Migration and Development

Remittances, Brain Drain, and the Politics of Movement

Carolina Torreblanca

University of Pennsylvania

Global Development: Intermediate Topics in Politics, Policy, and Data

PSCI 3200 - Spring 2026

Agenda

  1. Types of Migration
  2. Why Do People Migrate?
  3. Who Actually Migrates?
  4. Where Do They Go?
  5. Does Migration Drain Origin Countries?
  6. Yang (2008): Remittances and Household Investment

From Conflict to Migration

Last Class: The Consequences of Civil War

  • Conflict destroys human capital, changes behavior, politicizes people
  • One of the most immediate consequences of conflict is displacement
  • Today: what does migration mean for development?

Types of Migration

Forced vs Voluntary

  • Forced: conflict, persecution, disaster (refugees, asylum seekers, IDPs)
  • Voluntary: economic opportunity, family reunification, education
  • Of ~281 million international migrants, ~36 million are refugees or asylum seekers (~13%, UNHCR 2023)
  • Is the difference btw forced and voluntary clear cut?

The Scale

Why Do People Migrate?

The Place Premium

Clemens, Montenegro, and Pritchett (2008) estimate the wage gap for observably identical workers across borders:

  • A Bolivian worker earns 3x more just by working in the US
  • A Nigerian worker earns 7x more
  • These gaps dwarf returns to education, capital, or any other intervention
  • The biggest price distortion in the world economy is on wages!

So why doesn’t everyone move?

The Opportunity Cost of Staying

  • Same logic as the conflict literature: people weigh costs and benefits
  • The opportunity cost of staying is the wage you could earn elsewhere
  • If wages at home are low, the returns to migration are enormous
  • Migration is an investment: pay upfront costs now, earn returns later

So does that mean the poorest people migrate?

Who Actually Migrates?

What do you think?

If migration is driven by opportunity cost, should the poorest people be the most likely to migrate?

The conventional wisdom in policy circles says yes — and that development will reduce migration

  • Bill Clinton on NAFTA: “as the benefits of economic growth are spread in Mexico … there will be less illegal immigration” (White House 1993)
  • EU Commission (2008): migration policy should “focus much more on … improving the socio-economic situation in low-income and middle-income countries”

Migration as Investment

If migration is an investment, not everyone can afford it:

  • Moving costs money: transport, visas, fees, settling in a new place
  • You need information: where to go, how to get there, who can help
  • You need networks: a cousin, a friend, a community at the destination
  • The poorest people have the most to gain but cannot pay the upfront cost

Two Forces

So what happens as countries develop?

  • More financial reason to stay: wages at home rise, the gap with abroad shrinks
  • But easier to leave: people can afford the trip, transport is cheaper, diaspora networks grow, information flows improve

Which effect dominates?

The Mobility Transition

Clemens (2014) reviews 45 years of evidence:

  • At low incomes, the ability effect dominates: development makes migration more affordable \(\rightarrow\) emigration rises
  • Only after ~PPP $7,000-8,000 does the incentive effect kick in: wages at home are high enough that fewer people want to leave
  • Result: an inverted-U between development and emigration

Macro Evidence: The Migration Hump

Where Do They Go?

Does that mean migrants move to the richest countries?

What would you predict?

Not necessarily.

South-South Migration

South-South Migration

  • South-South migration (105M) is larger than South-North (82M) (UN DESA 2020)
  • Most migration happens between neighboring countries, not across continents
  • Afghan refugees go to Iran and Pakistan, not Germany
  • Proximity, language, networks, and lower barriers matter more than wage gaps

Major Migration Corridors

Does Migration Drain Origin Countries?

Remittances vs Foreign Aid

Remittances Are Huge — But Are They Productive?

  • Remittances exceeded $600 billion in 2022 — larger than all foreign aid (World Bank)
  • But money alone doesn’t guarantee development
  • Do families spend it on consumption (food, goods) or investment (education, business)?

This is what Yang (2008) studies.

Yang (2008)

Setting

  • The Philippines: one of the world’s largest migrant-sending countries
  • Millions of Filipino workers abroad, mostly in the Middle East, East Asia, and the US
  • They send remittances home to their families
  • What do families do with this money?

Research Question

Do remittances increase household investment in human capital and entrepreneurship in origin communities?

What would you predict? Do families spend remittances on consumption (food, goods) or investment (education, business)?

Hypothesis

  • H1: remittance income increases investment in education and entrepreneurship
  • Mechanism: remittances relax credit constraints that prevent poor households from making productive investments
  • Alternative: families might just consume more (better food, nicer house) without changing their long-run trajectory

Selection Problem

  • Migrants are not a random sample of the origin population
  • They tend to be younger, healthier, more educated, more risk-tolerant
  • This creates a measurement problem: if we compare migrants to non-migrants, we’re comparing different kinds of people
  • Sound familiar? (Connect to selection bias in Blattman and Annan)

The Identification Problem

  • Can’t just compare households that receive remittances to those that don’t
  • Households with migrants abroad are different: more connected, more entrepreneurial, possibly wealthier to begin with
  • Need exogenous variation in remittance income

What could serve as an instrument here?

Strategy

  • Filipino migrants work in many different countries
  • Exchange rate shocks: when the currency of the destination country appreciates against the Philippine peso, the same salary is worth MORE in pesos back home
  • These exchange rate movements are driven by global macro forces, not by individual migrant or household decisions
  • Instrument: exchange rate shocks specific to where a household’s migrant works

The Instrument in Detail

Yang uses the 1997 Asian financial crisis:

  • Some currencies appreciated against the peso (USD, Saudi riyal) \(\rightarrow\) remittances worth more
  • Others depreciated (Malaysian ringgit, HK dollar) \(\rightarrow\) remittances worth less
  • Each household gets a different shock depending on where their migrant works
  • Which country the migrant went to was decided before the crisis — not a response to it

Identifying Assumptions

Instrument valid if exchange rate shocks are unrelated to household type

  • Threat: richer households send migrants to dollar-pegged countries \(\rightarrow\) controls for pre-crisis household characteristics
  • Threat: crisis changes labor demand, migrants come home \(\rightarrow\) shows return rates don’t differ across destinations
  • Key: variation comes from which currency your migrant earns in, decided before the crisis

Data

  • Panel of ~1,600 Filipino households with migrants abroad
  • Track household spending, education enrollment, entrepreneurial activity
  • Match each household to the destination country of their migrant
  • Observe exchange rate changes over time

Results: Investment in Education

  • Positive exchange rate shocks (bigger remittances) lead to significantly MORE spending on education
  • Children in recipient households are more likely to stay in school
  • The effect is concentrated in households with school-age children

Results: Entrepreneurship

  • Households receiving larger remittances are more likely to start small businesses
  • More hours worked in self-employment
  • Remittances are functioning as startup capital

Results: NOT Just Consumption

  • The increase in investment is NOT just an increase in overall spending
  • Households shift the COMPOSITION of their spending toward productive investment
  • This is the key finding: remittances change behavior, not just income

Mechanisms

  • Why investment rather than consumption?
  • Remittances relax credit constraints: families couldn’t borrow to send kids to school or start a business, but now they can
  • This is a story about market failures in developing countries: if credit markets worked, families would already be investing

What Do We Learn?

  • Migration generates massive financial flows to developing countries
  • These flows fund human capital and entrepreneurship, not just consumption
  • Remittances may be doing more for development than foreign aid
  • But remember: only families who can send migrants abroad benefit

Discussion

Migration as Development Policy

  • If migration is this effective at reducing poverty and increasing investment, should development policy focus on enabling migration?
  • Remittances dwarf foreign aid: should we facilitate movement rather than send money?
  • If poverty causes conflict (MSS) and remittances alleviate poverty, could facilitating migration help prevent conflict?
  • But migration is politically toxic in destination countries: how do you square the development case with domestic politics?

Discussion

  • Yang shows remittances fund education, but the most educated people leave. Is migration a net positive or negative for origin countries?
  • Migration requires resources: the poorest can’t move. Does this make migration a regressive development tool?
  • How does this connect to the conflict classes: displacement as forced migration vs economic migration as choice?

Next Class

Data Wrangling Workshop

  • No readings
  • Bring your laptop