3 ﬁ
«1;...

 

I;
.
L.

 

 

 

7 .
(so...
iv

 

 

 

01;: .3. t
. r 3an ,71 r: .5
u 10:". M
(4‘, 3.
#5.:-
xtlz x
.1.

 

‘ i. .

w! r? was -
(.93...le
( l .

..

 

n.

v: «It,
21‘. {.Y.

 

 

Hit-Sis.

7
.2

4' rah 5
4. .s

This is to certify that the

dissertation entitled

INFLUENCE OF THE BOARD OF DIRECTORS
AT A STRATEGIC DECISION POINT FOR HIGH GROWTH FIRMS

presented by

REBECCA ANN LUCE

has been accepted towards fulﬁllment
of the requirements for

Ph - D - degree in Wnagement

 

ﬂquZZ/mww

0 Major professor

Date /3//%2

MS U is an Afﬁrmatiw Action/Equal Opportunity Institution 0 12771

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIHC/DateDuepGS-ots

 

INFLUENCE OF THE BOARD OF DIRECTORS
AT A STRATEGIC DECISION POINT FOR HIGH GROWTH FIRMS

By

Rebecca Ann Luce

A DISSERTATION

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Management

2002

ABSTRACT
INFLUENCE OF THE BOARD OF DIRECTORS
AT A STRATEGIC DECISION POINT FOR HIGH GROWTH FIRMS
By

Rebecca A. Luce

The design of this dissertation attempted to address some Of the
criticisms Of board composition research that have arisen due to the lack of
ﬁndings of a practically meaningful relationship between board composition and
ﬁrm performance. The ﬁve omissions noted in prior research of this nature that
are tackled in this study are: (1) contextual factors do not receive sufﬁcient
attention in designing studies regarding board inﬂuence; (2) the failure to
diversify the types of ﬁrms which are examined beyond the large corporation; (3)
the inconsistency with which board composition variables have been
Operationalized; (4) the failure of many studies to tie the categorization Of
directors to the theoretical foundations being relied upon; and (5) the efforts of
many researchers to link the composition of the board Of directors to
performance may fall Short due to the distal nature of the connection.

In this study I examined the resource dependence role Of the board Of
directors within the speciﬁc context Of relatively small ﬁrms about to enter a
period of high growth. I attempted to answer three primary research questions in
this dissertation. The ﬁrst was: How do the resources (in the form of knowledge

and experiences) represented on the board of directors inﬂuence the response

“of high growth ﬁrms to the internal organizational demands created by their
steadily increasing size? The second was: How does the extent Of the internal
organizational changes implemented by ﬁrms during a period Of high growth
affect subsequent ﬁrm performance? And the ﬁnal question was: How do the
knowledge and experience resources represented on the board impact
subsequent ﬁrm performance, both directly and indirectly, through the internal
organizational changes made by ﬁrms during a period of high growth? I
formulated six hypotheses to explore my three primary research questions, none

of which received more than weak support.

This dissertation is dedicated to my parents,
Wayne and Patricia Luce.

They instilled in me a life long love of learning
and a belief that I could accomplish anything I set out to do.

ACKNOWLEDGEMENTS

I was supported in my completion Of this dissertation by the commitment
of numerous people who unselﬁshly gave of their time and expertise to help me
acquire the skills to be a competent faculty member.

First Of all, I would like to thank Linn Van Dyne, whO'waS my ﬁrst mentor in
my graduate program at Michigan State. She encouraged me to persist in my
' studies and research efforts in spite Of the Obstacles I encountered along the
way.

Secondly, I would like to thank the co-Chairs of my dissertation committee,
Amy Hillman and Gerry McNamara. Both Amy and Gerry supported me not only
in the completion of my dissertation, but also in the development of my teaching
and research Skills throughout my program. They steadfastly believed in my
ability to become a capable professor Of management at every step along the
way. I also appreciate the contributions Of my two other committee members,
Bob Wiseman and John Wagner, whose involvement greatly strengthened the
quality of my dissertation.

My fellow doctoral students, Marcia Simmering, Michael Wesson, Jason
' Colquitt, and Jeff LePine, were always there for me to lean on and to offer advice
and reassurance when I needed it most. Thanks to all of them for being terriﬁc
role models as well as great friends.

Lastly, I owe a debt of gratitude to my colleagues at Texas Christian
University. They made me feel welcome and provided expertise and support in

helping me ﬁnish up my dissertation. I couldn’t have found a better home.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. viii
LIST OF FIGURES ................................................................................ ix
LIST OF APPENDICES ........................................................................... x
CHAPTER 1: INTRODUCTION ................................................................. 1
CHAPTER II: LITERATURE REVIEW, THEORY, AND HYPOTHESES ........... 12
Resource Dependence Role Of the Board of Directors ........................ 12
Firm strategic contingency-board composition link ..................... 17
Board composition-ﬁrm outcome link ...................................... 20
Resource dependence role Of the board in smaller ﬁrms ............ 27
Resource dependence rOle of the board during periods
of crisis and transition ......................................................... 33
High Growth Transition ................................................................. 38
Relationship Of the Composition Of the Board Of Directors to
Firms’ High Growth Transition Changes .......................................... 45
Implications for Firm Performance .................................................. 49
The Role of the Chief Executive Ofﬁcer ........................................... 54
Implications of the Rate of Growth Of Firms ...................................... 57
CHAPTER III: RESEARCH DESIGN AND METHODS OF ANALYSIS ............ 60
Research Design ........................................................................ 60
Sample ..................................................................................... 60
Data Collection and COdIng61
Independent Variable Measures ..................................................... 62
Board resources ................ ‘ ................................................ 62
CEO managerial qualiﬁcations .............................................. 63
Rate Of growth ................................................................... 66
Dependent Variable Measures ....................................................... 66
High growth transition infrastructure variables. . . . . . . . . . . . . . . ., ......... 66
Standardization and formalization .......................................... 70
Shared goals ..................................................................... 71
Professionalization of management ....................................... 73
Specialization .................................................................... 74
Firm performance ............................................................... 80
Control Variables ........................................................................ 84
Industry ............................................................................ 84
Firm Size .......................................................................... 86
Size and tenure Of the board ........... '. .................................... 87
CEO stock ownership ......................................................... 87

vi

TABLE OF CONTENTS (cont’d)

Control Variables continued

High growth transition changes ............................................. 88

Average industry performance .............................................. 88

Methods of Analysis .................................................................... 89
Variable Screening and Transformations ......................................... 90
CHAPTER IV: RESULTS ....................................................................... 92
Predictors Of High Growth Transition Infrastructure Variables ............... 92
Predictors of Subsequent Firm Performance ..................................... 92
Mediation Model ......................................................................... 94
Interaction Effects ....................................................................... 94
Supplementary Analyses .............................................................. 97

Post Regression Tests ............................................................... 100
CHAPTER V: DISCUSSION AND FUTURE RESEARCH ........................... 103
Key (Non)Findings ..... ' ................................................................ 103
Limitations of Study ................................................................... 107

Data collection procedures ................................................. 107
Infrastructure variables ...................................................... 111

Sample .......................................................................... 1 13

Model speciﬁcation and theory ............................................ 117

Directions for Future Research ................. -. ................................... 122
Present study research questions ........................................ 122

Firm development ........... ' .................................................. 124

Impact of investors ............................................................ 127

Survival .......................................................................... 129

Conclusion ............................................................................... 130
TABLES ................................ 132
FIGURES .......................................................................................... 171
APPENDICES .................................................................................... 175
REFERENCES .................................................................................. 196

vii

LIST OF TABLES

Table
3.1 TMT Titles Counted for Specialization Variable in a Sub-sample

of Thirty Firms ........................................................................... 133
3.2 Variable and Item-Level Reliability ................................................ 135
4.1 Descriptive Statistics and Correlations ........................................... 136

4.2 Regression Of High Growth Transition Infrastructure Changes
on Board Composition ................................................................ 142

4.3 Regression of Firm Performance on High Growth transition
Infrastructure Changes ............................................................... 149

4.4 Regression of Firm Performance on Board Composition .................... 153

4.5 Regression of High Growth Infrastructure Changes on Board

Composition Moderated by CEO Qualiﬁcations ................................ 157
4.6 Regression Of Firm Performance on High Growth Transition

Infrastructure Changes Moderated by Change in Firm Size ................ 164
4.7 Outlier Values Deleted ................................................................ 168

5.1 Summary of Results by Hypothesis ............................................... 169

viii

Figure

1.1

2.1

V 2.2

2.3

2.4

4.1

4.2

4.3

LIST OF FIGURES

General Theoretical Model ............................................................ 11
Theoretical Model A ..................................................................... 48
Theoretical Model B ...................................................... ‘ ............... 54
Theoretical Model C ..................................................................... 57
Complete Theoretical Model .......................................................... 59
Regression Of Executive Ofﬁcer Management Experience on the

Interaction of Business Experts and CEO General Management
Experience ............................................................................... 1 72

Regression of Executive Ofﬁcer Management Experience on the
Interaction Of Support Specialists and CEO General Management
Experience ............................................................................... 173

Regression Of Executive Ofﬁcer Management Experience on the
Interaction Of Business Experts and CEO Industry Experience ............ 174

LIST OF APPENDICES

Appendix

A Variable List with Data Sources and Time of Collection ..................... 176
B Data Collection and Coding Sheet ................................................ 178
C Data Collection Of High Growth Transition Infrastructure Variables ...... 185
D Executive Survey ...................................................................... 188
E Approval Of a Project Involving Human Subjects .............................. 190

CHAPTER 1
INTRODUCTION

Accompanying thedispersion of ownership that is Characteristic Of the
modern corporate form iS the rise Of the board of directors as a corporate
governance mechanism (Berle & Means, 1932). The existence Of the board Of
directors is an acknowledgment that the owners of the ﬁrm are not in a position
to assess for themselves whether or not their interests are being protected
(Fama & Jensen, 1983). Both the managerialist perspective and agency theory
take the general stance that when a ﬁrm’s owners (principals) are not directly
managing the business, they must entrust this activity to managers (agents), with
the potential result that managers’ self-interests will not coincide with those Of the
ﬁrm’s owners (Berle & Means, 1932; Jensen & Meckling, 1976). The costs of
monitoring the ﬁrm’s managers, to ensure they are acting appropriately on behalf
Of the owners, reduces the value of the ﬁrm from its theoretical potential (Jensen
& Meckling, 1976). The board of directors is a governance mechanism designed
to minimize monitoring costs, and thus retain as much Of the ﬁrm’s value as 2
possible for its shareholders (Fama & Jensen, 1983).

In their role as corporate watchdogs, boards of directors walk a ﬁne line.
They have a ﬁduciary responsibility to the ﬁrm’s owners to prOtect their
investment, yet they must generally accomplish this in a manner that does not
usurp the responsibility Of the ﬁrm’s management, which is the running Of the
business (Lorsch & Maclver, 1989; Vance, 1983). Zahra and Pearce (1989) in

their extensive review of the board of directors literature extant at that time,

discuss the resource dependence perspective as it relates to the capabilities of
boards Of directors to perform their functions. According to Zahra and Pearce,
the resource dependence approach “views boards as important boundary
spanners that make timely information available to executives” and “extract
resources for successful company operations” (1989: 297).' In their resource
dependence role, directors provide resources to the ﬁrm by virtue Of their varying
- skills and abilities, as well as their linkages with the external environment
(Pfeffer, 1972, 1973; Pfeffer & Salancik, 1978).

Zald (1969) points out that, in addition to board members serving as a
means of reducing external uncertainty due to interdependency with other
organizations, they also perform a valuable internal administrative role for the
ﬁrm. In this role, directors who possess knowledge and experience that relate to
the ﬁrm’s business can inﬂuence the strategic decision making process (Zald
1969). According to Zald, this type of board resource can be knowledge that
comes from “detailed familiarity with the speciﬁc organization or from general
expertise...” (104:1969). TO the degree that the resources represented by
' directors’ knowledge and expertise match the needs of the ﬁrm, the board is
A better positioned assist the ﬁrm in attaining its goals (Pfeffer, 1972, 1973;
Provan, 1980; Zald, 1967).

This dissertation examined the resource dependence role of the board Of
directors within the speciﬁc context of relatively small ﬁrms about to enter a
period of high growth. I attempted to answer three primary research questions in

this dissertation:

oHow do the resources (in the form Of knowledge and experiences)
represented on the board Of directors inﬂuence the response of high growth ﬁrms
to the internal organizational demands created by their steadily increasing size?

oHow does the extent of the internal organizational changes implemented
by ﬁrms during a period Of high growth affect subsequent ﬁrm performance?

oHow do the knowledge and experience resources represented on the
board impact subsequent ﬁrm performance, both directly and indirectly, through
the internal organizational changes made by ﬁrms during a period Of high
growth?

To underscore the contributions Of this study, I brieﬂy review the relevant
general ﬁndings thus far in research conducted on the inﬂuence Of boards Of
directors, and to highlight some Of the shortcomings that have been identiﬁed in
the strategic management literature regarding board of directors research.

The theoretical models Offered by the various perspectives on boards of
directors all suggest that the nature Of the directors serving on a ﬁrrn’s board, i.e.
the composition Of the board, is a critical factor in the ability of the board to fulﬁll
their functions effectively (Zahra & Pearce, 1989). Literally hundreds of studies
have investigated the relationship of board composition to various ﬁrm outcomes
(cf. Dalton, Daily, Ellstrand, & Johnson, 1998;Johnson, Daily & Ellstrand, 1996;
Zahra & Pearce, 1989). Studies of board composition adopt the presumption
that boards made up Of different types Of directors will handle board functions
differently due to the varying resources and perspectives they bring to their

positions (Pfeffer, 1972, 1973).

The litmus test of the efﬁcacy with which a board executes its roles is
ultimately ﬁrm performance. Perhaps due to the inherent difﬁculty in assessing
the manner in which the board performs its functions, researchers have often
sought to directly link board composition to ﬁrm performance, with mixed results
(Zahra & Pearce, 1989; Johnson et al., 1996). Some researchers have found
support for a relationship between the type of directors serving on a ﬁrm’s board
and ﬁrm performance (e.g. Barnhart, Marr, & Rosenstein, 1994; Baysinger &
Butler, 1985; Cochran, Wood, & Jones, 1985; Pearce & Zahra, 1992;
Schellenger, Wood, & Tashakori, 1989). Others, however, have failed to ﬁnd
such effects (e.g. Chaganti, Mahajan, & Sharma, 1985;Hermalin & Weisbach, '
1991; Zahra & Stanton, 1988). In an effort to clarify the results Of these
apparently contradictory ﬁndings, Dalton et al. (1998) recently performed a
meta-analysis Of the effects Of board composition on ﬁrm performance. Their
analysis, which included ﬁfty-four empirical studies and 159 different samples of
ﬁrms, found a corrected mean correlation Of .028 between board composition
and ﬁrm performance, leading them to conclude that “board composition has
virtually no effect on ﬁrm performance” (1998: 278). . I

The lack Of a practically meaningful relationship between board
composition and ﬁrm performance has attracted the attention of scholars
interested in the board Of directors as a governance mechanism. One of the
primary criticisms Of research being performed in this area is that contextual
fa_ct9_r_s_ do not receive sufﬁcient attention in designing studies regarding board

Inﬂuence (Zahra & Pearce, 1989). Governance scholars have pointed out that

boards’ inﬂuences are Often felt to the greatest degree when the ﬁrm faces a
crisis Of some sort calling for decisive action (Lorsch & Maclver, 1989; Mace,
1971; Zald, 1969), such as the dismissal Of a current CEO (Vance, 1983),
bankruptcy (Daily, 1995; Daily & Dalton, 1994), or takeover attempts (Ward,
1997). Zald emphasized that “it is during the handling Of major phase problems,
or strategic decision points, that board power is most likely to be asserted"
(1969: 107). A failure to specify the context Of the ﬁrms in a study may dilute the
Observed effects Of the board of directors.

A second concern raised in reviews Of the literature on board of directors
is the failure to diversify the types of ﬁrms, which are examined. Most studies of
the effects of boards Of directors on ﬁrm outcomes have focused on large
corporations, such as the Fortune 500 population (Zahra & Pearce, 1989). In the
samples of the 54 studies included in Dalton et al.’s (1998) meta-analysis, eighty
percent were large ﬁrms. Most board composition studies use a “convenience”
sample of the ﬁrms for which data is most readily available, which tend to be
larger corporations (Zahra & Pearce, 1989). However, the increased complexity
Of large, mature ﬁrms compared to smaller, younger ﬁrms may Cloud the effects
of the board Of directors (Daily 8 Dalton, 1993; Dalton et al., 1998). Zald (1969)
pointed out that when stock ownership is widely dispersed, (as it is apt to be in a
large, public company), it is likely board power will be lessened.
Correspondingly, the Observed effect Of the board’s inﬂuence in such ﬁrms is

likely to be diminished.

Third, a considerable source Of consternation among those reviewing
board Of director research ﬁndings thus far is the inconsistency with which board
composition variables have been Operationalized (Daily, Johnson, 8 Dalton,
1999; Dalton et al., 1998; Zahra & Pearce, 1989). As Daily et al. put it, “If
researchers have seen one measure of board composition, they assuredly have
not seen them all" (1999: 99). The often-used designation of “inside‘director”
categorization has numerously been Operationalized as “active employees of the
ﬁrm,” “active and former employees Of the ﬁrm,” and “active and former
employees Of the ﬁrm and relatives Of management” (Daily et al., 1999). This
lack of consistency has made it difﬁcult to draw conclusions across studies
regarding effects Of board composition.

Exacerbating this problem iS the failure of many studies to tie the
categorization of directors to the theoretical foundations being relied upon (Daily
et al., 1999). (For exceptions, see Baysinger & Butler, 1985; Baysinger 8.
Zardkoohi, 1986 and Hillman, Cannella, & Paetzold, 2000). As examples, Daily
at al. (1999) suggest that the “outside director" category (those who are
unafﬁliated with the ﬁrm) may be most relevant when the focus Of the study is the
resource dependencerole Of the ﬁrm, while the “independent/interdependent
director” distinction (hinging on whether or not the director was appointed by the
current CEO) may be more relevant to a study of the board’s function as a
monitoring mechanism. When the Operationalization Of board composition is not
matched carefully to the theory Of the study, reduced effect Sizes are likely to

result.

Additionally, the efforts Of many researchers to link the composition of the
board Of directors to performance may fall short due to the w of the
connection (Zahra & Pearce, 1989). Many factors in an organization have the
opportunity to intervene between action taken by the board and the ultimate
performance Of the ﬁrm. The management Of the ﬁrm by its executives is but
one major source of “noise” in drawing conclusions about board impact on
performance. Most of the theoretical models of board of director inﬂuence
contain “strategic outcomes” Of the execution Of board functions as intervening
variables between board composition and ﬁrm performance, yet few researchers
explicitly include them in their studies Of boards Of directors (Zahra & Pearce,
1989). A stronger relationship may be observed between boards of directors
and ﬁrm outcomes that are more proximal to the performance of the board’s
functions, with the inﬂuence of the board on ﬁrm performance modeled primarily
as an indirect effect felt through intermediate outcomes.

This dissertation attempts to address these criticisms Of prior board
research. The context for this study involves ﬁrms that are anticipating entrance
into a period of high growth. Zald (1969) and Zahra and Pearce (1989), in their
discussions of contexts when the board Of directors may be particularly valuable
to a ﬁrm, highlight the transition from one stage of the corporate life cycle to .
another. Although the existence Of a Speciﬁc sequence Of stages in a corporate
life cycle has been called into question by the results of empirical research (Miller
& Friesen, 1984b), there is evidence that as ﬁrms move from an entrepreneurial

stage Of development into a period of rapid growth, demands are placed on its

internal organization that necessitate Changes. to support ﬁrm performance at a
larger size (Chandler, 1962; Haire, 1959). This transition, if accomplished
successfully, is customarily Characterized by an increased prevalence of top
managers with professional experience in a corporate setting and increased
structural complexity Of the ﬁrm organizationally (Blau & Schoenherr, 1971;
Hambrick & Crozier, 1985; Penrose, 1959; Tashakori, 1980). The inﬂuence Of
the board of directors in the success Of a ﬁrm’s high growth transition, and the
changes within the ﬁrm that accompany it, form a part of the speciﬁc context for
this dissertation.

The ﬁrms being studied here differ from the typical “large ﬁrm” proﬁle of
the majority Of past board of director research. The size Of the ﬁrms making the
transition from entrepreneurial to more professional will be considerably different
in size from the large corporation commonly studied. TO illustrate this
divergence, we can compare ﬂrrns about to undergo an initial public Offering Of
their stock with the Fortune 500. Only twenty percent of the ﬁrms going public in
1988 that were included in a study by Welboume and Andrews (1996) had over
700 employees, while the average number Of people employed by a Fortune 500
ﬁrm in 1988 was 12,700 (Abelson & Jacob, 1989). As mentioned earlier, board
researchers and scholars have suggested that less complex, smaller ﬁrms may
provide an avenue for Observation of greater board effects on ﬁrm outcomes
than their large-ﬁrm counterparts (Daily & Dalton, 1993; Dalton et al., 1998; Zald,

1969).

This dissertation focuses on the administrative aspect of the resource
dependence role Of the board of directors, i.e. furnishing advice and counsel to
the CEO based on expertise directors bring to their positions. In an effort to
ensure that the operationalization Of board composition variables reﬂected this
resource provision aspect Of the board, I use the categorization of directors
developed by Hillman et al. (2000). In their study, Hillman et al. reviewed the
- resource dependence literature and arrived at designations for directors based
on the resources they brought to the board. The taxonomy they developed for
directors was based on directors’ experiences, expertise, and knowledge and
their occupational attributes. By using classiﬁcations of directors directly tied to
the speciﬁc role Of the board I examine in this dissertation, and building on the
use Of the categorization system established as predictive by Hillman et al.
(2000), I expect to increase the likelihood I would ﬁnd evidence Of the inﬂuence
of board resources on ﬁrm outcomes in my dissertation.

Lastly, my dissertation design addresses the issue of proximity of the
inﬂuence Of the board of directors by examining the relationship between board
composition and an outcome intervening between actions of the board of
I directors and ﬁrm performance, the transition to the growth stage of
organizational development. I present a mediation model showing how I expect
the expertise resources represented by directors to affect the ﬁrm’s high growth
transition, which in turn I predict will affect ﬁrm performance. I include a direct
effect Of the composition Of the board of directors on ﬁrm in the model based on

the results Of prior studies, but I anticipate that the primary impact of board

resources on performance will be indirect, through the internal organizational
infrastructure associated with high growth.

This dissertation employs a longitudinal research design in order to
adequately assess the inﬂuence of the board Of directors on Changes that may
(or may not) occur within the ﬁrm over time as it grows. I collected the data for
this dissertation from archival sources over the period Of four years, beginning
with the measurement Of board composition along with other “starting point”
variables. For four years following the starting point date, data were collected on
the degree to which a ﬁrm has implemented internal organizational Changes
associated with a high growth transition. Finally, subsequent ﬁrm performance
was assessed four years after the starting point, allowing time for the effects Of
the internal organizational changes on performance to be Observed. Figure 1.1
presents my general research model outlining the relationships between the
primary variables in this study.

The balance Of this dissertation is organized as follows: in Chapter II I
review the, primary literature underlying this dissertation, present the full model of
effects I examine in this dissertation, along with my hypotheses regarding the
relationships between the constructs. In Chapter III I describe the research
methodology I used to collect my data and tolempirically test my hypotheses.
Chapter IV presents the results of my analyses and Chapter V contains my

discussion of the results as well as directions for future research.

10

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Board High Growth Firm
—-> >-——->
Composition Transition Performance
Changes
Time I

Figure 1.1 General Theoretical Model

11

CHAPTER II
LITERATURE REVIEW, THEORY, AND HYPOTHESES

To provide a foundation for the investigation I describe in this dissertation,
I will next review the relevant research streams associated with the questions I
posed. The primary areas Of research related to this study are the resource
dependence role of the board of directors and organizational structure
Iiteratures. I will proceed ﬁrst by reviewing the research documenting the
resource dependence role Of the board, and then focus more narrowly on
research regarding the board of directors’ resource dependence role in smaller
ﬁrms as well as in those experiencing a period Of crisis or transition. Next, I will
present the research related to organizational structure and, more speciﬁcally,
ﬁrms” transitions in response to a period Of high growth.
Resource Dependence Role of the Board of Directors

The capacity of boards Of directors to be valuable resources to the chief
executive of their ﬁrms is one Of the primary ways by which boards serve the
interests Of the shareholders they are entrusted to protect (Mace, 1971; Pfeffer 8
Salancik, 1978; Zahra & Pearce, 1989). All organizations face uncertainty, with
prosperity and survival contingent on their ability to identify and cope with the
factors that contribute to that uncertainty (Pfeffer & Salancik, 1978; Selznick,
1949; Thompson, 1967). Having a board of directors that provides needed
resources is one manner in which ﬁrms can reduce uncertainty and be in a better
position to compete successfully (Pfeffer & Salancik, 1978). Selznick (1949)

originated the concept that organizations will be more successful if they gamer

12

the support of key constituencies in their environments by soliciting their
involvement in organizational activities, which he called "cooptation." Pfeffer and
Salancik describe the rationale behind the perspective that outside directors are
likely to coopt resources for use by the ﬁrms they serve as follows: “When an
organization appoints an individual to a board, it expects the individual will come
to support the organization, will concern himself with its problems, will favorably
present it to others, and will try to aid it” (1978: 163). Zald takes the position that
“to the degree that board members control or represent salient external
‘resources’ they are more powerful than if they do not control Such resources”
(1969: 100), thus giving them more latitude in executing their governance
responsibilities. Since the primary purpose Of the board of directors is to insure
the maximization of shareholder wealth (Fama 8 Jensen, 1983; Mace, 1971;
Vance, 1983), its ability to assist the ﬁrm’s management by supplying valuable
resources is directly related to serving the interests Of the board’s constituency.
Ferry (1999) reports that the representation of outside directors, (those
who are not part of the ﬁrm’s management), on corporate boards has steadily
increased since 1973. Today, the typical board in the United States is comprised
predominantly Of outside directors, with such directors holding nine Of the eleven
average number of board seats. This change in board composition may have
been spurred primarily by the focus on board reform over the past two decades,
due to powerful institutional investors’ perceptions that boards dominated by
insiders were tOO passive in the execution Of their discipline responsibilities

(Rechner, 1989; Ward, 1997). However, the result of increased outside director

13

' representation is boards that offer more resources than ever before to a ﬁrm’s
management. Since outside directors bring to their responsibilities a variety Of
experiences and knowledge on general business matters, as well as technical
expertise, the “outside board” presents considerable potential for contributions to
the running of the business (Fama 8 Jensen, 1983; Zahra 8 Stanton, 1988).

Board researchers have identiﬁed the resource dependence role Of the
board, its capacity to supplement a ﬁrm’s resource base, as one Of its key
functions (Johnson, et al., 1996; Mace, 1971; Zahra 8 Pearce, 1989). Boards Of
directors are potentially in a position to accomplish their resource dependence
role in two ways (Hillman et al., 2000; Pfeffer, 1973; Provan, 1980; Zald, 1969).
First, directors may provide linkages to the external environment that facilitate
the ability Of the ﬁrm to procure the resources it seeks (Pfeffer, 1972, 1973). For
example, a director from the ﬁnancial sector may have connections within the
banking industry that would make it easier for a ﬁrm to Obtain needed ﬁnancing
(Pfeffer, 1972). By virtue of their contacts with other organizations within the
ﬁrm’s domain, outside directors may be in a position to enhance the competitive
position of the ﬁrms they represent by making it possible to secure resources
that would otherwise be difﬁcult to procure (Charan, 1998; Pfeffer 8 Salancik,
1978; Thompson, 1967). When boards control access to crucial inputs, they.
reduce environmental uncertainty, and increase their inﬂuence within the.
organization (Zald, 1969).

The second manner in which outside directors can fulﬁll their resource

dependence role is in an administrative capacity, by furnishing advice and

14

counsel that derives from directors’ professional experiences and expertise
(Pfeffer, 1972,1973; Provan, 1980; Zald, 1969). When outside directors join a
ﬁrm’s board with either executive or technical expertise, they supplement the
knowledge and experience Of the ﬁrm’s top management and thereby provide an
additional resource base upon which the ﬁrm can draw for decision making
(Pfeffer, 1973; Pfeffer 8 Salancik, 1978; Provan, 1980; Zahra 8 Pearce, 1989).
The ability Of outside directors to supplement internal ﬁrm resources may assist
them in preventing business problems before they arise, a high priority Of large
institutional investors and governance advisors (Byrne, Grover, 8 Melcher,
1997). Research supports the existence and importance of both the advice and
counsel and external resource cooptation aspects of the board’s resource
dependence role, as I demonstrate in the following discussion.

As advisors, outside directors serve the interests Of the ﬁrm’s
Shareholders by providing counsel to CEOs and other top managers that
supplements their existing business knowledge (Mace, 1971; Zahra 8 Stanton,
1988). According to Mace’s interviews Of corporate CEOS, the role of directors. as
sounding boards is an integral part of their function: ”T he inputs Of outside
directors were valued by top management, who had generally regarded the
board members as ‘additional windows to the outside world” (1971: 39-40). The
mental capacity Of the top management team can be expanded by drawing upon
board expertise, resulting in a “collective wisdom” that is more valuable to the
ﬁrm than either the board or management functioning independently (Charan,

1998).

15

Using their professional expertise, outside board members can inﬂuence a
CEO’S decisions by “watching and counseling”, without directly interceding in the
management Of the business (Lorsch 8 Maclver, 1989), a tactic that should only
be employed when the ﬁrm’s management is clearly performing in a substandard
manner (Walsh 8 Seward, 1990). “Directors spend much more time advising the
CEO, a task that, while not as dramatic as replacing him, enables them to play
- what many consider to be their key normal duty" (Lorsch 8 Maclver, 1989: 64).
Mace (1971) found that Chief executives often regarded outside directors as their
peers in the business environment, and actively solicited their guidance in
making critical business decisions. Hermalin and Weisbach’s (1988) ﬁndings
suggest that new CEOS may appoint outside directors aS a means Of securing
needed advice and counsel.

While outside directors who currently or in the past served as executives
themselves may provide general business management experience and
perspective (Fama 8 Jensen, 1983), a ﬁrm may also the beneﬁt from the
expertise of those who come from more specialized professions, such as
' ﬁnance, law, or marketing (Mace, 1971; Pfeffer, 1972; Zald, 1969). These

1 “specialist” directors are In a position to Offer advice based on their technical
knowledge and, therefore, to make contributions to decisions in areas in which
managerial expertise is sparse (Baysinger 8 Butler, 1985; Baysinger 8
Zardkoohi, 1986; Hillman et al., 2000; Vance, 1983).

Two streams Of research investigate the Implications Of the resource

dependence perspective for the composition of the board Of directors. One

16

stream examines the effects Of contingencies facing the ﬁrm on the composition
Of the board, while the other looks at the effects Of board composition on ﬁrm
outcomes (Boyd, 1990). I ﬁrst discuss the results Of empirical studies that
identify ﬁrrns’ contextual demands and assess to what degree their needs
inﬂuence the composition Of their boards Of directors.

Firm strategic contingency-board composition link. While the
evidence Of the advice and counsel role of the board is largely descriptive, based
on interviews or experiences of researchers (Johnson et al., 1996), there is
considerable empirical research support for the position that the composition of
the board varies depending on the resource needs Of the ﬁrm. These studies
start with the assumption that if directors are indeed a factor in providing needed
resources for the beneﬁt of the ﬁrms on whose boards they sit, then the
backgrOunds Of the directors chosen by a particular ﬁrm should reﬂect its
resource needs as well as the potential Of the board to fulﬁll its functions
(Pearce 8 Zahra, 1992). Therefore, various aspects of board composition serve
as the dependent variables in this type Of study, with compositional factors acting
as proxies for the resources directors bring to the ﬁrm (Pfeffer, 1972, 1973).

Pfeffer (1972), in his landmark study of the resource dependence
perspective, examined the relationship between common contingencies facing
corporations, sUch as capital requirements or degree of regulation, and the
composition Of the ﬁrm’s board of directors. He found support for his
hypothesized linkages in a sample Of eighty large corporations across a variety

of industries. For instance, ﬁrms having needs to acquire ﬁnancial capital (i.e.

17

those with higher debt to equity ratios), were more likely to have outside directors
from financial institutions. Pfeffer (1973) Obtained similar results in his research
with a sample Of hospitals, where the composition of the board Of directors was
related to the environmental demands placed on the hospitals and the functions
administrators expected their boards to perform. Hospitals that were more
reliant on the environment for fund-raising were more likely to look to their
directors for linkages to the external environment than hospitals that had other
sources of funding available to them. In addition, there was some evidencethat
the board function Of providing external linkages was related to the composition
Of the board.

Baysinger and Zardkoohi (1986) classiﬁed directors into four categories,
based primarily on their likelihood Of contributing to decision control within a ﬁrm.
They hypothesized that regulated ﬁrms (utilities) were less in need Of directors
capable of performing decision control functions than unregulated ﬁrms, due to
the governmental restrictions on their operations. Baysinger and Zardkoohi’s
(1986) ﬁndings were consistent with the resource dependence perspective since
the composition Of boards differed between the two groups Of ﬁrms based on
their variation in resource requirements. Hermalin and Weisbach (1988) provide
additional evidence supporting the resource dependence view. They found that
an impending CEO succession was related to Changes in the types of directors
departing or being added to the board. Hermalin and Weisbach’s (1988) results
suggested that, as a CEO neared retirement, more insiders were added to the

board, presumably to give potential successors additional exposure to other

18

directors who would be making the ﬁnal CEO selection decision, and to give the
candidates experience with the directorship function. They also found that a
change in the ﬁrm’s product strategy or poor performance led to changes in
board composition.

Several ﬁrm contextual factors were linked to board size and to the
representation Of outside directors on the board in a longitudinal study Of large
ﬁrms conducted by Pearce and Zahra (1992) in their test Of Pfeffer’s (1972,
1973) strategic contingency view. For example, Pearce and Zahra (1992) had
expected that when environmental uncertainty facing a ﬁrm increases, a large
board and more outside directors will be observed. They supported this
contention by pointing out that ﬁrms facing uncertainty (1) will attempt to coopt
external constituencies to reduce uncertainty by including them on the board, (2)
will recognize the beneﬁts of a variety of director Skills and capabilities, and (3)
will seek to increase linkages with valuable resources in the environment.
According to the resource dependence perspective, these three Objectives would
be more likely to be met with a larger board and greater outsider representation.
As they had hypothesized, Pearce and Zahra ( 1992) found that increased
environmental uncertainty in a ﬁrm’s domain, diversiﬁcation strategy, higher
leverage structure, and past performance were all related to board size and the
level Of representation of outside directors.

Further indications Of the connection between resource needs and board
composition was provided by Hillman et al. (2000) who hypothesized that the

uncertainty presented by deregulation Of the airlines industry would dictate

19

strategic changes and, therefore, lead airline companies to alter the composition
of their boards Of directors accordingly. For instance, Hillman et al. (2000)
predicted that outside directors with executive expertise would be more likely to
appear on airline ﬁrms’ boards when a vacancy occurred following deregulation,
due to their capability to advise the ﬁrm on strategic direction. Supporting their
hypotheses, they found that the need for ﬁrms to change their corporate
strategies led to predicted alterations in the type of directors selected to serve on
airline companies’ boards.

Although not all Of these studies linking board composition to ﬁrms’
resource contingencies investigate whether or not directors actually make the
contributions expected of them based on the resources they bring to the ﬁrm,
their results are indicative of a link between ﬁrm resource needs and directors’
abilities to meet those needs. The second stream Of resource dependence
research tackles the issue of how board resources, as reﬂected in board
composition, are related to ﬁrm-level outcomes. These studies provide evidence
that the composition Of the board of directors affects the nature of their
contributions to their ﬁrms.

Board composition-ﬁrm outcome link. The ﬁrst set Of studies I review
within this second stream looks at ﬁrm outcomes that are intermediate between
board inﬂuence and ﬁrm performance. Zald (1967) and Provan (1980), in
research regarding the effect Of board composition on the level Of fund-raising by
nonproﬁt organizations, found that the presence Of directors who were inﬂuential

in the community was‘positively related to the amounts of contributions and fund

20

allocations the agencies received. In a study examining the cooptation Of
ﬁnancial resources by a ﬁrm through the composition Of its board of directors,
Stearns and Mizruchi (1993) found that the type and amounts Of funds a ﬁrm
borrowed was positively associated with the nature Of the ﬁnancial institutions its
directors represented. As Stearns and Mizruchi point out, "T he ﬁnancial
institution can use its position on the ﬁrm’s board to seek out potential business,
and the ﬁrm can use the board appointment to coopt the ﬁnancial institution into
providing funds should the ﬁrm require them” (1993: 614). Boeker and
Goodstein (1993) investigated the role of board composition On the selection Of a
new CEO. Their hypothesis that the choice Of a CEO from outside the ﬁrm
would be less likely when insiders dominated the board, since an outside
successor would portend greater disruption than someone with whom they were
already familiar, was supported.

I A few studies speciﬁcally examine the relationship between board
composition and strategic outcomes. Zahra and Stanton (1992), in an
exploratory study of the relationship Of board composition to strategic decisions,
discovered that the size of a ﬁrm’s board Of directors was positively associated
with increased research and development expenditures, and negatively related
to divestments, while the ratio Of outside directors was positively related to ‘
divestment frequency. Contrary to their expectations, Hill and Snell (1988) and
Baysinger, Kosnik, and Turk (1991) found that inside director representation on
the board was positively related to research and development spending by the

ﬁrm. Both sets of researchers had expected to ﬁnd that management directors

21

would be more risk-averse than outside directors, and therefore less likely to be
associated with high-risk investments, such as R 8 D. Hill and Snell (1988) were
also surprised to ﬁnd that increased representation Of outsiders was associated
with greater levels of diversiﬁcation. They had expected to ﬁnd that outside
directors would represent the shareholder view that there are more effective
means of minimizing risk than ﬁrm diversiﬁcation strategy. Although their
hypotheses were not supported, Hill and Snell Observed that “The results
certainly indicate that board composition matters. Thus, they contradict the
arguments of those who see the board as little more than a ‘rubber stamp’”
(1988: 588). In their discussion, Hill and Snell (1988) rely on a ﬁrm needs-board
resources argument to explain their ﬁndings: inside directors may provide
avenues for needed integration Of functions within a ﬁrm when it is spending
more on innovation, while more outside directors of the board may bring needed
breadth Of expertise to the ﬁrm that is diversifying. Since the design of their
study was cross-sectional, it is difﬁcult to assess the degree to which the board
composition of Hill and Snell’s (1988) ﬁrms was a reﬂection Of its resource needs
in managing strategies already decided upon, or whether the composition of the
board inﬂuenced the strategic decisions that were made. .

Judge and Zeithaml’s (1992) study assessed the effeCt Of board
composition variables on the degree to which directors became involved in their
ﬁrms’ strategic decisions. They used both archival records and personal
interviews with board members and CEOS in collecting their data. Again, results

obtained for the inﬂuence Of inside directors were counter to expectations.

22

Judge and Zeithaml (1992) had hypothesized that increased representation Of
inside directors would be associated with increased board involvement in
strategic decisions, due to management directors’ access to detailed internal
information regarding the ﬁrm’s business. In reﬂecting back on CEO interview
comments to explain their counter-intuitive results, Judge and Zeithaml (1992)
concluded that at least some CEOS may look to outside directors to make

. strategic contributions.

The second set Of studies in this stream examines the relationship Of
bottom-line ﬁrm outcomes and board composition. As mentioned earlier, there
are mixed results in the literature regarding board inﬂuences on ﬁrm
performance. Five of the studies discussed earlier also examined the effects Of
board composition on some type Of ﬁrm performance measure. Returning to
Pfeffer’s two studies, we ﬁnd support for the proposition that the resource
dependence role Of the board has an impact on ultimate outcomes. In his 1972
study, Pfeffer assesses the effect Of deviations from the “predicted” board
composition based on the strategic contingencies facing ﬁrms, and ﬁnds that
deviations are very Signiﬁcantly correlated with industry-adjusted performance
‘ measures (ratio Of net income to sales and ratio Of net income to Shareholders’
equity). Pfeffer (1973) uses several measures of organizational effectiveness for
the hospitals in his study, and also ﬁnds correlational evidence Of board
composition associations with additions in facilities or programs, increases in the
numbers Of beds, and increases in budget size. In their study of the effects Of

strategic contingencies on board composition, Pearce and Zahra (1992) went on

23

to examine the relationship between board composition measures and future
company performance. They found that both board Size and outsider
representation were related to multiple measures of ﬁrm performance, (ROA,
ROE, and EPS), averaged over a three-year period

These three studies assessed both aspects of the resource dependence
model described by Boyd: “First, composition Of the board should be affected by
environmental pressures and demands. Second, differences in board
composition Should affect a ﬁrm’s performance” (1990: 419). All three of them
found evidence for their hypotheses regarding the effects Of ﬁrm strategic
contingencies on board composition and the subsequent inﬂuence director
resources on ﬁrm performance. Their results suggest, as the resource
dependence perspective would predict, that when board resources are aligned
with ﬁrm needs, the ﬁrm experiences positive outcomes.

Two additional studies that were discussed above regarding interim ﬁrm
outcomes also investigated the effects Of board resources on ﬁrm performance.
Hill and Snell (1988) found that more outsiders on the board was positively
related to industry-adjusted three-year average ROA, in spite of the fact that their
predicted relationships between board composition and strategic outcome
variables were not supported. Judge and Zeithaml (1992) whose study looked at
the effects Of board composition on the board’s involvement in strategic decision
making, used structural equation modeling to determine if indirect effects Of
composition measures (board Size, insider representation) on ﬁnancial

performance could be Observed. They found a marginally Signiﬁcant relationship

24

between board involvement and industry-adjusted ROA (averaged over a five
yearpeﬁod)

A brief overview Of Six additional studies examining the relationship
between board composition and ﬁrm performance are given, three of which ﬁnd
such a relationship, and three, which do not. Baysinger and Butler (1985) ﬁnd
that there is a lagged effect Of changes in board composition on performance,
with an increase in the proportion Of directors having more independence
contributing positively to later performance. Because of the longitudinal design
of their study, Baysinger and Butler (1985) were able to establish that the effect
Of prior performance on board composition Changes was less than the effect of
composition on performance. Two studies used market measures Of ﬁrm
performance, designed to capture shareholder wealth as the dependent variable,
given its centrality in corporate governance theory. Schellenger et al. (1989) and ‘
Rosenstein and Wyatt (1990) both found positive relationships between the
representation of outside directors and shareholder annualized return on
investment and share-price reactions, respectively.

As a counterbalance to the studies ﬁnding a board cOmposition-ﬁrm
performance relationship, Chaganti et al. (1985) found no difference in the
proportion Of outside directors when they compared the boards Of twenty-one
matched pairs Of retailing ﬁrms, half Of which had failed, and the other half which
had not. In their exploratory study of the relationships between board
composition and ﬁrm outcomes, Zahra and Stanton (1988) found either no

relationship or a negative relationship between outside director ratio and multiple

25

measures of ﬁrm performance. Hermalin and Weisbach (1991) in a longitudinal
study with rigorous controls, found no relationship between board composition
and Tobin’s q (the ratio Of the market value of the ﬁrm to the replacement cost Of
its assets), a measure thought to capture ﬁrm value based on bOth tangible and
intangible assets.

As was pointed out in the Introduction, it is difﬁcult to determine why some
studies have found a relationship between board resources, as represented by
board composition measures, and ﬁrm performance, while other researchers
have failed to ﬁnd evidence of such a relationship. The differences in study
design and operationalization Of variables make comparisons problematic (Daily
et al., 1999). It seems safe to observe that much remains to be learned
regarding under what circumstances board inﬂuences on performance are most
likely to be felt.

It may be, as Wagner, Stimpert, and Fubara (1998) contend, that the
relationship between board composition and ﬁrm performance is more
complicated than assumed. The results Of their meta-analysis and follow up
study ﬁnds a curvilinear relationship between insider and outsider representation
and ﬁrm performance, with greater homogeneity of composition (either insiders
or outsiders) associated with higher performance, as measured by ROA.
However, as Hermalin and Weisbach (1991) commented, another reason for the
difﬁculty in detecting patterns of board composition-ﬁrm performance
relationships could be that different governance structures are Optimal for

different ﬁrms. This view is congruent with the resource dependence

26

perspective, and highlights the importance of specifying the speciﬁc strategic
contingencies, or contextual factors, that face the ﬁrms being studied, if
researchers hope to ﬁnd inﬂuences Of board resources on ﬁrm outcomes (Zahra
8 Pearce, 1989)

In Spite Of the contradictory ﬁndings regarding ﬁrm performance, we can
take the two resource dependence research streams together: (1) the studies
demonstrating a relationship between the contingencies facing ﬁrms, (which
deﬁne their resource needs), and the composition Of the board of directors,
(which represents the resources they bring to the ﬁrm), along with (2) the studies
ﬁnding relationships between board composition and ﬁrm outcomes dependent,
at least in part, on board resources, and ﬁnd ample evidence suggesting that
boards are inﬂuential from a resource dependence perspective in helping ﬁrms
Obtain valuable inputs when they are in need Of such assistance. While most of
these studies were conducted with ﬁrms that are large corporations, some board
scholars and researchers anticipate that the inﬂuences Of the board Of directors,
such as performance Of their resource dependence role, may be felt to a greater
degree in smaller ﬁrms (Daily 8 Dalton, 1992, 1993; Whisler, 1988).

Resource dependence role of the board in smaller ﬁrms. Mace’s
(1948) report based on interviews with small business presidents and directors
regarding the role of the board examined ﬁrms run by one or two people.
Although these ﬁrms are clearly entrepreneurial and smaller than those that are
the subject Of this dissertation, Mace’s Observations in this early treatment Of the

board in the small ﬁrm' context is echoed by later scholars and researchers and

27

’ is, therefore, worth recounting here. Mace concludes his book as follows: “It
would be misleading and erroneous, Of course, to conclude that competent
boards Of directors in small corporations are touchstones Of business success. It
can be concluded however, that active and able board members constitute one
source of management assistance to small corporation managements and this
source of help can be tapped for advice and counsel on any problem involved in
the Operations of the business” (1948: 91-92).

Mace (1948) points out that, unlike larger corporations, small businesses
usually do not have the specialized expertise in-houSe to dealwith many
business issues, such as technical problems that may arise or sophisticated
planning for the future. In addition, small business owners may have “blind
spots” due to their lack of breadth in education or experience. Due to these
limitations, the skills and managerial experiences of outside board members may
prove valuable to the president of a small ﬁrm who has capable directors upon
whom he can draw for supplemental knowledge and ideas. Mace (1948)
highlights the Objectivity of outside directors in emphasizing their potential value
in the small ﬁrm context, since they are not hampered by family ties or job
security concerns that may restrict the contributions Of others associated with the
ﬁrm.

Mace (1948) gives examples of when the board of directors may be
especially helpful to the small business president, who is likely to spend the
majority Of her time on the daily running Of the business itself. Outside board

members may detect changing circumstances in the ﬁrm’s environment that the

28

president doesn’t Observe due to her commitment to comfortable routines, which
have developed in Operating the business. Outside directors can also educate
the small business manager in the beneﬁts Of budgeting and planning with the
capability to forecast cash ﬂow, which may otherwise not seem necessary to the
company president who is preoccupied with day-tO-day business operations. In
general, active outside board members can supplement the limited managerial
business experience possessed by many, if not most, small business presidents.

Mace (1948) identiﬁes both dimensions of the resource dependence role
of the board, providing advice and counsel and linkages with the external
environment as importantin small ﬁrms. The outside board member “multiplies
the channels Of information brought to bear on company problems" and
“provides more avenues Of facts with regard to conditions outside the company
which otherwise might never be taken into account by the management" (Mace,
1948:33)

Although information regarding the long-term success of small businesses
is sketchy, the Small Business Administration (1998) reports that, of the new V
ﬁrms started during the 1976-1978 period (the most recent period for which such
information is available), only thirty percent of those employing ﬁve or more
people had survived eight years later. Welboume and Andrews (1996) found
that Of 136 nonﬁnancial ﬁrms undergoing an initial public Offering in 1988, only
sixty percent existed independently ﬁve years later. Since small ﬁrms seem to
be vulnerable to failure, they may beneﬁt to a greater degree from the

supplemental resources boards Of directors bring to the organization (Eisenhardt

29

8 Schoonhoven, 1990). Finkelstein and Hambrick (1996) make the Observation
that the amount Of managerial talent in smaller ﬁrms may be less than in larger
ﬁrms, creating a need for the contribution of the board of directors’ expertise.

Small ﬁrms may have more difﬁculty gaining access to crucial resources
and have fewer options for managing their resource dependencies, giving the
board Of directors an Opportunity to ﬁll this void (Pfeffer 8 Salancik, 1978). Daily
. and Dalton (1993), in their study Of governance structures in small corporations,
conclude that, for such ﬁrms, the resource dependence role of the board Of
directors may outweigh its monitoring function in importance. They state that
outside directors are “an effective means for overcoming the ‘Iiability Of size’” Of
the small business, since the expertise and resources outside directors offer may
“counter any disadvantages experienced as a result Of the modest resource base
experienced by many small ﬁrms" (Daily 8 Dalton, 1993: 76). Johnson et al.
(1996) agree that this resource dependence role of the board may be more
important in smaller ﬁrms, with the ﬁrms’ managers welcoming the breadth Of
knowledge Offered by the board.

Pfeffer and Salancik (1978) acknowledge the limited resources that may
plague small organizations, leading them to select and utilize board members for
their management Skills. Castaldi and Wortman (1984) also emphasize the
importance Of the small ﬁrm board as a resource, and identify ﬁve board
dimensions that are Of particular relevance for small ﬁrms. Among these

dimensions are the technical expertise, management expertise, and “special

30

economic service,” (speciﬁc competencies or contracts), directors may Offer the
small ﬁrm.

Zald (1969) makes the point that it is easier in smaller ﬁrms for the board
Of directors to understand how the business works and to be sufﬁciently well
informed to involve themselves in the ﬁrm’s decision making. Once the ﬁrm
becomes large and complex, directors are more reliant on management for
information regarding business operations (Fama 8 Jensen, 1983). The reduced
complexity Of small ﬁrms creates fewer constraints on activity and may permit
board inputs to be felt more readily (Daily 8 Dalton, 1993). Dalton et al. (1998)
expand on this theme, by stating that boards in smaller ﬁrms are positioned to
more readily fulﬁll their resource dependence role. “Moreover, we would expect
that boards Of smaller, less complex ﬁrms would enjoy more discretion with fewer
vested interests within the ﬁrm as well as external to the ﬁrm” (Dalton et al.,
1998: 274). They conclude that these factors may make it easier for board
contributions to be felt in smaller ﬁrms.

There is some empirical support for the view that board resources play an
important role within the context Of the small business. Mohan-Neill’s (1995)
study Of environmental scanning activities comparing larger and smaller ﬁrms
showed that smaller ﬁrms engage in less environmental scanning activities and
are therefore less informed regarding environmental issues. This lack Of reliable
information regarding the ﬁrm’s environment is likely to make board executive
expertise and cooptation Of external resources more valuable. Robinson’s

(1982) results suggest that this is the case. He found that small ﬁrms that

31

availed themselves of external expertise to assist with their planning processes
outperformed those not receiving such resources.

Judge and Zeithaml (1992) found that board involvement in the strategic
decision making process in a ﬁrm decreased with the increased complexity Of the
ﬁrm’s business, suggesting that boards of small ﬁrms may have the potential tO
make a greater contribution when the ﬁrm is smaller and less complicated, as
other researchers predicted (Fama 8 Jensen, 1983; Zald, 1969). In a study
regarding the impact of small ﬁrm boards on ﬁrm performance, Gilley, Ford, and
Coombs (1999) found that the strategic contributions of boards Of directors in
high growth ﬁrms had a greater impact on performance in smaller versus larger
ﬁrms. The amount of resources small ﬁrms possess does make a difference in
how well they perform. Brush and Chaganti (1998) found that when small ﬁrms
employing up to one hundred employees had more human and organizational
resources, such as more owner experience, long-range planning, staff skills, and
reporting systems, the net cash ﬂow of the ﬁrm was greater. Their results are
suggestive of the potential importance of board resources as a supplement to
limited managerial resources within the small ﬁrm. .

Finally, Daily and Dalton (1992) and Hambrick and Crozier (1985)
speciﬁcally examined the resource dependence role of the board in small
publicly held ﬁrms undergoing high growth, the context of this study. Both
studies found that the most successful high growth ﬁrms had a greater
representation of outside directors on their boards. Daily and Dalton concluded

that the board of directors “provides a sensible tool when striving for the goal Of

32

firm growth,” and that such ﬁrms “may need the strategic expertise and
resources that a board of directors can best supply” (1992: 376, 383).

Resource dependence role of the board during periods of crisis and
transition. Another context for the potentially increased impact of the board of
directors on ﬁrm outcomes to be observed is during a period of turmoil. The
resource dependence perspective suggests that, as uncertainty increases, which
is the case during crises and transitions, ﬁrms will make use of the resources
available through their boards of directors to reduce that uncertainty (Pfeffer,
1972, 1973; Pfeffer 8 Salancik, 1978). As Daily and Dalton point out in their
study of board composition and bankruptcy, the resource dependence role of
the board “would particularly underscore-the necessity Of having many external
representatives on a board in a time Of crisis as their presence would provide
access to valued resources and information, facilitate interﬁnn commitments,
and aid in establishing legitimacy (1994: 1606).

Both Mace (1971) and Lorsch and Maclver (1989), in their interview-
based research into practices of boards of directors, came to the conclusion that
the inﬂuence of the board may be at its greatest when the ﬁrm is undergoing
some type of crisis. Directors appear to be more willing to “make waves” in the
organizations on whose boards they Sit when the situation facing them calls for
some type of decisive action. Afﬁrmative board involvement will be dictated by
the nature Of the situation facing the ﬁrm, either due to its suddenness, e.g.
death or illness Of the CEO or a hostile takeover attempt, or its severity, e.g.

consistent and dramatic ﬁrm performance downturns indicating top management

33

Changes must be made (Lorsch 8 Maclver, 1989; Mace, 1971; Walsh 8 Seward,
1990)

Zald (1969) agrees that the board Of directors may have differing levels of
impact on a ﬁrm’S business depending on what is happening to the ﬁrm, but he
characterizes these critical junctures somewhat differently than Mace (1971) and
Lorsch and Maclver (1989). Zald describes “strategic decision points” that a ﬁrm
faces as “major phase problems” during which “board power is more likely to be
asserted” (1969: 107). Zald (1969) gives examples Of the type of events or
periods that may constitute a strategic decision point. One is the task of
choosing a successor to the current Chief executive. This responsibility may be
relatively routine if the successor Choice is Obvious and has the consensual
support of the board (Lorsch 8 Maclver, 1989), but the decision to maintain or
replace an incumbent CEO is an occasion upon which the board can make its
presence felt (Mizruchi, 1983; VanCil, 1987). The board has the responsibility to
accurately assess the degree of distress represented by poor ﬁrm performance
and to take action regarding the CEO if it determines he is detrimental to the
welfare of the shareholders (Walsh 8 Seward, 1990). Frederickson, Hambrick, 8
Baumrin (1988) present a theoretical model of the factors affecting the decision
to retain or dismiss a CEO, with board of director characteristics and beliefs as
two Of the four key elements likely to inﬂuence the outcome.

Another occasion for the board of directors to assert itself Is during what
Zald (1969) refers to as “life cycle problems.” By life cycle problems, Zald means

the organization is undergoing some kind Of major transition, such as initial

34

formation of the ﬁrm; a change in identity due to a merger, acquisition, or
alliance; or changes in organizational policies, such as strategic direction or
human resources philOSOphy. The board may need to step in to assure that
needed changes take place Should the ﬁrm’s management be hesitant, or to
make the difﬁcult choices involved in the firrn’s adjustment to its transformed or
tumultuous circumstances. Mace (1948) also identiﬁed this juncture in a ﬁrm’s
development as an Opportunity for outside director involvement and contribution
to company presidents. He points out that changes in the way in which the
business is run will arise as the ﬁrm grows, but that managers in small
businesses may not recognize the need to make the appropriate changes, or
have the capacity to execute them effectively. Mace observed several cases in
which “alert members of the boards Of directors provided useful assistance
during the expansion from intrinsically one-man businesses to larger enterprises”
(1948: 49). Furthermore, Johnson (1997) developed a theoretical model relying
on the resource dependence perspective which suggests that the resources
ﬁrms require from their boards of directors will vary depending on the ﬁrm’s life
cycle stage.

Aside from the theoretical and descriptive evidence of the board’s
inﬂuence during ﬁrm transitional periods, there is other empirical support for the
view that the situation faced by the ﬁrm does affect the level of board activity.
Corporate restructuring presents an occasion of signiﬁcant reorientation for a
ﬁrm that may lead to board action. Johnson, Hoskisson, and Hitt (1993) found

that “outside boards,” those made up predominantly Of outside directors, were

35

more likely to get involved in corporate restructuring decisions, (divestment of
more than two businesses), than “inside boards.” Their results suggest that
when a ﬁrm is undergoing a dramatic change, the board that is not management-
dominated becomes actively involved in the business. The level Of outside board
involvement became more accentuated when the strategic implementation of the
ﬁrm’s top managers was deﬁcient (Johnson et al., 1993). Boards were
apparently more inclined to take a “back seat” in the strategic decision making
aspect of restructuring, unless they perceived the-situation as sufﬁciently acute
so as to call for their involvement. As mentioned earlier, CEO succession is
another occasion where the board has the opportunity to become more assertive
in its role (Zald, 1969). Weisbach (1988) found that outside boards were more
inclined to respond to reports of poor ﬁrm performance by removing the chief
executive than inside boards. The differential activity level Of the outside boards
found in both the Johnson et al. (1993) and Weisbach (1988) studies implies that
they are drawing upon their outside expertise and, perhaps, increased Objectivity
compared to insiders, to make contributions to the ﬁrm during an important I
strategic event. A

While a decision to restructure a corporation or the selection Ofa new
chief executive are discrete, readily identiﬁable major events for a ﬁrm, declining
ﬁrm performance is a more gradual process, yet has even more critical
implications for the survival of the business. Gales and Kesner (1994) examined
the resource dependence perspective in the context of deteriorating ﬁrm

performance leading to bankruptcy declaration. AS Gales and Kesner (1994)

36

point out, ﬁrms in such dire circumstances are likely to draw on whatever
resources are available tO them, including those represented by their boards of
directors, to stave Off declines in their businesses. They found that both Size of
the board of directors and the representation of outsiders on their boards
decreased in ﬁrms headed toward bankruptcy, suggesting these ﬁrms may have
felt the effects of reduced board resources during this critical period. Similarly, in
~ a longitudinal study covering a period Of ten years, Daily and Dalton (1994) found
that ﬁve years prior to ﬁling bankruptcy, ﬁrms that eventually failed had higher
percentages Of afﬁliated directors on their boards, (those with Close personal or
professional ties with the company), than ﬁrms that survived. Daily and Dalton
(1994) used lack Of director afﬁliation as a proxy for board Objectivity and
independence, suggesting that the resources represented on the board
inﬂuenced ﬁrm performance at a critical juncture. Chaganti et al. (1985), in their
study comparing failed and nonfailed ﬁrms in the retailing industry, found that the
amount Of board resources available (as measured by Size of the- board)
inﬂuenced ﬁrm success or failure, with larger boards associated with lack of

‘ failure, although they did not ﬁnd any relationship between outside director

I representation and ﬁrm failure.

I found no studies that directly compared the effects Of board of director
composition on ﬁrm outcomes in periods of relative tranquillity versus periods of
turmoil. However, taken as a whole, the evidence collected by board scholars
and researchers suggests that board resources are activated and important to

ﬁrm outcomes during periods Of crisis, signiﬁcant reorientation, or transition. The

37

strategic decision point I examined in this dissertation is the transition
experienced by relatively small ﬁrms entering a period Of high growth. In the next
section I discuss the literature regarding organizational stmcture, with the
implications of this Speciﬁc transition in mind.
High Growth Transition

As Pfeffer points out, one of the strategies that ﬁrms can pursue to
“attempt to ensure their survival and continued growth” is to focus on “improving
the efﬁciency of the internal transformation process” (220:1972). The
organizational transition that is the focus Of this dissertation is one that is
engendered by rapid growth of the ﬁrm. Although organizational growth can be
conceptualized in a variety of ways (Stemp, 1970 ), the sense in which it is being
used here is “simple growth,” or an increase in the Size of the organization
(Boulding, 1953). Haire (1959) took the position that organizational growth
results in some predictable (“lawful”) processes that originate from within the
ﬁrm. His view suggests that, as a ﬁrm grows, its internal organization must also
evolve in order to adequately meet the increased demands generated by the
larger entity (Chandler, 1962; Haire, 1959; Khandwalla, 1973). Hambrick and
Crozier agreed that “small ﬁrms which are quickly becoming big must modify
some of their organizational arrangements” (1985: 37). AS Caplow (1964)
indicated, the points in time at which changes in size of the organization occur
affect the nature Of the demands placed on the ﬁrm. When an organization is
small to begin with, rapid growth will likely have a different impact than when the

organization has already become a behemoth (Caplow, 1964). The speciﬁc

38

context of this dissertation is the examination Of the effects of high levels of
growth on relatively small ﬁrms.

The need for the ﬁrm to make Changes during this period of time, that is to
move from the nature Of the organization as it exists, to the type Of organization it
must become to support its growth and continue to function effectively, can be
characterized as a transition (Greiner, 1972; Hambrick 8 Crozier, 1985). Miller
and F riesen deﬁned transition as “a package Of Changes that occur between the
onset Of the imbalance or stress and the time when some equilibrium or tranquil
interval is reached” (1984a: 128). In the case Of the ﬁrms that are the subject Of
this dissertation, rapid growth signals the onset of stress that begins the need for
transition. This transition would be complete upon attainment Of equilibrium, or
the tapering off Of the ﬁrm’s rate Of growth such that the internal organization of”
the ﬁrm is aligned with the organization’s size (Greiner, 1972; Miller 8 Friesen,
1984a). This deﬁnition implies there may be periods of time during a transition
when a temporary alignment of Size andiinternal organization exist. However, a
persistent high growth rate suggests the ﬁrm will continue to be in a transition
period due to the need for further organizational changes to accommodate its
rapid increases in size.

Lundberg (1984) further reﬁned transitions into three types: emergence,
transformational, and termination transitions. Emergence transitions encompass
the evolution Of an organization from a concept to a concrete entity. Termination
transitions arise with the onset Of organizational decline. Temporally positioned

between these two types of transition are transformational transitions, which are

39

associated with the middle stages of an organization’s development. One Of the
“precipitating conditions” for a transformational transition cited by Lundberg
(1984) is organizational growth. In this dissertation, I am interested in the
transformational transition associated with rapid growth as a ﬁrm leaves its “birth
stage” of organizational formulation, and enters a “growth stage” Of development
(Miller 8 Friesen, 1984b).

As ﬁrms experience signiﬁcant growth, the stress placed on the existing
organization Often results in “growing pains,” or symptoms that Signal the need
for the ﬁrm to make Changes to respond to its increasing size (Flamholtz, 1990). I
follow the lead Of Hambrick and Crozier (1985), who examined the Challenges
faced by ﬁrms undergoing rapid growth, and focus on factors internal to the ﬁrm
that are likely to affect the degree Of success it realizes from its growth. One of
the symptoms of organizational strain due to growth is captured by Robinson’s
(1934) managerial inefﬁciency hypothesis. According to Robinson (1934), as a
small ﬁrm grows in size, the capabilities Of the entrepreneur are stretched further
and further, until the Operational efﬁciency of the ﬁrm suffers due to the inability
Of one person to meet the organization’s managerial needs. Rapidly growing
organizations need to develop new coordination, control, planning, and
communication functions which were not necessary for the ﬁrm to perform
effectively when it was smaller (Haire, 1959; Hambrick 8 Crozier, 1985;
Kazanjian, 1988; Stanworth 8 Curran, 1976). Starbuck reinforced the difference
between the informational needs of small organizations, which can be met

informally, and the requirements of larger organizations to have information

40

“supplied regularly and collected in a systematic manner” (1965: 480). Hambrick
and Crozier pointed out that successful high growth ﬁrms “used all available
channels for acquiring crucial information about the environment” (1985: 39).
DeCanio, Dibble, and Amir-Ateﬁ, through use of Simulation technology, assert
that a “failure to take account of organizational [structure] adaptation” would have
a serious Impact on a ﬁrrn’s ability to implement its plans effectively (1292-
3:2000). Thesmar and Thoenig (2000) also emphasize the importance of
matching organizational structure to the needs Of the organization.

James (2000) reinforces the impact Of organizational structure on
behavior within organizations. He draws particular attention to three aspects Of
structure that are inﬂuential: (1) reward structure, (2) performance evaluation,
and (3) responsibility assignment. As James (2000) points out, there are
multiple avenues by which the infrastructure needs of an organization may be
addressed. For growing organizations, the prescriptions are in several key
areas. One is the addition Of “staff” personnel, or those individuals who bring
specialized knowledge into the ﬁrm (Haire, 1959; Khandwalla, 1973). “The two
main functions Of the staff are to provide information for control and coordination,
and to provide expertassistance beyond the Skill or training of line executives”
(Haire, 1959: 289). Penrose (1959) echoed this principle when she emphasizes
the need Of a growing ﬁrm for additional information to reduce the uncertainty
created by growth. Penrose (1959) recommended that ﬁrms experiencing growth
increase the heterogeneity Of their internal resources by bringing in people who

have expertise that is not already possessed by those currently employed. An

41

increase in specialists also facilitates the division of labor within a growing
organization, another need for ﬁrms, which become too large to be run effectively
by a Single manager (Chandler, 1962).

As Specialists are added tO the upper echelons of the organization, it
Simultaneously assumes a more functional structure, with increased horizontal
differentiation (Blau 8 Schoenherr, 1971; Hall, 1972; Kazanjian, 1988). A
functional structure responds to the higher level of complexity involved in mnning
a business of rapidly increasing size (Chandler, 1962). The more complicated
administrative tasks facing the larger organization require a more complex
organizational structure (than a pre-bureaucratic form) in order for the .
organization to perform well in the long run (Miller 8 Friesen, 1984b; Smith 8
Gannon, 1987; Tashakori, 1980).

A second means by which increased control and coordination is achieved
is through standardization and formalization of policies and procedures Of the
organization (Blau 8 Schoenherr, 1971; Hall, 1972). Standardization involves
deﬁning or specifying a procedure or policy that has regular application within the
ﬁrm, while formalization denotes the degree to which policies and procedures are
written (Pugh, Hickson, Hinings, 8 Turner, 1968). Increased levels of
standardization and formalization mediate between the information people need
to do their jobs and task accomplishment, thereby promoting consistency'of
operations within an organization that is adding people and/or units at a rapid

pace (Katz 8 Kahn, 1966).

42

In another response to Robinson’s (1934) managerial inefﬁciency
hypothesis, some researchers emphasize the need for growing organizations to
hire “professional managers” who have had experience in larger business
environments and, presumably, an increased capability to manage such a ﬁrm
compared to more entrepreneurially oriented managers (Hambrick 8 Crozier,
1985; Tashakori, 1980). Providing some empirical evidence for this view, Miller
and Toulouse (1986) found a signiﬁcant relationship between the proﬁtability Of
small ﬁrms in dynamic environments and their employment of professional
managers and technocrats. Penrose (1959) pointed out that the current
management team Of a ﬁrm places limits on its capacity to support growth
effectively. When those managers do not possess the qualiﬁcations to run a
business that is becoming increasingly large at a rapid pace, the capability of the
ﬁrm to grow and perform well is compromised (Penrose, 1959). Rapid growth
compOUnds this problem, as the ﬁrm must add managers with the requisite
expertise to run a larger organization prior to the need for them, if it is to succeed
(Hambrick 8 Crozier, 1985; Penrose, 1959).

Lastly, Dwyer (1970) identiﬁed the importance Of continuous
communication among the employees Of the organization as a key faCtor in an
effective transition during periods of high growth. “In a growth situation the
meshing Of individual goals with company goals IS imperative” (Dwyer, 1970: 98).
Hambrick and Crozier (1985) concurred, ﬁnding that successful high growth ﬁrms

were able to preserve the notable Characteristics that helped make them

43

successful in the ﬁrst place by maintaining high levels Of contact among key
personnel.

Empirical research on the characteristics Of ﬁrms in a “growth stage” of
development indicate that the speciﬁc internal attributes described above are
associated with such ﬁrms: organizational structure increases in complexity,
usually taking on a functional form; specialized areas Of the business are
' supervised by separate managers with expertise in the area; policies and
procedures are standardized and formalized; emphasis is placed on gathering
and processing information and monitoring performance; and communication
occurs between departments to facilitate coordination (Hanks, Watson, Jansen,
8 Chandler, 1993; Miller 8 Friesen, 1984b). .

For the transformational transition associated with high growth to occur,
the organization requires resources over and above those it needs to carry out its
normal business operations (Lundberg, 1984; Penrose, 1959). This places a
strain on the existing organization, because a rapidly growing small ﬁrm is
“suddenly much, much bigger, but without necessarily having any aptitude or
preparation for being big” (Hambrick 8 Crozier, 1985: 35). Firms, which are
A relatively small Often, lack a surplus of resources, which may lead them to draw
upon the supplemental resources provided by the board Of directors, if they are
available, during a period of rapid growth (Dwyer, 1970; Hambrick 8 Crozier,
1985)

Bearing in mind the demands placed on organizations as they enter a

period of transformational transition brought on by rapid growth, I next describe

44

the theoretical model Of relationships and accompanying hypotheses examined
in my dissertation. This section of Chapter II is organized by section of the
model, with the relevant theoretical underpinning presented in advance of the
hypotheses testing each set of relationships. At the conclusion of each section, |
illustrate that portion of the model related tO the relationships examined by the
hypotheses presented thus far.
Relationship of the Composition of the Board Of Directors to Firms’ High
Growth Transition Changes

The degree to which the board of directors provides additional resources
to a ﬁrm undergoing the transformational transition triggered by high growth may
be a key factor in the ﬁrm’s ability to do so successfully (Hambrick 8 Crozier,
1985; Whisler, 1988). The relationship between board performance Of its
resource dependence role and the ﬁrm’s implementation of the internal
organizational strategies required to support growth is one Of the primary
questions I explored in this study. Both the knowledge directors bring to the
board based on their professional experiences, and their valuable contacts with
other organizations, can be drawn upon by a ﬁrm’s CEO in the course of this
signiﬁcant transition in a ﬁrm’s life. “Boards can do management an invaluable
service by viewing the broader business landScape and helping management
recognize majOr opportunities and discontinuities that will affect the business”
(Charan, 1998: 8).

The view that the inﬂuence of the board has the potential to be felt more

during periods Of organizational turmoil can be combined with the resource

45

dependence perspective that the effectiveness of the board in executing its
responsibilities depends on the match Of director attributes with the
contingencies facing the organization. The result is a position suggesting that
the inﬂuence of the board in its resource dependence role, i.e. the provision of
resources to the ﬁrm, may be considerable when the ﬁrm is undergoing a
signiﬁcant transition, 3 the resources it possesses are well-matched to the needs
Of the ﬁrm’s ability to make the transition. This dissertation explores the resource
dependence role of the board Of directors in one such transition: the
transformational transition of a high growth ﬁrm to its changing internal
organizational requirements. .

In empirical studies assessing the linkage between resources brought to
the firm by the board of director and the resource needs of the ﬁrm, the means
by which the types of board resources provided by directors are measured is
most commonly through the backgrounds of the directors themselves (Johnson
et al., 1996)‘. The resources directors bring to the board are usually captured by
classifying them into one of several categories. The most common categories
are outside directors, inside directors, and afﬁliated directors. As pointed out
previously, board researchers have expressed concerns over the lack Of
consistency with which these measures of board resource provision have been

Operationalized (Daily et al., 1999; Zahra 8 Pearce, 1989). There have also

 

‘ Although some board researchers consider the resource dependence perspective and the
research on board interlocks, i.e. a director concurrently sits on the board of more than one
company, as one literature, Zahra and Pearce (1989) distinguish the two. In their view, the board
interlock approach is primarily concerned with a ﬁrm’s competitive relationships, while the
resource dependence perspective is more generally related to the external environment as a
whole. Since the resources secured through board interlocks are not relevant to the board

46

been criticisms leveled at the frequent lack Of congruence between the measure
of director background used and the theoretical foundations relied upon in the
study (Daily et al., 1999; Dalton et al., 1998).

Hillman et al. (2000) have addressed this issue related to the resource
dependence perspective by identifying four categories of directors, based on the
nature Of the ﬁrm’s resources needs that they address. Hillman et al.’s (2000)
ﬁndings suggest that ﬁrms undergoing deregulation in the airlines industry
Changed the composition Of their boards of directors to more closely match their
revised resource needs, in accordance with the authors’ predictions using their
four categories of directors. Two of these designations are relevant to the current
study. “Business Experts,” (current and former senior Ofﬁcers or professional
directors of large for-proﬁt ﬁrms), serve as sounding boards for top management,
and provide outside managerial business expertise. “Support Specialists,” (such
as lawyers, bankers, insurance executives, public relations experts), provide
specialized expertise in more technical areas Of the business. Whisler (1988), in
his article regarding the role of the board in ﬁrms making the transition to public
ownership, uses very Similar categories for the appropriate roles of outside
directors in such ﬁrms, as do Castaldi and Wortman (1984) in their article on the
boards in small corporations.

Business experts are in a position to offer general executive expertise in
when providing advice and counsel to the CEO. AS outside directors, they have

a long-term perspective and Objectivity that inside directors may lack. They are

 

inﬂuences being examined here, I adopt Zahra and Pearce's (1989) analysis, and characterize the
resource dependence perspective apart from the board interlock literature.

47

also likely to have contacts with organizations in the external environment that
may facilitate the acquisition of resources the ﬁrm needs to implement the plans
to make the transition to the growth stage Of development.

Hypothesis 1a: The number of business experts on the board Of directors

will be positively related to the extent to which the ﬁrm implements high

growth transition infrastructure.

Support Specialists possess expertise of a more technical nature than
business experts. However, they also possess an exposure to the broader
business environment that they can draw upon in their positions as directors for
growing ﬁrms. Their business experiences and Objectivity as to the ﬁrm’s long
term needs are also apt to lead them to offer advice and counsel as well as
external contacts related to increasing infrastructure complexity.

Hypothesis 1b: The number Of support Specialists on the board of

_ directors will be positively related to the extent to which the ﬁrm
implements high growth transition infrastructure.

The ﬁgure below illustrates the portion of the research model that

corresponds to Hypotheses 1a and 1b just described.

 

 

Board High Growth
Composition ——> Transition
Changes

 

 

 

 

 

 

Figure 2.1 Theoretical Model A

48

Implications for Firm Performance

Increasing organizational complexity and professionalization Of
management are associated with larger ﬁrms in the organizational literature
(Chandler, 1962; Haire, 1959; Penrose, 1959) as a means of supporting the
needs for coordination and control that are more pronounced in ﬁrms that have
greater numbers of employees. However, the empirical evidence that the
implementation Of such changes in ﬁrms that are relatively small but undergoing
rapid growth will lead to better ﬁrm performance than that of ﬁrms that ignore
such mandates is sparse. I F

Miller 8 Friesen (1983) examined the differences between succeszuI and
unsuccessful ﬁrms (based on growth in proﬁts and sales) at various stages of
development. Miller and Friesen (1983) used three variables associated with
information processing and three variables associated with decision making as
indicators of differences across the ﬁrms. Collapsed across stages, there were
Signiﬁcant differences between successful and unsuccessful ﬁrms on all six
variables. Beginning with the growth stage Of development, successful ﬁrms
showed a linear progression Of increased sophistication in decision-making and
information processing over their life cycle stages, while the unsuccessful ﬁrms
showed no such pattern. Miller and Friesen. (1983) concluded that ﬁrms need to
continuously improve their skills in these areas as they move through the.
different stages of development if they wish to remain successful.

Hambrick and Crozier (1985) studied ﬁrms that experienced a rapid

growth rate and drew conclusions about the attributes Of those they

49

Characterized as “stars” versus those considered “stumblers." They found that
the high growth ﬁrms with stronger performance demonstrated a tendency to hire
professional managers, i.e. those with experience in larger ﬁrms, in preparation
for the demands Of a larger organization. They also invested in the training and
development Of their middle managers, in order to develop managerial talent in-
house for future stafﬁng needs. Hambrick and Crozier (1985) also Observed that
the more successful high growth ﬁrms in their study took steps to maintain a high
level of communication and commitment throughout the organization as it grew.

Although the empirical evidence Of a relationship between internal
organizational changes customarily associated with growth and ﬁrm performance
is scanty, organizational theorists have extensively documented the degree to
which ﬁrms rely upon their internal resources to support growth (Chandler, 1962;
Galbraith, 1973; Penrose, 1959) and the importance of aligning their structures
with the strategic needs of the organization (Burns 8 Stalker, 1961; Lawrence 8
Lorsch, 1967; Woodward, 1958). The implication Of this alignment is that
managers who fail tO take notice of and respond to the Signals that their ﬁrm iS in
need of internal changes will pay a price for their neglect. There iS, therefore,
reason for me to expect that the degree to which a ﬁrm responds appropriately to
the transitional demands of its growth will have an effect on the subsequent
performance Of the ﬁrm.

Hypothesis 2: The extent to which the ﬁrm implements high growth

transition infrastructure will be positively related to subsequent ﬁrm

performance.

50

The relationship of board resources to ﬁrm performance has been
debated in the board Of director literature. Some researchers have found a
signiﬁcant link between the type Of directors Sitting on a ﬁrm’s board and the
performance Of the ﬁrm (Baysinger 8 Butler, 1985; Pearce 8 Zahra,1992;
Schellenger et al., 1989). There is countervailing evidence, however, that board
composition and ﬁrm-performance are unrelated (Chaganti et al., 1985; Hermalin
. 8 Weisbach, 1991; Zahra 8 Stanton, 1988). In their meta-analysis of board
composition and ﬁrm ﬁnancial performance using 54 empirical studies and 159
different samples of ﬁrms, Dalton et al. (1998) found no practically meaningful
relationship between composition and ﬁrm performance.

However, in a study speciﬁcally set in the same context as this
dissertation, smaller growth-oriented ﬁrms, Daily and Dalton (1992) found a
statistically signiﬁcant relationship between the composition Of the board Of
directors and ﬁrm performance as assessed by a market measure of
performance. The high-grth ﬁrms that performed the best had higher
proportions Of outside directors represented on their boards than their
counterparts. Similarly, Hambrick and Crozier found that successful high growth
ﬁrms “try to tap big-company expertise on their boards of directors” in advance of
problems surfacing, SO they could draw on the experience of their directors to
help avoid mistakes being made (1985: 37).

Although Hillman et al. (2000) did not test the relationship of board
composition to ﬁrm performance using their four director categories, their study

does address one of the criticisms being leveled at research regarding board

51

composition, the congruence Of measurement with underlying theory. As
reported earlier, they found that their categorizations Of directors were related to
ﬁrms undergoing changes in their resource needs. This study employs two Of
Hillman et al.'s (2000) measures of board composition based on the provision Of
resources by the board, business experts and support specialists. In light of the
mixed evidence regarding the relationship Of director type and ﬁrm performance,
and the use of theoretically relevant operationalizations Of board composition in
this study, I Offer the following hypotheses:

Hypothesis 3a: The number of business experts on the board of directors

will be positively and directly related to subsequent ﬁrm performance.

Hypothesis 3b: The number of support Specialists on the board Of

directors will be positively and directly related to subsequent ﬁrm

performance.

Although the board of directors literature has not been able to reliably
assess the relationship between board composition and ﬁrm performance, there
is evidence that the resources the board brings to the ﬁrm are important for
outcomes that are more proximal to board inﬂuence, such as selection of a new
CEO (Boeker 8 Goodstein, 1993), fund raising (Provan, 1980; Zald, 1967), and
borrowing practices (Stearns 8 Mizruchi, 1993). The nature of ﬁrms’ strategic
decisions have'also been successfully linked to board of director composition
(Baysinger, Kosnik, 8 Turk, 1991 ; Hill 8 Snell, 1988; Zahra 8 Stanton, 1992). It
is likely that these interim outcomes, given that they are integrally related to

important functions of the ﬁrm, ultimately have an effect on ﬁrm performance.

52

However, few studies have explicitly tested a mediation model of the inﬂuence of
board resources on ﬁrm performance.

Judge and Zeithaml (1992) examined the effects Of board composition on
an interim ﬁrm outcome, board involvement in strategic decision-making, and the
subsequent effect Of that outcome on ﬁrm performance. In their study using
structural equation modeling, Judge and Zeithaml (1992) found some evidence
of indirect effects Of board composition on ﬁrm performance. In another study
using structural equation modeling to assess board inﬂuences on corporate,
divestiture intensity, Hoskisson, Johnson, and Moesel (1994) found a signiﬁcant
positive relationship between the proportion Of outside directors on the board, '
and relative debt intensity of the ﬁrm. Relative debt intensity was, in turn,
signiﬁcantly negatively related to the ﬁrms' relative market performance, again
suggesting a mediated effect of board composition on performance.

In this research model, I have already hypothesized that the composition
Of the board Of directors inﬂuences the interim outcomes of the degree to which
a ﬁrm makes the internal organizational Changes associated with a successful
high growth transition. I have also presented the theoretical reasoning behind
the anticipated effects Of these transitional changes on subsequent ﬁrm
performance. I further expect that these high growth transition changes will
mediate the relationship between board composition and subsequent ﬁrm
performance, such that at least part Of the effect Of board composition on ﬁrm
performance will be felt through the degree to which it affects internal

organizational changes associated with high growth.

53

Hypothesis 4a: The relationship between the number Of business experts
on the board of directors and the subsequent performance Of the ﬁrm will
be partially mediated by the extent to which the ﬁrm implements high
growth transition infrastructure.

Hypothesis 4b: The relationship between the number of support
Specialists on the board of directors and the subsequent performance Of
the ﬁrm will be partially mediated by the extent to which the ﬁrm
implements high growth transition infrastructure.

Figure 2.2 illustrates the research model with the ﬁrm performance

implications added.

 

 

 

Board High Growth Firm
Composition —> Transition -——> Performance
Changes

 

 

 

 

 

 

 

 

 

 

Figure 2.2 Theoretical Model B

The Role of the Chief Executive Ofﬁcer

The Chief executive Of the ﬁrm also represents resources the ﬁrm may
draw upon in making a high growth transition. Much attention has been given in
the research literature to identiﬁcation of the differences in attributes between
“entrepreneurs” and “professional managers" (Daily 8 Dalton, 1992; Dyer, 1989;

Tashakori, 1980; Willard, Krueger, 8 Feeser, 1992). Robinson’s (1934)

“managerial inefﬁciency hypothesis” discussed earlier is consistent with the view
that the capabilities of the founder can only be stretched so far before ﬁrm
results will begin to deteriorate. The common conclusion is that entrepreneurs
who found ﬁrms must be replaced with professional managers in accordance
with the changing demands placed on the ﬁrm by its growth in order for it to
succeed long term (Dyer, 1989; Tashakori, 1980). However, Smith‘and Miner
(1983) identify two types of entrepreneur. The ﬁrst is considered a “craftsman”
who is “Characterized by narrowness in education and training, low social
awareness and involvement, a feeling Of incompetence in dealing with the social
environment, and a limited time orientation” (1983: 326). The second type of
entrepreneur is “opportunistic” and exhibits Characteristics that are the
counterpoint to those Of the craftsman entrepreneur, e.g. has a breadth of
education, is socially adept, and relates to a time frame in the future. Smith and
Miner.(1983) conclude that when the organization grows to the point Of needing
professional management, it is the orientation of the entrepreneur that is
relevant, not the fact that she founded the ﬁrm, in determining whether she
needs to be replaced for the ﬁrm to progress.

Rubenson andGupta (1992, 1996) take an approach that iscompatible
with that of Smith and Miner (1983). If the organizations needs are changing, aS
they are when ﬁrms are undergoing the birth-to-growth transition, then the
relevant question becomes, is the founder able to adapt? Rubenson and Gupta
(1996) suggest that the level of education, general management experience,

breadth Of functional experience, and industry experience Of the founder be used

55

as criteria in assessing the likelihood Of his ability to make the transition
successfully. They conclude that “it is only when the ﬁrm’s development calls for
more change that the founder is able or willing to make that the founder’s
replacement by a ‘professional manager’ Should be necessary (1996: 54).

The supplemental resource needs of the organization in making the
growth stage transition are, therefore, likely to be dependent on the managerial
qualiﬁcations Of its chief executive (Hambrick 8 Crozier, 1985; Mace, 1948).
When the CEO Of the ﬁrm has the background to draw upon to effect the ﬁrrn’s
transition successfully, less reliance may be placed On the board of directors to
supplement the ﬁrm’S deﬁcient resources (Golden 8 Zajac, 2001; Rubenson 8
Gupta, 1996; Westphal 8 Frederickson, 2001 ). Therefore, when CEOS already
possess sufﬁcient experience to identify and execute the Changes involved with
a high growth transition, I expect the inﬂuence of the board Of directors will be
reduced.

Hypothesis 5a: The relationship between the number of business experts

on the board Of directors and the extent to which the ﬁrm implements high

growth transition infrastructure will be weaker when the managerial
qualiﬁcations Of the CEO are greater.

Hypothesis 5b: The relationship between the number of support

specialists on the board Of directors and the extent to which the ﬁrm

implements high growth transition infrastructure will be weaker when the

managerial qualiﬁcations of the CEO are greater.

56

Figure 2.3 illustrates the addition of the moderating role Of the Chief

executive Ofﬁcer in the research model.

 

CEO
Managerial
Qualiﬁcations

 

 

 

 

 

 

Board High Growth Firm
Composition ———> Transition ._, Performance
Changes,

 

 

 

 

 

 

 

 

 

 

Figure 2.3 Theoretical Model C

Implications of the Rate of Growth of Firms

Firms that have undertaken plans to grow are likely to vary in how quickly
their ﬁrms undertake this strategy. For some ﬁrms, the plan may be to grow in a
controlled, gradual manner that does not overly tax the organization’s resources
(Churchill 8 Lewis, 1983). For others, their strategies may involve exploitation of
the immediate Opportunities presented by the marketplace as quickly as possible
to usurp the moves of any competitors that may seek to enter (Hambrick 8
Crozier, 1985). For these ﬁrms, a faster growth rate would be expected.

In addition, the faster the ﬁrm is growing, the more likely it is to have a
need to have these internal resources in place in order to perform well (Penrose,
1959). As Hambrick and Crozier point out, “the successful ﬁrms tend to

introduce these new systems gradually and Often in advance Of when they are

57

absolutely crucial” (1985: 37). A failure to allow time for the assimilation of new
people and other resources may make them less useful to the ﬁrm (Penrose,
1959)

Therefore, the growth rate Of the ﬁrm should have an effect on the
relationship between the degree to which it has accomplished a high growth
transition and subsequent ﬁrm performance. Since growth is dictating the
' necessity of the internal changes in order to support the expanding ﬁrm’s
requirements, the rate at which the ﬁrm is growing is likely to have an effect on
how the ﬁrm fares later on as a result of the effectiveness of these changes. For
firms that are growing more rapidly, the failure to implement internal
organizational changes will likely have a more detrimental impact on
performance. When ﬁrms are growing more slowly, the effects of the transitional
modiﬁcations are less likely to be felt on later ﬁrm performance, since the ﬁrm
will not be as burdened by the demands of its growth.

Hypothesis 6: The relationship between the extent to which the ﬁrm has

implemented high growth transition infrastructure and subsequent ﬁrm

performance will be stronger when the growth rate Of the ﬁrm is greater.

Figure 2.4 presents the full research model tested in this study, along with
the hypotheses representing each relationship that was tested. Next, Chapter III
presents the research design, data collection and coding procedures, the

variables included in this study, as well as the methods of analysis I used.

58

.30.: 363202.... 86.9.50 RN 952“.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:OEoanoo

 

 

 

 

 

 

2..
83. 539.0
E.
0:
39.2.0 PI
cocaEcotOn. i ocauuahmatﬁ I
ELK 22:39.... I Euom
N: :320 :9...
m:
98:35.50
Krona—Es.
Omo

£8509»:
cosmic:

 

 

 

59

CHAPTER III
RESEARCH DESIGN AND METHODS OF ANALYSIS

In this Chapter I describe the research design I used for this study, the
variables in the analyses, and the methods Of analysis | used to test my
hypotheses.
Research Design

Due to the relationships Of interest in this study, (1) the impact Of the
resources represented on the board Of directors on the degree to which a ﬁrm
makes future high growth transition changes, and (2) the effects Of both the
board of directors and the nature of the ﬁrm’s high growth transition on
subsequent performance, I selected a longitudinal research design. I expected
that examining these relationships over an extended period would provide a
more complete picture Of the relationships of the constructs to one another since
they require time to develop. The sources of data for this study are archival, with
data collected at varying points over a four-year period. I provide more
information regarding the data collection process below.
Sample

The proﬁle of ﬁrms I selected for testing the hypotheses Of this study are
ﬁrms that are poised to move into a period of growth. TO meet this criterion, the
sample was randomly drawn from the population of ﬁrms that made an initial
public Offering (IPO) of their stock in 1996. Firms undergoing an IPO are typically
seeking an infusion of cash to support future growth (Lipman, 1994; Weiss,

1988). I eliminated ﬁrms from the sample that (1) did not include growth as a

60

stated purpose of their IPO, (2) produced no good or service and, therefore, had
little need for an organizational structure, (3) were not US. ﬁrms, (4) had over
2,000 full-time employees (Miller and Friesen’s [1984b] empirical study capped
the size of growth stage ﬁrms at 2,000 employees), (5) were not still in business
as an independent entity in 1999 (Since this was the year for which I collected
subsequent performance data). Based on a power analysis for this study, data
on 250 - 300 ﬁrms were needed to detect an effect Size of .20 with 90%
likelihood. In the absence of any guidance from previous studies in this area, I
selected an effect size Of .20 since it is midway between a small and a medium
effect size according to the parameters set out in Cohen and Cohen, 1983. I
collected data on a total sample of 296 ﬁrms, although missing data reduced the
number Of ﬁrms for some analyses.
Data Collection and Coding

Data were collected from archival sources including the prospectus
associated with the ﬁrm’s IPO, 10Ks, and annual reports. See Appendix A for a
complete list Of the documents used and the timing of data collection for each of
the variables in this study. I was able to secure prospectuses and 10Ks for 1997
and 1998 for all ﬁrms included in my sample. For some ﬁrms, however, annual
reports were not available. (For 1997, annual reports for 50 ﬁrms were not
available; for 1998, annual reports for 104 ﬁrms were not available). Therefore, I
Included two dummy variables in my analyses, one for the 1997 annual report (1
if present, 0 if missing) and one for the 1998 annual report (1 if present, 0 if

missing).

61

A standardized format was used to record relevant information from the
archival sources and to code the data after it was collected. See Appendix B for
a copy Of my Data Collection and Coding Sheet. I describe the speciﬁc
procedures used to collect data for each Of the study variables below.
Independent Variable Measures

Board Resources. The two measures of board resources | used in this
study were outside directors who are either “business experts” or “support
specialists” (Hillman et al., 2000). Business experts are those directors who are
current and former senior Ofﬁcers or professional directors of large for-proﬁt
ﬁrms. Support specialists are directors who provide specialized expertise, such
as attorneys, ﬁnancial experts (including bankers and representatives of
investment ﬁrms), public relations personnel, and insurance company
representatives. TO measure business experts and support specialists, I
counted the number in each category sitting on the board. Board size was used
as a control variable to scale the raw number Of business experts and support
specialists.

Since I was searching for a relationship between board resources and the
changes in infrastructure subsequent to the ﬁrm’s IPO (I collected data on the
infrastructure variables at the end of the ﬁscal years 1997 and 1998), I wanted to
measure board composition at the point in time most likely to be associated with
the implementation of such changes. I found that board composition frequently
changed shortly after the IPO, perhaps based on the CEO’S or other board

members’ reassessments Of what types of directors were most suitable for the

62

growing ﬁrm. Sixty-two percent of the ﬁrms had some kind Of change in the
composition of their board of directors between the dates Of the 1996 IPO
prospectus and the 1997 proxy statement, although the size of the board
remained much the same (an average of 5.98 directors in 1996 and 6.18 in
1997). More Speciﬁcally, over a third Of the boards had a different number of
business experts and support specialists as Of the 1997 proxy compared to the
1996 IPO prospectus. With this degree Of change in the boards during this Short
period, I concluded that the 1996 board did not accurately represent the directors
most likely to inﬂuence future infrastructure changes. Therefore, I decided that
the board composition in early 1997 was more likely to be related to Changes in
infrastructure occurring in 1997 and 1998. I collected data regarding the
composition Of the board from the ﬁrm’s proxy statement ﬁled in 1997.1

CEO Managerial Qualiﬁcations. l adapted the criteria for measuring the
managerial qualiﬁcations Of the CEO from Rubenson and Gupta (1996).
Rubenson and Gupta’s model (1996) drew on the conceptual foundations Of
Hambrick and Mason (1984) which described the inﬂuence Of the backgrounds
and experiences Of key organizational decision makers on their perceptions and
behavior in their positions. I measured four areas of CEO managerial
qualiﬁcations with all data collected from the CEO biographical information
provided in the ﬁrm’s 1996 prospectus. First, the level of education of the. CEO
was obtained. Educational level is an important consideration in evaluating

managerial capability Since increased education is likely to enhance critical

 

‘ I also ran the analyses including the board resource variables with 1996 board data. There was
no difference in the results I obtained, so I maintained the 1997 board variables in my ﬁnal

63

thinking and the ability to consider issues in a systematic fashion and on a broad
scale. I measured the highest level of education obtained by the CEO using an
ordinal scale, where 0 = no college education, 1 = some college education, 2 =
undergraduate college degree, 3 = master’s degree, and 4 = advanced or
professional degree. Due to the lack of consistent reporting of CEO educational
background in company ﬁlings (data were available on only 149 out Of 296
CEOS), this variable was subsequently dropped from my analyses.

Second, I collected data on the years of general management experience
the CEO had prior to joining the IPO ﬁrm. Experience in a corporate setting at
the senior management level is likely to enable the CEO to anticipate the issues
that will confront her growing organization and deal with them more effectively. I
calculated the total number of years of prior experience the CEO had had in
corporate business positions with titles of CEO, COO, President, and Executive
Vice President. Prior experience as a Senior Vice President was also included if
the position entailed heading a business division of a large corporation.
Experience was not counted if the CEO had been the founder of the business or
if the organization was not a for-proﬁt business organization.

The third aspect Of CEO background I measured was breadth Of
functional experience. A diversity of functional experiences ”may enable the CEO
to identify problems and opportunities in a variety of areas Of the business while
she is in the process of assembling a complete management team of specialists.
I counted each major area of functional experience in which the CEO had

previously worked, including operations, manufacturing/production, sales and

 

analyses.

64

marketing, research and development, finance, business development and
strategic planning, or specialized experience in a discipline related to the IPO
ﬁrm’s business, e.g. engineering, information technology, medicine, real estate,
etc. An area Of functional experience was counted whether or not it had been
Obtained in the same industry as the IPO ﬁrm for which the CEO currently
worked. This data, like educational level, was not consistently available from

' company ﬁlings (data were Obtained on 174 Of 296 CEOS). Therefore, this
variable was also subsequently dropped from my analyses.

Lastly, I measured the number of years of industry experience the CEO
had prior to joining the IPO ﬁrm. Issues confronting the ﬁrm may be idiosyncratic
to the particular industry in which it competes and the CEO with previous
experience in that industry may be better able to understand and prioritize the
speciﬁc needs of her ﬁrm as it grows. l relied upon general industry descriptions
in calculating the number of years Of prior industry experience. The prospectus
provided a detailed description of the IPO ﬁrm’s business and the biographical
data for the CEO was sufﬁciently detailed to assess if the experience were in the
same or a different industry. For instance, the prospectus for IPO Company X
indicated it was in the business of providing optical storage systems. Its CEO
had previous experience as Director of Marketing for a duplicating machine
company and was President of a marketing services company. I concluded that
he had no prior industry experience. In IPO Company Y’s prospectus, on the
other hand, its business description indicated it sold motor vehicle installment

contracts. This company’s CEO had prior experience as the Executive Vice

65

President of an auto ﬁnance division of a large ﬁnance corporation, where he
managed a $1 billion portfolio Of automobile loan receivables. I considered this
experience to be in the same industry as the CEO’s current ﬁrm.

Rate of Growth. I measured the actual rate Of growth of the ﬁrm
subsequent to its IPO in two ways: growth in employees and growth in revenues.
Growth in employees and growth in revenues are the two primary means by
which previous studies in the business environment have assessed
organizational growth (Price 8 Mueller, 1986). AS discussed in Chapter 2, I
expected rate Of growth Of the ﬁrm to be a major inﬂuence on the relationship
between ﬁrm infrastructure development and subsequent ﬁrm performance. I
followed the lead Of other empirical studies of organizational growth (Cf. Baum,
Locke, 8 Smith, 2001; Hanks et al., 1993) and measured growth rate by
subtraCting ﬁrm employees, (or revenues) in 1996 from number Of employees (or
revenues) in 1998 and dividing by the 1996 level of the variable. I collected
information on both measures Of growth from the ﬁrm’s 1996 IPO prospectus
and its 1998 10K.

Dependent Variable Measures

High Growth Transition Infrastructure Variables. Miner (1982) points
out the problems associated with measuring variables to represent the types of
constructs inclirded in this study, especially given the archival nature of the data
collection process. The variables presented here are intended to be
representative of the issues raised in the literature describing the characteristics

of organizational infrastructure needed to support ﬁrm growth. Wherever

66

possible, I relied on deﬁnitions of constructs available in the research literature in
developing my measures.

The infrastructure measures l devised and the data collected were also
inﬂuenced by the sources Of data I had available for this study. Data for these
variables were collected at three different times. Time 1 was from the ﬁrms’
1996 IPO prospectuses. This measurement Of each variable served as a control
variable representing the degree to which the ﬁrm had implemented
infrastructure prior to its IPO. Time 2 data collection was from ﬁrms’ 10KS and
annual reports for 1997 with Time 3 collection from these same documents for
1998. Data for 1997 and 1998 were combined to form the high growth transition
infrastructure change dependent variables. By collecting data from these three
sets Of documents, I was able to capture the presence or absence Of
infrastructure in support Of the ﬁrm’s high growth transition just before and for
two years following the actual IPO event. ~

For measures coded in a binary fashion (0 if absent and 1 if present), the
maximum value was 1. This means that once one Of these measures was
detected as present in the ﬁrm, no further instance Of the measure was counted.
Appendix C contains a complete list Of the measures comprising each of the high
growth transition infrastructure variables, which documents and years were used
for data collection for each measure, what method was used to identify the
relevant data, how the measure was Operationalized for analysis, and an

example of the type of data collected for each measure.

67

For each measure, I developed a series Of “key words” to search for
evidence of the speciﬁc aspect of infrastructure. (Most of the SEC ﬁlings for this
sample Of ﬁrms were submitted as “Word” documents, which permitted the
search of their contents through the use key words). These key words were
identiﬁed based on a pilot study with ﬁrms not included in the sample. I
developed an initial list Of key words based on the organizational structure
literature, supplemented by a general reading Of company ﬁlings and documents.
For instance, to locate information regarding the “management development”
measure, I initially identiﬁed the following keywords: “training,” “development,”
“seminar,” “recruiting,” and “recruitment.” Similarly, I looked for evidence of the
existence Of a strategic planning process using keywords “strategy," “strategic
plan,” “planning,” and “decision making.”

The key words I initially developed were then used to perform actual
searches of documents and edited to a list of those that were most effective and
reliable in identifying evidence of each Of the measures Of interest. For instance,
the ﬁnal list of keyWords for management development became “management
development,” “train,” “seminar,” “recruit.” Through the use Of my original list of
keywords for management development, I discovered that using the keyword
“development” resulted in many “hits” in the documents on development of
products and services, which were not relevant to my search. Therefore, I
Speciﬁed “management development” as a keyword, which made the search far
more productive. I also substituted “train” for the keyword “training” that I had

used in the pilot study, Since adding the gerund unnecessarily restricted the

68

search. Evidence of management development programs could then be located

I, u I! u

if the words “train, trained, trains,” or "training” appeared in the document.
Similarly, “recruiting” and “recruitment” were Shortened to “recruit.” The keyword
“seminar” was maintained unchanged.

In using this methodology to look for indications of high growth transition
variables in sample ﬁrms, I was seeking to assess evidence Of infrastructure
development from three vantage points: (1) as an indication that the ﬁrm is
cognizant that changes need to be made to support growth, (2) as representative
of the changes a particular ﬁrm is in the process of making and, (3) as evidence
that the ﬁrm is sending Signals to its shareholders that it isundertaking the
Changes expected of it as it continues to grow.

I developed a total of fourteen measures that I expected to form ﬁve
variables representing key aspects Of infrastructure for high growth ﬁrms:
standardization and formalization (Six measures), Shared goals (ﬁve measures),
professionalization (one measure), management development (one measure)
and specialization (one measure). -I planned to create a numerical value for
variables with more than one underlying measure by summing the scores on the
measures that comprised it. As mentioned earlier, Appendix C contains detailed
information on the collection of data for the high growth transition infrastructure
variables. I describe the infrastructure variables and measures I used in this
study in detail below.

A total Of four recorders were used to collect the data for most Of the

infrastructure variables for this study: me and three graduate students who were

69

trained by me tO collect the data. The training process for the graduate students
was conducted in two stages. The first stage was an orientation and formal data
collection criteria session that lasted approximately an hour and a half. Graduate
students were given a detailed set Of instructions for collecting data, which I
thoroughly reviewed with them. I then gave each student several ﬁrms for which
they collected data using the Data Collection and Coding Sheet (Shown in
Appendix B). I also collected data on the same ﬁrms. I then met with each
graduate student to compare my data collection results to theirs. We discussed
any discrepancies, I answered questions, and we ﬁne-tuned their data collection
techniques, if necessary. Next, I describe each of the high growth transition
variables used in my study.

Standardization and formalization. As a ﬁrm grows, having policies and
procedures standardized is a critical means of insuring coordination and
consistency within an increasingly larger organization (Charan, Hofer, 8 Mahon,
1980). Pugh et al. describe formalization as the “extent to which rules,
procedures, instructions, and communications are written” (1968: 75). To
identify instances Of this type Of infrastructure development, I developed a list of
measures that serve as indicators Of a ﬁrm’s level of standardization and
formalization. | relied upon descriptions in Pugh et al. (1968), Hanks et al.
(1994), and Miller and Friesen (1984b) in arriving at the following measures for
this variable.

First, evidence was sought for the formalization of policies and procedures

within the ﬁrm. To assess the degree to which this aspect Of infrastructure

7O

existed within the ﬁrm, data collection focused on indications that policies and
procedures existed and were codiﬁed (such as the existence of quality control,
ISO standards, Operating manuals, etc.). This measure was coded as “0” if no
evidence was found and “1 ” if there was evidence of formalized policies and
procedures. Second, I sought evidence of employee training program since
training is a means by which companies can insure standardization Of the way in
which employees perform theirjobs. Examples Of ongoing employee training
programs were coded either 0 (absent) or 1 (present).

Next, I examined the degree tO which performance measurement and
compensation plans existed within the ﬁrm. I sought indications of objective
criteria for appraisal of performance and administration Of compensation
decisions for this measure. I identiﬁed two separate measures for performance
and compensation: those that related Speciﬁcally to top management (coded 0 or
1) and those that included employees below the top management level (coded 0
or 1). I also examined documents for the existence Of ﬁnancial reporting and,
budgets. Examples Of this measure consisted Of information systems reporting'of
ﬁrm performance, budgetary processes, and Systems of ﬁnancial controls. This
measure was also coded 0 or 1. Lastly, I sought indications that the ﬁrm had a
formal strategic planning process, coded 0 or 1. This variable was coded 1 only if
the documents showed evidence Of a complete strategic plan being in place, with
multiple areas Of the business being addressed.

Shared goals. A key aspect Of making a successful high growth transition

is the ability to sustain cohesion among the employees Of the ﬁrm, as well as

71

alignment Of employees’ interests with those Of the ﬁrm. Flamholtz (1990) and
Hambrick and Crozier (1985) emphasized the importance Of maintaining a
common bond within the ﬁrm that creates a sense of involvement and
commitment to the organization’s goals. I developed three measures related to
Shared goals, relying on Flamholtz (1990), Hambrick and Crozier (1985), and
Miller and Friesen (1984b).

First, I looked for evidence Of the existence of a mission or vision
statement. A published mission or vision statement available to employees is
one means of insuring a uniﬁed view of broad organizational goals. This
measure was coded as absent (0) or present (1 ). Second, I looked for company-
wide communication tools, such as newsletters, e—mail systems, or other
communication programs. The presence Of such systems would indicate efforts
are being made to keep employees consistently informed regarding critical
company decisions and events. Evidence Of this measure was coded 0 or 1.
Third, I sought indications Of three types Of risk sharing plans, including
employee stock option plans, employee bonus plans, and employee stock
I purchase plans. Ownership of a company’s stock and bonuses based on
performance are typical means by which ﬁrms attempt to align the interests Of
employees with those Of the company as a whole. To be counted as evidence of
shared goals, such plans had to include employees in general (not just top
management). Coding for these three measures was 0 or 1, with each Of three
plans (stock option program, bonus plan, and employee stock purchase plan)

counted as present or absent.

72

Professionalization Of management. Chandler (1962) and Parsons
(1960) both refer to the professionalization Of management in the modern
corporation as the increased presence of non-owner managers with the
capability to administer the ﬁrrn’s business within a formalized structure. More
speciﬁcally, professionalization Of management in the context Of a high growth
transition indicates senior management that uses “written plans, formal
evaluations, reward systems, and formal structure” in order to accomplish
organizational goals (Tashakori, 1980: 21-22). Professional managers are
“people who are adept at formal administration, planning, organization,
motivation, leadership, and control" (Flamholtz, 1990: 40). Professional
managers are characterized by having had experience in a corporate
environment and thereby gaining an understanding Of how to run a more
complex business (Daily 8 Dalton, 1992; Flamholtz 8 Randle, 1987; Hambrick 8
Crozier, 1985; Miller 8 Friesen, 1984b).

Based on the characterizations of professionalization in the literature, I
measured professionalization of management by averaging the number of years
Of previous management experience of the executive Ofﬁcers of the ﬁrm,
excluding the CEO. This measure provided an indication of the degree to which
the members of senior management had “professional” experience. This
information was collected from the biographical information provided in the ﬁrm’s
1996 IPO prospectus and its 1998 10K. (In some cases, this information was
not provided in the 10K document. Instead, it was incorporated by reference to

the ﬁrrn’s proxy statement. In these situations, I used biographical information

73

from the 1999 proxy statement for those Ofﬁcers who were employed as of the
end Of the 1998 ﬁscal year).

A second indicator of the ﬁrm’s professionalization of management that is
important for an effective growth stage transition, according to Flamholtz (1990),
is the introduction of management development programs to prepare future
managers for the ﬁrm’s growing needs. Hambrick and Crozier (1985) agree this
is a critical means Of building a professional team for the future. Therefore, a
second measure of professionalization of management I used was whetheror
not the ﬁrm had a program for developing managerial skills. This measure was
scored as “present” with a “1 ”or “absent” with a “0."

Because one Of these measures is continuous (years of corporation
experience) and one is dichotomous (management development program),
interpretation of a variable composed Of the two measures combined would be
very difﬁcult. Therefore, the measure Of years Of prior management experience
of the executive Ofﬁcer group is used in my analyses as one measure“ Of
professionalization (hereinafter referred to as the “professionalization" variable)
and the existence Of a management development program (hereinafter referred
to as the “management development” variable) is treated as a second measure
Of professionalization.

Specialization. Another aspect of making a successful high growth
transition is increasing specialization Of the functions of the organization (Miller 8
Friesen, 1984b). According to Pugh et al., Specialization is “concerned with the

division of labor within the organization, the distribution of ofﬁcial duties among a

74

number Of positions” (1968: 72-73). Specialization permits the ﬁrm to
accomplish two objectives important to making a successful growth stage
transition: (1 ) to add specialized expertise to the organization to handle speciﬁc
problems related to important areas of the business, and (2) to assist in the
development of a functional structure designed to permit the separation of tasks
under the supervision Of an expert in the area.

I measured this variable by counting the number of different reporting
relationships that were represented by the executive ofﬁcer group plus any other
members of the top management team the ﬁrm identiﬁed. I elected to take a
broad approach to measuring specialization because I wanted to capture the
dual aspects Of the variable as described above. That is, specialization
represents both “Specialized expertise” and “separation of responsibility.” Titles
of senior managers ranged from those that clearly indicated technical expertise,
such as “Vice President Of Immunology" or “Chief Technology Ofﬁcer” to those
that weighed in more heavily on the segmentation Of responsibility, such as
“Senior Vice President Of Eastern Operations". or “Vice President for lntemational
Sales.” There were also members Of senior management who managed a
distinct business segment, with titles such as “President, Shoe Division.” By
including these variations of Specialization, ( i.e. functional, product, and
geographic), I was also capturing the three basic types of organizational
structure identified by Mintzberg (1979).

Generally speaking, the levels Of titles represented by top management

and included in the specialization variable were “Chief (functional area) Ofﬁcer,”

75

"Executive Vice President," “Senior Vice President,” and “Vice President.” If an
individual with the title of “Director’ was speciﬁcally identiﬁed as a member Of the
top management team, he or she was also included. The CEO and COO titles
were not considered evidence of specialization since these functions are very
broad and consist primarily Of oversight Of the top management team. Table 3.1
in the Appendices presents a sampling of areas of responsibility included in the
specialization variable from a group of 30 ﬁrms. This table Shows how many
instances Of the title were counted and the general industry categories in which it
appeared. .

Specialization was measured with data collected from the ﬁrm’s 1996 IPO
prospectus and 1998 10K (or 1999 proxy statement, as described for data
collection for the professionalization variable above).

Following the primary data collection process, I recollected data on 25
(10%) of the ﬁrms for which data were collected by the graduate students so I
could assess inter-collector reliability. I used. Cohen’s Kappa (Cohen, 1960) to
measure degree of association of the two sets Of measurements at the item
level, Since the measures were operationalized as binary (0 = absent, 1 =
present). Cohen’s Kappa is considered a more accurate and conservative
measure of association than Simple percent agreement in scores, since it uses
the actual distribution of scores from the two sources to correct for chance in the
level of agreement (Cohen, 1960). In addition, I assessed the level Of
agreement at the variable level using Kendall’s tau-c. Kendall’s tau-C assesses

degree of association by comparing relative rankings of pairs of cases obtained

76

from multiple sources (Kendall, 1962). Cohen’s Kappa requires a symmetrical
table of scores from both raters, which is easier tO Obtain when the variable is
binary. Kendall’s tau-c is a more suitable measure of association with variables
that are not binary (my variables are the total Of the values Of their measures)
since the tables of raters’ scores do not have to be symmetrical (Kendall, 1962).
The results of the inter-collector reliability analyses are reported in Table
3.2 under the Cross Collection columns. Unfortunately, for some of the
measures (particularly for the 97-98 time period), the measure of association
could not be calculated. This was due to one or both of the collectors having
only one value (0 or 1) for all the cases on that measure. When all the Scores Of
a rater on a measure have just one value, a measure of association cannot be
Obtained. In these situations, I calculated a simple percent agreement
(uncorrected for chance), which is denoted in the table by a superscript.

. Following data collection, I numerically coded the data- based on the
values established for the measures in my study. Inter-coder reliability was
evaluated by having another Ph.D. student whom I trained code infrastructure
data from 30 ﬁrms (approximately 10% Of the sample). His training consisted Of
a review of the list of measures along with the criteria for each. I used the
instructions developed for the data collectors, a list of variables from SPSS, and
samples already coded by me for the training session. Prior to coding the 30
ﬁrms, we independently coded data on 10 other ﬁrms and compared results to
identify any misunderstandings in the coding criteria. Inter-coder agreement was

measured by the same methods I used for inter-collector agreement. The

77

results of the inter-coder reliability tests are reported in Table 3.2 under the Inter-
Coder Reliability columns.

I interpreted the acceptability of the level Of agreement for the inter-
collector and inter-coder results based on the guidance Of Fleiss (1981), who
takes the position that the reliabilities generated by measures of association
such as Cohen’s kappa and Kendall’s tau are subject to the same standards as
any reliability statistic, such as alpha. Nunnally (1967) provides standards for the
determining what level of reliability is satisfactory,’depending on the maturity Of
the research area and measurement tools available. For measures Of a
construct that are in the “early stages” of research, (such as the measures I
devised for this dissertatiOn), Nunnally states “reliabilities Of .50 to .60 will sufﬁce”
(1967: 226). Therefore, I used the criteria Of .60 as a cut-Off for interpreting the
results of the inter-collector and inter-coder reliabilities I calculated for the
measures underlying the standardization and formalization construct. On this.
basis, only the policies and procedures measure and the strategic planning
measure have sufﬁciently high reliability to include in my analyses. The
existence Of an overall standardization and fOrmalization cOnstruct was not
supported. Instead, I used the two reliable measures as separate variables in
my analyses. AS with the standardization and formalization Construct, some of
the measures underlying the shared goals construct had low inter-collector
and/or inter-coder reliability. Only the existence Of a mission statement and
existence of an employee stock purchase plan were sufﬁciently reliable for

inclusion in my analyses. My primary analyses including the infrastructure

78

variables thus included the following variables: policies and procedures, strategic
planning, mission statement, stock purchase plan, management development,
professionalization, and specialization.

As an external validity Check on the degree to which my use of archival
data collection sources for infrastructure variables were indicative Of the actual
level Of infrastructure development of the ﬁrms in my sample, I also collected
‘ data on these variables for a sub-sample Of 20 randomly selected ﬁrms. For
each of these ﬁrms, I sent a brief survey via e-mail to the highest-ranking
member of the current top management team who had been with the ﬁrm since
its IPO in 1996. See Appendix D for a copy Of the cover letter and survey used
to collect this data. '

This survey was necessarily retrospective. I asked the respondents to
indicate whether or not the named areas Of infrastructure were present or absent
at two different points in time: in 1996 at the time of the ﬁrm’s IPO and again in
1998 at the end of the company’s ﬁscal year. The results of the comparison Of
the survey data with the data I collected from archival sources are shown in
Table 3.2 in the External Validity columns. Cohen’s Kappa was again used to
assess degree of association. Where association could not be calculated, the
uncorrected percent agreement is reported (and noted as such in the table).
There is very little agreement between the executive survey and my archival data
collection measurement of the infrastructure variables. The only measure with
good reliability is employee stock purchase plan. Therefore, I ran a

supplemental set Of analyses using only this variable, management

79

development, professionalization, and specialization to see if it made a
difference in my results. These results are reported in Chapter 4.

Firm Performance. A wide variety of ﬁrm performance measures have
been used in research regarding the board of directors and ﬁrm performance
(Johnson et al., 1996). Both accounting measures Of performance and market-
based measures Of performance have been widely employed (Dalton et al.,
1998). Given the debate as to the efﬁcacy of one set Of performance measures
over another (Bentson, 1982; Chakravarthy, 1986; Venkatraman 8 Ramanujam,
1986), it seems prudent to include multiple measures of ﬁrm performance in
efforts to adequately assess the construct, as recommended by Zahra and
Pearce (1989) in their review Of the board Of directors literature.

The measurement Of performance in ﬁrms that have recently become
public Companies and are rapidly growing presents some special issues.
Traditional accounting measures, while they have the advantage of being widely
used, may not accurately reﬂect ﬁrm value when ﬁrms are making substantial
investments in and outlays for their businesses, as is frequently the case with
growing ﬁrms (Homgren, Sundem, 8 Stratton, 1999). One of the criticisms of
accounting measures of ﬁrm performance is that the manner in which they are
calculated is largely under the control Of the ﬁrm’s management and may vary
from ﬁrm to ﬁrm (Bentson, 1982). In ﬁrms which are inexperienced at reporting
their ﬁnancial results to the public, this concern may be magniﬁed, as the
management groups Of their respective ﬁrms are in the process Of developing

and ﬁne-tuning their newly-fonnatted accounting systems (Flamholtz, 1990;

80

Miller 8 Friesen, 1984b). On the other hand, for a firm with a recent inﬂux Of new
owners, such as those in this sample, the value attached to the ﬁrm by its
shareholders is likely to be a signiﬁcant factor in assessing ﬁrm performance.
Increases in market returns are one manner in which an expanding newly public
ﬁrm can ﬁnance its growth through acquisitions (Hof, 1999; Lipman, 1994).

As a result of these competing concerns, I used four measures of ﬁrm
performance in this study. I planned to use two frequently used accounting
measures Since they are useful for comparability Of results across studies: return
on assets (ROA) and return on equity (ROE). In addition, I intended to use two
market-based measure of performance: the ﬁrm's price/eamings (P/E) ratio and
the ratio Of a ﬁrm’s market value to its replacement value (MV/RV). A ﬁrm’s P/E
ratio provides an indication Of investors’ beliefs in the ﬁrm’S growth opportunities
and its value relative to its risk (Brealey 8 Myers, 1996). In a study of corporate
governance in growth-oriented ﬁrms, Daily and Dalton (1992) found a signiﬁcant
relationship between board composition and the ﬁrms’ P/E ratios. Tobin’s q is
frequently used to represent a measure Of the overall value Of a ﬁrm (including its
intangible assets) from the perspective of its shareholders (Tobin, 1969;
Lindenberg 8 Ross, 1981 ). Calculation of Tobin’s q, however, is complex and
requires the availability Of data to which I did not have access. Variaya, Kerin,
and Weeks (1987) compare the value creation (or destruction) implications of
Tobin’s q compared to the ratio Of market value to book value and concluded
they were equivalent from a theoretical point Of view. They also conducted an

empirical assessment of the market value to book value ratio as a measure Of

81

firm value to shareholders and found it behaved very similarly to Tobin’s q.
Therefore, I used the ratio Of a ﬁrm’s market value (price of a share of the ﬁrm’S
stock at ﬁscal year-end multiplied by the number of Shares Of common stock
outstanding at year-end) to book value (shareholder’s equity) as my second
performance variable designed to measure shareholder value of the ﬁrm.

Out Of necessity, variations Of three Of these performance measures were
used in my analyses. For ROE, twenty-seven ﬁrms had negative values for
shareholders’ equity (their liabilities exceeded their assets). This led to an
uninterpretable value for ROE. (If net income were also negative, the ﬁrm had a
positive ROE, which was misleading). Therefore, I decomposed the ROE
variable into a dependent variable (net income) and a control variable
(shareholders’ equity). Negative values for shareholders’ equity led to similar
difﬁculties in interpreting MV/BV. Since shareholders’ equity is the denominator
in this measure, negative values for shareholders’ equity would lead to negative
values for the ratio. Logically, a ﬁrm with a positive market value and negative
shareholders’ equity would have a higher MV/BV in the eyes of shareholders
than one with positive shareholders’ equity and the same market value, yet the
MV/BV ratio would be lower, Since it is a negative number. Therefore, I
decomposed MV/BV into a dependent variable Of market value and a control
variable of shareholders’ equity. Third, a negative earnings per Share value led
to values for a ﬁrrn’s P/E ratio that were also misleading. For example, a ﬁrm with
an earnings per Share Of -$.20 and a stock price of $5.00 per share would have a

lower P/E ratio than a ﬁrm with the same stock price but with earnings per share

82

of +$.20. Yet the poorer performing ﬁrm (with regard to eps) would be
considered to have better stock performance in light of its earnings than the ﬁrm
with the same stock price and positive eps. The source I used for industry data
calculated P/E ratio as market value (stock price multiplied by number of
common shares outstanding) divided by net income. I used the same measures
.at the ﬁrm level for the ﬁrm P/E measure when l decomposed it. The dependent
variable thus became market value and net income was used as a control
variable. The performance measure Of ROA remained in its typical composite
form in my analyses. .

Raw data for 1999 ﬁrm performance measures were Obtained from the
firrn’S 10K whenever possible. ROA was calculated by dividing net income by
total assets. The number of common Shares outstanding at the end of the year
was collected for the calculation of MV. Performance data for ﬁrms for which no
10K was available were collected from Compustat. Compustat was also used to
collect data on a ﬁrm’s Share price at year-end to calculate MV. If the data for a
ﬁrm were not available through Compustat, (some ﬁrms in my sample were not
included in the Compustat database), an on-Iine source was used, which
Obtained its information from the Primark database. This occurred primarily in
the collection Of ﬁscal year-end stock price information.

For 12 firms in my sample, performance data were available only through

the third quarter of 1999. Since I did have data for three quarters Of 1999 ﬁscal

83

year, I substituted fourth quarter performance data for ﬁscal year 1998 for the
fourth quarter 1999 data that were not available.2
Control Variables

Industry. I considered general industry type important to include as a
control variable in this study Since a ﬁrm’s industry may affect the manner in
which it makes a high growth transition, i.e. differing organizational structures
and/or management practices may be appropriate given the nature of the
business Of the ﬁrm. Burns and Stalker (1961) and Lawrence and Lorsch (1967)
were early researchers favoring a contingency approach to organizational
structure, whereby the nature of a ﬁrm’s internal design relates to industry in
which it competes. The more recent ﬁndings Of Thesmar and Thoenig (2000)
support this assumption. They found that the organizational structure Of ﬁrms
employing highly skilled labor were more decentralized and ﬂatter than those
relying more heavily on unskilled labor. They also found that ﬁrms that were in
more innovative industries (experienced more product changes) were likely to
have a functional structure that was more ﬂexible than that of less turbulent
industries. Similarly, Rajan and Zingales (2001) found differences in
organizational structure among different types Of industries based on whether
they were physical capital-intensive or human-capital intensive. Based on a .
review of the relevant literature, I made three distinctions in this study with regard
to industry type: manufacturing, service, and knowledge-intensive. Manufacturing

ﬁrms customarily have fewer employees than service ﬁrms, while knowledge-

 

2 I included a dummy variable representing my use of 4th quarter 1998 data for these 12 ﬁrms in
the ﬁrm performance analyses to determine if this data substitution affected my results. Since it

84

intensive ﬁrms tend to have less formalized structures and fewer employees than
either service or manufacturing.

In general, manufacturing ﬁrms are those that produced a tangible
product for sale from raw inputs. Examples Of manufacturing ﬁrms included in
my sample are a company that manufactures home improvement specialty tools
and one that is a ﬂat-rolled steel mini-mill. Service ﬁrms are those that provide
an intangible service, including distribution and sales of products. Examples
include a seafood restaurant and a company thatprovides linen service for
hospitals. Knowledge-intensive ﬁrms are those that rely heavily on intellectual
capital to produce a good or service, such as pharmaceutical ﬁrms, software
ﬁrms, and telecommunications ﬁrms. Examples include a ﬁrm that provides
information technology outsourcing services and another that creates
engineering software products.

Unfortunately, current broad Standard Industrial Classiﬁcation codes do
not distinguish between these industry types in a clear-cut fashion. For instance,
SIC 3 includes both manufacturing ﬁrms and knowledge-intensive ﬁrms; SIC
3714 includes a ﬁrm that designs, produces,and sells wheels and wheel
accessories, (3 manufacturing industry), while SIC 3577 includes ﬁrms that
develop technology-based color printing systems, (a knowledge-intensive
industry). Since each ﬁrm's prospectus contained a detailed description of its
business and four-digit SIC codes were available for all ﬁrms, I used this
information to Classify ﬁrms into one of the three industry categories, using the

basic descriptions of each given above. I created two dummy variables for

 

did not, this variable was excluded from my ﬁnal analyses in the interest of parsimony and power.

85

industry type, one for service ﬁrms and one for knowledge-intensive ﬁrms, with
manufacturing being the omitted category.

As a means of checking on the reliability Of my industry codings into the
three categories Of manufacturing, service and knowledge-intensive, a colleague
also coded each ﬁrm into one of these categories, based On a very brief
description Of its business. The Kendall’s tau-c for the cross-coding Of industry
- categories was .73 (p < .01 ). This level of reliability supports my original coding
scheme (Fleiss, 1981; Nunnally, 1967). To be conservative, I ran all analyses
twice, ﬁrst using my original industry codings and then with the codings arrived at
by my colleague.3

Firm Size. The size of the ﬁrm at the time of its IPO isian important
control variable since the Size of the ﬁrm is likely to be associated with the nature
of the subsequent implementation Of high growth transition infrastructure. The
' number Of employees Of an organization is the most frequently-used measure of
ﬁrm size in the organizational structure literature (Kimberly, 1976;-Price 8
Mueller, 1986) and was the indicator of size used by Blau and Schoenherr
' (1971) in their landmark study Of organizational changes in state employment

' security agencies. I recorded the number of employees of the ﬁrm as reported in
its 1996 IPO prospectus. The second most-frequentIy-used measure Of ﬁrm size
is revenues (Price 8 Mueller, 1986). Therefore, I also collected data on annual

revenues for fiscal 1996 from the ﬁrm’s 1996 10K.

 

3 The results were unchanged using my colleague’s coding of ﬁrm industry. Therefore, the results
of the analyses reported in Chapter 4 reﬂect my original industry codings.

86

Size and Tenure of the Board. The number of directors has been used
as an indication Of the resources represented by the board in a number of
studies of board composition (e.g. Chaganti et al., 1985; Pfeffer, 1972; Zahra 8
Stanton, 1988; also see Dalton et al., 1998). In addition, the size of the board is
likely to be related to the number Of business expert and support specialist
directors represented. The total number Of directors on the board was Obtained
from both the ﬁrm’s 1996 IPO prospectus and its 1997 proxy statement. Based
on the rationale I gave in the description Of the business expert and support
specialist measures, I used the board size in early 1997 in the analyses. In
addition, the length Of time the board has worked together may also have an
impact on its inﬂuence in infrastructure changes. Therefore, median board
tenure was also included as a control variable.4

CEO Stock Ownership. The amount of power wielded by the CEO may
inﬂuence the degree to which the members of the board of directors can affect
ﬁrm outcomes, such as high growth transition changes or ﬁrm performance.
(Pfeffer, 1981). The CEO may either support or refuse to support actions that
directors recommend. Finkelstein (1992) found that the percentage Of the ﬁrm’s
stock owned by the CEO was a reliable and valid measure Of the amount Of
power Of the CEO. Therefore, I collected the percentage of the company’s stock
owned by the CEO following the IPO from the 1996 prospectus to represent

CEO power.

 

‘ Analyses including board variables were also run using mean board tenure, rather than median.
There was no difference in the results.

87

High Growth Transition Changes. These variables were described
earlier in the Dependent Variable Measures section. They are: policies and
procedures, strategic planning, mission statement, stock purchase plan,
professionalization, management development, and specialization. The status of
the ﬁrm’s high growth transition prior to its IPO is likely to be related to the
degree to which it implements additional high growth transition changes, as well
as to subsequent ﬁrm performance and growth rate Of the ﬁrm. In measuring
these variables as control variables, I followed the same data collection
procedures as I used for the measurement Of their dependent variable
counterparts, except that the data were collected from the ﬁrrn’s 1996 IPO
prospectus.

Average Industry Performance. For each of the ﬁrm performance
measures, (ROA, ROE, PIE, and MV/BV), I collected average industry
performance for 1999 from lbbotson Associates’ Cost Of Capital 2000 Yearbook.
Whenever possible, I collected industry average performance data at the 4-digit
SIC code level, using the SIC codes for each of the ﬁrms included in the study.
Due to the inherent volatility of the performance data-Of the ﬁrms in my sample, I
wanted to use the most speciﬁc industry average data available to scale ﬁrm
performance. The more closely related the industry data were to the ﬁrm data,
the more appropriate the scaling was likely to be. Therefore, I used 4-digit data
as available and substituted fewer digit data for the remainder Of the ﬁrms’
industries. Four-digit SIC industry average data was available for 77% of the

sample (227 ﬁrms), 3-digit data for 14% (40 ﬁrms), 2-digit data for 8% (24 ﬁrms),

88

and the industry average data for the remaining 5 ﬁrms was at the 1-digit level.
Because not all industry data were at the 4-digit level, I included an indicator
variable in my analyses representing the number Of digits in the SIC code used
for the industry average performance data to determine if this difference affected
my results.5

Methods of Analysis

I used hierarchical logistic regression to test those hypotheses with binary
dependent variables, i.e. when the high growth transition variables Of policies
and procedures, strategic planning, mission statement, stock purchase plan, or
management development was the dependent variable. For all other analyses, I
used ordinary least squares hierarchical regression. For all hierarchical
regressions, the control variables were entered in Step 1, the independent
variables in Step 2, and, where appropriate, interaction terms in Step 3. For
Hypothesis 4, the mediation hypothesis, I planned to follow the guidelines
provided by Baron and Kenny (1986).

For the analyses using hierarchical logistic regression, the key statistics I
report at each step are: the unstandardized regression coefﬁcient (b), the Wald
statistic (along with its level Of statistical signiﬁcance), the percent Of cases
correctly Classiﬁed (by the dependent variable), and X2 (or change in X2) and its
level Of statistical signiﬁcance. For the analyses using hierarchical OLS
regression, I reported: the standardized regression coefﬁcient (B), the t-statistic

and its level Of statistical signiﬁcance, R2 (or change in R2), and the F-statistic

 

5 There was no difference in the results when this dummy variable was included in the analyses.
Therefore, it was dropped from my ﬁnal analyses in the interest Of parsimony and power.

89

and its level Of statistical Signiﬁcance. For all analyses, I was looking for a
statistically signiﬁcant change in explanatory value (either X2 or R2) at the ﬁnal
step Of the regression to ﬁnd support for my hypotheses. In addition, for those
hypotheses with interaction terms that appeared to be supported, I plotted the
interaction to determine if the relationship between the variables were in the
predicted direction.
Variable Screening and Transformations

Dependent variables were screened for skewness and kurtosis to assess
the deviation Of their distributions from normality. When dependent variables'
distributions varied from normality, I used the guidelines for variable
transformation described in Tabachnik and Fidell (1996). I ran the analyses with
both the untransformed variable values and the transformed variable values. I
assessed the normality Of the distribution of the residuals of the regression
(using the Lilliefors test, a modiﬁcation Of the Kolmogorov-Smimov test) to
determine which variation of the variable to use in the ﬁnal analyses, i.e. the
transformed value was used if it improved the normality of the distribution Of the
residuals. Transformed dependent variables used in the analyses are noted in
Table 4.1, which reports descriptive statistics and correlations.

Following the recommendation Of Aiken and West (1991), variables that I
combined to test for interaction effects were centered (the variable’s mean was
subtracted from the value Of the variable for all cases) to reduce the likelihood Of

multicollinearity arising in the regression analyses. This transformation was

90

' made prior to inclusion of these variables as main effects in the analyses and

prior to combining them to create the interaction terms used in the analyses.

91

CHAPTER IV

RESULTS

Table 4.1 Shows the descriptive statistics and correlations between the
variables used in the analyses. The means and standard deviations for variables
that were transformed for the analyses are reported based on the variables’
values prior to transformation for easier interpretation.

Predictors of High Growth Transition Infrastructure Changes

Hypothesis 1 predicted that the representation of resources on the board
of directors (the number Of business experts and the number of support
specialists) would be positively related to the development Of infrastructure
subsequent to the ﬁrm’s IPO. Five hierarchical logistic regression analyses were
run to test this hypothesis, one for each Of the dichotomous infrastructure
dependent variables Of policies and procedures, strategic planning, mission
statement, stock purchase plan, and management development program. In
addition, two hierarchical OLS regression analyses were run, one for
specialization and the other for prior management experience. Table 4.2 reports
the results Of these analyses. These tests Showed nO evidence Of the predicted
relationships, thus Hypothesis 1 was not supported.6

Predictors Of Subsequent Firm Performance

 

6 Since both business experts and support specialists were included in all analyses testing
hypotheses related to the board of directors’ inﬂuence, I do not distinguish between the “a” and “b”
versions of my hypotheses in reporting my results. Where the results of analyses indicate a
difference in the effects of business experts and support Specialists, I make the distinction in my
description of the results.

92

Hypothesis 2 proposed that the degree to which infrastructure was
developed post-IPO would have a positive relationship with subsequent ﬁrm
performance. Four OLS hierarchical regression analyses were used to test for
the relationship Of the seven high growth infrastructure variables with ﬁrm ROA,
net income (ROE), market value (P/E) and MV/BV. Table 4.3 reports the results
of these analyses. The addition of the infrastructure variables tO the regression
' equation led to signiﬁcantly improved prediction Of ROA (AR2 = .07, AF = 2.21, p
< .05). The only infrastructure variable relationship that provided any support for
Hypothesis 2 was the positive relationship between the degree Of Specialization
and ROA (B = .23, t = 2.66, p < .01). Therefore, I have to conclude that
Hypothesis 2 was not supported. 1

My third hypothesis predicted a positive relationship between board
resources (business experts and support specialists) and ﬁrm perforrnance.
Four hierarchical OLS regression analyses tested the relationships between
board resources and each Of the ﬁrm performance measures. Table 4.4 reports
these results. Of the four performance variables only the regression predicting
net income (ROE) had a statistically signiﬁcant change in the F statistic in step 2
I Of the hierarchical regression (AR2 = .02, AF = 3.28, p < .05). One Of the two
predicted relationships was statistically Signiﬁcant; the number Of business
experts on the board was positively related to net income (ROE) (B = .15, t =
2.14, p < .05). Since the contribution Of the board resource variables to
predicting subsequent ﬁrm performance is minimal, Hypothesis 3 was not

supponed.

93

Mediation Model

Hypothesis 4 predicted that the relationships, if any, between board
resources and subsequent ﬁrm performance would be mediated by high growth
transition infrastructure development post-IPO. To have a basis for testing this
hypothesis, I needed to ﬁnd a statistically significant relationship between board
resources and ﬁrm performance as well as a statistically Signiﬁcant relationship
between infrastructure and the same measure(s) Of ﬁrm performance. This did
not occur with any of the ﬁrm performance variables (only ROA had a positive
relationship with high growth infrastructure variables and only net income [ROE]
had a positive relationship with board resources). Therefore, analyses testing
Hypothesis 4 could not be performed.
Interaction Effects

“Hypothesis 5 predicted that greater CEO qualiﬁcations would weaken the
positive relationship between board resources and high growth transition
infrastructure development. Five hierarchical logistic regression analyses were
run to test this hypothesis, one for each of the ﬁve dichotomous elements Of
infrastructure. Two hierarchical OLS regressions were run for the dependent
infrastructure variables Of specialization and professionalization. The results from
these seven analyses are reported in Table 4.5 The only support for Hypothesis
5 was found in the analyses predicting professionalization Of the executive Ofﬁcer
group. The Change in F for step 3 of this regression, (the step in which the
interaction terms were introduced), was statistically signiﬁcant (AR2 = .05, AF =

3.59, p < .01).

94

Three Of the four interaction terms for this regression were statistically
signiﬁcant. First, the number of business experts on the board interacted with
years of CEO general management experience in predicting professionalization
(B = .40, t = 3.42, p < .01). The nature Of this relationship, however, did not
support Hypothesis 5 when the interaction was plotted (see Figure 4.1). When
CEOS had greater general management experience, the relationship between
business experts and management experience of the executive Ofﬁcers was
positive, not negative as predicted. Support specialist representation on the
board also interactively signiﬁcantly with CEO general management experience
(B = .35, t = 2.61, p < .01). The plot Of this interaction yielded a pattern similar to
the interaction Of CEO general management experience with business experts.
Figure 4.2 illustrates this plot. When CEOS had greater prior general
management experience, the relationship between support specialists and
executive Ofﬁcer management experience was positive, in contradiction to the
prediction Of Hypothesis 5. Lastly, business expert resources interacted with
prior CEO industry experience (B = -.19, t = -2.40, p < .01). Figure 4.3 shows the
plot of this interaction. The nature Of the interaction between business expert
representation and CEO industry experience generally supports Hypothesis 5.
When CEOS have more industry experience, the relationship between business
experts and executive Ofﬁcer management experience is weaker. Based on
these results, I have to conclude that there is only weak support for Hypothesis

5.

95

Hypothesis 6 also predicted interaction effects, this time between high
growth transition infrastructure variables and rate Of ﬁrm growth in predicting ﬁrm
performance. I expected to ﬁnd that higher rates Of ﬁrm growth would strengthen
the positive relationship between infrastructure and performance. Unfortunately,
when these hierarchical regressions were run using the rate of growth variables
described earlier, Signiﬁcant multicollinearity problems arose. The variance
inﬂation factor values went into the hundreds, well over the accepted maximum
value of ten for multicollinearity (Neter, Wasserrnan, 8 Kutner, 1990). One
interaction term was even excluded from the analysis due to redundancy.

TO contend with this problem, I controlled for number of employees and
level of revenues for the year of the lPOand used the number of employees and
level of revenues in 1998 as indicators Of size. I proceeded with this revised
model, expecting that the size Of the ﬁrm in 1998, (controlling for its original size
in 1996), would Similarly strengthen the positive relationship between
infrastructure and performance. When I tested this model, new multicollinearity
problems arose, this time with the strategic planning variable. Variance inﬂation
factors again were in the hundreds. In addition, the interaction term between
strategic planning and 1998 revenues was excluded from the analyses due to
redundancy. (The correlation between the Changes in strategic planning variable
and the strategic planning/revenues interaction term was -.85 and the correlation
between this interaction and the interaction Of strategic planning and employees
was .78). Therefore, I dropped the strategic planning variables (control, main

effect, and interaction terms) from the analyses testing Hypothesis 6 altogether.

96

There were no statistically signiﬁcant results, however, leading me to conclude
there was no support for Hypothesis 6. These results are reported in Table 4.6.
Supplementary Analyses

For some Of the analyses, I assessed the effects of substituting another
variable measuring the same construct to determine if it produced different
results in the analyses. I will ﬁrst discuss the results Of variable substitution with
regard to control variables. Nine dummy variables representing one-digit SICS
were substituted for the service and knowledge-intensive industry dummy
variables and the analyses were rerun. None Of the results Of the analyses were
changed when the SIC variables were included rather than the two dummy
variables of service and knowledge-intensive industries nor was the ﬁt of the
models improved. The results Of the analyses were also unaffected by the use
Of transformed (logged) versions Of the number Of employees and revenues, two
variables that were quite skewed in their distributions.

Based on the executive survey results of a sub-sample Of 20 ﬁrms on the
dichotomous infrastructure variables, only the variable of stock purchase plan
had acceptable reliability using Cohen’s Kappa (.73, p < .01). Therefore, I reran
the regression analyses with infrastructure variables as independent variables
including a reduced variable set, including only stock purchase plan,
specialization, management development, and executive Ofﬁcer management
experience. This included the analyses testing Hypotheses 2 and 6. There was
no Change in the overall ﬁt Of the models, but there was a Slight Change in the

results for two Of the analyses with regard to Hypothesis 2. Using the reduced

97

' set of infrastructure variables, both stock purchase plan and specialization were
Signiﬁcantly and positively related to subsequent ﬁrm ROA (rather than only
Specialization). The results of the revised analyses in testing Hypothesis 6 were
no stronger than when the full set Of infrastructure variables were included.

TO assess the explanatory value of having separate board resource
variables for business experts and support specialists, I re-ran the analyses
including board variables with a composite board resource variable, created by
adding together the number Of business experts and the number Of support
specialists on the board. There was no change in my results.

I also replaced 1997 board of director independent and control variables
with their counterparts from the 1996 board of director composition. There was
no substantial change in results. Also, due to the reduced number of ﬁrms for
which Specialization data were available for the full top management team, I
substituted 3 version of the specialization variable representing the executive
Officer group only. This variation produced weaker results than the original
specialization variable.

I tested the effects Of using a decomposed version Of ROA, with net
income as the dependent variable, controlling for total assets. Since I used
decomposed versions Of the other performance measures (out of necessity, ‘
given the nature of the data for this sample Of ﬁrms), I thought it was prudent to
examine ROA in a similar fashion. The ﬁt Of the models using this form Of ROA

was either the same or worse than using the composite form of the variable.

98

The last set of robustness analyses I ran were designed to examine the
appropriateness Of my research model and to determine if any Of the effects I
found were due to a misspeciﬁcation Of the relationships between the variables.
TO do this, I ﬁrst regressed business expert and support specialist representation
on the CEO industry and general management experience variables, controlling
for industry, ﬁrm size, CEO stock ownership, board tenure, and board size. I
found no relationship between the CEO independent variables and business
experts or support specialists. I then reversed the direction Of the analysis and
regressed the CEO variables on the board composition variables. In this case,
both business experts and support specialists were signiﬁcantly (and positively)
related to CEO industry experience (but not related to general management
experience). This raised the possibility that business experts and support
specialists on the board Of directors inﬂuenced the background experience of the
person in the CEO position, which, in turn, affected the Changes in infrastructure.
However, in testing Hypothesis 1, neither business experts nor support
specialists were related to any of the infrastructure change variables, eliminating
the mediation model alternative. This same (lack Of) ﬁnding also eliminated the
possibility that a spurious relationship between board resources and CEO
industry experience inﬂuenced the results of the analyses.

Lastly, I performed the same tests using the high growth transition
infrastructure variables as the predictors of subsequent ﬁrm size (employees and

revenues) and vice versa. The control variables were industry, CEO stock

99

ownership, employees and revenues in 1996, and the 1996 infrastructure
variables. None of these regressions produced statistically Signiﬁcant results.
Post Regression Tests

Following the running of the regressions testing my hypotheses, I looked
at the results of two tests to determine if multicollinearity might have inﬂuenced
my results. The statistical tests | used were the values Of the variance inﬂation
‘ factors, where a value in excess Of 10 indicates a problem (Neter et al., 1990),
and values Of the conditioning index where a value in excess Of 30 indicates a
problem (Belsley, Kuh, 8 Welsch, 1980; Tabachnick 8 Fidell, 1996). The only
evidence Of multicollinearity problems I found was noted earlier, when I ran the
regressions to test Hypothesis 6 using rate of ﬁrm growth as the moderating
variable. Substituting ﬁrm size for growth and deleting the strategic planning
variables minimized the multicollinearity in this set Of analyses. (There were still
a several high VIFS for the employee and revenue variables, but none higher
than 17).

To test for outlier cases for the regressions, I used casewise diagnostics
in SPSS to identify any dependent variable values in excess Of three standard
' deviations from the mean predicted value. I subsequently adjusted my screening
for dependent variable outliers to those with values in excess of four standard
deviations from the mean due to the pattern of values for many of my variables
(i.e. tOO many cases were Classiﬁed as outliers using the 3 standard deviation
criterion). l re-ran regressions where there were outliers identiﬁed to see if the ﬁt

Of the model were improved with these values deleted. By deleting these values,

100

the adjusted R2 for the model was improved or the distribution of the residuals
was improved or both were improved. Table 4.7 shows the cases for which
dependent variable values were deleted by hypothesis.

I assessed the degree Of ﬁt of the models used in the OLS regression
equations through several methods. First, I used the Lilliefors test (a
modiﬁcation Of the Kolmogorov-Smirnov test) to test for the normality Of the
distribution Of the studentized residuals from the regressions. Because this test
is quite sensitive when sample sizes get larger, (tending to readily ﬁnd a lack Of
normality as sample Size increases), I followed Tabachnik and Fidell’s (1996)
recommendation to also examine the residuals visually. Therefore, I also looked
at a frequency histogram Of the studentized residuals, a scatterplot Of the
standardized predicted values against the studentized residuals, and a normal P-
P plot Of the residuals with Observed cumulative probabilities of values plotted
against expected cumulative values.

For the OLS regressions, a Lilliefors statistic of .10 and below
corresponded to visual representations of a distribution Of residuals that
appeared to be normal. Where the Lilliefors statistic was higher than .10, I
tested the ﬁt of the model for curvilinearity in the relationship between the
independent and dependent variables to determine if this were the underlying
cause Of the lack Of normality. I did this by two methods: using the curve
estimator test for linear, quadratic, and exponential relationships available in
SPSS and by adding squared versions Of the independent variables to the

regressions to see if the ﬁt improved. Four out ofsixteen Of the OLS regression

101

equations produced Lilliefors statistics in excess of .10; none Of them had
improved ﬁt with nonlinear’models, however. In all cases, the models with poor
ﬁt were those with accounting measures Of performance as the dependent
variable (either ROA or net income [ROE]).

In the next chapter, I discuss the factors that may have contributed to my
lack of ﬁndings, how I might contend with these issues in future studies on this

topic, as well as directions for additional research in this domain.

102

CHAPTER V
DISCUSSION AND FUTURE RESEARCH

The two fundamental research questions I set out to answer in this study
were: (1) what is the interrelationship between board Of director resources and
the CEO in predicting the development of high growth transition infrastructure
and (2) what is the interrelationship between development of high growth
transition infrastructure and rate Of ﬁrm growth in predicting subsequent ﬁrm
ﬁnancial performance. My analyses produced little or no support for the
relationships predicted in my hypotheses. My results are reported in summary
form in Table 5.1. They are disappointing since they provide very little support
for my hypotheses.

In this chapter, I will ﬁrst describe the key ﬁndings from my analyses
based on my predictions. Next, I will discuss the limitations Of my study and
some potential reasons for my failure to ﬁnd support for my hypotheses. Third, I
will present some ideas for future research in this general domain and lastly, I
will provide a brief Conclusion.

Key (Non)Findings

The main effects hypotheses, (Hypotheses 1 — 3), served as a prelude to
the mediation and moderation hypotheses to follow. If my later hypotheses,
(Hypotheses 4 — 6), were descriptive Of the relationships between the variables,
then there should have been little in the way of statistically Signiﬁcant
relationships in the earlier main effects hypotheses. Table 5.1 summarizes my

ﬁndings by hypothesis for each Of the dependent variables. Out Of the 50

103

relationships investigated as main effects across three sets Of regression
analyses, only 3 (6%) were statistically signiﬁcant, a result that could easily have
occurred by chance. Further, in only one case was the second step of the
regression statistically Signiﬁcant. For Hypothesis 1, representation Of business
experts on the board Of directors was positively related to the infrastructure
variable Of employee stock purchase plan. For Hypothesis 2, specialization Of
the top management team was positively related to subsequent ROA (and the
second step of the regression was also statistically Signiﬁcant). For Hypothesis
3, representation of business experts was positively related to net income (ROE).
I had hoped to ﬁnd more support for my hypotheses when it came to the
second set of hypotheses. One implication Of the lack Of ﬁndings from testing
the ﬁrst three hypotheses was that there were no relationships I could test for the
mediation model I proposed in Hypothesis 4. Hypothesis 5 proposed that the
relationship between the resources on the board Of directors and the
development of high growth transition infrastructure post-IPO would be
moderated by the managerial qualiﬁcations of the CEO. I expected to ﬁnd that
when the CEO was more qualiﬁed, the relationship between business experts
and support specialists on the board and infrastructure development would be
weakened. My rationale was that the CEO would have less need of input from
the board when she had more experience coming into the job. Out of 76
predicted relationships, only 6 (8%) were statistically signiﬁcant. Of these
relationships, half occurred in the testing Of Hypothesis 5, where the third step Of

the hierarchical regreSsion Of market value (MV/BV) on board resources/CEO

104

qualiﬁcations interactions was signiﬁcant. Unfortunately, the speciﬁc nature of
the relationship between the variables in two Of these three cases was not as I
had envisioned. When CEOS had more general management experience prior
to joining the ﬁrm, the inﬂuence Of business experts and support Specialists on
professionalization was enhanced, not reduced. Only in the interaction of CEO
industry experience with business experts was the interaction generally in the
predicted direction, with CEOS who had more prior industry experience
apparently substituting their inﬂuence for that of the board.

More speciﬁcally, the plot Of this particular interaction shows that
professionalization is at the same high level when there are fewer business
experts on the board and the CEO has more industry experience O_r when there
are more business experts on the board and the CEO has less industry
experience. These relationships support my prediction of substitutability of board
expertise and CEO qualiﬁcations. However, professionalization Of the executive
Ofﬁcers is at its low—est when there are both more business experts on the board
and the CEO has more industry experience. I would have expected that it would
be the case that professionalization wOuld be lowest when there were fewer
business experts on the board and the CEO had less industry experience.

It is plausible to interpret this unanticipated relationship in a couple of .
ways. It may be that when CEOS with more years Of industry experience have
higher levels Of general management business expertise on their boards, they
see this as a resource that reduces the need for managers with previous

experience. In other words, business experts on the board and executive Ofﬁcer

105

experience are considered substitutable resources by industry-experienced
CEOS. Another possible interpretation is that the combination of more business
experts on the board and CEOS with more industry experience leads to an
impasse in decision making regarding professionalization of the executive ofﬁcer
group. Perhaps industry-experienced CEOS seek to bring in cronies from
companies where the CEO previously worked, while the business experts on the
board resist this effort.

Although these ﬁndings are very limited, they do suggest that some CEO
characteristics may supplement board resources to inﬂuence ﬁrm outcomes,
while others may have the opposite effect of suppressing the impact of the board
Of directors. Perhaps CEOS with more general management experience are
compatibly aligned with board members with business backgrounds and
capitalize on this mutuality of interest to support the ﬁrm’s internally directed
strategies. CEOS with more general management experience would likely also
have had more experience in working with boards and understand how to utilize
directors’ business backgrounds more effectively in accomplishing strategic ﬁrm
goals. 2 2

It seems likely that the nature Of the outcome makes a differenCe. One of
the motivations for this study was the fact that past studies have found little in
the way Of evidence that board composition inﬂuences ﬁrm performance. I also
failed to ﬁnd this relationship, while I found some slight encouragement for the
hypothesis that board resources, in interaction with CEO qualiﬁcations, have an

inﬂuence on a less distal outcome, such as the development of ﬁrm

106

infrastructure. The interaction of the board and CEO may also be a critical
component Of a model assessing board inﬂuences. While I found no statistically
signiﬁcant main effects of board resources on infrastructure, I did ﬁnd some
statistically Signiﬁcant interaction effects in predicting professionalization of the
executive Ofﬁcer group. Post hoc analyses including both bOard resources and
CEOqualiﬁcationS as main effects found no signiﬁcant relationships.

- Limitations Of Study

In the following sections of the Discussion, I will review what I consider, in
hindsight, to be the most problematic aspects Of my study. I will also present
alternative approaches that may address these issues in future research.

Data collection procedures. Using archival data collection procedures
for this study had some advantages and disadvantages. On the positive Side, it
allowed me to collect data longitudinally without having to wait a period of time to
‘ Obtain it. I also had access to comparable data from all ﬁrms in the sample, i.e.
SEC ﬁlings, and did not have to be concerned about a poor response rate to a
survey or biases that may result from respondents from within companies.

I Although same source bias may have been a problem, in that SEC ﬁlings were

I my primary source of data, I did not have to contend with issues of memory loss
over time, as I would have if I were attempting to collect historical data from
human sources through surveys or interviews. I also had the ability to key word
search the ﬁlings since they were submitted in Word document format to the
SEC starting in 1996, which gave me a consistent, thorough method for data

collection.

107

There were, Of course, signiﬁcant downsides to archival data collection
processes for this study. There were aspects of infrastructure development that
I could not measure by examining SEC ﬁlings that may be important predictors of
subsequent ﬁrm performance. Examples include delegation of authority,
reporting relationships within the organization, and corporate values. For some
of the data I was collecting, I was at the mercy Of what companies chose to
include in their ﬁlings. Some Of the data I was collecting fell into the category Of
required reporting, such as backgrounds Of the board of directors and executive
Ofﬁcers, CEO stock ownership post-IPO, and the existence Of stock Option and
stock purchase plans. Other data I collected seemed to be regularly reported,
although not speciﬁcally required, such as the existence of a strategic plan and
management compensation programs. Other data was strictly voluntary, such
as CEO educational and functional background, communication and training
programs, and employee compensation programs. This led to missing data
problems once I got to the analysis stage. Two CEO background variables had
to be eliminated due to an insufﬁcient number of cases with complete data to
maintain the power needed for my analyses.

TO the degree that ﬁrms varied in their reporting of this information, my
collection method yielded unreliable data. I attempted to minimize this inherent
problem by coding many of the infrastructure variables as either present or
absent, rather than attempting to assign a more reﬁned value. The reporting of
biographical information Of Ofﬁcers (used for the board Of director coding, the

CEO qualiﬁcations Of general management and industry experience, and the

108

specialization and professionalization variables) was typically Speciﬁc, leading to
effective data collection.

A complicating factor in the nature of information reporting in SEC ﬁlings
that I discovered once I was into the data collection process was that the
prospectus documents contained much more detailed information Of the kind I
was interested in collecting than did subsequent 10KS. Since the prospectus
serves as a publicity piece for the ﬁrm to attract prospective shareholders,
companies are more forthcoming with detailed information that they believe .
potential investors will ﬁnd informative in making their investment decisions.
Once the ﬁrm is public, it has less vested interest in establishing credibility with ‘
non-ﬁnancial information disclosure, since its stock price and ﬁnancial reports
provide the type of information investors want to know and a track record Of
some sorts has been established. TO the degree that some ﬁrms are not
performing as well as others in their industries, however, they may add more
infrastructure details to the 10K than the better performers in an effort to Offset
their results. If this does in fact occur, then it would make my data collection
methodology more conservative, since I would be less likely to ﬁnd a relationship
between infrastructure and subsequent performance.

A related reliability issue with my data collection process was the
disappointing inter-rater agreement level between the data the MBA students
collected, which was the majority Of my infrastructure data, and my own data
collection. I trained the students in advance and Checked their work early on,

making corrections in their methods where necessary. It appears, however, that

109

I Should have been more speciﬁc in my instructions. Alternatively, it could be
that my coding standards were not sufﬁciently detailed, since it was the coding Of
the data I compared. Although I did all the coding myself, it is possible that this
methodology was unreliable, rather than the data collection itself. If this were the
case, it is correctable by establishing a more rigorous set Of coding guidelines
and re-coding the data. (The low level Of agreement between the secondary
coder of a sub-sample Of ﬁrms and me indicates the coding guidelines | used
were a problem, absent another explanation for our lack of agreement).

Additionally, the results I received from a survey of a sub-sample Of 20
ﬁrms designed to Check on the external corroboration of my archival data
collection also suggested that my data collection was unreliable. Of the 11
dichotomous infrastructure variables I included on my survey, only the employee
stock purchase plan variable was supported with a reasonably high level Of
agreement between company respondents and my variable coding. The data I
collected from company respondents did suffer from reliance on memory since
the status of infrastructure development at the time of the IPO was
approximately six years earlier. As mentioned in the Results chapter, I re-ran my
analyses using only those variables I believed to be most reliable, but my
ﬁndings were not substantially different.

In retrospect, I believe that using real-time surveys completed by company
executives to gather data on infrastructure variables is much more likely to lead
to reliable data collection and variable measurement than the archival data

collection method I used, at least for the non-biographical variables.

110

Infrastructure variables. I could not use the individual measures of
infrastructure to form composite infrastructure variables as I had envisioned
when I designed this study. The reliabilities of the composite
standardization/formalization and shared goals scales were very low (for the
standardization scale, a = .20, for the shared goals scale, oi = .06). Therefore, I
resorted to using individual infrastructure measures as my infrastructure
' variables. TO the degree that I did ﬁnd any signiﬁcant results from the
infrastructure variable analyses, they were with the continuous variables Of
specialization and professionalization, rather than the dichotomous measures.
The present/absent coding of the dichotomous measures of infrastructure
created a ceiling effect since the maximum value for each was “1 .” Because I
was measuring infrastructure at two points in time, in 1996 and again for the
1997-1998 time period, once the infrastructure element had been observed, no
further development Of that aspect Of infrastructure was counted. This could
have created a problem in measuring the policies and procedures and strategic
planning variables, Since their average values in 1996 were .66 and .83,

I respectively, leaving little room for subsequent development in the 1997-1998
pedod.

In general, I found that the ﬁrms in my sample were further along in their
development Of infrastructure at the time Of their IPOS than I had expected. In
addition to the 1996 values of the policies and procedures and strategic planning
variables just mentioned, the average level of specialization in 1996 was 5.69

versus 6.05 in 1998, (an increase of only 6%) and the average level of

111

professionalization in 1996 was 6.89 versus 7.15 in 1998, (an increase of only
4%). Since the number of employees and the revenues Of the ﬁrms increased
substantially from 1996 to 1998 (over 200% for employees and over 100% for
revenues), it appears that ﬁrms preparing for their initial public Offerings Of stock
develop their infrastructure with an eye to future growth. There are numerous
handbooks available to guide executives through the preparation for an IPO
' (such as the one written by Blowers, Grifﬁth, 8 Milan, 1999 for Ernst and Young),
not to mention the availability Of advice from venture capital investors and
investment banks, two sources Of information with extensive experience with the
IPO process. In light Of this situation, I would restructure the design of my study
to use the combined value of the infrastructure variables (i.e. as of the end of
ﬁscal 1998) in my model, rather than controlling for the level of these variables in
1996. Logically, it is the relationship Of the total development Of infrastructure
with board resources, CEO qualiﬁcations, ﬁrm growth, and subsequent
performance that is likely to be important, not merely infrastructure development
subsequent to the ﬁrm’s IPO.

In addition, all the infrastructure measures used in the study were
4 weighted equally, which may or may not be representative of their actual
inﬂuence within the ﬁrm. From my review Of previous research in this area,
however, this is the method commonly used by researchers of organizational
structure (Blau 8 Schoenherr, 1971; Hanks et al., 1993; Smith 8 Gannon, 1987).

A possible approach to this issue would be to gather data from company

112

executives who have experienced an IPO and the subsequent growth Of their
ﬁrms regarding the relative importance of the various aspects Of infrastructure.

Sample. My choice of the population Of ﬁrms to study was designed to
address some Of the previous concerns of governance research on the effects of
board composition on ﬁrm outcomes. There were, however, issues that arose
during my study that caused difﬁculties because Of my Choice Of context.
Sampling across industry inevitably led to differences for which I could not
control. From my analyses, it appeared that the use Of the three industry
categories I used was as effective as SIC dummy variables in partialling out the
effects of industry. There iS little guidance in the literature for segregating ﬁrms
on the basis Of industry when board of director and infrastructure variables are
included in the study. I suspect that some other categorization of ﬁrms than the
two I uSed would be more effective in controlling for industry effects. Perhaps
the new NAIS industry codes will be more applicable to this type Of study than
the out-dated SICS. I might find more Signiﬁcant relationships by following a set
Of ﬁrms in the same general industry group over time, rather than attempting to
ﬁnd patterns across industries. The advantage Of collecting data from ﬁrms
across industries was the ability to generate a data set with a sufﬁcient number
of ﬁrms, while controlling for general external environment issues, such as the
economy, since all the IPOS occurred in the same year.

Another difﬁculty with the research design was the decision to use
subsequent ﬁrm performance as the dependent variable. This meant that any

ﬁrms no longer in existence as independent entities as Of 1999 were excluded

113

from my sample, creating a survivor bias. This approach restricted the sample of
ﬁrms due to the elimination of those with lower levels Of performance if I assume
that the ﬁrms that did not survive were poorer performers. By doing some follow
up research, however, it appeared that many, if not a majority, Of the ﬁrms that
did not “survive” until 1999 were acquired by or merged with other ﬁrms. With all
the companies going public in the late 90s (with money to spend based on funds
raised from new shareholders) and the high price/eamings ratio of many
companies’ stocks during this period, there was ample cash and/or market,
capitalization available in the ﬁnancial markets for ﬁrms to buy other ﬁrms. The
mergers and acquisitions that occurred were not necessarily due to the Operation
Of the market for corporate control because Of poor management, but simply a
desire on the part Of many companies to expand within a given industry and
create greater market power. Nevertheless, the generalizability Of my results is
limited to those ﬁrms surviving as independent entities three years out from their
IPOS.

Inclusion of the ﬁrms no longer in existence in 1999 might have added
richness and depth to my sample and my results, aswell as increased the
generalizability Of my ﬁndings. Finding a way to meaningfully measure the
dependent variable of performance was the Obvious problem. Perhaps some
method that takes into consideration the non-surviving ﬁrms’ performance prior
to acquisition or the price paid for the ﬁrm (as a multiple Of its stock price or
earnings) compared to the industry as a whole could be fashioned. A control

variable for the period covered (IPO date to date performance data were

114

collected) for all ﬁrms could be added to the regression model, permitting ﬁrms
that did not survive independently until 1999 to be included. There is also the
situation Of the ﬁrms that were in bankruptcy before 1999 to consider. Firms in
bankruptcy are exempt from ﬁling 10Ks, (unless they voluntarily choose to do
SO), which means performance data is only available up to the time Of bankruptcy
declaration. The best alternative here would seem to be the same as above,
using the latest performance data available and including the time period Since
IPO as a control variable.

The ﬁnal signiﬁcant issue I would like to raise with regard to my population
choice was the volatility Of the ﬁrm performance. The fact that almost 10% of the
ﬁrms had negative values for shareholders’ equity is an example of the high
degree Of variation in performance among these ﬁrms compared to a more
mature group of companies. This created measurement problems for three out
Of four of my dependent performance variables. Negative shareholders equity
combined with negative earnings (a likely combination) led to a positive ROE.
This is clearly not indicative Of the ﬁrm’s performance that ﬁts this description.
Negative shareholders' equity (book value) also led to uninterpretable MV/BVS,
since a negative book value would lead to a lower MV/BV performance when this
ﬁrm was actually performing better, (from a MV/BV standpoint), than one with a
positive BV. A comparable problem arose with negative earnings for calculation
of the P/E ratio. One hundred seventy-four ﬁrms in my sample (59% percent)
had negative earnings in 1999. If a company had negative earnings and was still

trading, then its P/E ratio would be negative. In a sense, however, it had a

115

higher P/E than a ﬁrm with a comparable stock price and positive earnings, since
it was performing as well in the market on ﬁnancial performance that was
weaker. My approach to this problem was to decompose the variables, using the
numerator Of the ratio as the dependent variable and controlling for the
denominator Of the ratio to scale the dependent variable.

Although this method Of measuring performance was unconventional and
may lead to less accumulatiOn of results of the research in this area, it had the
advantage Of breaking ﬁrm performance into its components, rather than using a
ratio dependent variable, which may lead to results that are less interpretable
(Cohen 8 Cohen, 1983). Since most of the ﬁrms in my sample were relatively
young, (their average age was just over 8 years), they were likely still struggling
to establish themselves three years after their IPOS. This could mean that they
were focusing on the most basic elements Of performance, such as revenue
generation, stock price and net income, rather than the ratios used as measuring
sticks by more mature companies.

Not being able to use the ratio form of three Of my performance variables
did produce a potential problem, however, in scaling for industry average
performance. I did not have data at the industry level that could be broken down
into the components of the performance variables, such as shareholder’s equity,
net income, and market value. Therefore, I could not directly scale each ﬁrm’s
performance tO that of its industry. I included the industry average performance
in ratio form as a control variable instead. Since ROA was the only performance

measure for which I had data at both the ﬁrm and industry level, I scaled ﬁrm

116

ROA directly, by dividing it by industry average ROA. l re-ran the analyses for
Hypotheses 2, 3, and 6, which are those containing performance dependent
variables. In all cases the R2 dropped Signiﬁcantly and the model ﬁt deteriorated.
The values Of my scaled ROA variable were distributed in a relatively normal
fashion, with no apparent outliers. Although this is a very limited test, my
ﬁndings do call into question the wisdom of dividing ﬁrm performance by industry
performance to scale it, a common practice among strategy researchers.

Model speciﬁcation and theory. Although a substantial proportion of the
total variance explained by my model was present after step one of the
hierarchical regression, i.e. was contributed by the control variables, surprisingly
few of the control variables were statistically signiﬁcant. In addition, the overall
variance explained (adjusted R2) averaged .28 for the moderation analyses
predicting specialization and professionalization infrastructure and .35 for the
moderation analyses predicting subsequent ﬁrm performance.7 This suggests
there is room for improvement with regard to the speciﬁcation Of my research
model.

Within the limitations of my data collection method, there are a number of
changes I would make to the model based on hindsight. First, as I mentioned
earlier, I believe the infrastructure variables(pre-IPO from 1996 and post-IPO
from 1997-1998) would be more explanatory if combined into one variable. In
retrospect, there is little theoretical reason to treat them separately. The ﬁrms in

my sample had implemented a signiﬁcant amount Of infrastructure prior to ﬁling

 

7 The average total R2 for these hypotheses is considerably higher, .37 and .50, respectively,
indicating the impact Of error on the explanatory value of the models.

117

their prospectuses, in preparation for going public, just as Penrose (1959)
advocated.

A difﬁculty with this approach, however, is determining the appropriate
measure of board Of director resources to use in predicting cumulative
infrastructure. The composition Of the board at the time of the prospectus would
likely be more related to total infrastructure development than the 1997 board,
which I used in my study. The tenure of the business experts and support
specialists on the board is an important factor to consider in determining their
inﬂuence on infrastructure development, since some infrastructure was likely
implemented prior to the directors’ appointments to the board. One possible
solution would be to weight the director categories by their average tenures.
Thus, a board with 3 business experts with an average tenure Of 2 years would
produce a weighted inﬂuence variable with a value of 6, whereas a board with 2
business experts with an average tenure of 5 years would have an inﬂuence
value of 10. Assuming that infrastructure has developed over a period Of time,
using board resource variables weighted by tenure seems appropriate. (A
similar approach to calculating inﬂuence was used by Walsh, Henderson, and
Deighton, 1988). '

Access to executives with IPO experience could assiSt with an
assessment Of the timing of board inﬂuence on infrastructure (or other interim
outcomes) Since they would know when directors began their association with
the ﬁrm. For example, a substantial number of the ﬁrms in my sample

(approximately 25%) appeared to have formed their board Of directors very close

118

to the time of the IPO, yet the directors themselves may have been advising the
CEO prior to their formal appointments as directors. I have been unable to ﬁnd
any studies supplying evidence that this is the case, so interviews with CEOS
may provide some insight into this issue.

The business expert and support specialist categories are more Closely
tied .to the resource dependence view than the more general categories of
' insider and outsider Often used in governance studies based on agency theory.
For instance, Fiegener, Brown, Dreux, and Dennis (2000), in their study of
privately-owned ﬁrms, found that adoption of outside boards was more
connected to the degree Of outside ownership of the ﬁrm than to the resources
such directors could provide. Their study suggests that the measurement of
these categories may require additional reﬁnement to be more predictive of
changes internal to a ﬁrm about to go public. (The lack of difference in the
results I Obtained with a combined board resource variable compared to
separate variables for business experts and support specialists suggests that
further reﬁnement may be informative). With regard to business experts, for
instance, the degree to which their executive background is in the same or a
related industry (in value Chain terms) may be a key distinction in their levels Of
inﬂuence. Additionally, the support specialist category is quite broad in its
deﬁnition, including those directors with backgrounds in virtually every area Of
running a business except general management. There may be an increase in
explanatory power if this variable is broken into more delineated categories, such

as ﬁnancial, operations, professional, technical, etc. For many of the ﬁrms in my

119

sample, their products or services were based on the development of technology
or on medical knowledge. These boards Often had engineers, chemists, or
physicians in director positions. These types Of directors would be likely to have
a different effect on infrastructure development than those directors with more
general functional business backgrounds, such as marketing or ﬁnance. The
ﬁndings of Golden and Zajac (2001), whose study evaluated the impact Of
boards on strategic Change in hospitals suggest that occupational heterogeneity
on the board is an important consideration in assessing its inﬂuence. Therefore,
a more reﬁned approach to board composition measurement, rather than the
typical broad-based categorization, may be more revealing.

The measurement of the CEO variables could also be expanded beyond
years Of prior general management and industry experience. Although my
efforts to collect data on educational level and prior functional experience were
not successful, there are other elements Of CEO background or status that may
affect the level of inﬂuence of board members. Examples are stock ownership
(as an independent variable, rather than a control variable), tenure with the
company, whether or not the CEO is also Chairman of the Board, and whether or
not the CEO is the founder Of the ﬁrm. Although there is limited evidence from
prior research that the latter two variables are inﬂuential in predicting CEO
effectiveness, they may be more meaningful when combined with additional
CEO attributes and when included in interaction with board composition

variables.

120

Other dependent variables may be relevant also. Besides the more
traditional ﬁnancial and market measures of performance I used in my study,
there are other performance measures that may be appropriate given the context
I chose. One is the degree to which a ﬁrm’s stock is underpriced for its IPO. If
infrastructure development serves as a signal to potential investors of the
readiness Of the ﬁrm for growth, those ﬁrms with appropriate infrastructure in
place prior to their IPOS may experience reduced underpricing in the market. A
recent study Of the signaling effect Of board membership on underpricing
suggests this may be a fruitful investigation (Certo, Daily, 8 Dalton, 2001).

Another possible dependent variable is ﬁrm growth. Increase in revenues
is Often considered an appropriate measure Of success in the literature studying
young, growth-oriented ﬁrms (Hanks et al., 1993; Smith 8 Gannon, 1987). Size V
at the time of the IPO may be an important predictor of infrastructure
development (the number of employees seems to be Slightly more important in
this regard than revenues, based on my results), but the level of future revenues
the ﬁrm generates may also be higher when supported by the presence of high
growth infrastructure. .

Additionally, an important control variable in this model might be the
amount Of money raised in the ﬁrm’s IPO. A portion Of the money raised by
Offering equity to the public is Often used to pay Off debt; this information is
usually reported in the ﬁrm’s 10K following its IPO. I could segregate the IPO
funds remaining following debt payment and use it as a control variable in

predicting both infrastructure development and growth. In addition, infrastructure

121

development may be differentially related to growth depending on whether that
growth is organic (generated by the ﬁrm itself) or based on acquisitions Of
existing ﬁrms. Less infrastructure may be required to increase revenues through
acquisition, since acquired ﬁrms possess their own infrastructure that can be
utilized by the acquirer to support its larger size.

Directions for Future Research

Present study research questions. In spite of the discouraging results
from this study, I believe the general research questions I asked are worth
exploring using improved methodology and a somewhat different conﬁguration Of
the variables. Recent studies by Golden and Zajac (2001) and Westphal and
Frederickson (2001) provide evidence that the balance of power between the
board Of directors and the CEO is an important consideration in evaluating the
inﬂuence Of the board on interim ﬁrm outcomes. Given the proper measurement
Of variables representing my constructs, I believe the general model I developed
for this study is a reasonable representation of the relationships I am interested
in exploring.

I plan to develop a theoretical basis for this research that is more reﬁned
than that presented here. Over time, as I have worked with the data I have
collected and discovered more about the nature Of the ﬁrms in my sample, I have
come to view the upper echelon inﬂuences on the development Of infrastructure
as an example Of success or failure in adaptation occurring within the ﬁrm. It
isn’t just a matter Of the degree to which directors provide access to valuable

outside resources in the form of their business expertise, as the traditional

122

resource dependence view suggests. The typical application of resource
dependence theory suggests that Obtaining critical resources (such as director
expertise) from outside the ﬁrm will reduce the uncertainty inherent in a ﬁrm’s
external environment, giving the ﬁrm an increased likelihood Of survival and
success. Implicit in this view is that the primary source of uncertainty is external
to the ﬁrm, that the management Of the ﬁrm has a ﬁrm grasp on the potentially
disruptive forces within the company itself.

In light Of what I have Ieamed about ﬁrms that are new to the public arena,
this is not likely to be true. The reduction of internal uncertainty for these
growing ﬁrms is also critical, since the management team Of most of them have
not experienced this particular transition before. The resources represented on
the board of directors, in interaction with those Of the CEO, may have the
potential to reduce internally generated uncertainty through the pooling Of
knowledge needed to identify the Changes that must be made to navigate this
transition successfully (Kuwada, 1998). Once the need to intemally adapt to the
ﬁrrn’s Changing Circumstances, (impending high growth), by making Changes
such as increasing infrastructure development, these upper echelon resources
are also critical to implementing the changes in an effective manner. DeCanio,
et al. (2000) emphasize the importance of completing such an organizational
structure transition appropriately for a ﬁrm to successfully adapt to Changing
Circumstances.

By modifying the resource dependence view to direct its focus toward

reducing uncertainty Within the ﬁrm and combining it with the literature on

123

adaptation, I think the theoretical foundation for my research questions becomes
more interesting and perhaps more readily testable.

Lastly, the likelihood of detecting the types Of relationships in this model
would be increased with more information regarding the processes behind the
results. My next steps with this line of research will be to conduct on-site
interviews with executives Of a handful Of recent IPO ﬁrms to gain insight into
how the top managers and board interact in achieving intemal strategic Changes
such as infraStructure development. For instance, I hypothesized in this study
that increased experience on the part Of the CEO, regardless Of its source, would
result in a diminished role Of the board on infrastructure development. It may be
that the interaction of the board and CEO is more reﬁned, i.e. that in some cases
(such as when the CEO’S and directors’ experience is general management-
based) that a supplementary effect occurs, where the CEO and the board pool
their resources to effect change. I would like on-site interviews to prepare me to
conduct qualitative research with a small number of ﬁrms, from which I can
induce a reﬁned theoretical framework to guide further research in this area.

Firm development. There are two speciﬁc avenues I am interested in
pursuing related to upper echelons questions in ﬁrms as they evolve over time.
The ﬁrst was triggered by my observation that the composition Of the boards Of
directors for ﬁrms with IPOS on the horizon or that have recently undergone an
IPO is very different from the composition Of the large, mature ﬁrms that are the
context for most board studies. Pfeffer (1972, 1973) established 30 years ago

that ﬁrms’ resource needs inﬂuence the make-up Of the board Of directors. I am

124

curious as to whether there are predictable changes that occur in board
resources as ﬁrms within an industry age and grow. TO answer this research
question, I plan to collect data regarding board and top management team
composition for ﬁrms within several different industries, i.e. traditional
manufacturing, service, research and development intensive, and technology-
based. This study would be cross-sectional, thus controlling for within-industry
effects, by examining ﬁrms Of varying stage of development at one point in time.
Flamholtz’s (1990) book, Growing Pains, presents a ﬁrm development pyramid
that describes the stages a ﬁrm needs to move through in priority order if it is to
succeed long-term. This type Of framework provides a potential basis for the
formulation Of hypotheses. In addition to these developmental factors, I expect
the resources represented by the CEO and the rest of the ﬁrm’s top
management team to affect the backgrounds Of the directors chosen for the
board. For instance, holding TMT expertise constant, younger, smaller ﬁrms '
more are likely to seek out directors with industry-speciﬁc expertise to assist the
company in developing its business strategy and market niche, the two lowest 2
levels on the Flamholtz pyramid. Golden and Zajac (2001) found a curvilinear
relationship between board attributes such as diversity of experience and size
and strategic decision involvement. Board demographics other than directors’
professional backgrounds may also vary as a ﬁrm grows and matures.
Additional infrastructure issues arise in the ﬁrm development context
besides the research questions related to upper echelons. For instance, is there

such a thing as too much infrastructure? Hambrick and Crozier (1985) advocate

125

maintaining an entrepreneurial culture if a ﬁrm is to enjoy long-term success, an
effort that is more likely to be realized if the ﬁrm remains less bureaucratic.
Organizational theorists and researchers have identiﬁed factors such as tOO
many layers of organizational structure and centralization as potential barriers to
adaptation (March 8 Simon, 1958; Miller, 1994; Perrow, 1970; Slevin 8 Covin,
1997). Can infrastructure be over-developed, even though a ﬁrm is still growing,
- to the point where it interferes with ﬁrm performance? Khandwalla’s (1973)
exploratory research highlights the importance Of aligning elements of
organizational structure with the ﬁrm’s size and the uncertainty Of its
environment. Slevin and Covin (1997) also support a contingency perspective.
They found that organic (decentralized, ﬂexible) versus mechanistic (hierarchical,
rule-oriented) organizational structures were better suited to ﬁrms with emergent
strategies and Operating in benign environments.

What are the other factors that impinge on this outcome? For instance,
can a ﬁrm handle more infrastructure successfully if its management team or
board possesses certain Characteristics? How does turnover of the top
management team impact the ﬁrm? Does it matter if the CEO and her top
I managers have worked together in the past or how long they have worked
together at the current ﬁrm?

The second ﬁrm development-based research question I would like to
investigate relates to a subset of ﬁrms that have an initial public Offering. These
are ﬁrms called “development-stage companies.” They are in the process of

developing products or services but have not yet reached the commercialization

126

stage. For instance, in 1996, there were 20 ﬁrms that had their IPOS with less
than $100,000 in revenues. These ﬁrms present an opportunity to study
infrastructure development over time, since they typically have very little in place
at the time of their IPOS. The focus Of investors in assessing the attractiveness
of the Opportunity to buy shares in these ﬁrms is on the viability of the concept
when (and if) it reaches the point of commercialization. Besides examining the
role of the board and top management team resources on infrastructure
development, development stage companies also provide a context for studying
how the type and the degree of infrastructure impacts subsequent performance.
In this case, subsequent performance may be the company’s success or failure
in bringing the product to market, as well as the revenues it generates or the
market Share It captures.

Impact of investors. One of the unique attributes Of board composition
in IPO ﬁrms is the representation of investors. Understandably, individuals or
ﬁrms that have invested in the ﬁrm prior to its IPO have a strong interest in. the
strategic decisions the ﬁrm makes and desire representation on the board for the
purposes Of monitoring top management. For the purposes Of this study,
investor directors were classiﬁed as support Specialists, based on their expertise
in ﬁnancial matters. Under this coding scheme, all investors were treated
similarly in terms Of their anticipated inﬂuence. It is likely, however, that the
nature of the investor sitting on IPO ﬁrm boards makes a difference in terms of
the director’s involvement in ﬁrm decisions. For instance, research published in

ﬁnance journals ﬁnds effects related to venture capitalist investment in IPO ﬁrms

127

on outcomes such as underpricing or Size of the Offering (Barry, Muscarella,
Peavy, 8 Vetsuypens, 1990; Gonnan 8 Sahlman, 1989). Venture capitalists are
generally considered more activist than other types of investors in their
involvement in intemal ﬁrm matters such as the background of the CEO, the
make-up Of the top management team, and the selection Of an investment bank
that ultimately represents the company in its IPO (Barry et al., 1990; Fried,
Bruton 8 Hisrich, 1998; Lerner, 1995).

The types of investors represented on the boards Of the IPO ﬁrms in ,my
sample ranged from the professional venture capitalist investor, to investors
representing small local (Often personally owned) investment ﬁrms, to corporate
entities (either parent companies or other ﬁrms with an interest in the IPO ﬁrm’s
success), to individual investors (“angel” investors), to family members, to the
current or former managers Of the ﬁrm itself. The size Of the ﬁnancial investment
Of these investors varies considerably, as well as their goals from their
investments. Venture capitalists are generally interested in realizing the gains
from their initial investment as soon after the IPO as is feasible without
jeopardizing the ﬁrm, while ﬁrm managers may be interested in the long term
value Of their investments. The ﬁnance literature contains contradictory ﬁndings
regarding the impact of stock ownership of Ofﬁcers and directors on ﬁrm
outcomes. Jain and Kini (1994) found that the equity retention by the original
entrepreneurs Of the company had a positive relationship with post-IPO
Operating performance Of the ﬁrm. On the other hand, Mikkelson, Partch, and

Shah (1997) found no such relationship in their study Of IPO ﬁrms.

128

The typical insider/outsider distinction is likely to be a less important
consideration in this research setting than other attributes of investor directors.
Based on the level of their investments, their investment goals, and their
professional backgrounds, some investors are likely to be more passive in their
involvement in infrastructure development or other strategic outcomes, (such as
individual investors or those representing small local ﬁrms), while others are apt
to take on a more active role in ﬁrm decision making (such as VCS, corporate
parents, and ﬁrm management). Characteristics and qualiﬁcations Of the top
management team also remain as moderating inﬂuences in this Situation. For
instance, venture capitalists are less likely to have an impact if the CEO and her
management team have corporate experience and/or a track record Of prior
business accomplishments.

Related to the role of investors on the board is the choice Of an
investment bank to take the ﬁrm through the IPO process. This factor may also
prove a moderating inﬂuence on interim outcomes such aS infrastructure
development pre-IPO. Depending on the length Of time the investment bank has
been retained and the type Of investment bank Chosen, it may have more or less
inﬂuence on how the ﬁrm prepares for its IPO. As mentioned earlier, investor
directors themselves may have an inﬂuence on this selection.

Survival. The ﬁnal direction for future research I will discuss involves
examining the upper echelon and infrastructure factors that predict survival
among IPO ﬁrms. As I mentioned earlier in regard to limitations Of the current

study, those ﬁrms no longer in existence as independent entities as of the end of

129

fiscal 1999 were automatically dropped from my sample. Considering that ﬁrms

that reach the point of Offering their stock to the public have already survived the
initial start up phase, it would be interesting to gain insight into why they succeed
or fail once they receive an infusion of outside funding.

Using a dependent variable that doesn’t necessitate ﬁrms being in
business as of a particular date in the future would lead to a more complete
sample and more generalizable results. The “survival” dependent variable could
take multiple “values:” dissolution, bankruptcy, acquired (good performance),
acquired (poor performance), independent survival (Change in Strategy), and
independent survival (original strategy), for instance. Jain and Kini (1999)
examined the differences in survival rate for post-IPO ﬁrms that fell into three
categories: survive as an independent ﬁrm, fail outright, or get acquired. Their
study focused on ﬁnancial factors such as risk and industry attractiveness
variables. I would like to examine managerial and internal preparation factors in
this regard, with a more speciﬁc delineation of outcomes, aS described above.
Conclusion

Although the lack Of results associated with the present study are
disappointing, to say the least, I believe this area or research has considerable
potential to contribute to our understanding of how businesses pass through the
high growth transition that typically comes on the heels Of an IPO and go on to
become stable, mature companies. If we can establish some of the key factors,
such as the interaction between a board of directors and a CEO, the

qualiﬁcations Of the other top managers, and the development of infrastructure to

130

support growth, that are associated with subsequent success or failure, we will
have made signiﬁcant strides in this area Of research.

We will also be in a position to assist and guide managers and investors
as they approach the IPO event and consider its implications with more
knowledge and understanding Of the potential consequences Of their decisions at
this critical juncture. Although it is highly unlikely there will ever be “one right
way” to handle this transition, conducting research that demonstrates the
superiority of certain strategies over others in regard to ﬁrm outcomes would be

Of signiﬁcant practical importance.

131

TABLES

132

Table 3.1 TMT Titles Counted for Specialization Variable
in a Sub-sample of Thirty Firms

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TITLE TOTAL MFGR SERVICE KNOWL

Chief Financial Ofﬁcer or 17 4 7 6

Finance 8 Administration .

Operations or Sipply Chain 13 2 6 5

Corporate or Business 11 1 4 6

Development ‘

Sales 8 Marketing 9 3 2 4

(Regional) Operations 9 4 2 3

Research 8 Development 8 2 0 6

or Chief Scientific Ofﬁcer

Sales or Sales Operations 7 2 3 2

Chief Accounting Officer or 6 1 5 0

Controller

Corporate Counsel 6 1 3. 2

Marketing or Public Relations 6 1 4 1

Human Resources or 5 0 2 3

Training

Chief Technology Ofﬁcer or 5 1 0 4

New Technology

Chief Medical Ofﬁcer or 4 0 0 4

‘ Drug Development

Business Division or 4 1 1 2

Strategic Business Unit

Quality Assurance or Quality 4 2 1 1

Control

Manufacturing or Production 4 2 1 1

Engineering 4 3 0 1
‘ Account Management or 4 1 2 1

Client Relations

Construction 3 0 3 0

Chemistry/Biology or 3 0 0 3

Immunology

Materials or Facilities 3 0 1 2

Accyisition

 

133

 

Table 3.1 (cont’d)

 

TITLE

TOTAL

MFGR

SERVICE

KNOWL

 

Product Development

 

Patents or Scientiﬁc
Licensing

0—8

O—E

N—t

 

Enterprise Consulting

 

Merchandising

 

Speciﬁc Product or Service

 

Regulatory Affairs

 

Franchising

 

Administration

 

International

 

Chief Information Ofﬁcer or
Information Tech.

NNNNNNNN

OO-‘OOOOO

ANOON-KO—e

 

Corporate Communications

 

Brewing Operations

 

Food 8 Beverage

 

Store Operations

 

Special Projects

 

Publishing

 

Health Care

 

Telecommunications

 

Real Estate

 

Credit 8 Collections

 

Tax

 

 

Education

 

 

.l-L-l—Lc-L—S—L—S-L-l-ﬁ—B

OOOOOOOOOOOO

 

 

O-‘OO-‘O-AOOOOO

 

134

 

 

EoEoEmm Eoan vouootooc: m_ umtoamu 28:ma ”noun-.328 on Co: 238 58683 Go 9:822 .
cozﬂoommm 9:39: 9 new: 83 6-39 m.__mocov_ F

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S. v at. ”mo. v a. 85865 333050 82:: .383. 9:200 65 mozmzﬁw
8 8 8 8 8 8 2
<2 <2 .8.. .8. :8. s .8. 88288 E88882
. cozanzmcgwmwunﬂm
:E. :8. :E. :5. .8. :8. 8E 8.85.. x85
8. 8. :8. :8. :8. :8. .880 x85
.8. 8. .8. .8. .8. .8. 8Com
8. .8. :8. .8.. .8.. :8. 880E880
8. NF. :8. .8.. .8.. :8. 8888 88:2
:8. :8. :8. :3. M80 8.25
8. 8. R. .8. .8. :5. ano w :8 8835
.8. .8. .8. .8. .8. .8. 85o a tea E88885.
8. or. :E. :8. .8. :8. 8E5:
.8.. 8. :8. :8. .8.. :8. 8.889.. .w 828
8. 8. .8. :8. .8. :8. 8E8... 2868
8. :8. 8. :8. .8. :8. 888m .888:
:8. :8. :8. :8. .cozﬁegcg
8-8 8 8-8 8 8-8 8 was. a 8.8:;
£8; 3:383. 5:588.
3:5me LEGO-.2:— 552.009.85

 

 

3:333”. 334-58. 2... 28:; a.» as.»

135

Table 4.1 Descriptive Statistics and Correlations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables N Mean SD.
1. Service industry 296 .28 .45
2. Knowledge-intensive industry 296 .47 .50
3. Number of employees (1996L 296 306 447
4. Revenues (1996) ($0003) 295 61,345 126,794
5. CEO stock ownership (%) 296 17.87 20.07
6. Board tenure 295 3.06 3.03
7. Board Size 296 6.18 1.66
8. Policies and procedures pre-IPO 296 .66 .47
9. Strategic planning pre-IPO 296 .83 .38
10. Mission statement pre-IPO 296 .03 .16
11. Stock purchase plan pre-IPO 296 .39 .49
12. Specialization pre-IPO 225 5.69 2.96
13. Management development pre-IPO 296 .03 .18
14. Professionalization pre-IPO 284 6.89 3.84
15. Annual report 97 296 .83 .37
16. Annual report 98 296 .65 .48
17. Business experts 296 1.38 1.18
18. Support specialists 296 2.18 1.36
19. Policies and procedures post-IPO 296 .11 .32
20. Strategic planning post-IPO 296 .02 .13
21. Mission statement post-IPO 296 .18 .38
22. Stock purchase plan post-IPO 296 .14 .34
23. Specialization post-IPOa 225 6.05 3.50
24. Management developmntpost-IPO 296 .02 .15
25. Professionalization post-IPO 280 7.15 3.84
26. CEO general mgt experience 237 3.54 4.86
27. CEO industry experience 4242 8.65 . 7.61
28. Number of employees (1998) 291 971 1,926
29. Revenues (1998) ($000’s) 295 126,021 266,731
30. Firm ROAa 292 -.15 .19
31. Firm net income ($000’S) 294 -2,331 27,777
32. Firm shareholders’ equity ($0003) 295 72,206 116,791
33. Firm market value ($000’s)a 288 1.80 .91
34. Industry average ROA 296 .06 .08
35. Industry average ROE 296 .01 .06
36. Industry average MV/Nl (PIE) 252 86.23 117.16
37. Industry average MV/BV 9.92 9.06

295

 

 

2'These variables were logged prior to their inclusion in analyses

 

136

 

Table 4.1 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Vars 1 2 3 4 5 6 7 8
1
2 -.58**
3 .34** -.30**
4 .12* -.20** .57**
5 .20" -.11* .17** .13*
6 -.15** .12* .01 -.06 -.12*

‘ 7 .02 -.10 .21** .15* -.11 -.06
8 .11 -.08 .05 -.08 .04 .13* .05
9 .08 .08 .18** .11 .08 .03 .02 .12*
10 .13* -.07 .13* -.03 .02 -.03 .01 -.01
11 -.10 .11 -.03 .01 -.06 .08 .01 -.07
12 .13* -.08 .27** .21 ** .17* -.03 .20** -.02
13 .22" -.14* .38** .18** .04 -.04 .04 .09
14 .08 -.14* .06 .03 .01 -.15* .10 -.04
15 -.09 .07 .19** .10 -.03 .12* .12* .05
16 .07 -.06 .26** .17** -.01 .13* .06 .03
17 -.01 -.01 .07 .15* -.07 .04 .39** -.07
18 -.01 -.03 -.06 -.12* -.21** -.11 .38** .07
19 -.08 .08 -.07 .07 -.01 -.13* -.01 -.41**
20 .04 -.02 -.03 -.05 -.02 -.02 .03 -.02
21 -.07 .12* .09 .03 -.08 .04 .01 .05
22 -.11 .01 .04 .04 -.08 .06 .12* -.03
23 .06 -.10 .23** .12 .05 .03 .12 .00
24 .05 -.01 .06 -.01 .06 -.03 -.07 -.03
25 .06 -.09 .15* .11 .00 -.10 .09 .04
26 .00 -.13* .09 .06 .01 -.16* .09 -.02
27 .00 .01 .05 -.03 -.06 .00 .08 -.11
28 .29** -.22** .68** .53** .18** -.1 1 .19** .05
29 .16** -.18** .49** .90** .16** -.09 .20** -.07
30 .14* -.09 . .22** .18** .00 .04 .07 -.01
31 .04 -.11 .22** . .31** .12* .00 -.03 -.01
32 .04 .02 .37** .46“ .09 -.1 1 .20** -.02
33 -.08 .17** .31 ** .29** .06 .01 .17** -.02
34 -.24** .27** -.14* -.12* -.02 .01 -.13* .08
35 .09 -.13* .00 .07 .12* -.12* -.10 .04
36 -.22** .34" -.14* -.13* -.21** .07 .03 -.01
37 -.38** .51** -.28** -.24** -.13* .01 -.12* -.14*
*p < .05
**p < .01

 

137

 

Table 4.1 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Vars 9 10 11 12 13 14 15 16
1

2

3

4

5

6

7

8

9

10 .02 ‘

11 .15** -.05

12 .09 .00 .20**

13 .08 .08 -.07 .12

14 -.10 .05 -.09 -.05 .07

15 .21** .07 .13* .06 .03 .04

16 .16** .08 .13* .07 .06 .04 .04

17 -.07 .07: .01 .08 -.01 .03 .03 .06
18 .07 -.02 .03 .11 -.03 .06 .03 -.07
19 -.41** -.06 -.02 .00 -.01 .01 -.04 -.01
20 -.29** .14* .00 .03 -.03 .02 .06 -.01
21 A .14* -.08 .02 -.05 .06 .01 .16*“ .21**
22 .07 -.01 -.31** -.04 -.02 .13* .10 .02
23 .16* -.04 .08 .49** .12 .00 -.10 .00
24 -.05 .11 -.03 .07 -.03 .06 .07 .07
25 .06 -.04 -.03 .02 .03 .57** .14* .09
26 -.10 .04 -.06 -.04 .01 .08 .10 .00
27 .03 .00 -.02 -.04 .00 .17* .21** .08
28 .15* .04 -.09 .22** .36** .03 ' .14* .19**
29 .12* .00 -.02 .21** .24** .05 .11 .17**
30 .11 .05 .10 .11 .05 -.07 .13* .20**
31 .04 .04 -.04 -.00 .11 -.01 .05 .10
32 .12* .00 .10 .12, .19** .01 .18** .25**
33 .26** .05 .28" .19** .09 -.06 .27** .33** -
34 .06 .03 .06 .05 -.08 -.11 .06 .05
35 -.06 .04 -.12* -.01 .05 .01 -.01 .01
36 -.05 -.07 .18** -.01 -.09 -.04 .09 -.04
37 .06 -.06 .20** -.01 -.15** -.06 .03 -.06
*p < .05

**p < .01

 

138

 

Table 4.1 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Vars 17 18 19 20 21 22 23 24
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18 -.30**

19 .03 -.08

20 .07 -.06 .12*

21 .01 .01 -.02 -.06

22 .15* .06 .11 .03 .00

23 .00 .07 .03 -.02 -.05 -.01

24 -.03 -.02 .02 .33** -.01 .13* -.02

25 .08 -.03 -.05 -.04 -.04 .15* -.06 .03
26 .10 .03 .01 -.06 .01 .03 .07 -.04
27 .11 .10 -.01 .02 -.08 .20** -.09 .22**
28 .04 -.07 .01 -.02 .07 .04 .23** .03
29 .16** -.08 .11 -.04 .01 .04 .15* -.01
30 .06 -.01 -.04 -.04 .04 .03 .20** .05
31 .13* -.15** -.06 -.04 .06 .09 -.03 .04
32 .08 -.06 .06 -.02 .13* -.05 .21** .03
33 .10 -.01 .02 .02 .16* -.02 .15* .01
34 -.07 -.01 .06 .03 .10 .06 -.07 .05
35 -.04 -.02 .02 .04 -.01 .06 -.01 .05
36 .06 .11 -.06 .01 .04 -.05 -.05 -.07
37 -.12* .06 .12* -.03 .11 -.02 -.02 -.07
*p<.05

**p<.01

 

139

 

Table 4.1 (cont’d)

 

Vars

25

26

27

28

29

30

31

32

 

 

 

 

 

 

 

 

 

(DmVODUI-hOON-t

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.19**

 

.15*

.11

-.06

 

.06

-.05

.68**

 

-.02

.07

.15*

.17*”

 

.06

.07

.16*”

.33“

.28“

 

.09

.05

.48"

.56**

.21**

.25"

 

.00

09

.27**

.33"

.33“

.19"

.63“

 

-.11

(.13*

-.10

-.10

.03

.17**

-.04

 

-.01

-.05

01

06

.06

.25“

-.02

 

.01

.02

-.13*

.13*

-.05

-.12

-.01

 

 

 

-.03

 

-.04

 

 

-.23**

 

-.06

 

-.10

 

-.02

 

 

-.21**

 

140

 

33 34 35

1
2
3
4
5
6
7
8
9

34
35
36
37
*p < .05
** < .01

.02
-.01

.14*

.07

.51 **
-.18**
.43“

-.47**
-.09

 

Table 4.1 (cont’d)

36

.37**

141

Table 4.2 Regression of High Growth Transition Infrastructure Changes

on Board Composition

 

Policies and Procedures

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Ste 3 1 Step 2

b Wald b Wald
Step 1: Controls
Service -.23 .12 -.18 .07
Knowledge .34 .42 .40 .58
Employees 96 .00 3.72* .00 4.04“
Revenues 96 .00 3.67 .00 3.53
CEO stockownership % .00 .02 .00 .09
Board tenure -.19 2.70 -.20 2.97
Board size .06 .24 .20 1.34
Annual report 97 -.38 .37 -.39 1.36
Annual report 98 .54 1.00 .54 .98
Policies 8 procedures 96 -2.87 25.95““ -2.84 25.30””
Strategic planning 96 '
Mission statement 96
Stock purchase plan 96
Management develop. 96
Specialization 96
Professionalization 96
% Correct 89.1
x" 57.3"

Step 2: IVs
Business experts ~.10 .20
Support Specialists -.30 2.11
% Correct 90.1
2.3

N 294
*p < .05
*3) < .01

 

142

 

Table 4.2 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Strategic Plannigg
Variables Ste 3 1 Step 2
b Wald b Wald
Step 1: Controls
Service -1 .06 .05
Knowledge 1 .76 .53
Employees 96 .04 2.61
Revenues 96 .00 1.58
CEO stockownership % .02 . .04
Board tenure -.39 .68
Board size .30 .31
Annual report 97 11.78 .00
Annual report 98 1.17 .29
Policies 8 procedures 96
Strategic planning 96 -60.22 .58
Mission statement 96
Stock purchase plan 96
Management develop. 96
Specialization 96
Professionalization 96
% Correct 98.9
x' 37.4“
Step 2: IVS
Business experts - NA NA
Support specialists NA NA
% Correct NA
NA
N 294
*p < .05
**p < .01

NA = Solution not found after 20 iterations

 

143

 

Table 4.2 (cont’d)

 

Mission Statement

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Ste 1 Step 2
b Wald b Wald -

Step 1: Controls
Service -.08 .02 -.09 .03
Knowledge .91 4.31* .91 429*
Employees 96 .00 4.88* .00 4.91 *
Revenues 96 .00 .70 .00_ .69
CEO stockownership "/0 -.01 2.17 -.01 1.94
Board tenure —.03 .26 -.03 .24
Board size -.07 .49 -.09 .49
Annual report 97 1.12 2.04 1.10 1.96
Annual report 98 1.19 6.69“ .1.19 6.76“
Policies & procedures 96
Strategic planning 96
Mission statement 96 -6.18 .24 -6.19 .24
Stock purchase plan 96
Management develop. 96
Specialization 96
Professionalization 96
"/o Correct 82.3
x“ 32.6“

Step 2: IVs
Business experts .03 .04
Support specialists .04 .06
% Correct 82.7
F .1
N 294
*p < .05
**p < .01

 

144

 

Table 4.2 (cont’d)

 

Stock Purchase Plan

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Ste D 1 Step 2
b Wald b Wald

Step 1: Controls

Service -1.53 6.01 ** -1.56 6.08“

Knowledge -.49 1.18 -.35 .58

Employees 96 .00 .93 .00 1.97

Revenues 96 .00 .44 .00 .72

CEO stockownership % -.01 1.58 -.01 .94

Board tenure .09 2.04 .08 1.55

Board size .20 2.99 -.04 .06

Annual report 97 1.08 2.29 .92 1.62

Annual report 98 -.05 .01 -.03 .00

Policies 8. procedures 96

Strategic planning 96

Mission statement 96

Stock purchase plan 96 -9.29 .40 -9.24 .39

Managment develop. 96

Specialization 96 '

Professionalization 96

“lo Correct 85.7

x" - 63.7“

Step 2: IVs

Business experts .41 3.75*

Support specialists .29 2.31 ,
. % Correct 85.0

4.3

N 294

*p < .05

**p < .01

 

145

 

Table 4.2 (cont’d)

 

Management Development

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Step 1 Ste 2
b Wald b Wald
Step 1: Controls
Service .41 .10 .39 .09
Knowledge .45 .14 .45 .14
Employees 96 .00 2.40 .00 2.39
Revenues 96 .00 1.03 .00 .98
CEO stockownership % .01 .45 .01 .47
Board tenure -.12 .49 -.12 .46
Board size -.51 2.40 -.55 1.65
Annual report 97 7.29 .04 7.27 .04
Annual report 98 .73 .40 ° .72 .39
Policies & procedures96
Strategic planning 96
Mission statement 96
Stock purchase plan 96 '
Management develop. 96 -8.35 .01 —8.31 .01
Specialization 96
Professionalization 96
% Correct 97.6
x“ 10.4
Step 2: IVs
Business experts .03 .00‘
Support specialists .08 .04
% Correct 97.6
.0
N 294
*p < .05
**p < .01

 

146

 

Table 4.2 (cont’d)

 

Specialization

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Ste 3 1 Step 2
B t B t

Step 1: Controls

Service -.09 -1.03 -.10 -1.05

Knowledge -.16 -1.84 -.16 -1.85

Employees 96 .11 1.24 .11 1.21

Revenues 96 -.05 -.62 -.05 -.54

CEO stockownership % .02 .25 .02 .31

Board tenure .08 1.12 .08 1.18

Board size .08 1.10 .07 .77

Annual report 97 -.15 -2.13* -.15 -2.10*

Annual report 98 .03 .46 .04 .49

Policies & procedure396

Strategic planning 96

Mission statement 96

Stock purchase plan 96

Management develop. 96

Siecialization 96 .45 6.28" .44 6.14“

Professionalization 96

R2 .30

F 6.50”

Step 2: IVs

Business experts .00 -.04

Support specialists .04 .43

AR: .00

AF .13
.30
.24

N 164

*p < .05

**p < .01

 

147

 

Table 4.2 (cont’d)

 

Professionalization

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Ste 2

B t B t
Step 1: Controls
Service -.04 -.52 -.03 -.47
Knowledge -.02 -.34 -.02 -.35
Employees 96 .06 .83 .06 .81
Revenues 96 .05 .74 .04 .56
CEO stockownership % -.01 -.25 -.02 -.43
Board tenure -.04 -.72 -.05 -.89
Board size -.02 -.45 -.01 -.19
Annual report 97 .05 .96 .06 .97
Annual report 98 .01 .21 .01 .16
Policies 8 procedures96
Strategic planning 96
Mission statement 96
Stock purchase plan 96
Management develop. 96
Specialization 96
Professionalization 96 .55 10.59” .55 10.59"
R2 .33
F 13.16”

Step 2: IVs
Business experts .02 .33
Support specialists -.06 -.86
AR2 .00
AF .84
.34
.31

N 272
*p < .05
**p < .01

 

148

 

Table 4.3 Regression of Firm Performance
on High Growth Transition Infrastructure Changes

 

Return on Assets

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Ste 2
B t B t

Step 1: Controls

Service .19 1.83 .23 2.31*

Knowledge -.02 -.18 .04 .35

Employees 96 .18 1.83 .16 1.59

Revenues 96 .08 .96 .12 1.39

CEO stock °/o .04 .57 .05 .65

Policies 96 -.20 -2.56** -.19 -2.22*

Planning 96 .14 1.90 .09 1.09

Mission 96 -.03 -.40 -.05 -.65

Stock purchase 96 -.02 -.28 .02 .28

Specialization 96 .00 -.01 -.11 -1.24

Mgt develop. 96 .02 .30 .05 .66

Professional. 96 -.06 -.76 .01 .10

Annual report 97 -.08 -1.03 -.04 -.51

Annual report 98 .29 3.69“ .29 3.74“

Industry Avg ROA .13 1.72 .16 2.10*

Shareholders’ equity

Industry Avg ROE

Net income

Industry Avg P/E

Industry Am MV/BV

R2 and F .29 3.84”

Step 2: IVs

Policies As -.01 -.15

Planning As -.06 -.81

Mission As -.09 -1.14

Stock purchase As .13 1.70

Specialization 98 .23 2.66“
Mgt develop. As .04 .54

Professional. 98 -.12 -1.19

AR2 and AF .07 2.21*

Total R2 .36

Adjusted R2 .26

N 156

*p < .05

**p < .01

 

149

 

Table 4.3 (cont’d)

 

Net Income (ROE)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Step 2

B t B t
Step 1: Controls
Service -.07 -.65 -.04 -.39
Knowledge -.01 -.09 -.00 -.03
Employees 96 .13 1.31 .09 .85
Revenues 96 .00 .01 .01 .14
CEO stock % .10 1.31 .12 1.56
Policies 96 -.07 -.92 -.14 -1.71
Planning 96 .01 .14 .01 .14
Mission 96 .05 .66 .05 .68
Stock purchase 96 .00 .02 .03 .40
Specialization 96 -.08 -1.05 -.08 -.96
Mgt develop. 96 .03 .33 .04 .44
Professional. 96 -.12 -1.63 -.18 -1.79
Annual report 97 .07 .86 .08 .99
Annual report 98 .06 .76 .06 .72
Industry Avg ROA
Shareholders” equity .14 1.85 .16 2.01 *
Industry Avg ROE .44 5.83" .43 5.60“
Net income

 

Industry Ag; P/E

 

Industry Avg MV/BV '

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

if and F .29 3.61”

Step 2: IVs

Policies As ' . -.18 -2.18*
Planning As .01 ' .07
Mission As .03 .43
Stock purchase As .08 .93
Specialization 98 -.01 -.10
Mgt develop. As .01 .15
Professional. 98 .08 .77
AR2 and AF .04 1.00
Total R2 .33

Adjusted R2 .21

N 157

*p < .05 '

**p < >01

 

150

 

Table 4.3 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Market Value (PIE)
Variables Step 1 Step 2
B t B t

Step 1: Controls

Service -.11 -.99 -.13 -1.15
Knowledge .16 1.42 .18 1.42
Employees 96 .19 1.86 .22 2.01*
Revenues 96 .16 1.71 .15 1.54
CEO stock % .13 1.51 .11 1.27
Policies 96 -.11 -1.33 -.08 -.80
Planning 96 .01 .11 .02 .22
Mission 96 .12 1.56 .13 . 1.51
Stock purchase 96 .24 2.71** .19 1.96*
Specialization 96 .10 1.12 .08 .82
Mgt develop. 96 .00 -.02 -.01 -.06
Professional. 96 -.01 -.18 .03 .30
Annual report 97 -.18 -2.19* -.17 -2.07
Annual report 98 .12 1.47 .12 1.45
Industry Avg ROA

Shareholders' equity

Industry Avg ROE

Net income .21 2.61“ .23 2.79
Industry Avg P/E .12 1.32 .13 1.44
Industry Avg MV/BV

R2 and F .33 '3.61**

Step 2: IVs

Policies As .09 .98
Planning As .00 -.03
Mission As -.04 -.47
Stock purchase As -.11 -1.26
Specialization 98 .05 .50
Mgt develop. As .00 -.02
Professional. 98 -.04 -.38
AR2 and AF .02 .51
Total R2 .35

Adjusted R2 .22

N 132

*p < .05

**p < >01

 

151

 

Table 4.3 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Market Value (MVIBV)

Variables Step 1 Step 2

B t B t
Step 1: Controls
Service -.07 -.84 —.10 -1.19
Knowledge .18 2.10* .16 1.84
Employees 96 .13 1.62 .17 2.03*
Revenues 96 .02 .33 .01 .08
CEO stock % .03 .53 .03 .42
Policies 96 -.10 -1.69 -.05 -.79
Planning 96 .01 .18 .02 .37
Mission 96 .09 1.57 , .08 1.38
Stock purchase 96 .15 2.41* .15 2.17*
Specialization 96 .03 .48 .07 .92
Mgt develop. 96 -.01 -.19 -.02 -.31
Professional. 96 -.05 -.77 .00 .05
Annual report 97 -.06 -1.01 -.08 -1.28
Annual report 98 .12 1.90 .12 1.93
Industry Avg ROA ,
Shareholders’ equity .59 9.51" .60 9.40"
Industry Avg ROE
Net' income
Industry Avg P/E
Industry Avg MV/BV -.01 -.18 -.04 -.49
R2 and F .56 10.86“
Step 2: IVs
Policies As .12 1.78
Planning As -.01 .21
Mission As -.03 -.46
Stock purchase As -.02 -.29
Specialization 98 -.08 -1.08 .
Mgt develop. As -.04 -.64
Professional. 98 -.06 -.75
Art2 and AF .02 .82
Total R7 .53
Adjusted R2 .50
N 154
*p < .05
**p < >01

 

152

 

Table 4.4 Regression of Firm Performance on Board Composition

 

Return on Assets

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Step 2

B t B t
Step 1: Controls
Service .13 1.69 .13 1.67
Knowledge .02 .24 .02 .23
Employees 96 .15 1.86 .15 1.90
Revenues 96 .09 1.22 .09 1.14
CEO stock °/o -.05 -.79 -.04 -.70
Board tenure .05 .90 .05 .87
Board size .03 .46 .01 .13
Industry AquOA .09 1.41 .09 1.42
Shareholders' equity '
Industry Avg ROE
Net income
Industry ﬂP/E
Industry Avg MV/BV
R2 .07
F 2.69”
Step 2: IVs
Business experts .04 .49
Support specialists .01 .16
AR2 .00
AF .13
Total R2 .07
Adjusted R2 .04
N 289
*p < .05 _
**p < .01

 

153

 

Table 4.4 (cont’d)

 

Net Income (ROE)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Step 2
B t B t

Step 1: Controls
Service -.08 -1.12 -.08 -1.21
Knowledge -.07 -1.07 -.08 -1.18
Employees 96 .08 1.02 .09 1.23
Revenues 96 .18 2.53“ .16 2.17*
CEO stock "/0 .05 .91 .06 .98
Board tenure .06 .99 .05 .80
Board size -.08 -1.39 -.14 -1.85
Industry Avg ROA
Shareholders’ equity .16 2.43* .16 2.47“
Industry Avg ROE .23 4.23" .23 4.24“
Net income
lndustrLAvg P/E
Industry Avg MV/BV
R2 .19
F 7.12**
Step 2: IVs
Business experts .15 2.14*
Support specialists -.00 -.02
AR2 .02
AF 1328*

' Total R2 .20

. Adjusted R2 .17
N 291
*p < .05
**p < .01

 

154

 

Table 4.4 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Market Value (PIE)

Variables Step 1 Step 2

B t B t
Step 1: Controls
Service -.06 -.86 -.07 -.86
Knowledge .20 2.59“ .20 2.60“
Employees 96 .29 3.61” .29 3.59“
Revenues 96 .12 1.49 .12 1.54
CEO stock % .10 1.65 .11 1.67
Board tenure .05 .87 .06 .91
Board size .09 1.55 .09 1.11
Industry Avg ROA
Shareholders’ equity
Industry Avg ROE
Net income .12 ' 1.92 .12 195*
Industry Avg P/E .13 2.04* .13 199*
Industry Avg MV/BV
R2 .23
F 7.61“
Step 2: IVs g
Business experts -.02 -.20
Support specialists .03 .34
AR2 .00
AF .17
Total R2 .23
Adjusted R2 .19
N 241
*p < .05
**p < .01

 

155

 

Table 4.4 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Market Value (MVIBV)

Variables Step 1 Step 2

B t B t
Step 1: Controls
Service -.03 -.56 -.04 -.64
Knowledge .18 3.02" .18 2.96"“r
Employees 96 .16 2.56“ .17 2.71**
Revenues 96 -.04 -.58 -.04 -.66
CEO stock % .05 1.09 .07 1.38
Board tenure .05 1.09 .05 1.14
Board size .04 .81 -.02 -.33
Industry Avg ROA .
Shareholders’ equity .58 10.82“ .58 10.91“
Industry Avg ROE
Net income
Industry Avg PIE
Industry Avg MV/BV .03 .53 .03 .55
R7 .45
F 25.79”
Step 2: IVs
Business experts .08 1.36
Support specialists .07 1.23
AR2 .00
AF 1.01
Total R2 .45
Adjusted R2 .43
N 284
*p < .05
**p < .01

 

156

 

Table 4.5 Regression of High Growth Transition Infrastructure Changes

on Board Composition Moderated by CEO Qualiﬁcations

 

Policies and Procedures

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 StepZ Step 3
b Wald b Wald b Wald

Step 1: Controls

Service -.37 .18 -.25 .08 -.28 .08

Knowledge .47 .47 .67 .81 .68 .70

Employees 96 .00 3.05 .00 3.27 .00 3.59

Revenues 96 .00 3.75* .00 3.63 .00 4.50*

CEO stock % .00 .02 .00 .00 .00 .02

Board tenure —.22 2.30 -.23 2.29 -.22 2.05

Board size .06 .15 .24 1.41 .26 1.60

Annual report 97 -.26 .12 -.10 .02 -.28 .11

Annual report 98 .51 .59 .47 .47 .63 .74

Policies/procedures 96 -4.01 14.57“ -4.00 14.43* -4.14 14.43“

% Correctly classified 89.0

X2 53.4"

Step 2: IVs

Business experts -.21 .60 -.25 .73

Support specialists -.32 1.45 -.37 1.79

CEO General mgt .03 .27 .23 1.79

CEO Industry experience -.01 .05 -.01 .03

% Correctly classified 90.0

- 1.9

Step 3: Interactions

BE * CEO General Mgt .06 1.41

SS * CEO General Mgt .06 .50

BE * CEO Industry Exp .03 .01

SS * CEO Industry Exp .03 .24

% Correctly classified 89.5
2.3

N 219

*p < .05

**p < .01

 

157

 

Table 4.5 (cont’d)

 

Strategic Plannin

 

Variables

Step 1

Step 2

Step 3

 

b Wald

b

Wald

Wald

 

Step 1: Controls

 

Service

-8.94 .

00

 

Knowledge

5.90 1.

34

 

Employees 96

.10 1.

89

 

Revenues 96

.00

.87

 

CEO stock %

.00

.00

 

Board tenure

-.54

.57

 

Board size

-.25

.06

 

Annual report 97

13.14

.00

 

Annual report 98

1.39

.13

 

StrategicJJIanning 96

-98.61

.07

 

 

% Correctly classiﬁed

99

.5

 

31.9"

 

 

Step 2: IVs

 

Business experts

NA

 

Support specialists

NA

 

CEO General mgt

NA

 

CEO Industry experience

NA

 

% Correctly classiﬁed

NA

 

AX:

NA

 

 

Step 3: Interactions

 

BE * CEO General Mgt

 

SS * CEO General Mgt

 

BE * CEO Industry Exp

 

SS * CEO Industry Exp

 

% Correctly classiﬁed

 

Ax2

 

 

 

 

 

 

 

 

 

*p < .05
**p < .01

NA = Solution not found after 20 iterations

 

158

 

Table 4.5 (cont’d)

 

Mission Statement

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Step 2 Step 3
b Wald b Wald b Wald

Step 1: Controls
Service .19 .10 .14 .05 .00 .00
Knowledge .86 3.06 .97 3.62 .96 3.39
Employees 96 .00 2.09 .00 3.05 .00 2.91
Revenues 96 .00 .54 .00 .69 .00 .75
CEO stock % -.01 1.23 -.01 1.01 -.01 .77
Board tenure -.02 .09 -.01 .04 -.01 .03
Board size -.07 .45 -.19 1.65 -.17 1.30
Annual report 97 1.52 1.96 1.65 2.28 1.67 2.31
Annual report 98 1.10 4.73* 1.23, 5.67* 1.18 5.16*

-6.03 .11 -6.30 .13 —6.45 .14
% Correctly classiﬁed 81.3
X2 22.8“”r
Step 2: IVs
Business experts .11 .30 .12 , .29
Support specialists .22 1.59 .26 1.74
CEO General mgt .01 .07 -.03 .04
CEO Industry experience -.07 5.65* -.07 556*
°/o Correctly classiﬁed 80.4 .

7.2
Step 3: Interactions
BE * CEO General Mgt .04 .01 ‘
SS * CEO General Mgt .02 .25
BE * CEO Industry Exp -.02 .37
SS * CEO Industry Exp .01 .26
% Correctly classiﬁed 80.1
1.4

N 219
*p < .05
**p < .01

 

159

 

Table 4.5 (cont’d)

 

Stock Purchase Plan

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Step 2 Step 3

b Wald b Wald b Wald
Step 1: Controls
Service -1.45 4.32* -1.56 4.32* -1.44 3.57
Knowledge -.73 2.24 -.72 1.86 -.63 1 .26
Employees 96 .00 .02 .00 .04 .00 .00
Revenues 96 .00 .04 .00 .02 .00 .01
CEO stock % -.01 .75 -.01 .14 .00 .07
Board tenure .10 1.62 .09 1.12 .10 1.47
Board size .14 1.27 -.07 .11 -.11 .25
Annual report 97 .88 1.38 .71 .85 .79 .96
Annual report 98 .11 .05 :05 .01 -.12 .05
Stock purchase plan 96 -9.33 .29 -9.31 .29 -9.43 .30
% Correctly classiﬁed 83.6

51.5“

Step 2: IVs
Business experts .25 1.75 .44 2.26
Support specialists .23 1.59 .25 .92
CEO General mgt .04 .88 -.05 .11
CEO Industry experience .03 4.64* .08 6.22“
% Correctly classiﬁed 84.9 ‘
Ax“ 9.0
Step 3: Interactions
BE * CEO General Mgt .05 2.50
SS * CEO General Mg -.03 .50
BE * CEO Industry EQ -.06 4.70*
SS * CEO Industry Exp -.03 1.06
% Correctly classiﬁed 87.7
Ax‘ 1.9
N 219
*p < .05
**p < .01

 

160

 

Table 4.5 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NA = Solution not found after 20 iterations

Management Development
Variables Step 1 Step 2 Step 3
b Wald b Wald b Wald

Step 1: Controls
Service -.95 .25 -2.26 .58
Knowledpe .88 .31 .93 .15
Employees 96 .00 3.65 .01 3.38
Revenues 96 .00 2.03 .00 2.54
CEO stock % .04 1.49 .06 2.10
Board tenure -.09 .17 -.28 .64
Board size -1.07 4.34" -2.79 2.54
Annual report 97 7.93 .01 6.81 .00
Annual report 98 10.04 .03 13.15 .02
Management develop. 96 -11.08 .00 -7.88 .00
% Correctly classiﬁed 97.7
x2 16.2
Step 2: IVs
Business experts 1.90 1.39
Support specialists .32 .06
CEO General mgt -.07 .32

. CEO Industry experience .33 4.80*

, % Correctly classiﬁed 97.7

10.8*

Step 3: Interactions
BE * CEO General Mgt

‘ SS * CEO General Mgt
BE * CEO Industry Expp
SS * CEO Industry EJX)
% Correctly classiﬁed NA
Ax"z NA
N
*p < .05
**p < .01

 

161

 

Table 4.5 (cont’d)

 

Specialization

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Step 2 Step 3
B t B t B t
Step 1: Controls
Service -.12 -1.09 -.11 -.93 -.09 -.75
Knowledge -.1 1 -1.08 -.09 -.79 -.08 -.76
Employees 96 .20 1.86 .19 1.68 .19 1.61
Revenues 96 -.02 -.25 -.01 -.11 -.01 -.12
CEO stock% .10 1.15 .11 1.18 .10 1.06
Board tenure .15 1.77 .17 1.90 .16 1.78
Board size .16 1.79 .13 1.16 .11 1.00
Annual report 97 -.17 -1.70 -.16 -1.63 -.18 -1.73
Annual report 98 .10 1.04 .09 .88 .11 1.00
Specialization 96 .38 4.50" .38 4.41" .38 4.34“
R2 .30
F 4.69“
Step 2: IVs
Business experts .01 .06 .00 -.04
Support specialists .05 .43 .03 .30
CEO General mt .12 1.39 .10 .40
CEO Industry experience -,05 -.60 -.04 -.45
AR2 .02
AF .61
Step 3: Interactions
BE * CEO General Mgt -.09 -.52
SS * CEO General Mgt .06 .27
BE * CEO Industry Exp .08 .61
SS * CEO Industry Exp -.09 -.94
AR2 .01
AF .50
N .33
*p < .05
**p < .01

 

162

 

Table 4.5 (cont’d)

 

Professionalization

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables Step 1 Step 2 Step 3

B t B t B t
Step 1: Controls
Service .01 .09 .02 .20 .02 .26
Knowledge -.04 -.53 -.04 -.49 -.04 -.58
Employees 96 -.05 -.58 -.06 -.72 -.06 -.75
Revenues 96 .09 1.17 .08 1.08 .07 .92
CEO stock % -.07 -1.09 -.09 -1.37 -.09 -1.49
Board tenure -.03 -.49 -.03 -.53 -.04 -.71
Board size -.08 -1.27 -.03 -.43 -.03 -.36
Annual report 97 .02 .33 .03 .41 . .05 .68
Annual report 98 -.01 -.18 -.02 -.25 -.05 -'.75
Professionalization 96 .58 9.74“ .58 9.53" .60 10.06"
R2 .35
F 10.75“
Step 2: IVs -
Business experts -.02 -.29 -.03 -.38
Support specialists -.10 -1.23 -.10 -1.27
CEO General mgt .02 .24 -.56 -3.13**
CEO Industry experience -.02 -.31 - -.02 -.27
AR2 .01
AF .49
Step 3: Interactions
BE * CEO General Mgt .40 3.42"
SS * CEO General Mgt .35 2.61”
BE * CEO Industry Exp -.19 -2.40**
SS * CEO Industry Exp -.05 -.79
Art2 .05
AF 3.59“
N .41
*p < .05
**p < .01

 

163

 

Table 4.6 Regression of Firm Performance on High Growth Transition
Infrastructure Changes Moderated by Change In Firm Size

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Return on Assets
Variables B t B t B t
Step 1: Controls
Service .23 2.21* .26 2.52“ .34 3.14”
Knowledge .04 .38 .08 .76 .07 .67
CEO stock % .05 .65 .03 .37 .07 .77
Policies 96 -.17 -2.13* -.14 -1.69 -.11 -1.24
Mission 96 -.02 -.31 -.06 -.83 -.03 -.41
Stock purchase 96 -.01 -.10 .03 .31 .00 .03
Specialization 96 .01 .10 -.11 -1.32 -.17 -1.84
Management develop. 96 .01 .12 .05 .59 -.02 -.19
Professionalization 96 -.06 -.79 -.01 -.06 -.07 -.63
Employees 96 .21 2.04* .27 2.25 .29 2.21*
Revenues 96 .09 1.08 -.14 -.77 -.23 -1.09
Annual report 97 -.09 -1.13 -.04 -.51 -.05 -.63
Annual report 98 .28 3.53“ .26 3.40“ .25 3.07“
Industry averaﬁ ROA .12 1.57 .16 2.02* .18 2.27'
R2 and F .27 3.64”
Step 2: IVs
Policies As ‘ -.02 -.19 .07 .76
Mission As -.07 -.90 -.05 -.63
Stock purchase As .16 1.97 .13 1.24
Specialization 98 .26 3.01“ .29 3.18"
Management develop. As .05 .70 .09 .98
Professionalization 98 -.10 -1.03 -.05 -.46
Employees 98 -.16 -1.38 -.27 -1.23
Revenues 98 .31 1.67 .45 1.85
AR2 and AF .10 2.58"
Step 3: Interactions
Policies*Emponees . .21 1.60
Mission*Emponees .14 1.27
Stock purchase*Emp. -.12 -.51
S ecialization*Emponees .14 .79
Mgt develop*Emp|oyees -.02 -.25
Professionalization*Erm). -.05 -.34
Policies*Revenues -.26 -1 .97"
Mission'Revenues .16 1.22
Stock purchase*Rev. .04 .21
Specialization*Revenues -.10 -1.01
Mgt develop*Revenues .04 .51
Professionalization*Rev. .19 1.64
AR2 and AF ' .05 .90
Total R2 .42
Adjusted R2 .26
N 152

 

 

 

 

 

 

 

 

*p< .05 **p< .01

 

 

164

Table 4.6 (cont’d)

 

Net Income (Return on Equity)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables B t B t B t
Step 1: Controls
Service -.06 -.64 -.02 -.23 .01 .11
Knowlepqpe .00 .02 .01 .06 .00 .01
CEO stock % .10 1.28 .11 1.39 .19 2.18*
Policies 96 -.06 -.85 -.12 -1.50 -.10 -1.11
Mission 96 .05 .63 .03 .34 .01 .14
Stock purchase 96 .03 .36 .04 .42 .11 1.19
Specialization 96 -.09 -1.13 -.09 -1.05 -.17 -1.84
Management develop. 96 .05 .57 .07 .81 -.01 -.06
Professionalization 96 —.12 -1.55 -.19 -1.86 -.18 -1.63
Employees 96 .14 1.35 .22 1.81 .35 2.65“
Revenues 96 .01 .13 -.16 -.84 -.49 -2.15*
Annual report 97 .07 .91 .08 1.06 .06 .82
Annual report 98 .06 .75 .04 .56 .02 .21
Shareholders’ equity .14 1.71 .16 1.71 .15 1.47
Industry average ROE .44 5.90“ .43 5.70" .44 5.80”
R2 and F .30 3.39"
Step 2: IVs
Policies As -.19 -2.34* -.24 -2.56**
Mission As .05 .59 .04 .52
Stock purchase As .07 .91 .03 .25
Specialization 98 .02 .21 .05 .54
Management develop. As .01 .19 .06 .71
Professionalization 98 .09 .94 .11 .97
Employees 98 -.23 -2.01* -.63 -2.84**
Revenues 98 .20 .98 .47 1.73
AR2 and AF .06 1.45
Step 3: Interactions
Policies*Employees —.08 -.60
Mission*Employees .10 .85
Stock purchase*Emp. .12 .51
Specialization*Employees .27 1 .55
_l\_II_gt develop*Emp|oyees .02 .18
Professionalization*Emp. -.15 -.93
Policies*Revenues -. 14 -1.04
Mission*Revenues .23 1.72
Stock purchase*Rev. -.17 -.86
Specialization*Revenues .02 .16
_M_gt develop*Revenues .04 .51
Professionalization*Rev. .17 1.45
AR2 and AF .07 1.13
Total R2 .42
Adjusted R2 .25
N 153

 

 

*p< .05 **p< .01

 

165

 

Table 4.6 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Market Value (PIE)
Variables B t B t B t
Step 1: Controls
Service -.1 1 -.97 -.23 -2.06* -.16 -1.40
Knowlepge .16 1.42 .13 1.13 .13 1.13
CEO stock % .13 1.49 .09 1.10 .09 .90
Policies 96 -.11 -1.32 -.09 -1.00 -.04 -.40
Mission 96 .12 1.55 .09 1.12 .10 1.14
Stock purchase 96 .23 2.60" .20 2.19* .16 1.75
Specialization 96 .10 1.17 .08 .81 .04 .37
Management develop. 96 .01 .09 .02 .24 .00 .04
Professionalization 96 -.01 -.10 .05 .46 .00 .00
Employees 96 .19 1.80 .31 2.31” .32 2.24*
Revenues 96 .16 1.70 -.44 -2.12* -.64 -2.61**
Annual report 97 -.18 -2.19* -.17 -2.15* -.17 -2.11*
Annual report 98 .12 1.41 .11 1.33 ' .12 1.45
Net income .22 2.73" .19 235* .14 1.52
Industry average P/E .10 1 .19 .14 1.52 .15 1.76
R2 and F .34 3.33"
Step 2: IVs
Policies As .08 .96 .33 218*
Mission As -.01 -.17 .00 -.02
Stock purchase As -.07 -.83 .02 .14
Specialization 98 .02 .18 .07 .79
Management develop. As .00 -.01 -.29 -1.41
Professionalization 98 -.03 -.26 .03 .22
Employees 98 -.06 -.47 ' -.06 -.21
Revenues 98 .66 3.16“ 1.29 4.10“
AR2 and AF .03 .96 .33 2.18*
Step 3: Interactions .09 2.03*
Policies*Employees .41 1.84
Mission*Emponees -.13 -.78'
Stock purchase*Emp. -.34 -1.31
Specialization*Emponees .18 .95
_M_gt develop*Emp|oyees .36 1.55
Professionalization*Emp. .30 1.62
Policies*Revenues -.38 -2.35*
Mission*Revenues .25 1.57
Stock purchase*Rev. .30 ' 1.47
Specialization*Revenues -.04 -.37
Mgt develop*Revenues .39 1.61
Professionalization*Rev. .08 .67
AR2 and AF .09 1.50
Total R2 .52
Adjusted R2 .34
N 128

 

 

*p < .05 **p< .01

 

166

 

Table 4.6 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Market Value (MVIRV)
Variables B t B t B t
Step 1: Controls
Service -.07 -.88 -.10 -1.14 -.03 -.31
Knowledge .16 1.94* .14 1.65 .16 1.81
CEO stock % .03 .56 .02 .32 .01 .08
Policies 96 .14 1.70 .27 2.74“ .25 2.33*
Mission 96 .02 .34 -.16 -1.03 -.23 -1.24
Stock purchase 96 -.11 -1.77 -.04 -.64 -.04 -.51
Specialization 96 .09 1.57 .07 1.10 .09 1.40
Management develop. 96 .14 2.23* .12 1.73 .10 1.39
Professionalization 96 .03 .51 .07 1.05 .03 .37
Employees 96 -.02 -.27 -.02 -.31 -.02 -.31
Revenues 96 -.05 -.77 -.01 -.15 -.04 -.44
Annual report 97 -.06 -1.05 -.08 -1.36 -.09 -1.48
Annual report 98 .12 1.90 .11 1.81 .13 2.03*
Shareholders’ equity .60 9.63" .60 8.06“ .56 6.75“
Industry average MV/RV .00 -.04 -.01 -.16 .00 .02
R2 and F .57 11.75"
Step 2: IVs
Policies As .12 1.86 .17 1.88
Mission As -.02 -.26 -.01 -.19
Stock purchase As -.02 -.27 -.04 -.45
Specialization 98 -.07 -1.03 -.05 -.64
Managpment develop As -.03 -.53 .00 .02
Professionalization 98 -.04 -.52 -.03 -.31
Employees 98 -.16 —1.69 -.18 -.95
Revenues 98 .19 1.11 .36 1.55
AR2 and AF .03 1.12
Step 3: Interactions .
Policies*Emponees .1 1 .88
Mission*Emponees .11 1.16
Stock purchase*Emp. -.23 -1.19
Specialization*Emponees -.05 - -.33
_Mgt develop*Emp|oyees -.04 -.59
Professionalization*Emp .16 1.20
Policies*Revenues -.15 -1.41
Mission*Revenues .10 .94
Stock purchase*Rev. .14 .83
Specialization*Revenues .01 .11
Mgt develop*Revenues .09 1.28
Professionalization*Rev. .06 .66
AR2 and AF .04 1.14
Total R2 .64
AdLusted R2 .53
N 150

 

 

*p< .05 **p < .01

 

167

 

Table 4.7 Outlier Values Deleted

 

 

Hypothesis Dependent Variable Values Deleted

2 ROA 75, 128. 193
Net Income (ROE) 162

3 ROA Same as H2
Net Income (ROE) Same as H2

6 ROA Same as H2
Net Income (ROE) Same as H2

 

168

 

Table 5.1 Summary of Results by Hypothesis

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Independent Dependent Variables Final step
Variables

Board Resources Policies and procedures NS

(Table 4.2) Strategic planning NA
Mission statement NS
Stock purchase plan NS
Management development NS
Specialization NS
Professionalization NS

High Growth ROA SS (p <. 05)

Transition Variables Specialization p < .05 Supports H

(Table 4.3) NI (ROE) NS
MV (PIE) NS
MV (MVIBV) NS

Board Resources ROA NS

(Table 44) NI (ROE) ss (p < .05)

Business Experts p < .05 Supports H

MV (PIE)
MV (MVIBV)

Board Resources Firm Performance NA

and
High Growth
Transition Variables

 

 

 

 

 

 

169

 

Table 5.1 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Independent Dependent Variables Final step
Variables
Interaction of Board Policies and procedures NS
Resources Strategic planning NA
and CEO Mission statement NS
Qualiﬁcations Stock purchase plan NS
(Table 4-5) Management development NA
Specialization NS
Professionalization SS (p < .01)
BE*CEO Gen Mgt Exp p < .01 No Support
SS*CEO Gen Mgt Exp p < .01 No Support
BE*CEO Industry Exp p < .01 Supports H '
Interaction of High ROA NS
Growth NI (ROE) NS
Transition Variables MV (PIE) NS
and Firm Size MV (MVIBV) NS

(Table 4.6)

 

 

 

 

 

170

 

FIGURES

171

.1 otooxm 30525 0..

 

 

A

A as. .928 821.1
as. .928 80 3 ii

 

.Illllll (I11 15. triti‘lilitl

monotooxm EoEomucus .2280 Omo can 389nm $335 Go 53022:. 2: co

oocotooxm EoEomacaE .8050 63333 ho 5.323”. 5. 2:9".

 

aoueyadxa tueweﬁeuew memo eApnoaxg

172

.: 80.38% .335 o._

 

 

62.19050 Own .1 + W
6.2 .9050 0.00 0.. IT

 

 

00:0...0oxm E05005...— .0._0:00 Omo 0:0 30.3000»). .5095 5 5:00.00... 05 5
00:0...0oxm “5:550:05. .0050 035003 .0 5.00050”. «6 0.52”.

 

ID V ('0 N 1-
aoualladxg tuewaﬁeuew .laolyo eAnnoaxa

(O

[x

173

=.. 0:25 88.30 o._

 

 

 

 

 

 

 

 

o
em. bases 80 .1 IT - no
.90. bases 000 o._ if . s
. m;
a...
. 0N
. m
. 0.0
r v
1 00
0

005.596 505:. 0m0 0:0 3.09am 00050.6 5 5.00205. 05 :o
00:0...0axm «5809055. .0050 0>zao0xm 5 5.00050”. 0.0. 0.50.".

eoueyedxg tuewabeuew .leojgo eAnnoexa

174

APPENDICES

175

Appendix A

Variable List with Data Sources and Time of Collection

 

VARIABLES

l

DATA SOURCE

I

COLLECTION DATE

 

Independent Variables

 

Board Composition
. Business Experts
0 Support Specialists

Prospectus (control)

Proxy statement

1990

1997

 

CEO Managerial
Qualiﬁcations

0 Education level

a General Mgt. Exp.
0 Functional Exp.

0 Industry Exp.

Prospectus

1 996

 

Growth Rate
a No. of Employees
0 Revenues

 

10K

 

1998

 

Dependent Variables

 

Professionalization of

Management

0 Prior management
experience of the
Executive Ofﬁcers

0 Management
development program

Prospectus (control)

10K, Proxy

10K, AR

1 996
1998

1998

 

Specialization
' 0 Areas of responsibility
of the TMT

Prospectus or AR
(control)

10Ks and ARs

1 996

1 998

 

 

Shared Goals

0 Mission statement

0 Communication

0 Employee bonus

0 Employee stock
option plan

0 Employee stock
purchase plan

 

Prospectus (control)

10Ks and ARs

 

1 996

1997,1998

 

176

 

Appendix A (cont’d)

 

Standardization and

Fonnalization

- Policies and
procedures
Employee training
Performance
measurement &
compensation plans
(Management and
Employee)

0 Financial reporting &
budgetary controls

0 Strategic planning

Prospectus (control)

10Ks and Ars

1 996

1997,1998

 

Firm Performance
0 RCMX
ROE
P/E ratio
MV/RV

 

10K
Compustat
Primark

 

1999

 

Control Variables

 

Industry Type

Prospectus

1 996

 

Firm Size
- No. of employees
0 Revenues

Prospectus
1 0K

1 996
1 996

 

Average Industry
Performance

0 ROA

o ROE

- PIE Ratio

0 MV/RV

Ibbotson’s

1 999

 

Board Controls
0 Size
0 Tenure

Prospectus
Proxy statement

1 996
1 997

 

 

CEO Stock Ownership
0 % of shares owned by
CEO following I PO

 

Prospectus

 

1 996

 

177

 

Appendix B
Data Collection and Coding Sheet

Firm No. Data Recorder Name

 

 

Date

96 Prosp_ 97 10k_ 97 AR_ 98 Proxy_ 99 10k_ 93 AR_ 99 Proxy—

Section I Firm Information

96 Prosp Firm Name

 

 

Ind. Code Ticker Prosp Date

 

 

 

 

96 Prosp Use of Proceeds: Growth?

96 Prosp, 10ks Size: No. Ees, Annual Revenues

lncorp Date

 

 

KW: employee, revenues

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

96 Ees 97 Ees 98 Ees 99 Ees
FT_ FT FT_ FT _
IPT_ IPT: |PT__ 1PT—
k‘tev Rev lRev hRev
Section II CEO Managerial Qualiﬁcations 96 Prosp
Name Dates: Joined Co.__ In Position—
a) Highest Education Level
HCode
b) Functional Exp. (list areas)
Fade
c) Industry Exp. (no. of yrs in same industry)
d) General Management Exp: Fade
Position Company Yrs.
Position Company Yrs.—
Position Company Yrs.__

 

 

178

Appendix B (cont’d)

 

 

 

p. 2 Recorder Firm
Section III Professional Management KW: Ofﬁcer Code

 

 

 

 

96 Prosp a) 1996 Corporate Experience of Executive Ofﬁcers (excl. CEO)

 

 

 

 

 

 

 

Name Title Yr Co. Yr Pos. Yrs
1.
2.
3.
4.
5.
6.
7.
8.
Code b) 1996 Management Development: Yes No

KW: management development, train, seminar, recruit

98 10k c) 1998 Corp Exp of Executive Ofﬁcers (excl. CEO) Code
or 99 -

 

 

 

Name Title Yr Co. Yr Pos. Yrs
Same as above list nos. :

 

 

 

Code d) Additional Management Development: Yes No

 

97 10k 97 AR 98 10k 98 AR (note source)

 

 

179

Appendix B (cont’d)
p. 3 Recorder Firm

 

Section IV Standardization and Formalization
96 Prosp KW: infrastructure

a) 1996 Formal Policies and Procedures

 

 

Code KW: polic, procedure, standard, manual, train

 

 

b) 1996 Performance Measurement/Comgnsation Plans

 

 

Code KW: performance, appraisal, review, personnel, compensation

 

 

c) 1996 Financial Reporting/BudgetIControls

 

 

Code KW: (information) system, budget, control

 

 

d) 1996 Strategic Plan
Code KW: strateg

 

 

 

 

180

p.4

Appendix B (cont’d)
Recorder Firm

 

Section IV Standardization and Formalization cont’d

KW: infrastructure 97 10K 97 AR 98 10K 98 AR (note source)

 

 

Code

 

 

ode

 

 

 

Code

 

 

 

 

Code

 

 

a) Additional Formal Policies and Procedures
KW: polic, procedure, standard, manual, train

b) Additional Performance Measurement/Compensation Plans

KW: performance, appraisal, review, personnel, compensation

c) Additional Financial Reporting/BudgetIControls
KW: (information) system, budget, control

d) Additional Strategic Plan
KW: strateg

181

p. 5
Section V

96 Prosp

 

 

Code

 

 

 

 

Code

 

 

 

 

Code

 

 

 

Code

 

 

 

Code

 

 

 

 

Code

 

 

Appendix B (cont’d)
Recorder Firm

 

Shared Goals

a) 1996 Mission Statement
KW: mission, vision (check “whole word only” on search)

b) 1996 Communication Tools
KW: newsletter, e-mail, intranet, communication

c) 1996 Risk Sharing Plans (Not incl. ofﬁcers/directors)
KW: 401, stock option, stock purchase, incentive,
commissions, bonus

97 10k 97 AR 98 10k 98 AR (Note source)

a) Additional Mission Statement

b) Additional Communication Tools

c) Additional Risk Sharing Plans

182

Appendix B (cont’d)
p. 6 Recorder Firm

Section VI Specialization KW: team, management, staff

 

 

 

 

 

  
  

 

 

 

 

 

 

 

 

 

 

 

 

96 AR a) 1996 TMT Functional Areas (Not incl. CEO, COO, Pres.)
ode ‘ Name Title
1. ‘
2.
3.
4.
- 5.
lCode 6.
Total 7.
8.
9.
10.
Code b) 1996 Staff ists
Name
98 AR c) 1998 TMT Functional Areas incl. CEO COO Pres.
Code Name Title
1.
2.
3.
4.
5.
Code 5-
Total 7-
8.
9.
10.
Code d) 1998 Staff

Name

 

 

 

183

Appendix B (cont’d)
p. 7 Recorder Firm

Section VII 1999 Firm Performance

 

Performance
ROA
ROE

PIE Ratio
MVIBV

 

 

 

Board of Directors 96 Prosp 97 Proxy

Name nd

 

 

 

 

 

 

 

 

 

Totals: BE SS IN COM 0TH
Act Capital Exp: 96 97 98 99
CEO Stock Ownership Post IPO: CEO #1 . CEO #2

 

 

184

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.00:.0.:.0E 0.

E9090 9.50 3:000 .5000

0>.0:0:90E00 <.. .0 .00::<

.00500 0:.:.0.. x00..s.~ 9.50 00.00 v.0.

0 0.0.0E00 0. 09.009 .0059 .0.00:0.0 mg
0.0 00>..0.:000.09 o .l. .:00:< 0500090 .0..00 00 00.000090 00.000090
00.200 0.:0.0:E0.2.. .. n .:0090 .0500 :9000 0.0.... 00.. 00040.0 0:0 00.0..00
5000:0500 0:0 :o..0~.0.00:0.m

o u :0..0~._0_000.w

05.9.5.2 .o 3.8.5 000 :89.

00.0w .0... .0> 0:0 000 0:. 0:.00.0x0 .00::<
0:0..0.000 .0> .E00. .:0E000:0E 00 v.0. E00.
.5050 .00090 .0> 00. 0:. :0 00.50909 . a2 .:0E000:0E 00.
00.00 .0:0..0:.0.:. .0>m 00.:0:0..0_9 0:...000. 0.00EOE E00. 00 00.000090 0:. .0 >....0.0:0009
.00....0 .0.0:0:.0 .0.:0.. .0 .00E0: .0.0.r .:0E000:0E 00. .0 .0... _0..|:00. .0 009...
:o..0~..0.00mm

...:0E000:0E E9090 .:0E00.0>00

3:000 0:0 .0:..00c000 .:0E000:0E 05005 00000

055000 :0 :0 .0 00:0.0.x0 0:. .00::<

.0::00.00 b00.>.0000 .0 00:00.5 00.500. ..0.09 ..0:.E00 00.00 .8.
0:0 .:0E000:0E 0.0.. ..:0E00.0>00 No.2 0E0.00.0
0. 000.50 0 0 E0090. .:0E000:0E 00 00.000090 .:0E00_0>00
0. 0:.:.0.. 0:.00:0.. F 0 .:0090 .0500 :9000 0.03 >3. a .:0E000:0.2

0... N 0 005.090 ms. .E... 00. 0:. 00 06.0
.> >:00E00 .0 0550.. 0. 5:0 0:0...000 m0 .0 v.0. 0.00.0.0 02.0090
22.9000 .0 5.09.0 .0..000:0E :. 09.5.5 >0.>. 0:. .0 005.090
00 00200 x .5. €00. 0.0050 02.0090 0.0050 02.0090 0:. :0 00 00.000090 .:0E000:0E
:00 0. 000. .05. E9“... 0:. 0.00.. .0 .0: 0>0 0: .. :0..0E.0.:. .00.:00.00.m 61.09.10. .20 .0 0.00>
5000.05.05.90

:0000:=:00.
0.0E0xw 0.5...0 50805000.: .0 00502 0.00> 0.:0E0000 0.:0..0>

 

 

00.:0..0> 05.00.300.55 550:0; 530.0 :0... .o :o..00..00 0.00 0 x.0:000<

185

 

0.5.00 5.5 ..0.0me

 

 

 

 

 

 

.50: 0. >00.0::00. 0:05 .5000
0:000 .:0..00...0.0 £390.. 0 >.0E.0 .0: .005...
>0. 0:0 .200 0:090 H00.:000 0: 0. 009: 00.3 v.0.
050.0000. >00.0::00. 0: 0. 00: :0.0 0.00.0..0. g4.
:.0.:.0E.. 00 :000 000.0 0 u 500:... 00.9.0 00 00.000090
00020:. :0.0 0.00.0..0 . n .:0090 .0500 :9000 05.5 >3. 3 0:.::0.0 0.00.0..0
.0500. 0000000 .00.:000
.00:0E.0..00 0.0.0 00 0. 0:...0E 5.0.000 .5000
>0 0.0.0 0:00.. 0:2 0:0 .0905. 0. >.000 0. .00::<
05500000 000.9500. 00: 0E0.0>0 0:...009. 00.00 v.0.
.o ..0.0 0E0.0>0 .9500 ..00000 3.». 0.00000
.:0E000:0E 0:0 0 n 500:0. .E0.0>0 .:0..0E.0.:.. 00 00.000090 0:0 0:..5000
..0.0:0:_. ..0:0..0.000.. . u 5000.0 .0:.00 :9000 0.02. >3. .0003 .0.0:0:.0
...:0E000:0E 5.:00 .00.:000 0: 0. .0>0. 090000E
.0. .000 00>..:00:. .:0E000:0E .0500 .5000 0.0.0000 N.
0000005050000. 30.0: 000>0.0E0 0. 0:0.00.EE00 .00::< 0E0.090
0:0 ..00:0E.0..00 >.000 0. 00: 0E0.090. .:0..00:00E00 00.0 v.0. :0..00:00E00
>:00E00 0:0 .000.>.0:. . ..0::00.00 .3039 >|0§a 0:0 05059.00
:0 0000: 000>0.0E0 o u .:000< ..00.0.000 05050.50 00 00.000090 .:0E000:0E
0. 0.00 00005: :000. . u 50090 0:00 :9000 0.02, >3. 30w 0:0 00>0_0Em
.0090 50.00.0500.
>9. :. 000>0.0E0 .0 .5000
0000 0000.305. 0:00x0 .00::<
0. 00.00050 0. 0:.:.0.. 00.3 v.0.
020.080.. .0 0:.:.0.. . >|02
00N.E0.000 02000. 0 u .:000< :.0.. 00 00.000090
..0.0 00.00 .0905... P u 50090 0:00 :9000 0.03 >3. .:9.:|00 0:.:.0.. 00>0_0Em.
0500 5.550550 0:0 :0..0~.0.00:0.0
0030000000.
0.0590 0.550 50505000.... .0 00:52 0.00> 05050000 0.:0..0>

 

 

 

 

 

 

 

.058. 0 0080.2

186

 

.2205

m 7.3. am 3388 05
£3, 53 gas on; new
x33 .3 2:0... ompmmm.
“a £95 oz; $28.95
$3.05 cm_n_ 3205.”.

89:38 3 2 _m>m_

AmmSmmoE
288mm my

53 03:23
x87. 028.35

 

 

 

 

 

 

 

 

 

 

xon 83?. ..o EmEommcmE 3E3 roam.
..mmmioEEm __m 33.05 26.8 93.8.an 2 .macc< cma cozao
55 530 xooaw 2an 2 van 25693 3:03 8.3 v.2. xooam 8.8.95.
83?. .o ..m¢m>o.aEm 622.85 .3223 g
92» .2 Egg >3 0 n E394 xoBm .cozao xooam 8 38835 cma
383383 93:00:... F u Emmen. 35m: 550m 295 >3. 38d mason 333.5.
@365me toamm
..mm:_._nmamo =9: .mEExm 53> “o: .macc<
.0 3:63.05 .umfozzmc {6888 m5 £53, 8 55:5 8.3 x9
9m 303930 03388 2 am: cozmgcsEEoov .__mE-m..._mzm_m>>mc g
ucm .mcoszEE o u Emmn< 52625858 8 38383
m.>cmano 9: __<.. P u E391 35m: 583 995 >3. glad cosmoEaEEoo
..cEmQEoo
53 $85 2
m_ mmmcm>==mano .65
EEEmE 9 cozao m_nm_>
>_co £58063 9: “m5 toqmm
35:09:00 om 022mm .m:cc<
new 6335 a 35255 8.3 x3
3 momaumfme g
5362:: m5 0 u Emmn< .563 628.8 8 38833
3 329030.. 6065.. P u Swami 35m: 553 203 >3. god. .coEmuﬂm co_mm__2
230 .9225
cozuuczcou_
29:53 33:5 22:23.32 to 3522 23> 3:25.30 o_na_..m>

 

 

6.23. o 59.0%?

187

Appendix D
Executive Survey

Date
Inside address

Dear:

I am a professor performing research in the area of Strategic

_ Management in the Neeley School of Business at Texas Christian University.
Over the last few years, I have been examining the organizational structure in ‘
companies that have recently had their initial public offerings of stock. [Name of
company] is one of the companies included in my research.

I have a very brief questionnaire (11 questions) regarding the
infrastructure your ﬁrm had in place at the time of your company’s IPO in 1996
and then again at the end of 1998.

Since you have been with [Name of company] during that time, I would
greatly appreciate your taking 5 minutes to complete the [attached or enclosed]
questionnaire and return it to me. Your name and your company’s identity will
not be revealed in anyway — I am only interested in generalized ﬁndings across a
large group of companies.

If there is someone in top or middle management who has been with your
company since its IPO and could respond to the questions accurately, I would
appreciate your forwarding the questionnaire to that individual. It will only take
about 5 minutes of his/her time to complete. Neither the name of the person
responding nor the name of your company will be revealed in my results — I am
only interested in generalized ﬁndings across a large group of companies.

Your company’s responses to my questionnaire are a critical part of my
research ﬁndings. I greatly appreciate your assistance.

Very truly yours,
Rebecca A. Luce
Assistant Professor of Management

M.J. Neeley School of Business
Texas Christian University

188

Appendix D (cont’d)

Your position/title:

Date you started with the company:
Instructions: For each of the areas of infrastructure below, please indicate to
what degree it was in place at two different points in time: when your company
had its IPO in 1996 and at the end of 1998. Use the following scale to rate the
level of implementation for each infrastructure area:

 

 

 

 

Present = 1 Absent = 0
IPO IN END OF
AREA OF INFRASTRUCTURE 1996 1998

 

—I

. Written policies and procedures

 

N

. Training programs for employees

 

3.

Guidelines to set performance objectives, evaluate
performance, and assign compensation for executives

 

Guidelines to set performance objectives, evaluate
performance, and assign compensation for employees

 

Information systems to produce ﬁnancial reports used for

managerial decision making

 

Strategic planning process

 

Written mission statement

 

Communication programs for getting information about
the company to employees

 

Stock option program that includes employees below the
execuﬁvelevel

 

10.

Cash bonus program based that includes
employees below the executive level

 

 

11.

Stock purchase plan that includes employees
below the executive level

 

 

 

Please FORWARD your completed on-Iine survey via e-mail to:
r.luce@cu.edu. (The document will not be sent if you use the Reply

button). Thanks so much for your participation. It will be very helpful for
my research.

189

 

 

MICFIEIIEAINASTATE Of a We“ ”Wing Human Subjects

UNIVERSITY

 

July 13, 2000

T0: Gerry MCNAMARA
N 475 N. Business Complex

RE: iRB# 00-206 CATEGORYzi-C
APPROVAL DATE: July 13, 2000

TITLE: INFLUENCE OF THE BOARD OF DIRECTORS ON HIGH GROWTH FIRMS
AT A STRATEGIC DECISION POINT

The University Committee on Research Involving Human Subjects' (UCRIHS) review of this
project is complete and I am pleased to advise that the rights and welfare of the human
subjects appear to-be adequately protected and methods to obtain informed consent are
appropriate. Therefore. the UCRIHS approved this project.

RENEWALS: UCRIHS approval ls valid for one calendar year. beginning with the approval
date shown above. Projects continuing beyond one year must be renewed with the green
renewal form. A maximum of four such expedited renewals possible. Investigators wishing to
continue a project beyond that time need to submit it again for a complete review.

REVISIONS: UCRIHS must review any changes in procedures Involving human subjects, prior
to Initiation of the change. If this is done at the time of renewal, please use the green renewal
form. To revise an approved protocol at any other time during the year, send your written
request to the UCRIHS Chair, requesting revised approval and referencing the project's lRBii
and title. Include in your request a description of the change and any revised hstrumenis.
consent forms or advertisements that are applicable.

PROBLEMS/CHANGES: Should either of the following arise during the course of the work.
notify UCRIHS promptly: 1) problems (unexpected side effects. complaints, etc.) Involving
human subjects or 2) changes in the research environment or new information indicating
greater risk to the human subjects than existed when the protocol was previously reviewed and

 

approved. '
HE mfg: If we can be of further assistance. please , contact us at 517 3552180 or via email:
UCRIHS@pilotmsu.edu. Please note that all UCRIHS forms are located on the web:
““0 http:Ilwww.msu-.eduluser/UCRIHSI
GRADUATE

STUDII Sincerely,

Michigan State Unlversity .
245 Adminishlion Building Ashlr Kumar, MD.
Elsi Lanslng, Michigan ,
40824-1046 AK' m
517,355,215), cc: Rebecca Luce _
FAX: 517/353—2976 N475 North Busmess Complex
immmmmusci/iicrllls

E-Mail: umIBOrhSUMU

The Michigan Silt: Murry

W I: mailman W
Emile/mm Action 190

MSU I: a Mum.

MW" mam

Appendix E (cont’d)

. APPLICATION FOR INITIAL REVIEW
(and 5 yr. renewal)

APPROVAL OF A PROJECT
INVOLVING HUMAN SUBJECTS

University Committee on Research Involving Human Subjects (UCRIHS)
David E Wright, Ph. 0., cm
248 Administration Building, Michigan State University
East Lansing, MI 48824-1048
PHONE (517) 355-2100 FAX (517) 353-2970
E-Mail - UCRIHsgmsuedu
WEB SITE - http:/Mww.rnsu.eduhtsarluaihal

Office Hours: M-F (8:00 Ara-Noon 8. 1:00-5:00 PM.)

 

DIRECTIONS: Please complete the questions on this application using the instructions and
definitions found on the attached sheets.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

REQUIRED IF APPLICABLE

1. Rggmnsigle PLgieg Invgﬁiggtgr; 2. Wm
(MSU Faculty or staff supervisor) (”Students M413 Provide Student IN")
Name: Gerry McNamara Name: Rebecga Luce . _
Social Security #2 . Student ID#: or $81! A256—231-37
Department Management Department: Management
College: Business College: Business
Mailing N475 North Business Complex Mailing N475 North Business Complex
Address: East Lansing, MI 48824 Address: East Lansing, MI 48824
Phone 517-353-4322 Phone: 517-353-3046
Fax: 517-432-11 1_1 Fax: 517-432-1311
Email: mcnamatﬁgmsuedu Email: lucerebegmsuedu
Ieaeeptrespensrwiryrereonductingmepmpoud Additionallnvestigatorlnforrnation
researchinaccordancewiththeprotectionsofhuman 3. Name:
subjects as specified by UCRIHS. including the Student ID# or SS#
supervision of faculty end student co-invesligators. 4. Name:

' ‘ Student ID#: or $83!
SIGN HERE: 5. Name:

. Student ID#: or S_S#

Note: Without signature, application can Moe processed

 

{.17-w 52:3!” «1611’s

‘ow5el' ”Ta "ﬁ‘g‘: Wimp“ "'
UCRIHS Correspondenceim Copies of correspondence wiil be sent to the primary and secondary investigators
only. If you would like additional investigators to receive correspondence, please provide further address hformation

.'_, “ﬂ . . ., 7
1..::e.:(_ 1k: .‘-.
.-‘.~s‘ gin-r .. AL

v»,

 

 

 

 

 

 

on a separate page.

6. TItle of Project; Influence of the Board of Directors on High Growth Firms at a Strategic
Decision Point

Form Revised 10000 (1)

FOR OFFICE USE ONLY

 

Subcommittee

 

191

Agenda ‘ -

 

10.

11.

12.

13.

14.

A d ‘ '
Have you ever received Prelinﬁﬁgrrly Jppbiosgffgi' this project?
No [ x] Yes [ ]
If yes, what IRB # was assigned to it?

 

Funding (if any)
if applicable. MSU Contracts and Grants app. and / or acct. #

 

 

Has this protocol been submitted to the FDA or are there plans to submit it to the FDA?
No [x] Yes [ ]
If yes, is there an IND #? No [ ] Yes [ ] IND #

 

Does this project involve the use of Materials of Human Origin (e.g., human blood or
tissue)? No [x] Yes [ ]

When would you prefer to begin data collection? asap

 

Please remember you may not begin data collection without UCTRIHS approval.

Category (Circle a,b. or 0 below and specify category for a and b. See instructions pp.
4—7) -

a. This proposal is submitted as EXEMPT from full review.
Specify category or categories: 1C

b. This proposal is submitted for EXPEDITWeview.
Specify category or categories:

c. This proposal is submitted for FULL sub-committee review.

 

 

 

Is this a Public Health Service funded, full review, multi-site project for which MSU is the
lead institution? No [x ] Yes [ ]
If yes. do the other sites have a Multiple Project Assurance IRB that will also review this
project? '
[ ] No. Please contact the UCRIHS ofﬁce for further information about meeting
the PHSINlH/OPRR regulations.
[ ] Yes. Please supply a copy of that approval letter when obtained.

Project Description (Abstract): Please limit your response to 200 words.

I am collecting information in a questionnaire format from a company representative regarding
the degree to which programs, policies. and procedures were in place within their organizations
at two different times: (1) when the ﬁrm went public and (2) at the end of the ﬁrm‘s 1978 fiscal
year. The questionnaire consists of a total of eight questions.

19 2

.Appendix E (conb'd) .
15. Procedures: Please descrIbe all preject actIVItIes to be used In collecting data from

human subjects. This also includes procedures for collecting materials of human origin
and analysis of existing data originally collected from human subjects
Respondents will be asked eight general questions requiring either a "yes" or "no" response
(3 questions) or a rating on a scale of 1 to 5 (from "Not at all" to "To a large extent"). Questions
cover topics ranging from the existence of management development programs and a mission
statement to standardization of ﬁnancial reporting. Questions may be asked by the investigator
over the phone or a questionnaire may be mailed to the respondent for completion.

16. Subject Population: Describe your subject population. (e.g., high school athletes.
women over 50 wlbreast cancer. small business owners )
An employee of a company that went public in 1996.

a. The study population may include (check all categories where subjects may be
included by design or incidentally):

Minors

Pregnant Women

Women of Childbearing Age

Institutionalized Persons

Students

Low Income Persons

Minorities

Incompetent Persons (or those
with diminished capacity) [ ]

HHHHHHH
X X ﬁ—‘H X X 0.5
Ha Ha

b. Number of subjects (including controls) 30

 

c. How will the subjects be recruited? (Attach appropriate number of copies of
recruiting advertisement, if any. See p. 13 of UCRIHS instructions)
Firms will be contacted and an appropriate respondent identiﬁed based on the information
requested by the investigator.

193

Appendix E (cont'd)
d. If you are associated with the subjects (e.g., they are your students, employees,
patients), please explain the nature of the association.
No association

e. If someone will receive payment for recruiting the subjects please explain the
amount of payment, who pays it and who receives it.
No payments will be made.

f. WIII the research subjects be compensated? [x] N0[ ] Yes.
If yes, details concerning payment, including the amount and schedule of
payments, must be explained in the informed consent.

9. W" the subjects incur additional ﬁnancial costs as a result of their participation in
this study? [x] No [ ]Yes. If yes, please include an explanation in the
informed consent.

h. VWI this research be conducted with subjects who reside in another country or
live in a cultural context different from mainstream US society? [x ] No [ ]
Yes.

(1) If yes, will there be any corresponding complications in your ability to
minimize risks to subjects, maintain their conﬁdentiality and/or assure
their right to voluntary informed consent as individuals?

[ ] No [ ] Yes.

(2) If your answer to h-1 is yes, what are these complications and how will
you resolve them?

17. How will the subjects' privacy be protected? (See Instructions p. 8-9.)
Only three items of information will be recorded about respondents:

1. the name of the employing company;

2. the title of the position of the respondent;
3. the respondent's hire date with the company.

194

A d1 E t' d
- 18. Risks and Beneﬁts for subjectgnglee fnstrtfcczztibnns p.) 9.)

A summary of the results of my study (dissertation) will be sent to the responding companies.
This information may be valuable to the management group of the company. There are no
risks for respondents.

19. Consent Procedures (See Instructions pp. 9-13.) '

Consent will be incorporated into the introductory message for the questionnaire. If the
respondent chooses to complete a mailed questionnaire, slhe will be indicating consent

by completing and returning the questionnaire. Where the respondent agrees to answer the
survey questions orally (by phone), slhe will be read an introductory message, including the
statement that answering the questions asked indicates consent.

 

 

CHECKLIST: Check off that you have included each of these items. If not applicable, state
NIA:

[ ] Completed and signed application .

[ j The correct number of copies of the application and instruments (e.g., surveys, interview
questions, questionnaires, etc.) according to\the category of review (See instructions pp. 14)

[ ] Consent form (or script for verbal consent), if applicable
[ na ] Advertisement, if applicable

[ ] One complete copy of the methods chapter of the research proposal (if available)

 

 

REFERENCES

196

REFERENCES

Abelson, R. & Jacob, R. 1989. The biggest blowout ever. Fortune, Apr. 24:
346-355.

Aiken, L. S. 8. West, 8. G. 1991. Multiple regression: Testing and interpreting
interactions. Thousand Oaks, CA: Sage.

Barney, J. B. 1997. Gaining and sustaining competitive advantage. Reading,
MA: Addison-Wesley.

Barnhart, S. W., Marr, M. W., 8. Rosenstein, S. 1994. Firm performance and
board composition: Some new evidence. Marggerial anrﬂecision Economicg,
15: 329-340.

Baron, R. M. 8. Kenny, D. A. 1986. The moderator-mediator variable distinction
in social psychological research: Conceptual, strategic, and statistical
considerations. Journal of Personality and Social Psychology, 51: 1173-1182.

Barry, C. B., Muscarella, C. J., Peavy, J. W. III, Vetsuypens, M. R. 1990. The
role of venture capital in the creation of public companies: The evidence from the
going-public process. Journal of Financial Economics, 27: 447-471.

Baum, J. R., Locke, E. A., 8. Smith, KG. 2001. A multidimensional model of
new venture growth. Academy of Management Journal, 44: 292-303.

Baysinger, B. D. 8. Butler, H. N. 1985. Corporate governance and the board-of
directors: Performance effects of changes in board composition. ﬂimal of Law.
Economics, and Organizations, 1: 101-124.

Baysinger, B. D., Kosnik, R. D., 8. Turk, T. A. 1991. Effects of board and
ownership structure on corporate r 8 d strategy. Academy of Management
Journal, 34: 205-214.

Baysinger, B. D. 8. Zardkoohi, A. 1986. Technology, residual claimants, and
corporate control. ﬂirnal of Law. Economics. and Organizations, 2: 339-349.

Belsley, D. A., Kuh, E., 8. Welsch, R. E. 1980. Regression diagnostics:
Identifyinq inﬂuential data and sogrces of collinearitv. New York: Wiley.

Bentson, G. 1982. Accounting numbers and economic values. Antitrust
Bulletin, Spring: 161-215.

Berle, A. A., Jr. 8. Means, G. C. 1932. The modern corporation and private
progem. New York: MacMillan.

197

Blau, P. M. 8. Schoenherr, R. A. 1971. The structure of organizations. New
York: Basic.

Blowers. S. C., Grifﬁth, P. H., Milan, T. L. 1999. The Ernst 8. Young LLP guide to
mlPO value iourrm. New York: Wiley.

Boeker, W. 8. Goodstein, J. 1993. Performance and successor choice: The
moderating effects of governance and ownership. Academy of Management
Journal, 36: 172-186. ’

Boulding, K. E. 1953. Toward a general theory of growth. Cana_dian JOLLI‘IBI of
Economics and Polﬂcal Science, 19: 326-340.

Boyd, 8. 1990. Corporate linkages and organizational environment: Atest of
the resource dependence model. Strategic Management Jogrn_a|, 11: 419-430.
Brealey, R. A. 8. Myers, 8. C. 1996. Princijles of corporate ﬁngnce. New York:
McGraw-Hill.

Brush, C. G. 8. Chaganti, R. 1998. Business without glamour? An analysis of
resources on performance by size and age in small service and retail ﬁrms.
Journal of Business Venturing, 14: 233-257.

Burns, T. 8. Stalker, G. M. 1961. The management of innovatiﬂ. New York:
Oxford University Press.

Byme, J. A., Grover, R., 8. Melcher, R. A. 1997. The best and worst boards.
Egsiness Week, Dec. 8: 90-99.

Caplow, T. 1964. Principles of organization. New York: Harcourt, Brace, 8.
World.

Castaldi, R. & Wortman, M S. Jr. 1984. Boards of directors in small
corporations: An untapped resource. American Journal of Small Business, 9(2):
. 1-10.

Certo, S. T., Daily, C. M., Dalton, D. R. 2001. Signaling ﬁrm value through
board structure: An investigation of initial public offerings. Entreprenegrship:
Theory 8. Practice, 26: 33-50.

Chaganti, R. S., Mahajan, V., 8. Sharma, 8. 1985. Corporate board size,
composition and corporate failures in retailing industry. Journal of Management
Studies, 22: 400-417.

Chakravarthy, B. S. 1986. Measuring strategic performance. Strategic
Management Journal, 7: 437-458.

198

Chandler, A. D. Jr. 1962. Strateﬂand structure. Cambridge, MA: M. I. T.
Press.

Charan, R. 1998. Boards at work: How corporate ppards creagte competitive
advantage. San Francisco: Jossey-Bass.

Charan, R., Hofer, C. W. 8. Mahon, J. F. 1980. From entrepreneurial to
professional management: A set of guidelines. Journal of Small Bgsinesg
Management, 18 (1): 1-10.

Churchill, N. C. 8. Lewis, V. L. 1983. The ﬁve stages of small business growth.
Harvard Business Review, 61(3): 30-50.

Cochran, P. L., Wood, R. A., 8. Jones, T. B. 1985. The composition of the board
of directors and incidence of golden parachutes. Academy of Maﬂgement
Journal, 28: 664-671.

Cohen, J. 1960. Acoefﬁcient of agreement for nominal scales. Educational
and Psychological Measgrement, 20: 37-46.

Cohen, J. 8. Cohen, P. 1983. Applied multiple regression/correlation analysis for
me behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum.

Daily, C. M. 1995. The relationship between board composition and leadership
structure and bankruptcy reorganization outcomes. Jngal of Management. 21:
1041-1056. .

Daily, C. M. 8. Dalton, D. R. 1992. The relationship between governance
structure and corporate performance in entrepreneurial ﬁrms. Journal of
Business Venturing, 7: 375-386. '

Daily, D. M. & Dalton, D. R. 1993. Board of directors leadership and structure:
Control and performance implications. Entreprengirshig Theory and Practice,
17: 65-81.

Daily, C. M. 8. Dalton, D. R. 1994. Bankruptcy and corporate governance: The
impact of board composition and structure. Academy of Manage_ment Jogrnal.
37: 1603-1617..

Daily, C. M., Johnson, J. L., 8. Dalton, D. R. 1999. On the measurements of
board composition: Poor consistency and a serious mismatch of theory and
operationalization. Decision Sciences. 30: 83-106.

Dalton, D. R., Daily, C. M., Ellstrand, A. E., & Johnson, J. L. 1998. Meta-

analytic reviews of board composition, leadership- structure, and ﬁnancial
performance. Strategic Management Journal, 19: 269-290.

199

DeCanio, S. J., Dibble, C., Amir—Ateﬁ, K. 2000. The importance of
organizational structure for the adoption of innovations. Management Science.
46: 1285-1299.

Dwyer, H. E. Jr. 1970. Organization for growth. In I. Stemp(Ed.). Cogporate
growth strategies: 96-130. New York: American Management Association.

Dyer, W. G. Jr. 1989. Integrating professional management into a family owned
business. Family Business Review, 2: 221-235.

Eisenhardt, K. M. 8. Schoonhoven, C. B. 1990. Organizational growth: Linking
founding team, strategy, environment, and growth among U. S. semiconductor
ventures, 1978-1988. Administrative Science Quarterly, 35: 504-529.

Fama, E. F. 8. Jensen, M. C. 1983. Separation of ownership and control. '
J_o_grnal of Law and Economics, 26: 301-325.

Ferry, R. M. 1999. Boardrooms yesterday, today, and tomorrow. Chief
Executive, March: 44-47.

Fiegener, M. K., Brown, B. M., Dreux, D. R., Dennis, W. J. 2000. The adoption
of outside boards by small private US ﬁrms. moorenegrship 8. Regignal
Development, 12:291-309.

Finkelstein, S. 1992. Power in top management teams: Dimensions,
measurement, and validation. Academy of Management Joym_al, 35: 505-538.
Finkelstein, S. 8. Hambrick, D. C. 1996; Sgategic IeadershipgTop executives
and their effects on organizations. St. Paul, MN: West.

Flamholtz, E. G. 1990. Growing pains: How to make the transition from an
entrepreneprship to a professionally managed ﬁrm. San Francisco: Jossey-
Bass.

Flamholtz, E. & Randle, Y. 1987. How to avoid choking on growth.
Management Review, 76 (5): 25-29.

Fleiss, J. 1981. Statistical methods for rates and proportigns. New York: Wiley.

Frederickson, J. W., Hambrick, D. C. 8. Baumrin, S. 1988. A model of CEO
dismissal. Academy of Management Review. 13: 255-270.

Fried, V. H., Bruton, G. D., & Hisrich, R. 1998. Strategy and the board of
directors in venture capital-backed ﬁrms. Journal ofBgsiness Venturing, 13:
493-503.

200

Galbraith,J. 1973. Designing complex giganizations. Reading, MA: Addison-
Wesley.

Gales. L. M. 8. Kesner, I. F. 1994. An analysis of board of director size and
composition in bankrupt organizations. Journal of Bgsings Resem, 30: 271-
282.

Gilley, K. M., Ford, R. H.. & Coombs, J. E. 1999. Director strategic contribution
and organizational performance in entrepreneurial ﬁrms. Academy of
Management Best Paper Proceedings.

Golden, B. R. 8. Zajac, E. J. 2001. When will boards inﬂuence strategy?
Inclination x power = strategic change. Stratagic Management Journal, 22:
1087-111.

Gorman, M. 8. Sahlman, W. A. 1989. What do venture capitalists do? Journal
of Bgsiness Venturing, 4: 231-248.

Greiner, L. E. 1972. Evolution and revolution as organizations grow. Harvard
ﬂsiness Review. 50(4): 37-46.

Haire, M. 1959. Biological models and empirical histories of the growth of
organizations. In M. Haire (Ed.). Modern organization theogy: 272-306. New
York: Wiley.

Hall, R. H. 1972. Organizations: Structpre angprocess. Englewood Cliffs, NJ:
Prentice-Hall.

Hambrick, D. C. 8. Crozier, L. M. 1985. Stumblers and stars in the management
of rapid growth. Journal oLB_usiness Venturing, 1: 31-45.

Hambrick, D. C. 8. Mason, P. A. 1984. Upper echelons: The organization as a
reﬂection of its top managers. Academy of Management Review, 16: 193-206.

Hanks, S. H., Watson, C. J., Jansen, E., 8. Chandler, G. N. 1993. Tightening
the life-cycle construct: A taxonomic study of growth stage conﬁgurations in high-
technology organizations. Bntreprenepgsm; Theory and Practice, 18: 5-29.

Hermalin, B. E. 8. Weisbach, M. S. 1988. The determinants of board
composition. Rand Journal of Economics. 19: 589-606.

Hermalin, B. E. 8. Weisbach, M. S. 1991. The effects of board composition and
direct incentives on ﬁrm performance. Financial Management, 20(4): 101-112.

201

Hill, C. W. L. & Snell, S. A. 1988. External control, corporate strategy, and ﬁrm
performance in research-intensive industries. Strategic Management Jogrnal, 9:
577-590.

Hillman, A., Carmella, A., 8. Paetzold, R. 2000. The resource dependence
role of corporate directors: Strategic adaptation of board composition in
response to environmental change. Mrnal of Management Studies, 37:
235-255.

Hof, R. D. 1999. Inside an internet IPO. Business Week, Sept. 6: 60-73.

Homgren, C. T., Sundem, G. L. 8. Stratton, W. O. 1999. Introduction to
management aocognting. Upper Saddle River, NJ: Prentice Hall.

Hoskisson, R. E., Johnson, R. A., & Moesel, D. D. 1994. Corporate divestiture
intensity in restructuring ﬁrms: Effects of governance, strategy, and performance.
Academy of Management Journal, 37: 1207-1251.

Jain, B. A. 8. Kini, O. 1994. The post-issue operating performance of IPO ﬁrms.
The Jopmal of Finance. 49: 1699-1726.

Jain, B. A. & Kini, O. 1999. The life cycle of initial public offering ﬁrms. Journal
of Business Finance 8. Accounting, 26: 1281-1317.

James, Jr., H. S. 2000. Reinforcing ethical decision making through
organizational structure. Journal of Business Ethics, 28: 43-58.

Jensen, M. C. & Meckling, W. H. 1976. Theory of the ﬁrm: Managerial behavior,
agency costs, and ownership structure. Journal of Financial Economics, 3: 305-
360. '

Johnson, J. L., Daily, C. M., 8. Ellstrand, A. E. 1996. Boards of directors: A
review and research agenda. Journal of Management, 22: 409-438.

Johnson, R. B. 1997. The board of directors over time: composition and the
organizational life cycle. lntemational Journal of Management, 14: 339-334.

Johnson, R. A., Hoskisson, R. E., 8 Hitt, M. A. 1993. Board of director '
involvement in restructuring: The effects of board versus managerial controls and
characteristics. Stratggic Management Journal, 14: 33-50.

Judge, W. Q. Jr. 8. Zeithaml, C. P. 1992. Institutional and strategic choice

perspectives on board involvement in the strategic decision process. Academy
of Management Joprnal, 35: 766-794.

202

Katz, D. 8. Kahn, R. L. 1966. The social psycholggy of organizations. New
York: Wiley.

Kazanjian, R. K. 1988. Relations of dominant problems to stages of growth in
technology-based new ventures. Academy of Management Jogmal, 31: 257-
279.

Kendall, M. G. 1962. Rank correlation methods (3d ed.) New York: Hafner.

Kerlinger, F. N. 1986. Foundations of behi/ioral research. New York: Holt.

Khandwalla, P. N. 1973. Viable and effective organizational designs of ﬁrms.
Academy of Management Journal, 16: 481-495.

Kimberly, J. R. 1976. Organizational size and the structuralist perspective: A
review, critique, and proposal. Administrative Science anrteny, 21: 571-597.

Kuwada, K. 1998. Strategic learning: The continuous side of discontinuous
strategic change. Organization Science, 9: 719-736.

Lawrence, P. R. 8. Lorsch, J. W. 1967. Organization and environment. Boston:
Harvard Business School Press.

Lerner, J. 1995. Venture capitalists and the oversight of private ﬁrms. Journal of
Finance, 50: 301 -31 8.

Lipman, F. D. 1994. Going public. Rocklin, CA: Prima Publishing.

Lindenberg, E. 8. Ross, S. 1981. Tobin’s q-ratio and industrial organization.
ﬂirnal of Business, 54: 1-32.

Lorsch, J. W. 8. Maclver, E. 1989. Pawns orjotentates: The reality of
America’s corporate boards. Boston, MA: Harvard Business School Press.

Lundberg, C. C. 1984. Strategies for organizational transitioning. In J. R.
Kimberly 8. R. E. Quinn (Eds). Managing organizational trarmons: 60-82. ,
Homewood, IL: lf'Wil'I.

Mace, M. L. 1948. The board of directors in small corporations. Boston, MA:
Harvard University.

Mace, M. L. 1971. Directors: Myth and reality. Boston, MA: Harvard University.

March, J. 8. Simon, H. 1958. Organization. New York: Wiley.

203

Mikkelson, W. H., Partch, M. M., 8. Shah, K. 1997. Ownership and operating
performance of companies that go public. Journal of Financial Boonomics, 44:
281-307.

Miller, D. 1994. What happens after success: The perils of excellence. Journal
of Management Studies, 31: 325-358.

Miller, D. 8. Friesen, P. H. 1983. Successful and unsuccessful phases of the
corporate life cycle. Organization Studies, 4: 339-356.

Miller, D. 8. Friesen, P. H. 1984a. Organizations: A quantum view. Englewood
Cliffs, NJ: Prentice-Hall.

Miller, D. 8. Friesen, P. H. 1984b. A longitudinal study of the corporate life cycle.
Management Science, 30: 1161-1183.

Miller, D. & Toulouse, J. M. 1986. Strategy, structure, and CEO personality in
small ﬁrms. American JOanal of Smallﬂisiness, Winter: 47-62.

Miner, J. B. 1982. Theories of organizational structure and process. Chicago:
Dryden.

Mintzberg, H. 1979. The structuring of organizations: A synthesis of th_e_
research. Englewood Cliffs, NJ: Prentice-Hall.

Mizruchi, M. S. 1983. Who controls whom? An examination of the relation
between management and boards of directors in large American corporations.
Academy of Management Review, 8: 426-435.

Mohan- Neill, S. l. 1995. The inﬂuence of ﬁrm ’5 age and size on its
environmental scanning activities. Journal of Small Business Management,
33(4). 10- 21.

Neter, J., Wasserman, W. 8. Kutner, M. 1990. Applied linear statistical models.
Boston: Irwin.

Nunnally, J. C. Pyschometric theory. New York: McGraw-Hill.

Parsons, T. 1960. Structure and process in mgiem socieﬂgs. Glencoe, IL:
Free Press.

Pearce, J. A. II 8. Zahra, S. A. 1992. Board composition from a strategic
contingency perspective. gggrnal of Management S_tgdies, 29: 411-438.

Penrose, E. T. 1959. The theory of the growth of the ﬁrm. Oxford: Basil
Blackwell. ,

204

Perrow, C. 1970. Organizational analysiszAsociological view. London:
Tavistock Books.

Pfeffer, J. 1972. Size and composition of corporate boards of directors: The
organization and its environment. AdministlrativeBCience (ﬂarteny. 17: 218-
228.

Pfeffer, J. 1973. Size, composition, and function of hospital boards of directors:
A study of organization-environment linkage. Administrative Science Quarteml.
18: 349-364.

‘ Pfeffer, J. 1981. Power in organizations. Marshﬁeld, MA: Pitman.

Pfeffer, J. 8. Salancik, G. 1978. The external control of organizations: A
resogce dependence perspectiLe. New York: Harper 8. Row.

Price, J. L. 8. Mueller, C. W. 1986. Hancﬁook of organizational measgrement.
Marshﬁeld, MA: Pitman.

Provan, K. G. 1980. Board power and organizational effectiveness among
human service agencies. Academy of Management Journal. 23: 221-236.

Pugh, D. S., Hickson, D. J., Hinings, C. R., & Turner, C. 1968. Dimensions of
organization structure. Administrative Science @arterly, 13: 65-105.

Rajan, R. 8. Zingales, L. 2001. The ﬁrm as a dedicated hierarchy: A theory of
the origins and growth of ﬁrms. The Q_uarterly Mmal of Economics. 116: 805-
851. -

Rechner, P. L. 1989. Corporate governance: Fact or ﬁction? Business
Horizons, 34(4), 11-15.

. Robinson, E. A. G. 1934. The problem of management and the size of ﬁrms.
genomic Journal, 44: 242-257.

Robinson, R. B. Jr. 1982. The importance of “outsiders” in small ﬁrm strategic
planning. Academy of Management Journal, 25: 80-93.

Rosenstein, S. 8. Wyatt, J. G. 1990. Outside directors, board independence,
and shareholder wealth. Journal of Financial Economics, 26: 175-191.

 

Rubenson, G. C. 8. Gupta, A. K, 1992. Replacing the founder: Exploding the
myth of the entrepreneur’s disease. B_usiness Honpns, 35(6): 53-57.

205

Rubenson, G. C. 8. Gupta, A. K. 1996. The initial succession: A contingency
model of founder tenure. Entrepreneurship Theory and Practice, 21: 21-35.

Selznick, P. 1949. TVA and the grass roots. Berkeley, CA: University of
California Press.

Schellenger, M. H.. Wood, D. D., 8. Tashakori, A. 1989. Board of director
compensation, shareholder wealth, and dividend policy. Journal of
Manament, 15: 457-467.

Slevin, D. P. 8. Covin, J. G. 1997. Strategy formation patterns, performance,
and the signiﬁcance of context. Journal of Management, 23: 189-209.

Small Business Administration. 1998. Small business answer card.
Http://www.sba.gov/ADVO/stats/answer.html: 1-6.

Smith, K. G. & Gannon, M. J. 1987. Organizational effectiveness in
entrepreneurial and professionally managed ﬁrms. gggmal of Small Bgsiness
Management, 25(3): 14-21.

Smith, N. R. 8. Miner, J. B. 1983. Type of entrepreneur, type of ﬁrm, and
managerial motivation: Implications for organizational life cycle theory. Strategic
Management Journal, 4: 325-340.

Stanworth, M. J. K. 8. Curran, J. 1976. Growth and the small ﬁrm - An
alternative view. The Jogmal of Management Studies, 13: 95-110.

 

Starbuck, W. H. 1965. Organizational growth and development. In J. G. March
(Ed.). Handbook of organi_zations: 451-533. Chicago: Rand IVIcNally.

Stearns, L. B. 8. Mizruchi, M. S. 1993. Board composition and corporate
ﬁnancing: The impact of ﬁnancial institution representation on borrowing.
Academy of Management Jopmal, 36: 603-618.

Stemp., I. 1970. Growth-For what? In I. Stemp(Ed.). Corporate growth
strategies: 9-32. New York: American Management Association.

Tabachnick, B, G. 8. Fidell, L. S. 1996. Using multivariate statistics. New York:
Harper Collins.

Tashakori, M. 1980. Management sgccession: From the owner-fognder to th_e_
professional president. New York: Praeger.

Thesmar, D. & Thoenig, M. 2000. Creative destruction and ﬁrm organization
choice. The Q_Igirterly Journal of Economics, 1152 1201-1237.

206

Thompson, J. D. 1967. Organizations in action. New York: McGraw-Hill.

Tobin, J. 1969. A general equilibrium approach to monetary theory. Journal of
Money. Credit. and Banking 1: 15-29.

Vance, S. C. 1983. Corporate leadership: Boards, directors. and strategy. New
York: McGraw Hill.

Vancil, R. F. 1987. Passing the baton: Managing the process of CE
succession. Boston, MA: Harvard Business School Press.

Varaiya, N., Kerin, R. A., & Weeks, D. 1987. The relationship between growth,
proﬁtability, and ﬁrm value. Strategic Management Jopmal, 8: 487-497.

Venkatraman, N. 8. Ramanujam, V. 1986. Measurement of business
performance in strategy research: A comparison of approaches. Academy of
Management Review, 11: 801-814.

Wagner, J. A. III, Stimpert, J. L., 8. Fubara, E. I. 1998. Board composition and'
organizational performance: Two studies of insider/outsider effects. Journal of
Manament Studies. 35: 1-23.

Walsh, J. P., Henderson, C. M., 8. Deighton, J. 1988. Negotiated belief
structures and decision performance: an empirical investigation. Organizational
Behavior and ligman Decision Processes, 42: 194-216.

Walsh, J. 8. Seward, J. 1990. On the efﬁciency of intemal and external
corporate control mechanisms. Academy of Management Review, 15: 421-458.

Ward, R. D. 1997. 21St century corporate boag. New York: John Wiley.

Weisbach, M. S. 1988. Outside directors and CEO turnover. Journal of
Financial Economics, 20: 431-460.

Weiss, M. 1988. Going public. Blue Ridge Summit, PA: Liberty House.
Welboume, T. M. 8. Andrews, A. O. 1996. Predicting performance of initial
public offerings: Should human resource management be in the equation?
Academy of Management Jogrnal, 39: 891-919.

Westphal, J. D. 8. Frederickson, J. W. 2001. Who directs strategic change?
Director experience, the selection of new CEOs, and change in corporate
strategy. Strategic Management Journal, 22: 1113-1137.

Whisler, T. L. 1988. The role of the board in the threshold ﬁrm. Family
Business Review. 1: 309-321.

207

Willard, G. E., Krueger, D. A., 8. Feeser, H. R. 1992. In order to grow, must the
founder go: A comparison of performance between founder and non-founder
managed high-grth manufacturing ﬁrms. Journal of Business Venturing, 7:
181-194.

Woodward, J. 1958. Management ang technology. London: Crown.
Zahra, S. A. 8. Pearce, J. A. II. 1989. Boards of directors and corporate ﬁnancial

performance: A review and integrative model. Journal of Management. 15: 291-
334.

Zahra, S. A. 8. Stanton, W. W. 1988. The implications of board of directors’
composition for corporate strategy and performance. lntemational Journal of
Mamement, 5: 229-237.

Zald, M. N. 1967. Urban differentiation, characteristics of boards of directors,
and organizational effectiveness. American Jogmal of Bociology, 73: 261-272.

Zald, M. N. 1969. The power and functions of boards of directors: A theoretical ‘
synthesis. American Jogrnal of Sociology, 75: 97-111.

208

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

  

1293 02328 7

   

I

 

—__