Emerging Trends in The Social and Behavioral Sciences

How Do Labor Market Networks Work?

Item

Title

How Do Labor Market Networks Work?

Author

Rubineau, Brian

Fernandez, Roberto M.

Research Area

Social Institutions

Topic

Markets

Abstract

The informal seeking and sharing of job opportunity information via contacts are the dominant mechanisms for both the supply and demand sides of the labor market. Despite many decades of scholarly scrutiny, we have established little certainty about the mechanisms through which labor market networks operate. Much of this uncertainty results from single‐perspective investigations of a fundamentally triadic process. Network‐mediated job search is not merely a version of the classic two‐way matching problem with some additional network factors but is rather a three‐way matching problem with three distinct agentic decision makers: the job seeker, the job screener, and the social contact acting as a connector. This essay summarizes what is currently known about the operation and consequences of labor market networks, their mechanisms, and their contextual dependencies. We show how the perspective of a triad of actors presents new opportunities for resolving current contradictory empirical findings and areas of ongoing debate. Progress on this topic requires both careful causal research isolating mechanisms affecting a particular actor and integrative research on how these mechanisms interact among the triad of actors.

Related Essays

Global Economic Networks (Sociology), Nina Bandelj et al.

Problems Attract Problems: A Network Perspective on Mental Disorders (Psychology), Angálique Cramer and Denny Borsboom

Migrant Networks (Sociology), Filiz Garip and Asad L. Asad

Interdependence, Development, and Interstate Conflict (Political Science), Erik Gartzke

The Future of Employment, Wages, and Technological Change (Economics), Michael J. Handel

The Development of Social Trust (Psychology), Vikram K. Jaswal and Marissa B. Drell

Herd Behavior (Psychology), Tatsuya Kameda and Reid Hastie

Emerging Trends in Social Network Analysis of Terrorism and Counterterrorism (Sociology), David Knoke

How Networks Form: Homophily, Opportunity, and Balance (Sociology), Kevin Lewis

Network Research Experiments (Methods), Allen L. Linton and Betsy Sinclair

Culture, Diffusion, and Networks in Social Animals (Anthropology), Janet Mann and Lisa Singh

Gender and Women's Influence in Public Settings (Political Science), Tali Mendelberg et al.

The Role of School-Related Peers and Social Networks in Human Development (Psychology), Chandra Muller

Social Relationships and Health in Older Adulthood (Psychology), Theodore F. Robles and Josephine A. Menkin

Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth Wilson

Identifier

etrds0168

extracted text

How Do Labor Market Networks
Work?
BRIAN RUBINEAU and ROBERTO M. FERNANDEZ

Abstract
The informal seeking and sharing of job opportunity information via contacts are
the dominant mechanisms for both the supply and demand sides of the labor market. Despite many decades of scholarly scrutiny, we have established little certainty
about the mechanisms through which labor market networks operate. Much of this
uncertainty results from single-perspective investigations of a fundamentally triadic
process. Network-mediated job search is not merely a version of the classic two-way
matching problem with some additional network factors but is rather a three-way
matching problem with three distinct agentic decision makers: the job seeker, the job
screener, and the social contact acting as a connector. This essay summarizes what is
currently known about the operation and consequences of labor market networks,
their mechanisms, and their contextual dependencies. We show how the perspective
of a triad of actors presents new opportunities for resolving current contradictory
empirical findings and areas of ongoing debate. Progress on this topic requires both
careful causal research isolating mechanisms affecting a particular actor and integrative research on how these mechanisms interact among the triad of actors.

The informal seeking and sharing of job opportunity information via contacts
are the dominant mechanisms for both the supply and demand sides of the
labor market. The importance and prevalence of these processes have been
well documented. Previous work has mostly focused on documenting the
importance of labor market networks via their consequences. Surprisingly,
few of these studies have clearly identified the mechanisms underlying labor
market networks’ operation or their contextual contingencies, and even
fewer studies have done so with any degree of causal certainty. Causally
identifying which mechanisms drive labor market networks and under what
conditions they are active is vital to effectively manage or leverage these
processes toward organizational or personal goals. For example, does the use
of informal contacts in job search improve the likelihood of receiving a job
offer? Do firms prefer referred applicants? Do referring employees generate
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

candidates that differ from the applicants generated by formal methods?
What are the scope conditions for any such effects? There are a host of
basic questions about how labor market networks work that still await
conclusive answers. This essay summarizes what is currently known
about the operation and consequences of labor market networks, their
mechanisms, and their contextual dependencies. Rather than seeking to
provide a comprehensive review (for recent reviews, see Castilla, Lan, &
Rissing, 2013a, 2013b; Topa, 2011), this essay focuses on identifying the open
questions that need to be answered in future research. Thus, the research
cited and discussed here are drawn disproportionately from very recent
scholarship providing insights into the causal processes involving labor
market networks. To preview our conclusion, we suggest that a substantial
challenge to past research in this area has been the multifaceted nature
of the phenomenon. Researchers tend to limit their attention to a narrow
subset of the job matching process, necessarily neglecting the effects of
other components. Network-mediated job search is not merely a two-way
matching problem with some additional network factors but is a three-way
matching problem with three distinct agentic decision makers. Progress in
this important field will depend on future research achieving two important
goals. Research must first be designed to isolate these various features of
network processes affecting each actor at the labor market interface. Second,
the findings from this research must be integrated to reveal the collective
operation of the three-actor system defining labor market networks.
THE THREE PLAYERS AND THEIR ROLES
The use of informal contacts in the job matching process requires a convergence in the actions of three players: the connector must provide job opportunity information to the job seeker, the job seeker must act on that information,
and the job screener must decide whether to extend a job offer to the job
seeker.1 These three players form a triad that is fundamental to the existence
and operation of labor market networks. Job seekers work on the supply side,
while job screeners work on the demand side of the labor market. Connectors may work on the supply side, the demand side, or as an intermediary
between the two sides. Connectors who use their contacts to seek out job
opportunities on behalf of a job seeker arguably operate on the supply side
of the labor market. Connectors who are firm employees recruiting for their
employer among their contacts arguably operate on the demand side. Some
1. Screening in organizations is often divided between initial evaluators (often performed by personnel in HR departments) and the hiring manager who makes the ultimate hiring decision. This arrangement
introduces the possibility of actor disagreement (Fernandez, 2010). For simplicity, we set aside issues of
possible misalignment.

How Do Labor Market Networks Work?

3

connectors may act simply as information conduits, not working on anyone’s
behalf. Scholarly attention has not been equally divided across these three
players. Studies examining the importance of networks for job seekers are
numerous and span over a half-century, studies adopting the hiring firm’s
perspective have emerged over the past two decades, and studies explicitly
examining the role of the connector are the most recent of all. In the next three
sections, we summarize and synthesize the research findings on labor market
networks from each of these perspectives.
JOB SEEKERS
How do labor market networks work for job seekers? What are their effects,
mechanisms, and contingencies? Does a job search including networks yield
distinct outcomes compared to a job search excluding networks? Although
there is broad scholarly consensus that network search is consequential for
job seekers, there is disagreement regarding some of those consequences.
Virtually all studies examining job search duration and receiving or accepting a job offer find that job seekers who use networks have shorter job
searches (e.g., Cingano & Rosolia, 2012) and are more likely to receive or
accept a job offer identified via networks (Fernandez & Greenberg, 2013;
Obukhova & Lan, 2013). Tests of the effects of network search on wages show
contradictory results. Some models suggest wage benefits to network job
search (Montgomery, 1991), and a number of empirical studies buttress that
finding (e.g., Hensvik & Nordström Skans, 2013). Other models suggest the
opposite (Krug & Rebien, 2012), and those models find empirical support
as well (Bentolila, Michelacci, & Suarez, 2010). Scholars have proposed
industry-level (Kugler, 2003) or nation-level (Pellizzari, 2010) factors as
contingencies to help resolve such contradictory findings. One of the earliest
contingencies is identified by Granovetter (1973), that is, whether the nature
of the job seekers’ relationships with their contacts is consequential. Indeed,
weaker ties tend to provide novel job opportunity information, and this
novel information mediates the relationship between tie strength and the
likelihood of getting a job via contacts (Yakubovich, 2005). More recently,
examining within-individual variation from data on the multiple job
searches of graduating MBAs Barbulescu (2014) found that while weak ties
were more helpful for learning about job opportunities and getting invited
for interviews, stronger ties were more helpful for converting interviews into
offers (see note 1). Another recently described contingency is the difference
between having contacts while engaging in a job search versus actively using
those contacts in a job search. Many studies posited that having better social
capital (measured in terms of social network opportunities for job search
help) would be a reasonable proxy for mobilizing that capital in job searches.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

But just as the marginal dollar remaining in a bank contributes little to
marginal productivity, unmobilized social capital has little opportunity to
assist a job seeker. Some studies finding positive social spillover effects
consistent with a social capital effect inferred such an effect explicitly (e.g.,
Cingano & Rosolia, 2012) even though no network mobilization data were
available. Other studies found social capital measures not to be associated
with job search outcomes (Mouw, 2003). Helping to resolve this puzzle using
data on multiple contemporaneous job searches, Obukhova and Lan (2013)
found that social capital did not predict contact use, and only contact use
was consequential in improving the chances of interviews and job offers.
Many other contradictory empirical findings still have not been resolved.
For example, does using networks in job search result in jobs that are better
matches with the interests and abilities of the job seeker (Franzen & Hangartner, 2006) or worse matches (Bentolila et al., 2010)? Similar debates continue
about the job satisfaction, performance, and turnover effects of network job
search. Even in areas of empirical agreement, mechanism questions remain.
For example, why does network job search result in shorter job search
durations? Using the framework recently proposed by Castilla et al. (2013a),
contacts may provide resources to the job seeker (e.g., job information or
advocacy), or signals about otherwise difficult-to-observe candidate traits
to the hiring firm (e.g., productivity or similarity to current employees).
Another possibility is that referrals change the temporal dynamics of the
candidate evaluation process. For example, when confronted with a pool of
applicants, employers may tend to respond to referrals first. This behavior
alone could generate many of the apparent beneficial effects of network job
search even absent any employer preferences for network applicants.
JOB SCREENERS: HIRING FIRMS
While the demand side of labor market networks has been understudied
relative to the supply side, a host of studies find that employers prefer referral applicants (e.g., Fernandez, Castilla, & Moore, 2000; Petersen, Saporta, &
Seidel, 2000). What are the mechanisms behind these preferences, and what
is the quality of the evidence supporting those mechanisms?
One question is whether job screeners have a simple preference for referral
applicants versus nonreferral applicants. In a within-individual study leveraging data from a firm where individuals had applied for jobs multiple times
sometimes as referral applicants and sometimes as nonreferral applicants,
Fernandez and Galperin (2014) find a clear preference for referral applicants.
Individuals first applying as nonreferral applicants and then later as referrals were more likely to be interviewed and offered jobs than individuals

How Do Labor Market Networks Work?

5

who remained nonreferral applicants in subsequent job applications. In contrast, individuals who first applied as a referral applicant and then later as
a nonreferral experienced significant decreases in their chances of interview
and job offer.
Why might such a preference manifest? Consistent with Castilla et al.’s
(2013a) framework, there are several signaling-related hypotheses for the
observed preferences. Several empirical studies find referral applicants are
more productive workers (Burks, Cowgill, Hoffman, & Housman, 2013;
Castilla, 2005; Pinkston, 2012), suggesting referral status is a quality signal.
Another common signaling explanation builds upon homophily—the tendency for individuals’ social contacts to be similar to themselves along many
dimensions. Because of homophily, current workers’ referral applicants will
be similar to their referrers—people whom the firm already has chosen to
employ. Many scholars have suggested that homophily on characteristics
important to the organization is a likely reason firms prefer referral applicants (Casella & Hanaki, 2008; Fernandez & Galperin, 2014; Hensvik &
Nordström Skans, 2013). However, field evidence suggests that the belief
that referrals provide benefit via homphily is contingent as some employers
avoid referrals because they see it as leading to cliques that are hard to control
(Bewley, 1999; Rees, Shultz, Hamilton, & Taylor, 1970). For these employers,
the signal value of referring would be negative. Other signaling explanations
do not rely on homophily. On the basis of results from their experimentally
constructed labor markets using students, Gërxhani, Brandts, and Schram
(2013) argue that the fact that another person has referred an applicant signals that the referral occupies a position in an informal information network
characterized by higher levels of trustworthiness, and these trustworthiness
effects would lead employers to prefer referrals. Lin, Zhang, Chen, Ao, and
Song (2009) suggest that referral applicants signal a higher level of social
capital than nonreferral applicants, arguing that this signal of social capital
is likely to be correlated with the social skills that are necessary for certain
types of jobs.
There are also numerous explanations for employers’ preferences for referral applicants that do not involve signalling. Several studies identify performance and productivity benefits for the referrer when their referral is hired
(Burks et al., 2013; Yakubovich & Lup, 2006). Thus, even absent signals from
referrers about referral quality, hiring referral applicants may boost productivity among current employees. In addition, recruiting via referrals is often
less costly (Fernandez & Castilla, 2001), and it yields workers with lower
turnover rates (Burks et al., 2013; Castilla, 2005; Neckerman & Fernandez,
2003). Further, referral applicants are more likely than nonreferral applicants
to become referrers (Fernandez & Fernandez-Mateo, 2006; Fernandez & Sosa,
2005). Preferring referral applicants may reflect an interest in encouraging

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

and investing in that more cost-effective mode of recruitment. Beyond the
anticipation of future savings from referral hires, some scholars posit that the
social ties with current employees entailed by the referring relationship confer additional cost, commitment, monitoring, or even performance benefits
that are not available from nonreferral applicants (Burks et al., 2013; Fernandez et al., 2000; Sterling, 2014).
One notable recent experimental study provides strong empirical support
for both signaling and nonsignaling explanations. Pallais and Sands (2014)
conducted three experiments using an online labor market where subjects
were paid employees—referrers, referrals, and nonreferrals—of three firms
created by the research team. They found strong evidence of homophily:
referral hires’ performance was strongly associated with the performance of
their referrers. They found strong evidence of quality signaling: controlling
for all characteristics observable at application and hire, referral hires outperformed nonreferral hires not only at their initial jobs but also in later work for
a second experimental firm. Finally, they also found strong referrer–referral
relationship effects: when working on team tasks, teams pairing referrers and
referrals outperformed teams with other pairing even when controlling for
all individual characteristics.
The two types of signaling mechanisms described suggest contingencies
as to when referral applicants may or may not be preferred. First, given that
being referred can provide important worker quality signals to the employer,
then in settings where worker quality is completely observable, then referral
applicants would have no signaling advantages over nonreferral applicants
[see Chua’s (2011) study of the state-sector labor market of Singapore].
Second, given the strength and consistency of homophily effects, it would be
understandable if the quality of the referrer affects the employer’s preference
for referral applicants. In their study of an electronically mediated contract
labor market, Yakubovich and Lup (2006) find precisely this outcome.
While referred applicants from higher performing employees received
significantly more favorable outcomes than nonreferral applicants, referral
applicants from lower performing employees received significantly less
favorable outcomes than nonreferral applicants. This homophily contingency, however, has more to do with the referrer than the job screener. In
the discussion of their experiment, Pallais and Sands (2014) describe the
mechanism yielding the enduring work performance premium from referral
hires versus nonreferral hires as “selection.” The actor doing the selecting,
the agent for this mechanism’s operation, is the referrer. That is, the referrer
is aware of otherwise unavailable quality and productivity information—not
wholly explained by homophily—and chooses to refer in part based on that
information. In this way, the referrer is a prescreener acting in a manner that
is likely to benefit the hiring firm. This filtering behavior is not a behavior

How Do Labor Market Networks Work?

7

or choice of the hiring firm, and prompts us to turn our attention to the
least-studied actor in the triad of labor market network actors—the person
connecting the job seeker to the hiring firm—the connector.
JOB CONNECTORS
In addition to the job seeker and job screener, a third role is played by actors
providing the intermediary connection between the supply and demand
sides of the labor market. While our focus here is on network accounts in
which individual actors play the role of network connectors, this function
can also be served by organizations working as labor market brokers
(Fernandez-Mateo, 2007). Connectors are defined by what they do: sharing
job opportunity information with job seekers within their networks. It is the
connector’s sharing of job information that turns a job seeker into a network
job seeker. How do labor market networks work for these job connectors?
Specifically with respect to employee referrals, there is evidence that
worker ability is related to the likelihood of engaging in referring. Hensvik
and Nordström Skans (2013) find high-ability/high-aptitude workers to be
overrepresented among employee referrers. Moreover, these connectors are
not passive conduits for job opportunity information (Marin, 2012; Smith,
2005), but are rather agentic decision makers whose behaviors and choices
directly influence labor market dynamics. In general, referrers see the
performance and success of their referrals as a reflection upon themselves
in the eyes of their employer; such reputation effects are likely to be much
weaker for nonemployee referrals. Smith (2005, 2010) examines the referring
behaviors of racial/ethnic minorities in high-poverty urban areas and finds
employed workers with job opportunity information are quite concerned
with the reputational impacts of their referrals, and thus are highly selective
in deciding with whom among their personal contacts they will share the
information. Recent research showing employers engage in shared wage
punishment of referrers and referrals (Heath, 2013) suggests that such reputational concerns are justified. In Beaman and Magruder’s (2012) experiment
in the Kolkata labor market for experiment subjects, subjects generated
other subjects via referring and under a variety of manipulated scenarios.
They found that referrers are aware of their contacts’ likely productivity,
and will select referral applicants on this worker quality dimension only if
the incentives are constructed such that the referral’s performance affects
the referrer directly. When referral hires’ behaviors have no impact on the
referrers’ work outcomes, the referrers are likely to refer close friends and
family without selecting on worker quality. To the extent that connectors
selectively share job opportunity information, they also explicitly filter their
contacts to select on additional productivity-associated characteristics that

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

may not be otherwise apparent to the employer. Thus, through the act of
sharing information, connectors implicitly provide some of the signaling
benefits valued by employers (Fernandez et al., 2000; Pallais & Sands, 2014).
This raises the intriguing possibility that firms can affect their labor market
outcomes by influencing the behavior of their referring employees.
We have previously suggested that referrers are the missing link to a more
complete understanding of labor market network dynamics (Rubineau &
Fernandez, 2013). Firms commonly offer referral bonuses as an attempt to
influence referring behavior, and as shown in Beaman and Magruder’s (2012)
Kolkata experiment, referrer behavior is sensitive to different referring incentives. The more recent set of studies examining referrer behaviors and outcomes consistently find that after their referral applicant is hired, referrers
are less likely to leave the firm (Burks et al., 2013). In addition, the performance effects of referral hires on their referrers remain strong (Burks et al.,
2013; Castilla, 2005).
Looking forward, numerous questions remain about how connectors
influence job matching. Past research suggests that the presence and quality
of additional assistance depends importantly on the nature of the relationship the referrer has with the job seeker. Unlike Granovetter’s Strength
of Weak Ties for job seekers, referrers appear to provide more assistance
to those job seekers with whom they are more strongly tied (Marin, 2012),
and having a strongly tied referrer was more strongly associated with
getting job offers. Other questions remain as to the role of the connector.
The research discussed suggests that providing incentives to refer leads the
connector to actively seek out candidates. But does this lead the connector
to draw on their existing stock of contacts—likely strong ties—or is this
accomplished by seeking out new contacts that are likely to be weaker ties?
Ongoing research being conducted by Fernandez uses survey vignettes
in an experimental set up to address this question. Other research by
Lin et al. (2009) suggests that the connector’s key point of influence is to
be found in relationships with screeners and hiring managers. Finally, a
promising direction for future research on connectors is to look at the effect
of referring on workplace composition (Trimble & Kmec, 2011). Fernandez
and colleagues (Fernandez & Fernandez-Mateo, 2006; Fernandez & Sosa,
2005) and other researchers (e.g., Beaman et al., 2013) find strong evidence
of demographic homophily between referrers and their referrals. While
past understandings of homophily suggest that this is likely to reinforce
workplace gender or racial segregation, more recent research by Rubineau
and Fernandez (2014) identify the conditions under which recruitment via
employee referrals desegregates. Moreover, they argue that firm policies can
create these conditions by managing referrer behaviors. To the extent that
the firm can incentive the underrepresented group to refer more than the

How Do Labor Market Networks Work?

9

overrepresented group, referring can increase the rate at which referring
desegregates. Recent empirical work supports this view. In their experimental study of referring in Malawi, Beaman, Keleher, and Magruder (2013) find
job segregation to be more sensitive to referring rates than to homophily.
CONVERGENCE AMONG THE PLAYERS
Earlier, we have focused upon each of the actors within the labor market network triad. In reality, all three of these actors work together interdependently
at the labor market interface. Focusing on a single player and setting aside the
effects of the other two players necessarily obscures many dynamics. Some
of the puzzles regarding findings from the perspective of a single actor can be
readily resolved by considering the dynamic interactions with other actors.
Given three actors, there are three pairs of actors that could be considered
together, and one full triad. The job seeker and the job screener pair is the
traditional focus of labor market research. Explicit consideration of the interactions with an agentic connector is needed to move our understanding of
labor market networks forward.
CONNECTORS AND SEEKERS
One area of debate is the absence of a consistent effect of individual social
capital measures on job search outcomes (Mouw, 2003). The using versus
having solution (Obukhova & Lan, 2013) likely represents only part of the
explanation. After all, it is reasonable to think that having more social capital makes it easier to use social capital. The other part of the explanation
may be in recognizing that the agency of the connector plays a key role in
creating a network job search. Connectors may not only be selective among
their job-seeking contacts in sharing job opportunity information, but may
actually induce contacts who are not active job seekers to apply for a particular position (Kmec, McDonald, & Trimble, 2010). Research also shows that
individuals referred to a job opportunity by a social contact feel some social
obligation to apply for the job (Sterling, 2014).
CONNECTORS AND SCREENERS
Just as connectors may create seekers, screeners—in the form of firm
policies—may create connectors. That is, the firm has some control over
whether and which employees learn about a job opportunity within the
firm. Identifying firm policies that could encourage particular employees to
engage in referring would create an important and powerful tool. Managing
referring behavior could have benefits in addition to the desegregating

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

effects discussed earlier. Upon engaging in referring, the referrer demonstrates higher performance and lower turnover. Although the speculative
explanation for this effect focuses on the social tie with the referral hire,
this effect could also be understood in terms of the referrer’s relationship
with the firm. In referring, the referrer is acting as a volunteer recruiter. The
referrer uses her own time to talk-up the firm to her contacts. Field evidence
reported in Fernandez and Galperin (2014) supports the idea that screeners
recognize this effort and are more likely to grant interviews to referrals out of
courtesy to the referring employee. But there might be consequences for the
referring employee as well. Festinger’s theory of cognitive dissonance (1957)
would suggest that taking this action would likely improve the referrer’s
view of and feelings toward the firm. This increased affective commitment to
the firm could contribute to the observed performance and turnover effects
of referring. To the extent that a dissonance mechanism contributes to these
effects, then larger monetary referral bonuses may decrease the performance
and turnover benefits of referring.
THE FULL TRIAD
One of the areas of mixed results regarding referral applicants is which
job search outcomes are consistently associated with network job searches.
Wages may be higher or lower than nonreferral applicants, or nonmonetary
outcomes may be associated with network job search. Consider the differing
behaviors of referrers depending on the incentive structure they face.
Although referrers tend to be aware of work-relevant characteristics, they
are likely to ignore these characteristics and refer on the basis of friendship
and family ties. They will only attend to these quality characteristics and
refer on the basis of likely worker quality when the incentives are structured
for them to do so. Because it is the hiring firm that commonly creates the
incentive structure for referring, variations across firms’ incentive structures
for referring could generate confusing and contradictory returns to referral
applicants.
The triad perspective offers another alternative explanation for the varied
wage premium finding. Much of the cost-effectiveness of the network
mode of recruiting comes from the behaviors of the referrers. These
referrer-generated savings may contribute to the sometimes-observed wage
benefits of network job seekers. The savings from connectors’ prescreening
of applicants, from referrers’ productivity and turnover benefits from referring, and from bypassing formal recruitment costs all accrue to the hiring
firm. In this case, the sometimes-observed wage bonus from screeners to
referral hires may be an efficiency wage (e.g., Kugler, 2003) rather than wage
benefits accruing from having a higher network-dependent reservation

How Do Labor Market Networks Work?

11

wage (Montgomery, 1991). Then contextual factors reducing a firm’s savings
from referral recruitment would be expected to reduce the apparent wage
benefits of network job search. Considering the three players together
yields an additional explanation for screeners’ apparent preferences for
referral applicants. Earlier, we also indicated that referrers could induce
non-job-seekers to apply for a particular job. Non-search applicants are
both disproportionately currently employed and disproportionately referral
applicants. These factors interact with screeners’ preferences. Job screeners
prefer currently employed workers to currently unemployed individuals
(Kroft, Lange, & Notowidigdo, 2013). Because of this, differences in the
proportion of currently employed versus nonemployed workers among the
job applicants may also contribute to the observed general preference for
referral applicants.
CONCLUSION
Despite many decades of scholarly scrutiny, there is little certainty about the
mechanisms through which labor market networks operate. Much of this
uncertainty results from single-perspective investigations of a fundamentally triadic process. The mechanics of a pulley system has multiple elements:
pulley, rope, mass, anchor, and force agent. The dynamics of the system cannot be understood by examining the pulley in isolation and setting aside its
interdependencies with the other elements. Yet this reductionist approach is
commonly applied to more complex and interdependent social systems such
as labor market networks. The triad of actors defining a network job search
are each of the agentic decision makers. The effects of network search on job
seekers is not independent of the behavioral patterns and choices of job connectors or the preferences of job screeners. The actions of the job connector,
although likely patterned, introduce complexities that cannot be captured in
a two-way matching process. If future research into labor market networks is
to inform strategies and policies at the firm or personal levels, mechanisms
affecting the behavior of any one actor need to be integrated with the mechanisms affecting the other two.
REFERENCES
Barbulescu, R. (2014). The strength of many kinds of ties: Unpacking the role of social
contacts across stages of the job search process. Working Paper, Paris: HEC Paris.
Beaman, L., & Magruder, J. (2012). Who gets the job referral? Evidence from a social
networks experiment. American Economic Review, 102(7), 3574–3593.
Beaman, L., Keleher, N., & Magruder, J. (2013). Do job networks disadvantage
women? Evidence from a recruitment experiment in rural Malawi (Working
paper). Evanston, IL: Northwestern University.

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Bentolila, S., Michelacci, C., & Suarez, J. (2010). Social contacts and occupational
choice. Economica, 77(305), 20–45.
Bewley, T. F. (1999). Why wages don’t fall during a recession. Cambridge, MA: Harvard University Press.
Burks, S., Cowgill, B., Hoffman, M., & Housman, M. (2013). “You’d Be Perfect for This”:
Understanding the value of hiring through referrals. The Institute for the Study of
Labor Discussion Papers, (7382).
Casella, A., & Hanaki, N. (2008). Information channels in labor markets: On the
resilience of referral hiring. Journal of Economic Behavior & Organization, 66(3),
492–513.
Castilla, E. J. (2005). Social networks and employee performance in a call center.
American Journal of Sociology, 110(5), 1243–1283.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013a). Social networks and employment:
Mechanisms (part 1). Sociology Compass, 7(12), 999–1012.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013b). Social networks and employment:
Outcomes (part 2). Sociology Compass, 7(12), 1013–1026.
Chua, V. (2011). Social networks and labour market outcomes in a meritocracy. Social
Networks, 33(1), 1–11.
Cingano, F., & Rosolia, A. (2012). People I know: Job search and social networks.
Journal of Labor Economics, 30(2), 291–332.
Fernandez, R. M. (2010). Creating connections for the disadvantaged: Networks and
labor market intermediaries at the hiring interface. Working paper. Retrieved from
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1576608.
Fernandez, R. M., & Castilla, E. J. (2001). How much is that network worth? Social
capital in employee referral networks. In N. Lin, K. Cook & R. S. Burt (Eds.), Social
Capital: Theory and Research (pp. 85–104). New York, NY: DeGruyter.
Fernandez, R. M., Castilla, E. J., & Moore, P. (2000). Social capital at work: Networks
and employment at a phone center. American Journal of Sociology, 105(5), 1288.
Fernandez, R. M., & Fernandez-Mateo, I. (2006). Networks, race, and hiring. American
Sociological Review, 71(1), 42–71.
Fernandez, R. M., & Galperin, R. V. (2014). The causal status of social capital in labor
markets. Research in the Sociology of Organizations, 40, 445–462.
Fernandez, R. M., & Greenberg, J. (2013). Race, network hiring, and statistical discrimination. Research in the Sociology of Work, 24, 81–102.
Fernandez, R. M., & Sosa, M. L. (2005). Gendering the job: Networks and recruitment
at a call center. American Journal of Sociology, 111(3), 859–904.
Fernandez-Mateo, I. (2007). Who pays the price of brokerage? Transferring constraint
through price setting in the staffing sector. American Sociological Review, 72(2),
291–317.
Festinger, L. (1957). A theory of cognitive dissonance. Palo Alto, CA: Stanford
University Press.
Franzen, A., & Hangartner, D. (2006). Social networks and labour market outcomes:
The non-monetary benefits of social capital. European Sociological Review, 22(4),
353–368.

How Do Labor Market Networks Work?

13

Gërxhani, K., Brandts, J., & Schram, A. (2013). The emergence of employer information networks in an experimental labour market. Social Networks, 35, 541–560.
Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78(6),
1360–1380.
Heath, R. (2013). Why do firms hire using referrals? Evidence from Bangladeshi garment
factories. Working Paper. University of Washington.
Hensvik, L. & Nordström Skans, O. (2013). Social networks, employee selection and labor
market outcomes (No. 2013: 15). Working Paper, IFAU-Institute for Evaluation of
Labour Market and Education Policy.
Kmec, J. A., McDonald, S., & Trimble, L. B. (2010). Making gender fit and “Correcting” gender misfits sex segregated employment and the nonsearch process. Gender
& Society, 24(2), 213–236.
Kroft, K., Lange, F., & Notowidigdo, M. J. (2013). Duration dependence and
labor market conditions: Evidence from a field experiment. Quarterly Journal of
Economics, 128(3), 1123–1167.
Krug, G., & Rebien, M. (2012). Network-based job search: An analysis of monetary and non-monetary labor market outcomes for the low-status unemployed.
Zeitschrift für Soziologie, 41(4), 316–333.
Kugler, A. D. (2003). Employee referrals and efficiency wages. Labour Economics,
10(5), 531–556.
Lin, N., Zhang, Y., Chen, W., Ao, D., & Song, L. (2009). Recruiting and deploying
social capital in organizations: Theory and evidence. Research in the Sociology of
Work, 19, 225–251.
Marin, A. (2012). Don’t mention it: Why people don’t share job information, when
they do, and why it matters. Social Networks, 34(2), 181–192.
Mouw, T. (2003). Social capital and finding a job: Do contacts matter? American Sociological Review, 68(6), 868–898.
Montgomery, J. D. (1991). Social networks and labor-market outcomes: Toward an
economic analysis. American Economic Review, 81(5), 1408–1418.
Neckerman, K., & Fernandez, R. M. (2003). Keeping a job: Network hiring and
turnover in a retail bank. Research in the Sociology of Organizations, 20, 299–318.
Obukhova, E., & Lan, G. (2013). Do job seekers benefit from contacts? A direct test
with contemporaneous searches. Management Science, 59(10), 2204–2216.
Pallais, A. & Sands, E. G. (2014). Why the referential treatment? Evidence from field experiments on referrals. Working Paper.
Pellizzari, M. (2010). Do friends and relatives really help in getting a good job? Industrial and Labor Relations Review, 63(3), 494–510.
Petersen, T., Saporta, I., & Seidel, M. D. L. (2000). Offering a job: Meritocracy and
social networks. American Journal of Sociology, 106(3), 763–816.
Pinkston, J. C. (2012). How much do employers learn from referrals? Industrial Relations: A Journal of Economy and Society, 51(2), 317–341.
Rees, A., Shultz, G. P., Hamilton, M. T., & Taylor, D. P. (1970). Workers and wages in
an urban labor market (pp. 201–202). Chicago, IL: University of Chicago Press.
Rubineau, B., & Fernandez, R. M. (2013). Missing links: Referrer behavior and job
segregation. Management Science, 59(11), 2470–2489.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Rubineau, B. & Fernandez, R. M. (2014). Tipping points: The segregating and desegregating effects of network recruitment. Working paper.
Smith, S. S. (2005). “Don’t put my name on it”: Social capital activation and
job-finding assistance among the black urban poor. American Journal of Sociology,
111(1), 1–57.
Smith, S. S. (2010). A test of sincerity: How black and latino service workers make
decisions about making referrals. The ANNALS of the American Academy of Political
and Social Science, 629(1), 30–52.
Sterling, A. D. (2014). Friendships and search behavior in labor markets. Management
Science, 60(9), 2341–2354.
Topa, G. (2011). Labor markets and referrals. In J. Benhabib, A. Bisin & M. O. Jackson
(Eds.), Handbook of social economics (pp. 1193–1221, Chapter 22). San Diego, CA:
Elsevier.
Trimble, L. B., & Kmec, J. A. (2011). The role of social networks in getting a job. Sociology Compass, 5(2), 165–178.
Yakubovich, V. (2005). Weak ties, information, and influence: How workers find jobs
in a local Russian labor market. American Sociological Review, 70(3), 408–421.
Yakubovich, V., & Lup, D. (2006). Stages of the recruitment process and the referrer’s
performance effect. Organization Science, 17(6), 710–723.

FURTHER READING
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013a). Social networks and employment:
Mechanisms (part 1). Sociology Compass, 7(12), 999–1012.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013b). Social networks and employment:
Outcomes (part 2). Sociology Compass, 7(12), 1013–1026.
Topa, G. (2011). Labor markets and referrals. In J. Benhabib, A. Bisin & M. O. Jackson
(Eds.), Handbook of social economics (pp. 1193–1221, Chapter 22). San Diego, CA:
Elsevier.

BRIAN RUBINEAU SHORT BIOGRAPHY
Brian Rubineau is an Assistant Professor of Organizational Behavior at the
Desautels Faculty of Management at McGill University. His research investigates how informal social dynamics contribute to inequalities in occupations
and labor markets. His research appears in the leading management and sociology journals Management Science and American Sociological Review, among
others. He is the recipient of multiple competitive research grants, and he has
been a Residential Research Fellow at the Institute for the Social Sciences at
Cornell University and a Graduate Fellow at the Institute for Quantitative
Social Science at Harvard University.
ROBERTO M. FERNANDEZ SHORT BIOGRAPHY
Roberto M. Fernandez is the William F. Pounds Professor of Management at
the MIT Sloan School of Management. He currently serves as the Co-Director

How Do Labor Market Networks Work?

15

of the MIT Sloan School’s PhD program in Economic Sociology. He has
extensive experience doing field research in organizations, including an
exhaustive 5-year case study of a plant retooling and relocation. His current
research is on networks, gender, and race inequality at the hiring interface.
He has received numerous research and teaching honors and awards, and
has recently been elected to the American Academy of Political and Social
Sciences.
RELATED ESSAYS
Global Economic Networks (Sociology), Nina Bandelj et al.
Problems Attract Problems: A Network Perspective on Mental Disorders
(Psychology), Angélique Cramer and Denny Borsboom
Migrant Networks (Sociology), Filiz Garip and Asad L. Asad
Interdependence, Development, and Interstate Conflict (Political Science),
Erik Gartzke
The Future of Employment, Wages, and Technological Change (Economics),
Michael J. Handel
The Development of Social Trust (Psychology), Vikram K. Jaswal and Marissa
B. Drell
Herd Behavior (Psychology), Tatsuya Kameda and Reid Hastie
Emerging Trends in Social Network Analysis of Terrorism and Counterterrorism (Sociology), David Knoke
How Networks Form: Homophily, Opportunity, and Balance (Sociology),
Kevin Lewis
Network Research Experiments (Methods), Allen L. Linton and Betsy Sinclair
Culture, Diffusion, and Networks in Social Animals (Anthropology), Janet
Mann and Lisa Singh
Gender and Women’s Influence in Public Settings (Political Science), Tali
Mendelberg et al.
The Role of School-Related Peers and Social Networks in Human Development (Psychology), Chandra Muller
Social Relationships and Health in Older Adulthood (Psychology), Theodore
F. Robles and Josephine A. Menkin
Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth
Wilson



How Do Labor Market Networks
Work?
BRIAN RUBINEAU and ROBERTO M. FERNANDEZ

Abstract
The informal seeking and sharing of job opportunity information via contacts are
the dominant mechanisms for both the supply and demand sides of the labor market. Despite many decades of scholarly scrutiny, we have established little certainty
about the mechanisms through which labor market networks operate. Much of this
uncertainty results from single-perspective investigations of a fundamentally triadic
process. Network-mediated job search is not merely a version of the classic two-way
matching problem with some additional network factors but is rather a three-way
matching problem with three distinct agentic decision makers: the job seeker, the job
screener, and the social contact acting as a connector. This essay summarizes what is
currently known about the operation and consequences of labor market networks,
their mechanisms, and their contextual dependencies. We show how the perspective
of a triad of actors presents new opportunities for resolving current contradictory
empirical findings and areas of ongoing debate. Progress on this topic requires both
careful causal research isolating mechanisms affecting a particular actor and integrative research on how these mechanisms interact among the triad of actors.

The informal seeking and sharing of job opportunity information via contacts
are the dominant mechanisms for both the supply and demand sides of the
labor market. The importance and prevalence of these processes have been
well documented. Previous work has mostly focused on documenting the
importance of labor market networks via their consequences. Surprisingly,
few of these studies have clearly identified the mechanisms underlying labor
market networks’ operation or their contextual contingencies, and even
fewer studies have done so with any degree of causal certainty. Causally
identifying which mechanisms drive labor market networks and under what
conditions they are active is vital to effectively manage or leverage these
processes toward organizational or personal goals. For example, does the use
of informal contacts in job search improve the likelihood of receiving a job
offer? Do firms prefer referred applicants? Do referring employees generate
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

candidates that differ from the applicants generated by formal methods?
What are the scope conditions for any such effects? There are a host of
basic questions about how labor market networks work that still await
conclusive answers. This essay summarizes what is currently known
about the operation and consequences of labor market networks, their
mechanisms, and their contextual dependencies. Rather than seeking to
provide a comprehensive review (for recent reviews, see Castilla, Lan, &
Rissing, 2013a, 2013b; Topa, 2011), this essay focuses on identifying the open
questions that need to be answered in future research. Thus, the research
cited and discussed here are drawn disproportionately from very recent
scholarship providing insights into the causal processes involving labor
market networks. To preview our conclusion, we suggest that a substantial
challenge to past research in this area has been the multifaceted nature
of the phenomenon. Researchers tend to limit their attention to a narrow
subset of the job matching process, necessarily neglecting the effects of
other components. Network-mediated job search is not merely a two-way
matching problem with some additional network factors but is a three-way
matching problem with three distinct agentic decision makers. Progress in
this important field will depend on future research achieving two important
goals. Research must first be designed to isolate these various features of
network processes affecting each actor at the labor market interface. Second,
the findings from this research must be integrated to reveal the collective
operation of the three-actor system defining labor market networks.
THE THREE PLAYERS AND THEIR ROLES
The use of informal contacts in the job matching process requires a convergence in the actions of three players: the connector must provide job opportunity information to the job seeker, the job seeker must act on that information,
and the job screener must decide whether to extend a job offer to the job
seeker.1 These three players form a triad that is fundamental to the existence
and operation of labor market networks. Job seekers work on the supply side,
while job screeners work on the demand side of the labor market. Connectors may work on the supply side, the demand side, or as an intermediary
between the two sides. Connectors who use their contacts to seek out job
opportunities on behalf of a job seeker arguably operate on the supply side
of the labor market. Connectors who are firm employees recruiting for their
employer among their contacts arguably operate on the demand side. Some
1. Screening in organizations is often divided between initial evaluators (often performed by personnel in HR departments) and the hiring manager who makes the ultimate hiring decision. This arrangement
introduces the possibility of actor disagreement (Fernandez, 2010). For simplicity, we set aside issues of
possible misalignment.

How Do Labor Market Networks Work?

3

connectors may act simply as information conduits, not working on anyone’s
behalf. Scholarly attention has not been equally divided across these three
players. Studies examining the importance of networks for job seekers are
numerous and span over a half-century, studies adopting the hiring firm’s
perspective have emerged over the past two decades, and studies explicitly
examining the role of the connector are the most recent of all. In the next three
sections, we summarize and synthesize the research findings on labor market
networks from each of these perspectives.
JOB SEEKERS
How do labor market networks work for job seekers? What are their effects,
mechanisms, and contingencies? Does a job search including networks yield
distinct outcomes compared to a job search excluding networks? Although
there is broad scholarly consensus that network search is consequential for
job seekers, there is disagreement regarding some of those consequences.
Virtually all studies examining job search duration and receiving or accepting a job offer find that job seekers who use networks have shorter job
searches (e.g., Cingano & Rosolia, 2012) and are more likely to receive or
accept a job offer identified via networks (Fernandez & Greenberg, 2013;
Obukhova & Lan, 2013). Tests of the effects of network search on wages show
contradictory results. Some models suggest wage benefits to network job
search (Montgomery, 1991), and a number of empirical studies buttress that
finding (e.g., Hensvik & Nordström Skans, 2013). Other models suggest the
opposite (Krug & Rebien, 2012), and those models find empirical support
as well (Bentolila, Michelacci, & Suarez, 2010). Scholars have proposed
industry-level (Kugler, 2003) or nation-level (Pellizzari, 2010) factors as
contingencies to help resolve such contradictory findings. One of the earliest
contingencies is identified by Granovetter (1973), that is, whether the nature
of the job seekers’ relationships with their contacts is consequential. Indeed,
weaker ties tend to provide novel job opportunity information, and this
novel information mediates the relationship between tie strength and the
likelihood of getting a job via contacts (Yakubovich, 2005). More recently,
examining within-individual variation from data on the multiple job
searches of graduating MBAs Barbulescu (2014) found that while weak ties
were more helpful for learning about job opportunities and getting invited
for interviews, stronger ties were more helpful for converting interviews into
offers (see note 1). Another recently described contingency is the difference
between having contacts while engaging in a job search versus actively using
those contacts in a job search. Many studies posited that having better social
capital (measured in terms of social network opportunities for job search
help) would be a reasonable proxy for mobilizing that capital in job searches.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

But just as the marginal dollar remaining in a bank contributes little to
marginal productivity, unmobilized social capital has little opportunity to
assist a job seeker. Some studies finding positive social spillover effects
consistent with a social capital effect inferred such an effect explicitly (e.g.,
Cingano & Rosolia, 2012) even though no network mobilization data were
available. Other studies found social capital measures not to be associated
with job search outcomes (Mouw, 2003). Helping to resolve this puzzle using
data on multiple contemporaneous job searches, Obukhova and Lan (2013)
found that social capital did not predict contact use, and only contact use
was consequential in improving the chances of interviews and job offers.
Many other contradictory empirical findings still have not been resolved.
For example, does using networks in job search result in jobs that are better
matches with the interests and abilities of the job seeker (Franzen & Hangartner, 2006) or worse matches (Bentolila et al., 2010)? Similar debates continue
about the job satisfaction, performance, and turnover effects of network job
search. Even in areas of empirical agreement, mechanism questions remain.
For example, why does network job search result in shorter job search
durations? Using the framework recently proposed by Castilla et al. (2013a),
contacts may provide resources to the job seeker (e.g., job information or
advocacy), or signals about otherwise difficult-to-observe candidate traits
to the hiring firm (e.g., productivity or similarity to current employees).
Another possibility is that referrals change the temporal dynamics of the
candidate evaluation process. For example, when confronted with a pool of
applicants, employers may tend to respond to referrals first. This behavior
alone could generate many of the apparent beneficial effects of network job
search even absent any employer preferences for network applicants.
JOB SCREENERS: HIRING FIRMS
While the demand side of labor market networks has been understudied
relative to the supply side, a host of studies find that employers prefer referral applicants (e.g., Fernandez, Castilla, & Moore, 2000; Petersen, Saporta, &
Seidel, 2000). What are the mechanisms behind these preferences, and what
is the quality of the evidence supporting those mechanisms?
One question is whether job screeners have a simple preference for referral
applicants versus nonreferral applicants. In a within-individual study leveraging data from a firm where individuals had applied for jobs multiple times
sometimes as referral applicants and sometimes as nonreferral applicants,
Fernandez and Galperin (2014) find a clear preference for referral applicants.
Individuals first applying as nonreferral applicants and then later as referrals were more likely to be interviewed and offered jobs than individuals

How Do Labor Market Networks Work?

5

who remained nonreferral applicants in subsequent job applications. In contrast, individuals who first applied as a referral applicant and then later as
a nonreferral experienced significant decreases in their chances of interview
and job offer.
Why might such a preference manifest? Consistent with Castilla et al.’s
(2013a) framework, there are several signaling-related hypotheses for the
observed preferences. Several empirical studies find referral applicants are
more productive workers (Burks, Cowgill, Hoffman, & Housman, 2013;
Castilla, 2005; Pinkston, 2012), suggesting referral status is a quality signal.
Another common signaling explanation builds upon homophily—the tendency for individuals’ social contacts to be similar to themselves along many
dimensions. Because of homophily, current workers’ referral applicants will
be similar to their referrers—people whom the firm already has chosen to
employ. Many scholars have suggested that homophily on characteristics
important to the organization is a likely reason firms prefer referral applicants (Casella & Hanaki, 2008; Fernandez & Galperin, 2014; Hensvik &
Nordström Skans, 2013). However, field evidence suggests that the belief
that referrals provide benefit via homphily is contingent as some employers
avoid referrals because they see it as leading to cliques that are hard to control
(Bewley, 1999; Rees, Shultz, Hamilton, & Taylor, 1970). For these employers,
the signal value of referring would be negative. Other signaling explanations
do not rely on homophily. On the basis of results from their experimentally
constructed labor markets using students, Gërxhani, Brandts, and Schram
(2013) argue that the fact that another person has referred an applicant signals that the referral occupies a position in an informal information network
characterized by higher levels of trustworthiness, and these trustworthiness
effects would lead employers to prefer referrals. Lin, Zhang, Chen, Ao, and
Song (2009) suggest that referral applicants signal a higher level of social
capital than nonreferral applicants, arguing that this signal of social capital
is likely to be correlated with the social skills that are necessary for certain
types of jobs.
There are also numerous explanations for employers’ preferences for referral applicants that do not involve signalling. Several studies identify performance and productivity benefits for the referrer when their referral is hired
(Burks et al., 2013; Yakubovich & Lup, 2006). Thus, even absent signals from
referrers about referral quality, hiring referral applicants may boost productivity among current employees. In addition, recruiting via referrals is often
less costly (Fernandez & Castilla, 2001), and it yields workers with lower
turnover rates (Burks et al., 2013; Castilla, 2005; Neckerman & Fernandez,
2003). Further, referral applicants are more likely than nonreferral applicants
to become referrers (Fernandez & Fernandez-Mateo, 2006; Fernandez & Sosa,
2005). Preferring referral applicants may reflect an interest in encouraging

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

and investing in that more cost-effective mode of recruitment. Beyond the
anticipation of future savings from referral hires, some scholars posit that the
social ties with current employees entailed by the referring relationship confer additional cost, commitment, monitoring, or even performance benefits
that are not available from nonreferral applicants (Burks et al., 2013; Fernandez et al., 2000; Sterling, 2014).
One notable recent experimental study provides strong empirical support
for both signaling and nonsignaling explanations. Pallais and Sands (2014)
conducted three experiments using an online labor market where subjects
were paid employees—referrers, referrals, and nonreferrals—of three firms
created by the research team. They found strong evidence of homophily:
referral hires’ performance was strongly associated with the performance of
their referrers. They found strong evidence of quality signaling: controlling
for all characteristics observable at application and hire, referral hires outperformed nonreferral hires not only at their initial jobs but also in later work for
a second experimental firm. Finally, they also found strong referrer–referral
relationship effects: when working on team tasks, teams pairing referrers and
referrals outperformed teams with other pairing even when controlling for
all individual characteristics.
The two types of signaling mechanisms described suggest contingencies
as to when referral applicants may or may not be preferred. First, given that
being referred can provide important worker quality signals to the employer,
then in settings where worker quality is completely observable, then referral
applicants would have no signaling advantages over nonreferral applicants
[see Chua’s (2011) study of the state-sector labor market of Singapore].
Second, given the strength and consistency of homophily effects, it would be
understandable if the quality of the referrer affects the employer’s preference
for referral applicants. In their study of an electronically mediated contract
labor market, Yakubovich and Lup (2006) find precisely this outcome.
While referred applicants from higher performing employees received
significantly more favorable outcomes than nonreferral applicants, referral
applicants from lower performing employees received significantly less
favorable outcomes than nonreferral applicants. This homophily contingency, however, has more to do with the referrer than the job screener. In
the discussion of their experiment, Pallais and Sands (2014) describe the
mechanism yielding the enduring work performance premium from referral
hires versus nonreferral hires as “selection.” The actor doing the selecting,
the agent for this mechanism’s operation, is the referrer. That is, the referrer
is aware of otherwise unavailable quality and productivity information—not
wholly explained by homophily—and chooses to refer in part based on that
information. In this way, the referrer is a prescreener acting in a manner that
is likely to benefit the hiring firm. This filtering behavior is not a behavior

How Do Labor Market Networks Work?

7

or choice of the hiring firm, and prompts us to turn our attention to the
least-studied actor in the triad of labor market network actors—the person
connecting the job seeker to the hiring firm—the connector.
JOB CONNECTORS
In addition to the job seeker and job screener, a third role is played by actors
providing the intermediary connection between the supply and demand
sides of the labor market. While our focus here is on network accounts in
which individual actors play the role of network connectors, this function
can also be served by organizations working as labor market brokers
(Fernandez-Mateo, 2007). Connectors are defined by what they do: sharing
job opportunity information with job seekers within their networks. It is the
connector’s sharing of job information that turns a job seeker into a network
job seeker. How do labor market networks work for these job connectors?
Specifically with respect to employee referrals, there is evidence that
worker ability is related to the likelihood of engaging in referring. Hensvik
and Nordström Skans (2013) find high-ability/high-aptitude workers to be
overrepresented among employee referrers. Moreover, these connectors are
not passive conduits for job opportunity information (Marin, 2012; Smith,
2005), but are rather agentic decision makers whose behaviors and choices
directly influence labor market dynamics. In general, referrers see the
performance and success of their referrals as a reflection upon themselves
in the eyes of their employer; such reputation effects are likely to be much
weaker for nonemployee referrals. Smith (2005, 2010) examines the referring
behaviors of racial/ethnic minorities in high-poverty urban areas and finds
employed workers with job opportunity information are quite concerned
with the reputational impacts of their referrals, and thus are highly selective
in deciding with whom among their personal contacts they will share the
information. Recent research showing employers engage in shared wage
punishment of referrers and referrals (Heath, 2013) suggests that such reputational concerns are justified. In Beaman and Magruder’s (2012) experiment
in the Kolkata labor market for experiment subjects, subjects generated
other subjects via referring and under a variety of manipulated scenarios.
They found that referrers are aware of their contacts’ likely productivity,
and will select referral applicants on this worker quality dimension only if
the incentives are constructed such that the referral’s performance affects
the referrer directly. When referral hires’ behaviors have no impact on the
referrers’ work outcomes, the referrers are likely to refer close friends and
family without selecting on worker quality. To the extent that connectors
selectively share job opportunity information, they also explicitly filter their
contacts to select on additional productivity-associated characteristics that

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

may not be otherwise apparent to the employer. Thus, through the act of
sharing information, connectors implicitly provide some of the signaling
benefits valued by employers (Fernandez et al., 2000; Pallais & Sands, 2014).
This raises the intriguing possibility that firms can affect their labor market
outcomes by influencing the behavior of their referring employees.
We have previously suggested that referrers are the missing link to a more
complete understanding of labor market network dynamics (Rubineau &
Fernandez, 2013). Firms commonly offer referral bonuses as an attempt to
influence referring behavior, and as shown in Beaman and Magruder’s (2012)
Kolkata experiment, referrer behavior is sensitive to different referring incentives. The more recent set of studies examining referrer behaviors and outcomes consistently find that after their referral applicant is hired, referrers
are less likely to leave the firm (Burks et al., 2013). In addition, the performance effects of referral hires on their referrers remain strong (Burks et al.,
2013; Castilla, 2005).
Looking forward, numerous questions remain about how connectors
influence job matching. Past research suggests that the presence and quality
of additional assistance depends importantly on the nature of the relationship the referrer has with the job seeker. Unlike Granovetter’s Strength
of Weak Ties for job seekers, referrers appear to provide more assistance
to those job seekers with whom they are more strongly tied (Marin, 2012),
and having a strongly tied referrer was more strongly associated with
getting job offers. Other questions remain as to the role of the connector.
The research discussed suggests that providing incentives to refer leads the
connector to actively seek out candidates. But does this lead the connector
to draw on their existing stock of contacts—likely strong ties—or is this
accomplished by seeking out new contacts that are likely to be weaker ties?
Ongoing research being conducted by Fernandez uses survey vignettes
in an experimental set up to address this question. Other research by
Lin et al. (2009) suggests that the connector’s key point of influence is to
be found in relationships with screeners and hiring managers. Finally, a
promising direction for future research on connectors is to look at the effect
of referring on workplace composition (Trimble & Kmec, 2011). Fernandez
and colleagues (Fernandez & Fernandez-Mateo, 2006; Fernandez & Sosa,
2005) and other researchers (e.g., Beaman et al., 2013) find strong evidence
of demographic homophily between referrers and their referrals. While
past understandings of homophily suggest that this is likely to reinforce
workplace gender or racial segregation, more recent research by Rubineau
and Fernandez (2014) identify the conditions under which recruitment via
employee referrals desegregates. Moreover, they argue that firm policies can
create these conditions by managing referrer behaviors. To the extent that
the firm can incentive the underrepresented group to refer more than the

How Do Labor Market Networks Work?

9

overrepresented group, referring can increase the rate at which referring
desegregates. Recent empirical work supports this view. In their experimental study of referring in Malawi, Beaman, Keleher, and Magruder (2013) find
job segregation to be more sensitive to referring rates than to homophily.
CONVERGENCE AMONG THE PLAYERS
Earlier, we have focused upon each of the actors within the labor market network triad. In reality, all three of these actors work together interdependently
at the labor market interface. Focusing on a single player and setting aside the
effects of the other two players necessarily obscures many dynamics. Some
of the puzzles regarding findings from the perspective of a single actor can be
readily resolved by considering the dynamic interactions with other actors.
Given three actors, there are three pairs of actors that could be considered
together, and one full triad. The job seeker and the job screener pair is the
traditional focus of labor market research. Explicit consideration of the interactions with an agentic connector is needed to move our understanding of
labor market networks forward.
CONNECTORS AND SEEKERS
One area of debate is the absence of a consistent effect of individual social
capital measures on job search outcomes (Mouw, 2003). The using versus
having solution (Obukhova & Lan, 2013) likely represents only part of the
explanation. After all, it is reasonable to think that having more social capital makes it easier to use social capital. The other part of the explanation
may be in recognizing that the agency of the connector plays a key role in
creating a network job search. Connectors may not only be selective among
their job-seeking contacts in sharing job opportunity information, but may
actually induce contacts who are not active job seekers to apply for a particular position (Kmec, McDonald, & Trimble, 2010). Research also shows that
individuals referred to a job opportunity by a social contact feel some social
obligation to apply for the job (Sterling, 2014).
CONNECTORS AND SCREENERS
Just as connectors may create seekers, screeners—in the form of firm
policies—may create connectors. That is, the firm has some control over
whether and which employees learn about a job opportunity within the
firm. Identifying firm policies that could encourage particular employees to
engage in referring would create an important and powerful tool. Managing
referring behavior could have benefits in addition to the desegregating

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

effects discussed earlier. Upon engaging in referring, the referrer demonstrates higher performance and lower turnover. Although the speculative
explanation for this effect focuses on the social tie with the referral hire,
this effect could also be understood in terms of the referrer’s relationship
with the firm. In referring, the referrer is acting as a volunteer recruiter. The
referrer uses her own time to talk-up the firm to her contacts. Field evidence
reported in Fernandez and Galperin (2014) supports the idea that screeners
recognize this effort and are more likely to grant interviews to referrals out of
courtesy to the referring employee. But there might be consequences for the
referring employee as well. Festinger’s theory of cognitive dissonance (1957)
would suggest that taking this action would likely improve the referrer’s
view of and feelings toward the firm. This increased affective commitment to
the firm could contribute to the observed performance and turnover effects
of referring. To the extent that a dissonance mechanism contributes to these
effects, then larger monetary referral bonuses may decrease the performance
and turnover benefits of referring.
THE FULL TRIAD
One of the areas of mixed results regarding referral applicants is which
job search outcomes are consistently associated with network job searches.
Wages may be higher or lower than nonreferral applicants, or nonmonetary
outcomes may be associated with network job search. Consider the differing
behaviors of referrers depending on the incentive structure they face.
Although referrers tend to be aware of work-relevant characteristics, they
are likely to ignore these characteristics and refer on the basis of friendship
and family ties. They will only attend to these quality characteristics and
refer on the basis of likely worker quality when the incentives are structured
for them to do so. Because it is the hiring firm that commonly creates the
incentive structure for referring, variations across firms’ incentive structures
for referring could generate confusing and contradictory returns to referral
applicants.
The triad perspective offers another alternative explanation for the varied
wage premium finding. Much of the cost-effectiveness of the network
mode of recruiting comes from the behaviors of the referrers. These
referrer-generated savings may contribute to the sometimes-observed wage
benefits of network job seekers. The savings from connectors’ prescreening
of applicants, from referrers’ productivity and turnover benefits from referring, and from bypassing formal recruitment costs all accrue to the hiring
firm. In this case, the sometimes-observed wage bonus from screeners to
referral hires may be an efficiency wage (e.g., Kugler, 2003) rather than wage
benefits accruing from having a higher network-dependent reservation

How Do Labor Market Networks Work?

11

wage (Montgomery, 1991). Then contextual factors reducing a firm’s savings
from referral recruitment would be expected to reduce the apparent wage
benefits of network job search. Considering the three players together
yields an additional explanation for screeners’ apparent preferences for
referral applicants. Earlier, we also indicated that referrers could induce
non-job-seekers to apply for a particular job. Non-search applicants are
both disproportionately currently employed and disproportionately referral
applicants. These factors interact with screeners’ preferences. Job screeners
prefer currently employed workers to currently unemployed individuals
(Kroft, Lange, & Notowidigdo, 2013). Because of this, differences in the
proportion of currently employed versus nonemployed workers among the
job applicants may also contribute to the observed general preference for
referral applicants.
CONCLUSION
Despite many decades of scholarly scrutiny, there is little certainty about the
mechanisms through which labor market networks operate. Much of this
uncertainty results from single-perspective investigations of a fundamentally triadic process. The mechanics of a pulley system has multiple elements:
pulley, rope, mass, anchor, and force agent. The dynamics of the system cannot be understood by examining the pulley in isolation and setting aside its
interdependencies with the other elements. Yet this reductionist approach is
commonly applied to more complex and interdependent social systems such
as labor market networks. The triad of actors defining a network job search
are each of the agentic decision makers. The effects of network search on job
seekers is not independent of the behavioral patterns and choices of job connectors or the preferences of job screeners. The actions of the job connector,
although likely patterned, introduce complexities that cannot be captured in
a two-way matching process. If future research into labor market networks is
to inform strategies and policies at the firm or personal levels, mechanisms
affecting the behavior of any one actor need to be integrated with the mechanisms affecting the other two.
REFERENCES
Barbulescu, R. (2014). The strength of many kinds of ties: Unpacking the role of social
contacts across stages of the job search process. Working Paper, Paris: HEC Paris.
Beaman, L., & Magruder, J. (2012). Who gets the job referral? Evidence from a social
networks experiment. American Economic Review, 102(7), 3574–3593.
Beaman, L., Keleher, N., & Magruder, J. (2013). Do job networks disadvantage
women? Evidence from a recruitment experiment in rural Malawi (Working
paper). Evanston, IL: Northwestern University.

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Bentolila, S., Michelacci, C., & Suarez, J. (2010). Social contacts and occupational
choice. Economica, 77(305), 20–45.
Bewley, T. F. (1999). Why wages don’t fall during a recession. Cambridge, MA: Harvard University Press.
Burks, S., Cowgill, B., Hoffman, M., & Housman, M. (2013). “You’d Be Perfect for This”:
Understanding the value of hiring through referrals. The Institute for the Study of
Labor Discussion Papers, (7382).
Casella, A., & Hanaki, N. (2008). Information channels in labor markets: On the
resilience of referral hiring. Journal of Economic Behavior & Organization, 66(3),
492–513.
Castilla, E. J. (2005). Social networks and employee performance in a call center.
American Journal of Sociology, 110(5), 1243–1283.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013a). Social networks and employment:
Mechanisms (part 1). Sociology Compass, 7(12), 999–1012.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013b). Social networks and employment:
Outcomes (part 2). Sociology Compass, 7(12), 1013–1026.
Chua, V. (2011). Social networks and labour market outcomes in a meritocracy. Social
Networks, 33(1), 1–11.
Cingano, F., & Rosolia, A. (2012). People I know: Job search and social networks.
Journal of Labor Economics, 30(2), 291–332.
Fernandez, R. M. (2010). Creating connections for the disadvantaged: Networks and
labor market intermediaries at the hiring interface. Working paper. Retrieved from
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1576608.
Fernandez, R. M., & Castilla, E. J. (2001). How much is that network worth? Social
capital in employee referral networks. In N. Lin, K. Cook & R. S. Burt (Eds.), Social
Capital: Theory and Research (pp. 85–104). New York, NY: DeGruyter.
Fernandez, R. M., Castilla, E. J., & Moore, P. (2000). Social capital at work: Networks
and employment at a phone center. American Journal of Sociology, 105(5), 1288.
Fernandez, R. M., & Fernandez-Mateo, I. (2006). Networks, race, and hiring. American
Sociological Review, 71(1), 42–71.
Fernandez, R. M., & Galperin, R. V. (2014). The causal status of social capital in labor
markets. Research in the Sociology of Organizations, 40, 445–462.
Fernandez, R. M., & Greenberg, J. (2013). Race, network hiring, and statistical discrimination. Research in the Sociology of Work, 24, 81–102.
Fernandez, R. M., & Sosa, M. L. (2005). Gendering the job: Networks and recruitment
at a call center. American Journal of Sociology, 111(3), 859–904.
Fernandez-Mateo, I. (2007). Who pays the price of brokerage? Transferring constraint
through price setting in the staffing sector. American Sociological Review, 72(2),
291–317.
Festinger, L. (1957). A theory of cognitive dissonance. Palo Alto, CA: Stanford
University Press.
Franzen, A., & Hangartner, D. (2006). Social networks and labour market outcomes:
The non-monetary benefits of social capital. European Sociological Review, 22(4),
353–368.

How Do Labor Market Networks Work?

13

Gërxhani, K., Brandts, J., & Schram, A. (2013). The emergence of employer information networks in an experimental labour market. Social Networks, 35, 541–560.
Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78(6),
1360–1380.
Heath, R. (2013). Why do firms hire using referrals? Evidence from Bangladeshi garment
factories. Working Paper. University of Washington.
Hensvik, L. & Nordström Skans, O. (2013). Social networks, employee selection and labor
market outcomes (No. 2013: 15). Working Paper, IFAU-Institute for Evaluation of
Labour Market and Education Policy.
Kmec, J. A., McDonald, S., & Trimble, L. B. (2010). Making gender fit and “Correcting” gender misfits sex segregated employment and the nonsearch process. Gender
& Society, 24(2), 213–236.
Kroft, K., Lange, F., & Notowidigdo, M. J. (2013). Duration dependence and
labor market conditions: Evidence from a field experiment. Quarterly Journal of
Economics, 128(3), 1123–1167.
Krug, G., & Rebien, M. (2012). Network-based job search: An analysis of monetary and non-monetary labor market outcomes for the low-status unemployed.
Zeitschrift für Soziologie, 41(4), 316–333.
Kugler, A. D. (2003). Employee referrals and efficiency wages. Labour Economics,
10(5), 531–556.
Lin, N., Zhang, Y., Chen, W., Ao, D., & Song, L. (2009). Recruiting and deploying
social capital in organizations: Theory and evidence. Research in the Sociology of
Work, 19, 225–251.
Marin, A. (2012). Don’t mention it: Why people don’t share job information, when
they do, and why it matters. Social Networks, 34(2), 181–192.
Mouw, T. (2003). Social capital and finding a job: Do contacts matter? American Sociological Review, 68(6), 868–898.
Montgomery, J. D. (1991). Social networks and labor-market outcomes: Toward an
economic analysis. American Economic Review, 81(5), 1408–1418.
Neckerman, K., & Fernandez, R. M. (2003). Keeping a job: Network hiring and
turnover in a retail bank. Research in the Sociology of Organizations, 20, 299–318.
Obukhova, E., & Lan, G. (2013). Do job seekers benefit from contacts? A direct test
with contemporaneous searches. Management Science, 59(10), 2204–2216.
Pallais, A. & Sands, E. G. (2014). Why the referential treatment? Evidence from field experiments on referrals. Working Paper.
Pellizzari, M. (2010). Do friends and relatives really help in getting a good job? Industrial and Labor Relations Review, 63(3), 494–510.
Petersen, T., Saporta, I., & Seidel, M. D. L. (2000). Offering a job: Meritocracy and
social networks. American Journal of Sociology, 106(3), 763–816.
Pinkston, J. C. (2012). How much do employers learn from referrals? Industrial Relations: A Journal of Economy and Society, 51(2), 317–341.
Rees, A., Shultz, G. P., Hamilton, M. T., & Taylor, D. P. (1970). Workers and wages in
an urban labor market (pp. 201–202). Chicago, IL: University of Chicago Press.
Rubineau, B., & Fernandez, R. M. (2013). Missing links: Referrer behavior and job
segregation. Management Science, 59(11), 2470–2489.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Rubineau, B. & Fernandez, R. M. (2014). Tipping points: The segregating and desegregating effects of network recruitment. Working paper.
Smith, S. S. (2005). “Don’t put my name on it”: Social capital activation and
job-finding assistance among the black urban poor. American Journal of Sociology,
111(1), 1–57.
Smith, S. S. (2010). A test of sincerity: How black and latino service workers make
decisions about making referrals. The ANNALS of the American Academy of Political
and Social Science, 629(1), 30–52.
Sterling, A. D. (2014). Friendships and search behavior in labor markets. Management
Science, 60(9), 2341–2354.
Topa, G. (2011). Labor markets and referrals. In J. Benhabib, A. Bisin & M. O. Jackson
(Eds.), Handbook of social economics (pp. 1193–1221, Chapter 22). San Diego, CA:
Elsevier.
Trimble, L. B., & Kmec, J. A. (2011). The role of social networks in getting a job. Sociology Compass, 5(2), 165–178.
Yakubovich, V. (2005). Weak ties, information, and influence: How workers find jobs
in a local Russian labor market. American Sociological Review, 70(3), 408–421.
Yakubovich, V., & Lup, D. (2006). Stages of the recruitment process and the referrer’s
performance effect. Organization Science, 17(6), 710–723.

FURTHER READING
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013a). Social networks and employment:
Mechanisms (part 1). Sociology Compass, 7(12), 999–1012.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013b). Social networks and employment:
Outcomes (part 2). Sociology Compass, 7(12), 1013–1026.
Topa, G. (2011). Labor markets and referrals. In J. Benhabib, A. Bisin & M. O. Jackson
(Eds.), Handbook of social economics (pp. 1193–1221, Chapter 22). San Diego, CA:
Elsevier.

BRIAN RUBINEAU SHORT BIOGRAPHY
Brian Rubineau is an Assistant Professor of Organizational Behavior at the
Desautels Faculty of Management at McGill University. His research investigates how informal social dynamics contribute to inequalities in occupations
and labor markets. His research appears in the leading management and sociology journals Management Science and American Sociological Review, among
others. He is the recipient of multiple competitive research grants, and he has
been a Residential Research Fellow at the Institute for the Social Sciences at
Cornell University and a Graduate Fellow at the Institute for Quantitative
Social Science at Harvard University.
ROBERTO M. FERNANDEZ SHORT BIOGRAPHY
Roberto M. Fernandez is the William F. Pounds Professor of Management at
the MIT Sloan School of Management. He currently serves as the Co-Director

How Do Labor Market Networks Work?

15

of the MIT Sloan School’s PhD program in Economic Sociology. He has
extensive experience doing field research in organizations, including an
exhaustive 5-year case study of a plant retooling and relocation. His current
research is on networks, gender, and race inequality at the hiring interface.
He has received numerous research and teaching honors and awards, and
has recently been elected to the American Academy of Political and Social
Sciences.
RELATED ESSAYS
Global Economic Networks (Sociology), Nina Bandelj et al.
Problems Attract Problems: A Network Perspective on Mental Disorders
(Psychology), Angélique Cramer and Denny Borsboom
Migrant Networks (Sociology), Filiz Garip and Asad L. Asad
Interdependence, Development, and Interstate Conflict (Political Science),
Erik Gartzke
The Future of Employment, Wages, and Technological Change (Economics),
Michael J. Handel
The Development of Social Trust (Psychology), Vikram K. Jaswal and Marissa
B. Drell
Herd Behavior (Psychology), Tatsuya Kameda and Reid Hastie
Emerging Trends in Social Network Analysis of Terrorism and Counterterrorism (Sociology), David Knoke
How Networks Form: Homophily, Opportunity, and Balance (Sociology),
Kevin Lewis
Network Research Experiments (Methods), Allen L. Linton and Betsy Sinclair
Culture, Diffusion, and Networks in Social Animals (Anthropology), Janet
Mann and Lisa Singh
Gender and Women’s Influence in Public Settings (Political Science), Tali
Mendelberg et al.
The Role of School-Related Peers and Social Networks in Human Development (Psychology), Chandra Muller
Social Relationships and Health in Older Adulthood (Psychology), Theodore
F. Robles and Josephine A. Menkin
Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth
Wilson


How Do Labor Market Networks
Work?
BRIAN RUBINEAU and ROBERTO M. FERNANDEZ

Abstract
The informal seeking and sharing of job opportunity information via contacts are
the dominant mechanisms for both the supply and demand sides of the labor market. Despite many decades of scholarly scrutiny, we have established little certainty
about the mechanisms through which labor market networks operate. Much of this
uncertainty results from single-perspective investigations of a fundamentally triadic
process. Network-mediated job search is not merely a version of the classic two-way
matching problem with some additional network factors but is rather a three-way
matching problem with three distinct agentic decision makers: the job seeker, the job
screener, and the social contact acting as a connector. This essay summarizes what is
currently known about the operation and consequences of labor market networks,
their mechanisms, and their contextual dependencies. We show how the perspective
of a triad of actors presents new opportunities for resolving current contradictory
empirical findings and areas of ongoing debate. Progress on this topic requires both
careful causal research isolating mechanisms affecting a particular actor and integrative research on how these mechanisms interact among the triad of actors.

The informal seeking and sharing of job opportunity information via contacts
are the dominant mechanisms for both the supply and demand sides of the
labor market. The importance and prevalence of these processes have been
well documented. Previous work has mostly focused on documenting the
importance of labor market networks via their consequences. Surprisingly,
few of these studies have clearly identified the mechanisms underlying labor
market networks’ operation or their contextual contingencies, and even
fewer studies have done so with any degree of causal certainty. Causally
identifying which mechanisms drive labor market networks and under what
conditions they are active is vital to effectively manage or leverage these
processes toward organizational or personal goals. For example, does the use
of informal contacts in job search improve the likelihood of receiving a job
offer? Do firms prefer referred applicants? Do referring employees generate
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

candidates that differ from the applicants generated by formal methods?
What are the scope conditions for any such effects? There are a host of
basic questions about how labor market networks work that still await
conclusive answers. This essay summarizes what is currently known
about the operation and consequences of labor market networks, their
mechanisms, and their contextual dependencies. Rather than seeking to
provide a comprehensive review (for recent reviews, see Castilla, Lan, &
Rissing, 2013a, 2013b; Topa, 2011), this essay focuses on identifying the open
questions that need to be answered in future research. Thus, the research
cited and discussed here are drawn disproportionately from very recent
scholarship providing insights into the causal processes involving labor
market networks. To preview our conclusion, we suggest that a substantial
challenge to past research in this area has been the multifaceted nature
of the phenomenon. Researchers tend to limit their attention to a narrow
subset of the job matching process, necessarily neglecting the effects of
other components. Network-mediated job search is not merely a two-way
matching problem with some additional network factors but is a three-way
matching problem with three distinct agentic decision makers. Progress in
this important field will depend on future research achieving two important
goals. Research must first be designed to isolate these various features of
network processes affecting each actor at the labor market interface. Second,
the findings from this research must be integrated to reveal the collective
operation of the three-actor system defining labor market networks.
THE THREE PLAYERS AND THEIR ROLES
The use of informal contacts in the job matching process requires a convergence in the actions of three players: the connector must provide job opportunity information to the job seeker, the job seeker must act on that information,
and the job screener must decide whether to extend a job offer to the job
seeker.1 These three players form a triad that is fundamental to the existence
and operation of labor market networks. Job seekers work on the supply side,
while job screeners work on the demand side of the labor market. Connectors may work on the supply side, the demand side, or as an intermediary
between the two sides. Connectors who use their contacts to seek out job
opportunities on behalf of a job seeker arguably operate on the supply side
of the labor market. Connectors who are firm employees recruiting for their
employer among their contacts arguably operate on the demand side. Some
1. Screening in organizations is often divided between initial evaluators (often performed by personnel in HR departments) and the hiring manager who makes the ultimate hiring decision. This arrangement
introduces the possibility of actor disagreement (Fernandez, 2010). For simplicity, we set aside issues of
possible misalignment.

How Do Labor Market Networks Work?

3

connectors may act simply as information conduits, not working on anyone’s
behalf. Scholarly attention has not been equally divided across these three
players. Studies examining the importance of networks for job seekers are
numerous and span over a half-century, studies adopting the hiring firm’s
perspective have emerged over the past two decades, and studies explicitly
examining the role of the connector are the most recent of all. In the next three
sections, we summarize and synthesize the research findings on labor market
networks from each of these perspectives.
JOB SEEKERS
How do labor market networks work for job seekers? What are their effects,
mechanisms, and contingencies? Does a job search including networks yield
distinct outcomes compared to a job search excluding networks? Although
there is broad scholarly consensus that network search is consequential for
job seekers, there is disagreement regarding some of those consequences.
Virtually all studies examining job search duration and receiving or accepting a job offer find that job seekers who use networks have shorter job
searches (e.g., Cingano & Rosolia, 2012) and are more likely to receive or
accept a job offer identified via networks (Fernandez & Greenberg, 2013;
Obukhova & Lan, 2013). Tests of the effects of network search on wages show
contradictory results. Some models suggest wage benefits to network job
search (Montgomery, 1991), and a number of empirical studies buttress that
finding (e.g., Hensvik & Nordström Skans, 2013). Other models suggest the
opposite (Krug & Rebien, 2012), and those models find empirical support
as well (Bentolila, Michelacci, & Suarez, 2010). Scholars have proposed
industry-level (Kugler, 2003) or nation-level (Pellizzari, 2010) factors as
contingencies to help resolve such contradictory findings. One of the earliest
contingencies is identified by Granovetter (1973), that is, whether the nature
of the job seekers’ relationships with their contacts is consequential. Indeed,
weaker ties tend to provide novel job opportunity information, and this
novel information mediates the relationship between tie strength and the
likelihood of getting a job via contacts (Yakubovich, 2005). More recently,
examining within-individual variation from data on the multiple job
searches of graduating MBAs Barbulescu (2014) found that while weak ties
were more helpful for learning about job opportunities and getting invited
for interviews, stronger ties were more helpful for converting interviews into
offers (see note 1). Another recently described contingency is the difference
between having contacts while engaging in a job search versus actively using
those contacts in a job search. Many studies posited that having better social
capital (measured in terms of social network opportunities for job search
help) would be a reasonable proxy for mobilizing that capital in job searches.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

But just as the marginal dollar remaining in a bank contributes little to
marginal productivity, unmobilized social capital has little opportunity to
assist a job seeker. Some studies finding positive social spillover effects
consistent with a social capital effect inferred such an effect explicitly (e.g.,
Cingano & Rosolia, 2012) even though no network mobilization data were
available. Other studies found social capital measures not to be associated
with job search outcomes (Mouw, 2003). Helping to resolve this puzzle using
data on multiple contemporaneous job searches, Obukhova and Lan (2013)
found that social capital did not predict contact use, and only contact use
was consequential in improving the chances of interviews and job offers.
Many other contradictory empirical findings still have not been resolved.
For example, does using networks in job search result in jobs that are better
matches with the interests and abilities of the job seeker (Franzen & Hangartner, 2006) or worse matches (Bentolila et al., 2010)? Similar debates continue
about the job satisfaction, performance, and turnover effects of network job
search. Even in areas of empirical agreement, mechanism questions remain.
For example, why does network job search result in shorter job search
durations? Using the framework recently proposed by Castilla et al. (2013a),
contacts may provide resources to the job seeker (e.g., job information or
advocacy), or signals about otherwise difficult-to-observe candidate traits
to the hiring firm (e.g., productivity or similarity to current employees).
Another possibility is that referrals change the temporal dynamics of the
candidate evaluation process. For example, when confronted with a pool of
applicants, employers may tend to respond to referrals first. This behavior
alone could generate many of the apparent beneficial effects of network job
search even absent any employer preferences for network applicants.
JOB SCREENERS: HIRING FIRMS
While the demand side of labor market networks has been understudied
relative to the supply side, a host of studies find that employers prefer referral applicants (e.g., Fernandez, Castilla, & Moore, 2000; Petersen, Saporta, &
Seidel, 2000). What are the mechanisms behind these preferences, and what
is the quality of the evidence supporting those mechanisms?
One question is whether job screeners have a simple preference for referral
applicants versus nonreferral applicants. In a within-individual study leveraging data from a firm where individuals had applied for jobs multiple times
sometimes as referral applicants and sometimes as nonreferral applicants,
Fernandez and Galperin (2014) find a clear preference for referral applicants.
Individuals first applying as nonreferral applicants and then later as referrals were more likely to be interviewed and offered jobs than individuals

How Do Labor Market Networks Work?

5

who remained nonreferral applicants in subsequent job applications. In contrast, individuals who first applied as a referral applicant and then later as
a nonreferral experienced significant decreases in their chances of interview
and job offer.
Why might such a preference manifest? Consistent with Castilla et al.’s
(2013a) framework, there are several signaling-related hypotheses for the
observed preferences. Several empirical studies find referral applicants are
more productive workers (Burks, Cowgill, Hoffman, & Housman, 2013;
Castilla, 2005; Pinkston, 2012), suggesting referral status is a quality signal.
Another common signaling explanation builds upon homophily—the tendency for individuals’ social contacts to be similar to themselves along many
dimensions. Because of homophily, current workers’ referral applicants will
be similar to their referrers—people whom the firm already has chosen to
employ. Many scholars have suggested that homophily on characteristics
important to the organization is a likely reason firms prefer referral applicants (Casella & Hanaki, 2008; Fernandez & Galperin, 2014; Hensvik &
Nordström Skans, 2013). However, field evidence suggests that the belief
that referrals provide benefit via homphily is contingent as some employers
avoid referrals because they see it as leading to cliques that are hard to control
(Bewley, 1999; Rees, Shultz, Hamilton, & Taylor, 1970). For these employers,
the signal value of referring would be negative. Other signaling explanations
do not rely on homophily. On the basis of results from their experimentally
constructed labor markets using students, Gërxhani, Brandts, and Schram
(2013) argue that the fact that another person has referred an applicant signals that the referral occupies a position in an informal information network
characterized by higher levels of trustworthiness, and these trustworthiness
effects would lead employers to prefer referrals. Lin, Zhang, Chen, Ao, and
Song (2009) suggest that referral applicants signal a higher level of social
capital than nonreferral applicants, arguing that this signal of social capital
is likely to be correlated with the social skills that are necessary for certain
types of jobs.
There are also numerous explanations for employers’ preferences for referral applicants that do not involve signalling. Several studies identify performance and productivity benefits for the referrer when their referral is hired
(Burks et al., 2013; Yakubovich & Lup, 2006). Thus, even absent signals from
referrers about referral quality, hiring referral applicants may boost productivity among current employees. In addition, recruiting via referrals is often
less costly (Fernandez & Castilla, 2001), and it yields workers with lower
turnover rates (Burks et al., 2013; Castilla, 2005; Neckerman & Fernandez,
2003). Further, referral applicants are more likely than nonreferral applicants
to become referrers (Fernandez & Fernandez-Mateo, 2006; Fernandez & Sosa,
2005). Preferring referral applicants may reflect an interest in encouraging

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

and investing in that more cost-effective mode of recruitment. Beyond the
anticipation of future savings from referral hires, some scholars posit that the
social ties with current employees entailed by the referring relationship confer additional cost, commitment, monitoring, or even performance benefits
that are not available from nonreferral applicants (Burks et al., 2013; Fernandez et al., 2000; Sterling, 2014).
One notable recent experimental study provides strong empirical support
for both signaling and nonsignaling explanations. Pallais and Sands (2014)
conducted three experiments using an online labor market where subjects
were paid employees—referrers, referrals, and nonreferrals—of three firms
created by the research team. They found strong evidence of homophily:
referral hires’ performance was strongly associated with the performance of
their referrers. They found strong evidence of quality signaling: controlling
for all characteristics observable at application and hire, referral hires outperformed nonreferral hires not only at their initial jobs but also in later work for
a second experimental firm. Finally, they also found strong referrer–referral
relationship effects: when working on team tasks, teams pairing referrers and
referrals outperformed teams with other pairing even when controlling for
all individual characteristics.
The two types of signaling mechanisms described suggest contingencies
as to when referral applicants may or may not be preferred. First, given that
being referred can provide important worker quality signals to the employer,
then in settings where worker quality is completely observable, then referral
applicants would have no signaling advantages over nonreferral applicants
[see Chua’s (2011) study of the state-sector labor market of Singapore].
Second, given the strength and consistency of homophily effects, it would be
understandable if the quality of the referrer affects the employer’s preference
for referral applicants. In their study of an electronically mediated contract
labor market, Yakubovich and Lup (2006) find precisely this outcome.
While referred applicants from higher performing employees received
significantly more favorable outcomes than nonreferral applicants, referral
applicants from lower performing employees received significantly less
favorable outcomes than nonreferral applicants. This homophily contingency, however, has more to do with the referrer than the job screener. In
the discussion of their experiment, Pallais and Sands (2014) describe the
mechanism yielding the enduring work performance premium from referral
hires versus nonreferral hires as “selection.” The actor doing the selecting,
the agent for this mechanism’s operation, is the referrer. That is, the referrer
is aware of otherwise unavailable quality and productivity information—not
wholly explained by homophily—and chooses to refer in part based on that
information. In this way, the referrer is a prescreener acting in a manner that
is likely to benefit the hiring firm. This filtering behavior is not a behavior

How Do Labor Market Networks Work?

7

or choice of the hiring firm, and prompts us to turn our attention to the
least-studied actor in the triad of labor market network actors—the person
connecting the job seeker to the hiring firm—the connector.
JOB CONNECTORS
In addition to the job seeker and job screener, a third role is played by actors
providing the intermediary connection between the supply and demand
sides of the labor market. While our focus here is on network accounts in
which individual actors play the role of network connectors, this function
can also be served by organizations working as labor market brokers
(Fernandez-Mateo, 2007). Connectors are defined by what they do: sharing
job opportunity information with job seekers within their networks. It is the
connector’s sharing of job information that turns a job seeker into a network
job seeker. How do labor market networks work for these job connectors?
Specifically with respect to employee referrals, there is evidence that
worker ability is related to the likelihood of engaging in referring. Hensvik
and Nordström Skans (2013) find high-ability/high-aptitude workers to be
overrepresented among employee referrers. Moreover, these connectors are
not passive conduits for job opportunity information (Marin, 2012; Smith,
2005), but are rather agentic decision makers whose behaviors and choices
directly influence labor market dynamics. In general, referrers see the
performance and success of their referrals as a reflection upon themselves
in the eyes of their employer; such reputation effects are likely to be much
weaker for nonemployee referrals. Smith (2005, 2010) examines the referring
behaviors of racial/ethnic minorities in high-poverty urban areas and finds
employed workers with job opportunity information are quite concerned
with the reputational impacts of their referrals, and thus are highly selective
in deciding with whom among their personal contacts they will share the
information. Recent research showing employers engage in shared wage
punishment of referrers and referrals (Heath, 2013) suggests that such reputational concerns are justified. In Beaman and Magruder’s (2012) experiment
in the Kolkata labor market for experiment subjects, subjects generated
other subjects via referring and under a variety of manipulated scenarios.
They found that referrers are aware of their contacts’ likely productivity,
and will select referral applicants on this worker quality dimension only if
the incentives are constructed such that the referral’s performance affects
the referrer directly. When referral hires’ behaviors have no impact on the
referrers’ work outcomes, the referrers are likely to refer close friends and
family without selecting on worker quality. To the extent that connectors
selectively share job opportunity information, they also explicitly filter their
contacts to select on additional productivity-associated characteristics that

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

may not be otherwise apparent to the employer. Thus, through the act of
sharing information, connectors implicitly provide some of the signaling
benefits valued by employers (Fernandez et al., 2000; Pallais & Sands, 2014).
This raises the intriguing possibility that firms can affect their labor market
outcomes by influencing the behavior of their referring employees.
We have previously suggested that referrers are the missing link to a more
complete understanding of labor market network dynamics (Rubineau &
Fernandez, 2013). Firms commonly offer referral bonuses as an attempt to
influence referring behavior, and as shown in Beaman and Magruder’s (2012)
Kolkata experiment, referrer behavior is sensitive to different referring incentives. The more recent set of studies examining referrer behaviors and outcomes consistently find that after their referral applicant is hired, referrers
are less likely to leave the firm (Burks et al., 2013). In addition, the performance effects of referral hires on their referrers remain strong (Burks et al.,
2013; Castilla, 2005).
Looking forward, numerous questions remain about how connectors
influence job matching. Past research suggests that the presence and quality
of additional assistance depends importantly on the nature of the relationship the referrer has with the job seeker. Unlike Granovetter’s Strength
of Weak Ties for job seekers, referrers appear to provide more assistance
to those job seekers with whom they are more strongly tied (Marin, 2012),
and having a strongly tied referrer was more strongly associated with
getting job offers. Other questions remain as to the role of the connector.
The research discussed suggests that providing incentives to refer leads the
connector to actively seek out candidates. But does this lead the connector
to draw on their existing stock of contacts—likely strong ties—or is this
accomplished by seeking out new contacts that are likely to be weaker ties?
Ongoing research being conducted by Fernandez uses survey vignettes
in an experimental set up to address this question. Other research by
Lin et al. (2009) suggests that the connector’s key point of influence is to
be found in relationships with screeners and hiring managers. Finally, a
promising direction for future research on connectors is to look at the effect
of referring on workplace composition (Trimble & Kmec, 2011). Fernandez
and colleagues (Fernandez & Fernandez-Mateo, 2006; Fernandez & Sosa,
2005) and other researchers (e.g., Beaman et al., 2013) find strong evidence
of demographic homophily between referrers and their referrals. While
past understandings of homophily suggest that this is likely to reinforce
workplace gender or racial segregation, more recent research by Rubineau
and Fernandez (2014) identify the conditions under which recruitment via
employee referrals desegregates. Moreover, they argue that firm policies can
create these conditions by managing referrer behaviors. To the extent that
the firm can incentive the underrepresented group to refer more than the

How Do Labor Market Networks Work?

9

overrepresented group, referring can increase the rate at which referring
desegregates. Recent empirical work supports this view. In their experimental study of referring in Malawi, Beaman, Keleher, and Magruder (2013) find
job segregation to be more sensitive to referring rates than to homophily.
CONVERGENCE AMONG THE PLAYERS
Earlier, we have focused upon each of the actors within the labor market network triad. In reality, all three of these actors work together interdependently
at the labor market interface. Focusing on a single player and setting aside the
effects of the other two players necessarily obscures many dynamics. Some
of the puzzles regarding findings from the perspective of a single actor can be
readily resolved by considering the dynamic interactions with other actors.
Given three actors, there are three pairs of actors that could be considered
together, and one full triad. The job seeker and the job screener pair is the
traditional focus of labor market research. Explicit consideration of the interactions with an agentic connector is needed to move our understanding of
labor market networks forward.
CONNECTORS AND SEEKERS
One area of debate is the absence of a consistent effect of individual social
capital measures on job search outcomes (Mouw, 2003). The using versus
having solution (Obukhova & Lan, 2013) likely represents only part of the
explanation. After all, it is reasonable to think that having more social capital makes it easier to use social capital. The other part of the explanation
may be in recognizing that the agency of the connector plays a key role in
creating a network job search. Connectors may not only be selective among
their job-seeking contacts in sharing job opportunity information, but may
actually induce contacts who are not active job seekers to apply for a particular position (Kmec, McDonald, & Trimble, 2010). Research also shows that
individuals referred to a job opportunity by a social contact feel some social
obligation to apply for the job (Sterling, 2014).
CONNECTORS AND SCREENERS
Just as connectors may create seekers, screeners—in the form of firm
policies—may create connectors. That is, the firm has some control over
whether and which employees learn about a job opportunity within the
firm. Identifying firm policies that could encourage particular employees to
engage in referring would create an important and powerful tool. Managing
referring behavior could have benefits in addition to the desegregating

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

effects discussed earlier. Upon engaging in referring, the referrer demonstrates higher performance and lower turnover. Although the speculative
explanation for this effect focuses on the social tie with the referral hire,
this effect could also be understood in terms of the referrer’s relationship
with the firm. In referring, the referrer is acting as a volunteer recruiter. The
referrer uses her own time to talk-up the firm to her contacts. Field evidence
reported in Fernandez and Galperin (2014) supports the idea that screeners
recognize this effort and are more likely to grant interviews to referrals out of
courtesy to the referring employee. But there might be consequences for the
referring employee as well. Festinger’s theory of cognitive dissonance (1957)
would suggest that taking this action would likely improve the referrer’s
view of and feelings toward the firm. This increased affective commitment to
the firm could contribute to the observed performance and turnover effects
of referring. To the extent that a dissonance mechanism contributes to these
effects, then larger monetary referral bonuses may decrease the performance
and turnover benefits of referring.
THE FULL TRIAD
One of the areas of mixed results regarding referral applicants is which
job search outcomes are consistently associated with network job searches.
Wages may be higher or lower than nonreferral applicants, or nonmonetary
outcomes may be associated with network job search. Consider the differing
behaviors of referrers depending on the incentive structure they face.
Although referrers tend to be aware of work-relevant characteristics, they
are likely to ignore these characteristics and refer on the basis of friendship
and family ties. They will only attend to these quality characteristics and
refer on the basis of likely worker quality when the incentives are structured
for them to do so. Because it is the hiring firm that commonly creates the
incentive structure for referring, variations across firms’ incentive structures
for referring could generate confusing and contradictory returns to referral
applicants.
The triad perspective offers another alternative explanation for the varied
wage premium finding. Much of the cost-effectiveness of the network
mode of recruiting comes from the behaviors of the referrers. These
referrer-generated savings may contribute to the sometimes-observed wage
benefits of network job seekers. The savings from connectors’ prescreening
of applicants, from referrers’ productivity and turnover benefits from referring, and from bypassing formal recruitment costs all accrue to the hiring
firm. In this case, the sometimes-observed wage bonus from screeners to
referral hires may be an efficiency wage (e.g., Kugler, 2003) rather than wage
benefits accruing from having a higher network-dependent reservation

How Do Labor Market Networks Work?

11

wage (Montgomery, 1991). Then contextual factors reducing a firm’s savings
from referral recruitment would be expected to reduce the apparent wage
benefits of network job search. Considering the three players together
yields an additional explanation for screeners’ apparent preferences for
referral applicants. Earlier, we also indicated that referrers could induce
non-job-seekers to apply for a particular job. Non-search applicants are
both disproportionately currently employed and disproportionately referral
applicants. These factors interact with screeners’ preferences. Job screeners
prefer currently employed workers to currently unemployed individuals
(Kroft, Lange, & Notowidigdo, 2013). Because of this, differences in the
proportion of currently employed versus nonemployed workers among the
job applicants may also contribute to the observed general preference for
referral applicants.
CONCLUSION
Despite many decades of scholarly scrutiny, there is little certainty about the
mechanisms through which labor market networks operate. Much of this
uncertainty results from single-perspective investigations of a fundamentally triadic process. The mechanics of a pulley system has multiple elements:
pulley, rope, mass, anchor, and force agent. The dynamics of the system cannot be understood by examining the pulley in isolation and setting aside its
interdependencies with the other elements. Yet this reductionist approach is
commonly applied to more complex and interdependent social systems such
as labor market networks. The triad of actors defining a network job search
are each of the agentic decision makers. The effects of network search on job
seekers is not independent of the behavioral patterns and choices of job connectors or the preferences of job screeners. The actions of the job connector,
although likely patterned, introduce complexities that cannot be captured in
a two-way matching process. If future research into labor market networks is
to inform strategies and policies at the firm or personal levels, mechanisms
affecting the behavior of any one actor need to be integrated with the mechanisms affecting the other two.
REFERENCES
Barbulescu, R. (2014). The strength of many kinds of ties: Unpacking the role of social
contacts across stages of the job search process. Working Paper, Paris: HEC Paris.
Beaman, L., & Magruder, J. (2012). Who gets the job referral? Evidence from a social
networks experiment. American Economic Review, 102(7), 3574–3593.
Beaman, L., Keleher, N., & Magruder, J. (2013). Do job networks disadvantage
women? Evidence from a recruitment experiment in rural Malawi (Working
paper). Evanston, IL: Northwestern University.

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Bentolila, S., Michelacci, C., & Suarez, J. (2010). Social contacts and occupational
choice. Economica, 77(305), 20–45.
Bewley, T. F. (1999). Why wages don’t fall during a recession. Cambridge, MA: Harvard University Press.
Burks, S., Cowgill, B., Hoffman, M., & Housman, M. (2013). “You’d Be Perfect for This”:
Understanding the value of hiring through referrals. The Institute for the Study of
Labor Discussion Papers, (7382).
Casella, A., & Hanaki, N. (2008). Information channels in labor markets: On the
resilience of referral hiring. Journal of Economic Behavior & Organization, 66(3),
492–513.
Castilla, E. J. (2005). Social networks and employee performance in a call center.
American Journal of Sociology, 110(5), 1243–1283.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013a). Social networks and employment:
Mechanisms (part 1). Sociology Compass, 7(12), 999–1012.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013b). Social networks and employment:
Outcomes (part 2). Sociology Compass, 7(12), 1013–1026.
Chua, V. (2011). Social networks and labour market outcomes in a meritocracy. Social
Networks, 33(1), 1–11.
Cingano, F., & Rosolia, A. (2012). People I know: Job search and social networks.
Journal of Labor Economics, 30(2), 291–332.
Fernandez, R. M. (2010). Creating connections for the disadvantaged: Networks and
labor market intermediaries at the hiring interface. Working paper. Retrieved from
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1576608.
Fernandez, R. M., & Castilla, E. J. (2001). How much is that network worth? Social
capital in employee referral networks. In N. Lin, K. Cook & R. S. Burt (Eds.), Social
Capital: Theory and Research (pp. 85–104). New York, NY: DeGruyter.
Fernandez, R. M., Castilla, E. J., & Moore, P. (2000). Social capital at work: Networks
and employment at a phone center. American Journal of Sociology, 105(5), 1288.
Fernandez, R. M., & Fernandez-Mateo, I. (2006). Networks, race, and hiring. American
Sociological Review, 71(1), 42–71.
Fernandez, R. M., & Galperin, R. V. (2014). The causal status of social capital in labor
markets. Research in the Sociology of Organizations, 40, 445–462.
Fernandez, R. M., & Greenberg, J. (2013). Race, network hiring, and statistical discrimination. Research in the Sociology of Work, 24, 81–102.
Fernandez, R. M., & Sosa, M. L. (2005). Gendering the job: Networks and recruitment
at a call center. American Journal of Sociology, 111(3), 859–904.
Fernandez-Mateo, I. (2007). Who pays the price of brokerage? Transferring constraint
through price setting in the staffing sector. American Sociological Review, 72(2),
291–317.
Festinger, L. (1957). A theory of cognitive dissonance. Palo Alto, CA: Stanford
University Press.
Franzen, A., & Hangartner, D. (2006). Social networks and labour market outcomes:
The non-monetary benefits of social capital. European Sociological Review, 22(4),
353–368.

How Do Labor Market Networks Work?

13

Gërxhani, K., Brandts, J., & Schram, A. (2013). The emergence of employer information networks in an experimental labour market. Social Networks, 35, 541–560.
Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78(6),
1360–1380.
Heath, R. (2013). Why do firms hire using referrals? Evidence from Bangladeshi garment
factories. Working Paper. University of Washington.
Hensvik, L. & Nordström Skans, O. (2013). Social networks, employee selection and labor
market outcomes (No. 2013: 15). Working Paper, IFAU-Institute for Evaluation of
Labour Market and Education Policy.
Kmec, J. A., McDonald, S., & Trimble, L. B. (2010). Making gender fit and “Correcting” gender misfits sex segregated employment and the nonsearch process. Gender
& Society, 24(2), 213–236.
Kroft, K., Lange, F., & Notowidigdo, M. J. (2013). Duration dependence and
labor market conditions: Evidence from a field experiment. Quarterly Journal of
Economics, 128(3), 1123–1167.
Krug, G., & Rebien, M. (2012). Network-based job search: An analysis of monetary and non-monetary labor market outcomes for the low-status unemployed.
Zeitschrift für Soziologie, 41(4), 316–333.
Kugler, A. D. (2003). Employee referrals and efficiency wages. Labour Economics,
10(5), 531–556.
Lin, N., Zhang, Y., Chen, W., Ao, D., & Song, L. (2009). Recruiting and deploying
social capital in organizations: Theory and evidence. Research in the Sociology of
Work, 19, 225–251.
Marin, A. (2012). Don’t mention it: Why people don’t share job information, when
they do, and why it matters. Social Networks, 34(2), 181–192.
Mouw, T. (2003). Social capital and finding a job: Do contacts matter? American Sociological Review, 68(6), 868–898.
Montgomery, J. D. (1991). Social networks and labor-market outcomes: Toward an
economic analysis. American Economic Review, 81(5), 1408–1418.
Neckerman, K., & Fernandez, R. M. (2003). Keeping a job: Network hiring and
turnover in a retail bank. Research in the Sociology of Organizations, 20, 299–318.
Obukhova, E., & Lan, G. (2013). Do job seekers benefit from contacts? A direct test
with contemporaneous searches. Management Science, 59(10), 2204–2216.
Pallais, A. & Sands, E. G. (2014). Why the referential treatment? Evidence from field experiments on referrals. Working Paper.
Pellizzari, M. (2010). Do friends and relatives really help in getting a good job? Industrial and Labor Relations Review, 63(3), 494–510.
Petersen, T., Saporta, I., & Seidel, M. D. L. (2000). Offering a job: Meritocracy and
social networks. American Journal of Sociology, 106(3), 763–816.
Pinkston, J. C. (2012). How much do employers learn from referrals? Industrial Relations: A Journal of Economy and Society, 51(2), 317–341.
Rees, A., Shultz, G. P., Hamilton, M. T., & Taylor, D. P. (1970). Workers and wages in
an urban labor market (pp. 201–202). Chicago, IL: University of Chicago Press.
Rubineau, B., & Fernandez, R. M. (2013). Missing links: Referrer behavior and job
segregation. Management Science, 59(11), 2470–2489.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Rubineau, B. & Fernandez, R. M. (2014). Tipping points: The segregating and desegregating effects of network recruitment. Working paper.
Smith, S. S. (2005). “Don’t put my name on it”: Social capital activation and
job-finding assistance among the black urban poor. American Journal of Sociology,
111(1), 1–57.
Smith, S. S. (2010). A test of sincerity: How black and latino service workers make
decisions about making referrals. The ANNALS of the American Academy of Political
and Social Science, 629(1), 30–52.
Sterling, A. D. (2014). Friendships and search behavior in labor markets. Management
Science, 60(9), 2341–2354.
Topa, G. (2011). Labor markets and referrals. In J. Benhabib, A. Bisin & M. O. Jackson
(Eds.), Handbook of social economics (pp. 1193–1221, Chapter 22). San Diego, CA:
Elsevier.
Trimble, L. B., & Kmec, J. A. (2011). The role of social networks in getting a job. Sociology Compass, 5(2), 165–178.
Yakubovich, V. (2005). Weak ties, information, and influence: How workers find jobs
in a local Russian labor market. American Sociological Review, 70(3), 408–421.
Yakubovich, V., & Lup, D. (2006). Stages of the recruitment process and the referrer’s
performance effect. Organization Science, 17(6), 710–723.

FURTHER READING
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013a). Social networks and employment:
Mechanisms (part 1). Sociology Compass, 7(12), 999–1012.
Castilla, E. J., Lan, G. J., & Rissing, B. A. (2013b). Social networks and employment:
Outcomes (part 2). Sociology Compass, 7(12), 1013–1026.
Topa, G. (2011). Labor markets and referrals. In J. Benhabib, A. Bisin & M. O. Jackson
(Eds.), Handbook of social economics (pp. 1193–1221, Chapter 22). San Diego, CA:
Elsevier.

BRIAN RUBINEAU SHORT BIOGRAPHY
Brian Rubineau is an Assistant Professor of Organizational Behavior at the
Desautels Faculty of Management at McGill University. His research investigates how informal social dynamics contribute to inequalities in occupations
and labor markets. His research appears in the leading management and sociology journals Management Science and American Sociological Review, among
others. He is the recipient of multiple competitive research grants, and he has
been a Residential Research Fellow at the Institute for the Social Sciences at
Cornell University and a Graduate Fellow at the Institute for Quantitative
Social Science at Harvard University.
ROBERTO M. FERNANDEZ SHORT BIOGRAPHY
Roberto M. Fernandez is the William F. Pounds Professor of Management at
the MIT Sloan School of Management. He currently serves as the Co-Director

How Do Labor Market Networks Work?

15

of the MIT Sloan School’s PhD program in Economic Sociology. He has
extensive experience doing field research in organizations, including an
exhaustive 5-year case study of a plant retooling and relocation. His current
research is on networks, gender, and race inequality at the hiring interface.
He has received numerous research and teaching honors and awards, and
has recently been elected to the American Academy of Political and Social
Sciences.
RELATED ESSAYS
Global Economic Networks (Sociology), Nina Bandelj et al.
Problems Attract Problems: A Network Perspective on Mental Disorders
(Psychology), Angélique Cramer and Denny Borsboom
Migrant Networks (Sociology), Filiz Garip and Asad L. Asad
Interdependence, Development, and Interstate Conflict (Political Science),
Erik Gartzke
The Future of Employment, Wages, and Technological Change (Economics),
Michael J. Handel
The Development of Social Trust (Psychology), Vikram K. Jaswal and Marissa
B. Drell
Herd Behavior (Psychology), Tatsuya Kameda and Reid Hastie
Emerging Trends in Social Network Analysis of Terrorism and Counterterrorism (Sociology), David Knoke
How Networks Form: Homophily, Opportunity, and Balance (Sociology),
Kevin Lewis
Network Research Experiments (Methods), Allen L. Linton and Betsy Sinclair
Culture, Diffusion, and Networks in Social Animals (Anthropology), Janet
Mann and Lisa Singh
Gender and Women’s Influence in Public Settings (Political Science), Tali
Mendelberg et al.
The Role of School-Related Peers and Social Networks in Human Development (Psychology), Chandra Muller
Social Relationships and Health in Older Adulthood (Psychology), Theodore
F. Robles and Josephine A. Menkin
Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth
Wilson



Media

How Do Labor Market Networks Work?