In the Dutch national cohort study pre-COOL, children (N=2500) participating in different types of early childhood education and care (ECEC) were followed from age 2 to age 12, at the end of primary school.  In two complementary analysis approaches, we assessed the effects of ECEC on children’s cognitive and social-emotional development. Growth modelling, encompassing the period from 2 to 6 years, revealed significant catch-up effects for children from low SES and immigrant families in language, executive function and self-control relative to a non-risk comparison group. Cross-lag path analysis was applied to a subsample of children (N=1200) in long day childcare to determine the effects of age of onset and intensity of daycare use on age 12 cognitive (reading, math) and social-emotional outcomes (externalizing behavior, self-regulation). The results revealed small but significant negative effects on social-emotional outcomes of intensive daycare in the period 0-2 years and positive effects of intensive daycare in the period 2-4 years on reading comprehension. Observed process quality and implemented early literacy activities were found to moderate the effects on onset and intensity. With higher quality, the negative social-emotional effects disappeared and the positive cognitive effects were stronger. Also child personality characteristics moderated the effects of onset and intensity, showing in general that children at risk were better off when starting later and using child care less intensively in the period 0-2 years. The findings will be discussed in the context of recent quasi-experimental research on very early daycare with middle to long-term follow-up assessments in Denmark, France, Germany, Italy, Norway and Quebec.