ECC3841 Chap.9 Development Through Social Networks
Development Through Social Networks
Development networks replace or supplement incomplete markets. Weak ties can carry nonredundant job and migration information; strong ties can provide money, housing and enforcement. Repeated communities support informal insurance, while peer observation can reduce uncertainty about technologies. These benefits are not universal: correlated shocks, exclusion and redundant information can weaken them.
Social learning can also create rational but incorrect cascades when early actions outweigh private signals.
What this chapter covers
- 01
Job information through contacts
- 02
Weak and strong ties
- 03
Migration information and assistance
- 04
Informal insurance and commitment
- 05
Correlated shocks
- 06
Technology adoption and social learning
- 07
Sequential information cascades
- 08
Network seeding and spillover measurement
Worked example · free
Compute early signal histories
- +1Correct agreement is 0.7 squared, equal to 0.49.
- +1Wrong agreement is 0.3 squared, equal to 0.09.
- +1Disagreement is two times 0.7 times 0.3, equal to 0.42.
- +1Check that the three probabilities sum to one.
Key terms
- Weak tie
- A less intensive connection often valuable for nonredundant information and reach.
- Strong tie
- A close relationship often capable of providing costly support and enforcement.
- Informal insurance
- Risk sharing through repeated community transfers outside a formal insurer.
- Social learning
- Updating beliefs or actions using information observed from others.
- Information cascade
- A sequence where public history leads later people to disregard private signals.
- Technology seeding
- Giving information or access to selected nodes to generate network spillovers.
- Market incompleteness
- A condition where formal markets do not fully provide information, credit, insurance or enforcement. Social connections can partly replace the missing service while excluding people outside the relevant community.
- Migration assistance
- Concrete help with travel, housing, documents or destination settlement supplied through ties. It differs from job information and is often more plausibly carried by trusted strong relationships.
- Risk sharing
- Transfers from members experiencing favorable shocks to members experiencing adverse shocks. Repetition and threatened exclusion can sustain cooperation when formal insurance and enforceable contracts are unavailable.
- Correlated shock
- An adverse event affecting many connected households at the same time. Local insurance performs poorly against correlated risk because potential helpers simultaneously need resources themselves.
- Adoption constraint
- A barrier such as credit, risk, skill or input access that prevents use of a technology despite awareness. Peer information cannot generate adoption when another constraint remains binding.
- Network interference
- A setting where one persons treatment changes another persons outcome through links. It is substantively important for spillovers but violates an assumption of independent individual outcomes.
- Commitment problem
- The temptation to refuse a promised transfer after receiving a favorable shock. Repeated interaction and future exclusion can support informal insurance when courts cannot enforce the arrangement.
- Treatment spillover
- An outcome change among untreated people caused by their connection to treated people. Estimating it requires a network exposure rule and a design that preserves a meaningful comparison.
- Credible messenger
- A network member whose information is trusted by intended recipients. Credibility can make a moderate-degree seed more effective than a highly connected outsider when adoption depends on belief as well as awareness.
- Binding constraint
- The barrier that currently prevents an outcome even if another friction is removed. Better information cannot create adoption when credit, risk, inputs or implementation skill remains the limiting factor.
Development Through Social Networks FAQ
Why do networks matter when markets are incomplete?
They can transmit information, enforce transfers and provide assistance when formal credit, insurance and labor-market institutions are limited. Link the network channel to the missing market and state the condition that limits the benefit.
Are weak ties always better for jobs?
Weak ties often provide novel information, but strong ties may offer credible referrals or material support. Their roles can be complementary. Link the network channel to the missing market and state the condition that limits the benefit.
How can an incorrect herd form?
Unlucky early private signals can create public evidence strong enough for later decision-makers to ignore their own information. Link the network channel to the missing market and state the condition that limits the benefit.
Who should receive a technology demonstration?
Choose nodes with credible, nonredundant reach under the actual learning mechanism. Raw degree alone may waste exposure inside an already informed cluster. Link the network channel to the missing market and state the condition that limits the benefit.
Exam move
Map each application to a market friction before discussing a network. Job contacts and migration ties move information; strong ties may provide housing or finance; repeated communities enforce informal insurance; observed adopters reduce uncertainty about technologies. State what the formal market fails to supply and what the relationship substitutes for.
Separate weak and strong ties by function rather than assuming one is better. Weak ties can carry nonredundant vacancies or destination information, while strong ties can make costly assistance credible. They may complement one another. For informal insurance, identify the shock, the transfer and the enforcement device.
Ask whether shocks are idiosyncratic or correlated, because a whole community cannot insure itself against a loss hitting everyone simultaneously. For social learning, distinguish private signals from observed actions. Enumerate early signal sequences, multiply independent probabilities and check that mutually exclusive events sum to one. Then translate signals into public histories.
Once choices respond to previous choices, later agreement may add little new information. State whether outcomes are ever revealed, because performance feedback can break a cascade. Technology-adoption evidence needs controls for common soil, prices, extension access and credit constraints. A positive peer correlation is not automatically learning.
In policy questions, distinguish the direct recipient, actual conversations, newly informed neighbors and eventual adopters. Measure the network before intervention and record intermediate transmission so a null effect can be diagnosed. Compare random seeding with degree, bridging and trust-based targeting. Marginal reach matters more than raw reach when neighborhoods overlap.
Distribution also matters: a network can improve allocation for insiders while excluding equally productive outsiders. During revision, build four one-page maps for jobs, migration, insurance and adoption. Put the market failure on the left, the tie function in the middle and the welfare benefit plus exclusion risk on the right.
Work one original cascade probability with a new signal accuracy and explain how unlucky early information can create a rational wrong herd. Then design an experiment that separates direct treatment from spillovers and names the interference assumption. End each answer with the condition under which the network stops helping, such as correlated shocks, low trust, redundant information or binding finance.
Working through Development Through Social Networks in ECC3841? Sia is AskSia’s AI Economics tutor — ask any ECC3841 Development Through Social Networks question and get a clear, step-by-step explanation grounded in how ECC3841 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
