Exploiting admission thresholds to the highly-reputed daycare system of Bologna, Italy, we show in a RDD that one additional month in daycare at age 0-2 reduces IQ by 0.5% (4.5% of a s.d.) at age 8-14 in a relatively affluent population. This finding is consistent with the hypothesis, suggested in psychology, that children in daycare experience fewer one-to-one interactions with adults, with negative effects in families where such interactions are of particularly high quality. We show in a model that when parents are offered the most preferred daycare program (as opposed to a less preferred one), daycare attendance increases and parents work more or reduce costly market care. At a high earning potential, this increase in family resources is attractive even if it comes at the cost of child IQ. The model lends structure to our RDD, and it is simulated to show that our estimate would be positive in a less advantaged population. (author abstract)