The Impact of International Migration on Food Consumption Pattern and Nutrition: Evidence from Bangladesh

The impact of migration on household wellbeing is a long-standing debate. Many authors find migration has positive impact on the family left behind, while others find negative. A positive channel of impact can be through remittance of income to household remaining in the country, while a negative impact pathway can be through potential labour loss and parental absence. However, household welfare impact can depend on remittance utilizations. Most empirical evaluations find that most of remittance income is used for investment or investment goods. This include education, housing, durable goods, status-oriented consumption goods and a lesser proportion is used for immediate family consumption. However, Quartey (2006) found in Ghana that the poorest migrant households maintain their consumption by remittance. Thus, Migration might affect food consumption pattern and nutrition through increased income (Karamba et. al, 2011). We have checked this hypothesis in a country where, International migration is a common occurrence. Though Bangladesh has achieved remarkable progress in poverty reduction in recent years, the nutritional status of its populace is still lagging behind (David, 2008). One cause is the inability of households to grow or purchase sufficient food for their needs (UNICEF). Like this, in many developing countries, international migrat remittances can play a vital role in increasing the purchasing power of migrant families including for food and can act as insurance during vulnerable periods. Migration is self-selected, hence there is a possibility of endogeneity in statistical evaluations of migration remittance impacts on household income and nutrition. The issue is that those who migrate may come from households who are better endowed with human and social capital characteristics that lead to better household welfare outcomes than households with no family migration and remittance income. The uniqueness of this study is, to overcome the challenges of self-selection in estimating remittance impact on household nutrition outcomes, this is the first study (to our best knowledge) that examined the impact of international migration on food consumption pattern specifically using the instrumental variable (IV) method. Moreover, this study is an important addition to the findings of existing migration literature on the utilization of remittance money and the relation between migration and nutrition. Using the data from a nationally representative survey of Bangladesh, we estimated determinants of household food expenditure, total calorie intake and mix of diet including proportion of calories as protein (e.g. meat, fish and dairy). Considering the degree of migration in Bangladesh, we hypothesized significant positive relationships between international migration overall household’s food calorie consumption and food consumption patterns that positively influence nutrition. We found that international migration households had significantly higher food expenditure and calorie intake from protein and oils/-spices than other households and we observed a noticeable behavioural shift both with food expenditure and calorie consumption pattern (ie. a shifting from grains to proteins and oils/-spices). However, total calorie consumption was non-significantly affected by international migration. Using IV to correct for self-selection reduced the estimate marginal impact of remittances on improved nutritional outcome compared to models that did not correct for self-selection bias.