3
Structural Holes versus Network
Closure as Social Capital
Ronald S. Burt
This chapter is about two network structures that have been argued to create
social capital. The closure argument is that social capital is created by
a network of strongly interconnected elements. The structural hole argument
is that social capital is created by a network in which people can broker
connections between otherwise disconnected segments. I draw from a
comprehensive review elsewhere (Burt 2000) to support two points in this
chapter: there is replicated empirical evidence on the socialcapital of structural
holes, and the contradiction between network closure and structural
holes can be resolved in a more general network model of social capital.
Brokerage across structural holes is the source of value added, but closure
can be critical to realizing the value buried in structural holes.
SOCIAL CAPITAL METAPHOR
The two arguments are grounded in the same social capital metaphor, so
it is useful to begin with the metaphor as a frame of reference. Cast in diverse
styles of argument (e.g.,Coleman 1990;Bourdieu & Wacquant 1992;
Burt 1992;Putnam 1993),socialcapital is a metaphor about advantage. Society
can be viewed as a market in which people exchange all variety of
goods and ideas in pursuit of their interests. Certain people, or certain
groups of people, do better in the sense of receiving higher returns to their
efforts. Some people enjoy higher incomes. Some more quickly become
31
32 Structural Holes versus Network Closure as Social Capital
prominent. Some lead more important projects. The interests of some are
better served than the interests of others. The human capital explanation
of the inequality is that the people who do better are more able individuals;
they are more intelligent, more attractive, more articulate, more skilled.
Socialcapital is the contextual complement to human capital.The social
capital metaphor is that the people who do better are somehow better connected.
Certain people or certain groups are connected to certain others,
trusting certain others, obligated to support certain others, dependent on
exchange with certain others.Holding a certain position in the structure of
these exchanges can be an asset in its own right. That asset is social capital,
in essence, a concept of location effects in differentiated markets. For
example, Bourdieu is often quoted in defining social capital as the resources
that result from social structure (Bourdieu & Wacquant 1992:119,
expanded from Bourdieu 1980):“social capital is the sum of the resources,
actual or virtual, that accrue to an individual or group by virtue of possessing
a durable network of more or less institutionalized relationships
of mutual acquaintance and recognition.” Coleman, another often-cited
source, defines socialcapital as a function of social structureproducing advantage
(Coleman 1990:302;from Coleman 1988:S98):”Socialcapital is defined
by its function.It isnot a singleentity but a variety of different entities
having two characteristics in common: They all consist of some aspect of a
social structure, and they facilitate certain actions of individuals who are
within the structure. Like other forms of capital, social capital is productive,
making possible the achievement of certain ends that would not be attainable
in its absence.” Putnam (1993:167) grounds his influential work
in Coleman’smetaphor, preserving the focus on action facilitated by social
structure: ”Socialcapital here refers to features of socialorganization, such
as trust, norms, and networks, that can improve the efficiencyof societyby
facilitating coordinated action.” I echo the above with a social capital
metaphor to begin my argument about the competitive advantage of structural
holes (Burt 1992:8,45).
So there is a point of general agreement from which to begin a discussion
of social capital. The cited perspectives on social capital are diverse in
origin and style of accompanying evidence, but they agree on a social-capital
metaphor in which social structure is a kind of capital that can create
for certain individuals or groups a competitive advantage in pursuing
their ends. Better connected people enjoy higher returns.
TWO NETWORK MECHANISMS
Disagreements begin when social capital as a metaphor is made concrete
with network models of what it means to be ”better” connected. Coimec-
33
Density Table of RelationsWithin and BetweenGroups
.65
.05 .25
.OO .01 .65 Group C (5 people and 8 ties; 5 strong, 3 weak)
Group A (5 people and 8 ties; 5 strong, 3 weak)
Group 8 (17 peopie and 41 ties; 27 strong, 14 weak)
Figure 1. Network around Robert and James.
tions are grounded in the history of a market. Certain people have met frequently.
Certain people have sought one another out. Certain people have
completed exchanges with one another. There is at any moment a network,
as illustrated in Figure 1, in which individuals are variably connected to
one another as a function of prior contact, exchange, and attendant emotions.
Figure 1is a generic sociogram and density table description of a network.
People are dots. Relationships are lines.Solid (dashed)lines connect
pairs of people who have a strong (weak) relationship.
In theory, the network residue from yesterday should be irrelevant to
market behavior tomorrow. I buy from the seller with the most attractive
offer. That seller may or may not be the seller I often see at the market, or
34 Structural Holes versus Network Closure as Social Capital
the seller from whom I bought yesterday. So viewed, the network in Figure
1 would recur tomorrow only if buyers and sellers come together as
they have in the past. The recurrence of the network would have nothing
to do with the prior network as a casual factor.Continuity would be a byproduct
of buyers and sellers seeking one another out as a function of supply
and demand.
Selecting the best exchange, however, requires that I have information
on available goods, sellers,buyers, and prices. Information can be expected
to spread across the people in a market, but jt will circulate within groups
before it circulatesbetween groups.A generic research finding in sociology
and social psychology is that information circulates more within than between
groups-within a work group more than between groups, within a
division more than between divisions, within an industry more than between
industries. For example, the sociogram in Figure 1and the density
table at the bottom of the figure show three groups (A,B,C), and the generic
pattern of ingroup relations stronger than relations between groups (diagonal
elements of the density table are higher than the off-diagonals, each
cell of the density table is the average of relations between individuals in
the row and individuals in the column).The result is that people are not simultaneously
aware of opportunities in all groups. Even if information is
of high quality, and eventually reaches everyone, the fact that diffusion occurs
over an interval of time means that individuals informed early or
more broadly have an advantage.
Structural Holes as Social Capital
Participation in, and control of, information diffusion underlies the social
capital of structural holes (Burt 1992).The argument describes social capital
as a function of brokerage opportunities, and draws on network
concepts that emerged in sociology during the 1970s,most notably Granovetter
(1973)on the strength of weak ties, Freeman (1977)on betweenness
centrality,Cook and Emerson (1978)on the benefits of having exclusive exchange
partners, and Burt (1980)on the structural autonomy created by
complex networks. More generally, sociological ideas elaborated by Simme1
(1955[1922])and Merton (1968[1957])on the autonomy generated by
conflicting affiliations are mixed in the hole argument with traditional economic
ideas of monopoly power and oligopoly to produce network models
of competitive advantage.
The weaker connections between groups in Figure 1are holes in the social
structure of the market. These holes in social structure-or more simply,
structural holes-create a competitive advantage for an individual
whose relationships span the holes. The structural hole between two
groups does not mean that people in the groups are unaware of one an-
other.It only means that the people are focused on their own activitiessuch
that they do not attend to the activities of people in the other group. Holes
are buffers, like an insulator in an electric circuit. People on either side of
a structural hole circulate in different flows of information. Structuralholes
are thus an opportunity to broker the flow of iizforiizntioii between people,
and corztrol the projects that bring together people from opposite sides of
the hole.
Structural holes separate nonredundant sources of information, sources
that are more additive than overlapping. There are two indicators of redundancy:
cohesion and equivalence.Cohesive contacts (contactsstrongly
connected to each other) are likely to have similar information and therefore
provide redundant information benefits. Structurally equivalent contacts
(contactswho link a manager to the same third parties) have the same
sources of information and therefore provide redundant information
benefits.
Robert and James in Figure 1have the same volume of connections, six
strong ties and one weak tie, but Robert has something more. James is tied
to people within group B, and through them to friends of friends all within
group B, soJames is well informed about cluster B activities. Robert is also
tied through friends of friends to everyone within group B, but in addition,
his strong relationship with person ”7” is a conduit for information on
group A, and his strong relationship with ”6” is a conduit for information
on group C. His relationship with 7 is for Robert a network bridge in that
the relationship is his only direct connection with group A. His relationship
with contact 6 meets the graph-theoretic definition of a network
bridge. Break that relationship and there is no connection between groups
B and C. More generally, Robert is a broker in the network. Network constraint
is an index that measures the extent to which a person’s contacts are
redundant (Burt 1992).James has a constraint score twice Robert’s (30.9
versus 14.8)and Robert is the least constrained of the people in Figure 1
(-1.4 Z-score). Network betweenness, proposed by Freeman (1977)’is an
index that measures the extent to which a person brokers indirect connections
between all other people in a network. Robert’sbetweenness score of
47.0 shows that almost half of indirect connections run through him. His
score is the highest score in Figure 1, well above average (47.0 is a 4.0
Z-score), and much higher than James’s5.2score, which is below average.
Robert’sbridge connections to other groups give him an advantage with
respect to information access. He reaches a higher volume of information
because he reaches more people indirectly. Further, the diversity of his contacts
across the three separate groups means that his higher volume of information
contains fewer redundant bits of information. Further still,
Robert is positioned at the crossroads of social organization so he is early
to learn about activities in the three groups. He corresponds to the ”opin-
36 Structural Holes versus Network Closure as Social Capital
ion leaders” proposed in the early diffusion literature as the individuals
responsible for the spread of new ideas and behaviors (Burt 1999a,b).
Moreover,Robert’smore diverse contacts mean that he is more likely to be
a candidate discussed for inclusion in new opportunities. These benefits
are compounded by the fact that having a network that yields such benefits
makes Robert more attractive to other people as a contact in their own
networks.
There is also a control advantage. Robert is in a position to bring together
otherwise disconnected contacts, which gives him a disproportionate
say in whose interests are served when the contacts come together.
Moreover, the holes between his contacts mean that he can broker communication
while displaying different beliefs and identities to each contact
(”robust action” in Padgett and Ansell 1993; see Brieger 1995on the connection
with structural holes). Simmel and Merton introduced the sociology
of people who derive control benefits from structural holes: The ideal
type is the tcrtizrs pzidcizs (literally,”thethird who benefits”),a person who
benefits from brokering the connection between others (seeBurt 1992,30-
32, for review). Robert in Figure 1is an entrepreneur in the literal sense of
the word-a person who adds value by brokering the connection between
others (Burt 1992,34-36; see also Aldrich 1999,Chap. 4; Thornton 1999).
There is a tension here, but not the hostility of combatants. It is merely uncertainty.
In the swirling mix of preferences characteristic of social networks,
where no demands have absolute authority, the tertizis negotiates
for favorable terms. Structural holes are the setting for tcvtizrs strategies,
and information is the substance.Accurate, ambiguous, or distorted information
is strategically moved between contacts by the tertizis. The information
and control benefits reinforce one another at any moment and
cumulate together over time.
Thus, individuals with contact networks rich in structural holes are the
individuals who know about, have a hand in, and exercise control over
more rewarding opportunities. The behaviors by which they develop the
opportunities are many and varied, but the opportunity itself is at all times
defined by a hole in the social structure. In terms of the argument, networks
rich in the entrepreneurial opportunities of structural holes are entrepreneurial
networks, and entrepreneurs are people skilled in building
the interpersonal bridges that span structural holes. They monitor information
more effectivelythan bureaucratic control. They move information
faster, and to more people, than memos. They are more responsive than a
bureaucracy, easily shifting network time and energy from one solution to
another (vividly illustrated in networks of drug traffic: Williams 1998;
Morselli2000; or health insurance fraud: Tillman& Indergaard 1999).More
in control of their surroundings, brokers like Robert in Figure 1can tailor
Xoiinld zyxwvutsS.Biirt 37
solutions to the specific individuals being coordinated, replacing the boilerplate
solutions of formal bureaucracy. To these benefits of faster, better
solutions, add cost reductions; entrepreneurial managers offer inexpensive
coordination relative to the bureaucratic alternative. Speeding the process
toward equilibrium, individuals with networks rich in structural holes operate
somewhere between the force of corporate authority and the dexterity
of markets, building bridges between disconnected parts of a market
where it is valuable to do so.
In sum, the hole prediction is that in comparisons between otherwise
similar people like Jamesand Robert in Figure 1,it is Robert who has more
socialcapital. His network across structuralholes give him broad, early access
to, and entrepreneurial control over, information.
Network Closure as Social Capital
Coleman’s (1988, 1990)view of social capital focuses on the risks associated
with being a broker. I will refer to Coleman’s view as a closure argument.
The key idea is that networks with closure-that is to say, networks
in which everyone is connected such that no one can escape the notice of
others, which in operational terms usually means a dense network-are
the source of social capital.
Network closure does two things for people in the closed network. First,
it affects access to information (Coleman 1990:310;cf. 1988:S104):”An important
form of social capital is the potential for information the inheres in
socialrelations. . . .Aperson who is not greatly interested in current events
but who is interested in being informed about important developments
can save the time required to read a newspaper if he can get the information
he wants from a friend who pays attention to such matters.” For example,
noting that information quality deteriorates as it moves from one
person to the next in a chain of intermediaries, Baker (1984;Baker & Iyer
1992)argues that markets with networks of more direct connections improve
communication between producers, which stabilizesprices, the central
finding in Baker’s (1984)analysis of a securities exchange.
Second, and this is the benefit more emphasized by Coleman, network
closure facilitates sanctions that make it lessrisky for people in the network
to trust one another. Illustrating the trust advantage with rotating-credit
associations, Coleman (1988:S103;1990:306-7; see Biggart 2000 for a closer
look at how such associations operate) notes, ”But without a high degree
of trustworthiness among the members of the group, the institution could
not exist-for a person who receives a payout early in the sequence of
meetings could abscond and leave the others with a loss. For example, one
could not imagine a rotating-credit association operating successfully in
38 Structural Holes versus Network Closure as Social Capital
urban areas marked by a high degree of social disorganization-or, in
other words, by a lack of social capital.” With respect to norms and effective
sanctions, Coleman (1990:310-11; cf. 1988:S104)says; ”When an effective
norm does exist,it constitutes a powerful, but sometimes fragile, form
of social capital.. . . Norms in a community that support and provide
effective rewards for high achievement in school greatly facilitate the
school’s task.” Coleman (1988:S107-8) summarizes: ”The consequence of
this closure is, as in the case of the wholesale diamond market or in other
similar communities, a set of effective sanctions that can monitor and
guide behavior. Reputation cannot arise in an open structure, and collective
sanctions that would ensure trustworthiness cannot be applied.” He
continues (1990:318);”The effect of closure can be seen especially well by
considering a system involving parents and children. In a community
where there is an extensive set of expectations and obligations connecting
the adults, each adult can use his drawing account with other adults to help
supervise and control his children.”
Coleman’s closure argument is prominent with respect to socialcapital,
but it is not alone in predicting that dense networks facilitate trust and
norms by facilitating effective sanctions. In sociology,Granovetter (1985,
1992:44)argues that the threat of sanctions makes trust more likely between
people who have mutual friends (mutual friends being a condition
of ”structural embeddedness”): “My mortification at cheating a friend of
long standing may be substantial even when undiscovered. It may increase
when the friend becomes aware of it. But it may become even more unbearable
when our mutual friends uncover the deceit and tell one another.”
There is an analogous argument in economics (the threat of sanctions creating
a ”reputation” effect,e.g.,Tullock 1985;Greif 1989):Mutual acquaintances
observing two people (a) make behavior between the two people
public, which (b)increases the salience of reputation for entry to future relations
with the mutual acquaintances, (c) making the two people more
careful about the cooperative image they display, which (d)increases the
confidence with which each can trust the other to cooperate. This chapter
is about socialcapital, so I focus on Coleman’sprediction that network closure
creates social capital. I have elsewhere discussed the network structures
that facilitate trust, showing that closure’s association with distrust
and character assassination is as strong as its association with trust (Burt
1999a,2001).
The closure prediction, in sum, is that in comparisons between otherwise
similar people likeJames and Robert in Figure 1,it is James who has
more social capital. Strong relations among his contacts give James more
reliable communication channels, and protect him from exploitation because
he and his contacts are more able to act in concert against someone
who violates their norms of conduct.
NETWORK EVIDENCE
39
Figure 2 contains graphs describing five study populations of managers. I
focus on these managers because on them I have detailed and comparable
network data. Managers in four of the Figure 2populations completed network
questionnaires in which they were asked to name (a) people with
whom they most often discussed important personal matters, (b)the people
with whom they most often spent free time, (c) the person to whom
they report in the firm, (d)their most promising subordinate, (e)their most
valued contacts in the firm, (f) essential sources of buy-in, (g)the contact
most important for their continued success in the firm, (h)their most difficult
contact, and (i)the people with whom they would discuss moving to
a new job in another firm. After naming contacts, respondents were asked
about their relation with each contact, and the strength of relations between
contacts (see Burt 1992:121-25, 199713; Burt, Hogarth, & Michaud,
2000, for item wording and scaling).
The horizontal axis of each graph in Figure 2 is a network constraint
index, C, that measures social capital. Network constraint measures the
extent to which a network is directly or indirectly concentrated in a single
contact. Constraint varies with three dimensions of a network: size,
density, and hierarchy (seeBurt 1992:50ff.,1995,1998,2000).Constraint is
low in large networks of disconnected contacts. Constraint is high in a
small network of contacts who are close to one another (density), or
strongly tied to one central contact (hierarchy). The index begins with a
measure of the extent to which manager i’s network is directly or indirectly
invested in his or her relationship with contact j: c , ~= (plj+
Cqplqpql)2,for q Z i,j, where p, is the proportion of i’s relations invested
in contact j. The total in parentheses is the proportion of i’s relations that
are directly or indirectly invested in connection with contact j. The sum
of squared proportions, ]clj,is the network constraint index C. I multiply
scores by 100.
As a frame of reference, network constraint is 27.9on average across the
841 observations in Figure 2, with a 10.5standard deviation. The network
around Robert in Figure 1is less constrained than average (C = 15).Robert
would appear to the far left in each Figure 2 graph. The network around
Tames is slightly more constrained than average (C = 31).
Association between performance and network constraint is a critical
test for the two leading network mechanisms argued to provide social
capital. More constrained networks span fewer structural holes, which
means less social capital according to the hole argument. rf networks t h t
span strtictliral holes are the source of social capital, then perforiizaizce shozild h w e
IZ rzegatizv association with network constraint. More constraint means more
network closure, and so more social capital according to the closure argu-
A
0
Izyxwvutsrqponmlkjihgf8"0 0
Probability of Poor Evaluation ;'
( 3 . 3 logit t-lesl) :
0 6
2 0 -
0 4
0 7 Probability
of
0 2 - Outstanding Evaluation
0 zyxwvutsrqponmlI - /*+
I , I , I
10 15 20 25 30 3 1 40 zyxwvutsrqponm45 50
0 Early Promotion
Y = 2 035 - 074(C)
o o o 0 r = - 4 0
Network Constraint
many ~ Structural Holes -- few
(manager C above. mean C in team below)
1.0-
0 0 -
1 0 -2
0 -
4.0
3 0 -
2 0 -
1.0-
0.0 -
1 0 -2
0
I I I , , . , , ,
Recognition of
'IQM Team
Achievements
Y = 7 525 - 2I0(C)
r = -.79
1;-54
P < 001
0
K0
-3 0 0
15 20 25 30 35 40 45
Network Constraint
many Structural Holes ~ few
(C for manager's network)
15 20 25 30 35 40 45 50
2 030
P .001
French Salary
0
D
E
o o Large Bonus
Y 0 438 - 02Z(C)
r = - 3 0
1 = - 3 7
P C 0010 zyxwv0 o 0
0 10 20 30 40 50 60 70 80
Network Constraint
many ___ Structural Holes -few
(C for officer's network)
0 10 20 30 40 50 60 70 80
1 I I , , I I , .
Probability of Small Bonus
(3.6 logit (-test)
/
Probability o f Large Bonua
(-2 7 log11I-tesl)
F
4 0
3 0
2.0
I.o
0 0
-1.0
-2 0
-3.0
0 6
0 5
0 4
0 3
0 2
0 1
0 0
Figure 2. Social capital matters.
40
Xol?ald s.Blfl’t 41
ment. rf network doslire zyxwvis the source of social capital, then perforiiiarzce shoirld
have a positive association with coizstraiizt.
The vertical axes in Figure 2 measure performance (explainedbelow for
each study population). Each graph in Figure 2 shows a strong negative association,
supporting the argument that structural holes are the source of
social capital.
Perfo mance zyxwvuEv a1uations
Graphs A and B show a negative association between network constraint
and performance evaluations. Figure 2A is based on a representative sample
of staff officers within the several divisions of a large financial organization
in 1996(Burt, Jannotta, & Mahoney 1998).The dependent variable
is job performance evaluation, taken from company personnel records.
Employees are evaluated at the end of each year on an A, B, C scaleof ”outstanding”
to ”poor” with plus and minus used to distinguish higher from
lower performances within categories. The evaluations stay with an employee
over time to affect future compensation and promotion. Women are
the majority of the several hundred employees in the staff function (76%)
of all officers within the function). Of 160 staff officers who returned network
questionnaires, the majority are women (69%).The results in Figure
2 are for the women (see Burt 2000:Table2, for the men). Graph A in Figure
2 shows how the probability of an ”outstanding” and a ”poor” evaluation
changes with network constraint. The graph is based on a logit
regression predicting the two extremes of evaluation with the middle category
as a reference point. Evaluations are adjusted for the four management
job ranks defined by the firm because more senior officers are more
likely to be evaluated as ”outstanding” (Burt, Jannotta, & Mahoney 1998:
84). Officers with less constrained networks, like Robert, have a significantly
higher probability of receiving an outstanding evaluation (-2.3 ttest).
The stronger effect is the tendency for officers living in the closeted
world of a constrained network to receive a ”poor” evaluation (3.3t-test).
Figure 2B is taken from Rosenthal’s (1996)dissertation research on the
socialcapital of teams.Troubledby the variable successof total quality management
(TQM) and inspired by Ancona and Caldwell’s (1992a, 199213)
demonstration that networks beyond the team are associated with team
performance, Rosenthal wanted to see whether the structure of external relationships
for TQM teams had the effect predicted by the hole argument.
She gained accessto a midwest manufacturing firm in 1994that was in the
process of using TQM teams to improve quality in all of its functions in its
several plants (a total of 165teams).Sheobserved operations in two plants,
then asked the senior manager responsible for quality in each plant to evaluate
the performance of each TQM team in his or her plant. Evaluations
42 Structural Holes versus Network Closure as Social Capital
were standardized within plants, then compared across plants to identify
functions in which team performance most varied. The study population
was teams assigned to a function with high successin someplants and low
success in other plants. Selecting two functionsfor study,Rosenthal sent to
each employee on the selected teams a network questionnaire; the survey
data were used to compute constraint in each person’s network within and
beyond the team.
The vertical axis in Figure 2B is the standardized team evaluation, and
the horizontal axis is average constraint on people in the team. The association
is as predicted by the hole argument, and quite striking (- .79correlation).
Teams composed of people whose networks extend beyond the
team to span structural holes in the company are significantly more likely
to be recognized as successful.
Pro zyxwvurnotions
Figure 2C shows a negative association between promotion and network
constraint. The data are taken from a probability sample of senior managers
in a large electronics manufacturer in 1989. Performance and network
data on these managers have been discussed in detail elsewhere
(Burt l992,1995,1997a,b, 1998).Survey network data were obtained on diverse
relationships using the questions described above. Performance and
background data on each manager were taken from company personnel
records. Company personnel records provided each manager’s rank (four
levels defined by the firm),date promoted to current rank, date entered the
firm, functional area of responsibility (defined by the firm as sales, service,
manufacturing, information systems, engineering, marketing, finance, and
human resources), and the usual personnel-file variables such as gender,
family, income, and so on.
Income in the study population was too closely tied to job rank to measure
the relative success of individual managers. Time to rank was a better
performance variable (Burt 1992:196-7).Whether promoted internally or
hired from the outside, people promoted to senior rank in large organizations
have several years of experience preceding their promotion. A period
of time is expected to pass before people are ready for promotion to senior
rank (seeMerton 1984,on socially expected durations). How much time is
an empirical question, the answer to which differs among individual managers.
Some managers are promoted early. Early promotion is the difference
between when a manager was promoted to his current rank and a
human-capital baseline model predicting the age at which similar managers
are promoted to the same rank to do the same work: E(age) - age.
Expected age at promotion E(age),is the average age at which managers
with specific personal backgrounds (education, race, gender, and senior-
Ro12flld s.Blll't 43
ity) have been promoted to a specificrank within a specific function (rank,
function, and plant location).Expected age at promotion is 12%of the population
variance in promotion age, and residuals are distributed in a bell
curve around expected promotion age (Burt 1992:126-31; 1995).The criterion
variable in Figure 2C is the early promotion variable standardized to
zero mean and unit variance.
Figure 2C contains the 170 most senior men responding to the survey
(seeBurt 1998:14,for the senior women).The negative associationbetween
early promotion and constraint is statistically significant (-5.4 t-test).Men
promoted early to their current senior rank tend to have low-constraint
networks (left side of the graph), while those promoted late tend to have
high-constraint networks (right side of the graph).
Compensation
Graphs D, E, and F show negative associations between compensation and
network constraint. Figure 2D contains 60 people who were a representative
sample of senior managers across functions in a division of a large
French chemical and pharmaceuticals company in 1997(Burt,Hogarth, &
Michaud 2000). Again, survey network data were obtained on diverse relationships
using the questions described above. Performance and background
data on managers in the study population were taken from
company personnel records. Seventy-twopercent of the study-population
variance in annual salaries canbe predicted from a manager's job rank and
age (salary slightly more associated with age than seniority).The residual z28% of salary variance defines the performance variable in Figure 2D. Relative
salary is based on the difference between a manager's salary and the
salary expected of someone in his rank at her age: salary - E(sa1ary).Associations
with other background factors are negligible with rank and age
held constant (Burt, Hogarth, & Michaud 2000). Relative salary is standardized
across all 85managers in the study population to zero mean and
unit variance (a score of 1.5,for example, means that the manager's salary
is one and a half standard deviations higher than the salary typically paid
to people in his rank at his age).The negative association between relative
salary and network constraint is statistically significant (-3.7 t-test). The
managers who enjoy salaries higher than expected from their rank and age
tend to be managers with networks that span structural holes in the firm.
Figure 2E contains investment officersin a financial organization in 1993
(Burt 1997a).The study population includes bankers responsible for client
relations, but also includes a large number of administrative and support
people who participate in the bonus pool. Performance, background, and
network data on the study population are taken from company records.
Seventy-three percent of the variance in annual bonus compensation,
44 Structural Holes versus Network Closure as Social Capital
which varies from zero to millions of dollars, can be predicted from job
rank (dummyvariables distinguishing ranks defined by the organization),
and seniority with the firm (years with the firm, and years in current job).
Salary is almost completely predictable from the same variables (95%of
salary variance). With rank and seniority held constant, there are no significant
bonus differencesby officergender, race, or other background factors
on which the firm has data.The residual 27% of bonus variance defines
the performance variable in Figure 2E. Relative bonus is based on the difference
between the bonus an officer was paid and the bonus typical for
someone in his rank, at her age, with his years of seniority at the firm:
bonus - E(bonus).I standardized relative bonus across all officers in the
study population to zero mean and unit variance (so a score of 1.5,for example,
means that an officer's bonus is one and a half standard deviations
higher than the bonus typically paid to people at his rank or her rank, age,
and seniority).Figure 2E contains a random sample of 147men analyzed
for social capital (see Burt 2000:Table2, for results on female bankers).
The work of this population requires flexible cooperation between colleagues.
It is impossible to monitor their cooperation through bureaucratic
chains of command because much of their interpersonal behavior is uiiknown
to their immediate supervisor. The firm is typical of the industry in
using peer evaluations to monitor employee cooperation. Each year, officers
are asked to identify the people with whom they had substantial or
frequent business dealings during the year and to indicate how productive
it was to work with each person. The firm uses the average of these peer
evaluations in bonus and promotion deliberations. The firm does not look
beyond the average evaluations. However, there is a network structure in
the evaluations that, according to social capital theory, has implications for
an officerI s performance, which in turn should affect his bonus (seeEccles
& Crane 1988,Chapter 8). From peer evaluations by the investment officers
and colleagues in other divisions of the firm, I identified the people
cited as productive contacts by each of the officers, and looked at evaluations
by each contact to see how contacts evaluated one another. I then
computed network constraint from the network around each officer.
What makes the study population analytically valuable is the time order
between the network and performance data. Socialcapital theory gives
a causal role to socialstructure.Consistent with the argument, I assume the
primacy of socialstructure for theoretical and heuristic purposes. I am limited
to assuming the primacy of social structure because the data collected
in the other Figure 2 study populations are cross-sectional and so offer no
evidence of causation (see Burt 1992:173-80, for discussion). It is difficult
to gather survey network data, wait forthe relative success of managers to
emerge over time, and then gather performance data. The network data on
the investment officerswere obtained in the routine of gathering peer evaluations
to affect bonus compensation five months later.
There is a negative association in Figure 2E between bonus compensation
and network constraint (-3.7 t-test). The managers who received
bonuses higher than expected from their rank and seniority tend to have
networks that span structural holes in the firm. The logit results in Figure
2F show that the association is even stronger than implied by the results in
Figure 2E. Thereis a triangular pattern to the data in Figure 2E.On the right
side of the graph, officers with the most constrained networks receive low
bonuses. On the left, officers receiving larger bonuses than their peers tend
to have low-constraint networks, but many officers with equally unconstrained
networks receive small bonuses. I attribute this to annual data.
The low-constraint networks that span structural holes provide better access
to rewarding opportunities, but that is no guarantee of exceptional
gains every year.There is a .47partial correlation between bonus in the current
year and bonus in the previous year (after rank and seniority are held
constant). Even the most productive officers can see a lucrative year followed
by a year of routine business. So, the logit results in Figure 2F more
accurately describe the social-capital effect for the investment officers.I divided
the officers into three bonus categories: large (bonus more than a
standard deviation larger than expected from rank and seniority), medium,
and small (bonus more than a standard deviation smaller than expected
from rank and seniority).Network constraint this year significantly
decreases the probability of a large bonus next year (-2.7 t-test), but the
stronger effect is the increased probability of receiving a low bonus next
year (3.6t-test).
Other Evidence
Across the five study populations in Figure 2, social capital results from
brokerage across structural holes, not from network closure. Elsewhere, I
review research based on less detailed network data, but research on a
broader diversity of substantive questions on a broader diversity of study
populations (Burt 2000).The conclusion of the review is the same as here:
closed networks-more specifically, networks of densely interconnected
contacts-are systematically associated with substandard performance.
For individuals and groups, networks that span structural holes are associated
with creativity and innovation, positive evaluations, early promotion,
high compensation and profits.
RETHINKING COLEMAN’S EVIDENCE
The most authoritative evidence in Coleman’s argument for closure as a
form of social capital comes from his studies of high-school students. He
argues that closure explains why certain students are more likely to drop
46 Structural Holes versus Network Closure as Social Capital
out of high school. When the adults in a child’s life are more connected
with one another, the closure argument predicts trust, norms, and effective
sanctions more likely among the adults, which means that the adults can
more effectively enforce their interest in having the child complete his or
her education.
Coleman (1988,1990:590-97) offers three bits of evidence to show that
children living within closed networks of adults are less likely to drop out
of high school:First, children in familieswith two parents and few children
are less likely to drop out of high school (two parents living together can
collaborate more effectively in the supervision of a child than two parents
living apart). Second, children who have lived in the same neighborhood
all their lives are less likely to drop out of high school (parents, teachers,
and other people in the neighborhood are more likely to know one another
and collaborate in the supervision of a child than can parents new to the
neighborhood). Third, children in Catholic and other religious private
schools are less likely to drop out (parent, teachers, and parents of the
child’s friends at the private schools are more likely-relative to adults in
the same roles in a public school-to know one another and collaborate in
the supervision of a child).
Two questions: First, is ”not dropping out of school” a productive performance
criterion for estimating social capital effects?Performance variation
around ”drop out” is probably driven by factors different from those
that determine variation at the other end of the performance continuum,
the ”stay-in-school-and-do-well” end of the continuum. For example, analyzing
data oii mathematics achievement from the National Education
Longitudinal Study survey of 9,241 students in 898 high schools, Morgan
and Sarensen (1999a,b:674) raise questions about the value of network closure:
”In contrast to [Coleman’s]basic hypotheses, our findings lead us to
conclude that the benefits offeredby the typical network configurations of
horizon-expanding schools outweigh those of norm-enforcing schools.”
Like Coleman before them, Morgaii and Sarensen have limited network
data available for their analysis,l but their two network variables do measure
closure of a kind, so the negative association between math scores and
”parents know parents” raises questions for scholars committed to the closure
argument.
Second, the accumulating evidence of brokerage as social capital invites
speculation about the role that brokerage could be playing in Coleman’s
evidence. Grant that children are less likely to drop out of school if they
have a constrained network in which friends, teachers, and parents are all
strongly connected to one another so as to eliminate opportunities for the
child to play contacts against one another. Constraint from parents and
teachers has positive long-term consequences for children, forcing them to
focus on their education. But is this socialcapital of the child or its parents?
Rorinld zyxwvutsS.B1ii.t 47
The evidence reviewed in this chapter is about the socialcapital of the person
at the center of the network. The social capital associated with higher
performance by adults comes from a network of disconnected contacts.At
some point on the way to adulthood, the child shaped by the environment
takes responsibility for shaping the environment, and is rewarded in proportion
to the value he or she adds to the environment. Constraint, positive
for the child, is detrimental to adults, particularly adults charged with
managerial tasks at the top of their firm. Moreover, the parental network
around their child defines only part of the social-capital effect on educational
achievement. Thecomplete story is about effectiveadult supervision
(closureargument) combined with parental ability to wrestle resources out
of society to support the child (holeargument).Whatever the effect of closure
providing adult control over the child, how much greater is the effect
of a parent network that spans structural holes at work such that the parents
bring home earlier promotions and higher compensation as illustrated
in Figure 2?
A POINT OF INTEGRATION
There remains an important role for closure. It can be critical to realizing
the value buried in structural holes. zyxExtesiza1 aPId I ntesrzaI COnstsaiizt
Begin with the table in Figure 3. Rows distinguish groups in terms of their
external network. Groups can be distinguished on many criteria. I have in
mind the two network criteria that define information redundancy (cohesion
and structural equivalence), but it is just as well to have in mind a
more routine group; a family, a team, a neighborhood, or some broader
community such as an industry. Some groups are composed of individuals
with many nonredundant contacts beyond the group-as illustrated by
the three-person sociograms at the top of the table. People in each of the
two groups have a total of six nonredundant contacts beyond the group.
With respect to network measurement, nonredundant contacts mean a lack
of externalconstraint on the group.Thehorizontal axis in Figure 2B, for example,
measures the average network constraint on individuals in TQM
teams. Low-constraint teams, to the left in the graph, were composed of
employees with many nonredundant contacts beyond their team. In spanning
structural holes beyond the team, their networks reached a diverse
set of perspectives, skills, or resources. They were the high-performance
teams. At the other extreme, to the right in Figure ZB, low-performance
teams were composed of individuals with redundant contacts beyond the
48 Structural Holes versus Network Closure as Social Capital
Figure 3. Social capital matters.
team. The sociogram at the bottom of Figure 3 is an illustration. The
group’s four contacts beyond the team are interconnected, and so are redundantby
cohesion.Such a team has accessto a single set of perspectives,
skills, or resources, and is expected not to see or successfully implement
new solutions, as illustrated in Figure 2B by their poor performance with
respect to TQM.
Columns distinguish groups in terms of network closure. Structural
holes between people or organizatioiis in a group weakens in-group communication
and coordination, which weakens group ability to take advantage
of brokerage beyond the group. Closure eliminates structural
holes within the team, which improves communication and coordination
within the team. The sociogram to the left of the table in Figure 3 shows a
group with disconnected elements within it. The two sociograms to the
right of the table show groups with all three elements connected. Density
or hierarchy can provide network closure, though hierarchy seems to be
the more potent form of closure (Burt2000).A leader with strong relations
to all members of the team improves Communicationand coordination despite
coalitions or factions separated by holes within the team.
Perfownance Surface
The graph at the top of Figure 3 shows group performance across the cells
of the table. Performance here is an undefined mixture of innovation, positive
evaluation, early promotion, compensation, and profit. Points A, B, C,
and D at the corners of the table in Figure 3 correspond to the same points
in the graph.
Performance is highest at the back of the graph (quadrant A), where ingroup
closure is high (oneclear leader, or a dense network connecting people
in the group) and there are many nonredundant contacts beyond the
group (membernetworks into the surrounding organization are rich in disconnected
perspectives, skills,and resources).Performance is lowest at the
front of the graph (quadrant C), where in-group closure is low (members
spend their time bickering with one another about what to do and how to
proceed) and there are few nonredundant contacts beyond the group
(members are limited to similar perspectives, skills, and resources).
Figure 3 is my inference from three bits of evidence, all of which are
reviewed in detail elsewhere (Burt 2000:Figure 5). In fact, the Figure 3
interaction between brokerage and closure is the concept of structural
autonomy from which the hole argument emerged (Burt 1980,1982,1992:
The first evidential bit comes from research with census data describing
the association between industry profit margins and market structure. Industry
profit margins increase with closure among industry producers and
38-45).
50 Structural Holes versus Network Closure as Social Capital
increase with the number of nonredundant suppliers and customer markets
(Burt 1992,Chapter 3; 2000:Figure6).Analogy with the market structure
research is productive in two ways: The market results are based on a
census of market conditions, so they include data on the performance-network
association at extremes not present in most samples of managers.
Second, the market results across a broader range of network conditions
show a nonlinear form of returns to network structure. The strongest network
effects occur with deviations from minimum network constraint.
With respect to network structure within a group, in other words, performance
should be weakened more by the first significant disconnection in
the group than by one more disconnection within an already disorganized
group. With respect to external structure, performance should be weakened
more by the entry of one strong perspective, or skill, or resource in
the surrounding organization than it is by the entry of another external
pressure on a group already frozenby external pressures.
Asecond bit of evidencefor the integration is Reagans and Zuckerman’s
(1999)study of performance in 223 corporate R&D units within 29 major
American firms in eight industries. They report higher levels of output
from units in which scientists were drawn from widely separate employee
cohorts (implying that their networks reached diverse perspectives, skills,
and resources outside the team) nizd there is a dense communication network
within the unit. Tenure diversity (or other kinds of diversity, see
Williams & O’Reilly 1998)can be disruptive because of the difficulties associated
with communicating and coordinating across different perspectives,
but when communication is successful (as implied by a dense
communication network within the team), team performance is enhanced
by the brokerage advantages of the team having access to more diverse information.Reagans
and Zuckerman’s finding is a segment somewhere between
points A and C on the performance surface at the top of Figure 3.
A third bit of evidence for the integration comes from the contingent
value of social capital to managers (Burt 1997a,2000:Figure6). Social capital
is most valuable to managers who hold relatively unique jobs (such as
CEO, divisional vice-president, or people managing ventures of a kind
new to their organization). These are people who have the most to gain
from the information and controlbenefits of social capital.The contingency
argument is that numerous peers define a competitive frame of reference
against which any one manager’s performance can be calibrated, so managers
doing similar work come to resemble one another in their efforts.
Burt (1997a,2000:Figure6) shows a nonlinear decline in the value of social
capital in proportion to the number of managers-peers-doing the same
work. Assume that network closure among peers decreases with the number
of peers; network closure among many people being more difficult to
sustain than closure among a few people. Then the negative association
Roiiald zyxwvutsS.Bzrrt 51
between peers and the value of social capital is a negative association between
closure and the value of social capital. The social capital of brokerage
across structural holes is again more valuable to a group where there
is network closure within the group-point A at the back of the graph in
Figure 3.Along the axis from point C to D in the graph, low closure means
poor communication and coordination within a group and such a group
can be expected to perform poorly,benefiting from external networks only
in the richest diversity of perspectives, skills, and resources.
Frame of Reference for Integrating
Research Resirlts
Figure 3 can be a useful frame of reference for integrating research results
across studies. A study can show exclusive evidence of social capital from
network closure or structural holes without calling either argument into
question.
For example, Greif (1989)argues that network closure was critical to the
success of the medieval Maghribi traders in North Africa. Each trader rail
a localbusiness in his own city that depended on salesto distant cities.Network
closure among the traders allowed them to coordinate so as to trust
one another, and so profitably trade the products of their disparate business
activities. The traders individually had networks rich in brokerage opportunities,
but they needed closure with one another to take advantage of
the opportunities. More generally, in an environment rich in diverse perspectives,
skills, and resources, group performance depends on people
overcoming their differences to operate as a group. Group performance
will vary with in-group closure, not brokerage, because brokerage opportunities
beyond the group are abundant for everyone (this is the Figure 3
surface from point A to point D).
Rosenthal’s (1996) study of TQM teams illustrates the other extreme.
People on the teams had been trained to act as a team and there was enthusiasm
for quality management in the firm-so the teams did not differ
greatly in their closure. Closure was high in all of them. Therefore, team
performance varied as illustrated in Figure 28 with a team’s external network.
If a cohesive team can see a good idea, it can act on it. With all teams
cohesive, those with numerous nonredundant contacts beyond the team
had the advantage of access to a broader diversity of perspectives, skills,
and resources. Several recent studies report high performance from groups
with external networks that span structural holes (see Burt 2000 for review):Geletkanycz
and Hambrick (1997)on higher company performance
when top managers have boundary-spanning relationships beyond their
firm and beyond their industry; Ahuja (1998)on the higher patent output
of organizations that hold broker positions in the network of joint ventures
52 Structural Holes versus Network Closure as Social Capital
or alliances at the top of their industry; Pennings, Lee, and Witteloostuijn
(1998)on the survival of accounting firms as a function of strong partner
ties to client sectors; Stuart and Podolny (1999)on the higher probability
of innovation from semiconductor firms that establish alliances with firms
outside their own technological area; McEvily and Zaheer (1999) on the
greater access to competitive ideas enjoyed by small job manufacturers
with more nonredundant sources of advice beyond the firm; Srarensen
(1999)on the negative effect on firm growth of redundant networks beyond
the firm; Hansen, Podolny, and Pfeffer (2000) on computer newproduct
teams completing their task more quickly when the team is
composed of people with more nonredundant contacts beyond the team;
Baum, Calabrese, and Silverman (2000)on the faster revenue growth and
more patents granted to biotechnology companies that have multiple
kinds of alliance partners at start-up; Koput and Powell (2000) on the
higher earnings and survival chances of biotechnology firms with more
kinds of activities in alliances with more kinds of partner firms; and
Podolny (2000)on the higher probability of early-stage investments surviving
to IPO for venture-capital firms with joint-investment networks of
otherwise disconnected partners. With Figure 3 in mind, these studies tell
me not that the closure argument is in error so much as that closure within
business groups is less often problematic than brokerage beyond the
group. More generally, the relative performance of cohesive groups will
vary with the extent to which a group is composed of people with networks
rich in structural holes, not network closure, because closure is high
for all of the groups (thisis the Figure 3 surface from point A to point B, illustrated
in Figure 2B).
In short, structural holes and network closure can be brought together
in a productive way. The integration is only with respect to empirical evidence.
The mechanisms remain distinct. Closure describes how dense or
hierarchical networks lower the risk associated with transaction and trust,
which can be associated with performance. The hole argument describes
how structural holes are opportunities to add value with brokerage across
the holes, which is associated with performance. The empirical evidence
reviewed supports the hole argument over closure. However, my summary
conclusion illustrated in Figure 3is that while brokerage acrossstructural
holes is the source of added value, closure can be critical to realizing
the value buried in the structural holes.
NOTE
1. For example, the ”density of student friendshipnetworks” to which they refer
in their conclusion is not a network density measure; it is a count of a
student’s closest friends named in an interview with the student’sparent (0
to 5, ”friends in school” variable in Morgan and Smrensen, 1999a:666-67).
”Friends in school” is an indicator of intergenerational network closure,
and, consistent with the closure argument, has a positive association with a
student’sgain in math scores to 12thgrade (primarily for studentsaveraged
across schools: Morgan and Serreiisen, 1999x669, 1999b:698; Carbonaro
1999:684-85). The ”density of parental networks” in Morgaii and Slzrrensen’s
conclusion is also a count. It is the number of the named close friends
for whom the interviewed parent claims to know one or both of the friend’s
parents (”parents know parents” variable). ”Parents know parents” is another
measure of intergenerational network closure, but in contradiction to
the closure argument, has a negative association with a student’s gain in
math scores (again primarily for students averaged across schools, Morgan
and Serrensen,1999a:669,1999b:698).Inferences are complicated by the fact
that ”friendsin school” is of course strongly correlated (.58)with ”parents
know parents.” More consequential, Morgan and Smrensen’s network variables
are enumerations by the parent, not the student.The student need not
agree with the parent’s selection of best friends, and the student’s network
can extend well beyond the view of his or her parents (recall that these are
high school students; see Hirschi 1972on the significance for delinquent behavior
of a boy’s friends unknown to his father).
REFERENCES
Ahuja, Gautam. 1998.Collaboration networks, structural holes, and innovation: a
longitudinal study. Paper presented at the annual meetings of the Academy of
Management.
Aldrich, Howard E. 1999. Orgaiiizatiolzs Evoluirig. Thousand Oaks, CA: Sage.
Aiicona, Deborah G., and David F. Caldwell. 1992a.Demography and design: Predictors
of new product team performance. Ol.g~7iZ12fltiolZSciciicc 3:321-41
. 1992b.Bridging the boundary: External activity and performance in organizational
teams. Advzirzistmtive Sczeizcc Quarter/!/ 37:634-65.
Baker, Wayne E. 1984.The social structure of a national securities market.Airiericaiz
Joziriml of Sociology 89:775-811.
Baker, Wayne E., and Ananth Iyer 1992.Information networks and market beliavior.Joiirtznl
ofMatlieniaticn1 Sociology 16:305-32.
Baum, Joel A. C., Tony Calabrese, and Brian S. Silverman. 2000. ”Don’t go it alone:
Alliance network composition and startups’ performance in Canadian biotechnology.”
Strategic Mamgcincnt Joiirnal21:267-94.
Biggart, Nicole Woolsey. 2000. Banking on each other: The situational logic of rotating
savings and credit associations. Paper presented at the 2000 Organization
ScienceWinter Conference.
Bourdieu, Piere. 1980.Le capital social: Notes provisoires. Actcs de 10 Rechcrclzc cii
Scieiiccs Socinles 3:2-3.
Bourdieu, Pierre, and Loi’cJ. D. Wacquant. 1992.Ari Iizzlitatioiz to R$ksiue Sociology.
Chicago, IL: University of Chicago Press.
54 Structural Holes versus Network Closure as Social Capital
Brieger, Ronald L. 1995. Socioeconomic achievement and the phenomenology of
Burt, Ronald zyxwvuS.1980. Autonomy in a social topology. Aiiiuicnii ]oi/rizal zyof Sociolocyy
achievement. Aizizziol RCZJ~CTUqf Sociology21:115-36.
85:892-925.
. 1982. Em~nidn Sfriicfiirol Tlirory $Actio/i. New York: Academic Press.
. 1992. Strirctiiral Holcs. Cambridge, MA: Harvard University Press.
. 1995.Le capital social, les trous structuraux, et l’entrepreneurtranslated by
.1997a. The contingent value of social capjtal. Adii/i~isfrnfi-i~(’Scicr7cc Q!/lir.
159%. A note on social capital and network content. Soci[7/ Nctcc~orks
. 1998.The gender of social capital. Rnfioi7alifyni7d Sncicfy 10:5-46.
. 1999a.Entrepreneurs,distrust, and third parties.Pp. 213-243 in Slinrcd Cos- zjritiori iii Orgnriiznfioiis, edited by Leigh L. Thompson, John M. Levine, and
David M. Messick. Hillsdale, NJ:Lawrence Erlbaum.
Emmanuel Lazega.. R(wicF r n i ~ p i i s ~dr Sociologi~~36599-628.
f c d y 42339-65.
191355-73.
. 1999b. The social capital of opinion leaders. Ai7iinls 566:37-54.
. 2000.The network structure of social capital. Pp. 345-423 in Rcse~7rcliiii Orgniriznfioiinl
BLdinzlioi; edited by Robert I. Sutton and Barry M. Staw. Greenwich,
C T JAI Press.
. 2001. Bandwidth and echo: Trust, information, and gossip in social networks.
In Networks n i i d Mnrlivfs, edited by Alessandra Casella and James E.
Iiaucli. New York: Russell Sage Foundation.
Burt, Ronald S.,Joseph E. Jannotta, and James T. Malioney. 1998. Personality correlates
of structural holes. Social Nafii~orks20:63-87.
Burt, Ronald S., Robin M. Hogarth, and Claude Michaud. 2000. The social capital
of French and American managers. Orpiiiznfioii Scio7ce 11:123-47.
Carbonaro, William J. 1999. ”Openiiing the debate on closure and schooling outcomes.”
Aiiici-icn717 Sociologicnlficz~i~7w64:682- 86.
Colenian, JamesS. 1988. Social capital in the creation of human capital. A1ii~ricr711
]oi/riml zyxwvutsqf Sociology 94:S95-S120.
. 1990. Foiiiidnfioii~qf Sosiril Tlico/y. Cambridge, MA: Harvard Uni\Tersity
Press.
Cook, Karen S., and Richard M. Emerson. 1978.Power, equity and commitment in
exchange networks. Aiiwricniz Sociologicril Rwiuu 43:712-39.
Eccles, Robert G., and Dwight B. Crane. 1988. Doiiig D P ~ ~ / S .Boston, MA: Harvard
Business School Press.
Freeman, Linton C. 1977.A set of measures of centrality based on betweenness. So-
cio/iicfr!/403-40.
Geletkanycz,Marta A., and Donald C. Hambrick. 1997.The external ties of top executives:implications
for strategicchoice and performance. Adriiiizistn7tiz~c.Scicizcc
Qiinrfuly42654-81.
Granovetter, Mark S.1973. The strength of w7eak ties. A~i~~ricriii]aiirrinl cf Sociology
. 1985.Economic action, social structure, and embeddedness.A/iicricniiloiir.1992.Problems
of explanation in economicsociology.Pp. 29-56 in Ncfc~orks
78:1360-80.
/ i n / of Sociology 91:481-510.
Rorznld zyxwvutS.Bzrrt 55
c1iiCIzyxwvutsrOrgarzizotioiz,edited by Nitin Nohria and Robert G. Eccles. Boston, MA:
Harvard Business School Press.
Greif, Avner. 1989. Reputation and coalition in medieval trade: evidence on the
Maghribi traders. ]ourm1of Ee0170ri1ieHistory 49:857-82.
Haiisen, Morten T., Joel M. Podolny, and Jeffrey Pfeffer.2000. So many ties, so
little time: a task contingency perspective on the value of social capital in organizations.
Paper presented at the 2000 Organization Science Winter Confer-
ence.
Hirsclii, Travis. 1972. Cnirsrs of DcIirzqiic/?cy.Berkeley, CA: University of California
Press.
Koput, Keimeth, and Walter W. Powell. 2000. Not your stepping stone: collaboration
and the dynamics of industry evolution in biotechnology. Paper presented
at the 2000 Organization Science Winter Conference.
McEvily, Bill, and Akbar Zalieer. 1999.Bridging ties: a source of firm heterogeneity
in competitive capabilities.SfrotegicMorzagc~rirozf]oi/riiol 20:1133-56.
Merton, Robert K. (1957) 1968. Continuities in the theory of reference group beliavior.Pp.
335-440 in Social Tlzcory zyxwmid Socir7I Sfri/cfiir.r.New York: Free Press.
. 1984.Socially expected durations: a case study of concept formation in sociology.
Pp. 262-83 in Cnuflict o/id Co/iscrzsiisedited by Walter W. Powell and
Richard Robbins. New York: Free Press.
Morgan, Stephen L. and Aage B. Smeiiseii. 1999a. “A test of Coleman’s social capital
explanation of school effects.” Airrcl-izn/zSociologici71 R w i m I 64:661-81.
. 1999b. ”Theory, measurement, and specification issues in models of iietwork
effects on learning.”Airi~ricaiiSocioloxicol IICU~CI(J64594-700.
Morselli, Carlo. 2000. ”Structuring Mr. Nice: entrepreneurial opportunities and
brokerage positioning in the cannabis trade.” Criir~c’,LOWnrzd Socinl Cl11711p7 33:
In Press.
Padgett, John F., and Cliristopher K.Ansell. 1993.Robust action and the rise of the
Medici, 1400-1434. Ariicricaii laiirml qf Sociology 98:1259-1319.
Pennings, JohannesM., Kyungniook Lee, and Arjen van Witteloostuijn. 1998.Human
capital, social capital, and firm dissolutioii.Acodc/iIy c,fMaizr7Sc~iircrifloiirPodolny,
Joel M. 2000. ”Networks as the pipes a i d prisms of the market.” Graduate
Scliool of Business, Stanford university.
Putiiam, Robert D. 1993.Mokiiig Dcriiocrncy Work.Princeton, NJ:Princeton University
Press.
Reagans, Ray, and Ezra W. Zuckerman. 1999. Networks, diversity, and performalice:
the social capital of corporate R&D units. Graduate School of Industrial
Administration, Carnegie Melloii University.
Rosenthal, Elizabeth A. 1996.Socinl Nctx7c7rIcs orid Tcnrrr P~i:foror-iirnircc.P1i.D. Dissertation,
Graduate School of Business, University of Chicago.
Simmel, Georg. [1922]1955.Corzflict m i d tlze W d of Gl-oiipAffiliotioiis (translated by
Kurt H.Wolff and Reinhard Bendix). New7 York: Free Press.
Smrensen,JesperB. 1999.Executive migration and interorgaiiizationalcompetition.
Sociol Scieiicr Resmrcli 28:289-315.
Stuart,Toby E., and JoelM. Podolny. 1999.Positionalcausesaiid correlatesof strategic
alliances in the seiniconductor industry. Pp. 161-82 in Rcscarclr i/i thc Soci-
/~al41:425-40.
56 Structural Holes versus Network Closure as Social Capital
ology zyxwvut(3O Y ~ O ~ Z ~ Z ~ ~ ~ O J Z S ,edited by Steven Andrews and David Knoke. Greenwich,
C T JAlPress.
Thornton, Patricia H. 1999.The sociology of entrepreneurship.Aiiiiual Res~iewof Sociolog!y
25:19-46.
Tillman, Robert, and Michael Indergaard. 1999.Field of schemes:health insurance
fraud in the small business sector. Social Problciiis 46:572-90.
Tullock, Gordon. 1985.Adam Smith and the prisoners’ dilemma. Qiinrterl~/]oiiuial
~ ? fEcoi zoiizics 100:1073-81.
Williams, Katherine Y., and Charles A. O’Reilly 111. 1998.Demography and dixTersity
in organizations: a review of 40 years of research. Pp. 77-140 in Rcscarcli ziiz 0 1 ; 9 a ~ i i ~ t i ( i ~ i n lRelznvior; edited by Barry M. Staw and L. L. Cummings.
Greenwich, CT JAI Press.
Williains, Phil. 1998. The nature of drug-trafficking networks. C ~ / I W J I ~History
971154-59.