Mileage may vary

I wouldn't have expected this. And I couldn't make any sense of it until I saw the line "investments in complements to production" and thought about it for a second. And then it made sense.

The World Bank's released a policy working paper on the effects of cash and in-kind food transfers in Mexico on student learning. Here's the abstract:

This paper studies the medium-term impact of early-life welfare transfers on children’s learning. It studies children who were exposed to the randomized controlled trial of the Mexico’s Food Support Program (the Programa de Apoyo Alimentario, PAL), in which households were assigned to receive cash, in-kind food transfers, or nothing (a control). The children are matched with administrative data on primary school standardized tests, which were taken four to 10 years after the experiment began. The findings show that in-kind transfers did not impact test scores, while cash transfers led to a significant and meaningful decrease in test scores. An analysis of the mechanisms driving these results reveals that both transfers led to an increase in child labor, which is likely detrimental to learning. In-kind food transfers, however, induced a greater consumption of several key micronutrients that are vital for brain development, which likely attenuated the negative impacts of child labor on learning. 

How could a cash transfer to families increase child labour? If the cash enables a rural family to buy more livestock that then requires more on-farm labour from the kids, which reduces the kids hours of schooling and increases their likelihood of attending a lower quality school.