Beruflich Dokumente
Kultur Dokumente
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
The Circulation of We develop a conceptual model of the circulation of cor-
Corporate Control: porate control-the instability in formal authority at the
top of large corporations. According to our model, chief
Selection of Functional
executive officer (CEO) selection is both a political con-
Backgrounds of New test for the top executive position and an ideological
CEOs in Large U.S. struggle among members of the firm's political coalition
Manufacturing over defining the corporate agenda and strategy. We ap-
Firms, 1981-1992 ply the model to analyze the selection of functional back-
grounds of 275 new CEOs in large U.S. manufacturing
firms from 1981 to 1992. Results show that the circula-
William Ocasio tion of control adds to previous explanations based on
Northwestern University
strategic contingencies and institutional isomorphism.
Hyosun Kim Furthermore, we find evidence of an ideological and po-
Massachusetts Institute of litical obsolescence of financial CEOs and a change in the
Technology strategic contingencies that previously favored finance
and the financial conception of control. Results suggest
that conceptions of control are best understood as styles,
rather than institutions, as these conceptions are neither
taken for granted nor isomorphic across industries or the
manufacturing sector, but are subject to sectoral varia-
tion, contestation, and change.'
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
problems come to the forefront and require attention in the
firm's environment (March, 1962; Cyert and March, 1963;
Ocasio, 1997). Which individuals and groups are in positions
of authority over the dominant coalition over the long run is
inherently unstable, as the outcome of recurring political
struggles will lead to sequential attention to the firm's issues
and problems, with resulting fluctuations in executive power.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
the corporation and what the critical problems are that the
organization must face. According to this view, resource de-
pendencies and structural contingencies are not objective
criteria but are historically situated social constructions that
are themselves the subject of conflict between contending
factions.4
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Table 1 summarizes key differences between institutions
and styles. White distinguished styles from institutions as
the former emerge under conditions of ambiguity and are
maintained by the structural positions of actors with a par-
ticular style. Styles, unlike institutions, are not taken for
granted but are subject to emergent contests for control
among actors with different styles. While both institutions
and styles provide cultural frames to guide organizational ac-
tion, they vary in the strength of their cultural assumptions,
the degree to which they are isomorphic across organization
fields, and in how they shape organizational inertia. Institu-
tions are relatively inert, and change is exogenous, infre-
quent, and episodic. Styles, in contrast, are subject to fre-
quent endogenous change and short-term fluctuations. The
greater cultural control that characterizes institutions relative
to styles is reflected in differences in the sources of power,
the stability of dominant elites, and the degree of elite hege-
mony over the organization and organizational field.
Table 1
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
ness of American industry to the excessive reliance on
portfolio management practices and short-term financial con-
trols at the expense of long-term investments in technology
and production. They associated this financial orientation
with the rise of finance (and legal) executives to CEO posi-
tions in large manufacturing companies. The conglomerate
movement of the 1960s and the portfolio management tech-
niques that became popular in the 1970s were presented as
reasons for "managing our way to economic decline" (Hayes
and Abernathy, 1980: 67).
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
Hypotheses
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
functional subunits. This suggests, during the period studied,
that finance CEOs became technically obsolete with a de-
cline in their ability to utilize prevailing conceptions of control
to meet changing organizational contingencies. Conse-
quently, the technical obsolescence of finance CEOs led to a
decline in their ability to gain power in firms with related or
unrelated diversification, multidivisional structures, or high
levels of merger and divestiture activity. Our predictions here
are consistent with Fligstein's (1987), who found that the
size and statistical significance of these variables varied dur-
ing different time periods in which the various conceptions
of control became dominant, consistent with the model of
the circulation of corporate control. We use these hypoth-
eses to examine whether Fligstein's findings of the domi-
nance of the financial conception of control from 1959 to
1979 can be replicated for the subsequent time period or
whether, consistent with technical obsolescence and the
circulation of corporate control, the challenges to the finance
conception limited the ability of these strategic contingen-
cies to favor the selection of new CEOs from financial back-
grounds:
Hypothesis 1: During the 1980s and early 1990s, the strategic con-
tingencies associated with the finance conception of control-multi-
divisional structures, related and unrelated diversification, and high
numbers of mergers and acquisitions-did not lead to increased
selection of new CEOs from finance, relative to other backgrounds.
Hypothesis 2: During the 1 980s and early 1 990s, the greater the
proportion of foreign imports at the industry level, the greater the
selection of production and operations CEOs relative to those from
finance.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Hypothesis 5: The likelihood of selection of a CEO from a particular
background will be directly related to the proportion of CEOs from
the same background selected by other firms in the sector during
the same year.
Sample
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
Measures
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
strategic contingencies affecting a CEO's functional back-
ground. Strategy and structure were identified using informa-
tion from annual reports and 1OKs. Following Rumelt (1974)
and Fligstein (1987), we coded corporate strategy as prod-
uct-dominant, related diversification, and unrelated diversifi-
cation (conglomerate) strategy. A firm was classified as prod-
uct-dominant when it had one line of business accounting
for more than 70 percent of revenues. Firms with related
diversification strategies had multiple lines of related busi-
nesses, with no single business accounting for 70 percent of
revenue. Firms with unrelated diversification strategies were
engaged in unrelated business, and no one business ac-
counted for 70 percent of revenue. These strategies were
coded as dummy variables, with product-dominant as the
omitted category.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
the determination of who controls the top executive posi-
tion. The functional backgrounds of previous CEOs were col-
lected the same way as functional backgrounds of new
CEOs. Operations and all other backgrounds of previous
CEOs was the omitted category.
Table 2
* The approximate cutoff for significance at the .05 level applies to correlations equal to or greater than .12 or equal to
or less than -.12.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
Modeling Procedure
RESULTS
Table 2 (continued)
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
-.19
-.30 -.35
.12 .07 -.15
-.28 .12 -.09 -.25
-.17 -.01 .08 -.87 .52
-.1 1 -.1 1 .12 -.08 .14 .07
.15 .15 -.32 .09 -.10 -.10 -.24
-.01 .11 -.05 .07 -.05 -.03 -.66 .03
-.05 -.12 .24 -.06 .01 .06 -.24 -.78 -.04
-.10 -.05 -.05 -.05 .03 .05 .07 -.1 1 -.02 .06
.18 -.01 .04 .08 -.20 -.12 -.11 .09 .11 -.06 -.30
-.08 .01 .01 -.01 .08 .02 .09 .00 .00 -.08 -.24 -.22
-.07 .06 .09 -.07 .08 .08 -.04 -.07 -.04 .12 -.36 -.32 -.26
.03 -.05 -.08 -.02 -.05 -.02 -.07 .04 .03 .01 .25 -.10 -.04 -.08
.11 -.07 .10 .11 -.09 -.14 -.06 -.07 .02 .10 -.22 .21 -.02 .10 -.28
-.04 .05 -.01 .05 .07 .02 .07 -.03 -.03 -.01 -.10 -.01 .09 .06 -.20 -.21
.05 .00 .00 -.08 .02 .09 .14 .00 -.09 -.07 .05 -.06 -.01 .01 -.27 -.30 -.22
.10 -.09 .16 -.14 .15 .15 .07 -.43 .02 .42 .06 -.03 -.07 .16 .07 .16 .01 -.12
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Table 3
Functional
backgrounds 1961 1971 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992
Production and
technical 24% 19% 20% 21% 22% 21% 22% 23% 23% 24% 25% 25% 25% 24%
Marketing and
sales 23% 22% 21% 23% 22% 22% 22% 20% 19% 18% 18% 18% 18% 18%
Finance and
legal 33% 31% 34% 30% 27% 26% 26% 23% 21% 21% 21% 22% 22% 21%
Finance 19% 19% 22% 20% 18% 18% 17% 17% 16% 16% 16% 17% 17% 16%
Legal 14% 12% 12% 10% 9% 8% 9% 6% 5% 5% 5% 5% 5% 5%
Operations and
others 20% 28% 24% 27% 29% 30% 31% 34% 37% 37% 36% 36% 36% 37%
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
Table 4
Total 70 57 43 25 80 275
Table 5
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
providing evidence of entrenchment at the organizational
level. No entrenchment was found, however, for CEOs from
finance and legal or operations and other backgrounds. The
finance and legal backgrounds, which had been dominant in
the 1960s and 1970s were not able to maintain control over
the top command position in the firm during the 1980s and
early 1 990s, consistent with the model of the circulation of
corporate control. The log linear analysis also shows that the
stability of functional backgrounds was significantly greater
for production and marketing than for other backgrounds.
Notably, operations backgrounds, which are now relatively
prevalent, were also not able to entrench themselves during
the period studied.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
Table 6
Independent Prod. & Mkt. & Chi-square Prod. & Mkt. & Chi-square
variable tech. sales Finance contrast tech. sales Finance contrast
Strategic contingencies
Multidivisional 0.334 -0.344 -0.054 2.23 0.326 -0.899 -0.241 1.20
(0.405) (0.408) (0.453) (0.447) (0.487) (0.494)
Related 0.611 0.246 -0.036 2.69 0.639 -0.574 0.351 4.61
diversif. (0.409) (0.435) (0.471) (0.451) (0.542) (0.526)
Unrelated -0.030 0.087 -0.984 2.93 -0.325 -0.240 0.874 1.86
diversif. (0.462) (0.491) (0.620) (0.503) (0.566) (0.653)
Mergers & -0.379-- -0.362-- -0.012 9.4600 -0.362-. -0.419-- 0.009 8.3700
divestitures (0.157) (0.158) (0.155) (0.172) (0.194) (0.176)
Log of - - - - -0.226 -0.38- 0.020 3.89
imports (0.192) (0.227) (0.246)
ROA - - - - 19.591000 0.507 14.4 10.1400
(6.643) (6.280) (8.232)
ROA * log of - - - - 6.61000 0.134 4.123 8.9200
imports (2.390) (2.207) (3.096)
Isomorphism
% Production 4.932000 -2.796 4.20100 17.44-- 4.44100 -0.652 3.734- 8.2300
& technical (1.666) (1.843) (2.061) (1.813) (2.079) (2.206)
% Marketing 2.974- 2.327 2.863 4.42 3.246- 1.844 2.805 3.82
& sales (1.679) (1.705) (2.053) (1.839) (1.929) (2.129)
% Finance -0.049 2.455 0.688 1.97 -0.697 2.413 -1.847 2.41
(2.000) (1.874) (2.355) (2.167) (2.226) (2.606)
Annual fluctuations
Annual percentage - - - - 2.827 4.858 3.502 2.97
prod. & tech. (2.553) (3.349) (3.220)
Annual percentage - - - - 1.054 8.096- 3.995 12.40'''
mkt. & sales (1.986) (2.413) (2.439)
Annual percentage - - - - 2.009 6.251 11.017- 6.28
finance (3.808) (4.768) (4.708)
Political dynamics
Prior CEO - - - - 0.95300 0.130 -0.880 9.9200
prod. & tech. (0.427) (0.634) (0.665)
Prior CEO - - - - -1.222- 0.918 -0.600 10.64-
mkt. & sales (0.595) (0.513) (0.576)
Prior CEO - - - - -0.532 0.486 -0.382 2.38
finance (0.532) (0.595) (0.546)
Second prior CEO - - - - -0.615 -0.406 -1.867-- 7.100
prod. & tech. (0.463) (0.599) (0.729)
Second prior CEO - - - - 0.107 1.790- -0.674 15.28-
mkt. & sales (0.552) (0.551) (0.637)
Second prior CEO - - - - 0.350 0.531 -0.723 1.05
finance (0.534) (0.690) (0.615)
Size
Log of assets 0.449000 0.35600 -0.058 11.71000 0.486000 0.217 0.180 7.41-
(0.157) (0.180) (0.197) (0.180) (0.217) (0.214)
Multiple CEO 0.110 0.215 0.417 1.14 -0.362 -0.419 0.009- 4.63
successions (0.347) (0.394) (0.406) (0.172) (0.194) (0.176)
Constant -5.832-- -3.346-- -2.019 -7.590---- -7.803000 -6.074--
(1.575) (1.655) (1.893) (2.155) (2.679) (2.684)
Chi-square 60.16 165.15 104.99--
D.f. 27 63 36
Op <.10; *-p < .05; 000p < .01; *---p < .001.
* Standard errors are in parentheses.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
operations backgrounds as CEOs from finance. The effects
of mergers and divestitures remain in model 2, with the vari-
able remaining significant at the .05 level.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
the baseline category) are statistically significant at only the
.10 level, the effects on marketing and sales relative to both
production and technical and finance CEOs are statistically
significant at the .05 level. This suggests that the persis-
tence of marketing CEOs is more likely to result in fewer
production and finance CEOs, rather than fewer CEOs from
operations and other backgrounds. At the corporate level,
the subunit power of production and technical and marketing
and sales become entrenched, as the subunit once in power
is more likely to maintain the position of CEO. For CEOs
with prior backgrounds in finance, the differences between
the categories are not statistically significant.
The results of the prior CEO variable provide support for hy-
pothesis 6a, which states that during the period studied,
CEOs from financial backgrounds will be less likely to be
succeeded by CEOs from the same background relative to
those from other backgrounds. To test this hypothesis for-
mally, we examine the differences between the coefficients
of entrenchment for production and technical and finance
CEOs (.953 and -.382, respectively). The chi-square contrast
between model 2 and a constrained model that equates the
value of the two coefficients is significant at the .01 level.
The chi-square contrast between model 3 and a constrained
model that equates the values of the prior marketing and
sales CEOs with the prior finance CEOs (with values of .918
and -.382, respectively, in model 2) is statistically significant
at the .05 level. Comparing the coefficients across equations
shows that stability in finance CEOs is significantly lower
and their impermanence in power is significantly higher than
for production and technical and marketing and sales CEOs.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
of results is inconsistent with the strong version of the
model of the institutionalization of power, in which political
dynamics lead to entrenchment of the power of functional
units at the level of the sector, the industry, or the firm. In-
stead, the overall pattern of results is more consistent with
the model of the circulation of control. We found, at the sec-
toral level, an overall decline in financial CEOs during the pe-
riod, a declining effect of strategic contingencies associated
with the finance conception of control, and annual fluctua-
tions in managerial fashions affecting the selection of func-
tional backgrounds of new CEOs. At the industry level, insti-
tutional isomorphism effects were only found for CEOs with
production and technical backgrounds. At the organizational
level, the ability of functional units to entrench themselves
varied among function, but these effects were significantly
lower for finance and operations than for production or
marketing.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
show instead that, during the period studied, the ability of
finance and operations CEOs to maintain their control and
command over formal authority was limited.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
are equally legitimate or that any individual style can gain
power in U.S. industrial corporations. To the degree that op-
posing functional groups within an organization are structur-
ally equivalent with members of incumbent groups, in-
creased circulation among contending styles is likely. In
many corporations, finance, production, marketing, and op-
erations executives below the rank of CEO or chief operating
officer are the structurally equivalent groups that are all part
of the firm's dominant coalition and potential candidates for
CEO. Managers and executives from other functions, human
resources, for example, are not structurally equivalent to the
dominant coalition members, their styles and conceptions of
control are less legitimate, and their exclusion from the CEO
position is highly institutionalized.
REFERENCES
Abrahamson, Eric Cyert, Richard M., and James G. DiMaggio, Paul D., and Walter W.
1996 "Management fashion." Acad- March Powell
emy of Management Review, 1963 A Behavioral Theory of the 1983 "The iron cage revisited: Insti-
21: 254-285. Firm. Englewood Cliffs, NJ: tutional isomorphism and col-
Prentice-Hall. lective rationality in organiza-
Boeker, Warren
1989 "The development and institu- Davis, Gerald F., Kristina A. Diek- tional fields." American
tionalization of subunit power mann, and Catherine H. Tinsley Sociological Review, 48: 147-
in organizations." Administra- 1994 "The decline and fall of the 160.
tive Science Quarterly, 34: conglomerate firm in the 1991 "Introduction." In Walter W.
388-410. 1980s: The deinstitutionaliza- Powell and Paul J. DiMaggio
tion of an organizational (eds.), The New Institutional-
Brealey, Richard A., and Stewart
form." American Sociological ism in Organizational Analysis:
C. Myers
Review, 59: 547-570. 1-38. Chicago: University of
1991 Principles of Corporate Fi-
Chicago Press.
nance, 4th ed. New York: Davis, Gerald F., and Suzanne K.
McGraw-Hill. Stout Donaldson, Gordon
1992 "Organization theory and the 1994 Corporate Restructuring: Man-
Cannella, Albert A., and Michael
market for corporate control." aging the Change Process
Lubatkin
Administrative Science Quar- from Within. Boston: Harvard
1993 "Succession as a sociopolitical
terly, 37: 605-633. Business School Press.
process: Internal impediments
to outsider selection." Acad- Deal, Terrence E., and Allan A. Fligstein, Neil
emy of Management Journal, Kennedy 1985 "The spread of the multidivi-
37: 763-793. 1982 Corporate Cultures: The Rites sional form." American Socio-
and Rituals of Corporate Life. logical Review, 50: 377-391.
Chaves, Mark
Reading, MA: Addison-Wes- 1987 "The intraorganizational power
1993 "Intraorganizational power and
ley. struggle: Rise of finance per-
internal secularization in Prot-
estant denominations." Ameri- sonnel to top leadership in
can Journal of Sociology, 99: large corporations,
1-48. 1919-1979." American Socio-
logical Review, 52: 44-58.
1990 The Transformation of Corpo-
rate Control. Cambridge, MA:
Harvard University Press.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Circulation of Control
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms
Vancil Richard F. Wernerfelt, Birger Womack, J. P., D. T. Jones, and D.
1987 Passing the Baton: Managing 1984 "A resource-based view of Roos
the Process of CEO Succes- the firm." Strategic Manage- 1990 The Machine That Changed
sion. Cambridge, MA: Harvard ment Journal, 5: 171-180. the World: The Story of Lean
Business School Press. Production. New York: Raw-
Westphal, James D., and Edward
son.
Virany, Beverly, Michael L. Tush- J. Zajac
man, and Elaine Romanelli 1995 "Who bJiall govern?: CEO/ Zald, Mayer N., and Michael A.
1992 "Executive succession and board power, demographic Berger
organizational outcomes in similarity, and new director 1978 "Social movements in organi-
turbulent environments: An selection." Administrative Sci- zations: Coup d'etat, insur-
organizational learning ap- ence Quarterly, 40: 60-83. gency, and mass move-
proach." Organization Sci- ments." American Journal of
White, Harrison C.
ence, 3: 72-91. Sociology, 83: 823-861.
1992 Identity and Control: A Struc-
Wade, James, Charles A. O'Reilly, tural Theory of Action. Prince- Zucker, Lynne G.
and Ike Chandratat ton, NJ: Princeton University 1977 "The role of institutionalization
1990 "Golden parachutes: CEOs Press. in cultural persistence."
and the exercise of social in- American Sociological Review,
fluence." Administrative Sci- 42: 726-743.
ence Quarterly, 35: 587-603.
This content downloaded from 130.56.64.29 on Sun, 26 Jun 2016 11:01:16 UTC
All use subject to http://about.jstor.org/terms