This paper is concerned with the extent to which household expenditure patterns are affected by Child Benefit (CB), a transfer payment that depends on the number of children in the household. Despite the fact that CB is cash, we find that it is spent differently than other income not on child assignable goods, but disproportionately on alcohol. We find, surprisingly, that this effect is much larger for couples that for lone mothers but this would be consistent with the idea that parents free-ride when it comes to child quality investment. Thus, our evidence suggests that the answer to our question is that - it is parents who benefit from CB. This would be consistent with two extreme views: parents place little weight on the welfare of their children; or parents place so much weight that they fully insure their children against shocks. We decompose CB variation into anticipated (inflation driven) variation, and (reform driven) surprises. We find that the alcohol spending result is driven by surprises consistent with the view that parents are altruistic towards their children.

Download full version (PDF 678 KB)