In a 19-year tracking panel survey, representative of the remote Kagera
region of Tanzania, we find that nearly all migration is domestic and
very few people have moved to developed countries. We show that a
migrant’s consumption drops in the face of a shock that hits others in
the extended family network, while the consumption of non-migrants
remains unaffected. This contradicts risk-sharing models based on
reciprocity, but is consistent with assistance driven by social norms.
Despite such unilateral insurance provision and with only minor welfare
differences at baseline, those who migrated out of the region (but
typically within the country) nearly tripled their consumption, while
those who remained at home grew by over 40%. Poverty was virtually
eliminated in this migrant group, but remained at 19% among nonmigrants.
We use these and other stylized facts to speak to three
commonly held views about migration: (i) that migration is a collective
decision made by the extended family or village, (ii) that the extended
family benefits substantially from ‘its’ migrants and (iii) that the
demands for assistance from the extended family form a serious
impediment to growth through migration. We show that these
statements are unlikely to be true for the (domestic) migration
observed in the population under study. What does become clear is
that, after all is said and done, the biggest beneficiaries of migration are
the migrants themselves.