ChapterPDF Available

Ecological Approaches to Organizations

September 2007

September 2007

DOI:10.4135/9781848608030.n3

In book: Handbook of Organization Studies

Authors:

Joel A. C. Baum

University of Toronto

Andrew V. Shipilov

INSEAD

Our goal is to assess and consolidate the current state-of-the-art in organizational ecology. To accomplish this we review major theoretical statements, empirical studies, and arguments that are now being made. Although we attempt to survey ecological approaches to organizations comprehensively, because ecological research now constitutes a very large body of work, and because other extensive reviews are available (Aldrich & Wiedenmayer, 1993; Barnett & Carroll, 1995; Baum, 1996; Baum & Amburgey, 2002; Baum & Rao, 2004; Carroll, Dobrev & Swaminathan, 2002; Galunic & Weeks 2002; Rao, 2002; Singh & Lumsden, 1990), we emphasize recent work that challenges and extends established theory and highlight new and emerging directions for future research that appear promising. Our appraisal focuses on two main themes - demographic processes and ecological processes.

Structural Inertia Theory (Adapted from Kell & Amburgey 1991: 593)

…

Age and size dependence failure studies, 1983-2003

Figures - uploaded by Andrew V. Shipilov

Content may be subject to copyright.

Content uploaded by Andrew V. Shipilov

Content may be subject to copyright.

Introduction: What

Organizational Ecology

Is and Isn’t

Until the mid-1970s, the prominent approach in

organization and management theory emphasized

adaptive change in organizations. In this view, as

environments change, leaders or dominant coalitions

in organizations alter appropriate organizational

features to realign their fit to environmental demands

(e.g. Lawrence and Lorsch 1967; Thompson 1967;

Child 1972; Chandler 1977; Pfeffer and Salancik 1978;

Porter 1980; Rumelt 1986). Since then, an approach to

studying organizational change that places more

emphasis on environmental selection processes, intro-

duced at about that time (Aldrich and Pfeffer 1976;

Hannan and Freeman 1977; Aldrich 1979; McKelvey

1982), has become increasingly influential. The

stream of research on ecological perspectives of orga-

nizational change has generated tremendous excite-

ment, controversy and debate in the community of

organization and management theory scholars.

Inspired by the question, Why are there so

many kinds of organizations? (Hannan and Freeman

1977: 936), organizational ecologists seek to explain

how social, economic and political conditions affect

the relative abundance and diversity of organizations

and attempt to account for their changing composi-

tion over time. Organizational ecologists are very

empirically oriented – that is, driven by the cumula-

tive research findings of an international community

of scholars that attempts to replicate and extend

empirical generalizations derived from theoretical

expectations. Although differences exist among indi-

vidual investigators, ecological research typically

begins with three basic observations: (1) diversity is a

property of aggregates of organizations that has no

analogue at the level of the individual organization,

(2) organizations often have difficulty devising and

executing changes fast enough to meet the demands

of uncertain, changing environments, and (3) the

community of organizations is rarely stable – organi-

zations arise and disappear continually. Given these

observations, organizational ecologists pursue expla-

nations for the diversity of organizations at higher

levels of analysis of the organizational population

and community and focus on rates of organizational

founding and failure and rates of creation and death

of organizational populations as sources of increas-

ing and decreasing diversity.

Organizations, populations and communities of

organizations constitute the basic elements of an

ecological analysis of organizations (Hannan and

Freeman 1977; 1989). A set of organizations engaged

in similar activities and with similar patterns of

resource utilization constitutes a population (Hannan

and Freeman 1977; 1989). Populations form as a

result of processes that isolate or segregate one set of

organizations from another, including technological

incompatibilities, institutional actions such as

government regulations and imprinting effects

(Stinchcombe 1965; McKelvey 1982; Hannan and

Freeman 1989; Baum and Singh 1994a). Populations

themselves develop relationships with other popula-

tions engaged in other activities that bind them into

organizational communities (Astley 1985; Fombrun

1986; Hannan and Freeman 1989). Organizational

communities are functionally integrated systems of

interacting populations. In an organizational com-

munity, the outcomes for organizations in any one

population are fundamentally intertwined with those

of organizations in other populations that belong to

the same community system.

Although organizational ecology has been a

prominent subfield in organization studies for more

than 25 years, numerous critics and skeptics remain.

Why? The debate centres primarily on assumptions

about the relative influences of organizational

1.2 Ecological Approaches to Organizations

JOEL A. C. BAUM AND ANDREW V. SHIPILOV

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 55

history, environment and strategic choice on

patterns of organizational change advanced by

structural inertia theory (Hannan and Freeman

1977; 1984). Structural inertia theory asserts that

existing organizations frequently have difficulty

changing strategy and structure quickly enough to

keep pace with the demands of uncertain, changing

environments and emphasizes that major organiza-

tional innovations often occur early in the life-histo-

ries of organizations and populations. Organizational

change and variability are thus regarded to reflect

primarily relatively inert (i.e. inflexible) organizations

replacing each other over time. To organizational

ecology’s critics and skeptics this means environmen-

tal determinism and loss of human agency (e.g. Astley

and Van de Ven 1983; Perrow 1986).

Do ecological approaches imply that the actions

of particular individuals do not matter for organiza-

tions? The answer is no, of course. One part of the

confusion about organizational ecology is that

determinism is mistakenly contrasted with volun-

tarism rather than with probabilism (Hannan and

Freeman 1989; Singh and Lumsden 1990). Leaving

aside whether their actions are intelligent or foolish,

carefully planned or seat-of-the-pants, individuals

can clearly influence their organization’s future –

but under conditions of uncertainty and ambiguity

there are severe constraints on the ability of individ-

uals to conceive and implement correctly changes

that improve organizational success and survival

chances reliably in the face of competition. Thus,‘in

a world of high uncertainty, adaptive efforts … turn

out to be essentially random with respect to future

value’ (Hannan and Freeman 1984: 150). A second

part of the confusion has to do with the level of

analysis. The actions of individuals matter more to

their organization than they do to their organiza-

tion’s population as a whole. The actions of particu-

lar individuals may thus not explain much of the

diversity in organizational populations.

Changes in organizational populations reflect

the operation of four basic processes: variation,

selection, retention and competition (Campbell

1965; Aldrich 1979; McKelvey 1982). Variations are

human behaviours. Any kind of change, intentional

or blind, is variation. Individuals produce variations

in, for example, technical and management compe-

tencies constantly in their efforts to adjust their

organization’s relationship to the environment.

Organizational variations provide the raw material

from which selection can be made. Some variations

prove more beneficial to organizations than others

in acquiring resources in a competitive environment

and are thus selected positively – not by the environ-

ment, but by managers inside organizations and

investors, customers and government regulators

in the resource environment (Burgelman 1991;

Burgelman and Mittman 1994; McKelvey 1994;

Meyer 1994; Miner 1994).

When successful variations are known, or when

environmental trends are identifiable, individuals

can attempt to copy and implement these successful

variations in their own organization, or they can

attempt to forecast, anticipate, plan and implement

policies in the context of the predictable trends

(Nelson and Winter 1982; DiMaggio and Powell

1983; McKelvey 1994). However, when successful

variations are unknown, because, for example, the

behaviour of consumers and competitors is unpre-

dictable, the probability of choosing the correct

variation and implementing it successfully is very

low. Even when effective variations are identifiable,

ambiguity in the causes of success may frustrate

attempts at imitation. Under such conditions, varia-

tions can be viewed as experimental trials, some of

which are consciously planned and some of which

are accidental, some of which succeed and some of

which fail (McKelvey 1994; Miner 1994). Whether

or not they are known, over time, successful varia-

tions are retained as surviving organizations come

to be characterized by them.

If the survival odds are low for organizations with

a particular variant, it does not mean that these

organizations are destined to fail. Rather, it means

the capacity of individuals to change their organiza-

tions successfully is of great importance (Hannan

and Freeman 1989). However, there are strong

constraints on the capacity of individuals to change

existing organizations successfully, including estab-

lished practices, norms and incentives, scarcity of

resources, competition and limits to individual

rationality. Nevertheless, the existence of these con-

straints does not mean that individuals are irrele-

vant to processes of organizational change – only

that there are limits on the influence of individuals’

actions on variability in organizational properties.

Ecological theory does not remove individuals

from responsibility for or control (influence, at least)

over their organization’s success and survival –

individuals do matter. Ecological theory does, how-

ever, assume that individuals cannot always (or often)

determine in advance which variations will succeed.

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 56

Ecological theory also stresses that individuals have

difficulty changing existing organizations’ strategies

and structures quickly enough to keep pace with the

demands of uncertain, changing environments.

Because the success of particular variations is

often unknown in advance and because there are

strong constraints on the ability of individuals to

change their organizations, ecological analysis for-

mulates issues of organizational change and vari-

ability at the population level. It attempts to

estimate the odds of success for any particular vari-

ation by studying the survival chances of organiza-

tions with the variation and the rate at which

organizations adopt the variation. Consequently, in

contrast to adaptation approaches, which explain

changes in organizational diversity exclusively in

terms of the cumulative strategic choices and changes

of existing organizations, ecological approaches also

highlight the creation of new organizations and the

demise of old ones in addition.

Also in contrast to most organizational research,

which concentrates on the largest and most success-

ful organizations at a particular point in time,

ecological research examines entire organizational

populations, from the largest and longest-lived, to

the smallest and shortest-lived members, over

extended periods of time. Why? Because it is often

not possible to learn about the processes that create

success and failure by examining only current

successes. The problem is that comparison between

success and failure is implicit. Successful organiza-

tions in an industry may have certain strategies and

structures – but without information on failures we

cannot be certain poor performers didn’t have the

very same ones. Moreover, if success and failure

depend on the environment, winners and losers

may change as conditions change. The success of a

strategy in an industry at one point in time doesn’t

mean success next year under different conditions.

Thus, concentrating on organizations that succeed

at a particular point in time can result in misleading

inferences about the factors that produced the suc-

cess. These are examples of the methodological

problem of sample-selection bias (Heckman 1979;

Berk 1983). For these reasons, organizational ecolo-

gists undertake exhaustive data collection efforts

to identify all organizations in a population or

community (Hannan and Freeman 1989).

Current ecological theory and research takes four

distinct levels (Carroll 1984a; Hannan and Freeman

1989; Baum and Singh 1994a). Research at level one

investigates intraorganizational ecology. Research at

this level focuses on variation, selection and reten-

tion processes within organizations. Intraorganiza-

tional ecology models seek to describe how

organizations’ internal contexts and individuals’

purposeful and accidental use of variation, selection

and retention processes in organizations influence

rates of creation, transformation and death of orga-

nizational routines and strategies. Research at level

two focuses on the demography of organizations.

Research at this level considers variations in vital

rates (i.e. rates of organizational founding, change

and failure) for organizational populations. It

examines variations in these rates over time and

among populations. It seeks to identify empirical

regularities in these rates and specify their relations

to organizational characteristics. Research at level

three concerns the population ecology of organiza-

tions. At this level, research concentrates on growth

and decline of individual populations. Population

ecology models seek to describe how vital rates of

one population are influenced by processes endoge-

nous to that population as well as interactions with

other populations. It also seeks to specify how these

rates are shaped by patterns of environmental

change. Research at level four concerns the commu-

nity ecology of organizations. Community ecology

research investigates the evolution of community

structures and examines how the links binding a set

of populations into a community affect the likeli-

hood of persistence and stability of the community

as a whole. Community ecology research also

emphasizes processes of creation and demise of

populations of organizational forms.

To date, ecological theory and research has

focused primarily on demographic and population

ecology analyses of rates of organizational found-

ing, failure and change and to a much less (but

increasing) extent on intraorganizational and com-

munity ecology (for recent reviews see Galunic and

Weeks 2002; Rao 2002; Baum and Rao 2004).

Organizational founding and failure figure promi-

nently in organizational ecology because they affect

the relative abundance and diversity of organiza-

tions. Ecological theory stresses the difficulty indi-

viduals have changing existing organizations’

strategies and structures to keep pace with the

demands of uncertain, changing environments.

Consequently, for organizational ecologists, the

rates at which new and diverse organizations are

created and the rates at which organizations of

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 57

various types disappear are central to the dynamics

of organizational demography.

The focus of ecological research is changing

rapidly, however, as researchers turn their attention

to processes of organization-level change and tests

of structural inertia theory (e.g. Barnett and Carroll

1995). As Hannan and Freeman (1977: 930) point

out, a full treatment of organization–environment

relations must cover both adaptation and selection.

Researchers have embarked on development of a

combined perspective that sees processes of adapta-

tion and selection as complementary and interacting.

The emerging view takes seriously the occurrence of

selection processes and combines it with the sys-

tematic study of organization-level changes that

may, under certain conditions, be adaptive. And,

consequently, that organizational change can best be

studied by examining how social and environmental

conditions and interactions within and among pop-

ulations influence the rates at which new organiza-

tions and new populations are created, existing

organizations and populations die out and individ-

ual organizations change (Singh and Lumsden 1990;

Baum and Singh 1994a).

Chapter Aims and Overview

Our goal is to assess and consolidate the current

state-of-the-art in organizational ecology. To

accomplish this we review major theoretical state-

ments, empirical studies and arguments that are

now being made. Although we attempt to survey

ecological approaches to organizations comprehen-

sively, because ecological research now constitutes a

very large body of work, and because other extensive

reviews are available (Singh and Lumsden 1990;

Aldrich and Wiedenmayer 1993; Barnett and Carroll

1995; Baum 1996; Baum and Amburgey 2002;

Carroll et al. 2002; Galunic and Weeks 2002; Rao

2002; Baum and Rao 2004), we emphasize recent

work that challenges and extends established theory

and highlight new and emerging directions for

future research that appear promising. Our appraisal

focuses on two main themes – demographic

processes and ecological processes – the main

research areas in each summarized in Table 1.

Demographic Processes

Ecological approaches to founding and failure

constitute a radical departure from traditional

approaches to founding and entrepreneurship and

business failure, which focus on individual initia-

tive, skills and abilities (Carroll 1984a; Romanelli

1991; Aldrich and Wiedenmayer 1993). The tradi-

tional traits approach to founding and entrepre-

neurship assumes that there is something about an

individual’s background or personality that leads

him or her to found an organization. Gartner’s

(1989) review of traits studies suggests that they

have hit a dead end: most are based on small, cross-

sectional samples, drawn from unknown popula-

tions whose generality is not clear, and typically do

not use multivariate analytical techniques, raising

doubts about the significance of occasionally signif-

icant ‘trait’ variables. Similarly, traditional business

policy research typically attributes organizational

failure to managerial inexperience, incompetence,

or inadequate financing (e.g. Dun and Bradstreet

1978). By concentrating on the ‘traits’ of entrepre-

neurs and managers, traditional approaches to

organizational founding and failure deflected

attention away from the volatile nature of organiza-

tional populations and communities (Aldrich and

Wiedenmayer 1993). Ecological approaches to

organizational founding and failure, by comparison,

emphasize contextual or environmental causes –

social, economic and political – that produce varia-

tions in organizational founding and failure rates

over time by influencing opportunity structures

that confront potential organizational founders and

resource constraints that face existing organizations.

Because existing organizations have histories and

structures that influence their rates of failure

and change, studying organizational failure and

change is complicated by the need to consider

processes at both organizational and population

levels. Although founding processes have typically

been conceived in ecological research as attributes

of a population, since no organization exists prior to

founding, recently founding processes too are

increasingly seen as occurring at both organiza-

tional and population levels. Demographic analysis

examines the effects of organizational characteris-

tics on rates of organizational founding, failure and

change.

Organizational Founding

Although organizational founding is an important

theme in ecological research, in large part, foundings

have been treated as identical additions to homo-

genous populations: the characteristics of new

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 58

Key references

Lomi (1995); Stuart and

Sorenson (2003a, b)

Baum and Haveman (1997)

Freeman et al. (1983);

Henderson (1999)

Bruderl and Schussler (1990);

Fichman and Levinthal (1991)

Baum (1989); Ingram (1993)

Barron et al. (1994);

Ranger-Moore (1997)

Freeman et al. (1983);

Mitchell (1994)

Hannan and Freeman (1984);

Amburgey et al. (1993);

Baum and Singh (1996);

Dobrev et al. (2001; 2003)

Amburgey and Miner (1992);

Amburgey et al. (1993);

Greve and Taylor (2000)

Baum and Ingram (1998);

Ingram and Baum (1997a)

Key predictions

Variation in social, institutional and economic conditions across regions produce

unobservable region-specific proneness to experiencing the founding of particular

organizational forms

Avoidance of direct competition pushes entrants away from similar organizations,

while complementary differences pull them together; agglomeration economies pull

entrants toward competitors

Liability of newness: failure rates decline with age as roles and routines are

mastered and links with external constituents established

Liability of adolescence: failure rates rise with age until initial buffering resource

endowments are depleted, then decline with further increases in age

Liability of obsolescence: failure rates increase with age as their original fit with the

environment erodes

Liability of senescence: failure rates increase with age as internal friction, precedent

and political pacts accumulate, impeding action and reliable performance

Liability of smallness: failure rates decline with size, which buffers organizations

from threats to survival

Change: structural inertia increases as organizations age and grow, lowering rates

of organizational change

Failure: the failure rate increases after a ‘core’ change, but then declines with the

passage of time; the disruptive effects of change increase (decrease) with

organizational age (size)

The rate of an organizational change of the same type increases with the number

of prior changes of the same type, but then declines with the passage of time

since the last change of the same type.

Initial increases in operating experience lower failure rates as organizations move

down learning curves for their routines, but further increases reduce

responsiveness to changing environmental demands, raising failure rates

Key variables

Spatial heterogeneity

Entrant’s similarity to

incumbents

Organizational age

Organizational size

Organizational change

Cumulative

organizational change

Organizational

operating experience

Demographic

processes

Organizational

founding

Age

dependence

Size

dependence

Structural

inertia

Organizational

momentum

Organizational

learning

Table 1 Major ecological approaches to organizational founding, failure and change

(Continued)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 59

Key references

Barnett (1997)

Freeman and Hannan (1983;

1987); Carroll (1985); Carroll

et al. (2002)

Carroll and Delacroix (1982);

Delacroix and Carroll (1983);

Delacroix et al. (1989)

Hannan and Freeman (1987;

1988; 1989); Hannan and

Carroll (1992); Hannan et al.

(1995)

Baum and Ingram (1998);

Ingram and Baum (1997a)

Key predictions

The greater an organization’s historical exposure to competition, the lower its

failure rate

Exploit a narrow range of resources and are favoured in fine-grained and

concentrated environments

Tolerate widely varying environmental conditions and are favoured in coarse-

grained, high variability environments

Initial increases in prior foundings signal opportunity, stimulating new foundings,

but further increases create competition, suppressing new foundings

Initial increases in prior deaths free up resources, stimulating new foundings, but

further increases signal a hostile environment suppressing new foundings

Initial increases in density increase the institutional legitimacy of a population,

increasing foundings and lowering failures, but further increases produce

competition, suppressing foundings and increasing failures

An organization’s failure rate declines as a function of the operating experience of

its population at the time of its entry and of the population’s increasing operating

experience after the organization’s founding

An organization’s failure rate declines as a function of its population’s history of

competitive outcomes (e.g. failures) at the time of its entry and of the population’s

increasing competitive experience after the organization’s founding

Key variables

Organizational

competitive experience

Specialist strategy

Generalist strategy

Prior foundings

Prior failures

Population density

Population operating

experience

Population competitive

experience

Demographic

processes

Ecological

processes

Niche width

dynamics

Population

dynamics

Density

dependence

Population-

level learning

Table 1 (

Continued

)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 60

organizations, which define their domains, have not

been of central interest. The absence of organization-

specific factors in studies of founding stands in sharp

contrast to ecological studies of failure and change, in

which issues of organizational demography have long

been conspicuous. Studying heterogeneity in found-

ing is more complicated than studying heterogeneity

in failure or change: organizational attributes cannot

be used as explanatory variables in analyses of found-

ing because they cannot be observed for organiza-

tions that do not yet exist (Delacroix and Carroll

1983). As a result, the organization itself cannot be

the focal point of study.

Ecological researchers initially sidestepped this

complication by considering the population itself as

the unit experiencing the events. The problem with

this approach is that variations in social and

economic conditions across a population’s environ-

ment will produce differences in founding rates. If not

properly accounted for, this heterogeneity will result

in specification bias. Some progress on this problem

has been made by researchers using differences among

foundings within populations to specify more fine-

grained population substructures into which organi-

zations are differentially founded. A variety of criteria

have be used for this purpose including legal form

(Ranger-Moore et al. 1991; Rao and Neilsen 1992;

Baum and Oliver 1996), core technology (Barnett

1990), customer base (Baum and Singh 1994a), strat-

egy (Brittain 1994) and geographical location

(Barnett and Carroll 1987; Carroll and Wade 1991;

Swaminathan and Wiedenmayer 1991; Cattani et al.

2003). In this approach, studying founding involves

first specifying potential organizational subpopula-

tions and then examining differences in the rate at

which these subpopulations receive new entrants.

Unfortunately, this approach suffers operational

problems (e.g. artificial, arbitrary and large numbers

of organizational niches, ambiguous risk-set defini-

tion) that can undermine efforts to operationalize the

approach empirically. Consequently, subpopulation

entry studies typically provide descriptive accounts of

organizational niches that were actually filled. Three

alternative approaches have recently appeared that

make it possible to examine how differentiation

within populations and variation in their environ-

ments affect the founding process.

The first begins with the observation that organi-

zational environments have spatial components –

geographical barriers, localized resource environ-

ments – that affect the dynamics of organizational

populations. Variation in social and economic condi-

tions across regions or local differences in a popula-

tion’s institutional history can produce differences in

intrinsic founding rates, or in unobservable region-

specific proneness to experiencing the founding of

particular organizational forms. If different segments

of an organizational environment cannot be consid-

ered equally at risk of experiencing foundings of

organizations of a given type, then we should expect

location dependence and unobserved heterogeneity

in organizational founding rates.

To address these problems empirically, Lomi

(1995) proposes a model of location dependence in

organizational founding rates. In contrast to other

work, Lomi’s approach does not require subdivision

of organizational populations on the basis of

abstract a priori categories; instead empirical corre-

ctions for unobserved heterogeneity are estimated

directly from data. His analysis of founding rates of

Italian cooperative banks based on the model indi-

cates that models neglecting unobserved heterogene-

ity across geographic regions tend to overestimate

the effects of ecological processes on founding rates.

The analysis revealed the existence of two distinct

segments within the population; when the founding

rate was allowed to vary within segments, evidence

of heterogeneous response to general population

processes was found. Lomi’s study contributes to a

growing literature showing that organizational pop-

ulations are internally differentiated and that vital

rates vary systematically across heterogeneous seg-

ments of the population. Bringing the role of spatial

factors to the forefront of ecological analyses

enriches our understanding of organizational

founding processes.

Sorenson and colleagues (Barnett and Sorenson

2002; Stuart and Sorenson 2003a, b) recently

pioneered a second approach to dealing with geo-

graphic heterogeneity in founding rates. Like

Lomi’s, their approach does not require arbitrary

subdivision of the population. Unlike Lomi’s, how-

ever, it makes use of continuous measures of the

distance between fine-grained geographic units (i.e.

single zip codes) at risk of founding and variables

of theoretical interest, rather than estimating cor-

rections from the data. This permits effects of par-

ticular variations in social and economic conditions

on the founding rate to be estimated directly.

Sorenson et al. have used the approach to examine

whether location-specific founding rates of biotech-

nology firms depend on geographic proximity to

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 61

other biotechnology firms, lead inventors of

patented biotechnologies, venture capital firms, or

universities with leading biotechnology labs (Stuart

and Sorenson 2003a), as well as biotechnology firms

that have recently made an initial public offering or

that have been acquired (Stuart and Sorenson

2003b). Their findings provide substantial evidence

of spatial heterogeneity, suggesting that the spatial

distribution of resources materially affects entre-

preneurs’ ability and propensity to create new

organizations.

A third approach, introduced by Baum and

Haveman (1997), complements traditional ecologi-

cal analyses that ask when a founding will occur – by

any kind of organization within a population, in any

location in that population’s niche – by taking orga-

nizational foundings as given and asking what kind

of organization is founded and where. Like

Sorenson et al.’s, their approach is relational, focus-

ing on organizations’ positions relative to each other

in the resource space, and so does not require any

specification of subpopulations where foundings

can occur. They are concerned with new organiza-

tions’ similarity to (or difference from) incumbent

organizations. Given that a new organization

appears, they examine how similar it is to neigh-

bouring organizations. They investigate key entre-

preneurial decisions, namely how close new

organizations should locate in product and geo-

graphic space to incumbent organizations in their

industry. Baum and Haveman thus go beyond

geographic heterogeneity to consider product het-

erogeneity as well. Their analysis of the roles of size,

price and location similarity on Manhattan hotel

foundings finds evidence of both avoidance of

direct competition pushing similar hotels apart and

pulling complementary hotels together (e.g. Hawley

1950; White 1981), as well as of agglomeration

economies and institutional forces pulling similar

hotels together (e.g. Hotelling 1929; DiMaggio and

Powell 1983). The ‘hotel districts’ resulting from

these process help reduce consumer search costs

regarding hotels’ locations (e.g. proximity to points

of tourist interest or business activity) and class (i.e.

economy or luxury).

Together, these studies refocus organizational

ecologists’ emphasis on temporal heterogeneity

and time dependence in organizational founding to

less studied issues of spatial heterogeneity and loca-

tion dependence. The approaches to the dynamics

of location choice they advance make it possible

to subject theoretical predictions from spatial

economics and organizational sociology to empirical

testing, improving our knowledge of key entrepre-

neurial decisions and processes, and the tendency

for firms to cluster (or not) in product and geogra-

phic space.

Age and Size Dependence

A central line of inquiry in ecological research has

been the effect of organizational ageing on failure.

Until recently, the predominant view was the liabil-

ity of newness (Stinchcombe 1965: 148–9), the

propensity of young organizations to have higher

failure rates. Underlying Stinchcombe’s (1965) lia-

bility of newness is the assumption that young orga-

nizations are more vulnerable because they have to

learn new roles as social actors and create organiza-

tional roles and routines at a time when organiza-

tional resources are stretched to the limit. New

organizations are also assumed to typically lack

broad bases of influence and endorsement, stable

relationships with important external constituents

and the legitimacy that years of experience in provid-

ing particular products or services confer on older

organizations. In a complementary viewpoint,

Hannan and Freeman (1984) suggest that selection

pressures favour organizations capable of demon-

strating their reliability and accountability. Reliabil-

ity and accountability require organizations to be

highly reproducible. This reproducibility, and the

structural inertia that it generates, are expected to

increase as organizations age. Since selection proces-

ses favour highly reproducible structures, older

organizations are predicted to be less likely to fail

than young organizations.

A second important line of research examines

how organizational size influences failure rates.

Larger organizations are thought to be less likely to

fail for a variety of reasons. Since inertial tendencies

in organizations increase with size, and since selec-

tion pressures favour structurally inert organiza-

tions for their reliability, large organizations are

expected to be less vulnerable to the risk of failure

(Hannan and Freeman 1984). The propensity of

small organizations to fail is also argued to result

from several liabilities of smallness, including prob-

lems of raising capital, recruiting and training a

workforce, meeting higher interest payments and

handling the administrative costs of compliance with

government regulations (Aldrich and Auster 1986).

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 62

Large size also tends to legitimate organizations to

the extent that stakeholders interpret organizational

size as an outcome of prior success and an indicator

of future dependability.

One major reason the liability of smallness is

studied is related to the liability of newness. Since

new organizations tend to be small, if, as the liabil-

ity of smallness predicts, small organizations have

higher failure rates, then liabilities of newness and

smallness are confounded and must be separated

empirically (Freeman et al. 1983). For example,

Levinthal (1991a) shows that a simple simulation

model in which change in organizational assets over

time is represented as a random walk replicates the

liability of newness in organizational mortality

rates. Thus, if organizational size increases with age

and failure decreases with size, then what appears as

a liability of newness may be an artifact of unmea-

sured size. This is one example of the well-known

problem of unobserved heterogeneity, which can

result in spurious negative age dependence in failure

rates (e.g. Tuma and Hannan 1984; Petersen and

Koput 1991).

Notably, although early ecological studies

supported the liability of newness hypothesis con-

sistently (e.g. Carroll and Delacroix 1982; Carroll

1983; Freeman et al. 1983), as Table 2 shows, later

studies find that, after controlling for contempora-

neous organizational size, failure rates do not gener-

ally decline with age. Since much of the original

support for the liability of newness comes from

studies in which organizational size is not con-

trolled, the early supportive findings may simply

reflect this. On the other hand, with few exceptions,

studies in Table 2 support the liability of smallness

prediction that organizational failure rates decline

with increased size.

Bigger may be Better,

but is Older Wiser?

These empirical findings have prompted two alter-

native theoretical perspectives on age dependence

that question the basic liability of newness argu-

ment. The first alternative, the liability of adolescence

hypothesis (Bruderl and Schussler 1990; Fichman

and Levinthal 1991), predicts an inverted U-shaped

relationship between age and failure. This model

observes that new organizations start with an initial

stock of assets (e.g. goodwill, psychological commit-

ment, financial investment) that buffers them from

failure during a honeymoon period – even when

early outcomes are unfavourable. The larger an

organization’s initial stock of assets, the longer the

expected duration the organization is buffered. As

this original stock is depleted, however, organiza-

tions face a liability of adolescence and those unable

to generate needed resource flows because, for

example, they were unable to establish necessary

roles and routines or develop stable relationships

with important external constituents, fail. After ado-

lescence, the future probability of failure declines,

since surviving organizations are those able to

acquire sufficient ongoing resources.

Liability of newness and adolescence arguments

provide divergent accounts of age dependence for

young organizations, but agree that failure rates

decline monotonically for older organizations.

Notably, processes underlying these models (e.g.

learning and creating new roles and routines, estab-

lishing relations with external constituents, depleting

initial endowments) occur early in the organizational

life cycle. These arguments thus have little to say

about organizations that are not new or adolescent.

The liability of ageing hypothesis identifies processes

that become important later in the organizational

life cycle and predicts an increasing rate of failure

for older organizations as a result of these processes

(Aldrich and Auster 1986; Baum 1989; Ingram 1993;

Barron et al. 1994; Ranger-Moore 1997). Thus, the

liability of ageing hypothesis complements and

extends liability of newness and adolescence

hypotheses (Baum 1989).

The liability of ageing argument begins with

another insight from Stinchcombe’s (1965: 153)

essay: ‘… the organizational inventions that can be

made at a time in history depend on the social tech-

nology available at that time’. Organizations thus

reflect the environment at the time of their founding.

The initial fit between organizations and their envi-

ronments erodes, however, as incomplete informa-

tion, bounded rationality, and inertial tendencies

make it difficult, if not impossible, for individuals to

keep their organizations aligned with rapidly chang-

ing and unpredictable environmental demands.

Environmental change also creates opportunities for

new organizations to enter and undermine the com-

petitive positions of established organizations (e.g.

Tushman and Anderson 1986). Thus, encountering a

series of environmental changes that decreases the

alignment between organizations and environments

exposes ageing organizations to an increased risk of

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 63

Table 2 Age and size dependence failure studies, 1983–2003

Population Age

Size

Size variable References

US Labor Unions, 1836–1985

−+Membership at founding Freeman et al. (1983); Hannan and Freeman

US Brewers, 1633–1988 − na (1989); Carroll and Hannan (1989a, b);

Argentina Newspapers, 1800–1900 − na Carroll and Wade (1991); Hannan and

Ireland Newspapers, 1800–1975 − na Carroll (1992)

San Francisco Newspapers, 1800–1975 − na

Little Rock Newspapers, 1815–1975 − na

Springfield Newspapers, 1835–1975 − na

Shreveport Newspapers, 1840–1975 − na

Elmira Newspapers, 1815–1975 − na

Lubbock Newspapers, 1890–1975 − na

Lafayette Newspapers, 1835–1975 − na

California Wineries, 1940–1985 0 − Storage capacity Delacroix et al. (1989); Delacroix and

Swaminathan (1991)

Iowa Telephone Companies, 1900–1929 0 0 Subscribers Barnett (1990); Barnett and Amburgey (1990)

Pennsylvania Telephone − 0

Companies, 1879–1934 +/− 0 Barnett (1997)

West German Business +/−−Employees at founding Bruderl and Schussler (1990)

Organizations, 1980–1989

Bavarian Brewers, 1900–1981 0 − Small firm dummy Swaminathan and Wiedenmayer (1991)

Toronto Day Care Centers, 1971–1989 +−Licensed capacity Baum and Oliver (1991; 1992); Baum and

Singh (1994b)

Toronto Nursery Schools, 1971–1987 −−Licensed capacity

US Immigrant Newspapers, 1877–1914 − na Olzak and West (1991)

African-American Newspapers, 1877–1914 − na

Manhattan Banks, 1840–1976 0 − Assets Banaszak-Holl (1992; 1993)

Manhattan Hotels, 1898–1990 +−Number of rooms Baum and Mezias (1992)

California S&Ls, 1970–1987 0 0 Assets Haveman (1992; 1993c)

US Mutual S&Ls, 1960–1987 +/− 0 Assets Rao and Neilsen (1992)

US Stock S&Ls, 1960–1987 +/− 0 Assets

US Cement Producers, 1888–1982 0/− na Anderson and Tushman (1992)

US Minicomputer Manufacturers, 1958–1982 +/– na

US Group HMOs, 1976–1991 0 − Enrolment Wholey et al. (1992)

(Continued)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 64

Table 2 (

Continued

)

Population Age

Size

Size variable References

US Independent Practice Assn. HMOs, 1976–1991 +−Enrolment

Finish Newspapers, 1771–1963 − na Amburgey et al. (1993)

US Brewers, 1878–1988 +−Production in Carroll et al. (1993)

1878 and 1879

US Microcomputer Manufacturers, 1975–1986 +−Units sold Ingram (1993)

US Integrated Circuit Manufacturers, 1971–1981 0 – Employees Loree (1993)

Medical Diagnostic Imaging Firms, 1953–1989 +/− 0 Corporate sales Mitchell and Singh (1993)

US Trade Associations, 1901–1990 0/−−Membership Aldrich et al. (1994)

US Credit Unions, 1980–1989

+−/+/− Assets Amburgey et al. (1994)

US Hotel Chains, 1896–1980 0 − Number of hotels Ingram (1994)

New York City Credit Unions, 1914–1990 +−Log real total assets Barron et al. (1994)

Medical Equipment Manufacturers, 1950–1990 −−Sales Mitchell (1994)

New York State Life Insurance & −−Number of policies sold Lehrman (1994)

Assessment Companies, 1881–1931

New York State Life Insurance −+Number of policies sold

Fraternal Societies, 1881–1931

US Automobile Producers, 1885–1981 −−Production capacity Carroll et al. (1996)

Niagara Falls Hotels, 1885–1991 0 − Number of rooms Ingram and Inman (1996)

NY State Life Insurance Companies, 1813–1935 +/0 − Assets Ranger-Moore (1997)

California Hospitals, 1980–1990 0 − Log number of beds Reuf (1997)

US Hospital Systems Software Firms, 1961–1991 na − Log sales Singh (1997)

US Hotel Chains, 1896–1990 +−Log number of components Ingram and Baum (1997a)

San Francisco Hospitals, 1945–1990 0 0 Log number of beds Reuf and Scott (1998)

US Breweries, 1845–1918 − na Wade et al. (1998)

Proprietary Strategists, US Personal +−Log sales Henderson (1999)

Computer Industry, 1975–1992

(Continued)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 65

Table 2 (

Continued

)

Population Age

Size

Size variable References

Standards Strategists, US Personal +/−−Log sales

Computer Industry, 1975–1992

New York Credit Unions, 1914–1990 +−Assets Barron (1999)

Ontario Independent Nursing Homes, 1971–1996 0 − Log number of beds Baum (1999)

Ontario Chain Nursing Homes, 1971–1996 + 0

US Breweries, 1938–1997 0 − Production capacity Carroll and Swaminathan (2000)

US Bicycle Industry, 1880–1916 − na na Dowell and Swaminathan (2000)

Israeli Worker Cooperatives, 1920–1992 0 na Cooperatives absorbed Ingram and Simons (2000)

US Wineries, 1941–1990 na −/+ Acres of vineyard owned Swaminathan (2001)

na − Log gallons of installed

capacity

French Auto Manufacturers, 1885–1981

−−Log production Dobrev et al. (2001); Dobrev

and Carroll (2003)

German Auto Manufacturers, 1885–1981

−−Log production

British Auto Manufacturers, 1885–1981

−−Log production

Law Firms/Attorneys in Silicon Valley, 1945–1996 0 − Number of partners Phillips (2001; 2002)

0 − Number of associates

MIT Start-Ups, 1980–1994 − 0 Log of cumulative revenues Shane and Stuart (2002)

Canadian Biotechnology Firms, 1991–2000 0 − Log R&D employees Baum and Silverman (2004)

US Auto Manufacturers, 1885–1981 −−Log production Dobrev and Carroll (2003);

Dobrev et al. (2003)

X/Y gives the signs of significant (p < 0.05) linear and squared terms, respectively, when estimated. X gives the sign of the effect of initial increases in age,

Y gives the sign of the effect for later increases.

See Hannan and Freeman (1989: 257–9) for an interpretation of this positive size effect.

Amburgey et al. (1994) test a cubic effect of size to examine the failure risk of mid-sized organizations.

Results differ somewhat in the two studies; Dobrev and Carroll (2003) exclude the smallest firms from their sample.

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 66

obsolescence and, concomitantly, failure. Ageing may

also bring about senescence: an accumulation of inter-

nal friction, precedent and political pacts that impede

action and reliable performance. Notably, obsoles-

cence does not require inertia to increase with orga-

nizational age – even if inertia is constant over time

there will be a liability of obsolescence. Obsolescence

and senescence thus pose separate risks: senescence is

a direct effect of ageing; obsolescence a result of envi-

ronmental change.

Available empirical evidence regarding age-

dependence is equivocal. Among studies in Table 2

that control for contemporaneous organizational

size, 12 populations exhibited a liability of newness,

11 a liability of ageing, six a liability of adolescence

and 15 no age dependence. Several explanations

may help account for the varied findings.

Two sample-selection problems may bias results

away from a liability of newness. First, the new orga-

nizations studied may be old new organizations,

that is organizations late in the process of emer-

gence (Katz and Gartner 1988). If researchers were

able to obtain data on organizations earlier in the

founding process (e.g. prior to formal incorpora-

tion – see, for example, Rao 2001), liability of new-

ness findings might be stronger. Secondly, in several

studies, left-censored organizations, that is organi-

zations founded before the start of the observation

period with known founding dates are included in

the analysis. Because they are already survivors, left-

censored organizations tend to be low-risk cases;

treating these cases as standard subjects can lead to

an underestimation of failure rates at shorter dura-

tions (Guo 1993).

It may also be that support for the liability of age-

ing is overstated. If organizational age coincides

with the amount of environmental change experi-

enced by an organization, and if the risk of failure

increases with cumulative environmental change,

then the probability of failure will increase, spuri-

ously, with age if accumulated environmental

change is uncontrolled (Carroll 1983: 313) – and it

typically is not. Thus, in the same way that negative

age dependence can result spuriously from uncon-

trolled size, positive age dependence (after control-

ling for size) may result spuriously from uncontrolled

organizational exposure to environmental change.

Notably, this implies that, after controlling for size

and environmental change, no age dependence

should be observed unless there is a liability of

senescence.

The more limited support for the liability of

adolescence hypothesis may have a benign expla-

nation: Because most researchers examine only

monotonic age dependence specifications, tests of

the liability of adolescence hypothesis are infre-

quent. Notably, however, six of the eight studies in

Table 2 that permit non-monotonic age dependence

and control for size find a liability of adolescence.

Hannan et al. (1998a, b) recently suggested that

divergent age dependence results might reflect non-

proportionality in age effects, combined with hetero-

geneity in the size distributions of populations

studied. And, although they find evidence that

patterns of age dependence vary for small and large

automobile producers in four European countries, no

systematic pattern emerges from their results. More

generally, divergent age dependence results might

reflect variation in age dependence across popula-

tions or subpopulations (Baum 1996). Henderson

(1999), for example, hypothesized contingent age

dependence effects based on an organization’s tech-

nology strategy – proprietary or standards-based.

Henderson’s analysis of sales growth and failure

among US personal computer manufacturers during

1975–92 provided strong support for his contingency

view, demonstrating how multiple patterns of age

dependence can operate simultaneously in a single

population. His study also revealed important trade-

offs between growth and the risk of failure resulting

from the joint effects of age and strategy.

To advance our understanding of these issues,

research on age dependence must move beyond the

use of age as a surrogate for all constructs underly-

ing the various age dependence models and begin to

test the models’ assumptions directly. Although

there has been much debate concerning the under-

lying source of the hazards facing new firms, most

of the research in this debate implicitly assumes that

new entrants are typified by a lack of stable rela-

tionships and sufficient resources. For example, the

liability of newness hypothesis assumes that a lack

of social approval, stability and sufficient resources

typifies recent entrants and that these shortcomings

increase their risk of failure, but organizational vari-

ation in these factors is rarely measured directly. Yet,

newly founded organizations display considerable

variation in their access to resources and stable rela-

tionships. Among US biotechnology firms, for

example, Barley et al. (1992) showed that startups

exhibited greater variability in alliances than estab-

lished firms, and Kogut et al. (1993) showed that

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 67

startups were more likely to develop alliances than

incumbents. To date, only a few studies other than

Henderson’s (1999) have explored the potential for

variation in liability of newness and smallness

effects. Singh et al. (1986b) found that external

legitimacy – measured as inclusion in community

directories or charitable registration – decreased the

liability of newness among voluntary social service

organizations. Reinforcing and extending these

results, Baum and Oliver (1991) showed that insti-

tutional linkages with municipal government and

community agencies moderated both liabilities of

newness and smallness for day care centres and

nursery schools. These studies, along with

Henderson’s, support the idea that the liability of

newness is attributable to the absence of stable rela-

tionships to important institutional actors and

suggest that if young organizations obtain early

legitimacy and access to resources through the

formation of institutional attachments to commu-

nity and public constituents, a liability of newness

may not be observed.

A further benefit of taking this approach is that

liabilities of newness, adolescence and obsolescence

can be treated as complementary rather than as com-

peting organizational processes. Thus, although we

still know very little about how ageing affects orga-

nizational failure, or the conditions under which

one, the other or some combination of these models

will predominate, several recent studies offer

promise for future progress.

Structural Inertia Theory

Structural inertia theory depicts organizations as

relatively inert entities for which adaptive response

is not only difficult and infrequent, but hazardous as

well. Consequently, change in individual organiza-

tions is viewed as contributing considerably less to

population-level change than organizational found-

ing and failure. Given the centrality of this theore-

tical position to ecological approaches, it is not

surprising that organizational ecologists have

amassed a wealth of studies on the influence of

organizational and environmental factors on rates

of organizational change, as well as the survival

consequences of different kinds of changes.

Organization theory focuses frequently on the

relative advantages of alternative configurations of

organizational features. Consequently, a great deal of

research on organizational change has concentrated

on the content of organizational change: A change to

a more advantageous configuration is considered

as adaptive, while a switch to a less advantageous

configuration is considered detrimental (Miller and

Friesen 1980; Amburgey et al. 1993). Complement-

ing this focus, Hannan and Freeman’s (1984) struc-

tural inertia theory offers a model of the process of

organizational change that considers both internal

and external constraints on change. Structural iner-

tia theory addresses two main questions: (1) How

changeable are organizations? and (2) is change

beneficial for organizations? Figure 1 summarizes

structural inertia theory.

How Changeable are

Organizations?

Hannan and Freeman (1977) pointed out that orga-

nizations face both internal and external constraints

on their capacity for change and that, given these

constraints, selection processes are important to

explaining change in organizational populations.

Building on this idea, Hannan and Freeman (1984)

adopt a somewhat different approach that takes

seriously the potential for organizational change by

viewing inertia as a consequence of rather than

antecedent to selection processes. Although some

kinds of organizational changes occur frequently in

organizations and sometimes these can even be rad-

ical changes, the nature of selection processes is

such that organizations with inert features are more

likely to survive (Hannan and Freeman 1984: 149).

The structure in structural inertia theory refers to

some, but not all, features of organizations. Hannan

and Freeman (1984: 156) emphasize core features of

organizations, which are related to ‘… the claims

used to mobilize resources for beginning an organi-

zation and the strategies and structures used to

maintain flows of scarce resources’. Core features

include, the goals, forms of authority, core technology

and market strategy of organizations. Peripheral

features protect an organization’s core from uncer-

tainty by buffering it and by broadening the organi-

zation’s connections to its environment. Peripheral

features include number and size of subunits, number

of hierarchical levels, spans of control, communica-

tion patterns and buffering mechanisms.

Hannan and Freeman (1984: 156) argue that core

features have higher levels of inertia than peripheral

features. Thus, in comparison to the probability of

change in peripheral features, the probability of

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 68

change in core features is very low (p. 157).

However, they do not claim that organizations never

change. Instead, they define inertia relative to envi-

ronmental change: ‘Structures of organizations have

high inertia when the speed of reorganization

[change in core features] is much lower than the

rate at which environmental conditions change’

(Hannan and Freeman 1984: 151). This formulation

poses some rather difficult measurement problems –

how can rates of organizational and environmental

change be matched? Consequently, most studies

reviewed here examine absolute rates of organiza-

tional change and report tests of structural inertia

theory, but not relative inertia theory (an exception

is Ruef 1997).

In addition to varying by facet of organizational

structure, Hannan and Freeman (1984) also pro-

pose that inertial pressures vary with organizational

age and size. Because older organizations have had

time to more thoroughly formalize internal rela-

tionships, standardize routines and institutionalize

leadership and power distributions, as well as

develop rich networks of dependencies and

commitments with other social actors, reproducibil-

ity of structure and inertia should increase with age

(Hannan and Freeman 1977; Pfeffer and Salancik

1978; Nelson and Winter 1982; Granovetter 1985;

Aldrich and Auster 1986). Thus, older organizations

should be most limited in their ability to adapt to

changing environmental demands, and so the prob-

ability of attempting change in core features to

decline with age (Hannan and Freeman 1984: 157).

Organizational size is also associated with resistance

to change. As organizations increase in size, they

emphasize predictability, formalized roles and con-

trol systems and organizational behaviour becomes

predictable, rigid and inflexible, increasing the level

of structural inertia (Downs 1967; Aldrich and

Auster 1986). Moreover, by buffering organizations

from failure, large size may reduce the impetus

for change (Levinthal 1994; Barnett 1997).

Consequently, the probability of attempting change

in core features declines with size (Hannan and

Freeman 1984: 159).

Ecological Approaches to Organizations

(+)

(+)(+)

(−)

(+)

(−)

(+)/(−)

Institutionalization

Reproducible

Structure

Standardized

Routines

Organizational

Age

Organizational

Size

Core Change

Attempt

Momentum

Failure

Inertia

Figure 2.1 Structural Inertia Theory (Adapted from Kell & Amburgey 1991: 593)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 69

Age and Size Dependence in

Rates of Organizational Change

Tests of age and size dependence in rates of organi-

zational change, presented in Table 3, are mixed

and, overall, appear to offer little support for struc-

tural inertia theory’s predictions. In their 1990

review, Singh and Lumsden used Hannan and

Freeman’s (1984) core-periphery distinction to

interpret the mixed findings available to them. They

speculated that rates of core feature change decrease

with age, while rates of peripheral feature change

increase with age. Does this distinction help account

for the mixed age (and size) dependence findings

available now? It appears not. For example, diversi-

fication – the development of new products or ser-

vices, often for new clients and frequently requiring

implementation of new administrative, production

or distribution technologies – is one core change

studied across multiple populations. Inspection of

Table 3 reveals little evidence that diversification is

related negatively to either age or size. What might

account for the findings? Below, we examine several

possibilities.

Fluidity of Age and Size

In contrast to structural inertia arguments, some

theoretical views suggest that organizations become

more fluid with age (Singh et al. 1988). Although

selection processes favour organizations that are fit

with their environment, the match between organiza-

tions and their environments is constantly eroding as

managerial bounded rationality, informational con-

straints and inertial pressures prevent organizations

from keeping pace with constantly changing environ-

ments. Thus, ‘… through a cumulative history of

having been alive, the stresses and strains of living

through multiple environmental changes cumulate

in organizations, increasing the pressure on organiza-

tions to change’ (Singh et al. 1988: 6). This implies a

discontinuous rather than incremental pattern of

organizational change (Miller and Friesen 1980;

Tushman and Romanelli 1985).

Some theoretical views also support the idea that

larger organizations are more fluid. Internal com-

plexity, differentiation, specialization and decentral-

ization, all features of large organizations, have each

been associated with the adoption of innovations

(Haveman 1993a). The slack resources available to

larger organizations may both enable them to initi-

ate change in response to environmental change

(Cyert and March 1963; Thompson 1967). Greater

size relative to other actors also increases market

power, lowering barriers to entry stemming from

scale economies (Bain 1956; Scherer and Ross 1990)

and reducing external political considerations

(Pfeffer and Salancik 1978; Aldrich and Auster

1986).

Age and size estimates in Table 3 support fluidity

and inertia predictions with approximately equal

frequency. Although Singh et al.’s (1988) original

study did not include left-censored organizations,

many subsequent studies that find evidence of

fluidity did. Because left-censored organizations,

founded before the observation period begins, are

not observed when they are youngest and smallest,

including these organizations in analyses without

appropriate corrections, can lead to an under-

estimation of rates of change at younger ages and

smaller sizes (Guo 1993). Moreover, if large organi-

zations are buffered by their resources from the risks

of change (Hannan and Freeman 1989), support for

the fluidity of size may reflect sample selection bias

resulting from right censoring: Small organizations

are not observed responding to changing environ-

ments because they fail prior to the realization of

such efforts. Once again, sample selection problems

may account, in part, for the mixed findings.

Repetitive Momentum

Although Hannan and Freeman (1984) do not

include prior changes in their theoretical model,

Amburgey and colleagues (Amburgey and Kelly

1985; Kelly and Amburgey 1991; Amburgey and

Miner 1992; Amburgey et al. 1993) suggest that a

complete understanding of organizational change

requires consideration of an organization’s history of

change. From an organizational learning perspec-

tive, making a change furnishes an organization

with the opportunity to routinize the change

(Nelson and Winter 1982; Levitt and March 1988).

Each time an organization engages in a particular

kind of change, it increases its competency at that

change. The more experienced an organization

becomes with a particular change, the more likely it

is to repeat the change. If a particular change

becomes causally linked with success in the minds

of organizational decision-makers – irrespective of

whether such a link actually exists – reinforcement

effects will make repetition even more likely. Thus,

once change is initiated, the change process itself

may become routinized and subject to inertial

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 70

Table 3 Rate of organizational change studies, 1985–2003

Population Type of change Age

Size # Prior Dynamic Reference

US Business Periodicals, 1774–1865 Ownership − 0 + na Amburgey and Kelly (1985)

Editor − 0 + na

Name − 00 na

Layout − 0 + na

Content 0 0 0 na

Voluntary Social Service Organizations, 1970–1982 Name ++ na na Singh et al. (1988; 1991);

Tucker et al. (1990a, b)

Sponsor 0 0 na na

Location + 0na na

Service area + 0na na

Goals + 0na na

Client groups 0 0 na na

Service conditions 0 0 na na

Chief executive ++ na na

Structure + 0na na

Silicon Valley Semiconductor Producers

Change in initial strategy + na na na Boeker (1989)

US Medical Diagnostic Imaging Firms, 1959–1988 Entry to emerging subfield na 0 na na Mitchell (1989)

Toronto Day Care Centres, 1971–1987 Specialist to generalist +− na na Baum (1990a)

Generalist to specialist +/− 0na na

US Health Maintenance Organizations

For-profit to non-profit + 0 na na Ginsberg and Buchholtz (1990)

US Airlines, 1962–1985 Business-level specialism − 0 + na Kelly and Amburgey (1991;

see also, Kelly 1988)

Business-level generalism 0 −+ na

Corporate-level specialism 0 0 + na

Corporate-level generalism 0 −+ na

Edmonton Gasoline Stations, 1959–1988 Domain enlargement 0 na na 0 Usher (1991); Usher and

Evans (1996)

Domain contraction 0 na na +

Niche migration +/− na na na

California Wineries, 1946–1984 Brand portfolio 0 −+ −Delacroix and

Swaminathan (1991)

Product line 0 0 0 0

Land ownership status 0 0 + 0

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 71

Table 3 (

Continued

)

Population Type of change Age

Size # Prior Dynamic Reference

Fortune

500 Companies, 1949–1977 Product-extension mergers na −+na Amburgey and Miner (1992);

Amburgey and Dacin (1994)

Conglomerate mergers na 0 + na

Horizontal mergers na 0 + na

Vertical integration na 0 + na

Product market diversification na 0 − 0

Structural decentralization na − 0 −

Finnish Newspapers, 1771–1963 Content − na +− Amburgey et al. (1993);

Miner et al. (1990)

Publication frequency − na +−

State Bar Associations, 1918–1950 Unification attempt −− na na Halliday et al. (1993)

US Private Liberal Arts Colleges, 1972–1986 Change to coed + 0 na na Zajac and Kraatz (1993)

Add graduate programme −+ na na

Add business programme 0 + na na

US Bank Holding Companies, 1956–1988 Related acquisition − na +− Ginsberg and Baum (1994)

Unrelated acquisition 0 na + 0

California S&Ls, 1977–1987 Real estate (entry rates) 0 + na na Haveman (1994; see also,

Haveman 1992; 1993a, b)

Non-residential mortgages 0 0 na na

Mortgage-backed securities 0 0 na na

Consumer lending 0 0 na na

Commercial lending 0 0 na na

Service companies + 0nana

US Trade Associations, 1900–1980 Change in organizing 0/0 0 na na Aldrich et al. (1994)

domain or goals

US Airlines, 1979–1986 Tactical competitive actions 0 0 + na Miller and Chen (1994)

Strategic competitive actions 0 + 0na

US Radio Stations, 1984–1993 Abandon format na − na na Greve (1995)

Abandon format and re-enter na 0 na na

US Airlines, 1985–1986 Propensity for action na − na na Chen and Hambrick (1995)

Action execution speed na − na na

Action visibility na + na na

Responsiveness na + na na

Response announcement speed na 0 na na

(Continued)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 72

Table 3 (

Continued

)

Population Type of change Age

Size # Prior Dynamic Reference

Response execution speed na + na na

Response visibility na − na na

Toronto Day Care Centres, 1971–1989 Market entry +− −+ Baum and Singh (1996)

Market exit 0 0 0 0

California Airlines, 1979–1984 Route entry +− na na Baum and Korn (1996)

Route exit −+ na na

US Radio Stations, 1984–1992 Innovative changes na 0 0 0 Greve (1996; 1998);

Greve and Taylor (2000)

Format changes na − 00

Satellite entry na 0 ++

Production changes na 0 0 0

California Hospitals, 1980–1990 Market change ++na na Ruef (1997)

US Semiconductor Producers, 1978–1992 Change in diversification 0 + na na Boeker (1997)

US Liberal Arts Colleges, 1971–1986 Adoption of new professional 0 + na na Kraatz (1998)

programme

Russian Construction Companies, 1989–1993 Spin-off of state-owned units na na + na Suhomlinova (1999)

into independent companies

US Voluntary Associations, 1955–1986 Protest to advocacy 0 0 + 0 Minkoff (1999)

Advocacy to protest 0 0 + 0

Advocacy to service ++ +−

Service to advocacy ++ +−

US Rural Hospitals, 1984–1991 Change in provided services 0 − na na D’Aunno et al. (2000)

Ontario Nursing Homes, 1971–1996 Acquisition na na na + Baum et al. (2000)

California Hospitals, 1978–1991 Domain enlargement 0 + na na Haveman et al. (2001)

CEO +− na na

500 Public US Corporations, 1963–1968 Diversifying acquisitions na ++na Palmer and Barber (2001)

Largest Dutch Firms, 1966–1994 Acquisition na + na na Vermeulen and Barkema

(2001)

French Auto Manufacturers, 1885–1981 Change in niche width 0 0 + na Dobrev et al. (2001)

German Auto Manufacturers, 1885–1981 Change in niche width 0 0 + na

British Auto Manufacturers, 1885–1981 Change in niche width ++ +na

US Auto Manufacturers, 1885–1981 Change in niche width −+ +na Dobrev et al. (2003)

X/Y gives the signs of significant (

< 0.05) linear and squared terms, respectively, when estimated.

Observation period dates not given.

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 73

forces. This creates repetitive momentum, that is the

tendency to maintain direction and emphasis of

prior actions in current behaviour (Miller and

Friesen 1980). Experience with change of a particu-

lar type is thus predicted to increase the likelihood

that the change will be repeated in the future

(Amburgey and Miner 1992).

To reconcile the idea that organizational change is

propelled by repetitive momentum with evidence

that organizations move from periods of change to

periods of stability (e.g. Miller and Friesen 1980;

Tushman and Romanelli 1985), Amburgey et al.

(1993) advance a dynamic effect of prior change.

Since organizational search processes begin with

recently utilized routines (Cyert and March 1963),

the likelihood of repeating a particular change should

be highest immediately after its occurrence, but

decline as time since the change was last made

increases. Combined, the main and dynamic effects

of prior change imply that the likelihood of repeating

a particular change jumps immediately after a change

of that type, the size of the jump increasing after each

additional change, but declines with increases in the

time since that type of change last occurred.

Support for repetitive momentum in organiza-

tional change is strong: Among the estimates in

Table 3, rates of change increase with the number of

prior changes of the same type in 29 of 40 empirical

tests. Estimates for the dynamic effect are weak,

however, with only seven of 22 tests supporting this

idea. Notably, studies that control for one or both

prior change effects account for most of the support

for age-dependent structural inertia – eight of 12

negative age coefficients occur in these studies.

Support for fluidity of ageing may thus reflect a

specification bias: Older organizations may be more

likely to change not because they are older, but

because they have accumulated experience with

change that is not accounted for. Overall, these find-

ings suggest strongly the merit of a broader view of

inertial forces in organizations – one that includes

momentum in the change process as well as inertia.

Although addressing left-censoring, right-censoring

and specification, bias issues can improve our

understanding of organization-level change proces-

ses incrementally, larger gains may be made by test-

ing the underlying arguments directly. Because, as

noted earlier, age and size coefficients reveal little

about the underlying organizational processes of

interest, we still know very little about how age and

size affect rates of change, or the conditions under

which fluidity, inertia or momentum will predomi-

nate. To learn what is really going on, studies using

more direct measures of the underlying organiza-

tional processes are needed. Fluidity and inertia

arguments are not necessarily competing. Indeed,

fluidity of ageing arguments rely on inertia to create

a gap between organizations and environments.

They may thus be complementary ideas, and the

underlying relationships they predict potentially

operate simultaneously.

Is Change Beneficial?

Perhaps the most striking aspect of structural

inertia theory is the relationship hypothesized

between change in core features and the liability of

newness (Stinchcombe 1965). Hannan and Freeman

(1984: 160) propose that attempting core change

produces a renewed liability of newness by robbing

an organization’s history of survival value. Attempt-

ing change in core features lowers an organization’s

performance reliability and accountability back to

that of a new organization by destroying or render-

ing obsolete established roles and routines, and by

disrupting relations with important external actors.

Attempting core change may also undermine an

organization’s acquired legitimacy by modifying its

visible mission. Since stakeholders favour organiza-

tions that exhibit reliable performance and account-

ability for their actions, Hannan and Freeman

(1984: 160) conclude that, frequently, attempts to

change core features to promote survival – even

those that might ultimately reduce the risk of failure

by better aligning the organization with its environ-

ment – expose organizations to an increased short-

run risk of failure. Thus, structural inertia theory

predicts that organizations may often fail precisely

as a result of their attempts to survive.

In addition to their effects on reproducibility and

inertia, organizational age and size also both affect

the likelihood of surviving the short-run shock of a

core change attempt. Because their internal struc-

tures and routines are more institutionalized and

their external linkages are more established, older

organizations are particularly likely to experience

disruption as a result of core change (Hannan and

Freeman 1984: 157). In contrast, larger organiza-

tions, although less likely to attempt core changes

in the first place, are less likely to die during a

core change attempt (p. 159). Large size can buffer

organizations from the disruptive effects of core

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 74

change by, for example, helping to maintain both

old and new ways of doing things during the transi-

tion period or to overcome short-term deprivations

and competitive challenges that accompany the

change attempt.

If an organization manages to survive the short-

run shock of a core change, Hannan and Freeman

(1984: 161) predict the risk of failure to decline over

time as performance reliability is re-established,

external relationships are restabilized and organiza-

tional legitimacy is reaffirmed. The rate of decline in

the failure rate after a core change is not specified by

the structural inertia model, however. If the rate of

decline in the failure rate continues at the same rate

as before the change, the organization will face the

short-term risk of change without any long-term

benefit. If the rate of decline is slower than before, the

organization will increase both its short- and long-

term risk of failure. If, however, the rate of decline is

faster than before the change, the organization will

benefit in the long-run from taking on the short-term

risks of change. Thus, although structural inertia

theory views core change as disruptive in the short-

run, it may, ultimately, be adaptive in the long-run if

the organization manages to overcome the hazards

associated with the initial disruption.

Thus, structural inertia theory frames the ques-

tion of whether organizational change occurs at the

population level (through founding and failure) or

at the level of individual organizations as an issue of

the rate of change of organizations relative to the

rate of change in the environment. Organizations

may be unable to respond to environmental change

either because they are unwilling or unable to

change or because they fail prior to the realization of

their change efforts.

Organizational Change

and Failure

Table 4 presents findings for studies examining the

survival consequences of organizational change. As

the table shows, organizations in the study popula-

tions do not necessarily fail as a result of their efforts

to change – but they do not necessarily improve their

organizational survival chances either. Do organiza-

tions operate in a world so uncertain that adaptive

efforts turn out to be essentially random with respect

to future value (Hannan and Freeman 1984: 150)?

Thirteen of the studies in Table 4 separate the short-

and long-run effects of change, but only six of these

also test for age variation, one for size variation and

one for both age and size variation in the disruptive

effects of change. Thus, despite the relatively large

number of studies, any conclusions drawn at this

point would be premature.

In addition to the need for future research on the

adaptiveness of organizational change that specifies

structural inertia theory predictions fully, future

research may also benefit by considering several

additional issues.

Left-censored organizations, founded before the

start of the observation period, are not observed

when they are youngest and smallest and, according

to structural inertia theory, when they are both most

likely to change and most vulnerable to the hazards of

change. Including these organizations in the analysis

can lead to under-estimation of the overall hazard of

change as well as variation in the hazard for organi-

zations of different ages and sizes. Moreover, if core

organizational change is as perilous in the short-term

as structural inertia arguments assert, unless data are

fine-grained, core changes may frequently not be

observed, because these deadliest of changes will

not be detected in the yearly data typically available.

This right censoring problem lowers estimates of the

hazardousness of change because the most hazardous

changes are not included in the analysis.

Although poor and superior performing organi-

zations are likely to experience different risks of

failure as well as rates and kinds of change (e.g.

Hambrick and D’Aveni 1988; Haveman 1992;

1993a, b; 1994; Greve 1998), ecological analyses

of the effects of organizational change on failure do

not typically include measures of ongoing organiza-

tional performance (Haveman et al. 2001). This

results in two potential problems. First, cause and

effect logic is blurred because some changes or types

of change are symptoms of organizational decline

rather than causes of organizational death. Secon-

dly, model estimates are prone to specification bias:

If rates of organizational change and failure are both

influenced by recent performance, a spurious rela-

tionship between change and failure will be

observed if prior performance is not controlled.

Although organization-specific indicators of per-

formance are difficult to obtain for entire popula-

tions over time, one option is to use organizational

growth and decline as a proxy performance measure

(Baum 1990a; Scott 1992: 342–62; Haveman 1993c;

Baum and Singh 1996). Others include focusing on

shorter time periods of interest (Haveman et al.

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 75

Table 4 Organizational change and failure studies, 1984–2003

Population Type of change Change* Dynamic x Age x Size Reference

US Newspapers, 1800–1975 Editor + 0 na na Carroll (1984b)

US Business Periodicals, 1774–1865 Ownership + na na na Amburgey and Kelly (1985)

Editor 0 na na na

Name 0 na na na

Layout 0 na na na

Content 0 na na na

Voluntary Social Service Organizations, 1970–1982 Sponsor + na na na Singh et al. (1986a)

Location − na na na

Service area + na na na

Goals 0 na na na

Client groups + na na na

Chief executive − na na na

Structure 0 na na na

US Airlines, 1962–1985 Business-level 0 na na na Kelly and Amburgey (1991;

specialism see also Kelly 1988)

Business-level 0 na na na

generalism

Corporate-level 0 na na na

specialism

Corporate-level 0 na na na

generalism

Peripheral change 0 na na na

Edmonton Gasoline Stations, 1959–1988 Domain enlargement 0 + na na Usher (1991); Usher

or contraction and Evans (1996)

California Wineries, 1946–1984 Brand portfolio 0 0 na na Delacroix and Swaminathan

increase (1991; see also Swaminathan

and Delacroix 1991)

Brand portfolio decrease 0 0 na na

Product line increase − 0nana

Product line decrease 0 0 na na

Land acquisition − 0nana

Land divestment 0 0 na na

Finnish Newspapers, 1774–1963 Content +−+na Amburgey et al. (1990; 1993)

Frequency +−+na

Layout 0 −+na

Location 0 −+na

Name +−+na

California S&Ls, 1977–1987 Residential mortgages − na na na Haveman (1992)

Real estate (+ invest) 0 na na na

(Continued)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 76

Table 4 (

Continued

)

Population Type of change Change* Dynamic x Age x Size Reference

Non-residential mortgages − na na na

Mortgage-backed securities 0 na na na

Cash and investment securities − na na na

Consumer lending 0 na na na

Commercial lending 0 na na na

Service companies 0 na na na

Iowa Telephone Companies, 1900–1917 Presidential succession +−0 na Haveman (1993c)

Managerial succession +−0na

US Medical Diagnostic Imaging Expand and survive in + na na na Mitchell and Singh (1993)

Firms, 1954–1989 new subfield new subfield

Expand and exit subfield − na na na

Toronto Day Care Centres, 1971–1989 Market entry 0 0 −+Baum and Singh (1996)

Market exit +−+−

US Liberal Arts Colleges, 1971–1986 Addition of a new 0 na na na Kraatz and Zajak (1996)

programme

California Hospitals, 1980–1990 Market change + na na na Ruef (1997)

US Voluntary Associations, 1955–1986 Protest to advocacy 0 na na na Minkoff (1999)

Advocacy to protest 0 na na na

Advocacy to service + 0nana

Service to advocacy + 0nana

US Computer Manufacturers, 1975–1992 Change from proprietary −+na na Henderson (1999)

to standards strategy

US Bicycle Industry, 1880–1916 New product introduction + 0 0 na Dowell and Swaminathan (2000)

French Auto Manufacturers, 1885–1981 Niche expansion 0 0 0 na Dobrev et al. (2001)

Niche contraction 0 0 0 na

Change in relative market position 0 0 na na

Change in absolute market position + 00na

German Auto Manufacturers, 1885–1981 Niche expansion 0 0 0 na

Niche contraction 0 0 0 na

Change in relative market position − 0nana

Change in absolute market position + 00na

British Auto Manufacturers, 1885–1981 Niche expansion − 00na

Niche contraction − 00na

Change in relative market position 0 0 na na

Change in absolute market position + 00na

US Auto Manufacturers, 1885–1981 Change in niche width − 0na− Dobrev et al. (2003)

* X gives the signs of significant (

< 0.05) coefficients.

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 77

2001) or modelling the effects of organizational

change directly on observable performance indica-

tors such as market share (Greve 1998; 1999).

A closely related issue is that all organizations are

assumed to be equally susceptible to the effects of

change on failure. Hannan and Freeman (1984)

have identified age and size as factors that alter the

exposure of organizations to the liabilities of change.

However, as noted above, to date, few studies have

accounted for this variability (see Table 4). Institu-

tional linkages (i.e. ties to important state and com-

munity institutions) may also provide a buffer

against change by conferring resources and legiti-

macy on organizations (Miner et al. 1990; Baum

and Oliver 1991; Baum and Mezias 1993). Like

unmeasured performance, unmeasured variation in

organizational susceptibility to the risks of change

can bias in model estimates.

A final issue is that, although researchers do dif-

ferentiate broad categories of changes, variation in

the ecological consequences of the changes is not

considered; rather, all instances of a particular cate-

gory of change are considered equivalent. For many

kinds of change, however, there may be significant

within-type differences with substantial survival

implications. One such difference is within-type

variation in the effect of changes on the intensity of

competition. For example, depending on how an

organization’s specific actions alter the size of its

domain relative to the number of organizations

competing over its domain, the organization’s diver-

sification activities can increase, decrease or leave

unchanged the intensity of competition the organi-

zation faces. Baum and Singh (1996), for example,

found that the effects of market domain changes

(both expansion and contraction) on the failure rate

in a population of day care centres depended on

how the changes affected the intensity of competi-

tion: Changes that lowered the intensity of compe-

tition improved organizational survival chances,

while those that increased the intensity of competi-

tion lowered survival chances. Such within-type

variation may help account for some earlier mixed

results in studies examining the adaptive conse-

quences of organizational change.

Reconciling Adaptation and

Selection

Although adaptationist and ecological views are fre-

quently presented as mutually exclusive alternatives

with very different implications for organization

studies, these views are not fundamentally incom-

patible. While ecological theory emphasizes the

predominance of selection over adaptation, the

complementarity of adaptive and ecological effects

is clearly reflected in the research reviewed here.

Research in Tables 3 and 4 does not support strong

ecological arguments: organizations appear to

change frequently in response to environmental

changes, and often without any harmful effects.

Moreover, rates of change are often not constrained

by age and size as predicted by structural inertia

theory.

At the same time, however, in contrast to a strong

adaptation view, survival consequences of change

appear more consistent with random groping or imi-

tation than calculated strategic action (Delacroix and

Swaminathan 1991; Baum and Singh 1996; Ruef 1997;

Baum and Korn 1999; Dobrev et al. 2001). In his study

of California hospitals, for example, Ruef (1997)

found little correspondence between organizational

change and ecological conditions. Hospitals did not

experience improved fitness as a result of their own

adaptive efforts, but rather as a result of their with-

drawal from unsuitable market niches, wariness of

competitors to enter incumbents’ markets and ecolog-

ical drift associated with competitors’ differentiation

strategies. Consequently, while studies provide clear

evidence of organizations’ adaptive potential, they

offer little systematic support for a belief in the funda-

mental adaptability of organizations.

Taken together, empirical evidence indicates

that the evolution of organizational populations

is shaped jointly by processes of selection and

adaptation and their interaction: Because organiza-

tional change can affect organizational failure, the

population-level result of combined adaptation and

selection is not the simple aggregate of each process

separately.

One interpretation for the absence of systematic

effects of organizational change on survival chances

may be anchored in the bounded rationality of

decision-makers in organizations (March and Simon

1958). Consider, for example, the case of market

niche change, one of the most frequently studied

organizational changes (see Table 4). It seems plau-

sible that decision-makers are more knowledgeable

about the environment they face in their current

market niches from past experience than they are

about the environments of other markets. And

expanding their markets would expose them to a

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 78

different environment. Thus, intentionally rational

expansions into new markets may sometimes have

the surprising consequence of exposing the organi-

zation to unanticipated environmental conditions

(e.g. greater competition). Of course, at other times,

expansion into new markets may prove beneficial ex

post. However, on average, there would be no effect.

The problem for organizations, then, is not that

adaptive changes are impossible (or even rare), it is

that organizational decision-makers are not, on

average, able to distinguish between beneficial and

detrimental organizational changes in advance. This

suggests that organizations more experienced with a

particular kind of change may be more likely to

benefit from it.

Another explanation is that non-bureaucratic

organizations make most of the changes (Baum and

Singh 1996). In the realm of adaptiveness, the

absence of bureaucracy can be viewed as both an

advantage (change is relatively easy) and a disad-

vantage (the ability to collect information on the

environment is weak). The combination of reduced

information gathering ability with a relatively

organic organizational structure could account for

our findings – a good deal of change, but not sys-

tematically effective change. Contained in this inter-

pretation is a paradox: Organizations that can

collect a lot of information often have organiza-

tional structures that are difficult to change, while

those that can change the fastest often have trouble

collecting relevant information to guide changes.

Thus, although simplicity might lead one to expect

organizations to possess a high potential for devel-

oping adaptive actions, in fact, it may lead them

only to be active. More generally, whether changes

in response to environments have adaptive conse-

quences may depend upon the nature of learning

processes in organizations (Levinthal 1991b).

Organizational Learning

Complementing the growing interest in organiza-

tional change and its consequences is an emerging

stream of ecological research that explores the

effects of organizations’ experiential learning on

their survival chances. Experience is one of the fun-

damental mechanisms facilitating organizational

learning (Cyert and March 1963; Huber 1991).

Organizations adjust their behaviour based on their

goals and past performance. Such ‘learning

by doing’ is widely held to be a critical source of

organizational knowledge, capabilities and improved

organizational performance (Argote et al. 1990). In

addition to advancing our understanding of how

experiential processes shape organizational survival,

this research contributes to an answer to the ques-

tion of how ageing influences organizational failure

by modelling more directly experiential learning

constructs invoked by the liability of newness and

ageing hypotheses.

Organizations’ Operating and

Competitive Experience

Organizations’ learning from their operating expe-

rience has been advanced as a source of production

efficiencies and through improved efficiency, a

source of sustainable competitive advantage (Yelle

1979; Argote 1999). There is substantial evidence

that organizations become more efficient at doing

something by repeatedly doing it. The learning

curve for operations has been demonstrated in

many manufacturing (Yelle 1979) and service orga-

nizations (Darr et al. 1995). Although learning

theorists recognize that experience may ultimately

influence organizational failure (e.g. Huber 1991),

the existing empirical literature typically uses effi-

ciency as the dependent variable (Yelle 1979; Darr

et al. 1995). Efficiency is linked to failure, with more

efficient organizations being less likely to fail, but

care must be taken in relating learning curve

research to organizational failure; operating experi-

ence is not a cure-all for organizations. Learning

from its own experience can constrain an organiza-

tion by leading it into competency traps, where it

focuses on perfecting routines that are ultimately

rendered obsolete by changing environmental

demands (Levinthal and March 1993).

Organizations must allocate energy between the

exploration of new routines and the exploitation of

old ones (March 1991). An organization that

engages in too much exploration will not harvest

the value of its current competencies; one that

engages in too much exploitation can stagnate.

Typically, short run rewards of exploitation drive

out exploration, since each increase in competence

at an activity increases the likelihood of benefiting

from that activity, while returns from exploration

are less certain. Given initial success with a routine,

organizations tend to continue using it because they

know how to, and because it is less risky than

exploring alternatives (Amburgey et al. 1993).

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 79

Consequently, organizational routines tend to be

determined more by initial conditions, experiences

and actions than by information gained from later

learning situations (Stinchcombe 1965; Levitt and

March 1988).

In the face of ambiguity and uncertainty, an

emphasis on exploitation can prevent organizations

from adjusting their routines too quickly in

response to idiosyncratic events and from engaging

in costly explorations (Levinthal 1991b). In the face

of production pressures and the need for reliability

and consistency of action, exploitive learning may

enhance performance by reducing variability in

the quality or efficiency of task performance (Yelle

1979; Hannan and Freeman 1984). Exploitation can

become harmful, however, if the criteria for organi-

zational success and survival change after an organi-

zation has learned. Then the organization may

perform poorly and even fail when doing well what

it learned in the past (Levitt and March 1988).

The notion of a competency trap suggests that

organizations may reduce their exploratory activity

prematurely and, in the case of a changing environ-

ment, not renew exploratory search and learning

activities, despite the fact that new opportunities

and threats are present. In this way, organizations’

experience contributes to the inertia that binds

them to routines of the past, leading them to

employ routines well beyond their point of useful-

ness and ultimately resulting in their failure

(Starbuck 1983; Henderson and Clark 1990; Miller

1990; Baum et al. 1995).

The idea that own experience improves efficiency,

which is strongly supported by empirical evidence,

is not inconsistent with the possibility that, simulta-

neously, overall organizational effectiveness decreases

if the organization does not adjust to new demands

(Baum and Ingram 1998). Based on these argu-

ments, Baum and Ingram (1998) predicted a non-

monotonic effect of an organization’s operating

experience on its failure rate. Initially, the rate of

failure decreases with operating experience as the

organization moves down the learning curve for its

favoured set of routines. Eventually, however, accu-

mulating experience causes the organization to

become less able to adapt and, as the demands of the

environment change, more likely to fail. Supporting

this prediction, they found that the failure rates of

Manhattan hotels first fell and then increased with

increases in their own operating experience, mea-

sured in terms of accumulated room-operating

years, discounted for possible decay in the value of

operating experience over time due to forgetting

and antiquation (Argote et al. 1990).

Ingram and Baum (1997b) replicated this finding

for US hotel chains, where operating experience was

computed as the (discounted) number of hotel-

years a hotel chain had accumulated between the

time it was founded and the current year. Notably,

in neither study did incorporating operating experi-

ence in the empirical model alter the pattern of age

dependence, suggesting that the liability of newness

is due to external endorsement (not internal

routinization), and that the liability of ageing is due

to obsolescence (not senescence). This suggests that

it is organization-environment processes and not

intraorganizational phenomena that account for

liabilities of newness and obsolescence.

Most studies of organizational learning opera-

tionalize organizational experience based on historical

operations, corresponding to operating experience.

When it involves interactions with competitors,

however, experience may also contribute to the

external capability of organizations by helping to

improve their models of markets and competitive

interaction, which contribute critically to effective

strategizing and success. Organizations thus face

both internal and external strategic demands

(Saloner 1994). Operational and competitive organi-

zational experiences correspond, respectively, to the

internal and external differentiation of strategic

demands, and together contribute to satisfying all of

the strategic and competitive demands faced by an

organization. Two recent studies advance measures

of organizations’ competitive experience to test the

idea that organizations become more adept com-

petitors by observing the competitive actions and

outcomes of other organizations.

Barnett et al. (1994) consider the average histori-

cal number of competitors a retail bank has faced in

its past (see also Barnett 1997). They found oppos-

ing effects of competitive experience accumulated

before and after bank deregulation on Illinois

banks’ return on assets. Before deregulation, com-

petitive experience was negatively related to the per-

formance of single unit banks; afterward it was

positively related to single unit banks’ performance.

Performance of multi-unit banks was not affected

by the average historical number of competitors

either before or after deregulation. Barnett et al.

(1994) interpret their findings as suggesting that

simpler organizational forms are more able and

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 80

likely to learn from their competitive experience

than more complex forms.

Elaborating this approach, Barnett and Hansen

(1996) modelled competitive organizational experi-

ence as the aggregate number of years that an orga-

nization had been exposed to competition from

rival organizations. They also differentiate between

an organization’s recent and distant competitive

experience, expecting that recent competitive expe-

rience (accumulated within the past 10 years) would

be beneficial, while older competitive experience

would be detrimental (accumulated more than 10

years earlier). Moreover, they consider not only an

organization’s own competitive experience, but also

the competitive experience accumulated by its com-

petitors. The latter is a sum of the competitors’

exposure to their own rivals (both recent and dis-

tant). Again using data on Illinois banks, they find

support for their ideas – a bank’s recent competitive

experience reduced its failure rate, while its distant

competitive experience increases it; and, moreover,

a bank’s competitors’ recent competitive experience

increased the bank’s failure rate while its competi-

tors’ distant competitive experience lowered it.

Barnett and Hansen view their results as supporting

the existence of a ‘Red Queen’ effect in which com-

peting organizations search for ways to improve

their performance, increasing their competitive

strength, triggering a further cycle of learning (and

so on), and, as a result, success and survival become

increasingly difficult.

Recent ecological studies of organizational learn-

ing have advanced our understanding of the role of

experiential processes in organizational adaptation

and selection. Experiential learning is a double-edged

sword – it can enhance organizational survival

chances while at the same time hindering adaptation;

it can strengthen competitive abilities while at the

same time making competition harder. To shine

additional light on the complexities and tradeoffs

associated with organizational learning, future

research is needed to refine and link typologies of

organizational learning – exploration vs exploitation,

local vs non-local, recent vs distal, operational vs

competitive – and to identify the types of experiences

that are most likely to trigger learning.

Between-organization differences in learning also

deserve examination. Organizations likely vary in

their capacities for experiential learning, and so,

while organizations with superior capacities can

improve their survival chances, the fates of other

weaker learners may be either unaffected (or even

harmed) by their efforts to learn. Differences in

internal information processing capabilities, for

example, may generate learning advantages (Cyert

et al. 1993). Incorporating heterogeneity in organi-

zational learning into the approaches outlined here

would extend them in an important way.

Ecological Processes

Niche Width Dynamics

Niche width theory (Hannan and Freeman 1977)

focuses on two aspects of environmental variability

to explain differential survival of specialists, which

possess few slack resources and concentrate exploit-

ing a narrow range of customers, and generalists,

which appeal to the mass market and exhibit toler-

ance for more varied environments. Variability

refers to environmental fluctuations about the mean

over time. Grain refers to the patchiness of these

fluctuations with frequent variations termed fine-

grained and periodic termed coarse-grained.

Niche width theory builds on the idea that a spe-

cialist well designed for a particular environmental

condition will always outperform a generalist in that

same condition, because the generalist must carry

extra capacity that sustains its ability to perform in

other environmental conditions. Thus, the specialist

‘maximizes its exploitation of the environment and

accepts the risk of having that environment change’

while the generalist ‘accepts a lower level of exploita-

tion in return for greater security’ (Hannan and

Freeman 1977: 948).

Niche width theory predicts that specialists per-

form better in stable or certain environments and

in fine grained environments. In contrast, when

environmental variability is high and coarse

grained, specialists have trouble outlasting the long

unfavourable periods and the generalist strategy

conveys advantage. The key prediction is that, in

fine-grained environments with large magnitude

variations relative to organizational tolerances, spe-

cialists out-compete generalists regardless of envi-

ronmental uncertainty. Specialists ride out the

fluctuations; generalists are unable to respond

quickly enough to operate efficiently. Thus, niche-

width theory challenges the classical contingency

theory prediction that uncertain environments

always favour generalists that spread their risk.

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 81

In contrast to niche width theory, which predicts

that for a given population one optimal strategy

exists, Carroll (1985) proposes that, in environ-

ments characterized by economies of scale, compe-

tition among generalists to occupy the centre of the

market where resources are most abundant frees

peripheral resources that are most likely to be used

by specialists. He refers to the process generating

this outcome as resource partitioning. His model

implies that, in concentrated markets with a few

generalists, specialists can exploit more of the avail-

able resources without engaging in direct competi-

tion with generalists. It also implies that economies

of scale and scope may be sufficiently large that they

overwhelm any costs of sustaining extra capacity to

maintain a broad niche, thus giving advantage to

generalists. Based on these ideas, resource partition-

ing theory yields the novel prediction that increas-

ing market concentration increases the failure rate

of generalists operating in the centre of the market

and lowers the failure rate of specialists operating at

its periphery.

Resource-partitioning theory has been sup-

ported in studies of newspapers (Carroll 1985;

1987; Dobrev 2000; Boone et al. 2002), American

breweries and microbreweries (Carroll and

Swaminathan 1992; 2000), Rural Cooperative Banks

in Italy (Freeman and Lomi 1994), US farm winer-

ies (Swaminathan 1995), Dutch auditing firms

(Boone et al. 2000), US feature film producers and

distributors (Mezias and Mezias 2000) and US and

European automobile producers (Dobrev et al.

2001; 2002). Each of these studies provides at least

partial support for the theory. For example, in a

study of founding rates of specialist farm wineries

over a 50-year period starting shortly after the

end of Prohibition, Swaminathan (1995) found that

farm winery foundings were lower in states with

more generalist mass-production wineries, sug-

gesting localized competition between these two

organizational forms. Consistent with resource par-

titioning theory, however, increasing concentration

of mass-producers increased the farm winery

founding rate.

Although the specialist-generalist distinction is

common in ecological research, tests of niche-width

theory’s predictions are scarce, and studies of

resource-partitioning theory do not typically explic-

itly contrast niche-width and resource-partitioning

predictions. Yet, a tension exists between these two

theories of niche width dynamics. They define

specialism and generalism differently and make

different claims regarding the tradeoff between

organizational fitness and niche width. The theories

also focus on different kinds of niches (Carroll et al.

2002). Niche width theory emphasizes fundamental

niches, that is, the set of environments in which an

organization can survive in the absence of competi-

tion from other entities. Resource partitioning

theory’s focus is on an organization’s resource loca-

tion relative to other organizations. As such, it

emphasizes realized niches – the subset of the funda-

mental niche in which an organization can survive

in the presence of competitors. This difference is

important because the absence of overlap between

realized niches takes on a very different meaning

depending on whether or not the underlying funda-

mental niches overlap.

Given these differences, it is an open question

whether the two theories differ fundamentally or

whether they can be integrated into a unified theory.

However, the recent flurry of theoretical efforts

(Hannan et al. 2003a; b), empirical work on

resource partitioning (for a review, see Carroll et al.

2002) and initial efforts to link the two theories

(Dobrev et al. 2001) provide a foundation for a

potential integration that would see niche width

theory’s emphasis on environmental dynamics

joined with resource partitioning’s emphasis on

scale advantages.

Population Dynamics and Density

Dependence

Recent research on founding and failure in organi-

zational ecology has paid considerable attention to

population-level processes of population dynamics,

the number of prior foundings and failures in a

population, and population density, the number of

organizations in a population. Although related,

these perspectives are not identical – population

dynamics focuses on how current founding and fail-

ure rates are more related to changes in density,

while density dependence focuses on levels of den-

sity itself (Tucker et al. 1988).

Density-dependent explanations for founding

and failure are broadly similar, though not identical.

Initial increases in population density can increase

the institutional legitimacy of a population. The

capacity of a population’s members to acquire

resources increases greatly when those controlling

resources take the organizational form for granted.

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 82

However, as a population continues to grow, the

nature of interdependence among a population’s

members becomes competitive. When there are few

organizations in a population, competition with

others for scarce common resources can easily be

avoided. However, as the number of potential

competitors grows, avoidance becomes more diffi-

cult. Combined, the mutualistic effects of initial

increases in density and the competitive effects of

further increases suggest curvilinear effects of den-

sity on founding and failure rates in organizational

populations (Hannan and Freeman 1989; Hannan

and Carroll 1992).

Hannan and Freeman (1989), Hannan and Carroll

(1992) and others provide substantial empirical sup-

port for the curvilinear relationships predicted by the

density dependence model in a variety of organiza-

tional populations. By comparison, although often

significant, population dynamics findings are mixed

(for reviews, see Singh and Lumsden 1990; Aldrich

and Wiedenmayer 1993). Moreover, as illustrated in

Table 5, when population dynamics and population

density are modelled together, with few exceptions

(e.g. Tucker et al. 1988; Delacroix et al. 1989), recent

studies find population dynamics effects are generally

weaker and less robust (for a review of earlier work,

see Singh and Lumsden 1990). Even Delacroix and

Carroll’s (1983) original Argentina and Ireland news-

paper population findings do not hold up when

density is introduced in a re-analysis of their data

(Carroll and Hannan 1989b).

There are several possible explanations for the

apparent dominance of density dependence

processes over population dynamics processes. One

is the more systematic character of density relative

to the transitory nature of changes in density that

result from ongoing foundings and failures. A

related explanation is that the effects of foundings

and failures on resource availability are more transi-

tory than the yearly data that is typically available

are able to detect. For example, if foundings and

failures keep pace with each other during a year,

density will remain stable, but population dynamics

will vary throughout the year (Aldrich and

Wiedenmayer 1993). A third explanation is the

much greater sensitivity of estimates for quadratic

specifications of prior foundings and failures to

outliers. This is especially true for prior foundings,

which are usually Poisson distributed (Hannan and

Freeman 1989). These issues need to be researched

more thoroughly before population dynamics

effects are abandoned, which has been the trend in

recent research (e.g. Baum and Mezias 1992; Rao

and Neilsen 1992; Amburgey et al. 1993).

Elaborations of the Density

Dependence Model

Although support for density dependence theory is

quite strong, it has not been without its critics.

Density dependence theory has received some criti-

cal attention for its proposed integration of ecolog-

ical and institutional perspectives (e.g. Zucker 1989;

Delacroix and Rao 1994; Baum and Powell 1995).

Some authors have questioned the implicit assump-

tion that each organization in a population influ-

ences and is influenced by competition equally (e.g.

Singh and Lumsden 1990; Winter 1990; Baum and

Mezias 1992). In a methodological critique, Petersen

and Koput (1991) argue that the negative effect of

initial increases in population density on the failure

rate may result from unobserved heterogeneity in

the population (for a reply on this point, see Hannan

et al. 1991). Singh (1993) observes that some of the

debate about density dependence stems from the

model’s main strength, its generality, which has

been achieved at the expense of precision of mea-

surement and realism of context. Singh (1993: 471)

concludes that ‘we have ample general supportive

evidence for density dependence, we may do well to

sacrifice some generality, provided it moves the

research toward greater precision and realism’.

Density effects are clear empirically, but the specific

conditions that generate legitimacy and competi-

tion are more ambiguous – defined by outcomes

rather than by substance. Thus, the precise interpre-

tation of the extensive density dependence findings

needs to be explored further.

Elaborations, respecifications and new measures

have been advanced to address questions raised

by the initial density dependence formulation.

Although Hannan and Carroll (1992: 38–9, 71–4)

have questioned some of these developments, they

appear to hold real promise for improving the pre-

cision and realism of the density dependence model

with respect to both legitimation and competition.

Additionally, the focus of several approaches on

organization-level differences within populations

may provide a bridge between organizational ecolo-

gists and researchers who focus primarily on the

organizational level of analysis. These develop-

ments, summarized in Table 6, are reviewed below.

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 83

Table 5 Population dynamics and density dependence studies, 1986–2003*

Population Found Fail Density Event References

US Labor Unions, 1836–1985 +/− na +/− Founding Hannan (1986); Hannan and Freeman (1989);

Carroll and Hannan (1989a, b);

Carroll and Swaminathan (1991);

Hannan and Carroll (1992)

na 0 −/+ Failure

US Brewers, 1633–1988 +/− 0 +/− Founding

+−−/+ Failure

San Francisco Newspapers, 1800–1975 0/0 −/++/− Founding

0/0 +/−−/+ Failure

Argentina Newspapers, 1800–1900 −/+ 0/−+/− Founding

0/0 +/0 −/+ Failure

Ireland Newspapers, 1800–1975 +/− 0/0 +/− Founding

0/0 +/−−/+ Failure

Little Rock Newspapers, 1815–1975 0/0 0/++/− Founding

+/− 0/0 0/0 Failure

Springfield Newspapers, 1835–1975 0/+ 0/++/− Founding

0/0 0/0 0/+ Failure

Shreveport Newspapers, 1840–1975 0/0 0/0 +/− Founding

+/0 0/0 0/0 Failure

Elmira Newspapers, 1815–1975 +/0 0/0 +/0 Founding

+/0 0/0 0/0 Failure

Lubbock Newspapers, 1890–1975 0/0 0 0/0 Founding

0/0 0 0/− Failure

Lafayette Newspapers, 1835–1975 na 0 0/0 Founding

0/0 0 0/0 Failure

California Wineries, 1940–1985 −−0/0 Failure Delacroix et al. (1989)

Pennsylvania Telephone Cos., 1879–1934 +−−/+ Founding Barnett and Amburgey (1990)

US semiconductor firms −/++/ −/+ Failure Freeman (1990)

Metro Toronto Day Care Centres, 1971–1989 + 0 +/− Founding Baum and Oliver (1992)

0 −−/+ Failure

US State Life Insurance Cos., 1759–1937 +/− na +/− Founding Ranger-Moore et al. (1991)

Manhattan Banks, 1840–1976 +/0 na +/− Founding Banaszak-Holl (1992; 1993)

NY State Life Insurance Cos., 1842–1904 −−0/+ Founding Budros (1993; 1994)

German Breweries, 1861–1988 −/0 na +/− Founding Carroll et al. (1993)

US Trade Associations, 1901–1990 0/0 0/0 +/− Founding Aldrich et al. (1994)

+−−/+ Failure

(Continued)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 84

Table 5 (

Continued

)

Population Found Fail Density Event References

US Medical Equipment Manufacturers, 1950–1990 + na −/+ Failure Mitchell (1994)

Manhattan Fax Transmission Cos., 1965–1992 −/+ 0 +/− Founding Baum et al. (1995)

Pre-dominant design cohort −−+/0 Failure

Post dominant design cohort −/+ 0 +/− Founding

00−/+ Failure

All US Immigrant Newspapers, 1877–1914 0 −+/− Founding West (1995)

Polish US Immigrant Newspapers, 1877–1914 −−+/+ Founding

Swedish US Immigrant Newspapers, 1877–1914 0 0 +/− Founding

Italian US Immigrant Newspapers, 1877–1914 − 0 +/− Founding

Belgian Automobile Manufacturers, 1885–1981 +/0 na +/− Founding Hannan et al. (1995);

Dobrev et al. (2001)

British Automobile Manufacturers, 1885–1981 +/0 na +/− Founding

0na−/0 Failure

French Automobile Manufacturers, 1885–1981 +/0 na +/− Founding

0na−/+ Failure

German Automobile Manufacturers, 1885–1981 0/0 na +/− Founding

0na0/+ Failure

Italian Automobile Manufacturers, 1885–1981 +/0 na +/− Founding

Niagara Falls Hotels, 1885–1991 + na +/− Founding Ingram and Inman (1996)

na 0/0 −/+ Failure

Massachusetts Railroads, 1826–1922 + 0 +/− Founding Dobbin and Dowd (1997)

Finnish Newspapers, 1771–1963 + na +/− Founding Dacin (1997)

New York State Life Insurance Companies, 1813–1985 + na 0/0 Failure Ranger-Moore (1997)

US Breweries, 1845–1918 +−0/− Founding Wade et al. (1998)

+−−/+ Failure

Dutch Accounting Firms, 1880–1990 0 + 0 Failure Pennings et al. (1998)

Egyptian Investment Firms, 1974–1989 +/0 0 0 Founding Messalam (1998)

US Computer Manufacturers, 1975–1992 − na 0 Failure Henderson (1999)

New York City Credit Unions, 1914–1990 0 na 0/+ Failure Barron (1999)

US Specialist Feature Film Producers, 1912–1929 +/− na −/+ Founding Mezias and Mezias (2000)

US Specialist Feature Film Distributors, 1912–1929 0 na −/+ Founding

US Wineries, 1941–1990 + na −/+ Failure Swaminathan (2001)

Bulgarian Newspapers, 1846–1992 +/− na +/− Founding Dobrev (2001)

Illionois Banks, 1900–1993 −++/− Founding Barnett and Sorenson (2002)

* Includes only analyses that estimate both population dynamics and density dependence effects; many additional studies estimate density dependence effects

alone. X/Y gives the signs of significant (

< 0.05) linear and squared terms, respectively.

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 85

Table 6 Elaborations of the density dependence model, 1989–2003

Model Key variables Nature of elaboration References

Institutional processes

Density delay Population density at founding Adds an imprinting effect of density at founding Carroll and Hannan (1989a);

to the original formulation. Helps explain Hannan and Carroll (1992)

the commonly observed decline in population

density in older populations

Institutional Relational density (number of Attempts to explain the legitimation of an Baum and Oliver (1992);

embeddedness linkages between a population organizational form in terms of endorsements Hybels et al. (1994)

and the institutional environment) by powerful actors and organizations

Non-density-based Certification contests and media- Models legitimation effects with non-density- Rao (1994); Hybels (1994);

measures of based content measures based measures of institutionalization Lamertz and Baum (1998);

legitimacy Deeds et al. (2004)

Competitive processes

Level of analysis City density; state density; Attempts to uncover the appropriate level of Carroll and Wade (1991);

regional density; national density analysis to study density-dependent Swaminathan and

(population density at various processes by comparing patterns of density Wiedenmayer (1991);

levels of geographic aggregation) dependence across multiple levels of analysis Hannan et al. (1995)

Localized Size similarity; price similarity; Re-specifies competition effect Hannan et al. (1990);

competition location similarity (population density of population density by allowing more Baum and Mezias (1992);

weighted by the size of differences similar organizations to compete at a greater Ranger-Moore et al. (1995);

in various organizational features) level of intensely Baum and Haveman (1997)

Organizational Overlap density; non-overlap Re-specifies population density by Baum and Singh (1994b, c);

niche overlap density (population density disaggregating it into competitive and Baum and Oliver (1996)

weighted by (non)overlap in mutualistic components using information on

resource needs of organizations) the overlap and non-overlap of resource

requirement of population members

Accounting for concentration

Mass dependence Population mass (population Re-specifies competition effect of population Barnett and Amburgey (1990)

density weighted by density by allowing larger organizations

organizational size) to generate stronger competition. Helps

explain the tendency toward concentration

in organizational populations

(Continued)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 86

Table 6 (

Continued

)

Model Key variables Nature of elaboration References

Changing basis Population age × population As the basis of competition in a population Baum (1995)

of competition density

, population mass, evolves from

-selection to

size similarity and differentiation, the strength of density-

concentration dependent competition declines, and

the strength of mass-dependent,

size-localized and concentration-based

competition increase

Competitive Organizational age × size; Organizations exposed to greater Barnett et al. (1994);

intensity average historical density competition over their lifetimes Barnett (1997)

generate the strongest competition

Coupled clocks Population age × population As a population matures, its density Hannan (1997);

density and density

becomes decoupled from legitimacy Hannan et al. (1998a, b);

and competition Dobrev et al. (2001)

Dynamic selection Mean of the log size distribution; As a population’s density approaches its Barron (1999)

organizational size × population carrying capacity, large organizations are

density better able to face the more intense

competition, increasing their survival

chances. The founding rate declines as a

population’s size distribution shifts right

and incumbents establish reputations,

customer loyalties and staff expertise

Scale-based Aggregate distance of an An organization’s competitive (dis) Dobrev and Carroll (2003)

selection organization to all larger advantage is a function of its position in

organizations in the population the size distribution of its population

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 87

Density and Institutional

Processes

Drawing on the neo-institutional literature (Meyer

and Rowan 1977; Zucker 1977; DiMaggio and

Powell 1983), organizational ecologists draw a dis-

tinction between cognitive and sociopolitical legiti-

macy (Aldrich and Fiol 1994). Zucker (1977) treats

institutionalization as a process, emphasizing that

legitimacy is a cognitive phenomenon reflected in

taken-for-granted assumptions. Meyer and Rowan

(1977) and DiMaggio and Powell (1983) stress that

legitimacy is embedded in relational networks and

normative codes of conduct. Thus, they viewed

institutionalization as both a sociopolitical process

through which certain organizational forms come

to be regarded as obligatory, and as a state in which

organizational forms are butressed by legal man-

date or by widely shared cultural, professional and

political norms and values. From a cognitive legit-

imacy perspective, an organizational form is legit-

imated ‘when there is little question in the minds

of actors that it serves as the natural way to effect

some kind of collective action’ (Hannan and

Carroll 1992: 34).

The sociopolitical approach emphasizes how

embeddedness in relational and normative contexts

influences an organizational form’s legitimacy by

signalling its conformity to social and institutional

expectations. Although institutionalists view these

two facets of legitimation as complementary and

fundamentally interrelated, density dependence

theory emphasizes only cognitive legitimacy.

Although cognitive legitimacy may be achieved

without sociopolitical approval, sociopolitical legit-

imacy is considered a vital source of, or impediment

to, cognitive legitimacy. Indeed, since contemporary

organizational populations rarely operate in isola-

tion from the state, the professions and broader

societal influences sociopolitical legitimacy cannot

be ignored (Baum and Oliver 1992; Baum and

Powell 1995).

Hannan and colleagues defend their exclusive

focus on population density with a series of interre-

lated arguments. The first claim is that legitimacy

defies measurement. ‘[R]econstructing the exact

details of changing levels of legitimation … over the

history of any one population demands attention to

all of the unique features of that population’

(Hannan and Carroll 1992: 37). Indeed, ‘[a] direct

measure of legitimation as we define it requires

learning what fraction of relevant individuals take a

particular organizational form for granted’

(Hannan and Carroll 1992: 38–9), thus severely lim-

iting the historical scope of analysis. This assertion

leads them to argue ‘… that theories of legitimation

and competition can be studied systematically and

comparatively only by testing their implications for

the relationships between other observables’

(Hannan and Carroll 1992: 39). Consequently, it

makes sense to study variables that can be easily

observed and compared across populations and

over long periods of time, such as density and vital

rates. This indirect measurement approach has gen-

erated criticism because conforming findings can-

not be interpreted precisely; density coefficients

reveal little about the theoretical explanations

designed to account for them (Singh 1993).

Critics question the legitimacy interpretation of

density effects, suggesting that legitimation is

invoked ex-post and that density estimates are

proxies for a wide range of other possible effects. In

her provocative commentary, Zucker (1989) takes

Hannan and colleagues to task for invoking the con-

cept of legitimation ex-post to explain the effects of

density on founding and failure rates, and suggests

that density estimates are proxies for other effects

(see also Petersen and Koput 1991; Miner 1993). She

advocates the use of more direct and precise mea-

sures of the underlying institutional processes.

Building on Zucker’s critique, Baum and Powell

(1995) argue that the point is not that legitimacy

must be measured directly, but rather that it should

be gauged using different observables from organiza-

tional density. Of course, there are sound method-

ological reasons for multiple measures. In their essay,

Delacroix and Rao (1994) provide a more general

critique that suggests that the density-dependence

conception of legitimation bundles together at least

three kinds of externalities – reputational, vicarious

learning, and infrastructural – only one of which is

related to legitimation.

These critiques appear to have led to the density-

as-process argument, in which legitimation is no

longer a variable to be measured, but a process that

relates density to founding and failure. Thus,

Hannan and Carroll (1992: 69) make the strong

claim that ‘growth in density controls … [legitima-

tion] processes – it does not reflect them’. These

competing proxy and process views suggest different

effects of adding covariates. According to Hannan

and Carroll (1992: 70–1), if density is an indirect

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 88

indicator, measuring legitimation more directly and

precisely would dampen the first-order effects of

density or lead them to disappear altogether. How-

ever, from the density-as-process view, the inclusion

of such covariates implies a sharpening and strength-

ening of density’s legitimation effects.

Institutional Embeddedness and

Sociopolitical Legitimacy

Baum and Oliver (1992) address exactly this ques-

tion (see also Singh 1993: 471). They argue that an

important limitation of the density dependence

model is that it focuses exclusively on cognitive

legitimacy and interdependencies among organiza-

tions within populations and neglects the evolution

of a population’s interdependencies with surround-

ing organizations and institutions. However, where

relations with community and government are

dense, these institutional actors may exert consider-

able influence over the conditions that regulate

competition for scarce resources and legitimacy in

the population. Baum and Oliver (1992) propose

an alternative hypothesis in which legitimation is

explained in terms of the institutional embedded-

ness of a population in its institutional environment

influences the population’s sociopolitical legitimacy

by signalling its conformity to social and institu-

tional expectations (Meyer and Rowan 1977;

Aldrich and Fiol 1994). Institutional embeddedness

refers to interconnections between a population and

its institutional environment (DiMaggio and Powell

1983; Fombrun 1986; 1988).

According to institutional theory, conformity to

the norms and social expectations of the institu-

tional environment improves an organization’s sur-

vival chances significantly (Meyer and Rowan 1977;

Scott and Meyer 1983; Oliver 1991). When organi-

zations establish ties to reputable societal institu-

tions, they typically obtain benefits that contribute

to their likelihood of survival. These benefits

include greater legitimacy and status (Scott and

Meyer 1983; Singh et al. 1986b; Baum and Oliver

1991), and enhanced resource access and pre-

dictability (Pfeffer and Salancik 1978; Aldrich and

Auster 1986; Miner et al. 1990). Thus, the endorse-

ment of an organization’s practices by a community

or state agency increases the organization’s legiti-

macy and enhances the organization’s ability to

attract clients and resources (Wiewel and Hunter

1985).

Organizational population growth is typically

accompanied by increasing institutional embedded-

ness (DiMaggio and Powell 1983; Zucker 1989). As

a population grows and its social or public impact

becomes more widely recognized, community advo-

cacy groups, government agencies, professional

associations and other social actors take an increas-

ingly active role in monitoring population

members’ activities, distributing endorsements and

rewards and shaping the rules and standards about

what are legitimate activities and outputs for the

population. Institutional theorists predict that this

increasing interconnectedness between a popula-

tion and its institutional environment enhances the

growth and survival of the population over time

(Scott and Meyer 1983). For example, Meyer and

Rowan (1977: 352) suggest that ‘the long-run sur-

vival prospects of organizations increase as state

structures elaborate and as organizations respond to

institutionalized rules … schools, hospitals and wel-

fare organizations show considerable ability to sur-

vive, precisely because they are … almost absorbed

by their institutional environments’.

Baum and Oliver model institutional embedded-

ness with relational density, defined as the number

of formal relations between the members of a pop-

ulation and key institutions in the population’s

environment. Key institutions refer to government

agencies and community organizations in a popula-

tion’s environment. Although initial estimates in

their study of Metropolitan Toronto day care centres

support the curvilinear density dependence predic-

tions for both founding and failure rates, legitima-

tion effects of initial increases in organizational

density disappear after inclusion of relational den-

sity and the relationship between density and

founding and failure rates became purely competi-

tive. Relational density, in contrast, exhibited either

predicted curvilinear effects or purely mutualistic

effects.

Baum and Oliver thus test the proxy-vs-process

prediction, and their results support the proxy view.

However, as Baum and Oliver (1992: 556) point out,

their results also support ecological explanations of

the underlying institutional processes. Since organi-

zational populations rarely operate in isolation

from state and community institutions, future

research incorporating both population and rela-

tional densities may provide further clarification of

the role of institutional processes in population

dynamics. Baum and Oliver’s findings were

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 89

replicated in Hybels et al.’s (1994) study of founding

of US biotechnology firms in which vertical (input

and output) strategic alliances are used to measure

industry embeddedness in relational and institu-

tional contexts. These studies suggest that the initial

density-as-proxy formulation of legitimacy was

more accurate and, in addition, that organizational

density may be a proxy for relational as well as (or

instead of) cognitive legitimacy.

Non-Density-Based Measures of

Legitimacy

Density dependence theory’s exclusive emphasis on

one facet of legitimation misses its multidimen-

sional nature (Baum and Powell 1995). Researchers

have, however, begun to measure other aspects of

legitimation and examine how diverse social

processes combine with organizational density to

contribute to the legitimacy of organizational

forms, and several promising non-density-based

alternatives to studying legitimation have recently

been examined.

In many industries, special purpose organizations

institute certification contests to evaluate products

or firms and rank-order participants according to

their performance on preset criteria. Certification

contests offer a common social test of products and

organizations that serve as a social diffusion mecha-

nism. Rao (1994) argues that cumulative victories in

such certification contests enhance organizational

reputations in the eyes of risk-averse consumers and

financiers, improving their access to resources and

their survival chances. Moreover, Rao argues that,

by increasing opportunities for certification and dif-

fusing knowledge about organizations and their

products, these contests establish the identity and

legitimacy of a product and its producers, lowering

the risk of producer failure. His analysis of the early

American automobile industry supports these ideas,

demonstrating that winning heavily publicized road

races improved the survival chances of individual

automobile manufacturers and, in addition, that the

cumulative prevalence of contests lowers the aggre-

gate failure rate. In addition to certification contests,

a wide range of accreditation, certification and cre-

dentialing activities signal reliability, raising the

sociopolitical legitimacy of organizational forms, as

well as contributing to their cognitive legitimacy by

spreading knowledge about them (Baum and Powell

1995).

Another basic source of information diffusion

about the activities of an organizational form are

the print media. Detailed archives of media cover-

age exist for many industries and content analyses

of these public records offer a potentially powerful

technique for operationalizing legitimation. Measure-

ment of this kind is used widely in social movement

research (e.g. Olzak 1992; Tilly 1993). Content-

based measures promise high comparability across

settings covered by the business press as well as tem-

poral comparability within a given context. Hybels

(1994) successfully employed media-based mea-

sures of legitimacy in an analysis of US biotechnol-

ogy firm foundings, as have Pollock and Rindova

(2003) and Deeds et al. (2004) in studies of US

biotechnology firm initial public offerings. In a

related study, Lamertz and Baum (1998) used media

accounts to track the legitimation of management

downsizing in Canada.

Density and Competitive

Processes

Density dependence theory assumes that the inten-

sity of competition depends on the number of orga-

nizations in a population. Some researchers,

however, question the assumption implicit in this

approach that all members of a population are

equivalent, with each member assumed to compete

for the same scarce resources and to contribute to

and experience competition equally (e.g. Winter

1990: 286). Although research demonstrates that

this assumption may be a reasonable starting

approximation, theory in organizational ecology

suggests that the intensity of competition between

organizations in a population is largely a function of

their similarity in resource requirements: the more

similar the resource requirements, the greater the

potential for intense competition (e.g. Hannan and

Freeman 1977; 1989; McPherson 1983).

1 If all orga-

nizations in a population are not equal competitors,

then population density may not provide the most

precise measure of the competition faced by differ-

ent organizations in a population.

Building on this logic, three elaborations of the

density dependence model – localized competition,

levels of analysis and organizational niche overlap –

that incorporate organizational differences explicitly

to specify competitive processes within organizational

populations more precisely have been advanced.

Each of these refinements to the original model

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 90

enriches the ecological approach to competition by

integrating ideas on population microstructures

into density dependence research.

Localized Competition

Following a long tradition in organizational sociol-

ogy, Hannan and Freeman (1977: 945–6) propose

that organizations of different sizes use different

strategies and structures. And, as a result, they pro-

pose that, although organizations of different sizes

are engaged in similar activities, large and small-sized

organizations depend on different mixes of resources.

This implies that organizations compete most

intensely with similarly-sized organizations. For

example, if large and small organizations depend on

different resources (e.g. large hotels depend on con-

ventions while small hotels depend on individual

travellers), then patterns of resource use will be spe-

cialized to segments of the size distribution.

Consequently, competition between large and small

organizations will be less intense than competition

among large or small organizations. Large organiza-

tions may, however, pose a threat to medium-sized

organizations. Whatever strategy medium-sized

organizations adopt to compete with large organiza-

tions will make them more vulnerable to competition

from small organizations and vice versa. Therefore,

the emergence of large organizations should be

accompanied by a decline in the number of

medium-sized organizations, while small ones flour-

ish as their most intense competitors are removed

from the environment. Thus, size-localized competi-

tion may also play a role in the consolidation of orga-

nizational populations over time.

Although size-localized competition did not

receive empirical attention initially (Hannan et al.

1990), studies of Manhattan banks (Banaszak-Holl

1992), Manhattan hotels (Baum and Mezias 1992),

US health maintenance organizations (Wholey et al.

1992) and Niagara Falls hotels (Ingram and Inman

1996) now provide empirical evidence of that size-

localized competition raises organizational failure

rates. Ranger-Moore et al. (1995) have also demon-

strated that size-localized competition dampens

organizational growth rates in the New York State

life insurance industry. These findings demonstrate

that the intensity of competition faced by organi-

zations in a population depends not only on the

number (i.e. density) of other organizations in the

population, but on their relative sizes as well. Baum

and Mezias (1992) generalize the size-localized

model to other organizational dimensions and show

that, in addition to similar-sized organizations,

competition within a population can be more

intense for organizations that are geographically

proximate and charge similar prices, lowering their

survival prospects. Baum and Haveman (1997)

showed that localized competitive processes also

shape key entrepreneurial decisions at the time of

founding, namely how near new organizations

should locate in product and geographic space to

other firms in their industry.

Future research on localized competition can

offer direct insights into the dynamics of organiza-

tional diversity. Localized competition models

imply a pattern of disruptive or segregating selec-

tion (Baum 1990b; 2006; Amburgey et al. 1994;

Baum and Haveman 1997) in which competition

between like entities for finite resources leads even-

tually to differentiation (see also, Durkheim 1933;

Hawley 1950: 201–3). This mode of selection, which

has not been emphasized in the ecological literature,

tends to increase organizational differentiation by

producing gaps rather than smooth, continuous

variation in the distribution of the members of a

population along some organizational dimension.

Level of Analysis

Recurrent patterns of organizational concentration

in space across different industries and national

contexts suggests that location may be a general

factor shaping the evolution of organizational pop-

ulations. If forces exist that give advantages to orga-

nizations located near other organizations or in

specific geographical areas, then the internal struc-

ture of organizational populations cannot be con-

sidered homogeneous and organizational birth and

death rates will vary systematically across locations.

It is essential, therefore, to select the appropriate

level of analysis to examine the dynamics of popula-

tions, since different levels of spatial aggregation

imply different assumptions about how general

processes of legitimation and competition unfold.

Although density dependence theory assumes

implicitly that geographically distant members of an

organizational population compete with each other

at intensity equal to neighbouring organizations,

several studies have refined this assumption by disag-

gregating population density according to geographic

proximity to explore the geographic boundaries on

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 91

competitive (and institutional) processes. For example,

Hannan and Carroll (1992), Carroll and Wade (1991)

and Swaminathan and Wiedenmayer (1991) analysed

density dependence in founding and failure rates of

US and German Breweries at a variety of level of geo-

graphic aggregation (e.g. city, state, nation). These

studies estimated the density dependence model sep-

arately for each level of analysis and then compared

coefficients across levels. In their analysis of the US

brewing industry, Carroll and Wade (1991) found

support for density dependence at the state and

regional levels but not city level, while Hannan and

Carroll (1992) found that founding rates of newspa-

per organizations depend on density in a manner

consistent with theory only in small metropolitan

areas. Although these studies vary in their support for

the density dependence model, estimates consistently

reveal stronger competitive effects among breweries

for populations defined at more local geographic

levels than for national populations as a whole, while

local and national legitimation effects were similar in

magnitude.

Taken together, these studies, along with other

research examining effects of geographic levels (e.g.

Rao and Neilsen 1992; Budros 1993; 1994; Baum and

Singh 1994b, c; Lomi 1995; Cattani et al. 2003)

robustly support Zucker’s (1989: 543) speculation that

‘smaller geographic areas should theoretically involve

more intense competition since they are more tightly

bounded resource areas’. Competitive processes in

organizational populations may thus often be hetero-

geneous, operating most strongly in local competitive

arenas, and therefore may generally be modelled more

precisely specified at a local geographic level. And, it

seems likely that the greater the geographic segmenta-

tion of a population’s environment, the more geo-

graphically localized competitive processes in the

population will be (Carroll and Huo 1986).

Geographic segmentation of organizational

environments occurs for a variety of reasons.

Institutional and political constraints may limit the

geographic scope of organizational activities, lead-

ing to distinct geographic markets in which organi-

zations can compete for resources (Hannan et al.

1995). The nature of a population’s activities can

also affect where potential consumers come from

when, for example, organizations provide services

on their premises (e.g. nursing homes, banks);

when transportation costs are high (e.g. brewers,

automobile manufacturers); when products are

made for specific locales (e.g. newspapers); or when

the demand for organizations’ products or services is

depends on location (e.g. hotels). Among recent

studies the level of analysis at which competition is

most intense appears to vary directly with the degree

of geographic segmentation: among Manhattan

hotels competition is most intense within a few city

blocks segmented by proximity to particular business

and tourist activities (Baum and Mezias 1992; Baum

and Haveman 1997), among bank branches within

city wards (Greve 2002), among newspapers within

metropolitan areas (Hannan and Carroll 1992),

among Dutch auditing firms within province

(Catttani et al. 2003), and among European automo-

bile producers within countries segmented by politi-

cal boundaries (Hannan et al. 1995).

Level of analysis issues are also relevant to insti-

tutional processes. Singh (1993: 467–8) speculates

that, while a local level of analysis may often be

more appropriate for analysing competition, it may

not to be so for legitimation processes, which are

likely to require much broader boundaries. Hannan

et al. (1995) estimated models of organizational

founding in the European automobile industry in

which density-dependent legitimation and compe-

tition were measured at different levels of analysis.

Supporting Singh’s (1993) speculation, they found

stronger density-dependent competition at the

country level, and some evidence (support for two

of five countries) of stronger density-dependent

legitimation at the European level. Thus, Hannan et

al.’s (1995) findings provide some support for the

idea that, in contrast to competitive processes, insti-

tutional processes in organizational populations

operate relatively homogeneously, and at a broad

geographic level serving to contextualize competi-

tive processes (Scott 1992; Tucker et al. 1992; but see

Lomi 2000; Cattani et al. 2003). This conclusion is

less well established, however, and may depend on

the type of institutional process.

Organizational Niche Overlap

Baum and Singh (1994b, c) advance and test a

resource overlap model in which the potential for

competition between any two organizations is

directly proportional to the overlap of their targeted

resource bases, or organizational niches. Following

McPherson (1983), Baum and Singh conceive each

member of a population as occupying a potentially

unique organizational niche that delineates its loca-

tion in a multidimensional resource space. The

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 92

organizational niche, which is defined by the

intersection of resource requirements and produc-

tive capabilities at the organization level, depends

on where the organization is located and what it

does (e.g. the clients it has the capacity to target,

how it responds to the environment). The organiza-

tional niche is a result, not a cause, of organizational

adaptation – it is ‘carved out’. Thus, the organiza-

tional niche is a dynamic concept. Depending on

the particular organizational niches they target,

organizations encounter different competitive land-

scapes. Attending to the internal structure of popu-

lation niches, and in particular the resource overlaps

of multiple organizational niches comprising a pop-

ulation, provides a simple way to isolate competitive

and non-competitive forces within a population.

Baum and Singh view the potential for competi-

tion between organizations in any two organizational

niches as directly proportional to the extent of the

overlap in their resource requirements. Potential

competition for each organization is measured using

overlap density, the aggregate overlap of an organiza-

tion’s resource requirements with those of all others

in the population (i.e. population density weighted

by the overlap in resource requirements). Overlap

density has a complementary property, non-overlap

density. Whereas overlap density captures the poten-

tial for competition by aggregating resource overlaps,

non-overlap density aggregates the resource non-

overlaps. Non-overlap density estimates the number

of organizations whose resource requirements do

not overlap with those of the focal organization.

Together, overlap and non-overlap densities disaggre-

gate competitive and non-competitive forces for each

organization in a population with respect to the

underlying resources.

Entrepreneurs are predicted to be unlikely to

target or to be capable of founding organizations in

parts of the resource space where overlap density is

high. Organizations operating in high overlap den-

sity conditions are also predicted to be less sustain-

able. Conversely, entrepreneurs are predicted to be

likely to target and be capable of founding organiza-

tions in parts of the resource space where non-

overlap density is high because of the absence of

direct competition for resources and the potential

for complementary demand enhancement. For

these reasons, high non-overlap density is also

expected to lower failure rates.

Baum and Singh (1994a, b) found support for

these predictions in a population of day care centres

in Metropolitan Toronto for which resource

requirements were defined in terms of the ages of

the children they had the capacity to enrol. These

studies indicate that organizations have different

likelihoods of being established and endure differ-

ent survival fates after founding as a function of the

locations they target in a multidimensional resource

space. If generalized to other populations, the disag-

gregation of population density into overlap and

non-overlap densities may help clarify the role of

population heterogeneity in interpretations of the

non-monotonic density dependence findings

(Hannan et al. 1991; Petersen and Koput 1991).

Overlap and non-overlap densities capture the

absolute presence and absence of resource overlaps

but fail to consider the situation in which organiza-

tions are competitive and non-competitive at the

same time. Baum and Oliver (1996) extended the

organizational niche overlap framework to consider

the relative influence of organizational niche over-

lap and non-overlap with non-overlap intensity,

which aggregates the ratio of an organization’s niche

non-overlap to overlap with other organizations in

the population. While low non-overlap intensity

suggests competitive effects between organizations

because they share common resource needs, recog-

nizing their potential for competition, organizations

occupying partially overlapping niches, and thus

relatively high non-overlap intensity, may be more

inclined to cooperate. Baum and Oliver (1996)

found relative as well as absolute effects of overlap

and non-overlap densities on organizational found-

ing, and, in particular, that interorganizational

cooperation is strengthened when it is combined

with an element of competition.

Although measures of organizational niche over-

lap are now quite common (e.g. Podolny et al. 1996;

Dobrev et al. 2001; 2003) and results generally sup-

portive, measures of non-overlap and non-overlap

intensity have not been replicated, largely because of

the difficulty of obtaining information on the dis-

tribution of production in the various niches.

Taking on questions about how interorganizational

competition and coexistence interact in a multidi-

mensional resource space represents an important

direction for future research.

Accounting for Concentration

The growth trajectories of diverse organizational

populations appear to follow a common path: The

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 93

number of organizations grows slowly initially, then

increases rapidly to a peak. Once this maximum is

reached, there is a decline in the number of popula-

tion members and increased concentration. In orga-

nizational ecology, the density dependence model is

used to account for the shape of the growth trajec-

tory to its peak (Hannan and Carroll 1992). Since

no organization or small group of organizations is

allowed to dominate (each organization in a popu-

lation is assumed to contribute to and experience

competition equally), the density dependence model

predicts logistic growth in numbers to an equilib-

rium level. However, it does not account for the

later decline in numbers and increase in concentra-

tion (Carroll and Hannan 1989a; Hannan and

Carroll 1992). This limitation of the original density-

dependence model formulation has received consid-

erable attention with five elaborations specifically

advanced to address this particular question.

Density Delay

In the density dependence model it is contempora-

neous population density – density at particular

historical times – that is the focus. Carroll and

Hannan (1989a) propose a refinement of the model

to include an additional, delayed population density

effect that helps explain the decline of populations

from their peak density. Carroll and Hannan

(1989a) suggest that organizations’ survival chances

are sensitive to population density levels at the time

of their founding. Specifically, they argue that orga-

nizations founded in high-density conditions expe-

rience persistently higher failure rates. High density

at founding creates a liability of resource scarcity that

prevents organizations from moving quickly from

organizing to full-scale operation. High density also

results in tight niche packing, forcing newly founded

organizations, which cannot compete head-to-head

with established organizations, to use inferior or

marginal resources. These founding conditions

imprint themselves on organizations, affecting their

viability throughout their existence. Carroll and

Hannan show that population density at the time of

organizational founding is positively related to fail-

ure rates in six of the seven populations they analyse

(Carroll and Hannan 1989a; Hannan and Carroll

1992). This means that organizations entering pop-

ulations in high-density conditions have persis-

tently elevated failure rates, contributing to an

explanation for the decline in population density

from its peak. This finding has been replicated in

several studies (e.g. Carroll et al. 1996; Ingram and

Inman 1996), however, several others have failed to

do so (e.g. Wholey et al. 1992; Aldrich et al. 1994).

Moreover, the density delay effect appears to pro-

duce an oscillating equilibrium, not a definite,

singular downturn (Hannan and Carroll 1992: 183).

It also does not permit the competitive strengths of

organizations to vary.

Mass Dependence

Barnett and Amburgey (1990) point out that various

perspectives in organization and management theory

suggest that larger organizations generate stronger

competition than their smaller rivals as a result of

their superior access to resources, greater market

power and economies of scale and scope (e.g.

Edwards 1955; Bain 1956; Starbuck 1965; Thompson

1967; Caves and Porter 1977; Pfeffer and Salancik

1978; Aldrich and Auster 1986; Scherer and Ross

1990; Winter 1990). If large organizations generate

stronger competition, then ecological models of pop-

ulation dynamics should reflect their greater signifi-

cance. Barnett and Amburgey (1990) advance an

elaboration of the density dependence model that

incorporates this possibility. They do this by model-

ling the effects of population mass, the sum of the

sizes of all organizations in the population, or, in

other words, population density weighted by organi-

zational size. If large organizations generate stronger

competition, then, after controlling for population

density, increases in population mass should have a

competitive effect, slowing the founding rate and

increasing the failure rate of smaller organizations.

By permitting the competitive strengths of orga-

nizations to vary as a function of their size, the mass

dependence model permits larger organizations in a

population to dominate by generating stronger

competition than smaller organizations, displacing

their population’s size in numbers. Mass-dependent

competition may thus help account for populations’

later declines in number and increases in concentra-

tion. Large organizations may therefore play a

central role in organizational ecology, not because

they are affected individually by selection pressures,

but because they have a disproportionate influence

on population dynamics (Barnett and Amburgey

1990). Unfortunately, mass dependence findings are

mixed. Some studies find the predicted competitive

effects (e.g. Banaszak-Holl 1992; 1993; Baum and

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 94

Mezias 1992). Others find either no or mixed effects

(e.g. Hannan and Carroll 1992) or mutualistic

effects (e.g. Barnett and Amburgey 1990).

Although contradictory findings appear to be

attributable to data limitations (Hannan and Carroll

1992: 130–1) or significant features of the study pop-

ulation (Barnett and Amburgey 1990: 98–9), a more

general explanation may be that the ability of organi-

zations to differentiate themselves from one another

varies from population to population (Baum and

House 1990). When a population’s members do not

possess the capacity for differentiation, the competi-

tive strength of larger organizations is felt throughout

the population, causing the population to become

increasingly concentrated. However, when meaning-

ful differentiation is possible, population density may

remain relatively high as differentiated specialists that

do not compete directly either with each other or the

large organizations in the population flourish (e.g.

Carroll 1985). A related explanation may be found in

strategic groups theory (e.g. Caves and Porter 1977),

which suggests that industry-wide inferences of

market power cannot be made when strategic groups

characterize competition, since mobility barriers dif-

ferentially protect strategic groups. Future research

examining the effects of organizational differentia-

tion and mobility barriers on competitive dynamics

may therefore improve our understanding of concen-

tration processes in organizational populations.

Competitive Intensity

Barnett (1997: 135) suggests that mixed evidence for

mass-dependence results from a ‘failure to think

about organizational size in an evolutionary way’.

His interpretation is that, while large organizations

may be less likely to fail, as much evidence shows

(see Table 2), this does not necessarily mean that they

are stronger competitors. If they are not, then increas-

ing mass does not necessarily increase the intensity

of competition experienced by the members of the

population. Barnett’s (1997) argument is that the

‘fitness’ of large organizations is equal to the average

of the fitness of their subunits. Since, protected

by their membership in a large organization, sub-

units with low fitness will be less likely to fail than

their independent counterparts, the average fitness

of large organizations will tend to be lower than

that of small organizations. Consequently, smaller,

not larger, organizations will provide the stiffest

competition.

Barnett (1997: 139) operationalizes the relative

weakness of large organizations by ‘allowing each

organization’s competitive intensity to develop over

time depending on its size’. This implies an interac-

tion between an organization’s size and its age,

reducing the competitive strength of a large organi-

zation as it grows. To test his idea, Barnett models

the effects of the sum of the ages of all organizations

in the population (i.e. population age), the sum of

the sizes of all organizations in the population (i.e.

population mass) and competitive intensity, the sum

of the age × size interaction for all population

members. Since Barnett’s model predicts greater

competition as organizations become older and as

population mass increases, population age and mass

should both lower the rate of founding and increase

the rate of failure, but since competitive intensity is

expected to decline as organizations grow it should

be positively related to the rate of founding and neg-

atively related to the failure rate. Barnett finds sup-

port for his model in analyses of founding and

failure among Pennsylvania telephone companies

and US breweries.

Although Barnett’s approach to explaining the

common concave pattern of population growth is

theoretically intriguing and holds promise as an

explanation for population concentration, there

appears to be little support for the model’s basic

premise, that older organizations generate stronger

competition. Stinchcombe’s (1965) liability of new-

ness hypothesis suggests that new organizations have

higher failure rates than other organizations, not that

older organizations have lower failure rates than

younger organizations. Organizational and environ-

mental change can reset the fitness of older organiza-

tions, muddling the relationship between age and

fitness. Even in the absence of such change,

Levinthal’s (1991a) finding that previously accumu-

lated capital can account for the liability of newness

suggests that the survivability of older organizations

may have little to do with their ageing. Mounting

empirical evidence that, after controlling for organi-

zational size, organizational failure rates increase with

age reinforces this point. The fitness of older organi-

zations may thus have little to do with their age.

Changing Basis of Competition

While Barnett’s competitive intensity model focuses on

heterogeneous competitive abilities among organiza-

tions to explain the concentration of organizational

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 95

populations, several other approaches emphasize a

population-wide shift in the basis of competition.

In the original statement of the density depen-

dence model in organization theory, Brittain and

Freeman (1980) described how organizational popu-

lations undergo a competitive transition from

‘r-selection’ to ‘K-selection’ as increasing population

density alters the basis of competition from first-

mover advantage to efficiency. The distinction

between these two bases of competition is grounded

in the Lotka Volterra growth model from bioecology:

= rN

K–N

dt K

where K is the carrying capacity of the popula-

tion’s environment, r is the natural growth rate of

the population, N is population density and t is a

time interval. In this equation, when population

density is initially low, the natural growth term r

dominates and organizations enter at a slow but

exponentially increasing rate. As population den-

sity increases, the carrying capacity term K

becomes dominant and the rate of growth slows.

Growth stops when population density equals the

carrying capacity.

In an emerging population, no organization

or small group of organizations is dominant and

new entrants are typically similar in size to estab-

lished organizations. Competition is dominated by

r-selection, which favours small organizations capa-

ble of moving quickly to exploit new resource oppor-

tunities in the resource-rich but dispersed and

uncertain environments that characterize low-den-

sity conditions. As the population continues to grow,

its markets become connected, demand becomes

more predictable and some organizations begin

making broad appeals to all customers to expand their

market shares. Along with the increased potential for

competition that accompanies increasing population

size (relative to the abundance of resources), this

begins to orient competition among the population’s

members toward cost and price reduction, favouring

larger population members capable of achieving effi-

ciency advantages.

When a population reaches the carrying capacity,

the resources available to its members are exploited

fully and competitive pressures shift to K-selection,

which favours larger organizations competing on

the basis of efficiency. The economic and competi-

tive advantages of the population’s large members

enable them to out-compete their smaller rivals,

which are crowded out by the asymmetric competi-

tion. This ‘shakeout’ decreases the number of orga-

nizations in the population, raises the level of

market concentration and creates a log-normal or

bimodal size distribution in which a few large orga-

nizations tend to dominate many smaller ones

(Hannan and Freeman 1977; Ijiri and Simon 1977).

Ultimately, however, increasing market concentra-

tion forces large organizations to compete increas-

ingly with one another for central markets capable of

sustaining their large-scale operations (Carroll 1985).

This specializes resource use to particular segments of

the size distribution and creates pockets of demand

that smaller, more specialized members of the popu-

lation may be able to exploit without engaging in

direct competition with the larger generalists. Thus,

resource partitioning localizes the competitive effects

of organizations – even large ones – to their particu-

lar segments of operations.

Competition at carrying capacity also initiates a

more general process of functional and territorial

differentiation within the population as competi-

tion pushes entrepreneurs to seek out distinct func-

tions in which they hold a competitive advantage

(Hawley 1950). This creates organizational sub-

groupings that fulfill complementary roles in which

they are dependent on, but non-competitive with,

each other and lowers the potential for competition

by reducing the number of direct competitors each

population member confronts.

The foregoing characterization of the evolution of

organizational populations suggests that the basis of

competition changes over the population life cycle

(Baum 1995). And, in particular, that the influence of

density-dependent competition weakens over a

population’s history and that the effects of size-

based competitive process (e.g. mass dependence,

size-localized competition) become increasingly

influential as efficiency-based advantages grow and

resource partitioning proceeds. Baum finds evidence

of such a shift in the basis of competition over time

in the Manhattan hotel industry. Although, to date,

no research has attempted to replicate his findings,

two related approaches have appeared, supporting

similar arguments and model specifications.

Coupled Clocks

Hannan (1997) advanced a related model in which

density becomes ‘decoupled’ from legitimacy and

Handbook of Organization Studies

(1)

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 96

competition as a population matures. Early in a

population’s history, density has a powerful effect

on the legitimacy of an organizational form. How-

ever, as the population grows older, density plays a

less important role in determining the extent to

which an organizational form is viewed as legiti-

mate. The mere fact of a population having been in

existence for some time is likely to play a role in its

being ‘taken-for-granted’. In addition, organizations

develop durable network ties with other kinds of

actors over time. ‘… as an organizational popula-

tion ages, its taken-for-grantedness presumably

comes to depend upon its network position’

(Hannan et al. 1998b: 305). Therefore, legitimacy is

relatively stable and not easily eroded, even if the

population density declines.

Competition also becomes decoupled as organi-

zations become more fixed into networks of

alliances, develop specializations and so on. In other

words, the population becomes more ‘structured’ as

it gets more mature. ‘As diverse forms of structure

develop, competition can shift from diffuse ecologi-

cal competition to focused, direct rivalry between

organizations with similar positions in the indus-

try…’ (Hannan et al. 1998b: 305). These arguments

imply that the effect of density on legitimation and

competition weaken as an industry ages, which

Hannan operationalizes as an interaction effect

involving population density, population density

squared and population age, and finds support for

his ideas among European automobile producers.

Dynamic Selection and Scale-Based

Competition

Barron (1999) has also proposed a variant of the

changing basis of competition model that empha-

sizes the organizational size distribution of a popu-

lation. In Barron’s version, as organizations grow in

scale they may be able to take deliberate steps to pre-

vent the formation of new competitors. And, even if

they do not, the fact that they already have well-

established exchange networks, reputations, cus-

tomer loyalties and staff expertise makes it more

difficult for new entrants. Therefore, as the size dis-

tribution of a population shifts to the right, Barron

predicts that the founding rate will decline. He tests

this prediction using the mean of the log size distri-

bution, which is approximately symmetric.

For failure, Barron’s argument is somewhat

different. In the early years of a population, when

density is low, the failure rate is high because the

legitimacy of the organizational form is also low.

However, competition is also weak. Under early

low-density conditions, therefore, he proposes that

failures are distributed randomly, with no organiza-

tions heavily disadvantaged relative to any other.

As density increases further, however, competition

increases and organizations best able to withstand

the more intense competition should have an

increasing advantage in terms of survival chances.

Barron’s argument implies an interaction between

the intensity of competition experienced by the

members of a population and some factor that con-

fers a survival advantage, the size of the advantage

increasing as competition intensifies. Thus, the gap

in survival chances of robust and frail organizations

should widen as the density increases. Barron uses

size to differentiate robust and frail organizations

and thus predicts that, as the number of organiza-

tions in a population increase, the survival benefits

of large size should increase. He tests his hypothesis

by interacting population density and the log of

organizational size. Barron tested and corroborated

the model in an analysis of founding and failure of

New York State credit unions during 1914–90.

Dobrev and Carroll (2003) have recently advanced

a related model, measuring what they term ‘scale-

based competition’ as the aggregate distance of an

organization from its larger competitors. Their

model is based on the idea that the aggregate dis-

tance of an organization from its larger competitors

captures the extent to which it can capitalize on

potential competitive advantages derived from its

scale of operations. Their analysis of European auto-

mobile producer failure in four countries supports

the model. Like Barron’s model, they emphasize an

organization’s location in the size distribution, but

distinct from it posit that the benefits of size are a

function of positional differences with larger, rather

than all other organizations.

Future research comparing the ability of these

models to explain the historical growth trajectories

of diverse organizational populations is needed to

refine or integrate these models into a more unified

theoretical and empirical approach. Baum and

Hannan, for example, share a common emphasis on

temporal variation in population density effects and

agree that density becomes increasingly decoupled

from population dynamics over a population’s

history. Hannan, however, does not specify what

population dynamics become coupled to, while

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 97

Baum identifies the increasing influence of size-based

competitive processes. Baum and Barron both

highlight how a shift toward efficiency-based com-

petition gives larger organizations a competitive

advantage over smaller organizations, leading them

to displace their population’s size in numbers, but

Barron’s empirical specification, which ties large

organizations’ competitive advantage to population

density, conflicts with the idea that population den-

sity becomes decoupled from competition over a

population’s history. Dobrev and Carroll’s approach

relies on the population’s evolving size distribution

to capture the increasing significance of scale-based

competition, but does not permit the strength of the

process to vary over time.

Connecting these models to Barnett’s competitive

intensity model is also important given that the dif-

ferent models emphasize different processes – hetero-

geneous competitive abilities among organizations vs

population-wide shifts in the basis of competition.

Future research is needed to determine the relative

significance of these two basic competitive processes

in accounting for concentration and to identify their

distinct roles in population evolution.

Population-Level Learning

In addition to organization-level experiential learn-

ing, collective experiences of organizations at the

population level lay the foundation for the popula-

tion-level learning. Although individual organiza-

tions may tend to engage in too much exploitation,

populations of organizations may still engage in

substantial exploration and generate new knowl-

edge that individual organizations may acquire for

themselves. At the population level, a lack of cohe-

sion (i.e. diverse goals and incentives) and authority

structures may allow the proliferation of new ideas

and routines (Miner and Haunschild 1995). Even

recklessly innovative organizations that quickly fail

can generate new knowledge that adds to the expe-

rience of the population (Ingram and Baum 1997a).

Although learning in established organizations

tends to focus on exploiting old routines rather than

on developing new ones, these organizations may, at

a relatively low cost, be able to exploit new routines

produced by the ongoing explorations and advances

of other organizations in their population. So, ‘the

best strategy for any individual organization is often

to emphasize the exploitation of successful explo-

rations of others’ (Levinthal and March 1993: 104).

By observing their population, organizations can

potentially learn a multiplicity of strategies, prac-

tices and technologies employed by other organiza-

tions. Thus, an organization’s own experience is not

the only opportunity for learning; organizations

may also learn from population experience – the

operating and competitive experience of others in

their population.

Population Operating and

Competitive Experience

The operating experience of other organizations in

the same population may contribute to internal effi-

ciency in a similar way to an organization’s own oper-

ating experience. An organization may be able to use

other organizations’ operating experience to improve

its efficiency at providing service and managing

employees and assets. By observing the internal oper-

ations of other organizations; reading about them in

trade journals; listening to lectures about them; or

by hiring the employees of other organizations, an

organization can gain ideas about how to efficiently

manage its own operations. The contribution of pop-

ulation operating experience to the organization’s

external capabilities, however, is probably more

important (Ingram and Baum 1997a). Industry oper-

ating experience provides additional evidence about

the consumers’ preferences. Just as an organization

can observe consumers’ responses to it, it can observe

(perhaps less accurately) consumers’ responses to

another organization. Every organization in a popu-

lation provides information about consumers’ prefer-

ences (White 1981).

Population experience has several advantages

over own experience for learning. Any one organi-

zation is limited in how much it can learn from its

own experience. The constraint on experimenting is

not just an organization’s resources, but also limits

on how much variance the organization’s internal

systems can handle and external constituents will

accept (Hannan and Freeman 1984). Unlike an

organization, a population can encompass a great

deal of variety without violating internal or external

standards of consistency and reliability, and there-

fore more varied in their experience than individual

organizations. Populations may thus engage in

exploration, even while the organizations within

them engage in exploitation of old routines.

For US hotel chains, Ingram and Baum (1997a)

operationalize population operating experience as

Handbook of Organization Studies

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 98

the operating experience (total chain units

operated) other US hotel chains had accumulated

since a given hotel chain was founded. They also

operationalize population experience at founding as

the operating experience other hotel chains had

accumulated at the time the hotel chain was

founded. To account for a possible decay in the

value of population operating experience over time

due to forgetting and antiquation, they discounted

these variables using weights that depreciated expe-

rience as various functions of its age. Controlling for

organizational operating experience, both popula-

tion experience at entry and accumulated since

entry had significant negative effects on hotel chain

failure rates. Baum and Ingram (1998) attempted to

replicate these findings for Manhattan hotels, mea-

suring population operating experience at founding

and since founding based on the operating experi-

ence (total rooms operated) other Manhattan hotels

had accumulated since a given hotel was founded.

Controlling for organizational operating experi-

ence, population operating experience at founding

had a significant negative effect on the hotel failure

rate; population operating experience accumulated

since a hotel’s founding did not.

Because organizational outcomes are also depen-

dent on the actions of competitors, organizations

also need a model of competitors. Operating expe-

rience can generate part of what is necessary for a

model of competitors by locating them in multidi-

mensional attribute space (White 1981). However,

with competition, an organization’s success is inter-

dependent with other organizations’ competitive

moves. How should a competitor’s moves be inter-

preted? How will the competitor respond to the

organization’s moves? What will it take to drive the

competitor from a market position or cause it to

fail? To answer these questions it is necessary to

observe competitive outcomes.

Ingram and Baum (1997a) suggest measuring

population competitive experience based on organi-

zational failures in a population. At the level of the

organization, Sitkin (1992) argues that small failures

can enhance long-term performance by facilitating

learning. This applies also at the population level. As

organizations fail, the population has an opportu-

nity to learn. Even organizations that break radically

from standards of effective practice and quickly fail

contribute to the experience of the population.

Indeed, since the failing organization disappears,

it is only the population that can learn from its

experience. Past failures are promising as a measure

of population competitive experience for at least

two reasons. First, organizational failures are usually

salient and well publicized events. Managers natu-

rally attend to failures and, unlike other sources of

population experience, since the failing organi-

zation is no longer trying to protect a competitive

future the details of its experience are more accessi-

ble. People who participated in the failed organiza-

tion facilitate dissemination of its experience when

they join other organizations. Secondly, failures are

rich in information that matters for competitive

strategy: what kills other organizations? By examin-

ing a failed organization’s successful rivals other

organizations can learn what competitive moves are

effective at driving others from the market.

Ingram and Baum (1997a) operationalized popu-

lation competitive experience based on the number

of failures observed by US hotel chains both at the

time of their founding and accumulated thereafter,

discounted to account for possible decay in the

value of the experience. Controlling for organiza-

tional and population operating experience, they

found that population competitive experience since

chain entry was negatively related to its failure;

whereas, population competitive experience at the

time of entry had no effect on chain failure.

These studies highlight the often neglected fact

that organizational forms are not homogenous

across time, providing evidence that successive

cohorts of new organizations are improved as a

function of the experience of the population (see

also Sorenson 2000). In population level learning,

variation and selective retention of bundles of

organizational routines as well as organizations are

important forces for population-level change. By

emphasizing both organizations’ learning of rou-

tines and organizational selection (i.e. founding and

failure) as mechanisms for population-level change,

population level learning admits both adaptation

and selection. That new organizations benefit from

the experience of their population suggests, for

example, that entrepreneurs are not strangers to the

populations they enter; they recognize the unex-

ploited potential of the organizations they chal-

lenge. Thus population level learning complements

and extends traditional ecological analyses of orga-

nizational founding and failure, which emphasize

that organizational change and variability reflect

primarily inert organizations replacing each other

but do not typically explore the implications of

Ecological Approaches to Organizations

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 99

entrepreneurs and ongoing organizations engaging

in vicarious selective learning of new routines.

Existing research would be complemented by

attention to processes within organizations that can

affect the capacity of organizations to learn from

their populations. Cyert et al. (1993), for example,

suggest that differences in internal information pro-

cessing capabilities can generate sustainable compet-

itive advantages. Cohen and Levinthal (1990) argue

that past preparation makes learning effectiveness

path dependent. Additionally, the implications of

organizations’ efforts to protect their own operating

experience as a source of competitive advantage (e.g.

Barney 1991; Peteraf 1993) for their population as a

whole – the tension between individual and collective

benefits from experience – would appear to be an

important topic for future research.

Conclusion: Progress, Problems

and Future Directions

As we have shown, organizational ecology is a vital

subfield within organization studies, its research

agenda expanding and methodological sophistica-

tion increasing constantly. However, what does

organizational ecology contribute to progress in orga-

nization studies? One way to answer this question is

to examine the problems organizational ecology

solves (Lauden 1984; Tucker 1994). According to

Lauden (1984: 15), scientific theories must solve two

kinds of problems: empirical problems, which are

substantive questions about the objects that consti-

tute the domain of inquiry, and conceptual problems,

which include questions about internal logical con-

sistency and conceptual ambiguity of theories

advanced to solve empirical problems, as well as the

methodological soundness of tests of theoretical

arguments. From this perspective, organizational

ecology’s contribution to progress can be defined in

terms of its capacity to accumulate solved empirical

problems while minimizing the scope of unsolved

empirical problems and conceptual problems.

As revealed in this review, the primary emphasis

of organizational ecology is the development of

theoretical explanations for specific empirical prob-

lems (e.g. age, size and density dependence in

organizational failure). Although organizational

ecology has advanced knowledge about a wide range

of empirical problems, few (if any) of these can be

considered solved conclusively. Of course, other

organization studies subdisciplines have not solved

these problems either. From a conceptual stand-

point, while examples of internal logical inconsis-

tencies are uncommon in ecological theory, and

those that are identified being tackled directly (see

for example, Péli 1997; Hannan 1998; Hannan et al.

2003a, b), instances of conceptual ambiguities are

more prevalent. Questions about the meaning of

central concepts such as population, founding,

failure and legitimacy have been raised frequently

(e.g. McKelvey 1982; Carroll 1984a; Astley 1985;

Delacroix and Rao 1994; Rao 2001). To be fair, such

ambiguities are not unique to organizational ecol-

ogy, but seem endemic to organization studies.

One persistent source of conceptual problems is

the methodological soundness of tests of theoretical

arguments. One area of frequent debate is the appro-

priateness of inferring underlying processes of legiti-

mation from population density estimates instead of

measuring the underlying construct directly (Zucker

1989; Baum and Oliver 1992; Hannan and Carroll

1992; Miner 1993). Inferring processes of learn-

ing from cumulative organizational and industry

experience estimates is subject to similar concerns

(Baum and Ingram 1998). In part, this problem stems

from organizational ecology’s use of large-scale,

historical databases in which, by necessity, measures

are frequently removed from concepts. Research on

age dependence and, to a lesser degree, research on

size dependence, also suffer from this problem

(Henderson 1999).

Although unsolved empirical problems and con-

ceptual problems are not uncommon in emerging

areas of scientific inquiry, the longer these problems

remain unresolved, the greater their importance

becomes in debates about the veracity of the theories

that generated them (Lauden 1984: 64–6). What pro-

duces organizational ecology’s problems? Although

organizational ecologists would like their theories to

maximize generality across organizational popula-

tions, realism of context and precision in measure-

ment of variables, in fact, no theory can be general,

precise and realistic all at the same time (McGrath

1982; Puccia and Levins 1985). Theories must sacri-

fice on some dimensions to maximize others. For

example, realistic theories may apply to only a limited

domain, while general theories may be inaccurate or

misleading for specific applications.

Historically, organizational ecologists appear to

have favoured a trade-off of precision and realism for

generality in their theories (Singh 1993). For example,

Handbook of Organization Studies

100

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 100

precision and realism are clearly sacrificed for gener-

ality in the original formulations of density depen-

dence and structural inertia theories. On the one

hand, this research strategy produces organizational

ecology’s main strength: A wealth of comparable

empirical evidence from diverse organizational set-

tings on a range of empirical problems unparalleled

in organization studies. On the other hand, it also

creates a major weakness: The large pool of coeffi-

cients for indirect measures such as age, size, popula-

tion density and population experience reveals little

about underlying theoretical explanations designed

to account for the empirical problems of interest.

This creates conceptual problems by fostering skepti-

cism regarding the veracity of inferred underlying

processes because conforming findings cannot be

interpreted precisely, and creates unsolved empirical

problems by making it difficult to account for non-

conforming findings on theoretical grounds.

The sacrifice of contextual realism and measure-

ment precision for generality thus underlies several

key problems in organizational ecology. By sacrific-

ing some generality for more precision and realism,

however, organizational ecologists have begun to

address this. Over the past decade, ecological theory

and analysis has become increasingly sensitive to the

need for more fine-grained theory and measure-

ment that captures in a more nuanced and direct

manner the underlying processes of interest.

Ecological research adopting this problem-solving

strategy has already contributed to the literature in

several important ways. Perhaps most notable are

elaborations of the original density dependence

model (see Table 6) help increase precision by mea-

suring underlying processes of competition and

legitimation more directly (Baum and Oliver 1992;

1996; Baum and Singh 1994b, c; Deeds et al. 2004).

They also enhance contextual realism by incorpo-

rating additional features of population such as

organizational size distributions (Barnett and

Amburgey 1990; Barron 1999; Dobrev and Carroll

2003), market niche structures (Baum and Singh

1994b, c) and competitive heterogeneity into the

model (Baum 1995; Barnett 1997). The result is a

more compelling and less problematic account of

the ecological dynamics of organizational popula-

tions. At the demographic level, research emphasiz-

ing greater measurement precision has shed new

light on the underlying causes of structural inertia

and age and size dependence in organizational

change and failure (e.g. Greve 1999; Henderson

1999). More robust organization-level measures

are necessary to establish the micro-foundations of

ecological theory.

Organizational ecology is making good progress

on its problems. Continued emphasis on increasing

precision and realism in theory and research is essen-

tial to enhancing the impact and influence of ecolog-

ical approaches to organizations. Getting closer to

research problems will reveal important aspects of

phenomena that more distanced researchers cannot

detect. Understanding anomalies – results that are

inconsistent with each other, theory or observation –

is crucial to specifying contingent predictions and

increasing precision. Letting research problems drive

the choice of research design and methodology

(rather than vice versa) will yield study designs best

suited to answering the questions posed. These shifts

in the ecological research orientation will foster new

kinds of questions that enhance links with other

research streams in organization theory and related

subfields such as strategy. In turn, these linkages will

help realize more of the great potential of ecological

approaches to contribute to theory and research in

organization studies, as well as to practice in public

policy, management and entrepreneurship.

Note

1. Strategic and economic models make the same pre-

diction (e.g. Porter 1980; Tirole 1988; Scherer and Ross

1990).

References

Aldrich, H.E. (1979) Organizations and Environments.

Englewood Cliffs, NJ: Prentice-Hall.

Aldrich, H.E. and Auster, E.R. (1986) ‘Even Dwarfs started

small: Liabilities of age and size and their strategic

implications’, Research in Organizational Behavior,8:

165–98.

Aldrich, H.E. and Fiol, C.M. (1994) ‘Fools rush in? The

institutional context of industry creation’, Academy of

Management Review, 19: 645–70.

Aldrich, H.E. and Pfeffer, J. (1976) ‘Environments of orga-

nizations’, Annual Review of Sociology, 2: 79–105.

Aldrich, H.E. and Wiedenmayer, G. (1993) ‘From traits to

rates: an ecological perspective on organizational

foundings’, Advances in Entrepreneurship, Firm

Emergence, and Growth, 1: 145–95.

Aldrich, H.E., Zimmer, C., Staber, U. and Beggs, J. (1994)

‘Minimalism, mutualism, and maturity: The evolution

Ecological Approaches to Organizations

101

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 101

of the American trade association population in the

20th century’, in J.A.C. Baum and J.V. Singh (eds),

Evolutionary Dynamics of Organizations.New York:

Oxford University Press. pp. 223–39.

Amburgey, T.L. and Dacin, T. (1994) ‘As the left foot fol-

lows the right? The dynamics of strategic and structural

change’, Academy of Management Journal, 37: 1427–52.

Amburgey, T.L. and Kelly, D. (1985) ‘Adaptation and selec-

tion in organizational populations: A competing risk

models’, Paper presented at the Academy of Management

meetings, San Diego, CA.

Amburgey, T.L. and Miner, A.S. (1992) ‘Strategic momen-

tum: The effects of repetitive, positional, and con-

textual momentum on merger activity’, Strategic

Management Journal, 13: 335–48.

Amburgey, T.L., Dacin, T. and Kelly, D. (1994) ‘Disruptive

selection and population segmentation: Interpo-

pulation competition as a segregating process’, in

J.A.C. Baum and J.V. Singh (eds), Evolutionary

Dynamics of Organizations.New York:Oxford Univer-

sity Press. pp. 240–54.

Amburgey, T.L., Kelly, D. and Barnett, W.P. (1990)

‘Resetting the clock: The dynamics of organizational

change and failure’, in J.L. Wall and L.R. Jauch (eds),

Academy of Management Proceedings. San Francisco,

CA: Academy of Management. pp. 160–4.

Amburgey, T.L., Kelly, D. and Barnett, W.P. (1993) ‘Resetting

the clock: The dynamics of organizational change and

failure’, Administrative Science Quarterly, 38: 51–73.

Anderson, P. and Tushman, M.L. (1992) ‘Technological,

ecological and economic determinants of industry

exit: A longitudinal study of the American cement

(1888–1980) and minicomputer (1958–1982) indus-

tries’. Unpublished manuscript, Johnson Graduate

School of Management, Cornell University.

Argote, L. (1999) Organizational Learning: Creating,

Retaining and Transferring Knowledge. Boston: Kluwer.

Argote, L., Beckman, S.L. and Epple, D. (1990) ‘The persis-

tence and transfer of learning in industrial settings’,

Management Science, 36: 140–54.

Astley, W.G. (1985) ‘The two ecologies: Population and

community perspectives on organizational evolution’,

Administrative Science Quarterly, 30: 224–41.

Astley, W.G. and Van De Ven, A.H. (1983) ‘Central per-

spectives and debates in organizational theory’, Admin-

istrative Science Quarterly, 28: 245–73.

Bain, J.S. (1956) Barriers to New Competition. Cambridge,

MA: Harvard University Press.

Banaszak-Holl, J. (1992) ‘Historical trends in rates of

Manhattan bank mergers, acquisitions, and failures’.

Unpublished manuscript, Center for Health Care

Research, Brown University.

Banaszak-Holl, J. (1993) ‘Avoiding failure when times get

tough: Changes in organizations’ responses to compet-

itive pressures’. Unpublished manuscript, Center for

Health Care Research, Brown University.

Barley, S.R., Freeman, J. and Hybels, R.C. (1992) ‘Strategic

alliances in commercial biotechnology’, in N. Nohria

and R. Eccles (eds), Networks and Organizations: Structure,

Form, and Action. Boston: Harvard Business School

Press. pp. 311–47.

Barnett, W.P. (1990) ‘The organizational ecology of a tech-

nological system’, Administrative Science Quarterly, 35:

31–60.

Barnett, W.P. (1997) ‘The dynamics of competitive inten-

sity’, Administrative Science Quarterly, 42: 128–60.

Barnett, W.P. and Amburgey, T.L. (1990) ‘Do larger organi-

zations generate stronger competition?’, in J.V. Singh

(ed.), Organizational Evolution: New Directions.

Newbury Park, CA: Sage. pp. 78–102.

Barnett, W.P. and Carroll, G.R. (1987) ‘Competition and

mutualism among early telephone companies’,

Administrative Science Quarterly, 32: 400–21.

Barnett, W.P. and Carroll, G.R. (1995) ‘Modeling internal

organizational change’, Annual Review of Sociology,

21: 217–36.

Barnett, W.P. and Hansen, M. (1996) ‘The red queen in

organizational evolution’, Strategic Management

Journal, 17(S): 139–57.

Barnett, W.P. and Sorenson, O. (2002) ‘The red queen in

organizational creation and development’, Industrial and

Corporate Change, 11: 289–325.

Barnett, W.P., Greve, H.R. and Park, D.Y. (1994) ‘An evolu-

tionary model of organizational performance’, Strategic

Management Journal, 15: 11–28.

Barney, J. (1991) ‘Firm resources and sustained competi-

tive advantage’, Journal of Management, 17(1): 99–120.

Barron, D.N. (1999) ‘The structuring of organizational

populations’, American Sociological Review, 64: 421–45.

Barron, D.N., West, E. and Hannan, M.T. (1994) ‘A time to

grow and a time to die: Growth and mortality of credit

unions in New York City, 1914–1990’, American Journal

of Sociology, 100: 381–421.

Baum, J.A.C. (1989) ‘Liabilities of newness, adolescence,

and obsolescence: Exploring age dependence in the

dissolution of organizational relationships and organi-

zations’, Proceedings of the Administrative Sciences

Association of Canada, 10(5): 1–10.

Baum, J.A.C. (1990a) ‘Inertial and adaptive patterns in the

dynamics of organizational change’, in J.L. Wall and L.R.

Jauch (eds), Academy of Management Proceedings.San

Francisco, CA: Academy of Management. pp. 165–9.

Baum, J.A.C. (1990b) ‘Why are there so many (few) kinds

of organizations? A study of organizational diversity’, in

C. Kirchmeyer (ed.), Proceedings of the Administrative

Sciences Association of Canada, 11(5): 1–10.

Baum, J.A.C. (1995) ‘The changing basis of competition in

organizational populations: The Manhattan hotel

industry, 1898–1990’, Social Forces, 74: 177–204.

Baum, J.A.C. (1996) ‘Organizational Ecology’, in S.R. Clegg,

C. Hardy and W.R. Nord (eds) Handbook of Organiza-

tion Studies. London: Sage. pp. 77–114.

Handbook of Organization Studies

102

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 102

Baum, J.A.C. (1999) ‘The rise of nursing home chains in

Ontario, 1971–1996’, Social Forces, 78: 543–84.

Baum, J.A.C. (2006) ‘Competitive and institutional

isomorphism in organizational populations’, in

W.W. Powell and D.L. Jones (eds), Bending the Bars of

the iron Cage: Institutional Dynamics and Processes.

Chicago: University of Chicago Press. In press.

Baum, J.A.C. and Amburgey, T.L. (2002) ‘Organizational

ecology’, in J.A.C. Baum (ed.), Companion to

Organizations. Oxford, UK: Blackwell. pp. 304–26.

Baum, J.A.C. and Haveman, H.A. (1997) ‘Love thy neigh-

bor? Differentiation and agglomeration in the

Manhattan hotel industry’, Administrative Science

Quarterly, 42: 304–38.

Baum, J.A.C. and House, R.J. (1990) ‘On the maturation

and aging of organizational populations’, in J.V. Singh

(ed.), Organizational Evolution: New Directions.

Newbury Park, CA: Sage. pp. 129–42.

Baum, J.A.C. and Ingram, P. (1998) ‘Survival-enhancing

learning in the Manhattan hotel industry, 1898–1980’,

Management Science, 44(7): 996–1016.

Baum, J.A.C. and Korn, H. (1999) ‘Dynamics of dyadic

competition’, Strategic Management Journal, 20: 251–78.

Baum, J.A.C. and Mezias, S.J. (1992) ‘Localized competi-

tion and organizational failure in the Manhattan hotel

industry, 1898–1990’, Administrative Science Quarterly,

37: 580–604.

Baum, J.A.C. and Mezias, S.J. (1993) ‘Competition, institu-

tional linkages and organizational growth’, Social

Science Research, 22: 131–64.

Baum, J.A.C. and Oliver, C. (1991) ‘Institutional linkages

and organizational mortality’, Administrative Science

Quarterly, 36: 187–218.

Baum, J.A.C. and Oliver, C. (1992) ‘Institutional embed-

dedness and the dynamics of organizational popula-

tions’, American Sociological Review, 57: 540–59.

Baum, J.A.C. and Oliver, C. (1996) ‘Toward an institutional

ecology of organizational founding’, Academy of

Management Journal, 39: 1378–427.

Baum, J.A.C. and Powell, W.W. (1995) ‘Cultivating an

institutional ecology of organizations: Comment on

Hannan, Carroll, Dundon, and Torres’, American

Sociological Review, 60: 529–38.

Baum, J.A.C. and Rao, H. (2004) ‘Dynamics of organiza-

tional populations and communities’, in M.S. Poole and

A.H. Van de Ven (eds), Handbook of Organizational

Change and Innovation. New York: Oxford University

Press. pp. 212–58.

Baum, J.A.C. and Silverman, B.S. (2004) ‘Picking winners

or building them? Alliance, intellectual, and human

capital as selection criteria in venture financing and

performance of biotechnology startups’, Journal of

Business Venturing, 19: 411–36.

Baum, J.A.C. and Singh, J.V. (1994a) ‘Organizational hierar-

chies and evolutionary processes: Some reflections on a

theory of organizational evolution’, in J.A.C. Baum and

J.V. Singh (eds), Evolutionary Dynamics of Organizations.

New York: Oxford University Press. pp. 3–22.

Baum, J.A.C. and Singh, J.V. (1994b) ‘Organizational niche

overlap and the dynamics of organizational founding’,

Organization Science, 5: 483–501.

Baum, J.A.C. and Singh, J.V. (1994c) ‘Organizational niche

overlap and the dynamics of organizational mortality’,

American Journal of Sociology, 100: 346–80.

Baum, J.A.C. and Singh, J.V. (1996) ‘Dynamics of organi-

zational responses to competition’, Social Forces, 74:

1261–97.

Baum, J.A.C., Korn, H.J. and Kotha, S. (1995) ‘Dominant

designs and population dynamics in telecommunica-

tions services: Founding and failure of facsimile service

organizations, 1969–1992’, Social Science Research, 24:

97–135.

Baum, J., Li, S.X. and Usher, J. (2000) ‘Making the next

move: How experiential and vicarious learning shape

the locations of chains’ acquisition’, Administrative Science

Quarterly, 45(4): 766–801.

Berk, R.A. (1983) ‘An introduction to sample selection bias

in sociological data’, American Sociological Review, 48:

386–98.

Boeker, W. (1989) ‘Strategic change: The effects of found-

ing and history’, Academy of Management Journal, 32:

489–515.

Boeker, W. (1997) ‘Strategic change: The influence of man-

agerial characteristics and organizational growth’,

Academy of Management Journal, 40: 152–70.

Boone, C., Brocheler, V. and Carroll, G. (2000) ‘Custom

service: Application and tests of resource-partitioning

theory among Dutch auditing firms for 1896 to 1992’,

Organization Studies, 21: 355–81.

Boone, C., Carroll, G.R. and van Witteloostuijin, A. (2002)

‘Environmental resource distributions and market par-

titioning: Dutch daily newspaper organizations from

1968 to 1994’, American Sociological Review, 67: 408–31.

Brittain, J.W. (1994) ‘Density-independent selection and

community evolution’, in J.A.C. Baum and J.V. Singh

(eds), Evolutionary Dynamics of Organizations.New

York: Oxford University Press. pp. 355–78.

Brittain, J.W. and Freeman, J.H. (1980) ‘Organizational

proliferation and density dependent selection’, in

J. Kimberly and R. Miles (eds), The Organizational Life

Cycle. San Francisco, CA: Jossey-Bass. pp. 291–338.

Bruderl, J. and Schussler, R. (1990) ‘Organizational mor-

tality: The liabilities of newness and adolescence’,

Administrative Science Quarterly, 35: 530–47.

Budros, A. (1993) An analysis of organizational birth

types: Organizational start-up and entry in the nine-

teenth-century life insurance industry’, Social Forces, 70:

1013–30.

Budros, A. (1994) ‘Analyzing unexpected density depen-

dence effects on organizational births in New York’s life

insurance industry, 1842–1904’, Organization Science,5:

541–53.

Ecological Approaches to Organizations

103

[AQ1]

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 103

Burgelman, R.A. (1991) ‘Intraorganizational ecology

of strategy-making and organizational adaptation:

Theory and field research’, Organization Science,2:

239–62.

Burgelman, R.A. and Mittman, B.S. (1994) ‘An intraorga-

nizational ecological perspective on managerial risk

behavior, performance, and survival: Individual, orga-

nizational, and environmental effects’, in J.A.C. Baum

and J.V. Singh (eds), Evolutionary Dynamics of

Organizations. New York: Oxford University Press.

pp. 53–74.

Campbell, D.T. (1965) ‘Variation and selective retention in

socio-cultural evolution’, in H.R. Barringer, G.I.

Blanksten and R.W. Mack (eds), Social Change in

Developing Areas: A Reinterpretation of Evolutionary

Theory. Cambridge, MA: Schenkman. pp. 19–48.

Carroll, G.R. (1983) ‘A stochastic model of organizational

mortality: Review and reanalysis’, Social Science

Research, 12: 303–29.

Carroll, G.R. (1984a) Organizational ecology’, Annual

Review of Sociology, 10: 71–93.

Carroll, G.R. (1984b) ‘Dynamics of publisher succession

in newspaper organizations’, Administrative Science

Quarterly, 29: 93–113.

Carroll, G.R. (1985) ‘Concentration and specialization:

Dynamics of niche width in populations of organiza-

tions’, American Journal of Sociology, 90: 1262–83.

Carroll, G.R. (1987) Publish and Perish: The Organizational

Ecology of Newspaper Industries.Greenwich CT:JAI

Press.

Carroll, G.R. and Delacroix, J. (1982) ‘Organizational mor-

tality in the newspaper industries of Argentina and

Ireland: An ecological approach’, Administrative Science

Quarterly, 27: 169–98.

Carroll, G.R. and Hannan, M.T. (1989a) ‘Density delay in

the evolution of organizational population: A model

and five empirical tests’, Administrative Science

Quarterly, 34: 411–30.

Carroll, G.R. and Hannan, M.T. (1989b) ‘Density depen-

dence in the evolution of newspapers organizations’,

American Sociological Review, 54: 524–41.

Carroll, G.R. and Huo,Y.P. (1986) ‘Organizational task and

institutional environments in ecological perspective:

Findings from the local newspaper industry’, American

Journal of Sociology, 91: 838–73.

Carroll, G.R. and Swaminathan, A. (1991) ‘Density depen-

dent organizational evolution in the American brewing

industry from 1633–1988’, Acta Sociologica, 34: 155–76.

Carroll, G.R. and Swaminathan, A. (1992) ‘The organiza-

tional ecology of strategic groups in the American

brewing industry from 1975–1990’, Corporate and

Industrial Change, 1: 65–97.

Carroll, G. and Swaminathan, A. (2000) ‘Why the micro-

brewery movement? Organizational dynamics of

resource partitioning in the U.S. brewing industry’,

American Journal of Sociology, 106: 715–62.

Carroll, G.R. and Wade, J.B. (1991) ‘Density dependence in

the evolution of the American brewing industry across

different levels of analysis’, Social Science Research, 20:

271–302.

Carroll, G.R., Bigelow, L., Seidel, M.-D. and Tsai, L. (1996)

‘The fates of de novo and de alio producers in the

American automobile industry, 1885–1981’, Strategic

Management Journal, 17: 117–38.

Carroll, G.R., Dobrev, S.D. and Swaminathan, A. (2002)

‘Organizational processes of resource partitioning’, in

R. Kramer and B. Staw (eds), Research in Organizational

Behavior, 24. Oxford, UK: JAI/Elsevier. pp. 1–40.

Carroll, G.R., Preisendoerfer, P., Swaminathan, A. and

Wiedenmayer, G. (1993) ‘Brewery and Braueri: The

organizational ecology of brewing’, Organization

Studies, 14: 155–88.

Cattani, G., Pennings, J.M. and Wezel, F.C. (2003) ‘Spatial

and temporal heterogeneity in founding patterns’,

Organization Science, 14: 670–85.

Caves, R.E. and Porter, M.E. (1977) ‘From entry barriers to

mobility barriers’, Quarterly Journal of Economics, 90:

241–61.

Chandler, A. (1977) The Visible Hand. Cambridge, MA:

Belknap.

Chen, M.-J. and Hambrick, D.D. (1995) ‘Speed, stealth,

and selective attack: How small firms differ from large

firms in competitive behavior’, Academy of Management

Journal, 38(2): 453–82.

Child, J. (1972) ‘Organization structure, environment and

performance: The role of strategic choice’, Sociology,

6: 1–22.

Cohen, W.M. and Levinthal, D.A. (1990) ‘Absorptive capa-

city: A new perspective on learning and innovation’,

Administrative Science Quarterly, 35: 128–52.

Cyert, R.M. and March, J.G. (1963) A Behavioral Theory of

the Firm. Englewood Cliffs, NJ: Prentice-Hall.

Cyert, R.M., Kumar, P. and Williams, J.R. (1993)

‘Information, market imperfections and strategy’,

Strategic Management Journal, 14(Winter Special

Issue): 47–58.

Dacin, M.T. (1997) ‘Isomorphism in context: The power

and prescription of institutional norms’, Academy of

Management Journal, 40: 46–81.

D’Aunno, T., Succi, M. and Alexander, J. (2000) ‘The role of

institutional and market forces in divergent organizational

change’, Administrative Science Quarterly, 45: 679–703.

Darr, E.D., Argote, L. and Epple, D. (1995) ‘The acquisi-

tion, transfer, and depreciation of knowledge in service

organizations: Productivity in franchises’, Management

Science, 42: 1750–62.

Deeds, D.L., Mang, P.Y. and Frandsen, M.L. (2004) ‘The

impact of industry and firm legitimacy on the flow of

capital into high technology ventures’, Strategic

Organization, 2: 9–34.

Delacroix, J. and Carroll, G.R. (1983) ‘Organizational

foundings: An ecological study of the newspaper

Handbook of Organization Studies

104

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 104

industries of Argentina and Ireland’, Administrative

Science Quarterly, 28: 274–91.

Delacroix, J. and Rao, H. (1994) ‘Externalities and ecologi-

cal theory: Unbundling density dependence’, in J.A.C.

Baum and J.V. Singh (eds), Evolutionary Dynamics

of Organizations. New York: Oxford University Press.

pp. 255–68.

Delacroix, J. and Swaminathan, A. (1991) ‘Cosmetic, spec-

ulative, and adaptive organizational change in the wine

industry: A longitudinal study’, Administrative Science

Quarterly, 36: 631–61.

Delacroix, J., Swaminathan, A. and Solt, M.E. (1989)

‘Density dependence versus population dynamics: An

ecological study of failings in the California wine

industry’, American Sociological Review, 54: 245–62.

DiMaggio, P.J. and Powell, W.W. (1983) ‘The iron cage

revisited: Institutional isomorphism and collective

rationality in organizational fields’, American Socio-

logical Review, 48: 147–60.

Dobbin, F. and Dowd, T. (1997) ‘How policy shapes com-

petition: Early railroad foundings in Massachusetts’,

Administrative Science Quarterly, 42: 501–29.

Dobrev, S.D. (2000) Decreasing concentration and

reversibility of the resource partitioning process:

Supply shortages and deregulation in the Bulgarian

newspaper industry, 1987–1992. Organization Studies,

21: 383–404.

Dobrev, S.D. (2001) ‘Revisiting organizational legitima-

tion: Cognitive diffusion and sociopolitical factors in

the evolution of Bulgarian newspaper enterprises,

1846–1992’, Organization Studies, 22: 355–81.

Dobrev, S.D. and Carroll, G.R. (2003) ‘Size (and competi-

tion) among organizations: Modeling scale-based

selection among automobile producers in four major

countries’, Strategic Management Journal, 24: 541–58.

Dobrev, S.D., Kim, T.-Y. and Carroll, G.R. (2002) ‘The evo-

lution of organizational niches’, Administrative Science

Quarterly, 47: 233–64.

Dobrev, S., Kim, T.-Y. and Carroll, G. (2003) ‘Shifting

gears, shifting niches: Organizational inertia and change

in the evolution of the U.S. automobile industry,

1885–1981’, Organization Science, 14: 264–82.

Dobrev, S., Kim, T.-Y. and Hannan, M. (2001) ‘Dynamics

of niche width and resource partitioning’, American

Journal of Sociology, 106: 1299–338.

Dowell, G. and Swaminathan, A. (2000) ‘Racing and back-

pedalling into the future: New product introduction

and organizational mortality in the US bicycle industry,

1880–1918’, Organization Studies, 21: 405–31.

Downs, A. (1967) Inside Bureaucracy. Boston: Little, Brown.

Dun and Bradstreet (1978) The Failure Record.New York:

Dun & Bradstreet Inc.

Durkheim, E. (1893) (1933 tr.) The Division of Labor in

Society. Glencoe, IL: Free Press.

Edwards, C.D. (1955) ‘Conglomerate bigness as a source of

power’, in National Bureau of Economic Research,

Business Concentration and Public Power: Princeton, NJ:

Princeton University Press. pp. 331–59.

Fichman, M. and Levinthal, D.A. (1991) ‘Honeymoons

and the liability of adolescence: A new perspective on

duration dependence in social and organizational rela-

tionships’, Academy of Management Review, 16: 442–68.

Fombrun, C.J. (1986) ‘Structural dynamics within and

between organizations’, Administrative Science Quarterly,

31: 403–21.

Fombrun, C.J. (1988) ‘Crafting an institutionally informed

ecology of organizations’, in G.R. Carroll (ed.),

Ecological Models of Organizations. Cambridge, MA:

Ballinger. pp. 223–39.

Freeman, J. (1990) ‘Ecological analysis of semiconductor firm

mortality’, in J.V. Singh (ed.), Organizational Evolution:

New Directions. Newbury Park, CA: Sage. pp. 53–77.

Freeman, J.H. and Hannan, M.T. (1983) ‘Niche width and

the dynamics of organizational populations’, American

Journal of Sociology, 88: 1116–45.

Freeman, J.H. and Hannan, M.T. (1987) ‘The ecology of

restaurants revisited’, American Journal of Sociology, 92:

1214–20.

Freeman, J. and Lomi, A. (1994) ‘Resource partitioning

and foundings of banking cooperative in Italy’, in J.A.C.

Baum and J.V. Singh (eds), Evolutionary Dynamics

of Organizations. New York: Oxford University Press.

pp. 269–93.

Freeman, J.H., Carroll, G.R. and Hannan, M.T. (1983) ‘The

liability of newness: Age dependence in organizational

death rates’, American Sociological Review, 48: 692–710.

Galunic, C. and Weeks, J.R. (2002) ‘Intraorganizational

ecology’, in J.A.C. Baum (ed.), Companion to

Organizations. Oxford, UK: Blackwell. pp. 75–97.

Gartner, W.B. (1989) ‘Some suggestions for research on

entrepreneurial traits and characteristics’, Entrepreneur-

ship: Theory and Practice, 14: 27–37.

Ginsberg, A. and Baum, J.A.C. (1994) ‘Evolutionary

processes and patterns of core business change’, in

J.A.C. Baum and J.V. Singh (eds), Evolutionary Dynam-

ics of Organizations. New York: Oxford University Press.

pp. 127–51.

Ginsberg, A. and Buchholtz, A. (1990) ‘Converting to for-

profit status: Corporate responsiveness to radical

change’, Academy of Management Journal, 33: 447–77.

Granovetter, M. (1985) ‘Economic action and social struc-

ture: the problem of embeddedness’, American Journal

of Sociology, 91: 481–510.

Greve, H.R. (1995) ‘Jumping ship: The diffusion of strat-

egy abandonment’, Administrative Science Quarterly,

40(September): 444–73.

Greve, H.R. (1996) ‘Patterns of competition: The diffusion

of a market position in radio broadcasting’,

Administrative Science Quarterly, 41: 29–60.

Greve, H.R. (1998) ‘Performance, aspirations and risky

organizational change’, Administrative Science Quarterly,

43: 58–86.

Ecological Approaches to Organizations

105

[AQ2]

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 105

Greve, H.R. (1999) ‘The effect of core change on perfor-

mance: Inertia and regression toward the mean’,

Administrative Science Quarterly, 44: 590–614.

Greve, H. (2002) ‘An ecological theory of spatial evolution:

Local density dependence in Tokyo Banking, 1894–1936’,

Social Forces, 80: 847–79.

Greve, H. and Taylor, A. (2000) ‘Innovations as catalysts for

organizational change: Shifts in organizational cogni-

tion and search’, Administrative Science Quarterly, 45:

54–83.

Guo, G. (1993) ‘Event history analysis for left-truncated

data’, Sociological Methodology, 23: 217–44.

Halliday, T., Powell, M.J. and Granfors, M.W. (1993) ‘After

minimalism: Transformations of state bar associations,

1918–1950’, American Sociological Review, 58: 515–35.

Hambrick, D.C. and D’Aveni, R.A. (1988) ‘Large corporate

failures as downward spirals’, Administrative Science

Quarterly, 33: 1–23.

Hannan, M.T. (1986) ‘A model of competitive and institu-

tional processes in organizational ecology’, Technical

Report 86-13, Department of Sociology, Cornell

University.

Hannan, M.T. (1997) ‘Inertia, density and the structure of

organizational populations: Entries in European auto-

mobile industries, 1886–1981’, Organization Studies, 18:

193–228.

Hannan, M.T. (1998) ‘Rethinking age dependence in orga-

nizational mortality: Logical formalizations’, American

Journal of Sociology, 104: 126–64.

Hannan, M.T. and Carroll, G.R. (1992) Dynamics of

Organizational Populations: Density, competition, and

legitimation. New York: Oxford University Press.

Hannan, M.T. and Freeman, J.H. (1977) ‘The population

ecology of organizations’, American Journal of Sociology,

83: 929–84.

Hannan, M.T. and Freeman, J.H. (1984) ‘Structural inertia

and organizational change’, American Sociological

Review, 49: 149–64.

Hannan, M.T. and Freeman, J.H. (1987) ‘The ecology of

organizational founding: American labor unions,

1836–1985’, American Journal of Sociology, 92: 910–43.

Hannan, M.T. and Freeman, J.H. (1988) ‘The ecology of

organizational mortality: American labor unions,

1836–1985’, American Journal of Sociology, 94: 25–52.

Hannan, M.T. and Freeman, J.H. (1989) Organizational

Ecology. Cambridge, MA: Harvard University Press.

Hannan, M.T., Barron, D. and Carroll, G.R. (1991) ‘On the

interpretation of density dependence in rates of organi-

zational mortality: A reply to Peterson and Koput’,

American Sociology Review, 56: 410–5.

Hannan, M.T., Carroll, G.R., Dobrev, S.D., Han, J. and

Torres, J.C. (1998a) ‘Organizational mortality in

European and American automobile industries, Part I:

Revising age dependence’, European Sociological Review,

14: 279–302.

Hannan, M.T., Carroll, G.R., Dobrev, S.D., Han, J. and Torres,

J.C. (1998b) ‘Organizational mortality in European and

American automobile industries, Part II: Coupled clocks’,

European Sociological Review, 14: 303–13.

Hannan, M., Carroll, G.R., Dundon, E. and Torres, J.C.

(1995) ‘Organizational evolution in a multinational

context: Entries of automobile manufacturers in

Belgium, Britain, France, Germany, and Italy’, American

Sociological Review, 60: 509–28.

Hannan, M.T., Carroll, G.R. and Pólos, L. (2003a) ‘The

organizational niche’, Sociological Theory, 21: 309–40.

Hannan, M.T., Carroll, G.R. and Pólos, L. (2003b) A for-

mal theory of resource partitioning, Working Paper,

Graduate School of Business, Stanford University.

Hannan, M.T., Ranger-Moore, J. and Banaszak-Holl, J.

(1990) ‘Competition and the evolution of organiza-

tional size distributions’, in J.V. Singh (ed.), Organiza-

tional Evolution: New Directions.Newbury Park,CA:

Sage. pp. 246–68.

Haveman, H.A. (1992) ‘Between a rock and a hard place:

Organizational change and performance under condi-

tions of fundamental environmental transformation’,

Administrative Science Quarterly, 37: 48–75.

Haveman, H.A. (1993a) ‘Organizational size and change:

Diversification in the savings and loan industry after

deregulation’, Administrative Science Quarterly, 38: 20–50.

Haveman, H.A. (1993b) ‘Follow the leader: Mimetic

isomorphism and entry into new markets’, Administra-

tive Science Quarterly, 38: 593–627.

Haveman, H.A. (1993c) ‘Ghosts of managers past:

Managerial succession and organizational mortality’,

Academy of Management Journal, 36: 864–81.

Haveman, H.A. (1994) ‘The ecological dynamics of orga-

nizational change: Density and mass dependence in

rates of entry into new markets’, in J.A.C. Baum and

J.V. Singh (eds), Evolutionary Dynamics of Organiza-

tions. New York: Oxford University Press. pp. 152–66.

Haveman, H.A., Russo, M.V. and Meyer, A.D. (2001)

‘Organizational environments in flux: The impact of

regulatory punctuations on organizational domains,

CEO succession and Performance’, Organization

Science, 12: 253–73.

Hawley, A.H. (1950) Human Ecology: A Theory of

Community Structure. New York: Ronald.

Heckman, J.J. (1979) ‘Sample selection bias as a specifica-

tion error’, Econometrica, 47: 153–61.

Henderson, A. (1999) ‘Firm strategy and age dependence:

A contingent view of the liabilities of newness, adoles-

cence, and obsolescence’, Administrative Science Quar-

terly, 44: 281–324.

Henderson, R.M. and Clark, K.B. (1990) ‘Architectural

innovation: The reconfiguration of existing product

technologies and the failure of established firms’,

Administrative Science Quarterly, 35: 9–30.

Hotelling, H. (1929) ‘Stability in competition’, Economic

Journal, 39: 41–57.

Huber, G.P. (1991) ‘Organizational learning: The con-

tributing processes and the literatures’, Organization

Science, 2: 88–115.

Handbook of Organization Studies

106

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 106

Hybels, R.C. (1994) Legitimation, population density, and

founding rates: The institutionalization of commercial

biotechnology in the U.S., 1971–1989. Unpublished

PhD dissertation, Cornell University.

Hybels, R.C., Ryan, A.R. and Barley, S.R. (1994) ‘Alliances,

legitimation, and founding rates in the U.S.

Biotechnology field, 1971–1989’. Paper presented at the

Academy of Management national meetings, Dallas,

TX.

Ijiri, Y. and Simon, H.A. (1977) Skew Distributions and the

sizes of Business Firms. Amsterdam: North Holland.

Ingram, P. (1993) ‘Old, tired, and ready to die: The age

dependence of organizational mortality reconsidered’.

Paper presented at the Academy of Management meet-

ings, Atlanta, GA.

Ingram, P. (1994) Endogenizing environmental change:

The evolution of hotel chains, 1896–1980. Unpublished

manuscript, Johnson Graduate School of Management,

Cornell University.

Ingram, P. and Baum, J.A.C. (1997a) ‘Opportunity and

constraint: Organizations’ learning from the operating

and competitive experience of industries’, Strategic

Management Journal, 18: 75–98 (Summer Special

Issue).

Ingram, P. and Baum, J.A.C. (1997b) ‘Chain affiliation and

the failure of Manhattan hotels, 1898–1980’, Adminis-

trative Science Quarterly, 42: 68–102.

Ingram, P. and Inman, C. (1996) ‘Institutions, intergroup

competition, and the evolution of hotel populations

around Niagara Falls’, Administrative Science Quarterly,

41: 629–58.

Ingram, P. and Simons, T. (2000) ‘State formation, ideo-

logical competition, and the ecology of Israeli workers

cooperatives, 1920–1992’, Administrative Science

Quarterly, 45: 25–53.

Katz, J. and Gartner, W.B. (1988) ‘Properties of emerging

organizations’, Academy of Management Review, 13:

429–41.

Kelly, D. (1988) ‘Organizational transformation and fail-

ure, 1962–1985. Unpublished PhD dissertation,

Graduate School of Business, Northwestern University.

Kelly, D. and Amburgey, T.L. (1991) ‘Organizational iner-

tia and momentum: A dynamic model of strategic

change’, Academy of Management Journal, 34: 591–612.

Kogut, B., Shan, W., and Walker, G. (1993) ‘Knowledge in

the network and the network as knowledge: The struc-

ture of new industries’, in G. Grabher (ed.), The

Embedded Firm: On the Social Economics of Industrial

Networks, London: Routledge. pp. 67–94.

Kraatz, M. (1998) ‘Learning by association? Inter-

organizational networks and adaptation to environ-

mental change’, Academy of Management Journal, 41:

29–60.

Kraatz, M. and Zajac, E. (1996) ‘Exploring the limits of the

new institutionalism: The causes and consequences of

illegitimate organizational change’, American Socio-

logical Review, 61: 812–36.

Lamertz, K. and Baum, J.A.C. (1998) ‘The legitimacy of

organizational downsizing in Canada: An analysis

of media accounts, 1988–1995’, Canadian Journal of

Administrative Sciences, 15: 93–107.

Lauden, L. (1984) Progress and its Problems. Berkeley, CA:

University of California Press.

Lawrence, P.R. and Lorsch, J.W. (1967) Organizations and

Environments: Managing Differentiation and Integra-

tion. Boston, MA: Harvard Business School Press.

Lehrman, W. (1994) ‘Diversity in decline: Institutional envi-

ronment and organizational failure in the American life

insurance industry’, Social Forces, 73: 605–35.

Levinthal, D.A. (1991a) ‘Random walks and organizational

mortality’, Administrative Science Quarterly, 36: 397–420.

Levinthal, D.A. (1991b) ‘Organizational adaptation and

environmental selection–Interrelated processes of

change’, Organization Science, 2: 140–5.

Levinthal, D.A. (1994) ‘Surviving Schumpeterian environ-

ments: An evolutionary perspective’, In J.A.C. Baum and

J.V. Singh (eds), Evolutionary Dynamics of Organizations.

New York: Oxford University Press. pp. 167–78.

Levinthal, D.A. and March, J.G. (1993) ‘The myopia of

learning’, Strategic Management Journal, 14: 95–112.

Levitt, B. and March, J.G. (1988) ‘Organizational learning’,

Annual Review of Sociology, 14: 319–40.

Lomi, A. (1995) ‘The population ecology of organizational

founding: Location dependence and unobserved hetero-

geneity’, Administrative Science Quarterly, 40: 111–44.

Lomi, A. (2000) Density dependence and spatial duality in

organizational founding rates: Danish commercial

banks, 1846–1989’, Organization Studies, 21: 433–61.

Loree, D.W. (1993) ‘Organizational mortality: The price

paid for institutional linkages in the semiconductor

industry’, Paper presented at the Academy of

Management meetings, Atlanta GA.

March, J.G. (1991) ‘Exploration and exploitation in orga-

nizational learning’, Organization Science, 2: 71–87.

March, J.G. and Simon, H.A. (1958) Organizations.New

York: Wiley.

McGrath, J.E. (1982) ‘Dilemmatics: The study of research

choices and dilemmas’, in J.E. McGrath, J. Martin and

R.A. Kulka (eds), Judgement Calls in Research. Beverly

Hills, CA: Sage. pp. 69–102.

McKelvey, B. (1982) Organizational Systematics.Berkeley,

CA: University of California Press.

McKelvey, B. (1994) ‘Evolution and organizational science’,

in J.A.C. Baum and J.V. Singh (eds), Evolutionary

Dynamics of Organizations.New York:Oxford Univer-

sity Press. pp. 314–26.

McPherson, J.M. (1983) ‘An ecology of affiliation’,

American Sociological Review, 48: 519–32.

Messalam, A. (1998) ‘The organizational ecology of invest-

ment firms in Egypt: Organizational founding’, Organi-

zation Studies, 19: 23–46.

Meyer, J.W. and Rowan, B. (1977) ‘Institutionalized orga-

nizations: Formal structure as myth and ceremony’,

American Journal of Sociology, 83: 340–63.

Ecological Approaches to Organizations

107

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 107

Meyer, M.W. (1994) ‘Turning evolution inside the organi-

zation’, in J.A.C. Baum and J.V. Singh (eds), Evolution-

ary Dynamics of Organizations.New York:Oxford

University Press. pp. 109–16.

Mezias, J.M. and Mezias, S.J. (2000) ‘Resource partitioning,

the founding of specialist firms, and innovation: The

American feature film industry, 1912–1929’,

Organization Science, 11: 306–22.

Miller, D. (1990) The Icarus Paradox: How Exceptional

Companies Bring about Their Own Downfall.New York:

Harper Collins.

Miller, D. and Chen, M.-J. (1994) ‘Sources and conseqences

of competitive inertia: A study of the U.S. airline indus-

try’, Administrative Science Quarterly, 39: 1–23.

Miller, D. and Friesen, P.H. (1980) ‘Momentum and revo-

lution in organizational adaptation’, Academy of

Management Journal, 22: 591–614.

Miner, A.S. (1993) ‘Review of Dynamics of Organizational

Populations: Density, competition, and legitimation,by

Michael T. Hannan and Glenn R. Carroll’, Academy of

Management Review, 18: 355–67.

Miner, A.S. (1994) ‘Seeking adaptive advantage:

Evolutionary theory and managerial action’, in J.A.C.

Baum and J.V. Singh (eds), Evolutionary Dynamics of

Organizations. New York: Oxford University Press.

pp. 76–89.

Miner, A.S. and Haunschild, P.R. (1995) ‘Population level

learning’, in L.L. Cummings and B.M. Staw (eds),

Research in Organizational Behavior, 17. Greenwich CT:

JAI Press. pp. 115–66.

Miner, A.S., Amburgey, T.L. and Stearns, T. (1990)

‘Interorganizational linkages and population dynamics:

Buffering and transformational shields’, Administrative

Science Quarterly, 35: 689–713.

Minkoff, D. (1999) ‘Bending with the wind: Strategic

change and adaptation by women’s and racial minority

organizations’, American Journal of Sociology, 104:

1666–703.

Mitchell, W. (1989) ‘Whether and when? Probability and

timing of entry into emerging industrial subfields’,

Administrative Science Quarterly, 34: 208–30.

Mitchell, W. (1994) ‘The dynamics of evolving markets:

The effects of business sales and age on dissolutions

and divestitures’, Administrative Science Quarterly,

39(4): 575–99.

Mitchell, W. and Singh, K. (1993) ‘Death of the lethargic:

Effects of expansion into new technical subfields on

performance in a firm’s base business’, Organization

Science, 4: 152–80.

Nelson R.R. and Winter, S.G. (1982) An Evolutionary

Theory of Economic Change. Cambridge, MA: Harvard

University Press.

Oliver, C. (1991) ‘Strategic responses to institutional

processes’, Academy of Management Review, 16: 145–79.

Olzak, S. (1992) Dynamics of Ethnic Competition and

Conflict. Stanford, CA: Stanford University Press.

Olzak, S. and West, E. (1991) ‘Ethnic conflict and the rise

and fall of ethnic newspapers’, American Sociological

Review, 56: 458–74.

Palmer, D. and Barber, B. (2001) ‘Challengers, elites, and

owning families: A social class theory of corporate

acquisitions in the 1960s’, Administrative Science

Quarterly, 46: 87–122.

Péli, G. (1997) ‘The niche hiker’s guide to population ecol-

ogy: A logical reconstruction of organizational ecol-

ogy’s niche theory’, Sociological Methodology, 27: 1–46.

Pennings, J., Lee, K. and van Witteloostuijn, A. (1998)

‘Human capital, social capital, and firm dissolution’,

Academy of Management Journal, 41: 425–40.

Perrow, C. (1986) Complex Organizations: A Critical Essay,

d edn. New York: Random House.

Peteraf, M.A. (1993) ‘The cornerstones of competitive

advantage: A resource-based view’, Strategic Manage-

ment Journal, 14: 179–91.

Petersen, T. and Koput, K.W. (1991) ‘Density dependence in

organizational mortality: Legitimacy or unobserved het-

erogeneity’, American Sociological Review, 56: 399–409.

Pfeffer, J. and Salancik, G.R. (1978) The External Control of

Organizations. New York: Harper and Row.

Phillips, D. (2001) ‘The promotion paradox: Organiza-

tional mortality and employee promotion chances in

Silicon Valley law firms, 1946–1996’, American Journal

of Sociology, 106: 1058–98.

Phillips, D. (2002) ‘A genealogical approach to organiza-

tional life chances: The parent-progeny transfer among

Silicon Valley law firms, 1946–1996’, Administrative

Science Quarterly, 47: 474–506.

Podolny, J.M., Stuart, T.E. and Hannan, M.T. (1996)

‘Networks, knowledge, and niches: Competition in the

worldwide semiconductor industry, 1984–1991’,

American Journal of Sociology, 102: 659–89.

Pollock, T.G. and Rindova, V. (2003) ‘Media legitimation

effects in the market for initial public offerings’,

Academy of Management Journal, 46: 631–42.

Porter, M.E. (1980) Competitive Strategy.New York:The

Free Press.

Puccia,C.J. and Levins, R.(1985) Qualitative Modeling of Com-

plex Systems. Cambridge, MA: Harvard University Press.

Ranger-Moore, J. (1997) ‘Bigger may be better, but is older

wiser? Organizational age and size in the New York life

insurance industry’, American Sociological Review, 62:

903–20.

Ranger-Moore, J., Banaszak-Holl, J.J. and Hannan, M.T.

(1991) ‘Density-dependent dynamics in regulated

industries: Founding rates of banks and life insurance

companies’, Administrative Science Quarterly, 36: 36–65.

Ranger-Moore, J., Breckenridge, R.S. and Jones, D.L.

(1995) ‘Patterns of growth and size-localized competi-

tion in the New York State life insurance industry,

1860–1985’, Social Forces, 73: 1027–49.

Rao, H. (1994) ‘The social construction of reputation:

Certification contests, legitimation and the survival of

Handbook of Organization Studies

108

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 108

organizations in the American automobile industry;

1895–1912’, Strategic Management Journal, 15(Special

Issue): 29–44.

Rao, H. (2001) ‘The power of public competition:

Promoting cognitive legitimacy through certification

contests’, in C. Schoonhoven and E. Romanelli (eds),

The Enntrepreneurship Dynamic. Stanford CA: Stanford

University Press. pp. 262–86.

Rao, H. (2002) ‘Interorganizational ecology’, in J.A.C.

Baum (ed.), Companion to Organizations. Oxford, UK:

Blackwell. pp. 541–56.

Rao, H. and Neilsen, E.H. (1992) ‘An ecology of agency

arrangements: Mortality of savings and loan associa-

tions, 1960–1987’, Administrative Science Quarterly, 37:

448–70.

Romanelli, E. (1991) ‘The evolution of new organizational

forms’, Annual Review of Sociology, 17: 79–103.

Ruef, M. (1997) ‘Assessing organizational fitness on a

dynamic landscape: An empirical test of the relative iner-

tia thesis’, Strategic Management Journal, 18: 837–53.

Ruef, M. and Scott, W.R. (1998) ‘A multidimensional

model of organizational legitimacy: Hospital survival

in changing institutional environments’, Administrative

Science Quarterly, 43: 877–904.

Rumelt, R. (1986) Strategy, Structure, and Economic Perfor-

mance. Boston, MA: Harvard Business School Press.

Saloner, G. (1994) ‘Game theory and strategic manage-

ment: Contributions, applications and limitations’, in

R.P. Rumelt, D.E. Schendel and D.J. Teece (eds),

Fundamental Issues in Strategic Management. Boston

MA: Harvard Business School Press. pp. 155–94.

Scherer, F.M. and Ross, D. (1990) Industrial Market

Structure and Economic Performance,3

d edn. Boston:

Haughton Mifflin.

Scott, W.R. (1992) Organizations: Rational, Natural, and

Open Systems,3

d edn. Englewood Cliffs, NJ: Prentice-

Hall.

Scott, W.R. and Meyer, J.W. (1983) ‘The organization of

societal sectors’, in J.W. Meyer and W.R. Scott (eds),

Organizational Environments: Ritual and Rationality.

Beverly Hills, CA: Sage. pp. 1–16.

Shane, S. and Stuart, T. (2002) ‘Organizational endow-

ments and the performance of university start-ups’,

Management Science, 48: 154–70.

Singh, J.V. (1993) ‘Review essay: Density dependence

theory—Current issues, future promise’, American

Journal of Sociology, 99: 464–73.

Singh, J.V. and Lumsden, C.J. (1990) ‘Theory and research

in organizational ecology’, Annual Review of Sociology,

16: 161–95.

Singh, J.V., House, R.J. and Tucker, D.J. (1986a)

‘Organizational change and organizational mortality’,

Administrative Science Quarterly, 31: 587–611.

Singh, J.V., Tucker, D.J. and House, R.J. (1986b)

‘Organizational legitimacy and the liability of newness’,

Administrative Science Quarterly, 31: 171–93.

Singh, J.V., Tucker, D.J. and Meinhard, A.G. (1988) ‘Are

voluntary social service organizations structurally

inert? Exploring an assumption in organizational ecol-

ogy’, Academy of Management Meeting, Anaheim, CA.

Singh, J.V., Tucker, D.J. and Meinhard, A.G. (1991)

‘Institutional change and ecological dynamics’, in W.W.

Powell and P.J. DiMaggio (eds), The New Institutional-

ism in Organizational Analysis. Chicago, IL: University

of Chicago Press. pp. 390–422.

Singh, K. (1997) ‘The impact of technological complexity

and interfirm cooperation on business survival’,

Academy of Management Journal, 40: 339–67.

Sitkin, S.B. (1992) ‘Learning through failure: The strategy

of small losses’, in B.M. Staw and L.L. Cummings (eds),

Research in Organizational Behavior, 14. Greenwich,

CT: JAI Press. pp. 231–66.

Sorenson, O. (2000) ‘The effect of population level learn-

ing on market entry: The American automobile indus-

try’, Social Science Research, 29: 307–26.

Starbuck, W.H. (1965) ‘Organizational growth and devel-

opment’, in J.G. March (ed.), Handbook of

Organizations. Chicago: Rand-McNally. pp. 451–533.

Starbuck, W.H. (1983) ‘Organizations as action genera-

tors’, American Sociological Review, 48: 91–102.

Stinchcombe, A.L. (1965) ‘Social structure and organiza-

tions’, in J.G. March (ed.), Handbook of Organizations.

Chicago, IL: Rand McNally. pp. 153–93.

Stuart, T.E. and Sorenson, O. (2003a) ‘The geography of

opportunity: Spatial heterogeneity in founding rates

and the performance of biotechnology firms’, Research

Policy, 32: 229–53.

Stuart, T.E. and Sorenson, O. (2003b) ‘Liquidity events and

the geographic distribution of entrepreneurial activity’,

Administrative Science Quarterly, 48: 175–201.

Suhomlinova, O. (1999) ‘Constructive destruction:

Transformation of Russian state-owned construction

enterprises during market transition’, Organization

Studies, 20: 451–83.

Swaminathan, A. (1995) ‘The proliferation of specialist

organizations in the American wine industry:

1941–1990’, Administrative Science Quarterly, 40: 653–80.

Swaminathan, A. (2001) ‘Resource partitioning and the

evolution of specialist organizations: The role of loca-

tion and identity in the U.S. wine industry’, Academy of

Management Journal, 44: 1169–85.

Swaminathan, A. and Delacroix, J. (1991) ‘Differentiation

within an organizational population: Additional evidence

from the wine industry’, Academy of Management

Journal, 34: 679–92.

Swaminathan, A. and Wiedenmayer, G. (1991) ‘Does the

pattern of density dependence in organizational mor-

tality rates vary across levels of analysis? Evidence from

the German brewing industry’, Social Science Research,

20: 45–73.

Thompson, J.D. (1967) Organizations in Action.New York:

Academic Press.

Ecological Approaches to Organizations

109

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 109

Tilly, C. (1993) European Revolutions, 1492–1992.

Cambridge, MA: Blackwell.

Tirole, J. (1988) The Theory of Industrial Organization.

Cambridge, MA: MIT Press.

Tucker, D.J. (1994) ‘Progress and problems in population

ecology’, in J.A.C. Baum and J.V. Singh (eds),

Evolutionary Dynamics of Organizations.New York:

Oxford University Press. pp. 327–36.

Tucker, D.J., Baum, J.A.C. and Singh, J.V. (1992) ‘The insti-

tutional ecology of human service organizations’, in

Y. Hasenfeld (ed.), Human Service Organizations.

Newbury Park, CA: Sage. pp. 47–72.

Tucker, D.J., Singh, J.V. and Meinhard, A.G. (1990a)

‘Organizational form, population dynamics, and insti-

tutional change: The founding patterns of voluntary

organizations’, Academy of Management Journal, 33:

151–78.

Tucker, D.J., Singh, J.V. and Meinhard, A.G. (1990b)

‘Founding characteristics, imprinting, and organiza-

tional change’, in J.V. Singh (ed.), Organizational

Evolution: New Directions. Newbury Park CA: Sage.

pp. 182–200.

Tucker, D.J., Singh, J.V., Meinhard, A.G. and House, R.J.

(1988) ‘Ecological and institutional sources of change

in organizational populations’, in G.R. Carroll (ed.),

Ecological Models of Organizations. Cambridge, MA:

Ballinger. pp. 127–51.

Tuma, N.B. and Hannan, M.T. (1984) Social Dynamics:

Models and Methods. New York: Academic Press.

Tushman, M.L. and Anderson, P. (1986) ‘Technological

discontinuities and organizational environments’,

Administrative Science Quarterly, 31: 439–65.

Tushman, M.L. and Romanelli, E. (1985) ‘Organizational

evolution: A metamorphosis model of convergence and

reorientation’, Research in Organizational Behavior,7:

171–222.

Usher, J.M. (1991) ‘Exploring the effects of niche crowding

on rates of organizational change and failure’.

Unpublished manuscript, Faculty of Business,

University of Alberta.

Usher, J.M. and Evans, M.G. (1996) ‘Life and death along

gasoline alley: Darwinian and Lamarckian processes in

a differentiating population’, Academy of Management

Journal, 39: 1428–66.

Vermeulen, F. and Barkema, H. (2001) ‘Learning through

acquisitions’, Academy of Management Journal,44:

457–76.

Wade, J.B., Swaminathan, A. and Saxon, M.S. (1998)

‘Normative and resource flow consequences of local

regulations in the American brewing industry,

1845–1918’, Administrative Science Quarterly, 43:

905–35.

West, E. (1995) ‘Organization building in the wake of eth-

nic conflict: A comparison of three ethnic groups’,

Social Forces, 73: 1333–63.

White, H.C. (1981) ‘Where do markets come from?’,

American Journal of Sociology, 87: 517–47.

Wholey, D.R., Christianson, J.B. and Sanchez, S.M. (1992)

‘Organizational size and failure among health mainte-

nance organizations’, American Sociological Review, 57:

829–42.

Wiewel, W. and Hunter, A. (1985) ‘The interorganizational

network as a resource’, Administrative Science Quarterly,

30: 482–96.

Winter, S.G. (1990) ‘Survival, selection, and inheritance in

evolutionary theories of organization’, in J.V. Singh

(ed.), Organizational Evolution: New Directions.

Newbury Park CA: Sage. pp. 269–97.

Yelle, L.E. (1979) ‘The learning curve: Historical review

and comprehensive survey’, Decision Sciences, 10:

302–28.

Zajac, E.J. and Kraatz, M.S. (1993) ‘A diametric forces

model of strategic change: Assessing the antecedents

and consequences of restructuring in the higher educa-

tion industry’, Strategic Management Journal, 14(S):

83–102.

Zucker, L.G. (1977) ‘The role of institutionalization in

cultural persistence’, American Sociological Review, 42:

726–43.

Zucker, L.G. (1989) ‘Combining institutional theory

and population ecology: No legitimacy, no history’,

American Sociological Review, 54: 542–5.

Handbook of Organization Studies

110

Clegg (New)-3384-Ch-02.qxd 1/30/2006 6:35 PM Page 110

Engaging Local People in Conserving the Socio-Ecological Production Landscape and Seascape by Practicing Collaborative Governance in Mao’ao Bay, Chinese Taipei

Chapter

Full-text available

Sep 2023

This study shows how collaborative governance actions have been practiced to initiate restoration efforts in northeastern Chinese Taipei’s Mao’ao Bay since 2019. The objectives of these actions were (1) to integrate different voices of multiple stakeholders to achieve harmony within specific social-ecological systems in coastal areas; (2) to empower local communities and stakeholders to facilitate management in the socio-ecological production landscape and seascape (SEPLS) involving coastal villages; and (3) to engage local people in restoring coastal ecosystems in the SEPLS by developing a collaborative governance mechanism. We established a public–private collaboration framework for managing Mao’ao Bay. Actions comprised three main activities. First, we identified stakeholders’ critical attitudes, issues, and development strategies through importance-performance analysis (IPA) and a focus group meeting. Second, Satoumi environmental education was organised to increase awareness among the local people on marine environmental conditions. Meanwhile, training courses enabled local residents to conduct intertidal/underwater ecological surveys. Third, an ecological survey team was established to conduct underwater investigations on economically important local seaweed species, such as Gelidium spp., and to verify the marine ecosystem condition step by step. Lessons learned from these actions include: (1) the techniques applied were useful for information gathering and sharing toward development of strategies; (2) environmental education and underwater survey training courses not only enhanced locals’ awareness of resource utilisation but also nurtured citizen scientists to help monitoring of coastal resources; (3) the public–private collaboration not only established governance of coastal resources, but also enhanced the local people’s capacity for long-term SEPLS management.KeywordsSocio-ecological production landscapes and seascapes (SEPLS)Collaborative governanceStakeholdersSatoumiParticipatory action research (PAR)Citizen scienceMao’ao BayChinese Taipei

Community-Based Approach to Wetland Restoration: Case Study of the Songor Wetland, Ghana

Chapter

Full-text available

Sep 2023

The Songor Wetland is a designated Ramsar site and is listed as a biosphere under the Man and Biosphere Programme. The wetland covers an area of 11,500 ha, extending 20 km along the coast and 8 km inland behind a narrow strip of sand dunes on which small communities are situated. It is a habitat and breeding ground for several species of both economic and ecological importance, thus provides multi-functional benefits relating to the ecosystem, the economy, and ecotourism. The wetland is threatened by overfishing, pollution, mangrove exploitation, invasive species, and bushfires. To better understand communal approaches to ensuring the ecological health of the wetland, including restoration activities, Conservation Alliance International (CA) conducted a study to ascertain the ecological health and the socio-economic and cultural status of the landscape, using qualitative and quantitative approaches including surveying 376 respondents. Results revealed that most of the respondents (53%) put higher premium on provisioning services, followed by supporting services (31%). The value placed on the wetland was influenced by the level of dependence on it for livelihoods. Community Resource Management Committees (CRMCs) and traditional beliefs accounted for more than 80% of restoration activities by the communities. Management practices undertaken through the CRMCs and based on traditional beliefs included mangrove restoration, clearing of aquatic weeds, location of disoriented turtles, pest control, and ecological monitoring. Moreover, the two most impacting land use types were mining and tourism, which have been exacerbated by impacts of the COVID-19 pandemic. Policymakers should offer support to the traditional authorities and the CRMCs to deepen local participation in management and restoration activities towards enhancing the ecological health of the wetland.KeywordsCoastalCommunity-based approachMangrovesRestorationSongor Wetland

Human Rights Vigilantes

Chapter

May 2023

This chapter surveys the different approaches to vigilante human rights enforcement, looking first at actions enabled by developments in the law and considering the impact of technological change. It looks at one particular aspect of the history of international human rights or the human rights movement, which is the autonomous enforcement of international human rights laws. It also reviews the body of international human rights law that has been continuously expanding in terms of the number of international treaties and conventions and in terms of the number of states that adhere to them. The chapter points out how non-governmental organizations (NGOs) have represented victims in regional international courts. It mentions a transnational approach to human rights enforcement that has depended on the legal doctrine of civil or criminal universal jurisdiction.

Foreign Direct Investment Speed and Firm Performance: Evidence from China

Article

Jan 2023

Jing Yan

Minding Competition in Complex Adaptive Social Systems: The Sociological Approach to Competition Law

Article

Jan 2024

Ioannis Lianos

Understanding Competition in Social Space: Religious Congregations in Manhattan, 1949 to 1999

Article

Jan 2024

Casey Homan

Competition between social units has long been central to sociological theories of change. Understanding it has become particularly important in the sociology of religion with the theory of religious economies, a market model of religious change. Existing empirical tests of the theory are limited by (1) ambiguity regarding which religious groups are expected to compete with which other groups, and/or (2) a neglect of the local level (competition among congregations). Using an original compilation of the life histories of religious congregations in Manhattan from 1949 to 1999, I conduct event-history analyses that avoid those limitations. The chief results are the following: (1) the more congregations there were near a given congregation that were theologically dissimilar to that congregation, the less likely that congregation was to advertise; (2) when there was an increase over time in the number of nearby congregations that were theologically similar to the focal congregation, that congregation became more likely to advertise; and (3) when there was an increase over time in the number of nearby congregations that were theologically dissimilar to the focal congregation, that congregation became less likely to advertise. Implications for the study of religion include modifications of religious-economies theory; broader implications speak to understanding the social units that compete and what drives competition.

Organizational non-conformity in an emerging economy: Exploring the non-adoption of credit rating in China

Article

Full-text available

Dec 2023
Asia Pac J Manag

Institutional scholarship tends to emphasize the tendency of organizations to conform to prevailing practices, but this study investigates Chinese firms’ non-conformity behavior in terms of not participating in credit rating. State ownership and firm status (in terms of age, size, and human capital) are all found to be useful predictors of this non-conformity. Building on institutional theory and resource dependence theory in an emerging market context, this study proposes that non-conformity would be high for state owned enterprises (SOEs) and for both low- and high-status firms, based on their evaluations of the legitimacy of credit rating and the relative power balance between the government and themselves. In contrast, middle-status firms would be less likely to show non-conformity behaviors. Moreover, the influence of state ownership and firm status on non-conformity would be further moderated by the degree of government intervention. The results from an empirical study of 2,708 manufacturing firms in China largely support these hypotheses.

Higher education and organizational theory: Systems, fields, markets, and populations in an increasingly global context

Chapter

Full-text available

Oct 2023

Jeroen Huisman

The categories commonly mobilized to think about education have long been associated with the notion of the nation state and have functioned as obstacles, rather than resources, for our understanding of how globalization plays out in this particular field. In the last two decades, both social theory and comparative politics have attempted to overcome these limitations in their own way. Social theory increasingly acknowledged education as a global phenomenon. Theories have been developed to describe a global society evolving across borders. They show how, through processes that remain debated (cultural isomorphism, capitalism, functional differentiation), a number of structural and semantic evolutions have spread across education systems. Part I of this Handbook is dedicated to presenting, discussing, and comparing three such theories of globalization and their implications for our understanding of education and education policy. Comparative politics has for its part concerned itself with developing a more complex, less unified, and “transformationalist” view of the state by acknowledging the fragmentation and distribution of its functions among distinct domains and levels. Part II gravitates around this global constellation, with chapters focusing on global reforms, norms, and ideas put forward by supranational organizations, on international accountability processes and on the ways in which nation states or local actors adopt, implement, or resist global ideas and reforms. The two parts reflect these disciplinary approaches to the relation between globalization and education. Together, these two approaches seek to provide a comprehensive overview of how globalization and education interact to result in distinct and varying outcomes across world regions.

Competition and Strategic Differentiation Among Transnational Advocacy Groups

Chapter

Oct 2023

Mette Eilstrup-Sangiovanni

Why do some transnational advocacy groups adopt radical, confrontational tactics whereas others focus on ‘inside’ lobbying and information provision? Why do some advocacy groups appeal to large global audiences while others approach decision-makers behind closed doors? Bringing together interest group studies and population ecology theory, this chapter examines how ecological dynamics affect strategic specialisation among transnational advocacy groups (TAGs). Building on the framework in Chap. 1, this chapter argues that increasing resource competition resulting from organisational crowding, along with the introduction of new legal and technological tools has led to growing strategic differentiation among TAGs, and has prompted a strategic division of labor whereby some groups (mainly larger, well-established and resource-rich groups) specialise in gaining political access and media attention, while others (mainly smaller, less established groups) focus on developing ‘niche’ agendas and strategies including, inter alia, radical protest, monitoring and enforcement, and litigation. The chapter illustrates the argument with quantitative data and comparative cases from the realm of transnational environmental conservation advocacy.

Ecological constraints on audience size in the digital media system: evidence from the longitudinal tracking data from 2019 to 2022

Article

Aug 2023
HUM COMMUN RES

The current high-choice media environment is characterized by increasingly intense competition for audience resources among media products. Drawing on research in organizational ecology and communication, audience behavior, and media economics, this study provides an ecological explanation for audience size in the digital media system. The analysis of Comscore's aggregate audience data on the use of 64 digital media platforms over a 38-month period from 2019 to 2022 shows that a platform's audience size is negatively associated with its niche overlap with another platform. This competitive effect of niche overlap on a platform's current audience size is moderated by its previous audience size (size dependence) and species membership in the environment (species dependence). The results also support the short-and long-term stability of audience size.

Evolutionary Processes and Patterns of Core Business Change

Chapter

Mar 1994

The book presents the latest research and theory about evolutionary change in organizations. It brings together the work of organizational theorists who have challenged the orthodox adaptation views that prevailed until the beginning of the 1980s. It emphasizes multiple levels of change - distinguishing change at the intraorganizational level, the organizational level, the population level, and the community level. The book is organized in a way that gives order and coherence to what has been a diverse and multidisciplinary field. (The book had its inception at a conference held at the Stern School of Business, New York University, January 1992.)

Progress and Problems in Population Ecology

Chapter

Mar 1994

David J Tucker

An Intraorganizational Ecological Perspective on Managerial Risk Behavior, Performance, and Survival: Individual, Organizational, and Environmental Effects

Chapter

Mar 1994

Organizational Hierarchies and Evolutionary Processes: Some Reflections on a Theory of Organizational Evolution

Chapter

Mar 1994

Minimalism, Mutualism, and Maturity: The Evolution of the American Trade Association Population in the 20th Century

Chapter

Mar 1994

Turning Evolution Inside the Organization

Chapter

Mar 1994

Marshall W Meyer

Disruptive Selection and Population Segmentation: Interpopulation Competition as a Segregating Process

Chapter

Mar 1994

Externalities and Ecological Theory: Unbundling Density Dependence

Chapter

Mar 1994

Density-Independent Selection and Community Evolution

Chapter

Mar 1994

Jack Brittain

Evolutionary Dynamics of Organizational Populations and Communities

Article

May 2021

Evolution is conceptualized as a multi-level phenomenon (subunit–organization–organizational field–national economy) that links organizational and ecological systems. Analysis of population and community-level evolution emphasizes the roles of institutional change (e.g., industry deregulation, globalization, market reforms), technological innovation cycles (e.g., technological discontinuities, dominant designs), entrepreneurs, and social movements as triggers of organizational variation. Institutional and technological change transforms the dynamics of organizational communities by shifting the boundaries of organizational forms, destabilizing or reinforcing existing community structures, giving rise to consensus and/or conflict oriented social movements, and creating opportunities for entrepreneurs and venture capitalists to shape new organizational forms.

Ecological Approaches to Organizations

Abstract and Figures

Recommended publications

A critical review of manufacturing and supply chain research through co-words analysis: 2004–2014

Individual Readiness for Organizational Change

ORGANISATIONAL CAPABILITIES, COMPETITIVE ADVANTAGE AND PERFORMANCE IN SUPPORTING INDUSTRIES IN VIETN...

Expanding the Conversation on the Community-Organizational Interface in the Context of Diversity