
so our measure — gross margin adjusted for industry membership — is a good proxy
for management skill.
RESEARCH PROPOSITIONS
This research study has a number of aims. We want to examine whether a
significant relationship exists between the new variable, comparative gross
margin as a proxy for management skill, and firm performance. We also want to
re-examine the influence of industry membership and diversification strategy on
performance using different data and measures. Finally, integrating these three
perspectives, we want to examine the relative influence of industry membership,
diversification strategy, and management skill on firm performance.
More specifically, this research examines the following propositions:
1) The choice of industry will have a significant influence
on the level of firm performance.
•
2) The choice of diversification strategy or the extent of
diversification may or may not have a significant
influence on the level of firm performance, but
3) management skill, as measured by the gross margin
adjusted for industry membership, will have a
significant influence on the level of firm performance.
METHODOLOGY
Data and Samples
All data required for this study were gathered from the Compustat
database. This database consists of financial and market performance data for
over 6000 firms. The database also includes financial data on the business
14
segments of these firms as required by the Financial Accounting Standards
Board's (1988) Statement of Financial Accounting Standards No. 14, "Financial
Reporting for Segments of a Business Enterprise."
We identified all of the firms in the 1989 Fortune "500" for which data
were available for the years 1984 through 1988. Most of the existing
diversification literature draws on samples which include data from the 1970s
and early 1980s, a time of business and economic volatility. The time frame
covered in this study (1984 through 1988), is marked by continuous economic
expansion, avoiding periods of wide cyclical and inflationary variations.
We created two samples — first, a sample of those firms which reported
results for two or more business segments during each of the five years 1984
through 1988, and a second sample consisting of the firms in the first sample
plus firms that reported results for only one business segment during the same
five year period. We felt this distinction was important, especially after
reviewing the papers by Schmalensee and Werner felt and Montgomery.
Schmalensee' s sample consisted of only multibusiness firms, while Wernerfelt and
Montgomery's sample, drawn from a sample developed by Lindenberg and Ross
(1981), consisted of both single and multibusiness firms. It is possible that
the more diverse sample used by Wernerfelt and Montgomery may have influenced
their findings. We therefore conducted our empirical tests on the two samples;
one consisting only of multibusiness firms (like Schmalensee), the other more
diverse, including both single and multibusiness firms (like Wernerfelt and
Montgomery) .
While samples drawn from the Fortune "500" are certainly not
representative of the entire population of business enterprises which remains
overwhelmingly atomistic, the largest industrial corporations do account for a
15
very large share of total business activity. Throughout the 1980s, for example,
the sales revenues of Fortune "500" firms have accounted for over 40 percent of
the total gross national product (Abelson & Jacob, 1989). As a result, an
interest in the factors influencing the performance of these large firms is
certainly warranted. While samples drawn from the Fortune "500" would be
inappropriate for some research questions, the issues raised in this paper would
seem to warrant use of samples drawn from this population.
For purposes of this study, industry is defined by four-digit SIC code.
We realize that industry is an elusive concept and that any definition is likely
to have advantages as well as limitations. One key advantage of defining
industries by four-digit SIC codes is that the Compustat database provides
aggregate data for nearly 300 industries defined by four-digit SIC codes. In
addition, defining industry by four-digit SIC code avoids the pitfalls of
defining industries too broadly. Particularly for the research questions raised
in this paper, a narrower definition of industry* is more conservative than a
broader definition.
Variables and Procedure
We used return on assets (ROA), where ROA is net income as a proportion
of total assets, to assess firm performance. While a variety of other
accounting and market measures could conceivably have been used to assess firm
performance, we agree with Holzmann, Copeland, and Hayya (1975) that ROA is
widely viewed and accepted by managers as a measure of firm performance and the
success of business strategies. Furthermore, Schmalensee and Hansen and
Wernerfelt also use rate of return measures to assess performance.
16
The influence of industry effects is assessed using average industry
return on assets (INDROA). Since large, multibusiness firms are likely to be
active in more than one industry, we felt the need to first identify the
industries in which our sample firms operate, and then determine the proportion
of each firm's activity in each industry. To do this, we identified from the
Compustat database the primary SIC codes of each firm's business segments. We
then calculated for each firm the weighted average of the industry ROAs for the
industries represented by these business segment SIC codes. The weighted
average was based on each segment's proportion of the firm's total sales.
Diversification (DIV) is assessed using a continuous measure of developed
by Davis and Duhaime (1989). Similar to the entropy measure developed by Palepu
(1985), this is a continuous measure which uses SIC classifications to identify
and evaluate the extent of diversification. . The Davis and Duhaime measure is
particularly useful for this study because it uses business segment data
available on the Compustat database to measure diversification.
Management skill (SKILL) is evaluated as each firm's gross margin adjusted
for average industry gross margin, where gross margin is operating income after
depreciation as a proportion of sales. Again, because multibusiness firms are
likely to be active in more than one industry, industry gross margin was
The extent of diversification (DIV) is the sum of measures for related
diversification (DR) and unrelated diversification (DU), where
DR = 2 {[ (SEGSALES/GRPSALES)*ln(GRPSALES/SEGSALES) ] * ( SEGSALES/TOTSALES ) }
DU =» Z [ (GRPSALES/T0TSALES)*ln(TOTSALES/GRPSALES) ]
where SEGSALES is sales for each segment of each company as reported by
Compustat, GRPSALES is total sales for all segments which share the same two-
digit SIC code in each company, and TOTSALES is total sales of each company.
17
calculated the same way we calculated INDROA. The SKILL variable then is the
firm's gross margin less this composite industry gross margin.
Two variables in this study, INDROA and SKILL, require the use of industry
averages. Other studies requiring firms' industry averages have used the
industry average of the primary or largest business segment. Since conditions
and performance levels can vary widely across the industries in which
multibusiness firms compete, this is an incomplete and possibly misleading
industry average for multibusiness firms. Our construction of composite
industry averages which are weighted averages of all industries in which a
multibusiness firm competes gives us greater confidence in the validity of our
results than if we had used previous methods.
Missing data reduced our multibusiness sample to 268 firms and our single
and multibusiness sample to 329 firms. Sample observations are five year
averages. Summary statistics and correlation matrices for these variables for
the sample of multibusiness firms and the sample of single and multibusiness
firms are shown in Table 1.
Insert Table 1 about here
Building on the work of Schmalensee, Wernerfelt and Montgomery, and Hansen
and Wernerfelt, we developed the following descriptive model:
ROA » bQ + b,( INDROA) + b2(DIV) ♦ bj( SKILL)
We tested this model on the two samples (multibusiness firms and single and
multibusiness firms). We then tested a number of sub-models, imposing various
restrictions, excluding one or more variables from the model.
18
RESULTS
The results for the sample of multibusiness firms are illustrated in
Figure 1, and the results for the sample of single and multibusiness firms are
illustrated in Figure 2. This form of presentation is identical to that used
by Schmalensee, Wernerfelt and Montgomery, and Hansen and Wernerfelt. In each
figure, results for the full model are shown at the bottom of the figure, and
the results of various restricted models are shown above this full model. As
with the earlier articles, the arrows correspond to restrictions excluding one
of the three effects, and the numbers next to the arrows are the probabilities
(P levels) at which an F-test would reject these restrictions.
Insert Figures 1 & 2 about here
The results of both tests confirm our propositions. First, note that the
R-square values of the full models in Figures 1 and 2 are quite high.
Furthermore, note that the very low P levels generated by tests for industry
effects (arrows pointing to the right) and management skill (arrows pointing to
the left) indicate the presence of both industry and management skill effects.
As in Schmalensee, the results for industry effects are quite strong — always
significant at the .0001 level. The management skill effects, however, are also
very strong — again, always significant at the .0001 level.
The results for diversification effects are also interesting. In the
sample of multibusiness firms, the high P levels indicate that diversification
effects are either not present or not particularly significant. In the sample
of single and multibusiness firms, however, the low P levels indicate that
19
diversification effects are present. Specifically, in these models higher
levels of diversification are associated with lower levels of firm performance.
This result closely conforms to the findings of Wernerfelt and Montgomery who,
as already noted, used a sample which included both single and multibusiness
firms.
Table 2 shows the incremental contribution of each effect to the adjusted
R-square of the full model for each sample. These values represent the
difference between the adjusted R-square of the full model and the adjusted R-
square of models with the effect of interest removed. The table illustrates
that the incremental contributions to the R-square made by management skill and
industry membership are roughly equal, while diversification makes a small
contribution to the R-square, but only in the sample of single and multibusiness
firms.
Insert Table 2 about here
DISCUSSION AND IMPLICATIONS
Like Schmalensee, Wernerfelt and Montgomery, and Hansen and Wernerfelt
before us, we find that industry effects are a major influence on firm
performance. Similarly, like Wernerfelt and Montgomery, we find that in a
sample of single and multibusiness firms, diversification effects are also an
influence on firm performance. The major contribution of this study, however,
is the introduction of a new variable to assess the importance of management
skill effects. This variable, comparative gross margin as a proxy for
20
management skill, proved to be a very highly significant influence on the level
of firm performance. In fact, the results reported here suggest that the
management skill effects are as important as industry effects in influencing
firm performance.
To the extent that our variable, management skill, reflects the quality
of management, these results are very reasonable. Selection of the industries
or markets in which to compete is likely to be an important influence on
performance, but just as important is the level of skill that management brings
to these industries. These results suggest that Prahalad and Bettis (1986) and
Kazan jian and Drazin (1987) are correct — movement into new markets will
influence performance, but these relationships are likely to be strongly
moderated by the level of management skill. This is why we see great variation
in the levels of performance enjoyed by both single business and highly
diversified firms, even after controlling for industry effects.
Firms enjoying high levels of performance, whether single business firms
or highly diversified firms, are much more likely to possess the requisite
skills and expertise to be effective in their market or markets. Our analysis
suggests that management skill effects are an important influence, and certainly
much more influential than diversification effects. As a result, we agree with
Ramanujam and Varadarajan (1989) that much of the recent diversification
literature has been incremental at best, and that continued traditional
diversification research (i.e. research assessing the relationship between
diversification strategy and performance) will lead to relatively few new
insights.
While management skill has enjoyed a central place in the field of
strategic management, it has been difficult to operational ize in the empirical
21
literature. We feel confident that comparative gross margin taps the presence
of management skill. It is also significant that we have been able to establish
the importance of this variable using public financial data. This variable may
have many additional applications. For example, the strategic groups and
governance literatures have typically used public financial data, but have not
explicitly incorporated management skill as a variable in studies done to date.
One possible application would be to use management skill as a way to
distinguish among firms within an industry's strategic groups.
As a result of the study, we feel that future research efforts in
strategic management should further examine management skill effects. For
example, more research is needed to assess what constitutes management skill,
how management skill is acquired, the relationship between management skill and
managerial characteristics, what management skills are needed in multibusiness
firms, and how a particular repertoire of skills moderates the relationships
between industry membership and diversification on the one hand and firm
performance on the other. To pursue these new research directions, qualitative
and field research methods may prove both necessary and worthwhile adjuncts to
data sets such as ours.
22
REFERENCES
Abelaon, R. , & Jacob, R. 1989. The biggest blowout ever. Fortune. 119(9):
346-353.
Bain, J. S. 1951. Relation of profit rate to industry concentration: American
manufacturing, 1936-1940. Quarterly Journal of Economics. 65(3): 293-
324.
Bain, J. S. 1956. Barriers to new competition. Cambridge, Massachusetts:
Harvard University Press.
Bettis, R. A. 1981. Performance differences in related and unrelated
diversified firms. Strategic Management Journal. 2(4): 379-394.
Bettis, R. A., & Hall, W. K. 1982. Diversification strategy, accounting
determined risk, and accounting determined return. Academy of Management
Journal. 25(2): 254-64.
Christensen, H. K. , & Montgomery, C. A. 1981. Corporate economic performance:
Diversification strategy versus market structure. Strategic Management
Journal. 2(4): 327-344.
David, P. A. 1975. Technical choice. Innovation and economic growth: Essays
on American and British experience in the nineteenth century. Cambridge,
England: Cambridge University Press.
Davis, R. , & Duhaime, I. M. 1989. Business level data disclosed under FASB No.
14: Effective use in strategic management research. In F. Hoy (Ed.),
Academy of Management Proceedings: 7-11. Washington: Academy of
Management.
Duhaime, I. M., & Grant, J. H. 1984. Factors influencing divestment decision-
making: Evidence from a field study. Strategic Management Journal. 5(4):
301-318.
Financial Accounting Standards Board. 1988. Accounting standards: Current
text (1988/89 ed. ) . Homewood, Illinois: Irwin.
Gahlon, J. M. , & Stover, R. D. 1979. Diversification, financial leverage and
conglomerate systematic risk. Journal of Financial and Quantitative
Analysis. 14(5): 999-1013.
Govindarajan, V. 1988. A contingency approach to strategy implementation at
the business-unit level: Integrating administrative mechanisms with
strategy. Academy of Management Journal. 31(4): 828-853.
Grant, R. M. , Jammine, A. P., & Thomas, H. 1988. Diversity, diversification,
and profitability among British manufacturing companies, 1972-84. Academy
of Management Journal. 31(4): 771-801.
23
Gupta, A. K. 1987. SBU strategies, corporate-SBU relations, and SBU
effectiveness in strategy implementation. Academy of Management Journal,
30(3): 477-500.
Hamermesh, R. O. 1977. Responding to divisional profit crises. Harvard
Business Review, 55(2): 124-130.
Hansen, G. S., & Wernerfelt, B. 1989. Determinants of firm performance: The
relative importance of economic and organizational factors. Strategic
Management Journal. 10(5): 399-411.
Hill, C. W. L.f & HosJcisaon, R. B. 1987. Strategy and structure in the
multiproduct firm. Academy of Management Review. 12(2): 331-341.
Hitt, M. A., & Ireland, R. D. 1986. Relationships among corporate level
distinctive competencies, diversification strategy, corporate structure
and performance. Journal of Management Studies. 23(4): 401-416.
Holzmann, O. J., Copeland, R. M. , & Hayya, J. 1975. Income measures of
conglomerate performance. Quarterly Review of Economics and Buaineas,
15(3): 67-78.
Hoskisson, R. E. 1987. Multidivisional structure and performance: The
contingency of diversification strategy. Academy of Management Journal,
30(4): 625-644.
Hounshell, D. A. 1984. From the American system to mass production: 1800-
1932. Baltimore: The Johns Hopkins University Press.
Kazanjian, R. K., & Drazin, R. 1987. Implementing internal diversification:
Contingency factors for organization design choices. Academy of
Management Review, 12(2): 342-354.
Layton, E. T., Jr. (Ed.). 1973. Technology and social change in America. New
York: Harper and Row.
Lindenberg, E. B., & Ross, S. A. 1981. Tobin's "q" ratio and industrial
organization. Journal of Business, 54(1): 1-32.
Lippman, S. A., & Rumelt, R. P. 1982. Uncertain instability: An analysis of
inter firm differences in efficiency under competition. Bell Journal of
Economics, 13(2): 418-438.
Lubatkin, M. 1987. Merger strategies and stockholder value. Strategic
Management Journal. 8(1): 39-59.
Mason, R. H. , & Goudzwaard, M. B. 1976. Performance of conglomerate firms:
A portfolio approach. Journal of Plnance, 31(1): 39-48.
Mayr, O. , & Post, R. C. (Eds.). 1981. Yankee enterprise: The rise of the
American system of manufactures. Washington: Smithsonian Institution
Press.
24
Melicher, R. , & Rush, D. F. 1973. The performance of conglomerate firms:
Recent risk and return experience. Journal of Finance, 28(2): 381-388.
Melman, S. 1983. Profits without production. New York: Alfred A. Knopf.
Michel, A., & Shaked, I. 1974. Does business diversification affect
performance? Financial Management. 13(4): 18-25.
Mueller, D. C. 1983. The determinants of persistent profits. Washington, DC:
United States Federal Trade Commission.
North, D. C. 1961. The economic growth of the United States: 1790-1860. New
York, New York: W. W. Norton and Company, Inc.
Palepu, K. G. 1985. Diversification strategy, profit performance, and the
entropy measure. Strategic Management Journal. 6(3): 239-255.
Porter, M. E. 1980. Competitive strategy. New York: The Free Press.
Porter, M. E. 1985. Competitive advantage. New York: The Free Press.
Porter, M. E. 1987. From competitive advantage to corporate strategy. Harvard
Business Review. 65(3): 43-59.
Prahalad, C. K. , & Bettis, R. A. 1986. The dominant logic: A new linkage
between diversity and performance. Strategic Management Journal, 7(6):
485-502.
Ramanujam, V., & Varadarajan, P. 1989. Research on corporate diversification:
A synthesis. Strategic Management Journal. 10(6): 523-551.
Rhoades, S. A. 1973. The effect of diversification on industry profit
performance in 241 manufacturing industries: 1963. Review of Economics
and Statistics. 55(2): 146-155.
Rosenberg, N., (Ed.). 1969. The American system of manufactures. Edinburgh,
Scotland: Edinburgh University Press.
Rumelt, R. P. 1974. Strategy, structure and economic performance. Cambridge,
Massachusetts: Harvard University Press.
Rumelt, R. P. 1982. Diversification strategy and profitability. Strategic
Management Journal. 3(4): 359-370.
Rumelt, R. P. 1984. Towards a strategic theory of the firm. In R. B. Lamb
(Ed.), Competitive strategic management: 556-570. Englewood Cliffs, New
Jersey: Prentice-Hall, Inc.
25
Rumelt, R. P. 1987. Theory, strategy, and entrepreneurship. In D. Teece
( Ed . ) , The competitive challenge: Strategies for Industrial innovation
and renewal: 137-158. Cambridge, Massachusetts! Ballinger Publishing
Company .
Scherer, F. M. 1980. Industrial market structure and economic performance.
Chicago: Rand McNally College Publishing Company.
SchmalenBee, R. 1985. Do markets differ much? American Economic Review.
75(3)t 341-351.
Snow, C. C. , & Hrebiniak, L. G. 1980. Strategy, distinctive competence, and
organizational performance. Administrative Science Quarterly, 25(2):
317-336.
Werner felt, B. 1984. A resource-based view of the firm. Strategic Management
Journal, 5(2): 171-180.
Wernerfelt, B., & Montgomery, C. A. 1988. Tobin's q and the importance of
focus in firm performance. American Economic Review, 78(1): 246-250.
Weston, J. F., Smith, K. V., & Shrieves, R. E. 1972. Conglomerate performance
using the capital asset pricing model. Review of Economics and
Statistics. 54(4): 357-363.
Wrigley, L. 1970. Divisional autonomy and diversification. Unpublished
doctoral thesis, Harvard Business School.
26
TABLE 1
Summary Statistics and Correlation Matrix
for the Sample of Multibusiness Finns (N=268)
Variable
Wean
.05063
Std. Dev.
.05493
Minimum
Maximum
ROA
-.40905
.20165
INDROA
.04744
.02841
-.06983
.10632
DIV
.80569
.32480
.03567
1.82378
SKILL
.00334
.03652
-.16025
.14104
ROA
INDROA
DIV
SKILL
ROA
1.00000
INDROA
.40833
1.00000
DIV
-.09465
-.13515
1.00000
SKILL
.34523
-.07687
.02770
1.00000
Summary Statistics and Correlation Matrix
for the Sample of Single and Multibusiness Firms (N=329)
Variable
Mean
Std. Dev.
Minimum
Maximum
ROA
.05486
.05680
-.40905
.23542
INDROA
.04881
.02864
-.06983
.10840
DIV
.65631
.42921
.00000
1.82378
SKILL
.00455
.03970
-.16025
.20681
ROA
INDROA
DIV
SKILL
ROA
1.00000
INDROA
.46219
1.00000
DIV
-.17066
-.15601
1.00000
SKILL
.44226
-.00756
-.03137
1.00000
27
ECKMAN
IDERY INC.
JUN95
To.FW N. MANCHESTER.
INDIANA 46962