This study uses a small, non-representative household survey from Tonga (438 households) to examine the income gains from international migration. All empirical studies that analyze the income gains from migration face the methodological problem of the non-random selection of migrants. To meet this problem, the authors use a migrant lottery system whereby New Zealand allows a certain quota of Tongans to migrate each year. This allows the authors to estimate the income gains from migration by comparing the incomes of 3 groups: migrants who were selected in the lottery, and migrated; those who were selected in the lottery, but did not migrate; and those who did not apply to the lottery. Results suggest that Tonga-to-New Zealand migrants are positively selected in terms of both observable and unobservable skills. Results also show that an instrumental variables approach works best in estimating the income gains from migration.