Public transfers, food consumption and dietary diversity. The case of Tarjeta Uruguay Social

Social protection systems in several Latin American countries are increasingly adopting prepaid magnetic cards to promote the consumption of food and other essential goods. However, little is known about how these transfers affect household expenditure and consumption patterns or their advantages over cash options in high- and middle-income countries. Based on a fuzzy regression discontinuity design that exploits programme assignment rules, we assess the effect of the Uruguayan Tarjeta Uruguay Social (TUS) on food expenditure patterns and a dietary diversity score (DDS). We also analyse non-food expenditure and its components, and three potential explanatory channels: the infra/extra-marginality of the transfer; labour market attachment of the beneficiaries; and self-reported consumption decisions within the household. We do not find effects regarding food expenditure and the DDS, which could be mainly associated with the infra-marginality of the transfer. However, poorer households receiving doubled TUS amounts show positive changes in DDS (3%) and expenditure on fruit (17%) and legumes (9%) relative to a control group. In addition, TUS increases spending on housing for all beneficiaries, which is also reflected in an improvement in the quality of housing materials. Although it does not change the balance between cash and credit purchases, it does lead to a reduction in indebtedness, appears to act as an income stabilizer. These elements suggest that, for most households, TUS behaves like a purely cash transfer.