The study aims to primarily examine the effects of parental internal migration on education, specifically on enrollment and educational expenditure within extremely impoverished households. The investigation complements the aim of exploring the impact on income, consumption expenditure, and asset holdings of ultra-poor households in order to test how the impact is distributed across household economic conditions and education. We adopted the difference-in-differences (DID) approach to explore the impact. We also used logistic regression to understand the factors determining internal migration. Results show that the household head's age, household size, and loan burden are the key predictors of internal migration. Internal migration significantly impacts the earnings of ultra-poor households. However, the migration induced expenditure is highly concentrated on nonfood expenses, particularly, the accumulation of business and non-business assets. It reduces the spending on children's education. Also, the migration of parents negatively affects the school enrollment, likley due to the lack of proper parental care. This suggests that internal migration increases the income of poor households and provides them with additional assets, both productive and non-productive. However, the impact of this migration on education remains relatively weak.