Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


The Effects of Social Networks on Employment and Inequality

This presentation describes a model of information transmission about job offers along social networks, and its effects on employement and inequality.

Rocío FB

on 10 December 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of The Effects of Social Networks on Employment and Inequality

The importance of social networks in labor markets is generalized and well documented.
A Simple Network Model
Lessons about policy in the presence of social networks
A Look at Policy Implications
Employment evolves as a function of past employment status and the network of connections.
The Dynamics and Patterns of Employment
The Effects of Social Networks on Employment and Inequality
By Anthoni Calvó-Armengol and Matthew O. Jackson
Duration Dependence and Persistence in Unemployment
Dropping Out and Inequality in Employment
The model is expanded by
the network by allowing agents to "drop out".
Begoña Candela Martínez
Rocío Fernández Bastidas
Granovetter (1973,1995) found in a survey of residents of a Massachussets town that over 50% of jobs were obtained through social contacts.
Rees (1966) found numbers of over 60%
in a similar study.
Many studies document similar figures for a variety of occupations, skill levels, and socioeconomic backgrounds.
In this paper, the role of social networks is taken as a way of obtaining
about job opportunities and explore its implications for the dynamics of employment.

Participation in the labor force is different across groups.

Sustained inequality in employment rates, among people remaining in the labor force.

Correlation between employment status inside a network.

Unemployment exhibits duration dependence and persistance.
Negative Conditional Correlation
"1" and "3" are
for job news in the short run
Correlation and Network Structure
a = 0.100
b = 0.015
Bridges and Asymmetries
Structure Matters: Densely Versus Closely Knit Networks
Average path length measures
: how connected among themselves the friends of a given agent are.
If a person has been unemployed for the last X periods, what is the probability she will be employed at the end of this period?
Proposition 3
: Under fine enough subdivisions of periods and starting under the steady-state distribution, the conditional
that an individual will become
in a given period is
with the
of their observed (individual)

This explanation is
complementary to:
Unobserved heterogeneity
Stigma effect
Decline in worker skills
during unemployment
Stickiness in the Dynamics
of Employment
attraction so that the closer it gets to one extreme (high or low employment), the greater is the pull from that extreme.

"Drop-out game"
An agent chooses to stay when the discounted expected future wages exceed the costs.
Drop-out percentages
Drop-out decisions depend on
how well connected
an agent is.
Positive correlation
of employment for
agents implies that more agents participating is better for an agent.
Decisions to stay in
the mark
et are
ategic complements
Drop-out game is
in pure
strategies exists.
Contagion effect
Example: Initial Conditions, Dropouts and Contagion
Contagion effect is larger for the worse starting state and for smaller networks.
A networked model can generate
persistent differences
among two social groups with
economic characteristics except that they differ in their starting state.
Example: Connected Social Groups and Dropouts
Drop-out rates significantly higher for agents initially unemployed, even though the network is connected.
Policies that affect current employment will have both
There is a
positive externality
between the status of connected agents.
If we want to improve the status of individuals, there will be two advantages to
the improvements
in tighter clusters
Transition probabilities of those directly involved improve.
Transition probabilities of those connected with these agents also improve.
Concentrated improvements lead to a greater improvement of the status of connections than disperse improvements.
Drop-out game
all agents decide to drop out. Consider two subsidies:
We pick agents distributed around the network.
We subsidize agents who are clustered together.
Direct impact of the subsidy
Direct impact of the subsidy +
Indirect contagion effect
Example: Concentration of Subsidies
Concentrating effects more locally can have a higher or lower impact, depending on the network configuration.
Employment is positively correlated across time and agents.
Unemployment exhibits duration dependence: probability of getting a job decreases with unemployment's spell.
A group's drop-out rate will be higher and probability of employment lower if staying in the market is costly and that group starts with a worse employment status.
: vector of agents' employment at the end of t.
Transition probabilities
: dependent on the network of relationships.
Employment follows a finite state
Markov process
This model has some features:
Agents decide whether to stay in the labor market network or drop out. This is a once-and-for-all decision (no reentry allowed).
Staying has an expected present value of costs c.
Social networks influence economic
success of individuals at least in part
due to the different
of individuals' networks.
Proposition 1:

Under fine enough subdivision of periods, the unique steady-state long-run distribution on employment is such that the employment statuses of any path-connected agents are positively correlated.
Proposition 2:

Under fine enough subdivisions of periods, starting under the steady-state distribution, the employment statuses of any two path-connected agents are positively correlated across arbitrary periods.
Example: Time Series of Employment for Networked Versus Disconnected Agents
Network structure influences the stickiness of the process and the duration of unemployment.
Simulation in Mathematica:
long-run unemployment rate
Full transcript