Causal-inference review of the effects of high-skilled emigration on origin countries. Synthesizes recent quasi-experimental evidence (DID, RD, SSIV, RA) on brain drain vs. brain gain mechanisms across human capital, innovation, trade/FDI, norms, political institutions, and population health.
A US visa expansion for Filipino nurses (2000-2006) caused 9 new nurses to be licensed in the Philippines for every 1 nurse who emigrated, increasing both the tertiary-educated nursing workforce and overall tertiary-educated labor stock. Net brain gain over 2000-2006: ~86,940 nurses (113,775 newly licensed minus 26,835 emigrants).
When the US H-1B visa cap was relaxed (raising US wages of Indian IT workers by ~10%), IT employment in India rose by 5.8%, indicating induced computer-science skill acquisition exceeded the outflow.
Top academic achievers from five small/middle-income countries (Ghana, New Zealand, Tonga, Micronesia, Papua New Guinea) earn $35,000-$79,000 USD per year more after migrating, equivalent to a 53-600% income increase. The review reports a range across countries, not a point estimate.
Domain: High-Skilled International Migration & Origin-Country Effects
Causal effects of high-skilled emigration from lower-income to higher-income countries on the human capital stock, innovation, entrepreneurship, trade/FDI, norms, political institutions, and welfare of origin (sending) countries. Synthesizes ‘brain drain’ vs. ‘brain gain’ mechanisms identified through randomized assignment, difference-in-differences, regression discontinuity, and shift-share IV designs.
Temporal scope: 1990-2024 (review covers studies primarily from 1990s through early 2020s) | Population: Lower- and middle-income migrant-sending countries (sub-Saharan Africa, Caribbean, Pacific, South Asia, Southeast Asia, Latin America, Eastern Europe), with high-skilled emigrants to OECD destinations. Featured cases: Philippines, India, China, Mexico, Cape Verde, Bangladesh, Malawi, Italy, Eastern European EU accession countries.
…and 25 more findings
A US visa expansion for Filipino nurses (2000-2006) caused 9 new nurses to be licensed in the Philippines for every 1 nurse who emigrated, increasing both the tertiary-educated nursing workforce and overall tertiary-educated labor stock. Net brain gain over 2000-2006: ~86,940 nurses (113,775 newly licensed minus 26,835 emigrants).
When the US H-1B visa cap was relaxed (raising US wages of Indian IT workers by ~10%), IT employment in India rose by 5.8%, indicating induced computer-science skill acquisition exceeded the outflow.
Top academic achievers from five small/middle-income countries (Ghana, New Zealand, Tonga, Micronesia, Papua New Guinea) earn $35,000-$79,000 USD per year more after migrating, equivalent to a 53-600% income increase. The review reports a range across countries, not a point estimate.
A one-standard-deviation increase in the post-2008-recession emigration rate of young, skilled Italians reduced new firm creation by 4.8%.
Following EU enlargement (2004/2007), tertiary-educated emigration from Eastern European new-EU countries raised home-market labor costs by 7%. (From the origin-welfare perspective this is a cost-side adjustment, not a welfare gain.)
Following EU enlargement, productivity in the home markets of Eastern European new-EU countries fell by 6% as tertiary-educated workers emigrated.
A 10% increase in a country's high-skilled migrant ethnic network in the US (driven by 1990 US Immigration Act variation) raised manufacturing output in the sending country by 3% (elasticity ~0.30).
A 1% increase in emigration from European countries (driven by EU mobility-law changes) increased patent applications in those origin countries by 0.64% over the subsequent two years (elasticity 0.64).
Returning Fulbright Fellows are cited 90% more in their home countries than a matched control group of nonmigrant scholars.
Return migration of a US-trained African scientist increases the publication output of their nonmigrant scientist colleagues at home by 12% through improved knowledge access and networks.
Internationally mobile Chinese scholars had 7.3% more international collaborators than their nonmobile counterparts.
Cross-border collaborations among researchers in new-EU countries fell after EU enlargement, as researchers with international linkages departed for other EU countries (qualitative DID, magnitude not numerically reported in review).
Chinese provincial talent-return programs increased the valuation and productivity of resident firms whose corporate boards added returnee directors (staggered rollout DID; qualitative magnitudes in review).
Indian Fortune 500 employees randomly assigned (by HR rotation uncorrelated with their characteristics) to managers with US return-migrant experience filed more US patents than peers under non-returnee managers.
A randomized intervention to integrate Cape Verdean immigrants in Portugal had spillover effects on their closest contacts in Cape Verde, increasing support for gender equity in household decision-making by 4-6 percentage points.
The same Cape Verde RCT increased electoral participation among migrants' closest home-country contacts by 12 percentage points relative to contacts of untreated migrants.
Bangladeshi visa-lottery winners (won the right to work in Malaysia) had spouses 148% more likely to be identified as the household head than spouses of unsuccessful entrants.
Bangladeshi visa-lottery winners' households showed a 75% increase in females holding exclusive household decision-making authority, vs unsuccessful-entrant households.
Eliminating Filipina women's ability to work as entertainers in Japan reduced mean household income by 0.5% in Filipino provinces (moving from 25th to 75th percentile in pre-policy province dependence).
Eliminating Filipina entertainer migration to Japan raised the rate of child labor by 2.8 percentage points (25th→75th percentile dependence).
A one-standard-deviation positive remittance shock (driven by exchange-rate movements in migrant destinations) increased Philippine origin-province income per capita by 1,349 PHP (real 2010 pesos) and expenditure per capita by 1,224 PHP — each 0.12 SD increases.
Across 53 African countries, IV-identified physician and nurse emigration rates do NOT lead to substantial reductions in domestic physician and nurse stocks (null finding).
The same 53-African-country IV study cannot detect any worsening of population health (infant mortality or disease prevalence) attributable to physician and nurse emigration (null finding).
Mexican households' marginal propensity to spend remittance income on health care is approximately 6%.
A 1-percentage-point increase in the share of Mexican households containing a return migrant from the US led to a 13% decrease in the share of households without health-care access (SSIV using interior US enforcement variation as the instrument).
The Nigerian government's random assignment of new doctors to rural communities (NYSC program) substantially reduced infant mortality and eliminated the rural-urban gap, indicating that within-country distribution of medical workers — not international emigration — is the binding constraint on rural population health.
Migration-induced changes in social norms around reproductive health in the Philippines led to reduced origin-community fertility and lower infant mortality in shift-share IV analysis using destination reproductive-health policy variation.
Sending international college students to study in high-quality democracies is associated with subsequent improvements in democratic quality in origin countries (cross-country panel association; suggestive but not causally identified).
Low-skilled migration of Malawians to South Africa benefited origin-area education and structural development in the long run, with returnee capital financing nonfarm investments and shifting rural workers from farming to nonfarm work.
Migration from Mexico to the US increased birth weights and reduced infant mortality in origin households, partly due to improved medical knowledge transmitted from migrants.
Vietnamese refugee resettlement in the US created exogenous diaspora variation that increased bilateral US-Vietnam trade flows decades later — natural-experiment evidence that ethnic networks lower trade frictions.
Japanese ethnic networks formed during WWII internment causally increased post-war trade between US counties and Japan, providing further natural-experimental evidence on the migrant-network → trade channel.
US counties with larger ancestry-based migrant networks send more foreign direct investment to migrants' origin countries (shift-share IV evidence on the diaspora→FDI channel).
name: batista-et-al-2025-brain-drain-gain version: 1.0.0 pax_type: paper author: Batista, Catia; Han, Daniel; Haushofer, Johannes; Khanna, Gaurav; McKenzie, David; Mobarak, Ahmed Mushfiq; Theoharides, Caroline; Yang, Dean license: CC-BY-4.0 published_by: Praxis Agent domain: high_skilled_migration constructs: - high_skilled_emigration_rate - human_capital_stock_origin - skill_acquisition_response - remittances_inflow - return_migration - diaspora_network_size - foreign_direct_investment_origin - bilateral_trade_volume - patent_output_origin - scientific_collaboration_intensity - firm_creation_rate - labor_productivity_origin - political_accountability - gender_norms_liberalization - fertility_rate_origin - infant_mortality_origin - household_health_access - household_income_origin - child_labor_origin - birth_weight_origin # … 2 more engines: counts: constructs: 22 findings: 33 propositions: 7 playbooks: 2 sources: 1